HPV16/18-positive cervical cancer cell lines HeLa, CaSki, SiHa and HPV-negative cervical cancer cells C33A and HT-3 were obtained from Shanghai Institute for Biological Sciences, Chinese Academy of Sciences Institute of Cell Resource Center (Shanghai, China). In the present study, we established the ectopically expressed HPV16 E6, E7 and E6/E7 cell models of HT-3E6, HT-3E7, HT-3E6/E7, C33AE6, C33AE7 and C33AE6/E7 by transfecting HPV16 E6, E7 and E6/E7 oncogenes with lentivirus vectors into the HPV-negative cervical cancer cell line C33A and HT-3 cells. The C33A-vector (C33A-V) and HT-3 vector (HT-3V) cells were established by transfecting C33A and HT-3 cells with lentivirus vectors that did not code for the HPV16 E6, E7, or E6/E7 proteins as controls. Stable transfectants were selected with 10 µg/ml puromycin for 3 weeks. The transfection efficiency was tested by western blotting. Western blotting showed that the transfected cells successfully expressed the E6, E7, or E6/E7 proteins (Fig. 1).

For the demethylation experiments, demethylation was induced with 5-Aza-CdR treatment at a concentration that was able to induce the demethylation of the DNA without killing the cells. CasKi, HT-3 and C33A cells were plated at a density of 8×10⁴ cells/25 cm² flask and treated with various concentrations (5 and 10 µM) of 5-Aza-CdR for 96 h.

bisulphate-modified DNA (2 µl) was used as template to amplify in 15 µl total reaction mixture which contained 2 µl DNA, 1.5 µl 10X PCR buffer, 0.75 µl each of the forward and reverse primers, 1.2 µl dNTPs mixture, 0.15 µl Hot Start Taq polymerase (Takara), and 9.4 µl PCR water. Primers for RASSF1A gene promoter amplification was from the literature (25). The primers for the methylated form were 5′-GAGAGCGCGTTTAGTTTCGTTTTC-3′ and 5′-ACCCGTACTTCGCTAACTTTAAAG-3′, and the primers for the unmethylated form were 5′- GAGAGTGTGTTTGTTTTGTTTTTG-3′ and 5′-CCCATACTTCACTAACTTTAAACC-3′. The PCR involved an initial denaturation at 94°C for 10 min, followed by 35 cycles consisting of 94°C for 30 sec, optimal annealing temperature 56°C for 30 sec, 72°C for 30 sec, with a final extension at 72°C for 10 min. double-distilled water as the blank control. PCR product (5.0 µl) was directly loaded on 2% agarose gels, stained with ethidium bromide, visualized and analysis under UV illumination. Each experiment was repeated three times.

The DNA was extracted by using SDS-proteinase K and purified with phenol:chloroform:isoamyl alcohol. A QIAampDNAFFPE tissue kit (Qiagen GmbH, Hilden, Germany) was used for DNA extraction from the paraffin-embedded tumor tissues. The bisulfite conversion of the DNA was performed using an EpiTect Fast DNA Bisulfite kit (Qiagen, Valencia, CA, USA) according to the instruction of the manufacturer. The methylation status of the promoter/exon 1 region of RASSF1A gene in the cervical cancer cells and cervical specimens was analyzed by BGS. BGS strand-specific primers BGSF (GGTTAAGTGTGTTGTTTTAGTAAT, from −271 to −246) and BGSR (CTACCCCTTAACTACCCCTTCC, from +413 to +434) gene were used to amplify a 704-bp region of RASSF1A gene. The purified PCR products were cloned into the pUC18-T vector, and six clones from each sample were randomly selected and sequenced.

Total RNA was extracted from cervical cancer cell lines and tissue samples using TRIzol (Invitrogen, Waltham, MA, USA). The cDNAs were synthesized from the templates in presence of reverse transcriptase and oligo(dT) 18 primers in accordance with the manufacturer's instructions. Forward primer (RASSF1A-F: GTGGGAGACACCTGACCTTT) and reverse primer (RASSF1A-R: TGAAGCCTGTGTAAGACCG) were design to generate a 118 bp PCR product at the appropriate annealing temperature. Double distilled water was used as a negative control and the housekeeping gene glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was amplified by primers (5′-GATGACCTTGCCCACAGCCT-3′ and 5′-ATCTCTGCCCCCTCTGCTGA-3′) to generate a 303 bp PCR product as the internal control. PCR reaction (20 µl) mixture was pre-denatured at 95°C for 2 min, and 35 cycles of 94°C for 40 sec, 55°C for 30 sec, and 72°C for 60 sec was performed with a final extension at 72°C for 10 min.

