Exploring The Neural Mechanism Of Childhood Maltreatment Affecting Brain Structural And Functional Networks In Major Depressive Disorder

Part Ⅰ Brain Structures Mediate the Relationship between Childhood Maltreatment and DepressionBackground:Previous studies have identified that both childhood maltreatment(CM)and major depressive disorder(MDD)are associated with brain structural alterations.However,the exact links between brain structures,CM,and MDD remains unclear.We therefore aimed to clarify the relationship between these three factors,especially the influence of different CM types.Methods:A total of 115 MDD patients and 69 healthy controls(HC)were enrolled in this study.All subjects were assessed by the Childhood Trauma Questionnaire(CTQ)and multidimensional neuropsychological scales,and completed structural magnetic resonance imaging(sMRI)scans.Preprocessing and automatically segment of sMRI data were conducted by FreeSurfer.Surface area and average thickness measures of all available 68(34 left and 34 right)cortical regions and volumes of all 14 subcortical brain structures(7 left and 7 right)were obtained.Partial correlation analysis was used to investigate the relationship between bran structures and depressive symptoms,and CM,especially different types of CM.Mediation analysis was used to identify whether brain structural alterations could mediate the relationship between childhood maltreatment and depression.Results:Compared to HC subjects,MDD patients were more likely to experience severe CM of all aspects,including childhood physical abuse(CPA),emotional abuse(CEA),physical neglect(CPN),emotional neglect(CEN),as well as sexual abuse(CSA).And there were extensive brain structural differences between MDD patients and HC patients,cortical surface area was significantly increased in MDD patients,while average cortical thickness and subcortical volume were significantly decreased.Besides,depression and CM were both associated with widely brain structural measures in MDD patients,and had common and specific brain structural basis.Mediation analysis found that the surface area of left precuneus and left rostral middle frontal gyrus(MFG)played key roles in mediating the relationship between CM and depressive symptoms.Specifically,the surface area of left precuneus significantly mediated the relationship between CEA and depressive symptoms,and the average thickness of right fusiform significantly mediated the relationship between CPN and depressive symptoms.Conclusions:This study provides convincing evidence for the relationship between CM,brain structures and depression,and for the first time explored the effects of different types of CM on brain structural alterations and the mediating role of on depression.The results confirmed that cortical surface area and average thickness in MDD patients could mediated the association between CM and depressive symptoms.Particularly,different types of CM could influence depression through different brain structures.These results further validate that brain structural alterations play an important role in the pathogenesis of neuropsychiatric diseases.Furthermore,our findings also add weight to the hypothesis that CM may be associated with a distinct subtype of MDD that might require special attention,care,and treatment.Part Ⅱ Neural Effects of Child Maltreatment on Large-scale Brain Networks in DepressionBackground:CM,as one of the major risk factor of depression,have been found to be associated with the alterations of brain structure and function in MDD patients.However,the neurobiological mechanism by which CM influences MDD at the network level is still unclear.Methods:A total of 141 participants,including 43 healthy control without childhood maltreatment(HCNCM),13 healthy control with childhood maltreatment(HCCM),35 major depressive disorder without childhood maltreatment(MDDNCM)and 50 major depressive disorder with childhood maltreatment(MDDCM),were enrolled.All subjects underwent a comprehensive neuropsychological battery and resting-state functional magnetic resonance imaging(rs-MRI)scans.Depression severity was assessed with the 17-items Hamilton Depression Scale(HAMD-17).The Childhood Trauma Questionnaire(CTQ)provided screening for a history of CM.We used the Power atlas to partition the brain of each participant into 10 well-established large-scale resting-state networks(RSNs).A one-way ANOVA test was used to determine the group comparisons of network metrics,and two-way ANOVA was used to investigate the interaction effects between CM experience and MDD symptoms on RSNs.The canonical correlation analysis(CCA)was used to link HAMD-17 subscores and CTQ subscores,to RSNs network variables.Then,mediation analyses were conducted to estimate the directional influence of brain network variables and CM to severity of depression.Lastly,we used support vector machine(SVM)models to classify MDD subgroups and HC group.Results:MDDCM patients and HCCM subjects had suffered serious CM experience than MDDNCM and HCNCM,especially in emotional abuse,physical neglect and emotional neglect.As the disease progressed,the cohesiveness of cingulo-opercular network(CON)presented obvious declining trends,while the others presented an overall trend of increasing first and then decreasing.Similarly,the connectedness of the ten networks was initially increasing and then declining.Besides,we found the network CCA mode and HAMD-17 CCA mode was significantly correlated(canonical correlation:r=0.783,p=0.002).The CTQ CCA mode and HAMD-17 CCA mode was also significantly correlated(canonical correlation:r=0.486,p<0.001).Interestingly,exploratory mediation analyses revealed CON,which showed the associations with both CTQ subscores and HAMD-17 subscores,could mediate the association between CTQ scores and HAMD-17 levels in all subjects.The conjunction analysis found there were total 12 connections between