MicroRNA-195 Suppresses Angiogenesis and Metastasis of Hepatocellular Carcinoma by Inhibiting the Expression of VEGF, VAV2, and CDC42

Authors

  • Ruizhi Wang,

    1. Key Laboratory of Gene Engineering of the Ministry of Education, State Key Laboratory of Biocontrol, School of Life Sciences, Sun Yat-sen University, Guangzhou, People's Republic of China
    Search for more papers by this author
  • Na Zhao,

    1. Key Laboratory of Gene Engineering of the Ministry of Education, State Key Laboratory of Biocontrol, School of Life Sciences, Sun Yat-sen University, Guangzhou, People's Republic of China
    Search for more papers by this author
  • Siwen Li,

    1. Key Laboratory of Gene Engineering of the Ministry of Education, State Key Laboratory of Biocontrol, School of Life Sciences, Sun Yat-sen University, Guangzhou, People's Republic of China
    Search for more papers by this author
  • Jian-Hong Fang,

    1. Key Laboratory of Gene Engineering of the Ministry of Education, State Key Laboratory of Biocontrol, School of Life Sciences, Sun Yat-sen University, Guangzhou, People's Republic of China
    Search for more papers by this author
  • Mei-Xian Chen,

    1. Department of Hepatobiliary Oncology, Sun Yat-sen University, Guangzhou, People's Republic of China
    Search for more papers by this author
  • Jine Yang,

    1. Key Laboratory of Gene Engineering of the Ministry of Education, State Key Laboratory of Biocontrol, School of Life Sciences, Sun Yat-sen University, Guangzhou, People's Republic of China
    Search for more papers by this author
  • Wei-Hua Jia,

    1. Bank of Tumor Resources, Cancer Center, Sun Yat-sen University, Guangzhou, People's Republic of China
    Search for more papers by this author
  • Yunfei Yuan,

    Corresponding author
    1. Department of Hepatobiliary Oncology, Sun Yat-sen University, Guangzhou, People's Republic of China
    • Address reprint requests to: Shi-Mei Zhuang, School of Life Sciences, Sun Yat-sen University, Xin Gang Xi Road 135#, Guangzhou 510275, People's Republic of China. E-mail: zhuangshimei@163.com or lsszsm@mail.sysu.edu.cn; fax: +86-20-84113469.

      Yunfei Yuan, Department of Hepatobiliary Oncology, Cancer Center, Sun Yat-sen University, Dongfengdong Road 651#, Guangzhou 510060, People's Republic of China. E-mail: yuanyf@mail.sysu.edu.cn; fax: +86-20-87343392.

    Search for more papers by this author
  • Shi-Mei Zhuang

    Corresponding author
    1. Key Laboratory of Gene Engineering of the Ministry of Education, State Key Laboratory of Biocontrol, School of Life Sciences, Sun Yat-sen University, Guangzhou, People's Republic of China
    • Address reprint requests to: Shi-Mei Zhuang, School of Life Sciences, Sun Yat-sen University, Xin Gang Xi Road 135#, Guangzhou 510275, People's Republic of China. E-mail: zhuangshimei@163.com or lsszsm@mail.sysu.edu.cn; fax: +86-20-84113469.

      Yunfei Yuan, Department of Hepatobiliary Oncology, Cancer Center, Sun Yat-sen University, Dongfengdong Road 651#, Guangzhou 510060, People's Republic of China. E-mail: yuanyf@mail.sysu.edu.cn; fax: +86-20-87343392.

    Search for more papers by this author

  • Supported by grants from the Ministry of Science and Technology of China (2010CB912803, 2011CB811305, 2011ZX09307-001-04), the Ministry of Health of China (2012ZX10002011), the National Natural Science Foundation of China (30925036, 81230049), and the Fundamental Research Funds for the Central Universities (12lgjc03, 12lgpy25).

  • Potential conflict of interest: Nothing to report.

Abstract

Hepatocellular carcinoma (HCC) is characterized by active angiogenesis and metastasis, which account for rapid recurrence and poor survival. There is frequent down-regulation of miR-195 expression in HCC tissues. In this study, the role of miR-195 in HCC angiogenesis and metastasis was investigated with in vitro capillary tube formation and transwell assays, in vivo orthotopic xenograft mouse models, and human HCC specimens. Reduction of miR-195 in HCC tissues was significantly associated with increased angiogenesis, metastasis, and worse recurrence-free survival. Both gain-of-function and loss-of-function studies of in vitro models revealed that miR-195 not only suppressed the ability of HCC cells to promote the migration and capillary tube formation of endothelial cells but also directly repressed the abilities of HCC cells to migrate and invade extracellular matrix gel. Based on mouse models, we found that the induced expression of miR-195 dramatically reduced microvessel densities in xenograft tumors and repressed both intrahepatic and pulmonary metastasis. Subsequent investigations disclosed that miR-195 directly inhibited the expression of the proangiogenic factor vascular endothelial growth factor (VEGF) and the prometastatic factors VAV2 and CDC42. Knockdown of these target molecules of miR-195 phenocopied the effects of miR-195 restoration, whereas overexpression of these targets antagonized the function of miR-195. Furthermore, we revealed that miR-195 down-regulation resulted in enhanced VEGF levels in the tumor microenvironment, which subsequently activated VEGF receptor 2 signaling in endothelial cells and thereby promoted angiogenesis. Additionally, miR-195 down-regulation led to increases in VAV2 and CDC42 expression, which stimulated VAV2/Rac1/CDC42 signaling and lamellipodia formation and thereby facilitated the metastasis of HCC cells. Conclusion: miR-195 deregulation contributes to angiogenesis and metastasis in HCC. The restoration of miR-195 expression may be a promising strategy for HCC therapy. (Hepatology 2013;58:642-653)

