Neurodevelopment and Schizophrenia

Since the early 1990s, developmental neurobiology has made important strides towards eluci-
dating the pathophysiology of psychiatric disorders. Nowhere has this link between basic science
and clinical insights become more evident than in the field of schizophrenia research. In this
volume, the editors bring together some of the most active investigators in this field. Each con-
tributor provides an up-to-date overview of the relevant research, including directions for further
investigation.
The book begins with a section on advances in developmental neurobiology. This is followed
by sections on etiological and pathophysiological developments, and models that integrate this
knowledge. The final section addresses the clinical insights that emerge from the developmental
models and sets the scene for future efforts at early detection and prevention of schizophrenia.
This book will be valuable to researchers in psychiatry and neurobiology, students in medicine
and psychology, and all mental health practitioners.

http://avaxhome.ws/blogs/ChrisRedfield
Neurodevelopment and
Schizophrenia

Edited by

Matcheri S. Keshavan
University of Pittsburgh and Western Psychiatric Institute and Clinic,
Pittsburgh, PA and Wayne State University, Detroit, MI, USA

James L. Kennedy
University of Toronto Centre for Mental Health and Addiction,
Toronto, Canada

and

Robin M. Murray
King’s College London and Institute of Psychiatry, London, UK
p u b l i s h e d b y t h e p r e s s s y n d i c at e o f t h e u n i v e r s i t y o f c a m b r i d g e
The Pitt Building, Trumpington Street, Cambridge, United Kingdom
cambrid ge university press
The Edinburgh Building, Cambridge CB2 2RU, UK
40 West 20th Street, New York, NY 10011–4211, USA
477 Williamstown Road, Port Melbourne, VIC 3207, Australia
Ruiz de Alarcón 13, 28014 Madrid, Spain
Dock House, The Waterfront, Cape Town 8001, South Africa
http://www.cambridge.org

C Cambridge University Press 2004

This book is in copyright. Subject to statutory exception and

to the provisions of relevant collective licensing agreements,
no reproduction of any part may take place without
the written permission of Cambridge University Press.

First published 2004

Printed in the United Kingdom at the University Press, Cambridge

Typefaces Minion 10.5/14 pt. and Formata System LATEX 2ε [tb]

A catalog record for this book is available from the British Library

Library of Congress Cataloging in Publication data

Neurodevelopment and schizophrenia / [edited by] Matcheri S. Keshavan,
James L. Kennedy, Robin M. Murray.
p. cm.
Includes bibliographical references and index.
ISBN 0 521 82331 5 (hardback)
1. Schizophrenia – Etiology. 2. Schizophrenia – Pathophysiology. 3. Developmental neurobiology.
4. Brain – Growth. I. Keshavan, Matcheri S., 1953– II. Kennedy, James L. III. Murray, Robin, M.
[DNLM: 1. Brain – growth & development. 2. Schizophrenia – etiology. 3. Brain – abnormalities.
4. Models, neurological. 5. Neurobiology – methods. WM 203 N49318 2004]
RC514.N4433 2004
616.89 8071 – dc22 2004045710

ISBN 0 521 82331 5 hardback

The publisher has used its best endeavors to ensure that the URLs for external websites referred to in this
book are correct and active at the time of going to press. However, the publisher has no responsibility for
the websites and can make no guarantee that a site will remain live or that the content is or will remain
appropriate.
Every effort has been made in preparing this book to provide accurate and up-to-date information that is
in accord with accepted standards and practice at the time of publication. Nevertheless, the authors, editors,
and publisher can make no warranties that the information contained herein is totally free from error, not
least because clinical standards are constantly changing through research and regulation. The authors,
editors, and publisher therefore disclaim all liability for direct or consequential damages resulting from the
use of material contained in this book. Readers are strongly advised to pay careful attention to information
provided by the manufacturer of any drugs or equipment that they plan to use.
 Contents

List of contributors page viii

Foreword xiii
Preface xvii

Part I Basic aspects

1 Genes and brain development 3

Timothy A. Klempan, Pierandrea Muglia, and James L. Kennedy

2 Brain development in healthy children and adolescents: magnetic

resonance imaging studies 35
Jay N. Giedd, Michael A. Rosenthal, A. Blythe Rose, Jonathan D. Blumenthal,
Elizabeth Molloy, Richard R. Dopp, Liv S. Clasen, Daniel J. Fridberg, and Nitin Gogtay

3 Cognitive development: functional magnetic resonance imaging studies 45

Beatriz Luna and John A. Sweeney

4 Cognitive development in adolescence: cerebral underpinnings, neural

trajectories, and the impact of aberrations 69
Stephen J. Wood, Cinzia R. De Luca, Vicki Anderson, and Christos Pantelis

5 Brain plasticity and long-term function after early cerebral insult: the
example of very preterm birth 89
Matthew Allin, Chiara Nosarti, Larry Rifkin, and Robin M. Murray

Part II Etiological factors

6 Do degenerative changes operate across diagnostic boundaries? The

case for glucocorticoid involvement in major psychiatric disorders 111
Carmine M. Pariante and David Cotter

7 Velo-cardio-facial syndrome (deletion 22q11.2): a homogeneous

neurodevelopmental model for schizophrenia 121
Stephan Eliez and Carl Feinstein

v
vi Contents

8 Can structural magnetic resonance imaging provide an alternative

phenotype for genetic studies of schizophrenia? 138
Colm McDonald and Robin M. Murray

9 Nutritional factors and schizophrenia 156

Sahebarao P. Mahadik

10 Schizophrenia, neurodevelopment, and epigenetics 174

Arturas Petronis

11 Early environmental risk factors for schizophrenia 191

Mary Cannon, Kimberlie Dean, and Peter B. Jones

12 Transcriptomes in schizophrenia: assessing altered gene expression with

microarrays 210
David A. Lewis, Karoly Mirnics, and Pat Levitt

13 Is there a role for social factors in a comprehensive development model

for schizophrenia? 224
Jane Boydell, Jim van Os, and Robin M. Murray

14 How does drug abuse interact with familial and developmental factors
in the etiology of schizophrenia? 248
Chih-Ken Chen and Robin M. Murray

Part III Pathophysiology

15 Developmental dysregulation of the dopamine system and the

pathophysiology of schizophrenia 273
Anthony A. Grace

16 The development of “mis-wired” limbic lobe circuitry in schizophrenia

and bipolar disorder 295
Francine M. Benes

17 Development of thalamocortical circuitry and the pathophysiology of

schizophrenia 310
Darlene S. Melchitzky and David A. Lewis

18 X chromosome, estrogen, and brain development: implications for

schizophrenia 330
Michael Craig, William Cutter, Ray Norbury, and Declan Murphy

19 Premorbid structural abnormalities in schizophrenia 347

Stephen M. Lawrie
vii Contents

20 Neurodegenerative models of schizophrenia 373

L. Fredrik Jarskog, John H. Gilmore, and Jeffrey A. Lieberman

21 Does disordered brain development occur across diagnostic

boundaries? 390
Christian W. Kreipke, David R. Rosenberg, and Matcheri S. Keshavan

Part IV Clinical implications

22 Can one identify preschizophrenia children? 415

Eugenia Kravariti, Paola Dazzan, Paul Fearon, and Robin M. Murray

23 High-risk studies, brain development, and schizophrenia 432

Matcheri S. Keshavan

24 Developmental models and hypothesis-driven early interventions in

schizophrenia 455
Matcheri S. Keshavan and Barbara A. Cornblatt

Index 473

The plates are situated between pages 190 and 191.

 Contributors

Matthew Allin Mary Cannon

Institute of Psychiatry at The Maudsley and Division of Psychological Medicine,
Department of Psychiatry, King’s College Institute of Psychiatry, De Crespigny
London, De Crespigny Park, London Park, Denmark Hill, London
SE5 8AF, UK SE5 8AF, UK

Vicki Anderson Chih-Ken Chen

Department of Psychology, University Department of Psychiatry, Chang Gung
of Melbourne and the Murdoch Memorial Hospital, 222 Mai-Chin Road,
Children’s Research Institute, Royal Keelung, Taiwan
Children’s Hospital, Melbourne,
Australia Liv S. Clasen
Child Psychiatry Branch, National Institute
Francine M. Benes of Mental Health, National Institutes of
Program in Structural and Molecular Health, Bethesda, MD 20892,
Neuroscience, McLean Hospital, Belmont USA
and Program in Neuroscience and
Department of Psychiatry, Harvard Barbara A. Cornblatt
Medical School, Boston, MA, Recognition and Prevention (RAP)
USA Program, 444 Lakeville Road, Lake Success,
NY 11042, USA
Jonathan D. Blumenthal
Child Psychiatry Branch, National Institute David Cotter
of Mental Health, National Division of Psychological Medicine,
Institutes of Health, Bethesda, Institute of Psychiatry, De Crespigny Park,
MD 20892, USA Denmark Hill, London SE5 8AF, UK

Jane Boydell Michael Craig

Institute of Psychiatry at The Maudsley and Division of Psychological Medicine,
Department of Psychiatry, King’s College Institute of Psychiatry, De Crespigny Park,
London, De Crespigny Park, London Denmark Hill, London
SE5 8AF, UK SE5 8AF, UK

viii
ix List of contributors

William Cutter Carl Feinstein

Division of Psychological Medicine, Stanford University School of Medicine,
Institute of Psychiatry, De Crespigny Park, 401 Quarry Road, Stanford, CA
Denmark Hill, London 94305–5719, USA
SE5 8AF, UK
Daniel J. Fridberg
Paola Dazzan Child Psychiatry Branch, National
Division of Psychological Medicine, Institute of Mental Health, National
Institute of Psychiatry, De Crespigny Institutes of Health, Bethesda, MD 20892,
Park, Denmark Hill, London USA
SE5 8AF, UK
Jay N. Giedd
Kimberlie Dean Child Psychiatry Branch, National
Division of Psychological Medicine, Institute of Mental Health, National
Institute of Psychiatry, De Crespigny Park, Institutes of Health, Bethesda, MD 20892,
Denmark Hill, London SE5 8AF, UK USA

Cinzia R. De Luca John H. Gilmore

Departments of Psychology and Psychiatry, Department of Psychiatry, UNC Conte
University of Melbourne and Melbourne Center for Neuroscience of Mental
Neuropsychiatry Centre, Sunshine Disorders, University of North Carolina
Hospital, Melbourne, School of Medicine, Chapel Hill, NC 27599,
Australia USA

Richard R. Dopp Nitin Gogtay

Child Psychiatry Branch, National Child Psychiatry Branch, National Institute
Institute of Mental Health, National of Mental Health, National Institutes of
Institutes of Health, Bethesda, MD 20892, Health, Bethesda, MD 20892,
USA USA

Stephan Eliez Anthony A. Grace

Division of Child and Adolescent Department of Neuroscience,
Psychiatry, University of Geneva School of 458 Crawford Hall, University of
Medicine, 41 Chemin Crts-de-Champel, Pittsburgh, Pittsburgh, PA 15206,
1206 Geneva, Switzerland USA

Paul Fearon L. Fredrik Jarskog

Division of Psychological Medicine, Department of Psychiatry, UNC Conte
Institute of Psychiatry at the Center for Neuroscience of Mental
Maudsley NHS Trust, De Crespigny Park, Disorders, University of North Carolina
Denmark Hill, London SE5 8AF, School of Medicine, Chapel Hill, NC 27599,
UK USA
x List of contributors

Peter B. Jones David A. Lewis

Department of Psychiatry, University of Departments of Psychiatry and
Cambridge, Addenbrooke’s Hospital, Neuroscience, University of Pittsburgh,
Cambridge CB2 2QQ, UK Pittsburgh, PA 15213, USA

James L. Kennedy
Jeffrey A. Lieberman
Centre for Addiction and Mental Health,
Department of Psychiatry, UNC Conte
University of Toronto, 250 College St,
Center for Neuroscience of Mental
Toronto, ON M5T1R8, Canada
Disorders, University of North Carolina
Matcheri S. Keshavan School of Medicine, Chapel Hill,
University of Pittsburgh School of NC 27599, USA
Medicine, Department of Psychiatry and
Western Psychiatric Institute and Clinic, Beatriz Luna
Pittsburgh and the Department of Laboratory of Neurocognitive
Psychiatry and Behavioral Neurosciences, Development, Western Psychiatric
Wayne State University School of Medicine, Institute and Clinic, University of
4201 St Antoine Boulevard, Pittsburgh Medical Center, 3501 Forbes
Detroit, MI 48201, Ave, Pittsburgh, PA 15213,
USA USA

Timothy A. Klempan
Sahebarao P. Mahadik
Centre for Addiction and Mental Health,
Department of Psychiatry and Health
University of Toronto, 250 College St,
Behavior, Medical College of Georgia and
Toronto, ON M5T1R8,
Medical Research Service, VA Medical
Canada
Center, 1 Freedom Way, Augusta,
Eugenia Kravariti GA 30904, USA
Division of Psychological Medicine,
Institute of Psychiatry, De Crespigny Park, Colm McDonald
Denmark Hill, London SE5 8AF, UK Division of Psychological Medicine,
Institute of Psychiatry, De Crespigny Park,
Christian W. Kreipke Denmark Hill, London SE5 8AF, UK
Cellular and Clinical Neurobiology,
Department of Psychiatry and Behavioral Darlene S. Melchitzky
Neurosciences, Wayne State University Department of Psychiatry, University of
School of Medicine, 540 E. Canfield, Pittsburgh, Pittsburgh, PA 15213 and
Detroit, MI 48201, USA Department of Biology, Mercyhurst
Stephen M. Lawrie College, Erie, PA 16546,
Division of Psychiatry, Royal Edinburgh USA
Hospital, Edinburgh EH10 5HF, UK
Karoly Mirnics
Pat Levitt Departments of Psychiatry and
Kennedy Center, Vanderbilt University, Neurobiology, University of Pittsburgh,
Nashville, TN 37235, USA Pittsburgh, PA 15213, USA
xi List of contributors

Elizabeth Molloy Arturas Petronis

Child Psychiatry Branch, National Institute The Krembil Family Epigenetics
of Mental Health, National Institutes of Laboratory, Centre for Addiction and
Health, Bethesda, MD 20892, USA Mental Health, 250 College St, Toronto,
ON M5T1R8, Canada
Pierandrea Muglia
Centre for Addiction and Mental Health, Larry Rifkin
University of Toronto, 250 College St, Institute of Psychiatry at The Maudsley and
Toronto, ON M5T1R8, Department of Psychiatry, King’s College
Canada London, De Crespigny Park, London
SE5 8AF, UK
Declan Murphy
Division of Psychological Medicine, A. Blythe Rose
Institute of Psychiatry, De Crespigny Park, Child Psychiatry Branch, National Institute
Denmark Hill, London SE5 8AF, UK of Mental Health, National Institutes of
Health, Bethesda, MD 20892, USA
Robin M. Murray
David R. Rosenberg
Institute of Psychiatry at The Maudsley and
Department of Psychiatry and Behavioral
Department of Psychiatry, King’s College
Neurosciences, Wayne State University
London, De Crespigny Park, London
School of Medicine, 4201 St Antoine
SE5 8AF, UK
Boulevard, Detroit, MI 48201, USA

Ray Norbury
Michael A. Rosenthal
Division of Psychological Medicine,
Child Psychiatry Branch, National Institute
Institute of Psychiatry, De Crespigny Park,
of Mental Health, National Institutes of
Denmark Hill, London SE5 8AF, UK
Health, Bethesda, MD 20892, USA

Chiara Nosarti John A. Sweeney

Institute of Psychiatry at The Maudsley and Department of Psychiatry, University of
Department of Psychiatry, King’s College Illinois at Chicago, The Psychiatric
London, De Crespigny Park, London Institute, 1601 W. Taylor St, Chicago,
SE5 8AF, UK IL 60612, USA

Christos Pantelis Jim van Os

Melbourne Neurospychiatry Centre Maastricht University, PO Box 616,
Department of Psychiatry, University of 6200 MD Maastricht, the Netherlands
Melbourne and Sunshine Hospital,
Melbourne, Australia Stephen J. Wood
Melbourne Neuropsychiatry Centre and
Carmine M. Pariante Department of Psychiatry, University of
Division of Psychological Medicine, Melbourne and Sunshine Hospital and The
Institute of Psychiatry, De Crespigny Park, Brain Research Institute, Melbourne,
Denmark Hill, London SE5 8AF, UK Australia
 Foreword

Although both Kraepelin and Bleuler noted that premorbid abnormalities in child-
hood could be present many years before a schizophrenic psychosis developed
(Marenco and Weinberger, 2000), until the late 1980s and 1990s, most people
viewed schizophrenia as an adult-onset mental illness. As a result, most biologi-
cal studies focused on a search for possible neurodegenerative changes that might
account for the onset of the condition. During the 1960s and 1970s (see Garmezy,
1974; Offord and Cross, 1969), evidence began to accumulate from developmen-
tally oriented follow-up and follow-back studies that abnormalities in interpersonal
relationships, neurodevelopmental immaturities, and attentional deficits in child-
hood all predicted the later onset of schizophrenia (see Rutter and Garmezy, 1983).
However, it was not until 1987 (Murray and Lewis, 1987; Weinberger, 1987) that
psychiatrists concerned with adult patients firmly took on board the notion that
schizophrenia might be a neurodevelopmental disorder. Since then, there has been
a veritable explosion of research tackling this proposition using a variety of research
strategies. In parallel, there has been an upsurge in studies of brain development
and function, giving rise to a much better understanding of brain plasticity and of
the role of neurotransmitters in both normal functioning and disease states. Clearly,
the time is ripe for a book that brings together the findings and concepts deriving
from both basic and clinic neuroscience, in order to examine the neurodevelop-
mental hypothesis in a critical but constructive fashion and then to consider the
implications for clinical practice. This book does just that, and does it extremely
well.
There is abundant consistent evidence that there are strong genetic influences on
the underlying liability to develop schizophrenia. Accordingly, one key challenge for
the neurodevelopmental hypothesis is to make explicit how neurodevelopmental
risk factors relate to genetic risk for schizophrenia. It is appropriate, therefore, that
the book begins with an account of what is known about genetic influences on brain
development, followed by reviews of what has been learned about normal brain
development from structural and functional magnetic resonance imaging studies.
Contrary to some people’s view that brain development is confined to infancy, it is
evident that there are important neural changes that extend into late adolescence
xiii
xiv Foreword

and early adult life. It remains the case that we have much to learn still about how
these changes relate to both normal and abnormal development, but they provide
a possible basis for the transition from the premorbid features of childhood to the
overt psychosis seen in adult life. This transition is also discussed with respect to
the evidence that there may be new abnormal brain changes that occur at this time.
The velo-cardio-facial syndrome is put forward as a possible model for the inter-
play between genes, brain, behavior, and cognition, and it is also suggested that
structural magnetic resonance imaging could provide a useful endophenotype for
genetic studies. This reflects the growing awareness that susceptibility genes prob-
ably do not act directly on mental disorders and that, therefore, it may be useful to
examine genetic effects on neurobiological abnormalities representing more prox-
imal effects of genes.
The neurodevelopmental hypothesis has been accompanied by indications of
the possible contributory causal role of epigenetic factors, malnutrition, pre- and
perinatal risk factors (including infection), minor physical anomalies, and adverse
rearing environment. The evidence on all of these is succinctly reviewed, with the
conclusion that the risk effects are probably real even though their effect size is
small.
Initially, with the advent of persuasive findings in genetics and in biological psy-
chiatry more generally, it became unfashionable to consider either social risk factors
or the effects of drug abuse: both of which had constituted a major focus of interest
in the 1960s and 1970s. However, animal studies and migration studies of humans
have shown that there is now reasonable evidence that social factors are influen-
tial, even though we do not understand how they operate. What is quite different
from the 1960s and 1970s is the appreciation that it is necessary to understand
how social factors may impact on brain development and that there may well be
a synergistic interplay between environmental risk and genetic vulnerability, with
the risks largely dependent on the presence of genetic susceptibility. Much the same
message derives from the study of recreational drugs, with the specific suggestion
that the drug effects associated with heavy early usage affect neurotransmitters
in ways that may precipitate the onset of psychosis when combined with genetic
susceptibility.
Further chapters provide a more detailed consideration of the possible role of the
dopamine system, of mis-wired limbic lobe or thalamocortical circuitry, of estrogen
and X-chromosome effects on brain development, of premorbid structural brain
abnormalities, and of neurodegenerative models.
From a clinical perspective, it is crucial to know whether the neurodevelopmen-
tal features are specific to schizophrenia or apply to a broader range of psychiatric
conditions. In frustrating fashion, the evidence suggests both substantial common-
alities and important specificities. A key chapter considers whether the extensive
xv Foreword

evidence of premorbid abnormalities means that it should now be possible to

identify children who are likely to develop schizophrenia. It is concluded that this
is not yet possible, yet combining risk factors from different functional domains
can usefully enhance the accuracy of predictive models. The final chapter discusses
how it may be possible for the pathophysiology to be explained by an integrative
neurochemical model and how there is now the beginning of a basis for possible
preventive interventions.
The book provides a rich intellectual meal of great interest and importance.
Nevertheless, we have to ask where all these creative ideas and empirical findings get
us. The book has two main achievements. First, it provides an excellent compilation
of what is known about the neurodevelopmental origins of schizophrenia, what the
findings explain, and what research challenges remain (together with invaluable
suggestions on how these challenges might be met). The evidence is compelling
that, in crucial ways, schizophrenia is a neurodevelopmental disorder and that
research using this concept has been immensely productive. There is the promise of
a better understanding of how neurodevelopmental features relate to genetic risk,
but that understanding has not yet been achieved. Similarly, although the onset in
late adolescence/adult life, long after premorbid manifestations have been evident in
childhood, is no longer quite the mystery that it was, the causal pathways have still to
be delineated. Nevertheless, the book clearly documents that neurodevelopmental
approaches are now mainstream in the study of schizophrenia and that much has
been learned in the last decade or so.
The second achievement is that the consideration of neurodevelopmental origins
of schizophrenia has thrown invaluable light on the broader issues involved in
normal and abnormal brain development. This gives the book an interest that
extends well beyond the world of schizophrenia clinicians and researchers. It is not
light bedtime reading, but the book does provide an engrossing read that is richly
rewarding.

Michael Rutter

REFERENCES

Garmezy, N. (1974). Children at risk: the search for the antecedents to schizophrenia. Schizophr
Bull 8: 14–55; 9: 90–125.
Marenco, S., Weinberger, D. R. (2000). The neurodevelopmental hypothesis of schizophrenia:
following a trail of evidence from cradle to grave. Dev Psychopath 12: 501–527.
Murray, R. M., Lewis, S. W. (1987). Is schizophrenia a neurodevelopmental disorder? Br Med
J 295: 681–682.
xvi Foreword

Offord, D. R., Cross, L. A. (1969). Behavioral anticedents of adult schizophrenia. Arch Gen
Psychiatry 21: 267–283.
Rutter, M., Garmezy, N. (1983). Developmental psychopathology. In Mussen’s Handbook of
Child Psychology, 4th edn, Vol. 4: Socialization, Personality, and Child Development, ed. M.
Hetherington, New York Wiley, pp. 775–911.
Weinberger, D. R. (1987). Implications of normal brain development for the pathogenesis of
schizophrenia. Arch Gen Psychiatry 44: 660–669.
 Preface

This volume arises out of a widely perceived need to take stock of our new know-
ledge about the developmental pathophysiology of schizophrenia. In 1997, we pub-
lished our first book on this topic, entitled Neurodevelopment and Psychopathology.
That volume, which was surprisingly well received, covered neurodevelopmen-
tal approaches to adult psychopathology, though many of the chapters concerned
schizophrenia. The enormous progress made in the subsequent years, particularly
on schizophrenia, has led us to focus the current volume exclusively on this disorder,
arguably the most debilitating of all psychiatric illnesses.
Since the early 1990s, our understanding of the developmental origins of
schizophrenia has “come of age,” with impressive advances emerging from both
the basic sciences and clinical studies. In retrospect, it is clear that the early neu-
rodevelopmental models that emerged in the mid-1980s were extremely simplistic
and too often relied on speculations about cellular and molecular mechanisms that
were not subsequently confirmed. Furthermore, to a large extent, they ignored the
contribution of psychology and certainly they included no mention of the role of
the social environment. We believe that the recent explosion of knowledge in both
neuroscience and cognitive science, as well as in imaging and epidemiology, has
allowed us to begin to remedy such deficiencies. Therefore, this new book brings
together many of the most productive and admired investigators in those areas of
research, individuals who we believe have contributed most to contemporary devel-
opmental models of schizophrenia. Each of the chapters provides a state-of-the-art
overview of the authors’ area of expertise, including directions for the future.
We start with a section on recent advances in developmental neurobiology. The
current state of our knowledge of genetics is reviewed, including the recent identi-
fication and apparent replication of susceptibility genes for schizophrenia (Ch. 1).
The latter is no small cause for rejoicing as claims for the identification of genes had
previously materialized and dematerialized with disturbing regularity, rather like
the sightings of alien spacecraft. Then, progress in our understanding of the normal
development of the human brain and its structural (Ch. 2), functional (Ch. 3), and
cognitive (Ch. 4) properties is successively outlined. Such a perspective is vital since
schizophrenia researchers have all too often attempted to outline the abnormal
xvii
xviii Preface

psychology and physiology underlying the condition before the normal had been
charted. The important field of brain plasticity and its limits is then discussed in
Ch. 5 with its implications for long-term functioning and psychopathology.
The next section contains overviews of pathophysiology and etiology of
schizophrenia. The lessons for schizophrenia from the study of unusual genetic
disorders such as the velo-cardio-facial syndrome are discussed, as is the impact of
genetic loading for schizophrenia on brain structure (Chs. 7 and 8). The important
and now well-documented effects of early environmental factors such as perinatal
complications (Ch. 11) and nutritional anomalies (Ch. 9) on risk of schizophrenia
are reviewed. Then the newer field of the effects of risk factors nearer to the onset
of psychosis is considered, such as psychosocial adversity (Ch. 13) and drug abuse
(Ch. 14) in pathogenetics, their epigenetic interaction (Ch. 10), and their impact
on gene expression (Ch. 12). Chapter 6 outlines a novel theory on the etiological
role of stress on schizophrenia and how its effects may be mediated by gluco-
corticoids. At a pathophysiological level, the developmental dysregulation of the
neurotransmitter systems such as dopamine (Ch. 15), and the limbic (Ch. 16)
and thalamocortical (Ch. 17) circuitry are critically reviewed. The important
role of X chromosome and estrogens in brain development and its relevance for
schizophrenia are discussed in Ch. 18. The possible premorbid neurodegenerative
changes in the schizophrenic illness (Ch. 20) and the commonalities versus differ-
ences between developmental/degenerative changes in common neuropsychiatric
disorders beginning in childhood/adolescence (Ch. 21) are discussed.
In the final section, some of the important clinical questions that drive pathophys-
iological research are considered. Can we identify preschizophrenia children? What
do studies of those at high genetic risk of the disorder teach us? Does understand-
ing of pathophysiology lead to specific predictions for preventive and therapeutic
approaches?
For too long, the origins of schizophrenia have been considered to be shrouded
in mystery. However, the exciting advances that have been made in schizophre-
nia research in the past decades, captured in this volume, have made the disease
more comprehensible, even though the puzzle continues to unravel. We also hope
that this volume will be of value for both researchers and practicing clinicians.
Throughout this volume, the implications of research findings for clinical practice
are discussed. The book will have served its purpose if the topics discussed herein
provided stimulation and new learning for a new generation of researchers and
clinicians in their efforts to inch toward better scientific knowledge and therapeutic
possibilities as applied to the patient with schizophrenia.
We wish to express our sincere gratitude to Drs Vaibhav Diwadkar, Debra
Montrose, Raj Rajarethinam, and Vandana Shashi for providing peer reviews of
the chapters in this book. We are in particular grateful to Karol L. Rosengarth for
her painstaking efforts in formatting and proofreading this work.
Part I

Basic aspects
 1

Genes and brain development

Timothy A. Klempan, Pierandrea Muglia, and James L. Kennedy
Centre for Addiction and Mental Health, University of Toronto, Toronto, Canada

The mechanisms underlying development of the mammalian central nervous sys-

tem (CNS) are of fundamental importance for research into psychiatric disorders.
The processes of neurulation, patterning, neuronal specification, and synaptogen-
esis, as well as the functional dynamics of neurotransmission, are governed by the
coordinated actions of products from a wide array of genes. Our knowledge of the
expression and complex interactions between the products controlling these pro-
cesses has been broadened by developmental studies using animal models (primarily
fruit fly and nematode but increasingly in the mouse); biochemical, histochemical,
and imaging studies; and analysis using high-throughput, non-selective techniques
such as microarray hybridization. Undoubtedly, the identification of novel brain-
expressed transcripts through the Human Genome Project has also provided a solid
framework for investigation of CNS function.
A neurodevelopmental etiology of schizophrenia is suggested by neuroimag-
ing and postmortem studies revealing significant and replicated lateral ventric-
ular enlargement, hippocampal and gray matter deficits, and cellular disarray,
independent of duration of illness and antipsychotic treatment. As these features
remain some of the best non-behavioral correlates of schizophrenia, the genetic
investigation of neurodevelopmental candidate genes, especially those with well-
characterized neural function and within chromosomal regions demonstrating
prior linkage or association with schizophrenia, is an important research focus.
This chapter will provide an overview of the major mechanisms involved in the
development of the mammalian CNS, with specific reference to the identity and pat-
terning of genes that are known to regulate these developmental phases (Table 1.1).
This pattern of gene expression will be related to the etiology of schizophrenia
through evidence provided by genetic, postmortem, imaging, electrophysiological,
and behavioral investigations of the disorder. In this fashion, a set of possible
determinants of both aberrant neurodevelopment and psychosis will be suggested.

Neurodevelopment and Schizophrenia, ed. Matcheri S. Keshavan et al. Published by Cambridge University
Press.

C Cambridge University Press 2004.

3
Table 1.1. Neurodevelopmental genes and reports of expression-based and genetic analysis
in schizophrenia

Gene Function Locus of interest Genetic and expression-based

analysis

Mash1 Neurulation -
Notch NOTCH4 (6p21.3) (Wei & Hemmings, 2000 and
see text)
Delta DLL1 (6q27)
Neurogenin NGN1 (5q23-q31)
NeuroD NEUROD (2q32)
Sonic Hedgehog SHH (7q36)
Wnt WNT1 (12q12-q13) (Cotter et al., 1998; Miyaoka
et al., 1999)
Krox20 Patterning EGR2 (10q21.1-q22.1)
Hox HOXB (17q21.3) (Kennedy and Kidd,
unpublished)
Dlx DLX1 (2q32)
Emx EMX2 (10q26.1)
Gbx GBX2 (2q36-q37)
Nkx TITF1 (14q13)
Otx OTX2 (14q21-q22)
Pax PAX6 (11p13) (Stober et al., 1999)
POU POU3F3 (3p14.2)
NCAM Cell migration/ NCAM1 (11q23.1) (Vicente et al., 1997; Doherty
neurite extension et al., 1990)
L1CAM
N-Cadherin NCAD (18q11.2)
Reelin RELN (7q22) (Fatemi et al., 2000; Guidotti
et al., 2000)
NGF Neuronal NGFB (1p13.1)
BDNF Survival BDNF (11p13) (Muglia et al., 2003)
NT-3 NTF3 (12p13) (Nanko et al., 1994 and see text)
NT-4/5 NTF5 (19q13.3)
GDNF GDNF (5p13.1-p12) (Lee et al., 2001)
CNTF CNTF (11q12.2) (Thome et al., 1996 and see text)
SNAP25 Presynaptic/ SNAP25 (20p11.2-p12) (Tachikawa et al., 2001; Wong
exocytosis et al., 2003)
Syntaxin STX1A (7q11.23) (A. Wong et al., unpublished data)
Synaptobrevin VAMP1 (12p)
Synapsin SYN3 (22q12.3) (Ohmori et al., 2000) and see text
Complexin CPLX2 (5q35.3) (Harrison and Eastwood, 1998
and see text)
Synaptophysin SYP (Xp11.23-p11.22)
Synaptotagmin SYT1 (12cen-q21)
5 Genes and brain development

Neurulation
The coordinated development of the human brain, from a single cell to some 1012
neurons with a possible 1000 connections between each, is a feat of unimaginable
intricacy. The function of individual groups of neurons within this framework in
the control of essential processes such as learning and memory, perception, mood,
and motor activity adds a further level of complexity. While the differentiative
pathways followed by neurons are topologically similar to those in other cell types,
one unique aspect of nervous system morphogenesis is the essential connectivity
of its units through axonal and dendritic outgrowth, allowing for interactions
in a highly plastic system. Despite the apparent difficulties of dissecting such a
detailed system, the changes involved in neuronal specification and overall CNS
development are becoming increasingly well understood. Much of our present state
of awareness in the field has been gathered through research using simple organisms
such as sea slug, squid and nematode, while recombinant DNA technology has
allowed generalization to vertebrates through investigations in mice. The cascade
of events proceeding from early gastrulation of the embryo to maturation can now
be adequately described.
The events leading to formation of the neural tube are known collectively as
neurulation. Two distinct developmental programs may be used: primary neurula-
tion, during which the chordamesoderm instructs the overlying ectoderm to divide,
invaginate, and separate from the surface to form the neural tube, and secondary
neurulation, during which a cylindrical zone of cells descends into the embryo and
hollows to form the tube. In mammals, this appears to occur in regionally differ-
entiated processes, with primary neurulation in the anterior region and secondary
neurulation posterior to somite 35.
The first signals responsible for the determination of anterior/posterior identity
in the ectoderm during gastrulation emanate from Hensen’s node, a small group
of cells at the anterior end of the primitive streak. The neural ectoderm is induced
through vertical signals from the mesodermal tissues beneath and adjacent to it.
The neural ectoderm gives rise to the neural plate, which is polarized through the
dorsal–ventral and anterior–posterior axes. The edges of the plate fold upwards
to join at the dorsal midline, producing the neural tube. Cells overlying the dorsal
midline of the neural tube form neural crest cells and migrate to form the brain and
spinal cord from the anterior and posterior portions, respectively. A commitment
to cell fates is revealed with the closure of the tube, with ventricular CNS generated
from the interior of the tube and epithelial cells at the interior periphery eventually
forming neurons and glia.
The brain becomes further divided through a series of constrictions into
the forebrain (prosencephalon), midbrain (mesencephalon), and hindbrain
6 T. A. Klempan, P. Muglia, and J. L. Kennedy

(rhombencephalon). Further compartmentalization results in regions within these

vesicles known as neuromeres, where mixing of cells becomes restricted to others
within the same region (as within rhombomeres of the hindbrain), laying the
ground plan for specification of neurogenesis. The forebrain gives rise to the telen-
cephalon (forming the olfactory lobes, hippocampus, basal ganglia, and cerebral
cortex in the adult brain following proliferation and folding) and the diencephalon
(forming the thalamus, epithalamus, hypothalamus, and the retina). The midbrain
forms the mesencephalon (including regions of connectivity between rostral and
caudal brain, the optic lobes, and tectum), while the hindbrain develops into the
metencephalon (producing the cerebellum and pons) and myelencephalon (form-
ing the medulla).
The localization and development of these highly specialized regions of the brain
is regulated principally through the actions of transcription factors: DNA-binding
proteins that direct the expression of specific genes. The phenotype of all neurons is
dictated by the precise set of genes expressed within the cell, allowing for functional
specialization. While the initial effect of transcription factors on the cell may be
brief, the influence of individual factors may be carried through numerous cell
divisions and persist beyond migration and diversification of a specific cell lineage.
The coordinated boundaries of expression of these factors serve to delineate clearly
morphological boundaries and to impart organization to the adult CNS. Much of
the information that we now possess regarding the determination of neural cell
fate and patterning in the early embryo has been made possible through research
conducted in Drosophila, Xenopus, and mice, where factors such as single neuroblast
ablation, ease of manipulation, and the ability to knock out specific genes are
advantageous. The extrapolation of findings from these organisms to humans is
facilitated by the remarkable degree of conservation of the factors controlling neural
fate across species.
The selection of distinct areas within the ectoderm to form neural precursor
cells is regulated through the expression of proneural genes, encoding transcrip-
tion factors of the basic helix-loop-helix (bHLH) class. The expression of these
genes provides the potential for the cell to become neural. Proneural genes are
typified by Mash1, encoding a factor in the central and peripheral nervous sys-
tems that selects for neural fate and acts in subsequent differentiation. Knockout
analysis of Mash1 reveals its role in neurogenesis within the ventral telencephalon
(Casarosa et al., 1999). Overexpression of the Xash3, the homologue of Mash1 in
Xenopus, leads to selection of neural over epidermal cell fate and ectopic neuronal
differentiation (Ferreiro et al., 1994). Lateral specification within the cluster of
proneural cells then allows specific cells within the equivalent group to become
neural. The transmembrane proteins encoded by the neurogenic genes Notch and
Delta permit the division between neuronal and epidermal cell fates to become
7 Genes and brain development

established in a mechanism incorporating cell-to-cell signaling and an amplified

feedback loop. Neural precursors express the Delta ligand, which, on binding to
the Notch receptor, represses further proneural gene expression and downregulates
the expression of Delta. When Notch transcription is absent from the embryo, all
ectodermal cells develop into neural precursors in Drosophila (Artavanis-Tsakonis
et al., 1983), while markers of neurogenesis are markedly increased in the absence
of Notch1 transcription in mouse (de la Pompa et al., 1997).
Neurogenins are proneural bHLH proteins that function as transcriptional acti-
vators of neuronal differentiation genes. The neurogenin Ngn1, for instance, pro-
motes neurogenesis and inhibits differentiation of neural stem cells into astrocytes
(Sun et al., 2001). Another neurogenin, Ngn2, restricts cell migration from the cor-
tex to the striatum (Chapouton et al., 2001) and is expressed in dorsal telencephalic
cells, defining a boundary through repression by Mash1, which defines the ventral
telencephalon (Parras et al., 2002). A regulator of differentiation known as NeuroD
is directly regulated by the neurogenins in many regions of the brain and maintains
expression in fully differentiated neurons (Lee, 1997). The overexpression of
NeuroD in Xenopus will direct the development of non-neural ectoderm into neur-
ons and accelerate the differentiation of neural precursors (Lee et al., 1995). A host
of additional factors act to convey positional information to groups of cells, instruct
the development of specific phenotypes through cascades of gene expression and
repression, and preserve the differentiated states of various cell lineages in the CNS.
Sonic hedgehog (Shh), a member of a vertebrate gene family corresponding to
the Drosophila gene hedgehog, is a secreted factor that acts in a concentration-
dependent fashion to induce cells of the floor plate and neural tube (Ericson et al.,
1995), in addition to its role in axial specification over various regions of the body.
Regionalization within the developing telencephalon is controlled by the ventraliz-
ing properties of Shh expression (Kohtz et al., 1998) and distinct dopaminergic and
serotonergic neuronal subpopulations are induced along the anterior–posterior axis
at different times by Shh signaling (Hynes and Rosenthal, 1999). Signaling mediated
by the Wnt family of glycoproteins is tightly connected to those controlled through
the hedgehog family. The expression of wnt1 (and the gene for fibroblast growth
factor 8) is coordinated with establishment of the mesencephalon–metencephalon
boundary; and mice lacking Wnt1 fail to develop midbrain structures and cerebel-
lum (McMahon and Bradley, 1990).
The recent description of a strong association between a promoter base-pair
substitution and the exon 1 (CTG)n repeat of NOTCH4 with schizophrenia (Wei
and Hemmings, 2000) in a region (6p21.3) previously associated with the disease
(Schwab et al., 1995; Straub et al., 1995) has ignited interest in this candidate gene. Of
six NOTCH homologues identified in vertebrates, NOTCH4 is the most divergent
phylogenetically (Kortschak et al., 2001) and shows a pattern of expression that
8 T. A. Klempan, P. Muglia, and J. L. Kennedy

is primarily endothelial and myocardial (Li et al., 1998) with minimal expression
in brain. A number of follow-up reports on NOTCH4 have failed to replicate the
original finding for both individual markers and haplotypes using case–control and
family-based association approaches (Fan et al., 2002; Imai et al., 2001b; Klempan
et al., 2001; McGinnis et al., 2001; Sklar et al., 2001; Swift-Scanlan et al., 2002; Ujike
et al., 2001). The extreme biases in transmission of NOTCH4 alleles witnessed in
the first study now appear more likely to be a false-positive association, although
an unusual population specific effect cannot be excluded since differences between
the African-American and European populations have been noted at the NOTCH4
locus (Luo et al., 2004).
Abnormalities of the Wnt signaling pathway have been suggested by several recent
studies which have described reductions of β-catenin and γ -catenin staining in the
CA3 and CA4 hippocampal subregions and increases in Wnt1 staining in these
regions of schizophrenic brains relative to controls (Cotter et al., 1998; Miyaoka
et al., 1999). Furthermore, levels of glycogen synthase kinase-3β (GSK-3β) are
significantly reduced in the prefrontal cortex of schizophrenia patients and Wnt
is known to act as a repressor of GSK-3β (Beasley et al., 2001). GSK-3β also par-
ticipates in apoptosis, a form of programmed cell death, and, therefore, aberrant
GSK-3β expression may provide a rationale for observations of irregular neuronal
distributions found in schizophrenia (Kozlovsky et al., 2002). Many of the signaling
components of the Wnt pathway have been localized and some of these map within
susceptibility regions for psychosis (Rhoads et al., 1999).

Segmentation
Patterning of the specialized cell groups that will eventually specify distinct regions
of the CNS is carried out by other groups of transcription factors, whose expression
is confined to discrete segments of the developing embryo. The compartmental-
ization granted by division of regions of the brain into segments (prosomeres in
the forebrain and rhombomeres in the hindbrain) prevents the mixing of various
lineages of cells and the activity of expressed genes and restricts the targets and
navigational properties of axons within these segments.
One gene, for example, that is known to be critical in the establishment of neural
tube boundaries is the zinc-finger transcription factor Krox20. Krox20 is expressed
in the neural plate in alternating segments (rhombomeres r3 and r5) prior to distinct
rhombomere formation, as revealed by lineage tracing studies in the chick. Mice that
are null for Krox20 display a disruption of segmental identity with a fused r2/r4/r6
region (Schneider-Manoury et al., 1997). The function of Krox20 is considered
comparable to that of the Drosophila pair-rule genes, which translate information
9 Genes and brain development

from previously expressed genes into periodic stripes of further expression. At the
early stages of segmental specification, however, there is very little conservation of
specific genes between flies and vertebrates.
One family of genes that is critical in the determination of regional identity along
the anterior–posterior axis is known as the homeobox (Hox) family. Based upon
homeotic genes in Drosophila, these master regulatory factors are strongly con-
served from fly to mammals. Homeobox proteins are characterized by a 60 amino
acid residue motif (the homeodomain), which binds to specific sequences of DNA.
The Hox genes show colinearity with positions of genes within clusters along the
chromosome corresponding to their domains of expression along the embryo.
Those genes located further 3 within the cluster are expressed both earlier during
development and in a more anterior location. The limits of rostral expression of
the Hox genes are strongly coordinated with the divisions between rhombomeres,
suggesting that they may be involved in the specification and/or maintenance of seg-
ment identity (Krumlauf, 1994). The influence of specific Hox members on actual
segment phenotype is illustrated by Hoxb1. The expression of Hoxb1 is particularly
strong in rhombomere r4 and loss of Hoxb1 in mutant mice transforms r4 to an r2
phenotype (Studer et al., 1996), while ectopic Hoxb1 expression in chick produces
the opposite result (Bell et al., 1999).
In addition to the Hox family of transcription factors, a number of other
homeodomain-containing proteins impart positional information to the devel-
oping brain. Mice lacking both Dlx1 and Dlx2, for instance, do not demonstrate
proper migration of cortical cells from subcortical regions and show disrupted cell
migration within the striatum (Anderson et al., 1997). In the absence of func-
tional Emx1 protein, the corpus callosum fails to develop, while Emx2 mutants
lack hippocampal dentate gyrus and Cajal–Retzius cells of the neocortex. These
Emx2 mutants are also similar to the reeler mouse, with disturbances of lamination
and neuronal migration (Mallamaci et al., 2000). The expression of Gbx2, required
in rostral hindbrain differentiation (Wassarman et al., 1997), acts in concert with
Otx2 to establish the isthmic signaling region (Broccoli et al., 1999). The function
of Otx2 is perhaps even more critical in neural induction, as targeted mutations of
Otx2 result in absence of rostral brain areas.
Another family of transcriptional regulatory factors expressed in the mammalian
forebrain with persistence into adulthood is known as the POU domain fam-
ily. Members of this group contain a POU homeodomain and a POU specific
domain and are generally expressed in restricted regions late in forebrain develop-
ment. The severe defects produced by mutations in many early patterning genes
(such as Hox) make these unlikely candidates for the subtle structural alterations
witnessed in schizophrenia; however, many members of the POU class of factors
10 T. A. Klempan, P. Muglia, and J. L. Kennedy

are expressed specifically in frontal cortical and hippocampal areas implicated in

disease processes (Turner et al., 1997). The POU-III subclass of factors includes
Brn-1, Brn-2, and SCIP (among others), which have been extensively studied in null
mutant mice. Brn-1 null mice show cellular disorganization of the hippocampus
and transitional cortex and disorganized cortical lamination, while Brn-2 mutants
are defective in hypothalamic development (Schonemann et al., 1995). Brn-1 and
Brn-2, coexpressed in layer II–V cortical neurons, have roles in the initiation of
radial migration (McEvilly et al., 2002). Reelin, also involved in radial migration of
cortical neurons, is reduced in a subpopulation of cortical plate neurons normally
colocalizing with Brn-1 expression in Brn-1 mutants, suggesting cross-regulation
between these pathways.
Transcriptional regulatory molecules can act in a hierarchical fashion to enhance
or inhibit the expression of downstream targets and sometimes may compensate
for others in their absence. The roles played by these factors can be diverse, acting
in neuroepithelial patterning, differentiation, and survival through cues given at
different times throughout embryogenesis. Numerous genes have been identified
as targets for these molecules, as seen in the enhancement of neuronal cell adhesion
molecule (NCAM) expression by Otx2 (Nguyen Ba-Charvet et al., 1999) and the
regulation of the SNAP25 promoter by Brn-3 POU transcription factors (Morris
et al., 1997). The cascade of gene expression induced by these factors would classify
them as interesting targets for investigation in schizophrenia, yet few studies have
been undertaken, perhaps because it is felt that the consequences of their misregula-
tion would be dire. One report has shown a mild association between a high-activity
variant of a Pax6 (paired-box family transcription factor) polymorphism, and para-
noid schizophrenia (Stober et al., 1999), while others have suggested that Pax6 may
contribute to schizophrenia through disrupted retinoic acid signaling (LaMantia,
1999).

Cell adhesion
In the developing brain, once a neuron has become committed to its phenotype it
must migrate to its proper layer of the maturing brain (Ruiz i Altaba, 1994). Cell
adhesion molecules (CAMs) are cell membrane proteins that mediate adhesion
between neural cells, exerting a key role in the morphogenesis, differentiation, and
migration of neurons as well as in the guidance of outgrowing axons in the develop-
ing brain (see Fig. 1.1). CAMs can be classified functionally into calcium-dependent
and calcium-independent groups. The calcium-dependent category contains at
least 80 proteins belonging to the cadherin superfamily, while the Ca2+ -independent
CAMs comprise the immunoglobulin (Ig) superfamily (IgSF). Most of the cadherin
11 Genes and brain development

NCAM
120

NCAM
180

NgCAM
L1

N-cadherin

CNR cadherin

Fig. 1.1. Members of the cadherin and immunoglobulin superfamilies of cell adhesion molecules.
Several prominent members of these transmembrane protein families are depicted, with
known roles in neurite outgrowth and morphogenesis. CNR, cadherin neuronal-related
receptor; NCAM, neuronal cell adhesion molecule; Ng CAM, neurogenin cell adhesion
molecule.

superfamily genes are expressed in the brain, and the protein structure is character-
ized by a unique domain named the cadherin motif, which is involved in calcium
binding (Takeichi, 1990). The cadherin motif (also called the EF motif) is repeated
a variable number of times in the different members of the superfamily (Yagi and
Takeichi, 2000). In addition to the classic cadherins, other members of the pro-
tocadherins superfamily are cadherin neuronal-related receptors (CNRs) and the
so-called seven-pass transmembrane cadherins. The seven-pass transmembrane
cadherins have a transmembrane structure similar to G-protein-coupled recep-
tors and appear to be involved in polarity orientation of the developing neuron,
as shown in Drosophila (Usui et al., 1999). The CNRs are coded by a cluster of
13 genes that map to 5q31.1 in humans and play a role in both the formation of
neuronal circuits at the synaptic level and the strengthening of the synapse during
long-term potentiation (LTP) (Yagi and Takeichi, 2000). It is interesting to note that
a significant linkage region for schizophrenia, derived from meta-analysis of data
from 20 genome scans, is located in a 30 cM stretch across 5q23–34 (Lewis et al.,
2003), which includes the CNR gene cluster. CNRs are distinguished from the other
12 T. A. Klempan, P. Muglia, and J. L. Kennedy

cadherins that exhibit homophilic interactions since they show etherophilic inter-
actions. They have been observed to bind to the Reelin protein and to the integrins
(Karecla et al., 1996; Senzaki et al., 1999). The genetic structure of the CNRs resem-
ble the organization of genes of the immunologic system, such as the T cell receptor
genes (Wu and Maniatis, 1999).
The calcium-independent CAMs (the IgSF) are characterized by a large amino-
terminal extracellular domain that contains a variable number of Ig motifs and
fibronectin repeats, which characterize the various members of the family sharing
a similar organization (Fields and Itoh, 1996). NCAM is the most extensively stud-
ied and characterized protein within the IgSF. NCAM contains five Ig folds and two
fibronectin repeats in the extracellular part of the protein (Crossin and Krushel,
2000). Further to its role during CNS development, NCAM appears to be involved
in the remodeling, regeneration, and maintenance of neuronal connections and
pathways in the adult brain (Kiss et al., 2001; Murase and Schuman, 1999). NCAM
also plays a role in learning and memory, and in guidance of brain development
in response to environmental stimuli (Fields and Itoh, 1996). Changes in NCAM
expression are also observed during synaptic remodeling and LTP in the rat hip-
pocampus (Fields and Itoh, 1996). NCAM is encoded by a single gene (NCAM1)
consisting of 26 exons located at 11q23.1 (Bello et al., 1989), which undergoes
alternative splicing generating at least 30 different forms with different spatial and
temporal patterns of expression in vertebrates (Akbarian et al., 1996). Three NCAM
isoforms (NCAM 180, NCAM 140 and NCAM 120) represent the most common
variants. NCAM 180 knockout mice have minor brain development abnormal-
ities, such as decreased size of the olfactory bulb, anterior ventricle enlargement,
and hippocampal dentate gyrus thinning (Andrews and Wood, 1986; Treloar et al.,
1997). In addition, the NCAM 180 knockouts exhibit behavioural deficits during
learning tasks and in the prepulse inhibition of startle (Wood et al., 1998). Both Hox
and Pax transcription factors appear to have a regulatory role on NCAM transcrip-
tion (Jones et al., 1993). NCAM functions as a cell–cell and cell–matrix homophilic
and heterophilic adhesion receptor; however, the mechanisms of interaction are not
completely understood. The binding of NCAM activates transmembrane-signaling
reactions and contributes to the initiation of cellular responses implicated in synap-
tic plasticity. The signaling downstream of neural NCAM not only impacts on
morphological changes but also can affect transcription factor activation and gene
induction (Krushel et al., 1999). Two main pathways are activated in response
to NCAM-mediated interactions, represented by the mitogen-activated protein
kinase and phospholipase C cascades (Crossin and Krushel, 2000). Other pathways
involve the interaction of the NCAM molecule with tyrosine kinase-linked recep-
tors (Beggs et al., 1997). The function of NCAM is also regulated by the molecule’s
content of polysialic acid (Fields and Itoh, 1996). The PSA residue makes the NCAM
13 Genes and brain development

less adhesive, thereby facilitating neuronal movement (Cunningham, 1995; Gower

et al., 1988). The addition of the PSA to NCAM is made possible by two sialyltrans-
ferases and their activity in directing polysialylated NCAM synthesis is controlled
at the mRNA level (Eckhardt et al., 1995; Nakayama et al., 1995). New interac-
tions and mechanisms are emerging from the study of NCAM. For example, it is
likely that NCAM binds to the brain-derived neurotrophic factor (BDNF) receptor
TrkB at the synapse level, since the deficient LTP observed in NCAM knockout
mice is recovered by addition of BDNF (Muller et al., 2000). It is also likely that
polysialylated NCAM can sensitize pyramidal cells to the action of BDNF as NCAM
knockout mice exhibit blunted activation of Trk B in response to BDNF (Muller
et al., 2000).
The role of NCAM in the schizophrenic brain has been investigated through
expression studies as well as molecular genetic association and linkage studies.
One study found a decreased number of neurons expressing the polysialylated
form of NCAM in the hippocampus of brains from individuals with schizophrenia
(Doherty et al., 1990). The abnormal levels of polysialylated NCAM in the brain of
schizophrenics might be responsible for the commonly observed abnormalities of
migration. This finding stimulated one group (Vicente et al., 1997) to investigate
whether variants in the NCAM gene were associated with susceptibility to develop
schizophrenia. Linkage analysis in five multiplex families was performed and an
association analysis was conducted in 71 nuclear families with one affected off-
spring. Linkage analysis of the polymorphic dinucleotide (CA) repeat in the region
of the NCAM gene, considering the disease as either autosomal dominant or reces-
sive, found no evidence of linkage. The association studies in the nuclear families
also did not find association between another (CA) repeat found at the NCAM
locus and schizophrenia. Subsequent to the initial report, a number of studies
have revealed NCAM alterations in schizophrenia, including decreased polysialy-
lated NCAM in the hippocampus (Barbeau et al., 1995), increased NCAM in the
prefrontal cortex and hippocampus (Honer et al., 1997), and increased NCAM
in cerebrospinal fluid (van Kammen et al., 1998). Several lines of evidence have
shown that other CAMs such as L1, fibroblast growth factor receptor, laminin, and
integrins interact with NCAM (Doherty and Walsh, 1996; Kristiansen et al., 1999;
Takei et al., 1999).
A major role in the layering of neurons into their specific target area of the CNS
once their migration is completed in several brain areas, including the hippocam-
pus, is ascribed to Reelin (Rice and Curran, 2001). The spontaneous mutation
occurring in the gene reelin in mice bred at the Institute of Animal Genetics in
Edinburgh (Falconer, 1951) has initiated five decades of research into reelin func-
tion and its involvement in brain development. The reeler mouse displays impaired
motor coordination, ataxia, and tremor evident since day 12 after birth. The reeler
14 T. A. Klempan, P. Muglia, and J. L. Kennedy

mutation is also associated with severe cerebellar hypoplasia, disorganization of

cerebral and cerebellar cortices, and disruption of other structures such as the
hippocampus (Rakic and Caviness, 1995). In spite of the altered neuronal position-
ing, most neuronal fibers make appropriate connections in the reeler brain although
with longer and more intricate paths (Steindler and Colwell, 1976). Yet, connec-
tions are aberrant in cerebellar neurons. Several known alleles in reelin determine
very similar phenotypes in the mouse (D’Arcangelo and Curran, 1998).
Reelin is a large protein coded by the reelin gene (Reln), located on chromo-
some 5 in the mouse (Goffinet and Dernoncourt, 1991). In humans, the gene
(RELN) maps to 7q22 (DeSilva et al., 1997). It contains 65 exons and encodes a
mRNA of approximately 12 kb, which translates into a 3461 amino acid residue
protein with 94% homology to the murine amino acid sequence. Reelin is abun-
dantly synthesized and secreted in the extracellular matrix during early stages of
brain development by the Cajal–Retzius neurons located in the marginal zone. It
has been shown in both rodents and primates that, once the Cajal–Retzius cells dis-
appear, a few months before gestation in primates (Rodriguez et al., 2000), Reelin is
synthesized throughout life by a select population of interneurons utilizing gamma-
aminobutyric acid (GABA) including cells within layers I and II of various cortical
areas (Rodriguez et al., 2000).
In the search for etiological factors underlying the neurodevelopmental anom-
alies described in schizophrenia brains, the expression pattern of Reelin has been
investigated. According to two independent findings (Fatemi et al., 2000; Guidotti
et al., 2000), schizophrenia brains show a 30 to 60% reduction of RELN expression
in the prefrontal cortex and in the hippocampus. Patients with bipolar disorder and
associated psychotic features also show a reduction in RELN expression (Guidotti
et al., 2000). Several polymorphisms have been described for RELN including a tri-
nucleotide repeat in the 5 untranslated region (UTR) region of the gene. One study
has found an association between this polymorphism and autism in a family-based
association study (Persico et al., 2001); however, at present, few studies (Akatone
et al., 2002) have investigated the possible role of RELN sequence variants in
schizophrenia.

Neuronal survival
With the development of the nervous system, as axons innervate their appropriate
target regions and connections become established, a naturally occurring cell death
takes place to refine the complex network of connections. This process serves to
establish a balance between the requirements of the target tissue and the characteris-
tics of the innervating neuronal population (Purves, 1988). The primary molecules
involved in this target-derived regulation of neuronal survival and differentiation
15 Genes and brain development

BDNF
NGF NT-4/5 NT-3 ALL

N T R

TrkA TrkB TrkC P75

Fig. 1.2. Specificity of neurotrophic factor binding to tyrosine kinase receptors. All factors activate
receptors of the trk class. While neurotrophin 3 acts weakly through TrkA and TrkB in addition
to TrkC, all factors can act through the low-affinity receptor p75NTR . BDNF, brain-derived
neurotrophic factor; NGF, nerve growth factor; NT, neurotrophin.

are the neurotrophins (see Fig. 1.2). The neurotrophin family of growth factors
display functional similarity through the promotion of neuronal survival, as first
seen with nerve growth factor (NGF) (Levi-Montalcini, 1987) and later for BDNF
(Leibrock et al., 1989). Neurotrophins now also include neurotrophin 3 (NT-3)
(Ernfors et al., 1990; Hohn et al., 1990), neurotrophin 4/5 (NT-4/5) (Berkmeier
et al., 1991; Ip et al., 1992), and neurotrophin 6 (NT-6) (Gotz et al., 1994), all cloned
through homology to NGF and BDNF. These also play key roles in neuronal differ-
entiation, cell body growth, promotion of neurite sprouting/extension, and, in the
mature CNS, the control of short-term synaptic transmission and LTP, considered
a mechanism for the processes of memory and learning (Thoenen, 1995).
Neurotrophin signaling is thought to occur primarily through the Trk family
of tyrosine kinase receptors. Binding of neurotrophin ligand induces transphos-
phorylation of Trk receptors, initiating signal transduction through phosphory-
lation of further intracellular proteins. Specificity is exhibited in the binding of
neurotrophins for individual Trk receptors: TrkA is a receptor for NGF, TrkB is a
receptor for BDNF and NT-4/5, and TrkC is a receptor for NT-3. A further receptor
known as p75NTR binds all neurotrophins with low affinity and acts selectively to
increase or decrease the response to different neurotrophins in neurons express-
ing Trk receptors. In the absence of Trk receptors, p75NTR controls an apoptotic
response to both NGF and BDNF in some neurons.
16 T. A. Klempan, P. Muglia, and J. L. Kennedy

The domains of expression of neurotrophins and their cognate high-affinity

receptors have been extensively mapped in both the peripheral and the central
nervous systems. The expression of TrkA and NGF in the CNS is localized pri-
marily to basal forebrain, striatal cholinergic neurons, and cerebellar Purkinje cells.
Neuronal populations responsive to BDNF and, to a lesser extent, NT-4/5 comprise
basal forebrain cholinergic, hippocampal, trigeminal mesencephalic, and substan-
tia nigra dopaminergic neurons, as well as motor neurons, retinal ganglion and
cerebellar granule cells. The expression of NT-3 and TrkC is initially strong in the
hippocampus and later becomes restricted to the dentate gyrus, basal forebrain, and
adrenergic neurons of the locus coeruleus. Each of these ligands and receptors have
also been knocked out in transgenic mice for assessment of functionality. While
profound losses of neurons are seen in the peripheral nervous system, deficits in the
CNS are generally quite mild and the situation is apparently more complex. It is cur-
rently believed that significant redundancy in neurotrophin activity occurs within
the CNS, as cross-talk between neurotrophins and receptors has been described and
receptor knockouts are generally more severe than their ligand counterparts (Snider,
1994). The neurotrophins play vital roles in cell survival and differentiation in all
regions of the brain and may be associated with reductions in cortical and whole
brain volume (Lawrie and Abukmeil, 1998) and with the high neuronal density
(Selemon et al., 1995) seen in morphometric analyses of the schizophrenic brain.
Significant alterations of BDNF protein levels have been seen in postmortem
studies of schizophrenia patients compared with normal controls, specifically
increases in cortical regions and decreases in the hippocampus (Brouha et al., 1996;
Durany et al., 2001), although others have seen the opposite effect in hippocampus
(Takahashi et al., 2000). It has also been observed that treatment with haloperidol
and risperidone decreases the expression of BDNF in frontal cortex, occipital cor-
tex, and hippocampus of rat brain (Angelucci et al., 2000; Takahashi et al., 2000)
and alters the level of TrkB expression in similar regions. These findings complicate
the interpretation of human expression studies, suggesting that results in the hip-
pocampus may be a confound of neuroleptic exposure, and further indicating that
these antipsychotic medications may be unlikely to have an effect through media-
tion of neurotrophin levels. Nevertheless, the reductions in BDNF-deficient mice
of immunoreactivity for neuropeptide Y, the calcium-binding proteins calbindin
and parvalbumin, somatostatin, and cholecystokinin indicate that neuronal matu-
ration and calcium trafficking is perturbed. Research using rat brain striatal slices in
vitro shows that BDNF induces release of GABA, dopamine, and serotonin (Goggi
et al., 2002). A direct induction of dopamine D3 receptor expression by BDNF has
also been described during early development and adulthood (Guillin et al., 2001),
linking molecular findings between two factors of potential etiological significance
in schizophrenia.
17 Genes and brain development

Genetic investigation of the gene BDNF in schizophrenia was initiated follow-

ing discovery of a dinucleotide repeat polymorphism located within what was
initially recognized as the promoter (now determined as intronic) region of the
gene (Proschel et al., 1992). The first published reports included a Japanese case–
control study (Sasaki et al., 1997a) and an Irish study using over 200 cases and
controls for association analysis and 250 families for linkage disequilibrium anal-
ysis (Hawi et al., 1998); both reports were negative for this polymorphism. While
the small sample size of the first study may limit any conclusions, the high power
of the sample from the second report discredits a role for BDNF (although vari-
ations in allele frequencies and population effects for this microsatellite may con-
tribute to these findings). Recently, additional studies have also been negative
for association of BDNF with schizophrenia; however, these have provided sup-
port for an effect of BDNF alleles on parietal lobe volume (Wassink et al., 1999)
and shown increased dinucleotide repeat marker length with treatment-responsive
schizophrenia (Krebs et al., 2000). The results of transmission disequilibrium
test-based analysis of a schizophrenia sample of Italian and Canadian families
(Muglia et al., 2003) have shown increased transmission of the 170 bp allele and
reduced transmission of the 174 bp allele to probands. Of further interest, a recent
study (Egan et al., 2003) has shown that the Val66Met polymorphism of BDNF
influences intracellular distribution of the precursor protein, alters hippocampal
activity as measured by functional magnetic resonance imaging (FMRI), and affects
performance on episodic memory tasks. The availability of extensive sequence data
for the BDNF gene should facilitate its further examination as a putative schizophre-
nia susceptibility factor.
The NT-3 gene, with significant hippocampal expression (Maisonpierre et al.,
1990), has been extensively examined following a positive association between a
highly polymorphic dinucleotide repeat polymorphism of NT-3 and schizophrenia
in a Japanese study (Nanko et al., 1994). This group also identified a missense muta-
tion in the NT-3 gene that was associated with severe schizophrenia (Hattori and
Nanko, 1995). Subsequent investigations of NT-3 have failed to replicate these find-
ings (Arinami et al., 1996; Jonsson et al., 1997; Nimgaonkar et al., 1995; Thome et al.,
1997a). One study has replicated the original finding, but only in male schizophren-
ics (Dawson et al., 1995). Recent reports have shown an association of illness with
NT-3 in female probands (Virgos et al., 2001) and a reduction of NT-3 expres-
sion in frontal and parietal cortical regions using enzyme-linked immunosor-
bant assays (Durany et al., 2001). These diverse results are difficult to reconcile,
although heterogeneity of samples and methods is a likely contributor to the
inconsistency.
Two additional molecules not structurally related to the neurotrophins that
serve to promote survival of neurons during development are glial cell-derived
18 T. A. Klempan, P. Muglia, and J. L. Kennedy

neurotrophic factor (GDNF) and ciliary neurotrophic factor (CNTF). GDNF is

related to transforming growth factor-β, which initiates induction of mesoderm
and is a trophic factor for midbrain dopaminergic (Lin et al., 1993), cholinergic
(Ha et al., 1996), and noradrenergic (Arenas et al., 1995) neurons. The role of
GDNF in differentiation of dopaminergic neurons has prompted its study as a puta-
tive susceptibility locus for schizophrenia; however, negative results were obtained
for a polymorphic trinucleotide repeat marker of GDNF in a recent association
study (Lee et al., 2001). CNTF is a cytokine that promotes the survival of various
populations of neurons during development (Sendtner et al., 1994) through acti-
vation of multiple signaling pathways. While the presence of a CNTF null mutation
initially appeared to associate with schizophrenia in one sample studied (Thome
et al., 1996), further investigation by other groups has been predominantly negative
(Arinami and Toru, 1996; Gelernter et al., 1997; Sakai et al., 1997b; Thome et al.,
1997a; Virgos et al., 2001). Some associations, however, have been made to schizoaf-
fective disorder (Tanaka et al., 1998) and other schizophrenia subtypes (Thome
et al., 1997b).

Neurotransmission
The process of neurotransmission at its most fundamental level involves regulated
exocytosis through recruitment of synaptic vesicles, fusion to the plasma mem-
brane and neurotransmitter release, and eventual re-establishment of the vesicle
pool through endocytosis. The processes and molecules associated with vesicu-
lar trafficking are also critical for cell division, fertilization, and secretion of hor-
mones and other proteins, and they are strongly conserved from yeast to human.
Exocytosis is regulated most centrally by a group of proteins known as soluble N-
ethylmaleimide-sensitive attachment factor protein receptors (SNAREs), which are
small (18–42 kDa) integral (or anchored) membrane proteins that interact with
each other through amphipathic helical domains known as SNARE motifs. The
three core SNARE proteins can be categorized into target-SNAREs (t-SNAREs)
such as SNAP-25 (synaptosomal-associated protein of 25 kDa) and the syntaxins,
and vesicle-SNAREs (v-SNAREs) such as the synaptobrevins (or VAMPs). While
this core complex of SNARE proteins is known to regulate vesicular fusion (Hodel,
1998), the cyclical and quantal aspects of neurotransmitter release are facilitated
through a host of synaptic protein interactions, which are seen to be quite strongly
location-specific among regions of the brain (Bark and Wilson, 1994). The array
of synaptic terminal proteins involved in generalized neurotransmission includes
the core complex and interactors, which are either soluble or bound to the synaptic
vesicle or plasma membrane, such as soluble SNAPs, N-ethylmaleimide sensitive
19 Genes and brain development

Synaptic
Synapsin
vesicle
ATP ADP

NSF
Synaptobrevin/VAMP

Syntaxin SNAP-25

Presynaptic membrane

Fig. 1.3. Synaptic terminal proteins involved in exocytosis. The core complex required for vesicle
fusion to the presynaptic membrane includes synaptosomal associated protein of 25 kDa
(SNAP-25), syntaxins, and synaptobrevins (or VAMP). Proteins such as N-ethylmaleimide-
sensitive factor (NSF), which recycle this complex, and synapsins, which interact between
the vesicle and the cytoskeleton, are also involved, as well as numerous others.

factor (NSF), synapsins, synaptophysins, synaptogyrins, synaptotagmin, dynamin,

and complexins, among others (see Fig. 1.3).
Vesicles are recruited in a calcium-dependent process to active zones on
the presynaptic membrane. It has been suggested that vesicle-bound synapsins
contribute during the recruitment phase by releasing actin following phosphory-
lation by calmodulin-dependent protein kinase II, permitting vesicles to untether
from the cytoskeleton. Vesicle docking at appropriate sites on the plasma membrane
is regulated through a strong connection between vesicle and membrane-bound
SNAREs. A slow, ATP-driven priming event then occurs, driven by interaction of
the ATPase NSF with the primary SNAREs through soluble α-SNAPs, forming a
stable 20S complex. A number of accessory proteins are capable of binding to indi-
vidual SNARE members, including munc-18, syntaphilin, and CDCrel-1, which
bind to syntaxin; synaptophysin, which binds to VAMP; and synaptotagmin, which
binds to SNAP-25. Changes in the degree of expression or interaction of specific
presynaptic proteins are likely to constitute a major element in the determination of
brain structure and function, and the genes encoding these proteins are becoming
a topic of investigation in schizophrenia research.
SNAP-25, a t-SNARE that binds to the plasma membrane on palmitoylation,
shows a pattern of expression coinciding with synaptogenesis during early embry-
onic differentiation (Catsicas et al., 1991; Lou and Bixby, 1995). SNAP-25 is most
strongly expressed in neurons of neocortex, hippocampus, anterior thalamic and
20 T. A. Klempan, P. Muglia, and J. L. Kennedy

pontine nuclei, and cerebellar granule cells (also pancreatic β-cells and adrenal
chromaffin cells), and it is associated with regions of synaptic plasticity. Antisense
oligonucleotides to SNAP-25 inhibit axonal elongation in rat cortical neurons,
specifically preventing complete terminal differentiation (Osen-Sand et al., 1993).
A strain of mouse designated coloboma, with a 2 cM deletion of a region of chro-
mosome 2 containing the Snap-25 gene, exhibits hyperactivity and delayed devel-
opment (Hess et al., 1996; Heyser et al., 1995). This strain also has abnormalities
in dopaminergic, serotonergic, and glutamatergic transmission in dorsal striatum
(Raber et al., 1997). This suggests that it may act as a model of attention-deficit
hyperactivity disorder and other psychiatric conditions. Hyperactivity is corrected
on introduction of a Snap-25 transgene into the strain, indicating the specificity of
this feature to the gene (Hess et al., 1996).
Postmortem assays of SNAP-25 immunoreactivity in schizophrenia have revealed
changes in the inferior temporal and prefrontal association cortices (Thompson
et al., 1998), with decreased expression in Brodmann areas 10 and 20 and increased
expression in area 9. A study of hippocampal connectivity found reduced cortical
SNAP-25 protein, most notably in the terminal fields of entorhinal cortex pro-
jections of schizophrenics (Young et al., 1998). More recently, decreased SNAP-25
expression has been seen in the cerebellum of schizophrenics (Mukaetova-Ladinska
et al., 2002), further implicating this molecule in the etiology of the disease. Altered
levels of SNAP-25 have also been witnessed in cerebral spinal fluid by quantitative
dot blotting (Thompson et al., 1999), with significant elevation of SNAP-25 over
that in control subjects and those with headaches. A further study was able to con-
firm the original SNAP-25 findings in area 10 (along with reduced synaptophysin
expression) yet showed no alteration in SNAP-25 mRNA expression (Karson et al.,
1999). Enhanced cortical expression of SNAP-25 has been identified through sub-
traction suppression polymerase chain reaction (PCR) in several rodent models of
schizophrenia, including haloperidol-treated versus control and Fischer 344 versus
Lewis rats (Wong et al., 2003).
The human SNAP-25 gene has been localized to 20p11.2-12 (Maglott et al., 1996),
near a region identified in a multi-center schizophrenia genome scan (Moises et al.,
1995). Three single nucleotide polymorphisms have been identified (Barr et al.,
2000) in the 3 untranslated region of SNAP-25, and the transmission of alleles
of these polymorphisms have been evaluated in 200 small, nuclear schizophrenia
families. Preliminary results indicate unbiased transmission of SNAP-25 alleles and
haplotypes within these families (Wong et al., 2003). A polymorphic tetranucleotide
repeat identified in the SNAP-25 promoter region also shows no association with
affection, subtypes, or family history in a recent case–control study (Tachikawa et al.,
2001); consequently, further investigation of additional markers and sample groups
is warranted.
21 Genes and brain development

The synapsins comprise a family of phosphoproteins localized to the cytoplas-

mic surface of synaptic vesicles. Until recently, two differentially spliced synapsin
genes (synapsin I at Xp11.23 and synapsin II at 3p25) have been identified, with
well-characterized roles in synaptic vesicle release (Jovanovic et al., 2000; Rosahl
et al., 1995) and efficiency of neurotransmission (Greengard et al., 1993). A third
member of this family, synapsin III, with neuron-specific expression and extensive
structural similarity to other members, was recently described (Kao et al., 1998)
and localized to chromosome 22q12-13, distal to the psychosis-associated dele-
tion region responsible for DiGeorge and velo-cardio-facial syndrome. A role of
synapsins in development of the presynaptic terminal is indicated by studies in rat
neuroblastoma cells, which showed that overexpression of synapsin IIb produces
an increase in the number of varicosities and the synaptic vesicle content of the
varicosities (Han et al., 1991).
Decreased expression of synapsin I protein, as determined through quantitative
Western blotting, has been observed in the hippocampus of postmortem brains
from individuals with schizophrenia (Browning et al., 1993). Further research
shows age-related abnormalities in expression of synapsin Ia and Ib, suggested
by data showing that mRNA expression of these isoforms in schizophrenics under
age 75 is approximately twofold higher in the left middle and superior temporal
gyrus than in age-matched controls, while normalizing in individuals above this
age (Tcherepanov and Sokolov, 1997). A preliminary investigation into the poss-
ible role of synapsin I variants in schizophrenia has given negative results for one
microsatellite polymorphism in over 100 families using both linkage- and linkage
disequilibrium-based approaches (A. M. Vicente and J. L. Kennedy, unpublished
data).
Reduced transcripts for synapsin II (among others, including NSF, synaptotag-
min 5, and synaptojanin I) were noted during microarray expression profiling of
schizophrenic versus matched control prefrontal cortical brain tissue (Mirnics et al.,
2000). This study demonstrated highly significant changes in expression of the
presynaptic gene group, among 250 other groups examined, which were verified
using in situ hyridization and a second cohort for confirmation. These changes
were quite consistent in the sample studied (nine of ten and ten of ten pairs showed
reduction of synapsin II and NSF, respectively) and appeared unrelated to admin-
istration of antipsychotic medication based on patient history and comparison of
arrays between haloperidol-treated and control monkeys. The report suggests the
presence of global rather than individual modifiers of presynaptic gene expression,
as potentially dictated by pre-existing changes in developmental molecules, aber-
rant cytoarchitecture, or survival/connectivity in the prefrontal cortex. It can be
suggested that schizophrenia is, in fact, a disease of the synapse, of polygenic inher-
itance, and determined by alterations in genes related to trafficking and regulated
22 T. A. Klempan, P. Muglia, and J. L. Kennedy

neurotransmission or their antecedents, with onset related to maturational pro-

cesses occurring in puberty (Mirnics et al., 2001). Combined with this finding, an
early study of synapsin II detected increased numbers of molecular weight variants
of the protein in postmortem brain tissue of schizophrenic and alcoholic individuals
in comparison with controls (Grebb and Greengard, 1990). A recent attempt to con-
firm the observations of the Mirnics study on NSF and synapsin II expression using
immunoblotting and real-time PCR has been negative (Imai et al., 2001a), raising
doubts as to the generalizability of the findings.
The localization of the synapsin III gene to a segment of chromosome 22q,
associated with a 25-fold elevation in rates of psychiatric disorders and elevated
numbers of deletions in individuals with schizophrenia (Murphy et al., 1999), has
made this synaptic candidate gene a promising area of study. However, to date,
at least six studies (Imai et al., 2001c; Klempan et al., 2002; Ohmori et al., 2000;
Ohtsuki et al., 2000; Stober et al., 2000; Tsai et al., 2002) using both linkage and
family-based association analysis of eight separate polymorphisms in schizophrenia
families have been resoundingly negative.
Complexins are SNARE accessory proteins that bind to syntaxin and limit its
interaction with other SNARE proteins. Complexin I is a marker of inhibitory
synapses while complexin II occurs in synapses that are primarily excitatory. Inves-
tigation of complexin expression identified a reduction in complexin protein and
mRNA in medial temporal lobe structures of schizophrenia patients (Harrison
and Eastwood, 1998). This was later confirmed specifically in the hippocampus and
found to be unaffected by antipsychotic treatment in rats (Eastwood and Harrison,
2000), although complexin mRNA is altered in other regions of the rat forebrain
by antipsychotics (Eastwood et al., 2000). Changes in cytoarchitecture and protein
expression in the anterior cingulate cortex prompted the examination of this region
for levels of complexins (Eastwood and Harrison, 2001); however, no changes were
observed in schizophrenia, although they were for mood disorders. This group
also identified reduced complexin II expression in the cerebellum (Eastwood et al.,
2001). A selective and progressive loss of complexin II and other SNARE proteins in
the striatum (and enrichment in the hippocampus) of patients with Huntington’s
disease has been observed, with possible functional consequences for this disorder
(Morton et al., 2001). No genetic investigations of the complexins as candidate
genes for schizophrenia have yet been reported in the literature.
A number of additional studies have revealed expression-based changes of synap-
tic terminal molecules in schizophrenia. Increases in levels of syntaxin within the
cingulate cortex have been detected in several studies (Gabriel et al., 1997; Honer
et al., 1997), along with changes in the anterior prefrontal cortex (Honer et al., 2002).
A negative correlation between age of individuals with schizophrenia and expres-
sion of many presynaptic proteins in the temporal cortex (Sokolov et al., 2000)
23 Genes and brain development

highlights the relative non-specificity of this process and suggests that changes may
arise from an abnormality of synaptic connectivity determined much earlier in
life. These quantifiable changes in protein levels reflect actual biochemical changes
at the presynaptic terminal, potentially leading to synaptic dysfunction, as loss of
neuropil is not evident in these regions. An imbalance in stoichiometry between
critical regulatory elements may have severe consequences for both response capa-
city following calcium influx and rates of vesicle fusion and recycling, leading to an
increase or deficit in neurotransmission at the synapse. These changes in neuronal
phenotype or maturational state may account for the disorganization and volume
loss witnessed in various brain regions.
Apart from the recent microarray-based analysis described (Mirnics et al., 2000;
also see Ch. 12), other reports are providing increasing evidence for cortical presy-
naptic dysfunction in schizophrenia, including findings of synaptic vesicle cluster-
ing (Soustek, 1989) and alterations of many presynaptic protein and mRNA levels.
In addition to a fundamental role in neurotransmission, these molecules are crit-
ical during early embryonic development, influence neurite outgrowth and growth
cone maturation, and regulate morphological plasticity (Hepp and Langley, 2001).
These changes in presynaptic protein expression are also affected by neurotrophic
factors such as BDNF (Tartaglia et al., 2001), thus pointing to the need for further
elaboration of epistasis in these pathways when examining genetic variants.

Conclusions
While a staggering number of genes are involved in the patterning and function of
the human brain, the rapid pace of human genome sequence analysis has facilitated
the isolation and further characterization of brain-expressed sequences. Many of the
proteins encoded by these genes will participate in complex biological activities
integral to the establishment and maintenance of proper neuronal connectivity,
which, in turn, provides a template for higher-order cognitive function. It is natural
to consider, therefore, given the demonstrated genetic liability for schizophrenia,
that a number of these loci are likely to be involved as either direct etiological
determinants or phenotypic modifiers of the disease. As most researchers would
now agree that schizophrenia is inherited in a decidedly non-Mendelian fashion,
the best question that we can ask at present would be what set of DNA alterations are
consistently linked to a particular phenotype of schizophrenia? Possible modifiers
of cognitive abilities are beginning to be found through assessment of candidate
gene influence on neuropsychological tests (Bilder et al., 2002; Egan et al., 2003)
and further genetic determinants of individual disease components will likely be
identified as the complexities of the human genome are unraveled.
24 T. A. Klempan, P. Muglia, and J. L. Kennedy

This chapter has outlined the role played in the developing brain by various
genes within broad functional categories. A myriad of other factors (both genetic
and environmental) await investigation, and the final understanding of schizophre-
nia etiology may only be known with improvements in methodologies for detection
of gene–gene and gene–environment interaction. Until fairly recently, the relative
permanency of individual cells and populations/networks within the brain was
undisputed; however, we now know that the CNS is highly plastic and responsive,
with the capacity for neurogenesis throughout life from discrete sets of neural stem
cells (Eriksson et al., 1998). We should perhaps not be surprised, therefore, to learn
that the complex set of phenotypes collectively known as schizophrenia is antici-
pated by an equally complex set of genetic alterations, from which only particular
combinations may be identified within any given individual at risk for the disorder.
Of particular interest in this regard is the recent description of epistasis between
alleles of the G72 and d-amino acid oxidase (DAOO) genes in conferring risk for
schizophrenia (Chumakov et al., 2002). Together with neuregulin 1 and dysbindin,
these genes may induce schizophrenia susceptibility through a common N-methyl-
d-aspartate receptor pathway (Cloninger, 2002), a pathway that could alter the
neurodevelopmental trajectory of the brain. If replicated, the finding of Chumakov
and colleagues would constitute the first significant observation of gene–gene inter-
action in the etiology of schizophrenia, with likely expansion as the interactions
between various susceptibility factors are uncovered in the years to come.

Acknowledgements
We thank the Canadian Institutes of Health Research for grant support and Mary
Smirniw for her assistance with this manuscript.

REFERENCES

Akatone, A., Kunugi, H., Tanaka, H., Nanko, S. (2002). Association analysis of polymorphic CGG
repeat in 5 UTR of the reelin and VLDLR genes with schizophrenia, Schizophr Res 58: 37–41.
Akbarian, S., Kim, J. J., Potkin, S. G., et al. (1996). Maldistribution of interstitial neurons
in prefrontal white matter of the brains of schizophrenic patients. Arch Gen Psychiatry 53:
425–436.
Anderson, S. A., Eisenstat, D. D., Shi, L., Rubenstein, J. L. R. (1997). Interneuron migration from
basal forebrain to neocortex: dependence on Dlx genes. Science 278: 474–476.
Andrews, P. L., Wood, K. L. (1986). Systemic baclofen stimulates gastric motility and secretion
via a central action in the rat. Br J Pharmacol 89: 461–467.
Angelucci, F., Mathe, A. A., Aloe, L. (2000). Brain-derived neurotrophic factor and tyrosine
kinase receptor TrκB in rat brain are significantly altered after haloperidol and risperidone
administration. J Neurosci Res 60: 783–794.
25 Genes and brain development

Arenas, E., Trupp, M., Akerud, P., Ibanez, C. F. (1995). GDNF prevents degeneration and promotes
the phenotype of brain noradrenergic neurons in vivo. Neuron 15: 1465–1473.
Arinami, T., Toru, M. (1996). No evidence for association between CNTF null mutant allele and
schizophrenia. Br J Psychiatry 169: 253.
Arinami, T., Takekoshi, K., Itokawa, M., Hamaguchi, H., Toru, M. (1996). Failure to find associ-
ations of the CA repeat polymorphism in the first intron and the Gly-63/Glu-63 polymor-
phism of the neurotrophin-3 gene with schizophrenia. Psychiatry Genet 6: 13–15.
Artavanis-Tsakonis, S., Muskavitch, M. A., Yedvobnick, Y. (1983). Molecular cloning of Notch,
a locus affecting neurogenesis in Drosophila melanogaster. Proc Natl Acad Sci USA 80:
1977–1981.
Barbeau, D., Liang, J. J., Robitalille, Y., Quirion, R., Srivastava, L. K. (1995). Decreased expression
of the embryonic form of the neural cell adhesion molecule in schizophrenic brains. Proc
Natl Acad Sci USA 92: 2785–2789.
Bark, I. C., Wilson, M. C. (1994). Regulated vesicular fusion in neurons: snapping together the
details. Proc Natl Acad Sci USA 91: 4621–4624.
Barr, C. L., Feng, Y., Wigg, K., et al. (2000). Identification of DNA variants in the SNAP-25 gene
and linkage study of these polymorphisms and attention-deficit hyperactivity disorder. Mol
Psychiatry 5: 405–409.
Beasley, C., Cotter, D., Khan, N., et al. (2001). Glycogen synthase kinase-3beta immunoreactivity
is reduced in the prefrontal cortex in schizophrenia. Neurosci Lett 302: 117–120.
Beggs, H. E., Baragona, S. C., Hemperly, J. J., Maness, P. F. (1997). NCAM140 interacts with the
focal adhesion kinase p125(fak) and the SRC-related tyrosine kinase p59(fyn). J Biol Chem
272: 8310–8319.
Bell, E., Wingate, R. J. T., Lumsden, A. (1999). Homeotic transformation of rhombomere identity
after localized Hoxb1 misexpression. Science 284: 2168–2171.
Bello, M. J., Salagnon, N., Rey, J. A., et al. (1989). Precise in situ localization of NCAM, ETS1,
and D11S29 on human meiotic chromosomes. Cytogenet Cell Genet 52: 7–10.
Berkmeier, L. R., Winslow, J. W., Kaplan, D. R., et al. (1991). Neurotrophin-5: a novel neurotrophic
factor that activates trk and trkB. Neuron 4: 189–201.
Bilder, R. M., Volavka, J., Czobor, P., et al. (2002). Neurocognitive correlates of the COMT
Val(158)Met polymorphism in chronic schizophrenia. Biol Psychiatry 52: 701–707.
Broccoli, V., Boncinelli, E., Wurst, W. (1999). The caudal limit of Otx2 expression positions the
isthmic organizer. Nature 401: 164–168.
Brouha, A. K., Weickert, C. S., Hyde, T. M., et al. (1996). Reductions in brain derived neurotrophic
factor mRNA in the hippocampus of patients with schizophrenia. In Proceedings of the Society
for Neuroscience. Washington, DC: Society for Neuroscience, pp. 1680.
Browning, M. D., Dudek, E. M., Rapier, J. L., Leonard, S., Freedman, R. (1993). Significant reduc-
tions in synapsin but not synaptophysin specific activity in the brains of some schizophrenics.
Biol Psychiatry 34: 529–535.
Casarosa, S., Fode, C., Guillemot, F. (1999). Mash1 regulates neurogenesis in the ventral telen-
cephalon. Development 126: 525–534.
Catsicas, S., Larhammar, D., Blomqvist, A., et al. (1991). Expression of a conserved cell-type-
specific protein in nerve terminals coincides with synaptogenesis. Proc Natl Acad Sci USA
88: 785–789.
26 T. A. Klempan, P. Muglia, and J. L. Kennedy

Chapouton, P., Schuurmans, C., Guillemot, F., Gotz, M. (2001). The transcription factor neuro-
genin 2 restricts cell migration from the cortex to the striatum. Development 128: 5149–5159.
Chumakov, I., Blumenfeld, M., Guerassimenko, O., et al. (2002). Genetic and physiological data
implicating the new human gene G72 and the gene for d-amino acid oxidase in schizophre-
nia. Proc Natl Acad Sci USA 99: 13675–13680.
Cloninger, C. R. (2002). The discovery of susceptibility genes for mental disorders. Proc Natl
Acad Sci USA 99: 13365–13367.
Cotter, D., Kerwin, R., al-Sarraji, S., et al. (1998). Abnormalities of Wnt signalling in schizophre-
nia: evidence for neurodevelopmental abnormality. Neuroreport 9: 1379–1383.
Crossin, K. L., Krushel, L. A. (2000). Cellular signaling by neural cell adhesion molecules of the
immunoglobulin superfamily. Dev Dyn 218: 260–279.
Cunningham, B. A. (1995). Cell adhesion molecules as morphoregulators. Curr Opin Cell Biol 7:
628–633.
D’Arcangelo, G., Curran, T. (1998). Reeler: new tales on an old mutant mouse. Bioessays 20:
235–244.
Dawson, E., Powell, J. F., Sham P. C., et al. (1995). An association study of a neurotrophin-3
(NT-3) gene polymorphism with schizophrenia. Acta Psychiatr Scand 92: 425–428.
de la Pompa, J. L. , Wakeham, A., Correia, K. M., et al. (1997). Conservation of the Notch signalling
pathway in mammalian neurogenesis. Development 124: 1139–1148.
DeSilva, U., D’Arcangelo, G., Braden, V. V., et al. (1997). The human reelin gene: isolation,
sequencing, and mapping on chromosome 7. Genome Res 7: 157–164.
Doherty, P., Walsh, F. S. (1996). CAM–FGF receptor interactions: a model for axonal growth.
Mol Cell Neurosci 8: 99–111.
Doherty, P., Fruns, M., Seaton, P., et al. (1990). A threshold effect of the major isoforms of NCAM
on neurite outgrowth. Nature 343: 464–466.
Durany, N., Michel, T., Zochling, R., et al. (2001). Brain-derived neurotrophic factor and neu-
rotrophin 3 in schizophrenic psychoses. Schizophr Res 52: 79–86.
Eastwood, S. L., Harrison, P. J. (2000). Hippocampal synaptic pathology in schizophrenia, bipolar
disorder and major depression: a study of complexin mRNAs. Mol Psychiatry 5: 425–432.
(2001). Synaptic pathology in the anterior cingulate cortex in schizophrenia and mood dis-
orders. A review and a Western blot study of synaptophysin, GAP-43 and the complexins.
Brain Res Bull 55: 569–578.
Eastwood, S. L., Burnet, P. W., Harrison, P. J. (2000). Expression of complexin I and II mRNAs
and their regulation by antipsychotic drugs in the rat forebrain. Synapse 36: 167–177.
Eastwood, S. L., Cotter, D., Harrison, P. J. (2001). Cerebellar synaptic protein expression in
schizophrenia. Neuroscience 105: 219–229.
Eckhardt, M., Muhlenhoff, M., Bethe, A., et al. (1995). Molecular characterization of eukaryotic
polysialyltransferase-1. Nature 373: 715–718.
Egan, M. F., Kojima, M., Callicott, J. H., et al. (2003). The BDNF val66met polymorphism affects
activity-dependent secretion of BDNF and human memory and hippocampal function. Cell
112: 257–269.
Ericson, J., Murh, J., Placzek, M., et al. (1995). Sonic hedgehog induces the differentiation of
ventral forebrain neurons: a common signal for ventral patterning within the neural tube.
Cell 81: 747–756.
27 Genes and brain development

Eriksson, P. S., Perfilieva, E., Bjork-Eriksson, T., et al. (1998). Neurogenesis in the adult human
hippocampus. Nat Med 4: 1313–1317.
Ernfors, P., Ibanez, C. F., Ebendal, T., Olson, L., Persson, H. (1990). Molecular cloning and neu-
rotrophic activities of a protein with structural similarities to nerve growth factor: develop-
mental and topographical expression in brain. Proc Natl Acad Sci USA 87: 5454–5458.
Falconer, D. S. (1951). Two new mutants trembler and reeler, with neurological actions in the
house mouse. J Genet 50: 192–201.
Fan, J. B., Tang, J. X., Gu, N. F., et al. (2002). A family-based and case–control association study
of the NOTCH4 gene and schizophrenia. Mol Psychiatry 7: 100–103.
Fatemi, S. H., Earle, J. A., McMenomy, T. (2000). Reduction in Reelin immunoreactivity in
hippocampus of subjects with schizophrenia, bipolar disorder and major depression. Mol
Psychiatry 5: 654–663, 571.
Ferreiro, B., Kintner, C., Zimmerman, K., Anderson, D. J., Harris, W. (1994). XASH1 genes
promote neurogenesis in Xenopus embryos. Development 120: 3649–3655.
Fields, R. D., Itoh, K. (1996). Neural cell adhesion molecules in activity-dependent development
and synaptic plasticity. Trends Neurosci 19: 473–480.
Gabriel, S. M., Haroutunian, V., Powchik, P., et al. (1997). Increased concentrations of presynaptic
proteins in the cingulate cortex of subjects with schizophrenia. Arch Gen Psychiatry 54:
559–566.
Gelernter, J., Van Dyck, C., van Kammen, D. P., et al. (1997). Ciliary neurotrophic factor null
allele frequencies in schizophrenia, affective disorders, and Alzheimer’s disease. Am J Med
Genet 74: 497–500.
Goffinet, A. M., Dernoncourt, C. (1991). Localization of the reeler gene relative to flanking loci
on mouse chromosome 5. Mamm Genome 1: 100–103.
Goggi, J., Pullar, I. A., Carney, S. L., Bradford, H. F. (2002). Modulation of neurotransmitter
release induced by brain-derived neurotrophic factor in rat brain striatal slices in vitro.
Brain Res 941: 34–42.
Gotz, R., Koster, R., Winkler, C., et al. (1994). Neurotrophin-6 is a new member of the nerve
growth factor family. Nature 372: 266–269.
Gower, H. J., Barton, C. H., Elsom, V. L., et al. (1988). Alternative splicing generates a secreted
form of N-CAM in muscle and brain. Cell 55: 955–964.
Grebb, J. A., Greengard, P. (1990). An analysis of synapsin II, a neuronal phosphoprotein, in
postmortem brain tissue from alcoholic and neuropsychiatrically ill adults and medically ill
children and young adults. Arch Gen Psychiatry 47: 1149–1156.
Greengard, P., Valtorta, F., Czernik, A. J., Benfenati, F. (1993). Synaptic vesicle phosphoproteins
and regulation of synaptic function. Science 259: 780–785.
Guidotti, A., Auta, J., Davis, J. M., et al. (2000). Decrease in reelin and glutamic acid decar-
boxylase67 (GAD67) expression in schizophrenia and bipolar disorder: a postmortem brain
study. Arch Gen Psychiatry 57: 1061–1069.
Guillin, O., Diaz, J., Carroll, P., et al. (2001). BDNF controls dopamine D3 receptor expression
and triggers behavioural sensitization. Nature 411: 86–89.
Ha, D. H., Robertson, R. T., Ribak, C. E., Weiss, J. H. (1996). Cultured basal forebrain cholinergic
neurons in contact with cortical cells display synapses, enhanced morphological features,
and decreased dependence on nerve growth factor. J Comp Neurol 373: 451–465.
28 T. A. Klempan, P. Muglia, and J. L. Kennedy

