SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. References
  7. Supporting Information

In our previous study, osteopontin (OPN) was identified as one of the leading genes that promote the metastasis of hepatocellular carcinoma (HCC). However, the mechanism by which OPN promotes metastasis of HCC is not understood. In this study, RNA interference mediated by viral vectors—which could induce a long-lasting down-regulation in gene expression—was applied to analyze the role of OPN in metastasis of HCC. Three lentiviral vectors encoding microRNA against OPN, Lenti.OPNi-1, Lenti.OPNi-2, and Lenti.OPNi-3, were constructed and found to down-regulate the OPN level by 62%, 78%, and 95%, respectively, in HCCLM3 cells which had an overexpression of OPN and a higher metastatic potential. Consequently, both Lenti.OPNi-2 and Lenti.OPNi-3 induced a significant decrease in matrix metalloproteinase (MMP)-2 and urokinase plasminogen activator expression, and led to an obvious inhibition of both in vitro invasion and in vivo lung metastasis of HCCLM3 cells (P < 0.001). Moreover, Lenti.OPNi-3, rather than Lenti.OPNi-2, could also suppress in vitro proliferation and in vivo tumor growth of HCCLM3. Smaller detectable tumors were found in only 50% of mice after implantation of Lenti.OPNi-3–transfected HCCLM3 cells (341 ± 502.6 mm3 versus >3500 mm3 in controls; P < 0.001). Lenti.OPNi-3, not Lenti.OPNi-2, significantly suppressed the MEK/ERK1/2 pathway in HCCLM3 cells. Recombinant OPN was found to induce translocation of p65 into the nucleus of HCC cells and activation of MMP-2 and MEK/ERK/1/2, which were suppressed by the nuclear factor κB (NF-κB) inhibitor pyrrolidine dithiocarbamate. Conclusion: OPN plays an important role in metastasis as well as tumor growth of HCC, in which different minimum threshold levels of OPN are needed. These effects may occur through activation of the mitogen-activated protein kinase and NF-κB pathways, and MMP-2. OPN could be a hopeful target for the control of HCC. (HEPATOLOGY 2008;48:1834-11842.)

Hepatocellular carcinoma (HCC) is the third leading cause of cancer death in the world, and the second in China.1, 2 The extremely poor prognosis of patients with HCC is largely due to the high rate of tumor recurrence or intrahepatic metastasis after surgical resection.3 Therefore, it is very important to search for molecular markers related to metastasis, which would provide new predictors as well as new targets for intervention of metastatic recurrence of HCC.

Osteopontin (OPN) is a noncollagenous, sialic acid–rich, glycosylated phosphoprotein that is involved in normal tissue remodeling processes as well as certain diseases.4–6 It has been shown that OPN also plays an important role in the development, invasion, and metastasis of malignancies.7–9 However, less definitive evidence has been reported in HCC.

In our previous gene expression profiling study, OPN was found to be overexpressed in metastatic HCC and was identified as one of the leading genes associated with metastasis of HCC.10 Moreover, the preoperative plasma level of OPN could serve as a predictor of tumor recurrence and prognosis of patients with HCC after operation.11 However, the real mechanism by which OPN mediates the metastasis of HCC is not clear.

RNA interference (RNAi) is a posttranscriptional gene silencing strategy that is specific for the evolutionarily conserved sequence, and lentiviral-based expression systems could mediate a long-lasting RNAi-based gene knockdown.12 Stegmeier13 found that polymerase II–transcribed short hairpin RNA displayed more efficiently in knockdown of gene expression when it was embedded in a microRNA (miRNA) context. In this study, we used the powerful lentiviral RNAi expression vectors with specific miRNA against OPN to block the expression of OPN, to investigate its effect on the proliferation and metastasis of HCC cells with different OPN expression levels and different metastatic potentials, and to explore the mechanism of OPN involved in the progression and metastasis of HCC.

Materials and Methods

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. References
  7. Supporting Information

Cell Lines.

Five human hepatoma-derived cell lines (HepG2, Hep3B, MHCC97L, MHCC97H, and HCCLM3) with different metastatic potentials and one untransformed liver cell line (CCL13) were used in this study. MHCC97-L, MHCC97-H, and HCCLM3 were established from the same parent human HCC cell line, MHCC97, in the authors' institution. These cell lines have an identical genetic background and stepwise increasing metastatic potentials (for example, MHCC97-L had the lowest ability of metastasis, and HCCLM3 had the highest).14, 15 They were cultured in Dulbecco's modified Eagle's medium (Gibco BRL, New York, NY) supplemented with 10% (vol/vol) fetal bovine serum (Hyclone, Logan, UT) at 37°C in a humidified incubator containing 5% CO2. HepG2 and Hep3B cells (low invasive) were obtained from American Type Culture Collection (Rockville, MD) and cultured in RPMI1640 medium supplemented with 10% fetal bovine serum.