Abbreviations
EC

endothelial cell

ELISA

enzyme-linked immunosorbent assay

ERK

extracellular signal-regulated kinase

GST

glutathione S-transferase

GTP

guanosine triphosphate

HCC

hepatocellular carcinoma

HUVEC

human umbilical vein endothelial cell

miRNA

microRNA

mRNA

messenger RNA

MVD

microvessel density

PBS

phosphate-buffered saline

RFS

recurrence-free survival

RT-PCR

reverse-transcriptase polymerase chain reaction

SDS

sodium dodecyl sulfate

SFM

serum-free medium for endothelial cells

TCM

tumor cell–conditioned medium

UTR

untranslated region

VEGF

vascular endothelial growth factor

VEGFR2

VEGF receptor 2.

Globally, hepatocellular carcinoma (HCC) is a common and highly lethal malignancy. Active angiogenesis and frequent metastasis are responsible for rapid recurrence and poor survival of HCC. Therefore, identifying molecules that can suppress angiogenesis and metastasis may provide novel targets for HCC therapies.

MicroRNAs (miRNAs) constitute a class of endogenous small noncoding RNAs that suppress protein expression by base-pairing with the 3′-untranslated regions (UTRs) of target messenger RNA (mRNA). miRNAs have been demonstrated to interact with various components of multiple cellular signaling pathways and to participate in a wide range of physiological and pathological processes, including tumorigenesis. Increasing evidence suggests that the dysregulation of miRNAs plays an important role in HCC development.[1-3] To date, a few miRNAs have been characterized to have proangiogenic (miR-221[4]) or antiangiogenic (miR-122,[5] miR-29b,[6] and miR-214[7]) activities or to possess prometastatic (miR-151,[8] miR-30d,[9] miR-210,[10] and miR-135a[11]) or antimetastatic (miR-122,[12] miR-124,[13] miR-139,[14] miR-125b,[15] miR-29b,[6] and miR-7[16]) functions in HCC.

miR-195 is down-regulated frequently in multiple cancer types, including HCC.[17-21] Studies from different groups have indicated a growth-suppressive function of miR-195.[17, 18, 21, 22] miR-195 has been shown to block the G1/S transition of the cell cycle by targeting CCND1/3, CDK4/6, and E2F3[17, 21] and to promote apoptosis by suppressing the expression of BCL2 and BCL-w.[18, 22] However, whether the dysregulation of miR-195 contributes to tumor angiogenesis or metastasis remains unclear. To date, only a few reports have explored the relationship of miR-195 to metastasis. Two research groups have employed the in vitro transwell system and disclosed that miR-195 might suppress the invasion of glioblastoma and breast cancer cells by targeting CCND3 and RAF1, respectively.[19, 21] In a study of miRNA profiling, miR-195 was down-regulated significantly in primary HCC tissues and tended to decrease further in portal vein tumor thrombi.[23] Clearly, more extensive investigations are required to verify the inhibitory role of miR-195 in tumor angiogenesis and metastasis.

Herein, we clarified the significance of miR-195 in tumor angiogenesis and metastasis of HCC by using in vitro assays, animal models and human specimens. We showed that the down-regulation of miR-195 in human HCC tissues was associated with increased angiogenesis, metastasis, and poor survival. Restoration of miR-195 in HCC cells significantly suppressed tumor angiogenesis and metastasis in vitro and in vivo. We further provided evidence that miR-195 exerted its antiangiogenic and antimetastatic effects by directly targeting multiple genes, including vascular endothelial growth factor (VEGF), VAV2, and CDC42. Our findings highlight the importance of miR-195 dysfunction in promoting HCC progression and recurrence and implicate miR-195 as a potential therapeutic target for HCC.

Materials and Methods

Human Tissue Specimens

Human HCC tissues were collected from 135 patients who underwent HCC resection at the Cancer Center of Sun Yat-sen University between 2001 and 2006. The patients had not received any local or systemic anticancer treatments prior to the surgery, and no postoperative anticancer therapies were administered prior to relapse. All patients were followed postoperatively to assess survival rates and to monitor for recurrence and metastases. The median follow-up time was 45 months. The relevant characteristics of the studied subjects are shown in Supporting Table 1. Informed consent was obtained from each patient, and the study was approved by the Institute Research Ethics Committee at the Cancer Center.

RNA Oligoribonucleotides and Vectors

All miRNA mimic and small interfering RNA duplexes (Supporting Table 2) were purchased from GenePharma (Shanghai, People's Republic of China). si-VEGF, si-VAV2, and si-CDC42 targeted the mRNAs that coded for human VEGF (NM_001171626.1), VAV2 (NM_003371), and CDC42 (NM_044472), respectively. The negative control RNA duplex (NC) for both miRNA mimic and small interfering RNA was nonhomologous to any human genome sequence. The sequence-specific miR-195 inhibitor (anti–miR-195) and its control (anti-NC) were from Dharmacon (Chicago, IL).

