| Part onebackground:nasopharyngeal carcinoma (NPC) is a common malignant tumor in head and neck in South China, especially in Guangdong and Guangxi province. Recent studies reveal that endogenic and ectogenic reactive oxygenic species (ROS) can create DNA damages, and hydroxyl free redical and superoxide anion, the main molecules of ROS, can directly react with DNA strains and lead to their oxidizing damaged changes such as breakage, modification and DNA-protein cross-linking. About 30,000 DNA lesions are produced in one human cell per day which determine the importance of repair of these lesions to maintain the genomic DNA integrity and prevent normal cells from carcinogenesis. Repair procedures of DNA damages is accurate and rigorous. And several pathways participate in this procedure among which base excision repair (BER) is the most important. Malfunctions of genes and proteins involved in BER may compromise the repair effectiveness of DNA damages, resulting in higher risk of cancer. Previous studies showed that single nucleotide polymorphism (SNP) is one of most important factors affecting the genetic and protein function. Only a few Investigations published of relationship between SNP in BER related genes and NPC, with their results being partly discrepancy. To further study the relationship between SNPs of relative genes in BER and NPC risk, a case-control study involving NPC patients and controls in our hospital was carried out.Materials and methods:1. Bioinformatics methods was used to screened the candidate key genes in BER:(1) according to published studies, key candidate genes in BER was selected to following SNP genotyping. (2) SNPs data of candidate genes were downloaded from the International HapMap Project database. (3) Haploview 4.2 software was used to identify tagSNPs; (4). Potential functions of tagSNPs chosen were evaluated in website http://snpinfo.niehs.nih.gov/snpinfo/snpfunc.htm; (5). Based on the results of previous studies, locations in function regions, candidate SNPs choosen by the above strategies were finally identified for following study.2. Candidate SNPs of genes in BER was functionally identified in 204 NPC cases and 255 age-and sex-matched normal subjects. Blood samples of these subjects were obtained, and the DNA was extracted following genotyping using matrix assisted laser desorption ionization-time of fly mass spectrum (MALDI-TOF-MS).3. Further functional verification of the hOGG1rs1052133 and MUTYH rs3219742 SNPs (which was identified using MALDI-TOF-MS) of BER in the larger cohort of 488 NPC patients and 573 age- and sex-matched cancer-free controls (including 204 NPC cases and 255 normal cases above). Detections of their genotype were performed using TaqMan real-time PCR.Results:1. Altogether 38 tagSNPs of 19 genes in BER were selected.2. Using MALDT-TOF-MS,34 of the 38 tagSNPs were successfully detected, but none of them was significantly associated with the difference between the frequency of mutant alleles and wild type allleles (all of the P values above 0.05). We selected 2 SNPs with the P values being the smallest, i.e. hOGG1 rs 1052133 and MUTYH rs3219472 for following study of further identification.3. In our further study verifying the role of hOGG1 rs1052133 and MUTYH rs3219472, we found that cases with GG genotype was significantly higher in the controls than that in NPC patients (OR=0.769,95%CI 0.595~0.993), especially in the subgroup of female (OR=0.571,95%CI 0.337~0.968),cases with no smoking history (OR=0.635,95%CI 0.464-0.866) or thosewith age above 40 (OR=0.706,95%CI 0.524-0.950). Our gene-gene interaction study indicated that subjects with rs1052133 CC genotype and rs3219472 AA genotype (CCAA) were at 2.887 times higher of NPC risk than those with GGGG,3.183 times higher risk than those with GGGA, and 3.392 times higher risk than those with GGAA.Conclusion:1.hOGGl rs 1052133 single nucleotide polymorphsimm may significantly correlated with NPC pathogenesis, especially in the population of women, those with age above 40, and those with no history of smoking.2.The hMUTYH rs3219472 SNP may exert a gene to gene interact with hOGG1 rs1052133 SNP to impact susceptibility to NPC.Part twoBackground:NPC is one of the most common maliganant tumors in South China, with an incidence of 20-30/100,000 per year. The mechanism of initiation and development of NPC remains unclear. Our results of Part one showed that individuals with rs 1052133 GG SNP would have a significant lower risk of NPC than that with CC, indicating a tight relationship between NPC susceptibility and rs 1052133 SNP. However, only a few relative studies reported regarding the corresponding mechanism. Therefore, we gathered the clinical samples including tumor tissues, serum and