Construction and Transfection of Lentivial Vectors with Specific miRNAs for OPN.

Three precursor miRNAs (pre-miRNA) sequences targeting to OPN (GenBank accession number NM_000582) were designed using an Internet application system (Invitrogen, CA) (Fig. 1A). Double-stranded oligonucleotide encoding pre-miRNA sequence and its mismatch mutants were annealed and inserted into pcDNATM6.2-GW/ EmGFP-miR expression vector (Invitrogen) containing human cytomegalovirus promoter and herpes simplex virus thymidine kinase polyadenylation signal. The pDONRTM221 vector was used as an intermediate to transfer the pre-miRNA expression cassette into the lentiviral expression plasmid (pLenti6/V5-DEST) using Gateway Technology (Invitrogen). The new miRNA expression vectors (pLenti6/V5-GW/EmGFP-miR) (Fig. 1B) and packaging vectors (ViraPowerTM Packaging Mix) (Invitrogen) were cotransfected into 293FT cells with Lipofectamine 2000 (Invitrogen). The culture supernatants were collected, concentrated, and used as a virus stock.

thumbnail image

Figure 1. Construction of lentiviral-mediated miRNA against OPN and their impact on OPN expression levels of HCC cells. (A) Predicted secondary structure of the pre-miRNA targeting OPN (OPNi-1, OPNi-2, and OPNi-3) and a control pre-miRNA (OPNi-3M) with two mismatch mutations (asterisks). (B) Lentiviral miRNA expression system. RRE, human immunodeficiency virus-1 Rev response element; PCMV, cytomegalovirus promoter; EmGFP, emerald green fluorescent protein; attB1 and attB2, two recombination sites; EM7, synthetic prokaryotic promoter for expression of the selection marker in Escherichia coli. The SV40 promoter allows high-level expression of the selection marker and episomal replication in cells expressing the SV40 large T antigen. (C) Inhibitory effects of lentiviral-mediated miRNA on OPN expression. OPN levels were down-regulated by 9%, 62%, 78%, and 95%, respectively, in HCCLM3 cells transfected with Lenti.OPNi-3M, Lenti.OPNi-1, Lenti.OPNi-2, or Lenti.OPNi-3. Phosphate-buffered saline–treated HCCLM3 cells served as mock control. (D) Flow cytometry revealed that Lenti.OPNi-3–transfected HCCLM3 cells were 100% positive for EGFP when the MOI was 50.

Download figure to PowerPoint

All lentiviral vectors expressed enhanced green fluorescent protein (GFP), which allowed for titering and measuring their infection efficiency in transfected cells. The viral titer was determined by counting GFP-positive cells after transfection. The lentiviral vectors were transfected into the HCC cells with a multiplicity of infection (MOI) from 30 to 50 in the presence of polybrene (6 μg/mL).

Detection and Localization of Tested Proteins.

The expression levels of proteins tested including OPN, MMP-2, MMP-9, urokinase plasminogen activator (uPA), glyceraldehyde-3-phosphate dehydrogenase (GAPDH), p65 nuclear factor κB (NF-κB), phospho-MEK1/2, phospho-ERK1/2, and ERK1/2 were evaluated via western blotting. Five to fifty micrograms of total protein extracted from the tested cells was separated via sodium dodecyl sulfate–polyacrylamide gel electrophoresis and transferred onto polyvinylidene fluoride membranes. The primary antibodies against OPN (R&D Systems, Minneapolis, MN), MMP-2, MMP-9 (Sigma, St. Louis, MO), uPA, glyceraldehyde-3-phosphate dehydrogenase, p65 (Santa Cruz Biotechnology, Santa Cruz, CA), phospho-MEK1/2, phospho-ERK1/2, and ERK1/2 (Cell Signal Tech, Danvers, MA) were used. To investigate whether RNAi for OPN resulted in the potential nonspecific activation of interferon (IFN) and its downstream effectors,16 expression of P-eIF2α was also assessed by Western blot analysis (Cell Signal Tech).