The expression vectors pRetroX-miR-195, pMir-vec-LUC, pc3-Gab-VEGF, pc3-Gab-VAV2, and pc3-Gab-CDC42 and firefly luciferase reporter plasmids pGL3cm-VEGF-3′UTR-Wt, pGL3cm-VAV2-3′UTR-Wt, pGL3cm-CDC42-3′UTR-Wt, pGL3cm-VEGF-3′UTR-Mut, pGL3cm-VAV2-3′UTR-Mut, and pGL3cm-CDC42-3′UTR-Mut were constructed as described in the Supporting Materials and Methods.

Tumor Cell Lines and Human Umbilical Vein Endothelial Cells

Human HCC (QGY-7703, MHCC-97H, MHCC-97L, Huh-7, and SMMC-7721), colorectal carcinoma (HCT-116), and transformed human embryonic kidney (HEK293T) cell lines were maintained in Dulbecco's modified Eagle's medium (Invitrogen Corp., Buffalo, NY) that was supplemented with 10% fetal bovine serum (FBS; Hyclone, Logan, UT). The QGY-7703 cell subline, which stably expressed miR-195 and firefly luciferase (QGY-miR-195-LUC), and the matched control line (QGY-control-LUC) were established with the Tet-off system (Clontech, Palo Alto, CA, USA), as described in Supporting Materials and Methods.

Human umbilical vein endothelial cells (HUVECs) were isolated and cultured in serum-free medium for endothelial cells (SFM; Invitrogen) as described.[6] HUVECs at passages 2-7 were used.

Analysis of Gene Expression

Gene expression was analyzed by semiquantitative reverse-transcriptase polymerase chain reaction (RT-PCR), real-time quantitative RT-PCR (qRT-PCR), western blotting, or immunohistochemical staining as described in the Supporting Materials and Methods.

Cell Transfections

Reverse transfections of RNA oligoribonucleotides were performed with Lipofectamine-RNAiMAX (Invitrogen). A total of 50 nM of RNA duplex or 200 nM of miRNA inhibitor were used for each transfection. Cotransfections of RNA duplex with plasmid DNA were performed with Lipofectamine 2000 (Invitrogen). HEK293T cells were transfected with plasmids by calcium phosphate precipitation. Packaging of the retroviral expression vectors and infections of the target cells were performed as described in the Supporting Materials and Methods.

Preparation of Tumor-Cell Conditioned Medium

Thirty-six hours after the reverse transfections with RNA oligonucleotides, the tumor cells were washed with 1× phosphate-buffered saline (PBS) and were cultured in SFM for 12 hours; tumor cell–conditioned medium (TCM) was collected subsequently as described in Supporting Materials and Methods. For all experiments, TCM loading was adjusted according to the number of live cells in each sample.

HUVEC Recruitment and Capillary Tube Formation Assays

Assays to determine the effects of tumor cells or TCM on the migration or capillary tube formation of HUVECs were performed as described in Supporting Materials and Methods.

In Vitro Tumor Cell Migration, Invasion, and Growth Assays

The migration and invasion of tumor cells were analyzed in 24-well Boyden chambers with 8-μm pore size polycarbonate membranes (Corning, NY). For invasion assays, the membranes were coated with Matrigel (3432-005-01, R&D Systems, MN) to form matrix barriers.

Nude Mouse Xenograft Studies and In Vivo Imaging

All experimental procedures involving animals were performed in accordance with the Guide for the Care and Use of Laboratory Animals (National Institutes of Health publication nos. 80-23, revised 1996) and according to the institutional ethical guidelines for animal experiments. For xenograft implantation experiments, 2 × 106 QGY-miR-195-LUC cells were resuspended in 25 μL of PBS/Matrigel (1:1) and were inoculated under the capsule of the left hepatic lobe of male BALB/c nude mice. miR-195 expression was silenced by administering drinking water that was supplemented with 10% sucrose and 2 mg/mL of doxycycline (Clontech). Bioluminescence imaging and tumor dissection were performed as described in Supporting Materials and Methods.

Luciferase Reporter Assay

QGY-7703 cells in a 48-well plate were cotransfected with 50 nM of miR-195 or NC duplex, 10 ng of pRL-TK (Promega, Madison, WI), and 50 ng of firefly luciferase reporter plasmid that contained either the wild-type or mutant 3′UTR of the target gene. Forty-eight hours after the transfections, the cell lysates were applied to luciferase assay as described.[3]

Enzyme-Linked Immunosorbent Assay

The levels of VEGF in the TCM were detected using enzyme-linked immunosorbent assay (ELISA) kits (R&D Systems) as instructed by the manufacturer.

Glutathione S-Transferase Pull-Down Assay

Lysates from QGY-7703 cells were incubated with glutathione-Sepharose bead–immobilized glutathione S-transferase (GST) or GST-PAK. Next, the bead-bound proteins were solubilized in sodium dodecyl sulfate (SDS) buffer and analyzed by immunoblotting. The experimental details are described in Supporting Materials and Methods.

