Loss of tumor suppressor miR-126 contributes to the development of hepatitis B virus–related hepatocellular carcinoma metastasis through the upregulation of ADAM9

Abstract

Hepatocellular carcinoma is the most common histological type of primary liver cancer, which represents the second leading cause of cancer-related mortality. MiR-126 was reported to be downregulated in hepatocellular carcinoma tissues, compared with its levels in noncancerous tissues. However, baseline miR-126 expression levels in hepatitis B virus–related hepatocellular carcinoma patients who did not undergo pre-operational treatment remains unknown since hepatitis B virus infection and pre-operational transcatheter arterial chemoembolization were shown to upregulate miR-126 expression. Here, we demonstrated that miR-126 is generally downregulated in a homogeneous population of pre-operational treatment–naïve hepatitis B virus–related hepatocellular carcinoma patients (84.0%, 84/100), and its expression is significantly associated with pre-operational alpha-fetoprotein levels (p < 0.05), microvascular invasion (p < 0.05), tumor metastasis (p < 0.05), as well as early recurrence (12 months after surgery; p < 0.01). Furthermore, the results of our study revealed that miR-126 is negatively correlated with ADAM9 expression in hepatitis B virus–related hepatocellular carcinoma patients. Overexpression of miR-126 was shown to attenuate ADAM9 expression in hepatocellular carcinoma cells, which subsequently inhibits cell migration and invasion in vitro. In addition, Cox proportional hazards regression model analysis showed that ADAM9 levels, tumor number, microvascular invasion, and tumor metastasis rate represent independent prognostic factors for shorter recurrence-free survival. In conclusion, we demonstrated that the loss of tumor suppressor miR-126 in hepatitis B virus–related hepatocellular carcinoma cells contributes to the development of metastases through the upregulated expression of its target gene, ADAM9. MiR-126-ADAM9 pathway–based therapeutic targeting may represent a novel approach for the inhibition of hepatitis B virus–related hepatocellular carcinoma metastases.

Keywords

MiR-126 hepatocellular carcinoma metastasis hepatitis B virus ADAM9

Introduction

Hepatocellular carcinoma (HCC) is the most common primary liver cancer type, the fifth most common type of all malignancies worldwide, and the second leading cause of cancer-related mortality.¹ Hepatic resection is the mainstay of treatment of noncirrhotic patients and cirrhotic patients with well-preserved liver function.² However, despite successful surgical resection and the use of anti-viral drugs, tumor recurrence complicates 70% of HCC cases in 5 years after the resection as a result of either intrahepatic and extrahepatic metastasis from the primary lesion or multicentric occurrence.^2,3 Undoubtedly, determining the molecular mechanism underlying HCC metastases may contribute to the improved prevention and management of post-operative recurrence after HCC resection.

Tumor metastases may be caused by genomic instability, and they are characterized by a high frequency of gene mutations which are believed to lead to the development of different cellular phenotypes, which further lead to the development of metastases.⁴ MicroRNAs (miRNAs), a class of small, endogenous, regulatory RNAs, are believed to collectively regulate one third of the genes by forming base pairs with 3′ untranslated region (3′-UTR) of the messenger RNA (mRNA) of specific genes.⁵ It was reported that the deregulation of many miRNAs affects carcinogenesis and tumor progression by epigenetically modulating critical oncogenes and tumor suppressor genes.⁶

MiR-126, also known as miR-126-3p, is derived from the seventh intron of its host gene, epidermal growth factor like domain 7 (EGFL7), which is located at human chromosome 9q34.3.⁷ It was reported that in the highly vascularized tissues, such as heart, liver, and lungs, and in the cells of the endothelial lineage, abundant levels of miR-126 can be detected, and it plays significant roles in developmental angiogenesis and vascular integrity.⁸ However, reduced miR-126 expression was observed in various types of cancers originating from colorectal, gastric, pancreatic, thyroid, breast, bladder, renal, and esophageal cells, with the exception of leukemias, where miR-126 expression varies depending on a subtype.⁹ As for miR-126 expression in HCC, Chen et al.¹⁰ demonstrated that the expression of miR-126 is lower in the tissues of HCC patients with post-liver transplantation recurrence than that in the patients in which the recurrence was not observed. Furthermore, it was shown that miR-126 expression is significantly downregulated in HCC tissues in contrast with the matching noncancerous tissues.^11,12 However, a recent study suggested that pre-operational transcatheter arterial chemoembolization (TACE) in HCC patients may lead to upregulation of miR-126 expression.¹³ To the best of our knowledge, previous studies did not specify whether the patients who had received pre-operational TACE treatment (and/or other therapeutic approaches) were excluded.^10–12 In addition, Ghosh et al.¹⁴ demonstrated that transfection of HepG2 cells with full-length hepatitis B virus (HBV) significantly induced miR-126 expression. These results caused us to consider the pre-treatment miR-126 expression levels in HBV-related HCC patients.

