Advertisement

Schizophrenia polygenic risk during typical development reflects multiscale cortical organization

  • Author Footnotes
    1 Montreal Neurological Institute, McGill University, Montreal, QC, Canada
    ,
    Author Footnotes
    2 Department of Psychiatry, Psychotherapy and Psychosomatics, Psychiatric Hospital, University of Zurich, Switzerland
    ,
    Author Footnotes
    ∗ equal contribution
    Matthias Kirschner
    Correspondence
    Corresponding authors: Matthias Kirschner, Department of Psychiatry, Psychotherapy and Psychosomatics, Psychiatric Hospital, University of Zurich, Switzerland,
    Footnotes
    1 Montreal Neurological Institute, McGill University, Montreal, QC, Canada
    2 Department of Psychiatry, Psychotherapy and Psychosomatics, Psychiatric Hospital, University of Zurich, Switzerland
    ∗ equal contribution
    Search for articles by this author
  • Author Footnotes
    1 Montreal Neurological Institute, McGill University, Montreal, QC, Canada
    ,
    Author Footnotes
    3 Institute of Neuroscience and Medicine (INM-1), Forschungszentrum Jülich, Jülich, Germany
    ,
    Author Footnotes
    ∗ equal contribution
    Casey Paquola
    Footnotes
    1 Montreal Neurological Institute, McGill University, Montreal, QC, Canada
    3 Institute of Neuroscience and Medicine (INM-1), Forschungszentrum Jülich, Jülich, Germany
    ∗ equal contribution
    Search for articles by this author
  • Author Footnotes
    1 Montreal Neurological Institute, McGill University, Montreal, QC, Canada
    Budhachandra S. Khundrakpam
    Footnotes
    1 Montreal Neurological Institute, McGill University, Montreal, QC, Canada
    Search for articles by this author
  • Author Footnotes
    1 Montreal Neurological Institute, McGill University, Montreal, QC, Canada
    ,
    Author Footnotes
    4 Institute of Psychology, Faculty of Social Sciences, Tartu, Estonia
    Uku Vainik
    Footnotes
    1 Montreal Neurological Institute, McGill University, Montreal, QC, Canada
    4 Institute of Psychology, Faculty of Social Sciences, Tartu, Estonia
    Search for articles by this author
  • Author Footnotes
    1 Montreal Neurological Institute, McGill University, Montreal, QC, Canada
    Neha Bhutani
    Footnotes
    1 Montreal Neurological Institute, McGill University, Montreal, QC, Canada
    Search for articles by this author
  • Benazir Hodzic-Santor
    Search for articles by this author
  • Author Footnotes
    2 Department of Psychiatry, Psychotherapy and Psychosomatics, Psychiatric Hospital, University of Zurich, Switzerland
    Foivos Georgiadis
    Footnotes
    2 Department of Psychiatry, Psychotherapy and Psychosomatics, Psychiatric Hospital, University of Zurich, Switzerland
    Search for articles by this author
  • Author Footnotes
    1 Montreal Neurological Institute, McGill University, Montreal, QC, Canada
    Noor B. Al-Sharif
    Footnotes
    1 Montreal Neurological Institute, McGill University, Montreal, QC, Canada
    Search for articles by this author
  • Author Footnotes
    1 Montreal Neurological Institute, McGill University, Montreal, QC, Canada
    Bratislav Misic
    Footnotes
    1 Montreal Neurological Institute, McGill University, Montreal, QC, Canada
    Search for articles by this author
  • Author Footnotes
    1 Montreal Neurological Institute, McGill University, Montreal, QC, Canada
    Boris Bernhardt
    Footnotes
    1 Montreal Neurological Institute, McGill University, Montreal, QC, Canada
    Search for articles by this author
  • Author Footnotes
    1 Montreal Neurological Institute, McGill University, Montreal, QC, Canada
    Alan C. Evans
    Footnotes
    1 Montreal Neurological Institute, McGill University, Montreal, QC, Canada
    Search for articles by this author
  • Author Footnotes
    1 Montreal Neurological Institute, McGill University, Montreal, QC, Canada
    Alain Dagher
    Correspondence
    Corresponding authors: Alain Dagher, Montreal Neurological Institute, McGill University, Montreal, QC, Canada,
    Footnotes
    1 Montreal Neurological Institute, McGill University, Montreal, QC, Canada
    Search for articles by this author
  • Author Footnotes
    ∗ equal contribution
    1 Montreal Neurological Institute, McGill University, Montreal, QC, Canada
    2 Department of Psychiatry, Psychotherapy and Psychosomatics, Psychiatric Hospital, University of Zurich, Switzerland
    3 Institute of Neuroscience and Medicine (INM-1), Forschungszentrum Jülich, Jülich, Germany
    4 Institute of Psychology, Faculty of Social Sciences, Tartu, Estonia
Open AccessPublished:August 23, 2022DOI:https://doi.org/10.1016/j.bpsgos.2022.08.003

      Abstract

      Background

      Schizophrenia is widely recognized as a neurodevelopmental disorder. Abnormal cortical development in otherwise typically developing children and adolescents may be revealed using polygenic risk scoring for schizophrenia (PRS-SCZ).

      Methods

      We assessed PRS-SCZ and cortical morphometry in typically developing children and adolescents (3–21 years, 46.8% female) using whole genome genotyping and T1-weighted MRI (n=390) from the Pediatric Imaging, Neurocognition and Genetics (PING) cohort. We contextualised the findings using (i) age-matched transcriptomics, (ii) histologically-defined cytoarchitectural types and functionally-defined networks, (iii) case-control differences of schizophrenia and other major psychiatric disorders derived from meta-analytic data of six ENIGMA working groups including a total of 12,876 patients and 15,670 controls.

      Results

      Higher PRS-SCZ was associated with greater cortical thickness, which was most prominent in areas with heightened gene expression of dendrites and synapses. PRS-SCZ related increases in vertex-wise cortical thickness were mainly distributed in association cortical areas, particularly the ventral attention network, while relatively sparing koniocortical type cortex (i.e. primary sensory areas).The large-scale pattern of cortical thickness increases related to PRS-SCZ mirrored the pattern of cortical thinning in schizophrenia and mood-related psychiatric disorders derived from the ENIGMA consortium. Age group models illustrate a possible trajectory from PRS-SCZ associated cortical thickness increases in early childhood towards thinning in late adolescence, the latter resembling the adult brain phenotype of schizophrenia.

      Conclusions

      Collectively, combining imaging-genetics with multi-scale mapping, our work provides novel insight into how genetic risk for schizophrenia impacts the cortex early in life.

      Introduction

      Schizophrenia is a multifaceted and heritable psychiatric disorder that is widely recognized to have a neurodevelopmental origin (
      • Birnbaum R.
      • Weinberger D.R.
      Genetic insights into the neurodevelopmental origins of schizophrenia.
      ,
      • Insel T.R.
      Rethinking schizophrenia.
      ,
      • Rapoport J.L.
      • Giedd J.N.
      • Gogtay N.
      Neurodevelopmental model of schizophrenia: update 2012.
      ). Abnormal brain development likely predates the onset of clinical symptoms, which typically emerge in early adulthood (
      • Kessler R.C.
      • Angermeyer M.
      • Anthony J.C.
      • DE Graaf R.
      • Demyttenaere K.
      • Gasquet I.
      • et al.
      Lifetime prevalence and age-of-onset distributions of mental disorders in the World Health Organization’s World Mental Health Survey Initiative.
      ). Genome-wide association studies (GWAS) support this hypothesis by showing schizophrenia-related genes are involved in multiple neurodevelopmental processes (
      • Birnbaum R.
      • Weinberger D.R.
      Genetic insights into the neurodevelopmental origins of schizophrenia.
      ,
      • Ripke S.
      • Neale B.M.
      • Corvin A.
      • Walters J.T.R.
      • Farh K.-H.
      • et al.
      Schizophrenia Working Group of the Psychiatric Genomics Consortium
      Biological insights from 108 schizophrenia-associated genetic loci.
      ). These genes may affect brain development, leading to vulnerability to environmental effects, and have been suggested to contribute to atypical cortical morphology, as previously observed in cohorts with a schizophrenia diagnosis (
      • Rapoport J.L.
      • Giedd J.N.
      • Gogtay N.
      Neurodevelopmental model of schizophrenia: update 2012.
      ).
      Childhood and adolescent brain development involves dynamic and complex structural changes that are shaped by genetic and environmental factors (
      • Shaw P.
      • Greenstein D.
      • Lerch J.
      • Clasen L.
      • Lenroot R.
      • Gogtay N.
      • et al.
      Intellectual ability and cortical development in children and adolescents.
      ,
      • Raznahan A.
      • Shaw P.
      • Lalonde F.
      • Stockman M.
      • Wallace G.L.
      • Greenstein D.
      • et al.
      How Does Your Cortex Grow?.
      ,
      • Khundrakpam B.S.
      • Lewis J.D.
      • Kostopoulos P.
      • Carbonell F.
      • Evans A.C.
      Cortical Thickness Abnormalities in Autism Spectrum Disorders Through Late Childhood, Adolescence, and Adulthood: A Large-Scale MRI Study.
      ,

      Park B, Bethlehem RA, Paquola C, Larivière S, Rodríguez-Cruces R, Vos de Wael R, et al. (2021): An expanding manifold in transmodal regions characterizes adolescent reconfiguration of structural connectome organization ((L. Q. Uddin, T. E. Behrens, & B. D. Fulcher, editors)). eLife 10: e64694.

      ,
      • Sydnor V.J.
      • Larsen B.
      • Bassett D.S.
      • Alexander-Bloch A.
      • Fair D.A.
      • Liston C.
      • et al.
      Neurodevelopment of the association cortices: Patterns, mechanisms, and implications for psychopathology.
      ). Longitudinal neuroimaging studies have consistently reported a global increase in cortical volume, thickness, and surface area that typically peaks in late childhood and is followed by decreases in adolescence (
      • Raznahan A.
      • Shaw P.
      • Lalonde F.
      • Stockman M.
      • Wallace G.L.
      • Greenstein D.
      • et al.
      How Does Your Cortex Grow?.
      ,
      • Sydnor V.J.
      • Larsen B.
      • Bassett D.S.
      • Alexander-Bloch A.
      • Fair D.A.
      • Liston C.
      • et al.
      Neurodevelopment of the association cortices: Patterns, mechanisms, and implications for psychopathology.
      ,
      • Tamnes C.K.
      • Herting M.M.
      • Goddings A.-L.
      • Meuwese R.
      • Blakemore S.-J.
      • Dahl R.E.
      • et al.
      Development of the Cerebral Cortex across Adolescence: A Multisample Study of Inter-Related Longitudinal Changes in Cortical Volume, Surface Area, and Thickness.
      ). At the same time, regional maturational trajectories are heterochronous, whereby sensory areas mature earlier than transmodal cortex (
      • Sydnor V.J.
      • Larsen B.
      • Bassett D.S.
      • Alexander-Bloch A.
      • Fair D.A.
      • Liston C.
      • et al.
      Neurodevelopment of the association cortices: Patterns, mechanisms, and implications for psychopathology.
      ,
      • Giedd J.N.
      • Blumenthal J.
      • Jeffries N.O.
      • Castellanos F.X.
      • Liu H.
      • Zijdenbos A.
      • et al.
      Brain development during childhood and adolescence: a longitudinal MRI study [no. 10].
      ,
      • Gogtay N.
      • Giedd J.N.
      • Lusk L.
      • Hayashi K.M.
      • Greenstein D.
      • Vaituzis A.C.
      • et al.
      Dynamic mapping of human cortical development during childhood through early adulthood.
      ), shaping large-scale patterns of cortical differentiation (

      Paquola C, Bethlehem RA, Seidlitz J, Wagstyl K, Romero-Garcia R, Whitaker KJ, et al. (2019): Shifts in myeloarchitecture characterise adolescent development of cortical gradients ((J. I. Gold, T. Satterthwaite, K. Mills, & B. Fulcher, editors)). eLife 8: e50482.