F-Actin Staining for Lamellipodia

Thirty-six hours after the transfections with RNA duplex, 2 × 104 QGY-7703 cells were added to coverslips that had been precoated with 240 μg Matrigel (R&D Systems) in 24-well plates. The cells were allowed to spread for 1 hour at 37°C and were fixed, permeabilized and stained with fluorescent phalloidin (Invitrogen, catalog no. A34055), a probe for filamentous actin.

Statistical Analysis

Recurrence-free survival (RFS) was calculated from the date of the HCC resection to the time of first recurrence. Patients who were lost to follow-up or who died from causes unrelated to HCC were treated as censored events. Kaplan-Meier plots and Cox proportional hazard regression analysis, which were applied to identify the prognostic factors, were performed with SPSS version 13.0 (SPSS Inc., Chicago, IL). Associations between the RFS and the molecular changes or clinical characteristics were analyzed initially by a univariate Cox proportional hazards regression analysis. Significant prognostic factors found in the univariate analysis were evaluated further by a multivariate Cox regression analysis.

The data are expressed as the mean ± SEM from at least three independent experiments. The values for the capillary tube formation and luciferase activity assays are from three independent experiments that were performed in duplicate. The differences between the groups were analyzed by Student t test when two groups were compared or by one-way analysis of variance when more than two groups were compared. Analyses were performed with GraphPad Prism, version 5 (GraphPad Software, Inc., San Diego, CA). Correlations between two variables were explored with the Spearman's correlation coefficient. All statistical tests were two-sided; P < 0.05 was considered statistically significant.

Results

Down-Regulation of miR-195 Is Associated With Worse Prognosis

Previously, we observed frequent down-regulation of miR-195 in HCC tissues.[17] To investigate the biological significance of this finding, we analyzed the correlation between miR-195 levels and the clinical features of HCC patients in this study. The Kaplan-Meier plots revealed an association of lower miR-195 levels with shorter RFS (P = 0.004; Fig. 1A). Multivariate Cox regression analysis further confirmed miR-195 down-regulation as an independent risk factor for RFS (HR, 1.773; P = 0.017; Supporting Table 1). Importantly, lower miR-195 levels were associated significantly with higher microvessel densities (MVDs; Fig. 1B) and the presence of metastasis (Fig. 1C), suggesting that miR-195 down-regulation may contribute to HCC progression by promoting tumor angiogenesis and metastasis.

Figure 1.

Down-regulation of miR-195 in HCC tissues is associated with a worse prognosis for HCC patients. (A) Kaplan-Meier plots revealed an association of lower miR-195 levels with a shorter RFS. The levels of mature miR-195 were analyzed using real-time qRT-PCR, and the median value of all 135 cases was chosen as the cutoff point for separating the miR-195–low expression and miR-195–high expression groups. (B) Microvessel density (MVD) was inversely correlated with miR-195 expression in HCC tissues. The median MVD value for all cases was chosen as the cutoff point for separating the low MVD (n = 67) and high MVD (n = 68) tumors. (C) Metastatic HCC displayed lower miR-195 expression levels. The absence (n = 77) or presence (n = 58) of venous invasion (tumor thrombus in the veins of adjacent nontumor tissues or in the portal vein, n = 52), lymph node metastasis (n = 3), or both (n = 3) are indicated with a minus sign (−) and plus sign (+), respectively. The central horizontal line represents the mean value; the error bars represent the SEM. *P < 0.05.

miR-195 Inhibits Tumor Angiogenesis and Metastasis In Vitro and In Vivo

Angiogenesis is a prerequisite for cancer growth and metastasis, and migration and invasion are key steps in the metastatic cascade. To clarify the effect of miR-195 on HCC angiogenesis, we first performed in vitro endothelial recruitment and capillary tube formation assays with two HCC cell lines, QGY-7703 and MHCC-97H. The endothelial recruitment assay, performed in Boyden chamber transwells, revealed that in the presence of HCC cells, many more HUVECs migrated through the transwell pores in comparison with those grown in the absence of tumor cells. However, the restoration of miR-195 expression significantly suppressed the ability of HCC cells to promote HUVEC migration (Fig. 2A and Supporting Fig. 1A). Furthermore, compared with the control media (SFM), TCM from NC-transfected or nontransfected HCC cells promoted the HUVECs to develop more capillary-like structures. Tube formation was reduced dramatically in HUVECs that were grown in TCM from miR-195 transfectants to a level comparable to that of HUVECs cultured in SFM (Fig. 2B and Supporting Fig. 1B). In contrast, the suppression of endogenous miR-195 in HCC cells resulted in enhanced HUVEC migration and capillary tube formation (Fig. 2C,D).