To further investigate the possible role of OPN in the regulation of signal pathway, HCCLM3 cells were incubated with 10 μM of recombinant human OPN (R&D Systems) at 37°C for 120 minutes, their nuclear and cytoplasm extracts were prepared to detect the level of p65 via western blotting as described.17 Immunofluorescence assay was also applied to localize the p65 protein in HCC cells. HCCLM3 cells were cultured on the glass slides and then treated with 10 μM of recombinant human OPN at 37°C for a period of 0-120 minutes, followed by washing and fixation with ice-cold methanol. The cells were then incubated with rabbit polyclonal anti-p65 primary antibody (1:50) and fluorescein isothiocyanate–conjugated anti-rabbit immunoglobulin G (1:100) (Santa Cruz Biotechnology) and were visualized under confocal microscopy (Leica, Heidelberg, Germany).

Detection of MMP-2 Activity by Gelatin Zymography.

To evaluate the effect of OPN on the activity of MMP-2, HCC cells transfected with the lentiviral vectors described above were incubated with different concentrations of recombinant human OPN in a serum-free medium at 37°C for 18 hours. The medium was collected to detect the protein concentrations using Bio-Rad assay and were loaded onto zymographic sodium dodecyl sulfate gel containing gelatin (1 mg/mL). Then the gel was incubated in renaturing buffer and developing buffer (Invitrogen) according to the manufacturer's instructions. The enzyme activity was visualized by staining the gel with Coomassie Blue R-250.

Detection of Cellular Proliferation by MTT Assay.

The cellular proliferation of transfected cells was measured via MTT assay. Briefly, 10 μL of MTT (5 mg/mL) (Sigma) was added to each well with a final volume of 100 μL of culture medium containing viable cells. After an additional incubation for 4 hours, the resulting formazan was dissolved in 100 μL of isopropanol with 40 mM hydrochloric acid. Spectrophotometric absorbance at 570 nm (for formazan dye) was measured with absorbance at 630 nm as a reference. Each assay was repeated three times.

In Vitro Invasion Matrigel Assays.

The invasive ability of HCC cells was determined using Matrigel (BD Pharmingen) coated 24-well transwell chambers with upper and lower culture compartments separated by polycarbonate membranes with 8-μm pore (Costar, New York, NY). The bottom chamber was filled with Dulbecco's modified Eagle's medium containing 10% fetal bovine serum as a chemoattractant. The transfected cells (5 × 104) were seeded on the top chamber and incubated at 37°C in 5% CO2 humidified air for 24 hours. The cells that migrated to the underside of the membrane were stained with Giemsa (Sigma) and counted with a microscope (Olympus, Tokyo, Japan).

In Vivo Assays for Tumor Growth and Metastasis.

HCCLM3 cells (5 × 106) transfected with Lenti.OPNi-2 or Lenti.OPNi-3 (at a MOI of 50) were implanted subcutaneously into the flank of nude mice (6 in each group, male BALB/c nu/nu, 4-6 weeks) (Institute of Materia Medica, CAS, Shanghai, China), and HCCLM3 cells treated with phosphate-buffered saline were used as a mock control. Tumor growth was monitored with tumor volume, which was calculated as described18: V (mm3) = width2 (mm2) × length (mm)/2. The mice were sacrificed 6 weeks later, and the lungs were removed. Consecutive sections were made for every tissue block of the lung and stained with hematoxylin-eosin. The incidence and classification of lung metastasis were calculated and evaluated independently by two pathologists. Based on the number of HCC cells in the maximal section of the metastatic lesion,10 the lung metastases were classified into four grades: grade I, ≤20 cells; grade II, 20-50 cells; grade III, 50-100 cells; and grade IV, >100 cells. These procedures were approved by The Animal Care and Use Committee of Fudan University, China.

Bioluminescence Imaging.

The tumor growth and metastasis were also monitored with bioluminescence imaging every 7 days for 6 weeks. In vivo bioluminescence imaging was performed using the NightOWL LB981 system (Berthold Technologies, Bad Wildbad, Germany). The emission spectrum is filtered using an HQ 520 bandpass filter (Chroma Technology) to enhance the GFP fluorescence relative to the autofluorescence signal from endogenous tissue. The mice were sedated (1.5 mg/kg acepromazine and torbugesic intramuscularly), placed in the light-tight chamber, and imaged with an exposure time of 100 milliseconds. Alterations in fluorescent signals were depicted on a graph. Images were acquired and processed using WinLight32 software.

Statistic Analysis.

All data are expressed as the mean ± standard deviation, and analyzed via analysis of variance and Student t test. The level of significance was set at P < 0.05.