Han, H. Q., Bahler, M., Greengard, P., Nichols, R. A., Rubin, M. R. (1991). Induction of formation
of presynaptic terminals in neuroblastoma cells by synapsin IIb. Nature 349: 697–700.
Harrison, P. J., Eastwood, S. L. (1998). Preferential involvement of excitatory neurons in medial
temporal lobe in schizophrenia. Lancet 352: 1669–1673.
Hattori, M., Nanko, S. (1995). Association of neurotrophin-3 gene variant with severe forms of
schizophrenia. Biochem Biophys Res Commun 209: 513–518.
Hawi, Z., Straub, R. E., O’Neill, A., et al. (1998). No linkage or linkage disequilibrium
between brain-derived neurotrophic factor (BDNF) dinucleotide repeat polymorphism and
schizophrenia in Irish families. Psychiatry Res 81: 111–116.
Hepp, R., Langley, K. (2001). SNAREs during development. Cell Tissue Res 305: 247–253.
Hess, E. J., Collins, K. A., Wilson, M. C. (1996). Mouse model of hyperkinesis implicates SNAP-25
in behavioral regulation. J Neurosci 16: 3104–3111.
Heyser, C. J., Wilson, M. C., Gold, L. H. (1995). Coloboma hyperactive mutant exhibits delayed
neurobehavioral developmental milestones. Brain Res Dev Brain Res 89: 264–269.
Hodel, A. (1998). Snap-25. Int J Biochem Cell Biol 30: 1069–1073.
Hohn, A., Leibrock, J., Bailey, K., Barde, Y. A. (1990). Identification and characterization of a
novel member of the nerve growth factor/brain-derived neurotrophic factor family. Nature
344: 339–341.
Honer, W. G., Falkai, P., Young, C., et al. (1997). Cingulate cortex synaptic terminal proteins and
neural cell adhesion molecule in schizophrenia. Neuroscience 78: 99–110.
Honer, W. G., Falkai, P., Bayer, T. A., et al. (2002). Abnormalities of SNARE mechanism proteins
in anterior frontal cortex in severe mental illness. Cereb Cortex 12: 349–356.
Hynes, M., Rosenthal, A. (1999). Specification of dopaminergic and serotonergic neurons in the
vertebrate CNS. Curr Opin Neurobiol 9: 26–36.
Imai, C., Sugai, T., Iritani, S., et al. (2001a). A quantitative study on the expression of synapsin II
and N-ethylmaleimide-sensitive fusion protein in schizophrenic patients. Neurosci Lett 305:
185–188.
Imai, K., Harada, S., Kawanishi, Y., et al. (2001b). The (CTG)n polymorphism in the NOTCH4
gene is not associated with schizophrenia in Japanese individuals. BMC Psychiatry 1: 1.
(2001c). Polymorphisms in the promoter and coding regions of the synapsin III gene. A lack
of association with schizophrenia. Neuropsychobiology 43: 237–241.
Ip, N. Y., Ibanez, C. F., Nye, S. H., et al. (1992). Mammalian neurotrophin-4: structure, chromo-
somal localization, tissue distribution, and receptor specificity. Proc Natl Acad Sci USA 89:
3060–3064.
Jones, F. S., Holst, B. D., Minowa, O., De Robertis, E. M., Edelman, G. M. (1993). Binding and
transcriptional activation of the promoter for the neural cell adhesion molecule by HoxC6
(Hox-3.3). Proc Natl Acad Sci USA 90: 6557–6561.
Jonsson, E., Brene, S., Zhang, X. R., et al. (1997). Schizophrenia and neurotrophin-3 alleles. Acta
Psychiatr Scand 95: 414–419.
Jovanovic, J. N., Czernik, A. J., Fienberg, A. A., Greengard, P., Sihra, T. S. (2000). Synapsins as
mediators of BDNF-enhanced neurotransmitter release. Nat Neurosci 3: 323–329.
Kao, H. T., Porton, B., Czernik, A. J., et al. (1998). A third member of the synapsin gene family.
Proc Natl Acad Sci USA 95: 4667–4672.
29 Genes and brain development

Karecla, P. I., Green, S. J., Bowden, S. J., Coadwell, J., Kilshaw, P. J. (1996). Identification of a
binding site for integrin alphaEbeta7 in the N-terminal domain of E-cadherin. J Biol Chem
271: 30909–30915.
Karson, C. N., Mrak, R. E., Schluterman, K. O., et al. (1999). Alterations in synaptic proteins
and their encoding mRNAs in prefrontal cortex in schizophrenia: a possible neurochemical
basis for “hypofrontality”. Mol Psychiatry 4: 39–45.
Kiss, J. Z., Troncoso, E., Djebbara, Z., Vutskits, L., Muller, D. (2001). The role of neural cell
adhesion molecules in plasticity and repair. Brain Res Brain Res Rev 36: 175–184.
Klempan, T. A., Trakalo, J. M., Pato, C. N., et al. (2001). Polymorphisms of the NOTCH4 gene
and schizophrenia. Am J Hum Genet 69 (Suppl. 4): 567.
Klempan, T. A., Trakalo, J., King, N., et al. (2002). Transmission of SYNAPSIN3 gene variants in
schizophrenia families. In Proceeding of the 52nd Annual Meeting of the American Society for
Human Genetics. Bethesda, MD: American Society for Human Genetics, p. 489.
Kohtz, J. D., Baker D. P., Corte, G., Fishell, G. (1998). Regionalization within the mammalian
telencephalon is mediated by changes in responsiveness to Sonic hedgehog. Development
125: 5079–5089.
Kortschak, R. D., Tamme, R., Lardelli, M. (2001). Evolutionary analysis of vertebrate Notch genes.
Dev Genes Evol 211: 350–354.
Kozlovsky, N., Belmaker, R. H., Agam, G. (2002). GSK-3 and the neurodevelopmental hypothesis
of schizophrenia. Eur Neuropsychopharmacol 12: 13–25.
Krebs, M. O., Guillin, O., Bourdell, M. C., et al. (2000). Brain derived neurotrophic factor (BDNF)
gene variants association with age at onset and therapeutic response in schizophrenia. Mol
Psychiatry 5: 558–562.
Kristiansen, L. V., Marques F. A., Soroka, V., et al. (1999). Homophilic NCAM interactions
interfere with L1 stimulated neurite outgrowth. FEBS Lett 464: 30–34.
Krumlauf, R. (1994). Hox genes in vertebrate development. Cell 78: 191–201.
Krushel, L. A., Cunningham, B. A., Edelman, G. M., Crossin, K. L. (1999). NF-kappaB activity
is induced by neural cell adhesion molecule binding to neurons and astrocytes. J Biol Chem
274: 2432–2439.
LaMantia, A. S. (1999). Forebrain induction, retinoic acid, and vulnerability to schizophrenia:
insights from molecular and genetic analysis in developing mice. Biol Psychiatry 46: 19–30.
Lawrie, S. M., Abukmeil, S. S. (1998). Brain abnormality in schizophrenia. A systematic and
quantitative review of volumetric magnetic resonance imaging studies. Br J Psychiatry 172:
110–120.
Lee, J. E. (1997). Basic helix–loop–helix genes in neural development. Curr Opin Neurobiol 7:
13–20.
Lee, J. E., Hollenberg, S. M., Lipnick, N., et al. (1995). Conversion of Xenopus ectoderm into
neurons by NeuroD, a basic helix–loop–helix protein. Science 268: 836–844.
Lee, K., Kunugi, H., Nanko, S. (2001). Glial cell line-derived neurotrophic factor (GDNF) gene
and schizophrenia: polymorphism screening and association analysis. Psychiatry Res 104:
11–17.
Leibrock, J., Lottspeich, F., Hohn, A., et al. (1989). Molecular cloning and expression of brain-
derived neurotrophic factor. Nature 341: 149–152.
30 T. A. Klempan, P. Muglia, and J. L. Kennedy

Levi-Montalcini, R. (1987). The nerve growth factor 35 years later. Science 237: 1154–1162.
Lewis, C. M., Levinson, D. F., Wise, L. H., et al. (2003). Genome scan meta-analysis of schizophre-
nia and bipolar disorder, part II: schizophrenia. Am J Hum Genet 73: 34–48.
Li, L., Huang, G. M., Banta, A. B., et al. (1998). Cloning, characterization, and the complete
56.8-kilobase DNA sequence of the human NOTCH4 gene. Genomics 51: 45–58.
Lin, L. H., Doherty, D. H., Lile, J. D., Bektesh, S., Collins, F. (1993). GDNF: a glial cell-derived
neurotrophic factor for midbrain dopaminergic neurons. Science 260: 1130–1132.
Lou, X. J., Bixby, J. L. (1995). Patterns of presynaptic gene expression define two stages of synaptic
differentiation. Mol Cell Neurosci 6: 252–262.
Luo, X. J., Klempan, T. A., Lappalainen, J., et al. (2004). NOTCH4 gene haplotype is associated
with schizophrenia in African-Americans. Biol Psychiatry 55: 112–117.
Maglott, D. R., Feldblyum, T. V., Durkin, A. S., Nierman, W. C. (1996). Radiation hybrid mapping
of SNAP, PCSK2, and THBD (human chromosome 20p). Mamm Genome 7: 400–401.
Maisonpierre, P., Belluscio, L., Friedman, B., et al. (1990). NT-3, BDNF, and NGF in the devel-
oping rat nervous system: parallel as well as reciprocal patterns of expression. Neuron 5:
501–509
Mallamaci, A., Mercurio, S., Muzio, L., et al. (2000). The lack of Emx2 causes impairment of Reelin
signaling and defects of neuronal migration in the developing cerebral cortex. J Neurosci 20:
1109–1118.
McEvilly, R. J., de Diaz, M. O., Schonemann, M. D., Hooshmand, F., Rosenfeld, M. G. (2002).
Transcriptional regulation of cortical neuron migration by POU domain factors. Science
295: 1528–1532.
McGinnis, R. E., Fox, H., Yates, P., et al. (2001). Failure to confirm NOTCH4 association with
schizophrenia in a large population-based sample from Scotland. Nat Genet 28: 128–129.
McMahon, A. P., Bradley, A. (1990). The Wnt-1 (int-1) proto-oncogene is required for the
development of a large region of the mouse brain. Cell 62: 1073–1085.
Mirnics, K., Middleton, F. A., Marquez, A., Lewis, D. A., Levitt, P. (2000). Molecular characteriza-
tion of schizophrenia viewed by microarray analysis of gene expression in prefrontal cortex.
Neuron 28: 53–67.
Mirnics, K., Middleton, F. A., Lewis D. A., Levitt, P. (2001). Analysis of complex brain disorders
with gene expression microarrays: schizophrenia as a disease of the synapse. Trends Neurosci
24: 479–486.
Miyaoka, T., Seno, H., Ishino, H. (1999). Increased expression of Wnt-1 in schizophrenic brains.
Schizophr Res 38: 1–6.
Moises, H. W., Kristbjarnarson, H., Wiese, C., et al. (1995). An international two-stage genome-
wide search for schizophrenia susceptibility genes. Nat Genet 11: 321–324.
Morris, P. J., Dawson, S. J., Wilson, M. C., Latchman, D. S. (1997). A single residue within the
homeodomain of the Brn-3 POU family transcription factors determines whether they
activate or repress the SNAP-25 promoter. Neuroreport 8: 2041–2045.
Morton, A. J., Faull, R. L. M., Edwardson, J. M. (2001). Abnormalities in the synaptic vesicle
fusion machinery in Huntington’s disease. Brain Res Bull 56: 111–117.
Muglia, P., Vicente, A. M., Verga, M., et al. (2003). Association between the BDNF gene and
schizophrenia. Mol Psychiatry 8: 147–148.
31 Genes and brain development

Mukaetova-Ladinska, E. B., Hurt, J., Honer, W. G., Harrington, C. R., Wischik, C. M. (2002).
Loss of synaptic but not cytoskeletal proteins in the cerebellum of chronic schizophrenics.
Neurosci Lett 317: 161–165.
Muller, D., Djebbara-Hannas, Z., Jourdain, P., et al. (2000). Brain-derived neurotrophic factor
restores long-term potentiation in polysialic acid-neural cell adhesion molecule-deficient
hippocampus. Proc Natl Acad Sci USA 97: 4315–4320.
Murase, S., Schuman, E. M. (1999). The role of cell adhesion molecules in synaptic plasticity and
memory. Curr Opin Cell Biol 11: 549–553.
Murphy, K. C., Jones, L. A., Owen, M. J. (1999). High rates of schizophrenia in adults with
velo-cardio-facial syndrome. Arch Gen Psychiatry 56: 940–945.
Nakayama, J., Fukuda, M. N., Fredette, B., Ranscht, B., Fukuda, M. (1995). Expression cloning of
a human polysialyltransferase that forms the polysialylated neural cell adhesion molecule
present in embryonic brain. Proc Natl Acad Sci USA 92: 7031–7035.
Nanko, S., Hattori, M., Kuwata, S., et al. (1994). Neurotrophin-3 gene polymorphism associated
with schizophrenia. Acta Psychiatr Scand 89: 390–392.
Nguyen Ba-Charvet, K. T., von Boxberg, Y., Godement, P. (1999). The mouse homeodomain
protein OTX2 regulates NCAM promoter activity. Brain Res Mol Brain Res 67: 292–295.
Nimgaonkar, V. L., Zhang, X. R., Brar, J. S., DeLeo, M., Ganguli, R. (1995). Lack of association
of schizophrenia with the neurotrophin-3 gene locus. Acta Psychiatr Scand 92: 464–466.
Ohmori, O., Shinkai, T., Hori, H., Kojima, H., Nakamura, J. (2000). Synapsin III gene polymor-
phisms and schizophrenia. Neurosci Lett 279: 125–127.
Ohtsuki, T., Ichiki, R., Toru, M., Arinami, T. (2000). Mutational analysis of the synapsin III gene
on chromosome 22q12-q13 in schizophrenia. Psychiatry Res 94: 1–7.
Osen-Sand, A., Catsicas, M., Staple, J. K., et al. (1993). Inhibition of axonal growth by SNAP-25
antisense oligonucleotides in vitro and in vivo. Nature 364: 445–448.
Parras, C. M., Schuurmans, C., Scardigli, R., et al. (2002). Divergent functions of the proneural
genes Mash1 and Ngn2 in the specification of neuronal subtype identity. Genes Dev 16:
324–338.
Persico, A. M., D’Agruma, L., Maiorano, N., et al. (2001). Reelin gene alleles and haplotypes as
a factor predisposing to autistic disorder. Mol Psychiatry 6: 150–159.
Proschel, M., Saunders, A., Roses, A. D., Muller, C. R. (1992). Dinucleotide repeat polymorphism
at the human gene for the brain-derived neurotrophic factor (BDNF). Hum Mol Genet 1:
353.
Purves, D. (1988) Body and Brain: A Trophic Theory of Neural Connections. Cambridge, MA:
Harvard University Press.
Raber, J., Mehta, P. P., Kreifeldt, M., et al. (1997). Coloboma hyperactive mutant mice exhibit
regional and transmitter-specific deficits in neurotransmission. J Neurochem 68: 176–186.
Rakic, P., Caviness, V. S., Jr. (1995). Cortical development: view from neurological mutants two
decades later. Neuron 14: 1101–1104.
Rhoads, A. R., Karkera, J. D., Detera-Wadleigh, S. D. (1999). Radiation hybrid mapping of genes
in the lithium-sensitive wnt signaling pathway. Mol Psychiatry 4: 437–442.
Rice, D. S., Curran, T. (2001). Role of the reelin signaling pathway in central nervous system
development. Annu Rev Neurosci 24: 1005–1039.
32 T. A. Klempan, P. Muglia, and J. L. Kennedy

Rodriguez, M. A., Pesold, C., Liu, W. S., et al. (2000). Colocalization of integrin receptors and
Reelin in dendritic spine postsynaptic densities of adult nonhuman primate cortex. Proc
Natl Acad Sci USA 97: 3550–3555.
Rosahl, T. W., Spillane, D., Missler, M., et al. (1995). Essential functions of synapsins I and II in
synaptic vesicle regulation. Nature 375: 488–493.
Ruiz i Altaba, A. (1994). Pattern formation in the vertebrate neural plate. Trends Neurosci 17:
233–243.
Sasaki, T., Dai, X. Y., Kuwata, S., et al. (1997a). Brain-derived neurotrophic factor gene and
schizophrenia in Japanese subjects. Am J Med Genet 74: 443–444.
Sakai, T., Sasaki, T., Tatsumi, M., et al. (1997b). Schizophrenia and the ciliary neurotrophic factor
(CNTF) gene: no evidence for association. Psychiatry Res 71: 7–10.
Schneider-Manoury, S., Seitanidou, T., Charnay, P., Lumsden, A. (1997). Segmental and neuronal
architecture of the hindbrain of Krox-20 mouse mutants. Development 124: 1215–1226.
Schonemann, M. D., Ryan A. K., McEvilly, R. J., et al. (1995). Development and survival of the
endocrine hypothalamus and posterior pituitary gland requires the neuronal POU domain
factor Brn-2. Genes Dev 9: 3122–3135.
Schwab, S. G., Albus, M., Hallmayer, J., et al. (1995). Evaluation of a susceptibility gene for
schizophrenia on chromosome 6p by multipoint affected sib-pair linkage analysis. Nat
Genet 11: 325–327.
Selemon, L. D., Rajkowska, G., Goldman-Rakic, P. S. (1995). Abnormally high neuronal density
in the schizophrenic cortex. A morphometric analysis of prefrontal area 9 and occipital area
17. Arch Gen Psychiatry 52: 805–818.
Sendtner, M., Carroll, P., Holtmann, B., Hughes, R., Thoenen, H. (1994). Ciliary neurotrophic
factor. J Neurobiol 25: 1436–1453.
Senzaki, K., Ogawa, M., Yagi, T. (1999). Proteins of the CNR family are multiple receptors for
Reelin. Cell 99: 635–647.
Sklar, P., Schwab, S. G., Williams, N. M., et al. (2001). Association analysis of NOTCH4 loci in
schizophrenia using family and population-based controls. Nat Genet 28: 126–128.
Snider, W. (1994). Functions of the neurotrophins during nervous system development: what
the knockouts are teaching us. Cell 77: 627–638.
Sokolov, B. P., Tcherepanov, A. A., Haroutunian, V., Davis, K. L. (2000). Levels of mRNAs
encoding synaptic vesicle and synaptic plasma membrane proteins in the temporal cortex
of elderly schizophrenic patients. Biol Psychiatry 48: 184–196.
Soustek, Z. (1989). Ultrastructure of cortical synapses in the brain of schizophrenics. Zentralbl
Allg Pathol 135: 25–32.
Steindler, D. A., Colwell, S. A. (1976). Reeler mutant mouse: maintenance of appropriate and
reciprocal connections in the cerebral cortex and thalamus. Brain Res 113: 386–393.
Stober, G., Syagailo, Y. V., Okladnova, O., et al. (1999). Functional PAX-6 gene-linked polymor-
phic region: potential association with paranoid schizophrenia. Biol Psychiatry 45: 1585–
1591.
Stober, G., Meyer, J., Nanda, I., et al. (2000). Linkage and family-based association study of
schizophrenia and the synapsin III locus that maps to chromosome 22q13. Am J Med Genet
96: 392–397.
33 Genes and brain development

Straub, R. E., MacLean, C. J., O’Neill, F. A., et al. (1995). A potential vulnerability locus for
schizophrenia on chromosome 6p24–22: evidence for genetic heterogeneity. Nat Genet 11:
287–293.
Studer, M., Lumsden, A., Ariza-McNaughton, L., Bradley, A., Krumlauf, R. (1996). Altered seg-
mental identity and abnormal migration of motor neurons in mice lacking Hoxb1. Nature
384: 630–634.
Sun, Y., Nadal-Vicens, M., Misono, S., et al. (2001). Neurogenin promotes neurogenesis and
inhibits glial differentiation by independent mechanisms. Cell 104: 365–376.
Swift-Scanlan, T., Lan, T. H., Fallin, M. D., et al. (2002). Genetic analysis of the (CTG)n NOTCH4
polymorphism in 65 multiplex bipolar pedigrees. Psychiatr Genet 12: 43–47.
Tachikawa, H., Harada, S., Kawanishi, Y., Okubo, T., Suzuki, T. (2001). Polymorphism of the 5 -
upstream region of the human SNAP-25 gene: an association analysis with schizophrenia.
Neuropsychobiology 43: 131–133.
Takahashi, M., Shirakawa, O., Toyooka, K., et al. (2000). Abnormal expression of brain-derived
neurotrophic factor and its receptor in the corticolimbic system of schizophrenic patients.
Mol Psychiatry 5: 293–300.
Takei, K., Chan, T. A., Wang, F. S., Deng, H., Rutishauser, U., Jay, D. G. (1999). The neural cell
adhesion molecules L1 and NCAM-180 act in different steps of neurite outgrowth. J Neurosci
19: 9469–9479.
Takeichi, M. (1990). Cadherins: a molecular family important in selective cell–cell adhesion.
Annu Rev Biochem 59: 237–252.
Tanaka, Y., Ujike, H., Fujiwara, Y., et al. (1998). Schizophrenic psychoses and the CNTF null
mutation. Neuroreport 9: 981–983.
Tartaglia, N., Du, J., Tyler, W. J., et al. (2001). Protein synthesis-dependent and -independent
regulation of hippocampal synapses by brain-derived neurotrophic factor. J Biol Chem 276:
37585–37593.
Tcherepanov, A. A., Sokolov, B. P. (1997). Age-related abnormalities in expression of mRNAs
encoding synapsin 1A, synapsin 1B, and synaptophysin in the temporal cortex of
schizophrenics. J Neurosci Res 49: 639–644.
Thoenen, H. (1995) Neurotrophins and neuronal plasticity. Science 270: 593–598.
Thome, J., Durany, N., Harsanyi, A., et al. (1996). A null mutation allele in the CNTF gene and
schizophrenic psychoses. Neuroreport 7: 1413–1416.
Thome, J., Durany, N., Palomo, A., et al. (1997a). Variants in neurotrophic factor genes and
schizophrenic psychoses: no associations in a Spanish population. Psychiatry Res 71: 1–5.
Thome, J., Jonsson, E., Foley, P., et al. (1997b). Ciliary neurotrophic factor null mutation and
schizophrenia in a Swedish population. Psychiatr Genet 7: 79–82.
Thompson, P. M., Sower, A. C., Perrone-Bizzozero, N. I. (1998). Altered levels of the synaptosomal
associated protein SNAP-25 in schizophrenia. Biol Psychiatry 43: 239–243.
Thompson, P. M., Rosenberger, C., Qualls, C. (1999). CSF SNAP-25 in schizophrenia and bipolar
illness. A pilot study. Neuropsychopharmacology 21: 717–722.
Treloar, H., Tomasiewicz, H., Magnuson, T., Key, B. (1997). The central pathway of primary
olfactory axons is abnormal in mice lacking the N-CAM-180 isoform. J Neurobiol 32: 643–
658.
34 T. A. Klempan, P. Muglia, and J. L. Kennedy

Tsai, M. T., Hung, C. C., Tsai, C. Y., et al. (2002). Mutation analysis of synapsin III gene in
schizophrenia. Am J Med Genet 114: 79–83.
Turner, E. E., Fedtsova, N., Jeste, D. V. (1997). Cellular and molecular neuropathology of
schizophrenia: new directions from developmental neurobiology. Schizophr Res 27: 169–
180.
Ujike, H., Takehisa, Y., Takaki, M., et al. (2001). NOTCH4 gene polymorphism and susceptibility
to schizophrenia and schizoaffective disorder. Neurosci Lett 301: 41–44.
Usui, T., Shima, Y., Shimada, Y., et al. (1999). Flamingo, a seven-pass transmembrane cadherin,
regulates planar cell polarity under the control of Frizzled. Cell 98: 585–595.
van Kammen, D. P., Poltorak, M., Kelley, M. E., et al. (1998). Further studies of elevated cere-
brospinal fluid neuronal cell adhesion molecule in schizophrenia. Biol Psychiatry 43: 680–
686.
Vicente, A. M., Macciardi, F., Verga, M., et al. (1997). NCAM and schizophrenia: genetic studies.
Mol Psychiatry 2: 65–69.
Virgos, C., Martorell, L., Valero, J., et al. (2001). Association study of schizophrenia with poly-
morphisms at six candidate genes. Schizophr Res 49: 65–71.
Wassarman, K. M., Lewandoski, M., Campbell, K., et al. (1997). Specification of the anterior
hindbrain and establishment of a normal mid/hindbrain organizer is dependent on Gbx2
gene function. Dev Suppl 124: 2923–2934.
Wassink, T. H., Nelson J. J., Crowe, R. R., Andreasen, N. C. (1999). Heritability of BDNF alleles
and their effect on brain morphology in schizophrenia. Am J Med Genet 88: 724–728.
Wei, J., Hemmings, G. P. (2000). The NOTCH4 locus is associated with susceptibility to
schizophrenia. Nat Genet 25: 376–377.
Wong, A., Macciardi, F., Klempan, T., et al. (2003). Identification of candidate genes for psychosis
in rat models, and possible association between schizophrenia and the 14-3-3η gene. Mol
Psychiatry 8: 156–166.
Wood, G. K., Tomasiewicz, H., Rutishauser, U., et al. (1998). NCAM-180 knockout mice display
increased lateral ventricle size and reduced prepulse inhibition of startle. Neuroreport 9:
461–466.
Wu, Q., Maniatis, T. (1999). A striking organization of a large family of human neural cadherin-
like cell adhesion genes. Cell 97: 779–790.
Yagi, T., Takeichi, M. (2000). Cadherin superfamily genes: functions, genomic organization, and
neurologic diversity. Genes Dev 14: 1169–1180.
Young, C. E., Arima, K., Xie, J., et al. (1998). SNAP-25 deficit and hippocampal connectivity in
schizophrenia. Cereb Cortex 8: 261–268.
 2

Brain development in healthy children

and adolescents: magnetic resonance
imaging studies
Jay N. Giedd, Michael A. Rosenthal, A. Blythe Rose, Jonathan
D. Blumenthal, Elizabeth Molloy, Richard R. Dopp, Liv S. Clasen,
Daniel J. Fridberg, and Nitin Gogtay
Child Psychiatry Branch, National Institute of Mental Health, Bethesda, USA

Any parent of a teenager can tell you that the brain of a 13 year old is different
than the brain of an 8 year old; yet, actually pinning down the neuroanatomical
substrates of those differences has proved elusive. Nature has gone through a great
deal of trouble to protect the brain. It is wrapped in a tough leathery membrane,
surrounded by a protective moat of fluid, and completely encased in bone. This has
shielded the brain from falls or attacks from predators, but it has also shielded the
brain from scientists.
Magnetic resonance imaging (MRI) has changed that. It allows for the acquisition
of exquisitely accurate pictures of the living growing human brain and it does so
without the use of ionizing radiation. This has opened the door not only for scanning
healthy children but also for acquiring repeated scans throughout development.
Using MRI, the team at the Child Psychiatry Branch of the National Institute
of Mental Health has been collecting brain MRI scans on healthy children and
adolescents since 1989. As of 2003, we have acquired over 300 scans from 150
healthy subjects. The data presented in this chapter will be largely drawn from this
cohort unless otherwise stated.

Total brain size

The relationship between brain size and function is complex. Evidence for the rel-
ative unimportance of brain size within a broad range of extremes is demonstrated
by observations that (a) healthy children with similar intelligence quotients (IQs)

Neurodevelopment and Schizophrenia, ed. Matcheri S. Keshavan et al. Published by Cambridge University
Press.

C Cambridge University Press 2004.

35
36 J. N. Giedd, M. A. Rosenthal, A. B. Rose, et al.

can have a 50% difference in brain volume, (b) there are robust differences in brain
sizes between males and females with similar functional capacity, and (c) there is
a paucity of established correlations between the size of any given brain structure
and a specific cognitive ability. However, from a computational science perspective
it seems likely that the number of neuronal connections in a structure reflects its
information processing capacity.
Also, modest positive correlation between IQ and total cerebral volume and
a possible relationship between hippocampal size and memory recall have been
reported. The intricacy of various neurochemical systems and the diversity of affer-
ent and efferent connections to the many distinct nuclei of most brain structures
make straightforward relationships between volumes of a single structure and per-
formance on a particular cognitive task uncommon. This supports the concept of
distributed neural systems, whereby functional attributes are not thought to lie so
much within a single structure as within a network of structures. Across species there
is a strong correlation between body size and brain size (Jerison, 1991), and animals
with large brains such as elephants and dolphins demonstrate greater behavioral
complexity.
Total brain volume is approximately 90% of its adult size by age six years. Data
of total brain size, as is the case for data of the subcomponents of the brain, is char-
acterized by a high degree of variability (Fig. 2.1). This large variability necessitates
large sample sizes to detect group differences (Lange et al., 1997).
MRI is adept at discerning gray matter, white matter, and fluid on brain images.
These boundaries are used to define the size and shape of a variety of brain structures
or regions.

White matter
White matter consists of myelinated axons (Fig. 2.2). The amount of white matter
in the brain increases throughout childhood and adolescence. Myelination greatly
speeds up transmission between neurons, up to 100 times the speed in unmyelinated
fibers. The greater speed of neuronal processing may facilitate cognitive complexity.
The myelinated axons comprising white matter can be projectional, connecting the
brain to the brainstem or spinal cord; associational, connecting one type of brain
part to another; or commissural, connecting like parts of the brain in the left and
right hemispheres. Diffusion tensor imaging, a structural brain imaging technique,
may help to discern these different types of white matter connection although our
group has not yet acquired diffusion tensor imaging scans on pediatric populations
(see Ch. 3 for a discussion of this approach).
The most conspicuous white matter component of the brain is the corpus cal-
losum (CC). It is the main connection between the left and right hemispheres and
37 Brain development in healthy children

1800

Volume in cubic cm 1600

1400

1200

1000
Female
Male

800
4 6 8 10 12 14 16 18 20 22
Age in years
NIH, DHHS.

Fig. 2.1. Total cerebral volume for 145 children and adolescents (ages 4–22) based on 243 brain
magnetic resonance imaging scans. (Jay N. Giedd, Child Psychiatry Branch, NIMH, NIH,
DHHS.)

Dendrites

Axon
Cell body
(the cell’s life
support center) Terminal
branches of
axon
Neuronal impulse

Myelin
sheath

Fig. 2.2. The neuron. (Donald Bliss, NIH Medical Arts and Photography Branch for Jay N. Giedd, Child
Psychiatry Branch, NIMH, NIH, DHHS.) (See plate section for color version.)
38 J. N. Giedd, M. A. Rosenthal, A. B. Rose, et al.

consists of approximately 180 million mostly myelinated axons (Tomasch, 1954).

The fibers usually connect similar areas in the left and right hemispheres. Because
they generally take the shortest route, a roughly topographic pattern is maintained,
with the anterior sections consisting of fibers connecting frontal brain areas, mid-
dle sections connecting middle cortical areas, and posterior sections connecting
posterior cortical areas.
There is some controversy about how tightly the spatial relationships of the
cortex are maintained in their CC representation, but for some of the areas studied,
such as the somatosensory regions, the spatial representation is highly preserved
(Innocenti et al., 1974; Spidalieri et al., 1985).
Although absent in monotrenes and marsupials, the CC is present in most ani-
mals from insectivores to higher primates. It appears to have evolved in parallel with
the neocortex (Kappers et al., 1936; Rapoport, 1990). In general, the CC functions
to integrate the activities of the left and right cerebral hemispheres, such as organ-
izing bimanual motor output (Zaidel and Sperry, 1977) and unifying the sensory
fields (Berlucchi, 1981; Shanks et al., 1975). The CC is also involved in memory
storage and retrieval (Zaidel and Sperry, 1974), attention and arousal (Levy, 1985),
language and auditory functions (Cook, 1986), and possibly in the perception of
consciousness (Joseph, 1980). Creativity and intelligence are linked to interhemi-
spheric integration (Bogen and Bogen, 1969) and the more difficult the cognitive
task the more critical interhemispheric integration becomes (Hellige et al., 1979;
Levy and Trevarthen, 1981).
These functions subserved by the CC continue to improve during childhood
and adolescence, highlighting interest in the structural changes shown to progress
during that developmental period (Giedd et al., 1996). Further developmental
interest stems from CC anomalies reported for several neuropsychiatric disorders
of childhood (Bigelow et al., 1983; Giedd et al., 1994; Hynd et al., 1990, 1991;
Njiokiktjien, 1991; Parashos et al., 1995; Peterson et al., 1994; Rosenthal and Bigelow,
1972).
In our cross-sectional study of 114 healthy pediatric subjects, total CC size
increased approximately 2% per year on average for the group (Giedd et al., 1996).
However, when individual children are followed longitudinally, the changes in spe-
cific subcomponents of the CC are shown to be occurring much more dramatically
(Thompson, 2000).
The CC changes occur in a front-to-back direction, with the anterior sections
reaching adult sizes sooner than the posterior sections. This was somewhat sur-
prising because, in general, frontal regions of the brain are thought to mature later.
Another white matter area that has particularly strong age effects is the left arcuate
fasiculus. This tract connects Wernicke’s area, involved in the reception of speech,
with Broca’s area, involved in the production of speech (Paus et al., 1999).
39 Brain development in healthy children

115
Parietal
110
Percentage Frontal

105
Temporal

100
Total brain

95

90
4 6 8 10 12 14 16 18 20 22
Age (in years)

Fig. 2.3. Standardized regional brain development for 145 children and adolescents (ages 4–22)
based on 243 brain magnetic resonance imaging scans. (Jay N. Giedd, Child Psychiatry
Branch, NIMH, NIH, DHHS.)

Gray matter
Gray matter consists largely of cell bodies and is home to the nucleus and
the dendrites, the antennae-like connections that receive input from other cells
(Fig. 2.2). Gray matter comprises the cortex (Latin for “bark”), the outermost layer
of the brain, and certain subcortical structures such as the basal ganglia and the
thalamus.
Initial cross-sectional studies showed a general decrease in the amount of cortical
gray matter during childhood, beginning at the earliest ages of the study design,
which was often around age 5 years (Jernigan et al., 1991; Pfefferbaum et al., 1994;
Sowell et al., 2002; Steen et al., 1997). However, when scans were acquired in the
same individuals at approximately two year intervals, cortical gray matter was shown
to follow an inverted U-type pattern (Fig. 2.3). In the frontal lobes, involved in
planning, organizing, strategizing, and other “executive” functions, the cortical
gray matter reaches its maximal thickness at 11.0 years in girls and 12.1 years
in boys (Giedd et al., 1997, 1999). Temporal lobe cortical gray matter peaks at
16.7 years in girls and 16.2 years in boys. Parietal lobe cortical gray matter peaks at
10.2 years in girls and 11.8 years in boys.
The thickening and thinning of gray matter is thought to reflect changes in the
size and complexity of neurons, not a change in the actual number. The increasing
size may reflect a process called arborization, as the cells grow extra branches, twigs,
and roots, thus growing “bushier” and making a greater number of connections
to other cells. The decreasing amount of gray matter may reflect the process of
pruning where certain connections are eliminated.
40 J. N. Giedd, M. A. Rosenthal, A. B. Rose, et al.

The forces guiding these processes of arborization and pruning are not well
understood. Genetics, nutrition, toxins, bacteria, viruses, hormones, and many
other factors have been shown to have an effect. One hypothesis for the pruning
phase is the “use it or lose it” principle, in which those connections that are used
well survive and flourish whereas those connections that are not used will wither
and die. If this hypothesis is correct, the activities of the child or teenager may have
a powerful influence on the ultimate physical structure of the brain.

Sex differences
Sex differences in the brain are particularly intriguing to psychiatrists studying chil-
dren and adolescents since nearly all psychiatric disorders of childhood onset have
different ages of onset, prevalence, and symptomatology between boys and girls.
Whether environmental or other genetic factors interact with normal differences
between boy and girl brains to account for some of these clinical differences is a
topic of active research (see Ch. 18 for a review of this topic).
In an analysis of 71 male and 50 female healthy subjects between the ages of 4
and 18 years, males had a 9% larger cerebral volume. This difference was consistent
across the entire age range of 4 to 18 years and is similar to that found in adult
postmortem brain (Dekaban and Sadowsky, 1978; Ho et al., 1980) and in vivo
imaging studies (Flaum et al., 1995).
Given the myriad of factors that determine brain structure size, and findings
of enlarged brains in patient groups such as those with fragile X syndrome (Reiss
et al., 1994) and autism (Piven et al., 1995), size difference should not be inter-
preted as imparting any sort of functional advantage or disadvantage. The fact that
gross structural size may be insensitive to sexual differences in receptor density
or connectivity between different neurons further emphasizes the complexity of
interpreting the implications of size differences in brain structures.
Other structures examined in that study included the lateral ventricles, caudate,
putamen, globus pallidus temporal lobe, amygdala, and hippocampus. Of these
structures, when adjusted for total cerebral volume difference by analysis of covari-
ance, only the basal ganglia demonstrated sex differences in mean volume, with the
caudate being relatively larger in females and the globus pallidus being relatively
larger in males.
The sexual dimorphism of the basal ganglia is interesting in light of the frequent
implication of basal ganglia involvement in neuropsychiatric disorders such as
attention-deficit hyperactivity disorder (Castellanos et al., 1994; Hynd et al., 1991)
and Tourette’s syndrome (Hyde et al., 1995; Peterson et al., 1993; Singer, 1993),
which have a higher incidence in males. Obsessive–compulsive disorder, which
41 Brain development in healthy children

has an approximately equal incidence for males and females in late adolescence
and adulthood, may be more common in boys in early childhood (Swedo et al.,
1989). Also, a developmental subtype of this disorder, which demonstrates a more
treatment-refractory course and neurological/basal ganglia abnormalities, is more
common in young boys (Blanes and McGuire, 1997).
Amygdala and hippocampal volume increased for both sexes, but with the amyg-
dala increasing significantly more in males than females and hippocampal volume
increasing more in females. The amygdala and hippocampus findings are consis-
tent with the preponderance of androgen receptors in the amygdala (Clark et al.,
1988) and of estrogen receptors in the hippocampus of rhesus monkeys (Sholl and
Kim, 1989). Further evidence of a relationship between estrogen and hippocampal
volume in rodents is the decreased fiber outgrowth and altered density of dendritic
spines in the hippocampus of gonadectomized adult females, which is reversed with
hormone replacement (Gould et al., 1990; Morse et al., 1986). Similarly, in humans,
women with gonadal hypoplasia have been found to have smaller hippocampi than
controls (Murphy et al., 1993).

Conclusions
MRI has allowed for the safe acquisition of brain scans in healthy children and
has helped to launch a new era of pediatric neuroscience. Characterization of nor-
mal brain development is imperative to assess the hypothesis that many of the
most severe neuropsychiatric disorders of childhood onset are manifestations of
deviations from that normative path. White matter volumes tend to increase in a
roughly linear fashion from ages 4 to 18 years and slopes do not differ significantly
by lobe of the brain. Gray matter volumes, by comparison, tend to follow an inverted
U-shaped developmental curve and differ by lobe. Sexual dimorphism in healthy
brain development may lead to differential vulnerability, which would account for
some of the clinical differences in childhood neuropsychiatric disorders.

REFERENCES

Berlucchi, G. (1981). Interhemispheric asymmetries in visual discrimination: a neurophysiolog-

ical hypothesis. Doc Opthal Proc Ser 30: 87–93.
Bigelow, L. H., Nasrallah, H. A., Rauscher, F. P. (1983). Corpus callosum thickness in chronic
schizophrenia. Br J Psychiatry 142: 284–287.
Blanes, T., McGuire, P. (1997). Heterogeneity within obsessive compulsive disorder: evidence
for primary and neurodevelopmental subtypes. In Neurodevelopment and Adult Psycho-
pathology, ed. M. S. Keshavan, R. Murray. Cambridge, UK: Cambridge University Press,
pp. 206–216.
42 J. N. Giedd, M. A. Rosenthal, A. B. Rose, et al.

Bogen, J. E., Bogen, G. M. (1969). The other side of the brain. 3, The corpus callosum and
creativity. Bull Los Ang Neurol Soc 34: 191–220.
Castellanos, F. X., Giedd, J. N., Eckburg, P., et al. (1994). Quantitative morphology of the caudate
nucleus in attention deficit hyperactivity disorder. Am J Psychiatry 151: 1791–1796.
Clark, A. S., MacLusky, N. J., Goldman-Rakic, P. S. (1988). Androgen binding and metabolism
in the cerebral cortex of the developing rhesus monkey. Endocrinology 123: 932–
940.
Cook, N. D. (1986). The Brain Code: Mechanisms of Information Transfer and the Role of the Corpus
Callosum. London: Methuen.
Dekaban, A. S., Sadowsky, D. (1978). Changes in brain weight during the span of human life:
relation of brain weights to body heights and body weights. Ann Neur 4: 345–356.
Flaum, M., Swayze, V. W., O’Leary, D. S., et al. (1995). Brain morphology in schizophrenia: effects
of diagnosis, laterality and gender. Am J Psychiatry 152: 704–714.
Giedd, J. N., Castellanos, F. X., Casey, B. J., et al. (1994). Quantitative morphology of the corpus
callosum in attention deficit hyperactivity disorder. [See comments] Am J Psychiatry 151:
665–669.
Giedd, J. N., Rumsey, J. M., Castellanos, F. X., et al. (1996). A quantitative MRI study of the
corpus callosum in children and adolescents. Brain Res Dev Brain Res 91: 274–280.
Giedd, J. N., Castellanos, F. X., Rajapakse, J. C., Vaituzis, A. C., Rapoport, J. L. (1997). Sexual
dimorphism of the developing human brain. Prog Neuropsychopharmacol Biol Psychiatry 21:
1185–1201.
Giedd, J. N., Blumenthal, J., Jeffries, N. O., et al. (1999). Brain development during childhood
and adolescence: a longitudinal MRI study. Nat Neurosci 2: 861–863.
Gould, E., Allan, M. D., McEwen, B. S. (1990). Dendritic spine density of adult hippocampal
pyramidal cells is sensitive to thyroid hormone. Brain Res 525: 327–329.
Hellige, J. B., Cox, J. P., Litvac, L. (1979). Information processing in the hemispheres: selective
hemisphere activation and capacity limitations. J Exp Psychol Gen 108: 251–259.
Ho, K. C., Roessmann, U., Straumfjord, J. V., Monroe, G. (1980). Analysis of brain weight. I.
Adult brain weight in relation to sex, age, and race. Arch Pathol Lab Med 104: 635–639.
Hyde, T. M., Stacey, M. E., Coppola, R., et al. (1995). Cerebral morphometric abnormalities in
Tourette’s syndrome: a quantitative MRI study of monozygotic twins. Neurology 45: 1176–
1182.
Hynd, G. W., Semrud-Clikeman, M., Lorys, A. R., Novey, E. S., Eliopulos, D. (1990). Brain
morphology in developmental dyslexia and attention deficit disorder/hyperactivity. Arch
Neurol 47: 919–926.
(1991). Corpus callosum morphology in attention deficit–hyperactivity disorder: morphome-
tric analysis of MRI. J Learn Disabil 24: 141–146.
Innocenti, G. M., Manzoni, T., Spidalieri, G. (1974). Patterns of somesthetic messages transferred
through the corpus callosum. Exp Brain Res 19: 447–466.
Jerison, H. J. (1991). Brain Size and the Evolution of Mind. New York: American Museum of
Natural History.
Jernigan, T. L., Trauner, D. A., Hesselink, J. R., Tallal, P. A. (1991). Maturation of human
cerebrum observed in vivo during adolescence. Brain 114: 2037–2049.
43 Brain development in healthy children

Joseph, R. (1980). Awareness, the origin of thought, and the role of conscious self-deception in
resistance and repression. Psychol Rep 46: 767–781.
Kappers, C. U. A., Huber, G. C., Crosby, C. C. (1936). The Comparative Anatomy of the Nervous
System of Vertebrates including Man, Vol. 2. New York: MacMillan.
Lange, N., Giedd, J. N., Castellanos, F. X., Vaituzis, A. C., Rapoport, J. L. (1997). Variability of
human brain structure size: ages 4 to 20. Psychiatry Res Neuroimaging 74: 1–12.
Levy, J. (1985). Interhemispheric collaboration: single mindedness in the asymmetric brain. In
Hemisphere Function and Collaboration in the Child, ed. C. T. Best. New York: Academic
Press, pp. 11–32.
Levy, J., Trevarthen, C. (1981). Color-matching, color naming and color memory in split brain
patients. Neuropsychology 19: 523–541.
Morse, J. K., Scheff, S. W., DeKosky, S. T. (1986). Gonadal steroids influence axonal sprouting in
the hippocampal dentate gyrus: a sexually dimorphic response. Exp Neur 94: 649–658.
Murphy, D. G. M., DeCarli, C. D., Daly, E., et al. (1993). Effects of the X chromosome on female
brain: a study of turner syndrome using quantitative magnetic resonance imaging. Lancet
342: 1197–1200.
Njiokiktjien, C. (1991). Pediatric Behavioral Neurology, Vol. 3 The Child’s Corpus Callosum.
Amsterdam: Suyi Publications.
Parashos, I. A., Wilkinson, W. E., Coffey, C. E. (1995). Magnetic resonance imaging of the corpus
callosum: predictors of size in normal adults. J Neuropsychiatry Clin Neurosci 7: 35–41.
Paus, T., Zijdenbos, A., Worsley, K., et al. (1999). Structural maturation of neural pathways in
children and adolescents: in vivo study. Science 283: 1908–1911.
Peterson, B., Riddle, M. A., Cohen, D. J., et al. (1993). Reduced basal ganglia volumes in Tourette’s
syndrome using three-dimensional reconstruction techniques from magnetic resonance
images. Neurology 43: 941–949.
Peterson, B. S., Leckman, J. F., Duncan, J. S., Wetzles, R., Riddle, M. A. (1994). Corpus callo-
sum morphology from magnetic resonance images in Tourette’s syndrome. Psychiatry Res
Neuroimaging 55: 85–99.
Pfefferbaum, A., Mathalon, D. H., Sullivan, et al. (1994). A quantitative magnetic resonance
imaging study of changes in brain morphology from infancy to late adulthood. Arch Neurol
51: 874–887.
Piven, J., Arndt, S., Bailey, J., et al. (1995). An MRI study of brain size in autism. Am J Psychiatry
152: 1145–1149.
Rapoport, S. I. (1990). Integrated phylogeny of the primate brain, with special reference to
humans and their diseases. Brain Res Rev 15: 267–294.
Reiss, A. L., Lee, J., Freund, L. (1994). Neuroanatomy of fragile X syndrome: the temporal lobe.
Neurology 44: 1317–1324.
Rosenthal, R. Bigelow, L. (1972). Quantitative brain measurements in chronic schizophrenia.
Br J Psychiatry 121: 259–264.
Shanks, M. F., Rockel, A. J., Powel, T. P. S. (1975). The commissural fiber connections of the
primary somatic sensory cortex. Brain Res 98: 166–171.
Sholl, S. A., Kim, K. L. (1989). Estrogen receptors in the rhesus monkey brain during fetal
development. Dev Brain Res 50: 189–196.
44 J. N. Giedd, M. A. Rosenthal, A. B. Rose, et al.

Singer, H., Reiss, A. L., Brown, J. E., et al. (1993). Volumetric MRI changes in basal ganglia of
children with Tourette’s syndrome. Neurology 43: 950–956.
Sowell, E., Trauner, D. A., Gamst, A., Jernigan, T. L. (2002). Development of cortical and subcor-
tical brain structures in childhood and adolescence: a structural MRI study. Dev Med Child
Neurol 44: 4–16.
Spidalieri, G., Franchi, G., Guandalini, P. (1985). Somatic receptive-field properties of single
fibers in the rostral portion of the corpus callosum in awake cats. Exp Brain Res 58: 75–81.
Steen, R. G., Ogg, R. J., Reddick, W. E., Kingsley, P. B. (1997). Age-related changes in the pedi-
atric brain: quantitative MR evidence of maturational changes during adolescence. Am J
Neuroradiol 18: 819–828.
Swedo, S. E., Rapoport, J. L., Leonard, H., Lenane, M., Cheslow, D. (1989). Obsessive–compulsive
disorder in children and adolescents. Clinical phenomenology of 70 consecutive cases. Arch
Gen Psychiatry 46: 335–341.
Thompson, P., Giedd, J. N., Blanton, R. E., et al. (2000). Growth patterns in the developing brain
detected by using continuum mechanical tensor maps. Nature 404: 190–192.
Tomasch, J. (1954). Size, distribution and number of fibers in the human corpus callosum. Anat
Rec 119: 119–135.
Zaidel, D., Sperry, R. W. (1974). Memory impairment after commissurotomy in man. Brain 97:
263–272.
(1977). Some long-term effects of cerebral commissurotomy in man. Neuropsychology 15:
193–204.
 3

Cognitive development: functional

magnetic resonance imaging studies
Beatriz Luna1 and John A. Sweeney2
1
Western Psychiatric Institute and Clinic, Pittsburgh, USA
2
The Psychiatric Institute, University of Illinois, Chicago, USA

Given the developmental etiology of schizophrenia and its neurobiological profile,

neurodevelopmental models promise to provide important insights into the causes
of this debilitating disorder (Hyde et al., 1992; Keshavan et al., 1994; Waddington
et al., 1991; Weinberger and Lipska, 1995). The onset of schizophrenia typically
presents in late adolescence or early adulthood and is marked by psychotic symp-
toms and cognitive impairments that are believed to have a neurodevelopmental
basis (Ch. 4). Although progress has been made in characterizing structural brain
impairments that may underlie these behavioral deficits, little is known about how
these may play out through development and how they may affect brain function.
Brain abnormalities may result in different adaptive mechanisms, especially given
developmental plasticity, that could affect brain function in a manner not evident
by information of the location of the structural impairment. Regions that are con-
nected to, but distant from, the location of injury could demonstrate impaired
function although the distant structure itself might not be intrinsically impaired.
Pediatric neuroimaging techniques can probe the integrity of brain function and
normal brain maturational processes and provide a window into possible abnor-
malities in neurocognitive development (Luna and Sweeney, 2001). First, however,
we must characterize the development of brain function in a healthy population,
especially during adolescence, in order to identify the processes that are vulnerable
to impairment in schizophrenia. This is a relatively new field and work has only
begun towards the understanding of developmental changes in brain function. In
this chapter, we will describe the initial investigations and discuss what has been
found regarding the changes in brain function that support the healthy maturation
of cognitive control of behavior.

Neurodevelopment and Schizophrenia, ed. Matcheri S. Keshavan et al. Published by Cambridge University
Press.

C Cambridge University Press 2004.

45
46 B. Luna and J. A. Sweeney

Brain and cognitive maturation

Basic cognitive processes, which are evident in infancy and show dramatic changes
throughout childhood (Diamond and Goldman-Rakic, 1989), continue to develop
throughout adolescence. This period of late childhood and adolescent development
is characterized by improvement in the capacity for abstract thought, planning, and
cognitive flexibility (Levin et al., 1991; Luciana and Nelson, 1998; Welsh et al., 1991).
As adult levels of basic cognitive processes are reached, we see improvements in
executive processes, such as working memory and voluntary response suppression,
that support goal-directed behavior.
Cognitive development occurs concurrently with significant changes in brain
maturation. The degree of cortical folding (Armstrong et al., 1995), overall size,
and regional functional specialization of the brain is, for the most part, in adult
form by early childhood (Caviness et al., 1996). However, significant refinements
of brain systems, including synaptic pruning, elaboration of dendritic arborization
(Changeux and Danchin, 1976; Huttenlocher, 1990), and increased myelination
(Jernigan et al., 1991; Pfefferbaum et al., 1994; Yakovlev and Lecours, 1967), con-
tinue into adolescence. Of special relevance to the maturation of executive abilities
is the continued elimination of synapses in the prefrontal cortex (layer III of the
middle frontal gyrus) throughout adolescence (Huttenlocher, 1990). The emer-
gence of higher-order cognitive control of behavior is believed to be subserved
by the integration of prefrontal cortex with other brain regions in widely dis-
tributed circuits (Chugani, 1998; Fuster, 1997; Goldman-Rakic, 1990; Thatcher,
1991). Huttenlocher’s (1990) findings, indicating a late stabilization of synaptic
pruning in prefrontal relative to visual cortex, have led to a traditional view that
brain maturation occurs last in frontal cortex. In contrast, histological studies and
recent magnetic resonance imaging (MRI) studies indicate that, similar to the non-
human primate, the human cortex develops in a concurrent fashion (Giedd et al.,
1999; Rakic et al., 1986; Yakovlev and Lecours, 1967). The effectiveness of prefrontal
control of behavior however, may continue to develop as other regions become more
specialized and myelination allows for the integration of function throughout the
brain.
In order to understand how cognitive development is supported by brain mat-
uration, it is crucial to investigate brain function. Changes in the microstructure
of the brain during childhood, and especially adolescence, though highly signifi-
cant from a functional perspective, are not readily observable in vivo with con-
ventional anatomic neuroimaging techniques (Ch. 2). Cognitive capabilities, how-
ever, improve at a relatively dramatic pace, indicating that robust changes in brain
microstructure and organization must also be occurring. Investigating brain func-
tion allows us to assess how brain processes develop relative to their structural
47 Cognitive development: fMRI studies

capacity. Brain weight is more than 90% of the adult level by school age (Caviness
et al., 1996); however, myelination and synaptic pruning continue throughout this
period without dramatic consequences on gross brain anatomy. Investigating brain
function allows us to see the effects of these structurally subtle, but significant,
processes. The elimination of redundant synaptic connections supports quicker
and more efficient regional neuronal processes, allowing for more focused brain
function with the increased capacity for more complicated neuronal operations.
Myelination speeds up neuronal transmission and allows distant brain regions to
participate more effectively in widely distributed circuitry, which, allows for top-
down, prefrontal executive control of behavior.
It has only recently become feasible to use neuroimaging procedures in healthy
pediatric populations. The available neuroimaging methods, such as positron emis-
sion tomography (PET) and single photon emission tomography, depend upon
ionizing radiation. Consequently, these two lines of investigation, cognitive devel-
opment and brain maturation, had remained separate with limited efforts at rec-
onciling or integrating them. With the advent of functional MRI (fMRI), which
allows non-invasive in vivo investigation of brain processes that underlie cognitive
function, we can now study the link between these psychological and neurobi-
ological phenomena. This technique has been used to investigate how the func-
tional organization of brain regions contributing to cognitive abilities changes with
development to support the emergence of adult-level cognition. Initial studies have
provided crucial information for delineating the brain circuitry underlying healthy
cognitive development.
Electroencephalography and PET were initially used to characterize developmen-
tal changes in brain function. Thatcher (1991) measured electroencephalography
coherence among neocortical regions and reported an increase in coherence activity
throughout adolescence, especially between frontal and other cortical areas. PET
results support these findings by indicating that local cerebral resting metabolic
rates decrease in frontal, parietal, and temporal regions throughout childhood
and only reach adult levels in adolescence (Chugani, 1998). These are significant
findings indicating a late integration of frontal regions to widely distributed brain
function; however, there are limitations to these methods and findings. The elec-
troencephalography studies lack sufficient spatial resolution to localize adequately
developmental changes in specific brain regions. Because PET is invasive, it has
rarely been used to study healthy pediatric populations.

Functional magnetic resonance imaging

Brain function can be imaged in vivo during cognitive activity using fMRI. Because
it is non-invasive, this method permits the characterization of changes in brain
activity that occur during development in pediatric populations. The minimal
48 B. Luna and J. A. Sweeney

Stimulus
Mirror IR Mirror

Microphone Rear Projection Screen

& Speakers Respiratory Projector
Belt

Head Coil

IR Camera Alarm Bell

Ball
H.R. Monitor

Fig. 3.1. Schematic of the MR environment. IR, infrared; H.R., heart rate.

health risk from repeated MRI studies also permits longitudinal fMRI studies of
developmental changes in brain–behavior systems. This is especially important as it
provides a basis for actually following developmental trajectories in individual sub-
jects, rather than utilizing only cross-sectional studies to investigate developmental
trends. Since development often proceeds in bursts rather than in a smooth, grad-
ual way (Thatcher, 1991), longitudinal studies provide a better way to characterize
patterns of development. For these reasons, fMRI is a valuable tool for bridging
cognitive and biological studies of brain development.
The principle of fMRI is that neuronal activation associated with cognitive
activity results in an increased metabolic demand; that metabolic demand, in turn,
brings increased blood flow to the active region. The increase in blood flow produces
a change in the ratio of oxygenated to deoxygenated blood, which changes the
magnetic properties of blood in a way that can be detected. Using sophisticated
image reconstruction and analysis procedures, these changes in regional magnetic
properties can be used to provide a detailed picture of brain regions involved in
performing a particular cognitive act. The psychological tasks subjects perform in
the MRI scanner need to be carefully selected to probe specific cognitive processes
of interest. A diagram of the MRI environment is presented in Fig. 3.1.
There are two primary fMRI methods used for identifying brain regions under-
lying specific cognitive processes: block design and event-related procedures. The
block design is a boxcar method used to identify the group of regions participating
in a cognitive process. Blocks of experimental trials are contrasted with alternating
blocks of baseline brain function; analyses identify the voxels where signal fluctuates
in accordance with changing block trial types. Baseline trials can vary from eyes
closed or fixation to trials identical to experimental trials except for the variable
of interest. The boxcar design provides the best characterization of specific brain
regions. The advantage of this method is that it is relatively easy to implement and
analyze, and it generates a relatively high signal-to-noise ratio, resulting in robust
49 Cognitive development: fMRI studies

brain activation. The limitation to this method is that one cannot disassociate the
different cognitive processes occurring in each experimental trial since these are
averaged across blocks, limiting the ability to investigate the relative contributions
of each of the brain regions in a circuit. Event-related designs are used to answer
this concern by acquiring brain images time-locked to the occurrence of each cog-
nitive event in a trial. In this manner, different brain regions can be associated with
specific cognitive steps in a task and regions that are primary to each cognitive pro-
cess can be identified. This method, however, typically has a significantly reduced
signal-to-noise ratio compared with block design studies, generating modest areas
of brain activation. Because of the limitations of each method, investigators will
often use both designs. A block design is used to identify all the brain regions in a
circuit; event-related designs are then used to disentangle the specific contributions
of each region to a cognitive process of interest.
Studies using fMRI to characterize brain function in healthy adults are relatively
straightforward. However, pediatric studies necessitate the comparison of brain
activation between different age groups, which requires additional methodological
considerations and statistical analyses. Initial studies reported differences in brain
activation determined from visual inspection between a group of children and
separate, but similar, adult studies (Casey et al., 1995; Hertz-Pannier et al., 1997).
Presently, pediatric studies study children and adult comparison groups in the same
study. Testing subjects ranging in age from childhood to adulthood provides the
greatest power, as regression analyses can identify brain regions that change with
age.

Changes in brain function supporting cognitive development

Two higher-order cognitive abilities crucial to the voluntary control of behavior are
working memory (Case, 1992; Fry and Hale, 1996) and voluntary suppression of
context-inappropriate responses (Bjorklund and Harnishfeger, 1990; Case, 1992;
Ridderinkhof and van der Molen, 1997; Wilson and Kipp, 1998). These processes
require brain circuitry that supports online calculations for the appropriate behav-
ioral responses to ongoing environmental demands. It is these cognitive processes
that have a protracted developmental sequence into adolescence and are impaired
in schizophrenia (Cohen et al., 1999; Merriam et al., 1999; Park and Holzman, 1992;
Weinberger et al., 1986).

Working memory
Working memory is the ability to maintain and manipulate information online
in order to guide goal-directed behavior (Baddeley, 1983). This circuitry has been
well delineated in the single-cell literature and is known to be subserved by a widely
50 B. Luna and J. A. Sweeney

distributed circuitry in which prefrontal regions play a primary role (Fuster, 1997;
Goldman-Rakic, 1992). Delay-dependent cells found in prefrontal cortex have the
unique feature of responding exclusively to the maintenance of information, as
distinct from sensory and motor responses. Brain regions with neurons having this
property and that are part of a circuitry that supports these processes are critical
for working memory.
Several pediatric studies have compared brain activation between children and
adults during performance of spatial and non-spatial working memory tasks. Work-
ing memory studies require that subjects retain a spatial location or a sequence of
non-spatial stimuli in memory and use this information to guide a future response.
These studies have either found similar brain function in children and adults or
increased activation in children relative to adults. In one spatial working memory
task, six children aged 6–11 years and six adults were asked to recall the location
of a visual stimulus that occurred in previous trials (Thomas et al., 1999). Despite
differences in performance, children demonstrated similar patterns of activation as
that in adults in their prefrontal and parietal regions. They did not, however, show
bilateral activation of cingulate cortex, temporal regions, and premotor regions,
which are known to underlie response preparation and were recruited by adults. In
another study, nine children aged 8–11 years performed a spatial working memory
task where they had to press a button corresponding to the location of a previ-
ous prompt (Nelson et al., 2000). The children activated the same brain regions as
adults, including dorsal prefrontal, anterior cingulate, and posterior parietal cor-
tices. These results suggest that the brain circuitry underlying working memory is
mainly in place by 8–11 years of age; however, differences in the relative contri-
bution of different regions of this circuitry may change with age. In a non-spatial
working memory n-back task, six children aged 9–11 years were required to press
a button when a letter was repeated with one intervening non-match letter (Casey
et al., 1995). Brain activation during these test trials were compared with blocks of
trials where subjects pressed a button whenever the letter “X” appeared. Coronal
slices focused on prefrontal regions. Results indicated a greater magnitude of acti-
vation in prefrontal cortex in children than adults. These results suggest that, while
the correct circuitry is accessed by children when they maintain information in
working memory, children rely on a greater contribution of prefrontal cortex than
do adults. Increased activation of prefrontal cortex could indicate either an imma-
ture specialization of the underlying neuronal processes, such as synaptic pruning,
or a higher reliance on prefrontal cortex during a task that is harder for children
to perform. (See below for issues regarding age-related performance differences in
developmental fMRI studies.) Increases in task difficulty are known to generate
increases in brain activation (Szameitat et al., 2002).
51 Cognitive development: fMRI studies

Other studies have found evidence for increased participation of brain regions
with age during working memory tasks. Thirteen subjects aged 9–18 years per-
formed a spatial working memory task during whole-brain imaging (Klingberg
et al. 2002). Subjects were presented with stimuli that appeared sequentially in a
4 by 4 grid for three or five steps. After a 1.5 second delay, subjects had to respond
if the present stimulus was in a location prompted in the previous sequence. Age-
related increases in activation were found in the superior frontal sulcus and intra-
parietal sulcus, which corresponded to increased working memory capacity. Similar
results were found in another spatial working memory task that was used to char-
acterize changes in brain function related to age progression in eight children, eight
adolescents, and seven adults (Kwon et al., 2002). A block design compared acti-
vation during trials where subjects had to press a key if an “O” appeared in the
same location as two previous trials (2-back) compared with rest trials where sub-
jects responded when the “O” stimulus appeared at center. Reaction time decreased
with age while accuracy remained stable. Multiple regression analyses were used
to correlate the magnitude of brain function with age. Results showed that brain
function increased with age throughout prefrontal and premotor regions as well as
throughout posterior parietal regions, while no age-related decreases in activation
were found. These results remained after factoring out performance differences
between age groups. These two studies are unique in that they looked at age as
a continuous variable from childhood to adulthood and, therefore, may be more
sensitive than studies on discrete age groups in identifying brain regions that show
subtle changes with age.
Taken together, these studies indicate that the basic brain circuitry underlying
spatial working memory is in place by childhood; however, with increasing age, there
are changes in the degree that different regions in this circuitry are recruited. The
prefrontal cortex, which is known to play a primary role in mature working memory,
demonstrates the most consistent change with age in these different studies.

Voluntary response suppression

The ability to choose, through response suppression and interference control, what
stimulus or plan to use to guide behavior is another essential aspect of higher exec-
utive cognition (Bjorklund and Harnishfeger, 1995). This ability, though present in
infancy, matures throughout adolescence, supporting increasing voluntary control
of behavior (Bjorklund and Harnishfeger, 1990). This is also a cognitive process
that is impaired in neurodevelopmental disorders, especially schizophrenia (Klein
et al., 2000; McDowell and Clementz, 2001; Sereno and Holzman, 1995).
Initial studies established that, like in the working memory task, children and
adults recruit a similar brain network during response suppression. Prefrontal
52 B. Luna and J. A. Sweeney

cortex was imaged in nine children aged 7–12 years and nine adults while they
performed a go–no-go task (Casey et al., 1997). Subjects had to button press in
response to a centrally presented letter (go-trials) except when the letter was an “X”
(no-go trials). Children made more false-alarm errors than adults. Children acti-
vated the middle frontal gyrus (Broca’s area 9, 10, 46) as adults did; however, they
showed a greater volume of activation. This study, as in the working memory results,
provides evidence that children rely on increased participation of prefrontal cortex
during response suppression, which could be a result of changes in the underlying
anatomy or of increased effort by children to complete the task.
Evidence for increasing participation of multiple brain regions with age
during performance of tasks requiring voluntary response suppression has also been
provided by whole-brain fMRI. In one study, 16 children aged 8–12 years and 16
adults performed a combination task with flanker and go–no-go trials (Bunge et al.,
2001). In the flanker trials, subjects pressed a button corresponding to the direction
of a central arrow flanked on both sides by two arrows pointing in the opposite
direction to the central arrow (response interference), two arrows pointing in the
same direction as the central arrow (no response interference), diamond shapes
(no response interference), or “X” symbols to indicate no response (no-go trials).
Children made more interference and response-suppression errors than adults.
Results from correct trials only indicated that children, as well as adults, recruited
prefrontal cortex during successful interference suppression, but in the opposite
hemisphere. During successful response inhibition, children, unlike adults, relied
more heavily on posterior rather than prefrontal regions. Children did not recruit
right ventrolateral prefrontal cortex as adults did during both tasks. These results
indicate that, with age, there is increased participation of prefrontal regions during
response-suppression tasks. A study on adolescents showed similar results. Nine
subjects aged 12–19 years were compared with eight healthy adults aged 22–40 years.
Dorsal cortical regions were imaged while subjects performed a stop task, where
they were required to press a button only if a picture of an airplane was not followed
by a picture of a bomb (Rubia et al., 2000). Performance did not differ between
age groups; however, adolescents demonstrated reduced activation of prefrontal
regions and increased activation in caudate and inferior frontal gyrus relative to
adults. Age-related increases in prefrontal function were also found during a Stroop
interference task (Adleman et al., 2002). Thirty subjects aged 7–22 years performed
a Stroop test where subjects had to read a list of color words printed in black
font, name the color of a series of “X” symbols, and say the color of the ink of
color words that were discordant for color font. Dorsal cortical regions of the
brain were imaged. Results indicated that parietal involvement in the Stroop task is
mature by adolescence, while prefrontal activation continues to increase into adult-
hood. These results support findings from working memory studies indicating that
53 Cognitive development: fMRI studies

specialization of prefrontal function changes with age in a way that extends or

elaborates its primary role in the voluntary control of behavior.
While it is generally accepted that prefrontal cortex plays a crucial role in higher-
order cognition, it is also now well established that higher-order executive cognitive
functions are subserved by widely distributed and integrated brain systems rather
than by the prefrontal cortex independently (Goldman-Rakic, 1988). It is possible
that, in conjunction with developmental improvements in the intrinsic computa-
tional capacity of prefrontal cortex, there is increased integration with other brain
regions as well. We do not yet know details about the maturation of the functional
integration of widely distributed circuitry.
Our own work has focused on characterizing changes in the distribution of brain
function from childhood through adolescence during the performance of a volun-
tary response suppression task (Luna et al., 2001). We conducted an fMRI study
of 36 healthy subjects ranging in age from 8 to 30 years to study developmental
changes in the brain substrate of antisaccade task performance. The antisaccade
task requires that the prepotent tendency to make an eye movement towards a new
visual stimulus be suppressed and instead a voluntary eye movement be generated
to the opposite location. Developmental studies have indicated that the number of
instances of response-suppression failures significantly decreases with age through
adolescence (Fischer et al., 1997; Munoz et al., 1998) suggesting that, the capacity to
suppress a response is present early on (a child can perform a correct antisaccade)
but efficiency (the ability to perform correct antisaccades consistently) continues
to improve into adolescence. In our fMRI study, the subjects were instructed to
look in the opposite direction of targets that appeared in the periphery when pre-
ceded by a red fixation cross-hair. These were compared with blocks of trials where
a green cross-hair indicated that subjects were to look toward peripheral targets.
Activation across whole brain was compared in children, adolescents, and adults.
Results indicated that, across the age span, subjects recruited a widely distributed
circuitry, including frontal, supplementary, and parietal eye fields; dorsolateral pre-
frontal cortex; thalamus; and striatum (Fig. 3.2). In contrast, adults demonstrated
increased activation in regions known to play a role in the establishment of the
preparatory state needed to perform antisaccades successfully including frontal and
parietal eye fields as well as the superior colliculus (Everling et al., 1999; Everling
and Munoz, 2000). Additionally, only adults recruited regions in the lateral cere-
bellum that underlie cognitive processes related to timing and learning (Kim et al.,
1994). Adolescents demonstrated increased activation of the dorsolateral prefrontal
cortex and striatum relative to children, apparently reflecting their ability to recruit
frontostriatal circuitry more effectively to perform the antisaccade task. Children,
compared with older subjects, relied more on posterior parietal regions, reflect-
ing increased use of visuospatial processing. Taken together, fMRI and behavioral
Fig. 3.2. Group activation maps (t ≥ 4.0) during an antisaccade task relative to a visually guided
prosaccade task superimposed on the structural anatomic image of a representative subject
(Female aged 26 years) warped into Talairach space. Columns show the average activation
for each age group. Rows depict the orientation (a,b, sagittal; d, axial; e,f, coronal) that
optimally illustrate activation in brain regions of interest. Ant-Cing, anterior cingulate; DM-
TH, dorsomedial thalamus; Pre-SMA, presupplementary area; SEF, supplementary eye fields;
Prec, precuneus; SC, superior colliculus; sFEF, superior precentral sulcus aspect of the frontal
eye field; IPS, intraparietal sulcus; BG, basal ganglia; DLPFC, dorsolateral prefrontal cortex;
SMG, supramarginal sulcus; Lat Cer, lateral cerebellum; DN, dentate nucleus. (Reprinted from
NeuroImage, Vol. 13, Luna, B., Thulborn, K. R., Munoz, D. P., Merriam, E. P., Garver, K. E.,
Minshew, N. J., et al., “Maturation of widely distributed brain function subserves cognitive
development” 786–793, Copyright 2001, with permission from Elsevier.) (See plate section
for color version.)
55 Cognitive development: fMRI studies

studies indicate that the increased integration of distant brain regions underlie
improved efficiency in the voluntarily cognitive control of behavior that occurs
throughout childhood and adolescence.
These results support other studies indicating that the increased participation
of brain regions that underlie executive and attentional control are important in
the maturation of voluntary response suppression. We extend these findings by
proposing that the recruitment of regions that underlie preparatory and refined
responses are also crucial to adult-level performance. Children are able to suppress
prepotent responses and interference, but what improves with age is the consis-
tency in which this behavioral state is maintained. We propose that this stage of
maturation, encompassing adolescence, may be characterized by increased reliance
on more widely distributed brain function that incorporates regions which are
not primary in the generating of a voluntary response but, instead, improve the
consistent control of behavior by a more concerted effort of the whole brain.
Developmental fMRI studies of both working memory and voluntary-response
suppression indicate that brain function does change with age and with increases
in performance; however, these changes are subtle and require further study. We
still need to characterize the relative contributions of localized brain regions and
the activities that are dependent on the integration of whole-brain processes. These
initial studies do show us that there are important changes occurring during adol-
escence that could be vulnerable either to emergence of schizophrenia or the neu-
rodevelopmental pathology that precedes it. With the background of methods and
findings from normative psychological developmental research, we can now begin
to explore possible failures in development that are associated with this disorder.
The occurrence of abnormalities of prefrontal processes in schizophrenia is now
widely supported by a range of histological, neuroimaging, and behavioral studies.
We have seen that this region is particularly sensitive to developmental changes and
may be vulnerable to impairment. Additionally, with neuroimaging techniques, we
can characterize the integrity of integrated brain function and begin to understand
the implications of localized impairment or failure to integrate the whole brain
and the relationship these might have to neurodevelopmental psychiatric disorders
(Luna and Sweeney, 1999; Luna et al., 2002). By studying at-risk populations, we
can examine when the developmental dysmaturation of executive abilities begins to
diverge most dramatically from normal processes (Ch. 23 reviews high-risk stud-
ies). As discussed below, we are still refining fMRI tools, especially with relevance
to pediatric populations; however, well-defined developmental experiments with
pediatric patient populations can begin to elucidate the nature of an impairment.
In the next section, we present how pediatric fMRI studies have informed us of
the development of other abilities and what they have shown us about impaired
development.
56 B. Luna and J. A. Sweeney

Development of brain function supporting language

Pediatric fMRI studies of language development have demonstrated similar trajec-
tories as those characterized in studies of the development of executive abilities.
Most pediatric fMRI studies have been performed to assist in the localization of
language areas to guide excision lesions to relieve epileptic seizures (Benson et al.,
1996; Hertz-Pannier et al., 1997; Stapleton et al., 1997). These studies provide fur-
ther evidence of changing brain function with increasing language proficiency with
age. The integrity of temporal regions has most typically been assessed using reading
and passive language listening tasks. Results demonstrate similar language circuitry
and lateralization for children and adults (Hertz-Pannier et al., 2001). Activation
of brain regions known to underlie sentence processing, including Broca’s and
Wernicke’s areas, as well as prefrontal and occipital cortices, was evident in chil-
dren aged 8–13 years. These children also showed increases in activation of left
temporal cortex during reading, as previously shown for healthy adults (Gaillard
et al., 2001a). Children also demonstrate recruitment similar to adults of frontal,
parietal, occipital, and cingulate cortices during auditory sentence comprehension
and verb generation (Booth et al., 1999). These results, like those from studies of
executive abilities, demonstrate that the basic circuitry needed for higher-order
function is in place by childhood.
Increases with age in activation within this circuitry have also been reported. Left
hemisphere lateralization of Broca’s area for semantic language function during a
verb generation task was found to increase significantly with age from 7 to 18 years,
concurrent with increases in linguistic proficiency (Holland et al., 2001). Evidence
from a fMRI study of 74 healthy subjects aged 7 to 17 years demonstrated no
changes in brain function with age in phonological processing, but age-related
changes in brain function in inferior frontal gyrus and precentral sulcus during
a semantic categorization task (Shaywitz et al., 2002). Frontal lobe activation was
compared in 10 children aged 8–13 years and 10 healthy adults aged 19–48 years
while they performed a verbal fluency task (Gaillard et al., 2000). Children were
found to activate the same frontal regions as adults including posterior inferior
frontal gyrus, dorsolateral prefrontal cortex, and cingulate cortex, but with a 60%
greater extent of activation. As mentioned above, this increase in the recruitment
of frontal regions could reflect the larger effort exerted by children to perform the
task or may be related to immature neuronal processes. Taken together, these results
confirm the primary role in the integration of frontal regions in the maturation of
higher-order cognition.
In order to delineate age changes in brain activation resulting from differences
in performance, children who performed at adult levels were compared with those
who performed worse than adults (Schlaggar et al., 2002). Results indicated that age
57 Cognitive development: fMRI studies

differences detected in lateral prefrontal regions and posterior extrastriate regions

were related to performance differences; however, those detected in neighboring
areas in more dorsal medial prefrontal regions (absent in children) and more ante-
rior extrastriate regions (increased activation in children) resulted only from age
differences and may reflect maturational changes in the brain that affect how regions
are recruited. These results show both increases and decreases in age-related activa-
tion. Greater activation in children relative to adults, as discussed above, may reflect
the increased effort by children to perform at adult levels. Decreases in activation
in children relative to adults may reflect decreased access to that region, immature
local circuitry, or the recruitment of an alternative circuitry.
The above results support findings from fMRI studies of cognitive develop-
ment showing continued changes in brain function with increases in development
throughout childhood and adolescence. It is striking how, in both cognitive and
language development, frontal regions appear to account for a significant number
of age-related differences.

Abnormal populations
Pediatric fMRI studies have already provided insight into the possible factors under-
lying the etiology of developmental abnormalities such as ADHD and dyslexia.

Attention-deficit hyperactivity disorder

Attention-deficit hyperactivity disorder (ADHD) is the most prevalent childhood
developmental disorder and is characterized by inattention, impulsivity, and hyper-
activity. Studies using fMRI have indicated impaired function of prefrontal cortex
and striatum. In one study, subjects aged 12–18 years who had been diagnosed with
ADHD performed a response-suppression task, which showed decreased func-
tion in prefrontal cortex while the premotor and parietal cortices and caudate
nucleus demonstrated similar activation as controls (Rubia et al., 2000). In another
study, 10 subjects aged 8–13 years diagnosed with ADHD were compared with
six age-matched controls (Vaidya et al., 1998). Subjects were tested before and
after methylphenidate treatment. Subjects performed a letter go–no-go task where
they refrained from responding on trials where an “X” appeared. Children with
ADHD demonstrated poorer inhibitory performance and increased frontal and
reduced striatal activation than controls. Methylphenidate improved performance
and normalized activation in prefrontal and striatal regions in patients. These results
provide evidence for frontostriatal involvement in ADHD.

Dyslexia
Dyslexia is an impairment in the ability to read that is not caused by reduced
intellectual capabilities. Neuroimaging studies indicate abnormalities in the
58 B. Luna and J. A. Sweeney

systems underlying phonological processes in dyslexic individuals, and fMRI has

indicated that dyslexic children fail to recruit parietotemporal and extrastriate
regions known to support orthographic processing (Temple et al., 2001). Simi-
lar results were obtained when a large number (74) of dyslexic subjects aged 7–18
years were compared with 70 normal readers aged 7–17 years (Shaywitz et al., 2002).
Dyslexic children had reduced activation of parietotemporal and occipitotemporal
regions compared with normal readers, instead relying on homologous right hemi-
sphere regions and the anterior cingulate cortex to perform the semantic category
task. These studies indicate a failure to recruit regions throughout circuits known
to underlie processes crucial to reading, primarily in parietotemporal regions, in
dyslexic subjects. Additionally, abnormalities in cortical visual region V5/MT have
also been found during motion processing in dyslexic subjects, indicating that, in
addition to abnormalities in phonological processing, impaired visual processing
may also underlie this disorder (Eden et al., 1996). Instead, dyslexics demonstrate
increased activation of Broca’s area as they struggle with phonological processing
(Georgiewa et al., 2002). In contrast to ADHD, these results illustrate abnormalities
associated with dysfunction of posterior brain processes.

Pediatric fMRI studies of schizophrenia

Given the very low incidence of childhood schizophrenia and the additional compli-
cations making that diagnosis in young children, it is difficult to test neurodevelop-
mental theories of schizophrenia directly on sufficient numbers of child patients.
Further, studies of affected children would make it difficult to disentangle dis-
ease effects from neurodevelopmental risk mechanisms. Pediatric offspring of a
schizophrenia parent are particularly well suited to investigate the developmental
nature of schizophrenia. Offspring of schizophrenia patients have a 10–16% risk
of developing schizophrenia–related disorders, compared with 1% in the general
population (Gottesman et al., 1982). We have initiated pediatric fMRI studies in the
offspring of individuals with schizophrenia (a high-risk group) to investigate brain
function underlying executive function in schizophrenia (Keshavan et al., 2002).
Brain function during a spatial working memory task, the oculomotor delayed
response task, were compared between four high-risk adolescents and four age-
and sex-matched healthy controls. In this task, subjects had to generate a saccade in
the absence of a visual stimulus to a location that had been previously indicated and
had been retained in working memory for several seconds. In order to isolate brain
function associated with working memory from oculomotor processes, the block of
oculomotor delayed response trials were compared with blocks of visually guided
saccade trials where subjects made similar saccadic eye movements but to visual
targets appearing in the periphery. During the working memory task, the high-risk
subjects had reduced prefrontal Broadmann’s areas 9 and 46 and parietal activation
59 Cognitive development: fMRI studies

relative to the healthy group. These results suggest that abnormalities in frontal and
parietal heteromodal association cortex may precede illness onset and represent
a neurodevelopmental failure that contributes to risk for the disorder, indicating
that studies of the high-risk group may be valuable in testing neurodevelopmental
models of schizophrenia.

Challenges to pediatric neuroimaging

Pediatric neuroimaging requires consideration of issues specific to the testing of
children. Non-compliance has to be considered and methods are used to engage
the child’s participation while ameliorating any fears. Training with the tasks that
will be presented in the scanner previous to the MRI session is often very helpful
to enable the child to focus on attainable goals when testing occurs. Using stimuli
and tasks that are particularly amenable to attract children’s attention can also be
of benefit during long tasks. The prospect of being “tested,” especially inside an
enclosed space such as an MR scanner, can be particularly stressful to children. In
order to reduce the anxiety and acclimate subjects to this constrained environment,
many laboratories now have subjects spend time inside a non-working model of a
scanner previous to the testing. In our experience, children and adolescents who
spend time inside a “mock” scanner prior to imaging studies have significantly
reduced heart rates and self-reported distress levels during subsequent imaging
studies (Rosenberg et al., 1997). Children who did not spend time in a “mock”
scanner had higher heart rates and self-reported distress levels in the actual scanner
than did simulation-trained subjects.
Head motion is one of the most significant sources of artifact in fMRI imag-
ing. fMRI activation maps are typically developed using more than 100 images
of the brain, which are acquired over the course of a study that may last 5 to 10
minutes. Head movement during image acquisition compromises the ability to
detect brain function because the brain tissue from which image data are obtained
varies over the course of a study. For this reason, head motion artifact can dramat-
ically underestimate brain function. This is particularly problematic with children
and patient groups, who can have a more difficult time than adults remaining still
(Logan, 1999; Poldrack et al., 2002; Thomas et al., 1999). Head motion is usually
corrected for by algorithms that align all images to a reference image (Cox, 1996;
Eddy et al., 1996; Woods et al., 1993). Head movement larger than the level of
resolution prescribed, displacement larger than a half to two-thirds of a voxel, can
often not be adequately corrected and such data cannot be included in analyses.
Therefore, restraints are often used to reduce head motion. A bite bar affords an
anchor that reduces head motion significantly; however, subjects, especially chil-
dren and patient groups, can actually move more with such devices because they do
60 B. Luna and J. A. Sweeney

not tolerate the discomfort. Less-restrictive systems include using pillows to pack
the head into the head coil, minimizing range of movement, and visual feedback
of head location to help subjects to reposition themselves to their original position
(Thulborn and Shen, 1999). Training subjects to keep their heads still through sen-
sory feedback in a mock scanner also helps to minimize head motion (Slifer, 1996).
Pediatric neuroimaging studies must report the amount of head movement found
in their subjects lest their results be confounded by this artifact. This is of par-
ticular concern when studies find a reduction of activation throughout the brains
of children relative to adults with no regions demonstrating similar magnitude of
activation.
Children have faster respiratory and heart rate cycles than adults. Respiration can
lead to physiological artifact by introducing variations in the static magnetic field,
which can lead to spurious and decreased activation (Kemna and Posse, 2001; Raj
et al., 2001). Increased pulse rate can result in motion within the brain, affecting
brainstem structures the most, but also basal ganglia and, to a lesser degree, the
cortical surface (Gaillard et al., 2001b). Algorithms are being devised to remove
this artifact (Frank et al., 2001) but there is still limited experience in using such
approaches with pediatric samples.
A great misconception is that there are dramatic differences in the gross mor-
phology of the child and adult brains that make any direct comparison of functional
responses between age groups seriously flawed. This view implies that children’s
brains would be warped to a higher degree than adults when warping brain struc-
ture into a common adult brain atlas such as the Talairach atlas (Talairach and
Tournoux, 1988). Studies have shown, however, that brain size and gross morphol-
ogy at age 5 is not significantly different from that of adults (Caviness et al., 1996;
Giedd et al., 1996; Reiss et al., 1996) and the variability in brain structure between
children and adults is less than the normal variance between adult subjects. Nasal
sinuses continue to form into adolescence, perhaps resulting in increasing sources
of artifacts in brain regions bordering these areas. Neck length also continues to
increase developmentally, raising concerns regarding the optimal placement of a
child’s head on a head coil designed for an adult (Gaillard et al., 2001b). This could
result in children not reaching the “sweet” spot of the head coil where optimal
images are obtained.

Differences in performance
A major concern of pediatric neuroimaging is understanding if age differences in
brain function result from differences in performance or actually reflect differences
in functional brain systems (Bookheimer, 2000). Some investigators choose to use
tasks where children perform at the same levels as adults (Casey et al., 1995).
The primary result in these studies, perhaps not surprisingly, has been similar
61 Cognitive development: fMRI studies

activation across ages. When differences in activation are found, these may reflect
underlying differences in anatomy or differences in the computational demand
required to equate performance across age groups. However, age differences in
brain activation where performance levels of the task differ does not exclude the
possibility that adult patterns of activation would be evident in the child brain if
performance was equated. These two approaches are useful in answering different
questions regarding development. The former approach is needed to identify the
presence of mature brain circuitry, while the latter allows the characterization of
the immature aspects of the brain systems supporting a specific cognitive process.
Although mature circuitry may be accessed by children given optimal conditions,
it is important to characterize the brain regions that are not being accessed when
performance fails.
There are methods available to gain the benefits of both approaches. One
approach is to identify the brain regions that demonstrate age-related changes owing
to differences in performance and those that are independent of performance dif-
ferences. One study identified performance-related activity by identifying regions
that had greater activation in children that performed worse than adults but not
in children with equivalent performance levels as adults (Schlaggar et al., 2002).
Brain regions that show age-related differences in children with both poor and
equivalent performance as adults are independent from performance differences
and may reflect differences in the underlying neuroanatomy. Another approach is
to change task difficulty parametrically. As task difficulty increases, differences in
brain activation resulting from age differences should emerge. The variables that
consistently differ between ages across different task difficulties can be identified
as demonstrating developmental changes. In this manner, we can identify regions
that demonstrate age differences owing to performance differences and those that
are independent of performance and may indicate differences in neuroanatomy
(Benson et al., 1996).

Future directions
The goal of pediatric neuroimaging is to characterize the interaction between brain
maturation and cognitive development. We still need to characterize how synaptic
pruning and myelination processes contribute to age-related changes in the func-
tional organization of brain systems. The next step in pediatric neuroimaging is to
gain more direct characterization of brain maturation while obtaining indices of
brain function.
Diffusion tensor imaging (DTI) provides an indirect measure of myelination in
vivo by distinguishing the motion properties of water molecules inside axonal tracts
(Basser et al., 1994; Conturo et al., 1996; Le Bihan et al., 2001). The confinement
62 B. Luna and J. A. Sweeney

produced by myelin and tract tissue forces water molecules to diffuse rapidly in
an anisotropic manner in the direction of the fibers, in contrast to non-directional
isotropic diffusion outside fiber tracts (Peled and Yeshurun, 2001). Similar to fMRI,
DTI is a non-invasive and relatively short procedure (about 10 minutes) and can be
acquired using the same echo planar imaging sequence as fMRI, which enables
the direct overlay of images, permitting high resolution for localizing regions.
Initial DTI studies indicated a continued increase in frontal white matter anisotropy
throughout childhood (Klingberg et al., 1999). The high resolution of DTI allows
the tracking of subtle white matter differences that are not evident in white matter
volume, as well as sensitivity to subtle and dynamic changes in brain disorders.
Decreased anisotropy in left parietotemporal regions have been reported for adult
subjects with reading disabilities, providing evidence for impaired communication
between visual- and language-processing regions (Klingberg et al., 2000) and a pos-
sible mechanism for the impaired brain function demonstrated in these regions (see
above). DTI has also demonstrated reduced anisotropy in schizophrenia patients,
indicating a lack of connective integrity in white matter volumes throughout the
neocortex (Agartz et al., 2001; Foong et al., 2000; Lim et al., 1999). Understand-
ing the changes in myelination and other potential changes in white matter tracts
that concur with cognitive development will be pivotal in characterizing possible
dysmaturation in schizophrenia.
Phosphorus MR spectroscopy (31 P MRS) allows non-invasive in vivo investi-
gation of phosphorus-containing metabolites, which provide an indirect measure
of synaptic pruning (Keshavan et al., 1991; Stanley et al., 1996). MRS probes the
integrity of cell membranes by indicating the levels of phosphomonoesters (PMEs),
which are membrane phospholipid precursors, and phosphodiesters (PDEs), which
are phospholipid breakdown products (Keshavan et al., 1991). This can provide
an indirect measure of age-related changes in synaptic integrity. MRS has pro-
vided evidence for decreases in PMEs and increases in PDEs in prefrontal cortex
in schizophrenia (Pettegrew et al., 1993). The concurrent acquisition of fMRI and
MRS data could potentially reveal the contribution changes in synaptic integrity to
age-related changes in brain functional organization.
Finally, in order to increase power, validity, and reliability, pediatric studies
should include a larger number of subjects with equal distribution of individu-
als across each age group. Given the variability in rates of maturation, sources of
individual difference should be reduced. Variables such as gender, intelligence quo-
tient, and puberty would decrease intersubject variability. Control groups should be
carefully characterized and be free of any neurobiological abnormalities, including
metabolic and neurological disorders as well as psychiatric illness in the subject and
first-degree relatives. Most importantly, longitudinal studies should be performed
in a well-characterized group of healthy subjects who could be tested annually
63 Cognitive development: fMRI studies

with the same cognitive tasks. In this manner, developmental trajectories can be
delineated and also stage-like characterization of development can be understood.
Only with carefully designed studies with a large number of subjects, as suggested,
will we be able to generate normative data that is sensitive to subtle dysmaturation
processes present in psychiatric disorders.

Conclusions
We have recently been able to characterize the interface between cognitive devel-
opment and brain maturation using fMRI, a non-invasive neuroimaging method.
This is important to the understanding of neurodevelopmental models of dysmat-
uration in brain disorders such as schizophrenia. Initial pediatric fMRI studies have
shown that the basic circuitry supporting the capacity to perform a task is estab-
lished by childhood. Changes in the participation of frontal cortex and the ability
to recruit widely distributed brain circuitry support increases in the efficiency of the
cognitive control of behavior with age. This pattern was observed in studies inves-
tigating the cognitive development of working memory and voluntary response
suppression, and language processing in healthy individuals. Studies of abnormal
pediatric populations demonstrate impairments in frontal circuitry in ADHD and
parietotemporal regions in dyslexia. When using fMRI with children, special pre-
cautions must be taken to eliminate factors that have the potential to confound the
results. The future direction of fMRI includes the ability to probe brain anatom-
ical changes supporting the brain functional changes that are observed during
development.