medical materials to evaluate the correlation of expressions and functions of hOGGl and its SNP at rs1052133 to the clinical and prognostic parameters. The results of this study would further clarify the mechanism of NPC occurrence and progression.Methods and materials:1. Subjects:(1) for immunofluorescent study,40 cases with NPC and 15 cases with chronic nasopharyngeal mucositis (CNM) were included. These cases were treated and diagnosed in Tumor Affiliated Hospital and First Affiliated Hospital of Guangxi Medical University from June 2013 to June 2014. Part of tumor tissues were retained for immunofluorescent detection in the process of biospy. (2) Another 99 cases of NPC patients were also included in our study. Peripheral blood samples were obtained from these patients and serum was harvested for ELISA assay to detect the 8-OHdG levels; (3) analysis of relationship between rs 1052133 SNP and NPC prognosis:we selected eligible NPC cases from the 488 cases who participate the our Part One study. Clinical and follow-up materials were analyzed.2. Methods:(1) data from of immunofluorescence study were interpretated by two independent experienced staffs in our laboratory, mean values read by them were obtained for following analysis. The fluoresence intensity and expression rate were detected under confocal scanning laser microscope and analyzed by its image data analyzing software; (2) ELISA assay:levels of 8-OHdG of the samples from 99 NPC cases were detected using ELISA kit according to the manufacturer’s instructions; (3)relationship between rs 1052133 SNP and NPC prognosis:altogether 371 cases fulfilling the inclusive criteria, all of whom received follow-up study after complete radiotherapy and/or chemotherapy; (4) statistics analysis:SPSS statistics analyzing software (15.0 version) were used to evaluate the data.Results:1. the expression intensity of 8-OHdG in NPC tissues was significantly higher than that in those of CNM (P=0.003); we also found that intensity of 8-OHdG in NPC tissues with GG and GC SNP was significantly higher than that in CNM tissues with the same SNP. And the intensity of 8-OHdG in NPC tissues with GG SNP had a tendency to be higher than that in CNM carrying the same SNP with a marginal significance (P=0.062). (3) the results of ELISA assay showed that serum 8-OHdG of NPC cases with GG SNP was significantly lower than that with GC or CC SNP (P=0.038); (4) In the study of prognosis of 371 NPC cases, we found a median follow-up period of 17 months (6-48 months). During follow-up,23 cases (6.2%) experienced disease progression and 12 cases (3.2%) died of NPC at the end of follow-up; In the univariate analysis, only T classification (P=0.026) significantly correlated with overall survival (OS), and T classification (P=0.018) and CC SNP genotype (P=0.043) were significantly related to tumor free survival (TFS). In the multivariate analysis, both T classification (HR 5.123,95% CI 1.199-21.886) and CC SNP genotype (HR 2.539,95%CI 1.068-6.033) were showed to be independent factors predicting shorter TFS. We also studied the possible factors affecting the prognosis of 339 NPC cases with late stage (III+IV stage). In the univariate analysis, CC SNP genotype (P=0.117) showed no significant correlation with OS, but a marginally significant one with TFS. In the multivariate analysis, CC SNP genotype (HR 2.368,95% CI 1.199-21.886) were showed to be related to shorter TFS with a marginal significance (P=0.061).Conclusions:1. Intensity of 8-OHdG in nasopharyngeal carcinoma tissues was significantly higher than that in chronic nasopharyngeal mucositis, indicating that 8-OHdG formation is related to NPC tumorgenesis and progression.2. Serum 8-OHdG level in NPC cases with hOGG1 rs 1052133 GG SNP were significantly lower than that in those wi th CC or GC SNP, suggesting that repairing function of hOGG1 encoded by hOGGl gene carrying GG SNP at rs 1052133 was significantly compromised compared with those carrying GC or CC SNP.3. hOGG1 rs1052133 CC SNP would be a unfavorable independent factor predicting tumor-free survival of NPC patients.Part 3 In vitro study of the relationship between hOGGl rs1052133 SNP and NPC.Background:the results of part 1 of our investigation showed that individuals with GG genotype would have significantly lower risk of NPC than those with GC or CC. This was consistent with previous studies but not with some others. Reasons for this discrepancy among these studies were below:(1) some studies involved relatively small number of subjects; (2) The mechanism for the influence of rs 1052133 SNP on the NPC carcinogenesis may be different among those from different ethnic population in different geographic regions. Some authors believed that GG genotype of rs 1052133 SNP may