ADAM9, also known as MDC9 (metalloprotease/disintegrin/cysteine-rich protein 9) or meltrin-γ, is one of the ADAM (a disintegrin and metalloproteinase) members.¹⁵ It was reported to be overexpressed in various cancer types, including breast,¹⁶ renal cell,¹⁷ pancreatic,^18,19 gastric,²⁰ prostate,²¹ and liver cancers,^22,23 and represents an independent prognostic marker for recurrence-free survival and/or overall survival in some types of cancer.^21,24 ADAM9 contains all protein domains discovered in most metalloproteinase disintegrins and plays critical roles in the proteolytic processing of multiple membrane-bound molecules, including cytokines, growth factors, chemokines, and adhesion molecules, via its metalloproteinase domain and in promoting cell migration and invasion through the interactions with integrins, via disintegrin-like domain.¹⁵ Considering its importance in cell–cell and tumor–stromal interactions, ADAM9 may represent an attractive therapeutic target in tumors where it is overexpressed.¹⁵

Here, we investigated miR-126 and ADAM9 expression patterns in a homogeneous population of HBV-related HCC patients who did not undergo pre-operational treatment. In the combination with cell-based assays, we attempted to elucidate the potential mechanism underlying the development of tumor metastases involving miR-126-ADAM9 pathway.

Materials and methods

Patients and tissue specimens

HCC tissues and corresponding noncancerous tissues (exceeding the edge of the tumor by at least 2 cm) were obtained from 100 consecutive treatment-naive HBV-related HCC patients who underwent curative resection for HCC in the Department of Hepatobiliary Surgery, Nanfang Hospital Affiliated to Southern Medical University, Guangzhou, China, between November 2010 and May 2015. All patients satisfied the following inclusion criteria: (a) they had not received any other treatments before this surgery; (b) HBV-related HCC, without hepatitis C virus (HCV) infection, was diagnosed pathologically and serologically; (c) the surgical margins were confirmed to contain no residual carcinoma tissue; and (d) complete medical records were available. The specimens were analyzed using real-time quantitative reverse transcription polymerase chain reaction (qRT-PCR) and immunohistochemical staining. This study and all protocols used were approved by the Ethical Committee of Nanfang Hospital at Southern Medical University, and they were performed according to the Declaration of Helsinki (6th revision, 2008). Informed consents were obtained from all the patients.

Patient follow-up

Patients were monitored for further survival analysis via outpatient follow-up or telephonic interviews. Metastasis was defined as the recurrence time less than 6 months after the surgery, the presence of lymph node invasion or extrahepatic invasion, and the pathological diagnosis of tumor embolism (including the portal vein, biliary tract, and hepatic vein thrombus). A relapse was diagnosed based on increased post-operative blood alpha-fetoprotein (AFP) levels and imaging tests (ultrasonic examination, computed tomography scan, or magnetic resonance imaging). Of the 100 patients enrolled in this study, 93 were eligible for prognosis and regression analysis, 2 (2%) died within the peri-operative period due to post-operative complications, and 5 (5%) were lost to the follow-up. The follow-up started at the date of surgery and was terminated on 31 August 2016.

Cell lines and culture conditions

Immortalized liver cell line LO2, liver cancer cell lines HepG2, Hep-3B, Huh7, SMMC-7721 (Institutes of Biological Sciences, Chinese Academy of Sciences, Shanghai, China), MHCC-97H, MHCC-97L, and MHCC-LM3 (obtained as a gift from Liver Cancer Research Institute, Zhongshan Hospital, Fudan University, Shanghai, China) were maintained in Dulbecco’s Modified Eagle’s Medium (DMEM; Gibco, CA, USA). SNU-387 and SNU-449 cells (American Type Culture Collection, VA, USA) were cultured in RPMI-1640 medium (Gibco). Both the culture media were supplemented with 10% fetal bovine albumin (FBS; Gibco). Cells were maintained at 37°C in a humidified atmosphere with 5% CO₂. Baseline miR-126 expression levels were determined in all cell lines mentioned above, and based on this, HepG2 and SNU-449 cell lines were selected for further experiments.

Cell transfection

Hsa-miRNA-126 mimic, negative control (NC) mimic, ADAM9 small interfering RNA (siRNA; target site: 5′-AAGAGCCUCUGAAAUGUGGAG-3′), and NC siRNA were synthesized by RiboBio (Guangzhou, China). HepG2 and SNU-449 cells were transfected with these mimic/siRNA (50 nM) using Lipofectamine 3000 transfection reagent (Invitrogen, CA, USA), according to the manufacturer’s instructions. The analyzed groups for miR-126 overexpressing assays were as follows: miR-126 (miR-126 mimic + Lipofectamine 3000), NC (NC mimic + Lipofectamine 3000), Mock (Lipofectamine 3000); for ADAM9 knockdown assays, si-ADAM9 (ADAM9 siRNA + Lipofectamine 3000), NC (NC siRNA + Lipofectamine 3000), and Mock (Lipofectamine 3000) were used. After 48–72 h of incubation, the cells were further analyzed.

