The Associations Of Genetic Variations In Nucleotide Excision Repair Genes With DNA Damage In Coke-oven Workers And Lung Cancer Risk

The abundant polycyclic aromatic hydrocarbons (PAHs) in cigarettes and coke ovenemission (COE) have been confirmed to be the main cause of lung cancer. Lung cancerof coke oven workers is one of the eight occupational tumors legislated in our country.However, recent researches demonstrated that DNA damage level may be different evenunder similar environments. Moreover, only a small portion of smokers and workerseventually develop cancer, which indicated the importance of individual susceptibility.The nucleotide excision repair (NER) is one of the most volatile repair pathways, whicheliminates a wide spectrum of DNA lesions including bulky base adducts induced bynumerous chemical compounds like PAHs. Therefore, to investigate the relationshipbetween NER genetic variations and DNA damage level as well as the associationbetween NER gene variations and lung cancer risk would be of great importance forunderstanding the underlying mechanism of PAHs carcinogenesis, screening ofsusceptible population and improving risk assessment among the exposed subjects. Somestudies have investigated the relationship between NER SNPs and DNA damage leveland lung cancer risk, but the results are still inconsistent. It may be attributed to theirsmall studying population or the fact that all reported SNPs in the candidate genes in a whole pathway were not taken into account.In the present study, we firstly used the HapMap data to select Tagging SNPs in theNER pathway and detected the association between these SNPs and DNA damage levelin the coke oven workers; secondly we investigated the potential functions of SNPslocated in the promoter region. Finally we detected the genotypes of the SNPs that canmodify DNA damage level in lung cancer patients.PartⅠThe associations of genetic variations in the NER geneswith PAHs-induced DNA damage in coke oven workersPAHs are the main carcinogenic components of COE and can cause DNA damage,which is considered as the early event in lung cancer development. NER plays animportant role in the repair of PAHs induced DNA damage. Therefore, to investigate therelationship between NER genetic variations and DNA damage level would be importantfor understanding the underlying mechanism of PAHs carcinogenesis and screening ofsusceptible population. In this section, with the HapMap Han Chinese data, wegenotyped 32 SNPs in 8 core genes involved in NER in 475 coke oven workers. SixSNPs in 4 genes were associated with DNA damage level, five in the high exposuregroup and the other one in the intermediate. In the intermediate exposure group, theOlive tail moment (OTM) in carriers of XPA rs1800975 GG and GA genotype (median0.30 and 0.39) was significantly lower than that of the AA genotype carriers (median0.53), P＜0.05. In the high-exposure group, two SNPs were associated with elevatedlevels of DNA damage while three with decreased levels of DNA damage. DNA-damagelevels elevated in subjects with the XPC rs2228001 GG genotype (median 0.42)compared with those with the TT genotype (median 0.33, P＜0.05); meanwhile, carriersof the DDB2 rs3781619 GG genotype had higher DNA-damage levels than those of the AA genotype (median 0.39 and 0.53, respectively, P=0.036). The OTM decreased incarriers of the XPC rs3731055 AA genotype (median 0.25) compared with those of theGG genotype (median 0.45) (P＜0.05). The OTM in carriers of XPD rs50871 GGgenotype was lower than those of the TT genotype (median 0.23 and 0.40, respectively,P trend=0.048), and a similar trend was found in the rs50872 TC and TT genotypecompared with the CC genotype (P trend=0.012). No significance was found in theassociation between polymorphisms in ERCC1, XPB and XPG genes and DNA damagelevel.The diplotype analysis revealed that, in the low exposure group, the DDB2TACGA/TACGA diplotype carriers had the highest OTM (median 0.75) compared withthat of the most widely distributed CTGAG/CTGAG diplotype (median 0.29), P＜0.05. TheOTM elevated in carriers of XPC TAA/TAA diplotype (median 0.65) compared withthose of the TGA/TAA diplotype (median 0.24), P＜0.05. In the intermediate exposuregroup, the OTM in carriers of XPA GTA/GTG diplotype were significantly higher thanthat in carriers with the GTA/GAG diplotype (median 0.82 vs. 0.36), and carriers of theATA/ATA diplotype also had higher OTM (median 0.62) than those of the GTA/GAGdiplotype (median 0.35) in the high exposure group, P＜0.05. In the high exposure group,the DNA damage in subjects with XPD TCCTC/TCTTC diplotype was significantlylower than that in subjects with TGCTG/TCCTC diplotype (median 0.25 vs. 0.53).Meanwhile, carriers of the XPF AC/AC diplotype had higher DNA damage level (median0.77) than carriers of TC/TC diplotype (median 0.35), P=0.010. No other diplotypes inNER genes were found to have significant association with DNA damage level.Further analysis of combinations of genetic variants was