Association Of Urinary Polycyclic Aromatic Hydrocarbon Metabolites With The Risk Of Cardiorespiratory Dysfunction In Community Residents And The Mediating Role Of Oxidative Damage Among Such Relationships

Polycyclic aromatic hydrocarbons(PAHs)are an important class of environmental contaminant that have an adverse impact on human health.In addition to carcinogenicity,numerous evidences suggest that long-term exposure to PAHs is also closely related to the occurrence and development of cardiopulmonary diseases.As reported by the World Health Organization,cardiopulmonary diseases have long been at the top of the mortality and the burden of disease in recent decades.Hence,in order to identify the high-risk population and carry out early prevention,we selected two indicators of lung function and 10-year risk of atherosclerotic cardiovascular disease(ASCVD)to evaluate the risk of early cardiopulmonary function lesion in the general population.In detail,lung function is not only an important index to characterize the health status of the respiratory system,but also closely related to the onset and death of cardiopulmonary diseases.Besides,as a relevant component of cardiovascular disease,ASCVD prediction models have been established in several countries to assess the risk of it in general populations.Correspondingly,exploring the association of PAH exposure with lung function,as well as with 10-year risk of ASCVD can help interpret the impact and extent of PAH exposure on cardiopulmonary lesions in community residents.Moreover,the mechanism of cardiopulmonary injury caused by PAH exposure is not completely clear,but previous research commonly believed that oxidative stress may play an important role in the health damage regarding to PAH exposure.Thus,we further explored the mediating role of oxidative damage in the association between PAH exposure and early cardiopulmonary impairment.On the other hand,due to the widespread existence of PAH emission sources,understanding the impact of common exposure sources on human internal exposure levels,is helpful to identify their important emission sources,and then reduce their exposure levels and corresponding health damage to a certain extent.Therefore,we conducted a study with 4812 subjects from the baseline of the Wuhan-Zhuhai cohort.The basic demographic characteristics and lifestyle relating to PAH exposures were collected by structured questionnaire surveys,and urinary levels of monohydroxy polycyclic aromatic hydrocarbon(OH-PAHs),the indicator for PAHs internal exposure,were also determined.Firstly,we investigated the effects of lifestyle exposure on urinary levels or species of OH-PAHs.Then,lung function test was performed,and the Prediction for ASCVD Risk in China(China-PAR)model was conducted to assess the risk of cardiopulmonary lesion for study population,and we further evaluated the relationships of urinary OH-PAHs with lung function,as well as with 10-year risk of ASCVD.After measuring the levels of oxidative damage biomarkers(8-hydroxy-2?-deoxyguanosine,8-oxod G and 8-iso-prostalandin-F2?,8-iso-PGF2?)in urine,interaction analysis and mediation analysis were used to explore the role of these oxidative damage indicators in the associations between PAH exposure and cardiopulmonary impairment,so as to provide clues for further exploration of the mechanism of caidiopulmonary damage by PAH exposures.In detail,this study included three parts as follows:Part?.Effects of lifestyle exposure on urinary levels of OH-PAHs in a community-based populationObjective:To investigate the single or combined effect of exposure sources in daily life on levels or species of urinary OH-PAHs.Methods:A total of 4812 participants from the Wuhan-Zhuhai cohort were recruited in this study.After excluding 720 subjects with potential occupational PAHs exposure,with missing data on basic information and urinary OH-PAHs concentrations,or with kidney diseases,4092 subjects were included.Detailed information on demographic data and lifestyle relating to PAH exposure sources was collected,including traffic exposure time,smoking status,cooking status and intake frequencies of foods.Concentrations of urinary10 kinds of OH-PAHs were determined by a gas chromatography-mass spectrometer,and the sum of all monohydroxy polycyclic aromatic hydrocarbons was expressed as?OH-PAHs.Based on the molecular weight of parent PAHs,these analytes were divided into two groups:sum of high molecular weight monohydroxy polycyclic aromatic hydrocarbon(?HMW OH-PAH)and sum of low molecular weight monohydroxy polycyclic aromatic hydrocarbons(?LMW OH-PAHs).Moreover,urinary?LMW OH-PAHs was classified into total hydroxynathalene(?OHNa),total hydroxyfluorene(?OHFlu)and total hydroxyphenanthrene(?OHPh).Then,covariance analysis was used to calculate and compare the adjusted geometric mean levels of urinary OH-PAHs under different exposure conditions with single exposure sources.Further,mixed linear models