REFERENCES

Adleman, N. E., Menon, V., Blasey, C. M., et al. (2002). A developmental fMRI study of the Stroop
color–word task. NeuroImage 16: 61–75.
Agartz, I., Andersson, J. L. R., Skare, S. (2001). Abnormal brain white matter in schizophrenia: a
diffusion tensor imaging study. Neuroreport 12: 2251–2254.
Armstrong, E., Schleicher, A., Omran, H., Curtis, M., Zilles, K. (1995). The ontogeny of human
gyrification. Cereb Cortex 5: 56–63.
Baddeley, A. D. (1983). Working memory. Philos Trans R Soc Lond B 302: 311–324.
Basser, P. J., Mattiello, J., Lebihan, D. (1994). Estimation of the effective self-diffusion tensor from
the NMR spin echo. J Magn Reson 103: 247–254.
Benson R. R., Logan W. J., Cosgrove G. R., et al. (1996). Functional MRI localization of language
in a 9-year-old child. Can J Neurol Sci 23: 213–219.
Bjorklund, D. F., Harnishfeger, K. K. (1990). The resources construct in cognitive development:
diverse sources of evidence and a theory of inefficient inhibition. Dev Rev 10: 48–71.
64 B. Luna and J. A. Sweeney

(1995). The evolution of inhibition mechanisms and their role in human cognition and behav-
ior. In Interference and Inhibition in Cognition, ed. F. N. Dempster, C. J. Brainerd. San Diego,
CA: Academic Press, pp. 141–173.
Bookheimer, S. Y. (2000). Methodological issues in pediatric neuroimaging. Mental Retard Dev
Disabil Res Rev 6: 161–165.
Booth, J. R., MacWhinney, B., Thulborn, K. R., et al. (1999). Functional organization of activation
patterns in children: whole brain fMRI imaging during three different cognitive tasks. Prog
Neuropsychopharmacol Biol Psychiatry 23: 669–682.
Bunge, S. A., Dudukovic, N. M., Thomason, M. E, Vaidya, C. J., Gabrieli, J. D. E. (2001). Immature
frontal lobe contributions to cognitive control in children: evidence from fMRI. Neuron 33:
301–311.
Case, R. (1992). The role of the frontal lobes in the regulation of cognitive development. Brain
Cogn 20: 51–73.
Casey, B. J., Cohen, J. D., Jezzard, P., et al. (1995). Activation of prefrontal cortex in children
during a nonspatial working memory task with functional MRI. NeuroImage 2: 221–229.
Casey, B. J., Trainor, R. J., Orendi, J. L., et al. (1997). A developmental functional MRI study of
prefrontal activation during performance of a go-no-go task. J Cogn Neurosci 9: 835–847.
Caviness, V. S., Kennedy, D. N., Bates, J. F., Makris, N. (1996). The developing human brain: a
morphometric profile. In Developmental Neuroimaging: Mapping the Development of Brain
and Behavior, ed. R. W. Thatcher, G. Reid Lyon, J. Rumsey, N. A. Krasnegor. New York:
Academic Press, pp. 3–14.
Changeux, J. P., Danchin, A. (1976). Selective stabilization of developing synapses as a mechanism
for the specification of neuronal networks. Nature 264: 705–712.
Chugani, H. T. (1998). A critical period of brain development: studies of cerebral glucose utiliza-
tion with PET. Prev Med 27: 184–188.
Cohen, J. D., Barch, D. M., Carter, C., Servan-Schreiber, D. (1999). Context-processing deficits
in schizophrenia: converging evidence from three theoretically motivated cognitive tasks.
J Abnorm Psychol 108: 120–133.
Conturo, T. E., McKinstry, R. C., Akbudak, E., Robinson, B. H. (1996). Encoding of anisotropic
diffusion with tetrahedral gradients: a general mathematical diffusion formalism and experi-
mental results. Magn Reson Med 35: 399–412.
Cox, R. W. (1996). AFNI: Software for analysis and visualization of functional magnetic resonance
neuroimages. Comput Biomed Res 29: 162–173.
Diamond, A., Goldman-Rakic, P. S. (1989). Comparison of human infants and rhesus monkeys
on Piaget’s AB task: evidence for dependence on dorsolateral prefrontal cortex. Exp Brain
Res 74: 24–40.
Eddy, W. F., Fitzgerald, M., Genovese, C. R., Mockus, A., Noll, D. C. (1996). Functional image
analysis software: computational olio. In Proceedings in Computational Statistics, ed. A. Prat.
Heidelberg: Physica-Verlag, pp. 39–49.
Eden, G. F., Van Meter, J. W., Rumsey, J. M., et al. (1996). Abnormal processing of visual motion
in dyslexia revealed by functional brain imaging. [See comments] Nature 382: 66–69.
Everling, S., Munoz, D. P. (2000). Neuronal correlates for preparatory set associated with pro-
saccades and anti-saccades in the primate frontal eye field. J Neurosci 20: 387–400.
65 Cognitive development: fMRI studies

Everling, S., Dorris, M. C., Klein, R. M., Munoz, D. P. (1999). Role of primate superior colliculus
in preparation and execution of anti-saccades and pro-saccades. J Neurosci 19: 2740–2754.
Fischer, B., Biscaldi M, Gezeck, S. (1997). On the development of voluntary and reflexive com-
ponents in human saccade generation. Brain Res 754: 285–297.
Foong, J., Maier, M., Clark, C. A., et al. (2000). Neuropathological abnormalities of the corpus
callosum in schizophrenia: a diffusion tensor imaging study. J Neurol Neurosurg Psychiatry
68: 242–244.
Frank, L. R., Buxton, R. B., Wong, E. C. (2001). Estimation of respiration-induced noise fluctu-
ations from undersampled multislice fMRI data. Magn Reson Med 45: 635–644.
Fry, A. F., Hale, S. (1996). Processing speed, working memory, and fluid intelligence: evidence
for a developmental cascade. Psychol Sci 7: 237–241.
Fuster, J. M. (1997). The Prefrontal Cortex. New York: Raven Press.
Gaillard, W. D., Hertz-Pannier, L., Mott, S. H., et al. (2000). Functional anatomy of cognitive
development: fMRI of verbal fluency in children and adults. Neurology 54: 180–185.
Gaillard, W. D., Pugliese, M., Grandin, C. B., et al. (2001a). Cortical localization of reading in
normal children: an fMRI language study. Neurology 57: 47–54.
Gaillard, W. D., Grandin, C. B., Xu, B. (2001b). Developmental aspects of pediatric fMRI: con-
siderations for image acquisition, analysis, and interpretation. NeuroImage 13: 239–249.
Georgiewa, P., Rzanny, R., Gaser, C., et al. (2002). Phonological processing in dyslexic children:
a study combining functional imaging and event related potentials. Neurosci Lett 318: 5–8.
Giedd, J. N., Snell, J. W., Lange, N., et al. (1996). Quantitative magnetic resonance imaging of
human brain development: ages 4–18. Cereb Cortex 6: 551–560.
Giedd, J. N., Blumenthal, J., Jeffries, N. O., et al. (1999). Brain development during childhood
and adolescence: a longitudinal MRI study. Nat Neurosci 2: 861–863.
Goldman-Rakic, P. S. (1988). Topography of cognition: parallel distributed networks in primate
association cortex. Annu Rev Neurosci 11: 137–156.
(1990). Parallel systems in the cerebral cortex: the topography of cognition. In Natural and
Artificial Parallel Computation, ed. M. A. Arbib, J. A. Robinson. New York: MIT Press,
pp. 155–176.
(1992). Working memory and the mind. Sci Am 267: 111–117.
Gottesman, I. I., Shields, J., Hanson, D. (1982). Schizophrenia: The Epigenetic Puzzle. New York:
Cambridge University Press.
Hertz-Pannier, L., Gaillard, W. D., Mott, S. H., et al. (1997). Noninvasive assessment of language
dominance in children and adolescents with functional MRI: a preliminary study. Neurology
48: 1003–1012.
Hertz-Pannier, L., Chiron, C., Vera, P., van de Morteele, P. F., et al. (2001). Functional imaging
in the work-up of childhood epilepsy. Childs Nerv Syst 17: 223–228.
Holland, S. K., Plante, E., Byars, A. W., et al. (2001). Normal fMRI brain activation patterns in
children performing a verb generation task. NeuroImage 14: 837–843.
Huttenlocher, P. R. (1990). Morphometric study of human cerebral cortex development.
Neuropsychologia 28: 517–527.
Hyde, T. M., Ziegler, J. C., Weinberger, D. R. (1992). Psychiatric disturbances in metachromatic
leukodystrophy. Arch Neurol 49: 401–406.
66 B. Luna and J. A. Sweeney

Jernigan, T. L., Trauner, D. A., Hesselink, J. R., Tallal, P. A. (1991). Maturation of human cerebrum
observed in vivo during adolescence. Brain 114: 2037–2049.
Kemna, L. J., Posse, S. (2001). Effect of respiratory CO2 changes on the temporal dynamics of the
hemodynamic response in functional MR imaging. NeuroImage 14: 642–649.
Keshavan, M. S., Kapur, S., Pettegrew, J. W. (1991). Magnetic resonance spectroscopy in psychi-
atry: potential, pitfalls, and promise. Am J Psychiatry 148: 976–985.
Keshavan, M. S., Anderson, S., Pettegrew, J. W. (1994). Is schizophrenia due to excessive synaptic
pruning in the prefrontal cortex? The Feinberg hypothesis revisited. J Psychiatr Res 28:
239–265.
Keshavan, M., Diwadkar, V., Spencer, S. M., et al. (2002). A preliminary functional magnetic
resonance imaging study in offspring of schizophrenic parents. Prog Neuropsychopharmacol
Biol Psychiatry 26: 1143–1149.
Kim, S. G., Ugurbil, K., Strick, P. L. (1994). Activation of a cerebellar output nucleus during
cognitive processing. Science 265: 949–951.
Klein, C., Heinks, T., Andresen, B., Berg, P., Moritz, S. (2000). Impaired modulation of the saccadic
contingent negative variation preceding antisaccades in schizophrenia. Biol Psychiatry 47:
978–990.
Klingberg, T., Vaidya, C. J., Gabrieli, J. D. E., Moseley, M. E., Hedehus, M. (1999). Myelination
and organization of the frontal white matter in children: a diffusion tensor MRI study.
Neuroreport 10: 2817–2821.
Klingberg, T., Hedehus, M., Temple, E., et al. (2000). Microstructure of temporo-parietal white
matter as a basis for reading ability: evidence from diffusion tensor magnetic resonance
imaging. Neuron 25: 493–500.
Klingberg, T., Forssberg, H., Westerberg, H. (2002). Increased brain activity in frontal and pari-
etal cortex underlies the development of visuospatial working memory capacity during
childhood. J Cogn Neurosci 14: 1–10.
Kwon, H., Reiss, R. L., Menon, V. (2002). Neural basis of protracted developmental changes in
visuo-spatial working memory. Proc Natl Acad Sci USA 99: 13336–13341.
Le Bihan, D., Mangin, J. F., Poupon, C., et al. (2001). Diffusion tensor imaging: concepts and
applications. J Magn Reson 13: 534–546.
Levin, H. S., Culhane, K. A., Hartmann, J., Evankovich, K., Mattson, A. J. (1991). Developmental
changes in performance on tests of purported frontal lobe functioning. Dev Neuropsychol
7: 377–395.
Lim, K. O., Hedehus, M., Moseley, M., et al. (1999). Compromised white matter tract integrity
in schizophrenia inferred from diffusion tensor imaging. Arch Gen Psychiatry 56: 367–374.
Logan, W. J. (1999). Functional magnetic resonance imaging in children. Semin Pediatr Neurol
6: 78–86.
Luciana, M., Nelson, C. A. (1998). The functional emergence of prefrontally-guided working
memory systems in four- to eight-year-old children. Neuropsychologia 36: 273–293.
Luna, B., Sweeney, J. A. (1999). Cognitive functional magnetic resonance imaging at very-high-
field: eye movement control. Top Magn Reson Imaging 10: 3–15.
(2001). Studies of brain and cognitive maturation through childhood and adolescence: a
strategy for testing neurodevelopmental hypotheses. Schizophr Bull 27: 443–455.
67 Cognitive development: fMRI studies

Luna, B., Thulborn, K. R., Munoz, D. P., et al. (2001). Maturation of widely distributed brain
function subserves cognitive development. NeuroImage 13: 786–793.
Luna, B., Minshew, N. J., Garver, K. E., et al. (2002). Neocortical system abnormalities in autism:
an fMRI study of spatial working memory. Neurology 59: 834–840.
McDowell, J. E., Clementz, B. A. (2001). Behavioral and brain imaging studies of saccadic per-
formance in schizophrenia. Biol Psychol 57: 5–22.
Merriam, E. P., Thase, M. E., Haas, G. L., Keshavan, M. S., Sweeney, J. A. (1999). Prefrontal cortical
dysfunction in depression determined by Wisconsin Card Sorting Test performance. Am J
Psychiatry 156: 780–782.
Munoz, D. P., Broughton, J. R., Goldring, J. E., Armstrong, I. T. (1998). Age-related performance
of human subjects on saccadic eye movement tasks. Exp Brain Res 217: 1–10.
Nelson, C. A., Monk, C. S., Lin, J., et al. (2000). Functional neuroanatomy of spatial working
memory in children. Dev Psychol 36: 109–116.
Park, S., Holzman, P. S. (1992). Schizophrenics show spatial working memory deficits. Arch Gen
Psychiatry 49: 975–982.
Peled, S., Yeshurun, Y. (2001). Superresolution in MRI: application to human white matter fiber
tract visualization by diffusion tensor imaging. Magn Reson Med 42: 29–35.
Pettegrew, J. W., Keshavan, M. S., Minshew, N. J. (1993). 31 P nuclear magnetic resonance spec-
troscopy: neurodevelopment and schizophrenia. Schizophr Bull 19: 35–53.
Pfefferbaum, A., Mathalon, D. H., Sullivan, E. V., et al. (1994). A quantitative magnetic resonance
imaging study of changes in brain morphology from infancy to late adulthood. Arch Neurol
51: 874–887.
Poldrack, R. A., Pare-Blagoev, E. J., Grant, P. E. (2002). Pediatric functional magnetic resonance
imaging: progress and challenges. Top Magn Reson Imaging 13: 61–70.
Raj, D., Anderson, A. W., Gore, J. C. (2001). Respiratory effects in human functional magnetic
resonance imaging due to bulk susceptibility changes. Phys Med Biol 46: 3331–3340.
Rakic, P., Bourgeois, J. P., Eckenhoff, M. F., Zecevic N, Goldman-Rakic, P. S. (1986). Concurrent
overproduction of synapses in diverse regions of the primate cerebral cortex. Science 232:
232–235.
Reiss, A. L., Abrams, M. T., Singer, H. S., Ross, J. L., Denckla, M. B. (1996). Brain development,
gender, and IQ in children: a volumetric imaging study. Brain 119: 1763–1774.
Ridderinkhof, K. R., van der Molen, M. W. (1997). Mental resources, processing speed, and
inhibitory control: a developmental perspective. Biol Psychol 45: 241–261.
Rosenberg, D. R., Sweeney, J. A., Gillen, J., et al. (1997). Magnetic resonance imaging of children
without sedation: preparation with simulation. J Am Acad Child Adolesc Psychiatry 36: 853–
859.
Rubia, K., Overmeyer S, Taylor, E., et al. (2000). Functional frontalisation with age: mapping
neurodevelopmental trajectories with fMRI. Neurosci Biobehav Rev 24: 13–19.
Schlaggar, B. L., Brown, T. T., Lugar, H. M., et al. (2002). Functional neuroanatomical differences
between adults and school-age children in the processing of single words. Science 296: 1476–
1479.
Sereno, A. B., Holzman, P. S. (1995). Antisaccades and smooth pursuit eye movements in
schizophrenia. Biol Psychiatry 37: 394–401.
68 B. Luna and J. A. Sweeney

Shaywitz, B. A., Shaywitz, S. E., Pugh, K. R., et al. (2002). Disruption of posterior brain systems
for reading in children with developmental dyslexia. Biol Psychiatry 52: 101–110.
Slifer, K. J. (1996). A video system to help children cooperate with motion control for radiation
treatment without sedation. J Pediatr Oncol Nurs 13: 91–97.
Stanley, J. A., Williamson, P. C., Drost, D. J., et al. (1996). An in vivo proton magnetic resonance
spectroscopy study of schizophrenia patients. Schizophr Bull 22: 597–609.
Stapleton, S. R., Kiriakopoulos, E., Mikulis, D., et al. (1997). Combined utility of functional MRI,
cortical mapping, and frameless stereotaxy in the resection of lesions in eloquent areas of
brain in children. Pediatr Neurosurg 26: 68–82.
Szameitat, A. J., Schubert, T., Mueller, K., von Cramon, D. Y. (2002). Localization of executive
functions in dual-task performance with fMRI. J Cogn Neurosci 14: 1184–1199.
Talairach, J., Tournoux, P. (1988). Co-Planar Stereotaxic Atlas of the Human Brain. New York:
Thieme Medical.
Temple, E., Poldrack, R. A., Salidis, J., et al. (2001). Disrupted neural responses to phonological
and orthographic processing in dyslexic children: an fMRI study. Neuroreport 12: 299–307.
Thatcher, R. W. (1991). Maturation of the human frontal lobes: physiological evidence for staging.
Dev Neuropsychol 7: 397–419.
Thomas, K. M., King, S. W., Franzen, P. L., et al. (1999). A developmental functional MRI study
of spatial working memory. NeuroImage 10: 327–338.
Thulborn, K. R., Shen, G. X. (1999). An integrated head immobilization system and high-
performance RF coil for fMRI of visual paradigms at 1.5 T. J Magn Reson 139: 26–34.
Vaidya, C. J., Austin, G., Kirkorian, G., et al. (1998). Selective effects of methylphenidate in
attention deficit hyperactivity disorder: a functional magnetic resonance study. Proc Natl
Acad Sci USA 95: 14494–14499.
Waddington, J. L., Torrey, E. F., Crow, T. J., Hirsch, S. R. (1991). Schizophrenia, neurodevelopment
and disease. Arch Gen Psychiatry 48: 271–273.
Weinberger, D. R., Lipska, B. K. (1995). Cortical maldevelopment, anti-psychotic drugs, and
schizophrenia: a search for common ground. Schizophr Res 16: 87–110.
Weinberger, D. R., Berman, K. F., Zec, R. F. (1986). Physiologic dysfunction of dorsolateral pre-
frontal cortex in schizophrenia: I. Regional cerebral blood flow evidence. Arch Gen Psychiatry
43: 114–124.
Welsh, M. C., Pennington, B. F., Groisser, D. B. (1991). A normative–developmental study of
executive function: a window on prefrontal function in children. Dev Neuropsychol 7: 131–
149.
Wilson, S. P., Kipp, K. (1998). The development of efficient inhibition: evidence from directed-
forgetting tasks. Dev Rev 18: 86–123.
Woods, R. P., Mazziotta, J. C., Cherry, S. R. (1993). Automated image registration. In Quantifi-
cation of Brain Function: Tracer Kinetics and Image Analysis in Brain PET, ed. K. Uemura,
N. A. Lassen, T. Jones, et al. Amsterdam: Elsevier Science, pp. 391–400.
Yakovlev P. I., Lecours A. R. (1967). Regional Development of the Brain in Early Life. Oxford:
Blackwell Scientific, pp. 3–70.
 4

Cognitive development in adolescence:

cerebral underpinnings, neural trajectories,
and the impact of aberrations
Stephen J. Wood, Cinzia R. De Luca, Vicki Anderson,
and Christos Pantelis
University of Melbourne, Melbourne, Australia

Children are not born with the full complement of cognitive skills they need for
adulthood; instead, they need to develop these abilities as they grow. Initial theor-
ies about this development espoused a behaviorist idea of the child as a passive
participant shaped by external forces, with more recent views considering the child
as developing through invariant and universal cognitive stages in an active quest to
make sense of their experience (Piaget, 1965). In this scheme, adolescence heralded
the final developmental progression in cognitive development, with the transition
from concrete to formal operational thought. This capacity to deal with abstract
concepts, “operating on operations” rather than reality, what is now referred to as
meta-cognition, or encompassed in the notion of executive functions, was consid-
ered to become available to the child around the age of 11 or 12 years (Stuss, 1992;
Travis, 1998).
Much of the detail of the influential cognitive model proposed by Piaget and
his proponents has now been amended and extended. Nevertheless, the notion
of a sequential and gradual accumulation of cognitive processes, along with the
limitations and capabilities of the child as they mature, has been invaluable to
the study of child development. An understanding of the normal development of
cognitive function is important in order to appreciate the pattern of impairments in
children and adolescents who suffer from illnesses of putative neurodevelopmental
origin. In particular, we need to know when cognitive functions are due to come “on-
line” during maturation, and how this differs across cognitive domains. Information
on the normal progression of skill acquisition allows for better characterization of
whether the onset of behavioral disturbances or symptoms of psychiatric or other
disorders reflect a deviation in functioning or a failure to make age-appropriate

Neurodevelopment and Schizophrenia, ed. Matcheri S. Keshavan et al. Published by Cambridge University
Press.

C Cambridge University Press 2004.

69
70 S. J. Wood, C. R. De Luca, V. Anderson, and C. Pantelis

gains. In this regard, the observation that the acute symptoms of schizophrenia
arise during this final phase of cognitive transition raises interesting questions
about the etiology of the disorder and how the underlying substrates of psychosis
are expressed.

Adolescence
Adolescence is a loose descriptive term referring to the period of transition from
childhood to adulthood. Although it comprises the teenage years and is heralded
by the onset of puberty, it is not defined by specific events. Indeed, while puberty
and adolescence are often used synonymously (Petersen, 1998), the two do not
necessarily co-occur (Dubas, 1991). For the purposes of this review, adolescence
will be taken to cover the years from ages 12 to 20, as distinct cognitive advances
in higher-order thinking are seen to occur around the age of 12, with a general
stabilization of most executive processes as individuals enter their twenties.
Adolescence is a time of major upheaval in behavioral and social domains as
the developing child seeks to attain the skills necessary for independence. The
major changes tend to be in the areas of social interaction, risk-taking behavior,
and intellectual expansion. The first two are more comprehensively reviewed by
Spear (2000), while the last is covered in detail. The greater emphasis placed on
exploring cognitive developments unique to adolescence derives from the belief that,
ultimately, the emergence of independent thinking, social and moral awareness,
impulsivity, and challenging of conceptual ideas are a reflection of the higher-order
executive abilities that become available to the child during the teenage years.

Brain development during adolescence

The adolescent period, which is associated with the greatest risk for psychotic dis-
orders, is accompanied by extensive, but localized, brain maturation, which takes
the form of myelination and cortical synapse elimination (Huttenlocher, 1984).
It has been suggested that as many as 30 000 synapses per second are lost dur-
ing primate adolescence (Rakic et al., 1994). The vast majority of these synapses
are excitatory in nature (Rakic et al., 1994), so it is unsurprising that removal of
these synapses is associated with a decline in brain rates of glucose metabolism
from the hypermetabolic state of childhood (Chugani et al., 1987). The elim-
ination of excess synaptic connections through apoptotic processes results in a
refinement of the neural circuitry, which is thought to strengthen the remaining
functional connections and reduce competition from suboptimal associations. As
a result, communication between distributed information systems is thought to be
71 Cognitive development in adolescence

vastly improved. However, the hard-wiring of neural networks, while improving

communication between distributed systems, has the added consequence of reduc-
ing the redundancy available in the brain.
Postmortem studies suggest that myelination begins during the second trimester
of gestation and continues well into the third decade of life (Benes et al., 1994;
Yakovlev and Lecours, 1967). Importantly, these studies show that myelination
does not occur concurrently in all brain regions: there is a graded progression of
maturation from inferior to superior and posterior to anterior, with the cerebellum
developing first and the frontal lobes last (Yakovlev and Lecours, 1967). Recent
advances in brain imaging have allowed the in vivo study of these maturational
changes in brain structure during adolescence, with the most consistent findings
being a significant decline in global gray matter volume and a contrasting increase
in global white matter volume (Giedd et al., 1999; Reiss et al., 1996; see also Ch. 2).
Klingberg et al. (1999) used diffusion tensor imaging to examine axonal integrity
in the frontal lobes of children and adults and found that white matter continues
to increase into the second decade of life in this region. These changes appear to be
regionally specific in that the white matter increase is located in the dorsal prefrontal
cortex and not in the orbitofrontal region (Reiss et al., 1996). However, automated
voxel-based methods have suggested that white matter change during adolescence
is seen most prominently in frontotemporal (specifically dorsal prefrontal–lateral
temporal) and corticospinal pathways (Paus et al., 1999). Of relevance, it is disrup-
tion in these frontotemporal pathways that is proposed to play a large role in the
pathology of schizophrenia.
These white matter changes, particularly those identified by Paus and colleagues
(1999), are consistent with reported findings of striatal gray matter volume loss over
the same period. These contrasting maturational changes are thought to enhance
neural transmission between the motor connections of the striatal pathway with
the frontal cortex, allowing for greater top-down mediation of behavioral out-
put (Sowell et al., 1999). Refinement of gray matter structure also follows het-
erochronous development, occurring latest in the prefrontal and parietal lobes
(Giedd et al., 1999). This development is associated with “thinning” of the cortex,
with reductions in gray matter most prominent in the dorsal, medial, and lateral
regions of the frontal lobes during adolescence (Sowell et al., 2001a, 2003). Despite
the loss of tissue, net brain volume continues to increase over adolescence through
a combination of regressive and progressive changes (Sowell et al., 2001b). Synaptic
pruning of inefficient or unused connections is thought to result in the shrinkage
of gray matter, while progressive myelination of these same areas leads to enhanced
communication between functionally distributed systems.
The magnitude of the maturational changes taking place during adolescence
holds implications for the emergence of abnormal cognitive processing and
72 S. J. Wood, C. R. De Luca, V. Anderson, and C. Pantelis

behaviors during this time. Pruning of connections and maximal activation of

frontal areas, along with their various functionally connected systems, means that
aberrantly formed associations will likely come to prominence during the adol-
escent years. The suggestion that schizophrenia involves disruption within the
frontotemporal–striatal system, with psychiatric symptoms arising late in adoles-
cence, conforms to the notion that such behavioral abnormalities correspond to the
anatomically related events that have been described above (Pantelis et al., 2002). It
is currently proposed that dysregulation of this developmental trajectory may actu-
ally be implicated in the emergence of symptoms, as individuals with schizophrenia
fail to make age-appropriate cognitive gains (Wood and Pantelis, 2001).

Cognitive changes
Cognition generally refers to “knowing” or “thinking” and includes a broad range
of abilities over and above what is thought of as intelligence. These include memory,
attention, executive functions, plus less-obvious skills such as appreciating humour
and perceiving another’s motivation: skills that are often deficient in individuals
with neurodevelopmental disorders such as schizophrenia. Basic cognitive pro-
cesses are in place by early childhood, and it has been shown that certain abilities
are inborn (Meltzoff and Moore, 1977). However, more complex behaviors develop
at a slower rate and some functions are not mastered until late in adulthood. This
holds implications for the observation of deficits in areas of higher-order func-
tions as individuals enter the adolescent period. Studies of cognitive development
tend to have proceeded with little attention to studies of brain development; how-
ever, given that the predominant brain regions developing during adolescence are
the frontal lobes, it is sensible to examine the cognitive development of functions
thought to be mediated by these regions. The frontal lobes have been given par-
ticular attention in the mediation of executive functions, that is, those interrelated
skills required for successful independent, volitional behavior in response to novelty
(Anderson, 1998). The functions subsumed under the executive umbrella include
higher-order processes such as strategic planning, problem solving, inhibitory con-
trol, cognitive flexibility, abstract thinking, concept formation, working memory,
and self-monitoring.
Cerebral organization and cognitive functioning models initially presented
executive function as an adult capacity that reaches maturity during adolescence,
approximating the stage of Piaget’s predicted transition from concrete to formal
operational thinking (Golden, 1981; Stuss, 1992; Travis, 1998). Research has now
identified a stage-like sequence of executive function development characterized by
“spurts” in executive abilities beginning from as young as 12 months of age, with
the majority of functions coming “online” around the age of 8 (Ardila and Rosselli,
1994; Case, 1992; Luciana and Nelson, 1998). These higher-order processes are
73 Cognitive development in adolescence

not homogeneous, however, and are thought to observe disparate developmental

trajectories. Simple planning, attentional set-shifting and hypothesis testing are
reportedly available to the child at an earlier stage than other abilities, including
temporal ordering and complex strategy formation (Chelune and Baer, 1986; Espy,
1997; Luciana and Nelson, 1998; Stuss, 1992). The development of working memory
capacity also follows a relatively slow but steady increase over the adolescent years
and into adulthood (Luciana and Nelson, 1998). The verbal modality is thought to
be slightly advanced in the amount of information it can hold online in an active
buffer (Nichelli et al., 2001). Cognitive theories of working memory systems pro-
pose that the ability to convert visuospatial information into phonological form
actually develops during adolescence, and that this is the preferred way to encode
information for temporary storage and manipulation (Hitch, 2002). As such, work-
ing memory abilities in both the verbal and the visuospatial modality are thought
to develop in unison into the adult years.
A number of influential developmentally oriented studies have provided evidence
to support the continuing development of executive functions into adolescence.
Welsh and Pennington (1988) assessed 140 children and adults from ages 3 to 28
years on several executive tasks. The data supported sequential development of these
skills over childhood and into adolescence. Performance on these tests was found to
be quite independent from intelligence quotient (IQ) scores, and adult levels were
reached sooner on simple planning and visual search tasks than in those measuring
complex planning and verbal fluency. Levin et al. (1991) also administered a range
of “executive” ability measures to a group of 52 normal children and adolescents.
The sample was divided into three age bands: 7–8, 9–12, and 13–15 years, with gains
across all tasks identified with increasing age. The improvements in performance
were described as reflecting progress in concept formation, mental flexibility, and
goal-setting skills through childhood. Scores on different measures also provided
evidence for discrepant maturation of distinct abilities, with some skills reaching
adult levels by 12 years of age, while others continued to show developmental
trends into mid-adolescence. A large-scale developmental study by Anderson et al.
(1995), involving 430 children aged 7–17 years, provided complementary findings
to those presented above. The data indicated significant improvements in executive
skills through middle to late childhood (7–11 years), with stabilization evident by
mid-adolescence.
We have recently conducted a study (De Luca et al., 2003) involving 193 individ-
uals ranging in age from 8 to 64 years, investigating the development of executive
functions across the lifespan, using the Cambridge Neuropsychological Test Auto-
mated Battery (CANTAB). In particular, we examined planning and organization,
working memory and strategy formation, and attentional set-shifting. In our study,
we demonstrated that performance on these tasks continues to develop throughout
74 S. J. Wood, C. R. De Luca, V. Anderson, and C. Pantelis

adolescence and is at its height between the ages of 20 and 29 years. Success on these
tasks is thought to result at least in part from the achievement and maintenance of
maximal short-term memory capacity in the 15–19 age range, since maturation of
this “capacity resource” contributes to a greater ability to define and order a strategic
plan (Hitch et al., 2001). However, while increases in abilities such as memory span,
inhibitory control, and speed of processing may allow for more efficient executive
control, advances in the last have been dissociated from the maturation of these
types of general, capacity-limited factor (Ridderinkhof and van der Molen, 1997).
In our study, attentional set-shifting ability was also tested using the intradimen-
sional/extradimensional task from the CANTAB, which assesses the ability of an
individual to establish the correct response set to positive and negative feedback, to
reverse this rule, and later to shift this learning to new exemplars within a previously
learned set. This measure of cognitive flexibility was found to follow a swifter mat-
urational path, and performance on this task did not differentiate any of the groups
(De Luca et al., 2003), confirming documented adult levels of set-shifting compet-
ence in 8 and 10 year old children (Anderson et al., 2001a; Chelune and Baer, 1986;
Luciana and Nelson, 1998). This would suggest that the ability to adjust responses
to feedback, and to shift them to new exemplars within a previously learned set,
is an executive skill acquired early in development, indicating that the purported
neural circuitry subserving this ability in the lateral and orbital prefrontal cortex
is connected and activated earlier than other executive systems (Rezai et al., 1993;
Roberts et al., 1998; Stuss and Benson, 1987). However, our use of an endpoint
score for this task may have obscured subtle differences in performance. Previous
research would suggest that children do in fact make a greater number of errors in
learning a response set, but they do not differ in the number of trials necessary to
advance at either level (Luciana and Nelson, 1998; Shanab and Yasin, 1979). Con-
sequently, they can competently complete such tasks by responding to the absolute
properties of the stimuli without relevant understanding of the underlying contin-
gency being tested. Adolescents and adults, by comparison, are found to experience
greater difficulty performing a shift in set because of their more profound under-
standing of the dimension characteristics of the task. Having learned the relevant
dimension governing the response rule, their performance is potentially impeded
as they are required to extinguish the original response set and transfer this learning
to new exemplars (Owen et al., 1993; Whitney and White, 1993). Indeed, patients
with schizophrenia are seen to perseverate on previously learned sets and perform
poorly on similar tests such as the Wisconsin Card Sorting Task. Their inability
to dissociate from one paradigm and shift to a new exemplar highlights deficits in
attentional control and, therefore, implicates dysfunction in the lateral and orbital
frontal systems that mediate this function. It is interesting to note that this skill
appears to be relatively unimpaired in first-episode schizophrenia, suggesting that
75 Cognitive development in adolescence

these brain areas, which develop earlier in life, are in fact well connected and
activated appropriately during development but may be affected later by the con-
tinued insult of acute psychotic episodes.
While the set-shifting component of the attentional system appears to develop
relatively early in childhood, the literature suggests that other attentional processes,
such as the ability to attend to meaningful stimuli and actively inhibit distractors,
continue to improve during adolescence (Anderson et al., 2001a). Greater control
in the triage of incoming information with internal needs and desires places less
demand on the capacity-limited attentional system, which, in turn, results in the
more efficient processing of relevant information. While attentional components
are found to dissociate both anatomically and in terms of their maturational course,
gradual increments in performance from 6 to 15 years of age have been found for
a range of these skills across modalities (Anderson et al., 2001a).
As is the case with the development of executive abilities, the prolonged trajectory
of attentional skills implies that deficits may not be obvious until later in devel-
opment. Within a developmental framework, the cumulative effects of attention
deficits, such as distractibility or the inability to shift-set, may have serious implica-
tions for the learning of new information and coordination of internal desires with
external demands at a time when these skills appear to be critical to the success of
the individual. In addition, impaired executive function will affect the adolescent’s
ability to plan homework or find classrooms, resulting in poorer scholastic perfor-
mance over and above its effect on actual learning. These practical consequences
highlight the dynamic interplay that exists between cognitive and psychological
factors, and how cognitive inefficiencies that arise during adolescence, already a
time of considerable emotional upheaval, feed feelings of personal inadequacy and
disrupt the usual process of self-discovery. It is not surprising then that the current
drive in the treatment of schizophrenia is to treat aggressively first-episode patients,
both to prevent the purported neurotoxic effects of repeated acute episodes and to
halt the snowballing effects of chronic illness and perceived personal failure.

Links between the emergence of cognitive skills and cerebral development

Much of the research on cognitive development and cerebral maturation fails to
consider the dynamic interplay that occurs between the structural and functional
advances of the developing brain. This is integral to an understanding of the nor-
mal capabilities and constraints of the individual across the lifespan, along with the
consequences of aberrant development at different times along the maturational
trajectory. Interesting parallels can be found in anatomical and intellectual devel-
opment from childhood through to adolescence, and changes in brain structure
centering on the frontal lobes have been implicated as the physiological substrate
for development of executive functions during this time (Anderson, 1998; Casey
76 S. J. Wood, C. R. De Luca, V. Anderson, and C. Pantelis

et al., 2000). Our findings of improved performance on tasks of working memory,

strategic ability, planning, and goal setting correspond to the development of func-
tional connections involving distributed systems of the frontal cortex. As stated
above, the frontal lobes are one of the latest maturing areas of the brain, with both
regressive events and myelination not fully active until the second decade of life,
and continuing beyond early adulthood (Giedd et al., 1999; Klingberg et al., 1999;
Luna and Sweeney, 2001). As such, executive skills are thought to become fully
available to the individual once the anterior portions of the brain have matured,
allowing for whole brain systems to be maximally activated.
Working memory, a higher-order capacity that peaks around 20–29 years of age,
is one of the functions that are thought to be limited in the child as a result of the
immaturity of a frontoparietal–limbic circuit. Functional imaging studies have been
informative in identifying a distributed neural system involved in the ability to hold
and manipulate information online in an active buffer. Several areas, including the
dorsolateral and ventrolateral prefrontal cortex, premotor cortex, and the posterior
parietal cortex, show increased activation on working memory tasks (Kwon et al.,
2002). The cingulate cortex and occipital areas are also considered to form part
of the visuospatial working memory circuitry (Klingberg et al., 2002; Rubia et al.,
2000). This distributed system is activated in both children and adults; however,
research looking at the development of visuospatial working memory capacity
have reported positive correlations between age and working memory-related brain
activity in several frontal and parietal areas into adulthood (Klingberg et al., 2002;
Kwon et al., 2002). In fact, Kwon et al. (2002) showed activation in areas correlating
to a left-hemisphere phonological rehearsal system and a right-hemisphere-based
visuospatial attentional system, suggesting concurrent development of verbal and
visual-based working memory modalities. These findings correspond to Baddeley’s
behavioral findings, which led to the development of his influential cognitive model
of working memory (Baddeley, 1998). Progression in functional capacities and
activation of neural systems are thought to correspond to the physiological changes
of synaptic pruning and myelination of frontal and parietal areas, as discussed
above, which continue into the second decade of life and beyond (Giedd et al.,
1999; Paus et al., 1999; Sowell et al., 1999).
Similar correlations in the development of attentional processes and the cir-
cuitry purported to mediate these functions are found in the literature. Attentional
processes are thought to be controlled by an anatomically distributed system pri-
marily involving the prefrontal cortex, cingulate gyrus, and posterior parietal cortex
(Heilman et al., 1997; Koski and Petrides, 2002). Areas such as the thalamus and
reticular activating system also form part of this system, although they are more
involved with the arousal/motivation aspects of attention. Improvements in the
volitionally driven skills of divided attention and inhibitory control during the
77 Cognitive development in adolescence

adolescent period are thought to reflect maturation of the connections in this func-
tional system. Myelination of the cingulate gyrus and prefrontal cortex does not
fully occur until at least the second decade of life (Giedd et al., 1999), and these
are key areas involved in the integration of emotional/biological information with
intentional, cognitively driven plans (Benes, 1989; Paus et al., 2001). Conversely,
reductions in the gray matter of the parietal and frontal cortices during late ado-
lescence are considered a marker of the refinement of these anterior–posterior
connections (Giedd et al., 1999; Sowell et al., 1999).
Functional magnetic resonance imaging (fMRI) studies looking at development
of selective attention and response inhibition skills have reported variable find-
ings, with both hypo- and hyperactivation found in these brain regions in children
and adolescents compared with adults (see Ch. 3). A steady increase in activation
of frontal, parietal, cingulate, striatal, and thalamic areas has been shown from
childhood to adulthood for tasks assessing response inhibition (Luna and Sweeney,
2001). These studies suggested that the power of activation corresponds to the
extent of functional maturation and, therefore, voluntary control over one’s actions
(Kwon et al., 2002; Rubia et al., 2000). In contrast, other fMRI studies have indi-
cated that essentially the same circuits are activated in children and adults, although
the activation is greater in children, both for the volume of cortex engaged to per-
form the task and the percentage signal change (Casey et al., 2000). These authors
argue that, taken together, the decrease in structural and functional areas mediating
attentional processes suggests that improvements in these skills during adolescence
correspond to anatomical maturation of frontal and parietal areas. These physio-
logical processes are thought to strengthen and enhance stable connections, leading
to a decrease in the brain tissue required to perform a task and removal of compet-
ing inputs that interfere with direct communication between these interconnected
regions (Klingberg et al., 2002). It is specifically these later maturing systems that
are found to be dysfunctional in both first-episode and chronic schizophrenia; they
will be discussed in greater detail in the section below dealing with schizophrenia.

Aberrations in development that impact on adolescence

The developing brain provides a model for the study of the impact of aberrations on
normal development. A unique aspect of the immature brain is the state of flux of the
young system, and its lack of fixed functional connections. Lesions of developmental
origin, therefore, have a dynamic impact on further maturation, rather than the
expected static losses that are seen with injury during adulthood. Proponents of
the vulnerability theories of childhood insult hold that children present a greater
risk for cognitive impairment as they can fail to acquire the skills necessary for new
learning and the organization of information. In other words, early disruption to
78 S. J. Wood, C. R. De Luca, V. Anderson, and C. Pantelis

the brain impacts most significantly on the development of higher-order functions

that rely on the intact primary processing of information and the integration of
this knowledge through robust associations between distributed functional systems.
This corresponds closely to the current models of schizophrenia, which hold that
a developmental lesion that disrupts the initial formation of system connections is
necessary to explain the myriad of deficits found in psychotic individuals.
Childhood injury is also distinct from similar lesions in adulthood in the observa-
tion of emerging deficits with age, or the notion of the child “growing into deficits.”
This extends from the fact that not all intellectual functions are fully mature in the
child, such as executive functions, which follow a prolonged maturational traject-
ory into young adulthood (Anderson et al., 2001a; De Luca et al., 2003; Luciana
and Nelson, 1998). Therefore, aberrations involving the anterior regions, which
mediate these executive and other skills, will not be immediately visible following
injury, as they are not yet available to the child. When these functions are due to
come “online” later during adolescence, dysfunction in various systems arise, and
this is observed as a widening gap in the performance of brain-damaged children
compared with their peers (Anderson et al., 2001b).

Developmental lesions
Research on the impact of developmental lesions supports the protracted and dis-
tinct consequences of disruption to the developing neural system. Of the limited
studies examining the outcome of frontal lesions in the young child, pervasive
behavioral and social deficits have been reported that are of a similar nature to
those seen in patients with schizophrenia. In line with the developmental perspec-
tive of frontal maturation, these deficits do not arise immediately following damage
in all cases but become progressively more evident as the child ages (Eslinger et al.,
1992; Marlowe, 1992). Many of the features are comparable to those described in
adult patients, such as a lack of initiative, disorganization, irritability, emotional
lability, and disinhibited behavior (Anderson et al., 1999; Eslinger and Damasio,
1985; Eslinger et al., 1992; Leduc et al., 1999). However, the severity of impairment
appears to be greater in individuals who acquired lesions in childhood, and a dis-
tinct antisocial presentation is reported in studies with adult follow-up (Price et al.,
1990). This is demonstrated as a failure to learn from experience, unresponsiveness
to punishment, increasing disruptive and demanding behavior, a lack of expressed
guilt and remorse, with some displaying verbal and physically abusive tendencies
(Eslinger and Damasio, 1985; Eslinger et al., 1992; Marlowe, 1992). The distinctive
feature that sets these individuals apart from sociopaths is their rather impulsive
and child-like performance of morally sanctioned acts, and the transparency with
which they attempt to cover up their indiscretions (Anderson et al., 1999). This
is interpreted as a reflection of the lack of insight and immature development of
79 Cognitive development in adolescence

executive skills in these individuals. It appears that damage to the anterior lobes
early in life disrupts the application of socially appropriate skills in real-life situa-
tions and compromises the establishment of adaptive behavior, the acquisition of
socially relevant knowledge, and the development of moral sensibilities (Anderson
et al., 1999; Eslinger et al., 1992).
Studies addressing the cognitive development of children with early lesions sup-
port the notion that these individuals experience more global deficits because they
fail to develop in a normal manner. In one of the first studies to look specifically
at cognitive outcomes of children with early brain insult, Mateer (1990) found
evidence of cognitive dysfunction, alongside intact or mildly depressed intellec-
tual ability, consistent with adult patterns of impairment. A number of more recent
studies have revealed a similar pattern of poor problem-solving skills, reduced plan-
ning and strategy use, and low self-monitoring in the adult survivors of early frontal
lobe damage (Anderson, 1988; Anderson and Jacobs, 2004; Anderson et al., 1999,
2000; Eslinger et al., 1992, 1997, 1999; Marlowe, 1992). Eslinger and colleagues
(1992) reported a pattern of delayed onset of impairments, with increasing diffi-
culties identified as executive skills failed to “come online” and mature at critical
stages throughout development. The progressive emergence of deficits as the child
reaches adolescence is believed to result from the combined effects of the failure to
make age-appropriate gains and the increasingly complex environmental demands
placed on the individual approaching adulthood. Interestingly, similar impairments
in both behavioral and cognitive domains have been described in many purported
neurodevelopmental and acquired pediatric conditions, including autism (Happe
and Frith, 1996), schizophrenia (Pantelis et al., 2002), attention-deficit hyperac-
tivity disorder (Barkley, 1996), head injury (Pentland et al., 1998), and epilepsy
(Anderson et al., 2001b).

Schizophrenia
Schizophrenia is a mental illness of putative neurodevelopmental origin (Murray
and Lewis, 1987; Weinberger, 1987) that represents an example of aberrant develop-
ment in the regions underpinning executive function. Because of this, we can use it as
a model to look at changes in cognitive performance over childhood/adolescence to
help to understand the neural development of brain regions mediating these skills.
One example of this is the retrospective study of Walker and colleagues (1994),
who used family home movies to study the neurodevelopment of children who
later developed schizophrenia. Although these movies demonstrated subtle abnor-
malities in motor function in early life, by the time of illness onset they were much
less prominent; instead these young people displayed disturbances of behavior and
cognitive function. This led the authors to propose a modular developmental tra-
jectory for schizophrenia, in which neuromotor dysfunction is most pronounced
80 S. J. Wood, C. R. De Luca, V. Anderson, and C. Pantelis

in early childhood and late in life, whereas psychotic symptomatology is most

pronounced in late adolescence and early adulthood. This hypothesis was based
on the literature on normal brain development suggesting that different circuits
within the brain are differentially activated at various points in the life course.
During infant life, the fiber tracts between motor cortex and subcortical structures
are highly myelinated and the regions they connect are more metabolically active
than either frontal or limbic cortices. In contrast, during late adolescence and early
adulthood, the motor cortex becomes less metabolically active relative to other
cortical regions, in particular limbic and frontal regions. During this period, the
frontal and limbic regions are the most active, and myelination of neural pathways
linking limbic regions is completed. It is hypothesized that during this period of
late adolescence/early adulthood the abnormalities of limbic interconnectivity are
primarily expressed behaviorally in psychotic symptoms. This would also be con-
sistent with the abnormalities in neuropsychological function, particularly those
affecting executive functions.
However, what is also apparent and not fully explained by this model is that
the early abnormalities are subtler than the later manifestations and are also less-
prominent features of the disorder. We propose that brain plasticity and the normal
maturational trajectory of various functions define the nature and extent of their
impairment over the course of the disorder. In particular, we propose that functions
that normally come online early in life (i.e. during infancy) such as sensory, motor,
and basic memory functions, when the brain is more adaptable, show fewer deficits
at illness onset than functions normally coming online in adolescence, such as
attention, working memory, and executive functions (Pantelis et al., 2001, 2003;
Wood and Pantelis, 2001).
Support for this model comes from our own work on the normal development
of executive function, presented above. We have found that attentional set-shifting
ability is fairly well established early in development but working memory skills are
not fully developed until at least the age of 20 (Fig. 4.1) (De Luca et al., 2003; Manly
et al., 2001; McKay et al., 1994; Rebok et al., 1997). Our model would suggest
that those cognitive functions which emerge earlier (attentional set-shifting in this
example) should be less compromised in schizophrenia than those that emerge later
(such as working memory). Indeed, this is consistent with the available evidence.
Using the set-shifting task from the CANTAB, Hutton et al. (1998) found that first-
episode patients were relatively unimpaired in set-shifting ability, while Elliott and
Sahakian (1995) found that patients with moderately severe schizophrenia showed
a specific deficit in set-shifting ability because of a tendency to perseverate. Further,
Pantelis and colleagues (1999) showed that a group of patients with very long-
standing schizophrenia performed even more poorly; these chronic schizophrenia
patients manifested perseverative responding and also had difficulty in generalizing
81 Cognitive development in adolescence

50

45

Total number of working memory errors 40

35

30

25

20

15

10

0
8 to 10 11 to 14 15 to 19 20 to 29 30 to 49 50 to 64

Age group
Fig. 4.1. Total number of working memory errors as a function of age group for males (
) and
females ( ). •

a rule that had been learned (representing failure at a simpler level of set-shifting
ability). We have now examined patients very early in the course of the illness (mean
illness duration of 95 days) and, despite obvious working memory difficulties,
set-shifting ability is less impaired (Mahony et al., 2001). Although there are no
published longitudinal studies available using this task, these findings suggest that
there is no or minimal impairment of set-shifting at first presentation, with a
possible slow decline over continuing illness.
Working memory, a late emerging aspect of executive functions, is clearly
impaired in patients with established illness (Pantelis et al., 1997) and also at the
first presentation of the disorder (Hutton et al., 1998; Proffitt, 2001), with no pro-
gression over time (Proffitt, 2001). In addition, there are deficits in young people
at ultra-high risk for psychosis prior to the onset of illness (Wood et al., 2003),
indicating that this ability has never been fully functional. Taken together with
the relatively late development of spatial working memory in normal individuals
(De Luca et al., 2003), these findings suggest that functions which normally develop
during the at-risk period for the development of psychosis manifest the most severe
deficits.
An important aspect of the maturational changes in the expressed sympto-
matology of schizophrenia is the highly contested argument of whether impair-
ments in cognitive functioning that present at the onset of psychosis represent a
82 S. J. Wood, C. R. De Luca, V. Anderson, and C. Pantelis

true regression in intellectual ability or, rather, a failure to progress developmen-

tally. As stated at the beginning of this chapter, knowledge of the normal maturation
of skills across the lifespan allows for improved categorization of deficits as a deter-
ioration in functioning in contrast to a failure to make age-appropriate gains.
Problems with extrapolating this information from study data occur when using
age-corrected IQ values to look at changes in performance over childhood and
early adolescence, a time when major cognitive developments are in progress. This
method appears to disguise stability in performance over time, with equivalent
raw scores at different ages placing children at opposite ends of the intelligence
range. This view is supported by studies such as that of Jacobs et al. (2001), who
found that despite decline in IQ over a number of years, the mental age scores of
older children with spina bifida remained stable, reflecting a halt in development
rather than a loss in function. Another recent study looking at post-psychotic IQ
changes in patients with childhood-onset schizophrenia used raw scores as an index
of dementia (Bedwell et al., 1999). Again, although there was a significant decline in
post-psychotic full-scale IQ scores, the raw scores obtained by these children for all
the subtests did not change over the testing interval of approximately 3 years. These
findings provide strong support for the idea that the decline in IQ scores observed
during adolescence for children with schizophrenia is the result of an inability to
acquire new information and abilities (Bedwell et al., 1999) rather than reflect-
ing cognitive deterioration. Such studies are important in addressing the ongoing
debate about whether schizophrenia is a disorder of early or late neurodevelopment
and/or involves neurodegenerative processes (Lieberman, 1999; Pantelis et al., 2003;
Velakoulis et al., 2000).

Conclusions and future directions

The development of cognitive function is a lengthy process that is not complete until
early adulthood, and during which different domains mature at different rates and
times. An understanding of normal cognitive development is particularly impor-
tant in order to appreciate the pattern of impairments in children and adolescents
who suffer from illnesses of putative neurodevelopmental origin. In this chapter,
we have discussed the gains in function made during adolescence, a time of major
upheaval in behavioral and social domains. First, we reviewed the structural brain
changes that occur during this time, after which we explored neuropsychological
development, with a special focus on executive functions. The development of these
skills is characterized by “spurts” beginning from as young as 12 months of age, with
the majority of functions coming online around 8 years of age. However, evidence
from our own and others’ studies suggest that working memory in particular is
slow to mature and may not reach maximal levels until early adulthood. We then
discussed links between cognitive and cerebral development, with an emphasis on
83 Cognitive development in adolescence

the impact of developmental lesions. Finally, we put forward a hypothesis explain-

ing the neuropsychological deficits in schizophrenia as an interaction between the
timing of illness onset and the timing of normal cognitive development. Specific-
ally, we have suggested that cognitive functions that mature around the time when
the illness first presents, such as working memory, are more impaired than those
functions that mature earlier.
The study of cognitive development through adolescence is of great importance
to our understanding of the neurobiology of disorders that first present at this
time (Luna and Sweeney, 2001). However, in order to deepen our knowledge of the
developmental period, it is now necessary to investigate how brain development is
related to improvements in cognitive function. One way in which this is being done
is through functional imaging (Casey et al., 2000), which allows an exploration of the
refinement of brain activation from childhood to adulthood. However, to date, these
studies have been cross-sectional only, and longitudinal studies are now required to
improve our understanding of the individual relationships between structural and
functional brain development. In addition, our understanding of the distributed
nature of the brain systems involved in higher-order cognition (Goldman-Rakic,
1988) means that our paradigms will need to explore the maturation of functional
integration within these networks (Luna et al., 2001).
Finally, we will need to explore new methods such as pharmacological MRI
(Vaidya et al., 1998) and carbon-13 spectroscopy (Shen et al., 1999). Both tech-
niques can potentially tell us about the development of neurotransmitter function
in greater detail than has been possible up to now, allowing an understanding of
adolescent development that will have enormous implications for the diagnosis and
treatment of mental illness.

REFERENCES

Anderson, S. W., Bechara, A., Damasio, H., Tranel, D., Damasio, A. R. (1999). Impairment of
social and moral behavior related to early damage in human prefrontal cortex. Nat Neurosci
2: 1032–1037.
Anderson, S. W., Damasio, H., Tranel, D., Damasio, A. R. (2000). Long-term sequelae of prefrontal
cortex damage acquired in early childhood. Dev Neuropsychol 18: 281–296.
Anderson, V. (1988). Recovery of function in children: the myth of cerebral plasticity. In Brain
Impairment. Proceedings from the Thirteenth Annual Brain Impairment Conference, ed. M.
Matheson, H. Newman. Sydney: University of Sydney, pp. 223–247.
(1998). Assessing executive function in children: biological, psychological and developmental
considerations. Neuropsychol Rehab 8: 319–349.
Anderson, V., Jacobs, R. (2004). Frontal lobe damage in children: interruptions to normal devel-
opment. In Enfance and Neuropsychologie: Interface entre la Recherche et la Clinique, ed. J.-P.
Laurent. Quebec: Quebec University Press, in press.
84 S. J. Wood, C. R. De Luca, V. Anderson, and C. Pantelis

Anderson, V., Lajoie, G., Bell, R. (1995). Neuropsychological Assessment of the School-aged Child.
Department of Psychology, Melbourne: University of Melbourne.
Anderson, V. A., Anderson, P., Northam, E., Jacobs, R., Catroppa, C. (2001a). Development of
executive functions through late childhood and adolescence in an Australian sample. Dev
Neuropsychol 20: 385–406.
Anderson, V., Northam, E., Hendy, J., Wrennall, J. (2001b). Developmental Neuropsychology.
Hove, UK: Psychology Press.
Ardila, A., Rosselli, M. (1994). Development of language, memory and visuospatial abilities in
5- to 12-year-old children using a neuropsychological battery. Dev Neuropsychol 10:
97–120.
Baddeley, A. (1998). Recent developments in working memory. Curr Opin Neurol 8: 234–238.
Barkley, R. A. (1996). Linkages between attention and executive functions. In Attention, Memory
and Executive Function, ed. G. R. Lyon, A. Krasnegor. Baltimore, MD: Paul H. Brookes, pp.
307–326.
Bedwell, J. S., Keller, K., Smith, A. K., et al. (1999). Why does postpsychotic IQ decline in
childhood-onset schizophrenia? Am J Psychiatry 156: 1996–1997.
Benes, F. M. (1989). Myelination of cortical–hippocampal relays during late adolescence.
Schizophr Bull 15: 585–593.
Benes, F. M., Turtle, M., Khan, Y., Farol, P. (1994). Myelination of a key relay zone in the hippo-
campal formation occurs in the human brain during childhood, adolescence and adulthood.
Arch Gen Psychiatry 51: 477–484.
Case, R. (1992). The role of the frontal lobes in the regulation of cognitive development. Brain
Cognit 20: 51–73.
Casey, B. J., Giedd, J. N., Thomas, K. M. (2000). Structural and functional brain development
and its relation to cognitive development. Biol Psychiatry 54: 241–257.
Chelune, G. J., Baer, R. (1986). Developmental norms of the Wisconsin Card Sorting Test. J Clin
Exp Neuropsychol 8: 219–228.
Chugani, H., Phelps, M., Mazziotta, J. (1987). Positron emission tomography study of human
brain functional development. Ann Neurol 22: 487–497.
De Luca, C. R., Wood, S. J., Anderson, V., et al. (2003). Normative data from the CANTAB.
I: Development of executive function over the lifespan. J Clin Exp Neuropsychol 25:
242–254.
Dubas, J. (1991). Cognitive abilities and physical maturation. In Encyclopedia of Adolescence,
Vol. 1, ed. R. Lerner et al. New York: Garland Publishing, pp. 133–138.
Elliott, R., Sahakian, B. J. (1995). The neuropsychology of schizophrenia: relations with clinical
and neurobiological dimensions. Psychol Med 25: 581–594.
Eslinger, P. J., Damasio, A. R. (1985). Severe disturbance of higher cognition after bilateral frontal
lobe ablation: Patient EVR. Neurology 35: 1731–1741.
Eslinger, P. J., Grattan, L. M., Damasio, H., Damasio, A. R. (1992). Developmental consequences
of childhood frontal lobe damage. Arch Neurol 49: 764–769.
Eslinger, P., Biddle, K., Grattan, L. (1997). Cognitive and social development in children with
prefrontal cortex lesions. In Development of the Prefrontal Cortex: Evolution, Neurology, and
Behavior, ed. P. S. Goldman-Rakic. Baltimore, MD: Brookes, pp. 295–336.
85 Cognitive development in adolescence

Eslinger, P., Biddle, K., Pennington, B., Page, R. (1999). Cognitive and behavioral development
up to 4 years after early right frontal lobe lesion. Dev Neuropsychol 15: 157–191.
Espy, K. (1997). The shape school: assessing executive function in preschool children. Dev
Neuropsychol 13: 495–499.
Giedd, J. N., Blumenthal, J., Jeffries, N. O., et al. (1999). Brain development during childhood
and adolescence: a longitudinal MRI study. Nat Neurosci 2: 861–863.
Golden, C. (1981). Luria Nebraska Children’s Battery: theory and formulation. In Neuropsycho-
logical Assessment of the School-aged Child, ed. G. W. Hynd, J. E. Obzut. London: Grune and
Stratton, pp. 277–302.
Goldman-Rakic, P. S. (1988). Topography of cognition: parallel distributed networks in primate
association cortex. Annu Rev Neurosci 11: 137–156.
Happe, F., Frith, U. (1996). The neuropsychology of autism. Brain 119: 1377–1400.
Heilman, K., Valenstein, E., Watson, R. (1997). Disorders of attention. In Contemporary Behavioral
Neurology, ed. M. Trimble, J. Cummings. London: Harcourt, pp. 127–138.
Hitch, G. J. (2002). Developmental changes in working memory: a multicomponent view. In
Lifespan Development of Human Memory, ed. P. Graf, N. Ohta. Cambridge, MA: MIT Press,
pp. 15–38.
Hitch, G. J., Towse, J., Hutton, U. (2001). What limits children’s working memory span? Theo-
retical accounts and applications for scholastic development. J Exp Psychol Gen 130: 184–
198.
Huttenlocher, P. (1984). Synapse elimination and plasticity in developing human cerebral cortex.
Am J Ment Defic 88: 488–496.
Hutton, S. B., Puri, B. K., Duncan, L.-J., et al. (1998). Executive function in first-episode
schizophrenia. Psychol Med 28: 463–473.
Jacobs, R., Northam, E., Anderson, V. (2001). Cognitive outcome in children with myelomeningo-
cele and perinatal hydrocephalus. J Dev Phys Disabil 13: 389–405.
Klingberg, T., Vaidya, C. J., Gabrieli, J. D. E., Moseley, M. E., Hedehus, M. (1999). Myelination
and organization of the frontal white matter in children: a diffusion tensor MRI study.
Neuroreport 10: 2817–2821.
Klingberg, T., Forssberg, H., Westerberg, H. (2002). Increased brain activity in frontal and pari-
etal cortex underlies the development of visuospatial working memory capacity during
childhood. J Cogn Neurosci 14: 1–10.
Koski, L., Petrides, M. (2002). Distractibility after unilateral resections from the frontal and
anterior cingulate cortex in humans. Neuropsychologia 40: 1059–1072.
Kwon, H., Reiss, A. L., Menon, V. (2002). Neural basis of protracted developmental changes in
visuo-spatial working memory. Proc Natl Acad Sci USA 99: 13336–13341.
Leduc, M., Herron, J. E., Greenberg, D. R., Eslinger, P. J., Grattan, L. M. (1999). Impaired awareness
of social and emotional competencies following orbital frontal lobe damage. Brain Cogn 40:
174–177.
Levin, H. S., Culhane, K. A., Hartmann, J., et al. (1991). Developmental changes in performance
on tests of purported frontal lobe functioning. Dev Neuropsychol 7: 377–395.
Lieberman, J. A. (1999). Is schizophrenia a neurodegenerative disorder? A clinical and neurobi-
ological perspective. Biol Psychiatry 46: 729–739.
86 S. J. Wood, C. R. De Luca, V. Anderson, and C. Pantelis

Luciana, M., Nelson, C. A. (1998). The functional emergence of prefrontally guided working
memory systems in four- to eight-year-old children. Neuropsychologia 36: 273–293.
Luna, B., Sweeney, J. A. (2001). Studies of brain and cognitive maturation through childhood
and adolescence: a strategy for testing neurodevelopmental hypotheses. Schizophr Bull 27:
443–455.
Luna, B., Thulborn, K. R., Munoz, D. P., et al. (2001). Maturation of widely distributed brain
function subserves cognitive development. NeuroImage 13: 786–793.
Mahony, K., Pantelis, C., Proffitt, T., et al. (2001). Comparison of set-shifting ability in patients
with first-episode psychosis and chronic schizophrenia. Schizophr Res 49: 114–115.
Manly, T., Anderson, V., Nimmo-Smith, I., et al. (2001). The differential assessment of children’s
attention: The Test of Everyday Attention for Children (TEA-Ch). Normative sample and
ADHD performance. J Child Psychol Psychiatry 42: 1065–1087.
Marlowe, W. B. (1992). The impact of a right prefrontal lesion on the developing brain. Brain
Cogn 20: 205–213.
Mateer, C. A. (1990). Cognitive and behavioral sequalae of face and forehead injury in childhood.
J Clin Exp Neuropsychol 12: 95.
McKay, K., Halperin, J., Schwartz, S., Sharma, V. (1994). Developmental analysis of three aspects of
information processing: sustained attention, selective attention and response organization.
Dev Neuropsychol 10: 121–132.
Meltzoff, A., Moore, M. (1977). Imitation of facial and manual gestures by human neonates.
Science 198: 74–78.
Murray, R. M., Lewis, S. (1987). Is schizophrenia a neurodevelopmental disorder? [Editorial]
Br Med J 295: 681–682.
Nichelli, F., Bulgheroni, S., Riva, D. (2001). Developmental patterns of verbal and visuospatial
spans. Neurol Sci 22: 377–384.
Owen, A. M., Roberts, A., Hodges, J., et al. (1993). Contrasting mechanisms of impaired atten-
tional set-shifting in patients with frontal lobe damage or Parkinson’s disease. Brain 116:
1159–1175.
Pantelis, C., Barnes, T. R., Nelson, H. E., et al. (1997). Frontal-striatal cognitive deficits in patients
with chronic schizophrenia. Brain 120: 1823–1843.
Pantelis, C., Barber, F., Barnes, T. R., et al. (1999). Comparison of set-shifting ability in patients
with chronic schizophrenia and frontal lobe damage. Schizophr Res 37: 251–270.
Pantelis, C., Yücel, M., Wood, S. J., McGorry, P. D., Velakoulis, D. (2001). The timing and
functional consequences of structural brain abnormalities in schizophrenia. NeuroSci News
4: 36–46.
Pantelis, C., Wood, S. J., Maruff, P. (2002). Schizophrenia. In Cognitive Deficits in Brain Disorders,
ed. A. M. Owen, J. Harrison. London: Martin Dunitz, pp. 217–248.
Pantelis, C., Yücel, M., Wood, S. J., McGorry, P. D., Velakoulis, D. (2003). Early and late neurode-
velopmental disturbances in schizophrenia and their functional consequences. Aust NZ J
Psychiatry 37: 399–406.
Paus, T., Zijdenbos, A., Worsley, K., et al. (1999). Structural maturation of neural pathways in
children and adolescents: in vivo study. Science 283: 1908–1911.
87 Cognitive development in adolescence

Paus, T., Collins, D. L., Evans, A. C., et al. (2001). Maturation of white matter in the human brain:
a review of magnetic resonance studies. Brain Res Bull 54: 255–266.
Pentland, L., Todd, J. A., Anderson, V. (1998). The impact of head injury severity on planning
ability in adolescence: a functional analysis. Neuropsychol Rehab 8: 301–317.
Petersen, A. (1998). Adolescence. In Behavioral Medicine and Women: A Comprehensive
Handbook, ed. E. Blechman, K. Brownell. New York: Guilford Press, pp. 45–50.
Piaget, J. (1965). Moral Judgment of the Child. New York: Free Press.
Price, B. H., Daffner, K. R., Stowe, R. M., Mesulam, M. M. (1990). The compartmental learning
disabilities of early frontal lobe damage. Brain 113: 1383–1393.
Proffitt, T. (2001). The stability of spatial working memory and problem solving deficits in first-
episode psychosis and established psychotic illness. Ph.D. Thesis, University of Melbourne,
Melbourne.
Rakic, P., Bourgeois, J.-P., Goldman-Rakic, P. S. (1994). Synaptic development of the cerebral
cortex: implications for learning, memory and mental illness. In Progress in Brain Research,
The Self-organizing Brain: From Growth Cones to Functional Networks, Vol. 102, ed. J. van
Pelt, M. Corner, H. B. Uylings, F. Lopes de Silva. Amsterdam: Elsevier, pp. 227–243.
Rebok, G., Smith, C., Pascualvaca, D., et al. (1997). Developmental changes in attentional per-
formance in urban children from eight to thirteen years. Child Neuropsychol 3: 47–60.
Reiss, A. L., Abrams, M. T., Singer, H. S., Ross, J. L., Denckla, M. B. (1996). Brain development,
gender and IQ in children: a volumetric imaging study. Brain 119: 1763–1774.
Rezai, K., Andreasen, N. C., Alliger, R. J., et al. (1993). The neuropsychology of the prefrontal
cortex. Arch Neurol 50: 636–642.
Ridderinkhof, K., van der Molen, M. (1997). Mental resources, processing speed and inhibitory
control: a developmental perspective. Biol Psychol 45: 241–261.
Roberts, A., Robbins, T. W., Weiskrantz, L. (1998). Discussions and conclusions. In The Prefrontal
Cortex: Executive and Cognitive functions, ed. A. Roberts, T. W. Robbins, L. Weiskrantz.
Oxford: Oxford University Press, pp. 221–242.
Rubia, K., Overmeyer, S., Taylor, E., et al. (2000). Functional frontalisation with age: mapping
neurodevelopmental trajectories with fMRI. Neurosci Biobehav Rev 24: 13–19.
Shanab, M., Yasin, A. (1979). Intradimensional and extradimensional shifts by Jordanian students.
J Gen Psychol 100: 199–213.
Shen, J., Petersen, K. F., Behar, K. L., et al. (1999). Determination of the rate of the glutamate/
glutamine cycle in the human brain by in vivo 13 C NMR. Proc Natl Acad Sci USA 96: 8235–
8340.
Sowell, E. R., Thompson, P. M., Holmes, C., Jernigan, T. L., Toga, A. (1999). In vivo evidence for
post-adolescent brain maturation in frontal and striatal regions. Nat Neurosci 2: 859–861.
Sowell, E. R., Thompson, P. M., Tessner, K. D., Toga, A. (2001a). Mapping continued brain
growth and gray matter density reduction in dorsal frontal cortex: inverse relationships
during postadolescent brain maturation. J Neurosci 21: 8819–8829.
Sowell, E. R., Delis, D., Stiles, J., Jernigan, T. L. (2001b). Improved memory functioning and
frontal lobe maturation between childhood and adolescence: a structural MRI study. J Int
Neuropsychol Soc 7: 312–322.
88 S. J. Wood, C. R. De Luca, V. Anderson, and C. Pantelis

Sowell, E. R., Peterson, B. S., Thompson, P. M., et al. (2003). Mapping cortical change across the
human life span. Nat Neurosci 6: 309–315.
Spear, L. (2000). The adolescent brain and age-related behavioral manifestation. Neurosci
Biobehav Rev 24: 417–463.
Stuss, D. T. (1992). Biological and psychological development of executive functions. Brain Cogn
20: 8–23.
Stuss, D. T., Benson, D. (1987). The frontal lobes and control of cognition and memory. In The
Frontal Lobes Revisited, ed. E. Perecman. New York: Irbn Press, pp. 141–154.
Travis, F. (1998). Cortical and cognitive development in 4th, 8th and 12th grade students: the
contribution of speed of processing and executive function to cognitive development. Biol
Psychol 48: 37–56.
Vaidya, C., Austin, G., Kirkorian, G., et al. (1998). Selective effects of methylphenidate in attention
deficit hyperactivity disorder: a functional magnetic resonance study. Proc Natl Acad Sci 95:
14494–14499.
Velakoulis, D., Wood, S. J., McGorry, P. D., Pantelis, C. (2000). Evidence for progression of brain
structural abnormalities in schizophrenia: beyond the neurodevelopmental model. Aust NZ
J Psychiatry 34(Suppl.): S113–S126.
Walker, E., Savoie, T., Davis, D. (1994). Neuromotor precursors of schizophrenia. Schizophr Bull
20: 441–451.
Weinberger, D. R. (1987). Implications of normal brain development for the pathogenesis of
schizophrenia. Arch Gen Psychiatry 44: 660–669.
Welsh, M. C., Pennington, B. F. (1988). Assessing frontal lobe functioning in children: views from
developmental psychology. Dev Psychol 4: 199–230.
Whitney, L., White, K. (1993). Dimensional shift and the transfer of attention. Q J Exp Psychol
46B: 225–252.
Wood, S. J., Pantelis, C. (2001). Does a neurodevelopmental lesion involving the hippocampus
explain memory dysfunction in schizophrenia? Z Neuropsychol 12: 61–67.
Wood, S. J., Pantelis, C., Proffitt, T., et al. (2003). Spatial working memory ability is a marker of
risk-for-psychosis. Psychol Med 33: 1239–1247.
Yakovlev, P., Lecours, A. (1967). The myelogenetic cycles of regional maturation of the brain. In
Regional Development of the Brain in Early Life, ed. A. Minkowski. Oxford: Blackwell, pp.
3–70.
 5

Brain plasticity and long-term function after

early cerebral insult: the example of very
preterm birth
Matthew Allin, Chiara Nosarti, Larry Rifkin, and Robin M. Murray
Institute of Psychiatry, King’s College, London, UK

Plastic : . . . easily influenced; impressionable . . . capable of being moulded or formed.

Collins Concise English Dictionary

Plasticity is an important phenomenon of vertebrate nervous systems throughout

life and is an adaptation that allows an organism to adjust its behavior so as to deal
with changes in its environment. In young animals, where both brain and behavior
are rapidly changing as the animal grows, plasticity is particularly marked. In our
own species, with our long period of postnatal development, plasticity allows for the
experience-dependent tuning of neural networks that is required for the acquisition
of complex behaviors such as our language and social interaction (Johnson, 2001).
This chapter will be concerned with developmental plasticity rather than adult
plasticity – indeed there is some evidence that these may have different underlying
mechanisms (Linkenhoker and Knudsen, 2002). We will discuss the recovery, or
sparing, of functions and the reorganization of brain structure that occurs as a
consequence of early brain injury, using the example of preterm birth. This is of
interest in the context of this book because of the neurodevelopmental theory of
schizophrenia (Murray and Lewis, 1987; Weinberger, 1987) and of the link between
obstetric complications and schizophrenia (Cannon et al., 2002; Lewis and Murray,
1987; Lewis et al., 1989).

A brief history of plasticity

The increased capacity of the immature brain to be ‘moulded or formed’ is part
of everyday experience. For example, as most of us can attest, many schoolchil-
dren and most adults find learning a second language very difficult. Those who

Neurodevelopment and Schizophrenia, ed. Matcheri S. Keshavan et al. Published by Cambridge University
Press.

C Cambridge University Press 2004.

89
90 M. Allin, C. Nosarti, L. Rifkin, and R. M. Murray

grew up in a bilingual environment, by comparison, have no trouble speaking

two languages fluently since they were exposed to both early in life, at a time
when their brains were more plastic. This period of plasticity for language acqui-
sition is relatively brief: infants respond to phonemes from widely disparate lan-
guages, but this ability begins to decline once they pass their first birthday, and
they acquire a preference for the phonemes of their parents’ language (Kolb,
1989). The “get them young” principle has also been applied to learning a musi-
cal instrument; for example, the intensive “Suzuki method” of teaching the violin
starts in early childhood, the earlier the better (motto: “classical music from age
zero”).
This idea of a “critical period” for learning has caused some parents to play
Mozart to their infants, both pre- and postnatally (Jones and Zigler, 2002). It will
be interesting to see whether this leads to a cohort of gifted composers in the next
two decades, since it is said that Mozart himself “. . . spent much time at the clavier,
picking out thirds . . .” from the age of 3. Whether or not this curious experiment
of fashion has any such effects, there is evidence to suggest that such early learning
alters the brain in measurable ways. For example, Pantev et al. (1998) found that
auditory cortex activation in response to the sound of a piano was greater in pianists
who had started playing before they were 9 years old. Learning to play after this age
did not have this effect. Early experience may also alter the structure of the brain,
in a way that is gross enough to be visible on magnetic resonance imaging (MRI)
scans later in life. For example Castro-Caldas et al. (1998) showed that the brains
of those who learned to read and write in childhood are structurally different from
those who did not (see also Frith, 1998).
The concept of “sensitive periods” in development dates back to Stockard’s experi-
ments in the 1920s in which he showed that embryos were better able to withstand
hypoxia at some stages of development than at others. Wiesel and Hubel (1963)
developed this concept in their classic experiments on the visual systems of cats.
Kittens deprived of the use of one eye lost the pattern of ocular dominance columns
in their visual cortex; adult cats do not reorganize their brains in this way. Similar
results are found in many other experimental models (Linkenhoker and Knudsen,
2002). Different functions, and the areas of the brain that support them, mature
at different rates and come “online” at different times; they may, therefore, have
different “critical periods.” For example, the acquisition of eye movement control
in humans takes place during the first 6 months of life and may depend on the
functional maturation of the parietal cortex (Johnson, 2001). At the other end of
the scale, there is evidence for a late period for language acquisition, in adolescence,
which may represent the fine tuning of frontal cortex and frontostriatal connections
(Grimshaw et al., 1998; Sowell et al., 1999).
91 Brain plasticity and long-term function

Plasticity after brain injury

The developmental plasticity of the young brain means that the effects of brain
injuries in infancy and childhood are different to the effects of an equivalent injury
in adulthood. One of the first scientific observations of this phenomenon was made
by the nineteenth century surgeon, neurologist, and anthropologist Paul Broca
(whose own brain has no further capacity for plasticity, residing as it does in a jar
in the Musee de l’Homme in Paris). In 1863, Broca famously identified the area of
left inferior frontal cortex that is necessary for the production of speech: if dam-
aged by a stroke or other insult, the result would be expressive dysphasia. However,
Broca also observed that if this area of the left hemisphere was damaged early in
life, speech could be preserved. He suggested that in this case expressive language
function could be taken over by the right hemisphere: the brain could remodel
itself if the injury occurred sufficiently early in development. More recent research
bears this out, suggesting that there is good functional recovery of language when
brain damage is sustained to the left hemisphere in infancy. If the injury is sus-
tained in middle childhood, however, adult-like aphasic symptoms occur (Hecaen,
1983).
It is now clear that the immature brain is capable of extensive reorganization,
with preservation of function in the face of injury (see review by Payne and Lomber,
2001). This was elaborated in the 1930s by Margaret Kennard (1936) and became
known as the Kennard principle, which states that the earlier in development the
brain is damaged the less severe is the resulting behavioral disturbance (Kolb, 1989).
This sparing of function could be the result of several processes: the recruitment of
pre-existing alternative synaptic connections and axonal pathways, the development
of new pathways so that the brain is ‘rewired’ (Elbert et al., 2001; Kolk, 2000),
or the learning of alternative cognitive strategies. There is some evidence for all
three possibilities. Interestingly, the behavioral disturbances can also be mitigated
somewhat in experimental animals by increasing the richness of the environment
in which they are reared (Kolb, 1989; Kolb and Whishaw, 1989). The drive towards
sparing functions is clearly a very strong one. Even children born with very little
cerebral cortex as a result of hydranencephaly can demonstrate some behaviors and
may be able to recognize faces even when primary visual (occipital) cortex is not
present (Shewmon et al., 1999).
However, there are some exceptions to the Kennard principle. For example,
Goldman (1971) reported that frontal cortex lesions in infant monkeys had little
effect on behavior initially but produced deleterious effects on behavior later in
life, as the frontal cortex matured. Sams-Dodd et al. (1997) subjected newborn
rats to lesions of the hippocampus. Again, these rats behaved apparently normally
92 M. Allin, C. Nosarti, L. Rifkin, and R. M. Murray

initially, but their behavior became grossly disturbed once they reached maturity. In
humans, Anderson et al. (1999) studied two young adults who received focal non-
progressive prefrontal damage before 16 months of age. These individuals showed
no behavioral abnormalities up to the age of 3 years but had severely impaired
social behavior despite normal basic cognitive abilities later on in life. Therefore,
it seems that the very plasticity that enables the young brain to overcome injury
so impressively, to reorganize its structure, and spare its essential functions may
have an adverse effect on later development. The neurodevelopmental hypothe-
sis of the etiology of schizophrenia postulates that a brain which is in some way
abnormally “set up” as a result of an early insult may start to exhibit functional
impairments later in life, particularly during the period of brain maturation and
acquisition of adult cognitive capabilities in adolescence and young adulthood
(Allin and Murray, 2002; Murray and Lewis, 1987; Weinberger, 1987). To examine
these issues further, we have been studying a cohort of individuals who were sub-
jected to risk of brain injuries around the time of birth as a result of being born
preterm.

The example of very preterm birth

From the public health point of view, one of the most important challenges to
early brain plasticity comes from preterm birth. Indeed, with the rapid advances in
neonatal care, ever more and ever smaller babies are being rescued. They are at great
risk of brain injury as a result of being born in an immature state, and their resilience
is tested to the limit in the first few weeks of life. Unfortunately, relatively little is
known about the long-term consequences of this. Do such individuals grow into
adolescents and adults still showing the neuropsychological scars of their preterm
birth?

The University College Hospital London study

We have been carrying out a longitudinal study of the consequences of preterm birth
in collaboration with the University College Hospital London (UCHL) Department
of Neonatal Paediatrics. This study combines detailed assessments of function from
the first days of life, through childhood, and into adolescence, together with assess-
ment of brain anatomy. In 1979–81, 172 infants born before 33 weeks of gestation
(which we shall define as very preterm [VPT]) and admitted consecutively to the
neonatal unit of UCHL within 5 days of birth survived and were discharged. Of this
cohort, four died within 24 months; the remaining 168 were enrolled for long-term
follow-up.
93 Brain plasticity and long-term function

Follow-up
Prospective assessments of neurological and cognitive status of these children were
carried out at 1, 4, and 8 years. At 14–15 years, 163 (97%) individuals were traced,
including 16 who were living abroad. Of the 147 living in the UK, 118 (80%)
agreed to attend for assessment. The cohort members who were unavailable for
investigation did not differ from those studied in birth weight, gestational age
at birth, sex ratio, mode of delivery, condition at birth, the need for mechanical
ventilation, or neonatal cranial ultrasonographic findings. At 18 years, 102 of the
VPT subjects were assessed again.

Patterns of perinatal brain injury

Being born before 33 weeks of gestation pitches a developmentally very immature
infant into a hostile world. There they are prey to infections, which their immune
systems are ill equipped to fight off. Their lungs are often immature, and they are
unable to take in as much oxygen as they need. As a result, they frequently need
supplementary oxygen, and even mechanical ventilation. It is difficult to provide
them with sufficient nutrition. Their brains contain immature and delicate blood
vessels, particularly in the remnants of their germinal matrix, and they do not
autoregulate their cerebral blood flow effectively. Hypoxia, along with unprotected
swings in systemic (and, therefore, cerebral) blood pressure, may cause these fragile
blood vessels to break (Hoon, 1995; Rosenbloom and Sullivan, 1996).
Several patterns of brain lesions are thus associated with preterm birth. Hemor-
rhage may occur from immature vessels in the germinal matrix (germinal matrix
hemorrhage [GMH]). This is the region adjacent to the ventricles that produces
the new neurons that migrate out, along highways made of radial glia, to form
the layers of the cerebral cortex. By 33 gestational weeks, the germinal matrix has
largely involuted, but remnants of it are still present, particularly around the frontal
horns of the lateral ventricles. GMH may also involve the ventricles (intraventric-
ular hemorrhage [IVH]). These bleeds may also be associated with periventricular
hemorrhagic infarction (PHI) of white matter in a localized and usually asym-
metric distribution (Volpe, 1998). Diffuse white matter damage is also common
and occurs adjacent to the ventricles (periventricular leukomalacia [PVL]) (Paneth
et al., 1994). PVL is a necrotic lesion, causing the death of axons and their developing
oligodendrocytes. It, therefore, also leads to incomplete and/or delayed myelination
(Volpe, 1998) and is the lesion most likely to be associated with severe handicap –
predominantly cerebral palsy. In addition to these lesions, vulnerable groups of
neurons, which include those undergoing active mitosis or migration and those
in an immature state of differentiation such as oligodendrocyte precursors, may
be adversely, and diffusely, affected (Back et al., 2001, 2002; Johnston, 1998). The
94 M. Allin, C. Nosarti, L. Rifkin, and R. M. Murray

postmortem studies of Marin-Padilla (1996, 1997, 1999) have demonstrated that

even localized lesions in the preterm brain can cause distant alterations of cell struc-
ture, dendritic arborization, and connectivity. So all these lesions may be expected
to have both localized and diffuse effects on brain structure.
The UCHL VPT subjects were among the first infants to have neonatal ultrasound
brain scanning performed, using relatively crude linear array apparatus. Even this
imaging technique, with its relatively poor spatial resolution, showed a high preva-
lence of brain lesions: 43% had abnormal cranial ultrasound scans (Stewart et al.,
1983). A follow-up study by Roth et al. (1993) of 206 VPT infants reported the
most frequently observed abnormality to be uncomplicated periventricular hem-
orrhage (PVH), which was present in 55 infants (27%). The next most common
abnormality was ventricular dilatation, present in 21 infants (10%). In 13 of these
infants, the dilatation had resolved on ultrasound by the time of discharge from the
neonatal unit. Five infants developed hydrocephalus, and three of these required
ventriculoperitoneal shunting. Thirteen (6%) had ultrasonographic evidence of
cerebral atrophy. This was localized and associated with intraparenchymal PVH in
seven infants but was more diffuse in six infants, in whom it was associated with
PVH or cystic PVL. So, brain injuries are clearly common in VPT infants, but until
recently it was not clear whether VPT individuals who survive the perinatal period
continue to show both focal and diffuse abnormalities of brain structure later in
life.

Can the sequelae of preterm birth still be seen in the brains of adolescents?
The relatively crude imaging resolution provided by ultrasound had demonstrated
a significant amount of brain pathology at birth in the UCHL VPT cohort. However,
some of these abnormalities had apparently resolved at the time of discharge, and the
extent to which the brain damage suffered by these infants resulted in permanent
change in brain structure was unclear. Therefore, at follow-up at 14–15 years,
structural magnetic resonance imaging (MRI) of the brain was performed on all
the participating VPT subjects. Stewart et al. (1999) reported on 72 VPT subjects
and 21 full-term controls. The scans were “blindly” rated according to a structured
format by two neuroradiologists. The scans were classified as normal (no detected
abnormality), equivocal (negligible ventricular dilatation, or negligible thinning
of the corpus callosum, or an isolated white matter hyperintensity), or abnormal
(more than one parenchymal lesion plus definite ventricular dilatation, or definite
thinning or atrophy of the corpus callosum plus reduced white matter or cortical
volume plus multiple areas of white matter signal changes, or intraparenchymal
cysts).
95 Brain plasticity and long-term function

Only 24% of the scans of these VPT adolescents were rated as normal; 21% were
equivocal and 56% abnormal. The equivalent percentages for the controls were 71%
normal, 24% equivocal and 5% abnormal. In those preterm individuals with an MRI
scan classified as abnormal, ventricular dilatation was observed in 80%, posterior
trigonal dilatation in 73%, thinning of the corpus callosum (particularly involving
the splenium) in 65%, abnormal white matter signal in 45%, and decreased white
matter volume in 25%. Ventricular dilatation and thinning of the corpus callosum
were correlated with each other and with other brain lesions and were considered
to be markers of hypoxic–ischemic damage. The white matter signal abnormalities
were thought to reflect scattered patchy gliosis secondary to ischemic damage (Hope
et al., 1988). In addition to these qualitative assessments of MRI scans, the volumes
of the whole brain and of various brain regions and structures were determined
using the Cavalieri stereological method, supported by the MEASURE software
package (Johns Hopkins University, Baltimore, USA). The VPT subjects showed a
6.0% decrease in whole brain volume and a 11.8% decrease in total cortical gray
matter volume. They also had a 42.0% increase in the size of the lateral ventricles.
Those VPT adolescents who had experienced PVH in the neonatal period had a
mean ventricular size twice as large as those who did not, indicating that these
changes were the sequelae of the early damage. Additionally, those who had PVH
and/or ventricular enlargement on neonatal ultrasound were especially likely to
have decreased white matter volumes (Nosarti et al., 2002). This is in agreement with
Kuban et al. (1999), who found that both IVH and ventriculomegaly were powerful
predictors of white matter damage in the first few weeks of life. Nosarti et al. (2002)
also found a positive correlation between white matter volume and gestational age:
the more immature the infant, the greater the damage to the white matter. This
may be a reflection of the developmental vulnerability of white matter between
24 and 32 gestational weeks, when oligodendrocyte progenitors are immature and
vulnerable to hypoxic injury (Back et al., 2001, 2002; du Plessis and Volpe, 2002).
Damage to these cell populations may disrupt the wave of white matter myelination,
which normally rapidly increases after 29 weeks of gestation (Kinney et al., 1988),
leading to a reduction of white matter (and also possibly to an increase in ventricular
volume).
In addition, the volumes of several brain structures were measured. VPT subjects
had reduced volumes of left and right hippocampus, independent of overall brain
volume reduction (Nosarti et al., 2002). These findings are likely to be caused by
hypoxic–ischemic damage (Kuchna, 1994; Mallard et al., 1999). The volume of the
cerebellum was also significantly smaller in VPT adolescents, again independent of
total brain volume (Allin et al., 2001). This may reflect damage to cerebellar granule
cells at a sensitive period in their development, when they are actively dividing and
migrating (Johnston, 1998; Sohma et al., 1995). VPT individuals also had smaller
96 M. Allin, C. Nosarti, L. Rifkin, and R. M. Murray

corpora callosa, particularly in the posterior portion (the splenium) (Nosarti et al.,
2001a).

What are the consequences of persisting brain abnormality?

In the follow-up assessments of the UCHL VPT cohort, assessments were made of
neurological, neuropsychological, and behavioral function. Some of these results
are summarized below. In general, the VPT individuals showed good functional
compensation in the face of abnormal brain structure, as might be predicted to be
the result of a plastic response to early injury. However, there are some areas where
function was compromised, sometimes in subtle ways, and there is evidence that
VPT individuals are more likely to fail to match the educational performance of
their term-born peers (Hadders-Algra et al., 1988; Hall et al., 1995; Msall et al., 1998;
Snider, 1998). They are more likely to be considered “clumsy” than their term-born
classmates (Goyen et al., 1998; Luoma et al., 1998), a finding that has been termed
“developmental coordination disorder” by some authors (Huh et al., 1998).

Neurology
At 1 year of age, 10% of the UCHL cohort had minor neurological impairments
not sufficient to cause disability, and 11% had major impairments with concomi-
tant disability (Stewart et al., 1989). The proportion of children in the cohort who
showed major neurological impairments remained relatively constant at both 4
(15%) and 8 (12%) years, in agreement with other studies (Tin et al., 1997). Ultra-
sound evidence of white matter damage, ventricular dilatation, or hydrocephalus
were powerful predictors of major disability at 1, 4 and 8 years (Costello et al., 1988;
Roth et al., 1993; Stewart et al., 1983).
In contrast to the relative stability of major abnormalities, “minor” abnormalities
increased in prevalence from 10% at age 1 to 23% at 8 years of age (Roth et al.,
1994). By the age of 14 years, 66% of the cohort had an abnormal or equivocal
neurological examination (Stewart et al., 1999). This may be because the range of
motor skills that normal individuals can perform (and which can, therefore, be
tested for) increases with age. An alternative explanation is that lesions of the brain
acquired early in development may have a more significant deleterious effect when
the affected neuronal systems reach developmental maturity, as discussed above
(Goldman, 1971; Sams-Dodd et al., 1997; see Ch. 4).

Neuropsychology
The mean full-scale intelligence quotient (IQ) of the VPT subjects was well within
the normal range at 8 years, indicating that neural plasticity had succeeded in
97 Brain plasticity and long-term function

preserving function in spite of the abnormalities of brain structure reported above.

However, evidence of more subtle deficits was present. Half of the VPT subjects
showed neuropsychological evidence of poor interhemispheric interaction, and this
was associated with poorer school performance (Stewart et al., 1983). On initial
investigation at 15 years, the VPT subjects were significantly impaired relative to
controls on only two measures of cognitive function: cognitive flexibility and phone-
mic verbal fluency (Rushe et al., 2001). Naming, spelling, IQ, visuomotor function,
verbal memory, and visuospatial memory were not significantly impaired.

Behavior
Behavior was assessed at 15 years, using the Rutter behavioral scale (Rutter et al.,
1981) and the social adjustment scale of Cannon-Spoor and colleagues (1982).
Behavioral abnormalities were significantly increased only in those preterm subjects
with abnormal MRI scans. Social adjustment problems were increased in subjects
with equivocal and abnormal MRI. Those with normal MRI scans showed similar
social adjustment to controls. At 18 years, there was an increased prevalence of
psychiatric illness in the VPT subjects compared with controls, but this difference
did not reach statistical significance (Rooney et al., 2001). In particular, there was a
2% rate of psychotic disorder in the VPT group, which exceeds the expected 0.2%
population prevalence for this age. This result should be interpreted with caution
in this relatively small study group but is of obvious interest in the context of the
neurodevelopmental theory of schizophrenia.

The link between brain structure and brain function

In spite of the high prevalence of structural brain abnormality, the association
between this and performance on specific cognitive and neurological tests is not
straightforward. For example, Stewart et al. (1999) reported that the cognitive
performance of the VPT subjects was unrelated to the presence of gross brain
abnormality, except that reading age was lower in those preterm individuals with
abnormal scans. The quantitative MRI results above were analyzed in relation to
neurodevelopmental outcome at 14–15 years by Nosarti et al. (2002). They could
not find a relationship between neurodevelopmental status and measures of brain
structure, including whole brain volume, lateral ventricular volume, total gray and
white matter volumes, and hippocampal volume. This may be because the presence
of an early brain injury had altered brain structure such that functions had become
aberrantly mapped in the brain (Moses and Stiles, 2002). As a result, the normal
adult structure–function relationship may no longer hold.
98 M. Allin, C. Nosarti, L. Rifkin, and R. M. Murray

There were some areas of brain structural abnormality, however, that proved
to be more directly associated with functioning. Reduced cerebellar volume was
significantly associated with deficits in executive and visuospatial function and
language (Allin et al., 2001), findings that are broadly consistent with the “cerebellar
cognitive–affective syndrome” described by Schmahmann and Sherman (1998).
This suggests that cerebellar abnormalities may underlie some of the cognitive
deficits found in VPT individuals. There was no association between cerebellar
volume and motor neurological signs, however; perhaps suggesting that plasticity
can allow motor functions to be spared, but at the expense of subtle abnormalities
in cognitive performance.
VPT individuals have been reported to have poorer interhemispheric interac-
tion (Roth et al., 1994), and it is possible that this is related to the reduced size
of the corpus callosum, which is the major white matter tract connecting the two
hemispheres. Nosarti et al. (2001a) found that verbal IQ, reading age, and verbal
fluency scores were positively associated with reduced midsagittal area of the cor-
pus callosum. Preliminary results of a functional MRI (fMRI) study conducted by
our group showed that VPT boys with callosal thinning exhibited differential lat-
eralization of phonological processing during a verbal fluency task compared with
full-term controls (Rushe et al., 1999). Another fMRI study used visual and auditory
tasks in a similar cohort of preterm boys in adolescence (Santhouse et al., 2002)
and showed that the VPT subjects with damaged corpora callosa had significantly
different activation patterns from the control group, and from a group of VPT
adolescents without callosal damage. The findings suggest that the VPT brains
are using alternative neural networks to circumvent the damaged corpus callo-
sum. This could be a demonstration of “rerouting plasticity,” as will be discussed
below.
The left and right caudate nuclei were reduced in volume in VPT individuals,
but this was not significant after controlling for total brain volume (Nosarti et al.,
2001b). However, reduced bilateral caudate volumes were associated with higher
scores on the “hyperactivity” index of the Rutter behavioral scale (Schachar et al.,
1981). This characterizes the predominantly hyperactive/impulsive subtype of
attention-deficit hyperactivity disorder (ADHD), defined by poor inhibitory regu-
lation, limitations in perseverance, perseveration, restlessness, overactivity, and
impaired self-awareness (Landau et al., 1999). This is consistent with evidence
implicating the basal ganglia in the pathogenesis of ADHD (Semrud-Clikeman
et al., 2000). This has been investigated further (Nosarti et al., 2002, 2004) in
a group of 16-year-old VPT boys using a go–no-go fMRI paradigm. The find-
ings again suggest that VPT individuals are activating different neural networks
than controls and using alternative strategies when performing these cognitive
tasks.
99 Brain plasticity and long-term function

Possible mechanisms of brain reorganization and sparing of function

At the neuronal level
A brain that has been damaged in the perinatal period has lost neurons and synaptic
connections in a focal and/or distributed pattern, and this represents a deficit in
processing power that the brain seeks to mitigate. In response to injury, the brain
produces growth factors that encourage adjacent neurons to sprout new neurites
and so form new synaptic contacts. The developing brain has a greater potential for
this kind of neuronal growth than the adult brain (Lindholm, 1994), and this may
provide the substrate for the reconstitution of cortical circuits (Ide et al., 1996).
This mechanism has been demonstrated in a perinatal lesion model in rodents by
Coltman et al. (1995), who observed collateral sprouting in the molecular layer
of the hippocampal dentate gyrus as early as 6 days after a lesion. The developing
brain has another advantage that may help it to repair itself: it already possesses
more neurons and synapses than it needs. During normal development, the brain
produces a large excess of both cells and synaptic connections (Chiron et al., 1992).
For example, the maximal density of synapses in human occipital cortex occurs at
4–12 months, and at this stage is around 150% of adult density (Johnson, 2001).
Many of these “extra” connections are later removed: a process known as synaptic
pruning (Black, 1998). These extra cells and synapses possess neurotrophin recep-
tors, which may allow them to be rescued from pruning by the increased availability
of growth factors released in response to injury (Lindholm, 1994). They could then
be used to reconstitute damaged circuits. Indeed, there is histological evidence of
alteration of cells, synapses and connections in preterm brains (Marin-Padilla, 1996,
1999). Synaptic pruning occurs at different ages in different parts of the brain. The
frontal lobe is generally held to be the latest to mature; through synaptic pruning,
the adult pattern is reached at approximately 18–20 years. In general, the earlier
the injury to the brain occurs, the greater is the availability of alternative neural
substrates to take over the role of any part that is damaged. However, there may
be adverse consequences of this kind of reorganization. Aram and Eisele (1992)
have suggested that one such consequence may be that complex cognitive func-
tions which are acquired later in development suffer through a lack of synaptic
sites.

At the level of neural networks

If a discrete perinatal lesion were to take out a particular cortical module, other areas
of the brain may be able to take over the functions of the lost area. Studies on the
visual system of the cat have demonstrated that this does occur in both adults and
neonates, but functional preservation is more effective and complete in neonates
(Spear, 1996). Much of modern neuroscience rests on the mapping of functions
100 M. Allin, C. Nosarti, L. Rifkin, and R. M. Murray

onto brain areas; this has been successful in many areas (a good example is Broca’s
area). This structure–function relationship, however, is altered in the case of early
brain injury because functions may be remapped onto other undamaged areas of
the brain (Moses and Stiles, 2002). Cioni et al. (2001) used fMRI to study two pairs
of monozygotic twins, where one of each pair had suffered a focal hemispheric
brain injury around the time of birth. They showed that the damaged area’s duties
had been taken over by undamaged areas in the same hemisphere adjacent to the
injury site, and also by areas in the other (intact) hemisphere. Such remapping
of functions can, therefore, occur locally and at more distant sites. The larger the
lesion, the more likely it is that the contralateral hemisphere will take over (Payne
and Lomber, 2001). This also depends on the maturational state of the system that
is lesioned. For example, language can be acquired by the right hemisphere as late
as 9 years if the left hemisphere is damaged (Vargha-Khadem et al., 1997). Bates
et al. (2001) compared language production in brain-injured children and adults.
Adults with left-hemisphere lesions showed severe language impairment and those
with right-hemisphere damage showed disinhibited and “empty” speech. Children
with similar unilateral brain damage showed no language impairment whatever
the side of the lesion (Rauschecker and Marler, 1987). However, this compensation
may not be complete: the areas of the brain that have taken over functions may not
have the right structural specialization to do the job as efficiently, and they may
be diverted from other functions, which may suffer as a result. Some evidence that
developmental compensation may not be complete comes from Weintraub and
Mesulam (1983), who found that individuals born with right-sided brain lesions
showed impairments later in life in spelling, arithmetic, prosody (the ability to vary
speech to convey emotional meanings), and social interactions. This adds to the
evidence that developmental plasticity may be a double-edged sword.

At the level of connectivity

If functions are remapped within the brain then, of necessity, the patterns of cor-
tical wiring – the white matter – must be altered. Altered connectivity may thus
be a secondary effect of particular neural networks being in a different location.
However, it is also possible for changes in connectivity to be a primary adaptation to
injury that may help to subserve effective performance. That is, rather than being
remapped, functions may be rerouted (Kolk, 2000). The example of language is
a good theoretical model of rerouting, as different pathways could be employed
in order to achieve the same results (Neville et al., 1993). For instance, sentence
comprehension could be performed by using a syntactic (structurally based) or
semantic (based on order and meaning of words) route, and reading aloud could
be performed using lexical or non-lexical processes. Functional rerouting could
be a consequence of two processes: automatic or strategic selection. In automatic
101 Brain plasticity and long-term function

selection, it is hypothesized that processing occurs in parallel along several routes

and output competition exists between the routes. According to this model, if
the route that normally secures output is impaired, another route takes its place.
In strategic selection, the individual makes use of alternative information (e.g.
from another sensory modality) (Finney et al., 2001) in carrying out the func-
tion of the damaged area. Studies of reading aloud in dyslexic children (Hendrix
and Kolk, 1996) and sentence production in agrammatic subjects (e.g., Kim and
Thompson, 2000) have found evidence for strategic rerouting. The structural sub-
strate underlying this rerouting is the white matter. Myelination of white matter
begins around term and is not fully complete until well into adult life (Benes et al.,
1994; Girard et al., 1991). This late maturation may allow for rerouting plasticity to
occur.

The limits of brain plasticity

The results of the UCHL study indicate that brain injuries acquired around the
time of birth in VPT babies cause abnormalities of brain structure that are still
detectable in adolescence. Despite this, the level of functioning of those individuals
who escape major physical disability is often not grossly impaired. In addition, it is
not easy to predict patterns of functional deficit from abnormalities of brain struc-
ture in these individuals. This is most likely to be the result of developmental neural
plasticity operating on any, or all, of the levels described above. This reorganization
seems to have spared functions; however, there are hints that it may predispose to
problems later in life. The animal studies mentioned above (Goldman, 1971; Sams-
Dodd et al., 1997) suggested that a brain lesion may remain relatively ‘silent’ until
later in life, only causing functional compromise when the neural system involved
reaches maturity. Some researchers have proposed similar mechanisms in humans
(Anderson et al., 1999; Rankin et al., 1981; Weinberger, 1987). There is evidence
that the prevalence of neuromotor and cognitive impairments increases with age
in VPT individuals (Palfrey et al., 1987; Roth et al., 1994). It will be important to
follow up this group into adulthood to determine whether they are at risk of late
complications of their brain injuries, and if so, what could be done to mitigate the
effects.

Conclusions
Plasticity is a fundamental property of vertebrate nervous systems. During devel-
opment, a high degree of plasticity allows the nervous system to be moulded to
best fit its environment, and this also enables the brain to compensate for injuries
that would cause permanent loss of function in an adult. With increasing age, this
102 M. Allin, C. Nosarti, L. Rifkin, and R. M. Murray

capacity for plasticity begins to decline, with the eventual relative “crystallization”
of brain and behavior, which allows the adult to display stable, adaptive behaviors.
The enormous resilience of the young brain to injury is demonstrated by individ-
uals born preterm, in whom functional impairments are relatively mild compared
with the abnormalities of brain structure that they show. We have drawn attention
to evidence that structural reorganization and remapping or rerouting of functions
has occurred in the brains of these individuals. Given that important events of brain
maturation and development continue into adulthood, there is the possibility that
these structural changes may adversely affect function later in life. There is some
evidence for this from animal models, and the severity of certain impairments in
the preterm population seems to increase with age. It will be important to continue
to follow up preterm individuals as they enter adulthood and to determine which
factors are associated with poor outcome. There may be ways of using the residual
plasticity of the child or even early adult brain to compensate for the late effects of
early brain lesions.