qRT-PCR

Total RNA was extracted from tissue specimens and cells using RNAiso Plus reagent (TaKaRa, Dalian, China), while complementary DNA (cDNA) was synthesized from total RNA using PrimeScript First-Strand cDNA Synthesis Kit (TaKaRa), according to the manufacturers’ instructions. The samples were analyzed using SYBR Real-Time PCR (TaKaRa) based on the LightCycler 480 II System (Roche, Basel, Switzerland). For the analysis of miR-126 expression, RT primer mix (containing oligo-dT primer and random hextamers) was replaced with RT primers (5′-GTCGTATCCAGTGCAGGGTCCGAGGTATTCGCACTGGATACGACCGCATT-3′ for miR-126 and ′-AACGCTTCACGAATTTGCGT-3′ for U6 small nuclear RNA (snRNA)) while preparing the template. The qRT-PCR primers were as follows—miR-126: 5′-TTGGCGGTCGTACCGTGAGTAAT-3′ (forward) and 5′-ATCCAGTGCAGGGTCCGAGG-3′ (reverse); U6: 5′-CTCGCTTCGGCAGCACA-3′ (forward) and 5′-AACGCTTCACGAATTTGCGT-3′ (reverse); ADAM9: 5′-AAAGGATGAAGAGGAAGAGCC-3′ (forward) and 5′-CCAAGTAGTTTGCCAGGAGAA-3′ (reverse); and glyceraldehyde 3-phosphate dehydrogenase (GAPDH): 5′-CAGGAGGCATTGCTGATGAT-3′ (forward) and 5′-GAAGGCTGGGGCTCATTT-3′ (reverse). U6 was used as an endogenous control for the evaluation of miR-126 expression, while ADAM9 levels were normalized to those of GAPDH. All primers were synthesized by Sangon (Guangzhou, China). Data were analyzed using the 2^−ΔΔCt method.²⁵

Western blot analysis

Total cell lysates were obtained from cells kept on ice after 48–72 h of incubation using radioimmunoprecipitation assay (RIPA) buffer supplemented with protease inhibitors, phenylmethanesulfonyl fluoride (PMSF) solution. Bicinchoninic acid (BCA) protein assay kit (Biocolor, Shanghai, China) was used to determine the protein concentration in the samples, and whole lysates were mixed with 5× sodium dodecyl sulfate (SDS) loading buffer (Beyotime, Shanghai, China) at a ratio of 1:4. After denaturation, equal amounts of protein from each sample were separated on a 10% SDS–polyacrylamide gel electrophoresis (PAGE) and transferred onto polyvinylidene difluoride (PVDF) membranes (Millipore, Bedford, Germany). Membranes were blocked with 5% non-fat milk for 1 h, which was followed by the incubation for overnight at 4°C with the following primary antibodies: anti-ADAM9 rabbit polyclonal antibody (1:1000; Abgent, CA, USA) and anti-ACTB rabbit polyclonal antibody (1:1000; BBI Life Sciences, Shanghai, China). The following day, the membranes were incubated with horseradish peroxidase (HRP)-conjugated anti-rabbit secondary antibody (1:10000; Earthox, Shanghai, China) at room temperature for 1 h. Subsequently, the immunoblotted bands were visualized with ECL Plus detection system (Millipore, MA, USA) and FluorChem E system (ProteinSimple, CA, USA).

Immunohistochemical staining

HCC tissue samples were fixed with 10% neutral formalin and embedded in paraffin. Immunohistochemical staining was performed on 4-µm-thick sections, as reported previously.²⁶ Anti-ADAM9 primary antibody (1:200; Abgent) was used for the detection of ADAM9 in the cytoplasm of HCC cells. Immunoreactivity was scored according to the staining intensity (0: negative, 1: weak, 2: moderate, and 3: strong) and the percentage of positive cells (1: 1%–25% positive cells, 2: 26%–49% positive cells, 3: 50%–75% positive cells, 4: >75% positive cells). Finally, ADAM9 expression was evaluated semi-quantitatively by multiplying staining intensity and positive cell percentage scores.

Cell proliferation assay

Cell Counting Kit-8 (CCK-8; Dojindo Molecular Laboratories, Tokyo, Japan) assay was used for determining the proliferation rates of the cultured cells, according to the manufacturer’s instructions. Cells were digested and seeded onto 96-well plates at a density of 2 × 10³ cells per well. After adherence, the cells were transfected with miR-126 mimic or NC mimic and cultured for the additional 0, 24, 48, 72, and 96 h. Optical density (OD) values were determined in six replicates in each group from three independent experiments using a microplate reader (SpectraMax Plus 384; Molecular Devices, CA, USA) at 450 nm.

Cell migration and invasion assay

Cells were collected and suspended in serum-free medium. Afterward, 50,000 cells were plated in the upper chambers of the transwell inserts (Corning, NY, USA) placed in 24-well plates, while the medium with 10% FBS was added to the lower chambers. After 18–24 h of incubation, the cells that migrated into the lower chambers were fixed in absolute methanol, stained with 0.2% crystal violet, and counted in five random fields. For the invasion assay, the inserts were pre-coated with invasion Matrigel (Corning, NY, USA) and diluted in the ration of 1:8 with the serum-free medium.

Target gene prediction

We searched for the potential target genes downstream of miR-126 using two web-based publicly available algorithms, miRanda (http://www.microrna.org) and TargetScan (http://www.targetscan.org), two database search engines that are commonly utilized to search for target sites for miRNAs.

