Effect Of Polycyclic Aromatic Hydrocarbons On Telomere Damage And Its Influencing Factors In Coke Oven Workers

Polycyclic aromatic hydrocarbons(PAHs)are the main components of coke oven emissions,which are human carcinogens(G1)and can cause occupational lung cancer in coke oven workers.Telomeres are DNA-protein complexes at the end of the human chromosome,and protect chromosome integrity with telomere binding proteins,telomere length can reflect the early damage of chromosomes.This research intends to investigate the association of three types of biomarkers: PAHs exposure,telomere length,and genetic susceptibility in coke oven workers.Objective1.To investigate the correlation between relative telomere length(RTL)of peripheral blood leucocytes and the indicators of external exposure and urinary metabolism,and analyze their influencing factors,and to explore the exposed biomarkers of telomere damage.2.To explore the molecular mechanism of telomere injury and discover the susceptible population by combining the m RNA expression levels of telomere regulation genes and the functional polymorphism loci in the telomere-related protein genes,c GAS/STING signaling pathway genes and coding mi RNA genes.Materials and Methods1 Study population Exposure group was from 544 workers exposed to coke oven emissions for more than one year in coke plant of He'nan.238 controls were healthy population without the exposure of occupational PAHs and other poison in the same period.Professional investigators collected the basic information for the subjects,mainly including general demographic characteristics and occupational history.2 Research methods HPLC method was used to determine four kinds of urinary OH-PAHs,soxhlet extractor method was to detect the concentration of coke oven emissions in the air,and the concentrations of 16 PAHs were determined by HPLC.Real-time fluorescent quantitative PCR method was applied to determine the RTL in human peripheral blood leukocytes DNA.q PCR method was to detect the m RNA expression levels of telomere-associated genes,Genotyping for the selected functional and susceptible SNP loci was performed using the flight mass spectrometry.3 Statistical analysis Data were input using Epi Data 3.1 software,and statistical analyses used SPSS21.0 software.For quantitative data,to select statistical analysis methods according to the distribution type of data.Qualitative data used chi-square test to compare the differences between groups.After appropriate adjustment of covariates,covariance analysis was used to analyze the influencing factors of urinary metabolic indicators and RTL.Spearman correlation analysis was used to analyze the correlation between RTL in peripheral blood leucocytes and the indicators of external exposure and urinary metabolism.After adjusting the covariates appropriately,covariance analysis was used to compare the RTL differences among different genotypes,multiple linear regression was used to analyze the trend of RTL with the number of mutant alleles.Haplotype was analyzed by Haploview and PHASE2.0.2 bioinformatics software.PLINK1.07 software was used to analyze the association between polymorphic loci and the m RNA expression levels of their corresponding genes.All statistical tests were two-sided,and the level of statistical significance was set at ?=0.05.Results1 Occupational exposure conditions of coke oven workers and its influencing factorsThe concentrations of each component for 16 PAHs showed the increasing trend in the order of furnace bottom,furnace side and furnace top,but the concentrations of only naphthalene,anthracene and benzo(k)fluoranthene had statistically significant differences in three sampling places(P<0.05).The total PAHs concentration ranged from 4.90 to 699.38 mg/L.There were significant differences in total cumulative exposure,1-hydroxypyrene,1-hydroxynaphthalene,2-hydroxynaphthalene and3-hydroxy-phenanthroline between the two groups,respectively(Z=22.093,P<0.001;Z=13.479,P<0.001;Z=3.240,P=0.001;Z=9.048,P<0.001;Z=13.752,P<0.001).Further covariance analysis showed that the concentrations of urinary1-hydroxypyrene in men(4.63±1.14 pg/?g Cre)and smokers(4.75±1.13 pg/?g Cre)were significantly higher than those in women(3.88±1.00 pg/?g Cre)and non-smokers(4.19±1.12 pg/?g Cre)in the exposure group(F=15.071,P<0.001;F=5.006,P=0.026).For the 1-hydroxynaphthalene,that of smokers(4.30±1.10 pg/?g Cre)was significantly higher than that in nonsmokers(3.86±1.26 pg/?g Cre)in the exposure group(F=7.822,P=0.006).To the urinary 2-hydroxynaphthalene,that of smokers(4.89±0.96 pg/?g Cre)was higher than that of non-smokers(4.16±0.97 pg/?g Cre)in the exposure group(F=37.930,P<0.001),the control group was the same result,smokers(4.05±1.15 pg/?g Cre)>non-smokers(3.17±0.82 pg/?g Cre)(F=11.40,P=0.001).The concentration of urinary 3-hydroxy-phenanthrene