ObjectivesThe early stressful environment,recent stressful environment,and the hypothalamicpituitary-adrenal axis-related genetic variation can interactively affect adolescent depressive symptoms.However,the existing studies mainly focus on individual genes,and their specific interactive model is still inconsistent.Based on this,this study firstly aims to investigate the occurrences of childhood maltreatment,recent stressful life events,and depressive symptoms among Chinese adolescents;then examine the interaction between childhood maltreatment,recent stressful life events and HPA axis-related multilocus genetic variation on adolescent depressive symptoms,and its specific interactive model;explore which subtype of stressful environment has a more robust interaction effect;and test whether multilocus genetic variation is more predictive than single-locus genetic variation.To provide evidence for prevention and intervention strategies of adolescent depressive symptoms.MethodsUsing longitudinal data from Genetic and Environmental Mechanisms of Adolescent Depression,827 secondary vocational students(50.2%boys)were selected,and a total of 3 assessments were conducted for 2 consecutive years.At T1 time,adolescent childhood maltreatment,stressful life events,and depressive symptoms were assessed by the Childhood Trauma Questionnaire Short Form(CTQ-SF),the Adolescent Self-rating Life Event Checklist(ASLEC),and the Children’s Depression Inventory(CDI).Oral exfoliated cell samples were collected using the Oragene DISCOVER tool.The HPA axis-related multilocus genetic variation was generated according to the relationship with the HPA axis function and using the multilocus genetic profile score(MGPS)approach(including CRHR1 rs 110402,NR3C1 rs41423247,NR3C2 rs5522,and FKBP5 rs1360780).At T2 and T3 times,reassessment for childhood maltreatment and depressive symptoms.Statistical analysis was performed using SPSS 23.0 and Mplus 7.4 software.The statistical methods included descriptive analysis,independent sample t test,one-way analysis of variance(ANOVA),repeated measures analysis of variance(rANOVA),Pearson correlation analysis,and hierarchical linear modeling(HLM).Results(1)Among adolescents in vocational schools,the incidence rates of physical abuse,emotional abuse,physical neglect,emotional neglect and sexual abuse in adolescents were 7.1%,5.7%,25.3%,19.1%,and 4.0%,respectively.Moreover,at T1、T2 and T3 times,98.2%、94.7%and 93.1%of adolescents had stressful life events,respectively;28.9%、21.5%and 14.0%of adolescents had depressive symptoms,respectively.(2)After controlling for gender and age,childhood maltreatment,stressful life events,and HPA axis MGPS had no significant three-way interaction on adolescent depressive symptoms(R12<0.238,R22<0.249,b=0.002,SE=0.001,P=0.110),but the interaction between childhood maltreatment and stressful life events was significant(b=0.004,SE=0.001,P=0.001),and they interacted with each other in a cumulative manner.(3)Among the subtypes of stressful life events,only interpersonal stress could interact with childhood maltreatment and HPA axis MGPS on adolescent depressive symptoms R12<0.240,R22<0.249,b=0.01,SE=0.004,P=0.015).Among the subtypes of childhood maltreatment,only emotional neglect could interact with interpersonal stress and HPA axis MGPS on adolescent depressive symptoms(R12<0.169,R22<0.152,b=0.02,SE=0.01,P=0.009).Specifically,among adolescents with less susceptible alleles,interpersonal stress significantly positively predicted adolescent depressive symptoms when the level of childhood maltreatment or emotional neglect was low(P<0.001);When the level of childhood abuse or emotional neglect was high,interpersonal stress had a stronger positive predictive effect on adolescent depressive symptoms(P<0.001).Among adolescents with more susceptible alleles,interpersonal stress significantly positively predicted depressive symptoms in adolescents with higher levels of childhood maltreatment or emotional neglect(P<0.001).When the level of childhood abuse or emotional neglect was low,interpersonal stress had no significant predictive effect on adolescent depressive symptoms(P≥ 0.758).In this study,the significant E1×E2×G interaction does not conform to the integrated model,but all conform to the cumulative stress hypothesis.(4)The multilocus genetic variation in E1×E2×G were more predictive than single-locus genetic variation.ConclusionThe incidences of childhood maltreatment,stressful life events,and depressive symptoms among vocational students were high,and their overall status was not optimistic.We highlight that interpersonal stress,as a recent stressful environment in adolescents,can interact with childhood maltreatment or emotional neglect and the HPA axis-related multilocus genetic variation to influence adolescent depressive symptoms.Regardless of the number of susceptibility genes carried by adolescents,childhood maltreatment or emotional neglect and interpersonal stress all affected adolescent depressive symptoms in a cumulative stress manner.This study is the first study to examine the influence of childhood maltreatment,stressful life events,and HPA axis-related multilocus genetic variation on adolescent depressive symptoms,providing new evidence for understanding the mechanism of adolescent depressive symptoms. |