This article is mentioned in:

Introduction

According to current research, 15-20% of all cases of cancer can be attributed to infectious agents, including Helicobacter pylori and human papillomavirus (HPV), followed by hepatitis B virus (HBV), hepatitis C virus (HCV) and the Epstein‑Barr virus (EBV) (1‑3). In HPV, specifically types 16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58 and 59 that are commonly referred to as high-risk (HR), are not only associated with cervical cancer (CC), the third most prevalent type of cancer worldwide in women in 2017, but also with other tumor types, including anal, penile, vulvar, vaginal and head and neck cancer (1,4).

Persistent infections with HR-HPVs are necessary, but not sufficient to cause cancer, indicating the existence of multistep actions in viral carcinogenesis that contribute to the characteristic hallmarks underlying the phenotype of tumors (3,5). Thus, it has been of great interest to study mechanisms by which persistent infections with these viruses contribute to cancer development. HPV has been demonstrated to induce a series of mechanisms that contribute to the evasion of the immune response and apoptosis‑activated cell death, and finally the transformation, proliferation and cellular immortalization of the host cell (6,7).

HPV life cycle disruption due to integration of the viral genome into the cellular genome (8,9) and activation of the cell methylation machinery have been demonstrated to be involved in the carcinogenic process of CC among other factors (10-12). Specifically, the HR‑HPV E7 oncoprotein has been reported to serve a crucial role in oncogenic transformation due to its ability to form complexes with members of the retinoblastoma protein (pRB) family and destabilize them (13,14), in addition to the ability to interact with other proteins, including histone deacetylase 1 (HDAC1) (15) and DNA methyltransferase 1 (Dnmt1) (16).

E7 binds to HDAC1 via its zinc finger‑like motif through an intermediary protein Mi2β, which is a component of the nucleosome remodeling and histone deacetylation (NURD) complex that has the ability to modify chromatin structure trough both deacetylation of histones and ATP-dependent nucleosome repositioning (17,18). The formation of this complex is necessary for the maintenance of viral episomes, controlling cell proliferation and extending cell life (15,17). Consistent with this, a chromatin immunoprecipitation assay in Ca Ski cells demonstrated that E7 and HDACs are associated with the major histocompatibility complex (MHC) class I promoter and histone deacetylation (19), as well as chromatin repression and the downregulation of MHC class I genes and MHC class I heavy chain, and repression of the genes encoding the transporter associated with antigen processing subunit 1 (TAP1) and low molecular weight protein 2 (LMP2) (20). Furthermore, an association with Dnmt1 is directed and mediated by the conserved region 3 (CR3) zinc‑finger region of E7, which is known to contribute to E7 transformation functions and stimulate the methyltransferase activity of Dnmt1, which may lead to aberrant genome methylation and cellular transformation as a consequence of the silencing of tumor-suppressor genes (16). Another study has demonstrated that in samples of normal cervix and cervical cancer, HPV types 16 and 18 activate the cell methylation machinery to methylate viral DNA, as well as the promoter regions of cellular genes, including cyclin A1, Rubicon-like autophagy enhancer, retinoic acid receptor β2, cadherin 1 (CDH1), death-associated protein kinase 1, human telomerase reverse transcriptase 1 (hTERT1), hTERT2, hypermethylated in cancer 1 and Twist Family BHLH Transcription Factor 1 (21). Therefore, previous evidence suggests that HR-HPV E7 serves an important role in the activation of the cellular methylation machinery, which regulates the transcription of viral and cellular genes either during their productive infection during its life cycle or during the carcinogenic process.

The mechanisms by which E7 is involved in the regulation of gene expression at the chromatin level are not well understood. It has been observed that a common characteristic of several cancer types associated with viruses is the decreased expression of CDH1, which encodes E-cadherin, through epigenetic mechanisms (22-24). In the case of cancer types associated with HPV infections, it has been demonstrated that HPV16 E7 suppresses the transcription of CDH1, which reduces protein expression of E-cadherin (25,26). In addition, HPV16 E7 has been reported to increase the amount and activity of Dnmt1 in NIKS cells, which are derived from foreskin keratinocytes transfected with HPV16 E7 or NIKS bearing episomal HPV16 DNA (26); however, NIKS cells not infected with HPV16 and NIKS bearing episomal HPV16 DNA did not exhibit any differences in the CpG methylation status of the CDH1 promoter regions, as all CpG sites were unmethylated (26). It is evident that HPV activates the methylation machinery via E7/Dnmt1; however, it is not clear how HPV induces CDH1 repression by epigenetic mechanisms.

Since HR-HPVs E7 has been demonstrated to interact with Dnmt1 and HDAC1, the aim of the current study was to determine the methylation pattern of the CDH1 promoter region in HeLa, SiHa and Ca Ski cell lines positive for HPV16 and HPV18. Additionally, associations with transcription factors Snai1 and Snai2, which are negative regulators of CDH1 expression and inducers of the epithelial-mesenchymal transition (EMT) process (27), were evaluated. HeLa, SiHa and Ca Ski cell lines were selected for the present study, as they are representative of the most frequent cancer types of the uterine cervix with positive HR-HPV infection with different viral loads and epithelial origins, including cervical adenocarcinoma, cervical squamous cell carcinoma and cervical epidermoid carcinoma (28-34).