Statistical analysis

All results are expressed as mean ± standard error of mean (SEM). Expression levels of miR-126 and ADAM9 in tumors and adjacent noncancerous liver tissue samples were compared using Mann–Whitney U test. The association between miR-126 expression and clinicopathological features was assessed by χ²-test. Student’s t test was used to analyze the differences between two groups; analysis of variance (ANOVA) was used to compare multiple groups; and Pearson’s correlation coefficient was used in the analysis of the correlation between two groups. Recurrence-free survival was compared via Kaplan–Meier method and analyzed using the log-rank test. Univariate and multivariate analyses were based on the Cox proportional hazards regression model. Statistical analyses were carried out with IBM SPSS Statistics 20.0 (IBM, IL, USA), and p < 0.05 was considered to be statistically significant.

Results

MiR-126 is frequently downregulated in HBV-related HCC tissue samples and HCC cells

Of 100 cases examined here, a decreased expression of miR-126 was observed in 84.00% (84/100) of patients (Figure 1(a) and (b), while 69.00% (69/100) showed a significant decrease (≤0.5-fold) in the levels of this miRNA in cancerous tissues, compared with those in the corresponding noncancerous tissues (Figure 1(c)). However, in only 8.00% (8/100) of cases, the expression of miR-126 was upregulated (≥two-fold; Figure 1(c)). In addition, miR-126 was shown to be downregulated in different HCC cell lines, especially in HepG2, SNU-387, and SNU-449 cells, compared with the expression levels in the immortalized liver LO2 cells (Figure 1(d)).

Figure 1.

Relative miR-126 expression levels in HBV-related HCC patients and HCC cell in vitro, and its association with recurrence-free survival. (a) Alterations in miR-126 expression in tumor tissues compared with that in normal tissues. (b) MiR-126 expression is generally downregulated in HBV-related HCC tissues, in comparison with that in the matching noncancerous tissues (***p < 0.001). (c) MiR-126 expression levels in the patient cohort analyzed here. (d) MiR-126 expression levels in a panel of HCC cells normalized to that determined in the human immortalized hepatic cell line LO2 (columns: mean of three independent experiments; bars: SEM (n = 3); ***p < 0.001). (e) Recurrence-free survival was compared between high and low miR-126 expression groups using Kaplan–Meier analysis (*p < 0.05).

MiR-126 downregulation is associated with poor prognosis in HBV-related HCC

We separated our patient cohort according to the median miR-126 expression (0.3345) into high miR-126 (n = 50) and low miR-126 groups (n = 50). The obtained results showed that miR-126 expression levels are prominently associated with pre-operational AFP levels (p = 0.025), microvascular invasion (MVI; p = 0.016), tumor metastasis rate (p = 0.026), and early recurrence (12 months after surgery; p = 0.008; Table 1). To determine whether miR-126 expression levels were correlated with recurrence-free survival of HBV-related HCC patients, Kaplan–Meier analysis was performed. As shown in Figure 1(e), patients belonging to the low miR-126 group had considerably shorter recurrence-free survival time than the patients in high miR-126 expression group (median recurrence-free survival time: 15.01 months vs 39.16 months; p = 0.047).

Table 1.

Relationships between relative miR-126 expression and clinicopathological parameters of patients included in this study.

Clinical parameters	Cases	Low	High	χ ²	p
Gender
Male	86	43 (50.0%)	43 (63.5%)	0.000	1.000
Female	14	7 (50.0%)	7 (50.0%)		1.000
Age (years)
<60	83	43 (51.8%)	40 (48.2%)	0.638	0.424
≥60	17	7 (41.2%)	10 (58.8%)		0.424
AFP (µg/L)
<400	59	24 (40.7%)	35 (59.3%)	5.002	0.025
≥400	41	26 (63.4%)	15 (36.6%)		0.025
Cirrhosis
Positive	76	36 (47.4%)	40 (52.6%)	0.877	0.349
Negative	23	14 (58.3%)	10 (41.7%)		0.349
Tumor number
=1	83	43 (51.8%)	40 (48.2%)	0.638	0.424
>1	17	7 (41.2%)	10 (58.8%)		0.424
Tumor size (cm)
≤3.0	24	12 (50.0%)	12 (50.0%)	0.000	1.000
>3.0	76	38 (50.0%)	38 (50.0%)		1.000
Tumor capsule
Positive	75	38 (50.7%)	37 (49.3%)	0.053	0.817
Negative	25	12 (48.0%)	13 (52.0%)		0.817
Differentiation
Well	14	3 (21.4%)	11 (78.6%)	5.360	0.069
Moderate	63	34 (54.0%)	29 (46.0%)
Poor	23	13 (56.5%)	10 (43.5%)
MVI
Positive	44	28 (63.6%)	16 (36.4%)	5.844	0.016
Negative	56	22 (39.3%)	34 (60.7%)		0.016
TNM stage
I + II	68	33 (48.5%)	35 (51.5%)	0.184	0.668
III + IV	32	17 (53.1%)	15 (46.9%)		0.668
BCLC stage
A	64	31 (48.4%)	33 (51.6%)	0.174	0.677
B + C	36	19 (52.8%)	17 (47.2%)		0.677
Metastasis^a
Positive	43	27 (62.8%)	16 (37.2%)	4.937	0.026
Negative	57	23 (40.4%)	34 (59.6%)		0.026
Early-recurrence (months)^b
<12	39	26 (66.7%)	13 (33.3%)	6.990	0.008
≥12	54	21 (38.9%)	33 (61.1%)		0.008

AFP: alpha-fetoprotein; MVI: microvascular invasion; TNM: tumor–node–metastasis; BCLC: Barcelona clinic liver cancer.