and covariance analysis were conducted to explore the associations of urinary metabolites levels with multiple PAH exposure sources.And standardized regression coefficients were also calculated to further compare the contributions of different sources to urinary OH-PAH levels.For analyses,sex,age,dietary PAH intake,smoking status,passive smoking status,self-cooking status,kitchen ventilation condition,traffic exposure time and region were adjusted.Result:In the present study,the mean age of all participants was 52.5 years,and the geometric mean of?OH-PAHs was 52.74?g/g Cr.Among them,the numbers of subjects from Wuhan and Zhuhai city were 2754 and 1338,respectively.Compared to participants from Zhuhai city,the proportions of males,smokers,passive smokers,and non-cookers were elevated in the Wuhan population.Besides,significantly increased time of traffic exposure and decreased levels of PAH in diet were observed in participants from Wuhan.Except for?HMW OH-PAH,urinary OH-PAHs levels in participants from Wuhan were obviously higher than those in Zhuhai(all P<0.05).The effects of single exposure source on urinary OH-PAHs showed that the adjusted geometric mean concentrations of urinary?HMW OH-PAH was significantly higher among self-cooking participants,when compared with nonself-cooking subjects.Smoker had increased levels of urinary?LMW OH-PAHs than non-smokers(all P<0.0001),especially for urinary?OHNa.Significantly increased adjusted levels of urinary?LMW OH-PAHs were observed among participant with higher exposure in diet,especially for urinary?OHNa and?OHFlu(all P<0.05).Participants who spent more than 30 min in traffic presented increased adjusted levels of urinary?OHFlu than participants with less than 30 min in traffic per day(P<0.05).Moreover,significantly decreased levels of urinary?LMW OH-PAHs were observed in participants with good kitchen ventilation(all P<0.05).The combined effects of any two exposure sources on urinary OH-PAH levels indicated that smoker always had markedly increased levels of urinary?LMW OH-PAHs and?OHNa compared to non-smokers,regardless of exposure from any other sources.Similarly,participant with both high levels of PAHs exposure in the diet and from any other sources showed significantly increased urinary?LMW OH-PAHs and?OHFlu levels compared with subjects in both low exposure groups.No significant combined effects of self-cooking and traffic exposure were observed on urinary OH-PAHs levels.The combined effects of any three exposure sources on urinary OH-PAHs concentrations showed that participants who coexposured to high levels of smoking,diet and cooking had highest the adjusted levels of urinary PAH metabolites,while those coexposured to high levels of cooking,diet and traffic showed a lowest adjusted level of urinary?LMW OH-PAHs.Individuals who coexposured to high levels of smoking,cooking and traffic showed a lowest level of urinary?HMW OH-PAH.In addition,the result of urinary?OHNa was similar to urinary?LMW OH-PAHs levels',whereas the results of urinary?OHFlu and?OHPh were consistent with that of urinary?HMW OH-PAH levels.The order of contributions to urinary?HMW OH-PAH levels was self-cooking>traffic exposure>cigarette smoking>poor kitchen ventilation>dietary intake.In contrast,urinary?LMW OH-PAHs was cigarette smoking>poor kitchen ventilation>dietary intake>traffic exposure>self-cooking.Among low melacular weight OH-PAHs,the rank to urinary?OHNa was consistent with urinary?LMW OH-PAHs.For urinary?OHFlu,the contributions in decreasing order were poor kitchen ventilation>dietary intake>traffic exposure>cigarette smoking>self-cooking.The order of contributions of PAH exposure sources to?OHPh levels was poor kitchen ventilation>traffic exposure>dietary intake>cigarette smoking>self-cooking.Conclusions:Different sources of PAH exposure could modify the species or the levels of urinary metabolites.Self-cooking was closely related to urinary?HMW OH-PAH levels,while smoking,dietary intake and traffic exposure were correlated to urinary?LMW OH-PAHs levels.In detail,cigarette smoking had a great influence on urinary levels of?OHNa,and higher levels of dietary intake was associated with?OHNa and?OHFlu.A long time spent in traffic was found to be a great contributor to urinary levels of?OHFlu.Moreover,good kitchen ventilation could efficiently reduce low molecular weight PAHs exposure among self-cooking participants.Part?.The relationship between urinary OH-PAHs and lung function in community population and the mediating role of oxidative damage among such relationship in a community populationObjective:To investigate the associations between different molecular weight of urinary OH-PAHs and lung function decrease,and further explore the mediating role of oxidative damage in such associations.Methods:Based on the study population from Part I,we further excluded 817 participants with respiratory system diseases,or with missing data on urinary 8-oxod G,8-iso-PGF2?and