Methylation Status Of HPV-16 Long Control Region And Its Effects On HPV-positive Head And Neck Cancer

Head and neck squamous cell carcinoma (HNSCC) is a group of common malignant tumors, including oropharyngeal cancers, laryngeal cancers, hypopharyneal cancers, oral cancers, and nasopharyngeal cancers. During the past decades, human papillomavirus (HPV) have been implicated in the pathlogenesis of cervical cancers and anal cancers. Persistent infection with high-risk HPV (hr-HPV), such as HPV-16/18, is an independent etiological factor other than excessive tobacco and alcohol consumption for HNSCC. Worldwide, about 20%-30% of HNSCC patients are detected with the infection of hr-HPV, particularly oropharyngeal SCC (OPSCC), of which 50%-70% of patients are infected with HPV-16. Many epidemiological studies revealed that the proportion of HNSCC patients with hr-HPV infection increased rapidly during past decades, especially in young patients who never smoke and drink, therefore, the carcinogenesis role of HPV infection in a subset of HNSCC patients is seriously addressed. Howere, the rates of HPV infection varied widely by geographical regions and ethnics groups, and by different detection methods. The screening of HPV-positive patients with head and neck tumors is of great significance for tumor prevention and therapy in a certain region. Also, it is a necessary step for further study involving HPV-related HNSCC patients. In this study, we adopted a high-sensitive PCR amplification to detect the HPV-DNA in fresh frozen specimens of head and neck tumors. We analyzed further the correlation between HPV infection and clinical characteristics.Laryngeal cancer (LC) is one of the most common lethal cancers worldwide, and 96%-98% LC patients were squamous cell carcinomas in pathology. Many researches reported the rate of HPV infection in patients with LC yielded widely varying estimates from 3% to 60%. Previous studies on the association between HPV infection and LC in Chinese population suggested a higher infection rate of HPV and possibly a more close association with LC risk. In a meta-analysis on worldwide HPV infection rate of 55 studies, including 3 from China, overall HPV prevalence in LC was 28.0%(95% CI,23.5%-32.9%). Notably, however, the HPV detection rates in the 3 studies from China were 58.8%,36.4%, and 41.67%, obviously higher than the mean rate of America and European. It should be noted that the number of researches on the HPV infection in LC patients in China is less than in Western countries and Japan, and largely varied rates of HPV infection were reported in these researches, which were possibly introduced by various sample sizes, sample storage methods, and detection methods. Therefore, HPV prevalence and LC cancer risk conferred by high-risk HPV infection remain largely uncertain in a Chinese population, necessitating an updated systematic review incorporating the most recent studies, to evaluate comprehensively the overall HPV infection rate and the cancer risk in Chinese LC patients.HPV-16 genome contains 8 open reading frame (ORF) coding for early protein (El, E2, E4, E5, E6, and E7) and late protein (L1 and L2), and a non-coding gene region which is also named as long control region (LCR) and performs a critical regulatory function in the maintenance of viral replication and life circle. Various regulatory elements including four E2 binding sites (E2BS-1, E2BS-2, E2BS-3, and E2BS-4) located in the LCR region. Usually, the sequence of LCR could be divided into 3 parts in structure,5’-LCR, Enhancer region, and Promoter region. E6 and E7 are two viral oncoproteins of HPV-16, which could maintain the malignant transformation through binding and degrading p53 and retinoblastoma (pRb) tumor suppressors, respectively. Whereas, E2 protein controlls the viral replication the expression of E6 and E7 through the binding to E2BSs. Therefore, LCR performs a vital function in the HPV-mediated malignant transformation.DNA methylation is an important pattern of epigenetic modifications, which could regulate gene expression at transcription level. Methylation usually occurs in the CpG islands or CpG sites, and could induce gene inactivation. Therefore, the methylation at the promoter of tumor suppressor genes is usually closely related with the tumorigenesis. Compared with HPV-negative HNSCC, a higher global methylation level was found in HPV-positive HNSCC, which may be involved in a specific signal pathway in carcinogenesis. It had been known for decades that HPV-genome was subjected to be methylated by the host methylation machinery, and methylation of HPV genome was related to the severity of cervical lesions, as previously reported. The methylation of HPV genome, especially LCR, could affect the viral cycle, transcription and replication of the viral genome. The binding of E2 protein to E2BSs was usually hindered by the methylation of E2BSs, and then E2 protein lost the repression function on E6 and E7 overexpression. Therefore, the methylation of CpG sites in LCR is closely associated with