Regulation of placenta growth factor by microRNA-125b in hepatocellular cancer

Abstract

Background & aims: microRNAs (miRNAs) are a class of small noncoding RNAs that can regulate gene expression by translation repression or mRNA degradation. Our aim was to evaluate the role of aberrantly expressed miRNAs in hepatocellular cancer (HCC).

Methods: miRNA expression in HCC tissues and cells was evaluated by qPCR array and Taqman miRNA assay. Cell proliferation, motility, invasion, and the angiogenesis index were quantitated using commercial assays. DNA methylation status, matrix metalloproteinases (MMPs) mRNA expression was quantitated by real-time PCR analysis.

Results: miRNA profiling identified a decrease in miR-125b expression in HCC tumor tissues and cell lines. The expression of miR-125b was significantly increased by the methylation inhibitor 5-aza-2'-deoxycytidine in HCC cells but not in normal controls, suggesting that the expression of miR-125b could be epigenetically modulated. Methylation-specific PCR revealed hypermethylation status of miR-125b in HCC cells compared to non-malignant controls. Cell proliferation, anchorage-independent growth, cell migration, invasion, and angiogenesis were significantly decreased by the introduction of miR-125b precursor in HCC cell lines. Placenta growth factor was identified as a target of miR-125b by bioinformatics analysis and experimentally verified using luciferase reporter constructs. Overexpression of miR-125b in HCC cells decreased PIGF expression, and altered the angiogenesis index. Furthermore, modulation of miR-125b also distorted expression of MMP-2 and -9, the mediators of enzymatic degradation of the extracellular matrix.

Conclusions: Our studies showing epigenetic silencing of miR-125b contributes to an invasive phenotype provide novel mechanistic insights and identify a potential target mechanism that could be manipulated for therapeutic benefit in HCC.

Figure 1. Aberrant miRNA expression in human HCC

(A) miRNA expression patterns in human HCC samples were analyzed using a self-organizing tree algorithm. A dendrogram showing 3 clusters was generated. miRNA expression in HCC tumor tissues is shown on the right axis relative to normal liver controls on the left axis. Cluster 3 comprised of 2 miRNA that were decreased in expression, including miR-125b and miR-122. (B) Relative miRNA expression profile between HCC Tumor vs. Normal Control tissues is shown. The expression of a panel of diverse updated miRNAs was evaluated by qPCR array (SBI). miR-125b and miR-122 are the most down-regulated miRNAs among the 318 miRNAs detected in HCC tumors.

Figure 2. miR-125b is silenced in human HCC

Total RNA was isolated from either normal liver and HCC tumors (Panel A), or from normal human hepatocytes, HepG2 and PLC/PRF-5 (Panel B), as well as normal and liver tumor endothelial cells (Panel C). The expression of a group of selected miRNAs from each cluster (miR-221 from cluster 1, miR-96 from cluster 2 and miR-125b & miR-122 from cluster 3) was assessed using real-time PCR. Panel D: Total RNA was isolated from HCC (t) and normal control (n). Real-time PCR analysis was performed, and the ratio of miR-125b to U6 small RNA expression in HCC samples was determined. The PCR products were verified by 1.8% agarose gel electrophoresis. Panel E: The expression of miR-125b is reduced HCC tissues. Expression of miR-125b in 19 HCC patients was examined by qPCR. The y axis indicates the fold change in the miR-125b level in HCC (T) and matched adjacent non-tumor tissue (N) relative to the median value of adjacent non-tumor tissue. The horizontal line indicates the median and interquartile range (25th to 75th percentiles). Cases are divided into two groups (x axis): HCC (T) and matched adjacent non-tumor tissue (N). Significant reduction of miR-125b in HCC tumors were detected (p=0.0058). Panel F: Positive correlation between miR-125b expression and HCC patients’ survival time. The survival time of 19 HCC patients were recorded through the follow-up in 19 HCC patients after surgery. The mature miR-125b level in HCC tumors was examined by real-time qPCR analysis and normalized to adjacent control liver tissues. Statistical analysis was performed using Pearson's correlation coefficient.