Figure 2.

miR-195 represses tumor angiogenesis in vitro. (A) An endothelial recruitment assay revealed the suppressive effect of miR-195 on the HCC cell–promoted migration of endothelial cells (ECs). HUVECs were seeded in the upper compartments of transwell chambers and QGY-7703 cells that were nontransfected (upper right) or transfected with NC (lower left) or miR-195 (lower right) were seeded in the lower compartments. The co-cultures were incubated for 12 hours in SFM. SFM (upper left) with no tumor cells in the bottom chamber was used as a control. (B) Restoration of miR-195 inhibited the HCC cell–promoted EC tube formation. HUVECs were cultured for 6 hours in the presence of SFM (upper left) or 100% TCM from QGY-7703 cells that were nontransfected (upper right) or transfected with NC (lower left) or miR-195 duplex (lower right). (C) Antagonism of miR-195 enhanced the HCC cell–induced EC migration. HUVECs in the upper compartments of transwell chambers were cocultured with QGY-7703 cells transfected with anti-NC or anti–miR-195, as in (A). (D) Antagonism of miR-195 increased the HCC cell–promoted EC tube formation. HUVECs were cultured for 6 hours in the presence of SFM or 100% TCM from QGY-7703 cells, which were nontransfected or transfected with anti-NC or anti–miR-195. Scale bars, 100 μm. *P < 0.05; **P < 0.01; ***P < 0.001.

To elucidate the role of miR-195 in HCC metastasis, the effects of miR-195 on the migration and invasion of HCC cells were analyzed initially in vitro. Transwell assays showed that both the migratory and invasive activities of HCC cells were suppressed by miR-195 expression (Fig. 3A,B and Supporting Fig. 2A,B) but were promoted when cellular miR-195 was neutralized by anti–miR-195 (Fig. 3C,D).

Figure 3.

miR-195 suppresses tumor migration and invasion in vitro. (A, B) Restoration of miR-195 inhibited HCC cell migration and invasion. QGY-7703 cells that were nontransfected (left) or transfected with NC (center) or miR-195 duplex (right) were added to transwell chambers without (A) or with (B) Matrigel coatings and incubated for 12 hours, followed by staining with crystal violet. (C, D) Antagonism of miR-195 enhanced HCC cell migration and invasion. QGY-7703 cells transfected with anti-NC or anti–miR-195 were added to transwell chambers without (C) or with (D) Matrigel coatings and incubated for 12 hours. Scale bars, 100 μm. *P < 0.05; **P < 0.01; ***P < 0.001.

To further validate the above findings in vivo, orthotopic liver implantations were conducted with the QGY-miR-195-LUC cell line (Supporting Fig. 3). Mice that were injected with QGY-miR-195-LUC were fed with doxycycline for 10 days before being divided into miR-195–on and miR-195–off groups, followed by doxycycline withdrawal in the miR-195–on group and continued doxycycline treatment in the miR-195–off mice for another 40 days. The tumor sizes remained similar between the two groups before the 40th day postimplantation but were reduced in the miR-195–on group on the 50th day (Supporting Fig. 4A,B; tumor incidence for miR-195–off versus miR-195–on groups: 6/8 versus 5/8). Furthermore, the miR-195–on group displayed fewer MVD in the xenografts and fewer metastases in the liver and lung compared with the miR-195–off group on the 50th day (Supporting Fig. 4C-E).

To exclude the potential confounding effect that decreased angiogenesis and metastasis in the miR-195–on group might result from smaller tumors with fewer metabolic demands and fewer cells, an independent experiment was performed to assess tumor angiogenesis and metastasis on the 40th day, the time point when the tumor incidence (miR-195–off versus miR-195–on groups: 9/15 versus 8/14) and tumor sizes (Supporting Fig. 5) were comparable between the groups. As observed in the results on the 50th day, the miR-195–on group had a much lower MVD (Fig. 4A), decreased metastasis incidence (miR-195–off versus miR-195–on groups: 7/9 versus 4/8), and reduced sizes and numbers of metastatic nodules in the liver (Supporting Fig. 6A,B and Fig. 4B,C) and lung (Supporting Fig. 6C,D and Fig. 4D,E) compared with the miR-195–off group.

Figure 4.

miR-195 suppresses tumor angiogenesis and metastasis in vivo. (A) Orthotopic xenograft tumors from miR-195–on mice displayed lower microvessel densities. On the 40th day postimplantation, tumor tissues from miR-195–off or miR-195–on mice were dissected and stained for CD146. Scale bar, 50 μm. (B-E) Restoration of miR-195 expression inhibited tumor metastasis in an orthotopic xenograft model. Intrahepatic metastases (B, C) and pulmonary metastases (D, E) were detected via ex vivo bioluminescent imaging (B, D) and histopathological analysis (C, E). For (B) and (D), on the 40th day postimplantation, mice were sacrificed 5 minutes after injection of luciferin, and the organs were collected and imaged ex vivo. Luciferase signals were quantified by measuring photons per second. For imaging of the intrahepatic metastases (B), the left hepatic lobe with primary tumors was removed before imaging. For (C) and (E), hematoxylin-eosin staining was performed on serial sections of the livers (C) and lungs (E) to detect the metastatic nodules. The numbers of tumor nodules were examined and counted under an anatomical microscope. Scale bar, 50 μm. For (A-E), QGY-miR-195-LUC cells were inoculated under the capsule of the left hepatic lobe of BALB/c nude mice. All mice were fed doxycycline for the first 10 days prior to division into the miR-195–on and miR-195–off groups, which was followed by doxycycline withdrawal in the miR-195–on group and continued doxycycline treatment in the miR-195–off mice for 30 days. The central horizontal line represents the mean value; the error bars represent the SEM. *P < 0.05; **P < 0.01.

