Genetic polymorphisms of metabolic enzyme genes associated with leukocyte mitochondrial DNA copy number in PAHs exposure workers

Abstract Background Polycyclic aromatic hydrocarbons (PAHs) exposure had been reported to be a risk factor of mtDNAcn in our early study. However, the effect of metabolic enzymes' genetic polymorphisms on mtDNAcn in PAHs‐Exposure workers has not been fully evaluated. Aim The aim of the study was to explore the effect of metabolic enzymes' genetic polymorphisms on mtDNAcn in PAHs‐Exposure. Methods and Results We investigated the effects of metabolic enzymes' genetic polymorphisms on mtDNAcn among 544 coke oven workers and 238 office staffs. The mtDNAcn of peripheral blood leukocytes was measured using the Real‐time quantitative polymerase chain reaction (PCR) method. PCR and restriction fragment length was used to detect five polymorphisms in GSTT1, GSTM1, GSTP1 rs1695, CYP2E1 rs6413432, and CYP2E1 rs3813867. The mtDNAcn in peripheral blood leukocytes was significantly lower in the exposure group than that in the control group (p < .001). The 1‐OHPYR had an increasing trend with the genotypes AA→AG → GG of GSTP1 rs1695 in the control group. Generalized linear model indicated that the influencing factors of mtDNAcn were PAHs‐exposure [β (95% CI) = −0.420 (−0.469, −0.372), p < .001], male [β (95% CI) = −0.058 (−0.103, −0.012), p = .013], and AA genotype for GSTP1 rs1695 [β (95% CI) = −0.051 (−0.095, −0.008), p = .020]. Conclusion The individuals carrying the AA genotype of GSTP1 rs1695 may have a lower mtDNAcn due to their weaker detoxification of PAHs.


| INTRODUCTION
Coke oven emissions (COEs) are generally derived from the incomplete combustion of organic matter and possess hazardous levels of fine particulate and polycyclic aromatic hydrocarbons (PAHs). Most chemicals of PAHs are classified as human carcinogens and the genotoxic potential of PAHs has been extensively studied. In addition to producing damage on the nuclear DNA, 1 PAHs have 40 to 90-fold higher affinity for mitochondrial DNA (mtDNA) than nuclear DNA and cause even higher level damage to mtDNA. 2 Several studies have found that PAHs may interfere mitochondrial biosynthesis and further alter mitochondrial DNA copy number (mtDNAcn). [3][4][5] Our previous study also found that mtDNAcn decreased with environmental PAHs-exposure. 6 Mitochondria play an important role in multiple cellular functions including oxidative phosphorylation, reactive oxygen species generation, calcium homeostasis, and apoptosis. Each mitochondrial contains 2-10 copies of mitochondrial DNA. Due to the lack of protective histones and DNA repair machinery, mtDNA is particularly more vulnerable to various kinds of environmental toxins. 7 The mtDNAcn will increase as compensation to early poison exposure, 8 however, when the increased mtDNAcn cannot maintain the mitochondria normal function, mitophagy will occur to remove dysfunctional ones, and the mtDNAcn will be reduced. 9 Therefore, mtDNAcn might be a sensitive and important target to the genotoxic of PAHs. Moreover, the alteration of mtDNAcn has also been found to be associated with various disease development. For example, alterations of mtDNAcn are interrelated with lung cancer risk 10 , and the decreased mtDNAcn may accelerate the aging process and causes age-related disorders. 11 After PAHs enter the human body, they are firstly metabolized to electrophilic active intermediates by phase Imetabolic enzymes, such as cytochrome P450 (CYP) monooxygenases. The metabolic intermediates may produce DNA adducts, leading to DNA mutations, alteration of gene expression, and even tumorigenesis. 12 Subsequently, the intermediates are converted into more polar and water-soluble products by phase II metabolic enzymes and then excreted from the body. 13 The phase II metabolic enzymes include glutathione S-transferases (GST), UDP glucuronyl transferases, NADPH quinone oxidoreductases, aldo-keto reductases, and epoxide hydrolases. Therefore, extensive polymorphism of metabolism enzymes genes may be one of the main reasons for the individual variable susceptibility to exogenous toxic substances. 14 Studies have shown that metabolic enzymes' genetic polymorphisms were associated with many health effects in PAHs exposure population. Whyatt et al. found that the CYP1A1 Mspl restriction site had higher DNA adduct levels among newborns with PAHs exposure. 15 One study suggested that the GSTM1 (−) was inversely associated with the DNA integrity in the men occupationally exposed to PAHs. 16 Our earlier study showed that GSTT1 (+) and GSTM1 (+) are the risk factors for oxidative stress in coke oven workers. 17 However, the influence of metabolic enzymes' genetic polymorphisms on mtDNAcn in PAHs-Exposure workers has not been studied yet. Therefore, we detected mtDNAcn to investigate the genotoxic of PAHs-exposure and screened GSTT1, GSTM1, GSTP1, and CYP2E1 gene to explore the role of metabolic enzymes genetic polymorphisms on mtDNAcn in PAHs-exposure.