REFERENCES

Allin, M., Murray, R. M. (2002). Schizophrenia: a neurodevelopmental or neurodegenerative

disorder? Curr Opin Psychiatry 15: 9–15.
Allin, M., Matsumoto, H., Santhouse, A. M., et al. (2001). Cognitive and motor function and the
size of the cerebellum in adolescents born very preterm. Brain 124: 60–66.
Anderson, S. W., Bechara, A., Damasio, H., Tranel, D., Damasio, A. R. (1999). Impairment of
social and moral behavior related to early damage in human prefrontal cortex. Nat Neurosci
11: 1032–1037.
Aram, D. M., Eisele, J. A. (1992). Plasticity and recovery of higher cognitive function follow-
ing early brain damage. In Handbook of Neuropsychology, Vol. 6, ed. J. Boller, J. Grafman.
Amsterdam: Elsevier, pp. 73–91.
Back, S. A., Luo, N. L., Borenstein, N. S., et al. (2001). Late oligodendrocyte precursors coincide
with the developmental window of vulnerability for human perinatal white matter injury.
J Neurosci 21: 1302–1312.
Back, S. A., Han, B. H., Luo, N. L., et al. (2002). Selective vulnerability of late oligodendrocyte
progenitors to hypoxia-ischaemia. J Neurosci 22: 455–463.
Bates, E., Reilly, J., Wulfeck, B., et al. (2001). Differential effects of unilateral lesions on language
production in children and adults. Brain Lang 79: 223–265.
Benes, F. M., Turtle, M., Khan, Y., Farol, P. (1994). Myelination of a key relay zone in the hippocam-
pal formation occurs in the human brain during childhood, adolescence, and adulthood.
Arch Gen Psychiatry 51: 477–484.
Black, J. E. (1998). How a child builds its brain: some lessons from animal studies of neural
plasticity. Prevent Med 27: 168–171.
103 Brain plasticity and long-term function

Cannon, M., Jones, P. B., Murray, R. M. (2002). Obstetric complications and schizophrenia:
historical and meta-analytical review. Am J Psychiatry 159: 1080–1092.
Cannon-Spoor, H. E., Potkin, S. G., Wyatt, K. J. (1982). Measurement of premorbid adjustment
in chronic schizophrenia. Schizophr Bull 8: 470–484.
Castro-Caldas, A, Petersson, K. M., Reis, A., Stone-Elander, S., Ingvar, M. (1998). The illiterate
brain: learning to read and write during childhood influences the structural organisation of
the brain. Brain 121: 1056–1063.
Chiron, C., Raynaud, C., Maziere, B., et al. (1992). Changes in regional cerebral blood flow during
brain maturation in children and adolescents. J Nucl Med 33: 696–703.
Cioni, G., Montanaro, D., Tosetti, M., Canapicchi, R., Ghelarduci, B. (2001). Reorganisation of
the sensorimotor cortex after early focal brain lesion: a functional MRI study in monozygotic
twins. Neuroreport 12: 1335–1340.
Coltman, B. W., Earley, E. M., Shahar, A., Dudek, F. E., Ide, C. F. (1995). Factors influencing mossy
fiber collateral sprouting in organotypic slice cultures of neonatal mouse hippocampus.
J Comp Neurol 362: 209–222.
Costello, A. M. de L., Hamilton, P. A., Baudin, J., et al. (1988). Prediction of neurodevelopmental
impairment at four years from brain ultrasound appearance of very preterm infants. Dev
Med Child Neurol 30: 711–722.
du Plessis, A. J., Volpe, J. J. (2002). Perinatal brain injury in the preterm and term newborn. Curr
Opin Neurol 15: 151–157.
Elbert, T., Heim, S., Rockstroh, B. (2001). Neural plasticity and development. In Handbook of
Developmental Cognitive Neuroscience, ed. C. A. Nelson, M. Luciana. Cambridge, MA: MIT
Press, pp. 191–202.
Finney, E. M., Fine, I., Dobkins, K. R. (2001). Visual stimuli activate auditory cortex in the deaf.
Nat Neurosci 4: 1171–1173.
Frith, U. (1998). Literally changing the brain. Brain 121: 1011–1012.
Girard, N., Raybaud, C., du Lac, P. (1991). MRI study of brain myelination. J Neuroradiol 18:
291–307.
Goldman, P. S. (1971). Functional development of the prefrontal cortex in early life and the
problem of plasticity. Exp Neurol 32: 366–387.
Goyen, T-A., Lui, K., Woods, R. (1998). Visual–motor, visual–perceptual and fine motor outcomes
in very-low-birth-weight children at 5 years. Dev Med Child Neurol 40: 76–81.
Grimshaw, G. M., Adelstein, A., Bryden, M. P., MacKinnon, G. E. (1998). First-language acqui-
sition in adolescence: evidence for a critical period for verbal language development. Brain
Lang 63: 237–255.
Hadders-Algra, M., Huisjes, H. J., Touwen, B. C. L. (1988). Perinatal risk factors and minor
neurological dysfunction: significance for behaviour and school achievement at 9 years. Dev
Med Child Neurol 30: 482–491.
Hall, A., McLeod, A., Counsell, C., Thomson, L., Mutch, L. (1995). School attainment, cognitive
ability and motor function in a total Scottish very-low-birthweight population at 8 years: a
controlled study. Dev Med Child Neurol 37: 1037–1050.
Hecaen, H. (1983). Acquired aphasia in childhood: revisited. Neuropsychologia 21: 581–587.
104 M. Allin, C. Nosarti, L. Rifkin, and R. M. Murray

Hendrix, A. W., Kolk, H. H. J. (1996). Strategic control in developmental dyslexia. Cogn

Neuropsych 14: 321–366.
Hoon, A. H. (1995). Neuroimaging in the high risk infant: relationship to outcome. J Perinatol
15: 389–394.
Hope, P. L., Gould, S. J., Howard, S., et al. (1988). Precision of ultrasound diagnosis of patho-
logically verified lesions in the brain of very preterm infants. Dev Med Child Neurol 30:
457–471.
Huh, J., Williams, H. G., Burke, J. R. (1998). Development of bilateral motor control in children
with developmental coordination disorders. Dev Med Child Neurol 40: 474–484.
Ide, C. F., Scripter, J. L., Coltman, B. W., et al. (1996). Cellular and molecular correlates to plas-
ticity during recovery from injury in the developing mammalian brain. Prog Brain Res 108:
365–377.
Johnson, M. H. (2001). Functional brain development in humans. Nat Rev Neurosci 2: 475–483.
Johnston, M. V. (1998). Selective vulnerability in the neonatal brain. Ann Neurol 44: 155–156.
Jones, S. M., Zigler, E. (2002). The Mozart effect: not learning from history. J Appl Dev Psychol
23: 355–372.
Kennard, M. A. (1936). Age and other factors in motor recovery from precentral lesions in
monkeys. Am J Physiol 115: 138–146.
Kim, M., Thompson, C. K. (2000). Patterns of comprehension and production of nouns and
verbs in agrammatism: implications for lexical organization. Brain Lang 74: 1–25.
Kinney, H. C., Brody, B. A., Kloman, A. S., Gilles, F. H. (1988). Sequence of central nervous
system myelination in human infancy. II. Patterns of myelination in autopsied infants.
J Neuropathol Exp Neurol 47: 217–234.
Kolb, B. (1989). Brain development, plasticity and behaviour. Am Psychol 44: 1203–1212.
Kolb, B., Whishaw, I. Q. (1989). Plasticity in the neocortex: mechanisms underlying recovery
from early brain damage. Prog Neurobiol 32: 235–276.
Kolk, H. H. J. (2000). Multiple route plasticity. Brain Lang 71: 129–131.
Kuban, K., Sanocka, U., Leviton, A., et al. (1999). White matter disorders of prematurity:
association with intraventricular hemorrhage and ventriculomegaly. The Developmental
Epidemiology Network. J Pediatr 134: 539–546.
Kuchna, I. (1994). Quantitative studies of human newborns’ hippocampal pyramidal cells after
perinatal hypoxia. Folia Neuropathol 32: 9–16.
Landau, Y. E., Gross-Tsur, V., Auerbach, J. G., van der Meere, J., Shalev, R. S. (1999). Attention-
deficit hyperactivity disorder and developmental right-hemisphere syndrome: congruence
and incongruence of cognitive and behavioral aspects of attention. J Child Neurol 14: 299–
303.
Lewis, S. W., Murray, R. M. (1987). Obstetric complications, neurodevelopmental deviance and
risk of schizophrenia. J Psychiatr Res 21: 413–422.
Lewis, S. W., Owen, M. J., Murray, R. M. (1989). Obstetric complications and schizophre-
nia: methodology and mechanisms. Schizophrenia: Scientific Progress, ed. S. Schutz, C. A.
Tomminga. New York: Oxford University Press, pp. 56–68.
Lindholm, D. (1994). Role of neurotrophins in preventing glutamate induced neuronal cell death.
J Neurol 242: S16–S18.
105 Brain plasticity and long-term function

Linkenhoker, B. A., Knudsen, E. I. (2002). Incremental training increases the plasticity of the
auditory space map in adult barn owls. Nature 419: 293–296.
Luoma, L., Herrgard, E., Martikainen, A. (1998). Neuropsychological analysis of the visuomotor
problems in children born preterm at ≤ 32 weeks of gestation: a 5 year prospective follow-up.
Dev Med Child Neurol 40: 21–30.
Mallard, E. C., Rehn, A., Rees, S., Tolcos, M., Copolov, D. (1999). Ventriculomegaly and reduced
hippocampal volume following intrauterine growth-restriction: implications for the aetiol-
ogy of schizophrenia. Schizophr Res 40: 11–21.
Marin-Padilla, M. (1996). Developmental neuropathology and impact of perinatal brain damage.
I. Hemorrhagic lesions of neocortex. J Neuropathol Exp Neurol 55: 758–773.
(1997). Developmental neuropathology and impact of perinatal brain damage. II: white matter
lesions of the neocortex. J Neuropathol Exp Neurol 56: 219–235.
(1999). Developmental neuropathology and impact of perinatal brain damage. III: gray matter
lesions of the neocortex. J Neuropathol Exp Neurol 58: 407–429.
Moses, P., Stiles, J. (2002). The lesion methodology: contrasting views from adult and child
studies. Dev Psychobiol 40: 266–277.
Msall, M. E., Buck, G. M., Schisterman, E. F., et al. (1998). Social and biomedical risks for 8 to 10
year educational outcomes of children born with extreme prematurity and without major
disability. AACPDM Abstracts. Dev Med Child Neurol Suppl H: 27.
Murray, R. M., Lewis, S. W. (1987). Is schizophrenia a neurodevelopmental disorder? Br Med J
295: 681–682.
Neville, H. J. (1993). Neurobiology of cognitive and language processing: Effects of early experi-
ence. In Brain Development and Cognition: A Reader, ed. M. H. Johnson. Oxford: Blackwell,
pp. 424–448.
Nosarti, C., Rifkin, L., Rushe, T. M., et al. (2001a). Corpus callosum size in adolescents who were
born very preterm. Pediatr Res Special Suppl 50: 15A.
Nosarti, C., Allin, M., Al-Asady, M., et al. (2001b). Behavioural and cognitive consequences of
caudate pathology in adolescents born very preterm. Neuroimage, 13: 339.
Nosarti, C., Al-Asady, M. H. S., Frangou, S., et al. (2002). Adolescents who were born very preterm
have decreased brain volumes. Brain 125: 1616–1623.
Nosarti, C., Rubia, K., Frearson, S., Rifkin, L., Murray, R. M. (2004). Altered neuronal organisation
of the brain of adolescents born very preterm during response inhibition. Schizophr Res
60(Special Suppl.): 230.
Palfrey, J. S., Singer, J. D., Walker, D. A., Butler, J. A. (1987). Early identification of children’s
special needs: a study in five metropolitan communities. J Pediatr 111: 651–659.
Paneth, N., Rudelli, R., Kazam, E., Monte, W. (1994). Brain Damage in the Preterm Infant.
Cambridge: MacKeith Press/Cambridge University Press.
Pantev, C., Oostenveld, R., Engelien, A., et al. (1998). Increased auditory cortical representation
in musicians. Nature 392: 811–814.
Payne, B. R., Lomber, S. G. (2001). Timeline: reconstructing functional systems after lesions of
cerebral cortex. Nat Rev Neurosci 2: 911–919.
Rankin, J. M., Aram, D. M., Horwitz, S. J. (1981). Language ability in right and left hemiplegia
children. Brain Lang 14: 292–306.
106 M. Allin, C. Nosarti, L. Rifkin, and R. M. Murray

Rauschecker, J. P., Marler, P. (1987). Imprinting and Cortical Plasticity: Comparative Aspects of
Sensitive Periods. New York: Wiley.
Rooney, M., Allin, M., Rifkin, L., et al. (2001). Comparison of psychopathology in prematurely
born adults from a 1979–1981 cohort, with full term born adults. Schizophr Res 49: 41.
Rosenbloom, L., Sullivan, P. B. (1996). The nutritional and neurodevelopmental consequences
of feeding difficulties in disabled children. In Clinics in Developmental Medicine, No. 104;
Feeding the Disabled Child, ed. P. B. Sullivan, L. Rosenbloom. Cambridge, UK: MacKeith
Press/Cambridge University Press, pp. 33–39.
Roth, S. C., Baudin, J., McCormick, D. C., et al. (1993). Relation between ultrasound appearance
of the brain of very preterm infants and neurodevelopmental impairment at eight years.
Dev Med Child Neurol 35: 755–768.
Roth, S. C., Baudin, J., Pezzani-Goldsmith, M., et al. (1994). Relation between neurodevelop-
mental status of very preterm infants at one and eight years. Dev Med Child Neurol 36:
1049–1062.
Rushe, T. M., Woodruff, P. W. R., Bullmore, E. B., et al. (1999). Lateralisation of language function
in adults born very preterm. Magn Reson Mater Phys, Biol Med 8(Suppl 1): 82.
Rushe, T. M., Rifkin, L., Stewart, A. L., et al. (2001). Neuropsychological outcome at adolescence
of very preterm birth and its relation to brain structure. Dev Med Child Neurol 43: 226–33.
Rutter, M., Tizard, J., Whitmore, K. (1981). Education, Health and Behaviour. Melbourne: Kreiger.
Sams-Dodd, F., Lipska, B. K., Weinberger, D. R. (1997). Neonatal lesions of the rat ventral
hippocampus result in hyperlocomotion and deficits in social behaviour in adulthood.
Psychopharmacology 132: 303–310.
Santhouse, A. M., Ffytche, D. H., Howard, R. J., et al. (2002). The functional significance of
perinatal corpus callosum damage: an fMRI study in young adults. Brain 125: 1782–1792.
Schachar, R., Rutter, M., Smith, A. (1981). The characteristics of situationally and pervasively
hyperactive children: implications for syndrome definition. J Child Psychol Psychiatry 22:
375–392.
Schmahmann, J. D., Sherman, J. C. (1998). The cerebellar cognitive–affective syndrome. Brain
121: 561–579.
Semrud-Clikeman, M., Steingard, R. J., Filipek, P., et al. (2000). Using MRI to examine brain–
behavior relationships in males with attention deficit disorder with hyperactivity. J Am Acad
Child Adolesc Psychiatry 39: 477–484.
Shewmon, D. A., Holmes, G. L., Byrne, P. A. (1999). Consciousness in congenitally decorticate
children: developmental vegetative state as a self-fulfilling prophecy. Dev Med Child Neurol
41: 364–374.
Snider, L. M. (1998). Preschool performance skills of extremely low birth weight children.
AACPDM Abstracts. Dev Med Child Neurol Suppl H: 27.
Sohma, O., Mito, T., Mizuguchi, M., Takashima, S. (1995). The prenatal age critical for the
development of the pontosubicular necrosis. Acta Neuropathol 90: 7–10.
Sowell, E. R., Thompson, P. M., Holmes, C. J., Jernigan, T. L., Toga, A. W. (1999). In vivo evidence
for post-adolescent brain maturation in frontal and striatal regions. Nat Neurosci 2: 859–
861.
Spear, P. D. (1996). Neural plasticity after brain damage. Prog Brain Res 108: 391–408.
107 Brain plasticity and long-term function

Stewart, A. L., Thorburn, R. J., Hope, P. L., et al. (1983). Ultrasound appearance of the brain in
very preterm infants and neurodevelopmental outcome at 18 months of age. Arch Dis Child
58: 598–604.
Stewart, A. L., Costello, A. M., Hamilton, P. A., et al. (1989). Relationship between neuro-
developmental status of very preterm infants at 1 and 4 years. Dev Med Child Neurol 31:
756–765.
Stewart, A. L., Rifkin, L., Amess, P. N., et al. (1999). Brain structure and neurocognitive and
behavioural function in adolescents who were born very preterm. Lancet 353: 1653–1657.
Tin, W., Wariyar, U., Hey, E. (1997). Changing prognosis for babies of less than 28 weeks’ gestation
in the north of England between 1983 and 1994. Br Med J 314: 107–111.
Vargha-Khadem, F., Carr, L. J., Isaacs, E., et al. (1997). Onset of speech after left hemispherectomy
in a nine-year-old boy. Brain 120: 159–182.
Volpe, J. (1998). Neurologic outcome of prematurity. Arch Neurol 55: 297–300.
Weinberger, D. R. (1987). Implications of normal brain development for the pathogenesis of
schizophrenia. Arch Gen Psychiatry 44: 660–669.
Weintraub, S., Mesulam, M. M. (1983). Developmental learning disabilities of the right hemi-
sphere: emotional, interpersonal, and cognitive components. Arch Neurol 40: 463–468.
Wiesel, T. N., Hubel, D. H. (1963). Single-cell responses in striate cortex of kittens deprived of
vision in one eye. J Neurophysiol 26: 1003–1017.
Part II

Etiological factors
 6

Do degenerative changes operate across

diagnostic boundaries? The case for
glucocorticoid involvement in major
psychiatric disorders
Carmine M. Pariante1 and David Cotter2
1
Institute of Psychiatry, King’s College, London, UK
2
Royal College of Surgeons, Dublin, Ireland

As it was initially proposed, the developmental theory of schizophrenia suggested

that an early insult to the developing brain could result in cerebral changes that
ultimately manifest as psychosis (Lewis and Murray, 1987; Weinberger, 1987).
The success of this theory has been in its capacity to explain many of the known
features of schizophrenia. Thus, the season of birth effect, the reported excess of
obstetric complications, minor physical anomalies, dermatoglyphic abnormalities,
and childhood developmental impairments may all be successfully and usefully
explained by the influence of environmental and epigenetic factors on normal fetal
brain development. Crucial supportive evidence for the theory was given by stud-
ies showing that classical degenerative brain changes were largely absent, by the
presence of microscopic brain changes indicative of abnormal brain development,
and by the absence of clear evidence for progressive ventricular dilatation among
schizophrenia subjects.
However, these latter findings are no longer secure. For example, the absence
of excess cortical gliosis is no longer accepted as indicating a lack of peri- or
postnatal cerebral inflammation, and firm cytoarchitectural evidence of abnormal
brain development has yet to be consistently presented (Harrison, 1999). Further-
more, some neuroimaging studies have shown evidence for progressive ventricular
enlargement over time and others have shown that hippocampal volume reductions
occur during the phase of a first psychotic episode (Lawrie et al., 2000; Pantellis
et al., 2000). Reversibility of reductions in superior temporal gyrus volume has also
been described during the first episode of psychosis (Keshavan et al., 1998), and pro-
gressive cortical reductions are observed in subjects with early-onset schizophrenia
(Rappaport et al., 1999). Therefore, evidence is accumulating that supports the
view that plastic changes in macroscopic cytoarchitecture may occur with illness.
Neurodevelopment and Schizophrenia, ed. Matcheri S. Keshavan et al. Published by Cambridge University
Press.

C Cambridge University Press 2004.

111
112 C. M. Pariante and D. Cotter

The microscopic basis of these cortical volume reductions are not known, but
clues are provided by recent neuropathological studies that have shown reductions
in cortical neuronal size, dendritic complexity, and synaptic proteins in schizophre-
nia (Harrison, 1999). Smaller neurons with fewer and less-elaborate branches would
result in a diminution in the amount of neuropil, more compacted cells, and
thus increased neuronal density. It is certainly feasible to suggest that these alter-
ations underlie the regional cortical volume reductions observed in schizophrenia.
The current challenge, however, is to understand the mechanism underlying these
changes. The term atrophic is preferred over degenerative because evidence of cell
death and of a gliotic response so characteristic of degenerative diseases is absent.
However, regardless of what term is used, the findings necessitate a re-evaluation of
the neurodevelopmental hypothesis of schizophrenia as it is currently understood.

Evidence for a common neuropathology in major depression,

bipolar disorder, and schizophrenia
It is possible that the microscopic neuropathological changes described in
schizophrenia may be vulnerability factors for schizophrenia. Alternatively, they
could be an intrinsic component or a consequence of the illness. Neuropatholog-
ical studies cannot tell which interpretation is correct. However, there are some
important similarities in the neuropathology of schizophrenia, major depressive
disorder (MDD) and bipolar disorder (BPD), which suggest that a common pro-
cess of change is involved in each disorder.
Macroscopic neuroanatomical investigations of brain pathology in schizophre-
nia, BPD, and MDD show differences that are generally quantitative rather than
qualitative. For example, ventricular dilatation and reduced hippocampal and
frontal brain volumes are seen in schizophrenia, but they are also present to a
lesser degree in MDD and BPD (McCarley et al., 1999). The single main departure
from this pattern is that the volume of the amygdala may be specifically enlarged in
BPD (Altshuler et al., 1998), possibly because of drug treatment. Microscopically,
reductions in dendritic spine density (Rosoklija et al., 2000), neuronal size (Cotter
et al., 2001; Rajkowska et al., 1999), and synaptic proteins (Eastwood and Harrison,
2001) have been described in mood disorders as well as in schizophrenia (Harrison,
1999). More recently, it has become apparent that glial cell loss may be a feature of
MDD, BPD, and schizophrenia (Cotter et al, 2001; Rajkowska et al., 1999) depend-
ing, possibly, on the presence of coexisting affective symptoms and on which region
of the brain is investigated.
This similar pattern of changes in cortical cellular architecture in schizophrenia
and mood disorders suggests that a common pathophysiology may underlie aspects
of these psychiatric diseases. What aspects of illness common to MDD, BPD, and
113 Glucocorticoids in major psychiatric disorders

schizophrenia could cause changes in keeping with the known cellular changes
described above? Glucocorticoid-related neurotoxicity is one such candidate.

A role for glucocorticoids in the neuronal changes of mood

disorders and schizophrenia?
There are several other lines of investigation that support the view that
glucocorticoid-related neurotoxicity may be implicated in depression and
schizophrenia. First, in vitro investigations have shown that high levels of gluco-
corticoid hormones result in reduced neuronal volume and dendritic arborization
(Sapolsky, 2000), and these latter changes have been observed in both disorders.
Second, elevated plasma glucocorticoid levels are associated with (largely reversible)
hippocampal volume reductions in MDD, post-traumatic stress disorder (PTSD),
Cushing’s disease, and normal aging, and such reductions have been observed in
the phase of a first psychotic episode (Lawrie et al., 2000; Pantellis et al., 2000).
Third, the functional effect of glucocorticoids on reducing hippocampal glial cell
activation and proliferation (Crossin et al., 1997) mirrors the glial deficit observed
in MDD, BPD, and possibly schizophrenia. Consequently, the glial deficit found in
these disorders may also relate to glucocorticoid effects.

Hypothalamic–pituitary–adrenal axis abnormalities in major depression

There is substantial evidence that hyperactivity of the hypothalamic–pituitary–
adrenal (HPA) axis is involved in the pathogenesis of mood disorder (Pariante, 2003;
Pariante and Miller, 2001). A significant percentage of depressed patients have been
shown to hypersecrete cortisol, the endogenous adrenal glucocorticoid in humans,
as manifested by increased 24-hour urinary free cortisol and elevated plasma and
cerebrospinal fluid (CSF) concentrations of cortisol (Dinan, 1996; Holsboer, 2000;
McAllister-Williams et al., 1998; McQuade and Young, 2000; Nemeroff, 1996).
Adrenal hypertrophy and increased pituitary volume has also been described in
these patients, and these findings have also been considered a marker of HPA axis
activation (Axelson et al., 1992; Nemeroff, 1996). Hyperactivity of the HPA axis in
major depression is driven by the hypersecretion of corticotropin-releasing factor
(CRF) in the hypothalamus (Holsboer, 2000; Nemeroff, 1996; Owens and Nemeroff,
1993). Most notable in this regard are the increased CSF concentrations of CRF
consistently found in drug-free depressed patients, and the decreased number of
CRF receptors in the frontal cortex of suicide victims (Banki et al., 1987; Heuser
et al., 1998; Nemeroff, 1996; Nemeroff et al., 1984; Owens and Nemeroff, 1993).
Moreover, depressed patients exhibit increased CRF messenger RNA (mRNA) and
protein in the paraventricular nucleus of the hypothalamus (postmortem samples)
114 C. M. Pariante and D. Cotter

(Purba et al., 1995; Raadsheer et al., 1995). Finally, a number of studies have provided
evidence that CRF may play a role not only in the endocrinopathy of depression but
also in the behavioral signs and symptoms of the disorder. Indeed, the behavioral
effects of CRF after direct central nervous system administration are remarkably
similar to the signs and symptoms of major depression (decreased libido, decreased
appetite, psychomotor alterations, and disturbed sleep).
These increased levels of CRF in the hypothalamus are related, at least in part, to
altered feedback inhibition by endogenous glucocorticoids (de Kloet et al., 1998;
Holsboer, 2000; McQuade and Young, 2000; Pariante and Miller, 2001; Pariante
et al., 2002). In fact, glucocorticoids interact with their receptors (the mineralocor-
ticoid receptor [MR] and the glucocorticoid receptor [GR]) in HPA axis tissues,
where they are responsible for feedback inhibition of the secretion of adrenocor-
ticotropic hormone (ACTH) from the pituitary and CRF from the hypothalamus
(de Kloet et al., 1998; McEwen, 2000; Nemeroff, 1996; Young et al., 1998). The MR
has a high affinity for endogenous glucocorticoids and is believed to play a role
in the regulation of circadian fluctuations of these hormones. In contrast to the
MR, the GR has a lower affinity for endogenous glucocorticoids and is believed to
be more important in the regulation of the response to stress, when endogenous
levels of glucocorticoids are high (de Kloet et al., 1998). Consistent with the fact
that patients with major depression exhibit impaired HPA negative feedback in
the context of elevated circulating levels of cortisol, when the negative feedback is
largely mediated by the GR, many studies have demonstrated that GR-mediated
feedback inhibition is impaired in major depression. Depressed patients have non-
suppression of cortisol secretion following dexamethasone in the dexamethasone
suppression test (DST), non-suppression of ACTH secretion following hydrocorti-
sone (fast-feedback test), and lack of inhibition of ACTH responses to CRF following
dexamethasone pretreatment (DEX/CRF test) (Heuser et al., 1994, 1996; Holsboer,
2000; Nemeroff, 1996; Ribeiro et al., 1993; Young et al., 1991). The only study that
specifically looked at MR-mediated negative feedback in depression found that this
pathway is intact (or possibly oversensitive) in these patients (Young et al., 2003).
In further support of the suggestion that patients with major depression exhibit
impaired GR-mediated HPA negative feedback, a number of studies have demon-
strated that GR function is also reduced in other tissues of depressed patients, as
shown by a decreased sensitivity to the effects of glucocorticoids on immune and
metabolic functions (Pariante and Miller, 2001). While there is no consistent evi-
dence of GR expression changes in blood mononuclear cells or fibroblasts from
depressed patients (Pariante and Miller, 2001), three studies that have examined
postmortem brains have found reduced corticosteroid receptor expression (Lopez
et al., 1998; Webster et al., 2002; Xing et al., 2004).
115 Glucocorticoids in major psychiatric disorders

Hypothalamic–pituitary–adrenal axis abnormalities in schizophrenia

In contrast to depression, the evidence linking HPA hyperactivity and GR dysfunc-
tion with schizophrenia is less clear. Sachar et al. (1970) suggested that patients
experiencing a first-episode psychosis were more likely to present with HPA abnor-
malities, because of the distress associated with its “dramatic and ego-dystonic”
nature. Several studies have confirmed that patients who are in the acute phase of
a psychotic disorder, with florid symptoms, newly hospitalized, or unmedicated,
have an elevated HPA axis activity, as shown by raised cortisol levels (Sachar et al.,
1970), non-suppression of cortisol secretion by dexamethasone in the DST and in
the DEX/CRF test (Coryell and Tsuang, 1992; Herz et al., 1985; Lammers et al.,
1995), elevated levels of CRF in the CSF (Banki et al., 1987), and abnormal volume
of the pituitary gland (Pariante et al., 2004a). Moreover, cortisol levels are positively
correlated with symptom severity in patients with psychosis or schizotypal person-
ality disorder (Walder et al., 2000; Walker et al., 2001) and negatively correlated
with performance on executive function and memory tasks (Walder et al., 2000).
Prospective studies also showed that most non-suppressor patients convert into
suppressors after a few weeks of antipsychotic treatment (Herz et al., 1985; Tandon
et al., 1991). For example, Herz et al. (1985) described that 11 out of 15 newly hos-
pitalized patients, who were acutely psychotic with “florid psychotic symptoms”
following a relapse, were DST non-suppressor, and that four out of five of these
non-suppressor patients converted to suppressor after a week of treatment. Finally,
patients who are clinically stable and receiving treatment tend to have a normal
HPA axis (Ismail et al., 1998; Tandon et al., 1991).
Interestingly, across studies, patients with schizophrenia seem to have “more”
HPA axis activation than normal individuals, but “less” HPA axis activation than
depressed patients. For example, Sharma et al. (1988) conducted a meta-analysis
comparing the results at the DST in patients with schizophrenia with those in
patients with major depression. They found that patients with schizophrenia had a
19% rate of non-suppression, which was higher than the 7% rate in normal controls
but lower than the 51% rate in patients with major depression. Similarly, Banki et al.
(1987) examined the CSF levels of CRF in controls, depressed patients, and patients
with schizophrenia; while depressed patients had levels of CRF levels that were
markedly elevated (double those of controls), patients with schizophrenia had CRF
levels that were only slightly, but significantly, higher than control. Recently, Webster
et al. (2002) and Xing et al. (2004) found decreased GR and MR mRNA in the frontal
cortex and hippocampus not only of patients with non-psychotic depression but
also in patients with BPD and schizophrenia. It is possible that HPA axis activation
in depression is part of the pathogenesis of the disorder–the increased levels of
CRF in the brain could be responsible for some of the depressive symptoms–while
116 C. M. Pariante and D. Cotter

HPA axis activation in schizophrenia is a temporary phenomenon and reflects a

stress-related activation of the axis. However, even if this was not a primary patho-
genetic event in schizophrenia, HPA axis activation could still have crucial clinical
effects on brain function (Cotter and Pariante, 2002). Hence, there is a need to clar-
ify whether there is HPA hyperactivity at any stage during the course of psychotic
disorders, especially around the time of the onset.

Cortisol: hero or villain?

Despite this evidence supporting the view that the interactions between glucocorti-
coid hormones and the brain may be abnormal in major depression and schizophre-
nia, there remains one big unanswered question: does the fact that depressed
patients have a hyperactive HPA axis actually mean that a lot of cortisol is flooding
their brain, and that the depressive symptoms are a consequence of this putative
“toxic” effect of cortisol? Or is the opposite true: that patients have a hyperactive
HPA axis as a compensatory mechanism because their brain is resistant to the effects
of circulating cortisol? The question is not trivial, especially in our quest for a more
effective treatment. In the first scenario, our recommendation should be the low-
ering of cortisol levels. In the second scenario, our recommendation should be the
administering of more cortisol. The available data are still failing to give a defini-
tive answer on this question, although certainly the evidence does not consistently
support the hypothesis that an excess of cortisol can cause depressive symptoms
and brain micro- and macroscopic changes. First, it is not at all clear whether
the mechanism of neurotoxicity in vivo is mediated through elevated or lowered
levels of cerebral glucocorticoids, for very low levels also have neurotoxic effects
(Pariante, 2003; Sapolsky, 2000). Second, the reduction of GR levels in the brains
of these patients, and the reduction of GR function in patients with depression
and in subjects experiencing chronic stress (Avitsur et al., 2001; Bauer et al., 2000;
Pariante and Miller, 2001) suggests that elevated plasma cortisol levels could rep-
resent a compensatory strategy. Third, recent studies indicate that levels of cortisol
in the brain of humans are regulated by efflux systems at the blood–brain barrier
(de Kloet et al., 1998; Pariante et al. 2001, 2003a,b), and that both GR and the cor-
tisol efflux systems may be influenced by psychotropic drugs (Pariante and Miller
2001; Pariante et al., 1997, 2001, 2003a,b, 2004b). This indicates that peripheral cor-
tisol levels, as often assumed in studies, may not necessarily dictate cerebral levels.
Finally, treatment with GR and MR agonists, including cortisol, has antidepressant
effects in humans (Bouwer et al., 2000), again suggesting that these patients may
not necessarily experience “too much cortisol” in their brain (DeBattista et al.,
2000; Dinan et al., 1997). In reality, whether patients with major depression (or
schizophrenia) have elevated or lowered activation of the GR in the brain is yet to
be fully elucidated.
117 Glucocorticoids in major psychiatric disorders

Conclusions
Evidence is accumulating that there are brain changes occurring during and pos-
sibly after the period of the first acute psychosis: that is, changes that are not
developmental in the traditional sense. Furthermore, these changes are not specific
to schizophrenia, either in terms of macroscopic or microscopic brain structure,
for they are also present, to a generally milder degree, in subjects with mood disor-
der, and they are in keeping with glucocorticoid-related brain changes. These brain
changes may be epiphenomena secondary to stress-related changes in glucocorti-
coid hormones and not primary pathogenetic pathways. Nevertheless, they could
have crucial clinical effects through diminishing neuronal and cortical function
and so complicate recovery from the primary illness. These changes may possibly
be reversed by therapies that protect from glucocorticoid-related neurotoxicity or
which act to promote neuroprotective cell signaling pathways. What is clear, how-
ever, is that the developmental theory of schizophrenia as it was first presented
(Lewis and Murray, 1987; Weinberger, 1987) is challenged, as it alone is insufficient
to explain the neuroanatomy of the disorder. Consequently, it is now important
to understand the illness in terms of both early and late events, which will involve
both developmental and atrophic processes.

REFERENCES

Altshuler, L. L., Bartzokis, G., Grieder, T., Curran, J., Mintz, J. (1998). Amygdala enlargement in
bipolar disorder and hippocampal reduction in schizophrenia: an MRI study demonstrating
neuroanatomic specificity. Arch Gen Psychiatry 55: 663–664.
Avitsur, R., Stark, J. L., Sheridan, J. F. (2001). Social stress induces glucocorticoid resistance in
subordinate animals. Horm Behav 39: 247–257.
Axelson, D. A., Doraiswamy, P. M., Boyko, O. B., et al. (1992). In vivo assessment of pituitary
volume with magnetic resonance imaging and systematic stereology: relationship to dexa-
methasone suppression test results in patients. Psychiatry Res 44: 63–70.
Banki, C. M., Bissette, G., Arato, M., O’Connor, L., Nemeroff, C. B. (1987). CSF corticotropin-
releasing factor-like immunoreactivity in depression and schizophrenia. Am J Psychiatry
144: 873–877.
Bauer, M. E., Vedhara, K., Perks, P., et al. (2000). Chronic stress in caregivers of dementia patients
is associated with reduced lymphocyte sensitivity to glucocorticoids. J Neuroimmunol 103:
84–92.
Bouwer, C., Claassen, J., Dinan, T. G., Nemeroff, C. B. (2000). Prednisone augmentation in
treatment-resistant depression with fatigue and hypocortisolaemia: a case series. Depress
Anxiety 12: 44–50.
Coryell, W., Tsuang, D. (1992). Hypothalamic–pituitary–adrenal axis hyperactivity and psychosis:
recovery during an 8-year follow-up. Am J Psychiatry 149: 1033–1039.
Cotter, D., Pariante, C. M. (2002). Stress and the progression of the developmental hypothesis of
schizophrenia. Br J Psychiatry 181: 363–365.
118 C. M. Pariante and D. Cotter

Cotter, D., Mackay, D., Landau, S., Kerwin, R., and Everall, I. (2001). Glial cell loss and reduced
neuronal size in the anterior cingulate cortex in major depressive disorder. Arch Gen Psychi-
atry 58: 545–553.
Crossin, K. L., Tai, M. H., Krushel, L. A., Mauro, V. P., Edelman, G. M. (1997). Glucocorticoid
receptor pathways are involved in the inhibition of astrocyte proliferation. Proc Natl Acad
Sci USA 94: 2687–2692.
de Kloet, E. R., Vreugdenhil, E., Oitzl, M. S., Joels, M. (1998). Brain corticosteroid receptor
balance in health and disease. Endocr Rev 19: 269–301.
DeBattista, C., Posener, J. A., Kalehzan, B. M., Schatzberg, A. F. (2000). Acute antidepressant effects
of intravenous hydrocortisone and CRH in depressed patients: a double-blind, placebo-
controlled study. Am J Psychiatry 157: 1334–1337.
Dinan, T. G. (1996). Noradrenergic and serotonergic abnormalities in depression: stress-induced
dysfunction? J Clin Psychiatry 57(Suppl 4): 14–18.
Dinan, T. G., Lavelle, E., Cooney, J., et al. (1997). Dexamethasone augmentation in treatment-
resistant depression. Acta Psychiatr Scand 95: 58–61.
Eastwood, S. L., Harrison, P. J. (2001). Synaptic pathology in the anterior cingulate cortex in
schizophrenia and mood disorders. A review and a western blot study of synaptophysin,
GAP-43, and the complexins. Brain Res Bull 55: 519–578.
Harrison, P. J. (1999). The neuropathology of schizophrenia: a critical review of the data and
their interpretation. Brain 122: 593–624.
Herz, M. I., Fava, G. A., Molnar, G., Edwards, L. (1985). The dexamethasone suppression test in
newly hospitalized schizophrenic patients. Am J Psychiatry 142: 127–129.
Heuser I. J., Yassouridis, A., Holsboer, F. (1994). The combined dexamethasone/CRH test: a
refined laboratory test for psychiatric disorders. J Psychiatr Res 28: 341–356.
Heuser, I. J., Schweiger, U., Gotthardt, U., et al. (1996). Pituitary–adrenal-system regulation and
psychopathology during amitriptyline treatment in elderly depressed patients and normal
comparison subjects. Am J Psychiatry 153: 93–99.
Heuser, I. J., Bissette, G., Dettling, M., et al. (1998). Cerebrospinal fluid concentrations of
corticotropin-releasing hormone, vasopressin, and somatostatin in depressed patients and
healthy controls: response to amitriptyline treatment. Depress Anxiety 8: 71–79.
Holsboer, F. (2000). The corticosteroid receptor hypothesis of depression. Neuropsychopharma-
cology 23: 477–501.
Ismail, K., Murray, R. M., Wheeler, M. J., O’Keane, V. (1998). The dexamethasone suppression
test in schizophrenia. Psychol Med 28: 311–317.
Keshavan, M. S., Haas, G. L., Kahn, C. E., et al. (1998). Superior temporal gyrus and the course
of early schizophrenia: progressive static or reversible? J Psychiatr Res 32: 161–167.
Lammers, C. H., Garcia-Borreguero, D., Schmider, J., et al. (1995). Combined dexamethasone/
corticotropin-releasing hormone test in patients with schizophrenia and in normal controls:
II. Biol Psychiatry 38: 803–807.
Lawrie, S. M., Whalley, H., Byrne, M., et al. (2000). Brain structure change and psychopathology
in subjects at high risk of schizophrenia. Schizophr Res 41: 11.
Lewis, S. W., Murray, R. M. (1987). Obstetric complications, neurodevelopmental deviance, and
risk of schizophrenia. J Psychiatr Res 21: 413–421.
119 Glucocorticoids in major psychiatric disorders

Lopez, J. F., Chalmers, D. T., Little, K. Y., Watson, S. J. (1998). A. E. Bennett Research Award.
Regulation of serotonin1A , glucocorticoid, and mineralocorticoid receptor in rat and human
hippocampus: implications for the neurobiology of depression. Biol Psychiatry 43: 547–573.
McAllister-Williams, R. H., Ferrier, I. N., Young, A. H. (1998). Mood and neuropsychological
function in depression: the role of corticosteroids and serotonin. Psychol Med 28: 573–584.
McCarley, R. W., Wible, C. G., Frumin, M., et al. (1999). MRI anatomy of schizophrenia. Biol
Psychiatry 45: 1099–1119.
McEwen, B. S. (2000). The neurobiology of stress: from serendipity to clinical relevance. Brain
Res 886: 172–189.
McQuade, R., Young, A. H. (2000). Future therapeutic targets in mood disorders: the glucocor-
ticoid receptor. Br J Psychiatry 177: 390–395.
Nemeroff, C. B. (1996). The corticotropin-releasing factor (CRF) hypothesis of depression: new
findings and new directions. Mol Psychiatry 1: 336–342.
Nemeroff, C. B., Widerlov, E., Bissette, G., et al. (1984). Elevated concentrations of CSF
corticotropin-releasing factor-like immunoreactivity in depressed patients. Science 226:
1342–1344.
Owens, M. J., Nemeroff, C. B. (1993). The role of corticotropin-releasing factor in the pathophysi-
ology of affective and anxiety disorders: laboratory and clinical studies. Ciba Found Symp 172:
296–308.
Pantellis, C., Velakoulis, D., Suckling, J., et al. (2000). Left medial temporal lobe volume reduction
occurs during the transition from high risk to first episode psychosis. Schizophr Res 41: 35.
Pariante, C. M. (2003). Depression, stress and the adrenal axis. J Neuroendocrinol 15: 811–812.
Pariante, C. M., Miller, A. H. (2001). Glucocorticoid receptors in major depression: relevance to
pathophysiology and treatment. Biol Psychiatry 49: 391–404.
Pariante, C. M., Pearce, B. D., Pisell, T. L., Owens, M. J., Miller, A. H. (1997). Steroid-independent
translocation of the glucocorticoid receptor by the antidepressant desipramine. Mol Phar-
macol 52: 571–581.
Pariante, C. M., Makoff, A., Lovestone, S., et al., (2001). Antidepressants enhance glucocorticoid
receptor function in vitro by modulating the membrane steroid transporters. Br J Pharmacol
134: 1335–1343.
Pariante, C. M., Papadopoulos, A. S., Poon, L., et al. (2002). A novel prednisolone suppression
test for the hypothalamic–pituitary–adrenal axis. Biol Psychiatry 51: 922–930.
Pariante, C. M., Hye, A., Williamson, R., et al. (2003a). The antidepressant clomipramine regulates
cortisol intracellular concentrations and glucocorticoid receptor expression in fibroblasts
and rat primary neurones. Neuropsychopharmacology 28: 1553–1561.
Pariante, C. M., Kim, R. B., Makoff, A., Kerwin, R. W. (2003b). The antidepressant fluoxetine
enhances glucocorticoid receptor function in vitro by modulating membrane steroid trans-
porters. Br J Pharmacol 139: 1111–1118.
Pariante, C. M., Vassilopoulou, K., Velakoulis, D., et al. (2004a). Pituitary volume in psychosis.
Br J Psychiatry 185: 5–10.
Pariante, C. M., Thomas, S. A., Lovestone, S., Makoff, A., Kerwin, R. W. (2004b). Do antidepres-
sants regulate how cortisol affects the brain? 2003 Curt Richter Award Paper. Psychoneuro-
endocrinology 29: 423–447.
120 C. M. Pariante and D. Cotter

Purba, J. S., Raadsheer, F. C., Hofman, M. A., et al. (1995). Increased number of corticotropin-
releasing hormone expressing neurons in the hypothalamic paraventricular nucleus of
patients with multiple sclerosis. Neuroendocrinology 62: 62–70.
Raadsheer, F. C., van Heerikhuize, J. J., Lucassen, P. J., et al. (1995). Corticotropin-releasing
hormone mRNA levels in the paraventricular nucleus of patients with Alzheimer’s disease
and depression. Am J Psychiatry 152: 1372–1376.
Rajkowska, G., Miguel-Hidalgo, J. J., Wei. J. (1999). Morphometric evidence for neuronal and
glial prefrontal cell pathology in major depression. Biol Psychiatry 45: 1085–1098.
Rappaport, J. L., Giedd, J. N., Blumenthal, J., et al. (1999). Progressive cortical change during
adolescence in childhood-onset schizophrenia. A longitudinal magnetic resonance imaging
study. Arch Gen Psychiatry 56: 649–654.
Ribeiro, S. C., Tandon, R., Grunhaus, L., Greden, J. F. (1993). The DST as a predictor of outcome
in depression: a meta-analysis. Am J Psychiatry 150: 1618–1629.
Rosoklija, G., Toomayan, G., Ellis, S. P., et al. (2000). Structural abnormalities of subicular
dendrites in subjects with schizophrenia and mood disorders. Arch Gen Psychiatry 57: 349–
356.
Sachar, E. J., Kanter, S. S., Buie, D., Engle, R., Mehlman, R. (1970). Psychoendocrinology of ego
disintegration. Am J Psychiatry 126: 1067–1078.
Sapolsky, R. (2000). The possibility of neurotoxicity in the hippocampus in major depression:
a primer on neuron death. Biol Psychiatry 48: 755–65.
Sharma, R. P., Pandey, G. N., Janicak, P. G., et al. (1988). The effect of diagnosis and age on the
DST: a metaanalytic approach. Biol Psychiatry 24: 555–568.
Tandon, R., Mazzara, C., DeQuardo, J., et al. (1991). Dexamethasone suppression test in
schizophrenia: relationship to symptomatology, ventricular enlargement, and outcome. Biol
Psychiatry 29: 953–964.
Walder, D. J., Walker, E. F., Lewine, R. J. (2000). Cognitive functioning, cortisol release, and
symptom severity in patients with schizophrenia. Biol Psychiatry 48: 1121–1132.
Walker, E. F., Walder, D. J., Reynolds, F. (2001). Developmental changes in cortisol secretion in
normal and at-risk youth. Dev Psychopathol 13: 721–732.
Webster, M. J., Knable, M. B., O’Grady, J., Orthmann, J., Weickert, C. S. (2002). Regional specificity
of brain glucocorticoid receptor mRNA alterations in subjects with schizophrenia and mood
disorders. Mol Psychiatry 7: 985–994, 924.
Weinberger, D. R. (1987). Implications of normal brain development for the pathogenesis of
schizophrenia. Arch Gen Psychiatry 44: 660–669.
Xing, G. Q., Russell, S., Webster, M. J., Post, R. M. (2004). Decreased expression of mineralo-
corticoid receptor mRNA in the prefrontal cortex in schizophrenia and bipolar disorder. Int
J Neuropsychopharmacol 7: 143–153.
Young, E. A., Haskett, R. F., Murphy-Weinberg, V., Watson, S. J., Akil, H. (1991). Loss of gluco-
corticoid fast feedback in depression. Arch Gen Psychiatry 48: 693–699.
Young, E. A., Lopez, J. F., Murphy-Weinberg, V., Watson, S. J., Akil, H. (1998). The role of
mineralocorticoid receptors in hypothalamic–pituitary–adrenal axis regulation in humans.
J Clin Endocrinol Metab 83: 3339–3345.
Young, E. A., Lopez, J. F., Murphy-Weinberg, V., Watson, S. J., Akil, H. (2003). Mineralocorticoid
receptor function in major depression. Arch Gen Psychiatry 60: 24–28.
 7

Velo-cardio-facial syndrome (deletion

22q11.2): a homogeneous
neurodevelopmental model
for schizophrenia
Stephan Eliez1 and Carl Feinstein2
1
University of Geneva School of Medicine, Geneva, Switerland
2
Stanford University School of Medicine, Stanford, USA

Since the pioneering reports by Barbara Fish in 1977 and of Daniel Weinberger
in 1987, there has been increasing awareness that schizophrenia is most likely a
neurodevelopmental disorder (Fish, 1977; Weinberger, 1987). According to this
model of pathogenesis, there are underlying deviations already present in the early
brain development of individuals who develop schizophrenia in adulthood. These
neurodevelopmental abnormalities do not present as classical schizophrenia but as
more subtle neurocognitive deficits. To test this theoretical model, several prospec-
tive longitudinal studies as well as other approaches that link earlier childhood
clinical data to onset of schizophrenia in adulthood have been utilized in the search
for early life indicators of vulnerability to this tragic disorder. Recent findings from
these studies consistently indicate that children at high genetic risk for schizophre-
nia do, in fact, have significantly elevated rates of neurocognitive and behavioral
problems (Cannon et al., 2002; Erlenmeyer-Kimling et al., 2000; Fish et al., 1992;
Hans et al., 1999; Marcus et al., 1993). In addition, closely related research has
demonstrated that, prior to the onset of schizophrenia in cohorts of adults, clear
indicators of neurocognitive impairments were present earlier in life (Reichenberg
et al., 2002). In a prospective longitudinal study of a population-based (non-high-
risk) cohort, significant neuromotor, cognitive, and language impairments were
present only among children who were later diagnosed with schizophreniform dis-
order or schizophrenia (Cannon et al., 2002). Most recently, subtle but similar
neurocognitive impairments have been found in both non-schizophrenia children
and adults with elevated familial risk for schizophrenia (Asarnow et al., 2002a,b).
In summary, there is now a very impressive body of data supporting the neurode-
velopmental hypothesis of schizophrenia.
Neurodevelopment and Schizophrenia, ed. Matcheri S. Keshavan et al. Published by Cambridge University
Press.

C Cambridge University Press 2004.

121
122 S. Eliez and C. Feinstein

Just prior to the research findings summarized above, in 1992, Shprintzen and col-
leagues reported that children with velo-cardio-facial syndrome (VCFS), a genetic
disability syndrome caused, in most cases, by a de novo 3 MB microdeletion at chro-
mosome region 22q11.2, were at remarkably high risk for developing schizophrenia
in adulthood. Of children with this medically complex chromosome 22 microdele-
tion syndrome, known mostly to pediatric specialists (and certainly obscure to
most schizophrenia researchers), 30% developed schizophrenia or schizoaffective
disorder by young adulthood. This finding has since been confirmed and extended
by successive studies, which now indicate that approximately 2% of all individuals
with adult-onset schizophrenia, and a possibly higher percentage of youngsters with
childhood-onset schizophrenia, have VCFS, as confirmed by reliable laboratory-
based genetic diagnostic methods.
There was initial reluctance among many mainstream researchers in schizophre-
nia to acknowledge this finding. Instead, it was argued that VCFS is a childhood
developmental disability that, in adults, merely mimics the phenotype of “true”
schizophrenia. Nevertheless, the accumulation of data confirming prior childhood
neurodevelopmental impairments in adults with schizophrenia suggests that a more
heuristic approach would be to view VCFS as a model type of disorder for studying
the developmental pathogenesis of schizophrenia. Unlike the clusters of subtle neu-
rocognitive deficits that are gradually emerging from other lines of research, VCFS
has a known molecular genetic cause that can be biologically validated in earliest
infancy. This makes possible detailed study of the interaction of known genetic
factors with brain development from infancy through adulthood, as assayed by
neuroimaging, neurophysiological, neuropsychological, and psychiatric diagnostic
techniques. Through such studies, at least one biologically validated developmen-
tal pathway to the development of schizophrenia could be elucidated. This, in
turn, would create opportunities to design early interventions for the prevention
or amelioration of the schizophrenia before its most destructive symptoms appear.
In this chapter, we summarize the available clinical, neuroimaging, and genetic
information regarding VCFS, so that it may be available for clinical research in
schizophrenia.

Medical aspects
VCFS was first described by Shprintzen et al. (1978) as a distinctive congenital pat-
tern presenting in early childhood as hypernasal speech (associated either with
cleft palate or submucosal cleft), cardiac defects, distinctive facial appearance,
learning disabilities, and behavioral disturbances. Several other contemporane-
ous authors had also described children with conotruncal heart defects, palate
123 Velo-cardio-facial (22q11.2 deletion) syndrome

malformations (and resulting hypernasal speech), numerous physical dysmor-

phisms, and both metabolic and hematological problems (Shprintzen, 1999). Con-
sequently, a number of other names are still used for similar or overlapping medical
syndromes, including DiGeorge syndrome, conotruncal anomaly face syndrome,
and CATCH 22 (cardiac anomalies, abnormal facies, thymic hypoplasia, cleft palate,
and hypocalcemia) (Bassett et al., 1998; Driscoll et al., 1993; Gong et al., 1996; Scam-
bler et al., 1992). More recently, the term, 22q11.2 deletion syndrome has been used
to describe this condition (Scambler et al., 1992).
Well over 100 medical manifestations or physical anomalies are associated with
VCFS, but there is great variability in the number, type, and severity of prob-
lems from individual to individual (Cohen et al., 1999; Goldberg et al., 1993;
Golding-Kushner et al., 1985; McDonald-McGinn et al., 1999; Ryan et al., 1997;
Vantrappen et al., 1999; a VCFS fact sheet published by R. Shprintzen can be down-
loaded at www.vcfsef.org). The most common clinical signs are clefting of the
secondary palate, long facies, retrognathia, prominent nose, malar flattening, car-
diac anomalies (ventriculoseptal defect, tetrology of Fallot, right-sided aortic arch),
hypoparathyroidism, thymic hypoplasia, small stature, hypotonia, slender hands
and digits, and learning disabilities or mental retardation. These medical anomalies
are present at birth and often result in serious perinatal illness requiring surgery or
other intensive intervention. However, mild cases, with no serious medical morbid-
ity, may escape detection at infancy. Developmental delays in the preschool period
may result in identification of VCFS at that time, but some children with VCFS and
most adults whose condition was unrecognized in childhood are never identified,
because of a low clinical index of suspicion.

Cognitive and behavioral phenotype in childhood developmental delays

Almost all toddler and preschool children with VCFS suffer from mild gross motor
delays, severe expressive and receptive language delays, and characteristic disabling
impairments in speech production (Gerdes et al., 1999; Golding-Kushner et al.,
1985; McDonald-McGinn et al., 1999; Scherer et al., 1999; Swillen et al., 1999).
These delays are not accounted for by the frequently occurring palatal and related
nasopharyngeal defects, and they are not associated with the presence or absence
of severe cardiac defects (Gerdes et al., 1999; Scherer et al., 1999). Motor delays
(particularly hypotonia) tend to be mild; however, early receptive and expressive
language development is almost always severely delayed, with very little vocabulary
development at 30 months of age (Gerdes et al., 1999). Measures of overall intelli-
gence in toddler and preschoolers with VCFS indicate a range of intelligence from
near-average to mild/moderate retardation, with the overall mean trending towards
the borderline to mildly retarded range by 30 months (Scherer et al., 1999). Other
124 S. Eliez and C. Feinstein

relatively common findings in preschool VCFS children include microcephaly and

feeding and swallowing problems.
School-age children and adolescents with VCFS suffer from decrements in overall
cognitive ability, learning disabilities, and speech and language problems (Golding-
Kushner et al., 1985; McDonald-McGinn et al., 1999; Moss et al., 1999; Swillen
et al., 1999). Children ages 6 years and older with VCFS have intelligence quotients
(IQ) that are lower than the population average. Approximately 40% perform in
the mild to moderately mentally retarded range of intelligence (Goldberg et al.,
1993; Golding-Kushner et al., 1985; Moss et al., 1999; Ryan et al., 1997; Swillen
et al., 1997). Learning disabilities are almost invariably present (Goldberg et al.,
1993; Moss et al., 1999; Ryan et al., 1997; Swillen et al., 1997).
The cognitive profile of VCFS in this age range is complex, with an admixture
of non-verbal learning disabilities and ongoing receptive and expressive language
disorders (Golding-Kushner et al., 1985; Moss et al., 1999). By middle childhood,
children with VCFS show considerable advances in aspects of their language devel-
opment, particularly speech, vocabulary, reading, word recognition, and spelling.
However, higher-order comprehension, abstract verbal understanding, and lan-
guage formulation remained very deficient, characterized by “immature, concrete
thought” (Golding-Kushner et al., 1985). Speech almost invariably remains hyper-
nasal. VCFS children have significantly lower mean performance IQ scores than
verbal IQ scores (based on the Wechsler Intelligence Scale for Children version III),
and it has been suggested that they might have a non-verbal learning disability (Moss
et al., 1999). However, scores on standardized measure of language functioning are
often significantly lower than verbal IQ scores. Of children with VCFS, 50% meet
full diagnostic criteria for specific language impairment. Therefore, it appears that
VCFS children have distinctive impairments in higher-order language, abstract
reasoning, visuaospatial reasoning, and arithmetic. Almost all require intensive,
long-term special educational services.
Unfortunately, there are no prospective data available regarding the developmen-
tal trajectory of cognitive dysfunction from childhood to adulthood. Information
about cognitive functioning in VCFS adults is very limited. A recent survey found
that 94% had evidence of mental retardation or learning disability (Cohen et al.,
1999).

Childhood behavioral phenotype

Children with VCFS have a high rate of maladaptive behaviors and social traits.
These include frequently occurring problems with overactive, impulsive, emo-
tionally labile, and disorganized behaviors, and social problems such as shyness
or disinhibition (Gerdes et al., 1999; Golding-Kushner et al., 1985; Swillen et al.,
1999). Papolos et al. (1996) reported that attention-deficit hyperactivity disorder
125 Velo-cardio-facial (22q11.2 deletion) syndrome

is very common, and also described high rates of separation-anxiety disorder and
obsessive–compulsive disorder. Dimensional behavior ratings utilizing the Child
Behavioral Checklist noted abnormally high scores in the subscales for social prob-
lems, attention problems, thought problems, withdrawn behaviors, and anxiety/
depression (Swillen et al., 1997, 1999). Children with VCFS are at an increased
risk for childhood-onset schizophrenia. Usiskin et al. (1999) recently reported that
6.4% of a cohort of youngsters with childhood-onset schizophrenia had VCFS.
However, it is also well known that rates of psychiatric disorder are elevated in
children with borderline and mild mental retardation, developmental language dis-
orders, and learning disabilities (Feinstein and Reiss, 1996; Feinstein and Weiner,
1997; Fraser and Nolan, 1994; Toppelberg and Shapiro, 2000). Consequently, it was
not clear that the high rates of psychopathology in VCFS children were distinc-
tively different from those of other children with similar degrees of developmental
impairment. Feinstein et al. (2002) compared psychiatric findings in a group of
VCFS children with a control group matched for reduced IQ and severe learning
disabilities. They found very high rates of specific phobias, generalized anxiety disor-
der, major depressive disorder, oppositional defiant disorder, higher than expected
rates of obsessive–compulsive disorder, and evidence of delusions and hallucina-
tions in a few subjects. However, VCFS patients did not have significantly higher
rates of any these disorders than the cognitively matched controls.
Despite the difficulties to date in finding specific cognitive or behavioral features
in children with VCFS that are specific indicators of later-onset psychosis, it may
be that a subset of symptoms within the total array are, indeed, specific markers.
Further longitudinal study of children with VCFS into adulthood offers the best
hope of identifying any specific markers present in childhood.

Psychiatric presentation in adulthood

Links between VCFS and adolescent- or adult-onset schizophrenia have been
demonstrated through clinical research on both disorders. Shprintzen’s group
reported, one decade after their early papers on VCFS, that 20–30% of VCFS chil-
dren 16 years or older had developed schizophrenia or schizoaffective disorder
(Pulver et al., 1994; Shprintzen et al., 1992). Papolos et al. (1996) also noted a high
rate of cyclothymia in adolescents with VCFS and reported that 64% met criteria
for bipolar spectrum disorders. The issue of pathological mood lability in patients
with VCFS remains a topic of interest, although it has been partially eclipsed for the
moment by the more extensive documentation of increased risk for schizophrenia.
Karayiorgou et al. (1995) reported two patients with 22q11.2 deletions in a
random sample of 100 schizophrenia patients. Gothelf et al. (1997) utilized a clinical
ascertainment strategy to identify adult hospitalized schizophrenia patients with
undiagnosed VCFS by selecting those patients with both heart and palate defects
126 S. Eliez and C. Feinstein

(or other significant congenital anomalies) and assaying for the VCFS chromosomal
deletion. This approach identified three VCFS patients with schizophrenia from a
pool of 20. Bassett and colleagues (Bassett and Chow, 1999; Bassett et al., 1998) have
reported association between VCFS and schizophrenia, using a similar approach
to that of Gothelf, identifying a group of 15 adult schizophrenia patients with at
least two of five common physical or developmental phenotypic features of VCFS.
Ten (67%) were found to have VCFS. The IQ scores of this group were within the
borderline to mildly retarded range. In addition to delusions, hallucinations, social
withdrawal, and adaptive impairment, most schizophrenia patients with VCFS had
a cluster of associated behavioral features, including impulsivity, frequent aggressive
or temper outbursts, mood lability, anxiety symptoms, and compulsive features.
Murphy et al. (1999) provided a decisive replication for previous findings of a
high rate of schizophrenia in adults with VCFS employing a structured psychiatric
research diagnostic algorithm for 50 VCFS patients: 30% were psychotic (24%
schizophrenia, 2% schizoaffective, 2% psychosis [not otherwise specified], and
2% rapid cycling bipolar disorder) and 12% also had a lifetime history of major
depressive disorder. The range of intelligence in the VCFS subjects was similar to
other studies (mean IQ scores in the borderline retarded range). There was no
statistical association between IQ, cardiac defects, or cleft palate problems and the
presence or absence of psychosis. No association was found between psychosis and
the allele for catechol-O-methyltransferase (COMT; see below).

Genetics
In 1992, Scambler et al. identified a microdeletion at chromosome 22q11.2 in a
majority of patients presenting with VCFS, as well as the other similar named
syndromes. Individuals with this deletion are hemizygous for all genes encompassed
by the deletion site. The 22q11.2 deletion is estimated to occur in at least 1 per 2000
to 4500 live births (Tezenas Du Montcel et al., 1996). In most affected individuals,
a de novo 3 Mb deletion at chromosome 22q11.2 is responsible for the syndrome;
however, in 10–25%, VCFS is inherited from a parent with VCFS (Carlson et al.,
1997a,b; Driscoll et al., 1993; Lindsay et al., 1995; Scambler et al., 1992; Shaikh
et al., 2000). There is some evidence that patients with inherited VCFS may have
more severe clinical manifestations (Ryan et al., 1997; Swillen et al., 1997). There
also appears to be a preponderance of maternally transmitted inheritance, perhaps
because of greater likelihood for having offspring in woman with VCFS (Murphy
et al., 1999; Ryan et al., 1997).
In fact, several microdeletions of different sizes can occur (presumably involving
different genes or different overlapping or non-overlapping regions in the deletion
127 Velo-cardio-facial (22q11.2 deletion) syndrome

site) in VCFS, all sited within the larger and by far the most commonly occurring
3 Mb 22q11.2 deletion region (Shaikh et al., 2000). The relationship between the
size and genetic content of the deletion and the resulting phenotoype is not well
understood. Moreover, not all patients presenting with features of VCFS have the
22q11.2 microdeletion. Clinic surveys have found the deletion in 76% of VCFS, 88%
of those with DiGeorge syndrome, and 84% of those with conotruncal anomaly
face syndrome (Driscoll et al., 1993; Fokstuen et al., 1998; Lindsay et al., 1995).
As diagnostic certainty will increase among clinicians, “non-deleted” individuals
with VCFS might appear as exceptions. Nevertheless, the ability to test for the
22q11.2 microdeletion greatly facilitates identifying and classifying the somatic,
cognitive, and psychiatric sequelae of hemizygosity for the chromosome 22q11.2
site.
At least 30 genes are encoded in the commonly deleted segment (Dunham
et al., 1999), a few of which are highly expressed in brain tissue and are likely
to be essential for normal brain development (Funke et al., 1997; Gottlieb et al.,
1997; Lachman et al., 1996a; Roberts et al., 1997; Yamagishi et al., 1999). Certain
specific genes within the 22q11.2 site have attracted intense interest because of
their known relationship to central nervous system structure or functioning. These
include the gene for the goosecoid-like protein (GSCL), the genes, Es2, UFDIL, the
gene for proline dehydrogenase (PDH), PIK4CA, and the gene for COMT.
Both the GSCL and ES2 genes play a role in mouse brain embryogenesis and thus
could be related to structural brain abnormalities observed in VCFS (Funke et al.,
1997; Gottlieb et al., 1997, 1998; Lindsay et al., 1998; Saint-Jore et al., 1998). Yamag-
ishi et al. (1998) measured the expression of a newly identified gene, UFDIL, which
has been localized to the commonly deleted region. Of 182 subjects with VCFS, all
had that specific gene deleted. Findings from this study indicate that UFDIL plays
a key role in the embryonic development of the heart, palate, frontonasal regions,
and brain. Specific polymorphisms of PIK4CA (Saito et al., 2003) and the gene
for PDH (Jacquet et al., 2002; Liu et al., 2002a) have been found to be potentially
more common among populations of individuals with schizophrenia or bipolar
disorder. Therefore, polymorphisms of PIK4CA and the gene for PDH have been
hypothesized to be also more frequent in individuals with VCFS and psychosis. So
far, no reports have clearly established, within the VCFS population, a link between
psychiatric symptoms and allelic variants.
More recently, three independent groups have targeted the T-box gene Tbx1
for gene inactivation in mouse models (Jerome and Papaioannou, 2001; Lindsay
et al., 2001; Merscher et al., 2001). Tbx1 homozygosity was perinatally lethal and
associated with several usual features of VCFS such as thymus and parathyroid
gland aplasia, major ear malformations, cleft palate, and conotruncal heart defects
128 S. Eliez and C. Feinstein

(McDermid and Morrow, 2002). Tbx1 is a member of the T-box-containing family

of transcription factors and also may modulate brain expression of several genes
expressed in gray and white matter brain tissue. More studies will be needed to
elucidate the function in humans of the gene considered today as the strongest
candidate for the malformative features associated with VCFS.
The COMT gene has attracted the greatest attention in psychiatry because of its
key role in neurotransmitter function. Its expression produces the enzyme involved
in the metabolism of catecholamines and catechol-containing drugs, including
dopamine, epinephrine, and norepinephrine (Cooper et al., 1996). The enzymic
activity of COMT varies substantially in humans (Palmatier et al., 1999; Wein-
shilboum, 1978) as a result of a single nucleotide polymorphism (known as the
COMT158 polymorphism) in the gene. The two resulting different alleles, COMT M
and COMT V , express two different versions of the COMT enzyme, with COMTM
having lower activity (three- to fourfold decreased) (Lachman et al., 1996b). Indi-
viduals can have any of three combinations of the alleles, accounting for variability
among humans in COMT enzyme activity. However, people with VCFS are hem-
izygous for the COMT allele and, therefore, are the only group where the possibil-
ity exists for possessing a single M allele (and, therefore, the only known human
group having the lowest possible level of COMT enzyme activity) (Lachman et al.,
1996a).
There have been numerous studies concerning the potential association of low-
enzymatic COMT polymorphism with psychiatric conditions, including bipolar
disorder, schizophrenia, and attention-deficit hyperactivity disorder (Bilder et al.,
2002; Kirov et al., 1998; Kunugi et al., 1997; Lachman et al., 1997, 1996b; Papolos
et al., 1998; Shifman et al., 2002; Strous et al., 1997; Weinberger et al., 2001).
VCFS has, at times, been referred to as a “human model” to assay the role of COMT
polymorphisms in psychiatric disorders (Lachman et al., 1996a; Papolos et al., 1996;
Shifman et al., 2002). Unfortunately, findings from these studies are frequently
contradictory and, therefore, the putative role of COMT in the pathogenesis of
psychiatric disorders remains unclear.
Studies of the relationship between VCFS and schizophrenia have recently stim-
ulated interest in further molecular genetic analysis of the 22q11.2 site, utilizing
both family- and population-based schizophrenia samples. Most recently, Liu et al.
(2002b,c), based on the VCFS data and on two independent schizophrenia genetic
linkage reports of high-susceptibility loci for schizophrenia in the 22q11 region,
reasoned that non-deletion variants of individual genes within the 22q11 region
might be related to risk of schizophrenia (Chakravarti, 2002). They undertook
a fine-grained molecular genetic search for schizophrenia susceptibility, utilizing
linkage analysis of single nucleotide polymorphisms from a 1.5 MB “schizophre-
nia critical region” in the 22q11 region. Their findings indicated more than one
129 Velo-cardio-facial (22q11.2 deletion) syndrome

haplotype within the segment, in which PRODH2 and DGCR6 are the only known
genes, where the schizophrenia susceptibility variant(s) most likely resides.

Neuroimaging
Since the early 1990s, magnetic resonance imaging (MRI) methodologies have been
utilized to investigate the effect of the 22q11.2 deletion on brain structure. Mitnick
et al. (1994) reported brain abnormalities, most commonly small cerebellar ver-
mises, among 9 of 11 children with VCFS. Reduced volume of the posterior fossa
was found in four, and cysts adjacent to the anterior horns of the ventricles were
found in three. Chow et al. (1999) studied 11 adults with VCFS and schizophrenia
and found that 90% exhibited bilateral white matter hyperintensities, distributed
mainly within the frontal lobes; 45% had either cavum septum pellucidum or
cavum vergae, and 36% had cerebellar hypoplasia. More recently, several cases of
cortical dysgenesis including uni/bilateral frontoparietal polymicrogyria have been
reported in children with 22q11 deletion syndrome (Bingham et al., 1998; Bird,
2001; Ghariani et al., 2002; Kawame et al., 2000; Worthington et al., 2000), possibly
resulting in hemiplegia and seizures in the most severely affected patient. The under-
lying pathogenic mechanism for the 22q11 deletion syndrome-associated cortical
dysgenesis is still unclear. Abnormal neuronal migration is one proposed explan-
ation, but vascular disruption has also been suggested as a potential mechanism
(Bird, 2001).
There are only a limited number of publications presenting MRI quantitative data
in VCFS (Chow et al., 2002; Eliez et al., 2000; Kates et al., 2001; van Amelsvoort
et al., 2001). Eliez et al. (2000) first reported significant alterations in cerebral
morphology among children and adolescents with VCFS; an 11% reduction in total
brain volume was detected related to both gray and white matter reduction. Frontal
lobe brain volume was relatively enlarged in subjects with VCFS. Most notable
was a significant reduction of parietal lobe gray matter in the left hemisphere and
alteration of cerebellar volumes and vermis (Eliez et al., 2001a). The dissociation
between the alteration of frontal and non-frontal regions in children with VCFS
was confirmed by Kates et al. (2001) with white matter tissue affected to a greater
extent than white matter. Interestingly, these authors also reported a reduction of
white matter in the left temporal cortex.
Investigating differences between 10 adults with VCFS and 13 age- and IQ-
matched controls (among which two met criteria for schizophrenia and four for
mood disorder), van Amelsvoort et al. (2001) described reduction of gray matter in
the cerebellum and left temporal lobe, and reduction of white matter in the superior
medial regions of the frontal lobe and in the frontotemporal regions. Another study,
comparing 14 adults with VCFS and schizophrenia with 14 normal gender- and
130 S. Eliez and C. Feinstein

Fig. 7.1. As suggested by several authors (McGlashan and Hoffman, 2000), the reduction of gray
matter volumes in schizophrenia (SZ) is principally a result of abnormal dendritic arboriz-
ation, giving rise to excessive pruning rather than decreased number of neurons. One might
speculate that identical neurohistological characteristics are responsible for the gray matter
volume decrease observed in individuals with the 22q11 deletion. To date, no detailed
postmortem neuroanatomical and neurohistological studies have been published on human
or animal models with velo-cardio-facial syndrome. These studies will be most valuable in
order to understand the relation between genetic alteration and changes in white and gray
matter tissue compartments.

age-matched controls (Chow et al., 2002), reported an overall decrease of gray but
not white matter. Lobar gray matter reduction was observed in the frontal and
temporal lobes bilaterally and in the left parietal and occipital lobes. Ventricular
cerebrospinal fluid was increased in individuals with VCFS and schizophrenia com-
pared with controls. Most of the brain alterations described in VCFS are common to
those reported in neuroimaging literature on schizophrenia (Shenton et al., 2001).
A review of most of the available neuroimaging studies of children with VCFS
would suggest that there is an early alteration of parietal lobe and cerebellum, and
that the decrease of temporal lobe gray matter and hippocampus can be observed
only in adults. However, in a study investigating volumetric changes with age in
VCFS, Eliez et al. (2001b) found that accelerated decrease in volumes of gray matter
in the temporal lobes and hippocampus already occurred during childhood and
adolescence. The accelerated decrease of gray matter in temporal and mesiotem-
poral structures is likely to lead to significant volumetric differences by the time
individuals with VCFS reach young adulthood. In another publication by the same
group (Eliez et al., 2001c), it has been suggested that individuals with a maternally
derived 22q11.2 microdeletion have an accelerated decrease of gray matter, leading
131 Velo-cardio-facial (22q11.2 deletion) syndrome

to significant reduction of total gray matter volumes and cognitive impairment

(Glaser et al., 2002). However, these results must be considered as preliminary
since previous molecular studies did not show evidences for imprinting in the
22q11 region. Nevertheless, such findings emphasize the potential importance of
epigenetic phenomenon like imprinting in the genetic factors contributing to brain
development and function in schizophrenia.
Interestingly, in a recently reported longitudinal MRI study of individuals with
early-onset schizophrenia, Thompson et al. (2001) showed that an alteration of
cortical gray matter first developed in the parietal regions, then progressed towards
frontal and temporal lobe, finally resulting in a severe overall gray matter reduction
over a course of 4 years. In patients with adult-onset schizophrenia, gray matter
volume reduction is already evident at the first clinical presentation of the disorder
and appears to explain, at least partially, the morbid (Gold et al., 1999; Gur et al.,
1999; Mohamed et al., 1999; Thompson et al., 2001) and premorbid cognitive
features (Cannon et al., 2002; Fuller et al., 2002) associated with this condition.
Further, differences in the size of the decrease in gray matter volume over time,
modulated by parental origin of the deletion and other factors (yet to be identified),
could explain the observed variability in the age of onset of schizophrenia in the
VCFS population. Figure 7.1, concordant with a developmental model proposed
by McGlashan and Hoffman (2000), illustrates the potential impact of gray matter
decrease rate on age of onset of clinical symptoms, duration of prodromal phase
before the expression of schizophrenia and the age at which the negative symptoms
are predominant.

Conclusions
In summary, we have described the medical, genetic, cognitive, and behavioral phe-
notype of VCFS from infancy to adulthood, with particular emphasis on the vari-
ability of the phenotype and the strong association between the 22q11.2 chromoso-
mal microdeletion and schizophrenia. VCFS represents a developmental model of a
genetically homogeneous subtype of schizophrenia and a unique window to under-
stand the interaction in children and adolescents between genes, brain, behavior,
and cognition to lead to adult psychiatric disorders.

REFERENCES

Asarnow, R. F., Nuechterlein, K. H., Asamen, J., et al. (2002a). Neurocognitive functioning and
schizophrenia spectrum disorders can be independent expressions of familial liability for
schizophrenia in community control children: the UCLA family study. Schizophr Res 54:
111–120.
132 S. Eliez and C. Feinstein

Asarnow, R. F., Nuechterlein, K. H., Subotnik, K. L., et al. (2002b). Neurocognitive impairments
in nonpsychotic parents of children with schizophrenia and attention-deficit/hyperactivity
disorder: the University of California, Los Angeles Family Study. Arch Gen Psychiatry 59:
1053–1060.
Bassett, A. S., Chow, E. W. (1999). 22q11 deletion syndrome: a genetic subtype of schizophrenia.
Biol Psychiatry 46: 882–891.
Bassett, A. S., Hodgkinson, K., Chow, E. W., et al. (1998). 22q11 deletion syndrome in adults
with schizophrenia. Am J Med Genet 81: 328–337.
Bilder, R., Volavka, J., Czobor, P., et al. (2002). Neurocognitive correlates of the COMT
Val(158)Met polymorphism in chronic schizophrenia. Biol Psychiatry 52: 701–707.
Bingham, P. M., Lynch, D., McDonald-McGinn, D., Zackai, E. (1998). Polymicrogyria in chro-
mosome 22 delection syndrome. Neurology 51: 1500–1502.
Bird, L. M. (2001). Cortical dysgenesis and 22q11 deletion. Clin Dysmorphol 10: 77.
Cannon, M., Caspi, A., Moffitt, T. E., et al. (2002). Evidence for early-childhood, pan-
developmental impairment specific to schizophreniform disorder: results from a longi-
tudinal birth cohort. Arch Gen Psychiatry 59: 449–456.
Carlson, C., Papolos, D., Pandita, R. K., et al. (1997a). Molecular analysis of velo-cardio-facial
syndrome patients with psychiatric disorders. Am J Hum Genet 60: 851–859.
Carlson, C., Sirotkin, H., Pandita, R., et al. (1997b). Molecular definition of 22q11 deletions in
151 velo-cardio-facial syndrome patients. Am J Hum Genet 61: 620–629.
Chakravarti, A. (2002). A compelling genetic hypothesis for a complex disease, PRODHH2/
DGCR6 variation leads to schizophrenia susceptibility. Proc Natl Acad Sci USA 99: 4755–
4756.
Chow, E., Zipursky, R. B., Mikulis, D., et al. (1999). MRI findings in adults with 22q11 deletion
syndrome (22qDS) and schizophrenia. Schizophr Res 36: 89.
Chow, E. W., Zipursky, R. B., Mikulis, D. J., Bassett, A. S. (2002). Structural brain abnormalities
in patients with schizophrenia and 22q11 deletion syndrome. Biol Psychiatry 51: 208–215.
Cohen, E., Chow, E., Weksberg, R., Bassett, A. S. (1999). Phenotype of adults with the 22q11
deletion syndrome. Am J Med Genet 86: 359–365.
Cooper, J., Bloom, F., Roth, R. (1996): The Biochemical Basis of Neuropharmacology, 7th edn.
New York: Oxford University Press.
Driscoll, D. A., Salvin, J., Sellinger, B., et al. (1993). Prevalence of 22q11 microdeletions in
DiGeorge and velocardiofacial syndromes: implications for genetic counselling and prenatal
diagnosis. J Med Genet 30: 813–817.
Dunham, I., Shimizu, N., Roe, B. A., et al. (1999). The DNA sequence of human chromosome
22. Nature 402: 489–495.
Eliez, S., Schmitt, J. E., White, C. D., Reiss, A. L. (2000). Children and adolescents with velocar-
diofacial syndrome: a volumetric MRI study. Am J Psychiatry 157: 409–415.
Eliez, S., Schmitt, J. E., White, C. D., Wellis, V. G., Reiss, A. L. (2001a). A quantitative MRI study
of posterior fossa development in velocardiofacial syndrome. Biol Psychiatry 49: 540–546.
Eliez, S., Blasey, C. M., Schmitt, E. J., et al. (2001b). Velocardiofacial syndrome: are structural
changes in the temporal and mesial temporal regions related to schizophrenia? Am J Psychi-
atry 158: 447–453.
133 Velo-cardio-facial (22q11.2 deletion) syndrome

Eliez, S., Antonarakis, S. E., Morris, M. A., Dahoun, S. P., Reiss,, A. L. (2001c). Parental origin
of the deletion 22q11.2 and brain development in velocardiofacial syndrome: a preliminary
study. Arch Gen Psychiatry 58: 64–68.
Erlenmeyer-Kimling, L., Rock, D., Roberts, S., et al. (2000). Attention, memory, and motor
skills as childhood predictors of schizophrenia-related psychosis. Am J Psychiatry 157:
1416–1422.
Feinstein, C., Reiss, A. L. (1996). Psychiatric disorders in mentally retarded children and adoles-
cents. Child Adolesc Psychiatr Clin N Am 5: 827–852.
Feinstein, C., Weiner, J. (1997). Developmental disorders of learning, motor skills, and commu-
nication. In Textbook of Child and Adolescent Psychiatary, ed. J. Weiner. Washington, DC:
American Psychiatry Press, pp. 281–300.
Feinstein, C., Eliez, S., Blasey, C., Reiss, A. L. (2002). Psychiatric disorders and behavioral prob-
lems in children with velocardiofacial syndrome: usefulness as phenotypic indicators of
schizophrenia risk. Biol Psychiatry 51: 312–318.
Fish, B. (1977). Neurobiologic antecedents of schizophrenia in children. Evidence for an inherited,
congenital neurointegrative defect. Arch Gen Psychiatry 34: 1297–1313.
Fish, B., Marcus, J., Hans, S. L., Auerbach, J. G., Perdue, S. (1992). Infants at risk for schizophrenia:
sequelae of a genetic neurointegrative defect. A review and replication analysis of pandsy-
maturation in the Jerusalem Infant Development Study. Arch Gen Psychiatry 49: 221–235.
Fokstuen, S., Arbenz, U., Artan, S., et al. (1998). 22q11.2 deletions in a series of patients with
non-selective congenital heart defects: incidence, type of defects and parental origin. Clin
Genet 53: 63–69.
Fraser, W., Nolan, M. (1994). Psychiatric disorders in mental retardation. In Mental Health in
Mental Retardation, ed. N. Bouras. Cambridge: Cambridge University Press, pp. 79–92.
Fuller, R., Nopoulos, P., Arndt, S., et al. (2002). Longitudinal assessment of premorbid cognitive
functioning in patients with schizophrenia through examination of standardized scholastic
test performance. Am J Psychiatry 159: 1183–1189.
Funke, B., Saint-Jore, B., Puech, A., et al. (1997). Characterization and mutation analysis of
goosecoid-like (GSCL), a homeodomain-containing gene that maps to the critical region
for VCFS/DGS on 22q11. Genomics 46: 364–372.
Gerdes, M., Solot, C., Wang, P. P., et al. (1999). Cognitive and behavior profile of preschool
children with chromosome 22q11.2 deletion. Am J Med Genet 85: 127–133.
Ghariani, S., Dahan, K., Saint-Martin, C., et al. (2002). Polymicrogyria in chromosome 22q11
deletion syndrome. Eur J Paediatr Neurol 6: 73–77.
Glaser, B., Mumme, D. L., Blasey, C., et al. (2002). Language skills in children with velocardiofacial
syndrome (deletion 22q11.2). J Pediatr 140: 753–758.
Gold, S., Arndt, S., Nopoulos, P., O’Leary, D. S., Andreasen, N. C. (1999). Longitudinal study
of cognitive function in first-episode and recent-onset schizophrenia. Am J Psychiatry 156:
1342–1348.
Goldberg, R., Motzkin, B., Marion, R., Scambler, P. J., Shprintzen, R. J. (1993). Velo-cardio-facial
syndrome: a review of 120 patients. Am J Med Genet 45: 313–319.
Golding-Kushner, K. J., Weller, G., Shprintzen, R. J. (1985). Velo-cardio-facial syndrome: lan-
guage and psychological profiles. J Craniofac Genet Dev Biol 5: 259–266.
134 S. Eliez and C. Feinstein

Gong, W., Emanuel, B. B., Collins, J. E., et al. (1996). A transcription map of the DiGeorge and
velocardiofacial syndrome minimal critical region. Hum Mol Genet 5: 789–800.
Gothelf, D., Frisch, A., Munitz, H., et al. (1997). Velocardiofacial manifestations and microdele-
tions in schizophrenic inpatients. Am J Med Genet 72: 455–461.
Gottlieb, S., Emanuel, B. S., Driscoll, D. A., et al. (1997). The DiGeorge syndrome minimal critical
region contains a goosecoid-like (GSCL) homeobox gene that is expressed early in human
development. Am J Hum Genet 60: 1194–1201.
Gottlieb, S., Hanes, S. D., Golden, J. A., Oakey, R. J., Budarf, M. L. (1998). Goosecoid-like, a gene
deleted in DiGeorge and velocardiofacial syndromes, recognizes DNA with a bicoid-like
specificity and is expressed in the developing mouse brain. Hum Mol Genet 7: 1497–1505.
Gur, R. E., Turetsky, B. I., Bilker, W. B., Gur, R. C. (1999). Reduced gray matter volume in
schizophrenia. Arch Gen Psychiatry 56: 905–911.
Hans, S. L., Marcus, J., Nuechterlein, K. H., et al. (1999). Neurobehavioral deficits at adolescence
in children at risk for schizophrenia: the Jersualem Infant Development Study. Arch Gen
Psychiatry 56: 741–748.
Jacquet, H., Raux, G., Thibaut, F., et al. (2002). PRODH mutations and hyperprolinemia in a
subset of schizophrenic patients. Hum Mol Genet 11: 2243–2249.
Jerome, L. A., Papaioannou, V. E. (2001). DiGeorge syndrome phenotype in mice mutant for the
T-box gene, Tbx1. Nat Genet 27: 286–291.
Karayiorgou, M., Morris, M. A., Morrow, B., et al. (1995). Schizophrenia susceptibility associated
with interstitial deletions of chromosome 22q11. Proc Natl Acad Sci USA 92: 7612–7616.
Kates, W. R., Burnette, C. P., Jabs, E. W., et al. (2001). Regional cortical white matter reductions
in velocardiofacial syndrome: a volumetric MRI analysis. Biol Psychiatry 49: 677–684.
Kawame, H., Kurosawa, K., Akatsuka, A., Ochiai, Y., Mizuno, K. (2000). Polymicrogyria is an
uncommon manifestation in 22q11.2 deletion syndrome. Am J Med Genet 94: 77–78.
Kirov, G., Murphy, K. C., Arranz, M., et al. (1998). Low activity allele of catechol-O-
methyltransferase gene associated with rapid cylcing bipolar disorder. Mol Psychiatry 3:
342–345.
Kunugi, H., Vallada, H., Sham, P., et al. (1997). Catechol-O-methyltransferase polymorphisms
and schizophrenia: a transmission disequilibrium study in multiply affected families.
Psychiatr Genet 7: 97–101.
Lachman, H. M., Morrow, B., Shprintzen, R., et al. (1996a). Association of codon 108/158
catechol-O-methyltransferase gene polymorphism with the psychiatric manifestations of
velo-cardio-facial syndrome. Am J Med Genet 67: 468–472.
Lachman, H. M., Papolos, D. F., Saito, T., et al. (1996b). Human catechol-O-methyltransferase
pharmacogenetics: description of a functional polymorphism and its potential application
to neuropsychiatric disorders. Pharmacogenetics 6: 243–250.
Lachman, H., Kelsoe, J., Moreno, L., Katz, S., Papolos, D. (1997). Lack of association of catechol-O-
methyltransferase (COMT) functional polymorphism in bipolar affective disorder. Psychiatr
Genet 7: 13–17.
Lindsay, E. A., Greenberg, F., Shaffer, L. G., et al. (1995). Submicroscopic deletions at 22q11.2:
variability of the clinical picture and delineation of a commonly deleted region. Am J Med
Genet 56: 191–197.
135 Velo-cardio-facial (22q11.2 deletion) syndrome

Lindsay, E. A., Harvey, E., Scambler, P., Baldini, A. (1998). ES2, a gene deleted in DiGeorge
syndrome, encodes a nuclear protein and is expressed during early mouse development,
where it shares an expression domain with a goosecoid-like gene. Hum Mol Genet 7: 629–
635.
Lindsay, E. A., Vitelli, F., Su, H., et al. (2001). Tbx1 haploinsufficiency in the DiGeorge syndrome
region causes aortic arch defects in mice. Nature 410: 97–101.
Liu, H., Heath, S. C., Sobin, C., et al. (2002a). Genetic variation at the 22q11 PRODH 2/DGCR6
locus presents an unusual pattern and increases susceptibility to schizophrenia. Proc Natl
Acad Sci USA 99: 3717–3722.
Liu, H., Abecasis, G. R., Heath, S. C., et al. (2002b). Genetic variation in the 22q11 locus and
susceptibility to schizophrenia. Proc Natl Acad Sci USA 99: 16859–16864.
Liu, H., Heath, S. C., Sobin, C., et al. (2002c). Genetic variation at the 22q11 PRODH2/DGCR6
locus presents an unusual pattern and increases susceptibility to schizophrenia. Proc Natl
Acad Sci USA 99: 3717–3722.
Marcus, J., Hans, S. L., Auerbach, J. G., Auerbach, A. G. (1993). Children at risk for schizophrenia:
the Jerusalem Infant Development Study II. Neurobehavioral deficits at school age. Arch Gen
Psychiatry 50: 797–809.
McDermid, H. E., Morrow, B. E. (2002). Genomic disorders on 22q11. Am J Hum Genet 70:
1077–1088.
McDonald-McGinn, D. M., Kirschner, R., Goldmuntz, E., et al. (1999). The Philadelphia story:
the 22q11.2 deletion. Report on 250 patients. Genet Couns 10: 11–24.
McGlashan, T. H., Hoffman, R. E. (2000). Schizophrenia as a disorder of developmentally reduced
synaptic connectivity. Arch Gen Psychiatry 57: 637–648.
Merscher, S., Funke, B., Epstein, J. A., et al. (2001). TBX1 is responsible for cardiovascular defects
in velo-cardio-facial/DiGeorge syndrome. Cell 104: 619–629.
Mitnick, R. J., Bello, J. A., Shprintzen, R. J. (1994). Brain anomalies in velo-cardio-facial syndrome.
Am J Med Genet 54: 100–106.
Mohamed, S., Paulsen, J. S., O’Leary, D., Arndt, S., Andreasen, N. (1999). Generalized cognitive
deficits in schizophrenia: a study of first-episode patients. Arch Gen Psychiatry 56: 749–754.
Moss, E. M., Batshaw, M. L., Solot, C. B., et al. (1999). Psychoeducational profile of the 22q11.2
microdeletion: a complex pattern. J Pediatr 134: 193–198.
Murphy, K. C., Jones, L. A., Owen, M. J. (1999). High rates of schizophrenia in adults with
velo-cardio-facial syndrome. Arch Gen Psychiatry 56: 940–945.
Palmatier, M. A., Kang, A. M., Kidd, K. K. (1999). Global variation in the frequencies of func-
tionally different catechol-O-methyltransferase alleles. Biol Psychiatry 46: 557–567.
Papolos, D. F., Faedda, G. L., Veit, S., et al. (1996). Bipolar spectrum disorders in patients diagnosed
with velo-cardio-facial syndrome: does a hemizygous deletion of chromosome 22q11 result
in bipolar affective disorder? Am J Psychiatry 153: 1541–1547.
Papolos, D. F., Veit, S., Faedda, G. L., Saito, T., Lachman, H. M. (1998). Ultra-ultra rapid cycling
bipolar disorder is associated with the low activity catecholamine-O-methyltransferase allele.
Mol Psychiatry 3: 346–349.
Pulver, A. E., Nestadt, G., Goldberg, R., et al. (1994). Psychotic illness in patients diagnosed with
velo-cardio-facial syndrome and their relatives. J Nerv Ment Dis 182: 476–478.
136 S. Eliez and C. Feinstein

Reichenberg, A., Weiser, M., Rabinowitz, J., et al. (2002). A population-based cohort study of
premorbid intellectual, language, and behavioral functioning in patients with schizophrenia,
schizoaffective disorder, and nonpsychotic bipolar disorder. Am J Psychiatry 159: 2027–
2035.
Roberts, C., Daw, S. C., Halford, S., Scambler, P. J. (1997). Cloning and developmental expression
analysis of chick Hira (Chira), a candidate gene for DiGeorge syndrome. Hum Mol Genet 6:
237–245.
Ryan, A. K., Goodship, J. A., Wilson, D. I., et al. (1997). Spectrum of clinical features associated
with interstitial chromosome 22q11 deletions: a European collaborative study. J Med Genet
34: 798–804.
Saint-Jore, B., Puech, A., Heyer, J., et al. (1998). Goosecoid-like (Gscl ), a candidate gene for
velocardiofacial syndrome, is not essential for normal mouse development. Hum Mol Genet
7: 1841–1849.
Saito, T., Stopkova, P., Diaz, L., et al. (2003). Polymorphism screening of PIK4CA: possible
candidate gene for chromosome 22q11-linked psychiatric disorders. Am J Med Genet 116:
77–83.
Scambler, P. J., Kelly, D., Lindsay, E., et al. (1992). Velo-cardio-facial syndrome associated with
chromosome 22 deletions encompassing the DiGeorge locus. Lancet 339: 1138–1139.
Scherer, N. J., D’Antonio, L. L., Kalbfleisch, J. H. (1999). Early speech and language development
in children with velocardiofacial syndrome. Am J Med Genet 88: 714–723.
Shaikh, T. H., Kurahashi, H., Saitta, S. C., et al. (2000). Chromosome 22-specific low copy repeats
and the 22q11.2 deletion syndrome: genomic organization and deletion endpoint analysis.
Hum Mol Genet 9: 489–501.
Shenton, M. E., Dickey, C. C., Frumin, M., McCarley, R. W. (2001). A review of MRI findings in
schizophrenia. Schizophr Res 49: 1–52.
Shifman, S., Bronstein, M., Sternfeld, M., et al. (2002). A highly significant association between
a COMT haplotype and schizophrenia. Am J Hum Genet 71: 1296–1302.
Shprintzen, R. J. (1999). Historic overview of VCFS. Genet Couns 10: 1–2.
Shprintzen, R. J., Goldberg, R. B., Lewin, M. L., et al. (1978). A new syndrome involving cleft palate,
cardiac anomalies, typical facies, and learning disabilities: velo-cardio-facial syndrome. Cleft
Palate J 15: 56–62.
Shprintzen, R. J., Goldberg, R., Golding-Kushner, K. J., Marion, R. W. (1992). Late-onset psychosis
in the velo-cardio-facial syndrome. Am J Med Genet 42: 141–142.
Strous, R., Bark, N., Parsia, S., Vovlavka, J., Lachman, H. (1997). Analysis of a functional catechol-
O-methyltransferase gene polymorphism in schizophrenia: evidence for association with
aggressive and antisocial behavior. Psychiatr Res 69: 71–77.
Swillen, A., Devriendt, K., Legius, E., et al. (1997). Intelligence and psychosocial adjustment in
velocardiofacial syndrome: a study of 37 children and adolescents with VCFS. J Med Genet
34: 453–458.
(1999). The behavioural phenotype in velo-cardio-facial syndrome (VCFS): from infancy to
adolescence. Genet Couns 10: 79–88.
Tezenas Du Montcel, S., Mendizabai, H., et al. (1996). Prevalence of 22q11 microdeletion [letter].
J Med Genet 33: 719.
137 Velo-cardio-facial (22q11.2 deletion) syndrome

Thompson, P. M., Vidal, C., Giedd, J. N., et al. (2001). Mapping adolescent brain change reveals
dynamic wave of accelerated gray matter loss in very early-onset schizophrenia. Proc Natl
Acad Sci USA 98: 11650–11655.
Toppelberg, C., Shapiro, T. (2000). Language disorders: a 10-year research update review. J Am
Acad Child Adolesc Psychiatry 39: 143–152.
Usiskin, S. I., Nicolson, R., Krasnewich, D. M., et al. (1999). Velocardiofacial syndrome in
childhood-onset schizophrenia. J Am Acad Child Adolesc Psychiatry 38: 1536–1543.
van Amelsvoort, T., Daly, E., Robertson, D., et al. (2001). Structural brain abnormalities associated
with deletion at chromosome 22q11: quantitative neuroimaging study of adults with velo-
cardio-facial syndrome. Br J Psychiatry 178: 412–419.
Vantrappen, G., Devriendt, K., Swillen, A., et al. (1999). Presenting symptoms and clinical features
in 130 patients with the velo-cardio-facial syndrome. The Leuven experience. Genet Couns
10: 3–9.
Weinberger, D. R. (1987). Implications of normal brain development for the pathogenesis of
schizophrenia. Arch Gen Psychiatry 44: 660–669.
Weinberger, D. R., Egan, M. F., Bertolino, A., et al. (2001). Prefrontal neurons and the genetics
of schizophrenia. Biol Psychiatry 50: 825–844.
Weinshilboum, R. M. (1978). Human biochemical genetics of plasma dopamine-beta-
hydroxylase and erythrocyte catechol-O-methyltransferase. Hum Genet Suppl 1: 101–112.
Worthington, S., Turner, A., Elber, J., Andrews, P. I. (2000). 22q11 deletion and polymicrogyria:
cause or coincidence? Clin Dysmorphol 9: 193–197.
Yamagishi, H., Garg, V., Matsuoka, R., Thomas, T., Srivastava, D. (1999). A molecular pathway
revealing a genetic basis for human cardiac and craniofacial defects. Science 283: 1158–1161.
 8

Can structural magnetic resonance imaging

provide an alternative phenotype for
genetic studies of schizophrenia?
Colm McDonald and Robin M. Murray
Institute of Psychiatry, King’s College, London, UK

There is overwhelming evidence from family, adoption, and twin studies that
schizophrenia has a genetic component to its etiology. Estimates of the heritability
of schizophrenia from modern twin studies are in the region of 80% (Cardno et al.,
1999), with the remaining variance attributed to non-shared environmental factors.
Despite two decades of molecular genetics research, however, there has been slow
progress towards identifying susceptibility genes for the illness, although there have
been linkage findings replicated at some sites on the genome and potential suscep-
tibility genes within these sites have recently been reported (Stefansson et al., 2002;
Straub et al., 2002). The relative lack of success in identifying genes of major effect
for the illness, combined with the pattern of inheritance of schizophrenia, has led
many researchers to favor a multifactorial model for susceptibility, whereby liability
is continuous within the population and multiple genes interact with environmental
stressors to propel the individual over a threshold for illness expression (Gottesman,
1991). Unlike other multifactorial diseases, such as diabetes mellitus and coronary
heart disease, genetic research into schizophrenia suffers from the added disad-
vantages of having an uncertain phenotype with as yet no validating biological
test, even at postmortem, and environmental risk factors that are unknown or of
minor effect. There is growing interest, therefore, in utilizing other approaches in
the search for susceptibility genes. Given a model of continuous liability within the
population, the use of the qualitative phenotype (affected/unaffected) is often likely
to be underpowered, but it may be possible to enhance the traditional categorical
approach to phenotypic definition by identifying genetically valid traits that can
be measured quantitatively. Candidates for such traits include symptom clusters or
schizotypal traits as well as biological markers of neurobiological dysfunction or
“endophenotypes.”