Results

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. References
  7. Supporting Information

Expression Levels of OPN in HCC Cell Lines.

Using western blotting, the expression levels of OPN in HCCLM3 and MHCC97-H cell lines were much higher than that in MHCC97L and Hep3B cells, while the lowest level of OPN among the 5 HCC cells tested was in HepG2 cells. These results provide further evidence that OPN correlates with the metastatic phenotype of HCC cells. Therefore, HCCLM3 was selected and used in the following RNAi experiments (Supplementary Fig. 1).

Identification of Specific and Efficient miRNA Sequence Against OPN.

Three lentiviral vectors expressing miRNA against OPN (Lenti.OPNi-1, Lenti.OPNi-2, and Lenti.OPNi-3) were constructed and the scrambled sequence was classified as Lenti.OPNi-3M. After being transfected with a MOI of 50 for 72 hours, the protein levels of OPN were down-regulated by 9%, 62%, 78%, and 95%, respectively, in HCCLM3 cells transfected with Lenti.OPNi-3M, Lenti.OPNi-1, Lenti.OPNi-2, or Lenti.OPNi-3 (Fig. 1C) compared with the mock control, while there was no significant difference in OPN protein levels between the mock control and Lenti.OPNi-3M. When the MOI was 50, almost 100% of the Lenti.OPNi-3–transfected HCCLM3 cells were positive for enhanced GFP as detected with flow cytometry (Fig. 1D).

Effect of Down-Regulation of OPN on In Vitro Proliferation and Invasion of HCCLM3 Cells.

Based on their effects on OPN expression, Lenti.OPNi-2 and Lenti.OPNi-3 were selected to evaluate the effects of down-regulation of OPN on the in vitro proliferation and invasion of HCCLM3 cells. It was found that Lenti.OPNi-3, rather than Lenti.OPNi-2, could significantly inhibit the proliferation of HCCLM3 cells (P < 0.001) (Fig. 2A). Similar results were found in MHCC97 cells, which also had a higher expression level of OPN (Supplementary Fig. 2A). These data suggest that even a lower expression level of OPN is sufficient for the proliferation of HCC cells. Moreover, miRNA-induced OPN down-regulation did not have a significant effect on the proliferation of HepG2 cells, which have a lower expression level of OPN (Supplementary Fig. 3A).

thumbnail image

Figure 2. Effect of lentiviral-mediated miRNA against OPN on in vitro proliferation and invasion of HCC cells. (A) The proliferative rates of different miRNA-transfected HCC cells were assessed via MTT assay. Lenti.OPNi-3, but not Lent.OPNi-2, significantly inhibited cellular proliferation compared with mock control. *P < 0.001. (B) The invasive activity of miRNA-transfected HCC cells assessed using Matrigel-coated transwell chambers. Both Lenti.OPNi-2 and Lenti.OPNi-3 significantly inhibited the invasion of HCCLM3 cells. *P < 0.001 compared with mock control.

Download figure to PowerPoint

In the Matrigel assays, the migrated cell numbers of HCCLM3 cells transfected with Lenti.OPNi-2 (9.2±1.3) and Lenti.OPNi-3 (7.8±1.5) were much lower than that of the mock cells (28.8±2.6, P < 0.001), which suggests that both Lenti.OPNi-2 and Lenti.OPNi-3 could significantly suppress the invasion of HCCLM3 cells. No significant difference was found between HCCLM3 cells transfected with Lenti.OPNi-3M (27±1.6) and mock control (28.8±2.6; P = 0.152) (Fig. 2B). Similar effects were also detected in MHCC97H cells; both Lenti.OPNi-2 and Lenti.OPNi-3 could significantly inhibit in vitro invasion of MHCC97-H cells (P < 0.001 compared with mock control) (Supplementary Fig. 2B,C).

Effect of Down-Regulation of OPN on In Vivo Tumor Growth and Metastasis of HCC.

