ObjectivesThe study aims to clarify the epigenetic mechanism of association between early parent-child separation and adolescent psychopathology.This study compares the possible effects of early parent-child separation pattern on telomere length and methylation level of stress-related genes(FKBP5 intron 5,NR3C1 promoter region,NR3C1 1F and OXTR exon 3)in adolescents to provide a basis for the prevention and intervention of the adverse effects of parent-child separation.MethodsIn December 2017,students from a local high school and a vocational technical college were recruited in rural areas of Fuyang,an area with intensive migrant work in Anhui.A total of 695 participants were included,with a mean age of 16.86(standard deviation,SD=1.52).The information was collected through questionnaire and interview,including demographic information,social economic status,adverse childhood experiences,parental warmth perception,as well as parent-child separation.Participants were classified into four categories:persistent parent-child separation,childhood parent-child separation,current parent-child separation and no parent-child separation experience according to exposure time of parent-child separation.The Mac Arthur Health&Behavior Questionnaire(HBQ)was used to assess internalizing and externalizing symptoms.DNA was collected by using buccal swabs for telomere length.BMI was calculated by height and weight,which was obtained from physical measurement.A total of 219 participants were recruited from the study for biological sample collection in December 2019.Buccal swabs were collected to detect telomere length,and 2ml saliva was collected to measure DNA methylation of stress-related genes(FKBP5 intron 5,NR3C1 promoter region,NR3C1 1F and OXTR exon 3).Corrected telomere attrition D was calculated by the formula(D=ρ(TL1-TLm1)-(TL2-TLm2);ρ=2r S1S2/(S12+S22)).The DNA methylation of the target gene was the arithmetic mean.One-way analysis of variance and chi-square tests were used to test the differences among different separation types on psychopathology,relative telomere length,as well as DNA methylation.Multivariate linear regression model was used to analyze the associations between early life parent-child separation and telomere length and psychopathological outcomes.We assessed the mediation effect of telomere length in the association between parent-child separation with psychopathology,using the“mediation”package in Stata statistical software.ResultsA total of 695 adolescents were recruited,consisted of 89 male and 606 female aged12.0-21.0 years.57.4%(399/695)of the participants experienced separation from both parents,among which 230 participants experienced persistent separation from both parents,and 136 participants separated from both parents during childhood.Compared to no separation group,there were more girls than boys in adolescents separated from parents.The second survey included 219 adolescents(14 male and 205 female).58.9%(129/219)of adolescents experienced parent-child separation,including 21.9%(48/219)in childhood parent-child separation,4.6%(10/219)in current parent-child separation and 32.4%(71/219)in persistent parent-child separation.Adolescents who reported childhood and persistent parent-child separation were found to have higher internalizing problems compared with those not separated from both parents(childhood separation vs.no separation:(3.28±0.46)vs.(3.13±0.47),F=0.936,P=0.002;persistent separation vs.no separation:(3.39±0.59)vs.(3.13±0.47),F=5.979,P<0.001).Similarly,childhood and persistent separation were associated with externalizing problems(childhood separation vs.no separation:(3.41±0.71)vs.(3.19±0.69),F=0.395,P=0.003;persistent separation vs.no separation:(3.46±0.74)vs.(3.19±0.69),F=0.740,P<0.001).However,current separation was not associated with psychopathology(internalizing problems:F=0.001,P=0.994;externalizing problems:F=17.013,P=0.452).Telomere length in both childhood and persistent parent-child separation were shorter compared with no separation group(childhood separation vs.no separation:(1.20±0.52)vs.(1.40±0.55),F=2.059,P<0.001;persistent separation vs.no separation:(1.21±0.55)vs.(1.40±0.55),F=0.163,P<0.001).However,there was no significant difference in telomere length between current separation group and no separation group(F=5.551,P=0.997).Telomere attrition was not associated with parent-child separation(F=0.495,P=0.686).DNA methylation of stress-related genes have no significantly association with parent-child separation(FKBP5 intron 5:F=0.196,P=0.899;NR3C1promoter region:F=1.421,P=0.239;NR3C1 1F:F=1.320,P=0.270;OXTR exon 3:F=0.168,P=0.918).Both childhood and persistent separation were associated with increased internalizing symptoms(childhood separation:β=0.13,95%CI:0.02,0.23;persistent separation:β=0.23,95%CI:0.14,0.31),increased externalizing symptoms(childhood separation:β=0.17,95%CI:0.03,0.32;persistent separation:β=0.23,95%CI:0.10,0.35),after adjusting for telomere length,age,sex,SES,BMI,adverse childhood experiences,maternal education background and parental warmth perception.Childhood or persistent separation has association with shorter telomere length(childhood:β=-0.16,95%CI:-0.26,-0.05;persistent:β=-0.13,95%CI:-0.22,-0.03),after adjusting for psychopathology,age,sex,SES,BMI,adverse childhood experiences,maternal education background and parental warmth perception.Mediation analysis showed that telomere length was a mediating factor for persistent parent-child separation and psychopathological symptoms.Telomere length was estimated to explain 15.2%and 12.7%of the total effect of separation on internalizing and externalizing symptoms.ConclusionAmong the adolescents in the areas with intensive migrant work in Anhui,childhood parent-child separation and persistent parent-child separation were significantly associated with internalizing and externalizing symptoms.Childhood and persistent parent-child separation were related to shorter telomere length,but not to DNA methylation of stress-related genes(FKBP5 intron 5,NR3C1 promoter region,NR3C11F and OXTR exon 3).Telomere length acts as a mediating factor accounting for the association between parent-child separation and psychopathological problems. |