Effects of miR-152 on Cell Growth Inhibition , Motility Suppression and Apoptosis Induction in Hepatocellular Carcinoma Cells

MicroRNAs (miRNAs) are endogenously expressed non-coding RNAs that are 20–24 nucleotides long and have been noted to regulate gene expression, cellular differentiation, development and disease (Papagiannakopoulos et al., 2008). Up to now more than a thousand miRNAs have been described in humans (Tsai et al., 2011; Mollaie et al., 2013). MiRNAs regulate various physiological processes, including the stability or translation efficiency of specific mRNAs. As individual miRNAs are capable of regulating a large number of different mRNAs (encoded by 250–500 target genes), there is a strong likelihood that approximately 20–80% of transcribed human genes are regulated by miRNAs. The efficacy in binding and neutralizing their targets depends on various parameters (e.g. primary sequence of the miRNA and target mRNA, three dimensional structure


Introduction
MicroRNAs (miRNAs) are endogenously expressed non-coding RNAs that are 20-24 nucleotides long and have been noted to regulate gene expression, cellular differentiation, development and disease (Papagiannakopoulos et al., 2008).Up to now more than a thousand miRNAs have been described in humans (Tsai et al., 2011;Mollaie et al., 2013).MiRNAs regulate various physiological processes, including the stability or translation efficiency of specific mRNAs.As individual miRNAs are capable of regulating a large number of different mRNAs (encoded by 250-500 target genes), there is a strong likelihood that approximately 20-80% of transcribed human genes are regulated by miRNAs.The efficacy in binding and neutralizing their targets depends on various parameters (e.g.primary sequence of the miRNA and target mRNA, three dimensional structure of the miRNA, co-factors, and so on) (Hunt et al., 2009;Tsai et al., 2011;Sun et al., 2013).
Hepatocellular carcinoma (HCC) is a common malignant neoplasm in Asian countries.The estimated number of new cases of HCC has risen to 564,300 and 548,600 patients with HCC have died, representing 97.2% of persons with this diagnosis (Kudo, 2011;Yamazaki et al., 2011;Zhang et al., 2013).The development and progression of HCC involves a multi-step, long-term process.Many reports have highlighted the investigation of genes and proteins underlying the development and progression of HCC (Chen et al., 2006;Chen et al., 2008a;Chen et al., 2008b;Hu et al., 2010;Karabork et al., 2010;Yang et al., 2010;Zhang et al., 2013), however, their sensitivity and specificity remain limited.Therefore, the identification of new biomarkers is urgently needed in order to understand the events causing hepatocarcinogenesis, also to relate various phenotypes in clinical features and prognosis.
Previously, we reported that tumor necrosis factor receptor super family 6B (TNFRSF6B, also named as Decoy receptor, DcR3/TR6/M68) is overexpressed in both HCC tissue and serum samples compared to other non-malignant liver diseases.TNFRSF6B plays an important role in cell growth and cellular apoptosis of HCC cells in vitro (Chen et al., 2008b;Chen et al., 2010a;Yang et al., 2010).We had predicted the possible miRNAs, which targets TNFRSF6B using different miRNA target prediction software.miR-152 appeared to be one of the potential miRNAs of TNFRSF6B due to the complementary sequences.miR-152 was found to be over-expressed in neuroblastoma cells than in mature neurons (Liu et al., 2012).Meanwhile, inconsistent underexpression of miR-152 was reported in other cancers (Hiroki et al., 2010;Chen et al., 2010c;Tsuruta et al., 2011;Song et al., 2011;Zhou et al., 2012;Chen et al., 2013d;Kohler et al., 2013;Lichner et al., 2013;Song et al., 2013;Azizi et al., 2014), facilitating its role as a tumor suppressor miRNA in these cancers.To our knowledge, only one study with 20 cases of HCC found that the expression of miR-152 was frequently downregulated in HBV-related HCC tissues in comparison with adjacent noncancerous hepatic tissues (Huang et al., 2010).However, the contribution of miR-152 in the tumorigenesis and progression of HCC has not been fully clarified.The relationship between miR-152 levels and the clinicopathological characteristics has not been reported either.We also hypothesized that TNFRSF6B could be a target gene of miR-152.Thus, in the current study, we investigated the expression of miRNA-152 in HCC and their paired adjacent noncancerous hepatic tissues in 89 cases of formalin-fixed paraffin-embedded (FFPE) surgically resected samples.We also studied the effect of miR-152 on the malignant genotypes of HCC cells in vitro.In addition, we studied the relationship between miR-152 and TNFRSF6B in HCC.