To further investigate the effect of down-regulation of OPN expression induced by miRNA on the in vivo tumor growth and metastasis of HCC, HCCLM3 cells transfected with the lentiviral vectors (Lenti.OPNi-2, Lenti.OPNi-3, or Lenti.OPNi-3M) or phosphate-buffered saline (as mock control) were subcutaneously implanted into nude mice. The mice were sacrificed, and the tumors and lungs were removed at 6 weeks postimplantation. Detectable tumor masses (341 ± 502.6 mm3) could be found in only 50% of mice in the Lenti.OPNi-3 group, while much larger tumors (>3500 mm3) were detected in all mice in the mock and Lenti.OPNi-3M controls, and in the Lenti.OPNi-2 group (P < 0.001) (Fig. 3A,B).

thumbnail image

Figure 3. The effect of lentiviral-mediated miRNA against OPN on the in vivo tumor growth and lung metastasis of HCCLM3 cells. (A,B) In vivo tumor growth in Lenti.OPNi-3 mice was significantly inhibited compared with mock controls, whereas no significant difference in tumor volumes was found in Lenti.OPNi-2 and Lenti.OPNi-3M mice. (C,D) Both the number and size of lung metastases in Lenti.OPNi-2 and Lenti.OPNi-3 mice were much decreased compared with mock controls; however, (C) no significant difference was observed in Lenti.OPNi-3M mice. (D) Representative lung tissue sections for grades I to IV (hematoxylin-eosin, magnification ×50). Classification of lung metastatic foci was based on the cell number in the metastatic lesion.

Download figure to PowerPoint

In Lenti.OPNi-2 and Lenti.OPNi-3 groups, enhanced fluorescent signal could be detected using the Luminescent and Fluorescent Image Analyzer (Berthold, Germany) in the subcutaneous implantation sites of HCCLM3 cells. However, the enhanced GFP signal from deep small metastatic lesions was too weak to be detected. Interestingly, fluorescent signals could also be detected at the inoculation site of mice in the Lenti.OPNi-3 group, even though no detectable tumor mass could be found (Supplementary Fig. 4). These suggested that there might still be a few HCC cells alive there even though the Lenti.OPNi-3 significantly suppresses the proliferation of HCCLM3 cells.

To examine the lung metastasis, 130-180 consecutive sections were made and stained with hematoxylin-eosin for each lung tissue block. The incidences of lung metastasis in the Mock controls, Lenti.OPNi-2, Lenti.OPNi-3, and Lenti.OPNi-3M groups were 100%, 66.7%, 50%, and 100%, respectively. The numbers of lung metastatic lesions in the Lenti.OPNi-2 and Lenti.OPNi-3 groups (1.2±1.3) were much decreased compared with those mock controls (14±2.4) and the Lenti.OPNi-3M controls (13.5±2.1) (P < 0.001). And also, their metastatic lesions were much smaller than that in the controls (most of the lung metastatic lesions in the Lenti.OPNi-2 and Lenti.OPNi-3 groups were grade I, while those in the controls were even grade III-IV) (Fig. 3C,D).

To evaluate the influence of different implantation sites on the formation of lung metastases, HCCLM3 cells were implanted in the thoracic and rear abdominal areas of mice in the mock controls. No significant difference in the incidence and degree (grade I-IV) of pulmonary metastasis was found between these two different implantation sites (see Supplementary Fig. 5).

The Mechanisms of OPN Regulating the Proliferation and Metastasis of HCC.

To explore the possible mechanisms of OPN in regulating tumor growth and metastasis of HCC, the expression levels of P-eIF2α (an IFN-response gene) and p65 NF-κB, and activation of MEK and ERK1/2 were investigated in HCC cells transfected with miRNA against OPN. Compared with mock controls, no significant alteration in P-eIF2α level was observed in the different transfectant groups, even though a significantly increased expression of P-eIF2α was detected in HCCLM3 cells treated with IFN-α (served as a positive control) (P < 0.05) (Supplementary Fig. 6), suggesting that the effects of down-regulation of OPN on tumor growth and metastasis of HCC are not related to the activation of IFN response genes.

Lenti.OPNi-3 was demonstrated to induce a significant decrease in the phosphorylation levels of MEK and ERK1/2; however, it did not induce a specific down-regulation of endogenous genes such as glyceraldehyde-3-phosphate dehydrogenase and p65 NF-κB. No significant alteration in OPN level and the phosphorylation levels of MEK and ERK1/2 was found in the cells transfected with Lenti.OPNi-3M or in mock controls (Fig. 4A,B). Moreover, it did not significantly influence the phosphorylation of MEK and ERK1/2 in HepG2 cells whose proliferation was not significantly inhibited by Lenti.OPNi-3 (Supplementary Fig. 3B). Similar results were observed in CCL13 cells. These suggest that lentiviral-mediated miRNA against OPN does not have any significant effect on the HCC cells with a constitutively low level of OPN.