      ).
      The neurodevelopmental hypothesis of schizophrenia posits that cortical maturation is perturbed, producing widespread cortical abnormalities (
      • van Erp T.G.M.
      • Walton E.
      • Hibar D.P.
      • Schmaal L.
      • Jiang W.
      • Glahn D.C.
      • et al.
      Cortical Brain Abnormalities in 4474 Individuals With Schizophrenia and 5098 Control Subjects via the Enhancing Neuro Imaging Genetics Through Meta Analysis (ENIGMA) Consortium.
      ). Differences in cortical morphometry are consistently reported across different stages and clinical phenotypes of the schizophrenia spectrum (
      • Wannan C.M.J.
      • Cropley V.L.
      • Chakravarty M.M.
      • Bousman C.
      • Ganella E.P.
      • Bruggemann J.M.
      • et al.
      Evidence for Network-Based Cortical Thickness Reductions in Schizophrenia.
      ,
      • Kirschner M.
      • Shafiei G.
      • Markello R.D.
      • Makowski C.
      • Talpalaru A.
      • Hodzic-Santor B.
      • et al.
      Latent Clinical-Anatomical Dimensions of Schizophrenia.
      ,
      • Jalbrzikowski M.
      • Hayes R.A.
      • Wood S.J.
      • Nordholm D.
      • Zhou J.H.
      • et al.
      ENIGMA Clinical High Risk for Psychosis Working Group
      Association of Structural Magnetic Resonance Imaging Measures With Psychosis Onset in Individuals at Clinical High Risk for Developing Psychosis: An ENIGMA Working Group Mega-analysis.
      ). Investigating neurodevelopmental features of schizophrenia requires a departure from classic case-control designs, however. Alternatively, focusing on genetic risk enables us to investigate neuroanatomical correlates in a large population-based cohort of children and adolescents, without interacting disease-related factors (e.g. medication, chronicity). Recent work shows an effect of polygenic risk scores for schizophrenia (PRS-SCZ) on cortical morphometry (
      • French L.
      • Gray C.
      • Leonard G.
      • Perron M.
      • Pike G.B.
      • Richer L.
      • et al.
      Early Cannabis Use, Polygenic Risk Score for Schizophrenia, and Brain Maturation in Adolescence.
      ,
      • Liu B.
      • Zhang X.
      • Cui Y.
      • Qin W.
      • Tao Y.
      • Li J.
      • et al.
      Polygenic Risk for Schizophrenia Influences Cortical Gyrification in 2 Independent General Populations.
      ,
      • Neilson E.
      • Bois C.
      • Gibson J.
      • Duff B.
      • Watson A.
      • Roberts N.
      • et al.
      Effects of environmental risks and polygenic loading for schizophrenia on cortical thickness.
      ,
      • Neilson E.
      • Shen X.
      • Cox S.R.
      • Clarke T.-K.
      • Wigmore E.M.
      • Gibson J.
      • et al.
      Impact of Polygenic Risk for Schizophrenia on Cortical Structure in UK Biobank.
      ), though not necessarily grey matter volume (
      • Van der Auwera S.
      • Wittfeld K.
      • Homuth G.
      • Teumer A.
      • Hegenscheid K.
      • Grabe H.J.
      No association between polygenic risk for schizophrenia and brain volume in the general population.
      ,
      • Auwera SV der
      • Wittfeld K.
      • Shumskaya E.
      • Bralten J.
      • Zwiers M.P.
      • Onnink A.M.H.
      • et al.
      Predicting brain structure in population-based samples with biologically informed genetic scores for schizophrenia.
      ,
      • Reus L.M.
      • Shen X.
      • Gibson J.
      • Wigmore E.
      • Ligthart L.
      • Adams M.J.
      • et al.
      Association of polygenic risk for major psychiatric illness with subcortical volumes and white matter integrity in UK Biobank.
      ). Thus far, studies have centred almost exclusively on adult cohorts. Only one study has investigated adolescents (aged 12–21 years) and noted an association of PRS-SCZ with globally decreased cortical thickness amongst cannabis users (
      • French L.
      • Gray C.
      • Leonard G.
      • Perron M.
      • Pike G.B.
      • Richer L.
      • et al.
      Early Cannabis Use, Polygenic Risk Score for Schizophrenia, and Brain Maturation in Adolescence.
      ). Discerning neurodevelopmental aspects of genetic risk for schizophrenia requires investigation of younger cohorts.
      Understanding the relation of genetic risk for schizophrenia to neurodevelopment can be further enhanced by contextualising imaging-derived phenotypes of polygenic risk with maps of cortical organization. At the cellular level, a range of processes associated with healthy cortical development, such as synaptic pruning, dendritic arborization and intracortical myelination, are implicated in the development of schizophrenia and may produce regional cortical disruptions (
      • Feinberg I.
      Schizophrenia: Caused by a fault in programmed synaptic elimination during adolescence?.
      ,
      • Sellgren C.M.
      • Gracias J.
      • Watmuff B.
      • Biag J.D.
      • Thanos J.M.
      • Whittredge P.B.
      • et al.
      Increased synapse elimination by microglia in schizophrenia patient-derived models of synaptic pruning [no. 3].
      ,
      • Sprooten E.
      • O’Halloran R.
      • Dinse J.
      • Lee W.H.
      • Moser D.A.
      • Doucet G.E.
      • et al.
      Depth-dependent intracortical myelin organization in the living human brain determined by in vivo ultra-high field magnetic resonance imaging.
      ,
      • Wei W.
      • Zhang Y.
      • Li Y.
      • Meng Y.
      • Li M.
      • Wang Q.
      • et al.
      Depth-dependent abnormal cortical myelination in first-episode treatment-naïve schizophrenia.
      ). Recent advances in RNA sequencing (RNAseq) of post mortem brain tissue (
      • Zhu Y.
      • Sousa A.M.M.
      • Gao T.
      • Skarica M.
      • Li M.
      • Santpere G.
      • et al.
      Spatiotemporal transcriptomic divergence across human and macaque brain development.
      ) allow discernment of the relative contribution of cell-types to patterns of atypical cortical morphometry (
      • Seidlitz J.
      • Nadig A.
      • Liu S.
      • Bethlehem R.A.I.
      • Vértes P.E.
      • Morgan S.E.
      • et al.
      Transcriptomic and cellular decoding of regional brain vulnerability to neurogenetic disorders [no. 1].
      ,

      Writing Committee for the Attention-Deficit/Hyperactivity Disorder, Autism Spectrum Disorder, Bipolar Disorder, Major Depressive Disorder, Obsessive-Compulsive Disorder, and Schizophrenia ENIGMA Working Groups (2021): Virtual Histology of Cortical Thickness and Shared Neurobiology in 6 Psychiatric Disorders. JAMA Psychiatry 78: 47–63.

      ). More complex interactions of microstructure and function on regional vulnerability may be captured by the groupings of cortical areas into cytoarchitectural types and functional networks. Indeed, recent studies of schizophrenia (
      • Kirschner M.
      • Shafiei G.
      • Markello R.D.
      • Makowski C.
      • Talpalaru A.
      • Hodzic-Santor B.
      • et al.
      Latent Clinical-Anatomical Dimensions of Schizophrenia.
      ,
      • Shafiei G.
      • Markello R.D.
      • Makowski C.
      • Talpalaru A.
      • Kirschner M.
      • Devenyi G.A.
      • et al.
      Spatial Patterning of Tissue Volume Loss in Schizophrenia Reflects Brain Network Architecture.
      ) and high PRS-SCZ in healthy adults (
      • Cao H.
      • Zhou H.
      • Cannon T.D.
      Functional connectome-wide associations of schizophrenia polygenic risk.
      ) suggest differential sensitivity of histological-defined cytoarchitectural types (
      • García-Cabezas M.Á.
      • Hacker J.L.
      • Zikopoulos B.
      A Protocol for Cortical Type Analysis of the Human Neocortex Applied on Histological Samples, the Atlas of Von Economo and Koskinas, and Magnetic Resonance Imaging.
      ) and functional networks (
      • Yeo B.T.T.
      • Krienen F.M.
      • Sepulcre J.
      • Sabuncu M.R.
      • Lashkari D.
      • Hollinshead M.
      • et al.
      The organization of the human cerebral cortex estimated by intrinsic functional connectivity.
      ). Finally, population-level effects of schizophrenia and other major psychiatric disorders can be used to illustrate the concordance of genetic risk for schizophrenia with disorder-related neuroanatomical phenotypes. Specifically, it can be tested how the association between genetic risk of schizophrenia and cortical morphometry in children and adolescents relates to shared and divergent neuroanatomical abnormalities across psychiatric disorders (

      Writing Committee for the Attention-Deficit/Hyperactivity Disorder, Autism Spectrum Disorder, Bipolar Disorder, Major Depressive Disorder, Obsessive-Compulsive Disorder, and Schizophrenia ENIGMA Working Groups (2021): Virtual Histology of Cortical Thickness and Shared Neurobiology in 6 Psychiatric Disorders. JAMA Psychiatry 78: 47–63.

      ). Taken together, multiple scales of cortical organization can be utilized to provide a comprehensive description of the regional variations of an imaging-derived phenotype, such as genetic risk for schizophrenia.
      Here we address the relationship between PRS-SCZ and cortical organization in a large population-based cohort of typically developing children and adolescents (3 – 21 years) derived from the Pediatric Imaging, Neurocognition and Genetics (PING) study (
      • Jernigan T.L.
      • Brown T.T.
      • Hagler D.J.
      • Akshoomoff N.
      • Bartsch H.
      • Newman E.
      • et al.
      The Pediatric Imaging, Neurocognition, and Genetics (PING) Data Repository.
      ). We hypothesised that higher PRS-SCZ would be associated with atypical cortical morphometry (thickness, surface area and volume). Then, we aimed to better understand the effect of PRS-SCZ on cortical morphometry by comparing the observed spatial patterns to cell-type specific gene expression, cytoarchitectural and functional differentiation, and cortical abnormalities seen in major psychiatric disorders. Finally, we examined age-group specific variations of high PRS-SCZ on cortical morphometry across different neurodevelopmental stages.

      Material and Methods

      Subjects

      Neuroimaging, demographic and genetic data of typically developing children and adolescents were derived from the PING study (
      • Jernigan T.L.
      • Brown T.T.
      • Hagler D.J.
      • Akshoomoff N.
      • Bartsch H.
      • Newman E.
      • et al.
      The Pediatric Imaging, Neurocognition, and Genetics (PING) Data Repository.
      ). The PING dataset is a wide-ranging, publicly shared data resource comprising cross-sectional data from 1493 healthy subjects. Participants were recruited and evaluated in the greater metropolitan areas of Baltimore, Boston, Honolulu, Los Angeles, New Haven, New York, Sacramento, and San Diego using local postings and outreach activities. Written informed consent was given by parents for all PING subjects below the age of 18. Participants between the ages of 7 and 17 gave additional child assent, and all participants 18 or older directly gave written informed consent. Exclusion criteria and more information about the PING cohort are described elsewhere (
      • Jernigan T.L.
      • Brown T.T.
      • Hagler D.J.
      • Akshoomoff N.
      • Bartsch H.
      • Newman E.
      • et al.
      The Pediatric Imaging, Neurocognition, and Genetics (PING) Data Repository.
      ). After quality control of genomic and imaging data (see below) a total of 390 participants from the PING dataset were included in this study. The mean age was 12.10 (SD=4.77) with an age range from 3 to 21 years, and 46.8 % were female (for more details see Table S1).