The harvested cells were lysed in RIPA lysis buffer supplemented with protease inhibitor cocktail. Each protein lysate (30 µg) was mixed with 5X SDS-PAGE sample loading buffer, and the mixtures were boiled for 5 min. The boiled mixtures (50 µg) were then fractionated on 12% SDS-PAGE gels and subsequently transferred onto PVDF membranes (Merck Millipore, Billerica, MA, USA). The primary antibodies used for the western blot analyses were anti-RASSF1A and anti-β-actin. β-actin was used as a housekeeping protein to normalize the protein loads.

This study was approved by the Institutional Review Board (IRB) of the Affiliated Hospital of Qingdao University. All human cervical tissue samples, including 70 HPV16-positive squamous cell carcinoma, 20 HPV-16-positive adenocarcinoma, 8 HPV-negative cervical cancers, 30 HPV16-positive cervical cancers with PACT (pre-operative adjuvant chemotherapy), 30 HPV16-positive normal cervical tissue, were obtained with written informed consent from the donors who underwent primary surgical treatment for cervical tumors or other benign uterine lesions at the Affiliated Hospital of Qingdao University between 2003 and 2015. All cases were reviewed by at least two pathologists to confirm the primary diagnoses.

The detection of HPV DNA was performed in cervical cancer samples by using MY09/11, GP5⁺/6⁺ and SPF1/2 methods. Both MY09/11 and GP5⁺/6⁺ were performed in a final reaction volume of 20 µl, containing 3 µl of the isolated DNA, 2 µl of the 10X Buffer, 0.8 µl of each deoxynucleoside triphosphate, 0.4 µl of forward and reverse primers, and 1.0 units of Taq DNA polymerase (Genebase Bioscience, Guangzhou, China). PCR conditions of MY09/11 were as follows: preheating for 5 min at 94°C was followed by 40 cycles of 1 30 sec at 94°C, 1 min at 45°C, and 1 min at 72°C and a final extension of 7 min at 72°C; PCR conditions of GP5⁺/6⁺ were as follows: preheating for 10 min at 94°C was followed by 40 cycles of 1 min at 94°C, 2 min at 40°C, and 1.5 min at 72°C and a final extension of 7 min at 72°C. Each PCR experiment was performed with positive and several negative PCR controls. The final PCR products were separated by electrophoresis in 1.8% agarose gel. SPF was performed in a final reaction volume of 40 µl, containing 4 µl of the isolated DNA, 4 µl of the 10X Buffer, 3.2 µl of each deoxynucleoside triphosphate, 0.5 µl of forward and reverse primers, and 2.5 units of Taq DNA polymerase (Genebase Bioscience). PCR conditions were as follows: preheating for 1 min at 94°C was followed by 40 cycles of 1 min at 94°C, 1 min at 45°C, and 1 min at 72°C and a final extension of 5 min at 72°C. Each PCR experiment was performed with positive and several negative PCR controls. The products of PCR were then fractionated on 12% polyacrylamide gel electrophoresis, PAGE gels. As for all methods of HPV-DNA detection and genotyping, see Table I for the detailed primer sequences and size of the product. If results of the three methods were all consistent with each other and negative, the samples were then demonstrated as HPV-negative samples; conversely, if HPV detection of the sample was positive in at least one of the three measures, the samples were defined as positive samples and were further detected by HPV type specific PCR for HPV16 and HPV18 (26,27). Details of PCR are available from the corresponding author.

Statistical tests were performed with SPSS version 17.0 (SPSS, Chicago, IL, USA). The correlation between the RASSF1A mRNA expression and the cell lines was analyzed by the analysis of variance (ANOVA), and multiple comparisons were analyzed by the LSD. The correlations between the clinicopathological characteristics and the methylation status of RASSF1A gene were tested using the χ² test. Differences were considered significant at P<0.05.

