The Regulation Of Genetic Damage By DNA Methylation And Mutations In Related Genes In Benzene Exposed Workers

Benzene is an important organic chemical and a confirmed human carcinogen. Chronic benzene exposure can cause hematotoxicity, including redued blood cell counts in peripheral blood, aplastic anemia, MDS and leukemia. Presently, chronic benzene poisoning accounted for one of the top three in occupational chronic poisoning, and which benzene induced leukemia is the first class in eight categories of occupational cancer. Now, the occupational exposure limit (OEL) of benzene in developed countries is 1 ppm (3.25mg/m3), while the PC-TWA in China is 6mg/m3.However there were several researches suggesting that hematotoxicity still could be observed when concentration of benzene in air was below 1ppm. Hence it is necessary to study benzene-induced hematotoxicity and gene-toxicity at current OELs. Susceptibility to these toxicities can affect by age, sex, health-and-nutrition level, especially the activity of environmental response genes and ability of DNA repair of individual, all of which can be attributed either to gene polymorphisms, or epigentics regulation. A pile of studies have been done to investigate the relationship between benzene-induced genetic damage/hematotoxicity, the gene polymorphisms and DNA methylation.Epigenetic regulation provides a new way for cancer. Especially DNA methylation is an important way to regulate the activity of the proteins, including the benzene exposed workers. Reports showed that DNA methylation levels in peripheral blood can be used as a biomarker of related bladder cancer, stomach cancer, breast cancer and colon cancer. And it can be a new biomarker of early damage or exposure of benzene exposed. Analyse the DNA methylation level of peripheral blood and genes related to damage repair from different benzene exposed workers will be usful to find new biomarkers. Furthermore, screening the polymorphism of DNA-methylation related genes is useful to discover its’ relationship with DNA methylation. Explore the relationship among the environmet, DNA methylation and gene is useful to discover new biomarkers of exposure or effect.In order to explore effect biomarkers under low level benzene exposure, this tudy evaluated the relationship between the cumulative exposure dose and benzene-indueed damage (hematologic toxicity and chromosomal damage). Two types of benzene exposure groups were inversted. Firstly,385 workers (male190 and female 195) were invested from six shoes factories from Zhejiang provience with average 27.3± 4.5 old (19-57olds). The orther 181 (male 144 and female 37) exposure workers were from painting workshop in Anhui’ car factory. There was 220 and 26 control from the same city as exposed workers in Zhejiang and Anhui separately. Furthermore,95 teachers were recuited from Shanghai as the external control. The peripheral blood, routine urine and the fuction of liver and kidney were detected in health examination. In addition to 118 controls in Wenzhou group, all of subjects were examined by cytokinesis-block micronucleus (CBMN) assay. And 35 workers in Anhui were followed up this time after the first investigation in 2009.The concentration of benzene in worksites was detected by the combination of point sampling and personal sampling. The cumulative exposure dose based on exactly calculated benzene concentration the duration in worksites for exposure assessment. According to the effective biomarkers of peripheral white blood cell count and genetic damage, the benchmark dose (BMD) method was applied to calculate the benchmark dose of benzene, for a reference in the determination of occupational exposure limit. The PCR-RFLP was used to detect polymorphisms in DNA-methylation related genes and metabolic enzyme genes, including DNMT3A (rs36012910, rs1550117, DNMT3A18, R882) DNMT3B (rs2424909, rs2424913, rs1569686), IDH2 (R140/R172), MGMT(rs 16906252), CYP2E1 (rs3813867, rs2031920, rs6413432, GSTT1/TM1), GSTP1 (rs1695) and mEH (rs1051740, rs2234922). We used high resolution melting (High Resolution Melt, HRM) to detect the methylation level in peripheral blood Linel, MGMT and Hmlh1, and MSP was applied to detect the methylation level MGMT, hMLHl, RAS, P53, P16, hMLH2, NQO1, GSTP1 in 140 benzene exposed workers and 48 controls.The benchmark dose software 2.2.1 (Version 2.2.1, US EPA) was used to calculate benchmark doses of benzene exposure. The BMDL was the lower 95% confidence limit on the BMD corresponding to 10% excess risk increase above the background. We used the 5-percentile of WBC count in the control workers as the cut-off point below which a worker’s WBC count (≤ 4.3 × 109) is considered reduced below the normal range. Thus, WBC count was converted to a dichotomous indicator of normal or reduced WBC count. The 95-percentile of the control’s MN frequency was used as the cut-off level, above which the subject’s MN frequency (>4%) is viewed as elevation above the normal range, i.e. a potential chromosomal damage. The lower confidence limit (BMDL) of BMD corresponding to a 10% excess increase risk above the control level was found to be 24.21 mg/m3-year,7.64 mg/m3-year for reduced WBC count, and MN frequency separately. Among BZ-exposed male, female, and male and female pooled, and was 7.41,4.41,6.76 ppm-year for reduced WBC count in the three groups, respectively. According to working lifetime of 40 years, the reference dose for benzene exposeure was 0.61 mg/m3,0.19mg/m3 for reduced WBC count, and MN frequency separately.The multivariate Poisson regression analysis showed that the micronucleus in demonstrated that the BZ-exposed workers