      Genomic data

      Genomic data processing and calculation of polygenic risk scores followed a recent publication from Khundrakpam and colleagues (
      • Khundrakpam B.
      • Vainik U.
      • Gong J.
      • Al-Sharif N.
      • Bhutani N.
      • Kiar G.
      • et al.
      Neural correlates of polygenic risk score for autism spectrum disorders in general population.
      ). Specifically, 550,000 single nucleotide polymorphisms (SNPs) were genotyped from saliva samples using the Illumina Human660W-Quad BeadChip. Details on imputation and preprocessing can be found in the Supplementary Methods. After SNP imputation and preprocessing, 4,673,732 variants were available for calculation of polygenic scores. Participants were filtered to have at least 0.95 loadings to the European genetic ancestry factor (coded as “GAF_europe” in the PING dataset), resulting in 526 participants. To capture and quantify population structure, the same participants were used to calculate the 10 principal components across the variants, excluding areas in high LD with each other (--indep-pairwise 50 5 0.2) with Plink 2.
      The PRS-SCZ was trained using results from latest GWAS on schizophrenia at the time of analysis (
      • Ruderfer D.M.
      • Ripke S.
      • McQuillin A.
      • Boocock J.
      • Stahl E.A.
      • Pavlides J.M.W.
      • et al.
      Genomic Dissection of Bipolar Disorder and Schizophrenia, Including 28 Subphenotypes.
      ). The GWAS was filtered for having imputation quality over 90. Polygenic scores were calculated with PRSice 2.30e (
      • Choi S.W.
      • O’Reilly P.F.
      PRSice-2: Polygenic Risk Score software for biobank-scale data.
      ). Clumping of the data was performed using PRSice default settings (clumping distance=250kb, threshold r2=.1). To calculate the PRS-SCZ we used the GWAS hits (p<5x10 08) cut-off criterion. This resulted in 86 variants common to the base and target datasets (Table S2). The choice of the GWAS hits threshold was made to increase the specificity of observed gene-brain associations for schizophrenia and to minimize the genetic overlap with other psychiatric disorders such as bipolar disorder, which increases with lower PRS significance thresholds including more SNPs. To illustrate this influence of PRS significance thresholds on the genetic overlap between schizophrenia and bipolar disorder, we calculated PRS for bipolar disorder (PRS-BIP) applying exactly the same processing pipeline (Supplementary Methods). In the present sample, PRS-SCZ and PRS-BIP did not correlate (r=.063) using the GWAS hits threshold chosen here, whereas applying lower significance thresholds, such as p=.05 and p=.1, resulted in moderate correlations (r=.254 and r=.286, respectively; Figure S1).

      Image acquisition and pre-processing

      Details on image acquisition and pre-pre-processing are described elsewhere (
      • Jernigan T.L.
      • Brown T.T.
      • Hagler D.J.
      • Akshoomoff N.
      • Bartsch H.
      • Newman E.
      • et al.
      The Pediatric Imaging, Neurocognition, and Genetics (PING) Data Repository.
      ). The CIVET processing pipeline, (http://www.bic.mni.mcgill.ca/ServicesSoftware/CIVET, page 2.1) (

      Ad-Dab’bagh Y, Einarson D, Lyttelton O, Muehlboeck J-S, Mok K, Ivanov O, et al. (2006): The CIVET Image-Processing Environment: A Fully Automated Comprehensive Pipeline for Anatomical Neuroimaging Research. 1.

      ) was used to compute cortical thickness, surface area and cortical volume measurements at 81,924 regions covering the entire cortex and quality control (QC) was performed by two independent reviewers (see Supplementary Methods for details). After QC of the total 526 subjects that passed filtering for European genetic ancestry, a total sample size of 390 participants remained for all subsequent analyses.

      Statistical analyses

      Association between PRS-SCZ and cortical morphometry

      To identify the association of PRS-SCZ with vertex-wise cortical thickness, surface area and cortical volume, general linear models (GLM) were applied using the SurfStat toolbox (http://www.math.mcgill.ca/keith/surfstat/) (
      • Worsley K.
      • Taylor J.
      • Carbonell F.
      • Chung M.
      • Duerden E.
      • Bernhardt B.
      • et al.
      SurfStat: A Matlab toolbox for the statistical analysis of univariate and multivariate surface and volumetric data using linear mixed effects models and random field theory.
      ). Each cortical feature was modelled as:
      Ti = intercept + β1PRS-SCZ + β2Age + β3Age2 + β4PRS-SCZ*Age + β5PRS-SCZ*Age2 + β6Sex + β7PC1-10 + β8Scanner + β9BrainVolume + εi (Eq.1)where i is a vertex, PRS-SCZ is the polygenic risk score for schizophrenia, Age is years at the time of scan, PC10 are the first 10 principal components of genomic data to account for population stratification, ε is the residual error, and the intercept and the β terms are the fixed effects. Models with quadratic age terms were chosen as they fit the data better than models with only lower degree age terms (Supplementary Methods). For each cortical feature, vertex-wise t-statistics of the association with PRS-SCZ (β1PRS-SCZ) were mapped onto a standard surface. To assess the significance of PRS-SCZ effects on each of the three different cortical features, whole-brain correction for multiple comparisons using Random field theory (RFT) at cluster-level p≤.01 (
      • Hayasaka S.
      • Phan K.L.
      • Liberzon I.
      • Worsley K.J.
      • Nichols T.E.
      Nonstationary cluster-size inference with random field and permutation methods.
      ,
      • Worsley K.J.
      • Taylor J.E.
      • Tomaiuolo F.
      • Lerch J.
      Unified univariate and multivariate random field theory.
      ) was applied. Please note, only cortical thickness showed a significant association with PRS-SCZ after RFT correction (see Results, Figure 1); all subsequent analyses were thus restricted to cortical thickness only.
      Figure thumbnail gr1
      Figure 1Association of PRS-SCZ with cortical morphometry. A) Scatterplot shows variability of PRS-SCZ scores across age range within the selected cohort. B) Probability distributions show the variation in vertex-wise t-values of the association of PRS-SCZ with each cortical feature. Dashed lines represent the mean values of each cortical feature. Only cortical thickness was significantly shifted from 0. C) Unthresholded (top) and thresholded (bottom) maps after RFT correction (P<0.01) show the association of PRS-SCZ with cortical thickness. Unthresholded maps for surface area and cortical volume are Supplementary Figures 2-3.
      Cellular composition of the cortex and PRS-SCZ effects on cortical thickness
      We evaluated how the observed pattern of PRS-SCZ effects on cortical thickness relates to regional variations in the cellular compositions of the cortex. Given prior evidence (
      • Herculano-Houzel S.
      • Watson C.R.
      • Paxinos G.
      Distribution of neurons in functional areas of the mouse cerebral cortex reveals quantitatively different cortical zones.
      ) and histological validation (Supplementary Methods), we focused on components of the neuropil, namely glial cell processes, axons, dendritic trees, neuron-to-neuron synapses i.e. in cortical tissue other than cell bodies or blood vessels. Neuropil-related gene expression was calculated by combining tissue-level RNAseq [available online at http://development.psychencode.org/] (
      • Li M.
      • Santpere G.
      • Imamura Kawasawa Y.
      • Evgrafov O.V.
      • Gulden F.O.
      • Pochareddy S.
      • et al.
      Integrative functional genomic analysis of human brain development and neuropsychiatric risks.
      ) with gene lists of cell-types, based on single cell RNAseq (
      • Zhu Y.
      • Sousa A.M.M.
      • Gao T.
      • Skarica M.
      • Li M.
      • Santpere G.
      • et al.
      Spatiotemporal transcriptomic divergence across human and macaque brain development.
      ), and gene lists of neuron-compartments, based on gene ontologies (
      • Ashburner M.
      • Ball C.A.
      • Blake J.A.
      • Botstein D.
      • Butler H.
      • Cherry J.M.
      • et al.
      Gene Ontology: tool for the unification of biology [no. 1].
      ,
      The Gene Ontology Consortium
      The Gene Ontology Resource: 20 years and still GOing strong.
      ). Tissue-level RNAseq provided expression levels of 60,155 genes in 11 neocortical areas (
      • Li M.
      • Santpere G.
      • Imamura Kawasawa Y.
      • Evgrafov O.V.
      • Gulden F.O.
      • Pochareddy S.
      • et al.
      Integrative functional genomic analysis of human brain development and neuropsychiatric risks.
      ). The areas were cytoarchitecturally defined in each specimen, supporting precise mapping and comparison across individuals. Crucially, we selected 12 brain specimens from the PsychEncode developmental dataset that were age-matched to the PING imaging cohort (3-21 years), because gene expression differs substantially between children and adults (
      • Kang H.J.
      • Kawasawa Y.I.
      • Cheng F.
      • Zhu Y.
      • Xu X.
      • Li M.
      • et al.
      Spatio-temporal transcriptome of the human brain [no. 7370].
      ). Single-cell RNAseq (
      • Zhu Y.
      • Sousa A.M.M.
      • Gao T.
      • Skarica M.
      • Li M.
      • Santpere G.
      • et al.
      Spatiotemporal transcriptomic divergence across human and macaque brain development.
      ) provided specificity scores for each gene to glial cell types. For each type, we weighted the genes by the specificity score, then calculated the average across genes, across specimens and within area. For each neuron-compartment, we defined a list of marker genes using The Gene Ontology knowledgebase, then calculated the average expression of marker genes in each area and specimen. The annotated terms used were “neuron_to_neuron_synapse”, “dendritic_tree” and “main_axon”. Next, we mapped the 11 areas to the cortical surface and extracted area-average PRS-SCZ effects on cortical thickness. The cortical areas were visually matched to nearest parcel in a 200 parcel decomposition of the Desikan-Killiany, as performed in previous work (
      • Paquola C.
      • Seidlitz J.
      • Benkarim O.
      • Royer J.
      • Klimes P.
      • Bethlehem R.A.I.
      • et al.
      A multi-scale cortical wiring space links cellular architecture and functional dynamics in the human brain.
      ). Finally, we tested the spatial similarity of cell-type specific gene expression with PRS-SCZ effects on cortical thickness using product-moment correlations. Statistical significance was determined relative to random reassignment permutation tests (10,000 repetitions).
      Aggregation of PRS-SCZ effects on cortical thickness by cytoarchitectural type and functional network
      Given the hierarchical properties of cortical development (
      • Sydnor V.J.
      • Larsen B.
      • Bassett D.S.
      • Alexander-Bloch A.
      • Fair D.A.
      • Liston C.
      • et al.
      Neurodevelopment of the association cortices: Patterns, mechanisms, and implications for psychopathology.
      ) and that disease-related imaging-phenotypes are guided by different modes of cortical organization (
      • Kirschner M.
      • Shafiei G.
      • Markello R.D.
      • Makowski C.
      • Talpalaru A.
      • Hodzic-Santor B.
      • et al.
      Latent Clinical-Anatomical Dimensions of Schizophrenia.
      ,
      • Shafiei G.
      • Markello R.D.
      • Makowski C.
      • Talpalaru A.
      • Kirschner M.
      • Devenyi G.A.
      • et al.
      Spatial Patterning of Tissue Volume Loss in Schizophrenia Reflects Brain Network Architecture.
      ,

      Park B, Kebets V, Larivière S, Hettwer MD, Paquola C, Rooij D van, et al. (2021, November 2): Multilevel neural gradients reflect transdiagnostic effects of major psychiatric conditions on cortical morphology. bioRxiv, p 2021.10.29.466434.

      ), we sought to contextualize the PRS-SCZ effects on cortical thickness by cytoarchitectural types and intrinsic functional networks. A whole-cortex atlas of cytoarchitectural types was acquired [https://github.com/caseypaquola/DMN] (

      Paquola C, Garber M, Frässle S, Royer J, Tavakol S, Rodriguez-Cruces R, et al. (2021, November 22): The Unique Cytoarchitecture and Wiring of the Human Default Mode Network. bioRxiv, p 2021.11.22.469533.

      ), which reflects an intersection of the classic Von Economo atlas of cortical areas (

      von Economo CF, Koskinas GN (1925): Die Cytoarchitektonik Der Hirnrinde Des Erwachsenen Menschen. J. Springer.