the tumorigenesis of HPV-related cancers.HNSCC and cervical cancers are two different groups of cancers with differences in biological characteristic, tumor etiology, therapy, and prognosis. In cervical cancers, inconsistences of methylation status were frequently reported in different researches. However, little is known on the methylation of HPV genome in head and neck cancers. To detect the methylation status of HPV LCR in OPSCC patients, especially the methylation of E2BS within LCR, and to investigate further the correlation between methylation of HPV LCR and clinical characteristics is helpful to explain the role of methylation in carcinogenesis and cancer progression. Furthermore, up to date, no study reported the methylation of HPV LCR in HPV-positive HNSCC cells, and the detection of HPV LCR in different cell lines could establish different experiment models for further research on carcinogenesis mechanism and demethylation treatment.E6 and E7 expression is mainly regulated by the binding of E2 protein to specific E2BS-3 and E2BS-4 located within the Promoter region of HPV-16 LCR, and some regulatory elements located within the Enhancer region, such as NF-1, AP1 and Tef also contribute to the regulation of E6 and E7 expression. As the methylation of CpG sites could lead to gene silencing, and the methylated CpG (meCpG) is potentially reversible, therefore, demethylating the meCpG sites of HPV genome thus to recovery the function of the regulatory elements not only is a valuable therapy target for cancers, but also is helpful in investigating the carcinogenesis role of methylated LCR for HPV-positive cancers. Recently, some researchers explored the demethylation of HPV-genome in HPV-positive cervical cancer cells, such as CaSki, SiHa and HeLa, revealing different responses on cell growth and physical properties as well as E6 and E7, which inspired us in the current study to explore the possible different roles of methylation status of HPV LCR.To date, however, no study reported the methylation status in HPV-positive head and neck cancer cells before and after demethylation treatment and the status variant in HNSCC cells. Therefore, we utilized HNSCC cells for the first time to map comprehensively the methylation status of the CpG sites within the LCR, and to explore further the effects of demethylating HPV 16 LCR on cell bioactivity and E6 and E7 expression, and to explore further the carcinogenesis mechanism in HPV-positive cancers. In addition, this study also evaluates the demethylation of meCpG sites of HPV-16 LCR as a valuable target in a subgroup of head and neck cancers.Chapter 1 Prevalence of HPV infection in head and neck cancers Part 1 Detection of HPV infection and clinical feature analysis in head and neck lesion tissuesObjective:The aim of the present study is to investigate the epidemiologic and clinical features of HPV in head and neck lesion tissues.Methods:A total of 180 specimens were obtained from patients with head and neck diseases treated at the Department of Otorhinolaryngology, Head and Neck Surgery, University of the Ryukyus between July 2013 and April 2014. Among theses samples, a total of 163 were obtained from head and neck tumors,14 samples were from inflammatory and atypical hyperplasia tissues, and 3 samples were obtained from laryngeal papilloma patients. The tumor specimens were consisted of 83 cases of squamous cell carcinoma,5 cases of laryngopharyngeal adenocarcinoma,16 cases of lymphoma,7 cancer cases of unknown primary origin,29 cases of parotid gland tumors,8 cases of external ear canal tumor, and 4 cases of submandibular gland tumor. All the biopsies were snap-frozen immediately once taken from the patients and stocked in the liquid nitrogen. Then, the DNA were extracted from the specimens and submitted to test the quality. HPV sequences were screened by a nested PCR using two consensus primer sets, GP5+/GP6+and MY09/MY11, which targeted L1 and El of HPV genome. DNA extraction from CaSki (HPV-16 positive) and HNO41 (HPV negative) cells were used as positive control and negative control, respectively. All positive PCR productions were purified and sequenced directly, and then the obtained sequences were aligned with those of known HPV types available from the GenBank database using the BLAST program, to confirm the infection of HPV and determine HPV types.Results:The highest rate of HPV infection (100%) was found in laryngeal papilloma tissues, HPV-6 was indentified in all specimens. The average infection rate of HPV in HNSCC tissues was 22.9%(19/64), HPV-16 was indentified in all these positive specimens. Among HNSCC patients, the HPV infection rate in oropharyngeal cancers was 36.0% (9/25), higher than the rates in other sublocation. Moreover,21.7% (5/23) of hypopharyneal cancers,20.0%(1/5) of nasopharyngeal cancers, and 17.4% (4/23) of oral cancers were all detected with HPV infection, but none of laryngeal cancers (7 cases) was infected with HPV. Also,1 of 5 laryngopharyngeal adenocarcinoma specimnes,1 of 14 inflammation and atypical hyperplasia