Figure 2. miR-125b is silenced in human HCC

Total RNA was isolated from either normal liver and HCC tumors (Panel A), or from normal human hepatocytes, HepG2 and PLC/PRF-5 (Panel B), as well as normal and liver tumor endothelial cells (Panel C). The expression of a group of selected miRNAs from each cluster (miR-221 from cluster 1, miR-96 from cluster 2 and miR-125b & miR-122 from cluster 3) was assessed using real-time PCR. Panel D: Total RNA was isolated from HCC (t) and normal control (n). Real-time PCR analysis was performed, and the ratio of miR-125b to U6 small RNA expression in HCC samples was determined. The PCR products were verified by 1.8% agarose gel electrophoresis. Panel E: The expression of miR-125b is reduced HCC tissues. Expression of miR-125b in 19 HCC patients was examined by qPCR. The y axis indicates the fold change in the miR-125b level in HCC (T) and matched adjacent non-tumor tissue (N) relative to the median value of adjacent non-tumor tissue. The horizontal line indicates the median and interquartile range (25th to 75th percentiles). Cases are divided into two groups (x axis): HCC (T) and matched adjacent non-tumor tissue (N). Significant reduction of miR-125b in HCC tumors were detected (p=0.0058). Panel F: Positive correlation between miR-125b expression and HCC patients’ survival time. The survival time of 19 HCC patients were recorded through the follow-up in 19 HCC patients after surgery. The mature miR-125b level in HCC tumors was examined by real-time qPCR analysis and normalized to adjacent control liver tissues. Statistical analysis was performed using Pearson's correlation coefficient.

Figure 3. Modulation of miR-125b expression altered cell migration and invasion

Panel A: miR-125b expression was assessed by real-time PCR in normal human hepatocytes transfected with either control or miR-125b precursors, or in HepG2 cells transfected with control or anti–miR-125b inhibitors. The ability of these constructs to modulate miR-125b expression was verified using miR-125b precursor in normal hepatocytes and with anti-miR-125b in HCC cells. Panel B&C: HCC cells were transfected with Pre-miR-125b (■) or with control precursor (□) (Panel B). Meanwhile, human liver sinusoidal endothelial cells were transfected with control and anti-miR-125b inhibitors (Panel C). Cell migration was assessed. Panel D: Cells were transfected with miR-125b or control precursor, and cell invasion was assessed after 72 hours using the QCM 96-well cell Invasion assay kit. * p<0.05 relative to controls.

Figure 4. Overexpression of miR-125b reduces HCC cell growth

Panel A: Cell proliferation was assessed using a viable cell assay and the proliferation index was assessed after 72 hours. Panel B: Cells were plated in agar wells in 96-well plates, and anchorage independent growth assessed fluorometrically after 7 days. * p<0.05 relative to controls. Panel C: Flow cytometric analysis of control and miR-125b overexpressed malignant hepatic cells to demonstrate the basis for the gating of viable, apoptotic and necrotic cells. The lower left quadrant shows the viable cells, the lower right quadrant represents the early apoptotic cells. The upper right quadrant represents nonviable, late apoptotic/necrotic cells, positive for Annexin V and APOTEST-FITC staining. The upper left quadrant shows nonviable necrotic cells/nuclear fragments.

Figure 5. Aberrant expression of PIGF in human HCC tumor sections and cells

Panel A: Tissue microarrays of paraffin-embedded HCC tumors with adjacent controls were stained with PIGF antibody and visualized by 3, 30-diaminobenzidine (Zymed Laboratories Inc.). Left panels, two representative adjacent control staining specimen with semi-quantitative scores negative (-); right panels, two representative HCC tumor specimen with strong-PIGF staining (semi-quantitative score +++). Figures are 100 magnifications. Panel B: Immunocytochemistry for PIGF was performed in normal human hepatocytes (N-Hep), HepG2 cells, LSECs and T-LECs. An increase in PIGF expression is observed in HepG2 and T-LECs when compared to N-Hep and LSECs respectfully.

Figure 6. miR-125b regulates expression of PIGF

(A) Western blot analysis and densitometric analysis of relative expression levels of PIGF and β-actin were performed in normal human hepatocytes and liver sinusoidal endothelial cells, was well as in HepG2 and T-LEC cell lines. The increase in PIGF was observed in both malignant cell lines. * p<0.05 relative to expression in normal controls. (B) Schematic of predicted miR-125b site in the 3’UTR of human PIGF. (C) Luciferase reporter constructs containing the miR-125b recognition sequence from the 3'-UTR of PIGF inserted downstream of the luciferase gene were generated. * p<0.05. (D) Cells were transfected with miR-125b or control precursor. Cell lysates were obtained after 48 hours, and western blots performed for PIGF and β-actin. Conditioned medium was collected, concentrated 20-fold by lyophilization and 10 μg analyzed by zymography to detect MMP-9 activity.