thumbnail image

Figure 4. The mechanisms involved in the effects of OPN on proliferation and metastasis of HCC cells. (A,B) The effect of down-regulation of OPN on the protein levels of MAPK pathway in HCCLM3 cells detected via western blotting. The phosphorylations of MEK and ERK1/2 were significantly inhibited in the Lenti.OPNi-3–transfected (P < 0.001) but not Lenti.OPNi-2–transfected HCCLM3 cells; no significant change was found in the p65 NF-κB level. (C) The exogenous recombinant OPN induced activation of the MAPK pathway in Lenti.OPNi-3–transfected HCCLM3 cells detected via western blotting. MEK and ERK were activated in OPN-treated cells, reaching the peak at 15 to 25 minutes. U0126 (a specific MEK inhibitor) could suppress the MEK activation induced by OPN. (D,E) The expression levels of MMP-2 and uPA, rather than MMP-9, were significantly decreased in Lenti.OPNi-2– or Lenti.OPNi-3–transfected HCCLM3 cells. (F,G) In the gelatin zymography, exogenous recombinant OPN markedly increased MMP-2 levels (both the 72-kDa pro and 66-kDa active forms) in the Lenti.OPNi-3–transfected HCCLM3 cells in a dose-dependent pattern (0-10μM). *P < 0.005.

Download figure to PowerPoint

To further investigate whether OPN induces phosphorylation of MEK and ERK1/2 in HCCLM3 cells, the Lenti.OPNi-3–transfected HCC cells were treated with 10μM exogenous recombinant OPN. Some of them were pretreated with specific MEK inhibitor U0126 (20μM), and those treated with OPN alone without pretreatment with U0126 were used as control. The cell lysates were analyzed via western blotting at different time points. MEK and ERK were found to be activated in OPN-treated cells, reaching the peak at 15 to 25 minutes, and MEK was not activated in the cells pretreated with U0126 (Fig. 4C).

OPN Induced Overexpression of MMP-2 and uPA and Activation of MMP-2.

The protein expression levels of MMP-2, MMP-9, and uPA—which have been shown to be associated with OPN in models of breast cancer and melanoma—were also evaluated. Interestingly, the expression levels of MMP-2 and uPA, rather than MMP-9, were decreased in HCCLM3 cells transfected with Lenti.OPNi-2 or Lenti.OPNi-3 (Fig. 4D,E).

To explore whether OPN could induce the production and activation of MMP-2 in Lenti.OPNi-3–transfected HCCLM3 cells, the cells were treated with 10 μM of OPN. The medium was collected to detect the activity of MMP-2 using gelatin zymography. The levels of both the pro- and active forms of MMP-2 in OPN-treated cells were significantly higher compared with those control cells without OPN treatment (Fig. 4F,G).

OPN Induced Translocation of p65 Subunit of NF-κB into the Nucleus of HCC Cells.

To further determine the effect of OPN on p65 subunit of NF-κB, the HCCLM3 cells were treated with 10 μM exogenous recombinant OPN in basal medium at 37°C for 0 to 120 minutes. Immunofluorescence analysis of NF-κB was performed at different time points and revealed that OPN induced translocation of p65 into nuclei of HCC cells in a time-dependent manner. In the cells treated with OPN, the majority of p65 staining resided in the cytoplasm up to 30 minutes. At 45 minutes and 90 minutes after treatment, translocation of p65 into the nucleus was observed, and reaching the peak at 120 minutes (Fig. 5A). Western blot analysis further confirmed the translocation of p65 into the nucleus (Fig. 5B). However, in the controls without OPN treatment, the p65 was majorly localized in the cytoplasm. These results suggest that OPN induces the translocation of p65 from cytoplasm to the nuclei of HCC cells.

thumbnail image

Figure 5. Effects of OPN on cellular localization of p65 unit of NF-κB in HCC cells. (A) Immunofluorescence assay for the localization of p65 protein in HCCLM3 cells. HCCLM3 cells grown on glass slides were treated with 10μM of OPN (for 0-120 minutes). p65 was mainly located in cytoplasm of HCC cells before OPN treatment (a). From 45 minutes (b) and 90 minutes (c) after OPN treatment, translocation of p65 into the nucleus was observed and increased; p65 was completely accumulated in the nucleus of HCC cells after OPN treatment for 120 minutes (d). (B) Western blotting (left) further confirmed the translocation of p65 into the nucleus. In the untreated HCC cells, p65 was detected in the cytoplasm (lane 1) but not in the nucleus (lane 2). In the OPN-treated cells, the cytoplasm p65 level was decreased (lane 3), while the p65 level in nucleus was obviously increased (lane 4). The changes are also represented in bar graphs (right). *P < 0.001. (C) Effect of the NF-κB inhibitor PDTC on expression levels of uPA and MMP-2 in HCC cells induced by OPN. HCCLM3 cells were preincubated with 10 μM of PDTC for 30 minutes (−, control media), followed by incubation with 10 μM of OPN for 90 minutes (for detection of p65 or N-p65) or 24 hours (for detection of uPA and MMP-2). OPN increased expression of MMP-2 and uPA and induced translocation of p65 into the nuclei of HCCLM3 cells. The NF-κB inhibitor PDTC significantly inhibited translocation of p65 and expression of MMP-2 and uPA in HCCLM3 cells induced by OPN. *P < 0.05. #P > 0.05. N-p65, nucleus p65.