Patients
This retrospective study included 89 cases of HCCs and their corresponding paraneoplastic liver FFPE tissues.The ages of HCC patients ranged from 29 to 82 years old, with an average age of 52 years.Clinicopathological information was obtained from medical records and summarized in Table 1.The corresponding paraneoplastic tissues were taken at least 2 cm from the cancerous node.All cases were initial hepatectomies and randomly chosen from the hepatectomies performed in the First Affiliated Hospital, Guangxi Medical University, P.R.China between March 2010 and December 2011.The study protocol was approved by the local Ethical Committee.Written informed consent to use the samples for research was obtained from the patients and clinicians.All samples were independently reviewed and diagnosed by two pathologists.

Immunohistochemistry
The procedure of immunohistochemistry for TNFRSF6B staining was as previously described (Chen et al., 2008b;Chen et al., 2010a;Yang et al., 2010).The positive signal for TNFRSF6B appeared yellow brown in the cytoplasm of the cells using 3,3'-diaminobenzidine.One or two of the most representative sections from each case were selected and stained with a rabbit polyclonal antibody against TNFRSF6B (TNFRSF6B<H-130>:sc-25464, 1:300 dilution) from Santa Cruz Biotechnology, Inc., Heidelberg, Germany, which is raised against amino acids 171-300 of TNFRSF6B of human origin.One hundred cells from five representative areas from each case were counted.According to immunodetection of stain intensity and number of positive cells, staining results were evaluted by two pathologists, who discussed each case until they reached a consensus.Staining intensity reached the standard of the relative staining intensity of most cells.The degree of staining was subdivided as follows: the staining intensity could range from 0 to 3 (0 = no staining; 1 = yellow or light brown, weak staining; 2 = brown, strong staining; and 3 = dark brown, intense staining), and the positive cells in the liver tissues observed ranged from 0 to 3 in percentage (0, no staining; 1, <30%, often focal or fine granular; 2, 30%-70%, linear or cluster; and 3 >70%, diffuse).Samples were scored by their summation: 0-1 (-); 2-3 (+); 4 (++); 5-6 (+++).Any staining score ≥2 (+) was considered as positive.

miR-152 expression in HCC FFPE tissues and its clinicopathological significance
The relative expression of miR-152 in HCC tissues was remarkably lower than that of their matched adjacent noncancerous hepatic tissues (P<0.001,Table 1, Figure 1A).When studying the relationship between miR-152 expression and other clinicopathological parameters, we found that the expression of miR-152 in the tissues of clinical TNM III and IV stages was significantly reduced compared to that in I and II stages (P=0.013, Figure 1A).miR-152 level was also significantly down-regulated in the groups of bigger tumor size and HBV positive infection than the corresponding groups (P=0.037,P=0.03, Figure 1A).miR-152 had no correlation with other characteristics, such as age, gender, histological differentiation grades, metastasis, cirrhosis, plasma AFP level, portal vein tumor embolus, tumor capsular infiltration or tumor nodes (Table 1).Fifty-five of 89 patients were followed up for 3-68 weeks and time-to-recurrence was collected.Time-to-recurrence for the cases of high miR-152 level (higher than the mean) was 59.29±3.41weeks, longer than that of the low miR-152 group (31.51±1.80).However, this bore no statistical significance.Thus, there was no relationship between the expression of miR-152 and the time-to-recurrence (P=0.127, Figure 1B).