Collectively, both the in vitro and in vivo studies suggest the suppressive effects of miR-195 on HCC angiogenesis and metastasis.

miR-195 Inhibits HCC Angiogenesis by Directly Targeting VEGF

We then explored the molecular mechanism behind the antiangiogenic function of miR-195. Putative targets of miR-195 were predicted with TargetScan. Among these, VEGF was chosen for further validation due to its well-known importance in tumor angiogenesis.[25] A dual-luciferase reporter assay revealed that the cotransfection of miR-195 significantly inhibited the activity of firefly luciferase reporter with wild-type 3′UTR of VEGF, whereas this effect was abrogated when the predicted 3′UTR binding site was mutated (Fig. 5A, and Supporting Fig. 7A). Moreover, both gain-of-function and loss-of-function analyses disclosed that miR-195 diminished the expression of cellular VEGF and the level of secreted VEGF in the TCM (Fig. 5B and Supporting Fig. 7B,C). Consistently, xenografts from the miR-195–on mice showed much lower VEGF levels compared with those from the miR-195–off controls (Supporting Fig. 7D). Additionally, the inverse correlation between miR-195 and VEGF expression was confirmed in human HCC tissues (Fig. 5C and Supporting Fig. 7E). These data indicate that miR-195 may negatively regulate VEGF expression by directly targeting its 3′UTR.

Figure 5.

miR-195 exerts its antiangiogenic function by targeting VEGF. (A) miR-195 inhibited the activity of a luciferase reporter that contained the wild-type 3′UTR of VEGF. NC or miR-195 duplexes were cotransfected with pRL-TK and a firefly luciferase reporter plasmid that contained either the wild-type (Wt) or mutant (Mut) 3′UTR of VEGF. pRL-TK that expressed Renilla luciferase was used as an internal control to calibrate the differences in the transfection and harvest efficiencies. The firefly luciferase activity of each sample was normalized to the Renilla luciferase activity. (B) Expression of miR-195 reduced the levels of cellular VEGF and secreted VEGF. HCC cells that were nontransfected or transfected with NC or miR-195 for 48 hours were analyzed via immunoblotting, and the TCM from the cells was analyzed via ELISA. Glyceraldehyde 3-phosphate dehydrogenase (GAPDH) was used as an internal immunoblotting control. (C) The level of miR-195 correlated inversely with VEGF expression in human HCC tissues. miR-195 and VEGF expression were analyzed in 90 HCC tissues from the HCC cohort described in Fig. 1. miR-195 was detected via qRT-PCR, and VEGF was detected via immunohistochemical staining. (D) Restoration of miR-195 in HCC cells attenuated VEGFR2 signaling in ECs. Twenty-four hours after seeding, HUVECs were grown in SFM for 12 hours and then cultured for 15 minutes in the presence of SFM (lane 1) or 100% TCM from QGY-7703 cells that were nontransfected (lane 2) or transfected with NC (lane 3) or miR-195 (lane 4), followed by immunoblotting for phosphor-Tyr1175-VEGFR2, VEGFR2, phosphor-T202/Y204-ERK1/2, and ERK1/2 expression. β-Actin was used as an internal control. (E) Introduction of VEGF antagonized the antiangiogenic effect of miR-195. HUVECs were cultured in TCM derived from QGY-7703 cells that were cotransfected with RNA duplex and an expression plasmid in the following combinations: NC/empty vector (first bar), miR-195/empty vector (second bar), NC/VEGF (third bar) or miR-195/VEGF (fourth bar). HUVEC capillary tube formation was assessed subsequently. (F) MVD correlated positively with VEGF expression in HCC tissues. The median MVD value of the 90 cases that were used in Fig. 5C was chosen as the cutoff point for separating the low MVD (n = 45) and high MVD (n = 45) groups. The central horizontal line represents the mean; error bars represent the SEM. For (B) and (D), results were reproduced in three independent experiments, and representative immunoblots are shown. *P < 0.05; **P < 0.01.

It has been demonstrated that tumor-secreted VEGF binds to VEGF receptor 2 (VEGFR2) in endothelial cells and induces the phosphorylation and activation of VEGFR2, which then phosphorylates extracellular signal-regulated kinase (ERK) and promotes angiogenesis.[24] Compared with the controls (SFM), HUVECs that were incubated with TCM from NC-transfected or nontransfected HCC cells displayed significantly increased phosphorylation of VEGFR2 and ERK, whereas the TCM-promoted VEGFR2 signaling was attenuated dramatically when TCM from miR-195 transfectants was applied (Fig. 5D and Supporting Fig. 8A). In contrast, coculture with the TCM from anti–miR-195 transfectants enhanced VEGFR2 signaling in HUVECs (Supporting Fig. 8B).

We further verified whether VEGF could mediate the antiangiogenic function of miR-195 and found that VEGF knockdown in HCC cells displayed a significantly reduced capacity to promote HUVEC migration and capillary tube formation (Supporting Fig. 9A-C), which phenocopied the effects of miR-195 expression. In contrast, the overexpression of VEGF in miR-195-transfected HCC cells attenuated the anti-angiogenic effects of miR-195 (Fig. 5E and Supporting Fig. 10A,B). Furthermore, higher VEGF levels were associated with higher MVD in human HCC tissues (Fig. 5F), corresponding to the correlation between lower miR-195 expression and higher MVD/VEGF levels in HCC tissues (Fig. 1B, 5C).