lead to malfunction of hOGG1 protein. And this may result in excessive accumulation of DNA damages and cell apoptosis which inhibit carcinogenesis. Some previous studies revealed that malignant tumor cells may exist in normal individuals without any clinical signs because of the control of immunity system of human body. The immune cells could induce tumor cells apoptosis by some mechanisms, such as ROS production to cause DNA damages in them. The study of our Part 2 showed that CC genotype SNP at rs1052133 could be an independent factor to predict NPC TFS. Studies have indicated that radiotherapy and chemotherapy can control malignant tumors by inducing apoptosis of tumor cells in the same way as photodynamic therapy (PDT). Based on what were suggested in our Part 1 and Part 2, we put forward the hypothesis below:(1) malignant tumors may exist in nasopharynx of a normal individual. These cells would not show any clinical signs because of the DNA damages caused by immune cell induced ROS, and development of tumor cells depend on the capacity repairing the DNA damages and the functions of BER pathway proteins such as hOGG1. Since GG genotype at rs 1052133 may lead to malfunction of hOGGl, this SNP may possibly affect the risk of NPC; (2) risk of and time to NPC recurrence after radical radiotherapy and/or chemotherapy depend on the self-repairing function of NPC cells. The higher capacity the NPC cells possess, the higher risk of NPC recurrence the patients would experience. Higher ability of hOGG1 with CC type at rs 1052133 may be partly responsible for shorter TFS of those patients harboring this SNP. To verify this hypothesis, we construct hOGGl (carrying CC and GG SNP at rs 1052133) transfected NPC cell models using lentinvirus as a vector. Following functional studies investigating their biological characteristics changes to further uncover the NPC inition and progression mechanism.Methods and materials:1.qPCR was used to detect hOGG1 mRNA expression cell lines of HK1, HONE1,5-8F,6-10B, CNE1, CNE2, NP69. Showing the lowest level of hOGG1 mRNA, HK1 was choosen as the studied cell line in the following investigation.2. Using vector of lentivirus, HK1 was transfected with hOGG1 (with CC and GG genotype SNP at rs 1052133) to construct cell models, namely OGG1-CC-HK1 and OGG1-GG-HK1. HK1 cells were also transfected with lentivirus without hOGG1 to establish a model named N-HK1 as a control. Fluorescence microscope was subsequently used to observe the transfection effectiveness. qPCR and western-blot (WB) was utilized to detect hOGG1 mRNA and protein expression. DNA sequencing methods was implemented to verify tranfected sequence with MTT assay used to proliferation evaluation for HK1, N-HK1, OGG1-CC-HK1 and OGG1-GG-HK1; (4) MTT assay was used to evaluate the proliferation of HK1, N-HK1, OGG1-CC-HK1 and OGG1-GG-HKl before and after PDT applying 5-ALA as photosensitizer, and their repair functions were also analyzed. In-cell Western (ICW) was used to check the changes 8-OHdG/hOGG1 ratio after PDT. ELISA assay was used to analyzed the 8-OHdG level in the supernatant of all the constructed cell models; (7) Flow cytometry was applied to detect the percentages of the constructed cell models in different stages in cell cycle; (8) we also used transcriptome methods to analyze the rs 1052133 SNP effects on mRNA changes of downstream genes.Results:1. Semiquantitative qPCR results revealed that a 2.3-fold higher level of hOGG1 mRNA in OGG1-GG-HK1 and OGG1-CC-HK1 compared to that in N-HK1. And data of WB showed that the level of hOGGl protein is higher in OGG1-GG-HK1 and OGG1-CC-HK1 than that in N-HK1;2. Genetic sequencing data indicated that a stable model of OGG1-GG-HK1 and OGG1-CC-HK1 had been successfully established carrying the transfected hOGGl gene with CC and GG genotypes at rs 1052133 without any other sequence variations.3. MTT assay results revealed that OD values of OGGl-GG-HK1 in day 2,3,4 after inoculation were significantly lower than that of OGG1-CC-HKl (P< 0.01).4. Detecting the repair function of the OGG1-GG-HK1 and OGG1-CC-HK1 after PDT, MTT assay was also used.The data showed that ratios of OD value of 12h,18h,24h and 36h after PDT to that of 6h after PDT in OGG1-GG-HKl were lower than those in OGG1-CC-HK1. The difference of ratios of 24h between the two cell models was significant (P< 0.05), indicating that former has lower repair capacity than the latter.6. Data from ICW showed that 8-OHdG/hOGG1 ratio in OGG1-GG-HK1 was significantly lower than that in OGG1-CC-HKl before PDT. However, this ratio increased in OGG1-GG-HK1 18h and 24h after PDT and was significantly higher than that in OGG1-CC-HK1, suggesting that weaker repair capacity in OGG1-GG-HK1 than in OGG1-CC-HK1 after oxidative