Neurodevelopment and Schizophrenia, ed. Matcheri S. Keshavan et al. Published by Cambridge University
Press.

C Cambridge University Press 2004.

138
139 Magnetic resonance imaging and phenotypes

Endophenotypes are objectively measured neurobiological abnormalities that

may represent more proximal effects of susceptibility genes than the clinical pheno-
type (Wickham and Murray, 1997). Such biological markers should fulfill a number
of criteria to be useful in genetic studies (Gershon and Goldin, 1986; Leboyer et al.,
1998): (i) be heritable themselves, (ii) be associated with the illness in the general
population, (iii) be state independent (i.e. be manifest whether or not the illness is
active), (iv) cosegregate with the illness within families (i.e. among relatives who
manifest the marker, the illness is more prevalent than among those relatives who
do not), (v) be measurable in both affected and unaffected subjects, and (vi) be
found more frequently among the biological relatives of patients than in healthy
controls. Putative endophenotypes for schizophrenia include abnormalities of the
auditory evoked response, such as the P300 and P50 waves (Frangou et al., 1997a;
Siegel et al., 1984); oculomotor dysfunction, such as on smooth pursuit and anti-
saccade tasks (Crawford et al., 1998; Holzman et al., 1988); neuropsychological
deficits (Cannon et al., 1994; Toulopoulou et al., 2003); and abnormalities of brain
structure. In this chapter, we review the value of structural brain deviations iden-
tified through magnetic resonance imaging (MRI) as potential endophenotypes in
genetic studies of schizophrenia and the evidence to date that a number of specific
brain deviations are linked to susceptibility genes for schizophrenia from studies
of patients and their unaffected relatives.

The heritability of brain structure

One criticism of applying the endophenotypic approach to schizophrenia is that
several of the putative markers have not yet themselves been demonstrated to be
heritable. A powerful way of examining the extent to which a trait is heritable is
to measure how commonly the trait is shared among monozygotic (MZ) twins,
who have identical genotypes, compared with dizygotic (DZ) twins, who share on
average 50% of their genes. One of the earliest studies of brain structure in normal
twins was a computed tomographic (CT) study of subjects from the Maudsley
Twin Series (Reveley et al., 1982, 1984a). In this study, measurements of ventricular
volume in normal MZ twins were much more highly correlated than those of DZ
twins, indicating strong genetic control over ventricular size, with estimates of
heritability for ventricular size using different methods all over 80%. There have
been a number of similarly designed twin studies reported in recent years that use
higher-resolution MRI and path analysis to model heritability. In contrast to the
study by Reveley and colleagues, the heritability of lateral ventricular volume was
reported to be low in two such studies (Baare et al., 2001a; Wright et al., 2002),
with most variation in ventricular volume being a result of common environmental
effects. However, Pfefferbaum et al. (2000) reported a high heritability of 79% for
140 C. McDonald and R. M. Murray

ventricular volume in 85 elderly twin pairs, and White et al. (2002) reported a
correlation of 0.85 for ventricular volume among a sample of MZ pairs (although
this did not significantly differ from the correlation in unrelated matched controls,
0.52, which was relatively high presumably because of the influence of age and
gender over ventricular volume). Ventricular volume is known to be affected by
environmental insults such as low birth weight (Nosarti et al., 2002) and the true
extent of genetic control over ventricular volume will require further studies.
Bartley and colleagues (1997) reported higher correlations among MZ than
DZ twins for measures of cerebral volume, and subsequent statistical modeling
demonstrated that this was most likely a result of additive genetic effects rather
than shared environmental effects, with heritability estimated at 94%. This high
degree of genetic control over brain volume has now been confirmed beyond doubt
by several other MRI studies of twins (Baare et al., 2001a; Carmelli et al., 1998;
Pennington et al., 2000; Pfefferbaum et al., 2000; Tramo et al., 1998; White et al.,
2002). In one of the largest such studies to date, which included siblings of the
twins in an attempt to dissect further the contribution of genetic as distinct from
common environmental effects (since even DZ twins share the same prenatal envi-
ronment), very high heritabilities were also found for global gray matter (82%) and
white matter (88%) (Baare et al., 2001a). In another large twin study, Pennington
and colleagues (2000) demonstrated high genetic control over factors representing
volumes of cortical and especially subcortical brain regions. In a sample of elderly
male twins, there was evidence for high genetic control over lobar brain volumes,
especially right frontotemporal regions, volume of white matter intensities, and
the size of the corpus callosum (Carmelli et al., 1998, 2002; Pfefferbaum et al.,
2000), but for lower genetic control over hippocampal volume compared with the
other brain regions, the heritability of which was estimated at 40% (Sullivan et al.,
2001). As with ventricular volume, hippocampal volume is known to be suscep-
tible to environmental insults such as early hypoxia (Rutherford et al., 1995). A
recent study of MZ twins found high correlations between twins for measures of
gray matter, white matter, lobar volumes, cortical depth, cerebellum, thalamus, cau-
date, and putamen (White et al., 2002). Other twin studies reported a high degree of
genetic control over the size of the corpus callosum (Oppenheim et al., 1989; Tramo
et al., 1998). By comparinson, there appears to be a poorer correlation between
MZ twins for measures of cortical gyrification (Bartley et al., 1997; Steinmetz
et al., 1994), gyral and sulcal curvature (White et al., 2002), and asymmetry of
the planum temporale (Steinmetz et al., 1995), indicating less genetic control over
these measures of brain structure.
Wright et al. (2002), in a reanalysis of the twins previously analyzed by Bartley
et al. (1997) with whole brain voxel-based morphometry and more novel statistical
analysis, found evidence for large genetic effects shared by several bilateral brain
141 Magnetic resonance imaging and phenotypes

regions, especially paralimbic structures and the temporoparietal cortex. Also, using
automated morphometry of the cortical surface, Thompson et al. (2001) reported
a high degree of genetic control over large areas of gray matter encompassing
the frontal lobes and the sensorimotor and linguistic neocortical regions. Conse-
quently, there is considerable evidence for a high degree of genetic control over
cerebral volume and many subregions within the adult human brain as assessed
by volumetric measurement of MRI scans, which supports the value of exploring
brain structure in vivo for potential endophenotypes of complex genetic illnesses
like schizophrenia. Future twin studies will continue to dissect the genetic and
environmental contributions towards other structures of interest to schizophrenia
researchers.

Brain structure and schizophrenia

Despite heterogeneous patient samples and methodologies, neuroimaging studies
consistently identify subtle volumetric deviations in a range of brain structures
when schizophrenia patients are compared with controls. Enlargement of the lat-
eral and third ventricles is the most prominent finding: a recent meta-analysis of
58 MRI studies reported a 26% enlargement of ventricular volume in schizophrenia
subjects (Wright et al., 2000). The illness is also characterized by a mild reduction
(2%) in global cerebral volume and further volume reductions in a number of
localized regions relative to cerebral volume. These regions include the hippocam-
pus, amygdala, parahippocampal gyrus, superior temporal gyrus, prefrontal cortex,
cingulate gyrus, thalamus, and insula (McCarley et al., 1999; Wright et al., 1999,
2000). In contrast to these volume reductions, increased volume of the basal ganglia
is reported in schizophrenia. The basal ganglia are rich in dopaminergic input and
the increased volume is usually attributed to the use of conventional antipsychotic
medication, which potently blocks dopamine receptors, since it is not found in
subjects who have had minimal antipsychotic treatment and the volume decreases
on switching to an atypical antipsychotic (Chakos et al., 1995). In general, the
neuroimaging studies are supported by morphological findings from postmortem
studies (Harrison, 1999).
Despite this evidence, the full nature and extent of structural abnormalities in
schizophrenia has yet to be elucidated. Most studies to date have focused on a
relatively small number of structures using a region-of-interest approach to mor-
phometry, which is necessarily hypothesis driven and restricted to the particular
regions the investigators choose to research; however, these studies ignore many
brain regions not chosen for measurement and frequently amalgamate functionally
distinct regions, for example within an entire lobe of the brain. More recently, some
researchers have adopted alternative techniques involving whole brain analysis.
142 C. McDonald and R. M. Murray

Goldstein et al. (1999), in a volumetric analysis of the entire neocortex divided into
parcellation units, identified several regions of volume deficit, most prominently
in the middle frontal gyrus, supramarginal gyrus, and paralimbic cortices, includ-
ing the anterior cingulate, paracingulate gyri, and the insula. Other recent studies
have used computational morphometry, a range of developing methodologies that
involves automated whole brain analysis of MRI images at the voxel level. These
studies have identified volumetric reduction in a range of distributed networks
usually most prominent in fronto-temporolimbic areas and including the dorso-
lateral prefrontal cortex, frontomedial cortex, superior temporal gyrus, thalamus,
cerebellum, insula, and limbic/paralimbic regions, as well as associated white mat-
ter tracts (Ananth et al., 2002; Sigmundsson et al., 2001; Suzuki et al., 2002; Wilke
et al., 2001; Wright et al., 1999).

State versus trait

Do the structural changes vary with the stage of the illness or its treatment? Some
recent studies report greater deviation of structural changes as the illness advances,
such as progressive ventricular and sulcal enlargement, and reduced volume of
frontal and temporal regions, especially in the early stages of the illness or in those
at the severe end of the spectrum (Gur et al., 1998; Lieberman et al., 2001). How-
ever, there are methodological difficulties associated with longitudinal quantitative
neuroimaging studies, including different resolution of imaging techniques, small
sample sizes, and inadequate matching of controls and subjects; further longitudi-
nal studies are required to elucidate fully whether brain deviations worsen over the
course of the illness (Weinberger and McClure, 2002).
Whether or not certain brain changes become more extensive over time, it is
apparent that they are not solely dependent on the presence of symptomatic episodes
to become manifest. Indeed various lines of evidence indicate that several of the
structural deviations result from impairment of normal neurodevelopment.
First, many brain abnormalities are detectable in patients at or near the onset
of psychosis, including enlargement of the lateral and third ventricles and reduced
volume of cortical gray matter, the thalamus, the left hippocampus/amygdala, and
the left posterior superior temporal gyrus (DeLisi et al., 1991; Fannon et al., 2000;
Hirayasu et al., 1998).
Second, some of the morphological abnormalities identified in adult schizophre-
nia patients suggest pathophysiological processes affecting the early develop-
ing brain. These include uncommon developmental brain lesions such as aque-
duct stenosis, arachnoid and septal cysts, and agenesis of the corpus callo-
sum, which occur with excess frequency in schizophrenia (Lewis, 1990). Normal
structural cerebral asymmetries such as those of the planum temporale and
143 Magnetic resonance imaging and phenotypes

fronto-occipital petalias also develop during fetal life, and some studies have
reported loss or reversal of such asymmetries in schizophrenia (Bilder et al., 1994;
Falkai et al., 1995). Similarly, gyrification of the prefrontal cortex achieves stability
soon after birth and right frontal cortical hypergyrification has been reported in
schizophrenia (Vogeley et al., 2001).
Third, a series of correlations have been reported between early environmental
risk factors for schizophrenia and structural deviations in the adult brain. These
include early reports (Murray et al., 1985) of an association between increased
ventricle-to-brain ratio on CT scans and exposure to obstetric complications,
although other studies did not find a relationship between obstetric complications
and lateral ventricular enlargement (McGrath and Murray, 1995). Other struc-
tural abnormalities, such as reduced hippocampal volume, have also been linked
to obstetric complications (McNeil et al., 2000; Stefanis et al., 1999).
Fourth, some of the brain abnormalities have also been reported in the unaf-
fected relatives of patients, which would indicate that such abnormalities are not
restricted to the pathological process of psychosis but are a manifestation of familial
risk factors, the most likely candidates being genes influencing neurodevelopment
(see below). Therefore, several of the brain deviations linked to schizophrenia are
probably related to aberrant neurodevelopment. Since at least some of the suscepti-
bility genes for schizophrenia are likely to influence neurodevelopmental processes
(Jones and Murray, 1991), such brain deviations could present suitable alternative
phenotypes to utilize in the search for such genes.

Brain abnormalities in unaffected relatives

First-degree relatives of schizophrenia subjects share 50% of their genes with their
affected relative on average; therefore, neurobiological traits of the illness that are
linked to susceptibility genes should be found more commonly among such rel-
atives. Three types of study design have examined the extent of structural brain
abnormalities in unaffected first-degree relatives. (i) Family studies assess unaf-
fected adult relatives, usually siblings and parents who tend to have lived beyond
the usual age of onset. (ii) High-risk studies assess offspring or adolescents with
first-degree relatives with schizophrenia; usually assessment is before the age of
onset of schizophrenia and hence 10–15% of these subjects would be expected to
develop the illness over time. (iii) Twin studies usually include groups of discordant
MZ twins and are thus designed to identify non-genetic traits of the illness, but
abnormalities in unaffected co-twins compared with normal twins could represent
manifestations of genetic risk, especially if also found more commonly in MZ than
DZ unaffected co-twins.
144 C. McDonald and R. M. Murray

Ventricles
Unaffected first-degree relatives of schizophrenia patients were first included in
CT studies as a comparison group for patients in order to reduce the considerable
genetic variation of ventricular size that exists in the population. These studies
generally demonstrated greater ventricular enlargement in patients than in their
unaffected relatives (DeLisi et al., 1986; Reveley et al., 1982; Silverman et al., 1998;
Weinberger et al., 1981; Zorrilla et al., 1997). Most did not include an independ-
ent control group and thus were not able to determine whether unaffected rela-
tives themselves displayed greater ventricular enlargement than controls. Of those
sibling CT studies that did include an independent control group, Weinberger
et al. (1981) reported significantly increased ventricle-to-brain ratio in unaffected
siblings compared with controls, whereas no significant differences were found in
studies by DeLisi et al. (1986) and Silverman et al. (1998) (the latter study also failing
to detect a significant difference in this ratio between the schizophrenia subjects and
controls). Reveley et al. (1982) reported a non-significant trend for the unaffected
twins from a small group of MZ pairs discordant for schizophrenia to have larger
ventricle-to-brain ratio than healthy twins. In a Danish cohort study of 97 adult
offspring of mothers with schizophrenia (who had actually largely lived through the
risk period for schizophrenia, with a mean age of 42 years) and 60 controls (Cannon
et al., 1993), the offspring had ventricular enlargement that increased in accordance
with likely genetic risk: from having no parent with schizophrenia, to one parent
affected, to both parents affected with a schizophrenia spectrum disorder.
This finding that ventricular enlargement is most prominent in those unaffected
relatives more likely to carry susceptibility genes for schizophrenia is supported
by the Maudsley Family Study, which used high-resolution MRI and volumet-
ric assessments of ventricular size (McDonald et al. 2002; Sharma et al., 1998).
McDonald et al. (2002) analysed MRI measurements that were performed on
66 subjects with schizophrenia, 96 of their unaffected first-degree relatives, and
68 controls. A gradient of likely genetic risk was provided by these families since the
patients were divided into those from multiply affected families and those with no
family history of psychosis. The relatives of patients from multiply affected families
included a group of parents who were considered especially likely to be unaffected
gene carriers (“presumed obligate carriers”), since they appeared to transmit the
liability for schizophrenia to their affected children by virtue of also having a sibling
and/or parent affected. The patients had significantly enlarged lateral ventricles, and
so did the presumed obligate carriers, whereas those relatives from singly affected
families had similar ventricular volume to the controls. Other MRI studies do
not report significant lateral ventricular enlargement in relatives of schizophrenia
patients (Cannon et al., 1998; Lawrie et al., 1999; Seidman et al., 1999; Staal et al.,
2000), although mild ventricular enlargement that falls short of significance is often
145 Magnetic resonance imaging and phenotypes

noted. Since these studies did not focus on relatives from multiply affected families,
the lack of significant results is unsurprising. The finding that ventricular volume
enlargement is present only in those relatives most likely to carry susceptibility genes
is consistent with data from another MRI study on offspring, which reported no
overall significant difference in ventricular volume between relatives and controls,
but increasing ventricular enlargement in unaffected relatives with a stronger fam-
ily history of schizophrenia (Lawrie et al., 1999). There is also evidence from twin
studies that non-genetic factors such as obstetric complications contribute to larger
lateral ventricles in schizophrenia (Baare et al., 2001b; McNeil et al., 2000; Reveley
et al., 1984b; Suddath et al., 1990). Recent family studies point to an interaction
between genetic risk for schizophrenia and hypoxic birth events upon ventricular
volume, since subjects most likely to be gene carriers demonstrate the greatest ven-
tricular enlargement in association with obstetric complications (Cannon et al.,
1993, 2002a; McDonald et al., 2002).
Third ventricular volume tends to correlate with lateral ventricular volume, and
several studies on relatives have found that third ventricular volume is increased in
size (Keshavan et al., 1997; Lawrie et al., 2001; Seidman et al., 1999; Staal et al., 2000).
As with lateral ventricular volume, there is evidence that third ventricular volume
increases with likelihood of carrying susceptibility genes in unaffected relatives
(Lawrie et al., 1999; McDonald et al., 2002).

Global cerebral and tissue volumes

Baare et al. (2001b) studied a number of morphometric measures in 15 MZ and 14
DZ twin pairs discordant for schizophrenia and 29 matched control twin pairs; they
report that both patients with schizophrenia from discordant MZ twin pairs and
their co-twins display decrements in intracranial volume compared with normal
twins, which are not shared by DZ discordant twin pairs. The authors interpreted
their findings as indicating that reduced early brain growth is linked to schizophre-
nia susceptibility genes. Whole brain volume was reduced in both patients and
unaffected co-twins from MZ and DZ twins pairs compared with control twins,
suggesting either a genetic or a shared environmental effect on brain size. Keshavan
et al. (2002) reported a trend for reduced brain volume in high-risk offspring. How-
ever, several other studies did not find unaffected relatives to have reduced cerebral
volume compared with controls (Lawrie et al., 2001; McDonald et al., 2002; Noga
et al., 1996; Seidman et al., 2002; Staal et al., 2000).
In the Danish cohort study, Cannon et al. (1993) also found that the offspring had
enlargement of sulcal cerebrospinal fluid (CSF) spaces (implying reduced volume
of the cerebral cortex), which increased in accordance with their likely genetic risk
(i.e. in a stepwise fashion from controls to those with one parent affected, to those
having both parents affected with a schizophrenia spectrum disorder). In contrast
146 C. McDonald and R. M. Murray

to ventricular volume, there was no evidence for an interaction between genetic

risk and birth complications upon CSF spaces in this study. In a later MRI study of
75 probands with schizophrenia, 60 of their siblings and 56 controls from a Finnish
cohort, Cannon et al. (1998) found that siblings had a similar decrement in global
gray matter and increase in CSF as probands, especially in frontotemporal regions,
whereas only patients, and not their siblings, had reduced white matter volume. In
a proportion of this cohort, fetal hypoxia was associated with reduced volume of
gray matter and increased sulcal spaces in siblings and probands, especially in the
temporal region (Cannon et al., 2002a), indicating a gene–environment interaction
similar to that reported for ventricular volume. Other studies did not detect reduced
global gray matter or white matter volumes in unaffected relatives (Cannon et al.,
2002a; Seidman et al., 1999; Sharma et al., 1998; Staal et al., 2000).

Subcortical structures
Studies of unaffected siblings and offspring have found the thalamus to be reduced
in volume compared with controls (Lawrie et al., 2001; Seidman et al., 1999; Staal
et al., 1998), consistent with the increased volume of the third ventricle identified
in the same studies. There was little evidence for abnormalities of the caudate or
lentiform nuclei from these studies (Lawrie et al., 2001; Seidman et al., 1999; Staal
et al., 2000).

Temporal lobe
No evidence has been reported in unaffected relatives for abnormality of the total
temporal lobe volume (Lawrie et al., 2001; McDonald et al., 2002; Schreiber et al.,
1999) nor for the superior temporal gyrus (Frangou et al., 1997b). However, several
groups have reported that the amygdala-hippocampal formation or the hippocam-
pus is reduced in volume in unaffected relatives or offspring (Baare et al., 2001b;
O’Driscoll et al., 2001; Schreiber et al., 1999; Steel et al., 2002; van Erp et al., 2002),
especially on the left (Lawrie et al., 2001, Keshavan et al., 2002; Seidman et al., 2002).
There have also been negative studies (Staal et al., 2000) and one study reporting
an enlarged hippocampus in those relatives likely to be transmitting genetic risk
(Harris et al., 2002). Schulze et al. (2003) also failed to detect reduced volume of the
hippocampus in unaffected relatives who took part in the Maudsley Family Study
but found that a history of obstetric complications was associated with reduced
volume of the left hippocampus. Evidence for a gene – environment interaction
for birth complications upon hippocampal volume, similar to that found for ven-
tricular volume in the same sample, was provided by van Erp et al. (2002), who
found that fetal hypoxia was associated with reduced volume of the hippocampus
in probands only and not in their siblings or controls.
147 Magnetic resonance imaging and phenotypes

Frontal lobe
No evidence has been reported for abnormality of the prefrontal lobe comparing
groups of unaffected siblings or offspring of schizophrenia subjects with controls
(Keshavan et al., 2002; Lawrie et al., 2001; Schreiber et al., 1999; Staal et al., 2000).
However Baare et al. (2001b) reported that both patients with schizophrenia from
discordant MZ twin pairs and their co-twins display decrements in frontal lobe
volume compared with normal twins whereas DZ discordant twin pairs do not,
strongly suggestive of a genetic etiology. Furthermore, although no group differ-
ences were found, Lawrie et al. (2001) reported that the prefrontal lobe volume
reduces as likelihood of carrying susceptibility genes increases among high risk
adolescents. Cannon et al. (2002b), in a computational morphometry study of
three-dimensional cortical maps from the MRI scans of 10 MZ and 10 DZ twin
pairs discordant for schizophrenia and 20 matched control twin pairs from a Finnish
cohort, reported evidence for reduced volume predominantly of the anterior and
superior prefrontal cortex in accordance with likely genetic risk among unaffected
twins.

Other structures
Sharma et al. (1999) reported that unaffected relatives from multiply affected fami-
lies had loss of normal fronto-occipital brain asymmetries compared with controls,
to a lesser degree than that found in affected probands, although a subsequent
study from the same Maudsley group, using a similar design, failed to replicate this
finding (Chapple et al., 2004). There is no evidence for abnormal volume of the
cerebellum in unaffected relatives compared with controls (McDonald et al., 2002;
Seidman et al., 1999; Staal et al., 2000) nor for midsagittal area reduction of the
corpus callosum (Chua et al., 2000; Narr et al., 2002). In the Finnish twin study,
significant vertical displacement of the corpus callosum (which was associated with
lateral ventricular enlargement) was found in the unaffected co-twins as well as in
the co-twins affected with schizophrenia in MZ twins but was not found in unaf-
fected DZ twins, suggesting that this structural deviation also represents a genetic
marker for schizophrenia (Narr et al., 2002).
The true nature of the morphological phenotype of schizophrenia is likely to
involve more widespread networks than those chosen for analysis by the region-
of-interest studies performed to date on unaffected relatives, and future studies
using whole brain analysis are likely to be helpful in more comprehensively inves-
tigating these. In a recent paper, Faraone et al. (2003) studied volumetric measure-
ments of the whole brain divided into parcellation units in relatives of patients
with schizophrenia and in controls. They identified three factors loaded by several
parcellation units throughout the neocortical ribbon as well as in subcortical and
limbic regions that could be used to describe an MRI phenotype which succeeded
148 C. McDonald and R. M. Murray

in discriminating relatives from controls and was most deviant in relatives from
more densely affected families.
The varying morphometric deviations reported in unaffected relatives most likely
reflect the effect of differing study designs, samples, and methodologies, as well as
the limited statistical power associated with many of the studies: since deviations in
relatives tend to be even more subtle than those found in probands, larger numbers
of subjects are required to demonstrate a statistically significant result in studies
of relatives than when comparing only probands with controls. Despite these dis-
agreements, work to date in this field does demonstrate that unaffected relatives
frequently display abnormalities of brain structure compared with controls that are
similar in nature to, but less severe than, those associated with schizophrenia. The
greatest evidence to date is for increased ventricular volume; reduced gray matter
volume, especially in frontotemporal regions; reduced volume of the thalamus; and
reduced volume of the amygdala–hippocampal complex. These “familial” abnor-
malities identified in probands and unaffected relatives are presumably genetic in
etiology but could also represent the impact of shared environmental factors. Fur-
ther evidence for the likely genetic origin of such traits can be inferred if they are
found most commonly in those relatives who are more likely to be gene carriers.
Studies that have examined likely genetic loading of unaffected relatives, based on
number of affected relatives or genetic relatedness to the proband, have generally
found the structural abnormalities in unaffected relatives to be more prominent
in those with a higher genetic loading (Cannon et al., 1993, 2002b; Faraone et al.,
2003; Lawrie et al., 1999, 2001; McDonald et al., 2002; Seidman et al., 2002). This
finding of increased impairment in relatives in accordance with their genetic load-
ing is in keeping with the predictions of the multifactorial model for transmis-
sion of the illness. Given the impact that the known environmental risk factor of
obstetric complications may have on ventricular and hippocampal volume, which
may act interactively with susceptibility genes, future studies should collect such
information where feasible.

Molecular genetics
Few published studies to date have attempted to link measures of brain morphology
in schizophrenia with specific genetic factors. Shihabuddin et al. (1996) found that
those family members with a marker allele on chromosome 5p14.1–13.1, which
had previously been linked to schizophrenia-related disorders in a single pedigree,
had larger ventricle-to-brain ratios and greater fronto-parietal sulcal atrophy than
those who lacked the marker allele. Other studies have examined relationships
between different allele frequencies of polymorphisms in genes likely to have a
role in neurodevelopment and MRI morphometric measures. In a small sample of
patients with psychosis, Kunugi et al. (1999) reported that hippocampal volume
149 Magnetic resonance imaging and phenotypes

was reduced in patients with schizophrenia who were carriers, compared with
non-carriers, of the A3 allele of a dinucleotide repeat polymorphism in the neu-
rotrophin 3 gene. Wassink and colleagues (1999, 2000) examined the relationship
between brain morphometry and alleles of polymorphisms in the genes for brain-
derived neurotrophic factor and tumor necrosis factor receptor II in patients with
schizophrenia, their parents, and controls. The polymorphisms were not related to
the clinical schizophrenia phenotype, but subjects with at least one copy of “allele
1” of the brain-derived neurotrophic factor polymorphism had larger parietal lobes
than those who did not (Wassink et al., 1999). Subjects with schizophrenia who were
homozygous for “allele 1” of the tumor neurosis factor receptor II polymorphism
had larger ventricles and smaller frontal lobes than subjects with at least one copy
of “allele 2” (Wassink et al., 2000). Meisenzahl et al. (2001) examined the associa-
tion of MRI measurements from a sample of male patients with schizophrenia and
controls with a polymorphism in the promoter region of the gene for interleukin-
1β. The polymorphism had no influence on brain morphology in controls, but
schizophrenia patients who were “allele 2” carriers of the polymorphism had gray
matter volume deficits in frontotemporal regions and generalized white matter
deficits compared with non-carriers. In a similar sample, Rujescu et al. (2002)
examined a common biallelic polymorphism, Met129Val, in the gene coding the
prion protein (PRNP). Allele frequencies did not differ between patients and con-
trols, but homozygosity for methionine was significantly associated with decreased
white matter volume and enlarged CSF volume in the subjects, independent of
diagnosis.
The results of these molecular genetic studies must be considered very tentative
in the absence of replication. However, they demonstrate the value of using brain
morphometry as an alternative phenotype in genetic studies of schizophrenia and
suggest that a richer understanding of the actions of potential susceptibility genes
could be achieved through such a design.

Conclusions
The study of the relationship between genetics and neuroimaging represents an
emerging field of research in complex neuropsychiatric disorders. Further studies
using more advanced methodology are likely to help to clarify the brain struc-
tural manifestations of schizophrenia susceptibility genes. Research to date indi-
cates that there is a high degree of genetic control over brain morphometry, that
schizophrenia is associated with several subtle deviations of brain morphology,
which are frequently neurodevelopmental in origin, and that unaffected relatives
also demonstrate some of these deviations. The brain morphological phenotype of
schizophrenia is likely to be increasingly used as a valuable alternative to the clinical
150 C. McDonald and R. M. Murray

phenotype in molecular genetic studies and should help to simplify the genetic and
biological complexity of the illness, and thus facilitate the search for susceptibility
genes.

REFERENCES

Ananth, H., Popescu, I., Critchley, H. D., et al. (2002). Cortical and subcortical gray matter
abnormalities in schizophrenia determined through structural magnetic resonance imaging
with optimized volumetric voxel-based morphometry. Am J Psychiatry 159: 1497–1505.
Baare, W. F., Hulshoff Pol, H. E., Boomsma, D. I., et al. (2001a). Quantitative genetic modeling
of variation in human brain morphology. Cereb Cortex 11: 816–824.
Baare, W. F., van Oel, C. J., Hulshoff Pol, H. E., et al. (2001b). Volumes of brain structures in
twins discordant for schizophrenia. Arch Gen Psychiatry 58: 33–40.
Bartley, A. J., Jones, D. W., Weinberger, D. R. (1997). Genetic variability of human brain size and
cortical gyral patterns. Brain 120: 257–269.
Bilder, R. M., Wu, H., Bogerts, B., et al. (1994). Absence of regional hemispheric volume asym-
metries in first-episode schizophrenia. Am J Psychiatry 151: 1437–1447.
Cannon, T. D., Mednick, S. A., Parnas, J., et al. (1993). Developmental brain abnormalities in the
offspring of schizophrenic mothers. I. Contributions of genetic and perinatal factors. Arch
Gen Psychiatry 50: 551–564.
Cannon, T. D., Zorrilla, L. E., Shtasel, D., et al. (1994). Neuropsychological functioning in siblings
discordant for schizophrenia and healthy volunteers. Arch Gen Psychiatry 51: 651–661.
Cannon, T. D., van Erp, T. G. M., Huttunen, M., et al. (1998). Regional gray matter, white matter,
and cerebrospinal fluid distributions in schizophrenic patients, their siblings, and controls.
Arch Gen Psychiatry 55: 1084–1091.
Cannon, T. D., van Erp, T. G., Rosso, I. M., et al. (2002a). Fetal hypoxia and structural brain
abnormalities in schizophrenic patients, their siblings, and controls. Arch Gen Psychiatry 59:
35–41.
Cannon, T. D., Thompson, P. M., van Erp, T. G., et al. (2002b). Cortex mapping reveals regionally
specific patterns of genetic and disease-specific gray-matter deficits in twins discordant for
schizophrenia. Proc Natl Acad Sci USA 99: 3228–3233.
Cardno, A. G., Marshall, E. J., Coid, B., et al. (1999). Heritability estimates for psychotic disorders:
the Maudsley twin psychosis series. Arch Gen Psychiatry 56: 162–168.
Carmelli, D., DeCarli, C., Swan, G. E., et al. (1998). Evidence for genetic variance in white matter
hyperintensity volume in normal elderly male twins. Stroke 29: 1177–1181.
Carmelli, D., Swan, G. E., DeCarli, C., et al. (2002). Quantitative genetic modeling of regional
brain volumes and cognitive performance in older male twins. Biol Psychol 61: 139–155.
Chakos, M. H., Lieberman, J. A., Alvir, J., et al. (1995). Caudate nuclei volumes in schizophrenic
patients treated with typical antipsychotics or clozapine. Lancet 345: 456–457.
Chapple, B., Grech A., Sham, P., et al. (2004). Normal cerebral asymmetry in familial and non-
familial schizophrenic probands and their unaffected relatives. Schizophr Res 67: 33–40.
151 Magnetic resonance imaging and phenotypes

Chua, S. E., Sharma, T., Takei, N., et al. (2000). A magnetic resonance imaging study of cor-
pus callosum size in familial schizophrenic subjects, their relatives, and normal controls.
Schizophr Res 41: 397–403.
Crawford, T. J., Sharma, T., Puri, B. K., et al. (1998). Saccadic eye movements in families multiply
affected with schizophrenia: the Maudsley Family Study. Am J Psychiatry 155: 1703–1710.
DeLisi, L. E., Goldin, L. R., Hamovit, J. R., et al. (1986). A family study of the association of
increased ventricular size with schizophrenia. Arch Gen Psychiatry 43: 148–153.
DeLisi, L. E., Hoff, A. L., Schwartz, J. E., et al. (1991). Brain morphology in first-episode
schizophrenic-like psychotic patients: a quantitative magnetic resonance imaging study.
Biol Psychiatry 29: 159–175.
Falkai, P., Bogerts, B., Schneider, T., et al. (1995). Disturbed planum temporale asymmetry in
schizophrenia. A quantitative post-mortem study. Schizophr Res 14: 161–176.
Fannon, D., Chitnis, X., Doku, V., et al. (2000). Features of structural brain abnormality detected
in first-episode psychosis. Am J Psychiatry 157: 1829–1834.
Faraone, S. V., Seidman, L. J., Kremen, W. S., et al. (2003). Structural brain abnormalities among
relatives of patients with schizophrenia: implications for linkage studies. Schizophr Res 60:
125–140.
Frangou, S., Sharma, T., Alarcon, G., et al. (1997a). The Maudsley Family Study, II: endogenous
event-related potentials in familial schizophrenia. Schizophr Res 23: 45–53.
Frangou, S., Sharma, T., Sigmudsson, T., et al. (1997b). The Maudsley Family Study IV. Normal
planum temporale asymmetry in familial schizophrenia – A volumetric MRI study. Br J
Psychiatry 170: 328–333.
Gershon, E. S. Goldin, L. R. (1986). Clinical methods in psychiatric genetics. I. Robustness of
genetic marker investigative strategies. Acta Psychiatr Scand 74: 113–118.
Goldstein, J. M., Goodman, J. M., Seidman, L. J., et al. (1999). Cortical abnormalities in
schizophrenia identified by structural magnetic resonance imaging. Arch Gen Psychiatry
56: 537–547.
Gottesman, I. I. (1991). Schizophrenia Genesis: The Origins of Madness. New York: H Freeman.
Gur, R. E., Cowell, P., Turetsky, B. I., et al. (1998). A follow-up magnetic resonance imaging study
of schizophrenia. Relationship of neuroanatomical changes to clinical and neurobehavioral
measures. Arch Gen Psychiatry 55: 145–152.
Harris, J. G., Young, D. A., Rojas, D. C., et al. (2002). Increased hippocampal volume in
schizophrenics’ parents with ancestral history of schizophrenia. Schizophr Res 55: 11–17.
Harrison, P. J. (1999). The neuropathology of schizophrenia: a critical review of the data and
their interpretation. Brain 122: 593–624.
Hirayasu, Y., Shenton, M. E., Salisbury, D. F., et al. (1998). Lower left temporal lobe MRI volumes
in patients with first-episode schizophrenia compared with psychotic patients with first-
episode affective disorder and normal subjects. Am J Psychiatry 155: 1384–1391.
Holzman, P. S., Kringlen, E., Matthysse, S., et al. (1988). A single dominant gene can account
for eye tracking dysfunctions and schizophrenia in offspring of discordant twins. Arch Gen
Psychiatry 45: 641–647.
Jones, P., Murray, R. M. (1991). The genetics of schizophrenia is the genetics of neurodevelopment.
Br J Psychiatry 158: 615–623.
152 C. McDonald and R. M. Murray

Keshavan, M. S., Montrose, D. M., Pierri, J. N., et al. (1997). Magnetic resonance imaging and
spectroscopy in offspring at risk for schizophrenia: Preliminary studies. Prog Neuropsy-
chopharmacol Biol Psychiatry 21: 1285–1295.
Keshavan, M. S., Dick, E., Mankowski, I., et al. (2002). Decreased left amygdala and hippocampal
volumes in young offspring at risk for schizophrenia. Schizophr Res, 58: 173–183.
Kunugi, H., Hattori, M., Nanko, S., et al. (1999). Dinucleotide repeat polymorphism in the
neurotrophin-3 gene and hippocampal volume in psychoses. Schizophr Res 37: 271–273.
Lawrie, S. M., Whalley, H., Kestelman, J. N., et al. (1999). Magnetic resonance imaging of brain
in people at high risk of developing schizophrenia. Lancet 353: 30–33.
Lawrie, S. M., Whalley, H. C., Abukmeil, S. S., et al. (2001). Brain structure, genetic liability, and
psychotic symptoms in subjects at high risk of developing schizophrenia. Biol Psychiatry 49:
811–823.
Leboyer, M., Bellivier, F., Nosten-Bertrand, M., et al. (1998). Psychiatric genetics: search for
phenotypes. Trends Neurosci 21: 102–105.
Lewis, S. W. (1990). Computerised tomography in schizophrenia 15 years on. Br J Psychiatry
9(Suppl.): 16–24.
Lieberman, J., Chakos, M., Wu, H., et al. (2001). Longitudinal study of brain morphology in first
episode schizophrenia. Biol Psychiatry 49: 487–499.
McCarley, R. W., Wible, C. G., Frumin, M., et al. (1999). MRI anatomy of schizophrenia. Biol
Psychiatry 45: 1099–1119.
McDonald, C., Grech, A., Toulopoulou, T., et al. (2002). Brain volumes in familial and non-familial
schizophrenic probands and their unaffected relatives. Am J Med Genet 114: 616–625.
McGrath, J., Murray, R. (1995). Risk factors for schizophrenia: from conception to birth. In
Schizophrenia, ed. S. Hirsch, D. R., Weinberger. Oxford: Blackwell Scientific, pp. 187–205.
McNeil, T. F., Cantor-Graae, E., Weinberger, D. R. (2000). Relationship of obstetric complica-
tions and differences in size of brain structures in monozygotic twin pairs discordant for
schizophrenia. Am J Psychiatry 157: 203–212.
Meisenzahl, E. M., Rujescu, D., Kirner, A., et al. (2001). Association of an interleukin-1beta
genetic polymorphism with altered brain structure in patients with schizophrenia. Am J
Psychiatry 158: 1316–1319.
Murray, R. M., Lewis, S. W., Reveley, A. M. (1985). Towards an aetiological classification of
schizophrenia. Lancet i: 1023–1026.
Narr, K. L., Cannon, T. D., Woods, R. P., et al. (2002). Genetic contributions to altered callosal
morphology in schizophrenia. J Neurosci 22: 3720–3729.
Noga, J. T., Bartley, A. J., Jones, D. W., et al. (1996). Cortical gyral anatomy and gross brain
dimensions in monozygotic twins discordant for schizophrenia. Schizophr Res 22: 27–40.
Nosarti, C., Al-Asady, M. H., Frangou, S., et al. (2002). Adolescents who were born very preterm
have decreased brain volumes. Brain 125: 1616–1623.
O’Driscoll, G. A., Florencio, P. S., Gagnon, D., et al. (2001). Amygdala-hippocampal volume and
verbal memory in first-degree relatives of schizophrenic patients. Psychiatry Res 107: 75–85.
Oppenheim, J. S., Skerry, J. E., Tramo, M. J., et al. (1989). Magnetic resonance imaging morphol-
ogy of the corpus callosum in monozygotic twins. Ann Neurol 26: 100–104.
153 Magnetic resonance imaging and phenotypes

Pennington, B. F., Filipek, P. A., Lefly, D., et al. (2000). A twin MRI study of size variations in
human brain. J Cogn Neurosci 12: 223–232.
Pfefferbaum, A., Sullivan, E. V., Swan, G. E., et al. (2000). Brain structure in men remains highly
heritable in the seventh and eighth decades of life. Neurobiol Aging 21: 63–74.
Reveley, A. M., Reveley, M. A., Clifford, C. A., et al. (1982). Cerebral ventricular size in twins
discordant for schizophrenia. Lancet i: 540–541.
Reveley, A. M., Reveley, M. A., Chitkara, B., et al. (1984a). The genetic basis of cerebral ventricular
volume. Psychiatr Res 13: 261–266.
Reveley, A. M., Reveley, M. A., Murray, R. M. (1984b). Cerebral ventricular enlargement in
non-genetic schizophrenia: a controlled twin study. Br J Psychiatry 144: 89–93.
Rujescu, D., Meisenzahl, E. M., Giegling, I., et al. (2002). Methionine homozygosity at codon
129 in the prion protein is associated with white matter reduction and enlargement of CSF
compartments in healthy volunteers and schizophrenic patients. Neuroimage 15: 200–206.
Rutherford, M. C., Pennock, J. M., Schwieso, J. E., et al. (1995). Hypoxic ischaemic encephalopa-
thy: early magnetic resonance findings and their evolution. Neuropediatrics 26: 183–191.
Schreiber, H., Baur-Seack, K., Kornhuber, H. H., et al. (1999). Brain morphology in adolescents
at genetic risk for schizophrenia assessed by qualitative and quantitative magnetic resonance
imaging. Schizophr Res 40: 81–84.
Schulze, K., McDonald, C., Frangou, S., et al. (2003). Hippocampal volume in familial and
nonfamilial schizophrenic probands and their unaffected relatives. Biol Psychiatry 53: 562–
570.
Seidman, L. J., Faraone, S. V., Goldstein, J. M., et al. (1999). Thalamic and amygdala–hippocampal
volume reductions in first-degree relatives of patients with schizophrenia: an MRI-based
morphometric analysis. Biol Psychiatry 46: 941–954.
Seidman, L. J., Faraone, S. V., Goldstein, J. M., et al. (2002). Left hippocampal volume as a
vulnerability indicator for schizophrenia: a magnetic resonance imaging morphometric
study of nonpsychotic first- degree relatives. Arch Gen Psychiatry 59: 839–849.
Sharma, T., Lancaster, E., Lee, D., et al. (1998). Brain changes in schizophrenia – volumetric MRI
study of families multiply affected with schizophrenia: The Maudsley Family Study 5. Br J
Psychiatry 173: 132–138.
Sharma, T., Lancaster, E., Sigmundsson, T., et al. (1999). Lack of normal pattern of cerebral
asymmetry in familial schizophrenic patients and their relatives: the Maudsley Family Study.
Schizophr Res 40: 111–120.
Shihabuddin, L., Silverman, J. M., Buchsbaum, M. S., et al. (1996). Ventricular enlargement
associated with linkage marker for schizophrenia-related disorders in one pedigree. Mol
Psychiatry 1: 215–222.
Siegel, C., Waldo, M., Mizner, G., et al. (1984). Deficits in sensory gating in schizophrenic patients
and their relatives. Evidence obtained with auditory evoked responses. Arch Gen Psychiatry
41: 607–612.
Sigmundsson, T., Suckling, J., Maier, M., et al. (2001). Structural abnormalities in frontal, tem-
poral, and limbic regions and interconnecting white matter tracts in schizophrenic patients
with prominent negative symptoms. Am J Psychiatry 158: 234–243.
154 C. McDonald and R. M. Murray

Silverman, J. M., Smith, C. J., Guo, S. L., et al. (1998). Lateral ventricular enlargement
in schizophrenic probands and their siblings with schizophrenia-related disorders. Biol
Psychiatry 43: 97–106.
Staal, W. G., Hulshoff, H. E., Schnack, H., et al. (1998). Partial volume decrease of the thalamus
in relatives of patients with schizophrenia. Am J Psychiatry 155: 1784–1786.
Staal, W. G., Pol, H. E. H., Schnack, H. G., et al. (2000). Structural brain abnormalities in patients
with schizophrenia and their healthy siblings. Am J Psychiatry 157: 416–421.
Steel, R. M., Whalley, H. C., Miller, P., et al. (2002). Structural MRI of the brain in presumed
carriers of genes for schizophrenia, their affected and unaffected siblings. J Neurol Neurosurg
Psychiatry 72: 455–458.
Stefanis, N., Frangou, S., Yakeley, J., et al. (1999). Hippocampal volume reduction in schizophre-
nia: effects of genetic risk and pregnancy and birth complications. Biol Psychiatry 46: 697–
702.
Stefansson, H., Sigurdsson, E., Steinthorsdottir, V., et al. (2002). Neuregulin 1 and susceptibility
to schizophrenia. Am J Hum Genet 71: 877–892.
Steinmetz, H., Herzog, A., Huang, Y., et al. (1994). Discordant brain-surface anatomy in monozy-
gotic twins. N Engl J Med 331: 951–952.
Steinmetz, H., Herzog, A., Schlaug, G., et al. (1995). Brain asymmetry in monozygotic twins.
Cereb Cortex 5: 296–300.
Straub, R. E., Jiang, Y., MacLean, C. J., et al. (2002). Genetic variation in the 6p22.3 gene DTNBP1,
the human ortholog of the mouse dysbindin gene, is associated with schizophrenia. Am J
Hum Genet 71: 337–348.
Suddath, R. L., Christison, G. W., Torrey, E. F., et al. (1990). Anatomical abnormalities in the
brains of monozygotic twins discordant for schizophrenia. N Engl J Med 322: 789–794.
Sullivan, E. V., Pfefferbaum, A., Swan, G. E., et al. (2001). Heritability of hippocampal size
in elderly twin men: equivalent influence from genes and environment. Hippocampus 11:
754–762.
Suzuki, M., Nohara, S., Hagino, H., et al. (2002). Regional changes in brain gray and white matter
in patients with schizophrenia demonstrated with voxel-based analysis of MRI. Schizophr
Res 55: 41–54.
Thompson, P. M., Cannon, T. D., Narr, K. L., et al. (2001). Genetic influences on brain structure.
Nat Neurosci 4: 1253–1258.
Toulopoulou, T., Morris, R. G., Rabe-Hesketh, S., et al. (2003). Selectivity of verbal memory
deficit in schizophrenic patients and their relatives. Am J Med Genet 116: 1–7.
Tramo, M. J., Loftus, W. C., Stukel, T. A., et al. (1998). Brain size, head size, and intelligence
quotient in monozygotic twins. Neurology 50: 1246–1252.
van Erp, T. G., Saleh, P. A., Rosso, I. M., et al. (2002). Contributions of genetic risk and fetal
hypoxia to hippocampal volume in patients with schizophrenia or schizoaffective disorder,
their unaffected siblings, and healthy unrelated volunteers. Am J Psychiatry 159: 1514–1520.
Vogeley, K., Tepest, R., Pfeiffer, U., et al. (2001). Right frontal hypergyria differentiation in affected
and unaffected siblings from families multiply affected with schizophrenia: a morphometric
MRI study. Am J Psychiatry 158: 494–496.
155 Magnetic resonance imaging and phenotypes

Wassink, T. H., Nelson, J. J., Crowe, R. R., et al. (1999). Heritability of BDNF alleles and their
effect on brain morphology in schizophrenia. Am J Med Genet 88: 724–728.
Wassink, T. H., Crowe, R. R., Andreasen, N. C. (2000). Tumor necrosis factor receptor-II: heri-
tability and effect on brain morphology in schizophrenia. Mol Psychiatry 5: 678–682.
Weinberger, D. R., McClure, R. K. (2002). Neurotoxicity, neuroplasticity, and magnetic resonance
imaging morphometry: what is happening in the schizophrenic brain? Arch Gen Psychiatry
59: 553–558.
Weinberger, D. R., DeLisi, L. E., Neophytides, A. N., et al. (1981). Familial aspects of CT scan
abnormalities in chronic schizophrenic patients. Psychiatry Res 4: 65–71.
White, T., Andreasen, N. C., Nopoulos, P. (2002). Brain volumes and surface morphology in
monozygotic twins. Cereb Cortex 12: 486–493.
Wickham, H. Murray, R. M. (1997). Can biological markers identify endophenotypes predispos-
ing to schizophrenia? Int Rev Psychiatry 9: 355–364.
Wilke, M., Kaufmann, C., Grabner, A., et al. (2001). Gray matter-changes and correlates of disease
severity in schizophrenia: a statistical parametric mapping study. Neuroimage 13: 814–824.
Wright, I. C., Ellison, Z. R., Sharma, T., et al. (1999). Mapping of grey matter changes in
schizophrenia. Schizophr Res 35: 1–14.
Wright, I. C., Rabe-Hesketh, S., Woodruff, P. W. R., et al. (2000). Meta-analysis of regional brain
volumes in schizophrenia. Am J Psychiatry 157: 16–25.
Wright, I. C., Sham, P., Murray, R. M., et al. (2002). Genetic contributions to regional variability
in human brain structure: methods and preliminary results. Neuroimage 17: 256–271.
Zorrilla, L. T., Cannon, T. D., Kronenberg, S., et al. (1997). Structural brain abnormalities in
schizophrenia: a family study. Biol Psychiatry 42: 1080–1086.
 9

Nutritional factors and schizophrenia

Sahebarao P. Mahadik
Medical College of Georgia and VA Medical Center, Augusta, USA

Proper nutrition, in biological terms, is the dietary intake of essential nutrients,

including amino acids, lipids, minerals, vitamins, and an adequate energy sup-
ply for every cell type in the body. Such nutrition enables each cell to make the
unique proteins, lipids, and nucleic acids to carry out its function from conception
to death as dictated by the genetic program. Additionally, properly nourished cells
can adapt (i.e. change their cellular machinery through changes in gene expression)
in response to environmental factors such as heat, cold, infections, toxins, hypoxia,
etc. Poor nutrition can be considered as undernutrition, a significant reduction in
caloric intake in general rather than of one or more essential nutrients, or malnu-
trition, a lack of essential nutrients with no accompanying lack in caloric intake, is
a state in which a cell lacks or has a reduced intake of one or more essential nutri-
ents. This state can impair the ability of the cell to adapt to altered environmental
situations and can lead to the disruption of normal physiological function, which
is a disease state (e.g. cardiovascular, hepatic, diabetic, or brain disorders/diseases)
depending on the cell type.
The role of nutrition in the development and maintenance of a healthy body
and mind has been considered throughout the ages. The ancient Indian scrip-
ture Bhagavad Gita (c. 600 BC) (Ch. 6, verse 17, and Ch. 17 verses 6, 8, 9 and
10) stated that proper nutrition builds the “noble character,” which can further
improve the nutrition, from existing knowledge and experimentation and becomes
a nutritional standard for better civilization. There is a prevailing general belief
that the body and mind have synergistic influences on the choice of food eaten
and thus on nutrition. The role of nutritional factors in schizophrenia has been
considered for a long time. One hypothesis states that schizophrenia may be a
developmental systemic metabolic disorder involving membrane phospholipids
(Horrobin, 1996, 1998; Horrobin et al., 1994; Rotrosen and Wolkin, 1987). How-
ever, schizophrenia has been primarily characterized as a brain disorder giving rise

Neurodevelopment and Schizophrenia, ed. Matcheri S. Keshavan et al. Published by Cambridge University
Press.

C Cambridge University Press 2004.

156
157 Nutritional factors and schizophrenia

to serious behavioral disturbances, which cause the patient major suffering and
disability.
At this stage, the role of nutrition in the etiology of schizophrenia is circumstan-
tial and speculative; however, nutrition may play a critical role in its variable course
and outcome. Large nutritional differences, resulting from the quality and quan-
tity of food available, exist between cultures and socioeconomic classes (Mahadik
et al., 1999a). Cultures evolve by selection of food that supports the healthy body
and mind. An editorial by Murphy (1984) suggested that diseases such as general
paresis, peptic ulcer, and schizophrenia could be caused as well as cured by civiliza-
tion. Civilization, which is conceived through a process rather than established as a
state, combines single human individuals into one great unity (Freud, 1961). Torrey
(1980) has suggested a close correlation between the prevalence of schizophrenia
and the degree to which a civilization has advanced; he favored a viral theory
of the etiology of schizophrenia, since specific types of virus can be propagated,
modified, and transmitted from generation to generation. Others have blamed
the diet of a civilization for the diseases it suffers. The pathogenic factors that cause
these diseases may be incidental, rather than an essential aspect of civilization. The
nutrition, lifestyle, and viruses are unique and essential aspects of each culture
and socioeconomic class. However, since the incidence of schizophrenia is similar
in all cultures, while the course and outcome vary widely (see below), nutritional
factors may influence the underlying pathophysiology of the disease to varying
degrees.
Multinational studies coordinated by the World Health Organization (WHO)
have found that the incidence, prevalence, and manifestation of schizophrenia are
similar across different countries and cultures, but the course and outcome of the
disease vary widely from mild in developing countries to chronic in developed
countries (Jablensky et al., 1991; Sartorius et al., 1986; WHO, 1973). These findings
have led to the notion of a “universality” of schizophrenia prevalence but variability
in outcome. Several studies also have shown that the outcome of schizophrenia is
related to the socioeconomic class of the patient within a culture (Eaton, 1985;
Nandi et al., 1980). Several variables, such as social stigma, family support, type of
treatment, and family history, examined in the WHO studies could not satisfac-
torily account for the cross-national/cross-cultural variability in outcome (Craig
et al., 1997; Edgerton and Cohen, 1994; Karno and Jenkins, 1993). Recently, the
influence of culture and socioeconomic classes on various facets (i.e. epidemiol-
ogy, etiology, phenomenology, course, and outcome) of schizophrenia has been
discussed extensively (Kulhara and Chakrabarti, 2001). In addition to these factors,
viral infections, famines, and season of birth, which have all been found to be asso-
ciated with schizophrenia and its course and outcome, can significantly affect the
availability and metabolism of essential nutrients during the early stages of human
158 S. P. Mahadik

development (Horrobin et al., 1994; Mahadik et al., 1999a). Therefore, this chapter
will focus primarily on the role of essential nutritional factors such as amino acids,
lipids, vitamins, and minerals on the course and outcome of schizophrenia.

Nutrition in brain and behavioral development and maintenance

Both under- and malnutrition have been found to impair the brain and behavioral
development, as well as affect the health of the body and mind in the adult (Chafetz,
1990; Coleman and Gillberg, 1996; Dhopeshwarkar, 1983; Edelson, 1988). Gener-
ally, undernutrition refers to a significant reduction in caloric intake and not to a
reduction in essential nutrients. Although severe prenatal undernutrition has been
found to affect the brain growth and maturation of the fetus, often resulting in cog-
nitive and psychomotor retardation, undernutrition per se may not be related to
schizophrenia, since schizophrenia patients who have adequate caloric intake, both
in developed countries and in the upper socioeconomic classes of underdeveloped
countries, can suffer from a more unfavorable course and outcome. However, it
is likely that undernutrition during infections or famine may also result in mal-
nutrition at critical stages of fetal brain development. This may explain the higher
incidence of schizophrenia reported in a population in the decades following a viral
epidemic (Mednick et al., 1988) or famine (Hoek et al., 1998; Hulshoff et al., 2000).
Malnutrition refers to the lack of essential nutrients, such as essential amino acids,
lipids, vitamins, and trace elements and may be more relevant to schizophrenia
than undernutrition. It is suggested that schizophrenia may be associated with the
reduced availability or lack of some essential nutrients during a critical stage of
brain development, and a continued lack of these essential nutrients after birth and
through adolescence may contribute to an unfavorable course and outcome of the
disease.

Nutritional factors and schizophrenia

As indicated above, establishing the direct role of specific nutritional factors in
the etiology of schizophrenia and identifying strategies to prevent the disease can
be very difficult tasks at present. However, knowledge of the specific nutritional
factors that can improve the course and outcome of the disease may be critical.
Based on the nutritional data of the Food and Drug Administration (FDA), there
should be relatively less malnutrition in the USA (in those citizens receiving the
recommended daily allowance of essential amino acids, vitamins, and minerals)
(e.g. selenium, zinc, copper, manganese). However, the dietary intake of omega-3
essential fatty acids (EFAs) and antioxidant vitamins (e.g. vitamins C and E) has
steadily declined since the 1950s, primarily because of a decline in food resources
containing these nutrients and the intake of these foods (Simopoulos, 1991).
159 Nutritional factors and schizophrenia

Furthermore, studies indicate that the need for these nutrients has increased because
of significant changes in lifestyle during this period. Therefore, dietary availability
of essential amino acids and most vitamins and minerals may not be relevant to
schizophrenia. Nevertheless, a low prenatal vitamin D hypothesis for schizophrenia
has been proposed (McGrath, 1999). However, since this hypothesis is primarily
based on the higher prevalence of schizophrenia among individuals born in winter
months or at higher latitudes, this may still be related to some other dietary factors
(e.g. EFAs and antioxidants) that are reduced under these circumstances (Mahadik
et al., 1999a). Furthermore malnutrition is probably not related to defective utiliza-
tion (absorption, transport, and incorporation into biologically active molecules)
of amino acids, vitamins, and minerals, since such defects are often fatal. This brings
us to the primary issue of dietary availability of essential lipids, specifically essential
polyunsaturated fatty acids (EPUFAs) and antioxidants. The former are needed
to synthesize membrane phospholipids, particularly neural membrane phospho-
lipids, which are highly enriched in EPUFAs (Horrocks et al., 1982). Antioxidants
are needed to protect these EPUFAs from alteration by free radicals (Mahadik and
Gowda, 1996; Mahadik et al., 2001).
Since the 1950s, schizophrenia has been treated with a wide range of antipsy-
chotic drugs, and these are going to be the drugs of choice for the treatment of
schizophrenia for a long time. However, Khan et al. (2001) recently analyzed the
full FDA database on clinical trials of antipsychotic drugs and reported that the aver-
age reduction in symptoms was 17.3% when a conventional antipsychotic drug was
administered, and 16.6% when a newer antipsychotic drug was administered. This,
and the earlier finding by Hegarty et al. (1994) that long-term favorable outcomes
in terms of employment and reintegration into the community do not significantly
differ from 100 years ago, suggests that current therapeutic strategies have limita-
tions in correcting the neuropathophysiology of schizophrenia and in improving
the psychopathology associated with the disease. In addition, antipsychotic treat-
ments, in general, are associated with numerous morbidities, including weight gain,
insulin resistance, cardiovascular problems, and abnormal lipid metabolism, which
result in a diminished quality of life and even in increased mortality (McIntyre
et al., 2001). Increasing evidence suggests that dietary supplementation with selec-
tive essential nutritional factors, such as, essential fatty acids and antioxidants, may
significantly improve the course and outcome of schizophrenia (Fenton et al., 2000;
Horrobin, 1998; Mahadik and Evans, 1997; Mahadik and Gowda, 1996; Mahadik
and Scheffer, 1996; Mahadik et al., 2001; Reddy and Yao, 1996).

Essential polyunsaturated fatty acids and schizophrenia

Burr and Burr (1929) established the dietary essentiality of the EPUFAs. There
are two types, omega-6 and omega-3, so named for the position of the carbon-6
160 S. P. Mahadik

and carbon-3 chemical double bonds on each molecule, respectively. Algae and
plants exclusively make these EPUFAs. All animals actually consume the precursors
of EPUFAs (linolenic acid [LA; C18:2n6] and α linolenic acid [ALA; C18:3n3] or
their predominant functional analogs (arachidonic acid [AA] and docosahexaenoic
acid [DHA], respectively). AA and DHA make 50% of the fatty acids that are
attached to brain phospholipids, which comprise almost 60% of brain mass (Suzuki,
1981).

Altered fatty acid metabolism in schizophrenia

Extensive work on membrane phospholipids and their EPUFAs in schizophrenia
has established that AA and DHA metabolism is altered beginning in the very
early stages of schizophrenia (Horrobin, 1998; Horrobin et al., 1994; Keshavan
et al., 1993; Mahadik et al., 1994, 1996a,b; Rotrosen and Wolkin, 1987; Yao, 1999;
Yao et al., 1994). Christinsen and Christinsen (1988) were the first to identify
the relationship between the outcomes of schizophrenia patients and variation in
the intake of dietary fat. They observed that 98% of the variation in the course
of schizophrenia could be explained by variations in fat intake. The course of
schizophrenia was better in developing countries with a low-fat diet in which the
major dietary source of fat was from vegetables and seafood (low saturated fatty acid
and high EPUFA contents) compared with the course of the disease in developed
countries, with high-fat diets in which the major source of the fat consumed was
from land animals and birds (high saturated fatty acids and low EPUFA contents).
Glen et al. (1994) reported reduced levels of AA and DHA in patients with pre-
dominantly negative symptoms. Recently, a few studies have found that the levels
of AA and DHA in the membranes of red blood cells correlated with the sever-
ity (based on symptomatic scores) of schizophrenia (Arvindakshan et al., 2003a;
Khan et al., 2002); such changes parallel changes in brain membranes under a
variety of pathophysiological conditions. In addition, improving the levels of AA
and DHA in red blood cell membranes by supplementation or by increased dietary
intake has been found to correlate with improved clinical outcomes in patients
with schizophrenia (Arvindakshan, 2003; Arvindakshan et al., 2003a; Peet and
Horrobin, 2002; Peet et al., 1995, 2001). These studies strongly indicate that AA
and DHA may have a critical role in the pathophysiology and symptomatology
of schizophrenia. However, the Western diet is very rich in AA. Therefore, only
DHA, and its precursors ALA and eicosapentaenoic acid (EPA) may be critical for
schizophrenia.

Precursors of brain phospholipids and neurodevelopment in schizophrenia

Both AA and DHA are critical for brain and behavioral development (Innis, 1991;
Simopoulos, 1991; Wainwright, 1992), which has consistently been found to be
161 Nutritional factors and schizophrenia

abnormal in patients with schizophrenia (Bloom, 1993; Keshavan and Murray,

1997; Murray, 1994; Raedler et al., 1998; Weinberger, 1996). Dietary depletion of
AA and DHA during pregnancy in animals, and the resulting lack of availabil-
ity to the fetus, has been shown to cause reduced body weight, head size, brain
weight, and cognitive deficits (Crawford, 1992; Neuringer et al., 1986; Wainwright,
1992; Yamomoto et al., 1987). Similar changes, including cognitive deficits, are
also found in patients with schizophrenia (Chua and McKenna, 1995; Lawrie and
Abukmeil, 1998; Tollefson, 1996). Specifically, these studies indicate differential
reductions in volumes of certain brain regions (probably secondary to reduced
number of neurons and their processes), disorganized neuronal networks, and
increased ventricular size; all of which are believed to predate the onset of ill-
ness. In addition, an increased “pruning” (excessive removal of nerve endings
and processes during pubertal maturation) has been proposed (Feinberg, 1990;
Keshavan et al., 1994), which is supported by postmortem studies (Glantz and
Lewis, 2000; Selemon et al., 1995). AA and DHA are selectively enriched in the
neuropils and synapses of the cortex, hippocampus, and basal ganglia (Carlson
et al., 1986; O’Brien and Sampson, 1965). Schizophrenia has been associated
with significant morphological abnormalities in these brain regions (Heckers,
1997).

Membrane signal transduction

AA and DHA are critical constituents of neuronal membrane phospholipids
(Horrocks et al., 1982; Thompson, 1992). Using nuclear magnetic resonance spec-
troscopy, altered brain membrane phospholipid metabolism has been reported in
both first-episode psychotic patients and chronic medicated schizophrenia patients
(Fukuzako et al., 1996; Pettegrew et al., 1991; Stanley et al., 1995). These phospho-
lipids are critical for maintaining membrane fluidity appropriate for function-
ing of membrane receptors. Specific phospholipids are hydrolyzed by receptor-
mediated processes, generating intermediates that act as second messengers
(e.g. diacylglycerol, inositol polyphosphates, AA, prostaglandins, and cytokines)
(Axelrod, 1990; Berridge, 1981; Horrobin et al., 1994; Rana and Hokin, 1990).
Particularly, dopamine receptor activation of the AA cascade seems to be a basis
for D1 /D2 receptor synergism (Piomelli et al., 1991). Several investigators have
also reported altered phosphoinositol turnover after thrombin receptor activa-
tion of platelets in schizophrenia (Essali et al., 1990; Kaiya et al., 1989; Yao et al.,
1992). Altered membrane receptor-mediated signal transduction of several neu-
rotransmitters and growth factors has been considered in schizophrenia (Hudson
et al., 1993). This is probably a result of alterations in second messenger-derived
membrane phospholipids, since the effects of several transmitters (e.g. dopamine,
serotonin, acetylcholine, norepinephrine, glutamate, and gamma-aminobutryric
162 S. P. Mahadik

acid) (Axelrod, 1990) and growth factors (Virdee et al., 1994) are mediated by these
mechanisms.

Other physiological roles of phospholipid precursors relevant to schizophrenia

In addition to the possible role of AA and DHA in the structure and function of
neural membranes in schizophrenia, these fatty acids, combined with antioxidants,
play critical roles in several important cellular processes and medical morbidities.
DHA has been shown to prevent apoptosis under a variety of pathophysiological
conditions (Akbar and Kim, 2002; Brand et al., 2000). This may be important
since increased apoptosis has been proposed in schizophrenia (Margolis et al.,
1994). DHA is effectively converted into EPA, a precursor for prostaglandins that
plays an important role as an anti-inflammatory factor; abnormal prostaglandin
metabolism has been reported in schizophrenia (Horrobin et al., 1994). DHA has
also been found to regulate a large number of genes for transcription factors, genes
involved in sterol, including cholesterol metabolism, and neuroprotective genes
such as trophic factors and anti-apoptotic factors (Ntambi and Bene, 2001). The
role of DHA in apoptosis and inflammatory processes, combined with its beneficial
effects in controlling weight gain, diabetes, hypertension, autoimmune disorders,
certain cancers, and preventing neurodegeneration (Simopoulos, 1991), makes it a
very important nutritional factor in schizophrenia, which is associated with many
of these morbidities.

Factors affecting membrane polyunsaturated fatty acid

metabolism in schizophrenia
Intake of polyunsaturated fatty acids
Both prenatal and postnatal membrane EPUFA status is critical in the etiology of
schizophrenia. Many factors influence the dietary intake of EPUFAs by an individ-
ual: the season in which a person is conceived and born (which significantly affects
the availability of food rich in EPUFAs), family size, birth order, whether the person
was breast-fed as an infant, culture, socioeconomic status, and domicile (urban or
rural) (Mahadik et al., 1999a). Breast-feeding (mother’s milk is very rich in DHA,
which was not present in bottle formula until recently) has been thought to be
associated with lower incidence and better outcome of schizophrenia (McCreadie,
1997). Similarly, the levels of EPUFA in membranes are influenced by a patient’s
dietary availability and intake of EPUFAs and antioxidants, lifestyle, medications,
family support, and illness (Mahadik and Evans, 1997; Mahadik et al., 1999a). This
is particularly important since the quality and quantity of a patient’s diet depends
on the family or care providers. Lifestyle, which includes habits such as alcohol
consumption, smoking, and drug use, and level of physical activity influences the
163 Nutritional factors and schizophrenia

metabolism of EPUFAs (Mahadik et al., 2001). However, since the 1970s with
increased migration of population and very significant changes in food production
and delivery, some of the influences of season, culture, and family-related factors
are being attenuated.

Oxidative breakdown of polyunsaturated fatty acids

Peroxidative breakdown of EPUFA by free radicals, also called reactive oxygen
species (e.g. O2 •− , • OH, OH− , • NO and ONOO− ), may be a very important fac-
tor in the pathophysiology of schizophrenia (Mahadik and Mukherjee, 1996; Reddy
and Yao, 1996). EPUFAs are selectively susceptible to peroxidation. Increased oxida-
tive stress and lipid peroxidation is associated with schizophrenia (Mahadik et al.,
1999b). Furthermore, typical antipsychotic drugs have pro-oxidant effects and atyp-
ical drugs have antioxidant effects (Parikh et al., 2002). Membrane AA and DHA
levels in patients treated with atypical antipsychotic drugs such as clozapine have
been found be higher than levels in patients taking haloperidol (Horrobin, 1999;
Khan et al., 2002). In addition, some newer antipsychotic drugs have been found to
increase the patient’s caloric intake significantly, which can increase the oxidative
stress and peroxidative EPUFA breakdown. This drug effect is unhealthy and pro-
oxidant, particularly when combined with a lifestyle involving little or no exercise,
high levels of smoking and drinking alcohol, and a high-fat diet (Brown et al.,
1999; Scottish Schizophrenia Research Group, 2000). In animals, caloric restriction
has been shown to reduce the oxidative cellular damage associated with a regular
high-caloric diet (Sohal et al., 1994) and also to improve learning and memory
(Forster et al., 1996). However, peroxidative breakdown of EPUFA can be effec-
tively prevented by dietary supplementation of antioxidants such as β-carotene,
quinones, flavones, lycopenes, and vitamins E and C, which contain antioxidants
to prevent oxidative damage to the membrane lipid (hydrophobic) and cytoso-
lic (hydrophilic) constituents of the EPUFA (Mahadik and Gowda, 1996). Several
studies have reported that supplementation with vitamin E primarily improves the
tardive dyskinesia symptomatology (Adler et al., 1999; Mahadik and Gowda, 1996;
Mahadik and Scheffer, 1996 Reddy and Yao, 1996). Only one study found that sup-
plementation with vitamin E reduced the lipid peroxides, increased the membrane
EPUFAs, and improved the clinical outcome of schizophrenia (Peet et al., 1993).
Recent studies have found that reduced levels of EPUFAs and increased levels of
peroxides correlated with the severity of schizophrenia (Arvindakshan et al., 2003a;
Khan et al., 2002). The antioxidants primarily prevent the peroxidation of EPUFAs.
Therefore, dietary supplementation with a combination of antioxidants (e.g. vita-
min E and C) and EPUFAs, particularly EPA and DHA, has been suggested to
improve the outcome of schizophrenia (Mahadik et al., 2001). One study that
used such supplementation has reported very significantly improved and sustained
164 S. P. Mahadik

outcomes in a racially homogenous group of schizophrenia patients (Arvindakshan

et al., 2003b).

Strategies to address nutritional deficits in schizophrenia

At present the primary concerns for nutritional factors in schizophrenia relate to
intake of EPUFAs, particularly the omega-3 EFAs, and antioxidants, such as vita-
mins E and C. It may be important to examine the therapeutic value of these supple-
ments as an adjunctive to pharmacotherapy. However, all the EPUFA supplementa-
tion studies so far have been carried out with patients with chronic schizophrenia,
who have been treated with a variety of antipsychotic drugs. These studies have
indicated several critical issues that must be considered in designing and carrying
out further research.

Factors influencing supplementation

Age and duration of illness
The age of the patient may be an important issue, since age has been associated
with the body’s antioxidant defense. Furthermore, the number of years of illness
and treatment, particularly with typical antipsychotic drugs, may lead to a state
of membrane pathology that may be difficult to correct. Fenton et al. (2000), who
did not find any therapeutic effect from administering EPUFAs, have suggested
supplementation in younger patients in their early stages of illness.

Adjunctive medication
Typical antipsychotic drugs such as haloperidol have pro-oxidant properties
(Jeding et al., 1995). These medications can also affect EPUFA metabolism
(Horrobin et al., 1994). The newer atypical antipsychotic drugs have antioxidant
effects, particularly clozapine, which has been found to improve membrane EPUFA
levels (Arvindakshan et al., 2003a; Horrobin, 1999; Khan et al., 2002). Recently, we
have found that haloperidol reduced the antioxidant enzymes and increased the
lipid peroxides in the rat brain; however, none of the atypical drugs such as cloza-
pine, olanzapine, and risperidone did this (Parikh et al., 2002).

Type of essential polyunsaturated fatty acid

Earlier studies have reported the efficacy of administering omega-6 fatty acids
(Vaddadi et al., 1989). Recently, Peet and Horrobin (2002) reported that the omega-
3 fatty acid EPA alone might be adequate to correct the membrane EPUFA deficits
and improve psychopathology in schizophrenia. However, the use of a mixture of
EPA and DHA may still be preferred, since a large percentage of EPA is converted
165 Nutritional factors and schizophrenia

to prostaglandin and may not be adequate as a substrate for DHA, which is a

predominant omega-3 fatty acid in neuronal membranes.

Dose and the quality of omega-3 fatty acid

Dose is significantly influenced by the patient’s lifestyle and dietary status. Fur-
thermore, since the omega-3 EPUFA preparations are unstable, the quality must
be carefully monitored. If not balanced with dietary antioxidants, high doses of
EPUFAs may increase the levels of peroxides, which are known to be toxic to sev-
eral plasma membrane functions including neurotransmitter signal transduction
(Rafalowska et al., 1989).

Duration of treatment
Earlier studies have suggested that supplementation for at least 4 months is required
to restore the level of EPUFA in red blood cells and brain membranes to a steady state
(Mahadik and Evans, 1997). However, one study indicated that supplementation
with EPA had therapeutic effects within a few weeks, indicating a possible indirect
role for EPA and/or its metabolites (Peet and Horrobin, 2002).

Placebo control
Since psychiatric disorders show a significant placebo effect on symptom reduction,
well-designed, placebo-controlled, dose-ranging replication studies would be crit-
ical to establish the therapeutic benefits of omega-3 EPUFAs (Peet and Horrobin,
2002).

Preferred supplementation components

Based on the concept that the ongoing oxidative breakdown of cell membrane
EPUFAs must be stopped and their levels maintained to restore and preserve neural
membrane function in schizophrenia, the use of a combination of antioxidants to
prevent peroxidative damage and EPUFAs to restore membrane phospholipids
may be necessary for optimal treatment of schizophrenia. Unfortunately, studies
reported so far involve the administration of either an antioxidant, which has been
exclusively vitamin E alone (Adler et al., 1999; Mahadik and Mukherjee, 1996; Reddy
and Yao, 1996), or of EPUFAs accompanied by only trace amounts of antioxidants
(Fenton et al., 2000; Mahadik and Evans, 1997). Only one study has reported using
a dietary supplementation with EPA (180 mg) and DHA (120 mg) plus vitamins E
(400 iu) and C (500 mg) (Arvindakshan et al., 2003b). There was a highly significant
reduction in symptomatology in patients with chronic schizophrenia after taking
the supplement twice a day for 4 months. It is also important to note that all the
patients in the study were racially homogeneous, and they had previously had a
very similar dietary intake of EPUFAs (based on diet questionnaire) and lifestyle.
166 S. P. Mahadik

When and what should be treated with antioxidants and essential

polyunsaturated fatty acids?
The above discussion indicates that oxidative cell injury and altered EPUFA
metabolism could conceivably predate the illness, contribute to the onset of psy-
chopathologies, and continue and even worsen during the course of schizophrenia.
Supplementation with antioxidants and EPUFAs may, therefore, play a critical role
in altering the onset and improving the course and outcome of schizophrenia. Since
typical antipsychotic drugs are still the drugs of choice for the effective control of
psychosis, the most serious symptom of schizophrenia, adjunctive treatment with
antioxidants and EPUFAs may prove important in the early course of schizophre-
nia, in its outcome, and in the reduction of serious comorbidities. EPUFAs have
also been found to be lower in children with attention-deficit hyperactivity dis-
order (Arnold et al., 1994; Stevens and Burgess, 1999), a disorder that may be
a premorbid indicator of a person’s risk for developing schizophrenia (Keshavan
et al., 2003a). Interestingly, a recent study has suggested that children at genetic risk
for schizophrenia had evidence for decreased membrane phospholipid metabolism
in the prefrontal cortex, as examined by phosphorus-31 magnetic resonance spec-
troscopy (Keshavan et al., 2003b). These alterations were more marked in children
and adolescents with attention-deficit hyperactivity disorder and other neurode-
velopmental disorders. These observations suggest the possible benefit of very early
supplementation with antioxidants and EPUFAs early in schizophrenia and in indi-
viduals at risk for this illness.

Conclusions
Since proper nutrition regulates the normal development of body and mind, and
maintains their health throughout a person’s life, it can play a critical role in influ-
encing the pathophysiology, and thereby the psychopathology, of schizophrenia.
However, based on global epidemiological data, nutrition can primarily be con-
sidered as a means of improving the course and outcome of the disease. Proper
nutrition is a complex issue, particularly for schizophrenia patients, whose dietary
intake can be affected by lifestyle, illness, medications, support services, and other
yet unknown metabolic genetic defects. Reports on diet in developed countries such
as the USA indicate that there is malnutrition (lack or reduced intake) with respect to
dietary intake of omega-3 EPUFAs and antioxidants. This and the increased oxida-
tive stress and reduced levels of membrane phospholipid omega-3 EFAs from very
early stages of schizophrenia illness suggest that early nutritional supplementation
of omega-3 EFAs and antioxidants could significantly improve the clinical course
and outcome of patients; such supplementation may also potentially diminish the
167 Nutritional factors and schizophrenia

likelihood of other medical morbidities such as weight gain, hypertension, diabetes,

immune disorders, and some cancers associated with schizophrenia.

REFERENCES

Adler, L. A., Rotrosen, J., Edson, R., et al. (1999). Vitamin E treatment for tardive dyskinesia. Arch
J Psychiatry 56: 836–841.
Akbar, M., Kim, H. Y. (2002). Protective effects of docosahexaenoic acid in staurosporine-induced
apoptosis: involvement of phosphatidylinositol-3 kinase pathway. J Neurochem 82: 655–665.
Arnold, L. E., Kleykamp, D., Votolato, N., Gibson, R. A., Horrocks, L. (1994). Potential link
between dietary intake of fatty acids and behavior: pilot exploration of serum lipids in
attention-deficit hyperactivity disorder. J Child Adoles Psychopharmacol 4: 171–182.
Arvindakshan, M. (2003). The role of membrane essential fatty acids in schizophrenia outcome.
Ph.D. Thesis, Pune University India.
Arvindakshan, M., Sitasawad, S., Debsikdar, V., et al. (2003a). Membrane essential polyunsatu-
rated fatty acids (EPUFA) and schizophrenia outcome: EPUFA and lipid peroxide levels in
never-medicated and medicated schizophrenics. Biol Psychiatry 53: 56–64.
Arvindakshan, M., Ghate, M., Ranjekar, P. K., Evans, D.R., Mahadik, S. P. (2003b). Supplemen-
tation with a combination of omega-3 fatty acids and antioxidants (vitamins E and C)
improves the outcome of schizophrenia. Schizophr Res 62: 195–204.
Axelrod, J. (1990). Receptor-mediated activation of phospholipase A2 and arachidonic acid
release in signal transduction. Biochem Soc Trans 18: 503–507.
Berridge, M. J. (1981). Phosphatidylinositol hydrolysis: a multifunctional transducting mecha-
nism. Mol Cell Endocrinol 24: 115–140.
Bloom, F. (1993). Advancing neurodevelopmental origin for schizophrenia. Arch Gen Psychiatry
50: 224–227.
Brand, A., Gil, S., Yavin, E. (2000). N-Methyl bases of ethanolamine prevent apoptotic cell death
induced by oxidative stress in cells of oligodendroglia origin. J Neurochem 74: 1596–1604.
Brown, S., Birtwistle, J., Roe, L., Thompson, C. (1999). The unhealthy lifestyle of people with
schizophrenia. Psychol Med 29: 697–701.
Burr, G. O., Burr, M. M. (1929). A new deficiency disease produced by the rigid exclusion of fat
from the diet. J Biol Chem 82: 345–367.
Carlson, S. E., Rhodes, P. G., Ferguson, M. G. (1986). Docosahexaenoic acid status of preterm
infants at birth and following feeding with human milk or formula 1–3. Am J Clin Nutr 44:
798–804.
Chafetz, M. D. (1990). Nutrition and Neurotransmitters: The Nutrient Bases of Behavior. Engle-
wood Cliffs, NJ: Prentice Hall.
Christinsen, O., Christinsen, E. (1988). Fat consumption and schizophrenia. Acta Psychiatr Scand
78: 587–591.
Chua, S. E., McKenna, P. J. (1995). Schizophrenia: a brain disease? A critical review of structural
and cerebral abnormalities in the disorder. Br J Psychiatry 166: 563–582.
168 S. P. Mahadik

Coleman, M., Gillberg, C. (1996). The Schizophrenias. A Biological Approach to the Schizophrenia
Spectrum Disorders. New York: Springer.
Craig, T. J., Siegel, C., Hopper, K., Lin, S., Sartorius, N. (1997). Outcome in schizophrenia and
related disorders compared between developing and developed countries. Br J Psychiatry
170: 229–233.
Crawford, M. A. (1992). Essential fatty acids and neurodevelopmental disorder. In Neurobiology
of Essential Fatty Acids, ed. N. G. Bazan. New York: Plenum Press, pp. 307–314.
Dhopeshwarkar, G. A. (1983). Nutrition and Brain Development. New York: Plenum Press.
Eaton, W. W. (1985). Epidemiology of schizophrenia. Epidemiol Cal Rev 7: 105–126.
Edelson, E. (1988). Nutrition and the Brain. New York: Chelsea House.
Edgerton, R. B., Cohen, A. (1994). Culture and schizophrenia: the DOSMD challenge. Br J
Psychiatry 164: 222–231.
Essali, M. A., Das, R., de Belleroche, J., Hirsch, S. R. (1990). The platelets polyphosphoinosi-
tide system in schizophrenia: the effects of neuroleptic treatment. Biol Psychiatry 28: 478–
487.
Feinberg, I. (1990). Cortical pruning and the development of schizophrenia. Schizophr Bull 16:
567–568.
Fenton, W. S., Hibbeln, J., Knable, M. (2000). Essential fatty acids, lipid membrane abnormalities,
and the diagnosis and treatment of schizophrenia. Biol Psychiatry 47: 8–21.
Forster, M. J., Dubey, A., Dawson, K. M., et al. (1996). Age-related losses of cognitive and motor
skills in mice are associated with oxidative protein damage in the brain. Proc Natl Acad Sci,
USA 93: 4765–4769.
Freud, S. (1961). Civilization and its discontents. In The Standard Edition of the Com-
plete Psychological Work of Sigmund Freud, Vol. 21, ed. J. Strachey. New York: Norton,
pp. 59–145.
Fukuzako, H. Fukuzako, T., Takeuchi, K., et al. (1996). Phosphorus magnetic resonance spec-
troscopy in schizophrenia: correlation between membrane PL metabolism in the temporal
lobe and positive symptoms. Progr Neuropsychopharmacol Biol Psychiatry 20: 629–640.
Glantz, L. A., Lewis, D. A. (2000). Decreased dendritic spine density on prefrontal cortical pyra-
midal neurons in schizophrenia. Arch Gen Psychiatry 57: 65–73.
Glen, A. I. M., Glen, E. M. T., Horrobin, D. F., et al. (1994). A red cell membrane abnormality in
a subgroup of schizophrenic patients: evidence for two diseases. Schizophr Res 12: 53–61.
Heckers, S. (1997). Neuropathology of schizophrenia: cortex, thalamus, basal ganglia, and
neurotransmitter-specific projection system. Schizophr Bull 23: 403–421.
Hegarty, J. D., Baldessarini, R. J., Tohen, M., Waternaux, C., Oepen, G. (1994). One hundred years
of schizophrenia: a meta-analysis of the outcome literature. Am J Psychiatry 151: 1409–1416.
Hoek, H. W., Brown, A. S., Susser, E. (1998). The Dutch famine and schizophrenia spectrum
disorders. Soc Psychiatry Psychiatr Epidemiol 33: 373–379.
Horrobin, D. F. (1996). Schizophrenia as a membrane lipid disorder which is expressed through-
out the body. Prostaglandins, Leukot Essent Fatty Acids 55: 3–8.
(1998). The membrane phospholipid hypothesis as a biochemical basis for the neurodevelop-
mental concept of schizophrenia. Schizophr Res 30: 193–208
169 Nutritional factors and schizophrenia

(1999). The effects of antipsychotic drugs on membrane phospholipids: a possible novel mech-
anism of action of clozapine. In Phospholipid Spectrum Disorders in Psychiatric, ed. M. Peet,
A. L. Glen, D. Horrobin. Carnforth, UK: Marius Press, pp. 113–117.
Horrobin, D. F., Glen, A. I. M., Vaddadi, K. S. (1994). The membrane hypothesis of schizophrenia.
Schizophr Res 13: 195–208.
Horrocks, L. A., Ansell, G. B., Porcellati, G. (ed.) (1982). Phospholipids in the Nervous System,
Vol. 1, Metabolism. New York: Raven Press.
Hudson, C. J., Young, L. T., Li, P. P., Warsh, J. J. (1993). CNS signal transduction in the patho-
physiology and pharmacology of affective disorders and schizophrenia. Synapse 13: 278–
293.
Hulshoff Pol, H. E., Hoek, H. W., Susser, E., et al. (2000). Prenatal exposure to famine and brain
morphology in schizophrenia. Am J Psychiatry 157: 1170–1172.
Innis, S. M. (1991). Essential fatty acids in growth and development. Progr Lipid Res 30: 39–
103.
Jablensky, A., Sartorius, N., Ernberg, G., et al. (1991). Schizophrenia: manifestations, incidence
and course in different cultures. Psychol Med Monogr Suppl 20: 1–97.
Jeding, I., Evans, P. J., Akanmu, D., et al. (1995). Characterization of the potential antioxidant
and pro-oxidant actions of some neuroleptic drugs. Biochem Pharmacol 49: 359–365.
Kaiya, H., Nishida, A., Imai, A., Nakashima, S., Nozawa, Y. (1989). Accumulation of diacylglyc-
erol in platelet phosphoinositides turnover in schizophrenia: a biological marker of good
prognosis? Biol Psychiatry 26: 669–676.
Karno, M., Jenkins, J. H. (1993). Cross-cultural issues in the course and treatment of schizophre-
nia. Psych Clin North Am 16: 339–350.
Keshavan, M. S., Murray, R. M. (1997). Neurodevelopment and Adult Psychopathology. Cambridge,
UK: Cambridge University Press.
Keshavan, M. S., Mallinger, A. G., Pettegrew, J. W., Dippold, C. (1993). Erythrocyte membrane
phospholipids in psychotic patients. Psychiatr Res 49: 9–95.
Keshavan, M. S., Anderson, S., Pettegrew, J. W. (1994). Is schizophrenia due to excessive synaptic
pruning in the prefrontal cortex? The Feinberg hypothesis revisited. J Psychiatr Res 28:
239–265.
Keshavan, M. S., Sujata, A., Mehra, M., Montrose, D. M., Sweeney, J. A. (2003a). Psychosis prone-
ness and ADHD in young relatives of schizophrenia patients. Schizophr Res 59: 85–92.
Keshavan, M. S., Stanley, J. A., Montrose, D. M., Minshew, N. J., Pettegrew, J. W. (2003b). Pre-
frontal membrane phospholipid metabolism of child and adolescent offspring at risk for
schizophrenia or schizoaffective disorder: an in vivo 31 P MRS study. Mol Psychiatry 8: 316–
323.
Khan, A., Khan, S. R., Leventhal, R. M., Brown, W. A. (2001). Symptom reduction and suicide
risk among patients treated with placebo in antipsychotic clinical trials: an analysis of the
Food and Drug Administration database. Am J Psychiatry 158: 1449–1454.
Khan, M. M., Evans, D. R., Gunna, V., et al. (2002). Reduced erythrocyte membrane essential
fatty acids increased lipid peroxides in schizophrenia at the never-medicated first-episode
of psychosis and after years of treatment with antipsychotics. Schizophr Res 58: 1–10.
170 S. P. Mahadik

Kulhara, P., Chakrabarti, S. (2001). Culture and schizophrenia and other psychotic disorders.
Psychiatr Clin North Am 24: 449–464.
Lawrie, S. M., Abukmeil, S. S. (1998). Brain abnormality in schizophrenia: a systematic and
quantitative review of volumetric magnetic resonance imaging studies. Br J Psychiatry 172:
119–120.
Mahadik, S. P., Evans, D. (1997). Essential fatty acids in the treatment of schizophrenia. Drugs
Today 33: 5–17.
Mahadik, S. P., Gowda, S. (1996). Antioxidants in the treatment of schizophrenia. Drugs Today
32: 1–13.
Mahadik, S. P., Mukherjee, S. (1996). Free radical pathology and the antioxidant defense in
schizophrenia. Schizophr Res 19: 1–18.
Mahadik, S. P., Scheffer, R. E. (1996). Oxidative injury and potential use of antioxidants in
schizophrenia. Prostaglandins, Leukot Essent Fatty Acids 55: 45–54.
Mahadik, S. P., Mukherjee, S., Correnti, E., et al. (1994). Distribution of plasma membrane
phospholipids and cholesterol in skin fibroblasts from drug-naive patients at the onset of
psychosis. Schizophr Res 13: 239–247.
Mahadik, S. P., Mukherjee, S., Horrobin, D., et al. (1996a). Plasma membrane phospholipid fatty
acid composition of cultured skin fibroblasts from schizophrenic patients: comparison with
bipolar and normal controls. Psychiatr Res 63: 133–142.
Mahadik, S. P., Shendarkar, N. S., Scheffer, R., Mukherjee, S., Correnti, E. E. (1996b). Utilization
of precursor essential fatty acids in culture by skin fibroblasts from schizophrenic patients
and normal controls. Prostaglandins, Leukot Essent Fatty Acids 55: 65–70.
Mahadik, S. P., Mulchandani, M., Hegde, M. V., Ranjekar, P. K. (1999a). Cultural and socioeco-
nomic differences in dietary intake of essential fatty acids and antioxidants-effects on the
outcome. In Phospholipid Spectrum Disorders in Psychiatric, ed. D., Horrobin, A. L. Glen,
M. Peet. Carnforth, UK: Marius Press, pp. 167–179.
Mahadik, S. P., Sitasawad, V., Mulchandani, M. (1999b). Membrane peroxidation and the neu-
ropathology of schizophrenia. In Phospholipid Spectrum Disorders in Psychiatric, ed. D.
Horrobin, A. L. Glen, M. Peet. Carnforth, UK: Marius Press, pp. 99–111.
Mahadik, S. P., Evans, D., Lal, H. (2001). Oxidative stress and the role of antioxidant and ω-
3 essential fatty acid supplementation in schizophrenia. Prog Neuropsychopharmacol Biol
Psychiatry 25: 463–493.
Margolis, R. L., Chung, D.-M., Post, R. M. (1994). Programmed cell death: Implications for
neuropsychiatric disorders. Biol Psychiatry 35: 946–956.
McCreadie, R. G. (1997). The Nithsdale Schizophrenia Surveys 16: breast-feeding and schizophre-
nia: preliminary results and hypothesis. Br J Psychiatry 170: 334–337.
McGrath, J. (1999). Hypothesis: is low vitamin D a risk-modifying factor for schizophrenia?
Schizophr Res 40: 173–177.
McIntyre, R. S., McCann, S. M., Kennedy, S. H. (2001). Antipsychotic metabolic effects: weight
gain, diabetes mellitus and lipid abnormalities. Can J Psychiatr 46: 272–281.
Mednick, S. A., Machon, R. A., Huttunen, M. O., Bonett, D. (1988). Adult schizophrenia following
prenatal exposure to an influenza epidemic. Arch Gen Psychiatry 45: 189–192.
Murphy, H. B. M. (1984). Editorial: diseases of civilization? Psychol Med 14: 487–490.
171 Nutritional factors and schizophrenia

Murray, R. M. (1994). Neurodevelopmental schizophrenia: the rediscovery of dementia precox.

Br J Psychiatry 165(Suppl. 25): 6–12.
Nandi, D. N., Mukherjee, S. P., Boral, G. C., et al. (1980). Socioeconomic status and mental
morbidity in certain tribes and castes in India: a cross-cultural study. Br J Psychiatry 136:
73–85.
Neuringer, M. Connor, W. E., Lin, D. S., Barstad, L., Luck, S. (1986). Biochemical and functional
effect of prenatal and postnatal omega-3 fatty acid deficiency on retina and brain in rhesus
monkeys. Proc Natl Acad Sci USA 83: 4021–4025.
Ntambi, J. M., Bene, H. (2001). Polyunsaturated fatty acid regulation of gene expression. J Mol
Neurosci 16: 273–278.
O’Brien, J. S., Sampson, E. L. (1965). Lipid composition of the normal human brain: gray matter,
white matter, and myelin. J Lipid Res 6: 537–544.
Parikh, V. Khan, M. M., Mahadik, S. P. (2002). Differential effects of antipsychotics on expression
of antioxidant enzymes and membrane lipid peroxidation in rat brain. J Psychiatr Res 37:
43–51.
Peet, M., Laugharne, J., Rangarajan, N., Renolds, G. P. (1993). Tardive dyskinesia, lipid peroxi-
dation, and sustained amelioration with vitamin E treatment. Int Clin Psychopharmacol 8:
151–153.
Peet, M., Laugharne, J., Mellor, J. E., Ramchand, C. N. (1995). Essential fatty acid defi-
ciency in erythrocyte membranes from chronic schizophrenic patients, and the clinical
effects of dietary supplementation. Prostaglandins, Leukotr Essent Fatty Acids 55: 119–
122.
Peet, M., Brind, J., Ramchand, C. N., Shah, S., Vankar, G. K. (2001). Two double-blind placebo-
controlled pilot studies of eicosapentaenoic acid in the treatment of schizophrenia. Schizophr
Res 49: 243–251.
Peet, M., Horrobin, D. F., in association with the E-E Multicenter Study Group (2002). A dose-
ranging exploratory study of the effects of ethyl-eicosapentaenoate in patients with persistent
schizophrenic symptoms. J Psychiatric Res 36: 7–18.
Pettegrew, J. W., Keshavan, M. S., Minshew, N. J. (1991). Alterations in brain high-energy phos-
phate and membrane phospholipid metabolism in first-episode, drug-naive schizophrenics.
A pilot study of the dorsal prefrontal cortex using in vivo phosphorus 31 nuclear magnetic
resonance spectroscopy. Arch Gen Psychiatry 48: 563–568.
Piomelli, D., Pilon, C., Giros, B., et al. (1991). Dopamine activation of the arachidonic acid
cascade as a basis for D1 /D2 receptor synergism. Nature 353: 164–167.
Raedler, T. J., Knable, M. B., Weinberger, D. R. (1998). Schizophrenia as a developmental disorder
of the cerebral cortex. Curr Opin Neurobiol 8: 157–161.
Rafalowska, U., Liu, G.-J., Floyd, R. A. (1989). Peroxidation induced changes in synaptoso-
mal transport of dopamine and gamma-aminobutyric acid. Free Rad Biol Med 6: 485–
492.
Rana, R. S., Hokin, L. E. (1990). Role of phosphoinositols in transmembrane signaling. Physiol
Rev 70: 115–164.
Reddy, R., Yao, J. (1996). Free radical pathology in schizophrenia: a review. Prostaglandins, Leukot
Essent Fatty Acids 55: 33–43.
172 S. P. Mahadik

Rotrosen, J., Wolkin, A. (1987). Phospholipid and prostaglandin hypotheses of schizophrenia. In

Psychopharmacology: The Third Generation of Progress, ed. H. Y. Meltzer. New York. Raven
Press, pp. 759–764.
Sartorius, N., Jablensky, A., Korten, A., et al. (1986). Early manifestations and first-contact
incidence of schizophrenia in different cultures. Psychol Med 16: 909–928.
Scottish Schizophrenia Research Group (2000). Smoking habits and plasma lipid peroxides and
vitamin E in never medicated first-episode schizophrenic patients with schizophrenia. Br J
Psychiatry 176: 290–293.
Selemon, L. D., Rajkowska, G., Goldman-Rakic, P. S. (1995). Abnormally high neuronal density
in the schizophrenic cortex. Arch Gen Psychiatry 52: 805–818.
Simopoulos, A. P. (1991). Omega-3 fatty acids in health and disease, and in growth and devel-
opment. Am J Clin Nutr 54: 438–463.
Sohal, R. S., Ku, H.-H., Agarwal, S., Forster, M. J., Lal, H. (1994). Oxidative damage, mitochondrial
oxidant generation and antioxidant defenses during aging, and in response to food restriction
in the mouse. Mech Aging Dev 74: 121–133.
Stanley, J. A., Williamson, P. C., Drost, D. J., et al. (1995). An in vivo study of the prefrontal cortex
of schizophrenic patients at different age of illness via phosphorous magnetic resonance
spectroscopy. Arch Gen Psychiatry 52: 399–406.
Stevens, L. J., Burgess, J. R. (1999). Essential fatty acids in children with attention-
deficit/hyperactivity disorder. In Phospholipid Spectrum Psychiatric Disorders, ed. D.
Horrobin, A. L. Glen, M. Peet. Carnforth, UK: Marius Press, pp. 263–269.
Suzuki, K. (1981). Chemistry and metabolism of brain lipids. In Basic Neurochemistry, 3rd edn,
ed. D. J. Seigel, R. W. Albers, B. W. Agranoff, et al. Boston, MA: Little Brown, pp. 355–370.
Tollefson, G. D. (1996). Cognitive function in schizophrenic patients. J Clin Psychiatry 57(Suppl.
11): 31–39.
Thompson, G. A. (1992). The Regulation of Membrane Lipid Metabolism, 2nd edn. Boca Raton,
FL: CRC Press.
Torrey, E. F. (1980). Schizophrenia and Civilization. New York: Jason Aronson.
Vaddadi, K. S., Courtney, P., Gilleard, C. S., Manku, M. S., Horrobin, D. F. (1989). A double-blind
trial of essential fatty acid supplementation in patients with tardive dyskinesia. Psychiatr Res
27: 313–323.
Virdee, K., Brown, B. L., Dobson, P. R.-M. (1994). Stimulation of arachidonic acid release from
Swiss 3T3 cells by recombinant basic fibroblast growth factor: independence from phos-
phoinositide turnover. Biochim Biophys Acta 16: 193–205.
Wainwright, P. E. (1992). Do essential fatty acids play a role in brain and behavioral development?
Neurosci Biobehav Rev 16: 193–205.
Weinberger, D. R. (1996). On the plausibility of “the neurodevelopmental hypothesis” of
schizophrenia. Neuropsychopharmacology 14: 1S–11S.
WHO (World Health Organization) (1973). The International Pilot Study of Schizophrenia,
Vol. 1. Geneva: World Health Organization.
Yamamoto, N., Saitoh, M., Moriuchi, A., Nomura, M., Okuyama, H. (1987). Effect of dietary
γ -linolenate/linoleate balance on brain lipid compositions and learning ability of rats.
J Lipid Res 28: 144–151.
173 Nutritional factors and schizophrenia

Yao, J. K. (1999). Red blood cell and platelet fatty acid metabolism in schizophrenia. In Phospho-
lipid Spectrum in Psychiatric Disorders, ed. D. Horrobin, A. L. Glen, M. Peet. Carnforth, UK:
Marius Press, pp. 57–71.
Yao, J. K., Yasaei, P., van Kammen, D. P. (1992). Increased turnover of platelet phosphatidylinositol
in schizophrenia. Prostaglandins, Leukotr Essent Fatty Acids 46: 39–46.
Yao, J. K., van Kammen, D. P., Welker, J. A. (1994). Red cell membrane dynamics in schizophrenia:
II fatty acid composition. Schizophr Res 13: 217–226.
 10

Schizophrenia, neurodevelopment,
and epigenetics
Arturas Petronis
Centre for Addiction and Mental Health, Toronto, Canada

The neurodevelopmental theory of schizophrenia is based on the hypothesis that

early brain insults affect brain development and eventually cause dysfunction of
the mature brain, predisposing to schizophrenia. The basis for these ideas formed
early in the twentieth century when clinical psychiatrists insightfully speculated that
schizophrenia could be resulting from cerebral maldevelopment (Kraepelin, 1919).
Over the last several decades, a myriad of clinical, epidemiological, morphologi-
cal, and molecular studies have investigated brain development to find evidence
concerning possible neurodevelopmental changes in schizophrenia (reviewed in
Lewis and Levitt, 2002; Woods, 1998). Although the results of such studies are not
necessarily fully consistent, there is converging evidence that at least a subgroup
of schizophrenia patients exhibit subtle developmental abnormalities that presum-
ably occur during embryogenesis. Changes have been detected in brain asymmetry,
neuronal cell migration and clustering, density of neurons in various brain regions,
and their cytoarchitectural organization and synapse formation. Schizophrenia-
associated morphological differences of the brain have also been documented at the
macroscopic level, including larger lateral brain ventricules and changes of the vol-
ume of cortex and subcortical structures. Consistently with the brain morpholog-
ical aberrations, prospective and retrospective behavioral studies of schizophrenia
patients showed that, long before they became affected, such individuals exhib-
ited a higher incidence of minor neuromotor abnormalities, delayed attainment
of developmental milestones, and various other subtle behavioral and intellectual
abnormalities. Beyond the brain, schizophrenia patients are also reported to have
a higher prevalence of minor physical anomalies, such as low-set ears, furrowed
tongue, high-arched palate, curved fingers, and greater distance between the eyes,
among numerous others. Since both the skin and the brain originate from the ecto-
derm, the minor physical anomalies can be considered as external signs of damage to

Neurodevelopment and Schizophrenia, ed. Matcheri S. Keshavan et al. Published by Cambridge University
Press.