specimens, and 1 of 16 laryngopharyngeal malignant lymphoma specimens were detected with HPV-16 infeciton. However, none of HPV-positive case was found in thyroid tumors (11 cases), parotid gland tumors (29 cases), submandibular gland tumors (4 cases), external ear canal tumor (8 cases) and cancers of unknown primary origin (7 cases). Study on clinical features in HNSCC patients revealed that, compared with HPV-negative patients, HPV-positive patients were most likely had an early T stage (P= 0.034), and a higher proportion of low-age patients. (P= 0.074).Conclusions:In conclusion, laryngeal papilloma patients were exactly associated with HPV-6 infection. A subgroup of HNSCC patients were related with HPV-16 infection, the association was more obvious in oropharyngeal cancers. However, the association between HPV infection and laryngeal cancers was unable to determine as the limitation of sample size. No significant associations were observed between HPV infection and thyroid tumor, and parotid gland tumor.Part 2 Prevalence and laryngeal cancer risk in Chinese patients with HPV infection:a meta-analysisObjective:The purposes of this meta-analysis were to estimate high-risk HPV (hr-HPV) infection rate in Chinese laryngeal cancer (LC) patients and assess the LC risk conferred by hr-HPV infection in this population.Methods:We searched both the English and Chinese databases for studies published in either English or Chinese up to October 2013, and systemically reviewed 22 original research articles which met the inclusion criterias. The related data were extracted in accordance with the projects. Both the HPV-16/18 infection rate in the LC group (all 22 studies) and the LC risk conferred by infection with HPV-16/18 (a subgroup of 12 case-control studies) were meta-analyzed by R 3.0 software.Results:① A total of 1477 LC cases from the 22 articles were enrolled in this research, the infection rate of HPV-16/18 was 30.1% (95% CI:24.2%-36.8%). ② In order to evaluate the cancer risk introduced by the infection of high-risk HPV-16/18, a total of 1040 LC and 735 cases free of tumors were divided into the LC group and the control group, respectively. The infection rates of HPV-16/18 were 31.1% and 5.0% in LC group and the control group, respecitively. There was a strong association between high-risk HPV-16/18 infection and LC (P<0.001; OR= 8.07,95% CI: 5.67-11.48), indicating infection with high-risk HPV-16/18 significantly increased the LC risk in a Chinese population.Conclusions:A high HPV-16/18 prevalence was found in Chinese LC patients, and the infection of HPV-16/18 significantly increased the LC risk.Chapter 2 Methylation status of the CpG Sites within the HPV-16 LCR in HPV-positive cancer cells and OPSCC tissuesObjective:To explore the various methylation statuses of CpG sites within LCR in three HPV-16 positive cancer cells, and to detect the methylation status of CpG sites within the HPV-16 long control region (LCR) in oropharyngeal squamous cell carcinoma (OPSCC) infected with HPV-16.Methods:DNA Extraction from UM-SCC47, CaSki and SiHa cells, and OPSCC specimens which were positive of HPV-16 infection which were confirmed by DNA sequencing in Chapter 1 and a previous study of our department, a total of 19 OPSCC specimens were researched in this study. All the specimens were taken from cancers origined from tonsils and tongue base, and the majority of these cases were in advantage stage, including 0 case of stage Ⅰ,1 case of stage Ⅱ,1 case of stage III,17 cases of IVA stage and 1 case of IVC stage. Histological differentiation classification of these specimens was 6 cases of well differentiation,10 cases of moderate differentiation, and 3 cases of poor differentiation. The DNA extracted from cells and tissues were bisulfite-converted using a bisulfite kit. The bisulfite-modified DNA was amplified in 3 amplicons with bisulfite-sequencing PCR (BSP) primer assays targeted 5’-LCR region, Enhancer region, and Promoter region, respectively. The BSP products were purified and then cloned into the pCRTM4-TOPO(?) vectors. For each amplicon, at least 6-8 (6 for tissues and 8 for cells) individual E. coli plaques were cultured for plasmid DNA extraction, and then sequenced individually (6-8 sequencings).Results: ① The DNA was completely bisulfite-conversed, and then BSP amlificated and sequenced. Therefore, every CpG site had at least 6-8 CpG clones. The methylation frequency of every CpG site could be calculated by the number of meCpG clones dividing the number of total meCpG clones. A total of 15 CpG sites were identified in CaSki and SiHa cells and in all the OPSCC specimens. However, two nucleotide (nt) changes at nt7435 and nt31 altered the presences of CpG site there, therefore,13 CpG sites were contained in the LCR of UM-SCC47 cells. ②The methylation rate of HPV-16 LCR were 79.8%(83/104) in UM-SCC47 cells and 90% (108/120) in CaSki cells, respectively. However, the LCR region of SiHa cells was completely unmethylated (0/120). In the 19 OPSCC specimens, the mean methylation rate of the LCR was 