Materials and methods

Cell lines

As reported by the American type culture collection, HeLa cells are derived from a female African-American patient with uterine cervical adenocar-cinoma and are reported to contain 10-50 integrated copies of HPV18 per cell (30-36). SiHa cells are derived from a female Asian patient with grade II cervical squamous cell carcinoma and are reported to contain 1-2 integrated copies of HPV16 per cell (29,30,33,36). Ca Ski cells are derived from a female Caucasian patient with cervical epidermoid carcinoma and are reported to contain 500-600 integrated copies of HPV16 per cell (28,30,36). HaCaT is a non-tumorigenic immortalized human epidermal cell line derived from skin keratinocytes. All cell lines were authenticated through STR DNA profiling (ID no. DP0297) by the University of Colorado DNA Sequencing & Analysis Core. HeLa, SiHa and HaCaT cell lines (ATCC) were cultured in DMEM (cat. no. 12800-058; Gibco; Thermo Fisher Scientific, Inc.); Ca Ski cells were cultured in RPMI medium (cat. no. 31800‑014; Gibco; Thermo Fisher Scientific, Inc.). All cell lines were supplemented with 10% fetal bovine serum (FBS; cat. no. 16000‑044; Gibco; Thermo Fisher Scientific, Inc.) and 1X penicillin-streptomycin (cat. no. 15140122; Gibco; Thermo Fisher Scientific, Inc.) and were incubated in a humidified chamber at 37°C with 5% CO2.

Untransfected MCF-7 cells and MCF-7 cell stable clones transfected with a pcDNA 3.1 expression vector (Invitrogen; Thermo Fisher Scientific, Inc) with the bicistronic E6/E7 region from HPV18 (MCF-7 pE6/E7) were kindly provided by Dr Erick de la Cruz Hernández (Juarez Autonomous University of Tabasco, Villahermosa, Mexico) and were cultured with 800 µg/ml Geneticin in DMEM/F12 for 3 weeks as previously described (37-39). Untransfected C33-A cells and C33-A cell stable clones transfected with a pcDNAE7 plasmid (C33-A pE7/HPV16) were kindly provided by Dr Patricio Gariglio (CINVESTAV-IPN, Mexico City, Mexico) and were cultured with 800 µg/ml Geneticin in DMEM for 2 weeks as previously described (40).

Treatments with 5‑aza‑2'‑deoxycytidine (5‑AzadC) and trichostatin A (TSA)

5-AzadC causes DNA demethylation or hemi-demethylation that results in gene activation by inhibiting Dnmt activity (41). TSA has been used as a histone deacetylase inhibitor, which causes histone hyperacetylation that leads to chromatin relaxation and modulation of gene expression (42). 5-AzadC and TSA were obtained from Sigma-Aldrich; Merck KGaA (cat. nos. A3656 and T8552, respectively) and resuspended in DMSO (Sigma-Aldrich; Merck KGaA; cat. no. D8418) to obtain working stock solutions (10.0 mM 5-AzadC and 1.0 mM TSA). The 5-AzadC and TSA stock solutions were aliquoted, protected from light and stored at ‑80°C for later use. A total of 4.5×105 HeLa and 5×105 SiHa cells were seeded in p60 boxes in triplicate and were treated with 5 and 10 µM 5-AzadC and 200 and 500 nM TSA. Untreated cells or cells treated with DMSO were used as controls. The total volume of culture medium was 3 ml, which was supplemented with 10% FBS and did not contain any antibiotics. Assays were performed protected from light and the cells were incubated for 48 h at 37°C with 5% CO2. The culture medium was replaced after 24 h due to the half-life of 5-AzadC and TSA.

Transfection with short interfering RNA (siRNA)

siRNAs targeting HPV16 and HPV18 E7 were designed as previously described (43,44) and obtained from Ambion (Thermo Fisher Scientific, Inc.; cat. nos. s237642 and s237640). Silencer® Select GAPDH siRNA (Hs, Mm, Rn) (cat. no. 4390849; Ambion; Thermo Fisher Scientific, Inc.) was used as a positive transfection control to select the transfection agent (Lipofectamine® 2000 or siPORT™ NeoFX™) and to optimize gene silencing without affecting cell viability according to the manufacturer's protocol and as previously described (45-50). siRNAs were resuspended in UltraPure™ DNase/RNasefree distilled water (cat. no. 10977‑015; Thermo Fisher Scientific, Inc.) to obtain a working stock of 10 µM. siRNAs were then aliquoted and stored at ‑80°C for later use.

Transfection with siRNA was performed in triplicate using the siPORT™ NeoFX™ kit (cat. no. AM4511; Ambion; Thermo Fisher Scientific, Inc.) according to the manufacturer's protocol. A total of 7.5×104 HeLa and 8.0×104 SiHa cells/well were seeded in a 12-well plate. HeLa cells were transfected with 30 nM siRNA against HPV18 E7, whereas SiHa cells were transfected with 30 nM siRNA against HPV16 E7. In addition, 30 nM siRNA targeting GAPDH was transfected as a positive control in both cell lines. Cells subjected to treatment with siPORT™ NeoFX™ Transfection Agent with Opti-MEM® I (cat. no. 31985070; Thermo Fisher Scientific, Inc.) were used as a reference control to obtain the relative values, and untransfected cells were used as a reference control for statistical comparisons. Culture medium supplemented with 12% FBS and no antibiotics was added to make up a final transfection volume of 1.2 ml. Cells were incubated for 48 h at 37°C with 5% CO2 and subsequently the extraction of nucleic acids and proteins was performed. Transfection efficiency was determined by measuring the expression of E7 and GAPDH at the mRNA level.