Metastasis was defined as recurrence time <6 months, the presence of lymph node invasion or extrahepatic invasion, or tumor embolism (including portal vein, biliary tract, and hepatic vein thrombus).

Of 100 patients recruited, 93 cases were included in the prognosis and regression analysis, since 2 died within the peri-operative period because of post-operative complications and 5 were lost to the follow-up.

MiR-126 overexpression inhibits HCC cell migration and invasion in vitro

In order to functionally analyze miR-126 in vitro, HepG2 and SNU-449 cells were transfected with miR-126 mimic. We confirmed that the intracellular levels of miR-126 were successfully upregulated, compared with those in the NC group (approximately 800-fold higher levels in HepG2 and 1100-fold higher in SNU-449 cells; Figure 2(a)). No significant alterations in the proliferative rates of these cells were observed in vitro (Figure 2(b)), but, as presented in Figure 2(c), migratory potential and invasiveness of the cells overexpressing miR-126 were shown to be significantly decreased, compared with those of the cells transfected with the NC (p < 0.01, for all).

Figure 2.

Effects of miR-126 overexpression on proliferation, migration, and invasion of HCC cells in vitro. (a) Confirmation of miR-126 overexpression in HepG2 and SNU-449 cells after 48–72 h of incubation (columns: mean of three independent experiments; bars: SEM (n = 3); ***p < 0.001). (b) Proliferative rate of the transfected cells. Data represent three independent experiments and are shown as mean ± SEM (n = 3). (c) Migratory and invasion potentials of the transfected HepG2 and SNU-449 cells (original magnification: 100×; right: histograms showing quantification results; columns: mean of four randomly selected field; bars: SEM (n = 4); **p < 0.01; ***p < 0.001).

MiR-126 inhibits HCC metastasis through the post-transcriptional regulation of ADAM9 expression

As an attempt to explore the mechanisms underlying the observed effects, we searched for the potential target genes downstream of miR-126 using miRanda and TargetScan. Among the potential candidates, ADAM9 was the most likely candidate responsible for the development of tumor metastasis (Figure 3(a)). Based on the bioinformatics analyses, previous studies demonstrated that ADAM9 represents one of the target genes of miR-126.^27–32 However, very little is known about the effects of miR-126 on ADAM9 expression in HBV-related HCC patients. Therefore, ADAM9 mRNA and protein expression levels were determined in patients enrolled in this study using qRT-PCR and immunohistochemical staining, respectively. The obtained results showed that ADAM9 levels in the cancer tissue samples were generally higher than those in the matching noncancerous tissues (p = 0.039; Figure 3(b) and (c)). A significant inverse correlation between miR-126 and ADAM9 mRNA (r = −0.292, p = 0.024) and protein expression levels (r = −0.269, p = 0.027) was observed as well (Figure 3(d)). Furthermore, after restoring miR-126 expression in HepG2 and SNU-449 cells, ADAM9 expression was suppressed, especially at the post-transcriptional level (Figure 3(e)). In addition, following the effective ADAM9 knockdown by transient transfection with siRNA (Figure 4(a) and (b)), the migratory potential and invasiveness of the transfected cells decreased as well (p < 0.001 for all; Figure 4(c)).

Figure 3.

Correlation between miR-126 and ADAM9 expression levels. (a) MiR-126 predicted target site at ADAM9 3′-UTR. (b and c) Relative ADAM9 expression in cancer tissue samples and noncancerous tissues (n = 100, *p < 0.05). (d) Correlation between ADAM9 and miR-126 expression levels (mRNA—n = 100, r = −0.292, *p < 0.05; protein—n = 56, r = –0.269 *p < 0.05; original magnification: 200×). (e) ADAM9 expression following the overexpression of miR-126 in HepG2 and SNU-449 cells (columns: mean of three independent experiments; bars: SEM (n = 3); *p < 0.05).

Figure 4.

Effects of ADAM9 silencing on migratory potential and invasiveness of HCC cells in vitro. (a and b) Successful downregulation of ADAM9 expression (columns: mean of three independent experiments; bars: SEM (n = 3); ***p < 0.001). (c) Results of cell migration and invasion assays performed using the transfected HepG2 and SNU-449 cells (original magnification: 100×; right: histograms showing the results of the quantification; columns: mean of four randomly selected field; bars: SEM (n = 4); ***p < 0.001).

