The Association Of Prenatal Exposure To PM_2.5 And PM₁ With Preterm Birth And Role Of Maternal LINE-1 Methylation In Guangzhou,China

BackgroundPreterm birth?PTB?is an important risk factor for infant mortality and morbidity.In recent years,many epidemiological studies have observed a positive association of prenatal exposure to air particulate matters with the risk of PTB.However,there are some scientific questions that need further study.First,most previous studies were conducted to investigate the effects of PM₁₀ and PM_2.5,and very few studies were conducted to assess the impacts of extreme fine particulate matters such as PM₁ which may have more toxic effects.Second,even though many previous studies have attempted to identify the susceptible windows,the findings remain controversial.The time scales of susceptible exposure window used in previous studies mostly were pregnancy trimester and gestational months,while more delicate time scales like gestational weeks were rarely used.Third,the assessment methods of prenatal exposure to air pollutants needs to be further improved.Most previous studies directly used environmental data from monitoring sites,and few considered individual exposure.Fourth,the biological mechanisms of air pollution leading to PTB have not been totally understood.Several previous studies suggested that methylation alternation of genes such as long interspersed nucleotide element-1?LINE-1?may be an important biological mechanism of air particulate matters inducing PTB.However,there is no direct epidemiological evidence.ObjectiveWe aimed to accurately assess prenatal exposure to ambient particulate matters(PMs,PM_2.5 and PM₁),estimate the effects of exposure to PMs with different sizes during pregnancy on PTB,identify the susceptible exposure windows,and further illustrate the roles of LINE-1methylation in the associations between PMs and PTB.MethodsThe Birth Cohort Study on Prenatal Environments and Offspring Health?PEOH?has been ongoing since 2016 in Guangzhou,China.Guangzhou Panyu Central Hospital was the study setting,the largest hospital in Panyu district,Guangzhou.Pregnant women were initially recruited as study subjects from the outpatient department of obstetrics if they fulfill the following criteria:?1?gestational weeks from 1 to 13;?2?aged 18 to 50 years;?3?the following diseases were not present:chronic diseases such as diabetes and heart disease,psychiatric disease,and other serious diseases.A total of 4,928 pregnant women were recruited and followed up during their delivery in the hospital,650 participants were lost to follow-up due to factors,such as the relocation,and other reasons.A total of 4,278?86.8%?pregnant women were successfully followed up,7stillbirths and 78 twins were excluded,43 participants with missing key variable information were excluded,and 49 participants with missing exposure level of environmental pollutants were excluded,and a total of4101 cases were finally obtained.Each individual weekly exposure to PM_2.5 from 3 months before pregnancy to childbirth was assessed using a spatiotemporal land use regression?ST-LUR?model and a individual indoor and outdoor time-activity model,and the weekly PM₁ exposure was estimated by employing a generalized additive model?GAM?.A distributed lag nonlinear model incorporated with a Cox proportional hazard model?DLNM-Cox model?was applied to assess the effect of weekly-specific maternal PMs exposure on PTB risk,and identify susceptible exposure windows.A nested case-control study was established based on the PEOH cohort study data and PTB outcomes.Maternal LINE-1 methylation levels?%5mC?were tested using bisulfite-PCR pyrosequencing.A multiple-linear regression model was employed to investigate the associations between PMs exposure and maternal LINE-1methylation levels.We also evaluated the associations between maternal LINE-1 methylation levels and PTB risk by using a logistic regression model.All the analyses were two-sided test,P<0.05 was the significance level,and all the statistical analyses were used R software.ResultsA total of 4,101 subjects were included in this study,among which234 were PTB,with an average age of 32.4 years and an average gestational age of 38.6 weeks.The incidence rate of PTB was 5.7%?234/4,101?.The result for the ST-LUR model on PM_2.5 exposure assessment showed that R² was 88.86%,RMSE was5.53?g/m³.The weekly average concentrations of PM_2.5 and PM₁ during the entire study period were 32.07?g/m³ and 27.93?g/m³,respectively.The risk of PTB was positively associated with PM_2.5 exposure during the 13th to 20th gestational weeks,and the strongest association was observed during 16th gestational week,compared with the first quartile?Q₁?of PM_2.5,a significantly highest risk of PTB was found for the fourth quartile?Q₄?,and the hazard risk?HR?was 1.18?95%CI:1.04-1.35?.The risk of PTB was positively associated with PM₁ exposure during the 12th to 20th gestational weeks,and the strongest association was observed during 16th gestational week,compared with the Q₁ of PM_2.5,a significantly highest risk of PTB was found for the Q₄,and HR was 1.20?95%CI:1.03-1.39?.Maternal blood LINE-1 methylation level had significantly negative associations with average PM₁ concentrations during 12th to 20th gestational weeks??=0.86%5mC for per 10?g/m³,P=0.034?,but no statistically significant in PM_2.5.5 exposure.The risk of PTB was negatively associated with maternal blood LINE-1 methylation level?OR=0.97 for per1%5mC,95%CI:0.87-1.09?,but the associations were not statistically significant.Conclusion?1?The ST-LUR model can accurately evaluate the exposure level of air particulate matters during pregnancy;?2?Higher maternal prenatal exposure(>P₇₅)to PM_2.5 and PM₁ were associated with increased risk of PTB,and the effects of PM₁ on PTB was higher than that of PM_2.5.?3?The 13th to 20th and 12th to 20th gestational weeks might be the susceptible exposure window of PM₁ and PM_2.5,respectively.?4?Maternal LINE-1 methylation levels decreased might be an underlying epigenetic mechanism of PMs increasing PTB risk.