RASSF10 suppresses hepatocellular carcinoma growth by activating P53 signaling and methylation of RASSF10 is a docetaxel resistant marker.

Hepatocellular carcinoma (HCC) is one of the most common malignances and the second leading cause of cancer related death worldwide. RASSF10 is located on chromosome 11p15.2, a region that shows frequent loss of heterozygosity (LOH) in several cancer types. Our previous study found that RASSF10 suppresses colorectal cancer growth by activating P53 signaling. To explore the epigenetic changes and the mechanism of RASSF10 in human HCC, 69 cases of primary HCC, twenty cases of normal liver tissue samples and 17 HCC cell lines were involved in this study. We found that RASSF10 was methylated in 82.6% (57/69) of human primary HCC and methylation of RASSF10 was significantly associated with tumor size (P < 0.05) and TNM stage (P < 0.05). The expression of RASSF10 was regulated by promoter region methylation. Restoration of RASSF10 expression suppressed cell proliferation, induced apoptosis and G2/M phase arrest, as well as sensitized cells to docetaxel and activated P53 signaling in HepG2 and QGY7703 cells. In conclusion, we demonstrated that RASSF10 is frequently methylated in human HCC and its methylation is a potential docetaxel resistant marker. Our data also indicate that RASSF10 suppresses human HCC growth by activating P53 signaling.


INTRODUCTION
Hepatocellular carcinoma (HCC) is the fifth most common cancer in men and the ninth most common cancer in women, and it is the second leading cause of cancer related death worldwide [1]. The mechanisms involved in HCC remain unclear. Accumulations of genetic and epigenetic changes are regarded as important mechanisms for many cancers [2][3][4][5]. Aberrant epigenetic changes have been frequently found in human HCC [6][7][8][9][10][11]. The Ras-association domain family (RASSF) proteins are associated with Ras-like small GTP binding proteins and participate in a range of cellular processes including cell growth, adhesion, migration, differentiation and apoptosis [12,13]. The family of genes comprises 10 members that are subdivided into C-terminal (RASSF1-6) and N-terminal (RASSF7-10) RASSF genes [14]. The C-terminal RASSF proteins harbor Ras association domains and Salvador/RASSF/Hippo (SARAH) domains in the C-terminus, whereas the N-terminal RASSF proteins contain Ras association domains in the N-terminus and lack the SARAH domain [15]. RASSF10 is located on chromosome 11p15.2 [16]. Loss of heterozygosity (LOH) frequently occurs in this region in several types of cancer [17][18][19][20][21]. RASSF10 is frequently methylated in many human cancers [13,14,[22][23][24][25][26]. Our previous study found that RASSF10 suppresses colorectal cancer growth by activating P53 signaling. In this study, we explored the epigenetic changes and the role of RASSF10 in HCC progression.

RASSF10 is frequently methylated in human primary hepatocellular carcinoma
To determine the methylation status of RASSF10 in human primary HCC, 69 cases of primary HCC and 20 cases of normal liver tissue samples were examined by MSP. RASSF10 was methylated in 82.6% (57/69) of primary HCC samples, but no methylation was detected in normal liver tissue samples ( Figure 3A). As shown in Table1, methylation of RASSF10 was significantly associated with tumor size (P < 0.05) and tumor stage (P < 0.05), but no association was found between RASSF10 methylation and age, gender, HBV infection, cirrhosis, lymph node metastasis and cell differentiation. The expression of RASSF10 was evaluated by immunohistochemistry (IHC) in 31 cases of available matched primary HCC and adjacent tissue samples. Staining of RASSF10 was mainly localized in the cytoplasm and its expression was significantly reduced in primary HCC compared to adjacent tissue samples (p<0.05, Figure 3B and C). In 31 cases of available primary hepatic cancer samples, loss or reduced expression of RASSF10 was found in 24 cases. Of these 24 case samples, 22 cases were methylated and 2 cases were unmethylated. Loss or reduced expression of RASSF10 was significantly associated with promoter region hypermethylation (p<0.05, Figure 3D). These results indicate that RASSF10 expression is regulated by promoter region methylation in primary HCC.

Restoration of RASSF10 expression suppresses proliferation of HCC cells
To evaluate the effects of RASSF10 on HCC progression, cell viability and colony formation assays were employed. As shown in Figure 4A, cell viability was determined by the MTT assay. The OD values were 0.498 ± 0.017 vs. 0.432 ± 0.021 (P<0.05) in HepG2 cells and 0.519 ± 0.028 vs. 0.320 ± 0.014 (P<0.05) in QGY703 cells before and after restoration of RASSF10 expression. The cell viability was significantly reduced. The colony numbers were 121 ± 8 vs. 64 ± 4 in HepG2 cells and 362 ± 22 vs. 222 ± 34 in QGY7703 cells before and after re-expression of RASSF10. The colony number was significantly reduced after re-expression of RASSF10 in HCC cells (all P<0.05, Figure 4B). These results demonstrate that RASSF10 suppresses the proliferation of HCC cells.