      ,
      • Scholtens L.H.
      • de Reus M.A.
      • de Lange S.C.
      • Schmidt R.
      • van den Heuvel M.P.
      An MRI Von Economo - Koskinas atlas.
      ) with a recent re-analysis of Von Economo micrographs that categorized the areas according to type (
      • García-Cabezas M.Á.
      • Hacker J.L.
      • Zikopoulos B.
      A Protocol for Cortical Type Analysis of the Human Neocortex Applied on Histological Samples, the Atlas of Von Economo and Koskinas, and Magnetic Resonance Imaging.
      ). Cortical types synopsize degree of granularity, from high laminar elaboration in koniocortical areas, six identifiable layers in eulaminate III-I, poorly differentiated layers in dysgranular and absent layers in agranular.
      Functional networks were defined based on the Yeo atlas [https://github.com/ThomasYeoLab/CBIG] (
      • Yeo B.T.T.
      • Krienen F.M.
      • Sepulcre J.
      • Sabuncu M.R.
      • Lashkari D.
      • Hollinshead M.
      • et al.
      The organization of the human cerebral cortex estimated by intrinsic functional connectivity.
      ). The atlas reflects clustering of cortical vertices according to similarity in resting state functional connectivity profiles, acquired in 1000 healthy young adults. We assessed whether the PRS-SCZ effects on cortical thickness were stronger or weaker within each cortical class or functional network, relative to spin permutations that preserve spatial autocorrelation (
      • Alexander-Bloch A.F.
      • Shou H.
      • Liu S.
      • Satterthwaite T.D.
      • Glahn D.C.
      • Shinohara R.T.
      • et al.
      On testing for spatial correspondence between maps of human brain structure and function.
      ,
      • Burt J.B.
      • Helmer M.
      • Shinn M.
      • Anticevic A.
      • Murray J.D.
      Generative modeling of brain maps with spatial autocorrelation.
      ) (see Supplementary Methods for details).
      Pattern similarity between PRS-SCZ effects and cortical abnormalities in major psychiatric disorders
      We assessed whether cortical thickness differences of PRS-SCZ relate to patterns of cortical thickness abnormalities observed in major psychiatric disorders including schizophrenia, bipolar disorder, major depressive disorder (MDD), attention deficit hyperactivity disorder (ADHD), autism spectrum disorder (ASD), obsessive compulsive disorder (OCD). To this end, the PRS-SCZ related t-statistic map was parcellated into 64 Desikan-Killiany (DK) atlas regions (
      • Desikan R.S.
      • Ségonne F.
      • Fischl B.
      • Quinn B.T.
      • Dickerson B.C.
      • Blacker D.
      • et al.
      An automated labeling system for subdividing the human cerebral cortex on MRI scans into gyral based regions of interest.
      ) and then correlated with the corresponding Cohen’s d maps derived from recently published meta-analyses by the ENIGMA schizophrenia (
      • van Erp T.G.M.
      • Walton E.
      • Hibar D.P.
      • Schmaal L.
      • Jiang W.
      • Glahn D.C.
      • et al.
      Cortical Brain Abnormalities in 4474 Individuals With Schizophrenia and 5098 Control Subjects via the Enhancing Neuro Imaging Genetics Through Meta Analysis (ENIGMA) Consortium.
      ), bipolar disorder (
      • Schmaal L.
      • Hibar D.P.
      • Sämann P.G.
      • Hall G.B.
      • Baune B.T.
      • Jahanshad N.
      • et al.
      Cortical abnormalities in adults and adolescents with major depression based on brain scans from 20 cohorts worldwide in the ENIGMA Major Depressive Disorder Working Group.
      ), major depressive disorder (
      • Hibar D.P.
      • Westlye L.T.
      • Doan N.T.
      • Jahanshad N.
      • Cheung J.W.
      • Ching C.R.K.
      • et al.
      Cortical abnormalities in bipolar disorder: an MRI analysis of 6503 individuals from the ENIGMA Bipolar Disorder Working Group.
      ), attention deficit hyperactivity disorder (
      • Hoogman M.
      • Muetzel R.
      • Guimaraes J.P.
      • Shumskaya E.
      • Mennes M.
      • Zwiers M.P.
      • et al.
      Brain Imaging of the Cortex in ADHD: A Coordinated Analysis of Large-Scale Clinical and Population-Based Samples.
      ), obsessive compulsive disorder (
      • Boedhoe P.S.W.
      • Schmaal L.
      • Abe Y.
      • Alonso P.
      • Ameis S.H.
      • Anticevic A.
      • et al.
      Cortical Abnormalities Associated With Pediatric and Adult Obsessive-Compulsive Disorder: Findings From the ENIGMA Obsessive-Compulsive Disorder Working Group.
      ) and autism spectrum disorder (
      • van Rooij D.
      • Anagnostou E.
      • Arango C.
      • Auzias G.
      • Behrmann M.
      • Busatto G.F.
      • et al.
      Cortical and Subcortical Brain Morphometry Differences Between Patients With Autism Spectrum Disorder and Healthy Individuals Across the Lifespan: Results From the ENIGMA ASD Working Group.
      ) working groups implemented in the ENIGMA toolbox (
      • Larivière S.
      • Paquola C.
      • Park B.
      • Royer J.
      • Wang Y.
      • Benkarim O.
      • et al.
      The ENIGMA Toolbox: multiscale neural contextualization of multisite neuroimaging datasets.
      ). Specifically, spatial pattern similarity of cortical DK maps was examined using product-moment correlations. Statistical significance accounting for spatial autocorrelation were assessed with the spin permutation tests (10,000 repetitions) (
      • Alexander-Bloch A.F.
      • Shou H.
      • Liu S.
      • Satterthwaite T.D.
      • Glahn D.C.
      • Shinohara R.T.
      • et al.
      On testing for spatial correspondence between maps of human brain structure and function.
      ,
      • Burt J.B.
      • Helmer M.
      • Shinn M.
      • Anticevic A.
      • Murray J.D.
      Generative modeling of brain maps with spatial autocorrelation.
      ) implemented in the ENIGMA toolbox (
      • Larivière S.
      • Paquola C.
      • Park B.
      • Royer J.
      • Wang Y.
      • Benkarim O.
      • et al.
      The ENIGMA Toolbox: multiscale neural contextualization of multisite neuroimaging datasets.
      ). The medial wall was assigned as a NaN and not included in any permuted correlation (

      Markello RD, Misic B (2020): Comparing spatially-constrained null models for parcellated brain maps. bioRxiv 2020.08.13.249797.

      ). Statistical significance was deemed where pspin<.025 (two-tailed test) and false discovery rate (FDR) (pFDR<.05) was applied to control for multiple comparisons (n=6).

      Age group effects of PRS-SCZ on cortical thickness

      We examined the distinctiveness of PRS-SCZ related cortical thickness differences in different neurodevelopmental stages by dividing the sample into three age groups; early childhood (3-9 years, n=145), early adolescence (10-15 years, n=155) and late adolescence (16-21 years, n=116) (
      • Sawyer S.M.
      • Azzopardi P.S.
      • Wickremarathne D.
      • Patton G.C.
      The age of adolescence.
      ). The PRS-SCZ effect on cortical thickness was evaluated within each group using the abovementioned GLM (Eq.1), however, the age term was centred to the mean of the group to focus on the effect within the specified developmental stage (
      • Khundrakpam B.S.
      • Lewis J.D.
      • Kostopoulos P.
      • Carbonell F.
      • Evans A.C.
      Cortical Thickness Abnormalities in Autism Spectrum Disorders Through Late Childhood, Adolescence, and Adulthood: A Large-Scale MRI Study.
      ). Note that there was no correlation between age and PRS-SCZ scores (r=.058, Figure 1A). We specifically examined whether the association of the PRS-SCZ effect with the adult cortical thickness phenotype of SCZ changed across age groups. To do so, we compared the product-moment correlation coefficients between cortical maps (
      • Meng X.
      • Rosenthal R.
      • Rubin D.B.
      Comparing correlated correlation coefficients.
      ).

      Results

      Polygenic risk for SCZ is associated with greater cortical thickness

      To test the association between PRS-SCZ and cortical morphometry in typically developing children and adolescents, we used T1-weighted MRI and whole genome genotyping (n=390) from the PING cohort (3–21 years, mean±sd=12.1±4.7 years, 46% female) (Table S2). Vertex-wise GLMs related cortical thickness with PRS-SCZ, controlling for age, sex, the first 10 principal components of genetic variants (to account for population stratification), scanner, and total brain volume. We found that higher PRS-SCZ was significantly associated with greater cortical thickness (Random Field Theory (RFT) corrected, P<.01) but not surface area or cortical volume (Figure 1B, Figures S2-3). Overall, the unthresholded t-statistic map revealed that higher PRS-SCZ was associated with widespread increases in cortical thickness in association cortex, but reduced cortical thickness in sensory areas (Figure 1C top). Higher PRS-SCZ was associated with significantly thicker cortex in the left insula, left superior temporal gyrus and left inferior parietal lobule (Figure 1C bottom, RFT corrected, P<.01). These results suggest a significant effect of PRS-SCZ on cortical thickness but not surface area or cortical volume in typically developing children and adolescents. As such, subsequent analyses are restricted to cortical thickness.
      In a next step, we sought to examine how the PRS-SCZ related cortical thickness increase in typically developing children and adolescents relates to different levels of cortical organization including 1) cell type specific gene expression 2) cytoarchitectural and functional systems and 3) cortical pattern of case-control differences from schizophrenia and other major psychiatric disorders.

      Alignment with cell-type specific gene expression

      Histological examinations have reported a null or minimal relationship between cortical thickness and neuron number in healthy brain samples (
      • Collins C.E.
      • Airey D.C.
      • Young N.A.
      • Leitch D.B.
      • Kaas J.H.
      Neuron densities vary across and within cortical areas in primates.
      ,
      • Cahalane D.J.
      • Charvet C.J.
      • Finlay B.L.
      Systematic, balancing gradients in neuron density and number across the primate isocortex.
      ,
      • Carlo C.N.
      • Stevens C.F.
      Structural uniformity of neocortex, revisited.
      ). Instead, regional variations in cortical thickness show a strong association with neuropil (
      • Carlo C.N.
      • Stevens C.F.
      Structural uniformity of neocortex, revisited.
      ), the portion of cortical tissue that excludes cell bodies or blood vessels (

      Braitenberg V, Schüz A (1998): Density of Axons. In: Braitenberg V, Schüz A, editors. Cortex: Statistics and Geometry of Neuronal Connectivity. Berlin, Heidelberg: Springer Berlin Heidelberg, pp 39–42.

      ). We examined whether cortical thickness differences related to PRS-SCZ mirrored regional variations in the neuropil composition of the cortex, in order to generate hypotheses on the neuropil components affected by PRS-SCZ. To this end, we first validated the relationship between cortical thickness and neuropil using tabular data and photomicrographs of Nissl stains [r=.49, pspin=.018, Figure 2A, (

      von Economo CF, Koskinas GN (1925): Die Cytoarchitektonik Der Hirnrinde Des Erwachsenen Menschen. J. Springer.

      ,

      Economo C von (2009): Cellular Structure of the Human Cerebral Cortex ((L. C. Triarhou, editor)). Karger. Retrieved January 16, 2021, from https://www.karger.com/Book/Home/247637

      )]. In contrast, neuronal density was not correlated with histologically-defined cortical thickness (r=.08 pspin=.678), which aligns with previous work (
      • Cahalane D.J.
      • Charvet C.J.
      • Finlay B.L.
      Systematic, balancing gradients in neuron density and number across the primate isocortex.
      ,
      • Carlo C.N.
      • Stevens C.F.
      Structural uniformity of neocortex, revisited.
      ). Then, we estimated regional variations in neuropil-related gene expression, based on six cellular components (astrocytes, microglia, oligodendrocytes, axons, dendritic trees, neuron-to-neuron synapses) by combining tissue-level RNAseq (
      • Li M.
      • Santpere G.
      • Imamura Kawasawa Y.
      • Evgrafov O.V.
      • Gulden F.O.
      • Pochareddy S.
      • et al.
      Integrative functional genomic analysis of human brain development and neuropsychiatric risks.
      ) with single-cell RNAseq for cell-types (
      • Zhu Y.
      • Sousa A.M.M.
      • Gao T.
      • Skarica M.
      • Li M.
      • Santpere G.
      • et al.
      Spatiotemporal transcriptomic divergence across human and macaque brain development.
      ) and gene ontologies for neuron-compartments (
      • Ashburner M.
      • Ball C.A.
      • Blake J.A.
      • Botstein D.
      • Butler H.
      • Cherry J.M.
      • et al.
      Gene Ontology: tool for the unification of biology [no. 1].
      ,
      The Gene Ontology Consortium
      The Gene Ontology Resource: 20 years and still GOing strong.
      ) (Figure 2B). Correlating PRS-SCZ effects on cortical thickness with neuropil-related gene expression, we found that PRS-SCZ effects are significantly associated with gene expression for dendritic trees (r=.755, pperm=.006), synapses (r=.618, pperm=.005), and at a trend-level with axons (r=.481, pperm=.069) (Figure 2C). In contrast, no significant correlation was observed with gene expression related to glial components of neuropil (Figure 2C). Together, these analyses suggest greater cortical thickness with higher PRS-SCZ is observed in areas with greater dendritic and synaptic density.
      Figure thumbnail gr2
      Figure 2Decoding spatial patterns of PRS-SCZ on cortical thickness. A) Correlation of histological measurements of cortical thickness and neuropil from Von Economo & Koskinas (

      von Economo CF, Koskinas GN (1925): Die Cytoarchitektonik Der Hirnrinde Des Erwachsenen Menschen. J. Springer.