Effect of miR-152 on the malignant phenotypes in HCC cells
Transfection efficiency of miR-152 inhibitor and miR-152 mimic was first confirmed using real time RT-qPCR (data not shown).Western blot also confirmed that miR-152 mimic could downregulate the protein expression of two known targets (Wnt-1 and DNMT1 (Huang et al., 2010;Huang et al., 2014)) of miR-152 in HCC (data not shown), which further supported the success of the transfection system.The effect of miR-152 on cell growth was detected using three independent assays, including MTS tetrazolium assay, fluorimetric resorufin viability assay and Hoechst 33342/propidium iodide (PI) double fluorescent chromatin staining, respectively.MTS tetrazolium assay revealed that cell proliferation increased in HepG2 cells from 72 h post transfection compared to blank and negative controls with miR-152 inhibitor.miR-152 inhibitor exerted weaker influence on HepB3 or SNU449 cells.However, after transfection with the miR-152 mimic, a dramatic reduction in cell proliferation was achieved at 72 and 96 h in all the three cell lines tested.The effect was almost equal for all three cell lines, which led to a 25% inhibition of cell growth at 96 h (Figure 2).The cell growth inhibitory effect showed a time dependent manner.To verify these results, the effect on cell viability was assessed by using a fluorimetric resorufin viability   .2014.15.12.4969 miR-152 Cell Growth Inhibition, Motility Suppression and Apoptosis Induction in Hepatocellular Carcinoma Cells assay (data not shown) and Hoechst 33342/PI double fluorescent chromatin staining (data not shown), which largely mirrored the results from MTS assay.The effect of miR-152 on cell growth suppression also showed a dose dependent manner in all three cell lines tested (data not shown).
To validate whether miR-152 is able to influence apoptosis, the CellTiter-Blue assay was multiplexed with a fluorescent caspase-3/7 assay.The results showed that with the miR-152 inhibitor, caspase-3/7 activity was slightly less than the blank and negative controls, but indicated no significant change in HepG2 and SNU449 cells.However, with the miR-152 mimic, caspase-3/7 activity was markedly enhanced in all three HCC cell lines tested (Figure 3) with a time and dose dependent manner.The time and dose dependent effect on apoptosis was confirmed microscopically by Hoechst 33342 and PI double fluorescent staining (Figure 4).The most potent effect of miR-152 mimic on apoptosis was observed in HepG2 cells, with a 3-fold stronger as compared to the mock control 96 h post transfection.
We further investigated the role of miR-152 in the motility of HCC cell with wound-healing model.Similar as the effect on cell growth, miR-152 inhibitor produced limited impact on the cell motility (data not shown).miR-152 mimic could reduce the cell motility and decelerate the spread speed of HCC cells (Figure 5).

Contribution of miR-152 in relevant cellular signaling
To investigate the contribution of miR-152 in the regulation of cellular signaling, we examined the signaling of ERK and AKT pathways, as well as the proved targets of miR-152 Wnt-1 and DNMT-1 by using western blot, which are related to cell survival, apoptosis and invasion.These pathways were slightly enhanced with miR-152 inhibitor transfection.However, the phospho-ERK1/2 and phospho-AKT were down-regulated notably by miR-152 mimic 96 h post-transfection (data not shown).