C Cambridge University Press 2004.

174
175 Schizophrenia,neurodevelopment, and epigenetics

the ectoderm during embryogenesis, which serves as indirect support for aberrant
neurodevelopment.
What could be the cause(s) of the above brain development and schizophrenia-
related aberrations? A large group of putative etiological factors has been sug-
gested, investigated, and categorized into environmental and genetic groups. The
first group includes various obstetric complications such birth traumas, maternal
viral infection during pregnancy, pre-eclampsia, and deficiencies in nutrition. The
second group puts the emphasis on DNA sequence variation in the genes that may
play a role in neurodevelopment. So far, none of the above factors has been proven
to be in a cause–effect relationship with schizophrenia. In this chapter, the idea that
developmental changes in schizophrenia can be caused and/or mediated by epigen-
etic factors is suggested. It is argued that shifting the emphasis from the traditional
“gene–environment” dichotomy to epigenetics may provide a cohesive theoretical
framework for the myriad of fragmented phenomenological and molecular find-
ings in schizophrenia and lead to a series of new molecular strategies, designs, and
approaches.

The basics of epigenetics

By definition, epigenetics refers to regulation of gene expressions that are con-
trolled by heritable but potentially reversible changes in DNA methylation and/or
chromatin structure (Henikoff and Matzke, 1997) (Fig. 10.1). In most organisms,
cytosines can be of two functional states: unmodified cytosines (C) and cytosines
that are methylated at the 5-position of the pyrimidine ring (met C). DNA methyla-
tion is an enzymatic reaction performed by several types of DNA methyltransferase,
namely DNMT1, DNMT3A, and DNMT3B (reviewed in Bestor, 2000). DNMT1 is
ubiquitously expressed and its primary role is maintaining the DNA methylation
pattern in the somatic cells by methylating hemimediated CpG sites after DNA
replications. DNMT3A and DNMT3B are required to initiate de nova methylation
and establish new DNA methylation patterns (Bestor, 2000). The evidence for DNA
demethylase has been controversial so far. It is interesting to note that, although
the overwhelming majority of organisms from single-cell prokaryotes to complex
multicellular mammals have cytosine methylation, among the rare exceptions that
do not exhibit DNA methylation there are three major model organisms: Drosophila,
yeast, and Caenorhabditis elegans. A large number of genes exhibit an inverse cor-
relation between the degree of methylation and gene expression, which supports
an increasing body of experimental evidence suggesting that epigenetic modifica-
tion is intimately involved in the regulation of expression of genes (Holliday, 1996;
Jackson-Grusby et al., 2001; Razin and Shemer, 1999; Siegfried et al., 1999; Singal
176 A. Petronis

A A AA A
AAAAAA
AAAA
A A AA A
A A AA A
A A AA A

Fig. 10.1. DNA and histone modifications in regulation of gene expression. DNA is wrapped around
histone complexes (gray cylinders) to form nucleosomes. Depending on DNA and histone
modifications (small circles and squares) chromatin can be transcriptionally competent
(A) or not (B). PolyA sequences represent mRNA molecules.

and Ginder, 1999; Stancheva and Meehan, 2000; Yeivin and Razin, 1993) although
this is not universal (Walsh and Bestor, 1999).
One of the mechanisms of epigenetic regulation of genes is related to methy-
lation of the binding sites for transcription factors, serving to change the affinity
of these factors for the regulatory sequences of these specific genes (Riggs et al.,
1998). In addition to the “critical site” effects of met C, the overall ratio of met Cs
over unmethylated cytosines in a gene-regulatory region also contributes to gene
activity. This type of regulation seems to be linked to another mechanism of epi-
genetic regulation, namely various types of histone modification. Histones are
key components of nucleosomes represented by four highly conservative proteins:
H2A, H2B, H3, and H4. The role of mediators between DNA modification and
chromatin modification is performed by a group of met C-binding proteins such
as MECP2, MBD1, and MBD2, among others (reviewed in Li, 2002). The met C-
binding proteins recruit different chromatin remodeling proteins and transcription
complexes to the loci of methylated DNA. For example, MECP2 binds to methy-
lated DNA and attracts histone deacetylases that hypoacetylate histones. Tran-
scriptionally competent chromatin is normally enriched with acetylated histones,
while transcriptionally silent chromatin is deacetylated (reviewed in Robertson
177 Schizophrenia,neurodevelopment, and epigenetics

and Wolffe, 2000). The interaction of DNA methylation and histone acetylation
demonstrates that the two types of epigenetic regulation act in concert. In addition
to acetylation, nucleosomal histones can be modified by methylation, phosphory-
lation, ubiquitylation, and possibly other, yet undiscovered, ways. It is important to
note that histone modification targets are numerous residues of lysine, arginine, and
serine. For example, Lys9, Lys14, Lys18, and Lys23 of H3 and Lys5, Lys8, Lys12, and
Lys16 of H4, together with lysines on H2A and H2B, can be acetylated; whereas Lys4,
Lys9, Lys27, Arg2, Arg17, and Arg26 of H3 and Lys20 and Arg3 of H4 can be methy-
lated (Li, 2002). It is evident that a large number of combinations of various types of
histone modification are possible even over a short stretch of chromatin. Further-
more, the same DNA sequence may be in a wide variety of functional states. This
provided the basis for the concept of the “histone code” (Jenuwein and Allis, 2001).
In addition to their putative role in regulation of autosomal biallelically expressed
genes, epigenetic mechanisms of transcriptional repression are of primary impor-
tance for X chromosome inactivation and genomic imprinting, as well as in the tran-
scriptional inactivation of endogenous retroviruses and other “parasitic” sequences
(reviewed in Wolffe and Matzke, 1999). Furthermore, epigenetic factors play a role
in DNA mutagenesis and repair as well as in DNA recombination and possibly
replication.
Epigenetic patterns are transmitted similarly to DNA sequences, from maternal
chromatids to daughter chromatids during mitotic divisions, and transmission of
the epigenetic status is called the epigenetic inheritance system (Maynard Smith,
1990). Unlike DNA sequences, which exhibit nearly complete interclonal fidelity,
epigenetic status usually exhibits only partial stability; over time, this can result in
substantial changes across identical sequences. DNA and chromatin modifications
do react to the extracellular environment. Stochastic factors in the cell also con-
tribute to epigenetic differences in the cells of the same line. After mitotic division,
the daughter chromosomes do not necessarily carry identical epigenetic patterns in
comparison with the parental chromosomes, and, although belonging to the same
group of cells (myocytes, lymphocytes, and epithelial cells), these cells exhibit small
quantitative differences in terms of their phenotypes and functions. Over time, sub-
stantial epigenetic differences may be accumulated across the cells of the same cell
line or the same tissue (Riggs et al., 1998). This represents a fundamental difference
between epigenetic systems and DNA sequence-based hereditary factors. The par-
tial stability of epigenetic modification is called epigenetic metastability. Epigenetic
metastability also applies to the germline. Although it has been generally accepted
that the parental epigenetic profile is replaced by a new one during the maturation
of gametes, there is increasing evidence demonstrating that some epigenetic signals
escape erasure and can be transmitted from one generation to another (Rakyan
et al., 2002).
178 A. Petronis

Relevance of epigenetics to schizophrenia

Two key aspects of epigenetic modification of the genome make epigenetics very
relevant to schizophrenia (as well as numerous other complex non-Mendelian dis-
eases) (Petronis [2001] and references therein for this section). First, epigenetic
modifications of DNA and chromatin “orchestrate” the activity of the genome,
including regulation of gene expression. While a DNA sequence provides the infor-
mation of what specific protein has to be synthesized, it is absolutely necessary that
each gene is regulated accordingly to the type and the needs of a specific cell. While
relatively small deviations in the expression pattern account for normal individual
variations, more significant ones can be as detrimental to a cell as the mutant DNA
sequences that encode dysfunctional proteins.
Epigenetic metastability is the second aspect that makes epigenetics relevant to
various non-Mendelian irregularities of schizophrenia. Epigenetic regulation of
genes undergoes significant reorganization during development and aging as well
as under the influence of extracellular factors (e.g. hormonal status of an organism)
or environmental factors (reviewed in Jablonka and Lamb, 1995). Epigenetic regu-
lation represents the dynamic feature of a gene and genome, whereas most of the
DNA sequences (especially the coding ones) do not change during the life of an
individual.
Age-dependent epigenetic changes may shed a new light on the relatively late age
of onset in schizophrenia as well as the periods of critical age of susceptibility (early
twenties and late sixties in both sexes, plus late forties in women). It is evident that
the ages of the increased incidence coincide with major hormonal rearrangements
in the organism. Epigenetic status of a gene is one of the targets of hormone action.
Changes in endocrine status may lead to significant epigenetic rearrangements.
Various hormones, including sex hormones, have a significant impact on gene
expression, and this is achieved by changing chromatin conformation and/or local
patterns of gene methylation. A disease process may be provoked by hormone-
mediated epigenetic changes in the critical genes. In the absence of solid evidence
for the involvement of the X chromosome, the same mechanism may contribute to
the clinical differences of schizophrenia in males and females (reviewed in Seeman,
1997; see Ch. 18). In this respect, changes in the hormonal milieu during maturation
and involution can substantially affect the regulation of genes, which at least to some
extent is achieved via their epigenetic modifications.
Parent-of-origin effects present with differential risk to the offspring of being
affected with a disease depending on which of the two parents is affected with that
specific disease. In molecular genetic studies, parental effects present with coseg-
regation of the disease with either maternal or paternal alleles and/or with differ-
ences in the lod scores (measure of genetic linkage) in the families with maternal
179 Schizophrenia,neurodevelopment, and epigenetics

versus paternal transmission of the disease (reviewed in Petronis, 2000). Parental

origin effect is one of the classical epigenetic mechanisms of differential regulation
of gene activity, called genomic imprinting.
Epigenetic metastability may shed a new light on the non-Mendelian mode of
transmission of schizophrenia as well as on the presence of both sporadic and
familial cases of the disease (in the absence of phenotypic differences between the
two groups). As a rule, the clinical phenotypes in sporadic schizophrenia are indis-
tinguishable from familial ones. Meiotically stable epimutations can segregate like
quasi-Mendelian factors and may represent familial disease, while the epimutations
that are erased during gametogenesis may be the cause of sporadic cases. In sum-
mary, the epigenetic theory puts the emphasis on epigenetic changes (or absence of
such changes) in specific genomic loci during meiosis, while the traditional DNA
sequence-oriented paradigm infers that different genes are operating in familial
and sporadic cases (major genes and minor genes, respectively).
The most straightforward example of the value of the epigenetic theory is the
new interpretation of discordance of monozygotic twins, one of the key mysteries
in complex traits. The rate of monozygotic twin discordance in schizophrenia in a
combined twin sample is about 50% (Cardno and Gottesman, 2000). Although two
co-twins contain the same DNA sequence in the overwhelming majority of somatic
cells, substantial epigenetic differences may be accumulated over numerous cell
divisions during development and aging in such co-twins. The epigenetic non-
identity may result in differential epigenetic regulation of disease-relevant genes,
which would result in only one twin reaching the “threshold” of clinical symptoms
(Petronis et al., 2003).
In addition, the putative epigenetic misregulation-based disease mechanisms are
consistent with numerous epidemiological, clinical, and molecular findings (Chen
et al., 2002; Petronis et al., 2002; Tremolizzo et al., 2002). Of particular interest to
the topic of this book is the contribution of epigenetic factors in embryogenesis
and development with regard to premorbid changes in the brains of schizophrenia
patients. The next section is dedicated to the overview of the role of epigenetic
factors in development.

Epigenetics and development

There are numerous links between development and epigenetics. Etymologically,
the term epigenetics originates from epigenesis, a theory of development, which
proposed that the embryo at its early stage of development is not differentiated
and embryogenesis proceeds by increasing levels of complexity (Holliday, 1994).
Epigenesis contrasted with preformation, a theory which assumed that complexity
already existed in the very early embryo.
180 A. Petronis

The idea of a putative relationship between gene activity and development was
entertained by numerous biologists during the twentieth century. For example, T.
H. Morgan hypothesized that “genes . . . are changing in some way as development
proceeds in response to that part of the protoplasm in which they lie, and that these
changes have reciprocal influence on the protoplasm” (Morgan, 1934). In the 1950s,
C. H. Waddington coined the term “epigenetics,” which referred to the develop-
mental processes that “connect” the genotype to phenotype, or the processes by
which genotype gives a rise to phenotype (reviewed in Slack, 2002). These ideas
originated long before the advent of molecular biology and represented predomi-
nantly a theoretical, although inspiring, framework that influenced various other
fields of scientific activity including psychiatric research (Gottesman and Shields,
1982; Woolf, 1997). Waddington is best known for the concept of “epigenetic land-
scape.” The uneven surface of the landscape with “hills” and “valleys” represents
a simplified version of multidimensional and intricate non-linear interactions of
various molecules within the cell during the embryonic development (Fig. 10.2A).
The rolling ball symbolizes a phenotype of a differentiating cell, which at each
developmental step acquires new features. The end result of this process is a huge
variety of highly specialized cells (Fig. 10.2B).
The idea of the primary role of epigenetic factors in development was raised
again in two seminal papers published in 1975 (Holliday and Pugh, 1975; Riggs,
1975). At that time, the presence of epigenetic DNA modification (methylation)
was already known and there was experimental evidence suggesting that epige-
netic modification may have an impact on the regulation of gene activity. The two
papers represented the first systematic attempt to synthesize the available evidence
regarding linkage of epigenetic regulation to both differentiation of cells and main-
tenance of the differentiated cell states. There are two key questions. First, how do
the pluripotent undifferentiated embryonic cells become highly specialized.
Second, how is the specific phenotype of a specialized cell maintained over numer-
ous mitotic divisions. For example, ectoderm cells differentiate into two very dif-
ferent phenotypes, neurons and keratocytes, and the latter would never convert
into the former, and vice versa. What keeps the cell “locked” in a particular regu-
lated state? Understanding the molecular principles of the developmental programs
remains one of the main challenges for developmental biologists. Although there
are no final answers, epigenetics seems to be one of the best candidates for a molec-
ular mechanism that would be consistent with a wide variety of developmental
events in a cell.
Maintenance of the epigenetic profile is consistent with the phenotypic stability of
somatic cells. In mammals, methylation occurs predominantly at the symmetrical
dinucleotide CpG positions, which after DNA replication become hemimethylated.
This hemimethylated DNA is the favored substrate for DNMT1, which efficiently
181 Schizophrenia,neurodevelopment, and epigenetics

Fig. 10.2. Epigenetics and cell differentiation. (A) Waddington’s epigenetic landscape represents the
imaginary path of an undifferentiated pluripotent embryonic cell during development.
(B) The end result of cell differentiation is a wide variety of highly specialized cells that look
different and perform different functions, although they contain the same DNA sequences.
(See plate section for color version.)

recognizes unmethylated CpGs in the newly synthesized strands that are comple-
mentary to the methylated CpG and proceeds to methylate them (Bestor, 2000).
In tissue culture, fidelity of maintenance methylation in mammalian cells was 97–
99.9% and de novo methylation activity was 3–5% per mitosis (Riggs et al., 1998).
Fidelity of maintenance methylation in different types of cell in vivo is unknown
but it is very possible that it is sufficient to assure clonal continuity of myocytes,
hepatocytes, and the many other kinds of specialized cell. The ways by which
182 A. Petronis

various types of epigenetic modification of histones can be inherited in mitotic cells

have also been discussed (Wolffe, 1994). All this provides an excellent mechanism
that faithfully maintains specific expression patterns in various types of somatic
cell.
The question of the role of epigenetic factors in differentiation of the pluripo-
tent embryonic cell into a highly specialized one in the developing embryo is far
more complex. Longitudinal studies of epigenetic changes in specific embryonic
cells with respect to their increasing differentiation are still technically challeng-
ing and there have been only a few studies investigating the relationship between
epigenetic changes in the specific genes and changes in such gene activity during
cell differentiation. The best known case is the epigenetic changes in regulation
of the β-globin gene family during the development of red blood cells. DNA and
chromatin modification analyses have shown major differences in epigenetic regu-
lation of transcriptionally active and inactive globin genes (Litt et al., 2001; Singal
and Ginder, 1999). More importantly, there are suggestions that such cell-specific
epigenetic changes in the chromatin structure of the β-globin locus precede the
activation of individual genes; therefore, transcriptional competence (the poten-
tial for expression) is detected before the actual transcription occurs (reviewed
in Wolffe and Barton, 2000). Such experimental findings are consistent with the
developmental role of epigenetic changes in the genome. However, studies of the
methylation state of the promoter of the mouse gene for skeletal α-actin showed
no correlation between methylation and expression patterns during development
and differentiation (Warnecke and Clark, 1999). It is important to note that only
a relatively small genomic region of the gene was investigated; consequently, given
our very superficial understanding the critical regions of epigenetic regulation,
rejection of the role of epigenetic factors is premature.
Indirect evidence implicating the role of epigenetic modifications in development
derives from the surprising complexity of epigenetic changes during gametogene-
sis and embryogenesis (Fig. 10.3). The spermatozoon and oocyte genomes exhibit
major differences in their methylation patterns. In mice studies, mRNA for Dnmt1
is present at high levels in post-mitotic germ cells but undergoes alternative splicing
of sex-specific 5 exons, which controls the production and localization of enzyme
during specific stages of gametogenesis (Mertineit et al., 1998). An oocyte-specific
5 exon is associated with the production of very large amounts of active Dnmt1 pro-
tein, which is truncated at the N-terminus and sequestered in the cytoplasm during
the later stages of oocyte growth, while a spermatocyte-specific 5 exon interferes
with translation and prevents production of mouse Dnmt1 during the prolonged
crossing-over stage of male meiosis (Mertineit et al., 1998). During embryonic
development, the genome is subjected to major changes in epigenetic regulation
(reviewed in Reik et al., 2001). There are developmental periods of reprogramming
183 Schizophrenia,neurodevelopment, and epigenetics

Fig. 10.3. (A) Methylation reprogramming in the germline. Primordial germ cells (PGCs) in the mouse
become demethylated early in development. Remethylation begins in pro-spermatogonia
in male germ cells and after birth in growing oocytes. Some stages of germ cell develop-
ment are shown. (B) Methylation reprogramming in preimplantation embryos. The paternal
genome is demethylated by an active mechanism immediately after fertilization. The mater-
nal genome is demethylated by a passive mechanism that depends on DNA replication. Both
are remethylated around the time of implantation to different extents in embryonic (EM)
and extraembryonic (EX) lineages. Methylated imprinted genes and some repeat sequences
(upper dashed line) do not become demethylated. Unmethylated imprinted genes (lower
dashed line) do not become methylated. (Reproduced with permission from Reik, W., Dean,
W., and Walter, J. (2001). Epigenetic reprogramming in mammalian development. Science,
293: 1089–1093. Copyright 2001 American Association for the Advancement of Science.)

of methylation patterns in vivo, when a substantial part of the genome is demethy-

lated and after some time remethylated in a cell- or tissue-specific pattern. Repro-
gramming of DNA methylation in early embryos occurs both by active and passive
mechanisms (Reik et al., 2001).
It is highly unlikely that such a sophisticated machinery of epigenetic changes
would be operating for no reason or that such a reason would not be related to
184 A. Petronis

development. Consistent with this idea, genome-wide alterations in methylation

induced by knockouts of the genes encoding the epigenetic regulators often result in
embryo death or developmental defects. For example, inactivation of Dnmt1 in the
Dnmt1-/- knockouts resulted in a significant reduction of total met C in the genome
after several rounds of DNA replication and developmental arrest at an early stage of
embryogenesis (reviewed in Li, 2002). Interestingly, it has been shown that Dnmt3a,
and Dnmt3b are also required for the stable inheritance of DNA-methylation par-
ent in mouse embryonic stem cells. By comparison, Mbd2-/- knockouts were viable
and fertile but showing defective maternal behavior (reviewed in Li, 2002). Ani-
mals lacking MeCP2 exhibited complex neurological defects, including tremor,
ataxia, hindlimb clasping, stereotypic forelimb motions, increased anxiety-related
behavior, and seizures. In humans, mutations of MECP2 cause a neurodevelop-
mental disorder known as Rett syndrome (van den Veyver and Zoghbi, 2001).
MECP2 is located on the X chromosome, with male carriers of mutant MECP2
being lethal. Girls affected with Rett syndrome have normal development for the
first 18 months, which is followed by a period of regression that is characterized
by loss of purposeful hand use and speech, and by mental handicap. In addition to
Rett syndrome, two other classical epigenetic diseases, Prader–Willi and Angelman
syndromes, exhibit numerous brain anomalies such as abnormal cortical develop-
ment, microencephaly, and ventricular dilation (reviewed in Schumacher, 2001).
Finally, epigenetic inactivation of FMR1 is a key factor in etiopathogenesis of frag-
ile X syndrome(El-Osta, 2002). Pathology, studies from the brains of patients and
from Fmr1 knockout mice show abnormal dendritic spines and altered synaptic
plasticity (El-Osta et al., 2001).
Further supportive evidence for the role of epigenetic changes in development
derives from nuclear cloning experiments in mammals. Although a somatic cell
nucleus from adult tissue can initiate embryonic development after being trans-
planted into an enucleated oocyte, cloning is extremely inefficient. Most cloned
embryos die shortly after implantation and the few that survive to birth frequently
have developmental abnormalities (Rideout et al., 2001). Studies of DNA methy-
lation have shown that genome-wide demethylation before implantation is less
efficient in cloned embryos, and that de novo methylation occurs at an earlier stage
in cloned embryos than in normal embryos (Reik et al., 2001). The abnormal repro-
gramming of DNA methylation could result in the failed reactivation of the genes
that are essential for embryonic development.
Further evidence relating epigenetics and embryogenesis derives from the action
of some teratogenic agents, changing epigenetic regulation. An example could be
sodium valproate, a medication commonly used for treatment of epilepsy that also
exhibits teratogenic effects. It has been recently shown that valproic acid inhibits
histone deacetylase, causing hyperacetylation of histones in cultured cells, which
185 Schizophrenia,neurodevelopment, and epigenetics

results in activation of transcription from diverse exogenous and endogenous pro-

moters (Gottlicher et al., 2001; Phiel et al., 2001). Non-teratogenic analogs of val-
proic acid do not affect histone deacetylase and do not activate transcription. Based
on these observations, it can be proposed that inhibition of histone deacetylase is
the cause of valproic acid-induced developmental defects.
All the above considerations provide the basis for further elaboration and exper-
imental analysis of the ideas suggested by Morgan and Waddington (Morgan, 1934;
Slack, 2002). Morgan’s “dynamic genes” can now be seen as “hard core” coding DNA
sequences packed in their malleable epigenetic environment, which regulates the
activity of such sequences. Waddington suggested that the epigenetic landscape is
formed by the action of various genes and on the “underside” of the landscape, such
genes can be visualized as pegs stuck in the ground, each with a longer or shorter
rope attached to the peaks and troughs, respectively (see Fig. 3C in Slack, 2002).
Since it is now known that DNA sequences do not change during development and
that such sequences are identical nearly in all somatic cells of an individual, it is
more likely that the peaks and troughs are formed by the epigenetic modifications
of such genes (Fig. 10.4).
In conclusion, although direct proof of the role of epigenetics in development
is yet to come, epigenetic regulation may serve as a convenient candidate for the
mechanistic explanation of cell differentiation and development.

The synthesis: epigenetics, development, and schizophrenia

The putative role of epigenetic factors in development adds an important aspect
to the epigenetic theory of schizophrenia. Since embryogenesis is related to major
epigenetic rearrangements and the genesis of highly specialized cells includes thou-
sands and thousands of small epigenetic steps, it is very likely that deviation from
the “ideal” epigenetic scenario is quite common. Because of its complexity, the
brain is likely to be the most susceptible organ and even mild epigenetic malfunc-
tion might lead to a wide variety of small morphological and functional changes
in the developing brain. Epigenetic aberrations (epimutations) may originate from
any of the three sources acting individually or in combination. First, because of
epigenetic metastability, epimutations can be inherited through the germline.
Second, epigenetic misregulation can be caused by environmental factors, such
as exposure of an embryo or fetus to various toxic agents, some medications, mal-
nutrition, or a mother’s infections during pregnancy. Third, epigenetic aberrations
may be generated by stochastic events in the embryonic cells. In the last case,
there is neither inherited predisposition nor exposure to the hazardous factors but
the developmental program goes awry for various non-deterministic events of the
molecular world. An example of a stochastic event is where hemimethylated DNA
186 A. Petronis

Fig. 10.4. Waddington’s epigenetic landscape: a view from “behind.” The peaks and troughs are formed
by the epigenetic status of various genes (A, B, C, etc.) in the cell, and this determines cellular
phenotype and functions.

remained asymmetrically methylated because there was no DNMT1 in the close

vicinity at that specific time; this effect would be amplified during the subsequent
mitotic divisions. Prenatal epigenetic misregulation would cause subtle aberrations
in neuronal cell migration, density of neurons, and their cytoarchitectural organ-
ization but would not lead to major dysfunction of brain activity. Since the age
of onset in most psychoses is delayed for a relatively long time, it is possible that
inherited and/or acquired epimutations are “tolerated” by the brain cells during
the childhood and adulthood of an individual at risk. It takes several decades of
age-dependent epigenetic changes (which are also a part of the postnatal devel-
opmental program) until the epigenetic misregulation of a critical gene(s) reaches
a sufficient level that the compensatory mechanisms of a cell are not able to deal
with the problem, resulting in the manifestation of psychotic symptoms. External
factors such as stress, other diseases, and urban life may or may not play a role in
adding to the “critical mass” of epigenetic misregulation. It is more likely, however,
that the endogenous environment of an individual (i.e. changes in the gonadal and
187 Schizophrenia,neurodevelopment, and epigenetics

adrenal hormone homeostasis during puberty) is the key factor that changes from
a predisposing epimutation to a disease-causing one. Environmental factors may
contribute to the progression of pre-epimutation, but it is more likely that develop-
mental changes and stochastic events play a more important role (Petronis, 2001).
Consistent with the latter, some first episodes of schizophrenia come without any
herald of the disease and in the absence of hazardous environmental factors. Sever-
ity of epigenetic misregulation may fluctuate over time, which in clinical terms
will be treated as disease remissions and relapses. The age-dependent epigenetic
changes may reach epigenetic reversion to a near normal state, which is seen as
a decline of psychotic symptoms and partial recovery from the disease (Petronis,
2001).
There is one key issue that the epigenetic theory treats differently to the neu-
rodevelopmental approach. From the epigenetic point of view, age- and hormone-
dependent neurochemical changes rather than structural changes in the brain are
the main disease mechanism. Brain morphological aberrations are more likely to
be just the reporters of mild deviation in the developmental program rather than
factors causing or predisposing to schizophrenia. There is strong reason to believe
that those early and minor developmental aberrations can be fully compensated
by the developing brain, which exhibits surprising plasticity and compensatory
potential (Kolb, 1995; Woods, 1998). The shift from the brain structural peculiarities
to the functional ones is consistent with the rare but very informative cases of
monozygotic twins discordant for schizophrenia, where brain anomalies are more
severe in the unaffected twin (Torrey et al., 1999, Fig. 6.4 on p. 114). The subtle
localized non-progressive prenatal brain developmental aberrations are just like
superficial scars, which do not do any harm but only remind of an old minor
injury. The problem of latency of the neurodevelopmental changes – one of the
main unclear issues in the neurodevelopmental theory – can now be reformulated
into the latency of epigenetic misregulation of some critical genes.
The heuristic value of the epigenetic model of schizophrenia lies in the possibil-
ity of integrating a variety of unrelated data into a new theoretical framework that
provides the basis for new experimental approaches in the study of this disease.
As it was attempted to show above, epigenetics is likely to be intimately related
to development and, therefore, represents an important aspect in the interpreta-
tion of the neurodevelopmental findings in schizophrenia. In one of her seminal
papers on “dynamic genome,” Barbara McClintock (1951) wrote: “If the ordered
processes of development are deranged, then genes which usually become active at
very specific times may instead be activated spasmodically or in random fashion
during development.” Fifty years later we raise a reverse question: can abnormally
expressed genes derange development? The answer to this question should provide
new insights into the mystery of schizophrenia.
188 A. Petronis

Acknowledgements
I thank Dr. A. Wong and Mr. Z. A. Kaminsky (CAMH, Toronto, Canada) for their
valuable comments and advice. This work was supported by grants from the Ontario
Mental Health Foundation, the National Alliance for Research on Schizophrenia
and Depression, and the Stanley Foundation.

REFERENCES

Bestor, T. H. (2000). The DNA methyltransferases of mammals. Hum Mol Genet 9: 2395–
2402.
Cardno, A. G., Gottesman, I. I. (2000). Twin studies of schizophrenia: from bow-and-arrow
concordances to Star Wars Mx and functional genomics. Am J Med Genet 97: 12–17.
Chen, Y., Sharma, R. P., Costa, R. H., Costa, E., Grayson, D. R. (2002). On the epigenetic regulation
of the human reelin promoter. Nucl Acids Res 30: 2930–2939.
El-Osta, A. (2002). FMR1 silencing and the signals to chromatin: a unified model of transcrip-
tional regulation. Biochem Biophys Res Commun 295: 575–581.
El-Osta, A., Baker, E. K., Wolffe, A. P. (2001). Profiling methyl-CpG specific determinants on
transcriptionally silent chromatin. Mol Biol Rep 28: 209–215.
Gottesman, I., Shields, J. (1982). Schizophrenia: The Epigenetic Puzzle. Cambridge, UK: Cambridge
University Press.
Gottlicher, M., Minucci, S., Zhu, P., et al. (2001). Valproic acid defines a novel class of HDAC
inhibitors inducing differentiation of transformed cells. EMBO J 20: 6969–6978.
Henikoff, S., Matzke, M. A. (1997). Exploring and explaining epigenetic effects. Trends Genet 13:
293–295.
Holliday, R. (1994). Epigenetics: an overview. Dev Genet 15: 453–457.
(1996). DNA methylation in eukaryotes: 20 years on. In Epigenetic Mechanisms of Gene Regu-
lation, ed. V. Russo, R. Martienssen, A. Riggs. Cold Spring Harbor, NY: Cold Spring Harbor
Laboratory Press, pp. 5–27.
Holliday, R., Pugh, J. E. (1975). DNA modification mechanisms and gene activity during devel-
opment. Science 187: 226–232.
Jablonka, E., Lamb, M. (1995). Epigenetic Inheritance and Evolution. Oxford: Oxford University
Press.
Jackson-Grusby, L., Beard, C., Possemato, R., et al. (2001). Loss of genomic methylation causes
p53-dependent apoptosis and epigenetic deregulation. Nat Genet 27: 31–39.
Jenuwein, T., Allis, C. D. (2001). Translating the histone code. Science 293: 1074–1080.
Kolb, B. (1995). Brain Plasticity and Behavior. Mahwah, NJ: Lawrence Erlbaum Associates.
Kraepelin, E. (1919). Dementia Praecox and Paraphrenia. Edinburgh: Livingstone.
Lewis, D. A., Levitt, P. (2002). Schizophrenia as a disorder of neurodevelopment. Annu Rev
Neurosci 25: 409–432.
Li, E. (2002). Chromatin modification and epigenetic reprogramming in mammalian develop-
ment. Nat Rev Genet 3: 662–673.
189 Schizophrenia,neurodevelopment, and epigenetics

Litt, M. D., Simpson, M., Gaszner, M., Allis, C. D., Felsenfeld, G. (2001). Correlation between
histone lysine methylation and developmental changes at the chicken beta-globin locus.
Science 293: 2453–2455.
Maynard Smith, J. (1990). Models of a dual inheritance system. J Theor Biol 143: 41–53.
McClintock, B. (1951). Chromosome organization and genic expression. Genes and Mutations.
Cold Spring Harb Symp Quant Biol XVI: 13–47.
Mertineit, C., Yoder, J. A., Taketo, T., et al. (1998). Sex-specific exons control DNA methyltrans-
ferase in mammalian germ cells. Development 125: 889–897.
Morgan, T. H. (1934). Embryology and Genetics. New York: Columbia University Press.
Petronis, A. (2000). The genes for major psychosis: aberrant sequence or regulation? Neuropsy-
chopharmacology 23: 1–12.
(2001). Human morbid genetics revisited: relevance of epigenetics. Trends Genet 17: 142–146.
Petronis, A., Paterson, A. D., Kennedy, J. L. (1999). Schizophrenia: an epigenetic puzzle? Schizophr
Bull 25: 639–655.
Petronis, A., Popendikyte, V., Kan, P. X., Sasaki, T. (2002). Major psychosis and chromosome 22:
genetics meets epigenetics. CNS Spectrums 7: 209–214.
Petronis, A., Gottesman, I. I., Kan, P. X., et al. (2003). Monozygotic twins exhibit numerous
epigenetic differences: clues to twin discordance? Schizophr Bull 29: 169–178.
Phiel, C. J., Zhang, F., Huang, E. Y., et al. (2001). Histone deacetylase is a direct target of valproic
acid, a potent anticonvulsant, mood stabilizer, and teratogen. J Biol Chem 276: 36734–
36741.
Rakyan, V. K., Blewitt, M. E., Druker, R., Preis, J. I., Whitelaw, E. (2002). Metastable epialleles in
mammals. Trends Genet 18: 348–351.
Razin, A., Shemer, R. (1999). Epigenetic control of gene expression. Results Probl Cell Differ 25:
189–204.
Reik, W., Dean, W., Walter, J. (2001). Epigenetic reprogramming in mammalian development.
Science 293: 1089–1093.
Rideout, III, W. M., Eggan, K., Jaenisch, R. (2001). Nuclear cloning and epigenetic reprogramming
of the genome. Science 293: 1093–1098.
Riggs, A., Xiong, Z., Wang, L., LeBon, J. M. (1998). Methylation dynamics, epigenetic fidelity
and X chromosome structure. In Epigenetics, Vol. Novartis Foundation Symposium 214:
Epigenetics, ed. A. Wolffe. Chichester, UK: Wiley, pp. 214–227.
Riggs, A. D. (1975). X inactivation, differentiation, and DNA methylation. Cytogenet Cell Genet
14: 9–25.
Robertson, K. D., Wolffe, A. P. (2000). DNA methylation in health and disease. Nat Rev Genet 1:
11–19.
Schumacher, A. (2001). Mechanisms and brain specific consequences of genomic imprinting in
Prader–Willi and Angelman syndromes. Gene Funct Dis 1: 7–25.
Seeman, M. V. (1997). Psychopathology in women and men: focus on female hormones. Am J
Psychiatry 154: 1641–1647.
Siegfried, Z., Eden, S., Mendelsohn, M., et al. (1999). DNA methylation represses transcription
in vivo. Nat Genet 22: 203–206.
Singal, R., Ginder, G. D. (1999). DNA methylation. Blood 93: 4059–4070.
190 A. Petronis

Slack, J. M. (2002). Conrad Hal Waddington: the last Renaissance biologist? Nat Rev Genet 3:
889–895.
Stancheva, I., Meehan, R. R. (2000). Transient depletion of xDnmt1 leads to premature gene
activation in Xenopus embryos. Genes Dev 14: 313–327.
Torrey, E. F., Bowler, A. E., Taylor, E. H., Gottesman, I. I. (1999). Schizophrenia and Manic
Depressive Disorder. The Biological Roots of Mental Illness as Revealed by the Landmark Study
of Identical Twins. New York: Basic Books.
Tremolizzo, L., Carboni, G., Ruzicka, W. B., et al. (2002). An epigenetic mouse model for molecular
and behavioral neuropathologies related to schizophrenia vulnerability. Proc Natl Acad Sci
USA 99: 17095–17100.
van den Veyver, I. B., Zoghbi, H. Y. (2001). Mutations in the gene encoding methyl-CpG-binding
protein 2 cause Rett syndrome. Brain Dev 23(Suppl. 1): S147–S151.
Walsh, C. P., Bestor, T. H. (1999). Cytosine methylation and mammalian development. Genes
Dev 13: 26–34.
Warnecke, P. M., Clark, S. J. (1999). DNA methylation profile of the mouse skeletal alpha-actin
promoter during development and differentiation. Mol Cell Biol 19: 164–172.
Wolffe, A. P. (1994). Inheritance of chromatin states. Dev Genet 15: 463–470.
Wolffe, A. P., Barton, M. C. (2000). Developmental regulation of chromatin function and gene
expression. In Chromatin Structure and Regulation, ed. S. C. R. Elgin, J. L. Workman. Oxford:
Oxford University Press, pp. 182–202.
Wolffe, A. P., Matzke, M. A. (1999). Epigenetics: regulation through repression. Science 286:
481–486.
Woods, B. T. (1998). Is schizophrenia a progressive neurodevelopmental disorder? Toward a
unitary pathogenetic mechanism. Am J Psychiatry 155: 1661–1670.
Woolf, C. M. (1997). Does the genotype for schizophrenia often remain unexpressed because of
canalization and stochastic events during development? Psychol Med 27: 659–668.
Yeivin, A., Razin, A. (1993). Gene methylation patterns and expression. In DNA Methylation:
Molecular Biology and Biological Significance, ed. J. Jost, H. Saluz. Basel: Birkhauser Verlag,
pp. 523–568.
 11

Early environmental risk factors for

schizophrenia
Mary Cannon1,2 , Kimberlie Dean1 , and Peter B. Jones3
1
Institute of Psychiatry, King’s College, London, UK
2
Royal College of Surgeons in Ireland, Dublin, Ireland
3
University of Cambridge, UK

The existence of early environmental risk factors for schizophrenia is central to the
notion of schizophrenia as a neurodevelopmental disorder (Marenco and Wein-
berger, 2000; McDonald et al., 2000) and these risk factors represent some of the
most challenging and interesting targets of schizophrenia epidemiology. This chap-
ter discusses prenatal and perinatal risk factors for schizophrenia. Childhood devel-
opmental impairment and later environmental and psychosocial risk factors are
dealt with in Chs. 22 and 13, respectively.

Prenatal risk factors for schizophrenia

Time and place of birth
The risk of developing schizophrenia has consistently been found to be increased
(1.5 to 2-fold) among those born in cities compared with those born in rural
areas (Lewis et al., 1992; Marcelis et al., 1998; Mortensen et al., 1999; Takei et al.,
1995; Torrey and Bowler, 1990). Studies that have teased apart the effects of urban
birth and urban living (Marcelis et al., 1999) have found that the greatest risk
was for those born in urban areas, with no additional effect of later urban resi-
dence, indicating that the urban risk factor or factors associated with an increase
in risk of schizophrenia appear to act in early life rather than around the time
of illness onset. Linear trends for risk of schizophrenia with increasing popula-
tion density of area of birth have been noted (Marcelis et al., 1998; Mortensen
et al., 1999). There is also consistent evidence for a small (5–8%) winter–spring
excess of births for both schizophrenia and mania/bipolar disorder (Cotter et al.,
1996; McGrath and Welham, 1999; Mortensen et al., 1999; Torrey et al., 1997).
The cause or causes of the “season of birth” effect and the “urban birth” effects
are not as yet clear. Both are proxy variables that could encompass a large

Neurodevelopment and Schizophrenia, ed. Matcheri S. Keshavan et al. Published by Cambridge University
Press.

C Cambridge University Press 2004.

191
192 M. Cannon, K. Dean, and P. B. Jones

number of possible risk factors, including prenatal and childhood infections, the
prevalence of which tends to increase with increasing population density and in
winter.

Prenatal influenza
The early evidence regarding the role of prenatal infections has come from ecolog-
ical, or population-association studies. In 1988, Mednick and colleagues demon-
strated that the offspring of women who were in the second trimester of pregnancy
during the 1957–8 influenza A2 pandemic in Helsinki were about twice as likely
to be hospitalized with a diagnosis of schizophrenia as those not exposed dur-
ing pregnancy or exposed earlier or later in pregnancy. There have been many
attempts to replicate this finding for the 1957 epidemic using an ecological design
(Erlenmeyer-Kimling et al., 1994; Izumoto et al., 1999; Kendell and Adams, 1991;
Kendell and Kemp, 1989; Kunugi et al., 1995; McGrath et al., 1994; Mednick et al.,
1990; O’Callaghan et al., 1991; Torrey et al., 1992). The effect sizes demon-
strated in these studies are small: somewhere between 1.5 and 2.0 (Cannon and
Jones, 1996), and not all ecological studies replicated the association (Kendell and
Kemp, 1989; Selten and Slaets, 1994; Selten et al., 1999; Susser et al., 1994; Torrey
et al., 1988; Westergaard et al., 1999). However, the balance of evidence does suggest
that prenatal exposure to the 1957 epidemic was associated with a raised incidence
of schizophrenia, particularly if the exposure occurred in the second trimester of
pregnancy (reviewed by McGrath et al., 1995a; Wright et al., 1999). Studies of
longer-term trends in the association between the timing of influenza epidemics
and the birth dates of people with schizophrenia have yielded generally positive
results in most (Adams et al., 1993; Barr et al., 1990; Sham et al., 1992; Takei
et al., 1994) but not all (Morgan et al., 1997; Selten and Slaets, 1994; Torrey et al.,
1988) cases. In most studies where the sexes are examined separately, the positive
association was found mainly or exclusively in females (Adams et al., 1993; Izumoto
et al., 1999; Kendell and Kemp, 1989; McGrath et al., 1994; Mednick et al., 1990;
O’Callaghan et al., 1991; Takei et al., 1994), but no satisfactory explanation has yet
been offered for this.
The ecologic design has undoubted limitations, including the so-called “ecolog-
ical fallacy” (we cannot be certain that the individuals in the population who were
exposed to influenza in utero are the same individuals who are diagnosed with
schizophrenia as adults) and the possibility of unknown confounding with other
factors, such as maternal fever or medication.
However, the application of cohort and case–control methodology has not
shed much light on the problem. Two case–control studies replicated the associa-
tion between second-trimester exposure to influenza and later schizophrenia but
relied on maternal recall for information about the exposure (Stöber et al., 1992;
193 Early environmental risk factors for schizophrenia

Wright et al., 1995). Two cohort studies have failed to support the influenza–
schizophrenia association (Cannon et al., 1996; Crow and Done, 1992) but had low
power to examine this effect (Adams and Kendell, 1996).

Prenatal rubella and other prenatal infections

Interest in prenatal infection as a risk factor for schizophrenia is not restricted to
influenza. Influenza lends itself to study because of the frequent occurrence of well-
defined epidemics. Brown et al. (2000a) showed that second-trimester exposure to
a wide variety of respiratory infections (including influenza, pneumonia, tubercu-
losis and acute bronchitis among others) was associated with significant (twofold)
increased risk of schizophrenia spectrum disorders. These results indicate that sev-
eral infections, both bacterial and viral, may increase the risk of schizophrenia
through some common pathogenic mechanism.
Brown and colleagues (2000b) investigated a cohort of individuals who were
serologically documented to have sustained in utero exposure to rubella and found
that the rubella-exposed subjects, most of whom were exposed in the first trimester,
had a substantially higher risk (relative risk, 5.2) of developing non-affective psy-
choses than those who were not exposed, independent of hearing status. The cohort
methodology with proof of individual exposure status is robust and the effect size for
the association between prenatal rubella and schizophrenia is much larger than that
reported for prenatal influenza. These findings suggest that non-affective psychosis
may be a remote neuropsychiatric consequence of prenatal rubella. An ecologi-
cal study from Finland by Suvisaari and colleagues (1999) found an association
between second-trimester exposure to poliovirus infection and later schizophre-
nia. Jones and colleagues (1999) have followed a large group of individuals whose
mothers were identified during pregnancy as suffering from identified viral infec-
tions (Fine et al., 1985) and have found evidence relating neurotropic virus exposure
with a range of adverse central nervous system (CNS) outcomes, including mental
retardation, epilepsy, and psychosis.
The availability of stored prenatal serum from birth cohorts in the USA who
are now in the period of risk for schizophrenia has opened up a new research
avenue for investigating prenatal infection as a risk factor for psychiatric illness
(Susser et al., 2000). Using stored prenatal serum samples from the Providence
cohort of the National Collaborative Perinatal Project, Buka et al. (2001) found a
strong association between maternal antibodies to herpes simplex virus type 2 gG2
glycoprotein and later psychosis (odds ratio, 5.8; 95% confidence interval, 1.7–
19.3), and have subsequently replicated this finding in another sample from the
same cohort (Buka et al., 2003). Analysis of prenatal serum from a birth cohort in
Oakland, California (Prenatal Determinants of Schizophrenia Study) has produced
the first serological evidence of an association between prenatal influenza infection
194 M. Cannon, K. Dean, and P. B. Jones

in the first half of pregnancy and an increased (threefold) risk of adult schizophrenia
(Brown et al., 2003).

Neonatal and early childhood infection

The time window during which early exposure to infection may exert an effect
appears to extend into childhood. The North Finland 1966 Birth Cohort con-
tains all live births (12 058 in total), to women in northern Finland (Oulu and
Lapland provinces) during 1966 (Rantakallio, 1969). Rantakallio and colleagues
(1997) identified through record linkage all cohort members who had been hospit-
alized during childhood for a CNS infection (mainly encephalitis and meningitis).
They then identified those members of the cohort who had been hospitalized for
psychiatric illness. Of 145 in the group with serologically diagnosed childhood CNS
infections who had survived to age 16 years, four (2.8%) developed schizophrenia,
compared with 0.7% of the unexposed majority. This fourfold relative risk may
even underestimate the true risk because of the restriction to severe exposure. An
estimate of the population attributable fraction (the amount by which the popu-
lation burden of schizophrenia would be reduced if the effect of the exposure, if
causal, were removed) is around 4%. As with any causal inference, one must con-
sider the possibility of reverse causality. The CNS may already have been abnormal
in the individuals who developed infection and schizophrenia, and these abnormal-
ities may have made them liable to CNS infection. Unknown genetic or epigenetic
factors may be necessary components of the causal pathway.

Prenatal malnutrition
A series of ecological studies of exposure to prenatal famine (Susser and Lin, 1992,
1994; Susser et al., 1996) have demonstrated dose–response relationships between
maternal nutritional deprivation during the Nazi blockade of the Netherlands in
the winter of 1944–5 and risk of later schizophrenia in the offspring. The initial
finding to emerge from these studies was that birth cohorts exposed to the famine in
early but not late gestation had a twofold increase in risk for schizophrenia (Susser
and Lin, 1992; Susser et al., 1996). A subsequent study using military conscription
data demonstrated that prenatal famine during early gestation was associated with
a twofold elevation in risk for schizoid or schizotypal personality disorders (Hoek
et al., 1996). The authors postulated that severe nutritional deprivation is the etio-
logical mechanism involved (Butler et al., 1999; Hoek et al., 1999). This possibility
has received indirect support from a finding that a short interval between siblings
is associated with an increased risk of schizophrenia (Westergaard et al., 1999).
There is biological plausibility for this putative cause; congenital CNS defects in
this population were related to famine exposure in a similar fashion (Susser et al.,
195 Early environmental risk factors for schizophrenia

1985). Interaction with maternal genetic effects or exposure to maternal stress may
also be components of any causal association (Koenig et al., 2002).

Rhesus incompatibility
Rhesus (Rh) incompatibility, characterized by an Rh-negative mother preg-
nant with an Rh-positive fetus, has been associated with an elevated risk for
schizophrenia. Hollister et al. (1996) used data on males from the Danish Perinatal
Cohort Study and found an elevated risk for schizophrenia among offspring of
Rh-incompatible pregnancies compared with Rh-compatible pregnancies. Rhesus
incompatibility can give rise to hemolytic disease of the newborn, which results,
among other things, in childhood neuromotor abnormalities and behavioral dis-
orders such as emotional instability. It is possible that schizophrenia is yet another
possible consequence of rhesus hemolytic disease (Hollister and Brown, 1999).
Rhesus hemolytic disease occurs most commonly in mothers who have already
delivered a Rh-positive child, thus triggering the production of the antibody against
the Rh(D) antigen. As hypothesized, the risk for schizophrenia among males in the
Danish Perinatal Cohort Study was increased over threefold in second and later-
born offspring from Rh-incompatible pregnancies, but there was no increased risk
for first-born offspring (Hollister et al., 1996).

Minor physical anomalies

The term minor physical anomalies (MPAs) is used to refer to a group of subtle but
observable defects of the head, face, hands, and feet. The concept encompasses both
qualitative and quantitative abnormalities, such as facial asymmetries, the presence
of clefts, misshapen appendages, and a variety of abnormal skull measurements.
These MPAs are known to arise from abnormalities occurring in ectodermal devel-
opment before the third trimester of pregnancy. The precise cause of such abnormal
development is not yet known but is likely to involve the interacting effects of both
genes and the prenatal environment. Examination of MPAs is ideally suited to gain-
ing insight into the prenatal period because, once produced in utero, they persist
largely unchanged throughout life. MPAs can act as a marker of neurodevelopmen-
tal disorder since the brain is also ectodermally derived.
There is now a strong body of evidence demonstrating that the prevalence of
MPAs is elevated in sufferers of schizophrenia compared with controls (Gualtieri
et al., 1982; Ismail et al., 1998, 2000; Lane et al., 1997; McGrath et al., 1995b, 2002).
These findings lend weight to the neurodevelopmental model of schizophrenia.
Patients with schizophrenia demonstrate an excess of multiple dysmorphic features
of the craniofacial and skull base regions (Lane et al., 1997; McGrath et al., 2002),
including high palate, abnormal palatal ridges, abnormal supraorbital ridges, pres-
ence of bifid tongue, presence of epicanthus, narrow mouth, widened anterior ear
196 M. Cannon, K. Dean, and P. B. Jones

helix, ear protrusion, abnormal eye fissure inclination, abnormal binocular diam-
eter, wide skull base, and hypoplastic ear lobe size. A novel method for studying
craniofacial dymorphology is the use of three-dimensional laser surface scanning
and geometric morphometric analysis (Hennessy et al., 2002), and its application
to psychosis is in progress. Elevated rates of MPAs have also been reported in
other psychoses (McGrath et al., 2002) and disorders such as schizotypal personal-
ity disorder (Weinstein et al., 1999); MPAs appear to be relatively independent of
obstetric complications in their association with later development of schizophre-
nia (McNeil and Cantor-Graae, 1999, 2000). MPAs are increased in those at genetic
risk for schizophrenia in some (Lawrie et al., 2001; Schiffman et al., 2002) but not
all (Griffiths et al., 1998) studies that have examined this issue. The association
between MPAs and schizophrenia does not appear to be confounded by ethnicity
(Dean et al., 2003).

Prenatal exposure to stress

Maternal experience of extremely stressful events, such as death of spouse
(Huttunen and Niskanen, 1978), war (van Os and Selten, 1998), tornado (Kinney
et al., 1999), or nuclear explosion (Imamura et al., 1999) appears to increase the
risk of schizophrenia in offspring who are in midgestation at the time (for review
see Kinney, 2001). More subtle forms of prenatal stress may also have an effect.
In the 1966 North Finland Birth Cohort, both maternal depression in late preg-
nancy (Jones et al., 1998) and “un-wantedness” of a pregnancy by the mother
(Myhrman et al., 1996) were independently associated with a modest increase
in the risk of schizophrenia in the offspring. Many potential confounders, such
as maternal age, physical conditions, and socioeconomic status, were taken into
account, but a diluted genetic effect cannot be excluded.
There is a considerable animal literature concerning the effects of prenatal stress
on brain development and behavior of the offspring (reviewed by Koenig et al.,
2002; Suomi, 1997). Animals exposed to prenatal stress produce elevated levels of
cortisol in response to stressful situations, indicative of enhanced hypothalamic–
pituitary–adrenal axis reactivity. These effects are robust and persistent, indicating
that this axis can be “programmed” during fetal life. Prenatal stress is associated
with smaller head circumference, neonatal neurological impairment (Lou et al.,
1994), and behavior problems in childhood (O’Connor et al., 2002). Stress during
birth also appears to have an effect (Taylor et al., 2000). Prenatal stress in human
subjects is also associated with neonatal neurological impairment (Lou et al., 1994),
increased response of the infant to stress (Taylor et al., 2000), and behavior problems
in childhood (O’Connor et al., 2002), which further support the heuristic value of
a prenatal stress model for schizophrenia. Prenatal stress is associated with low
birth weight and preterm birth (Lou et al., 1994), which have been reported as
197 Early environmental risk factors for schizophrenia

risk factors for schizophrenia. This effect is possibly mediated by uterine artery
vasoconstriction and reduced placental blood flow (Teixeira et al., 1999). Prenatal
stress may be particularly pathogenic in those already at genetic risk of schizophrenia
(Mednick and Schulsinger, 1968; Mednick et al., 1971).

Perinatal risk factors: obstetric complications

There is a large literature on the possible role of obstetric complications in
schizophrenia (reviewed by Cannon, 1997; McNeil, 1988, 1995; McNeil et al., 2000).
Before 1997, most of the studies showing an association were of case–control design
using maternal recall for assessment of the exposure. They also had small sample
sizes and were prone to various forms of bias. Geddes and Lawrie (1995) carried
out a meta-analysis of the results published from 16 case–control studies and two
cohort studies and reached three main conclusions: (i) a pooled odds ratio of 2.0
(95% confidence interval 1.6–2.4) suggested that obstetric complications did have a
small effect on increasing risk for schizophrenia; (ii) there was evidence for selection
and publication bias; (iii) there was significant heterogeneity between the results
from the case–control studies and the two birth cohort studies.
Because of methodological problems associated with the classic case–control
design, particularly when measuring such early exposures, cohort and population-
based study designs have now been exploited to investigate the relationship between
obstetric complications and later schizophrenia. Results of these population-based
studies are given in Table 11.1. They all have the following characteristics: large and
psychiatrically well-defined samples of schizophrenia patients were drawn from
population-based registers; standardized, prospectively collected obstetric infor-
mation was obtained from birth records or registers; controls were from the general
population with obstetric complication information from the same source and con-
text; demographic confounding factors were controlled through either matching
or statistical adjustment. It can be seen from Table 11.1 that these methodological
advances have not confirmed the earlier findings. Moreover, although the designs
of these studies are broadly similar, their results are quite diverse and no consensus
has yet been reached. The methodological reasons for this include lack of defi-
nition of exposures, confounding by period and cohort effects, and insufficient
statistical power (for further detail see Zornberg et al. (2000a) and Cannon et al.
(2002a)). The issue of interactive effects is especially problematic. Pregnancy, birth,
and neonatal complications do not act independently of each other. Pregnancy fac-
tors that have been associated with schizophrenia, such as rhesus incompatibility
and prenatal stress (see above), can increase the risk for hypoxic–ischemic dam-
age during delivery. Current studies have negligible statistical power to detect such
interactive effects. Some studies try to overcome these problems by examining only
198 M. Cannon, K. Dean, and P. B. Jones

Table 11.1. Summary of results of population-based studies of obstetric complications

and schizophrenia

Study Country Design (cohort or Findings

case–control)

Done et al. UK Cohort: NCDS No overall association between risk factors for perinatal
(1991) death and later schizophrenia. Low maternal weight
and medications given to baby were associated with
narrowly defined schizophrenia
Buka et al. US Cohort: NCPP Non-significant increased risk of psychosis in those
(1993) (Providence exposed to chronic fetal hypoxia (OR, 2.6)
center)
Sacker et al. UK Cohort: NCDS Re-analysis of Done et al. (1991). The following
(1995) variables were associated with increased risk of
narrowly defined schizophrenia: low maternal
weight; maternal psychological problems; smoking in
pregnancy; poor antenatal attendance; rhesus
negative; parity > 2; previous births < 2500 g;
bleeding in pregnancy; untrained person delivering;
baby’s weight < 2500 g; other drugs given to baby
Jones et al. Finland Cohort: 1966 North Low birth weight (OR, 2.4; CI, 1–5.6): combination of
(1998) Finland Birth low birth weight and prematurity (OR, 3.5; CI,
Cohort 1.3–9.6) and perinatal brain damage (OR, 6.9; CI,
2.9–16.3) were associated with schizophrenia
Hultman et al. Sweden Case–control Schizophrenia was associated with multiparity (OR,
(1999) 2.0); bleeding during pregnancy (OR, 3.5); winter
birth (OR, 1.4); SGA (males only OR, 3.2); parity > 4
(males only OR, 3.6)
Dalman et al. Sweden Cohort Increased risk of pre-eclampsia (OR, 2.5); vacuum
(1999) extraction (OR, 1.7); malformations (OR, 2.4);
parity of 1 (OR, 1.3); bleeding during pregnancy
(OR, 2.0); threatened premature delivery (OR, 2.3);
gestational age < 32 weeks (OR, 2.7); prolonged
delivery (OR, 1.6); uterine inertia (OR, 2.4; ponderal
index < 20 (OR, 3.4); respiratory illness (OR, 1.5);
birth weight < 2500 g (males only OR, 2.2); birth
weight < 1500 g (females only OR, 6.0); SGA (males
only OR, 1.9). Preeclampsia remained significant
after adjusting for all other complications
Kendell et al. Scotland Case–control No association found between any OC and
(2000) schizophrenia
Kendell et al. Scotland Case–control Emergency Cesarian section (OR, 3.7; CI, 1.02–13.1)
(2000) and labor > 12 hours were associated with
subsequent schizophrenia
199 Early environmental risk factors for schizophrenia

Table 11.1. (cont.)

Study Country Design (cohort or Findings

case–control)

Byrne et al. Ireland Case–control No overall differences in complication rates. Cesarian

(2000) section and narrow maternal pelvis commoner
among cases. Males with early-onset schizophrenia
had greater frequency and severity of OCs than
controls.
Cannon et al. US Cohort: NCPP Hypoxia-related complications > 3 increased the risk of
(2000) (Philadephia schizophrenia (OR, 3.84) and particularly early-onset
center) schizophrenia (OR, 7.3)
Rosso et al. Finland Case–control Risk of early-onset schizophrenia increased (OR, 2.16;
(2000) CI, 1.3–3.5) per hypoxia-related OC
Zornberg et al. US Cohort: NCPP Re-analysis of Buka et al. (1993). Composite variable
(2000b) (Providence entitled “hypoxic–ischaemia-related fetal/neonatal
centre) complications” associated with schizophrenia (OR,
4.56; CI, 2.42–8.6)
Dalman et al. Sweden Case–control Signs of asphyxia at birth were associated with increased
(2001) risk for schizophrenia (OR, 4.4; CI, 1.9–10.3) after
adjusting for other complications and confounders

NCPP, National Collaborative Perinatal Project: NCDS, National Child and Development Study; OR, odds
ratio; CI, confidence interval; SGA, small for gestational age; OC, obstetric complication.

one exposure based on a prior hypothesis, and these have showed significant effects
for the putative risk increasing mechanism of hypoxic–ischemic damage (Buka
et al., 1993; Cannon et al., 2000; Rosso et al., 2000; Zornberg et al., 2000b).
The standardized fashion of reporting results and the methodological similarities
of the population-based studies listed in Table 11.1 lend themselves to a meta-
analytic approach. Meta-analysis provides a method for integrating quantitative
data from multiple studies by using a weighted average of the results in which
larger studies have more influence than smaller studies. It improves the estimates
of effect size, increases the statistical power, and helps to make sense out of studies
with conflicting conclusions (Egger et al., 1997; Fleiss, 1993).
Cannon et al. (2002a) carried out a meta-analytic synthesis of the prospective
“population-based” studies that had published data on individual obstetric com-
plications and found that three groups of complications were significantly associ-
ated with schizophrenia: (i) complications of pregnancy (bleeding, diabetes, rhe-
sus incompatibility, pre-eclampsia); (ii) complications of delivery (uterine atony,
asphyxia, emergency cesarean section); and (iii) abnormal fetal growth and develop-
ment (low birth weight, congenital malformations, reduced head circumference).
200 M. Cannon, K. Dean, and P. B. Jones

Pooled estimates of effect sizes were generally small (odds ratios 2). Interactive
effects and independence of obstetric risk factors could not be examined using this
design. An individual patient-data meta-analysis on these population-based stud-
ies, such as that carried out by Geddes et al. (1999) on case–control studies, may
help to elucidate such issues.
The period of risk during which hypoxic brain damage may lead to later
schizophrenia appears to extend beyond birth. In the North Finland 1966 cohort, a
group of CNS insults that had hypoxia as a common mechanism was identified and
termed perinatal brain damage (Rantakallio et al., 1987). Of 125 survivors of such
perinatal brain damage, six (4.8%) developed schizophrenia in adult life: a sevenfold
relative risk (Jones et al., 1998). The estimate of the proportion of schizophrenia
in the general population that may be attributable to this mechanism was 5–8% in
this study.

Neonatal risk factors: early rearing environment

Human brain development does not stop after birth; the brain continues to develop
substantially during the first 2 years of life, during which time it is particularly
susceptible to adverse environmental effects, not only those of a physical nature
but also from interpersonal affective experiences (Schore, 2001; Siegel, 2001). The
central role of the early rearing environment in influencing later emotional and
psychological outcomes has long been recognized (Bowlby et al., 1944; Kendler
et al., 1992; Murray et al., 1996; Nemeroff, 1999; Rutter, 2002) and is also rele-
vant to schizophrenia (Schiffman et al., 2001). Unwantedness of the pregnancy
(Myhrman et al., 1996), antenatal depression (Jones et al., 1998), atypical mother–
infant interaction (Cannon et al., 2002b), poor mothering (Jones et al., 1994), and
early parental loss (Agid et al., 1999) have been variously associated with increased
risk of schizophrenia. Conversely, a good adoptive family environment appears to
act as a protective factor among adoptees at genetic risk for the disorder (Tienari,
1991; Wahlberg et al., 1997).

Conclusions
It seems that many pre- and perinatal risk factors are somehow involved in increas-
ing the risk for schizophrenia in later life. The best evidence to date is for prenatal
(probably second-trimester) exposure to influenza and other respiratory infections,
prenatal rubella, hypoxia-related obstetric complications and low birth weight or
intrauterine growth retardation. Evidence is less secure for prenatal stress or prena-
tal malnutrition, principally because of the difficulties in obtaining suitable samples
in which to examine these exposures.
201 Early environmental risk factors for schizophrenia

Table 11.2. Estimate of approximate effect sizes for pre- and perinatal risk
factors for schizophrenia

Pre- or perinatal risk factor Approximate effect size

(RR or OR)

Place or time of birth

Winter birth 1.15
Urban birth 1.5–2.4

Infection
Prenatal influenza 2.0
Prenatal respiratory infection (2◦ ) 2.1
Prenatal rubella 5.2
Prenatal poliovirus (2◦ ) 1.05
Neonatal and childhood CNS infection 4.0

Malnutrition
Prenatal famine (1◦ ) 2.0

Prenatal stress
Bereavement of spouse 6.2
Flood (2◦ ) 1.8
War (2◦ )
“Unwantedness” 2.4
Maternal depression (3◦ ) 1.8

Obstetric complications
General 2.0
Rhesus incompatibility 2.8
Hypoxia-related 2.1–4.4
Perinatal brain damage 7.0
Low birth weight (< 2500 g) 1.6
Pre-eclampsia 2.5

RR, relative risk; OR, odds ratio; 1◦ , 2◦ , 3◦ , first, second, and third trimester,
respectively.

The effect sizes for these prenatal and perinatal risk factors are small, with odds
ratios or relative risks of approximately 2 (Table 11.2). Similarly, genetic studies
since the early 1990s have suggest that multiple genes with small effect sizes are
involved in causation of schizophrenia. These and other findings indicate that we
are likely to be dealing with interactive effects of prenatal and genetic factors (Jones
and Murray, 1991; Tsuang, 2000). Recent findings confirm earlier reports that
increasing paternal age is associated with an increased risk of schizophrenia (Hare
202 M. Cannon, K. Dean, and P. B. Jones

and Moran, 1979; Malaspina et al., 2001). The most parsimonious interpretation
attributes the effect to mutations in the male sperm line prior to conception. These
mutations, in turn, are to a large extent environmentally determined, by both the
cumulative life experience (age) and specific environmental exposures (e.g. toxins)
of the biological father.
Techniques for investigating genetic and environmental risk factors are beginning
to converge, with case–control and cohort designs being used for genetic associa-
tion studies. Molecular genetic studies are including measures of environment and
cohort studies are collecting both DNA samples and information on early and later
environmental risk factors (Caspi et al., 2002, 2003). Over the next few years, we
should see the first reports of studies examining precisely measured genetic and
early environmental causes of schizophrenia in the same populations.

Acknowledgements
This work was made possible by the support of the Wellcome Trust (MC), NARSAD
(MC) and the Theodore and Vada Stanley Foundation (PBJ).

REFERENCES

Adams, W., Kendell, R. E. (1996). Influenza and schizophrenia (letter). Br J Psychiatry 169:
791–792.
Adams, W., Kendell, R. E., Hare, E. H. (1993). Epidemiological evidence that maternal influenza
contributes to the aetiology of schizophrenia: an analysis of Scottish, English and Danish
data. Br J Psychiatry 163: 522–534.
Agid, O., Shapira, B., Zislin, J., et al. (1999). Environment and vulnerability to major psychiatric
illness: a case control study of early parental loss in major depression, bipolar disorder and
schizophrenia. Mol Psychiatry 4: 163–172.
Barr, C. E., Mednick, S. A., Munk-Jorgensen, P. (1990). Exposure to influenza epidemics during
gestation and adult schizophrenia: a 40 year study. Arch Gen Psychiatry 47: 869–874.
Bowlby, J. (1944). Forty four juvenile thieves: their characters and home life. Int J Psychoanalysis
25: 19–52, 107–127.
Brown, A. S., Cohen, P., Greenwald, S., Susser, E. (2000a). Nonaffective psychosis after prenatal
exposure to rubella. Am J Psychiatry 157: 438–443.
Brown, A. S., Schaefer, C. A., Wyatt, R. J., et al. (2000b). Maternal exposure to respiratory
infections and adult schizophrenia spectrum disorders: a prospective birth cohort study.
Schizophr Bull 26: 287–296.
Brown, A. S., Begg, M., Gravenstein, S., et al. (2003). Serologic evidence for prenatal influenza in
the etiology of schizophrenia. Schizophr Res 60: 34.
203 Early environmental risk factors for schizophrenia

Buka, S. L., Tsuang, M. T., Lipsitt, L. P. (1993). Pregnancy/delivery complications and psychiatric
diagnosis. A prospective study. Arch Gen Psychiatry 50: 151–156.
Buka, S. L., Tsuang, M. T., Torrey, E. F., et al. (2001). Maternal infections and subsequent psychosis
among offspring. Arch Gen Psychiatry 58: 1032–1037.
Buka, S. L., Tsuang, M. T., Goldstein, J. M., et al. (2003). Maternal exposure to herpes simplex
virus type 2 and psychosis among adult offspring: replication and reanalysis. Schizophr Res
60: 35.
Butler, P. D., Printz, D., Klugewicz, D., Brown, A. S., Susser, E. S. (1999). Plausibility of early
nutritional deficiency as a risk factor for schizophrenia. In Prenatal Exposures in Schizophre-
nia, ed. E. S. Susser, A. S. Brown, J. M. Gorman. Washington, DC: American Psychiatric
Press, pp. 163–193.
Byrne, M., Browne, R., Mulryan, N., et al. (2000). Labour and delivery complications and
schizophrenia. Case–control study using contemporaneous labour ward records. Br J Psy-
chiatry 176: 531–536.
Cannon, M., Jones, P. (1996). Schizophrenia: neuroepidemiology of schizophrenia. J Neurol
Neurosurg Psychiatry 61: 604–613.
Cannon, M., Cotter, D., Sham, P. C., et al. (1996). Schizophrenia in an Irish sample following
prenatal exposure to the 1957 influenza epidemic: a case-controlled, prospective follow-up
study. Schizophr Res 11: 95.
Cannon, M., Jones, P. B., Murray R. M. (2002a). Obstetric complications and schizophrenia.
Historical and meta-analytic review. Am J Psychiatry 159: 1080–1092.
Cannon, M., Caspi, A., Moffott, T. E., et al. (2002b). Evidence for early-childhood, pan-
developmental impairment specific to schizophreniform disorder. Results from a longi-
tudinal birth cohort. Arch Gen Psychiatry 59: 449–456.
Cannon, T. D. (1997). On the nature and mechanisms of obstetric influences in schizophrenia:
a review and synthesis. Int Rev Psychiatry 9: 387–397.
Cannon, T. D., Rosso, I. M., Hollister, J. M., et al. (2000). A prospective cohort study of
genetic and perinatal influences in the etiology of schizophrenia. Schizophr Bull 26: 351–
366.
Caspi, A., McClay, J., Moffitt, T. E., et al. (2002). Role of genotype in the cycle of violence in
maltreated children. Science 297: 851–854.
Caspi A., Sugden, K., Moffitt, T. E., et al. (2003). Influence of life stress on depression: moderation
by a polymorphism in the 5-HTT gene. Science 301: 386–389.
Cotter, D., Larkin, C., Waddington, J. L., O’Callaghan, E. (1996). Season of birth in schizophrenia:
clue or cul-de-sac. In The Neurodevelopmental Basis of Schizophrenia, ed. J. L. Waddington,
P. B. Buckley. Austin, TX: Landes, pp. 17–33.
Crow, T. J., Done, J. (1992). Prenatal exposure to influenza does not cause schizophrenia. Br J
Psychiatry 161: 390–393.
Dalman, C., Allebeck, P., Culberg, J., Grunewald, C., Köster, M. (1999). Obstetric complications
and the risk of schizophrenia: a longitudinal study of a national birth cohort. Arch Gen
Psychiatry 56: 234–240.
Dalman, C., Thomas, H. V., David, A. S., et al. (2001). Signs of asphyxia at birth increase the risk
of schizophrenia. A population-based case–control study. Br J Psychiatry 179: 403–408.
204 M. Cannon, K. Dean, and P. B. Jones

Dean, K., Fearon, P., Dazzan, P., et al. (2003). Increased minor physical anomalies in all ethnic
groups in the AESOP first-onset study. Schizophr Res 60.
Done, J., Johnstone, E. C., Frith, C. D., et al. (1991). Complications of pregnancy and delivery in
relation to psychosis in adult life: data from the British Perinatal Mortality Survey. Br Med
J 302: 1576–1580.
Egger, M., Davey Smith, G., Phillips, A. N. (1997). Meta-analysis: principles and procedures.
Br Med J 315: 1533–1537.
Erlenmeyer-Kimling, L., Folnegovic, Z., Hrabic-Zerjavic. V., Borcic, B., Folnegovic-Smalc, V.
(1994). Schizophrenia and prenatal exposure to the 1957 influenza epidemic in Croatis. Am
J Psychiatry 151: 1496–1498.
Fine, P. E. M., Adelstein, A. M., Snowman, J., Clarkson, J. A., Evans, S. M. (1985). Long term
effects of exposure to viral infections in utero. Br Med J 290: 509–511.
Fleiss, J. L. (1993). The statistical basis of meta-analysis. Stat Meth Med Res 2: 121–145.
Geddes, J. R., Lawrie, S. M. (1995). Obstetric events in schizophrenia: a meta-analysis. Br J
Psychiatry 167: 786–793.
Geddes, J. R., Verdoux, H., Takei, N., et al. (1999). Schizophrenia and complications of
pregnancy and labor: an individual-patient data meta-analysis. Schizophr Bull 25: 413–
423.
Griffiths, T. D., Sigmundsson, T., Takei, N., et al. (1998). Minor physical anomalies in familial
and sporadic schizophrenia: the Maudsley family study. J Neurol Neurosurg Psychiatry 64:
56–60.
Gualtieri, C. T., Adams, A., Shen, C. D., Loiselle, D. (1982). Minor physical anomalies in alcoholic
and schizophrenic adults and hyperactive and autistic children. Am J Psychiatry 139: 640–
643.
Hare, E. H., Moran, P. A. (1979). Raised parental age in psychiatric patients: evidence for the
constitutional hypothesis, Br J Psychiatry 134: 169–177.
Hennessy, R. J., Kinsella, A., Waddington, J. L. (2002). 3D laser surface scanning and geometric
morphometric analysis of craniofacial shape as an index of cerebro-craniofacial morpho-
genesis: initial application to sexual dimorphism. Biol Psychiatry 51: 507–514.
Hoek, H. W., Susser, E., Buck, K. A., et al. (1996). Schizoid personality disorder after prenatal
exposure to famine. Am J Psychiatry 153: 1637–1639.
Hoek, H. W., Brown, A. S., Susser, E. S. (1999). The Dutch famine studies: prenatal nutritional
deficiency and schizophrenia. In Prenatal Exposures in Schizophrenia, ed. E. S. Susser, A. S.
Brown, J. M. Gorman. Washington, DC: American Psychiatric Press, pp. 135–161.
Hollister, J. M., Brown, A. S. (1999). Rhesus incompatibility and schizophrenia. In Prenatal
Exposures in Schizophrenia, ed. E. S. Susser, A. S. Brown, J. M. Gorman. Washington, DC:
American Psychiatric Press, pp. 197–214.
Hollister, J. M., Laing, P., Mednick, S. A. (1996). Rhesus incompatibility as a risk factor for
schizophrenia in male adults. Arch Gen Psychiatry 53: 19–24.
Hultman, C. M., Sparen, P., Takei, N., Murray, R. M., Cnattingius, S. (1999). Prenatal and
perinatal risk factors for schizophrenia, affective psychosis and reactive psychosis of early
onset: case–control study. Br Med J 318: 421–426.
205 Early environmental risk factors for schizophrenia

Huttunen, M., Niskanen, P. (1978). Prenatal loss of father and psychiatric disorders. Arch Gen
Psychiatry 35: 427–431.
Imamura, Y., Nakane, Y., Ohta, Y., Kondo, H. (1999). Lifetime prevalence of schizophrenia among
individuals exposed to atomic bomb radiation in Nagasaki. Acta Psychiatr Scand 100: 344–
349.
Ismail, B., Cantor-Graae, E., McNeil, T. F. (1998). Minor physical anomalies in schizophrenic
patients and their siblings. Am J Psychiatry 155: 1695–1702.
(2000). Minor physical anomalies in schizophrenia: cognitive, neurological and other clinical
correlates. J Psychiatr Res 34: 45–56.
Izumoto, Y., Inoue, S., Yasuda, N. (1999). Schizophrenia and the influenza epidemics of 1957 in
Japan. Biol Psychiatry, 46: 119–124.
Jones, P., Murray, R. (1991). The genetics of schizophrenia is the genetics of neurodevelopment,
Br J Psychiatry 158: 615–623.
Jones, P., Rodgers, B., Murray, R., Marmot, M. (1994). Childhood developmental risk factors for
schizophrenia in the British 1946 birth cohort. Lancet 344: 1398–1402.
Jones, P. B., Rantakallio, P., Hartikainen, A. L., et al. (1998). Schizophrenia as a long-term outcome
of pregnancy, delivery and perinatal complications: a 28-year follow-up of the 1966 North
Finland general population birth cohort. Am J Psychiatry 155: 355–364.
Jones, P. B., Pang, D., Piriach, S., Fine, P. E. M. (1999). Prenatal viral infection and subsequent
mental illness: a long-term cohort study of 6152 subjects. Schizophr Res 36: 45.
Kendell, R. E., Adams, W. (1991). Unexplained fluctuations in the risk for schizophrenia by month
and year of birth. Br J Psychiatry 158: 758–763.
Kendell, R. E., Kemp, J. W. (1989). Maternal influenza in the aetiology of schizophrenia. Arch
Gen Psychiatry 46: 878–882.
Kendell, R. E., McInneny, K., Jusczak, E., Bain, M. (2000). Obstetric complications and schizophre-
nia: two case–control studies based on structured obstetric records. Br J Psychiatry 174:
516–522.
Kendler, K. S., Neale, M. C., Kessler, R. C., Heath, A. C., Eaves, L. J. (1992). Childhood parental
loss and adult psychopathology in women. A twin study perspective. Arch Gen Psychiatry
49: 109–116.
Kinney, D. K. (2001). Prenatal stress and risk for schizophrenia. Int J Mental Health 29: 62–71.
Kinney, D. K., Hyman, W., Greetham, C., Tramer, S. (1999). Increased relative risk for schizophre-
nia and prenatal exposure to a severe tornado. Schizophr Res 36: 45–46.
Koenig, J. I., Kirkpatrick, B., Lee, P. (2002). Glucocorticoid hormones and early brain development
in schizophrenia. Neuropsychopharmacology 27: 309–318.
Kunugi, H., Nanko, S., Takei, N., et al. (1995). Schizophrenia following in utero exposure to the
1957 influenza epidemics in Japan. Am J Psychiatry 152: 450–452.
Lane, A., Kinsella, A., Murphy, P., et al. (1997). The anthropometric assessment of dysmorphic
features in schizophrenia as an index of its developmental origins. Psychol Med 27: 1155–
1164.
Lawrie, S. M., Byrne, M. Miller, P. et al. (2001). Neurodevelopmental indices and the development
of psychotic symptoms in subjects at high risk of schizophrenia. Br J Psychiatry 178: 524–530.
206 M. Cannon, K. Dean, and P. B. Jones

Lewis, G., David, A., Andreasson, S., Allebeck, P. (1992). Schizophrenia and city life. Lancet 340:
137–140.
Lou, H., Hansen, D., Nordentoft, M., et al. (1994). Prenatal stressors of human life affect fetal
brain development. Dev Med Child Neurol 36: 826–832.
Malaspina, D., Harlap, S., Fennig, S., et al. (2001). Advancing paternal age and the risk of
schizophrenia. Arch Gen Psychiatry 58: 313–412.
Marcelis, M., Takei, N., van Os, J. (1999). Urbanization and risk for schizophrenia: does the effect
operate before or around the time of illness onset? Psychol Med 29: 1197–1203.
Marcelis, M., Navarro-Mateu, F., Murray, R., Selten, J. P., van Os, J. (1998). Urbanization and
psychosis: a study of 1942–1978 birth cohorts in the Netherlands. Psychol Med 28: 871–
879.
Marenco, S., Weinberger, D. R. (2000). The neurodevelopmental hypothesis of schizophrenia:
following a trail of evidence from cradle to grave. Dev Psychopathol 12: 501–527.
McDonald, C., Fearon, P., Murray, R. M. (2000). Neurodevelopmental hypothesis of schizophre-
nia 12 years on: data and doubts. In Childhood Onset of Adult Psychopathology, ed. J. L.
Rapaport. Washington, DC: American Psychiatric Press, pp. 193–222.
McGrath, J. J., Welham, J. L. (1999). Season of birth and schizophrenia: a systematic review and
meta-analysis of data from the Southern Hemisphere. Schizophr Res 35: 237–242.
McGrath, J. J., Pemberton, M. R., Welham, J. L., Murray, R. M. (1994). Schizophrenia and the
influenza epidemics of 1954, 1957 and 1959: a southern hemisphere study. Schizophr Res
14: 1–8.
McGrath, J. J., Castle, D., Murray, R. M. (1995a). How can we judge whether or not prenatal expo-
sure to influenza causes schizophrenia. In Neural Development and Schizophrenia. Theory
and Research, ed. S. A. Mednick, J. M., Hollister. New York: Plenum Press, pp. 203–214.
McGrath, J. J., van Os, J., Hoyos, C., et al. (1995b). Minor physical anomalies in psychoses:
associations with clinical and putative aetiological variables. Schizophr Res 18: 9–20.
McGrath, J., El-Saadi, O., Grim, V., et al. (2002). Minor physical anomalies and quantitative
measures of the head and face in patients with psychosis. Arch Gen Psychiatry 59: 458–464.
McNeil, T. F. (1988). Obstetric factors and perinatal injuries. In Handbook of Schizophrenia,
Vol. 3: Nosology, Epidemiology and Genetics, ed. M. T. Tsuang, J. C. Simpson. North Holland:
Elsevier Science, pp. 319–311.
(1995). Perinatal risk factors and schizophrenia: selective review and methodological concerns.
Epidemiol Rev 17: 107–112.
McNeil, T. F., Cantor-Graae, E. (1999). Does preexisting abnormality cause labor–delivery com-
plications in fetuses who will develop schizophrenia? Schizophr Bull 25: 425–435.
(2000). Minor physical anomalies and obstetric complications in schizophrenia. Aust NZ J
Psychiatry 34(Suppl.): S65–S73.
McNeil, T., Cantor-Graae, E., Ishmail, B. (2000). Obstetric complications and congenital mal-
formations in schizophrenia. Brain Res Rev 31: 166–178.
Mednick, S. A., Schulsinger F. (1968). Some premorbid characteristics related to breakdown in
children with schizophrenic mothers. J Psychiat Res 6(Suppl. 1): 267–291.
Mednick, S. A., Mura, E., Schulsinger, F., Mednick, B. (1971). Perinatal conditions and infant
development in children with schizophrenic parents. Soc Biol 18: S103–S113.
207 Early environmental risk factors for schizophrenia

Mednick, S. A., Machon, R. A., Huttunen, M. O., Bonett, D. (1988). Adult schizophrenia following
prenatal exposure to an influenza epidemic. Arch Gen Psychiatry 45: 189–192.
Mednick, S. A., Machon, R. A., Huttunen, M. O. (1990). An update on the Helsinki influenza
project. Letter to Editor. Arch Gen Psychiatry 47: 292.
Morgan, V., Castle, D., Page, A., et al. (1997). Influenza epidemics and incidence of schizophrenia,
affective disorders and mental retardation in Western Australia: no evidence of a major effect.
Schizophr Res 26: 25–39.
Mortensen, P. B., Pedersen, C. B., Westergaard, T., et al. (1999). Effects of family history and place
and season of birth on the risk of schizophrenia. N Engl J Med 340: 603–608.
Murray, L., Fiori-Crowley, A., Hooper, R., Cooper P. J. (1996). The impact of post natal depression
and associated adversity on early mother–infant interactions and later infant outcome. Child
Dev 67: 2512–2526.
Myhrman, A., Rantakallio, P., Isohanni, M., Jones, P. B., Partanen, U. (1996). Unwantedness of a
pregnancy and schizophrenia in the child. Br J Psychiatry 169: 637–640.
Nemeroff, C. (1999). The pre-eminent role of early untoward experience on vulnerability to major
psychiatric disorders: the nature–nurture controversy revisited and soon to be resolved. Mol
Psychiatry 4: 106–108.
O’Callaghan, E., Sham, P., Takei, N., Glover, G., Murray, R. M. (1991). Schizophrenia after
prenatal exposure to 1957 A2 influenza epidemic. Lancet 337: 1248–1250.
O’Connor, T., Heron, J., Golding, J., Beveridge, M., Glover, V. (2002). Maternal antenatal anx-
iety and children’s behavioural/emotional problems at 4 years. Br J Psychiatry 180: 502–
508.
Rantakallio, P. (1969). Groups at risk in low birth weight infants and perinatal mortality. Acta
Paediatr Scand Suppl 193: 1–71.
Rantakallio, P., von Wendt, L., Koivu, M. (1987). Prognosis of perinatal brain damage: a prospec-
tive study of a one year birth cohort of 12 000 children. Early Hum Dev 15: 75–84.
Rantakallio, P., Jones, P., Moring, J., von Wendt, L. (1997). Association between central nervous
system infections during childhood and adult onset schizophrenia and other psychoses: a
28-year follow-up. Int J Epidemiol 26: 837–843.
Rosso, I. M., Cannon, T. D., Huttunen, T., et al. (2000). Obstetric risk factors for early-onset
schizophrenia in a Finnish birth cohort. Am J Psychiatry 157: 801–807.
Rutter, M. (2002). The interplay of nature, nurture and development. Arch Gen Psychiatry 59:
996–1001.
Sacker, A., Done, D. J., Crow, T. J., Golding, J. (1995). Antecedents of schizophrenia and affective
illness: obstetric complications. Br J Psychiatry 166: 734–741.
Schiffman, J., Abrahamson, A., Cannon, T., et al. (2001). Early rearing factors in schizophrenia.
Int J Mental Health 30: 3–16.
Schiffman, J., Ekstrom, M., LaBrie, J., et al. (2002). Minor physical anomalies and schizophrenia
spectrum disorders: a prospective investigation. Am J Psychiatry 159: 238–243.
Schore, A. N. (2001). Effects of early relational trauma on right brain development, affect regu-
lation, and infant mental health. Infant Ment Health J 22, 201–269.
Selten, J. P. C. J., Slaets, J. P. J. (1994). Evidence against maternal infection as a risk factor for
schizophrenia. Br J Psychiatry 164: 674–676.
208 M. Cannon, K. Dean, and P. B. Jones

Selten, J. P., van der Graaf, Y., van Duursen, R., Gispen-de Wied, C. C., Kahn, R. S. (1999).
Psychotic illness after prenatal exposure to the 1953 Dutch Flood Disaster. Schizophr Res 35:
243–245.
Sham, P. C., O’Callaghan, E., Takei, N., et al. (1992). Schizophrenia following pre-natal exposure
to influenza epidemics between 1939 and 1960. Br J Psychiatry 160: 461–466.
Siegel, D. J. (2001). Toward an interpersonal neurobiology of the developing mind: attachment
relationships, “mindsight”, and neural integration. Infant Ment Health J 22: 67–94.
Stöber, G., Franzek, E., Beckmann, J. (1992). The role of maternal infectious diseases during
pregnancy in the aetiology of schizophrenia in offspring. Eur Psychiatry 7: 147–152.
Suomi, S. J. (1997). Long-term effects of different early rearing experiences on social, emotional,
and physiological development in nonhuman primates. In M. S. Keshavan and R. M. Murray
(Eds.) Neurodevelopment and Adult Psychopathology, Ch. 8, ed. M. S. Keshavan, R. M. Murray.
Cambridge: Cambridge University Press, pp. 104–116.
Susser, E., Lin, S. P. (1992). Schizophrenia after prenatal exposure to the Dutch Hunger Winter
of 1944–1945. Arch Gen Psychiatry 49: 983–988.
(1994). Schizophrenia after prenatal exposure to the Dutch hunger winter of 1944–1945. Arch
Gen Psychiatry 51: 333–334.
Susser, E. S., Lin, S. P., Brown, A. S., Lumey, L. H., Erlenmeyer-Kimling, L. (1994). No relation
between risk of schizophrenia and prenatal exposure to influenza in Holland. Am J Psychiatry
151: 922–924.
Susser, E. S., Neugebauer, R., Hoek, H. W., et al. (1996). Schizophrenia after prenatal famine. Arch
Gen Psychiatry 53: 25–31.
Susser, E. S., Schaefer, C., Brown, A., Begg, M., Wyatt, R. J. (2000). The design of the Prenatal
Determinants of Schizophrenia Study (PDS). Schizophr Bull 26: 257–273.
Susser, M., Hauser, W. A., Keity, J. L., Paneth, N., Stein, Z. (1985). Quantitative estimates of
prenatal and perinatal risk factors for perinatal mortality, cerebral palsy, mental retardation
and epilepsy. In Prenatal and Perinatal Factors Associated with Brain Disorder, ed. J. Freeman.
Washington, DC: Institute of Health, pp. 359–432.
Suvisaari, J., Haukka, J., Tanskanen, A., Hovi, T., Lönnqvist, J. (1999). Association between
prenatal exposure to poliovirus infection and adult schizophrenia. Am J Psychiatry 156:
1100–1102.
Takei, N., Sham, P., O’Callaghan, E., et al. (1994). Prenatal exposure to influenza and the devel-
opment of schizophrenia: is the effect confined to females? Am J Psychiatry 151: 117–119.
Takei, N., Sham, P., O’Callaghan, E., Glover, G., Murray, R. M. (1995). Schizophrenia: increased
risk associated with winter and city birth: a case–control study in 12 regions within England
and Wales. J Epidemiol Community Health 49: 106–107.
Taylor, A., Fisk, N., Glover, V. (2000). Mode of delivery and later stress response. Lancet 355:
120.
Teixeira, J. M. A., Fisk, N. M., Glover, V. (1999). Association between maternal anxiety in preg-
nancy and increased uterine artery resistance index: cohort based study. Br Med J 318:
153–157.
Tienari, P. (1991). Interaction between genetic vulnerability and family environment: the Finnish
adoptive family study of schizophrenia. Acta Psychiatr Scand 84: 460–465.
209 Early environmental risk factors for schizophrenia

Torrey, E. F., Bowler, A. (1990). Geographical distribution of insanity in America: evidence for
an urban factor. Schizophr Bull 16: 591–604.
Torrey, E. F., Rawlings, R., Waldman, I. N. (1988). Schizophrenic births and viral diseases in two
states. Schizophr Res 1: 73–77.
Torrey, E. F., Bowler, A. E., Rawlings, R. (1992). Schizophrenia and the 1957 influenza epidemic.
Schizophr Res 6: 100–107.
Torrey, E. F., Miller, J., Rawlings. R., Yolken, R. H. (1997). Seasonality of birth in schizophrenia
and bipolar disorder: a review. Schizophr Res 28: 1–38.
Tsuang, M. (2000). Schizophrenia: genes and environment. Biological Psychiatry 47: 210–220.
van Os, J., Selten, J.-P. (1998). Prenatal exposure to maternal stress and later schizophrenia: the
May 1940 invasion of the Netherlands. Br J Psychiatry 172: 324–326.
Wahlberg, K.-E., Wynne, L. C., Oja, H., et al. (1997). Gene–environment interaction in vulnera-
bility to schizophrenia: findings from the Finnish Adoptive Family Study of Schizophrenia.
Am J Psychiatry 154: 355–362.
Weinstein, D. D., Diforio, D., Schiffman, J., Wlaker, E., Bonsall, R. (1999). Minor physical anoma-
lies, dermatoglyphic asymmetries and cortisol levels in adolescents with schizotypal per-
sonality disorder. Am J Psychiatry 156: 617–623.
Westergaard, T., Mortensen, P. B., Pedersen, C. B., Wohlfahrt, J., Melbye, M. (1999). Exposure to
prenatal and childhood infections and the risk of schizophrenia. Arch Gen Psychiatry 56:
993–998.
Wright, P., Rifkin, L., Takei, N., Murray, R. (1995). Maternal influenza, obstetric complications
and schizophrenia. Am J Psychiatry 152: 1714–1720.
Wright, P., Takei, N., Murray, R. M., Sham, P. C. (1999). Seasonality, prenatal influenza exposure
and schizophrenia. In Prenatal Exposures in Schizophrenia, ed. E. S. Susser, A. S. Brown,
J. M. Gorman. Washington, DC: American Psychiatric Press, pp. 89–112.
Zornberg, G., Buka, S. L., Tsuang, M. T. (2000a). The problem of obstetrical complications and
schizophrenia. Schizophr Bull 26: 249–256.
(2000b). Hypoxic-ischaemia-related fetal/neonatal complications and risk of schizophrenia
and other nonaffective psychoses: a 19 year longitudinal study. Am J Psychiatry 157: 196–
202.
 12

Transcriptomes in schizophrenia: assessing

altered gene expression with microarrays
David A. Lewis1 , Karoly Mirnics1 , and Pat Levitt2
1
University of Pittsburgh, Pittsburgh, USA
2
Vanderbilt University, Nashville, USA

The diversity of implicated etiological factors and the heterogeneity of clinical

features in schizophrenia have been associated, not surprisingly, with a wide vari-
ety of findings regarding the nature of the brain abnormalities that underlie the
pathophysiology of this illness (Lewis and Lieberman, 2000). Because all of these
abnormalities are either caused by, or reflected in, changes at the molecular level, the
molecular neuropathology of schizophrenia is certain to provide keys to improved
treatment in the near term and disease prevention in the more distant future.
Characterizing the molecular neuropathology of schizophrenia requires, at least in
part, knowledge of the full complement of genes showing altered expression in the
brains of affected individuals. The advent of new technology has made it possible
to determine transcriptomes: the collection of all mRNA transcripts expressed in
a given brain region or cell type. As the product of genomic DNA, the transcrip-
tome is a central determinant of the phenotype of a given brain region or cell
type (Brown and Bottenstein, 1999; Duggan et al., 1999). Therefore, determining
the schizophrenia-related transcriptomes is likely to provide critical insight into
the pathogenetic and pathophysiological processes of this illness. In this chapter,
we review the current status of the technology to assess transcriptomes, its asso-
ciated strengths and limitations, and initial findings resulting from the use of this
approaches to study schizophrenia.