13.1%(194/1481). Moreover, on the whole, the methylation status of LCR in OPSCC could be classified into 3 types:a hypermethylation type (methylation rate> 30%), a hypomethylation type (5%-30%) and an unmethylated type (methylation rate<5%). In UM-SCC47 and CaSki cells, the mean methylation rate in 5’-LCR-Promoter region was higher than that in Enhancer region, (91.7% vs. 76.3%) and (100% vs.86.4%), respectively. The mean methylation rate in 5’-LCR and Promoter regions were 14.6%(58/396) and 16.6%(95/577) in OPSCC specimens, which were higher than the mean methylation rate of 8.1% in Enhancer region (P= 0.002 and P< 0.001). The CpG site located at nt7862 tended to be hypomethylated in UM-SCC47 and CaSki cells and in OPSCC specimens. Compared with the CpG sites outside the E2BS, the CpG sites within E2BSs other than E2BS-2 (nt7862) were hypermethylated in UM-SCC47 cells (97.9% vs.68.8%, P< 0.001), CaSki cells (100% vs.91.1%, P= 0.060) and in OPSCC patients (17.5% vs.11.2%, P= 0.001).Conclusions:Two methylation types of HPV-16 LCR were found in HPV-positive cancer cells; though the mean methylation rate of the LCR was relatively low in OPSCC tissues, various methylation types could also be classified.Chapter 3 Effects of demethylating CpG sites within HPV-16 LCR on head and neck cancer cellsObjective:To detect the effects of demethylating CpG sites within HPV-16 LCR in UM-SCC47 cells, including cell bioactivities and oncogenic E6 and E7 expression. This research also aims to analyze the responses upon demethylation treatment in HPV-16 positive cancer cells with different methylation types of LCR.Methods:① BSP was used to detect the change of CpG methylation status, and a nested PCR (a combination of BSP and MSP) was carried out to evalute the overall methylation status of the LCR and methylation variation after 5-aza-2’-deoxycytidine (5-aza-dc) treatment. ② Total RNA was extracted from the cells and converted to cDNA; Both RT-PCR and quantitative real-time PCR (qRT-PCR) were adopted to analyze E6 and E7 mRNA expression after 5-aza-dc treatment or E6 and E7 knockdown. The expression of E6 and E7 were detected by a standard curve method and calculated as E6/β-actin and E7/β-actin.③ Immunocytochemistry was adopted to evaluate the expression of E6 and E7 oncoproteins in cells with different treatment. ④ We use MTS and trypan blue staining assays to exam the cell growth and cell proliferation after 5-aza-dc treatment and E6 and E7 knockdown.⑤ Cell apoptosis were determined by Annexin-V and 7-AAD staining, and the cell cycle arrest was evaluated by prodidium iodide staining.⑥ We further sought to determine whether the demethylation-induced apoptosis and cell cycle arrest were mediated by the decrease of E6 and E7, we transfected siRNA using a Lipofectamine RNAiMax to silence E6 and E7, and then detected the growth inhibition, apoptosis, and cell cycle arrest.Results: ⑦ In the present study, the result of nested PCR revealed that the treatment with 0.5 μM 5-aza-dc for 96 h showed the optimal demethylation potency, which was in line with the result by BSP and TA cloning. ② After treatment with 0.5 μM 5-aza-dc for 96 h, the rates of meCpGs within the LCR in UM-SCC47 and CaSki cells were significantly decreased to 19.2%(20/104; P< 0.001) and 29.2% (35/120; P < 0.001), respectively, whereas, no change was found in SiHa cells (0%,0/120). ③ Upon demethylation treatment, the mRNA expressions of E6 and E7 were significantly decreased in UM-SCC47 (P= 0.034 and P= 0.008, respectively) and CaSki cells (P= 0.017 and P= 0.023, respectively), but not in SiHa cells. Furthermore, we detected E6 and E7 proteins by immunocytochemistry, E6 and E7 proteins were localized in the cytosol and nucleus of UM-SCC47 and CaSki cells, the percentage of stained cells and staining intensity were significantly decreased after demethylation. ④ Upon demethylation, UM-SCC47 and CaSki cells whose meCpG sites were obviously demethylated, responded distinctly during growth, apoptosis (P = 0.026 and P= 0.001), and cell cycle arrest at S (P= 0.048 and P= 0.003) and G2/M phase (P= 0.002 and P= 0.044), however, the growth of SiHa cells was also slightly affected by the demethylation reagent, no significant changes were observed on apoptosis or cell cycle arrest at S and G2/M. ⑤ At 96 h after knockdown of E6 and E7, the growth of UM-SCC47, CaSki, and SiHa cells were inhibited (P= 0.044, P= 0.006, and P= 0.022, respectively), moreover, apoptosis (P= 0.026, P= 0.036, and P = 0.016, respectively) and G2/M stage arrest (P= 0.027, P= 0.035 and P= 0.012, respectively) were observed. However, the S-phase arrest of SiHa cells was most likely related with the inherent characteristics of the drug. Conclusion:HPV-16 positive cancer cells with different methylation types of the LCR responded distinctly to the demethylation treatment, suggesting there may have different carcinogenesis mechanisms in HPV-positive cancers. In addition, our findings suggested that demethylating the meCpGs in HPV16 LCR might be a valuable target for a subgroup of HPV16-positive HNSCC patients.