in men(3.12±1.08pg/?g Cre)was higher than that in women(2.49±0.87 pg/?g Cre)(F=13.305,P<0.001).2 Influencing factors of RTL in human peripheral blood leucocytesRank sum test showed that peripheral blood DNA RTL in exposure group0.75(0.51,1.08)were significantly shorter than that of the control group1.05(0.76,1.44)(Z=7.692,P<0.001).After data conversion,covariance analysis found the adjusted RTL in the older age group 5.39(95%CI,5.16-5.62)was significantly longer than that of the lower age group 4.74(95%CI,4.56-4.90)(F=13.385,P<0.001),that of smoking group 5.24(95%CI,4.99-5.48)was also significantly longer than that in non-smoking group 4.95(95%CI,4.84-5.05)in the control group(F=4.432,P=0.037).Spearman correlation analysis showed that total cumulative exposure,1-hydroxypyrene,2-hydroxynaphthalene and 3-hydroxy-phenanthrene was a negative correlation with RTL(r=-0.307,P<0.001;r=-0.212,P<0.001;r=-0.110,P=0.02;r=-0.251,P<0.001).3 The effects of telomere-related genetic polymorphisms and its m RNA expression levels on RTLEight of the 29 SNPs in these genes were found to have an effect on RTL.After adjusting the covariate of affecting RTL,covariance analysis showed that RTL of individuals carrying the mutation homozygous AA(4.86±0.96)in TERT rs2736109 locus was longer than that of the wild homozygous GG(4.53±0.85)in the exposure group(P=0.024),and that of heterozygous AG individuals in this locus was shorter than that of wild homozygous GG(5.14±0.67)in the control group(P=0.001).In the exposure group,RTL of the insertion mutation homozygous CC genotype(4.81±0.81) for TERT rs3215401 polymorphism locus was significantly longer than that of the wild homozygous deletion genotype(4.54±0.82)(P=0.014).For the TERT rs2736100 locus,we found that RTL of heterozygous GT individuals(4.63±0.80)was longer than that of wild homozygous TT(4.45±0.83)in the exposure group(P=0.028).For the TNKS rs1055328 locus,RTL of individuals carrying the mutation homozygous CC(5.02±0.71)and heterozygous CG(5.08±0.68)were longer than that of wild homozygous GG(4.54±1.01)in the control group,respectively(P=0.021;P=0.005).For the rs9329202 locus in TNKS gene,RTL of heterozygous AG individuals(4.50±0.86)was shorter than that of wild homozygous GG(4.72±0.79)in the exposure group(P=0.030).In the control group,RTL of heterozygous CT genotype(4.90±0.67)for TEP1 rs1760897 polymorphism locus was significantly shorter than that of the wild homozygous CC genotype(5.28±0.60)(P=0.030),that of mutation homozygous CC genotype(5.30±0.53)for TEP1 rs1760903 polymorphism locus was longer than that of the wild homozygous TT genotype(5.00±0.75)(P=0.021),that of mutation homozygous TT genotype(5.27±0.52)for TEP1 rs1760904 polymorphism locus was longer than that of the wild homozygous CC genotype(4.98±0.75)(P=0.032).Trend test showed that RTL had an increasing trend in the order of genotype-/-?-/C?CC for the TERT rs3215401 polymorphism locus in the exposure group(Ptrend=0.048).To the TERT rs2736100 polymorphism locus,RTL showed an increasing trend along with the order of genotype TT,GT and GG in the exposure group(Ptrend=0.044).For the TEP1 rs938886 polymorphism locus,RTL showed an increasing trend according with the order of CC,CG and GG(Ptrend=0.014).Haplotype analysis showed that SNPs in TERT gene combined into three haplotypes according to the sequence of rs3215401 and rs2735940,covariance analysis found that RTL of DT/CT(4.32±0.72),DC/DC(4.48±0.79)and DC/CT(4.57±0.85)diplotye individuals was significantly shorter than that of CT/CT diplotye(4.37±0.94)in the exposure group(P=0.011;P=0.002;P=0.004;P=0.011).TNKS genetic polymorphism loci formed three haplotype in the order of rs17734024 and rs1055328,we found that RTL of individuals carrying GC/AC diplotye(4.47±0.81)was shorter than that of GG/GG diplotye(4.76±0.86)in the exposure group (P=0.026),and those of GC/GC(5.11±0.74),GC/GG(5.03±0.71)and GG/AC(5.23±0.55)diplotye were significantly longer than that of GG/GG diplotye(5.14±0.67)in the control group,respectively(P=0.005;P=0.011;P=0.006).Two haplotype blocks were composed of SNPs in TEP1 gene locus,we found that RTL of CT/GC(4.90±0.75)diplotye individuals was significantly shorter than that of GC/GC diplotye(5.14±0.70)in the control group for the four haplotypes formed according to the order of rs938886 and rs1713449 in the Block1(P=0.011),and those of CT/CT(4.97±0.76)and CT/TC(4.84±0.78)diplotye were shorter than that of TC/TC diplotye(5.29±0.51)in the control group for the four haplotypes formed in the order of rs1760904 and rs1760903,respectively(P=0.012;P=0.001)After the statistical analysis,we found that the m RNA expression levels of TPP1?TERF1 and TERF2 genes in the exposure group were significantly lower than those in the control group(Z=8.396,P<0.001;Z=15.605,P<0.001;Z=2.679,P=0.007),and the m RNA expression levels of POT1 gene in the exposure group was significantly higher than that of the controls(Z=1.983,P=0.047).Spearman correlation analysis showed that the m RNA expression levels of TPP1 and TERF2 genes was a negative correlation with RTL in the exposure group(r=-0.136,P=0.001;r=-0.103,P=0.016).After adjusting the covariate,TERF2 gene m RNA expression level increased along with the increase of minimum allele C of TERF2rs153045 locus in the exposure group(?