RASSF10 induces G2/M phase arrest and sensitizes HCC cells to docetaxel
The role of RASSF10 in the cell cycle was analyzed by flow cytometry. As shown in Figure 5A, the distribution of cell phase in RASSF10 unexpressed and re-expressed HepG2 cells was 64.22 ± 3.0% vs. 54.57 ± 0.7% in G0/G1 phase, 22.88 ± 0.3% vs. 24.96 ± 1.1% in S phase and 12.9 ± 2.7% vs. 20.48 ± 1.7% in G2/M phase. In QGY7703 cells, the cell phase distribution was 54.65 ± 0.4% vs. 42.26 ± 1.8% in G0/G1 phase, 31.54 ± 1.3% vs. 35.28 ± 2.1% in S phase and 13.81 ± 1.0% vs. 22.45 ± 0.5% in G2/M phase before and after restoration of RASSF10 expression. The G0/G1 phase was significantly reduced and the G2/M phase was significantly increased before and after re-expression of RASSF10 in HCC cells (all P<0.05). The G1/S checkpoint was not obviously affected by RASSF10, as S phase was not changed significantly (P>0.05). The effect of RASSF10 on the arrest of the G2/M checkpoint was further validated by detecting G2/M phase related proteins in HCC cells. The expression levels of cyclin B1 and cdc-2, important G2/M checkpoint regulators, were dramatically reduced after re-expression of RASSF10 in HepG2 and QGY7703 cells ( Figure 5B). Docetaxel, a microtubule inhibitor, exerts its effects on the G2/M checkpoint. To determine whether RASSF10 is involved in docetaxel sensitivity, we examined the cell viability of RASSF10 unexpressed and re-expressed HepG2 and QGY7703 cells after treatment with docetaxel. The IC 50 of docetaxel was 2.759 ± 0.079 vs. 1.008 ± 0.066 μg/ml (P < 0.05) in HepG2 cells and 4.115 ± 0.172 vs. 2.044 ± 0.144 μg/ml (P < 0.05) in QGY7703 cells before and after re-expression of RASSF10 ( Figure 5C). These results indicate that RASSF10 sensitizes HCC cells to docetaxel.

RASSF10 induces cell apoptosis and suppresses HCC cell growth by activating P53 signaling
To explore the role of RASSF10 in apoptosis, flow cytometry was performed. The percentage of apoptotic   Chi-square test; * P<0.05 was regarded as a significant difference. www.impactjournals.com/Genes&Cancer apoptosis in HCC cells ( Figure 6A). To further explore the mechanisms of RASSF10 in HCC cell apoptosis, survivin, capase-3, cleaved capase-3, P21 and bcl-2 levels were examined by western blot before and after re-expression of RASSF10. The expression levels of cleaved capase-3 and P21 were increased, while surviving, capase-3 and bcl-2 were reduced after re-expression of RASSF10 in HepG2 and QGY7703 cells ( Figure 6B). Our previous study reported that RASSF10 suppresses colorectal cancer growth by activating P53 signaling [13]. To understand the mechanism of RASSF10 in HCC, MDM2, P53, P21, bcl-2 and BAX, the key components of P53 signaling, were examined by western blot before and after re-expression of RASSF10 in HepG2 and QGY7703 cells. The expression levels of P53, P21 and BAX were increased, and MDM2 and bcl-2 were reduced after re-expression of RASSF10 ( Figure 6C). These results indicate that P53 signaling is activated by RASSF10 in HCC cells. To further validate the growth inhibiting effect of RASS10 by activating p53 signaling, siRNA knockdown technique was applied in RASSF10 highly expressed DKO cells. As shown in Figure 6D, the expression of P53, P21 and BAX were reduced, MDM2 and bcl-2 were increased by knocking down RASSF10. The results further suggest that RASSF10 suppresses HCC cell growth by activating p53 signaling.

DISCUSSION
RASSF10 has been reported to be frequently methylated in different malignancies and is regarded as a tumor suppressor [13,[22][23][24]27]. However, the mechanism of RASSF10 in human HCC remains unclear. Our study found that RASSF10 was frequently methylated in human HCC and its expression was regulated by promoter region methylation. Methylation of RASSF10 was associated with tumor size and stage. Restoration of RASSF10 suppressed cell proliferation and induced apoptosis in human HCC. It has been reported that knockdown of RASSF10 increases mitosis in A549 lung cancer cells [26]. Our previous study also found that RASSF10 induces G2/M arrest in human colorectal cancer and esophageal cancer [13,25]. The effects of RASSF10 on G2/M arrest were verified in human HCC in this study. The sensitivity of HepG2 and QGY7703 cells to docetaxel increased after re-expression of RASSF10. These results suggest that methylation of RASSF10 may serve as a docetaxel resistant marker in human HCC. Similar to our previous study in human colorectal cancer, RASSF10 suppressed human HCC proliferation by activating p53

signaling.
In conclusion, RASSF10 is frequently methylated in human HCC and the expression of RASSF10 is regulated by promoter region methylation. Methylation of RASSF10 is associated with tumor size and TNM stage, and it may serve as a docetaxel resistant marker in human HCC. RASSF10 suppresses human HCC growth by activating P53 signaling.