Assessing transcriptomes with microarrays

Although a variety of approaches are available to assess transcriptomes, this review
will focus on the use of DNA microarrays: high-density arrangements of single-
or double-stranded DNA sequences (termed “probes”) that are immobilized at
specific locations on a solid surface or membrane (DeRisi et al., 1996; Lockhart

Neurodevelopment and Schizophrenia, ed. Matcheri S. Keshavan et al. Published by Cambridge University
Press.

C Cambridge University Press 2004.

210
211 Transcriptomes in schizophrenia

et al., 1996; Schena et al., 1995). Microarray technology is based on the well-
established principle of complementary hybridization between nucleic acids; each
microarray probe is capable of recognizing the complementary sequence through
base pairing (Southern et al., 1999). The “target” (e.g. mRNA extracted from a tissue
sample) is hybridized to the immobilized DNA sequence probes on the array sur-
face, an approach made possible by the miniaturization process and development of
high-throughput printing robots. After a series of high-stringency washes, only the
highly complementary probe–target complexes remain tightly bound. The amount
of the retained label is then quantified over each probe spot, with the amount of
retained label or the signal intensity providing a direct reflection of the abundance
of that mRNA species in the sample (Cheung et al., 1999).
Thus, microarrays permit an extensive and relatively rapid screening of the
expression levels of thousands of mRNA transcripts. Indeed, it is now possible to
analyze the expression profile in cells from a schizophrenia patient using microar-
rays that represent a majority of the human genome. Although this approach
is not informative regarding the potential contribution of translational or post-
translational processing events to the disease state, proteins in the postmortem
brain are likely to be less intact than mRNA, and their large-scale analysis is much
more complex. Furthermore, microarrays analyze transcripts that are intrinsic to
the harvested brain region, whereas protein profiles reflect intrinsic and extrinsic
sources, making it more difficult to discern the anatomical source of the observed
disease-related changes.
Two types of microarray platform have been used in studies of schizophrenia.
Complementary DNA (cDNA) microarrays are composed of relatively long cDNA
probes (typically ranging from approximately 100 to 2000 base pairs) that are
usually DNA clones amplified by the polymerase chain reaction (PCR) or expressed
sequence tags (ESTs) derived from the 3 end of RNA targets. Thousands of these
probes are each deposited or synthesized in very small, unique locations on a solid
support (e.g. glass or membrane structure) (Duggan et al., 1999; Schena and Davis,
1999). The sample analysis can be performed using a single label or a dual fluores-
cent method. In a dual-color analysis, two different fluorescent tags (e.g. cyanin 3
and 5 [Cy3 and Cy5, respectively]) are used to label, through a reverse transcription
procedure (Schena et al., 1996; DeRisi et al., 1997), the mRNA species present in
each of the two tissue samples to be compared (e.g. a subject with schizophrenia
and a matched normal control). The samples containing the labeled targets are
combined and hybridized together onto the same microarray. Using two independ-
ent scans that are selective for each of two incorporated fluorescent tags’ emission
wavelengths, a high-resolution fluorescent intensity measurement is obtained from
each spot on the microarray. This signal is quantified, standardized, and compared
between the two images, and the intensity differences are interpreted as relative
212 D. A. Lewis, K. Mirnics, and P. Levitt

differences in transcript abundance between the two samples (e.g. schizophrenia

and control).
The other general type of platform, synthetic oligonucleotide probe arrays, con-
sists of high-density, two-dimensional arrays of synthetic oligonucleotides. The
GeneChip® , manufactured under the proprietary technology of Affymetrix, is cur-
rently the most widely utilized (Lockhart and Barlow, 2001). These glass-supported
arrays are constructed by spatially arranging light-directed combinatorial chemical
synthesis of thousands of different oligonucleotides (Lipshutz et al., 1999; Lockhart
et al., 1996). This photochemical, solid-phase technology facilitates a high spa-
tial resolution and precision of probe placement, which thus permits very high-
density throughput for gene expression analyses. For example, the human U133
GeneChip® contains probe spots as small as 324
m2 on a single 1.28 cm × 1.28 cm
array, such that up to 80% of the human genome is represented across two arrays.
The expression of each gene is interrogated with multiple probes, and the presence
of the transcript of interest is detected using a sophisticated mathematical model-
ing process. Then, after a routine standardization procedure, expression differences
across different arrays (i.e. each array interrogates one sample) are detected using
a statistical algorithm.

Experimental design in microarray studies

The study of gene expression in human brain tissue, the area of interest in
schizophrenia, requires the use of postmortem human brain specimens. Thus, such
studies require an understanding of both the advantages and the limitations intrin-
sic to the use of postmortem material. First, the successful application of microarray
technology to the study of schizophrenia depends upon the use of well-characterized
tissue specimens, in which potential confounds are identified and addressed in the
experimental design (Lewis, 2002). Three general issues are important to consider.
First, the relevant demographic and diagnostic information needs to be obtained
in a standardized fashion for both cases and comparison subjects. Ideally, this
information is acquired from both the review of available medical records and
through structured interviews conducted with surviving relatives and other infor-
mants (Glantz and Lewis, 1997). The latter typically provides data that are not
available from the medical records, as well as an additional way to verify the infor-
mation in these records through collateral sources. In addition to demographic
variables, these investigations yield the clinical characterization of the illness (e.g.
diagnostic signs and symptoms, age of onset, duration of illness, family history of
illness), the treatment of the illness (e.g. medications at the time of death, history
of treatment with other medications, length and number of hospitalizations), and
factors that may be comorbid with the primary diagnosis of interest (e.g. history of
alcohol or other substance abuse, nicotine exposure). This information is essential
213 Transcriptomes in schizophrenia

for the selection of appropriate comparison subjects, including both normal con-
trols and subjects with other disorders who share some features (e.g. medication
history) with the schizophrenia subjects under investigation. In addition, the
detailed clinical information suggests the types of parallel study that may need to be
conducted in animal model systems, where factors such as medication exposure can
be assessed in a controlled fashion. For example, we have used macaque monkeys
treated chronically with antipsychotic drugs in a manner that directly mimics their
use in humans as one way of assessing the potential contribution of these thera-
peutic agents to our microarray findings in subjects with schizophrenia (Mirnics
et al., 2000).
Second, in order to avoid the introduction of potential confounds secondary
to geographic factors, differences in the handling of tissue specimens, or other
variables, all tissue specimens ideally are obtained from the same source. A stan-
dardized approach to blocking the brain facilitates examination and sampling for
neuropathological studies and permits a uniform dissection of the region(s) of
interest at the time that a microarray study is initiated. Photographic documen-
tation of the tissue blocks provides a means to confirm and record that the same
macroscopic landmarks are present in the tissue blocks from all subjects. These
procedures also facilitate the design of cutting and sampling procedures to best
approximate the acquisition of systematic and uniform random samples of the
region of interest. In addition, the preparation of slide-mounted sections from the
tissue blocks that are used for the isolation of RNA provides Nissl-stained material
that can be used to verify that the tissue block contains the characteristic cytoarchi-
tecture of the brain region targeted for study. Other sections are used to confirm
array findings for individual transcripts by in situ hybridization.
Third, it is important to assess variables that may affect the expression or integrity
of mRNA. These variables include agonal state events, the cause and manner of
death, and the postmortem interval (PMI), the period between the time of death
and the freezing of the tissue specimens. Although most mRNA species exhibit
minimal to modest degradation following PMIs of more than 24 hours (comparable
to freshly frozen animal brain tissue), premortem events can have a substantial
negative impact. In particular, antemortem hypoxia and/or acidosis, conditions
frequently found in individuals who die in hospital following medical complications
or sustained periods of illness, appear to be associated with the loss of at least some
mRNA species. Available studies suggest that postmortem brain pH provides an
easily measured and reliable assessment of the severity of these factors (Harrison
et al., 1995). Because the factors monitored by brain pH or PMI may not have
linear effects on mRNA integrity, we prefer to match individual pairs of subjects as
closely as possible on these variables, in addition to including pre- and postmortem
conditions as covariates in statistical analyses.
214 D. A. Lewis, K. Mirnics, and P. Levitt

Microarray data analysis and interpretation

Biological sensitivity differs from the nominal microarray sensitivity of microar-
rays. That is, at a sensitivity of 1 in 250 000 copies of mRNA, both cDNA and
oligonucleotide microarrays can theoretically detect even the lowest abundance
mRNA species in a uniform cell population. However, the complexity introduced
by the many different cell types present in a given region of the brain can result
in a dilution effect such that the relative abundance of mRNAs expressed in only a
subset of cells falls below microarray detection limits. Consequently, the biological
sensitivity may be substantially lower than the nominal microarray sensitivity, and
some relatively rare transcripts (e.g. neurotransmitter receptors), which are readily
identified by other molecular techniques, such as in situ hybridization, will not be
detected by microarrays. Furthermore, cross-hybridization between RNA species
can obscure differences in expression levels for the transcript of interest.
Initially, microarray experiments used “fold change” (i.e. the ratio of the expres-
sion level for a given gene in a subject with schizophrenia relative to that of the
matched control subject) as a descriptor of the magnitude of the measurement
difference across samples. Unfortunately, the reliability of “fold changes” depends
on many parameters, such as probe length and properties, absolute abundance of a
molecule, data standardization, modeling, type of tissue specimen, and array plat-
form. In addition, “fold changes” may be relatively insensitive to marked disease
processes. For example, the loss of 20% of the neurons in a region may produce
only a 0.2-fold change in the transcripts that are expressed in all neurons. Con-
sequently, other statistical approaches are being developed and used (see below)
to uncover expression changes of modest magnitude that, nonetheless, may have
great importance for the disease process under study.
Microarray studies of brain regions do not reveal the cellular source or cause
of any changes in gene expression. For example, a given decrease in the level of a
transcript could be the result of a consistent downregulation of expression across all
cells or the complete absence of expression in a subset of cells. Furthermore, parallel
expression changes in different transcripts may occur in the same cell type or may be
present in different, but functionally related, cell classes. Obviously, obtaining this
type of information is critical for the biological interpretation of the findings, and it
demonstrates the need for microarray data to be supplemented with anatomically
based molecular approaches.
Although changes in protein levels may occur without expression changes in
the underlying transcripts, changes in gene expression usually result in protein
level changes, albeit not necessarily of the same magnitude. However, the sources
of proteins and transcripts may be different in a given brain region. That is,
mRNA signal originates predominantly from cell bodies in the harvested tissue,
whereas proteins are also found in the axon terminals that project to a given region.
215 Transcriptomes in schizophrenia

Consequently, apparent discrepancies in transcript versus the protein change in

tissue homogenates may reflect differences in the anatomical source of the change,
rather than the absence of a change in protein level in association with a change in
transcript expression.

Gene expression changes associated with schizophrenia

In postmortem brain specimens from subjects with schizophrenia, a difference in
the expression level of one or a cluster of genes may have several different meanings.
First, given the apparent polygenic nature of schizophrenia, some transcript dif-
ferences may result from DNA sequence variants that confer disease susceptibility
through changes in expression level of the encoded mRNA. Second, some changes
may represent secondary effects that either contribute to the pathophysiology of
the illness or represent a compensatory response. Finally, some expression changes
may result from the treatment of schizophrenia or be associated with factors that
frequently accompany schizophrenia (e.g. depression, substance abuse or nicotine
exposure). Therefore, in this section, we review existing transcriptome findings
from studies of schizophrenia and consider their potential significance from these
perspectives.
The first microarray study of schizophrenia used a cDNA platform containing
about 8000 genes and ESTs to examine area 9 of the dorsolateral prefrontal cortex
(Mirnics et al., 2000). The initial evaluation of these data involved the a-priori
categorization of the represented transcripts into approximately 250 gene groups.
This approach was used in order to exploit the power of microarray data to reveal
the presence of disease-related transcriptome changes among functionally related
genes, relationships that are not apparent from examining lists of “most-changed”
genes or conducting classical hierarchical analyses. In our own studies, we used a
supervised approach, assuming that a single gene expression deficit in a functional
pathway would be associated with altered expression of the other genes belong-
ing to the same pathway. Although 95% of detectable transcripts showed normal
levels of expression in schizophrenia, this analysis revealed a prominent deficit in
the expression of members of a group of genes (termed the PSYN group) whose
protein products contribute to the mechanisms involved in the presynaptic release
of neurotransmitters. These findings appear to be related to the disease process
of schizophrenia and are not a consequence of its treatment, since (i) compara-
ble findings were observed in individuals with schizophrenia who were on or off
antipsychotic medications at the time of death, and (ii) no deficits in the expres-
sion of PSYN genes were found in monkeys exposed chronically to haloperidol in
a manner, and with serum levels, that mimiced clinical use. In addition to a highly
significant effect at the level of the gene group, the expression deficits of individual
216 D. A. Lewis, K. Mirnics, and P. Levitt

genes within the PSYN group appeared to be subject specific. That is, although
all subjects with schizophrenia showed decreased expression of multiple PSYN
genes, the specific combination of affected transcripts varied from subject to sub-
ject. This phenotypic diversity may reflect a continuum of molecular phenotypes
in schizophrenia, perhaps similar to the variability in the clinical manifestations
of the illness across subjects. In addition to this general molecular diversity seen
across subjects, consistent expression deficits in several genes, such as those for
N-ethylmaleimide-sensitive factor (NSF), synapsin 2, and vesicular ATPase, were
found in almost all subjects with schizophrenia. NSF and its attachment proteins
are critical to provide energy for synaptic vesicle fusion; synapsin II regulates vesicle
availability for exocytosis; and vesicular ATPase is responsible for synaptic vesicle
homeostasis. These expression deficits were confirmed by in situ hybridization,
indicating that they did not represent artifacts of the microarray data analysis or
interpretation procedures.
Associated with these alterations in PSYN genes, robust decreases were also found
in two groups comprising much smaller numbers of genes represented on the arrays.
These two gene groups encode proteins that are specifically related to neurotrans-
mission for gamma-aminobutyric acid (GABA) and glutamate, which together
account for over 95% of cortical synapses. In the glutamate transmission gene
group, glutamate receptor 2 (AMPA2) showed the largest expression decrease, and
the 67 kDa isoform of glutamic acid decarboxylase (GAD67 ) was the most robustly
and consistently affected gene related to GABA neurotransmission (Mirnics et al.,
2000). These single gene findings replicated previous reports using more conven-
tional methods (Akbarian et al., 1995; Eastwood et al., 1995; Volk et al., 2000). The
etiology of these changes remains unknown, yet it is possible that these group-wise
changes reflect changes secondary to the broader alterations in presynaptic function.
For example, a reduction in synaptic activity may lead to the findings of reduced
prefrontal brain-derived neurotrophic factor (BDNF) signaling in schizophrenia
(Hashimoto et al., 2002; Weickert et al., 2003). This, in turn, may account for the
marked and selective reduction in GAD67 mRNA expression observed in the sub-
population of prefrontal GABA neurons (Hashimoto et al., 2003) that express TrkB,
the receptor for BDNF.
Because of the high metabolic demands placed on neurons by the processes
involved in synaptic communication, and given the in vivo evidence for alterations
in prefrontal cortical metabolism in schizophrenia (Berman et al., 1986; Buchs-
baum et al., 1992), we used the well-accepted listing of metabolic groups from the
Kyoto Encyclopedia of Genes and Genomes (KEGG) to identify 70 different clusters
of genes whose protein products are involved in metabolic functions (Middleton et
al., 2002). Remarkably, only five of these metabolic gene groups displayed signifi-
cant alteration in expression in the prefrontal cortex of subjects with schizophrenia.
217 Transcriptomes in schizophrenia

These included the mitochondrial malate shuttle system, transcarboxylic acid cycle,
ornithine/polyamine, aspartate/alanine, and ubiquitin metabolism groups. Veri-
fication of these microarray data by in situ hybridization confirmed reduced
mRNA levels for soluble malate dehydrogenase (MAD1), mitochondrial glutamate–
oxaloacetate transaminase type 2, ornithine decarboxylase antizyme inhibitor,
and ornithine aminotransferase. Interestingly, assessment of expression levels of
individual genes in these pathways for changes in haloperidol-exposed monkeys
revealed an unexpected increase in the expression of MAD1, raising the intriguing
possibility that the antipsychotic treatment may counteract the MAD1 expres-
sion decrease observed in schizophrenia. The MAD1 expression increase was most
prominent in the deep cortical layers, including layer 5, which contains the highest
density of dopamine D2 receptors (Lidow et al., 1989). Together, these findings
suggest that altered prefrontal metabolism may be an important component of the
disease process in schizophrenia, contributing to or reflecting alterations in presyn-
aptic function in the same brain region, and that the reversal of at least some of
these abnormalities may be involved in the therapeutic effects of antipsychotic
medications.
The same microarray data set also revealed multiple changes in genes that play
a role in chemical neurotransmission from the postsynaptic side. In addition to
changes in a subset of glutamate and GABA receptors, we found a marked and
highly consistent reduction in the transcript for a regulator of G-protein signal-
ing protein (RGS4) (Mirnics et al., 2001a). RGS proteins are GTPase-activating
proteins that facilitate hydrolysis of GTP bound to Gα -protein and thus limit the
duration and timing of signaling through G-protein-coupled receptors (De Vries
et al., 2000). Hence, downregulation of RGS4 may compensate for decreased synap-
tic efficacy in the prefrontal cortex by increasing the duration of signaling following
activation of G-protein-coupled receptors. However, a similar magnitude of RGS4
transcript reduction was also present in motor and visual cortices of the same sub-
jects with schizophrenia, regions that differ substantially from the prefrontal cortex
in structure, connectivity, and function. In addition, these changes in RGS4 expres-
sion were not observed in the prefrontal cortex of subjects with major depression
or in monkeys exposed chronically to haloperidol. Consequently, these findings
suggest that decreased RGS4 expression in schizophrenia is not restricted to the
prefrontal cortex, and that this decrease may be specific to the disease process and
not a consequence of potentially confounding factors such as comorbid depression
or antipsychotic medications. Together, these observations support the hypothe-
sis that RGS4 represents a schizophrenia susceptibility gene. Interestingly, RGS4 is
present at chromosomal locus 1q21–22, which has been linked to schizophrenia in
some cohorts (Brzustowicz et al., 2000). Consistent with this hypothesis, evidence
of association between allelic variations in the 5 region of RGS4 and schizophrenia
218 D. A. Lewis, K. Mirnics, and P. Levitt

was found in three family samples, although the pattern of transmission differed
across them (Chowdari et al., 2002). Although replication studies are required to
confirm this finding, these studies do exemplify how expression information from
microarray studies, in concert with positional and functional data, can be used to
identify schizophrenia susceptibility genes.
The concomitant reduction in expression of presynaptic, energy metabolism,
and postsynaptic genes in the same subjects strongly suggests that these findings
are interrelated components of the disease process of schizophrenia, although the
direction of causality among them is unclear at present. It is possible that a primary
deficit in the PSYN group, leading to decreased or inefficient synaptic activity, could
lead to secondary downregulation of energy metabolism and altered signaling at
the postsynaptic membrane. Indeed, we have proposed a testable model (Mirnics
et al., 2001b) which argues that a primary reduction in synaptic efficacy could
account for both the developmental trajectory of schizophrenia and the findings
of decreased markers of cortical synaptic number (Glantz and Lewis, 1997, 2000).
Specifically, we suggested that impaired synaptic function in certain cortical cir-
cuits would initially be compensated, at least to a certain degree, by the normal
exuberant production of synapses, but that these connections would be rendered
more vulnerable to additional alterations as a result of later developmental events.
For example, the adolescence-related pruning of cortical excitatory synapses
(Huttenlocher, 1979; Rakic et al., 1986) would not only reduce the total com-
plement of cortical synapses but might also do so to a greater degree than normal
because of the diminished function of the affected circuits (Mirnics et al., 2001b).
In this regard, the evidence of cognitive abnormalities during childhood of indi-
viduals who later become ill with schizophrenia may be considered to reflect both
the disease process and a risk factor for later developmental disturbances required
for the clinical manifestations of the illness (Lewis and Levitt, 2002). Once pruning
is completed, normally during the second decade of life (although the temporal
trajectory could itself be altered in schizophrenia), the manifestations of the ill-
ness would be expected to be relatively stable; however, the effects of normal aging
might be more severe, perhaps accounting for the non-Alzheimer’s dementia seen
in elderly individuals with schizophrenia (Arnold et al., 1998). Consistent with this
hypothesis, altering neuronal excitability during synaptogenesis has a profound
effect on the numbers of synapses that form, but this effect is not seen if excitability
is altered after synaptogenesis (Burrone et al., 2002).
As in all studies of schizophrenia, replication of findings in additional subjects
and by other investigators is essential. Other investigators, using different cohorts
of subjects, observed some of our microarray findings. For example, Sklar and co-
workers (Sklar, 2001) reported that NSF was decreased in the prefrontal cortex of
schizophrenia subjects; an analysis of single cells from subjects with schizophrenia
219 Transcriptomes in schizophrenia

showed decreased expression levels of several synaptic markers (Hemby et al., 2002);
and altered expression of genes involved in ubiquitinization was reported by Vawter
and colleagues (2001, 2002). Such replication studies are challenging for microar-
ray techniques for a variety of reasons including a lack of standardized methods
and modes of analysis and differences in sensitivity of different platforms. Indeed,
the rapid advances in microarray technology suggest that in the short-term we
are more likely to see improvement in, rather than replication of, results (Branca,
2003). One of the biggest challenges to replication of findings from gene investi-
gators at present is the fact that different sets of gene are represented on different
microarrays. For example, using a microarray containing 1127 brain-related genes
to probe prefrontal area 9 from subjects with schizophrenia, Vawter et al. (2002)
found reduced expression of several of the PSYN (e.g. NSF, vesicular ATPase, synap-
togyrin), GABA (e.g. GAD67 ), and glutamate (e.g. AMPA2) genes identified in our
initial study. However, perhaps because the specific genes constituting these groups
in this study were different and smaller in number than those used in our study,
significant effects at the level of gene groups were not observed.
The impact that the genes represented on the arrays can have on the outcome of
a study may also be reflected in a study using a different platform, the Affymetrix
GeneChip. Investigators from the Mt Sinai group reported deficient expression of
six genes whose protein products are thought to play important roles in myelina-
tion (Hakak et al., 2001). Consistent with these observations, we found reduced
expression of proteolipid protein 1 (PLP1) across the majority of subjects with
schizophrenia (Pongrac et al., 2002). This PLP1 expression deficit did not appear
to be related to antipsychotic medication since it was not found in haloperidol-
treated monkeys. PLP1 appears to have a dual function, participating in both glial
cell development and the assembly of myelin (Schneider et al., 1992). These deficits
in myelin-related genes also raise the interesting question of the possible factors that
may account for such an alteration in the transcriptome. Although such deficits are
frequently discussed as evidence of alterations in transcription, they may be caused
by more fundamental problems. For example, the Mt Sinai group has also observed
a decreased number of glial cells in layer 3 and the underlying white matter in area
9 of subjects with schizophrenia (Hof et al., 2003), suggesting that the reduction
in myelin-related transcripts may be the consequence of a process that leads to
decreased numbers of oligodendrocytes.
Another major study, using a custom-made candidate gene array comprising 300
genes, found upregulated expression of members of the apo L lipoprotein family
(apo L1, apo L2 and apo L4, which play a central role in cholesterol transport) in
several independent cohorts of subjects with schizophrenia (Mimmack et al., 2002).
Although the function of apo L proteins in the brain is unknown, the findings
are of particular interest given that the genes for apo L proteins are located on
220 D. A. Lewis, K. Mirnics, and P. Levitt

chromosome 22q12, a susceptibility locus for schizophrenia (Waterworth et al.,

2002).

Conclusions and future directions

The application of transcriptome-based methods to studies of the molecular neu-
ropathology of schizophrenia are clearly in their infancy, and like newborns they
offer great promise for the future. However, careful rearing is essential to fulfill this
promise. Ongoing improvements in the number of genes represented on arrays,
in the sensitivity for detecting rare transcripts within samples and subtle differ-
ences between samples, and for the statistical management of data mining and
interpretation are certain to advance the field. In addition, single cells of defined
phenotypes from the diseased brain can now also be analyzed for complex gene
expression patterns in a high-throughput fashion. Using laser-capture dissection
or sharp electrode harvest of identified single cells, members of a subclass of cells
can be pooled and analyzed using high-throughput genomic techniques (Hemby
et al., 2002; Kamme et al., 2003; Luo et al., 1999). The characterization of gene
expression patterns in neuronal populations defined by a shared axonal projection
target and/or electrophysiological properties will advance our understanding of
specific neuronal phenotypes by placing them in the context of circuits and neural
networks, and thus inform how transcriptome alterations may contribute to the
disturbances in brain functions that are manifest as schizophrenia.

Acknowledgements
The interactions between our laboratories that are reflected in this chapter were
supported by NIMH Conte Center Grant MH45156 and by MH43784.

REFERENCES

Akbarian, S., Kim, J. J., Potkin, S. G., et al. (1995). Gene expression for glutamic acid decarboxylase
is reduced without loss of neurons in prefrontal cortex of schizophrenics. Arch Gen Psychiatry
52: 258–266.
Arnold, S. E., Trojanowski, J. Q., Gur, R. E., et al. (1998). Absence of neurodegeneration and
neural injury in the cerebral cortex in a sample of elderly patients with schizophrenia. Arch
Gen Psychiatry 55: 225–232.
Berman, K. F., Zec, R. F., Weinberger, D. R. (1986). Physiological dysfunction of dorsolateral
prefrontal cortex in schizophrenia. II. Role of neuroleptic treatment, attention and mental
effort. Arch Gen Psychiatry 43: 126–135.
Branca, M. (2003). Genetics and medicine: putting gene arrays to the test. Science 300: 238.
221 Transcriptomes in schizophrenia

Brown, P. O., Bottenstein, D. (1999). Exploring the new world of the genome with DNA microar-
rays. Nat Genet 21S: 33–37.
Brzustowicz, L. M., Hodgkinson, K. A., Chow, E. W. C., Honer, W. G., Bassett, A. S. (2000).
Location of a major susceptibility locus for familial schizophrenia on chromosome 1q21–
q22. Science 288: 678–682.
Buchsbaum, M. S., Haier, R. J., Potkin, S. G., et al. (1992). Frontrostriatal disorder of cerebral
metabolism in never-medicated schizophrenics. Arch Gen Psychiatry 49: 935–942.
Burrone, J., O’Byren, M., Murthy, V. N. (2002). Multiple forms of synaptic plasticity triggered
by selective suppression of activity in individual neurons. Nature 420: 414–418.
Cheung, V. G., Morley, M., Aguilar, F., et al. (1999). Making and reading microarrays. Nat Genet
21S: 20–32.
Chowdari, K. V., Mirnics, K., Semwal, P., et al. (2002). Association and linkage analyses of RGS4
polymorphisms in schizophrenia. Hum Mol Genet 11: 1373–1380.
DeRisi, J., Iyer, V. R., Brown, P. O. (1997). Exploring the metabolic and genetic control of gene
expression on a genomic scale. Science 278: 680–686.
DeRisi, J., Penland, L., Brown, P. O., et al. (1996). Use of a cDNA microarray to analyse gene
expression patterns in human cancer. Nat Genet 14: 457–460.
De Vries, L., Zheng, B., Fischer, T., Elenko, E., Farquhar, M. G. (2000). The regulator of G protein
signaling family. Annu Rev Pharmacol Toxicol 40: 235–271.
Duggan, D. J., Bittner, M., Chen, Y., Meltzer, P., Trent, J. M. (1999). Expression profiling using
cDNA microarrays. Nat Genet 21S: 10–14.
Eastwood, S. L., McDonald, B., Burnet, P. W. J., et al. (1995). Decreased expression of mRNAs
encoding non-NMDA glutamate receptors GluR1 and GluR2 in medial temporal lobe neu-
rons in schizophrenia. Mol Brain Res 29: 211–223.
Glantz, L. A., Lewis, D. A. (1997). Reduction of synaptophysin immunoreactivity in the prefrontal
cortex of subjects with schizophrenia: regional and diagnostic specificity. Arch Gen Psychiatry
54: 943–952.
(2000). Decreased dendritic spine density on prefrontal cortical pyramidal neurons in
schizophrenia. Arch Gen Psychiatry 57: 65–73.
Hakak, Y., Walker, J. R., Li, C., et al. (2001). Genome-wide expression analysis reveals dysregulation
of myelination-related genes in chronic schizophrenia. Proc Natl Acad Sci USA 98: 4746–
4751.
Harrison, P. J., Heath, P. R., Eastwood, S. L., et al. (1995). The relative importance of premortem
acidosis and postmortem interval for human brain gene expression studies: selective mRNA
vulnerability and comparison with their encoded proteins. Neurosci Lett 200: 151–154.
Hashimoto, T., Volk, D. W., Buchheit, S. E., Lewis, D. A. (2002). Expression of BDNF and trkB
mRNAs in prefrontal cortex of subjects with schizophrenia. Soc Neurosci Abstr 28: 703.7.
Hashimoto, T., Volk, D. W., Eggan, S. M., et al. (2003). Gene expression deficits in a subclass
of GABA neurons in the prefrontal cortex of subjects with schizophrenia. J Neurosci 23:
6315–6326.
Hemby, S. E., Ginsberg, S. D., Brunk, B., et al. (2002). Gene expression profile for schizophrenia:
discrete neuron transcription patterns in the entorhinal cortex. Arch Gen Psychiatry 59:
631–640.
222 D. A. Lewis, K. Mirnics, and P. Levitt

Hof, P. R., Haroutunian, V., Friedrich, Jr., V. L., et al. (2003). Loss and altered spatial distribution
of oligodendrocytes in the superior frontal gyrus in schizophrenia. Biol Psychiatry 53: 1075–
1085.
Huttenlocher, P. R. (1979). Synaptic density in human frontal cortex: developmental changes and
effects of aging. Brain Res 163: 195–205.
Kamme, F., Salunga, R., Yu, J., et al. (2003). Single-cell microarray analysis in hippocampus CA1:
demonstration and validation of cellular heterogeneity. J Neurosci 23: 3607–3615.
Lewis, D. A. (2002). The human brain revisited: opportunities and challenges in postmortem
studies of psychiatric disorders. Neuropsychopharmacology 26: 143–154.
Lewis, D. A., Levitt, P. (2002). Schizophrenia as a disorder of neurodevelopment. Annu Rev
Neurosci 25: 409–432.
Lewis, D. A., Lieberman, J. A. (2000). Catching up on schizophrenia: natural history and neuro-
biology. Neuron 28: 325–334.
Lidow, M. S., Goldman-Rakic, P. S., Rakic, P., Innis, R. B. (1989). Dopamine D(2) receptors in the
cerebral cortex: distribution and pharmacological characterization with [(3)H]raclopride.
Proc Natl Acad Sci USA 86: 6412–6416.
Lipshutz, R. J., Fodor, S. P. A., Gingeras, T. R., Lockhart, D. J. (1999). High density synthetic
oligonucleotide arrays. Nat Genet 21S: 20–24.
Lockhart, D. J., Barlow, C. (2001). DNA arrays and gene expression analysis in the brain. In
Methods in Genomic Neuroscience, ed. H. R. Chin, S. Moldin. New York: CRC Press, pp.
143–170.
Lockhart, D. J., Dong, H., Byrne, M. C., et al. (1996). Expression monitoring by hybridization to
high-density oligonucleotide arrays. Nat Biotechnol 14: 1675–1680.
Luo, L., Salunga, R. C., Guo, H., et al. (1999). Gene expression profiles of laser-captured adja-
cent neuronal subtypes. Nat Med 5: 117–122.
Middleton, F. A., Mirnics, K., Pierri, J. N., Lewis, D. A., Levitt, P. (2002). Gene expression profiling
reveals alterations of specific metabolic pathways in schizophrenia. J Neurosci 22: 2718–2729.
Mimmack, M. L., Ryan, M., Baba, H., et al. (2002). Gene expression analysis in schizophrenia:
reproducible up-regulation of several members of the apolipoprotein L family located in a
high-susceptibility locus for schizophrenia on chromosome 22. Proc Natl Acad Sci USA 99:
4680–4685.
Mirnics, K., Middleton, F. A., Marquez, A., Lewis, D. A., Levitt, P. (2000). Molecular characteriza-
tion of schizophrenia viewed by microarray analysis of gene expression in prefrontal cortex.
Neuron 28: 53–67.
Mirnics, K., Middleton, F. A., Stanwood, G. D., Lewis, D. A., Levitt, P. (2001a). Disease-specific
changes in regulator of G-protein signaling 4 (RGS4) expression in schizophrenia. Mol
Psychiatry 6: 293–301.
Mirnics, K., Middleton, F. A., Lewis, D. A., Levitt, P. (2001b). Analysis of complex brain disorders
with gene expression microarrays: schizophrenia as a disease of the synapse. Trends Neurosci
24: 479–486.
Pongrac, J., Middleton, F. A., Lewis, D. A., Levitt, P., Mirnics, K. (2002). Gene expression profiling
with DNA microarrays: advancing our understanding of psychiatric disorders. Neurochem
Res 27: 1049–1063.
223 Transcriptomes in schizophrenia

Rakic, P., Bourgeois, J. P., Eckenhoff, M. F., Zecevic, N., Goldman-Rakic, P. S. (1986). Concurrent
overproduction of synapses in diverse regions of the primate cerebral cortex. Science 232:
232–235.
Schena, M., Davis, R. W. (1999). Microgenes, genomes and chips. In DNA Microarrays: A Practical
Approach, ed. M. Schena. Oxford: Oxford University Press, pp. 1–16.
Schena, M., Shalon, D., Davis, R. W., Brown, P. O. (1995). Quantitative monitoring of gene
expression patterns with a complementary DNA microarray. Science 270: 467–470.
Schena, M., Shalon, D., Heller, R., et al. (1996). Parallel human genome analysis: microarray-based
expression monitoring of 1000 genes. Proc Natl Acad Sci USA 93: 10614–10619.
Schneider, A., Montague, P., Griffiths, I., et al. (1992). Uncoupling of hypomyelination and glial
cell death by a mutation in the proteolipid protein gene. Nature 358: 758–761.
Sklar, P. (2001). Microarray analysis of the Stanley brain collection. Abstract. In 7th Symposium on
the Neurobiology and Neuroimmunology of Schizophrenia and Bipolar Disorder. Washington,
DC: Stanley Foundation.
Southern, E., Mir, K., Shchepinov, M. (1999). Molecular interactions on microarrays. Nat Genet
21: 5–9.
Vawter, M. P., Barrett, T., Cheadle, C., et al. (2001). Application of cDNA microarrays to examine
gene expression differences in schizophrenia. Res Bull 55: 641–650.
Vawter, M. P., Crook, J. M., Hyde, T. M., et al. (2002). Microarray analysis of gene expression in
the prefrontal cortex in schizophrenia: a preliminary study. Schizophr Res 58: 11–20.
Volk, D. W., Austin, M. C., Pierri, J. N., Sampson, A. R., Lewis, D. A. (2000). Decreased
GAD67 mRNA expression in a subset of prefrontal cortical GABA neurons in subjects with
schizophrenia. Arch Gen Psychiatry 57: 237–245.
Waterworth, D. M., Bassett, A. S., Brzustowicz, L. M. (2002). Recent advances in the genetics of
schizophrenia. Cell Mol Life Sci 59: 331–348.
Weickert, C. S., Hyde, T. M., Lipska, B. K., et al. (2003). Reduced brain-derived neurotrophic
factor in prefrontal cortex of patients with schizophrenia. Mol Psychiatry 8: 592–610.
 13

Is there a role for social factors in

a comprehensive developmental model
for schizophrenia?
Jane Boydell1 , Jim van Os2 , and Robin M. Murray1
1
Institute of Psychiatry King’s College London, UK
2
Maastricht University, Maastricht, the Netherlands

In the 1950s and 1960s, there was much extravagant discussion of the role of social
factors in the etiology of schizophrenia. However, there was little scientific basis
to this speculation, and it was swept away by the demonstration that people with
schizophrenia showed abnormalities of brain structure on computed tomographic
scans (Johnstone et al., 1976). A decade later, the neurodevelopmental model of
schizophrenia was proposed, and it subsequently became the dominant etiological
and pathogenetic model (Murray and Fearon, 1999; Murray et al., 1992). As a
result of these two developments, researchers have come to regard schizophrenia as
a brain disease, and social factors have been largely ignored as putative etiological
agents.
It is increasingly clear, however, that the neurodevelopemental model, an essen-
tially neurological concept, does not explain all the available data about schizophrea-
nia. One consequence has been a revival during the 1990s, particularly in Europe, of
research into the role of social factors as causal agents in schizophrenia. It is oppor-
tune, therefore, to draw together this disparate research and to examine critically
whether any of it stands up to scientific scrutiny. We will review both those social
factors that are postulated to operate early in life and those which may act more
proximal to the onset of the disorder (see Table 13.1 for an overview). First we will
briefly consider how animal research has informed on psychiatric conditions.

Animal models of isolation rearing and social stress

Brain development is to an important degree dependent on experience (Wiesel
and Hubel, 1974); conversely, altered brain development can influence the way
individuals interact with their environment (Fishbein, 2000). Some of the strongest

Neurodevelopment and Schizophrenia, ed. Matcheri S. Keshavan et al. Published by Cambridge University
Press.

C Cambridge University Press 2004.

224
225 Social factors and development of schizophrenia

Table 13.1. Selected social risk factors for schizophrenia

Risk factor Measure Effect size Outcome Type of study Author

of effect (95% CI)

Unwanted child OR 2.4 (1.2–4.8) DSM III-R Birth cohort Myhrman et al.
schizophrenia (1996)
Poor mothering OR 2.65 (1.2–5.6) Schizophreniform Birth cohort Cannon et al. (2002)
disorders DSM IV
Early parental loss OR 3.8∗ DSM III-R Case–control Agid et al. (1999)
schizophrenia
Childhood abuse OR 7.3 (1.1–49) Positive psychotic Population Janssen et al. (2003)
symptoms cohort
reaching clinical
significance
City birth RR 2.4 (2.13–2.7) ICD 8 schizophrenia Population Mortensen et al.
cohort (1999)
City upbringing RR 2.75 (2.31–3.28) ICD 8 schizophrenia Population Pedersen and
cohort Mortensen (2001)
Ethnic minority IRR 4.4 (2.49–7.75) RDC Schizophrenia Ecological Boydell et al. (2001)
status
Racial harassment OR 2.86 (1.69–4.83) Psychosis diagnosed Cross-sectional Karlsen and Nazroo
(verbal abuse) by semistructured (2002)
interview

CI, confidence interval; OR, odds ratio; RR, relative risk; IRR, incidence rate ratio; DSM III-R, Diagnostic and
Statistical Manual III–Revised; ICD, International Statistical Classification of Diseases, Injuries, and Causes of
Death.
∗
p = 0.001.

evidence for the role of social factors in brain development has come from animal
experiments. For example, there is ample evidence that rats reared in isolation for a
period after birth have structural and physiological differences in their hippocampi
compared with rats reared in groups (Whitaker-Azmitia et al., 2000). Isolation
reared rats also show behavioral changes, anxiety, learning deficits, and sensory
changes. Furthermore, Matthews et al. (2001) found increased levels of dopamine
in the dorsal and ventral striatum in adult rats who were separated from their
mothers early in life. There is also evidence that rats reared in isolation, but not
other rats, respond to antipsychotic drugs in such a way as to normalize their
behavior (Heidbreder et al., 2001).
Therefore, by analogy, these data suggest that the social factors in the envi-
ronment of the developing child may similarly affect physical brain development.
These factors are now being revisited by psychiatric researchers searching for the
brain–environment interactions that shape vulnerability to conditions such as
226 J. Boydell, J. van Os, and R. M. Murray

psychosis. For example, experiences embedded in social relationships may alter the
prefrontal brain systems that mediate emotional self-regulation (Lyons et al., 2002);
similarly, the early social environment appears to program aspects of neurobiolog-
ical development that, in turn, affect behavioral, emotional, cognitive, and physio-
logical development (Sanchez et al., 2001). The early social environment has been
shown to induce synaptic changes that may be indicative, and perhaps the cause,
of alterations of behavioral and cognitive capacities (Ovtscharoff and Braun, 2001).
Although much of this work comes from animal research, there is evidence that
the early social environment can mediate the establishment of neural networks
that regulate a child’s response to stress and capacity for self-control (DiPietro,
2000). Furthermore, studies of exposure to institutional deprivation have shown
dramatic effects on a range of child behavioral outcomes, with clear dose–response
relationships between adverse outcomes and length of exposure (Beckett et al.,
2002). As Eisenberg (1995) states “The human brain is constructed socially.”

Family factors
Mother–child relationship
The British 1946 cohort study followed up all children (5362) born in 1 week in
the UK. By age 43, there were 30 cases of schizophrenia. The quality of the mother-
child relationship at 4 years of age, as rated by health visitors, was one of the most
powerful risk factors for later schizophrenia; a poor relationship (less skilled and
less understanding of the child) carried a sixfold increase in risk of schizophrenia
(Jones et al., 1994). Similarly, the Dunedin study (Cannon et al., 2002), which
followed up about 1000 children from birth to age 26 with comprehensive medical
and psychiatric assessments, found that the mothers of children who developed
schizophreniform disorders were rated as having poorer attitudes and behavior
towards their children at age 3 (odds ratio [OR], 2.65; 95% confidence interval
[CI], 1.2–5.6).
The British 1946 and the Dunedin studies are particularly interesting as they are
population-based and not high-risk studies. However, we do not know the direction
of the relationship. Is the increased risk an effect of poor mothering or is it that the
preschizophrenic child could not form a normal relationship with the mother?

Unwantedness
Several studies have claimed that being an unwanted child is associated with emo-
tional and social deprivation (Kubicka et al., 1995; Myhrman, 1992). Myhrman
et al. (1996), therefore, investigated whether unwantedness of pregnancy was a
risk factor for schizophrenia in the child as defined in the Diagnostic and Statistical
227 Social factors and development of schizophrenia

Manual III (DSM III-R; American Psychiatric Association, 1987). Their sample was
the Northern Finland 1966 Birth Cohort, which comprised 11 017 individuals who
were followed up until they were 28. Their mothers had been asked by a midwife
(when 6 or 7 months pregnant) whether the pregnancy was wanted, mistimed but
wanted, or unwanted. The risk of schizophrenia among the unwanted group was
considerably raised even after adjusting for sociodemographic, pregnancy (includ-
ing depression), and perinatal variables (OR, 2.4; 95% CI, 1.2–4.8). Unfortunately,
genetic risk was not considered in the study and, as with much social research, there
may have been residual confounding.

Family communication deviance

In the 1950s, several research groups (Bateson et al., 1956; Lidz, 1958; Wynne et al.,
1958) claimed that communication deviance in parents of people with schizophre-
nia had a causal role in the development of the disorder. However, Hirsch and
Leff (1975), who comprehensively reviewed this work, were not convinced by the
methodology used in much of it. Furthermore, most authorities have considered
that, if such abnormalities do exist, they reflect an expression of genetic loading
rather than independent causal factors. One recent study, however, has suggested
that family communication deviance may be of etiological importance. Wahlberg
et al. (1997, 2000) found a significant interaction between high genetic risk for
schizophrenia and communication deviance in their adoptive parents. The high
risk but not the control adoptees showed greater evidence of thought disorder if
their adoptive parents showed communication deviance. This provocative claim
awaits replication, and the reader could be forgiven for being skeptical given the
previous history of research on this topic.

Dysfunctional family environment

In the Finnish Adoptive Family Study, the risk of developing schizophrenia spectrum
disorders was higher in those adopted-away offspring of schizophrenia parents who
were exposed to a dysfunctional adoptive family-rearing environment (Tienari,
1991). The most recent report from this study suggests that the adopted-away
offspring may have lower risk than children who remain with their schizophrenia
parents (Tienari et al., 2000). Similarly, the Danish–American adoption studies
suggested that the high-risk adopted-away offspring had a lower chance of later
developing schizophrenia than those who remained with their biological parents
(Rosenthal et al., 1971).

Communal upbringing
In the Israeli High Risk Study, Mirsky et al. (1985) carried out a follow-up study of
46 children with high genetic risk, half of whom were brought up communally
228 J. Boydell, J. van Os, and R. M. Murray

on a kibbutz (where their parents lived) and half in family homes, and com-
pared them with controls. Children with known genetic risk for schizophrenia
were significantly more likely to develop a psychotic disorder and, interestingly,
also an affective disorder if they were brought up on a kibbutz, rather than a
family home. The control kibbutz children did not have a higher risk (although
the numbers were small), suggesting genetic vulnerability to the social environ-
ment in the high-risk children. Similar findings emerged from the Copenhagen
High Risk Project, which followed up 207 high-risk children and compared them
with controls (Mednick et al., 1987). The high-risk children who had been sep-
arated from their parents were either brought up in public care institutions or
foster placements with families. The latter group had a better outcome (especially
for males) but of course the placement was not random. Nevertheless, these two
high-risk studies are among the few studies into social factors during childhood
that are less likely to have been confounded by the inherited characteristics of the
child.

Early parental loss

Another aspect of childhood that has been examined is parental loss through
death or separation. Agid et al. (1999), in a case–control study, found that per-
manent parental loss before the age of 8 years was significantly associated with later
schizophrenia (OR, 4.3), particular for loss of the mother (OR, 6). The association
was stronger for parental death than separation, making confounding less likely
as an explanation for the findings. The DSM III-R diagnoses were based on struc-
tured interviews, making misdiagnosis less likely. Similarly, Mallett et al. (2002)
found that separation from both parents in childhood distinguished UK African–
Caribbean subjects with schizophrenia from African–Caribbean controls (OR, 5;
95% CI, 1.09 22.82).

Expressed emotion
It is well known that social overstimulation (Wing, 1978) and high expressed emo-
tion within families (Vaughn and Leff, 1976, 1982) are associated with relapse in
schizophrenia. Some assume that these factors are also implicated in the first episode
(Wing, 1978). As far as we are aware, there is no evidence to support or refute
this.

Childhood abuse
Several investigators have claimed an association between trauma and psychosis
(Read et al., 2001). First, studies of patients with psychosis have demonstrated that
they recall a high incidence of trauma in their lifetimes (Ross et al., 1994), and those
229 Social factors and development of schizophrenia

who experience positive symptoms of schizophrenia are especially likely to give a

history of childhood trauma (Ellason and Ross, 1997; Ross et al., 1994). Conversely,
patients who give a history of abuse are particularly likely to experience positive
symptoms (Heins et al., 1990; Sansonnet-Hayden et al., 1987). In a recent study,
childhood abuse was a significant predictor of hallucinations, even in the absence
of adult abuse (Read et al., 2003).
Second, in patients with other diagnoses, such as bipolar disorder (Hammers-
ley et al., 2003; Read and Argyle, 1999) or post-traumatic shock disorder (Butler
et al., 1996), a history of child abuse has also been found to “cooccur” with a high
frequency of auditory hallucinations and delusions. Dissociative identity disor-
der, which is assumed to be a disturbance resulting from severe childhood abuse
(Putnam et al., 1986), may present with Schneiderian first rank symptoms, par-
ticularly in the form of auditory hallucinations (Ross et al., 1989). It has even
been suggested that, of all diagnostic categories, psychosis displays the strongest
associations with child abuse (Bryer et al., 1987; Swett et al., 1990).
The experience of abuse may create a biological (Read et al., 2001) or psycholog-
ical (Garety et al., 2001) vulnerability for the development of psychotic symptoms,
including subclinical psychotic experiences such as low-grade delusional ideation
and isolated auditory hallucinations (Johns and van Os, 2001). In the general popu-
lation, childhood sexual abuse is related to schizotypy, including perceptual aber-
rations (Startup, 1999), which are 10 times more common in adults who were
maltreated as children (Berenbaum, 1999). In both clinical and non-clinical popu-
lations, the diagnostic group with the highest rate of childhood abuse consistently
reported the most Schneiderian symptoms (Ross and Joshi, 1992). In a population-
based cross-sectional study, Bebbington (2003) found increased odds of psychosis
(diagnosed by semistructured interview) in those reporting childhood sexual abuse
and other forms of trauma. Therefore, several studies have now found evidence for
an association between abuse and psychotic experiences in non-clinical samples
(Ross and Joshi, 1992; Startup, 1999).
The difficulty with almost all of the above studies is that they are cross-sectional
in nature and rely on psychotic people giving an accurate picture of their child-
hood. However, there have been attempts to go beyond such bias-prone studies.
For example, in a recent 3 year longitudinal study of a general population sample
of 4045 subjects aged 18–64 years with no previous lifetime presence of psychotic
or psychosis-like symptoms, baseline reported childhood abuse predicted devel-
opment of clinically relevant positive psychotic symptoms associated with need
for care (OR, 11.5; 95% CI, 2.6–51.6). This association remained after adjustment
for demographic variables, a range of other reported risk factors, and presence of
any lifetime psychiatric diagnosis at baseline (OR, 7.3; 95% CI, 1.1–49.0) (Janssen
et al., 2004).
230 J. Boydell, J. van Os, and R. M. Murray

The urban effect

City birth
Many studies have found an increased incidence of schizophrenia in cities (e.g. Farris
and Dunham, 1939). For decades it was suggested that this could be because people
at higher risk or in the early stages of schizophrenia moved into the city (social
drift). Now several studies have shown a significant effect for urban birth and/or
urban upbringing (Allardyce et al., 2001; Lewis et al., 1992; Marcelis et al., 1998,
1999). One of the most impressive was carried out by Mortensen et al. (1999),
who investigated the effect of place on risk of admission with schizophrenia, as
defined by the International Statistical Classification of Diseases, Injuries and Causes
of Death (ICD) 8 (World Health Organization, 1967), in a large Danish population-
based cohort of 1.75 million. The relative risk associated with birth in Copenhagen
compared with birth in rural areas was 2.40 (95% CI, 2.13–2.7), and there was a
clear dose–response relationship for urbanicity in that the larger the town of birth,
the greater the risk. A family history of schizophrenia did not explain or affect the
results. Indeed, Mortensen and colleagues (1999) calculated that the population
attributable fraction for urban birth was 34.6% (which compared with 9% and 7%,
respectively, for having a mother or father who had schizophrenia). Therefore, the
effect of urban birth was much larger than the effect of having a parent affected
(though we should point out that having an affected parent does not equate with
genetic predisposition).

City upbringing
It is important but very difficult to separate place of birth and place of upbringing.
Astrup and Odegard (1961) found a stronger effect of city upbringing amongst
those who had moved to the city. A recent analysis of the Danish population cohort
just described has now addressed this question. Pedersen and Mortensen (2001)
used a comprehensive national registration system that accurately recorded every
change of residence to show that schizophrenia risk increased with the number of
years (between 0 and 20 years of age) that an individual lived in an urbanized area,
and with increasing degree of urbanization. Relative risk for those who had spent
their entire childhood in the capital was 2.75 (95% CI, 2.31–3.28) compared with
those who had always lived in the most rural areas. There was no evidence that
urbanicity was particularly toxic for any particular age group. The dose–response
relationship found in this study suggests that etiological factors of a pervasive and
long-term (either continuous or repeated) nature are operating in urbanized areas.

Social drift and social residue theories

When considering the effect of urbanization, we cannot totally discount the effects
of “social drift”: people with pre- or prodromal schizophrenia move into urban
231 Social factors and development of schizophrenia

areas. However, this effect cannot account for the results of recent studies that
looked at place of birth and upbringing. For example, Dauncey and colleagues
(1993) investigated where patients had lived 5 years before their first admission
in Nottingham, UK and were unable to find evidence for systematic geographic
drift. Furthermore, it is unlikely that the drift occurred in the previous parental
generation, as the magnitude of this movement would need to have been extremely
high to explain the findings. For example, in the Danish study, Mortensen and
colleagues (1999) calculated that nearly 50 000 children born in the capital and
its suburbs needed to have a parent who transmitted a genetic risk equal to that
transmitted by a parent with diagnosed schizophrenia to account for the urban
excess. Furthermore, a family history of schizophrenia did not explain or effect the
urban–rural difference in this and other studies (van Os et al., 2002). This is also
relevant for the social residue theory (that those at greater risk are left behind in
an area as it becomes less desirable because they do not have the resources to move
out).

Possible explanations
No adequate explanation has been found for the urban excess of schizophrenia.
Infectious agents were once considered likely etiological factors but they are unlikely
to account for much of the risk-increasing effect, given the fact that the urban envi-
ronment is not only associated with increased rates of schizophrenia but also with
increases in much more prevalent expressions of non-clinical psychotic experi-
ences (van Os et al., 2001). Other physical factors associated with city living such
as lead pollution require further investigation. However, while there is no direct
evidence that the urban excess is caused by social factors, the fact that its effect is
so widespread across the population is compatible with the notion that part of the
excess risk represents a psychological reaction to factors in the wider social environ-
ment. Certainly some of the major differences between urban and rural areas are
to do with social cohesion and social support. For example, it has been shown that
the physical and mental health of children growing up in an urban environment is
adversely influenced by low levels of social cohesion in the neighborhood (Drukker
et al., 2003), and within the city, the incidence of schizophrenia has been shown to
vary as a function of the social neighborhood environment (van Os et al., 2000).
Although the evidence for an early effect is strong, there may also be an effect
of urbanization around the time of onset for non-early-onset schizophrenia and
there is evidence of a cumulative effect of urban exposure throughout childhood
(Castle et al., 1993). As we will see below, social isolation, deprivation, and adverse
life events are associated with increased risk of psychosis. As individuals in big cities
may be more likely to be exposed to these factors (Takano and Nakamura, 2001), it is
possible that some of the urban excess might be attributed to such factors. However,
it is likely that the social factors that increase the risk do not act independently of
232 J. Boydell, J. van Os, and R. M. Murray

genetic risk, and a recent report suggested that the urban environment acts by
increasing the likelihood that schizophrenia is expressed in those who carry genetic
susceptibility (van Os et al., 2003).

Social isolation
During childhood
Numerous reports from Bleuler onwards have shown that children destined to
develop schizophrenia tend to be solitary, lack friends, and indeed often prefer to
be alone. For example, the British 1946 cohort study of Jones et al. (1994), described
above, found that preference for solitary play at age 4 and 6 was associated with
later schizophrenia (OR, 2.1 and 2.5, respectively). Self-reported anxiety at age 13
and teacher-rated anxiety at age 15 both showed linear associations with later risk
for schizophrenia (Jones et al., 1994).

Moving schools in adolescence

A fascinating finding to emerge from the Danish study of Pedersen and Mortensen
(2001), discussed above, was that change in municipality (and, therefore, school)
increased risk of schizophrenia whereas change of address during childhood within
the municipality (usually, therefore, without changing school) was not associated
with increased risk. Moves during early teenage years appeared to have the greatest
effect, and the more the moves, the greater was the risk. It is possible that the social
stress of having to seek out and join a new peer group, or perhaps not being able to
join a peer group, adversely affected the childrens’ social development.

In young adult life

The Swedish conscript study also looked at the role of premorbid personality at a
later stage of development. This study examined a cohort of 50 087 army conscripts
(approximately 98% of the entire male population born in Sweden 1949 to 1950).
The investigators adjusted for family history of psychiatric illness, intelligence quo-
tient (IQ), city upbringing and diagnosis of non-psychotic disorder at conscription.
There was a significantly increased risk of later developing schizophrenia in young
men who felt they were more sensitive than their peers, had fewer than two close
friends, preferred small groups, and did not have a girl friend (OR, 2.38, 1.7, 2.05,
and 3.48, respectively); there was also a risk, to a lesser extent, for other psychoses
(Malmberg et al., 1998). Once again, this raises the question of whether these are an
expression of a schizoid or schizotypal personality or whether they are in themselves
independent risk factors. Until proven otherwise, it is wise to consider that both
may be true: individuals with a schizoid or schizotypal personality may be less able
233 Social factors and development of schizophrenia

to make social relationships, and then the social isolation itself may propel them
further toward frank psychosis.

At time of onset
Hare (1956) reported that social isolation, as measured by proportion of single
person households in a geographical area, was associated with increased rates of
schizophrenia. The findings were not accounted for by movement of people with,
or developing, schizophrenia into the area. There has been a recent resurgence of
interest in these findings. Thornicroft et al. (1993) noted that clustering of individ-
uals with schizophrenia in deprived areas occurs only in urban areas and suggested
that social isolation is an important mediator of this. However, it is difficult to
distinguish between cause and effect in this context. Related to these ideas is the
theory that disruption of social networks decreases an individual’s capacity to cope
with psychosocial stress and increases the risk of schizophrenia.
Van Os et al. (2000) found that people who were single had a slightly higher
risk of developing psychosis if they lived in a neighborhood with fewer single
people compared with a neighborhood with many other single people. The authors
suggested that single status might give rise to perceived (or actual) social isolation
if most other people are living with a partner. The question of whether social
isolation may increase risk of schizophrenia (or whether a close relationship may
be protective) is also raised by Jablensky and Cole (1997), who showed that marriage
had a protective effect and that this was not simply a consequence of better adjusted
males being able to marry.

Migration and ethnic minority status

As early as 1932, Ødegård (1932) showed that Norwegian migrants to the USA had
an increased risk of psychosis. One early interpretation was that this resulted from
selective migration; those at risk, less connected with their families, are more likely
to leave their homeland. An alternative was that the crucial factor was the alien-
ation and suspicion engendered in the migrant by their unfamiliar surroundings.
Ødegård (1932) wrote:

Everywhere you are surrounded by people with strange and unfamiliar ways. . . . They do not
seem to be as friendly as the people at home and many of them do their best to profit by your lack
of experience. Even if you have not had any disagreeable experiences yourself, your imagination
is stirred by all the stories you have heard about how crooked and dangerous they may be.
You notice that your own appearance, clothing and language points you out to everyone as a
greenhorn.

Studies from different countries and continents have now shown that the inci-
dence of psychosis is high amongst many migrant groups (Castle et al., 1991;
234 J. Boydell, J. van Os, and R. M. Murray

Harrison et al., 1988, 1997; King et al., 1994; Murphy, 1977; Selten and Sijben,
1994; Selten et al., 1998, 2001; Thomas et al., 1993; van Os et al., 1996a,b). Even
when migration between very similar cultures is considered, there is still evidence of
an increased risk. For example, Bruxner et al. (1997) found higher rates of hospital
admission with psychosis in British, Irish and southern European migrants to West.
Australia; the risk persisted many years after migration.

Methodological issues
There are many methodological pitfalls in migration and minority group research,
for example the “category fallacy” (that the categories of mental illness used in
one culture cannot be applied to another), misdiagnosis (whether psychiatrists in
the host country may wrongly diagnose schizophrenia in migrants because of preju-
dice or a lack of understanding of their cultural norms), or inaccurate estimation of
the denominator (migrants may be under-represented in general population data).
However, studies that have overcome these problems have still found an excess of
psychosis among migrant groups (Hickling et al., 1999; Mortensen et al., 1997;
Sharpley et al., 2001; van Os et al., 1996a,b; Wessely et al., 1991).
Increased rates of schizophrenia amongst migrants and ethnic minorities might
also be attributed (at least in part) to the urban effect, as most migrants live in cities
and often in the most deprived areas. However, Malzberg (1969) found that con-
trolling for urbanicity reduced but did not explain the excess of psychosis amongst
migrants to the USA, and similar findings were reported more recently in the
Netherlands (van Os et al., 2001)

African-Caribbeans in the UK
Most work has investigated the increased rates of psychosis in the African-
Caribbeans in the UK and the Netherlands. Genetic predisposition cannot be the
sole explanation since the increased risk is not shared by the population of origin
in the Caribbean (Bhugra et al., 1997; Mahy et al., 1999); in addition the morbid
risk for second-generation siblings is much higher than for their first-generation
equivalents (Hutchinson et al., 1996; Sugarman and Crawford, 1994). Selective
migration has also been largely ruled out. For example, Selten et al. (2002) found
that, even if the entire Surinamese population had migrated to the Netherlands,
this still would not account for the absolute number of cases amongst Surinam
migrants unless the incidence had increased. Increased exposure, or susceptibil-
ity, to neurodevelopmental insult such as obstetric complications (Glover, 1989;
Hutchinson et al., 1997) and viral infections (Selten et al., 1998) have largely been
excluded as possible explanations. McGuire et al. (1995) found no difference in drug
abuse between their white and African- Caribbean patients in South London. Simi-
larly, Selten et al. (1997) found that consumption of cannabis was lower amongst
235 Social factors and development of schizophrenia

Caribbean migrants to the Netherlands than amongst the native population, but
their incidence rate of schizophrenia was higher.

The second generation

There are now several studies showing higher rates of schizophrenia in the children
of migrants. This phenomenon has been described when the migration took place
in very different circumstances. Children born in Greenland to Danish mothers, in
the study described above, had a relative risk of 3.71 for schizophrenia (Mortensen
et al., 1999). Malzberg (1969) found that English-born migrants to the USA had
lower rates than the native USA-born individuals but the second generation (i.e.
children of English-born migrants) had higher rates than both the native born and
their own parents. Some studies suggest that second-generation (i.e. children of)
migrants from the Caribbean to the UK seem to have higher rates than the first
generation (Harrison et al., 1988).
A natural experiment occurred when the entire population of Yemenite Jews
migrated to Israel as they were in danger in the Yemen and they believed the
Messianic Era had begun. The adults did not integrate into Israeli society and
their lifestyle was considered primitive. Weingarten and Orren, 1983 found a high
prevalence of schizophrenia among their offspring. Children who had been born in
the Yemen and those who were born in Israel both seem to have been at higher risk.
Since the entire population moved to Israel, selective migration cannot account for
the findings. This study was small and there might have been an element of social
residue (i.e. the more able might have moved out of the area studied). There is no
satisfactory explanation as to why there are higher rates of psychosis in children of
migrants, but the range of countries and circumstances in which this phenomenon
has been described is suggestive of a socially induced phenomenon.

Discrimination
The combination of more affective symptoms and a less-deteriorated course has
led to the idea that factors in the social environment might cause psychosis in those
migrants predisposed but who might not otherwise develop the disease (McKenzie
et al., 1995). Racism (overt and institutionalized), social isolation, and reduced
social networks are among the factors that some consider may contribute (Hutchin-
son et al., 1999). Boydell et al. (2001) have found that incidence rates of schizophre-
nia increased in ethnic minorities as the proportion of ethnic minorities in the
locality fell, suggesting social experience (perhaps of isolation or discrimination)
contributes to the development of the disorder.
Janssen et al. (2003) measured subjective experience of discrimination and sub-
sequent development of psychotic illness 3 years later. Experience of discrimination
strongly predicted for the development of delusional ideation (OR, 2.3; 95% CI,
236 J. Boydell, J. van Os, and R. M. Murray

1.2–4.2). In another analysis in the same sample, it was shown that the effect of
ethnic minority status on psychosis was no longer significant when controlling for
experience of discrimination (Janssen et al., 2002). Karlsen and Nazroo (2002) stud-
ied the experience of racial harassment and perception of discrimination among
a UK representative sample of 5000 people from ethnic minorities. They found
significantly increased OR values for a number of health problems but particularly
psychosis (experience of verbal abuse OR, 2.86 [95% CI, 1.69–4.83] experience of
physical attack OR, 4.77 [95% CI, 2.32–9.8]).
Some people may be more susceptible to the deleterious effects of such events
than others because they are less able to derive support from those around them
(Sharpley et al., 2001). Mallett et al. (1998) demonstrated that one of the main
distinguishing features of first-onset patients of Caribbean origin with psychosis in
London was that they lived alone and additionally had been separated from their
mother at an early age.

Unemployment
Bhugra et al. (1997) found that unemployment was particularly high amongst peo-
ple of Caribbean origin first presenting with schizophrenia in the UK. They then
went on to compare the unemployment rates in African-Carribean people present-
ing with their first episode in London and Trinidad, relative to local unemployment
rates. In London, the rate was much higher amongst the patients than in the local
population, but this was not the case in Trinidad (Bhugra et al., 2000). It is pos-
sible, therefore, that the high unemployment rate amongst African-Caribbeans in
London (and subsequent social isolation and stigmatization) might have etiological
relevance. Alternatively, the social milieu in Trinidad might be more sympathetic
to those at risk of developing schizophrenia.

Life events
The first study to find an association between schizophrenia and life events reported
an excess of events limited to 3 weeks before onset (Brown and Birley, 1968).
Prospective studies have also found an association between life events and relapse
of psychosis (Malla et al., 1990; Ventura et al., 1989). Bebbington (2000) carried out
a comprehensive case–control study in London and found a significant excess of
events and, importantly, independent life events in the 3 months prior to onset of
schizophrenia. Not all studies have concurred however (Chung et al., 1986; Jacobs
and Myers, 1976).
There is evidence that the kinds of stress that affect individuals with schizophrenia
vulnerability are not so much the major life events but rather the hassles and stress
that individuals experience in the flow of daily life (Malla et al., 1990). Studies using
237 Social factors and development of schizophrenia

intensive field methods to examine subjective experience of stress in the flow of daily
life have found that individuals with schizophrenia, and individuals with genetic
susceptibility to schizophrenia, display greater levels of emotional reactivity to small
daily life stressors than do control subjects (Myin-Germeys et al., 2001). Individuals
with schizophrenia may be more sensitive to such stressors than individuals with
affective disorder (Myin-Germeys et al., 2003a). In addition, part of the sensitivity
to daily life stressors may be determined by prior exposure to major life events
(Myin-Germeys et al., 2003b), suggesting synergistic environment–environment
interactions.
Other research suggests that some individuals carry traits that make them more
likely to experience life events, such as the personality trait neuroticism (van Os and
Jones, 1999). This finding appears to be relevant to schizophrenia, as several studies
have demonstrated that neuroticism is a risk factor for schizophrenia (Krabbendam
et al., 2002; van Os and Jones, 2001) and is more prevalent in the first-degree relatives
of patients with schizophrenia (Maier et al., 1994). Therefore, the role of life events
and daily life hassles as a risk factor for schizophrenia is complex. Stress may be
generated in part by underlying personality traits, whose genetic contribution may
overlap with that of schizophrenia. Alternatively, individuals with vulnerability to
schizophrenia may be more sensitive to the effects of stress, while sensitivity to stress
is also determined by the degree of prior exposure to stress, possibly including stress
in early life (Janssen et al., 2004).

Socioeconomic factors, deprivation, and inequality

Several studies have found a relationship (not necessarily linear) between depri-
vation and incidence rates of psychosis (Croudace, 2000), prevalence rates of
schizophrenia (Moser, 2001) and admission rates for schizophrenia (Boardman
et al., 1997; Harrison et al., 1995; Koppel and McGuffin, 1999). The influence of
economic deprivation might be exerted via biological mechanisms such as diet
or overcrowding, but recent work has suggested that inequality may also have an
effect. Boydell et al. (2003) found that the incidence rate of schizophrenia as defined
by Research Diagnostic Criteria in deprived (but not in affluent) areas in London
increased as inequality within the local area increased (incidence rate ratio, 3.79)
after adjusting for age, sex, absolute deprivation, and ethnicity.

Interaction between social and other etiological factors

Gene–environment interaction
There is an apparent paradox in that schizophrenia appears highly heritable and
yet, as discussed above, many environmental and social factors, appear to play a
role. The predictive power of each of these environmental factors, however, is low
238 J. Boydell, J. van Os, and R. M. Murray

(i.e. most people exposed to each of the risk factors remain well and never develop
the illness). A unifying explanation seems to be that the environmental factors
operate upon genetic risk. There are many forms that this interaction could take:
synergistic, additive, multiplicative (van Os and Sham, 2003). We have reviewed
evidence from the Finnish and Danish adoption studies and the Israeli High Risk
Study that the social environment can operate on genetic factors to increase risk
of schizophrenia. Genes might also influence the likelihood of exposure to some
social risk factors. According to the model proposed by van Os and Marcelis (1998),
individuals vary in their sensitivity to adverse environmental circumstances, and
genetically sensitive individuals are more likely to develop psychiatric illness when
exposed to certain environments than those without genetic predisposition. In the
case of schizophrenia, being different in some way from one’s neighbors, and/or
experiencing discrimination seem to be important factors.

Social factors and cognitive processing

An important implication of any social theory is that the effects of social factors
may operate by impacting not only on brain development but also on psychological
processes that may contribute to the symptoms of schizophrenia. The development
of cognitive theories of psychosis in general and delusion formation in particular has
been a major advance in the understanding of schizophrenia. Detailed discussion of
this topic (reviewed by Blackwood et al., 2001) is outside the scope of this chapter,
but abnormal cognitive processing could be the mechanism by which a range of
social factors exerts an effect. People with persecutory delusions selectively attend
to threatening information, tend to jump to conclusions, attribute negative events
to external personal causes, and have difficulty in understanding others’ intentions,
motivations, and states of mind (Blackwood et al., 2001). It is plausible that people
with this cognitive style would be even more likely to develop delusions when
subjected to social adversity, than if they lived in a more benign social milieu.
Attributional style, in particular, has been identified as a pathway through which
discrimination and racial harassment could lead to psychosis (Sharpley et al., 2001).

Social causation versus social selection

Any discussion of a possible causal role of social factors needs to be considered within
the old debate on social selection versus social causation. There has been ongoing
debate as to whether social factors influence the development of schizophrenia (cau-
sation) or whether individuals at risk choose adverse social environments (selec-
tion). This has been a theme running throughout this chapter; a number of the
studies reviewed have produced findings that cannot be accounted for by social
selection. The question has now been largely superseded by discussion of more
complex interaction but we include it for completeness. The most influential paper
239 Social factors and development of schizophrenia

to address the causation/selection issue investigated the relative effects of social

status and ethnicity on risk for schizophrenia amongst Israeli-born people of north
African descent. Social status had a greater effect, leading the authors to conclude
that social selection is the more important factor (Dohrenwend et al., 1992). How-
ever, the rates were low in this study and so ethnicity might not have been associated
with a causal etiological agent whereas it might elsewhere. At the ecological level,
Boydell et al. (2001) found non-white ethnic minorities to be at higher risk when
they were in a smaller minority even though they lived in an area of higher social
status. Hence it is likely that opposite results would have been found had the study
been carried out in London.

Conclusions
A range of social factors have been reported to be implicated in the etiology of
schizophrenia, but the evidence for many of them remains scanty. Even for those
where the evidence is more substantial (e.g. urban living, social isolation, and
discrimination), it is unclear how these factors contribute to the risk. Indeed, it is
likely that more than one mechanism operates, and there may be many different
interactions with pre-existing genetic/neurodevelopmental vulnerabilities. It is now
clear, however, that, in order to understand the causes of schizophrenia, the role of
the social environment cannot continue to be ignored. In saying this, we are not
proposing an oppositional social instead of biological approach, which we consider
as futile as arguing whether poverty or mycobateria cause tuberculosis! Rather, we
suggest that both social and biological factors need to be studied as well as their
interaction.
We need to recognize that (i) social factors can impact on brain development,
(ii) some social factors give rise to psychological vulnerabilities, and (iii) many
social factors act over the life course, creating developmental liabilities. We have
presented evidence that early social adversity can affect brain development. For
example, structural abnormalities can be demonstrated in adult life in individuals
with a history of childhood abuse (Anderson et al., 2002; Bremner et al., 1997; Stein
et al., 1997). Now that evidence is emerging that childhood abuse may contribute
to schizophrenia risk, it is attractive to speculate that some of the neuroimaging
findings associated with abuse are relevant to schizophrenia; for example, decreased
hippocampal volume has been reported in both. It is possible that the social environ-
ment creates psychological vulnerabilities that act additively to the risk function
in combination with genetic or non-genetic neurodevelopmental impairments.
For example, chronic exposure to discrimination may give rise to a psychological
vulnerability to delusion formation, which in individuals with additional neurode-
velopmental impairment may result in overt psychotic disorder. Finally, it is also
240 J. Boydell, J. van Os, and R. M. Murray

plausible that the social environment and neurodevelopmental impairments rein-

force each other over the life course, creating enduring vulnerabilities. For example,
a child who is poorly coordinated may not join in at playtime and so forms fewer
friends. This pattern may be exacerbated in those who also have schizotypal traits,
leading to the developing adolescent becoming increasingly isolated and, later,
through his lack of social support in adult life, more susceptible to the effects of
social adversity and formation of early psychosis-like experiences.
The challenge for schizophrenia researchers in the coming decade is first to
distinguish those candidate social factors that do contribute to schizophrenia risk
from those that do not and, second, to identify the interplay between these factors,
genetic susceptibility, and their respective effects on, and interactions with, brain
development.

REFERENCES

Agid, O., Shapira, B., Zislin, J., et al. (1999). Environment and vulnerability to major psychiatric
illness: a case–control study of early parental loss in major depression, bipolar disorder and
schizophrenia. Mol Psychiatry 4: 163–172.
Allardyce, J., Boydell, J., van Os, J., et al. (2001). A comparison of the incidence of schizophrenia
in rural Dumfries and Galloway and urban Camberwell 179: 335–340.
American Psychiatric Association (1987). Diagnostic and Statistical Manual III–Revised.
Washington, DC: American Psychiatric Press.
Anderson, C. M., Teicher, M. H., Polcari, A., et al. (2002). Abnormal T2 relaxation time in the
cerebellar vermis of adults sexually abused in childhood: potential role of the vermis in
stress-enhanced risk for drug abuse. Psychoneuroendocrinology 27: 231–244.
Astrup C, Odegard, O. (1961). Internal migration and mental illness in Norway. Psychiatr Q 34:
116–130.
Bateson, G., Jackson, D., Haley, J., et al. (1956). Toward a theory of schizophrenia. Behav Sci 1:
251–264.
Bebbington, P. (2003). In Schizophrenia – not just biological. In Proceedings of Conference held
at the Institute of Psychiatry. London: Institute of Psychiatry, p. 2.
Beckett, C., Bredenkamp, D., Castle, J., et al. (2002). Behavior patterns associated with institutional
deprivation: a study of children adopted from Romania. J Dev Behav Pediatr 23: 297–303.
Berenbaum, H. (1999). Peculiarity and reported childhood maltreatment. Psychiatry 62: 21–35.
Bhugra, D., Leff, J., Mallett, R., et al. (1997). Incidence and outcome of schizophrenia in Whites,
African Caribbeans and Asians in London. Psychol Med 27: 791–798.
Bhugra, D., Hilwig, M., Mallett, R., et al. (2000). Factors in the onset of schizophrenia: a com-
parison between London and Trinidad samples. Acta Psychiatr Scand 101: 135–141.
Blackwood, N., Howard, R., Bentall, R., et al. (2001). Cognitive neuropsychiatric models of
persecutory delusions. Am J Psychiatry 158: 527–539.
241 Social factors and development of schizophrenia

Boardman, A., Hodgson, R., Lewis, M., et al. (1997). Social indicators and the prediction of
psychiatric admission in different diagnostic groups. Br J Psychiatry 171: 457–462.
Boydell, J., van Os, J., McKenzie, K., et al. (2001). Incidence of schizophrenia in ethnic minorities
in London: ecological study into interactions with environment. Br Med J 7325: 1336–1338.
Boydell, J., van Os, J., McKenzie, K., Murray, R. (2003). The influence of inequality on the
incidence of schizophrenia: an ecological study. Schizophr Bull 60: Suppl. 33.
Bremner, J. D., Randall, P., Vermetten, E., et al. (1997). Magnetic resonance imaging-based
measurement of hippocampal volume in posttraumatic stress disorder related to childhood
physical and sexual abuse: a preliminary report. Bio Psychiatry 41: 23–32.
Brown, G. W., Birley, J. L. (1968). Crises and life changes and the onset of schizophrenia. J Health
Soc Behav 9: 203–214.
Bruxner, G., Burvill, P., Fazio, S., et al. (1997). Aspects of psychiatric admissions of migrants to
hospitals in Perth. West Aust N Z J Psychiatry 31: 532–542.
Bryer, J. B., Nelson, B. A., Miller, J. B., et al. (1987). Childhood sexual and physical abuse as
factors in adult psychiatric illness. Am J Psychiatry 144: 1426–1430.
Butler, R. W., Mueser, K. T., Sprock, J., et al. (1996). Positive symptoms of psychosis in posttrau-
matic stress disorder. Biol Psychiatry 39: 839–844.
Cannon, M., Caspi, A., Moffitt, T., et al. (2002). Evidence for early-childhood, pan-developmental
impairment specific to schizophreniform disorder. Arch Gen Psychiatry 59: 449–456.
Castle, D., Wessley, S., Der, G., Murray, R. M. (1991). The incidence of operationally defined
schizophrenia in Camberwell, 1965–1984. Br J Psychiatry 159: 790–794.
Castle, D. J., Scott, K., Wessely, S., et al. (1993). Does social deprivation during gestation and
early life predispose to later schizophrenia? Soc Psychiatry Psychiatr Epidemiol 28: 1–4.
Chung, R. K., Langeluddecke, P., Tennant, C., et al. (1986). Threatening life events in the onset of
schizophrenia, schizophreniform psychosis and hypomania. Br J Psychiatry 148: 680–685.
Croudace, T. J., Kayne, R., Jones, P. B., et al. (2000). Non-linear relationship between an index of
social deprivation, psychiatric admission prevalence and the incidence of psychosis. Psychol
Med 30: 177–185.
Dauncey, K., Giggs, J., Baker, K., et al. (1993). Schizophrenia in Nottingham: lifelong residential
mobility of a cohort. Br J Psychiatry 163: 613–619.
DiPietro, J. A. (2000). Baby and the brain: advances in child development. Annu Rev Publ Health
21: 455–471.
Dohrenwend, B. P., Levav, I., Shrout, P. E., et al. (1992). Socioeconomic status psychiatric disorders
the causation-selection issue. Science 255: 946–952.
Drukker, M., Kaplan, C. D., Feron, F. J. M., et al. (2003). Children’s health-related quality of life,
neighbourhood socio-economic deprivation and social capital. A contextual analysis. Soc
Sci Med 57: 825–841.
Eisenberg, L. (1995). The social construction of the human brain. Am J Psychiatry 152: 1563–
1575.
Ellason, J. W., Ross, C. A. (1997). Childhood trauma and psychiatric symptoms. Psychol Rep 80:
447–450.
Farris, R., Dunham, H. (1939). Mental Disorders in Urban Areas. Chicago, IL: University of
Chicago Press.
242 J. Boydell, J. van Os, and R. M. Murray

Fishbein, D. (2000). The importance of neurobiological research to the prevention of psy-

chopathology. Prev Sci 1: 89–106.
Garety, P. A., Kuipers, E., Fowler, D., et al. (2001). A cognitive model of the positive symptoms
of psychosis. Psychol Med 31: 189–195.
Glover, G. (1989). Why is there a high rate of schizophrenia in British Caribbeans? Br J Hosp Med
42: 48–51.
Hammersley, P., Dias, A., Todd, G., et al. (2003). Childhood trauma and hallucinations in bipolar
affective disorder: a preliminary investigation. Br J Psychiatry 182: 543–547.
Hare, E. (1956). Mental illness and social conditions in Bristol. J Ment Sci 102: 349–357.
Harrison, G., Owens, D., Holten, A., et al. (1988). A prospective study of severe mental disorder
in Afro-Caribbean patients. Psychol Med 18: 643–657.
Harrison, G., Glazebrook, C., Brewin, J., et al. (1997). Increased incidence of psychotic disorders
in migrants from the Caribbean to the United Kingdom. Psychol Med 27: 799–806. [See
comments in Psychol Med (1998). 28: 496–497.]
Harrison, J., Barrow, S., Creed, F., et al. (1995). Social deprivation and psychiatric admission rates
among different diagnostic groups. Br J Psychiatry 167: 456–462.
Heidbreder, C. A., Foxton, R., Cilia, J., et al. (2001). Increased responsiveness of dopamine to
atypical but not typical antipsychotics in the medial pre-frontal cortex of rats reared in
isolation. Psychopharmacology 156: 338–351.
Heins, T., Gray, A., Tennant, M., et al. (1990). Persisting hallucinations following childhood
sexual abuse. Aust N Z J Psychiatry 24: 561–565.
Hickling, W., McKenzie, K., Mullen, R., et al. (1999). A Jamaican psychiatrist evaluates diagnoses
at a London hospital. Br J Psychiatry 175: 283–285.
Hirsch, S. R., Leff, J. P. (1975). Maudsley Monograph No. 22. Abnormalities in Parents of
Schizophrenics. London: Oxford University Press.
Hutchinson, G., Takei, N., Fahy, T. A., et al. (1996). Morbid risk of schizophrenia in first-degree
relatives of white and African-Caribbean patients with psychosis. Br J Psychiatry 169: 776–
780.
Hutchinson, G., Takei, N., Bhugra, T., et al. (1997). Increased rate of psychosis among African-
Caribbeans in Britain is not due to an excess of pregnancy and birth complications. Br J
Psychiatry 171: 145–147.
Hutchinson, G., Mallett, R., Fletcher, H., et al. (1999). Are the increased rates of psychosis reported
for the population of Caribbean origin in Britain an urban effect? Int Rev Psychiatry 11: 122–
128.
Jablensky, A., Cole, S. W. (1997). Is the earlier age at onset of schizophrenia in males a confounded
finding? Results from a cross-cultural investigation. Br J Psychiatry 170: 234–240.
Jacobs, S., Myers, J. (1976). Recent life events and acute schizophrenic psychosis: a controlled
study. J Nerv Ment Dis 162: 75–87.
Janssen, I., Hanssen, M., Bak, M., et al. (2002). Evidence that ethnic group effects on psychosis
risk are confounded by experiences of discrimination. Schizophr Res 53: 34.
Janssen, I., Hansen, M., Bak, M., et al. (2003). Discrimination and delusional ideation. Br J
Psychiatry 182: 71–76.
243 Social factors and development of schizophrenia

Janssen, I., Krabbendam, L., Bak, M., et al. (2004). Childhood abuse as a risk factor for psychotic
experience. Acta Psychiatr Scand in press.
Johns, L. C., van Os, J. (2001). The continuity of psychotic experiences in the general population.
Clin Psychol Rev 21: 1125–1141.
Jones, P., Rodgers, B., Murray, R., et al. (1994). Child developmental risk factors for adult
schizophrenia in the British 1946 birth cohort. Lancet 344: 1398–1402.
Johnstone, E. C., Crow, T. J., Frith, C. D., et al. (1976). Cerebral ventricular size and cognitive
impairment in chronic schizophrenia. Lancet, ii: 924–926.
Karlsen, S., Nazroo, J. (2002). Relation between racial discrimination and health among ethnic
minority groups. Am J Publ Health 92: 624–631.
King, M., Coker, E., Leavey, G. et al. (1994). Incidence of psychotic illness in London: a comparison
of ethnic groups. Br Med J 309: 1115–1119.
Koppel, S., McGuffin, P. (1999). Socioeconomic factors that predict psychiatric admissions. Psy-
chol Med 29: 1235–1241.
Krabbendam, L., Janssen, I., Bak, M., et al. (2002). Neuroticism and low self-esteem as risk factors
for psychosis. Soc Psychiatry Psychiatr Epidemiol 37: 1–6.
Kubicka, L., Matejcek, Z., David, H.P., et al. (1995). Children from unwanted pregnancies in
Prague, Czech Republic revisted at age 30. Acta Psychiatr Scand 91: 361–369.
Lewis, G., David, A., Andreasson, S., et al. (1992). Schizophrenia and city life. Lancet 340: 137–
140.
Lidz, T. (1958). Schizophrenia and the family. Psychiatry 21: 21–27.
Lyons, D. M., Afarian, H., Schatzberg, A. F., et al. (2002). Experience-dependent asymmetric
variation in primate prefrontal morphology. Behav Brain Res 136: 51–59.
Mahy, G., Mallett, R., Leff, J., Bhugra, D. (1999). First-contact incidence rate of schizophrenia on
Barbados. Br J Psychiatry 175: 28–33.
Maier, W., Minges, J., Lichtermann, D., et al. (1994). Personality variations in healthy relatives of
schizophrenics. Schizophr Res 12: 81–88.
Malla, A. K., Cortese, L., Shaw, T. S., et al. (1990). Life events and relapse in schizophrenia. A one
year prospective study. Soc Psychiatry Psychiatr Epidemiol 25: 221–224.
Mallett, R., Hutchinson, G., Leff, J., et al. (1998). Social conditions and schizophrenia in African-
Caribbeans: Edward Hare revisited. In Proceedings of Winter Meeting 39 of the Royal College
of Psychiatrists. London: Royal College of Psychiatrists.
Mallett, R., Leff, J., Bhugra, D., Pang, D., Zhao, J. (2002). Social environment, ethnicity and
schizophrenia. A case-control study. Soc Psychiatry Psychiatr Epidemiol 37: 329–335.
Malmberg, A., Lewis, G., David, A., et al. (1998). Premorbid adjustment and personality in people
with schizophrenia. Br J Psychiatry 172: 308–313.
Malzberg, B. (1969). Are immigrants psychologically disturbed? In Changing Perspectives in Men-
tal illness, ed. S. Plog, R. Edgerton. New York: Holt, Rinehart and Winston, pp. 395–421.
Marcelis, M., Navarro-Mateu, F., Murray, R., et al. (1998). Urbanization and psychosis: a study
of 1942–1978 birth cohorts in the Netherlands. Psychol Med 28: 871–879.
Marcelis, M., Takei, N., van Os, J., et al. (1999). Urbanization and risk for schizophrenia: does
the effect operate before or around the time of illness onset? Psychol Med 29: 1197–1203.
244 J. Boydell, J. van Os, and R. M. Murray

Matthews, M. K., Dalleny, J. W., Matthews, C., et al. (2001). Periodic maternal separation of
neonatal rats produces region- and genetic-specific effects on bioamine content in post-
mortem adult brain. Synapse 40: 1–10
McGuire, P., Jones, P., Harvey, M., et al. (1995). Morbid risk of schizophrenia in relatives of
patients with cannabis-associated psychosis. Schizophr Res 15: 277–281.
McKenzie, K., van Os, J., Fahy, T., et al. (1995). Psychosis with good prognosis in Afro-Caribbean
people now living in the United Kingdom. [See comments] Br Med J 311: 1325–1328.
Mednick, S.A., Parnas, J., Schulsinger, F., et al. (1987). The Copenhagen High Risk Project 1962–
1986. Schizophr Bull 133: 485–495.
Mirsky, A. F., Silberman, E. K., Latz, A., et al. (1985). Adult outcomes of high-risk children:
differential effects of town and kibbutz rearing. Schizophr Bull 11: 150–156.
Mortensen, P. B., Cantor-Graae, E., McNeil, T. F., et al. (1997). Increased rates of schizophrenia
among immigrants: some methodological concerns raised by Danish findings. Psychol Med
274: 813–820.
Mortensen, P., Pedersen, C., Westergaard, T., et al. (1999). Effects of family history and place and
season of birth on the risk of schizophrenia. N Eng J Med 340: 603–608. [Sec correspondence
in N Engl J Med (1999) 341: 370–372].
Moser, K. (2001). Inequalities in treated heart disease and mental illness in England and Wales
1994–1998. Br J Gen Pract 51: 438–444.
Murphy, H. B. (1977). Migration, culture and mental health. Psychol Med 7: 677–684.
Murray, R., Fearon, P. (1999). The developmental “risk factor” model of schizophrenia. J Psychiatr
Res 33: 497–499.
Murray, R. M., O’Callaghan, E., Castle, D. J., et al. (1992). A neurodevelopmental approach to
the classification of schizophrenia. Schizophr Bull 18: 319–332.
Myhrman, A. (1992). Unwanted pregnancy, its occurrence and significance for the family and
the child. Acta Univ Ouluen Ser 249: 1–46.
Myhrman, A., Rantakallio, P., Isohanni, M., et al. (1996). Unwantedness of pregnancy and
schizophrenia in the Child. Br J Psychiatry 169: 637–640.
Myin-Germeys, I., van Os, J., Schwartz, J. E., et al. (2001). Emotional reactivity to daily life stress
in psychosis. Arch Gen Psychiatry 58: 1137–1144.
Myin-Germeys, I., Peeters, F., Havermans, R., et al. (2003a). Emotional reactivity to daily life stress
in psychosis and affective disorder: an experience sampling study. Acta Psychiatr Scand 107:
124–131.
Myin-Germeys, I., Krabbendam, L., Delespaul, P. A., et al. (2003b). Do life events have their
effect on psychosis by influencing the emotional reactivity to daily life stress? Psychol Med
33: 327–333.
Ødegård, O. (1932). Emigration and insanity: a study of mental disease among Norwegian born
population in Minnesota. Acta Psychiatr Neurol Scand 7(Suppl. 4): 1–206.
Ovtscharoff, Jr., W., Braun, K. (2001). Maternal separation and social isolation modulate the
postnatal development of synaptic composition in the infralimbic cortex of Octodon degus.
Neuroscience 104: 33–40.
Pedersen, C. B., Mortensen, P. B. (2001). Evidence of a dose–response relationship between
urbanicity during upbringing and schizophrenia risk. Arch Gen Psychiatry 58: 1039–1046.
245 Social factors and development of schizophrenia

Putnam, F. W., Guroff, J. J., Silberman, E. K., et al. (1986). The clinical phenomenology of multiple
personality disorder: review of 100 recent cases. J Clin Psychiatry 47: 285–293.
Read, J., Argyle, N. (1999). Hallucinations, delusions, and thought disorder among adult psychi-
atric inpatients with a history of child abuse. Psychiatr Serv 50: 1467–1472.
Read, J., Perry, B. D., Moskowitz, A., et al. (2001). The contribution of early traumatic events to
schizophrenia in some patients: a traumagenic neurodevelopmental model. Psychiatry 64:
319–345.
Read, J., Agar, K., Argyle, N., et al. (2003). Sexual and physical abuse during childhood and
adulthood as predictors of hallucinations, delusions and thought disorder. Psychol Psychother
Theory Res Pract 76: 1–22.
Rosenthal, D., Wender, P., Kety, S., et al. (1971). The adopted-away offspring of schizophrenics.
Am J Psychiatry 128: 307–311.
Ross, C. A., Joshi, S. (1992). Schneiderian symptoms and childhood trauma in the general popu-
lation. Compr Psychiatry 33: 269–273.
Ross, C. A., Norton, G. R., et al. (1989). Multiple personality disorder: an analysis of 236 cases.
Can J Psychiatry Rev Can Psychiatry 34: 413–418.
Ross, C. A., Anderson, G., Clark, P., Wozney, K. (1994). Childhood abuse and the positive symp-
toms of schizophrenia. Hosp Community Psychiatry 45: 489–491.
Sanchez, M. M., Ladd, C. O., Plotsky, P. M. (2001). Early adverse experience as a developmen-
tal risk factor for later psychopathology: evidence from rodent and primate models. Dev
Psychopathol 13: 419–449.
Sansonnet-Hayden, H., Haley, G., Marriage, K., et al. (1987). Sexual abuse and psychopathology
in hospitalized adolescents. J Am Acad Child Adolesc Psychiatry 26: 753–757.
Selten, J. P., Sijben, N. (1994). First admission rate for schizophrenia in immigrants to the Nether-
lands: the Dutch National Register. Soc Psychiatry Psychiatr Epidemiol 29: 71–72.
Selten, J. P., Slaets, J. P., Kahn, R. S., et al. (1997). Schizophrenia in Surinamese and Dutch
Antillean immigrants to the Netherlands: evidence of an increased incidence. Psychol Med
27: 807–811.
(1998). Prenatal exposure to influenza and schizophrenia in Surinamese and Dutch Antillean
immigrants to the Netherlands. Schizophr Res 30: 101–103.
Selten, J. P., Veen, N., Feller, W., et al. (2001). Incidence of psychotic disorders in immigrant
groups to the Netherlands. Br J Psychiatry 178: 1–7.
Selten, J. P., Cantor-Graae, E., Slaets, J., et al. (2002). Ødegårds selection hypothesis revisited:
schizophrenia in Surinamese immigrants to the Netherlands. Am J Psychiatry 159: 669–671.
Sharpley, M., Hutchinson, G., McKenzie, K., et al. (2001). Understanding the excess of psychosis
among the African-Caribbean population in England. Review of current hypotheses. Br J
Psychiatry 178(Suppl. 40): 60–68.
Startup, M. (1999). Schizotypy, dissociative experiences and childhood abuse: relationships
among self-report measures. Br J Clin Psychol 38: 333–344.
Stein, M. B., Koverola, C., Hanna, C., et al. (1997). Hippocampal volume in women victimized
by childhood sexual abuse. Psychol Med 27: 951–959.
Sugarman, P., Crawford, D. (1994). Schizophrenia in the Afro-Caribbean community. Br J Psy-
chiatry 164: 474–480.
246 J. Boydell, J. van Os, and R. M. Murray

Swett, Jr., C. , Surrey, J., Cohen, C., et al. (1990). Sexual and physical abuse histories and psychiatric
symptoms among male psychiatric outpatients. Am J Psychiatry 147: 632–636.
Takano, T., Nakamura, K. (2001). An analysis of health levels and various indicators of urban
environments for healthy cities projects. J Epidemiol Community Health 55: 263–270.
Thomas, C. S., Stone, K., Osborn, M., et al. (1993). Psychiatric morbidity and compulsory
admission among UK born Europeans, Afro-Caribbeans and Asians in central Manchester.
Br J Psychiatry 163: 91–99.
Thornicroft, G., Bisoffi, G., De Salva, D., et al. (1993). Urban–rural differences in the associa-
tions between social deprivation and psychiatric service utilization in schizophrenia and all
diagnoses: a case-register study in Northern Italy. Psychol Med 23: 487–496.
Tienari, P. (1991). Interaction between genetic vulnerability and family environment: The Finnish
adoptive family study of schizophrenia. Acta Psychiatrica Scand 84: 460–465.
Tienari, P., Wynne, L., Laksy, K., et al. (2000). Finnish Adoptive Family Study. Acta Psychiatr
Scand 101 433–443.
van Os, J., Jones, P. B. (1999). Early risk factors and adult person–environment relationships in
affective disorder. Psychol Med 29: 1055–1067.
(2001). Neuroticism as a risk factor for schizophrenia. Psychol Med 31: 1129–1134.
van Os, J., Marcelis, M. (1998). The ecogenetics of schizophrenia: a review. Schizophr Res 32:
127–135.
van Os, J., Sham, P. (2003). Gene–environment correlation and interaction in schizophrenia. In
The Epidemiology of Schizophrenia, ed. R. Murray, P. Jones, E. Susser, J. van Os, M. Cannon.
Cambridge, UK: Cambridge. University Press.
van Os, J., Castle, D. J., Takei, N., et al. (1996a). Psychotic illness in ethnic minorities: clarification
from the 1991 census. Psychol Med 26: 203–208.
van Os, J., Takei, N., Castle, D. J., et al. (1996b). The incidence of mania: time trends in relation
to gender and ethnicity. Soc Psychiatry Psychiatr Epidemiol 31: 129–136.
van Os, J., Driessen, G., Gunther, N., et al. (2000). Neighbourhood variation in incidence of
schizophrenia. Evidence for person–environment interaction. Br J Psychiatry 176: 243–248.
van Os, J., Hanssen, M., Bijl, R. V., et al. (2001). Prevalence of psychotic disorder and com-
munity level of psychotic symptoms: an urban–rural comparison. Arch Gen Psychiatry 58:
663–668.
van Os, J., Hanssen, M., de Graaf, R., et al. (2002). Does the urban environment independently
increase the risk for both negative and positive features of psychosis? Soc Psychiatry Psychiatr
Epidemiol 37: 460–464.
van Os, J., Hanssen, M., Bak, M., et al. (2003). Do urbanicity and familial liability coparticipate
in causing psychosis? Am J Psychiatry 160: 477–482.
Vaughn, C. E., Leff, J. P. (1976). The influence of family and social factors on the course of
psychiatric illness. Br J Psychiatry 129: 125–137.
(1982). Patterns of emotional responses in the relatives of schizophrenic patients. Schizophr
Bull 7: 43–44.
Ventura, J., Nuechterlein, K. H., Lukoff, D., et al. (1989). A prospective study of stressful life
events and schizophrenic relapse. J Abnorm Psychol 98: 407–411.
247 Social factors and development of schizophrenia

Wahlberg, K. E., Wynne, L., Oja, H., et al. (1997). Gene–environment interaction in vulnerability
to schizophrenia: findings from the Finnish Adoptive Family Study of Schizophrenia. Am J
Psychiatry 154: 355–362.
Wahlberg, K. E., Wynne, L., Oja, H., et al. (2000). Thought disorder index of Finnish adoptees
and communication deviance of their adoptive parents. Psychol Med 30: 127–136.
Weingarten, M. A., Orron, D. E. (1983). Schizophrenia in a Yemenite immigrant town in Israel.
Int J Soc Psychiatry 29: 249–254.
Wessely, S., Castle, D., Der, R., et al. (1991). Schizophrenia and Afro-Caribbeans. A case-control
study. Br J Psychiatry 159: 795–801.
Whitaker-Azmitia, P., Zhou, F., Hobin, J., et al. (2000). Isolation rearing of rats produces deficits
as adults in the serotonergic innervation of hippocampus. Peptides 21: 1755–1759.
Wiesel, T. N., Hubel, D. H. (1974). Ordered arrangement of orientation columns in monkeys
lacking visual experience. J Comp Neurol 158: 307–318.
Wing, J. K. (1978). The social context of schizophrenia. Am J Psychiatry 135: 1333–1339.
World Health Organization (1967). Manual of International Statistical Classification of Diseases,
Injuries and Causes of Death, 8th edn. Geneva: World Health Organization.
Wynne, L., Ryckoff, I., Day, J., et al. (1958). Pseudo-mutuality in the family relations of
schizophrenics. Psychiatry 21: 205–220.
 14

How does drug abuse interact with familial

and developmental factors in the etiology
of schizophrenia?
Chih-Ken Chen1 and Robin M. Murray2
1
Chang Gung Memorial Hospital, Keelung, Taiwan
2
Institute of Psychiatry, King’s College, London, UK

Schizophrenia is frequently described as a neurodevelopmental disease with a strong

genetic component (Tsuang et al., 2001). While it may be an exaggeration to call
schizophrenia a neurodevelopmental disease, it is clear that it has a major develop-
mental component. Furthermore, family, twin, and adoption studies suggest that
65–85% of the liability to schizophrenia can be attributed to genes (Cannon et al.,
1998; Cardno et al., 1999; Kendler and Diehl, 1993). As outlined elsewhere in this
volume (see Ch. 11), the environmental risk factors for schizophrenia can be sum-
marized as operating either early in life or later nearer the onset of frank psychosis.
In this chapter, we focus on the role of drug abuse as one of the later factors, and
on how it interacts with familial and developmental factors.