had significantly increased MN frequency compared with the controls (FR (95%CI):1.005 (1.002,1.008), P< 0.01), and showed a cumulative exposure dose--esponse relationship. And the WBC in the exposed workers indicated a statistically significant decrease compared with the control (contact group vs:5.61 ± 1.58 vs 6.47 ± 1.40, P<0.01), and showed a cumulative exposure dos--response relationship too. Furthermore, the DNA repair capacity in exposed workers was lower than the controls. The 3AB index in exposed was 0.22, significantly lower than the control group (0.37) by one way analysis of variance (P< 0.01).The mthylation of LINE1, MGMT and hMLH1 were detected by HRM method. The LINE1 methylation in the controls (54.09±5.84)% was higher than the benzene exposure group (48.16±8.07)% with significant statistical difference. However, both of the MGMT (contro vs exposed:4.89 ± 2.14% vs 6.81 ± 2.14%, P<0.05) and hMLH1 (contro vs exposed:10.61 ± 4.84% vs 11.94 ± 7.81%, P<0.05) in the control group were lower than the exposure with significant statistical difference. The results from MSP indicated that the methylation status of GSTP1, hMLHl, MGMT and RAS in the control group were 8%,8%,29% and 6% separately, and the methylation rate in the contact group were 30%,34%,61% and 35% separately. The methylation rates in exposed were significiant higher than controls.Effect of genetic polymorphisms on methylation showed that, the R882C locus (R882C wild type vs; 45.06 ± 6.97% vs49.35 ± 8.39%, P<0.05) and mutant R882H (R882H wild type vs; 44.04 ± 3.13 vs49.35 ± 8.39%, P<0.05) can lead to reduced LINE1 methylation rate than the wild type. The heterozygous genotype in rs36012910 (heterozygous mutant vs wild-type; 46.37 ±4.55% vs48.58 ± 8.56%, P=0.266) represent lower LINE1 methylation than the wild type, but there were no statistical significance. Diplotype analysis showed that LINE1 methylation in DNMT3A genotype AGG/GGG, was lower than wild type, with statistical significance (P=0.037). The MGMT promoter methylation in heterozygous rs36012910 locus, was higher than the wild type, but the P value near the edge of point of departure (P=0.079).Multiple linear regression analysis demonstrated that the heterozygous of DNMT3A rs36012910 (wild-type vs heterozygous:5.66 ± 1.79 ×109 vs 6.34 ± 2.51 ×109, P<0.01), and DNMT3B rs2424913 (wild-type vs mutant:5.75 ±1.63 × 109 vs6.78 ± 3.17 ×109, P<0.05) had a significantly increased white blood cell count compared with the wild type. Univariate Poisson regression revealed that DNMT3A sites of R882C and R882H genotype micronucleus frequency (FR (95% CI); 1.14 (1.02,1.29), P=0.024) was significantly higher than the wild type. The CYP2E1 rs3813867 mutant allele (CC+GC (FR 1.15,95% CI 1.02-1.29; P= 0.020) and rs2031920 variant allele (CT+TT) (FR=1.23,95% CI:1.09-1.37,p<0.01) was not only associated with higher MN frequency, but also rs3813867 (mutant vs wild-type:5.74 ± 1.69 ×109 vs5.34 ± 1.29 × 109, P<0.05) and rs2031920 (mutant vs wild-type:5.75 ±1.63 × 109 vs 5.32 ±1.43 × 109, P<0.05) related to reduced WBC counts significantly compared with the wild genotype separately.We conducted a follow up of 35 benzene exposed workers in car painting workshop. The white blood cell counts in 2013 was significantly higher than that of 2009 (2013 vs 2009:5.39 ± 2.75 × 109 vs 4.62±1.44 × 109; P<0.01), the difference was statistically significant; the same trend in red blood cell (2013 vs 2009: 5.22 ± 0.37 × 10’2vs 4.04 ± 0.41 × 1012; P<0.01) and hemoglobin (2013 vs 2009: 151.43 ± 12.12 g/L vs 135.94 ± 14.53 g/LP<0.01). The rate of abnormal white blood cell count was 71% in 2009, and 40% in 2013. the 2013 was statistically significant lower than 2009 by chi square difference test. However, the micronucleus frequency in 2013 was higher than in 2009 (2013 vs 2009:3.51 ± 1.88‰vs 2.86 ± 2.07‰), though with no significant statistically difference. The rate of abnormal MN frequency was 26% in 2009, and 51% in 2013, with no significant statistically difference.The methylation rate of LINE1 in 2013 was much higher than in 2009 (2013 vs 2009:51.31 ± 7.63% vs 45.10 ± 9.95%; P<0.01). But the promoter methylation rate in HMLH1 is lower than in 2009 (2013 vs 2009:11.45 ± 5.53% vs 9.28 ± 4.32%; P=0.081).In conclusion, benzene can induce genetic damage and hematologic toxicity in the current occupation health standard of our country, and benzene exposure can lead to the changes of overall methylation rate (LINE1), may also lead to the reduction in promoter methylation of environmental response genes, such as GSTP1, hMLH1, MGMT and RAS. The mutation in Methylation related gene could regulate the level of methylation of methylation, DNMT3A/3B mutations can lead to decreased LINE1 methylation rate, although not statistically significant. R882 mutations can lead to reduced LINE1 methylation and, with reduced white blood cell counts and increased MN frequency. The mutation was found in two sites of DNMT3A/3B (rs36012910 and rs2424913) were the protective factors for the body, but the methylation rate of LINE1 is lower. There was no correlation with two of them. To discover the effect of mutations in DNA methylation related gene with the regulation of methylation and health also need to further expand the sample size and the detection of the methylation sites. Follow up observation found that when workers exposed to benzene by high concentration changed into the very low (<6mg/m3) environment, peripheral hematologic toxicity gradually restored, but the genetic damage still exists, even becoming seriously than before. But with the change of the exposure dose, the level of DNA methylation may change too.