      ). Cortical thickness is shown in pink and on the y-axis. Neuropil is shown in green and on the x-axis. B) Gene expression varies across glial and neuron-related compartments of neuropil in eleven cortical regions. C) Correlation of neuropil-related gene expression with the PRS-SCZ effects showed significant association with dendrites and synapses, compared with null distributions from permutation testing (grey). Abbreviations. A1C: primary auditory cortex. DFC: dorsal frontal cortex. IPC: inferior parietal cortex. ITC: inferior temporal cortex. M1C: primary motor cortex. MFC: medial frontal cortex. OFC: orbital frontal cortex. S1C: primary somatosensory cortex. STC: superior temporal cortex. V1C: primary visual cortex. VFC: ventral frontal cortex.

      Contextualisation by cytoarchitectural types and functional networks

      Next, we aimed to determine whether the PRS-SCZ is preferentially associated with certain cytoarchitectural types or functional networks. Based on established atlases of cytoarchitectural and functional differentiation (
      • García-Cabezas M.Á.
      • Hacker J.L.
      • Zikopoulos B.
      A Protocol for Cortical Type Analysis of the Human Neocortex Applied on Histological Samples, the Atlas of Von Economo and Koskinas, and Magnetic Resonance Imaging.
      ,
      • Yeo B.T.T.
      • Krienen F.M.
      • Sepulcre J.
      • Sabuncu M.R.
      • Lashkari D.
      • Hollinshead M.
      • et al.
      The organization of the human cerebral cortex estimated by intrinsic functional connectivity.
      ), we found that the PRS-SCZ effect was stronger compared to null models in the ventral attention network (median±SD=1.64±0.88, pspin=.004). Conversely, PRS-SCZ effect was weaker or more negative compared to null models in the koniocortical type (i.e. primary sensory areas) (median±SD=0.08±1.10, pspin=.008) (Figure 3).
      Figure thumbnail gr3
      Figure 3Aggregation of PRS-SCZ effect within cytoarchitectural types and functional networks. Raincloud plots show the distribution of the PRS-SCZ effect on cortical thickness stratified by cytoarchitectural type (
      • García-Cabezas M.Á.
      • Hacker J.L.
      • Zikopoulos B.
      A Protocol for Cortical Type Analysis of the Human Neocortex Applied on Histological Samples, the Atlas of Von Economo and Koskinas, and Magnetic Resonance Imaging.
      ) and functional network (
      • Yeo B.T.T.
      • Krienen F.M.
      • Sepulcre J.
      • Sabuncu M.R.
      • Lashkari D.
      • Hollinshead M.
      • et al.
      The organization of the human cerebral cortex estimated by intrinsic functional connectivity.
      ). Relative to spin permutation null models, the koniocortical cortical type encompassed significantly lower t-statistics, whereas the VAN encompassed significantly higher t-statistics. DAN: dorsal attention network. VAN: ventral attention network. DMN: default mode network.

      Cortical thickness signatures of PRS-SCZ and major psychiatric disorders

      We assessed whether PRS-SCZ effects on cortical thickness relate to abnormal cortical thickness patterns observed in case-control meta-analyses (Cohen’s d maps) of schizophrenia and other psychiatric illnesses. The PRS-SCZ related cortical thickness increase showed a negative correlation with schizophrenia-related cortical abnormalities (r=-.326, pspin=.0022). In addition, we found similar negative correlations to cortical abnormalities in BD (r=-.466, pspin<.001), MDD (r=-.538, pspin<.001) and ADHD (r=-.430, pspin<.001) but not OCD or ASD (Figure 4). To further test whether the correlation between PRS-SCZ related cortical thickness differences and schizophrenia-related cortical thinning was significantly different than the correlations observed in non-SCZ psychopathologies, we applied pairwise comparisons of the correlation coefficients of schizophrenia with those from BD, MDD, and ADHD using the R package cocor (
      • Diedenhofen B.
      • Musch J.
      cocor: A Comprehensive Solution for the Statistical Comparison of Correlations.
      ) and the confidence interval test from Zou (
      • Zou G.Y.
      Toward using confidence intervals to compare correlations.
      ). The confidence intervals for each comparison of correlations r included zero (correlation r: SCZ vs BD, 9% CI [-0.01 0.30]; SCZ vs MDD, 95% CI [-0.03 0.45]; SCZ vs ADHD 95% CI [-0.20 0.39]). This shows that the correlation of PRS-SCZ cortical thickness differences with the schizophrenia-related cortical thickness abnormalities was not significantly different than the correlations with each of the other major psychiatric disorders. Altogether, cortical regions showing PRS-SCZ related greater thickness are those with the strongest thinning across disease maps of schizophrenia and genetically related affective disorders (e.g. BD, MDD) and ADHD.
      Figure thumbnail gr4
      Figure 4Pattern similarity of PRS-SCZ on cortical thickness with major psychiatric disorders. Cortical surfaces show the effect size of schizophrenia, bipolar disorder and major depressive disorder, attention deficit hyperactivity disorder, obsessive compulsive disorder and autism spectrum disorder diagnosis on cortical thickness from ENIGMA meta-analyses of each disorder (
      • Larivière S.
      • Paquola C.
      • Park B.
      • Royer J.
      • Wang Y.
      • Benkarim O.
      • et al.
      The ENIGMA Toolbox: multiscale neural contextualization of multisite neuroimaging datasets.
      ). Statistical significance was deemed where pspin<0.025 (two-tailed test) and false discovery rate (FDR) (pFDR<0.05) was applied to control for multiple comparisons (n=6). Scatterplots show the correlation of each map with PRS-SCZ effect on cortical thickness.

      Age group effects of PRS-SCZ

      Schizophrenia-related genes are implicated in neurodevelopmental processes and as such the effect of PRS-SCZ on cortical thickness likely varies with age. Although the cross-sectional nature of this cohort prohibits mapping individual trajectories of cortical development, we sought to approximate developmental variation in the effect of PRS-SCZ by estimating age group effects of early childhood (3-9 yrs), early adolescence (10-15 yrs), and late adolescence (16-21 yrs) in the cohort. Due to the smaller sample sizes within each age group, effects were in general smaller compared to the main effect and did not survive multiple comparison correction, with the exception of a small cluster in the right rostral anterior cingulate cortex for the early childhood group. Higher PRS-SCZ was associated with greater cortical thickness in early childhood, however the pattern differed in the older age groups (Figure 5). We detected a significant difference in the correlation coefficients between early childhood and late adolescence (z=2.84, p=.002), as well as early adolescence and late adolescence (z=1.84, p=.033). Furthermore, PRS-SCZ related cortical thickness increase in early childhood correlated negatively with schizophrenia-related cortical abnormalities, whereas PRS-SCZ related cortical thinning in late adolescents correlated positively (Figure 5). To further inspect the age-related change in the PRS-SCZ effect on cortical thickness, we repeated the analysis using the entire cohort and iteratively shifting the age-centring from 3-21 in one-year intervals. Higher PRS-SCZ was associated with greater cortical thickness in the 3-6 yrs age-centred models (Figure S4) closely resembling the results from the main analysis (Figure 1C) and survived multiple comparison correction (Figure S4 bottom, RFT corrected, P<.01). Finally, the observed non-linear relationship of PRS-SCZ on cortical thickness across different age groups were confirmed when examining the interaction effect of PRS-SCZ-by-Ageˆ2 on cortical thickness across the entire age range 3-21 yrs (Figure S5).
      Figure thumbnail gr5
      Figure 5Age group based PRS-SCZ effect on cortical thickness. Cortical surfaces show unthresholed maps of PRS-SCZ effect on cortical thickness within age groups. Line plot shows how the relationship (linear regression with 95% confidence intervals) of PRS-SCZ with the SCZ-related pattern of abnormalities (Figure 4) changes from negative to positive across the age groups. Significant difference in correlation coefficients is shown by * for p<0.05 and ** for p<0.01.

      Discussion

      Combining imaging-genetics with multi-scale mapping, we characterised the effect of PRS-SCZ on cortical morphometry across different scales of cortical organization. We find that higher PRS-SCZ was associated with greater cortical thickness in typically developing children and adolescents, while surface area and cortical volume showed only subtle associations with PRS-SCZ. We further provide evidence that PRS-SCZ preferentially affects areas with heightened expression of dendrites and synapses and that the PRS-SCZ related cortical differences accumulate in cytoarchitecturally and functionally defined cortical systems. We also find that the PRS-SCZ related cortical pattern mirrors cortical thinning related to schizophrenia and other major psychiatric disorders. Finally, age group models suggest a potential trajectory from PRS-SCZ associated cortical thickness increase in early childhood towards thinning in late adolescence spatially resembling the adult brain phenotype of schizophrenia.
      Our cell-type specific gene expression approach enabled cross-modal exploration of the relationship of genetic risk for schizophrenia with expression levels of neurons and glia. Our findings support and extend upon post-mortem analyses, which demonstrate abnormal dendritic and synaptic density in individuals with schizophrenia (see (
      • Berdenis van Berlekom A.
      • Muflihah C.H.
      • Snijders G.J.L.J.
      • MacGillavry H.D.
      • Middeldorp J.
      • Hol E.M.
      • et al.
      Synapse Pathology in Schizophrenia: A Meta-analysis of Postsynaptic Elements in Postmortem Brain Studies.
      ) for a recent meta-analysis). Another recent study showed that differences in cortical thickness across multiple psychiatric disorders (including schizophrenia) are associated with pyramidal-cell gene expression, a gene set enriched for biological processes of dendrites (e.g. dendritic arborization and branching) (
      • Srinivas K.V.
      • Buss E.W.
      • Sun Q.
      • Santoro B.
      • Takahashi H.
      • Nicholson D.A.
      • Siegelbaum S.A.
      The Dendrites of CA2 and CA1 Pyramidal Neurons Differentially Regulate Information Flow in the Cortico-Hippocampal Circuit.
      ) as well as synaptic function (

      Writing Committee for the Attention-Deficit/Hyperactivity Disorder, Autism Spectrum Disorder, Bipolar Disorder, Major Depressive Disorder, Obsessive-Compulsive Disorder, and Schizophrenia ENIGMA Working Groups (2021): Virtual Histology of Cortical Thickness and Shared Neurobiology in 6 Psychiatric Disorders. JAMA Psychiatry 78: 47–63.