Correlation between miR-126 and ADAM9 levels and recurrence-free survival of the HBV-related HCC patients

To determine the prognostic value of miR-126 and ADAM9 levels and of some common clinicopathological parameters associated with recurrence-free survival time in HBV-related HCC patients, univariate and multivariate analyses were used. In the univariate analysis, miR-126 (p = 0.05) and ADAM9 expression (p = 0.006) were shown to be correlated with recurrence-free survival of the patients enrolled in this study, together with the additional five pathological parameters, tumor size (p = 0.004), tumor number (p = 0.001), histological differentiation (p < 0.001), MVI (p < 0.001), and the occurrence of metastases (p < 0.001; Table 2). Using the multivariate analysis, ADAM9 level (p = 0.033), tumor number (p = 0.014), MVI (p = 0.037), and tumor metastasis (p = 0.001) were further identified as independent prognostic factors of shorter recurrence-free survival, but miR-126 expression level (p = 0.71) was not confirmed as an independent prognostic factor (Table 2).

Table 2.

Univariate and multivariate analysis of recurrence-free survival of the patients included in this study (n = 93).

Clinical parameters	Univariate				Multivariate
Clinical parameters	HR	95% CI		p	HR	95% CI		p
Gender (male vs female)	0.498	0.178	1.392	0.184
Age (<60 vs ≥60)	0.891	0.398	1.994	0.779
AFP (µg/L) (<400 vs >400)	1.371	0.766	2.453	0.288
Cirrhosis (positive vs negative)	1.013	0.514	1.996	0.970
Capsule (positive vs negative)	1.258	0.606	2.614	0.538
Tumor size (cm) (≤3.0 vs >3.0)	5.504	1.705	17.767	0.004	2.611	0.797	8.882	0.111
Tumor number (=1 vs >1)	2.934	1.513	5.689	0.001	2.566	1.212	5.433	0.014
Differentiation (well, moderate, and poor)	2.860	1.750	4.673	0.000	1.333	0.763	2.330	0.312
MVI (positive vs negative)	5.801	2.938	11.455	0.000	2.444	1.056	5.659	0.037
Metastasis^a (positive vs negative)	9.154	4.610	18.175	0.000	4.352	1.897	9.981	0.001
MiR-126 (low vs high)	0.553	0.306	1.001	0.050	0.886	0.469	1.675	0.710
ADAM9 mRNA (low vs high)	2.387	1.285	4.433	0.006	2.092	1.060	4.131	0.033

HR: hazard ratio; CI: confidence interval; AFP: alpha-fetoprotein; MVI: microvascular invasion; TNM: tumor–node–metastasis; BCLC: Barcelona clinic liver cancer.

Of 100 patients recruited, 93 cases were eligible for the prognosis and regression analysis, since 2 died within the peri-operative period because of post-operative complications and 5 were lost to the follow-up.

Metastasis was defined as recurrence time <6 months, the presence of lymph node invasion or extrahepatic invasion, or tumor embolism (including portal vein, biliary tract, and hepatic vein thrombus).

Discussion

Metastatic disease, occurring during both early and late development and progression of primary tumor, remains largely incurable, and it represents the leading cause of cancer-related mortality.³³ In order to develop novel therapeutic approaches for the treatment of metastases, a deeper understanding of molecular mechanism underlying their development is required. In the past years, many miRNAs were shown to regulate tumor metastases by targeting genes involved in cell proliferation, apoptosis, migration, invasion, cellular senescence, and DNA damage responses.³⁴

Previous studies demonstrated that miR-126 is universally downregulated in HCC cells in vitro³⁵ and in vivo.^11,12 Recently, the transfection of HepG2 cells with full-length HBV (qRT-PCR and enzyme-linked immunosorbent assay (ELISA) confirmed in vitro replication of HBV and expression of viral proteins, respectively)¹⁴ and pre-operational TACE treatment were shown to significantly induce miR-126 expression,¹³ but no underlying mechanisms were elucidated. Therefore, determining the pre-treatment expression levels of miR-126 and the clinical significance of this molecule in HBV-related HCC patients are necessary. Here, we determined that miR-126 expression is generally decreased in cancer tissues of HBV-related HCC patients who did not undergo pre-operational treatment and that miR-126 expression correlates with MVI rate, pre-operational tumor metastasis, and early HCC recurrence after surgery. Combined with the results of miR-126 overexpressing cell-based assays, we showed that this miRNA markedly attenuates cell migration and invasion, while the cell proliferation rate was not affected. However, in some studies, the restored expression of miR-126 in HCC cells was reported to lead to a decrease in the proliferation rate through SOX2 and EGFL7 targeting.^12,36 In conclusion, our results indicate that the downregulation of miR-126 may contribute to the metastasis of HBV-related HCC as well as early recurrence.