lung function index,then a total of 3275 participants were included.Lung function tests were measured for participants from the baseline and three years'follow-up,according to the American Thoracic Society recommendations.Urinary levels of 8-oxod G and 8-iso-PGF2?were determined by high performance liquid chromatography and enzyme-linked immunosorbent assay kits,respectively.The classification of urinary OH-PAHs was consistent with that from Part I.In the cross-sectional study,mixed linear models were conducted to explore the associations between urinary OH-PAHs,urinary 8-oxod G or 8-iso-PGF2?,and lung function index.The relationship between urinary OH-PAHs and lung function in non-smokers was estimated by sensitivity analysis.In addition,stratified analyses were employed to investigate the modification effects of sex,age,body mass index,smoking status,drinking status,regular physical activity and region on the correlations between urinary OH-PAHs and oxidative damage biomarkers.Then,the potential role of urinary 8-oxod G or 8-iso-PGF2?among the relationship of lung function decrease with PAH exposures was conducted by mediation effect models.Further in follow-up study,restrictive cubic spline models were used to analysis the relationships between the baseline urinary OH-PAHs levels,urinary oxidative damage biomarkers and lung function index three years later,respectively.Sex,age,height,weight,education level,smoking amounts,passive smoking status,drinking status,self-cooking status,regular physical activity and region were adjusted in all analyses.Result:The results from cross-sectional study showed negative dose-response relationships between urinary OH-PAHs and lung function index after adjusting for potential confounders.Each one-unit increase in log-transformed values of urinary?OH-PAHs,?HMW OH-PAH and?LMW OH-PAHs were associated with-20.89 m L(95%CI:-40.14 m L,-1.63 m L),-22.57 m L(95%CI:-40.27 m L,-4.86 m L)and-18.75 m L(95%CI:-37.53 m L,0.03 m L)(P=0.05)decrease in forced expiratory volume in one second(FEV₁),respectively(all P<0.05).Each one-unit increase in log-transformed value of urinary?OH-PAHs and?HMW OH-PAH were associated with-22.31 m L(95%CI:-43.62 m L,-1.24 m L)and-22.36 m L(95%CI:-43.97 m L,-0.75 m L)decrease in forced vital capacity(FVC)(all P<0.05).Further,the categorical variable analysis also showed significant monotonic FEV₁ decrease for urinary?OH-PAHs,?HMW OH-PAH and?LMW OH-PAHs,and FVC decrease for urinary?HMW OH-PAH(all P for trend<0.05).However,no significant associations between urinary OH-PAHs and FEV₁/FVC were observed.Moreover,similar results were observed among non-smokers.With adjustment for potential confounders,positively dose-response relationships between urinary OH-PAHs and oxidative damage biomarkers were discovered.Each one-unit increase in log-transformed values of urinary?OH-PAHs,?HMW OH-PAH and?LMW OH-PAHs were associated with 0.31(95%CI:0.25,0.36),0.18(95%CI:0.13,0.23)and 0.30(95%CI:0.24,0.36)increase in log-transformed urinary 8-oxod G levels,as well as with 0.31(95%CI:0.27,0.35),0.26(95%CI:0.22,0.29)and 0.30(95%CI:0.26,0.33)increase in log-transformed urinary 8-iso-PGF2?levels,respectively(all P<0.05).In the categorical variable model,we found significant monotonic increase of urinary 8-oxod G and 8-iso-PGF2?with urinary?OH-PAHs,?HMW OH-PAH and?LMW OH-PAHs(all P for trend<0.0001)levels.Furthermore,age,sex,body mass index,smoking status,drinking status,regular physical activity and regions all modified the associations between urinary OH-PAHs and oxidative damage levels(all P for interaction<0.0001).A significant negative dose-response relationship between urinary 8-oxod G level and FVC was found.Each 1-unit increase in log-transformed level of urinary 8-oxod G was significantly associated with a-17.51 m L(95%CI:-31.09 m L,-3.92 m L)decrease in FVC.The categorical variable analysis only showed a significantly negative dose-response relationship of urinary 8-oxod G with FVC(P for trend=0.034).No significant association between urinary 8-iso-PGF2?and lung function index was observed.No interaction effects between urinary OH-PAHs and oxidative damage biomarkers were found.In the mediation models,urinary 8-oxod G mediated 12.5%of the association between urinary?HMW OH-PAH and FVC decrease,while urinary 8-iso-PGF2?did not show a mediating role in such relationship.The value of FVC after three years'follow-up showed a downward trend with the increased level of baseline urinary?HMW OH-PAH,or urinary 8-oxod G,despite no statistically difference in the above relationship was observed.Conclusions:Urinary levels of OH-PAHs were negatively associated with lung function decrease,and positively related to oxidative damage.DNA oxidative damage may play an important role in the association between high molecular weight PAH exposure and lung function.More prospective studies are needed to verify this finding.Part?The relationship between urinary OH-PAHs