      ). The present findings extend this work by demonstrating a relationship between gene expression of dendrites and synapses with PRS-SCZ related cortical differences during neurodevelopment.
      System-specific contextualization revealed that PRS-SCZ effects on cortical thickness were spatially distributed with the ventral attention network being preferentially sensitive to PRS-SCZ, while koniocortical type cortex was mostly spared from its influence. This cortical thickness pattern of PRS-SCZ closely mirrors recent observations in patients with schizophrenia showing stronger brain abnormalities in the ventral attention network, while primary cortex, as defined by von Economo, was relatively spared (
      • Shafiei G.
      • Markello R.D.
      • Makowski C.
      • Talpalaru A.
      • Kirschner M.
      • Devenyi G.A.
      • et al.
      Spatial Patterning of Tissue Volume Loss in Schizophrenia Reflects Brain Network Architecture.
      ). Altogether, these findings demonstrate that system-specific differentiations of PRS-SCZ related cortical thickness differences during neurodevelopment reflect cortical abnormalities of schizophrenia suggesting some neuroanatomical continuity between polygenic risk and clinical phenotype.
      Longitudinal data and case-control meta-analysis have shown that the development of psychosis in high-risk adolescents is associated with progressive loss of cortical thickness in several areas of the association cortex (
      • Jalbrzikowski M.
      • Hayes R.A.
      • Wood S.J.
      • Nordholm D.
      • Zhou J.H.
      • et al.
      ENIGMA Clinical High Risk for Psychosis Working Group
      Association of Structural Magnetic Resonance Imaging Measures With Psychosis Onset in Individuals at Clinical High Risk for Developing Psychosis: An ENIGMA Working Group Mega-analysis.
      ,
      • Cannon T.D.
      • Chung Y.
      • He G.
      • Sun D.
      • Jacobson A.
      • van Erp T.G.M.
      • et al.
      Progressive Reduction in Cortical Thickness as Psychosis Develops: A Multisite Longitudinal Neuroimaging Study of Youth at Elevated Clinical Risk.
      ). Of note, the areas implicated in these studies overlap considerably with those showing increased cortical thickness in early childhood and more pronounced cortical thinning in adolescents with higher PRS-SCZ in the current study. We further observed that the pattern of PRS-SCZ related cortical thickening was associated with areas of cortical thinning in schizophrenia, BD, MDD, ADHD. This cross-disorder overlap notably mirrors the genetic and phenotypic correlation between these disorders (

      Writing Committee for the Attention-Deficit/Hyperactivity Disorder, Autism Spectrum Disorder, Bipolar Disorder, Major Depressive Disorder, Obsessive-Compulsive Disorder, and Schizophrenia ENIGMA Working Groups (2021): Virtual Histology of Cortical Thickness and Shared Neurobiology in 6 Psychiatric Disorders. JAMA Psychiatry 78: 47–63.

      ,
      • Consortium T.B.
      • Anttila V.
      • Bulik-Sullivan B.
      • Finucane H.K.
      • Walters R.K.
      • Bras J.
      • et al.
      Analysis of shared heritability in common disorders of the brain.
      ) and is in line with recent work in neurodevelopmental cohorts reporting a transdiagnostic mode of genome–connectome covariation (
      • Taquet M.
      • Smith S.M.
      • Prohl A.K.
      • Peters J.M.
      • Warfield S.K.
      • Scherrer B.
      • Harrison P.J.
      A structural brain network of genetic vulnerability to psychiatric illness [no. 6].
      ) and shared features of frontotemporal dysconnectivity of general psychopathology (
      • Alnæs D.
      • Kaufmann T.
      • van der Meer D.
      • Córdova-Palomera A.
      • Rokicki J.
      • Moberget T.
      • et al.
      Brain Heterogeneity in Schizophrenia and Its Association With Polygenic Risk.
      ). Collectively, these findings support the relevance of transdiagnostic gene-brain and brain-psychopathology phenotypes during neurodevelopment. While we do not know the cause of increased cortical thickness in our sample, converging evidence supports the idea that reduced cortical thickness in adults with schizophrenia results from loss of neuropil, and specifically synapses. For example, post mortem studies in schizophrenia demonstrate synaptic loss (
      • Berdenis van Berlekom A.
      • Muflihah C.H.
      • Snijders G.J.L.J.
      • MacGillavry H.D.
      • Middeldorp J.
      • Hol E.M.
      • et al.
      Synapse Pathology in Schizophrenia: A Meta-analysis of Postsynaptic Elements in Postmortem Brain Studies.
      ); many genes implicated in schizophrenia are associated with synapses or synaptic pruning (
      • Ripke S.
      • Neale B.M.
      • Corvin A.
      • Walters J.T.R.
      • Farh K.-H.
      • et al.
      Schizophrenia Working Group of the Psychiatric Genomics Consortium
      Biological insights from 108 schizophrenia-associated genetic loci.
      ,
      • Kirov G.
      • Pocklington A.J.
      • Holmans P.
      • Ivanov D.
      • Ikeda M.
      • Ruderfer D.
      • et al.
      De novo CNV analysis implicates specific abnormalities of postsynaptic signalling complexes in the pathogenesis of schizophrenia [no. 2].
      ,

      Consortium TSWG of the PG, Ripke S, Walters JT, O’Donovan MC (2020): Mapping genomic loci prioritises genes and implicates synaptic biology in schizophrenia. medRxiv 2020.09.12.20192922.

      ); regional variations in cortical thickness correlate with neuropil (
      • Carlo C.N.
      • Stevens C.F.
      Structural uniformity of neocortex, revisited.
      ). It is conceivable that the PRS-SCZ is associated with delayed pruning and an excess of synapses for age, which in turn may render the affected brain regions vulnerable to catastrophic synaptic loss during the emergence of psychosis.
      The association of PRS-SCZ with greater cortical thickness in early childhood raises the question of how the genetic risk of schizophrenia contributes to abnormal developmental trajectories. Given that the transmodal areas identified in the present analysis, such as the insula, exhibit modest cortical thinning from 3-21 years (
      • Ball G.
      • Beare R.
      • Seal M.L.
      Charting shared developmental trajectories of cortical thickness and structural connectivity in childhood and adolescence.
      ,
      • Sotiras A.
      • Toledo J.B.
      • Gur R.E.
      • Gur R.C.
      • Satterthwaite T.D.
      • Davatzikos C.
      Patterns of coordinated cortical remodeling during adolescence and their associations with functional specialization and evolutionary expansion.
      ,
      • Zielinski B.A.
      • Prigge M.B.D.
      • Nielsen J.A.
      • Froehlich A.L.
      • Abildskov T.J.
      • Anderson J.S.
      • et al.
      Longitudinal changes in cortical thickness in autism and typical development.
      ), our results align with either an amplified trajectory (i.e.: higher peak, steeper decline) and/or delayed cortical thinning in early childhood. Related to the complexity and heterochronicity of cortical maturation during childhood and adolescents (
      • Ball G.
      • Beare R.
      • Seal M.L.
      Charting shared developmental trajectories of cortical thickness and structural connectivity in childhood and adolescence.
      ), polygenic disorders can involve multiple types of abnormal trajectories, occurring simultaneously or sequentially (
      • Di Martino A.
      • Fair D.A.
      • Kelly C.
      • Satterthwaite T.D.
      • Castellanos F.X.
      • Thomason M.E.
      • et al.
      Unraveling the Miswired Connectome: A Developmental Perspective.
      ,
      • Klingler E.
      • Francis F.
      • Jabaudon D.
      • Cappello S.
      Mapping the molecular and cellular complexity of cortical malformations.
      ). Amplified or delayed trajectory of transmodal area morphometry may represent a core motif of cortical development in children with high PRS-SCZ. The PRS-SCZ includes multiple genetic factors, however, and their individual variation may produce heterogeneity in cortical development within high PRS-SCZ individuals.
      The lack of associations between PRS-SCZ and surface area is in line with previous observations in adults (
      • Neilson E.
      • Shen X.
      • Cox S.R.
      • Clarke T.-K.
      • Wigmore E.M.
      • Gibson J.
      • et al.
      Impact of Polygenic Risk for Schizophrenia on Cortical Structure in UK Biobank.
      ) and meta-analytic evidence showing weaker surface area abnormalities compared to cortical thickness in schizophrenia (
      • van Erp T.G.M.
      • Walton E.
      • Hibar D.P.
      • Schmaal L.
      • Jiang W.
      • Glahn D.C.
      • et al.
      Cortical Brain Abnormalities in 4474 Individuals With Schizophrenia and 5098 Control Subjects via the Enhancing Neuro Imaging Genetics Through Meta Analysis (ENIGMA) Consortium.
      ). The divergent effects of PRS-SCZ on cortical thickness and surface area observed in the present study, might further be explained by different underlying genetic architecture of both cortical features. The mechanisms underlying cortical thickness versus surface area are often placed in the context of Rakic’s radial unit hypothesis, which proposes that surface area reflects the number of cortical columns while thickness reflects the volume of each column (
      • Rakic P.
      • Ayoub A.E.
      • Breunig J.J.
      • Dominguez M.H.
      Decision by division: making cortical maps.
      ). Using bioinformatic analyses, Grasby and colleagues revealed that cortical thickness is influenced by genetic variants reflecting processes during mid-fetal development including myelination, branching and pruning, while total surface area has been related to altered gene regulation in neural progenitor cells during fetal development (
      • Grasby K.L.
      • Jahanshad N.
      • Painter J.N.
      • Colodro-Conde L.
      • Bralten J.
      • Hibar D.P.
      • et al.
      The genetic architecture of the human cerebral cortex.
      ). Collectively, these findings suggest that schizophrenia-related genetic variations exert greater influence on neurodevelopmental processes altering cortical thickness than surface area.
      The present study should be interpreted in the light of the cross-sectional nature of the dataset that limits the ability to map individual longitudinal trajectories. The convergence of the observed findings with gene expression and neuroanatomical studies of schizophrenia support multiscale continuity between polygenic risk and clinical phenotype of schizophrenia. However, given the low familial risk and relative absence of other biological or environmental risk factors for schizophrenia in the study cohort, interaction between PRS-SCZ and other biological and environmental risk factors could not directly be assessed and warrants further investigation. Additionally, although mapping polygenic risk profiles on neuroimaging-derived phenotypes is a useful approach to further our understanding of genetic influence on neuroanatomical signatures related to schizophrenia risk, this method is limited by the fact that it does not allow direct translation into underlying biological mechanism. Future research could therefore be enhanced by larger datasets with longitudinal designs and longer follow-up to determine which individuals will develop psychosis or other mental disorders. Finally, the observed PRS-SCZ related cortical thickness increase in early childhood (age 3-9) highlights the need for large-scale initiatives targeting this age range.

      Conclusions

      The present study provides novel evidence on the cellular basis and developmental trajectory of cortical thickness differences related to genetic risk for schizophrenia that may help to refine the neurodevelopmental hypothesis of schizophrenia. More generally, the present work illustrates how maps of cortical organization can enrich descriptions of imaging-derived phenotypes related to genetic risk for mental illnesses. Altogether, this integrative framework combining imaging-genetics and multi-scale mapping could advance our understanding of the complex associations between individual genetic profiles and cortical organization across multiple psychiatric and neurological conditions.

      Acknowledgments

      MK acknowledges funding from the Swiss National Science Foundation (P2SKP3_178175). AD is supported by the Canadian Institutes of Health Research Foundation Scheme. NBA is supported by grants from Brain Canada (238990, 243030), CFREF/HBHL Innovative Ideas (247613), Coutu Research Fund (241177), CFREF/HBHL Discovery (247712). An earlier version of this article has been posted as preprint on biorxiv.org (https://www.biorxiv.org/content/10.1101/2021.06.13.448243v1).

      References

        • Birnbaum R.
        • Weinberger D.R.
        Genetic insights into the neurodevelopmental origins of schizophrenia.
        Nat Rev Neurosci. 2017; 18: 727-740
        • Insel T.R.
        Rethinking schizophrenia.
        Nature. 2010; 468: 187-193
        • Rapoport J.L.
        • Giedd J.N.
        • Gogtay N.
        Neurodevelopmental model of schizophrenia: update 2012.
        Mol Psychiatry. 2012; 17: 1228-1238
        • Kessler R.C.
        • Angermeyer M.
        • Anthony J.C.
        • DE Graaf R.
        • Demyttenaere K.
        • Gasquet I.
        • et al.
        Lifetime prevalence and age-of-onset distributions of mental disorders in the World Health Organization’s World Mental Health Survey Initiative.
        World Psychiatry. 2007; 6: 168-176
        • Ripke S.
        • Neale B.M.
        • Corvin A.
        • Walters J.T.R.
        • Farh K.-H.
        • et al.
        • Schizophrenia Working Group of the Psychiatric Genomics Consortium
        Biological insights from 108 schizophrenia-associated genetic loci.
        Nature. 2014; 511: 421-427
        • Shaw P.
        • Greenstein D.
        • Lerch J.
        • Clasen L.
        • Lenroot R.
        • Gogtay N.
        • et al.
        Intellectual ability and cortical development in children and adolescents.
        Nature. 2006; 440: 676-679
        • Raznahan A.
        • Shaw P.
        • Lalonde F.
        • Stockman M.
        • Wallace G.L.
        • Greenstein D.
        • et al.
        How Does Your Cortex Grow?.
        J Neurosci. 2011; 31: 7174-7177
        • Khundrakpam B.S.
        • Lewis J.D.
        • Kostopoulos P.
        • Carbonell F.
        • Evans A.C.
        Cortical Thickness Abnormalities in Autism Spectrum Disorders Through Late Childhood, Adolescence, and Adulthood: A Large-Scale MRI Study.
        Cereb Cortex. 2017; 27: 1721-1731
      1. Park B, Bethlehem RA, Paquola C, Larivière S, Rodríguez-Cruces R, Vos de Wael R, et al. (2021): An expanding manifold in transmodal regions characterizes adolescent reconfiguration of structural connectome organization ((L. Q. Uddin, T. E. Behrens, & B. D. Fulcher, editors)). eLife 10: e64694.