Bisulfite conversion and DNA methylation analysis

Total genomic DNA was isolated from the treated and transfected cell lines using the Wizard® Genomic DNA Purification kit (cat. no. A1120; Promega Corporation) and 1.5 µg genomic DNA was treated with bisulfite according to the manufacturer's protocol of the EZ DNA Methylation-Gold™ kit (cat. no. D5006; Zymo Research Corp.). Methylation of CpG sites at the CDH1 promoter region was analyzed by the bisulfite sequencing PCR (BSP) protocol (Fig. 1A) (10-12,51,52) or using oligonucleotides for MSP-protocol provided by Dr Alfonso Dueñas-Gonzalez (INCan-UNAM, Mexico City, Mexico); the PCR conditions for MSP-protocol were the same as those used for the BSP protocol. PCR was performed with a total volume of 25 µl, containing 1X PCR Gold Buffer, 1 mM dNTP, 2 mM MgCl2, 10 pMol forward CDH1-BSP (5'-TTT TAG TAA TTT TAG GTT AGA GGG TTA T-3') and reverse CDH1-BSP (5'-AAA CTC ACA AAT ACT TTA CAA TTC C-3') oligonucleotides, 1 U AmpliTaq Gold® DNA Polymerase (cat. no. 4338856; Applied Biosystems; Thermo Fisher Scientific, Inc.) and 300 ng bisulfite‑treated DNA. The thermocycling conditions were as follows: 95°C for 7 min, followed by 35 cycles of 95°C for 35 sec, 57°C for 35 sec and 72°C for 60 sec, and a final extension at 72°C for 7 min. PCR products were treated using ExoI (cat. no. EN0581; Thermo Fisher Scientific, Inc.) and SAP (cat. no. EF0651; Thermo Fisher Scientific, Inc.) enzymes. The treated PCR products were sequenced using the BigDye® v3.1 Cycle Sequencing kit (cat. no. 4337455; Applied Biosystems; Thermo Fisher Scientific, Inc.) according to the manufacturer's protocol in an ABI PRISM™ 3100-Avant Genetic Analyzer (Applied Biosystems; Thermo Fisher Scientific, Inc.). The BSP oligonucleotides were designed by MethPrimer v2.0 software (The Li Lab; PUMCH; Chinese Academy of Medical Sciences) using GenBank sequence DQ335132.1 for the CDH1 gene (53). The sequencing data obtained from BSP were analyzed using Chromas v2.6.4 software (Technelysium Pty., Ltd.) and Lasergene v7 package (DNASTAR, Inc.).

Figure 1 Methylation of the CDH1 gene promoter region. (A) The location of different elements; the E-box and GC box motifs that regulate transcription of CDH1. Red vertical lines indicate each CpG site contained in the promoter region. (B) A lollipop methylation diagram of bisulfite sequencing of the CDH1 promoter in the HeLa, SiHa, Ca Ski and HaCaT cell lines. Black indicates a methylated CpG site; white indicates an unmethylated CpG site. (C) Quantification of the methylation frequency of the CDH1 promoter in the HeLa, SiHa and Ca Ski cell lines. (D) Quantitative PCR was performed to measure the expression levels of CDH1, SNAI1 and SNAI2 genes in the indicated cell lines. *P<0.05 and **P<0.01 vs. HaCaT. (E) Protein expression levels of E-cadherin, Snai1 and Snai2 were assessed by western blot analysis in HPV-positive cell lines. GAPDH was used as a loading control. CDH1, cadherin 1; Dmnt1, DNA methyltrans-ferase 1; AP‑2, activating protein 2; ZEB1/2, zinc finger E‑box‑binding homeobox 1/2.

RNA extraction and reverse transcription‑quantitative (RT‑q) PCR

Total RNA was extracted from the treated and transfected cells using TRIzol® reagent (cat. no. 15596026; Invitrogen; Thermo Fisher Scientific, Inc.). The RNA was treated with DNase I (cat. no. EN0521; Invitrogen; Thermo Fisher Scientific, Inc.), and purified with Direct‑zol™ RNA MicroPrep (cat. no. R2060; Zymo Research Corp.) according to the manufacturer's protocol. Complementary DNA (cDNA) was obtained from the purified RNA using the SuperScript™ IV First-Strand Synthesis system (cat. no. 18091050; Invitrogen; Thermo Fisher Scientific, Inc.) according to the manufacturer's protocol. The RNA‑primer mix was incubated at 65°C for 5 min and 4°C for 1 min, and the RT reaction mix was incubated at 55°C for 15 min and 80°C for 10 min to inactivate the reaction and placed on ice for subsequent use. Subsequently, 60 ng cDNA was subjected to qPCR to determine the expression levels of the genes of interest using the primers listed in Table I. qPCR conditions were as follows: 95°C for 10 min, followed by 40 cycles of 95°C for 20 sec, 60°C for 30 sec and 72°C for 35 sec, and a final extension at 72°C for 7 min. This was followed by a melting curve analysis at 65‑95°C. All qPCR assays were analyzed using Rotor-Gene Q Series v2.1.0 software (Qiagen, Inc.).

Table I Primer sequences used for the BSP and qPCR assays.

Table I

Primer sequences used for the BSP and qPCR assays.