| Study population and epidemiological data
A random cluster sampling method was adopted to enroll the subjects in this study. A total of 544 workers exposed to COEs for more than 1 year were recruited as the exposure group from the Henan Anyang Iron and Steel Group, Henan, China. Their workplaces were in five representative locations, including auxiliary production, office personnel, oven bottom, oven side, and oven top. Also, 238 healthy workers without a history of exposure to occupational poison were enrolled from the same region as the control group. Exposure-group subjects were selected based on the following inclusion criteria: (a) age from 18 to 60, (b) who worked in the coke plant and were occupationally exposed to COEs for more than 1 year, and (c) provided informed consent. Subjects were excluded if they had major organ function failure or a tumor, were pregnant, or were lactating. Except for the history of occupational exposure, the inclusion and exclusion criteria for the control group were similar to those in the exposure group.
Detailed information on general demographic characteristics, professional history, and biological samples from each participant was collected by trained interviewers. The blood was collected using Na 2 EDTA and heparin anticoagulants, and the urine (the end-of-work urine of the occupational exposure population and morning urine of the control group) was retained with 50 mL centrifuge tubes. The study protocol and consent form from all subjects were subjected to approval by the Ethics Committee of Zhengzhou University, China.
More detailed information is described in our previous study. 6

| Detection of genetic polymorphisms
The GSTT1 and GSTM1 genotype were determined as previously described. 20 PCR-restriction fragment length polymorphism (PCR-RFLP) method was used to detect other loci for genotyping, including GSTP1 rs1695, CYP2E1 rs6413432, and CYP2E1 rs3813867. 21 The primers are shown in Table 1. To ensure the accuracy of the experiment, the positive control, and the negative control were set both in the process of PCR and restriction enzyme digestion. What's more, 10 % of each gene polymorphic loci samples were selected randomly for repeat and the concordance was 100%. The representative gel pictures for PCR-RFLP were shown in Figure 1.

| Statistical analysis
Baseline survey data were entered using EpiData 3.1 software. All analyses were performed using SPSS 25.0 software (SPSS, Inc., Chicago, Illinois). The data of four OH-PAHs were converted by natural logarithm to satisfy normal distribution. After adjusting appropriate adjustments for gender, age (years), smoking status, drinking status, and BMI, covariance analysis was used to analyze the effects of the general characteristics and gene polymorphism on OH-PAHs or mtDNAcn. Multiple linear regression analyzed the trend of OH-PAHs change with mutant allele loci. The generalized linear model (GLM) analyzed the influencing factors of mtDNAcn by adjusting the smoking index, drinking status, and BMI. All statistical tests were twosided, and the level of statistical significance was set at α = .05.

| Individual characteristics and mtDNAcn
The cohort consists of 544 COEs-exposed workers as the expo-  (Table 2).
3.2 | The effects of gender, age, smoking, drinking, and BMI on mtDNAcn After adjusting appropriate adjustments of gender, age (years), smoking index, drinking status, and BMI, covariance analysis showed that no factors were related to the mtDNAcn (p > .05). However, the mtDNAcn of all layers had significant differences between the two groups (p < .001). As reported in our early study, 6 the results are shown in Table 3.

| Effects of genetic polymorphisms on 1-OHPYR
The genotype distribution for each genetic polymorphism locus did not deviate from the Hardy-Weinberg balance (p > .05; Table 4), and the allele frequencies were similar to those of Asians in the International Human Genome HapMap Project, suggesting the control samples had representativeness.
The differences in 1-OHPYR among metabolic enzyme genes polymorphisms are shown in Table 5. With the adjustment of the covariates affecting 1-OHPYR, covariance analysis showed that the 1-OHPYR in T A B L E 1 Primers sequences for gene polymorphism

Polymorphism
Primers sequences Note: Forward represents the upstream primer and reverse represents the downstream primer. a ALB is a control gene used in multiple PCR for GSTT1 and GSTM1.
non-deletion for GSTM1 was significantly higher than that in deletion in the exposure group (p = .024), the 1-OHPYR in AA for GSTP1 rs1695 was significantly higher than that in AG genotype in the control group (p = .044). After adjusting the covariates (gender, age, smoking index, drinking status, and BMI), the trend test of multiple linear regression revealed that the 1-OHPYR had an increasing trend with the genotypes AA ! AG ! GG in the control group (p = .013).