These results suggest that miR-195 may repress tumor angiogenesis by inhibiting VEGF in HCC cells and subsequently abrogating the proangiogenesis signaling of VEGF/VEGFR2 in endothelial cells.

miR-195 Represses HCC Metastasis by Negatively Regulating VAV2 and CDC42 Expression

Next, the mechanism by which miR-195 inhibited tumor metastasis was elucidated. Among the predicted targets of miR-195, VAV2 and CDC42 stood out as attractive candidates because they promote cell motility and their up-regulation is associated with metastasis.[25] Dual-luciferase reporter analysis and immunoblotting assays revealed that miR-195 directly suppressed the expression of VAV2 and CDC42 (Fig. 6A,B and Supporting Fig. 11A-C). Consistently, xenografts from the miR-195–on mice had much lower VAV2 and CDC42 levels compared with controls (Supporting Fig. 11D). Furthermore, miR-195 down-regulation correlated with the overexpression of VAV2 and CDC42 in human HCC specimens (Fig. 6C and Supporting Fig. 11E).

Figure 6.

miR-195 suppresses metastasis by targeting VAV2 and CDC42. (A) miR-195 inhibited the activity of a luciferase reporter that contained the wild-type 3′UTR of VAV2 or CDC42. NC or miR-195 duplexes were cotransfected with pRL-TK and a firefly luciferase reporter plasmid that contained the wild-type (Wt) or mutant (Mut) 3′UTR of VAV2 or CDC42. Luciferase activity analysis was performed as in Fig. 5A. (B) Expression of miR-195 reduced the protein levels of endogenous VAV2 and CDC42. QGY-7703 cells that were nontransfected (lane 1) or transfected with NC (lane 2) or miR-195 (lane 3) for 48 hours were analyzed via immunoblotting. (C) The level of miR-195 correlated inversely with VAV2 and CDC42 expression in human HCC tissues. The expression of miR-195, VAV2, and CDC42 was analyzed in the 90 HCC tissues used in Fig. 5C. miR-195 was detected via qRT-PCR and VAV2 and CDC42 by immunohistochemical staining. (D) Metastatic HCC displayed higher VAV2 and CDC42 expression. The absence (n = 51) or presence (n = 39) of venous invasion (n = 37) or lymph node metastasis (n = 1) or both (n = 1) are indicated with a minus sign (−) and plus sign (+), respectively. The central horizontal line represents the mean; error bars represent the SEM. (E) Restoration of miR-195 decreased the activity of Rac1. Lysates from QGY-7703 cells that were nontransfected (lane 2) or transfected with NC (lane 3) or miR-195 (lane 4) were incubated with glutathione-Sepharose bead-immobilized GST-PAK protein. The bead-bound proteins were solubilized in SDS buffer and analyzed by immunoblotting for GTP-bound Rac1. A lysate of QGY-7703 cells incubated with bead-immobilized GST (lane 1) was used as a negative control. Aliquots of each lysate were saved and used for the detection of total Rac1. (F) Overexpression of miR-195 inhibited lamellipodia formation in QGY-7703 cells. QGY-7703 cells that were nontransfected or transfected with NC, miR-195, si-VAV2, or si-CDC42 for 48 hours were added to Matrigel-coated plates and incubated for 1 hour, followed by staining for filamentous actin. For (B) and (E), results were reproduced in three independent experiments, and representative immunoblots are shown. *P < 0.05; **P < 0.01.

Next, we showed that, similar to the phenotype induced by miR-195 expression, the silencing of either VAV2 or CDC42 obviously decreased the motility of HCC cells (Supporting Fig. 12A,B), whereas the overexpression of either in miR-195 transfectants abrogated the inhibitory effects of miR-195 on cell migration (Supporting Fig. 13A,B). Moreover, metastatic HCC displayed higher levels of VAV2 and CDC42 (Fig. 6D), which was in agreement with the inverse correlation of miR-195 expression with VAV2/CDC42 levels and HCC metastasis.

VAV2 is known to act as a guanine nucleotide exchange factor that activates Rac1 and CDC42 by facilitating the exchange of guanosine diphosphate to guanosine triphosphate (GTP), and GTP-bound Rac1 and CDC42 activate multiple cytoskeletal proteins to induce actin polymerization and lamellipodia formation, which are required for metastasis.[26] A GST pull-down assay revealed that miR-195 restoration decreased the amounts of GTP-bound Rac1 (Fig. 6E). Furthermore, miR-195-transfection, like silencing of VAV2 and CDC42, resulted in a dramatic reduction in the fraction of cells with lamellipodia (Fig. 6F, Supporting Fig. 14), indicating that miR-195 may suppress metastasis by suppressing VAV2 and CDC42, which in turn attenuates VAV2/Rac1/CDC42 signaling.