Gene	Assay Primer sequences (5'→3')
CDH1	BSP F: TTTTAGTAATTTTAGGTTAGAGGGTTAT
	R: AAACTCACAAATACTTTACAATTCC
CDH1	qPCR F: GTCAGTTCAGACTCCAGCCC
	R: AAATTCACTCTGCCCAGGACG
SNAI1	qPCR F: ACCACTATGCCGCGCTCTT
	R: GGTCGTAGGGCTGCTGGAA
SNAI2	qPCR F: GACCCTGGTTGCTTCAAGGA
	R: TGTTGCAGTGAGGGCAAGAA
E7 HPV16	qPCR F: CAGCTCAGAGGAGGAGGATG
	R: TGCCCATTAACAGGTCTTCC
E7 HPV18	qPCR F: TGAAATTCCGGTTGACCTTC
	R: CACGGACACACAAAGGACAG
GAPDH	qPCR F: AAGGTCGGAGTCAACGGATTTG
	R: CCATGGGTGGAATCATATTGGAA
HPRT	qPCR F: GGACTAATTATGGACAGGACTG
	R: GCTCTTCAGTCTGATAAAATCTAC

[i] BSP, bisulfite sequencing PCR; qPCR, quantitative PCR.

To obtain expression levels of CDH1, SNAI1, SNAI2, HPV16 E7 and HPV18 E7 in the HeLa, SiHa, Ca Ski and HaCaT cells, a commercial sample of RNA extracted from normal cervix negative for HPV (Human Cervix Total RNA; cat. no. AM6992; Ambion; Thermo Fisher Scientific, Inc.) was used as a reference. The ΔCq values for each gene were normalized to the reference gene GAPDH using the 2−ΔΔCq method (54). The commercial sample of normal cervix negative for HPV was set as 1, and the results are not presented. For statistical analysis, the HaCaT cell line was used for comparison. For experiments involving the treatment of HeLa and SiHa cells with 5-AzadC and TSA, cells treated with DMSO were used as a reference, but data were not included in the graphs. For experiments involving the transfection of HeLa and SiHa cells with siRNAs, the cells treated with siPORT™ NeoFX™ Transfection Agent with Opti-MEM® I were used as a reference, but results were not included in the graphs. ΔCq values of CDH1, SNAI1, SNAI2, HPV16 E7 and HPV18 E7 were normalized using the 2−ΔΔCq method (39) with GAPDH as reference for 5-AzadC and TSA treatments, and HPRT as reference for experiments involving siRNA transfections. For statistical analysis, untreated (Unt) HeLa and SiHa cells were used for comparisons.

Protein extraction and western blot analysis

Proteins were obtained using a lysis buffer containing 5 mM EDTA, 150 mM NaCl, 5 mM Tris-HCl pH 9.0, 1% Nonidet-P40 and 1.2 mg/ml cOmplete™ protease inhibitor cocktail (Roche Applied Science). Protein extracts were forced through a 22-gauge needle 10 times and centrifuged for 10 min at 17,000 × g at 4°C. Protein concentration was determined using the Bradford method. Subsequently, 30 mg protein was loaded and separated on 12% SDS-PAGE gels followed by transfer to nitrocellulose membranes. Membranes were blocked with 5% nonfat dry milk in 1X TBS with 0.1% Tween‑20 (TBST) at 4°C for 1 h with gentle agitation and incubated overnight at 4°C with antibodies against E-cadherin (cat. no. sc-8426; 1:1,000), GAPDH (cat. no. sc-48167; 1:1,000), β-actin (cat. no. sc-1616; 1:1,000; all from Santa Cruz Biotechnology, Inc.), Snai1 (cat. no. L70G2; 1:1,000; Cell Signaling Technology, Inc.) and Snai2 (cat. no. C19G7; 1:1,000; Cell Signaling Technology, Inc.) diluted in TBST with 5% BSA (cat. no. 9998; Cell Signaling Technology, Inc.). Subsequently, membranes were incubated with secondary antibodies for 2 h at room temperature, including goat anti-mouse IgG-horseradish peroxidase (HRP; cat. no. sc-2005; 1;10,000), donkey anti-rabbit IgG-HRP (cat. no. sc-2313; 10,000) and donkey anti-goat IgG-HRP (cat. no. sc-2020; 1:5,000; all from Santa Cruz Biotechnology, Inc.) diluted in TBST with 5% non-fat dry milk. Immobilon Western Chemiluminescent HRP substrate (EMD Millipore) was used for protein detection, and images were acquired using C-DiGit Blot scanner equipment (Li-Cor Biosciences) and processed in the Image Studio™ Lite version 5.2 software (Li-Cor Biosciences).

Nitrocellulose membranes were incubated with two primary antibodies in the following manner: i) Incubation was performed as described against a primary antibody, including E-cadherin, Snai1 or Snai2 with their respective secondary antibody; ii) images were acquired; iii) membranes were washed 3×5 min with TBST at room temperature; iv) membranes were re-probed with a second primary antibody, including GAPDH or β-actin with their respective secondary antibody; and v) images were acquired.

Statistical analysis

Statistical analysis was performed using GraphPad Prism v4.0 (GraphPad Software, Inc.). One-way ANOVA with Turkey's post hoc test were used to evaluate significant differences in gene expression and methylation levels, and results were presented as the mean ± SD. P<0.05 was considered to indicate a statistically significant difference.

Results

Different methylation patterns in the CDH1 promoter region are present in HPV16‑ and HPV18‑positive cancer cell lines

A common site‑specific methylation pattern in certain CpG islands (−160, −150, −131 and −122) of the CDH1 promoter region was detected in HeLa and SiHa cells, whereas other CpG islands (−45, −136, −105, −103, −83, −57, −52, −45, −36, −13, +6 and +9) were identified to be methylated in HeLa only (Fig. 1B). Of note, Ca Ski cells did not exhibit methylation of any of the 17 CpG sites of the CDH1 promoter that were analyzed (Fig. 1B). Quantification of the methylation levels in the CDH1 promoter region indicated that HeLa presented a methylation frequency of 88.24%, SiHa cells exhibited a methylation frequency of 17.65% and Ca Ski cells demonstrated no methylation (frequency, 0%; Fig. 1C).