        • Sydnor V.J.
        • Larsen B.
        • Bassett D.S.
        • Alexander-Bloch A.
        • Fair D.A.
        • Liston C.
        • et al.
        Neurodevelopment of the association cortices: Patterns, mechanisms, and implications for psychopathology.
        Neuron. 2021; 109: 2820-2846
        • Tamnes C.K.
        • Herting M.M.
        • Goddings A.-L.
        • Meuwese R.
        • Blakemore S.-J.
        • Dahl R.E.
        • et al.
        Development of the Cerebral Cortex across Adolescence: A Multisample Study of Inter-Related Longitudinal Changes in Cortical Volume, Surface Area, and Thickness.
        J Neurosci. 2017; 37: 3402-3412
        • Giedd J.N.
        • Blumenthal J.
        • Jeffries N.O.
        • Castellanos F.X.
        • Liu H.
        • Zijdenbos A.
        • et al.
        Brain development during childhood and adolescence: a longitudinal MRI study [no. 10].
        Nature Neuroscience. 1999; 2: 861-863
        • Gogtay N.
        • Giedd J.N.
        • Lusk L.
        • Hayashi K.M.
        • Greenstein D.
        • Vaituzis A.C.
        • et al.
        Dynamic mapping of human cortical development during childhood through early adulthood.
        PNAS. 2004; 101: 8174-8179
      2. Paquola C, Bethlehem RA, Seidlitz J, Wagstyl K, Romero-Garcia R, Whitaker KJ, et al. (2019): Shifts in myeloarchitecture characterise adolescent development of cortical gradients ((J. I. Gold, T. Satterthwaite, K. Mills, & B. Fulcher, editors)). eLife 8: e50482.

        • van Erp T.G.M.
        • Walton E.
        • Hibar D.P.
        • Schmaal L.
        • Jiang W.
        • Glahn D.C.
        • et al.
        Cortical Brain Abnormalities in 4474 Individuals With Schizophrenia and 5098 Control Subjects via the Enhancing Neuro Imaging Genetics Through Meta Analysis (ENIGMA) Consortium.
        Biol Psychiatry. 2018; 84: 644-654
        • Wannan C.M.J.
        • Cropley V.L.
        • Chakravarty M.M.
        • Bousman C.
        • Ganella E.P.
        • Bruggemann J.M.
        • et al.
        Evidence for Network-Based Cortical Thickness Reductions in Schizophrenia.
        Am J Psychiatry. 2019; 176: 552-563
        • Kirschner M.
        • Shafiei G.
        • Markello R.D.
        • Makowski C.
        • Talpalaru A.
        • Hodzic-Santor B.
        • et al.
        Latent Clinical-Anatomical Dimensions of Schizophrenia.
        Schizophr Bull. 2020; https://doi.org/10.1093/schbul/sbaa097
        • Jalbrzikowski M.
        • Hayes R.A.
        • Wood S.J.
        • Nordholm D.
        • Zhou J.H.
        • et al.
        • ENIGMA Clinical High Risk for Psychosis Working Group
        Association of Structural Magnetic Resonance Imaging Measures With Psychosis Onset in Individuals at Clinical High Risk for Developing Psychosis: An ENIGMA Working Group Mega-analysis.
        JAMA Psychiatry. 2021; https://doi.org/10.1001/jamapsychiatry.2021.0638
        • French L.
        • Gray C.
        • Leonard G.
        • Perron M.
        • Pike G.B.
        • Richer L.
        • et al.
        Early Cannabis Use, Polygenic Risk Score for Schizophrenia, and Brain Maturation in Adolescence.
        JAMA Psychiatry. 2015; 72: 1002-1011
        • Liu B.
        • Zhang X.
        • Cui Y.
        • Qin W.
        • Tao Y.
        • Li J.
        • et al.
        Polygenic Risk for Schizophrenia Influences Cortical Gyrification in 2 Independent General Populations.
        Schizophr Bull. 2017; 43: 673-680
        • Neilson E.
        • Bois C.
        • Gibson J.
        • Duff B.
        • Watson A.
        • Roberts N.
        • et al.
        Effects of environmental risks and polygenic loading for schizophrenia on cortical thickness.
        Schizophr Res. 2017; 184: 128-136
        • Neilson E.
        • Shen X.
        • Cox S.R.
        • Clarke T.-K.
        • Wigmore E.M.
        • Gibson J.
        • et al.
        Impact of Polygenic Risk for Schizophrenia on Cortical Structure in UK Biobank.
        Biological Psychiatry. 2019; https://doi.org/10.1016/j.biopsych.2019.04.013
        • Van der Auwera S.
        • Wittfeld K.
        • Homuth G.
        • Teumer A.
        • Hegenscheid K.
        • Grabe H.J.
        No association between polygenic risk for schizophrenia and brain volume in the general population.
        Biol Psychiatry. 2015; 78: e41-42
        • Auwera SV der
        • Wittfeld K.
        • Shumskaya E.
        • Bralten J.
        • Zwiers M.P.
        • Onnink A.M.H.
        • et al.
        Predicting brain structure in population-based samples with biologically informed genetic scores for schizophrenia.
        American Journal of Medical Genetics Part B: Neuropsychiatric Genetics. 2017; 174: 324-332
        • Reus L.M.
        • Shen X.
        • Gibson J.
        • Wigmore E.
        • Ligthart L.
        • Adams M.J.
        • et al.
        Association of polygenic risk for major psychiatric illness with subcortical volumes and white matter integrity in UK Biobank.
        Sci Rep. 2017; 7https://doi.org/10.1038/srep42140
        • Feinberg I.
        Schizophrenia: Caused by a fault in programmed synaptic elimination during adolescence?.
        Journal of Psychiatric Research. 1982; 17: 319-334
        • Sellgren C.M.
        • Gracias J.
        • Watmuff B.
        • Biag J.D.
        • Thanos J.M.
        • Whittredge P.B.
        • et al.
        Increased synapse elimination by microglia in schizophrenia patient-derived models of synaptic pruning [no. 3].
        Nature Neuroscience. 2019; 22: 374-385
        • Sprooten E.
        • O’Halloran R.
        • Dinse J.
        • Lee W.H.
        • Moser D.A.
        • Doucet G.E.
        • et al.
        Depth-dependent intracortical myelin organization in the living human brain determined by in vivo ultra-high field magnetic resonance imaging.
        NeuroImage. 2019; 185: 27-34
        • Wei W.
        • Zhang Y.
        • Li Y.
        • Meng Y.
        • Li M.
        • Wang Q.
        • et al.
        Depth-dependent abnormal cortical myelination in first-episode treatment-naïve schizophrenia.
        Human Brain Mapping. 2020; 41: 2782-2793
        • Zhu Y.
        • Sousa A.M.M.
        • Gao T.
        • Skarica M.
        • Li M.
        • Santpere G.
        • et al.
        Spatiotemporal transcriptomic divergence across human and macaque brain development.
        Science. 2018; 362https://doi.org/10.1126/science.aat8077
        • Seidlitz J.
        • Nadig A.
        • Liu S.
        • Bethlehem R.A.I.
        • Vértes P.E.
        • Morgan S.E.
        • et al.
        Transcriptomic and cellular decoding of regional brain vulnerability to neurogenetic disorders [no. 1].
        Nature Communications. 2020; 11: 3358
      3. Writing Committee for the Attention-Deficit/Hyperactivity Disorder, Autism Spectrum Disorder, Bipolar Disorder, Major Depressive Disorder, Obsessive-Compulsive Disorder, and Schizophrenia ENIGMA Working Groups (2021): Virtual Histology of Cortical Thickness and Shared Neurobiology in 6 Psychiatric Disorders. JAMA Psychiatry 78: 47–63.

        • Shafiei G.
        • Markello R.D.
        • Makowski C.
        • Talpalaru A.
        • Kirschner M.
        • Devenyi G.A.
        • et al.
        Spatial Patterning of Tissue Volume Loss in Schizophrenia Reflects Brain Network Architecture.
        Biological Psychiatry. 2020; 87: 727-735
        • Cao H.
        • Zhou H.
        • Cannon T.D.
        Functional connectome-wide associations of schizophrenia polygenic risk.
        Mol Psychiatry. 2020; : 1-9
        • García-Cabezas M.Á.
        • Hacker J.L.
        • Zikopoulos B.
        A Protocol for Cortical Type Analysis of the Human Neocortex Applied on Histological Samples, the Atlas of Von Economo and Koskinas, and Magnetic Resonance Imaging.
        Front Neuroanat. 2020; 14https://doi.org/10.3389/fnana.2020.576015
        • Yeo B.T.T.
        • Krienen F.M.
        • Sepulcre J.
        • Sabuncu M.R.
        • Lashkari D.
        • Hollinshead M.
        • et al.
        The organization of the human cerebral cortex estimated by intrinsic functional connectivity.
        J Neurophysiol. 2011; 106: 1125-1165
        • Jernigan T.L.
        • Brown T.T.
        • Hagler D.J.
        • Akshoomoff N.
        • Bartsch H.
        • Newman E.
        • et al.
        The Pediatric Imaging, Neurocognition, and Genetics (PING) Data Repository.
        NeuroImage. 2016; 124: 1149-1154
        • Khundrakpam B.
        • Vainik U.
        • Gong J.
        • Al-Sharif N.
        • Bhutani N.
        • Kiar G.
        • et al.
        Neural correlates of polygenic risk score for autism spectrum disorders in general population.
        Brain Commun. 2020; 2https://doi.org/10.1093/braincomms/fcaa092
        • Ruderfer D.M.
        • Ripke S.
        • McQuillin A.
        • Boocock J.
        • Stahl E.A.
        • Pavlides J.M.W.
        • et al.
        Genomic Dissection of Bipolar Disorder and Schizophrenia, Including 28 Subphenotypes.
        Cell. 2018; 173 (e16): 1705-1715
        • Choi S.W.
        • O’Reilly P.F.
        PRSice-2: Polygenic Risk Score software for biobank-scale data.
        Gigascience. 2019; 8https://doi.org/10.1093/gigascience/giz082
      4. Ad-Dab’bagh Y, Einarson D, Lyttelton O, Muehlboeck J-S, Mok K, Ivanov O, et al. (2006): The CIVET Image-Processing Environment: A Fully Automated Comprehensive Pipeline for Anatomical Neuroimaging Research. 1.