=0.100,P=0.023).4 The association between cGAS/STING signal pathway genetic polymorphisms and RTLOnly one of the 8 SNPs in these genes was found to have an effect on RTL.After adjusting the covariate,covariance analysis showed that RTL of the heterozygous-/A genotype(4.42±0.85)for c GAS rs34413328 deletion polymorphism locus was significantly shorter than that of the wild homozygous AA genotype(4.65±0.82)in the exposure group,and the RTL showed a decreasing trend in the order of genotype AA ?-/A ?-/-(Ptrend=0.008).SNPs in a c GAS gene formed four haplotype according to the order of rs610913 and rs34413328,covariance analysis showed the RTL of AA/AA(4.71±0.74)and AA/CA(4.67±0.86)diplotye individuals was significantly longer than that of AA/CD diplotye(4.42±0.83)in the exposure group, that of AA/CA diplotye individuals(5.14±0.67)was significantly longer than that of AA/CD diplotye(4.90±0.77)in the control group(P=0.016).5 The association of coding mi RNA genetic polymorphisms and RTLWe found that six of the 15 SNP sites in the coding mi RNA genes had an effect on RTL.After adjusting the covariate of affecting RTL,covariance analysis showed that RTL of individuals carrying the heterozygous AG(5.06±0.71)in mi R-210rs11246190 locus was longer than that of the wild homozygous AA(4.88±0.79)in the control group(P=0.026).For the mi R-145 rs353291 locus,we found that RTL of mutation homozygous GG individuals(4.67±0.85)was longer than that of wild homozygous AA(4.51±0.783)in the exposure group(P=0.042).For the mi R-30 a rs763354 locus,RTL of individuals carrying the mutation homozygous AA(5.26±0.58)was longer than that of wild homozygous GG(4.92±0.75)in the control group(P=0.018).In the control group,RTL of insertion mutation homozygous TT genotype(5.12±0.71)for mi R-30 a rs111456995 polymorphism locus was longer than that of the wild homozygous deletion genotype(4.81±0.73)(P=0.038).For the rs2222722 locus in coding mi R-30 a gene,RTL of heterozygous CT individuals(4.66±0.83)was longer than that of wild homozygous TT(4.40±0.90)in the exposure group(P=0.008),and RTL of mutation homozygous CC genotype(4.87±0.73)and heterozygous CT genotype(4.95±0.78)were significantly shorter than that of wild homozygous TT(5.25±0.57)in the control group,respectively(P=0.007;P=0.010).For the mi R-197 rs1889470 polymorphism locus,RTL of mutation homozygous AA individuals(4.60±0.82)was longer than that of wild homozygous GG(4.29±1.04)in the exposure group(P=0.046).Trend test showed that RTL had an increasing trend in the order of genotype AA and AG for the mi R-210 rs11246190 polymorphism locus in the control group(Ptrend=0.026).To the mi R-145 rs353291 polymorphism locus,RTL showed an increasing trend along with the order of genotype AA,AG and GG in the exposure group(Ptrend=0.029).For the mi R-30 a rs763354 polymorphism locus,RTL showed an increasing trend according with the order of GG,AG and AA in the control group(Ptrend=0.017).For mi R-30 a rs2222722 polymorphism locus,RTL had a decreasing trend in the order of genotype TT,CT and CC in the control group(Ptrend=0.017).The results of haplotype analysis showed that SNPs in coding mi R-210 gene combined into seven haplotypes according to the sequence of rs7935908,rs7395206 and rs12364149,covariance analysis found that RTL of TTC/TTC(4.93±0.69),TTC/CCC(5.12±0.80),TTC/CCG(5.07±0.56)and CCG/CCG(5.19±0.57)diplotye individuals was significantly longer than that of TTC/TCC diplotye(4.37±0.94)in the control group(P=0.011;P=0.002;P=0.004;P=0.011).Conclusions1.Coke oven emissions could cause the shortening of human peripheral blood leukocytes RTL.RTL had a tendency to become shorter with the increasing concentrations of 1-hydroxypyrene,2-hydroxynaphthalene and3-hydroxy-phenanthrene.These three indicators may be used as biomarkers of PAHs exposure.2.In combination with regulation mechanism of telomere length,the m RNA expression levels of shelterin gene TPP1,TERF1,TERF2 and POT1 may be related to telomere length.TERF2 gene m RNA expression level may be affected by the genetic variation of TERF2 rs153045.3.The genetic variation of TERT rs2736109,TERT rs3215401,TERT rs2736100,TNKS rs9329202,c GAS rs34413328,mi R-145 rs353291,mi R-30 a rs2222722 and mi R-197 rs1889470 affected the telomere length in peripheral blood leukocytes of exposure workers.These may be susceptible loci of telomere damage induced by coke oven emissions.