Based on previous studies (25-28), initial experiments were conducted using the C33-A, C33-A transfected with pE7/HPV16 (C33-A pE7/HPV16), MCF-7 and MCF-7 transfected with pE6/E7 from HPV18 (MCF-7 pE6/E7) cell lines as HPV-negative cancer models. Validation of C33-A and MCF-7 cell stable clone selection with pE7/HPV16 and pE6/E7, respectively, was performed by evaluating E7 mRNA expression by RT-PCR (Fig. S1B). However, C33-A vs. C33-A pE7/HPV16 and MCF-7 vs. MCF-7 pE6/E7 did not exhibit any differences in the methylation of CDH1 promoter regions (Fig. S1A) or in the expression of CDH1 at mRNA and protein levels (Fig. S1B and C). As the MCF-7 cell line is an adenocarcinoma that derives from the mammary gland and exhibits an epithelial phenotype with a high expression level of CDH1, similar to that observed in Ca Ski cells, and since the C33-A cell line originally does no express CDH1, C33-A and MCF7 cells were eliminated from the study; neither the effect of oncoprotein E7 on the suppression of CDH1 expression, nor the methylation patterns in the CDH1 promoter could be evaluated in these cells.

Methylation levels in the CDH1 promoter are associated with the mRNA and protein expression levels of CDH1

Analysis of CDH1 expression in the different cell lines demonstrated a significant decrease of the CDH1 mRNA level in the HeLa (P<0.001) and SiHa (P<0.01) cell lines compared with the HaCaT cell line (Fig. 1D). A similar decrease was observed at the protein level (Fig. 1E). By contrast, the Ca Ski cell line exhibited high CDH1 expression, similar to the control HaCaT cell line, at the mRNA and protein level (Fig. 1D and E).

CDH1 expression level is associated with the expression level of SNAI1

The present study measured the expression of SNAI1 and SNAI2 to test the hypothesis that the expression of CDH1 is regulated by the transcription factors Snai1 and Snai2, which mediate EMT by negatively regulating the expression of CDH1 (27,55-57). The results revealed that the mRNA expression level of SNAI1 is significantly increased in HeLa (P<0.0001) and SiHa cells (P<0.001), but significantly reduced in Ca Ski cells (P<0.01) compared with HaCaT cells (Fig. 1D). This result was also reflected at the protein level (Fig. 1E). No significant differences were observed in the expression of SNAI2 at the mRNA (Fig. 1D) and protein (Fig. 1E) level among all cell lines. Since the Ca Ski cell line exhibits a non-mesenchymal phenotype, a high expression level of E-cadherin and low expression levels of Snai1 and Snai2, this cell line was excluded from further analysis.

Treatment with 5‑AzadC and TSA affect methylation and re‑expression of CDH1

HeLa and SiHa cells were treated with different concentrations of 5-AzadC and TSA, followed by analysis of the expression levels of CDH1, SNAI1, and SNAI2. HeLa and SiHa cells treated with the vehicle (DMSO) were used for normalizing the expression values, as well for performing comparisons in statistical analysis.

The results demonstrated that in the HeLa cell line, the methylation pattern was maintained at CpG sites -103, -105, and -160 of the CDH1 promoter region following treatment with 5-AzadC and TSA. By contrast, in the SiHa cell line, this region was completely demethylated following treatment with 5 or 10 µM 5-AzadC or 500 nM TSA, and only CpG site -52 was methylated following treatment with 200 nM TSA (Fig. 2A and C). Compared with untreated HeLa and SiHa cells, a decrease in methylation of 76.48% was observed following 5 or 10 µM 5-AzadC treatment, whereas following treatment with 200 or 500 nM TSA, a decrease in methylation of 68.63% was observed. By contrast, in SiHa cells, a decrease of 21.57% was observed following treatment with 5 or 10 µM 5-AzadC or 500 nM TSA, whereas following treatment with 200 nM TSA, a 17.65% decrease in methylation was observed. Therefore, it was demonstrated in the two cell lines that treatment with 5‑AzadC or TSA significantly diminished the level of CDH1 methylation (P<0.001; Fig. 2B and D). However, no significant differences were observed between the two treatments in diminishing the methylation levels of CDH1.

Figure 2 Changes in the methylation pattern of the CDH1 promoter in HeLa and SiHa cell lines following treatment with 5-AzadC and TSA. (A and C) Lollipop methylation diagrams demonstrating changes in the methylation pattern in CpG sites of the CDH1 promoter region in (A) HeLa and (C) SiHa cell lines following treatments with 5-AzadC and TSA compared with Unt and vehicle-treated cells. Black indicates a methylated CpG site; white indicates an unmethyl-ated CpG site. (B and D) Quantification of the methylation frequency of the CDH1 promoter in (B) HeLa and (D) SiHa cell lines following treatment with the indicated concentrations of 5-AzadC and TSA compared with untreated cells.**P<0.01 vs. untreated cells. Unt, untreated; CDH1, cadherin 1; 5-AzadC, 5-aza-2'-deoxycytidine; TSA, trichostatin A.