Download figure to PowerPoint

To evaluate the association of this translocation with the OPN effects on uPA and MMP-2, HCCLM3 cells were preincubated with pyrrolidine dithiocarbamate (PDTC), a NF-κB inhibitor, for 30 minutes followed by incubation with OPN. OPN could significantly increase the expression of MMP-2 and uPA and induce translocation of p65 into the nuclei of HCCLM3 cells. The effects of OPN on the p65 translocation and expression of MMP-2 and uPA in HCCLM3 cells were obviously inhibited by the NF-κB inhibitor, PDTC (Fig. 5C). These results provide evidence that OPN could regulate the expression of MMP-2 and uPA and induce translocation of p65 into the nuclei of HCC cells.

Discussion

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. References
  7. Supporting Information

OPN is a secreted phosphoglycoprotein that binds the integrin αvβ and CD44 families of receptors. Recently, elevated expression of OPN has been associated with tumor invasion, progression, or metastasis of malignant tumors.8, 19 In addition, OPN has been shown to activate the molecular mechanisms regulating the migration and invasion of tumor cells in the models of breast cancer, colon cancer, and melanoma. In HCC, an elevated plasma level of OPN is regarded as a potential prognostic biomarker, and overexpression of OPN is closely correlated with intrahepatic metastasis, early recurrence, and a worse prognosis.10, 11, 20, 21 However, there are limited data in the literature to understand the mechanism by which OPN regulates the proliferation and metastasis of HCC.

First, we wanted to know after knock-down of OPN, the possible alterations in the proliferative and invasive activities of HCCLM3 cells with a constitutive OPN overexpression and higher metastatic potential. Both the present study and our previous studies have shown that OPN is overexpressed in HCC cell lines with higher metastatic potential. So, in this study, three lentiviral vectors expressing miRNA against OPN (Lenti.OPNi-1, Lenti.OPNi-2, and Lenti.OPNi-3) were constructed to block the expression of OPN in HCC cells. The data showed that the lentiviral-mediated RNAi could efficiently down-regulate the expression of OPN. Consequently, the down-regulation of OPN induced by RNAi led to an obvious inhibition of both in vitro invasion and in vivo lung metastasis of HCC. This is consistent with previous reports10, 20–22 and provides further evidence to support that OPN plays an important role in the metastasis of HCC.

Another interesting finding of this study is the role of OPN down-regulation on tumor growth, which has drawn more attention recently.23–25 In this study, although both Lenti.OPNi-2 and Lenti.OPNi-3 could significantly suppress the invasion and metastasis, only Lenti.OPNi-3 could inhibit in vitro proliferation and in vivo tumor growth of HCC. Since the down-regulation degrees in OPN expression were different in HCCLM3 cells transfected with these two kinds of RNAi (78% by Lenti.OPNi-2, and 95% by Lenti.OPNi-3), these imply that the necessary level of OPN for tumor growth is much lower than that for invasion and metastasis of HCC cells, and a very low level of OPN is enough for tumor growth. To our knowledge, this is the first report that OPN expression is also critical for tumor proliferation and growth of human HCC besides its important role in metastasis of HCC.

Second, to further define the mechanism of these observations, molecular markers in the downstream of OPN-related signaling pathways were investigated. The mitogen-activated protein kinase (MAPK) pathway plays an integral role in coordinating growth and survival signaling and may play an important role in tumor development and progression. Increasing evidence suggests that this pathway is abnormally regulated in HCC and plays a central role in tumorigenesis and maintenance of tumor growth.26–30 In addition, targeted inhibition of this pathway has been regarded as an alternative approach for the treatment of HCC.31, 32 In this study, Lenti.OPNi-3, rather than Lenti.OPNi-2, could significantly inhibit activation of the MAPK pathway, as evidenced by inhibiting the phosphorylation of the downstream ERK1/2. These indicate that OPN might be through regulating the MAPK pathway to affect the proliferation and HCC tumor growth.