| Effects of genetic polymorphisms on mtDNAcn
As shown in Table 6, there were no statistically significant differences in mtDNAcn among different genotypes in loci of the metabolic enzyme genes. The mtDNAcn in AG + GG for GSTP1 rs1695 was slightly higher than that in the AA genotype in the exposure group (p = .077).

| The influencing factors on mtDNAcn
The influencing factors were screened by GLMs with mtDNAcn as the dependent variable, PAHs exposure, gender, age, GSTT1, GSTM1,  Table 7).

| DISCUSSION
Workers in coke oven plants are exposed to a wide variety of volatile organic compounds and particulates, especially PAHs. The PAHs and their metabolic intermediates might cause damage to the mitochondria. 23 Increasing evidence indicates that PAHs-exposure may relate The GSTP1 rs1695 is located on exon 5 of chromosome 11q13, and contains a wild-type G allele and mutant A allele. The transition of an A allele to G allele in GSTP1 rs1695 confers increased conjugating activity 2829 and may also with lower levels of genotoxicity in PAHs exposure. Our study found that the 1-OHPYR had an increasing trend with the genotypes AA ! AG ! GG of GSTP1 rs1695 in the control group. Moreover, our results firstly showed that AA for GSTP1 rs1695 was a risk factor for mtDNAcn in PAHs-exposure. Therefore, we inferred that the toxicity of PAHs may be influenced by GSTP1 rs1695 polymorphisms resulting in the different alteration of mtDNAcn.
Though we also analyzed the association between mtDNAcn and polymorphisms in GSTT1, GSTM1, there was no significant difference.
The possible reason is that mitochondrial copy number is affected by many factors, and the polymorphism of the metabolic enzyme gene has a modest effect on mtDNAcn. However, our result showed that the 2-OHNAP and 3-OHPHE in non-deletion for GSTM1 were significantly higher than that in deletion in the exposure group. This confirms that metabolic enzyme gene polymorphism may alter the toxicity of PAHs.
CYP2E1 is an important member of CYP450 system and is located on Homo sapiens chromosome 10q24.3. The CYP2E1 rs3813867 is a G/C polymorphism located at 1259 position in the 5 0 -flanking region and the CYP2E1 rs6413432 (T > A) located on intron 6, which can affect the CYP2E1 gene expression level and its enzyme activity. 30 Therefore, polymorphism of CYP2E1 may affect the activity of the enzyme express leading to individual differences in PAHs metabolism.
Nan et al. observed that CYP2E1 polymorphism was an important factor influencing the levels of 1-OHPYR and 2-OHNAP in urinary from coke oven workers. 31 In this study, there was no significant difference between OH-PAHs and CYP2E1 polymorphism, that may be because the PAHs are mostly activated by CYP1A and CYP1B. 32 Polymorphism of CYP2E1 may also influence the genotoxic of environmental toxins. Guang et al. 33 found that the CYP2E1 rs3813867 mutant allele was associated with higher micronuclei among benzene-exposed shoe workers. Jheneffer et al. 34 reported that alcoholics who heterozygous in the CYP2E1 rs3813867 showed higher DNA damage (tail length and olive tail moment). Jing et al. 35 found heterozygous in the CYP2E1 rs6413432 had shorter telomere lengths among benzene-exposed shoe workers. However, mtDNAcn change was not associated with CYP2E1 polymorphism in the present study.
In the study, we have analyzed a larger number of samples, which is the advantage of the study. However, several limitations of the present study need to be considered.  In 63 male healthy steel workers, Hou et al. 36 found that particulate matter exposure was positively associated with mtDNAcn. However, Hou et al. 37 observed decreased mtDNAcn in association with increased exposure to particulate matter in a repeated-measure study.
Wong et al. 38 and Wang et al. 39 also found that particulate matter exposure was linked with decreased mtDNAcn. Therefore, the relationship between particulate matter exposure and mtDNAcn needs to be further studied. Due to lack of data on particulate matter exposure, we did not analysis the effect of particulate matter on mtDNAcn and we would consider it in further study. Hence, further research is needed to address these questions.
In conclusion, the individuals carrying the AA genotype of GSTP1 rs1695 may have a lower mtDNAcn due to their weaker detoxification of PAHs.

ACKNOWLEDGMENTS
The authors express their gratitude to all the individuals who volunteered to participate in this study. Note: GLMs was used to analyze the influencing factors of mtDNAcn adjusted for smoking index, drinking status, and BMI. Abbreviation: PAHs, polycyclic aromatic hydrocarbons.