In the HeLa cell line, treatments with 10 µM 5-AzadC (P=0.004) and 500 nM TSA (P=0.03) significantly increased the expression of CDH1 mRNA and protein (Fig. 3A and G). In addition, 500 nM TSA significantly increased the mRNA expression of SNAI1 (P=0.01, Fig. 3B) in HeLa cells without notable changes in protein expression (Fig. 3I). In the SiHa cell line, the mRNA expression of CDH1 significantly increased following treatment with 5-AzadC at 5 and 10 µM (P<0.05 and P<0.01, respectively) or 500 nM TSA (P=0.004; Fig. 3D); a similar effect was observed at the protein level (Fig. 3H). In addition, the expression of SNAI1 mRNA was significantly increased following treatment with 10 µM 5-AzadC (P=0.004) or 500 nM TSA (P=0.01) (Fig. 3E) in SiHa cells without notable changes at the protein level (Fig. 3J), similar to that observed in HeLa.

Figure 3 Changes in the expression levels of CDH1, SNAI1 and SNAI2 in HeLa and SiHa cells following treatment with 5-AzadC and TSA. (A-F) Quantitative PCR analyses of CDH1, SNAI1 and SNA2 mRNA expression levels in (A-C) HeLa and (D-F) SiHa cells following treatments with 5-AzadC and TSA compared with untreated cells; *P<0.05 and **P<0.01 vs. untreated cells. (G and H) Protein expression of E-cadherin in (G) HeLa and (H) SiHa cell lines. (I and J) Protein expression of Snai1 in (I) HeLa and (J) SiHa cell lines following the indicated treatments. Unt, untreated; CDH1, cadherin 1; 5-AzadC, 5-aza-2'-deoxycytidine; TSA, trichostatin A.

5-AzadC and TSA also increased the mRNA expression level of SNAI1 in the HeLa (Fig. 3B) and SiHa cell lines (Fig. 3E); however, no changes were observed in SNAI1 protein levels in the two cell lines (Fig. 3I and J). The expression of SNAI2 was not significantly modified at the mRNA level under any treatment condition in the tested cell lines (Fig. 3C and F).

Suppression of E7 by siRNA modifies the methylation patterns of the CDH1 promoter and induces CDH1 expression in HeLa and SiHa cell lines

To examine the involvement of E7 in the CDH1 methylation and expression patterns, HeLa and SiHa cell lines were transfected with siRNA against E7, which resulted in 57.9 and 42.5% reduction in the E7 mRNA level, respectively (Fig. 5A and C). In the HeLa cell line, E7 silencing led to demethylation of CDH1 CpG promoter sites located between +9 and -45, as well as at -103 and between -122 and -136 (Fig. 4A). In addition, in the SiHa cell line, total demethylation of the CDH1 promoter was observed following E7 silencing (Fig. 4C). Therefore, partial silencing of E7 in HeLa and SiHa cells yielded a significant decrease in CDH1 promoter methylation compared with untreated control cells (P<0.001; Fig. 4B and D). Increased CDH1 mRNA and protein levels were observed following E7 silencing in HeLa (Fig. 5B and F) and SiHa (Fig. 5B and G) cells. The use of Silencer® Select GAPDH siRNA as a positive control siRNA excluded the possibility that changes in the methylation pattern of the CDH1 promoter were due to the siRNA transfection conditions as GAPDH silencing did not induce significant changes in the methylation pattern of the CDH1 promoter in HeLa (P=0.7140; Fig. 4B) and SiHa (P=0.3248; Fig. 4D) cells; therefore, the changes in the methylation pattern were likely due to E7 silencing.

Figure 5 Silencing of E7 from HPV18 and HPV16 modifies the expression of CDH1 and SNAI1 genes in HeLa and SiHa cell lines. (A-E) The relative mRNA expression levels of (A) E7, (B) CDH1, (C) GAPDH, (D) SNAI1 and (E) SNAI2 in HeLa and SiHa cell lines following transfection with siRNA against E7 and GAPDH. **P<0.01 vs. untreated cells. E-cadherin, Snai1, GAPDH and β-actin protein expression levels in (F and H) HeLa and (G and I) SiHa cells following transfection with siRNA targeting E7 and GAPDH. CDH1, cadherin 1; siRNA, small interfering RNA; HPV, human papilloma virus.

Figure 4 Silencing of E7 from HPV18 and HPV16 modifies the methylation pattern of the CDH1 promoter in HeLa and SiHa cells. (A and C) Lollipop meth-ylation diagrams demonstrating changes in the methylation pattern in CpG sites of the CDH1 promoter region in (A) HeLa and (C) SiHa cell lines following transfection with siRNA targeting E7 and GAPDH compared with Unt and vehicle control cells. Black indicates a methylated CpG site; white indicates an unmethylated CpG site. (B and D) Methylation frequency of the CDH1 promoter in (B) HeLa and (D) SiHa cells following transfection with siRNA targeting E7 from HPV18, E7 from HPV16 and GAPDH compared with untreated cells. **P<0.01 vs. untreated cells. CDH1, cadherin 1; Unt, untreated; siRNA, small interfering RNA; HPV, human papilloma virus.

Suppression of E7 inhibits SNAI1 and SNAI2 expression in HeLa and SiHa cells

Silencing of E7 not only induced the expression of CDH1, but also significantly decreased the expression of SNAI1 and SNAI2 (P<0.001) in HeLa and SiHa cells (Fig. 5D, E, H and I). This suggested that E7 may be not only involved in suppressing the expression of CDH1, but may also regulate the expression of SNAI1 and SNAI2, which negatively regulate CDH1. No significant changes were observed in the mRNA expression of E7, CDH1, SNAI1 and SNAI2 following GAPDH silencing in HeLa and SiHa cells (P>0.5; Fig. 5A, B, D and E). However, silencing of GAPDH with siRNA resulted in a decrease in the protein expression of Snai1 in HeLa cells (Fig. 5H).