To further clarify the role of OPN in the regulation of MAPK signaling pathway in HCC, exogenous recombinant OPN was used to treat the HCCLM3 cells. MEK and ERK1/2 were found to be activated in the OPN-treated HCC cells, and the inhibitor of MEK1/2 (U0126) could inhibit the activation of this pathway triggered by OPN. These data provide further evidence that OPN promotes the proliferation of HCC by activating MAPK pathway. Interestingly, Lenti.OPNi-3 could significantly inhibit activation of the MAPK pathway in HCCLM3 cells, but not in HepG2 and CCL13 cells. This may explain why the proliferation rate of HepG2 and CCL13 cells with a lower expression level of OPN was not affected by Lenti.OPNi-3.

Finally, the downstream molecules of OPN—such as NF-κB translocation, expression MMP-2, uPA, and MMP-9 and pro-MMP-2 activation—were evaluated to explore the possible mechanisms of OPN involved in the invasion and metastasis of HCC. MMP-2 and uPA were found to be significantly inhibited in OPN-silenced HCC cells, and the levels of both pro- and active MMP-2 were increased after treatment with OPN. However, there was no significant change in MMP-9. These data suggest that MMP-2 rather than MMP-9 might be one of the important molecules involved in the effect of OPN on invasion and metastasis of HCC.

Moreover, OPN could induce a nuclear accumulation of NF-κB in a time-dependent manner in HCCLM3 cells. In most cells, NF-κB is present as a latent inactive complex in the cytoplasm, but upon activation, NF-κB rapidly translocates into nuclei and activates gene expression.33 Previous studies have demonstrated that OPN stimulates NF-κB–mediated uPA secretion and MMP-2 activation in breast cancer and melanoma cells.34, 35 Currently, no data are available for the functional role of OPN in the expression regulation of NF-κB, MMP-2, and uPA, which are associated with HCC invasion and metastatic recurrence.36–39 The OPN-induced MMP-2 and NF-κB activation implicates a role of OPN in HCC metastasis.

Taken together, based on both in vitro and in vivo investigations, we show that OPN plays an important role in metastasis as well as HCC tumor growth. Moreover, the necessary level of OPN for tumor growth is much lower than that for HCC metastasis, and a very low level of OPN is sufficient for tumor growth. OPN might regulate tumor growth through activation of the MAPK pathway. Furthermore, inducing the NF-κB (p65) translocation and the production/activation of MMP-2 may be an important mechanism by which OPN is involved in HCC metastasis. These data suggest that OPN could be a hopeful target for the control of metastasis as well as HCC tumor growth.

References

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. References
  7. Supporting Information

Supporting Information

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. References
  7. Supporting Information

Additional Supporting Information may be found in the online version of this article.

FilenameFormatSizeDescription
HEP_22531_SupportingFigure1.tif312KSupp Fig.1 Osteopontin (OPN) expression levels in HCC cell lines with different metastatic potentials analyzed by Western blot. GAPDH was used as an internal control. Amount of OPN is indicated as relative fold. OPN expression level was closely related to the metastatic potentials.
HEP_22531_SupportingFigure2.tif10226KSuppFig.2 The effect of lentiviral-mediated miRNA against OPN on in vitro proliferation and invasion of MHCC97-H cells.
HEP_22531_SupportingFigure3.tif1518KSuppFig.3 The lentiviral-mediated miRNA against OPN did not have any significant effect on the HepG2 cells.
HEP_22531_SupportingFigure4.tif5161KSuppFig.4 Visual and fluorescent images of subcutaneous tumors in nude mice implanted with HCCLM3 cells.
HEP_22531_SupportingFigure5.tif8244KSuppFig. 5 The comparison of lung metastases between the mock controls mice with implantation of HCCLM3 cells in the thoracic area and those with HCCCLM3 cells implanted in the rear abdominal area.
HEP_22531_SupportingFigure6.tif1231KSuppFig.6 The protein levels of IFN-induced gene (P-eIF2α) in the HCCLM3 cells.
HEP_22531_SupportingFigure7.tif3938KSuppFig. 7 The expression levels of OPN, pMEK, pERK1/2, MMP-2, uPA and MMP7hyphen;9 detected in the tumors at the end of the in vivo experiments.

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.