The development of cancer metastasis is a complex, multistage process that begins with the local invasion of primary tumor cells, which involves the interaction between cancer cells and surrounding extracellular matrix components.³⁷ Du et al.¹¹ demonstrated that the deregulation of miR-126 expression promotes HCC metastasis and neoangiogenesis by targeting LRP6 and PIK3R2, respectively. To understand the potential alternative mechanism underlying HCC metastasis, we focused on its target gene, ADAM9, considering the strong correlation between its expression and tumor metastasis. Here, we showed that miR-126 expression is inversely correlated with that of ADAM9 at both mRNA and protein levels in treatment-naive HBV-related HCC patients and that the overexpression of miR-126 in HCC cells in vitro leads to a decrease in ADAM9 expression mainly at the post-transcriptional level. Furthermore, the effects of miR-126 overexpression on tumor cell migratory potential and invasiveness were shown to be consistent with those induced by ADAM9 silencing. In addition, several previous studies confirmed that miR-126 targets the ADAM9 gene and demonstrated the role of ADAM9 in cell migration and invasion.^27–32 Therefore, our results indicate that the loss of miR-126 in tumor cells contributes to the development of metastases in HBV-related HCC partially through the upregulation of ADAM9 expression.

HCC recurrence after liver transplantation is regarded as metastasis-related recurrence.³⁸ MiRNAs with the most significantly altered levels in patients with post-liver transplantation HCC recurrence were shown to be miR-19a, miR-886-5p, miR-126, miR-223, miR-24, and miR-147, and the presence of this high-risk six-miRNA signature was shown to have high sensitivity and specificity in predicting post-liver transplantation HCC recurrence.^10,39 Herein, our Cox proportional hazard model analysis demonstrated that ADAM9 expression level, tumor number, MVI, and tumor metastasis, but not miR-126 expression, represent independent prognostic factors for shorter recurrence-free survival time, which indicates that a decrease in miR-126 expression takes roles in the process of tumor metastasis through upregulating the expression of its metastasis-related target genes in HBV-related HCC, but alone, in the complex network of signaling pathways, it may be insufficient for the induction of tumor metastasis. To the best of our knowledge, the regulation of ADAM9, that dysregulated in the complex, multistage process of HCC progression involves multiple miRNAs, such as miR-203,⁴⁰ miR-1274a,⁴¹ and miR-126. Remarkably, we observed that miR-223, one of the six miRNAs mentioned above, was also predicted to negatively regulate ADAM9 expression according to the results of miRanda analysis.

Taken together, the obtained results demonstrate that miR-126 levels are significantly decreased in HBV-related HCC patients without pre-operational treatment. The loss of tumor suppressor miR-126 in cancer cells contributes to HCC progression by upregulating the expression of ADAM9, an oncogene that was significantly associated with HCC metastasis and recurrence. The results obtained here may help develop novel therapeutic approaches for the prevention of HBV-related HCC metastasis.

Footnotes

Acknowledgements

The authors thank Professor Weizhong Wu, Liver Cancer Research Institute, Zhongshan Hospital, Fudan University, Shanghai, China, for providing the HCC cell lines MHCC-LM3, MHCC-97H, and MHCC-97L.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by Science and Technology Planning Project of Guangdong Province, China (No. 2013B022000069).

References

Lafaro

Demirjian

Pawlik

. Epidemiology of hepatocellular carcinoma. Surg Oncol Clin N Am 2015; 24: 1–17.

Forner

Llovet

Bruix

. Hepatocellular carcinoma. Lancet 2012; 379: 1245–1255.

Tung-Ping Poon

Fan

Wong

. Risk factors, prevention, and management of postoperative recurrence after resection of hepatocellular carcinoma. Ann Surg 2000; 232: 10–24.

Steeg

. Targeting metastasis. Nat Rev Cancer 2016; 16: 201–218.

Hammond

. An overview of microRNAs. Adv Drug Deliv Rev 2015; 87: 3–14.

Lin

Gregory

. MicroRNA biogenesis pathways in cancer. Nat Rev Cancer 2015; 15: 321–333.

Wang

Aurora

Johnson

. The endothelial-specific microRNA miR-126 governs vascular integrity and angiogenesis. Dev Cell 2008; 15: 261–271.

Meister

Schmidt

. MiR-126 and miR-126*: new players in cancer. ScientificWorldJournal 2010; 10: 2090–2100.

Ebrahimi

Gopalan

Smith

. MiR-126 in human cancers: clinical roles and current perspectives. Exp Mol Pathol 2014; 96: 98–107.

10.

Chen

Miao

Fan

. Decreased expression of miR-126 correlates with metastatic recurrence of hepatocellular carcinoma. Clin Exp Metastasis 2013; 30: 651–658.

11.

Cao

. MiR-126-3p suppresses tumor metastasis and angiogenesis of hepatocellular carcinoma by targeting LRP6 and PIK3R2. J Transl Med 2014; 12: 259.

12.

Wang

. MicroRNA-126 inhibits tumor proliferation and angiogenesis of hepatocellular carcinoma by down-regulating EGFL7 expression. Oncotarget 2016; 7: 66922–66934.

13.

Song

. Inhibition of microRNA-126 promotes the expression of Spred1 to inhibit angiogenesis in hepatocellular carcinoma after transcatheter arterial chemoembolization: in vivo study. Onco Targets Ther 2016; 9: 4357–4367.

14.

Ghosh

Datta

. Hepatic miR-126 is a potential plasma biomarker for detection of hepatitis B virus infected hepatocellular carcinoma. Int J Cancer 2016; 138: 2732–2744.

15.

Peduto

. ADAM9 as a potential target molecule in cancer. Curr Pharm Des 2009; 15: 2282–2287.