Human tissue samples and cell lines
A total of 69 cases of primary hepatocellular carcinoma were surgically resected. Twenty cases of normal liver tissue samples were collected from noncancerous patients. All samples were stored at -80°C. Tumor stage was determined according to the American Joint Committee on Cancer (AJCC) TNM Staging for liver tumors (7th ed., 2010). Thirty-one cases of cancer tissue paraffin blocks were available with matched adjacent tissue samples. All samples were collected under the guidelines approved by the institutional review board of the Chinese PLA General Hospital. HCC cell lines SNU182, SNU387, HBXF344, SNU475, HepG2, PLC/ PRF/5, Huh7, BEL7402, LM3, QGY7703, SNU449, SMMC7721, Sk-hep1, Huh1, BEL7404, BEL7405 and QSG7701 cel lines were previously established from human primary HCC. DKO is a DNMT1 and DNMT3b double knockout cell stemed from HCT116 cell, a colorectal cancer cell line. All cells were maintained in 90% RPMI 1640 (Invitrogen, Carlsbad, CA) supplemented with 10% fetal bovine serum.

DNA extraction, methylation-specific PCR (MSP) and bisulfite sequencing (BSSQ)
DNA was prepared by the proteinase K method.
Bisulfite treatment was carried out as previously described [28]. Primers for MSP and BSSQ were designed around the transcription start site.

Immunohistochemistry
Immunohistochemistry (IHC) was performed in primary HCC and paired adjacent tissue samples. The dilution of RASSF10 antibody (Lifespan Bioscience, WA, USA) was 1:25. Rabbit polyclonal antibody against RASSF10 was used. IHC was performed and evaluated as described previously [9]. The staining intensity and extent of the staining area were graded according to the German semi-quantitative scoring system as described previously [29,30].

Expressional vector construction and transfection
The CDS region of RASSF10 was amplified by RT-PCR and cloned into the pcDNA3.1 expression vector. Primer sequences are as follows: 5'-GATCGAATTC GCCACCATGGATCCTTCGGAAAAGAAGATATC-3' (F) and 5' -GATCTCTAGACTACACAAGGGATTCGCACATG-3' (R). RASSF10 cDNA was then subcloned into the pLenti6-GFP lentivirus expression vector. RASSF10 expressing lentiviral or empty vectors were packaged using the ViraPower TM lentiviral expression system (Invitrogen, San Diego, CA, USA). Lentivirus was added to the supernatant of HepG2 and QGY7703 cell culture medium, and RASSF10 stably expressed cells were selected by blasticidin (2 μg/ml, Invitrogen).

Colony formation assay
Cells were seeded into 6-well culture plates at a density of 800 cells per well in triplicate and cultured for 2 weeks. Cells were then fixed with 75% ethanol for 30 minutes, stained with 0.2% crystal violet (Beyotime, Nanjing, China) for 20 minutes and counted.

Cell viability assay
Cells were seeded into 96-well plates at 2×10 3 cells/ well, and the cell viability was measured by the MTT assay at 0, 24, 48 and 72h (KeyGENBiotech, Nanjing, China). Absorbance was measured on a microplate reader (Thermo Multiskan MK3, MA, USA) at a wavelength of 490 nm.
The sensitivity of HCC cells to docetaxel was evaluated by detecting the IC 50 of RASSF10 unexpressed and re-expressed gastric cancer cells treated with docetaxel at concentrations of 0, 0.5, 1, 5, 10, 20, 50 and 100 µg/ ml for 48 hours. IC 50 was defined as the concentration that was required for 50% inhibition of cell growth. The

Flow cytometry analysis
For cell cycle analysis, cells were fixed with 70% ethanol and stained with 50 mg/ml propidium iodide (KeyGEN Biotech, Jiangsu, China). The cells were then sorted by a FACS Caliber (BD Biosciences, San Jose, CA) and analyzed by the Modfit software (Verity Software House, ME, USA).
For apoptosis analysis, the Annexin V-FITC/PI Apoptosis Detection Kit (KeyGen Biotechnology, China) was used according to manufacturer's instructions. Each sample was analyzed by flow cytometry with a FACScan Flow Cytometer (Becton-Dickinson Biosciences, Mansfield, MA).

Western blot
Western blot was performed as described previously [28]. Antibodies were diluted according to