assessed by logisticregression models. For the at-risk genotypes, the ORs for individuals with four and more risk alleles tended to be high, though not statistically significant, in all three groupsdivided by 1-hydroxypyrene levels compared with individuals having one or no riskgenotypes. For the relatively protective genotypes, the ORs in individuals with three andmore variant alleles decreased significantly compared with those with one or no variantalleles in the intermediate (OR=0.25 (0.10-0.67) and 0.31 (0.12-0.83) for 3 and≥4 variantalleles, respectively) and high (OR=0.20 (0.06-0.66) and 0.15 (0.04-0.57) for 3 and≥4variant alleles, respectively) exposure groups, compared with the individuals with 0-1alleles in the same group; the trend for increased number of protective variant allele withdecreased OR was statistically significant (P=0.01 for the intermediate and P=0.0007 forthe high exposure groups).These results demonstrated that workers with variant alleles of XPC rs2228001 andDDB2 rs3781619 had higher DNA damage level, while the OTM decreased in subjectswith XPA rs1800975 GA and GG, XPC rs3731055 AA, XPD rs50871 GG and rs50872 TCgenotypes.PartⅡFunctional investigation of SNPs located in the promoter ofXPA and XPCDifferences in the sequence structure due to the variations in the promoter regionmay give rise to the variety of binding affinity of specific transcriptional factor, and thencause the disparity in efficiency of transcription and translation. To explore the possiblefunctional impact of the SNPs on the XPA and XPC gene that had shown their effects onDNA damage level in the multivariate analysis of covariance, plasmids were constructedwith luciferase as reporter gene and transfected in cultured cells. Relative luciferaseactivity (RLA) and luciferase expression were detected. The RLA containing XPArs1800975-G promoter was remarkably higher than that of the rs1800975-A containing promoter in the three types of cell lines (all P＜0.001). Similarly, greater RLA wasobtained in the construct with XPC rs3731055-A allele compared to that with XPCrs3731055-G allele in 16HBE (P=0.08), A549 (P=0.013) and HepG2 (P＜0.01) cell lines.The luciferase expression was significantly up-regulated in the XPA rs1800975-Gcontaining promoter in all three cell lines, with the rs1800975-G allele resulting in 1.66,9.08 and 8.73 times higher in luciferase expression compared to the rs1800975A allele in16HBE (P=0.020), A549 and HepG2 cell lines (P＜0.01). Compared with constructcontaining the XPC rs3731055-G allele, the XPC rs3731055-A allele containingfragment had 3.11, 2.93 and 4.08 fold higher luciferase expression in 16HBE (P=0.08),A549 (P=0.025) and HepG2 cell lines (P＜0.01). These data suggested that the XPArs1800975-G and XPC rs3731055-A allele may enhance promoter activity and mRNAexpression level.PartⅢAssociations between NER SNPs and lung cancersusceptibilityLung cancer is a complex multifactor process, while one of the early events ofwhich is DNA damage. In order to determine whether the SNPs that can modify DNAdamage level were associated with susceptibility of lung cancer, we conducted acase-control study of 1152 patients and 1152 matched cancer free controls in south China.After adjustments of age, gender, smoking status, drink and family history of cancer, wefound that the DDB2 rs3781619 GA and GG genotypes were associated with increasedrisk of lung cancer (OR=1.19, 95%CI=0.99-1.42, P=0.067 and OR=1.31,95%CI=1.04-1.68, P=0.040, respectively). Joint effect was found in the rs3781619genotypes with smoking and family history of lung cancer. Subjects carryingrs3781619GA+GG genotypes, smoking and having family history of lung cancer had significantly higher risk of lung cancer (OR=8.04, 95%CI=4.72-13.72, P＜0.01).However, XPA rs1800975, XPD rs50871 and rs50872 were not significantly associatedwith lung cancer risk. Additionally, there were no significant associations between thefour SNPs (XPA rs1800975, XPD rs50871, XPD rs50872 and DDB2 rs3781619) andprognosis of lung cancer.In summary, we selected Tagging SNPs in the NER genes with the HapMap dataand detected the associations of these SNPs with DNA damage level in the coke ovenworkers and risk of lung cancer, and investigated the potential functions of SNPs in thepromoter region. Our results indicated that NER gene variations may moderate DNAdamage level and lung cancer risk, and SNPs in the promoter may have an impact on thepromoter activity and expression.However, lung cancer is a complex process involving a variety of factors. Thusassociations of variations in other pathways with lung cancer susceptibility in coke ovenworkers need to be detected. Secondly, protein expression and protein-proteininteractions in the related population need to be investigated in numerous pathways.Finally, further researches of the association of combining effect of SNPs and othermolecular markers such as copy number variation and microRNAs with lung cancersusceptibility are warranted in the future.