16.

O’Shea

McKie

Buggy

. Expression of ADAM-9 mRNA and protein in human breast cancer. Int J Cancer 2003; 105: 754–761.

17.

Fritzsche

Wassermann

Jung

. ADAM9 is highly expressed in renal cell cancer and is associated with tumour progression. BMC Cancer 2008; 8: 179.

18.

Grutzmann

Luttges

Sipos

. ADAM9 expression in pancreatic cancer is associated with tumour type and is a prognostic factor in ductal adenocarcinoma. Br J Cancer 2004; 90: 1053–1058.

19.

Yamada

Ohuchida

Mizumoto

. Increased expression of ADAM9 and ADAM15 mRNA in pancreatic cancer. Anticancer Res 2007; 27: 793–799.

20.

Carl-McGrath

Lendeckel

Ebert

. The disintegrin-metalloproteinases ADAM9, ADAM12, and ADAM15 are upregulated in gastric cancer. Int J Oncol 2005; 26: 17–24.

21.

Fritzsche

Jung

Tolle

. ADAM9 expression is a significant and independent prognostic marker of PSA relapse in prostate cancer. Eur Urol 2008; 54: 1097–1106.

22.

Le Pabic

Bonnier

Wewer

. ADAM12 in human liver cancers: TGF-beta-regulated expression in stellate cells is associated with matrix remodeling. Hepatology 2003; 37: 1056–1066.

23.

Tannapfel

Anhalt

Hausermann

. Identification of novel proteins associated with hepatocellular carcinomas using protein microarrays. J Pathol 2003; 201: 238–249.

24.

Tao

Qian

Tang

. Increased expression of a disintegrin and metalloprotease-9 in hepatocellular carcinoma: implications for tumor progression and prognosis. Jpn J Clin Oncol 2010; 40: 645–651.

25.

Schmittgen

Livak

. Analyzing real-time PCR data by the comparative C(T) method. Nat Protoc 2008; 3: 1101–1108.

26.

Liu

Huang

Jiang

. Methylenetetrahydrofolate dehydrogenase 2 overexpression is associated with tumor aggressiveness and poor prognosis in hepatocellular carcinoma. Dig Liver Dis 2016; 48: 953–960.

27.

Hamada

Satoh

Fujibuchi

. MiR-126 acts as a tumor suppressor in pancreatic cancer cells via the regulation of ADAM9. Mol Cancer Res 2012; 10: 3–10.

28.

Felli

Felicetti

Lustri

. MiR-126&126* restored expressions play a tumor suppressor role by directly regulating ADAM9 and MMP7 in melanoma. PLoS One 2013; 8: e56824.

29.

Jia

Castillo-Martin

Bonal

. MicroRNA-126 inhibits invasion in bladder cancer via regulation of ADAM9. Br J Cancer 2014; 110: 2945–2954.

30.

Jiang

Zhang

. MiR-126 inhibits cell growth, invasion, and migration of osteosarcoma cells by downregulating ADAM-9. Tumour Biol 2014; 35: 12645–12654.

31.

Liu

Jiang

. DNMT1-microRNA126 epigenetic circuit contributes to esophageal squamous cell carcinoma growth via ADAM9-EGFR-AKT signaling. Clin Cancer Res 2015; 21: 854–863.

32.

Wang

Yuan

. MiR-126 regulated breast cancer cell invasion by targeting ADAM9. Int J Clin Exp Pathol 2015; 8: 6547–6553.

33.

Turajlic

Swanton

. Metastasis as an evolutionary process. Science 2016; 352: 169–175.

34.

Zhang

Yang

Wang

. Microenvironmental regulation of cancer metastasis by miRNAs. Trends Cell Biol 2014; 24: 153–160.

35.

Wong

Lung

Law

. MicroRNA-223 is commonly repressed in hepatocellular carcinoma and potentiates expression of Stathmin1. Gastroenterology 2008; 135: 257–269.

36.

Zhao

Zhang

. MiR-126 inhibits cell proliferation and induces cell apoptosis of hepatocellular carcinoma cells partially by targeting SOX2. Hum Cell 2015; 28: 91–99.

37.

Albelda

. Role of integrins and other cell adhesion molecules in tumor progression and metastasis. Lab Invest 1993; 68: 4–17.

38.

Schlitt

Neipp

Weimann

. Recurrence patterns of hepatocellular and fibrolamellar carcinoma after liver transplantation. J Clin Oncol 1999; 17: 324–331.

39.

Han

Zhong

Teng

. Identification of recurrence-related microRNAs in hepatocellular carcinoma following liver transplantation. Mol Oncol 2012; 6: 445–457.

40.

Wan

Shen

. MiR-203 suppresses the proliferation and metastasis of hepatocellular carcinoma by targeting oncogene ADAM9 and oncogenic long non-coding RNA HULC. Anticancer Agents Med Chem 2016; 16: 414–423.

41.

Zhou

Liu

. MicroRNA-1274a, a modulator of sorafenib induced a disintegrin and metalloproteinase 9 (ADAM9) down-regulation in hepatocellular carcinoma. FEBS Lett 2011; 585: 1828–1834.