Discussion

Malignant tumors, including HCC, are characterized by high vascularity and frequent metastasis. Angiogenesis is critical for tumor progression, whereas metastasis is the major cause of tumor recurrence and patient death. Therefore, miRNAs that possess antiangiogenic or antimetastatic activities may provide novel targets for anticancer therapies. Based on in vitro and in vivo evidence, we propose that miR-195 is capable of suppressing HCC angiogenesis and metastasis and the down-regulation of miR-195 may facilitate HCC progression.

Most publications have focused on the regulatory function of miR-195 in cell proliferation and apoptosis.[17, 18, 22] Here, we identified the antiangiogenic function of miR-195 based on the following evidence: in vitro studies showed that restoration of miR-195 inhibited the capacity of HCC cells to promote endothelial cell migration and tube formation; in vivo mouse models revealed that the induced expression of miR-195 in HCC cells significantly reduced angiogenesis in orthotopic xenograft tumors; in human HCC tissues, MVD was inversely associated with the level of miR-195; and gain-of-function and loss-of-function studies showed that miR-195 repressed tumor angiogenesis by targeting VEGF, one of the most important proangiogenic factors secreted by tumor cells. Tumor cells are the critical initiators and promoters of angiogenesis. Our data suggest that miR-195 down-regulation in HCC cells may result in enhanced VEGF levels in the tumor microenvironment, which subsequently activated VEGFR2 signaling in endothelial cells and thereby promoted angiogenesis.

Antimetastatic activity is another function of miR-195 that we identified in this study. Recently, two groups employed the in vitro transwell system and showed that miR-195 suppressed the invasion of glioblastoma and breast cancer cells through Matrigel.[19, 21] Herein, we disclosed that miR-195 suppressed HCC metastasis, based on observations from human specimens as well as in vitro and in vivo models. Importantly, we presented evidence that the induction of miR-195 expression markedly decreased the intrahepatic and pulmonary metastasis of orthotopic xenograft HCC tumors and that the down-regulation of miR-195 in human HCC tissues was associated with enhanced metastasis. Furthermore, we identified VAV2 and CDC42 as two novel targets that were at least partly responsible for the antimetastatic function of miR-195. This study, together with those from other groups, suggests a crucial inhibitory function of miR-195 in tumor migration, invasion, and metastasis.

Previously, we showed that miR-195 overexpression inhibited growth and that the introduction of miR-195 duplex into MHCC-97L or HCT-116 cells led to significant reductions in both the incidence and sizes of subcutaneous xenograft tumors.[17] Consistently, we observed a correlation between decreased miR-195 level and increased Ki-67–positive HCC cells in human specimens (Supporting Fig. 15). One could argue that the fewer cells and lower metabolic demands in smaller tumors might lead to decreased angiogenesis and metastasis. However, our data clearly suggest that miR-195 can directly repress tumor angiogenesis and metastasis. As shown in this study (Fig. 3 and Supporting Figs. 2, 16, and 17), miR-195 significantly suppressed the in vitro migration of all examined cell lines prior to the appearance of the growth inhibitory effects of miR-195, although the extent of the growth inhibition was variable among the different tumor cell lines. Furthermore, for our in vivo study, we chose QGY-7703, an HCC cell line that displays less growth inhibition by miR-195 (Supporting Figs. 16 and 17), to create a subline (QGY-miR-195-LUC) with the Tet-off inducible expression of miR-195. Moreover, miR-195 expression was restored at 10 days postimplantation of the QGY-miR-195-LUC cells, which allowed the xenograft tumors to establish in the miR-195–on mice and limited the differences in tumor sizes between the miR-195–off and miR-195–on groups. Based on in vivo imaging analyses, we decided on a time point (the 40th day postimplantation) at which the tumor sizes of the two groups were similar. At this time point, we detected significant decreases in angiogenesis and metastasis in the miR-195–restored tumors. In accordance with the antiangiogenic and antimetastatic activities of miR-195, a mechanistic investigation revealed that miR-195 directly targeted the proangiogenic factor VEGF and the prometastatic factors VAV2 and CDC42. Knockdowns of each target phenocopied the effects of miR-195 restoration, whereas the introduction of each target antagonized the function of miR-195. Together, these data suggest that the antiangiogenic and antimetastatic effects of miR-195 are independent of its growth-suppressive function.

Interestingly, miR-195 significantly inhibited the proliferation of HCC lines, such as MHCC-97L, MHCC-97H, Hep3B, and SMMC-7721, but had fewer inhibitory effects on the proliferation of other HCC lines such as QGY-7703 and Huh-7 (Supporting Figs. 16 and 17). Previously, we showed that miR-195 suppressed proliferation in MHCC-97L by repressing the phosphorylation of Rb and thereby attenuating the transcription of S-phase genes.[17] Herein, we found that miR-195 expression significantly reduced the levels of both phosphorylated Rb protein and mRNA of S-phase genes in MHCC-97L cells, but not in QGY-7703 cells (Supporting Fig. 18). The underlying mechanisms that determine these different responses to miR-195 expression might reflect differences in cellular contexts and remain to be elucidated.

miR-195 has been shown to promote apoptosis or inhibit proliferation in multiple cancer types.[17, 18, 22] Together with the findings from this study, it is intriguing to find that a single miRNA can regulate several phenotypes of cancer cells and thereby affect different stages of cancer (Supporting Fig. 19). Such an miRNA may represent a promising molecular target for anticancer therapies.

Ancillary