As housekeeping genes GAPDH and HPRT were used to normalize the expression levels of the genes studied, no significant differences were noted in the reduction of SNAI1 expression at the mRNA level (Fig. 5D). Further analysis of the expression levels of GAPDH, CDH1 and SNAI1 genes normalized against β-actin demonstrated that GAPDH expression was 2.6-fold higher in HaCaT, 1.6-fold higher in HeLa, 3.3-fold higher in SiHa and 3.8-fold higher in Ca Ski cells compared with a commercial sample of RNA extracted from normal cervical tissue (Fig. S2).

Discussion

HR-HPV activates the cell methylation machinery, which not only methylates its own genome, but also the promoter regions of cellular genes (21). Laurson et al (26) have demonstrated that E7 induces the expression of Dnmt1 and suppresses the expression of CDH1. Reduction of E-cadherin has been reported to contribute to the persistence of HPV, which is in agreement with reports that E7 interacts with and induces the expression of Dnmt1 and triggers its de novo methylation activity (16) (Fig. 6A). The results of the present study agree with previous findings by Laurson et al (26), in which E7 from HPV16 suppressed the expression of CDH1. However, in this previous study, no significant differences were observed in the methylation pattern of the CDH1 promoter of NIKS cells overexpressing E7, as the 17 CpG sites analyzed were not methylated (23). In addition, treatment with 5-AzadC re-established the expression of CDH1 at the mRNA and protein levels (26); however, the concentration of 5-AzadC used was not stated and the changes in the expression of other cellular genes were not discussed.

Figure 6 A schematic model demonstrating how high-risk HPV E7 may induce epithelial-mesenchymal transition via Snai1. (A) A schematic model of proteins that interact with or are induced by E7. The loss of E2 function, either from the integration or methylation of the E2 binding sites in the HPV long control region, leads to the dysregulated expression of the oncoproteins E6 and E7. E7 serves an important role in regulating several pathways; E7 induces cellular proliferation via pRB and E7 not only induces the expression of Dnmt1 and HDAC1, but also physically interacts with them, which promotes epigenetic regulation of HPV and cellular genes. The results of the present study indicated that E7 may not only suppresses the expression of CDH1 through the methylation of its promoter region, but also induce the expression of Snai1, which is a negative regulator of CDH1 expression. (B and C) Proposed models of how E7 may regulate the expression of CDH1. HPV, human papilloma virus; CDH1, cadherin 1; Dmnt1, DNA methyltransferase 1; HDAC1, histone deacetylase type 1; HIF-1α, hypoxia-inducible factor 1α; pRB, retinoblastoma protein.

The present study evaluated the re-expression of CDH1 at the mRNA and protein levels in HeLa and SiHa cell lines treated with 5-AzadC and TSA, as well as changes in the methylation pattern of the promoter region of CDH1. Additionally, although TSA is not a demethylating agent, demethylation in the promoter of CDH1 following treatment with TSA was observed. This result was an agreement with previous studies that have reported that TSA induces DNA demethylation and proposed that a change in chromatin modification, including the deacetylation of histones induced by an HDAC inhibitor, such as TSA, may render a gene susceptible to DNA demethylation (58,59).

In addition, an increase in SNAI1 mRNA level was observed after treatment with 500 nM TSA in HeLa cells, whereas in SiHa cells an increase in SNAI1 mRNA level was observed after treatment with 10 µM 5-AzadC and 500 nM TSA; however, no apparent changes were observed in SNAI1 protein level in the two cell lines. This result was in agreement with studies in which such treatments not only re-established the expression of CDH1, but also induced upregulation of SNAI1 and SNAI2 at the mRNA level (60,61), which was likely due to modifications in the methylation pattern in the promoter regions of SNAI1 and SNAI2 (62). The role of SNAI1 regulation by epigenetic mechanisms is largely unknown. A previous study has demonstrated that the treatment with 5‑AzadC in fibroblast cell IMR90, induced pluripotent stem cells from IMR90, BeWo and HTR8/SVneo cell lines induces a greater expression of SNAI1 and SNAI2 at the mRNA level and that the regulation of the two genes is mediated by DNA methylation of their first intron and not due to DNA methylation of their promoter region; however, this previous study did not determine the expression of these genes at the protein level (62). The present study did not determine the methylation status of the SNAI1 promoter region as it is transcriptionally active in HeLa and SiHa cells. The differences observed in the effect on SNAI1 expression at the mRNA and protein levels by treatment with 5-AzadC and TSA indicated that other factors may regulate the expression of SNAI1 in the two cell lines.

On the other hand, the results published by Laurson et al (26) suggested that in the NIKS-cell model, suppression of CDH1 expression by E7 was independent of the methylation status of the CDH1 promoter region, which was observed in the SiHa cell line in the current study. The previous study also suggested that repression of CDH1 may be regulated via SNAI2 (SLUG); however, it was reported that the expression of SNAI2 was not altered by the presence of E7 (26). The results of the present study revealed that while SNAI1 mRNA and protein was expressed in HeLa and SiHa cells, SNAI2 mRNA expression was barely detectable in HaCaT and Ca Ski cells. SNAI2 has been demonstrated to be upregulated in HaCaT cells during the process of cell motility and wound-healing (63).