Correlation between miR-152 level and TNFRSF6B protein expression in HCC
We compared the complementarity of miR-152  noncancerous hepatic tissues.In the current study, we found the concordant lowered expression of miR-152 in a larger patient size (89 cases) in HCC, which confirmed that miR-152 expression is frequently downregulated in HCC.This, together with the report of Huang, et al (Huang et al., 2010), suggested that miR-152 may have a tumor-suppressive capacity in the hepatocarcinogenesis and progression of HCC.Next, we investigated the relationship between miR-152 level and clinicopathological parameters in HCC.Huang, et al (Huang et al., 2010), reported that miR-152 is significantly down-regulated in HBx transgenic mice and HCC cell lines expressing HBx.In the current study, we found the concordant relationship between miR-152 and HBV infection.Lower expression was observed in HBV positive group than in the negative one.We also found that the downregulation of miR-152 was related to the clinical stages of HCC.In the advanced stages III and IV, the miR-152 expression was markedly lower than that in the early I and II stages (P<0.05).Moreover, miR-152 expression was reduced in the group with larger tumor size, which is represented for tumor growth and deterioration.Thus, the result in current study reveals a relation between miR-152 and the progression of HCC.It may be valuable to examine miR-152 expression for the clinical prediction of progression and prognosis of HCC.
Since miR-152 expression is closely related to the HCC progression, next we investigated the contribution of miR-152a to cell growth, apoptosis and motility in HCC cells.The miR-152 mimic decelerated the cell growth and cell motility in all the cell lines tested (HepG2, HepB3 and SNU449).Additionally, miR-152 mimic enhanced the caspase-3/7 activity and induced apoptosis in HCC cell lines.These results were partially in agreement with Huang, et al (Huang et al., 2010), who also found an inhibition of cell migration and invasion, as well as induced cell apoptosis caused by ectopic miR-152 expression.However, they did not find any significant role for miR-152 in cell proliferation.The contradiction may be due to the products from different company (Gene pharma vs Ambion), transfection reagent (Lipofectamine 2000 vs combiMAGnetofection), concentration of mimic (50nM vs 200nM) and incubating time (24h vs 96h).Another study validated that miR-152 inhibition promoted, while miR-152 mimics inhibited cell proliferation (Huang et al., 2014), in line with the findings of current study.The biological effect of miR-152 could be explained by the downregulation of the pathways as was shown by western blot, especially ERK1/2 and AKT.Together with the findings of previous reports, the results of current study suggest that the enhanced expression of miR-152 by gene transfection could reverse the malignant phenotypes of HCC cells and these data ascertain a tumor-suppressive role and a prospective therapeutic target of miR-152 for HCC patients.
The mechanisms of miR-152 being a tumor suppressor could be related to different target genes.Diverse target genes of miR-152 have been demonstrated in several cancers.miR-152 has been proved to target the DNMT1 in neuroblastoma (Das et al., 2010), endometrial cancer (Tsuruta et al., 2011), ovarian cancer (Xiang et al., 2014), sequence to the 3'-untranslated region of TNFRSF6B with Targetscan (http://www.targetscan.org/),miRTar (http:// mirtar.mbc.nctu.edu.tw/human/) and miRanda (http:// www.microrna.org/microrna/home.do),respectively.Partial complementarity between miR-152 and TNFRSF6B was observed by all these miRNA target prediction methods (data not shown).Furthermore, the cases of stronger TNFRSF6B expression showed lower miR-152 level (P=0.001).A negative relationship was observed between the miR-152 level and the intensity of TNFRSF6B expression (r=-0.347,P=0.002).The possible relationship was finally verified in vitro.TNFRSF6B protein level was indeed downregulated with the transfection of miR-152 mimic into HepG2 cells for 96 h.