        • Worsley K.
        • Taylor J.
        • Carbonell F.
        • Chung M.
        • Duerden E.
        • Bernhardt B.
        • et al.
        SurfStat: A Matlab toolbox for the statistical analysis of univariate and multivariate surface and volumetric data using linear mixed effects models and random field theory.
        NeuroImage. 2009; 47: S102
        • Hayasaka S.
        • Phan K.L.
        • Liberzon I.
        • Worsley K.J.
        • Nichols T.E.
        Nonstationary cluster-size inference with random field and permutation methods.
        Neuroimage. 2004; 22: 676-687
        • Worsley K.J.
        • Taylor J.E.
        • Tomaiuolo F.
        • Lerch J.
        Unified univariate and multivariate random field theory.
        NeuroImage. 2004; 23: S189-S195
        • Herculano-Houzel S.
        • Watson C.R.
        • Paxinos G.
        Distribution of neurons in functional areas of the mouse cerebral cortex reveals quantitatively different cortical zones.
        Front Neuroanat. 2013; 7https://doi.org/10.3389/fnana.2013.00035
        • Li M.
        • Santpere G.
        • Imamura Kawasawa Y.
        • Evgrafov O.V.
        • Gulden F.O.
        • Pochareddy S.
        • et al.
        Integrative functional genomic analysis of human brain development and neuropsychiatric risks.
        Science. 2018; 362eaat7615
        • Ashburner M.
        • Ball C.A.
        • Blake J.A.
        • Botstein D.
        • Butler H.
        • Cherry J.M.
        • et al.
        Gene Ontology: tool for the unification of biology [no. 1].
        Nature Genetics. 2000; 25: 25-29
        • The Gene Ontology Consortium
        The Gene Ontology Resource: 20 years and still GOing strong.
        Nucleic Acids Res. 2019; 47: D330-D338
        • Kang H.J.
        • Kawasawa Y.I.
        • Cheng F.
        • Zhu Y.
        • Xu X.
        • Li M.
        • et al.
        Spatio-temporal transcriptome of the human brain [no. 7370].
        Nature. 2011; 478: 483-489
        • Paquola C.
        • Seidlitz J.
        • Benkarim O.
        • Royer J.
        • Klimes P.
        • Bethlehem R.A.I.
        • et al.
        A multi-scale cortical wiring space links cellular architecture and functional dynamics in the human brain.
        PLOS Biology. 2020; 18e3000979
      5. Park B, Kebets V, Larivière S, Hettwer MD, Paquola C, Rooij D van, et al. (2021, November 2): Multilevel neural gradients reflect transdiagnostic effects of major psychiatric conditions on cortical morphology. bioRxiv, p 2021.10.29.466434.

      6. Paquola C, Garber M, Frässle S, Royer J, Tavakol S, Rodriguez-Cruces R, et al. (2021, November 22): The Unique Cytoarchitecture and Wiring of the Human Default Mode Network. bioRxiv, p 2021.11.22.469533.

      7. von Economo CF, Koskinas GN (1925): Die Cytoarchitektonik Der Hirnrinde Des Erwachsenen Menschen. J. Springer.

        • Scholtens L.H.
        • de Reus M.A.
        • de Lange S.C.
        • Schmidt R.
        • van den Heuvel M.P.
        An MRI Von Economo - Koskinas atlas.
        Neuroimage. 2018; 170: 249-256
        • Alexander-Bloch A.F.
        • Shou H.
        • Liu S.
        • Satterthwaite T.D.
        • Glahn D.C.
        • Shinohara R.T.
        • et al.
        On testing for spatial correspondence between maps of human brain structure and function.
        Neuroimage. 2018; 178: 540-551
        • Burt J.B.
        • Helmer M.
        • Shinn M.
        • Anticevic A.
        • Murray J.D.
        Generative modeling of brain maps with spatial autocorrelation.
        NeuroImage. 2020; 220117038
        • Desikan R.S.
        • Ségonne F.
        • Fischl B.
        • Quinn B.T.
        • Dickerson B.C.
        • Blacker D.
        • et al.
        An automated labeling system for subdividing the human cerebral cortex on MRI scans into gyral based regions of interest.
        NeuroImage. 2006; 31: 968-980
        • Schmaal L.
        • Hibar D.P.
        • Sämann P.G.
        • Hall G.B.
        • Baune B.T.
        • Jahanshad N.
        • et al.
        Cortical abnormalities in adults and adolescents with major depression based on brain scans from 20 cohorts worldwide in the ENIGMA Major Depressive Disorder Working Group.
        Mol Psychiatry. 2017; 22: 900-909
        • Hibar D.P.
        • Westlye L.T.
        • Doan N.T.
        • Jahanshad N.
        • Cheung J.W.
        • Ching C.R.K.
        • et al.
        Cortical abnormalities in bipolar disorder: an MRI analysis of 6503 individuals from the ENIGMA Bipolar Disorder Working Group.
        Molecular Psychiatry. 2018; 23: 932-942
        • Hoogman M.
        • Muetzel R.
        • Guimaraes J.P.
        • Shumskaya E.
        • Mennes M.
        • Zwiers M.P.
        • et al.
        Brain Imaging of the Cortex in ADHD: A Coordinated Analysis of Large-Scale Clinical and Population-Based Samples.
        AJP. 2019; 176: 531-542
        • Boedhoe P.S.W.
        • Schmaal L.
        • Abe Y.
        • Alonso P.
        • Ameis S.H.
        • Anticevic A.
        • et al.
        Cortical Abnormalities Associated With Pediatric and Adult Obsessive-Compulsive Disorder: Findings From the ENIGMA Obsessive-Compulsive Disorder Working Group.
        AJP. 2017; 175: 453-462
        • van Rooij D.
        • Anagnostou E.
        • Arango C.
        • Auzias G.
        • Behrmann M.
        • Busatto G.F.
        • et al.
        Cortical and Subcortical Brain Morphometry Differences Between Patients With Autism Spectrum Disorder and Healthy Individuals Across the Lifespan: Results From the ENIGMA ASD Working Group.
        Am J Psychiatry. 2018; 175: 359-369
        • Larivière S.
        • Paquola C.
        • Park B.
        • Royer J.
        • Wang Y.
        • Benkarim O.
        • et al.
        The ENIGMA Toolbox: multiscale neural contextualization of multisite neuroimaging datasets.
        Nat Methods. 2021; 18: 698-700
      8. Markello RD, Misic B (2020): Comparing spatially-constrained null models for parcellated brain maps. bioRxiv 2020.08.13.249797.

        • Sawyer S.M.
        • Azzopardi P.S.
        • Wickremarathne D.
        • Patton G.C.
        The age of adolescence.
        Lancet Child Adolesc Health. 2018; 2: 223-228
        • Meng X.
        • Rosenthal R.
        • Rubin D.B.
        Comparing correlated correlation coefficients.
        Psychological Bulletin. 1992; 111: 172-175
        • Collins C.E.
        • Airey D.C.
        • Young N.A.
        • Leitch D.B.
        • Kaas J.H.
        Neuron densities vary across and within cortical areas in primates.
        Proc Natl Acad Sci U S A. 2010; 107 (–15932)15927
        • Cahalane D.J.
        • Charvet C.J.
        • Finlay B.L.
        Systematic, balancing gradients in neuron density and number across the primate isocortex.
        Front Neuroanat. 2012; 6https://doi.org/10.3389/fnana.2012.00028
        • Carlo C.N.
        • Stevens C.F.
        Structural uniformity of neocortex, revisited.
        PNAS. 2013; 110: 1488-1493
      9. Braitenberg V, Schüz A (1998): Density of Axons. In: Braitenberg V, Schüz A, editors. Cortex: Statistics and Geometry of Neuronal Connectivity. Berlin, Heidelberg: Springer Berlin Heidelberg, pp 39–42.

      10. Economo C von (2009): Cellular Structure of the Human Cerebral Cortex ((L. C. Triarhou, editor)). Karger. Retrieved January 16, 2021, from https://www.karger.com/Book/Home/247637

        • Diedenhofen B.
        • Musch J.
        cocor: A Comprehensive Solution for the Statistical Comparison of Correlations.
        PLOS ONE. 2015; 10e0121945
        • Zou G.Y.
        Toward using confidence intervals to compare correlations.
        Psychological Methods. 2007; 12: 399-413
        • Berdenis van Berlekom A.
        • Muflihah C.H.
        • Snijders G.J.L.J.
        • MacGillavry H.D.
        • Middeldorp J.
        • Hol E.M.
        • et al.
        Synapse Pathology in Schizophrenia: A Meta-analysis of Postsynaptic Elements in Postmortem Brain Studies.
        Schizophrenia Bulletin. 2020; 46: 374-386
        • Srinivas K.V.
        • Buss E.W.
        • Sun Q.
        • Santoro B.
        • Takahashi H.
        • Nicholson D.A.
        • Siegelbaum S.A.
        The Dendrites of CA2 and CA1 Pyramidal Neurons Differentially Regulate Information Flow in the Cortico-Hippocampal Circuit.
        J Neurosci. 2017; 37: 3276-3293
        • Cannon T.D.
        • Chung Y.
        • He G.
        • Sun D.
        • Jacobson A.
        • van Erp T.G.M.
        • et al.
        Progressive Reduction in Cortical Thickness as Psychosis Develops: A Multisite Longitudinal Neuroimaging Study of Youth at Elevated Clinical Risk.
        Biological Psychiatry. 2015; 77: 147-157
        • Consortium T.B.
        • Anttila V.
        • Bulik-Sullivan B.
        • Finucane H.K.
        • Walters R.K.
        • Bras J.
        • et al.
        Analysis of shared heritability in common disorders of the brain.
        Science. 2018; 360https://doi.org/10.1126/science.aap8757
        • Taquet M.
        • Smith S.M.
        • Prohl A.K.
        • Peters J.M.
        • Warfield S.K.
        • Scherrer B.
        • Harrison P.J.
        A structural brain network of genetic vulnerability to psychiatric illness [no. 6].
        Mol Psychiatry. 2021; 26: 2089-2100
        • Alnæs D.
        • Kaufmann T.
        • van der Meer D.
        • Córdova-Palomera A.
        • Rokicki J.
        • Moberget T.
        • et al.
        Brain Heterogeneity in Schizophrenia and Its Association With Polygenic Risk.
        JAMA Psychiatry. 2019; https://doi.org/10.1001/jamapsychiatry.2019.0257
        • Kirov G.
        • Pocklington A.J.
        • Holmans P.
        • Ivanov D.
        • Ikeda M.
        • Ruderfer D.
        • et al.
        De novo CNV analysis implicates specific abnormalities of postsynaptic signalling complexes in the pathogenesis of schizophrenia [no. 2].
        Molecular Psychiatry. 2012; 17: 142-153
      11. Consortium TSWG of the PG, Ripke S, Walters JT, O’Donovan MC (2020): Mapping genomic loci prioritises genes and implicates synaptic biology in schizophrenia. medRxiv 2020.09.12.20192922.

        • Ball G.
        • Beare R.
        • Seal M.L.
        Charting shared developmental trajectories of cortical thickness and structural connectivity in childhood and adolescence.
        Hum Brain Mapp. 2019; 40: 4630-4644
        • Sotiras A.
        • Toledo J.B.
        • Gur R.E.
        • Gur R.C.
        • Satterthwaite T.D.
        • Davatzikos C.
        Patterns of coordinated cortical remodeling during adolescence and their associations with functional specialization and evolutionary expansion.
        PNAS. 2017; 114: 3527-3532
        • Zielinski B.A.
        • Prigge M.B.D.
        • Nielsen J.A.
        • Froehlich A.L.
        • Abildskov T.J.
        • Anderson J.S.
        • et al.
        Longitudinal changes in cortical thickness in autism and typical development.
        Brain. 2014; 137: 1799-1812
        • Di Martino A.
        • Fair D.A.
        • Kelly C.
        • Satterthwaite T.D.
        • Castellanos F.X.
        • Thomason M.E.
        • et al.
        Unraveling the Miswired Connectome: A Developmental Perspective.
        Neuron. 2014; 83: 1335-1353
        • Klingler E.
        • Francis F.
        • Jabaudon D.
        • Cappello S.
        Mapping the molecular and cellular complexity of cortical malformations.
        Science. 2021; 371https://doi.org/10.1126/science.aba4517
        • Rakic P.
        • Ayoub A.E.
        • Breunig J.J.
        • Dominguez M.H.
        Decision by division: making cortical maps.
        Trends in Neurosciences. 2009; 32: 291-301
        • Grasby K.L.
        • Jahanshad N.
        • Painter J.N.
        • Colodro-Conde L.
        • Bralten J.
        • Hibar D.P.
        • et al.
        The genetic architecture of the human cerebral cortex.
        Science. 2020; 367https://doi.org/10.1126/science.aay6690