The results of the present study demonstrated that following silencing of E7 from HPV16 and HPV18, CDH1 expression was recovered in HeLa and SiHa cells, which is in agreement with a previous study by Caberg et al (25). This previous study reported that following 24-h transfection of SiHa cells with siRNA against HPV16 E7, CDH1 was upregulated and an increase of the Retinoblastoma protein (pRB), which is responsible for a major G1 check point, blocking S-phase entry and cell growth, and activating protein 2α were detected without changes in the mRNA expression levels of SNAI1 and SNAI2 (25). By contrast, the present study demonstrated that following 48-h transfection with siRNA against E7, changes in the methylation pattern of the CDH1 promoter region were observed in HeLa and SiHa cells. In addition, an increase in the mRNA and protein expression levels of CDH1 were identified, as well as a decrease in the mRNA and protein expression levels of SNAI1 and SNAI2. Therefore, the current results suggested that E7 not only suppressed the expression of CDH1 via the methylation of its promoter, but also regulated the expression of SNAI1, a negative regulator of CDH1 involved in EMT and associated with metastasis (27,55-57). These observations are also concordant with the mechanism of action reported for other oncogenic viruses, where the X protein of HBV (HBx), core protein of HCV and latent membrane protein 1 (LMP1) of EBV promote EMT and metastasis by inducing the expression SNAI1 and suppressing CDH1 expression (64-67).

Following silencing GAPDH with siRNA, SNAI1 expression was partially suppressed in HeLa cells at the protein level, but not at mRNA level. This was consistent with a previous study that demonstrated that the interaction of GAPDH with Sp1 resulted in increased expression of Snai1, which promoted the proliferation and metastasis of cancer cells, and that suppression of GAPDH with shRNA resulted in a significant decrease of Snai1 in the HCT116 and LoVo cell lines (68). This suggested that GAPDH may serve a role in the metastasis of cervical adenocarcinoma (HeLa) by affecting EMT through the upregulation of Snai1 expression mediated by Sp1, similar to its role reported in colon cancer (68), but further studies are required to verify this.

It is currently unknown how HPV may activate the expression of SNAI1; however, it has been reported that in other cancer types associated with virus, HBx, core and LMPI proteins increased SNAI1 expression through the activation of the PI3K/Akt and MAPK pathways by transforming growth factor-β (TGF-β) action (64,69-72), which is in agreement with a previous study by Peinado et al (73) suggesting that TGF-β induces SNAI1 transcription through MAPK and PI3K.

In the present study, the signaling pathways involved in regulating SNAI1 expression were not determined; however, previous studies have reported that HR-HPV infection activates the PI3K/Akt/mTOR pathway (74), E7 from HPV upregulates Akt activity though the pRB protein (75) and TGF-β stimulates EMT and tumor invasion in SiHa cells (76).

In summary, the results of the present demonstrated that HR-HPV E7 may regulate the expression of CDH1 by two different pathways, in which Snai1 is involved. The first pathway involves hypermethylation of the CDH1 promoter region and the expression of Snai1, as observed in the HeLa cell line. The second pathway involves hypomethylation of the CDH1 promoter region with expression of Snai1 as observed in SiHa cell line, suggesting that CDH1 and SNAI1 may be considered as biomarkers of metastasis in uterine cervical cancer. Therefore, based on the present results and previous evidence that E7 interacts with Dnmt1 and HDAC1 (15-17), it would be beneficial to determine if E7 from HR‑HPV may interact with Snai1 to form a co-repressor complex with either Dnmt1 or HDAC1 in the CDH1 promoter (Fig. 6B and C), which may explain the suppression of CDH1 expression during the EMT process.

Supplementary Data

Funding

The present study received additional support of the Fundación Miguel Alemán, A.C. to AGC. PRC received a fellowship from Programa de Movilidad Internacional de Estudiantes de la Coordinación de Estudios de Posgrado (CEP-UNAM) during his doctoral stay at the Universidad Autónoma de Madrid (UAM).

Availability of data and materials

The datasets used and analyzed during the current study are available from the corresponding author on reasonable request.

Authors' contributions

PRC performed all experiments, interpreted the data and wrote the manuscript. VAV, GDBO and CCPM performed western blot analysis and interpreted the data. AC and AGC conceived and designed the study, interpreted the data and edited the manuscript. All authors have read and approved the final manuscript.

Ethics approval and consent to participate

Not applicable.

Patient consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.

Acknowledgments

This study is part of the doctoral dissertation project of Pedro Rosendo Chalma, a Doctoral student from Programa de Doctorado en Ciencias Biomédicas (PDCB), Instituto de Investigaciones Biomédicas, Universidad Nacional Autónoma de México (UNAM) and received a fellowship from Consejo Nacional de Ciencia y Tecnología (CONACyT; grant no. 203376). The present study was supported by CONACyT (grant no. 253804 to AG-C) and MINECO (grant no. SAF2013-44739-R to AC). The authors would like to thank Ms. Miriam C. Guido Jiménez (UNAM) and Ms. Raquél López Paniagua (INCan) for their technical support, Dr Erick de la Cruz Hernández (Juarez Autonomous University of Tabasco) and Dr Patricio Gariglio (CINVESTAV-IPN) for providing the cell lines and Dr Alfonso Dueñas Gonzalez (INCan-UNAM) for providing the oligonucleotides for CDH1 used in the MSP-Protocol.

References