and 10-year ASCVD risk in community population and the mediating role of oxidative damage among such relationship in a community populationObjective:To explore the associations between different molecular weight and species of urinary OH-PAHs and 10-year risk of ASCVD,and further evaluate the mediating role of oxidative damage in such relationship.Methods:Based on the study population in Part I,we further excluded 1000 participants whose age beyond the prediction model(35?74 years old)or with ASCVD,a total of3052 participants were remained.The China-PAR prediction model was used to assess the study participants'10-year risk of ASCVD.According to the percentage of 5%,all subjects were classified into low-risk(<5%)group and high-risk(?5%)group.The classification of urinary OH-PAHs was similar to which of Part I.Then,logistic regression models were used to investigate the associations of urinary OH-PAHs and 10-year ASCVD risk.Stratified analyses were further used to evaluate whether sex,age,body mass index,smoking status,drinking status,regular physical activity and region could modify such relationships,respectively.Logistic regression models and mixed linear models were used to investigate the association between urinary OH-PAHs and 10-year ASCVD risk,8-oxod G or 8-iso-PGF2?,respectively.Moreover,interaction effects and mediation analysis were used to assess the role of oxidative damage biomarkers in such relationship between urinary OH-PAHs and 10-year ASCVD risk.For analyses,sex,body mass index,education level,smoking amounts,passive smoking status,drinking status,self-cooking status,regular physical activity and region were adjusted in all models.Result:This study population included 668 subjects in high-risk group and 2384 subjects in low-risk group.Those in high-risk group of 10-year ASCVD were composed of a large proportion of males,older,more smokers,more drinkers and higher body mass index(all P<0.0001).Besides,the levels of fasting blood glucose,blood pressure,and lipid-related indicators were significantly higher in high-risk group when compared with those in low-risk group(all P<0.0001).With adjustment for these potential confounders,for each 1-unit increase in log-transformed urinary?OH-PAHs and?LMW OH-PAHs,there were 21.1%(95%CI:5.7%-39.1%)and 22%(95%CI:6.5%-40.2%)increased risk of 10-year ASCVD,and the categorical analysis presented a significantly positive dose-response relationship of?OH-PAHs,?LMW OH-PAHs with 10-year ASCVD risk(all P for trend<0.0001).Furtherly,among low molecular weight OH-PAHs,each 1-unit increase in log-transformed urinary?OHNa and?OHFlu were associated with 15.8%(95%CI:1.9%-31.7%)and 24.6%(95%CI:11.5%-39.2%)increase in 10-year ASCVD risk in the continuous variable model,respectively.Consistent with?LMW OH-PAHs,significantly positive dose-response relationships of?OHNa and?OHFlu with 10-year ASCVD risk were observed in the categorical variable models(all P for trend<0.001).Besides,sex,age,body mass index,cigarette smoking,alcohol drinking,physical activity and region could notably modify such relationships(all P for interaction<0.0001).No significant relationship between?HMW OH-PAH and 10-year ASCVD risk was found,either in the continuous or categorical variable model.Similar to the results from Part?,significantly positive dose-response relationships between different molecular weight or species of urinary OH-PAHs and oxidative damage biomarkers were observed(all P for trend<0.0001).After adjusting for potential confounders,each 1-unit increase in log-transformed urinary 8-oxod G was associated with a 13.1%(95%CI:4.1%-21.6%)increase in 10-year ASCVD risk,and the categorical variable model presented a significantly positive dose-response relationship.Each 1-unit increase in log-transformed urinary 8-iso-PGF2?was associated with a 16.0%(95%CI:1.5%-35.2%)increase in 10-year ASCVD risk,whereas no significant dose-response relationship of urinary 8-iso-PGF2?with 10-year ASCVD risk was observed.No interaction effects between urinary OH-PAHs and oxidative damage biomarkers were found.Mediation effect models showed that urinary 8-oxod G mediated 15.6%of the association between urinary?OH-PAHs and 10-year ASCVD risk.Particularly,urinary 8-oxod G mediated 14.5%of urinary?LMW OH-PAHs and 10-year ASCVD risk.Furtherly,among low molecular weight OH-PAHs,urinary 8-oxod G also played a mediated role of 12.6%and 10.6%in the associations of urinary?OHNa and?OHFlu with 10-year ASCVD risk,respectively.However,we did not find the potential role of urinary 8-iso-PGF2?among such relationships.Conclusions:Significant positive associations of urinary OH-PAHs with urinary 8-oxod G and 10-year risk of ASCVD were observed in a general urban adult population.Our results indicated that oxidative DNA damage pathway represented by urinary 8-oxod G maybe a possible mechanism underlying higher risk of ASCVD with low molecular weight PAHs exposure,especially for naphthalene and fluorene.