the neural correlates of the CTQ scales and subscales and HAMD-17 scales and subscales on the 10 RSNs.The SVM model revealed that the 12 functional connections represented higher capacity to discriminate disease spectrum(area under the ROC curve between 0.88-0.91).Conclusions:The present study is among the first detailed investigations of the network basis of CM and MDD,using the multivariate patient symptom-brain network association methods.These results deepen our understanding of the neuroimaging mechanisms of CM and MDD,and provide new insight into diagnostic of depression.Further,CCA can be employed more generally to investigate multivariate correlation profiles in other psychiatric disorders.Part Ⅲ Dynamic Functional Connectivity Alterations in Depression and the Correlation with Childhood MaltreatmentBackground:Traditional rs-fMRI studies,based on the assumption that intrinsic fluctuations remain relatively stable throughout the scan,have been widely used to measuring abnormalities in spontaneous brain activity and brain networks in neuropsychiatric disorders,including depression.Recently,dynamic alterations of functional connectivity have been suggested to better reflect the functional capacity of nervous system,and may serve as biomarkers for disease.However,the dynamic functional connectivity(DFC)alterations of MDD and the influence of CM are not clear.Methods:We evaluated 183 subjects matched for age and sex,including HCNCM(n=49),HCCM(n=20),MDDNCM(n=48),MDDCM(n=66).All subjects completed the multidimensional neuropsychological assessment and rs-fMRI scan.DFC was constructed using the sliding-window approach based on rs-fMRI data.Clustering analysis was used to assess the reoccurring functional connectivity patterns(states)of DFC.Partial correlation analysis was used to investigate the relationship between temporal properties of DFC state and depressive symptoms and CM.Results:Firstly,dynamic analysis suggested two different connectivity ’States’across the entire group:a less frequent,integrated state with stronger connected functional inter-network components,State 1,and a more frequent,segregated brain state characterized by the predominance of intra-network connections,State 2.In MDD patients,the State 2 occurred 9%more often than in HCs.In MDDNCM group,the separated state 2 had highest frequency,while integrated state 1 had lowest frequency among the four groups.Further,the mean dwelled time(MDT)of integrated State 1 was the shortest in MDDNCM,and the longest of segregated State 2,compared to other groups.The number of transitions(NT)showed an obvious decrease trend with the change of MDD and CM spectrum,and the MDDCM group had the lowest NT.Partial correlation analysis demonstrated that the NT of states was negatively correlated with childhood physical abuse,the MDT of State 1 was positively correlated with childhood emotional abuse,the fractional windows(FW)was significantly correlated to childhood sexual abuse.In addition,CM experience of MDD patients can affect different inter-and intra-network under States 1 and State 2.Conclusions:This study is the first to investigate DFC of MDD and CM spectrum population.We demonstrated that the temporal properties of DFC states altered in MDD patients,and the CM experience could influence these alterations.The results suggest that the more severe the childhood emotional abuse,the longer the MDT of State 1.The more severe the childhood physical abuse,the less frequent the transitions between the two states.The more severe the childhood sexual abuse,the higher proportion of state 1 and the lower proportion of state 2.Therefore,we considered the DFC approach,especially the temporal properties of DFC state,may be an objective biomarker for monitoring the influence of CM experience on MDD.Further studies on DFC alterations could help to better understand the progressive dysfunction of networks of MDD influenced by CM experience.Part Ⅳ Identification of MicroRNA-9 Linking the Effects of Childhood Maltreatment on Depression Using Amygdala ConnectivityBackground:CM is regarded as an important risk factor for MDD.However,the neural links corresponding to the process of early CM experience producing brain alterations and then leading to depression later remain unclear.Methods:To explore the neural basis of the effects of CM on MDD and the potential role of microRNA-9(miR-9)in these processes,we recruited 40 unmedicated MDD patients and 34 HC subjects to complete rs-fMRI scans and peripheral blood miR-9 tests.The neural substrates of CM,miR-9,and depression,as well as their interactive effects on intrinsic amygdala functional connectivity(AFC)networks were investigated in MDD patients.Two-step mediation analysis was separately employed to explore whether AFC strength mediates the association among CM severity,miR-9 levels,and depression.A SVM model of machine learning was used to distinguish MDD patients from HC group.Results:MDD patients showed higher miR-9 levels that were negatively correlated with CM scores and depressive severity.Overlapping effects of CM,miR-9,and depressive severity on bilateral AFC networks in MDD patients were primarily located in the prefrontal-striatum pathway and limbic system.The connection of amygdala to prefrontal-limbic circuits could mediate the effects of CM severity on the miR-9 levels,as well as the impacts of miR-9 levels on the severity of depression in MDD patients.Furthermore,the SVM model,which integrated miR-9 levels,CM severity,and AFC strength in prefrontal-limbic regions,had good power in differentiating MDD patients from HC group(accuracy 85.1%).Conclusions:MiR-9 may play a crucial role in the process of CM experience-produced brain changes targeting prefrontal-limbic regions and that subsequently leads to depression.The present neuroimaging-epigenetic results provide new insight into our understanding of MDD pathophysiology.