damage.7. Data from transcriptome study showed that hOGG1 mRNA level was significantly lower (by 2.6-fold) in OGG1-GG-HK1 than in OGG1-CC-HK1 24h after PDT. Nevertheless, The results showed that hOGGl mRNA level in OGG1-CC-HKl increased significantly 24h after PDT by 2.23-fold. However, hOGGl mRNA level in OGG1-GG-HK1 was kept in the same level 24h after PDT as that before PDT, leading to its lower level by 2.6-fold, as a result, than that in OGG1-CC-HKl 24h after PDT. This suggested a higher self-repairing capacity of OGG1-CC-HK1 than OGG1-GG-HK1 after the oxidative damage. Our transcriptome method also indicated a significant 2-fold increase of SLC39A10 mRNA before and a 2.4-fold one after PDT in OGG1-CC-HK1. And our data also showed a decrease before PDT with no change after PDT. In OGG1-GG-HK1. mRNA of DUSP6, DUSP4 and LAPTM5 were down-regulated in OGG1-CC-HK1 and up-regulated in OGG1-GG-HK1 with higher levels of the latter than that of the former after PDT. Like DUSP6, DUSP4 and LAPTM5 mRNA, GADD34 was also down-regulated in OGG7-CC-HK1 and up-regulated in OGG7-GG-HK1. Nevertheless, a changeover from down-regulation to up-regulation of this mRNA was showed after PDT in OGG1-CC-HK1, with no change for OGG1-GG-HKlConclusions:1. Stable NPC cell models, named OGG1-GG-HK1 and OGG1-CC-HK1 were successfully established.2. Repair capacity of OGG1-GG-HK1 was significantly weaker than that of OGG1-CC-HK1 after PDT, and this may be associated with decreased mRNA level of hOGG1 in OGG1-GG-HK1 under oxidant stress.3. SNP at rs 1052133 of hOGGl may alter the characteristics of apoptosis of HK1 cells through the regulation of its downstream genes including SLC39A10, DUSP6, DUSP4,LAPTM5 and GADD34.Part four Functional effect of SNP at rs1052133 of hOGG1 in NPC-bearing nude miceBackground:Our in vitro studies in Part 3 showed that SNP at rs1052133 of hOGGl plays a important role in repair capacity of HK-1, but this results should be further verified by in vivo investigation. Animal models are very useful in research investigating carcinogenesis mechanism, which is categorized by spontaneous tumor model, induced tumor model, genetic modification tumor model and transplanted tumor model. And transplanted tumor model is widely used, and divided by homotransplanted tumor model and heterotransplanted tumor model. Nude mouse is the most widely used in transplanted tumor model and has some advantages:(1) stable tumor formation rate with low individual variance rate; (2) model construction is relatively simple and not time-saving; (3) Tumor transplanted can maintain it original biological characteristics even generations after construction; (4) high tumor formation rate dued to immunodeficiency of animals used. Studies have indicated that results from in vitro investigations are of relatively limited significance using mono-layers of tumor cell. In vivo studies using animal model can provide 3-dimension microenvironment for tumor cells interacting with each other and matrix. In the study of this part, we construct NPC transplanted nude mice models carrying OGG1-CC-HK1 and OGG1-GG-HK1, and the effect of SNP at rs1052133 on HK1 cells in vivo was evaluated. We believe that the results would further clarify the mechanisms of NPC tumorgenesis, progression, recurrence and metastasis.Materials and Methods:1. Thawing, cultivation, generation passage, counting of HK1,N-HK1, OGG1-GG-HK1 and OGG1-CC-HK1.2. Nude mice were randomly divided into 4 groups, and each has 5 nude mice transplanted with HK1, N-HK1,OGG1-GG-BK1 and OGG1-CC-HK1 respectively.3. Cells suspensions containing HK1, N-HK1, OGG1-GG-HK1 and OGG1-CC-HK1 were injected subcutaneously into axillary of nude mice. Transplanted tumor sizewas measureed very 2 days. Tumors were removed after nude mice execution and their weights were checked at the 30th day.4. hOGG1 mRNA level was measured using real-time fluorescent PCR.5. SNPs at rs1052133 in HK1, N-HK1, OGG1-GG-HK1 and OGG1-CC-HK1 was evaluated by genetic sequencing method.Results:Formation of transplanted tumor was successfully established in all of the cell models of HK1, N-HK1, OGG1-GG-HK1 and OGG1-CC-HK1. Transplanted tumor formation began at 5 to 8 days, growed slowly in 3 weeks and rapidly in the 4th week after inoculation. In the 30th day, the mean weight of tumors from nude mice with OGGl-GG-HK1 and OGGl-CC-HK1 (0.591±0.23 g)was higher than that with HK1 and N-HK1(0.360±0.16g) (P=0.021).Conclusion:1. Nude mice model bearing OGG1-GG-HK1 and OGG1-CC-HK1 were successfully established, providing the base for further investigation of the relationship between SNP at rs 1052133 of hOGGl and NPC occurrence and development.2. Overexpression of hOGG1 could promote proliferation of HK1 cell lines transplanted in nude mice model. |