Genetic And Environmental Contributions To Regional Cortical Surface Area In Adolescents:a MRI Twin Study

Objective: Cortical surface area measures are considered to befunctionally relevant and particular in etiology, development, andbehavioral correlates compared with other size characteristics. we utilizebivariate twin methods to identify global and regionally specific geneticfactors,which influence cortical surface areas particularly duringadolescence.Methods： A cross-sectional twin design was used. We recruitedadolescent twins(age:12-18);Intelligence testing was performed using theChina-Wechsler Intelligence Scale for Children (C-WISC);All participantsdivided handness in Edingburgh Handedness Inventory; Sagittal T1-weighted brain images were acquired on GE3.0T MRI scanner.Wecalculated Regional Cortical Surface Area(total surface area、12lobarcortical regions and66subregions of interest) by Freesurfer softwarepackage and the Desikan-Killiany atlas；the blood sample was collected toidentify zygosity of twins. Bivariate twin analysis(using both total andregional surface area measures) of structural equation modeling (SEM) estimate the proportion of phenotypic variance of Regional CorticalSurface Area due to genetic and environmental influences. This approachallows for the estimation of ’unadjusted’ and ’adjusted’ genetic andenvironmental effects on a specific Regional of interest(ROI), whichincludes and doesn’t include those genetic and environmental effects sharedwith total surface area and identify global and regionally specific geneticand environmental factors influence cortical surface areas, particularlyduring adolescence.Results: The heritability of total surface area was substantial,with nearzero estimates of shared and unique environmental effects (MZcorrelation=0.90, DZ correlation=0.33;A=0.91[95%confidence interval{CI}=0.63;0.95]; C=0.01[95%CI=0.00;0.28]; E=0.08[95%CI=0.05;0.16]). we also found moderate heritability for regional surface areas beforeadjustment.Under models with ACE influences on12lobar surface areas,bilateral frontal areas heritabilities were the highest(left frontal areas：MZcorrelation=0.88, DZ correlation=0.28;A=0.90[95%{CI}=0.65;0.94];C=0.00[95%CI=0.00;0.25]; E=0.10[95%CI=0.06;0.20] right frontalareas：MZ correlation=0.89, DZ correlation=0.26;A=0.91[95%{CI}=0.68;0.95]; C=0.00[95%CI=0.00;0.23]; E=0.09[95%CI=0.05;0.17]) andbilateral medial temporal surface areas heritabilities were the lowest(leftmedial temporal surface area：MZ correlation=0.63, DZ correlation=0.56;A=0.13[95%{CI}=0.00;0.73]; C=0.47[95%CI=0.00;0.72]; E=0.40[95% CI=0.24;0.63] right medial temporal surface area：MZ correlation=0.73,DZ correlation=0.47;A=0.40[95%{CI}=0.00;0.82]; C=0.30[95%CI=0.00;0.72]; E=0.30[95%CI=0.17;0.51]). Under a model containing Aand E influences on total surface area and A, C, and E influences on eachregion, the average heritability across all regions after adjusting for totalsurface area was0.23, a49%reduction compared with the average estimatewithout adjustment for total surface area. Under a model with A and Eeffects estimated for both total surface area and ROI area,50%lobars andsome ROIs(26/66) estimates of heritability remained significant afteradjustment for total surface area.Conclusions: Although the heritability of all regions estimates inadolescents were not exactly equal to middel-age men, for both groups thetendency of heritability changed in the same degree in most regions.Bothglobal and regionally specific genetic factors influence cortical surfaceareas during adolescence also.