Discussion
Aberrant expression of miR-152 has been reported in various classes of malignancies.In neuroblastoma cells, miR-152 was found to be over-expressed compared to mature neurons (Liu et al., 2012).However, underexpression of miR-152 was examined in ovarian cancer (Zhou et al., 2012), bladder cancer (Kohler et al., 2013), prostate cancer (Lichner et al., 2013;Song et al., 2013), endometrial cancer (Hiroki et al., 2010;Tsuruta et al., 2011), pancreatic cancer (Azizi et al., 2014) and gastrointestinal cancers (Chen et al., 2010c;Song et al., 2011).Thus the polarity of the implication of miR-152 in malignancies differs between an oncogenic versus a suppressive role in various tumor types.To determine whether miR-152 was expressed differentially in human primary liver cancer, Huang et al. (2010) assessed miR-152 expression in 20 pairs of human HBV-related HCC tissues and pair-matched normal hepatic tissues by realtime PCR.Among the 20 HBV-related HCC samples analyzed, the miR-152 levels were significantly decreased in 18 HCC samples (90%) in comparison with the adjacent  doi.org/10.7314/APJCP.2014.15.12.4969 miR-152 Cell Growth Inhibition, Motility Suppression and Apoptosis Induction in Hepatocellular Carcinoma Cells malignant cholangiocarcinoma (Braconi et al., 2010) and in HCC (Huang et al., 2010), which was confirmed in the current study.We also proved that Wnt-1 was another target of miR-152 in HCC, which was consistent with Huang et al. (2014).The cholecystokinin B receptor (CCKBR) (Chen et al., 2010c), colony stimulating factor-1 (CSF-1) (Woo et al., 2012) and transforming growth factor-alpha (TGFα) (Zhu et al., 2013) were also found as putative targets of miR-152 in different cancers (Chen et al., 2010c).However, none of these genes were influenced by miR-152 mimic in HepG2 cells in the current study (data not shown).We have demonstrated the oncogenic role of TNFRSF6B in HCC previously (Chen et al., 2008b;Chen et al., 2010a;Yang et al., 2010).As predicted by several in silico methods for target gene prediction, including Targetscan, miRTar and miRanda, TNFRSF6B was identified as one of the high-scoring candidate genes of miR-152 targets.The TNFRSF6B-encoded mRNA contains a 3'-UTR element that is partially complementary to miR-152, indicating that miR-152 would directly target this site.Then we found TNFRSF6B protein expression was inversely correlated with the levels of miR-152 in HCC.More importantly, with in vitro data, we indeed found that TNFRSF6B protein expression was downregulated by miR-152 mimic transfection.
In conclusions, together with previous reports, the current observations support the notion that miRNA-152 is a tumor suppressive miRNA that plays a role in the oncogenesis and progression of human HCC, by targeting Wnt-1, DNMT1 and TNFRSF6B.miR-152 expression in HCC FFPE samples could be a prognostic biomarker for HCC.Further in vivo test to explore the value of miR-152 as a therapeutic tool on HCC is warranted.

Figure 2 .
Figure 2. Time Dependent Effect of miR-152 on Cell Proliferation in HCC Cell Lines.HepG2, HepB3 and SNU449 cells (2.5×10 3 cells per well in 96-well-plate) were cultured for 24 h and then transfected with miR-152 inhibitor, miR-152 mimic and their negative controls (200nM) up to another 96 h.Cell proliferation was assessed per day with MTS assay (CellTiter96 Aqueous One Solution Cell Proliferation Assay).* P<0.05, ** P<0.01, compared to blank and negative controls at the same time point

Figure 4 .
Figure 4. miR-152 Mimic Suppressed Cell Growth and Induced Apoptosis with Hoechst 33342/ Propidium Iodide (PI) Double Fluorescent Chromatin Staining.HepG2 cells (2.5×10 3 cells per well in 96-well-plate) were cultured for 24 h then transfected with miR-152 inhibitor, miR-152 mimic and their negative controls (200nM) up to another 96 h.The effect on apoptosis was assessed and compared to mock and negative controls, ×200

Figure 5 .
Figure 5. miR-152 Mimic Inhibited Cell Motility.HepG2 cells (2.5×10 3 cells per well in 96-well-plate) were cultured for 24 h then a wound was made.The cells were transfected with miR-152 mimic and negative control (200nM) up to another 96 h.Pictures were taken every 48 h post transfection.The effect on cell motility was assessed and compared to mock and negative control, ×100

Table 1 . Correlation Between the Expression of miR-152 and Clinicopathological Parameters in HCC(X±S)
∆ Paired samples t-test was performed