RASSF1A promoter methylation status was detected in five cervical cancer cell lines by MSP. The results showed that RASSF1A promoter was hypermethylated in two HPV-negative cell lines (C-33A, HT-3), but not in the three HPV-positive cell lines (HeLa, SiHa and Caski) (Fig. 2A). RASSF1A mRNA and protein expression were detected in unmethylated cell lines HeLa, SiHa and CasKi, whereas no RASSFIA expression was detected in the two HPV-negative cell lines HT-3 and C33A with hypermethylated promoters (Fig. 2A). TA clone and BGS results of RASSF1A in these five cell lines showed densely methylated CpG sites C-33A and HT-3 cell lines, while only scattered methylated CpG sites were detected in HeLa, SiHa and Caski (Fig. 2B). BGS analysis confirmed the MSP results.

To determine whether methylation status of the RASSF1A gene promoter directly mediates its transcriptional expression, the two cell lines HT-3 and C33A with a complete methylation promoter and CasKi with a complete hypomethylation were treated with 5-Aza-dc of various concentrations (5 and 10 µM) for 96 h. A gradual restoration of RASSF1A expression was detected in HT-3 and C33A cells with two different concentrations (5 and 10 µM) of 5-Aza-dc treatment by real-time quantitative PCR (Table II) and western blotting (Fig. 3A), concomitant with the demethylation of its promoter detected by BGS (Fig. 3B). No significant changes in the expression and promoter methylation of RASSF1A were detected in CasKi cell before and after being treated with 5-Aza-dc of different concentrations. As the results in Table II show, after being treated with 5-Aza-dc of various concentrations (5 and 10 µM), RASSF1A mRNA expression was restored to different levels in HT-3 and C33A cell lines. While the expression was not influenced by the demethylating agent 5-aza-dC in CasKi cell line. Thus, in effective concentration range, RASSF1A mRNA expression may be induced by 5-Aza-CdR concentration dependently in HT-3 and C33A cell lines.

BGS results (Fig. 4A) showed that among the HPV16 E6, E7 or E6/E7-transfected cell lines, RASSF1A promoter methylation status was relatively downregulated in the HT-3E6/E7 cells but not in the HT-3E6, HT-3E7, HT-3V, C33A-V, C33AE6, C33AE7 and C33AE6/E7 cells compared with the primary HT-3 and C33A cells. The RT-PCR and western blot results (Fig. 4B) revealed that among the HPV16 E6, E7 or E6/E7-transfected cell lines, RASSF1A re-expressed in HT-3E6/E7 cells but not in the HT-3E6, HT-3E7, HT-3V, C33A-V, C33AE6, C33AE7 and C33AE6/E7 cells, which is consistent with the BGS result (Fig. 4A).

All 90 cases of cervical cancer patients were positive in previous HC2 test. The HPV DNA was detected in at least one of the MY09/11, GP5⁺/6⁺ and SPF1/2 methods (Fig. 5A) in all the tumor tissues. Of the 14 HPV-negative cervical samples previously identified by HC2 clinically, we retested the HPV DNA infection by combining the MY09/11, GP5⁺/6⁺ and SPF1/2 methods. The results showed that only 8/14 (57.14%) was confirmed all negative combining MY09/11, GP5⁺/6⁺ and SPF1/2 methods, 9/14 (64.29%) was negative with MY09/11 method, 8/14 (57.14%) was confirmed negative with GP5⁺/6⁺ and SPF1/2 method, respectively (Fig. 5B). Thus, there was a false negative rate (FNR) of 6.25% by HC2 method, when confirmed HPV detection combining the MY09/11, GP5⁺/6⁺ and SPF1/2 methods.

MSP analysis indicated RASSF1A hypermethylation was found in 16 of 98 (16%) cancer samples and none in 30 normal cervical samples (χ²=4.20; P=0.04). The relations between the methylation status of RASSF1A and the clinicopathological characteristics in the cervical cancer patients were analyzed. The methylation status of RASSF1A promoter was not associated with neoadjuvant chemotherapy, tumor size, smoking or hormone receptors (P>0.05; Table III). Of the 70 SCC, the result showed a statistically reversed relationship between RASSF1A methylation and HPV16 infection (P<0.05). While of the 20 adenocarcinoma (AC), no significant differences was found between RASSF1A methylation and HPV16 infection. Examples of MSP and BGS results are shown in Fig. 6A and B. DNA sequencing analyzed by chromas software confirmed the results (Fig. 6C).