Association of drug abuse and psychosis

Numerous studies have found that those with psychosis commonly use drugs. In
the USA, a study from the Eastern Pennsylvania Psychiatric Institute (Muesser et al.,
1990) reported that, among schizophrenia patients, 47% used alcohol, 42% used
cannabis, 25% used stimulants, 18% used hallucinogens, 7% used sedatives, and
4% used narcotics. A Veterans Administration inpatient study reported that 60% of
those admitted with schizophrenia had a current or past history of drug abuse (Selzer
and Lieberman, 1993). A meta-analysis concluded that substance-use disorders
occur in approximately 40–50% of schizophrenia individuals in the USA. These
rates are much higher than those for the general population. For example, in the US
Epidemiologic Catchment Area Study, the rate of lifetime substance-use disorder in
the general population was 17%, but it was 48% among persons with schizophrenia
(Regier et al., 1990).
Neurodevelopment and Schizophrenia, ed. Matcheri S. Keshavan et al. Published by Cambridge University
Press.

C Cambridge University Press 2004.

248
249 Drug abuse and development of schizophrenia

Elsewhere in the world, rates of substance abuse are generally lower, but again
consumption is more frequent in those with psychosis than in the rest of the popu-
lation. In the UK, in South London, the 1-year prevalence rate among patients with
psychotic illnesses was found to be 36.3% for any substance problem and 15.8%
for drug problems (Menezes et al., 1996). A more recent study from Scotland
(McCreadie, 2002) revealed that problem use of drugs by people with schizophre-
nia was greater than in the general population for both the previous year (7% versus
2%) and at any time previously (20% versus 6%). Cantwell et al. (1999) examined
168 subjects with first-episode psychosis in the UK and reported that 37% of the
sample met criteria for drug use, drug misuse, or alcohol misuse; 8.4% of the sub-
jects received a primary diagnosis of substance-related psychotic disorder. The rates
of substance use among individuals with schizophrenia in Continental Europe are
similar to this and rarely as high as those reported in US studies (Cassano et al.,
1998; Modestin et al., 1997; Soyka et al., 1993; Verdoux et al., 1996a).

Direction of the association

The associations between use of certain drugs and schizophrenia are generally
not disputed. However, there are several possible explanations for these observed
associations (Blanchard et al., 2000; Murray et al., 2003; Thornicroft, 1990).

Self-medication
A long-standing notion has been that substance abuse is often the product of
psychopathology, resulting from the patient’s attempt to “self-medicate” dysphoric
symptoms (Khantzian, 1985). Patients suffering from schizophrenia are postulated
to use substances either to improve their negative, depressive, or anxiety symptoms
(DeQuardo et al., 1994; Dixon et al., 1991; Mueser et al., 1998; Schneier and Siris,
1987) or to alleviate the side effects of antipsychotic drugs (Dixon et al., 1991; Lopez
and Jeste, 1997). For example, small amounts of cannabis are said to attenuate
anxiety, depression, and negative symptoms in schizophrenia (Linszen et al., 1994;
Peralta and Cuesta, 1992). However, empirical studies (Fischman and Schuster,
1982; Griffith et al., 1972) on the effects of cannabis and stimulants have often
failed to produce any evidence to substantiate the self-medication hypothesis.

Common factor hypothesis

The common factor hypothesis proposes that substance abuse and schizophrenia
share common etiological factors, which may include low socioeconomic status,
certain personality traits (Mueser et al., 1998), or genetic vulnerability (Blanchard
et al., 2000). Mueser et al. (1998) speculate that antisocial personality may underlie
both schizophrenia and substance misuse, but this argument has not been supported
250 C.-K. Chen and R. M. Murray

by cohort studies (Jones et al., 1994). So far there is little evidence for a common
genetic vulnerability for schizophrenia and drug abuse.

Vulnerability hypothesis
As with other common chronic diseases such as diabetes and coronary artery disease,
many consider that the appropriate etiological model (Murray and Fearon, 1999)
for schizophrenia is one involving multiple genes and environmental risk factors.
As discussed elsewhere in this book, genetic and/or early environmental factors are
postulated to cause the development of anomalous neural networks, which interact
in the growing child with inherited schizotypal traits to establish a trajectory towards
an increasingly solitary and deviant lifestyle. The vulnerablity hypothesis postulates
that stressors, including the use of drugs, then precipitate the individual across the
threshold for expression of psychosis (e.g. Bebbington et al., 1993; Jones, et al.,
1994; Tsuang et al., 2001; Verdoux et al., 1996b).

Drug abuse as a cause of psychosis

Breakey et al. (1974) suggested that taking certain drugs (amphetamines, cannabis,
or hallucinogens) might play a precipitating role in the onset of schizophrenia,
bringing the disorder on earlier and increasing its likelihood in patients who other-
wise seemed somewhat less-constitutionally vulnerable than the average psychotic
individual. Obviously, different drugs may have different associations with psy-
chosis because of their varying actions on the brain. In this review, we will concen-
trate on drugs that have been shown to have a clear association with schizophrenia,
namely the psychostimulants cocaine, amphetamine, and methamphetamine, as
well as cannabis and phenyclidine (PCP).

Amphetamine-induced psychosis
Amphetamine psychosis is a reaction that usually occurs following chronic use
of amphetamines, though it can occur after only a short period of amphetamine
consumption. Patients with amphetamine psychosis show a picture very similar
to paranoid schizophrenia (Bell, 1965; Connell, 1958). Indeed, amphetamine psy-
chosis so closely resembles schizophrenia, especially the paranoid type, that it has
been viewed as a model of schizophrenia (Crow et al., 1976; Snyder, 1973).
Connell (1958) and Bell (1965) originally conceptualized amphetamine psy-
chosis as lasting no more than 1 week after the urine becomes negative for
amphetamines. However, the psychosis does not always remit so readily, and use of
methamphetamine has been reported in several studies to result in psychotic symp-
toms lasting several months or more (Iwanami et al., 1994; Nakatani et al., 1989; Sato
et al., 1983; Tatetsu et al., 1956). Methamphetamine is a derivative of amphetamine,
251 Drug abuse and development of schizophrenia

with similar but more pronounced psychotropic properties (National Institute on

Drug Abuse, 1998). Methamphetamine comes in many forms and can be smoked,
snorted, orally ingested, or injected. In the Pacific region, especially Japan and
Taiwan, abuse of methamphetamine and the resultant psychosis are common (Chou
et al., 1999; Suwaki et al., 1997). Tatetsu (1963) reported that a substantial minority
of patients with methamphetamine psychosis did not recover within the first month
but rather went on to develop a chronic schizophrenic-like state. Almost 15% of
Japanese patients with methamphetamine psychosis since 1945 took 5 or more years
to recover in the absence of ongoing methamphetamine use (Konuma, 1994). Such
prolonged psychotic symptoms are hypothesized by some investigators to represent
a triggering or unmasking of vulnerability to psychosis by the amphetamine abuse
(Bell, 1973; Flaum and Schultz, 1996; Gold and Bowers, 1978).
In Connell’s original report (1958) 9 of 42 subjects developed psychosis after a
single dose of amphetamine. Connell considered amphetamine to be a true hal-
lucinogen that was directly related to the syndrome and he, therefore, regarded
amphetamine psychosis as a distinct disorder. Several prospective studies, in which
amphetamine psychosis was experimentally induced in non-schizophrenia drug
abusers, indicated that amphetamine does not simply activate latent schizophrenia
since the psychosis can be produced in apparently normal subjects (Angrist and
Gershon, 1970; Bell, 1973; Griffith et al., 1969).

Cannabis-induced psychosis:
Of the plant-based drugs, the illicit consumption of cannabis is most widespread
(United Nations International Drug Control Programme, 1997). Many studies
(Dixon et al., 1991; Linszen, et al., 1994; Longhurst, 1997; Mathers et al., 1991;
Rolfe et al., 1993; Schneier and Siris, 1987) have reported a positive association
between cannabis use and psychosis. Furthermore, in a prospective cohort study of
psychotic patients, Linszen et al. (1994) found that significantly more and earlier
relapses occurred in the cannabis-abusing group. In all but one of their subjects,
cannabis abuse preceded the onset of the first psychotic symptoms by at least
1 year. The authors, therefore, concluded that cannabis abuse, and particularly
heavy abuse, can elicit relapse in patients with schizophrenia and related disorders,
and it may possibly precipitate the initial psychosis.
Thornicroft (1990), in a review of the evidence for an association between
cannabis and psychosis, concluded that there is no convincing support for a separate
clinical diagnosis of “cannabis psychosis.” A retrospective study of 272 psychotic
inpatients (70 with cannabis-related psychosis) also failed to find any consistent
pattern of symptoms associated with cannabis use (Imade and Ebie, 1991). How-
ever, arguments against the use of this particular diagnostic label should not be
confused with arguments against any association between the drug and psychosis
252 C.-K. Chen and R. M. Murray

(Thomas, 1993). Chronic psychosis following prolonged cannabis abuse may reflect
the precipitation of schizophreniform disorder or schizophrenia in a vulnerable
individual, rather than a separate diagnosis of cannabis psychosis (Thornicroft,
1990). To establish conclusively whether cannabis consumption actually causes the
onset of schizophrenia, one must obtain prospective data and establish whether
there is evidence of a dose–response relationship between the amount consumed
and the risk of psychosis.
The first prospective cohort study was reported in 1987. Andreasson et al. (1987)
followed up almost 50 000 individuals conscripted into the Swedish Army and
reported the relative risk of developing schizophrenia was 2.4 for cannabis users
compared with non-users, rising to 6.0 for heavy users. An extension of this Swedish
Conscript Study has recently been carried out (Zammit et al., 2002). Consistent with
previous findings, this showed that “heavy cannabis users” by the age of 18 years
were 6.7 times more likely than non-users to be diagnosed with schizophrenia in the
ensuing 27 years. The risk was reduced but remained significant after controlling
for other potential confounding factors such as disturbed behavior, low intelligence
quotient score, growing up in a city, cigarette smoking, other drug use, and poor
social integration.
However, this study was largely ignored for 15 years. Then, in the Netherlands,
van Os and his colleagues (2002) examined the effect of cannabis use on psychotic
symptoms among the general population, assessing 4045 subjects at baseline and
again 3 years later. Compared with non-users, individuals using cannabis at baseline
were nearly three times more likely to manifest psychotic symptoms at follow-up.
This risk remained significant after statistical adjustment for a range of confounders.
There was a dose–response relationship with the highest risk (odds ratio, 6.81) being
observed for the highest level of cannabis use. Further analysis revealed that lifetime
history of cannabis use at baseline, as opposed to use of cannabis at follow-up, was a
stronger predictor of psychosis 3 years later than cannabis use at the later point. This
suggests that the association between cannabis use and psychosis was not merely a
short-term effect of cannabis use leading to an acute psychotic episode.
The Swedish and Dutch studies were open to criticism on a number of points
(Johnson et al., 1988; Negrete, 1989), notably that both the cannabis use and the
schizophrenia may have been related to an underlying predisposing factor such as
premorbid personality. This criticism has been at least partly overcome in a study
from New Zealand, which followed-up a general population birth cohort of 1037
individuals born in Dunedin, New Zealand in 1972–1973 (96% follow-up rate at age
26). Although much smaller, this study has information on self-reported psychotic
symptoms at age 11 years (i.e. before the likely onset of cannabis use) and then
information on cannabis use (self-reports) at 15 and 18 years. Thus, it allows the
age of onset of cannabis use to be examined in relation to later psychiatric outcome.
253 Drug abuse and development of schizophrenia

Results showed that both cannabis users by age 15 and cannabis users by age 18
had elevated schizophrenia symptoms at age 26, which remained significant after
controlling for psychotic symptoms predating the onset of cannabis use (Arseneault
et al., 2002). The effect was much stronger with earlier use, and onset of cannabis use
by age 15 was associated with an increased likelihood of meeting diagnostic criteria
for schizophreniform disorder at age 26. Indeed, 10.3% of those using cannabis at
age 15 in this cohort were diagnosed with schizophreniform disorder at age 26, as
opposed to 3% of the controls.

Phencyclidine-induced psychosis
PCP is a dissociative anesthetic originally employed in human anesthesia in the late
1950s. It can produce hallucinations, relaxation, feeling of dissociation from the
environment, and sometimes intense euphoria (Jacobs et al., 1987). Further char-
acterization of the psychomimetic actions of PCP was facilitated by the epidemic
of PCP abuse that occurred in the USA during the late 1960s and 1970s. Although
hallucinations, agitation, and paranoid delusions occur after PCP administration,
the most striking and consistent behavioral effects of PCP are alterations in body
image, disorganization of thought, negativism, and apathy (Javitt and Zukin, 1991).
Javitt and Zukin (1991) suggested that the symptoms induced by PCP are similar
to the “four A’s” proposed by Bleuler as the primary symptoms of schizophrenia:
affective blunting, ambivalence, autism, and disturbance of association. They also
suggested that PCP psychosis may provide a neurochemical model corresponding
uniquely to the Bleulerian conception of schizophrenia.
PCP interacts selectively with a specific binding site (PCP receptor) that is associ-
ated with the N-methyl-D-aspartate (NMDA)-type excitatory amino acid receptor.
Occupation of its receptor by PCP induces non-competitive inhibition of NMDA
receptor-mediated neurotransmission (Snell et al., 1988). Primary dysfunction of
NMDA receptor functioning could potentially account for the deficit in cognitive
functioning associated with schizophrenia, as well as for the dopaminergic hyper-
activity and dysregulation associated with acute schizophrenia (Kleinman et al.,
1988).
In short, PCP psychosis has been an attractive fashionable model for psychosis.
Unfortunately, there have been few studies of its role in precipitating long-lasting
schizophrenia-like psychoses in real life.

Lysergic acid diethylamide-induced psychosis

Hallucinogens, by comparison, do not generally create psychoses that truly simu-
late schizophrenia. Drugs like lysergic acid diethylamide (LSD) and mescaline
evoke chemical psychosis with characteristic hallucinogenic distortions of per-
ception such as synesthesia (a condensation of two sensations); for instance, a
254 C.-K. Chen and R. M. Murray

person will “taste” green, “smell” grey or “see” music (Jacobs et al., 1987; Slaby,
1991). However, the picture in some patients remains sufficiently complex to ren-
der LSD psychosis very difficult to distinguish from acute schizophrenia (Vardy
and Kay, 1983). Many believe that LSD alone does not produce an enduring psy-
chosis but rather precipitates schizophrenia in predisposed personalities (Slaby,
1991). Strassman (1984) suggested that the development of chronic LSD psychosis
is multifactorial in origin and that premorbid factors such as family history and per-
sonality characteristics, as well as exposure to a multitude of drugs, play important
roles.

Individual differences in liability to developing drug-related psychosis

It is clear that a number of drugs in different classes can cause, or increase the
susceptibility for, a state of chronic psychosis. Why do some individuals develop
psychotic symptoms after taking a specific drug, while others use regularly over
long periods and remain unscathed? There is a striking paucity of studies that have
examined the predisposing factors systematically. It is obvious that the drug abuse,
which is at least partly genetically determined, plays a critical role in substance-
induced psychosis. Nevertheless, in addition to factors related to drug consumption,
a series of variables relating to a patient’s personality and genetic predisposition
seem crucial in determining the response to a given agent (Post, 1975).
In short, the risk for developing a substance-abuse problem is not equally dis-
tributed in the general population. Freud (1885) was cognizant of the importance
of predisposing characterisitics in relation to cocaine’s effects, stating:

I could not fail to note, however, that the individual disposition plays a major role in the effects of
cocaine, perhaps a more important role than with other alkaloids. The subjective phenomena after
ingestion of cocaine differ from person to person, and only a few persons experience, like myself, a
pure euphoria without alteration. Others already experience slight intoxication, hyperkinesia and
talkativeness after the same amount of cocaine, while still others have no subjective symptoms
of the effects of cocaine at all.

Genetic determinants of response to stimulants are also highlighted by the demon-

stration of significant differences in response to amphetamine in animals, for exam-
ple, hyperthymic responses in different strains of mice (Jori and Garattini, 1973).
Again, the effects of an acute systemic injection of methamphetamine on motor
behavior (stereotypy, locomotor activity, and rearing) and extracellular dopamine
in the ventral striatum are different in two different strains of rats (Camp et al.,
1994).
255 Drug abuse and development of schizophrenia

Childhood development and personality

Tsuang et al. (1982) studied premorbid schizoid/paranoid personality and family
psychiatric history of drug abusers with psychosis. They compared hospital records
for four groups: (i) 72 drug abusers with psychosis lasting less than 6 months before
admission, (ii) drug abusers with psychosis of a longer duration, (iii) drug abusers
without psychosis, and (iv) patients with schizophrenia and atypical schizophrenia
without drug abuse. Their study examined the role of premorbid personality, family,
and genetic factors by comparison of multiple subgroups, and thus it provides
a good model for studying predisposing factors to substance-induced psychosis.
They reported that more drug abusers with prolonged psychosis have schizoid or
paranoid premorbid personality than drug abusers without psychosis and those
with short-term psychosis.
Vardy and Kay (1983) compared family history, manifest symptoms, premorbid
adjustment, and profiles on an extensive test battery between patients with LSD
psychosis and those with matched first-episode schizophrenia. The two groups
were distinguished on some clinical features but were equivalent in premorbid
adjustment and in number of subsequent rehospitalizations. The authors, there-
fore, suggested that in most respects the patients with LSD psychosis were funda-
mentally similar to those with schizophrenia in geneology, phenomenology, and
course of illness, and the study supported a model of LSD psychosis as a drug-
induced schizophreniform reaction in persons vulnerable to psychosis (Vardy and
Kay, 1983).
Schizotypy is related genetically to schizophrenia and is often considered to
represent a phenotypic expression of familial–genetic liability to schizophrenia
(Battaglia et al., 1997). A few investigators have reported correlations of drug abuse
and schizotypal personality traits in non-clinical individuals (Dumas et al., 2002;
Kwapil, 1996; Williams et al., 1996). Dumas et al. (2002) reported that cannabis users
exhibit more psychotic-like schizotypal traits compared with never-users, regardless
of their anxiety or depression dimension levels. Verdoux et al. (2002) found in a
non-clinical population that the acute effects of cannabis were modified by the
subject’s level of psychosis-proneness. Those with low vulnerability for psychosis
experienced enhanced feelings of pleasure after taking cannabis, while those with
high vulnerability for psychosis were more likely to respond with increased hostility
and suspiciousness (Verdoux et al., 2002).
We compared childhood premorbid characteristics between 88 metham-
phetamine users with a lifetime diagnosis of methamphetamine psychosis and 116
methamphetamine users without psychosis by interviewing their parents with the
Premorbid Schizoid and Schizotypal Traits (PSST) schedule. Subjects with exper-
ience of psychotic symptoms before their first use of methamphetamine were
256 C.-K. Chen and R. M. Murray

6.0

Mean PSST
5.5

5.0

4.5

4.0

3.5
MNP MIP-B MIP-P

Fig. 14.1. Premorbid schizoid/schizotypal traits between methamphetamine users with and without
psychosis. MNP, methamphetamine users with no experience of psychotic symptom; MIP-B,
subjects with brief methamphetamine psychosis; MIP-P, subjects with prolonged metham-
phetamine psychosis; PSST, score of Premorbid Schizoid and Schizotypal Traits.

excluded. Compared with their non-psychotic counterparts, those with metham-

phetamine psychosis had a significantly higher mean PSST score (Chen et al., 2003).
Indeed, we found a linear correlation between premorbid schizoid/schizotypal
personality traits and liability to psychosis after methamphetamine abuse
(Fig. 14.1). If schizotypy is one phenotypic expression of the familial–genetic
liability to schizophrenia or other psychotic disorders, a hypothesis can be pro-
posed that a continuum of vulnerability to psychosis accounts, at least to a cer-
tain degree, for individual differences in liability to drug-induced psychosis. In
addition, it is reasonable to speculate that the patients with drug-induced psy-
chosis, particularly if prolonged, may share some common vulnerability with
schizophrenia.
Of course, one must remember that childhood problem behaviors may also pre-
cede substance abuse (Tarter et al., 1990). Boyle et al. (1992) concluded that conduct
disorder significantly predicts later abuse of marijuana and hard drugs. Many other
studies have reported a link between childhood conduct disorder and antisocial
behaviors on the one hand and later substance abuse on the other. However, no
such link with childhood conduct disorder has been reported for schizophrenia,
and the deviations that precede substance abuse appear different from those that
precede psychotic disorders. Indeed, there is some evidence that individuals with
257 Drug abuse and development of schizophrenia

schizophrenia and comorbid substance abuse have better premorbid adjustment

levels than those with schizophrenia only (Arndt et al., 1992).
A possible explanation for the latter finding is that those schizophrenia patients
with marked childhood impairment are less likely to have acquired the social skills
and initiative necessary first to expose themselves to drug use and then to be able
to afford the cost of prolonged consumption of drugs, which are usually expensive.
Alternatively, it could be that those with little or no neurodevelopmental impair-
ment are likely to need more precipitating stressors such as drug abuse before
becoming psychotic, while those with significant premorbid deviation require little
or no additional stress to develop frank psychosis.

Familial predisposition
There have been only a few competent studies of the relationship between familial
risk of psychosis and the occurrence and duration of drug-related psychoses. Tsuang
et al. (1982) revealed that the familial risks of schizophrenia and affective disorder
were greater in drug abusers with prolonged psychosis than those without psychosis
or with psychosis of short duration. McGuire et al. (1995) compared the lifetime
morbid risk of psychiatric disorder among the first-degree relatives of patients
admitted with acute psychosis who were cannabis positive on urinary screening
with that of psychotic controls who screened negatively for all substances. Patients
with cannabis-associated psychosis had a significantly increased familial morbid
risk for schizophrenia. This might be taken to imply that cannabis abuse acts on
genetic predisposition to trigger psychosis.
In an early study, Tatetsu et al. (1956) reported that the prevalence of schizophre-
nia in the families of methamphetamine psychotics (4.7% for siblings, 3.4% for
parents) was lower than that for schizophrenia patients (13.1% for siblings, 10.1%
for parents) but higher than that for normal controls (0.6% for siblings, 0.25 for
parents). Their interpretation was that methamphetamine psychosis is not geneti-
cally identical to schizophrenia, and that the higher prevalence of schizophrenia in
families of patients with methamphetamine psychosis than in families of normal
controls may have resulted from a small number of schizophrenia patients being
included in the group of methamphetamine psychotics. An alternative formulation
is that the psychosis predisposition of methamphetamine psychotics falls between
that of schizophrenia patients and that of the normal population.
We assessed 445 methamphetamine users with the Diagnostic Interview for
Genetic Studies and the Family Interview for Genetic Study. Familial morbid risk
for psychotic disorders and other mental illnesses of first-degree relatives was com-
pared after age correction between the methamphetamine users with a lifetime
diagnosis of methamphetamine psychosis and those without. This study obtained
psychiatric information for 1983 first-degree relatives of the methamphetamine
258 C.-K. Chen and R. M. Murray

users. The relatives of those with methamphetamine psychosis had higher morbid
risks for schizophrenia than those of methamphetamine abusers without psychosis
(3.1% versus 0.6%). The familial morbid risk for schizophrenia was even higher
(6.5%) among those with prolonged methamphetamine psychosis.
Therefore, it appears that familial loading is an important factor in determining
not only the occurrence of psychosis associated with drug abuse but also its dura-
tion. Furthermore, these findings are consistent with the notion of an underlying
continuum of genetic liability that has schizophrenia as only one of its possible
outcomes (Tsuang et al., 2001; van Os et al., 1998).

Molecular genetic studies of substance-associated psychosis

Dopamine receptor genes
Numerous studies have looked for genes associated with schizophrenia, and many
others for possible associations with substance-use disorders, but until recently
no unambiguous findings had been found (but see Harrison and Owen, 2003).
Dopamine receptor (DRD) genes have been much studied in both conditions.

Dopamine D2 receptor gene

Persico et al. (1996) reported that polysubstance users with histories of heavy daily
preferential psychostimulant use more often displayed one or two copies of the TaqI
A1 and B1 markers at the dopamine D2 receptor (DRD2) locus. Ralph et al. (1999)
reported that DRD2 is essential for the disruption of prepulse inhibition produced
by amphetamine in mice. Similar disruptions in prepulse inhibition have, of course,
also been reported in schizophrenia (Braff et al., 1992).

Dopamine D3 receptor gene

DRD3 is reported to control the expression of sensitization to drugs of abuse (Guillin
et al., 2001) and to modulate drug-cue controlled cocaine-seeking behavior (Pilla
et al., 1999). Staley and Mash (1996) reported that elevated levels of DRD3 in
the reward circuitry of the brain are associated with chronic cocaine abuse. DRD3
expression is controlled by brain-derived neurotropic factor (Guillin et al., 2001),
and interestingly, elevated levels of this factor have been found in the anterior
cingulate cortex and the hippocampus of patients with schizophrenia (Takahashi
et al., 2000). There are isolated reports of an association of DRD3 polymorphism
with substance abuse in schizophrenia (Krebs et al., 1998) and with opiate depend-
ence (Duaux et al., 1998).

Dopamine D4 receptor gene

Substance abuse is associated with novelty seeking, a heritable human personality
trait that has been reported as associated with the seven-repeat variant of the DRD4
variable number of tandem repeats (VNTR) (Ebstein et al., 1996). Kotler et al.
259 Drug abuse and development of schizophrenia

(1997), therefore, analysed the DRD4 VNTR in opiate-dependent subjects from

Israel and found a significant excess of the seven-repeat allele. Other studies have
failed to replicate the association of DRD4 with substance abuse (Franke et al., 2000;
Li et al., 2000); however, recently, we have found marginally significant differences
in the DRD4 exon III VNTR (C. K. Chen et al., unpublished data) between 445
methamphetamine abusers and 416 normal controls.
Other genes
In the sample of 445 methamphetamine abusers, an association was noted between
methampthetamine abuse and the allele giving rise to a valine or methionine at one
position in catechol-O-methyltransferase (COMT) (T. Li et al., unpublished data).
This is interesting as individuals with the Val/Val genotype break down dopamine
in the frontal cortex more rapidly than those with the Met/Met genotype, and
there is experimental evidence that those with the former genotype experience
fewer adverse reactions to amphetamine than those with the latter. One might,
therefore, speculate that, among individuals who try methamphetamine, those
with the Met/Met genotype may be more likely to experience adverse reactions
and, therefore, not persist in taking the drug.
We did not find genetic differences between methamphetamine abusers with and
without psychosis: indeed, this is the general finding so far. However, Kikuchi et al.
(2001) found significant differences in both the DRD1 and the DRD5 locus in 107
methamphetamine users (96 with methamphetamine psychosis) compared with
192 healthy controls; it is not clear whether these polymorphisms are associated
with liability to psychosis or to methamphetamine abuse. Using the same sample,
Ujike et al. (2003) genotyped four types of dopamine transporter gene polymor-
phism. There were no significant differences in these four polymorphisms between
methamphetamine users and controls. However, patients with prolonged metham-
phetamine psychosis (lasting more than 1 month after therapy) showed significant
excess in the non-10-repeat alleles of the VNTR in the 3 untranslated region of
the dopamine transporter gene compared with those with brief methamphetamine
psychosis.
Cubells et al. (2000) found an association of cocaine-induced paranoia and a
haplotype associated with low dopamine β-hydroxylase activity, while low levels of
dopamine β-hydroxylase protein in plasma or cerebrospinal fluid were previously
reported to be associated with greater vulnerability to positive psychotic symptoms
in several psychiatric disorders (Ewing et al., 1977; Meltzer et al., 1976; Meyers
et al., 1999).
Summary
All the molecular genetic study results discussed above are preliminary; both the
positive and the negative findings need replication in larger samples. Nevertheless,
rapid advances in molecular techniques will certainly promote progress in the search
260 C.-K. Chen and R. M. Murray

for the candidate genes of these disorders. It is likely that certain drugs change
the expressions of genes related to neurotransmitter systems such as dopamine or
glutamic acid and also for transcription factors, cell proliferation, apoptosis, cell
adhesion, and the synapse (Ito, 2002). Hopefully, identifying these alterable gene
expressions may facilitate the discovery of the genes that underlie the development
of drug-associated psychosis.

Animal models of the interaction of drugs with perinatal brain damage

Patients with schizophrenia have been consistently shown to have experienced more
pregnancy and labour complications than have normal controls (Cantor-Graae
et al., 1994; Geddes and Lawrie, 1995; Lewis and Murray, 1987; O’Callaghan et al.,
1992); such complications are associated with damage to the hippocampus and
amygdala (Stefanis et al., 1999). Animal researchers have attempted to model such
effects. In one of the best known models, Lipska and her colleagues (1992) showed
that rats with neonatal damage of the ventral hippocampus display in adulthood
a variety of abnormalities said also to occur in schizophrenia; for example, they
show exaggerated behavioral response to amphetamine compared with controls.
This provides an animal model of the interaction of drugs and perinatal insults in
developing psychosis. Lipska and her colleagues (2002) concluded that transient loss
of ventral hippocampal function during a critical time in maturation of intracortical
connections permanently changes the development of neural circuits mediating
certain dopamine- and NMDA-related behaviors.
Other studies have revealed that animals who were subject to Cesarean section
birth (Vaillancourt and Boksa, 1998, 2000) or perinatal anoxia (Brake et al., 1997)
are more prone to develop a sensitized response to amphetamine. Moreover, Berger
et al. (2000) demonstrated the interaction of genetic predisposition, birth compli-
cations, and amphetamine-induced locomotion by revealing different patterns of
association between Cesarean section birth and amphetamine-induced locomotor
activity in strains of rats differing in their genetic constitution.

Dopamine sensitization
Behavioral sensitization is a phenomenon whereby exposure to a given stimu-
lus such as a drug or a stressor results in an enhanced response at subsequent
exposures (Wolf et al., 1993). Repeated cocaine use leads to progressive sensitiza-
tion, with short-term psychotic symptoms triggered more rapidly and after smaller
amounts of cocaine (Bartlett et al., 1997). The process of sensitization does not
necessarily require subsequent exposures to exactly the same stimulus that induced
the process. For example, Gorriti et al. (1999) showed that chronic treatment with
261 Drug abuse and development of schizophrenia

(−)-9 -tetrahydrocannabinol, the active ingredient of cannabis, resulted in sensi-

tization to the effects of d-amphetamine on locomotion, exploration and stereotyp-
ies in rats. In an analogous manner, methamphetamine psychosis shows a marked
tendency toward recurrence, which can be provoked by methamphetamine re-use
at low dose or a wide variety of stressors, even after long-term abstinence (Sato
et al., 1983). Interestingly, patients with schizophrenia are particularly sensitive
to the psychosis-inducing effects of amphetamine in doses that are subpsychoto-
genic in normals (Lieberman et al., 1987), and they show an exaggerated release of
dopamine in the ventral striatum following amphetamine challenge (Breier et al.,
1997; Laruelle, 2000).
It seems likely that dopamine plays a modulatory role on corticoventrostriatal
circuits, and that the psychotomimetic actions of dopaminergic agents and PCP
may result from interference with these circuits (Carlsson, 1988). Positron emission
tomography studies have suggested that dopamine transporter density in the cau-
date/putamen is reduced in methamphetamine users (McCann et al., 1998; Sekine
et al., 2001; Volkow et al., 2001). Sekine et al. (2001) showed that the reduction of
dopamine transporter may be long lasting, even if methamphetamine use ceases,
and that persistent psychiatric symptoms in methamphetamine users, including
psychotic symptoms, may be attributable to the reduction of dopamine transporter
density.
Dopamine sensitization has been proposed to play a critical role in the dis-
ease process in schizophrenia. For example, Lieberman and his colleagues (1997)
postulated that the symptoms of schizophrenia may be caused by deficits in neu-
ral regulation resulting in a pathological condition of neurochemical sensitiza-
tion analogous to the preclinical model of pharmacologically induced behavioral
sensitization. Dopamine sensitization may also explain the association between
schizophrenia and abuse of psychostimulants and cannabis. It seems likely that
familial predisposition to psychosis and perinatal insult increase the vulnerability
to, dopamine sensitization, and that individuals with these predispositions may be
more sensitive to the psychosis-inducing effects of certain drugs.

Conclusions
It is generally not disputed that the use of certain drugs, particularly cannabis
and psychostimulants, which have major effects on dopamine is associated with
schizophrenia. We regard the vulnerability hypothesis as the most likely explanation
for this association. Early environmental insults often compound the genotype for
psychosis to produce developmental deviance, but this is in itself sometimes not
sufficient to produce a psychotic illness. Later environmental factors, such as drug
abuse, then act on the underlying predisposition to precipitate frank psychosis.
262 C.-K. Chen and R. M. Murray

Recent research suggests that dopamine sensitization may underlie both craving
and the onset of drug-associated psychosis. Whether an individual develops psy-
chosis after drug abuse may be determined by various factors, such as drug-use
patterns, premorbid mental health, and the individual’s place on a continuum of
psychosis proneness. The liability for psychosis of each individual falls within a
normal distribution. A drug abuser with low liability to psychosis may use psy-
chostimulant drugs regularly for long periods without developing psychosis or, at
worst may have just brief psychotic symptoms. By comparison, an individual with
heavy liability may develop psychosis after using drugs only a few times and may
have a prolonged psychosis in spite of subsequently remaining abstinent from the
drug.

REFERENCES

Andreasson, S., Alleback, A., Engstrom, A., Rydberg, U. (1987). Cannabis and schizophrenia: a
longitudinal study of Swedish conscripts. Lancet i: 1483–1485.
Angrist, B. M., Gershon, S. (1970). The phenomenology of experimentally induced amphetamine
psychosis: preliminary observations. Biol Psychiatry 2: 95–107.
Arndt, S., Tyrrell, G., Flaum, M., Andreasen, N. C. (1992). Comorbidity of substance abuse and
schizophrenia: the role of premorbid adjustment. Psychol Med 22: 379–388.
Arseneault, L., Cannon, M., Poulton, R., et al. (2002). Cannabis use in adolescence and risk for
adult psychosis: longitudinal prospective study. Br Med J 325: 1212–1213.
Bartlett, E., Hallin, A., Chapman, B., Angrist, B. (1997). Selective sensitization to the psychosis-
inducing effects of cocaine: a possible marker for addiction relapse vulnerability? Neuropsy-
chopharmacology 16: 77–82.
Battaglia, M., Cavallini, M. C., Macciardi, F., Bellodi, L. (1997). The structure of DSM-III-R
schizotypal personality disorder diagnosed by direct interviews. Schizophr Bull 23: 83–92.
Bebbington, P., Wilkins, S., Jones, P., et al. (1993). Life events and psychosis. Initial results from
the Camberwell Collaborative Psychosis Study. Br J Psychiatry 162: 72–79.
Bell, D. S. (1965) Comparison of amphetamine psychosis and schizophrenia. Br J Psychiatry III:
701–707.
(1973). The experimental reproduction of amphetamine psychosis. Arch Gen Psychiatry 29:
35–40.
Berger, N., Vaillancourt, C., Boksa, P. (2000). Genetic factors modulate effects of C-section birth
on dopaminergic function in the rat. Neuroreport 11: 639–643.
Blanchard, J. J., Brown, S. A., Horan, W. P., Sherwood, A. R. (2000). Substance use disorders in
schizophrenia: review, integration, and a proposed model. Clin Psychol Rev 20: 207–234.
Boyle, M. H., Offord, D. R., Racine, Y. A., et al. (1992). Predicting substance use in late adolescence:
results from the Ontario child health study follow-up. Am J Psychiatry 149: 761–767.
Braff, D. L., Grillon, C., Geyer, M. A. (1992). Gating and habituation of the startle reflex in
schizophrenic patients. Arch Gen Psychiatry 49: 206–215.
263 Drug abuse and development of schizophrenia

Brake, W. G., Boksa, P., Gratton, A. (1997). Effects of perinatal anoxia on the acute locomotor
response to repeated amphetamine administration in adult rats. Psychopharmacology 133:
389–395.
Breakey, W. R., Goodell, H., Lorenz, P. C., McHugh, P. R. (1974). Hallucinogenic drugs as pre-
cipitants of schizophrenia. Psychol Med 4: 255–261.
Breier, A., Su, T. P., Saunders, R., et al. (1997). Schizophrenia is associated with elevated
amphetamine-induced synaptic dopamine concentrations: evidence from a novel positron
emission tomography method. Proc Natl Acad Sci USA 94: 2569–2574.
Camp, D. M., Browman, K. E., Robinson, T. E. (1994). The effects of methamphetamine and
cocaine on motor behavior and extracellular dopamine in the ventral striatum of Lewis
versus Fischer 344 rats. Brain Res 668: 180–193.
Cannon, T. D., Kaprio, J., Lonnqvist, J., Huttunen, M., Koskenvuo, M. (1998). The genetic
epidemiology of schizophrenia in a Finnish twin cohort. A population-based modeling
study. Arch Gen Psychiatry 55: 67–74.
Cantor-Graae, E., McNeil, T. F., Sjostrom, K., Nordstrom, L. G., Rosenlund, T. (1994). Obstetric
complications and their relationship to other etiological risk factors in schizophrenia. A
case–control study. J Nerv Ment Dis 182: 645–650.
Cantwell, R., Brewin, J., Glazebrook, C., et al. (1999). Prevalence of substance misuse in first-
episode psychosis. Br J Psychiatry 174: 150–153.
Cardno, A. G., Marshall, E. J., Coid, B., et al. (1999). Heritability estimates for psychotic disorders:
the Maudsley twin psychosis series. Arch of Gen Psychiatry 56: 162–168.
Carlsson, A. (1988). The current status of the dopamine hypothesis of schizophrenia. Neuropsy-
chopharmacology 1: 179–186.
Cassano, G. B., Pini, S., Saettoni, M., Rucci, P., Dell’Osso, L. (1998). Occurrence and clinical
correlates of psychiatric comorbidity in patients with psychotic disorders. J Clin Psychiatry
59: 60–68.
Chen, C. K., Lin, S. K., Sham, P., et al. (2003). Premorbid characteristics and comorbid-
ity of methamphetamine users with and without psychosis. Psychol Med 33: 1407–
1414.
Chou, P., Liou, M. Y., Lai, M. Y., Hsiao, M. L., Chang, H. J. (1999). Time trend of substance use
among adolescent students in Taiwan, 1991–1996. J Formos Med Assoc 98: 827–831.
Connell, P. H. (1958). Amphetamine Psychosis London: Chapman, Hall.
Crow, T. J., Johnstone, E. C., Deakin, J. F., Longden, A. (1976). Dopamine and schizophrenia.
Lancet ii: 563–566.
Cubells, J. F., Kranzler, H. R., McCance-Katz, E., et al. (2000). A haplotype at the DBH locus,
associated with low plasma dopamine beta-hydroxylase activity, also associates with cocaine-
induced paranoia. Mol Psychiatry 5: 56–63.
DeQuardo, J. R., Carpenter, C. F., Tandon, R. (1994). Patterns of substance abuse in schizophrenia:
nature and significance. J Psychiatr Res 28: 267–275.
Dixon, L., Haas, G., Weiden, P. J., Sweeney, J., Frances, A. J. (1991). Drug abuse in schizophrenic
patients: clinical correlates and reasons for use. Am J Psychiatry 148: 224–230.
Duaux, E., Gorwood, P., Griffon, N., et al. (1998). Homozygosity at the dopamine D3 receptor
gene is associated with opiate dependence. Mol Psychiatry 3: 333–336.
264 C.-K. Chen and R. M. Murray

Dumas, P., Saoud, M., Bouafia, S., et al. (2002). Cannabis use correlates with schizotypal person-
ality traits in healthy students. Psychiatry Res 109: 27–35.
Ebstein, R. P., Novick, O., Umansky, R., et al. (1996). Dopamine D4 receptor (D4DR) exon III
polymorphism associated with the human personality trait of novelty seeking. Nat Genet
12: 78–80.
Ewing, J. A., Mueller, R. A., Rouse, B. A., Silver, D. (1977). Low levels of dopamine beta-hydroxylase
and psychosis. Am J Psychiatry 134: 927–928.
Fischman, M. W., Schuster, C. R. (1982). Cocaine self-administration in humans. Fed Proc 41:
241–246.
Flaum, M., Schultz, S. K. (1996). When does amphetamine-induced psychosis become
schizophrenia? [Clinical conference] Am J Psychiatry 153: 812–815.
Franke, P., Nothen, M. M., Wang, T., et al. (2000). DRD4 exon III VNTR polymorphism-
susceptibility factor for heroin dependence? Results of a case–control and a family-based
association approach. Mol Psychiatry 5: 101–104.
Freud, S. (1885). On the general effects of cocaine. [Lecture before Psychiatric Union, 5 March,
1885. Translated to English.] Drug Depend 5: 15–17.
Geddes, J. R., Lawrie, S. M. (1995). Obstetric complications and schizophrenia: a meta-analysis.
Br J Psychiatry 167: 786–793.
Gold, M. S., Bowers, Jr. M. B. (1978). Neurobiological vulnerability to low-dose amphetamine
psychosis. Am J Psychiatry 135: 1546–1548.
Gorriti, M. A., de Rodriguez, F., Navarro, M., Palomo, T. (1999). Chronic (−)-delta9-
tetrahydrocannabinol treatment induces sensitization to the psychomotor effects of
amphetamine in rats. Eur J Pharmacol 365: 133–142.
Griffith, J. D., Cavanaugh, J., Oates, J. A. (1969). Schizophreniform psychosis induced by large-
dose administration of d-amphetamine. J Psychedel Drugs 2: 42–48.
Griffith, J. D., Cavanaugh, J., Held, J., Oates, J. A. (1972). Dextroamphetamine. Evaluation of
psychomimetic properties in man. Arch Gen Psychiatry 26: 97–100.
Guillin, O., Diaz, J., Carroll, P., et al. (2001). BDNF controls dopamine D3 receptor expression
and triggers behavioral sensitization. Nature 411: 86–89.
Harrison, P. J., Owen, M. J. (2003). Genes for schizophrenia? Recent findings and their patho-
physiological implications. Lancet 361: 417–419.
Imade, A. G. T., Ebie, J. C. (1991). A retrospective study of symptom patterns of cannabis-induced
psychosis. Acta Psychiatr Scand 83: 134–136.
Ito, C. (2002). Analysis of overall gene expression induced by amphetamine and phencyclidine:
novel targets for the treatment of drug psychosis and schizophrenia. Curr Pharm Des 8:
147–153.
Iwanami, A., Suga, I., Kaneko, T., Sugiyama, A., Nakatani, Y. (1994). P300 component of event-
related potentials in methamphetamine psychosis and schizophrenia. Prog Neuropsychophar-
macol Biol Psychiatry 18: 465–475.
Jacobs, M. R., Fehr, K. O., Cox, T. C., et al. (1987). Drugs and Drug Abuse. A Reference Text.
Toronto: Addiction Research Foundation.
Javitt, D. C., Zukin, S. Z. (1991). Recent advances in the phencyclidine model of schizophrenia.
Am J Psychiatry 148: 1301–1308.
265 Drug abuse and development of schizophrenia

Johnson, B. A., Smith, B. L., Taylor, P. (1988). Cannabis and schizophrenia. Lancet i: 593.
Jones, P., Rodgers, B., Murray, R., Marmot, M. (1994). Child developmental risk factors for adult
schizophrenia in the British 1946 birth cohort. Lancet 344: 1398–1402.
Jori, A., Garattini, S. (1973). Catecholamine metabolism and amphetamine effects on sensitive
and insensitive mice. In Frontiers in Catecholamine Research, ed. E. Usdin, S. H. Snyder. New
York: Pergamon Press, pp. 939–941.
Kendler, K. S., Diehl, S. R. (1993). The genetics of schizophrenia: a current, genetic-epidemiologic
perspective. Schizophr Bull 19: 261–285.
Khantzian, E. J. (1985). The self-medication hypothesis of addictive disorders: focus on heroin
and cocaine dependence. Am J Psychiatry 142: 1259–1264.
Kikuchi, K., Inada, T., Iijima, Y., et al. (2001). Association between dopamine D1 receptor fam-
ily (DRD1, DRD5) gene polymorphisms and methamphetamine psychosis. In Proceed-
ings of the International College of Neuro-Psychopharmacology Regional Meeting, Hiroshima,
p. 398.
Kleinman, J. E., Casanova, F. M., Jaskiv, G. E. (1988). The neuropathology of schizophrenia.
Schizophrenia Bulletin 14: 209–216.
Konuma, K. (1994). Use and abuse of amphetamines in Japan. In Amphetamine and its Analogs
pp. 415–435.
Kotler, M., Cohen, H., Segman, R., et al. (1997). Excess dopamine D4 receptor (D4DR) exon III
seven repeat allele in opioid-dependent subjects. Mol Psychiatry 2: 251–254.
Krebs, M. O., Sautel, F., Bourdel, M. C., et al. (1998). Dopamine D3 receptor gene variants and
substance abuse in schizophrenia. Mol Psychiatry 3: 337–341.
Kwapil, T. R. (1996). A longitudinal study of drug and alcohol use by psychosis-prone and
impulsive-nonconforming individuals. J Abnorm Psychol 105: 114–123.
Laruelle, M. (2000). The role of endogenous sensitization in the pathophysiology of schizophrenia:
implications from recent brain imaging studies. Brain Res Brain Res Rev 31: 371–384.
Lewis, S. W., Murray, R. M. (1987). Obstetric complications, neurodevelopmental deviance, and
risk of schizophrenia. J Psychiatr Res 21: 413–421.
Li, T., Zhu, Z. H., Liu, X., et al. (2000). Association analysis of polymorphisms in the DRD4 gene
and heroin abuse in Chinese subjects. Am J Med Genet 96: 616–621.
Lieberman, J. A., Kane, J. M., Alvir, J. (1987). Provocative tests with psychostimulant drugs in
schizophrenia. Psychopharmacology 91: 415–433.
Lieberman, J. A., Sheitman, B. B., Kinon, B. J. (1997). Neurochemical sensitization in the patho-
physiology of schizophrenia: deficits and dysfunction in neuronal regulation and plasticity.
Neuropsychopharmacology 17: 205–229.
Linszen, D. H., Dingemans, P. M., Lenior, M. E. (1994). Cannabis abuse and the course of recent-
onset schizophrenic disorders. Arch of Gen Psychiatry 51: 273–279.
Lipska, B. K., Jaskiw, G. E., Chrapusta, S., Karoum, F., Weinberger, D. R. (1992). Ibotenic acid
lesion of the ventral hippocampus differentially affects dopamine and its metabolites in the
nucleus accumbens and prefrontal cortex in the rat. Brain Res 585: 1–6.
Lipska, B. K., Halim, N. D., Segal, P. N., Weinberger, D. R. (2002). Effects of reversible inactivation
of the neonatal ventral hippocampus on behavior in the adult rat. J Neurosci 22: 2835–
2842.
266 C.-K. Chen and R. M. Murray

Longhurst, J. G. (1997). Cannabis and schizophrenia. Br J Psychiatry 171: 584–585.

Lopez, W., Jeste, D. V. (1997). Movement disorders and substance abuse. Psychiatr Serv 48: 634–
636.
Mathers, D. C., Ghodse, A. H., Caan, A. W., Scott, S. A. (1991). Cannabis use in a large sample
of acute psychiatric admissions. Br J Addict 86: 779–784.
McCann, U. D., Wong, D. F., Yokoi, F., et al. (1998). Reduced striatal dopamine transporter density
in abstinent methamphetamine and methcathinone users: evidence from positron emission
tomography studies with [11 C ]WIN-35,428. J Neurosci 18: 8417–8422.
McCreadie, R. G. (2002). Use of drugs, alcohol and tobacco by people with schizophrenia: case–
control study. Br J Psychiatry 181: 321–325.
McGuire, P. K., Jones, P., Harvey, I., et al. (1995). Morbid risk of schizophrenia for relatives of
patients with cannabis-associated psychosis. Schizophr Res 15: 277–281.
Meltzer, H. Y., Cho, H. W., Carroll, B. J., Russo, P. (1976). Serum dopamine-beta-hydroxylase
activity in the affective psychoses and schizophrenia. Decreased activity in unipolar psy-
chotically depressed patients. Arch of Gen Psychiatry 33: 585–591.
Menezes, P. R., Johnson, S., Thornicroft, G., et al. (1996). Drug and alcohol problems
among individuals with severe mental illness in south London. Br J Psychiatry 168: 612–
619.
Meyers, B. S., Alexopoulos, G. S., Kakuma, T., et al. (1999). Decreased dopamine beta-hydroxylase
activity in unipolar geriatric delusional depression. Biol Psychiatry, 45: 448–452.
Modestin, J., Nussbaumer, C., Angst, K., Scheidegger, P., Hell, D. (1997). Use of potentially
abusive psychotropic substances in psychiatric inpatients. Eur Arch Psychiatry Clin Neurosci
247: 146–153.
Mueser, K. T., Yarnold, P. R., Levinson, D. F., et al. (1990). Prevalence of substance abuse in
schizophrenia, demographic and clinical correlates. Schizophr Bull 6: 10–41.
Mueser, K. T., Drake, R. E., Wallach, M. A. (1998). Dual diagnosis: a review of etiological theories.
Addict Behav 23: 717–734.
Murray, R. M., Fearon, P. (1999). The developmental “risk factor” model of schizophrenia. J
Psychiatr Res 33: 497–499.
Murray, R. M., Grech, A., Phillips, P., Johnson, S. (2003). What is the relationship between
substance abuse and schizophrenia? In The Epidemiology of Schizophrenia, ed. R. M. Murray,
P. Jones, E. Susser, J. van Os, M. Cannon. Cambridge, UK: Cambridge University Press,
pp. 317–342.
Nakatani, Y., Yoshizawa, F., Yamada, H., et al. (1989). Methamphetamine psychosis in Japan: a
survey. Br J Addict 84: 1548–1549.
National Institute on Drug Abuse (1998). Methamphetamine abuse and addiction. (Publication
98–4210). Bethesda, MD: National Institutes of Health.
Negrete, J. C. (1989). Cannabis and schizophrenia. Br J Addiction 84: 349–351.
O’Callaghan, E., Gibson, T., Colohan, H. A., et al. (1992). Risk of schizophrenia in adults born
after obstetric complications and their association with early onset of illness: a controlled
study. Br Med J 305: 1256–1259.
Peralta, V., Cuesta, M. J. (1992). Influence of cannabis abuse on schizophrenic psychopathology.
Acta Psychiatr Scand 85: 127–130.
267 Drug abuse and development of schizophrenia

Persico, A. M., Bird, G., Gabbay, F. H., Uhl, G. R. (1996) D2 dopamine receptor gene TaqI A1 and
B1 restriction fragment length polymorphisms: enhanced frequencies in psychostimulant-
preferring polysubstance abusers. Biol Psychiatry 40: 776–784.
Pilla, M., Perachon, S., Sautel, F., et al. (1999). Selective inhibition of cocaine-seeking behavior
by a partial dopamine D3 receptor agonist. Nature 400: 371–375.
Post, R. M. (1975). Cocaine psychosis: a continuum model. Am J Psychiatry 132: 225–231.
Ralph, R. J., Varty, G. B., Kelly, M. A., et al. (1999). The dopamine D2 , but not D3 or D4 , receptor
subtype is essential for the disruption of prepulse inhibition produced by amphetamine in
mice. J Neurosci 19: 4627–4633.
Regier, D. A., Farmer, M. E., Rae, D. S., et al. (1990). Comorbidity of mental disorders with alcohol
and other drug abuse. Results from the Epidemiologic Catchment Area (ECA) Study. J Am
Med Assoc 264: 2511–2518.
Rolfe, M., Tang, C. M., Sabally, S., et al. (1993). Psychosis and cannabis abuse in The Gambia. A
case–control study. Br J Psychiatry 163: 798–801.
Sato, M., Chen, C. C., Akiyama, K., Otsuki, S. (1983). Acute exacerbation of paranoid psychotic
state after long-term abstinence in patients with previous methamphetamine psychosis. Biol
Psychiatry 18: 429–440.
Schneier, F. R., Siris, S. G. (1987). A review of psychoactive substance use and abuse in schizo-
phrenia. Patterns of drug choice. J Nerv Ment Dis 175: 641–652.
Sekine, Y., Iyo, M., Ouchi, Y., et al. (2001). Methamphetamine-related psychiatric symptoms
and reduced brain dopamine transporters studied with PET. Am J Psychiatry 158: 1206–
1214.
Selzer, J. A., Lieberman, J. A. (1993). Schizophrenia and substance abuse. Psychiatr Clin North
Am 16: 401–412.
Slaby, A. E. (1991). Dual diagnosis: fact or fiction? In Dual Diagnosis in Substance Abuse, ed. M.
S. Gold, A. E. Slaby. New York: Marcel Dekker, pp. 3–28.
Snell, L. D., Yi, S.-J., Johnson, K. M. (1988). Comparison of the effects of MK-801 and phencycli-
dine on catecholamine uptake and NMDA-induced norepinephrine release. Eur J Pharmacol
145: 223–226.
Snyder, S. H. (1973). Amphetamine psychosis: a “model” schizophrenia mediated by cate-
cholamines. Am J Psychiatry 130: 61–67.
Soyka, M., Albus, M., Kathmann, N., et al. (1993). Prevalence of alcohol and drug abuse in
schizophrenic inpatients. Eur Arch Psychiatry Clin Neurosci 242: 362–372.
Staley, J. K., Mash, D. C. (1996). Adaptive increase in D3 dopamine receptors in the brain reward
circuits of human cocaine fatalities. J Neurosci 16: 6100–6106.
Stefanis, N., Frangou, S., Yakeley, J., et al. (1999). Hippocampal volume reduction in schizophre-
nia: effects of genetic risk and pregnancy and birth complications. Biol Psychiatry 46:
697–702.
Strassman, R. J. (1984). Adverse reactions to psychedelic drugs. A review of the literature. J Nerv
Ment Dis 172: 577–595.
Suwaki, H., Fukui, S., Konuma, K. (1997) Methamphetamine abuse in Japan: its 45 year history
and the current situation. In Amphetamine Misuse: International Perspective on Current
Trends, ed. H. Klee. Amsterdam: Harwood Academic, pp. 199–214.
268 C.-K. Chen and R. M. Murray

Takahashi, M., Shirakawa, O., Toyooka, K., et al. (2000). Abnormal expression of brain-derived
neurotrophic factor and its receptor in the corticolimbic system of schizophrenic patients.
Mol Psychiatry 5: 293–300.
Tarter, R. E., Laird, S. B., Kabene, M., Bukstein, O., Kaminer, Y. (1990). Drug abuse severity in
adolescents is associated with magnitude of deviation in temperament traits. Br J Addiction
85: 1501–1504.
Tatetsu, S. (1963). Methamphetamine. Folia Psychiatr Neurol Jap Suppl 7: 377–380.
Tatetsu, S., Goto, A., Fujiwara, T. (1956). The Methamphetamine Psychosis. Tokyo: Igakushoin.
Thomas, H. (1993). Psychiatric symptoms in cannabis users. Br J Psychiatry 163: 141–149.
Thornicroft, G. (1990). Cannabis and psychosis: is there epidemiological evidence for an associ-
ation? Br J Psychiatry 157: 25–33.
Tsuang, M. T., Simpson, J. C., Kronfol, Z. (1982). Subtypes of drug abuse with psychosis. Demo-
graphic characteristics, clinical features, and family history. Arch Gen Psychiatry 39: 141–147.
Tsuang, M. T., Stone, W. S., Faraone, S. V. (2001). Genes, environment and schizophrenia. Br J
Psychiatry 178: 18–24.
Ujike, H., Harano, M., Inada, T., et al. (2003). Nine- or fewer repeat alleles in VNTR poly-
morphism of the dopamine transporter gene is a strong risk factor for prolonged metham-
phetamine psychosis. Pharmacogenom J 3: 242–247.
United Nations International Drug Control Programme (1997). World Drug Report. New York:
Oxford University Press.
Vaillancourt, C., Boksa, P. (1998). Caesarean section birth with general anesthesia increases
dopamine-mediated behavior in the adult rat. Neuroreport 9: 2953–2959.
(2000). Birth insult alters dopamine-mediated behavior in a precocial species, the guinea pig.
Implications for schizophrenia. Neuropsychopharmacology 23: 654–666.
van Os, J., Jones, P., Sham, P., Bebbington, P., Murray, R. M. (1998). Risk factors for onset and
persistence of psychosis. Soc Psychiatry Psychiatr Epidemiol 33: 596–605.
van Os, J., Bak, M., Hanssen, M., et al. (2002). Cannabis use and psychosis: a longitudinal
population-based study. Am J Epidemiol 156: 319–327.
Vardy, M. M., Kay, S. R. (1983). LSD psychosis or LSD-induced schizophrenia? A multimethod
inquiry. Arch Gen Psychiatry 40: 877–883.
Verdoux, H., Mury, M., Besancon, G., Bourgeois, M. (1996a). [Comparative study of substance
dependence comorbidity in bipolar, schizophrenic and schizoaffective disorders.] Encephale
22: 95–101.
Verdoux, H., van Os, J., Sham, P., et al. (1996b). Does familiality predispose to both emergence
and persistence of psychosis? A follow-up study. Br J Psychiatry 168: 620–626. [Published
erratum appears in Br J Psychiatry (1996). 169: 116].
Verdoux, H., Gindre, C., Sorbara, F., Tournier, M., Swendsen, J. D. (2002). Cannabis use and the
expression of psychosis vulnerability in daily life. Schizophr Res 53 (Suppl.): 225.
Volkow, N. D., Chang, L., Wang, G. J., et al. (2001). Association of dopamine transporter
reduction with psychomotor impairment in methamphetamine abusers. Am J Psychiatry
158: 377–382.
Williams, J. H., Wellman, N. A., Rawlins, J. N. (1996). Cannabis use correlates with schizotypy
in healthy people. Addiction 91: 869–877.
269 Drug abuse and development of schizophrenia

Wolf, M. E., White, F. J., Nassar, R., Brooderson, R. J., Khansa, M. R. (1993). Differential develop-
ment of autoreceptor subsensitivity and enhanced dopamine release during amphetamine
sensitization. J Pharmacol Exp Ther 264: 249–255.
Zammit, S., Allebeck, P., Andreasson, S., Lundberg, I., Lewis, G. (2002). Self reported cannabis
use as a risk factor for schizophrenia in Swedish conscripts of 1969: historical cohort study.
Br Med J 325: 1199.
Part III

Pathophysiology
 15

Developmental dysregulation of the

dopamine system and the pathophysiology
of schizophrenia
Anthony A. Grace
University of Pittsburgh, Pittsburgh, USA

Evidence suggests that the dopamine (DA) system plays a key role in the pathophys-
iology and treatment of schizophrenia. Therefore, drugs that increase DA trans-
mission are known to exacerbate this disorder and mimic paranoid psychosis in
normal individuals (Angrist et al., 1974, 1980), and drugs that are effective antipsy-
chotic agents all have DA receptor-blocking properties in common (Creese et al.,
1976; Grace et al., 1997; Seeman, 1987). However, despite substantial effort, there
has been little direct evidence for a pathology within at least the subcortical DA
system (Beuger et al., 1996; Post et al., 1975; Van Kammen et al., 1986). Nonetheless,
schizophrenia patients show significantly higher levels of DA release in the stria-
tum in response to low doses of amphetamine, particularly in patients in whom the
amphetamine leads to an exacerbation of their psychosis (Laruelle, 2000; Laruelle
and Abi-Dargham, 1999). This has led to the suggestion that the DA system may
be abnormally regulated in the brains of schizophrenia patients, leading to the
psychopathological state. One system in particular that has attracted attention as a
potential source of this dysregulatory influence is the cortical glutamatergic system.
This is based on evidence that drugs that interfere with glutamate transmission exert
a symptom-specific exacerbation of schizophrenia in susceptible patients and will
mimic the entire spectrum of schizophrenia symptoms in some normal individuals
(Goff and Coyle, 2001; Jentsch and Roth, 1999; Olney and Farber, 1995). There is
also evidence for a pathology within the limbic cortex of schizophrenia patients
(Weinberger, 1999).
Substantial evidence exists to suggest that the pathological processes contributing
to schizophrenia occur early in life, supporting the contention that schizophrenia
is a developmental disorder. Thus, the incidence of schizophrenia is higher in indi-
viduals with a genetic predisposition to the disorder, with the rate of occurrence

Neurodevelopment and Schizophrenia, ed. Matcheri S. Keshavan et al. Published by Cambridge University
Press.

C Cambridge University Press 2004.

273
274 A. A. Grace

being proportional to the relative amount of shared genetic background. However,

unlike true genetic diseases such as Huntington’s disease, schizophrenia does not
appear to be completely genetically determined. Even in identical twins, which
should have identical genetic make-up, the concordance rate for schizophrenia is
only 50% (Kendler, 1983). This is approximately 50 times higher than that found
in the general population but still is not supportive of a fully expressed genotype.
In addition, it is known that birth insults will often correlate with higher incidence
of schizophrenia in the offspring. For example, there is a significantly higher inci-
dence of births of individuals who later develop schizophrenia where the mother
was exposed to areas of famine or severe influenza infections during the second
trimester (McDonald and Murray, 2000; Pilowsky et al., 1993; Susser et al., 1996).
Moreover, the incidence of birth of schizophrenia patients is higher during spring
months, which corresponds to the sensitive second trimester occurring during the
winter months when illness is more common (Torrey et al., 1997). Together, this
evidence suggests that an insult during the second trimester, which is a time when
limbic cortices are developing, combined with a genetic susceptibility may lead to a
predisposition for schizophrenia illness (Bayer et al., 1999; Lewis and Levitt, 2002;
Weinberger, 1996).

The prefrontal cortex

Role in the pathophysiology of schizophrenia
One question that has been difficult to fit into a pathological model of schizophrenia
relates to the age of onset of symptoms. From the above, it appears that the trigger
for the illness is present during gestation. However, although the prodromal state
may be present for variable intervals prior to the onset of psychosis, the first episode
of psychotic symptoms typically occurs during late adolescence or early adulthood
(DeLisi, 1992). One possibility is that the developmental insult causes the limbic
system to reorganize in a type of temporary compensatory manner, which subse-
quently fails to perform when the subject reaches adolescence (Grace and Moore,
1998). One focus for a source of this delayed development is the prefrontal cortex,
based in part on the protracted development of this region, which extends well into
adolescence (Lewis, 1997). Subjects at high risk for the development of schizophre-
nia as well as first-degree relatives of schizophrenia patients often show disturbed
performance in tasks that measure prefrontal cortical activity, typically referred to
as tests of executive function or working memory (Cohen and Servan-Schreiber,
1992; Goldman-Rakic, 1999). Performance of working memory tasks are associated
with activation of the prefrontal cortex (Goldman-Rakic, 1995). Moreover, working
memory within the prefrontal cortex is believed to depend on its DA innervation
(Goldman-Rakic, 1996), which has been shown to be pathologically decreased by
275 Developmental dysregulation of the dopamine system

as much as 33% in this structure in the brains of schizophrenia patients (Akil et al.,
1999). Together, these data suggest that pathology within the prefrontal cortex may
be an etiological factor in the onset of schizophrenia. However, although prefrontal
cortical lesions in adults may mimic some of the negative symptoms of schizophre-
nia, there is little evidence that such an insult will lead to psychosis.

Role in stress and emotional reactivity

A deficit in prefrontal cortical function would be expected to produce several spe-
cific disturbances in brain function. In particular, our studies have shown that the
prefrontal cortex exerts an important suppressive role on the basolateral amygdala
(Rosenkranz and Grace, 1999, 2001), a structure known to be important in the
expression of emotion (LeDoux, 2000). Stimulation of the prefrontal cortex was
found to activate interneurons preferentially within the basolateral amygdala; as a
consequence, it exerted a suppressive effect over excitatory afferents arising from the
sensory association cortex (Rosenkranz and Grace, 2001, 2002a). One interpreta-
tion of this finding is that the prefrontal cortex will suppress emotional responses to
stimuli that experience has determined to be benign or non-threatening in nature.
The prefrontal cortex was also found to suppress conditioned responses to noxious
stimuli at the cellular level (Rosenkranz and Grace, 2002b). The amygdala also
appears to play a role in the reaction to chronic stressors. Via the projections from
the central nucleus of the amygdala to the hypothalamus (Pitkanen et al., 1997), the
amygdala is positioned to exert influence over the hypothalamic–pituitary–adrenal
(HPA) axis, which among other things would provide a link between stress-induced
emotional responses and release of stress hormones such as glucocorticoids. Indeed,
it is likely that the amygdala is involved in a variety of psychiatric disorders rang-
ing from depression to attention-deficit hyperactivity disorder (Breier et al., 1992;
Drevets, 1999; Goddard and Charney, 1997; Grace, 2000; Lawrie and Abukmeil,
1998; Ninan, 1999; Tebartz van Elst et al., 1999). This structure is also capa-
ble of showing plasticity in response to stress, including conditioned associations
(McKernan and Shinnick-Gallagher, 1997; Ono et al., 1995; Pare and Collins, 2000;
Rogan et al., 1997; Rosenkranz and Grace, 2002b).
Exposure to stress is known to activate several central systems, principally the
noradrenergic system of the locus coeruleus. Upon presentation of a stressor, there
is an increase in locus coeruleus neuron firing (Abercrombie and Jacobs, 1987),
norepinephrine levels and metabolites at postsynaptic sites (Abercrombie et al.,
1988; Shanks et al., 1991; Thierry et al., 1968), and the capacity to synthesize nor-
epinephrine (Chang et al., 2000; Sabban and Kvetnansky, 2001; Serova et al., 1999).
Corticotropin-releasing hormone (CRH) is the primary regulator of adrenocor-
ticotropic hormone release from the pituitary during stress (Vale et al., 1981),
although evidence has demonstrated a prominent extrahypothalamic role for CRH
276 A. A. Grace

as well (Valentino et al., 1993). Therefore, intraventricular injection of CRH is

known to elicit numerous stress-like behaviors (Dunn and Berridge, 1987; Kalin,
1985; Sherman and Kalin, 1986), and studies have shown that CRH injected intra-
ventricularly or into the locus coeruleus will activate noradrenergic neuron fir-
ing and norepinephrine release (Curtis et al., 1997; Finlay et al., 1997; Page and
Abercrombie, 1999; Smagin et al., 1995; Valentino and Foote, 1986; Valentino et al.,
1983, 1993). In addition, administration of CRH antagonists will attenuate stress-
evoked increases in locus coeruleus neuronal activity and norepinephrine release
(Emoto et al., 1993; Lechner et al., 1997; Smagin et al., 1997; Valentino and Wehby,
1988; Valentino et al., 1991) and behavioral responses (Smagin et al., 1996) to select
stressors.
The amygdala is known to provide a CRH input to the locus coeruleus, both
directly and indirectly, which synapses preferentially in the pericoerulear region, an
area that contains dendritic processes of locus coeruleus neurons (Van Bockstaele
et al., 1996, 1998, 1999). This region is also the recipient zone for the afferents
from the prefrontal cortex to the locus coeruleus (Ennis et al., 1998; Zhu and
Aston-Jones, 1996). Based on this anatomical arrangement, it has been suggested
that direct CRH and glutamate inputs to the locus coeruleus proper from regions
such as the perigigantocelluaris region and Barrington’s nucleus (Aston-Jones et
al., 1986; Valentino et al., 1996) act to convey messages that are of immediate
physiological survival value (i.e. the “systemic” stressors; (Herman and Cullinan,
1997), whereas inputs to the pericoerulear region, which would contact major
dendrites of locus coeruleus neurons (Ennis et al., 1998), would act more in a
modulatory role (Herman and Cullinan, 1997). In this way, inputs to the dendrites
could, in turn, modulate the response of locus coeruleus neurons to the afferents
that carry physiological information (e.g., the perigigantocelluaris) onto processes
more proximal to the locus coeruleus somata. Such a modulatory function would
be expected to subserve the types of emotion or cognition activity associated with
psychopathology or exposure to chronic stress or drug abuse.
Our studies show that rats exposed to chronic cold stress exhibit increased
responses to acute stressors within the basolateral amygdala (Correll et al., 2001) and
the locus coeruleus (Jedema and Grace, 2003; Mana and Grace, 1997; Ramsooksingh
et al., 2002). In addition, following chronic stress exposure, the prefrontal cortex
has less inhibitory action over the locus coeruleus (Ramsooksingh et al., 2001).
When considered in the light of the role of the prefrontal cortex in suppress-
ing amygdala responses (Rosenkranz and Grace, 1999, 2001), it is clear that the
prefrontal cortex plays an essential role in regulating emotional expression, and
that disruption of this prefrontal cortical influence would be expected to result
in a pathological increase in emotional responsivity of the subject. The prefrontal
cortex has also been reported to play a role in other aspects of the response to
277 Developmental dysregulation of the dopamine system

stress. Studies have shown that exposure to a stressor leads to increased DA release
in the prefrontal cortex, with substantially smaller increases in DA within subcor-
tical structures (Abercrombie et al., 1989; Imperato et al., 1989). However, if the
DA innervation of the prefrontal cortex is disrupted, two consequences are noted.
First, there is a decrease in the spontaneous activity of ventral tegmental DA neu-
rons (Harden et al., 1998); second, there is an increase in the stress-induced DA
response within subcortical limbic structures such as the nucleus accumbens (King
et al., 1997).
Together, these studies suggest that, in the absence of a normally functioning
prefrontal cortex, the system would over-react to potentially anxiogenic stimuli,
resulting in increased levels of stress. Indeed, such a model is consistent with obser-
vations of high-risk individuals by E. C. Johnstone (personal communication), who
found that those adolescents who eventually go on to develop schizophrenia typ-
ically show a higher reactivity to stress compared with those who do not develop
psychosis. What consequence would such an increase in stress produce? One region
that may be of particular significance in this regard is the ventral hippocampus.
One pathological finding that has been reported with some consistency is a
decrease in ventral hippocampal volume in schizophrenia patients, particularly
when comparing afflicted versus normal monozygotic twins (Suddath et al., 1990).
A focus on the ventral hippocampus is also supported by prospective studies, in
which an increase in hippocampal volume is observed in high-risk individuals
prior to the onset of psychosis (Pantelis et al., 2003; Phillips et al., 2002). Such
a condition suggests that the ventral hippocampus may undergo some type of
pathological alteration during the first episode that would lead to the expression
of psychosis. A potential etiological factor involving developmental disruption of
the ventral hippocampus or limbic cortex has been exploited in the advancement
of animal models of this disorder.

Developmental models of schizophrenia

An impediment to advancing our understanding of the relationship between limbic
pathophysiology and the etiology of schizophrenia is the lack of an appropriate ani-
mal model for this disorder. Because schizophrenia is a uniquely human affliction,
it is difficult to approximate its complex psychological and cognitive components.
Furthermore, neuroanatomical studies of schizophrenia brains have failed to yield
concrete pathological correlates of this disorder. Instead, pathological changes are
reported to be rather variable and diffuse, involving mainly limbic and frontal
cortical regions (Harrison, 1999). However, as reviewed above, several epidemio-
logical results suggest that gestational factors may contribute to schizophrenia. This
has led to the suggestion that it may arise from an interaction between a genetic
278 A. A. Grace

predisposition combined with a developmental insult (Bayer et al., 1999; Wein-

berger, 1996).
Recent studies have suggested the presence of an early disruption in cortical
development in schizophrenia (Jones, 1997; Weinberger and Lipska, 1995). Such a
pathology is thought to lead to the widespread but subtle pathological changes in
the cerebral cortex in schizophrenia, which include a decrease in cortical thickness
(Harrison, 1999; McCarley et al., 1999) with no significant alteration in total neuron
number (Thune and Pakkenberg, 2000) and the absence of evidence for neurode-
generation (Arnold et al., 1998). This is consistent with the reported increase in
neuronal density accompanied by reduced neuropil (e.g. Harrison, 1999; Selemon
et al., 1995). Studies have also reported decreases in structures related to synaptic
transmission (Eastwood et al., 2000; Garey et al., 1998; Glantz and Lewis, 1997;
Mirnics et al., 2000). Such alterations are most commonly reported to be present in
medial temporal and frontal cortices (Harrison, 1999; McCarley et al., 1999). More-
over, the behavioral symptoms associated with schizophrenia are most often corre-
lated with pathology within limbic regions of the frontal lobes and their associated
target structures (Arnold, 1997; Bilder et al., 1995; Liddle et al., 1992; Tamminga
et al., 1992). Studies have also shown that the mesolimbic DA system is hyperreac-
tive in schizophrenia (Laruelle, 2000; Laruelle and Abi-Dargham, 1999), a state that
has been proposed to be secondary to pathology within limbic cortex (Bertolino
et al., 1999; Grace, 1991; Weinberger and Lipska, 1995).
One approach to examining the consequences of developmentally related patho-
logical alterations in an animal model was proposed by Lipska and Weinberger
(Lipska et al., 1993; Weinberger and Lipska, 1995). These investigators showed that
lesions created in the ventral subiculum of the hippocampus in neonatal rats led to
behaviors in the adult animal consistent with hyperactivation within the mesolim-
bic DA system (Lipska et al., 1993). Similarly, in primates, early temporal lobe lesions
were found to result in alterations in metabolism in the prefrontal cortex and con-
sequent abnormalities in striatal DA that are analogous to the changes observed in
the rat (Lipska and Weinberger, 2000; Saunders et al., 1998). These studies showed
that an alteration in a specific neuronal circuit in the neonatal animal could lead to
pathology within limbic circuits in the adult that is consistent with what would be
expected in schizophrenia; however, the pathology produced by this manipulation
was markedly dissimilar from the more subtle anatomical alterations observed in
the brain of the schizophrenia patient (Harrison, 1999), which presumably was the
result of genetically derived pathological brain development.
It is unclear how one may be able to duplicate a human genetically based alter-
ation in brain development using an animal model. However, we have found
that administration of a developmental mitotoxin methyl azoxymethanol acetate
(MAM) (Nagata and Matsumoto, 1969) to pregnant dams may approximate this
279 Developmental dysregulation of the dopamine system

condition. Administration of MAM to the pregnant dam at gestational day 17,

which is when the proliferation and migration of neurons within the limbic cortex
is occurring (Bayer and Altman, 1991), disrupted the normal development of these
limbic circuits. The resultant adult offspring demonstrated a number of the anatom-
ical and behavioral characteristics that would be predicted for an animal model of
this disorder. Anatomically, there is a selective decrease (10–30%) in the thickness
of the paralimbic and limbic cortical regions but no change in anterior sensori-
motor cortex or subcortical regions (Moore and Grace, 1997; Moore et al., 2001a),
which is in concert with that described in schizophrenia, allowing, of course, for
the anatomical differences between cortical regions when determining homologies
between rat and human brain. Behaviorally, these rats showed spontaneously occur-
ring orofacial dyskinesias and a hyperresponsiveness to novelty (Ghajarnia et al.,
1998; Moore et al., 2001a); such behaviors are typically associated with lesions of
the frontal lobes (Edwards, 1970; Gunne et al., 1982; Waddington, 1990) or admin-
istration of glutamate antagonists to the ventral striatum (Kelley and Delfs, 1994;
Kelley and Swanson, 1997). There was also an increased response to phenyclidine
and to amphetamine, and the rats also showed deficits in prepulse inhibition of
the startle response (Ghajarnia et al., 1998): a characteristic that has been found in
human schizophrenia subjects (Braff et al., 1992).
In physiological studies, the MAM-treated rats exhibited a loss of hippocampal-
driven bistable activity in the nucleus accumbens. As a result, the hippocampus
no longer exerted its normal gating influence over the accumbens (O’Donnell and
Grace, 1995). Moreover, we have found that the modulatory actions of DA in the
prefrontal cortex of MAM-treated rats is significantly attenuated (Lavin and Grace,
1997). As a consequence, the amygdala exhibited an abnormally large influence
within the accumbens. Such a condition suggests that, in this animal model, emotion
plays a much greater role in guiding behavior when compared with goal-directed
(prefrontal cortex) or context-dependent (hippocampal) cues (Grace and Moore,
1998).

Possible origins of the delayed onset of schizophrenia

One issue that remains to be considered is why a genetically or developmentally
based pathology would not be present at birth, but instead show an onset that is
delayed for several decades. One possibility is that the delayed pathology in the
hippocampus, as suggested by imaging studies (Pantelis et al., 2003; Phillips et al.,
2002), may be related to the changes in stress placed upon the system secondary to
prefrontal cortical disruption. So, some studies have shown that the hippocam-
pus is particularly susceptible to damage as a result of stress (Sapolsky et al.,
1985). Moreover, the hippocampus is involved in modulating the actions of the
280 A. A. Grace

HPA axis in response to glucocorticoids (Sapolsky et. al. 1990), which are released
in response to stress (Cullinan et al., 1995). Indeed, in an animal model of this
disorder, Lipska et al. (1993) have shown that rats with neonatal damage to the hip-
pocampus exhibit hyperresponsiveness to stress. This interaction between stress
and hippocampal damage has been proposed as a model for delayed onset of disor-
ders such as schizophrenia (Walker and Diforio, 1997). In addition, a pathology of
prefrontal cortex function may also place increased demand on the hippocampus.
It has been suggested that some of the higher-order functions that are mediated by
specific cortical regions in the adult, such as the prefrontal cortex, may be mediated
by subcortical regions during developmental periods prior to the maturation of
the cortex under question (Goldman and Rosvold, 1972). Specifically, it has been
proposed that, early in life, the hippocampus may subsume some of the work-
ing memory functions of the individual while the prefrontal cortex is developing
during early life (Diamond, 1990), with this responsibility being off-loaded to the
prefrontal cortex as it reaches functional maturity. One consequence of a prefrontal
pathology would be that this region may not be capable of taking over such func-
tions from the hippocampus. Indeed, this is supported by data showing that some
working memory functions in the schizophrenia patient are mediated by the hip-
pocampus (Meyer-Lindenberg et al., 2001; Weinberger et al., 1993). Therefore, at
a time when stress may be taking an increasing toll on the hippocampus, it will
also be placed under abnormally high functional demand because of the increased
working memory load placed upon it. Indeed, the increased hippocampal volume
prior to the first episode of psychotic symptoms (Pantelis et al., 2003; Phillips et al.,
2002) may be an indicator of such a pathological condition.
The result is that maturation-related stressors could trigger a cascade whereby
prefrontal dysfunction interrupts coping mechanisms for stress at the level of the
amygdala, which, via projections from the central nucleus of the amygdala to the
hypothalamus, could augment stress hormone release. The hyperactivated sub-
cortical DA system could serve to overwhelm the system further with a flooding
of sensory input, thus increasing the stress-induced responses of the system. As
reviewed above, the glucocorticoids released in response to stress are known to
damage the hippocampus (Sapolsky et al., 1985), and it is known that the hip-
pocampus is involved in modulating the glucocorticoid response to stress (Hoschl
and Hajek, 2001; McEwen, 1998; Sapolsky, 1992; Sapolsky et al., 1990). There-
fore, the multiple systems that contribute to an augmented stress response and
the stress-induced release of glucocorticoids, when combined with increased func-
tional demand placed on the hippocampus, could lead to hippocampal damage,
which would reduce the hippocampal modulation of the stress response, thus lead-
ing to further heightened response to stress and more hippocampal damage in a
positive feedback system. Such an interaction could explain the delayed onset of
281 Developmental dysregulation of the dopamine system

Fig. 15.1. A model of the developmental pathophysiology of schizophrenia. This model consists of
a series of interconnected loops, with the common endpoint being a disruption in the
regulation of phasic dopamine (DA) release. One feature of this model is that the positive
feedback loop could be initiated from a number of starting points. For example, a perinatal
deficit in DA in the prefrontal cortex (PFC) is proposed to result in (i) disruption of tonic
DA regulation in the nucleus accumbens; (ii) removal of regulation of affective responses
within the basolateral amygdala; and (iii) a failure to offload working memory demand
from the hippocampus during maturation. This would lead to abnormally potentiated stress
responses, which would drive hippocampal pathology and DA imbalance further. Similarly,
an initiation within the hippocampal system would also be expected to exacerbate the
stress response, ultimately leading to disruption of PFC regulation and subcortical DA system
responsivity. Note that each of these feedback pathways loops through the stress response.
Thus, one potential means to alter the course of this pathological process would be to treat
the exacerbated stress response prior to the initiation of the damaging positive feedback
cascade that is proposed to lead to irreversible system damage and psychosis.

this pathology (Fig. 15.1). This is likely to have important implications for how it
ultimately affects DA system function.

Pathophysiological impact on dopamine transmission

A triumvirate of pathophysiology is proposed to take place: (i) decreased pre-
frontal cortical function as a result of an extant loss of DA modulation or in
response to stress; (ii) increased subcortical stress response secondary to the loss of
prefrontal cortical modulatory influences; and (iii) stress-induced damage to the
282 A. A. Grace

hippocampus. Moreover, each of these processes is known to affect DA system func-

tion in a specific manner. For example, we have found that rats with lesions of the
DA innervation of the prefrontal cortex exhibit decreases in spontaneous firing and
increases in burst firing of DA neurons in the ventral tegmental area (Harden
et al., 1998). Furthermore, inactivation of the ventral hippocampus (Floresco
et al., 2001) as well as chronic cold stress (Moore et al., 2001b) will also decrease
spontaneous activity of DA neurons in this region. On first assessment, such a
condition may suggest that these antecedents would decrease DA neuron function;
however, a model of DA system regulation (Grace, 1991) combined with recent
studies into the regulation of DA system activity (Floresca et al., 2003) would sug-
gest that the opposite is likely to take place.
Our studies have shown that the activity states of DA neurons can be differentially
modulated by afferents to the ventral tegmental area. Thus, we found that inactiva-
tion of the ventral pallidal afferents to DA neurons causes an increase in the number
of spontaneously active DA neurons without affecting their overall firing pattern or
rate of discharge. This is thought to be mediated via the fast-firing ventral pallidal
gamma-aminobutyric acid (GABA)-ergic afferents to DA neurons (Chrobak and
Napier, 1993; Yang and Mogenson, 1985), which we predict holds a subset of neu-
rons in a hyperpolarized, non-firing state (Floresco et al., 2001). However, when
ventral pallidal firing is inactivated via local microinjection of the GABA A/B agonists
muscimol and baclofen, these non-firing neurons are released from inhibition. In
contrast, activation of the pedunculopontine cholinergic/glutamatergic afferents
to the ventral tegmental area had no effect on the number of spontaneously firing
DA neurons; instead, it only caused an increase in burst firing. Since burst firing is
believed to be mediated via N-methyl-d-aspartate (NMDA) receptors, this suggests
that activation of NMDA receptors on spontaneously firing DA neurons would
cause them to begin burst firing. In contrast, the neurons that are being held in a
hyperpolarized state by the ventral pallidal GABAergic afferents would not be acti-
vated by NMDA as a result of voltage-dependent magnesium block of this ionotropic
receptor site (Mayer et al., 1984). As a consequence, we found that we could sepa-
rately regulate DA neuron population activity and firing pattern via manipulation
of these afferents (Floresco et al., 2001, 2003).
Being able to regulate activity states of the ventral tegmental DA neurons indepen-
dently enabled us to examine the impact of changes in activity states on DA release
in the nucleus accumbens. When the ventral pallidum was inactivated (to increase
the number of spontaneously active DA neurons), we observed a 50% increase in
extracellular DA in the accumbens. In contrast, when the pedunculopontine
tegmentum was activated (to increase burst firing), there was no detectable change
in extracellular DA. However, by using microdialysis measures, we were measuring
overflow of DA from the synapse rather than DA release per se. For this reason,
283 Developmental dysregulation of the dopamine system

we tested the effects of blockade of DA uptake on accumbens DA levels. Admin-

istration of the DA uptake blocker nomifensine locally through the microdialysis
probe caused a 10-fold increase in baseline levels of extracellular DA. Under these
conditions, increasing DA population activity (via ventral pallidal inactivation) still
caused a 50% increase in extracellular DA over baseline levels. However, in marked
contrast, activation of burst firing (via pedunculopontine activation) now caused
a greater than 300% increase in extracellular DA in the accumbens (Floresco et al.,
2003).
These data suggest that extracellular DA levels are regulated primarily by the
number of DA neurons firing. We have defined this previously as “tonic” DA levels
(Grace, 1991), since this appears to be the low-level, constant amount of DA present
in the extrasynaptic space. Levels of DA in this region are quite low (estimated to
be approximately 5–40 nM); such levels are clearly too low to impact directly on
postsynaptic DA receptors within the synapse, which are believed to be exposed to
many times higher concentrations (Wightman and Robinson, 2002). However, this
level is sufficient to stimulate presynaptic receptors (Richfield et al., 1989), such
as the inhibitory presynaptic DA autoreceptors present on the DA terminal. As a
consequence, tonic DA levels are proposed to downregulate DA released by spike
activity (Fig. 15.2B).
In contrast, the DA released by burst firing appears to be confined to the vicinity
of the synapse by the reuptake process, which is consistent with the location of the
dopamine transporters at the borders of the synaptic junction (Nirenberg et al.,
1997) where they can most effectively limit diffusion of DA out of the synapse. This
“phasic” (Grace, 1991) DA level is proposed to be the actual DA signal that mediates
behaviorally relevant DA transmission, which is consistent with the burst activation
reported to occur during behavioral tasks (Schultz, 1998). In addition to being high
in amplitude, this DA signal is also brief in duration, as it is rapidly inactivated via
reuptake into the DA terminal before it can leave the synaptic space (Fig.15.2A).
Phasic DA is proposed to be modulated by tonic DA, in which stimulation of
presynaptic DA autoreceptors by tonic DA will downmodulate burst firing-induced
phasic DA release (Fig. 15.2C).
Consequently, a decrease in DA neuron population activity produced by disrup-
tion of the hippocampus, by repeated stress, or by lesion of the prefrontal cortex DA
innervation will result in attenuation of tonic DA release. As a result, stimuli that
cause burst firing-mediated phasic DA release would be expected to release much
more DA than would occur normally, owing to this loss of tonic downregulation.
This would be accentuated by the increased burst firing noted as a consequence of
prefrontal cortical DA lesions. Therefore, one consequence of stress-related patho-
logical changes within the prefrontal–amygdala–hippocampal–HPA axis would be
hyperactivation of the DA system in response to stimuli, analogous to that reported
284 A. A. Grace

Fig. 15.2. The dopamine (DA) system is under a complex set of regulatory influences. (A) In response
to a behaviorally salient stimulus, the DA neuron emits bursts of action potentials (1), leading
to a massive release of DA (in micro- to millimolar concentrations) into the synaptic cleft,
where it can stimulate postsynaptic receptors (2). The DA that is released is then rapidly taken
back up into the terminal by the dopamine transporter (DAT; 3) before it can escape the
synaptic cleft. The DAT achieves this because it is located at the periphery of the synaptic cleft
285 Developmental dysregulation of the dopamine system

to take place in schizophrenia patients (Laruelle, 2000; Laruelle and Abi-Dargham,

1999).

Conclusions
I have presented a model whereby stress is proposed to interact with an existing
pathology within the limbic system to cause the delayed emergence of the posi-
tive symptoms of schizophrenia. The involvement of stress is demonstrated by the
finding that, of the children at risk for developing schizophrenia, those that do go
on to show the pathology exhibit higher levels of anxiety and stress premorbidly
(E. C. Johnstone, personal communication). Moreover, stress is known to exacer-
bate the symptoms of schizophrenia and lead to a relapse in patients under remis-
sion (Birley and Brown, 1970; Norman and Malla, 1993a,b). One implication of
this model is that, given that the DA system does not appear to be abnormal prior
to the first episode of psychotic symptoms, prophylactic treatment with antipsy-
chotic drugs may not be an effective course of action for preventing the ultimate
transition to schizophrenia. In contrast, drugs that act to reduce indices of stress or
anxiety may be more effective at preventing the triggering of the cascade of events
that eventually lead to the development of schizophrenia.

Acknowledgements
This work was supported by USPHS MH01055, MH57440, and MH45156. I thank
Stan Floresco for his help in producing the figures used in this manuscript.

(Nirenberg et al., 1997), where it can effectively limit diffusion of released transmitter.
(B) In contrast, slow, irregular firing of DA neurons (1) leads to substantially lower lev-
els of DA release into the synaptic cleft; this DA release is not as potently affected by
the DAT (2) but is modulated by glutamatergic (glu) afferents that synapse in the vicin-
ity of the DA terminal (3). As a result, a small amount of DA will escape the synaptic cleft
(4) or be released from non-synaptic sites on the terminal (5). This low (approximately
20 nM) concentration of tonic DA (6) is then free to diffuse away from the synaptic terminal.
(C) Although the levels of tonic DA (1) are too low to stimulate postsynaptic receptors, this
concentration is sufficient to activate D2 autoreceptors on the DA terminal, which would
cause an attenuation of burst firing-dependent phasic DA release (2). As a result, there is a
tonic inhibition in the amplitude of the phasic DA release (3). Systems that will attenuate
tonic DA release, either decreasing tonic DA neuron activity (i.e. less DA neurons active) or a
pathological decrease in cortical glutamatergic drive, would thereby remove tonic inhibition
of phasic DA release (Grace, 1991).
286 A. A. Grace

REFERENCES

Abercrombie, E. D., Jacobs, B. L. (1987). Single-unit response of noradrenergic neurons in the

locus coeruleus of freely moving cats. I. Acutely presented stressful and nonstressful stimuli.
J Neurosci 7: 2837–2843.
Abercrombie, E. D., Keller, Jr., R. W., Zigmond, M. J. (1988). Characterization of hippocam-
pal norepinephrine release as measured by microdialysis perfusion: pharmacological and
behavioral studies. Neuroscience 27: 897–904.
Abercrombie, E. D., Keefe, K. A., DiFrischia, D. S., Zigmond, M. J., (1989). Differential effect
of stress on in vivo dopamine release in striatum, nucleus accumbens, and medial frontal
cortex. J Neurochem 52: 1655–1658.
Akil, M., Pierri, J. N., Whitehead, R. E., et al. (1999). Lamina-specific alterations in the dopamine
innervation of the prefrontal cortex in schizophrenic subjects. Am J Psychatry 156: 1580–
1589.
Angrist, B., Santhananthan, G., Wilk, S., Gershon, S. (1974). Amphetamine psychosis: behavioral
and biochemical aspects. J. Psychiatr Res 11: 13–23.
Angrist, B., Rotrosen, J., Gershon, S. (1980). Differential effects of amphetamine and neuroleptics
on negative vs. positive symptoms in schizophrenia. Psychopharmacology 72: 17–19.
Arnold, S. E. (1997). The medial temporal lobe in schizophrenia. J Neuropsychiatry Clin Neurosci
9: 460–470.
Arnold, S. E., Trojanowski, J. Q., Gur, R. E., et al. (1998). Absence of neurodegeneration and
neural injury in the cerebral cortex in a sample of elderly patients with schizophrenia. Arch
Gen Psychiatry 55: 225–232.
Aston-Jones, G., Ennis, M., Pieribone, V. A., Nickell, W. T., Shipley, M. T. (1986). The brain
nucleus locus coeruleus: restricted afferent control of a broad efferent network. Science 234:
734–737.
Bayer, S. A., Altman, J. (1991). Neocortical Development. New York: Raven Press.
Bayer, T. A., Falkai, P., Maier, W. (1999). Genetic and non-genetic vulnerability factors in
schizophrenia: the basis of the “two hit hypothesis”. J Psychiatr Res 33: 543–548.
Bertolino, A., Knable M. B., Saunders R. C., et al. (1999). The relationship between dorsolateral
prefrontal N-acetylaspartate measures and striatal dopamine activity in schizophrenia. Biol
Psychiatry 45: 660–667.
Beuger, M., van Kammen, D. P., Kelley, M. E., Yao, J. (1996). Dopamine turnover in schizophrenia
before and after haloperidol withdrawal. CSF, plasma, and urine studies, Neuropsychophar-
macology 15: 75–86.
Bilder, R. M., Bogerts, B., Ashtari, M., et al. (1995). Anterior hippocampal volume reductions
predict frontal lobe dysfunction in first episode schizophrenia. Schizophr Res 17: 47–58.
Birley, J., Brown, G. W. (1970). Crisis and life changes preceding the onset or relapse of acute
schizophrenia: clinical aspects. Br J Psychiatry 16: 327–333.
Braff, D. L., Grillon, C., Geyer, M. A. (1992). Gating and habituation of the startle reflex in
schizophrenic patients. Arch Gen Psychiatry 49: 206–215.
Breier, A., Buchanan, R. W., Elkashef, A., et al. (1992). Brain morphology and schizophrenia.
A magnetic resonance imaging study of limbic, prefrontal cortex, and caudate structures.
Arch Gen Psychiatry 49: 921–926.
287 Developmental dysregulation of the dopamine system

Chang, M. S., Sved, A. F., Zigmond, M. J., Austin, M. C. (2000). Increased transcription of the
tyrosine hydroxylase gene in individual locus coeruleus neurons following footshock stress.
Neuroscience 101: 131–139.
Chrobak, J. J., Napier, T. C. (1993). Opioid and GABA modulation of accumbens-evoked ventral
pallidal activity. J Neur Transm Gen Sect 93: 123–143.
Cohen, J. D., Servan-Schreiber, D. (1992). Context, cortex and dopamine: a connectionist
approach to behavior and biology in schizophrenia. Psychol Rev 99: 45–77.
Correll, C. J., Rosenkranz, J. A., Grace, A. A. (2001). Chronic cold stress alters basal neuronal firing
rate and response to acute stressors of basolateral amygdala neurons in rats. Soc Neurosci
Abstr 27: 2540.
Creese, I., Burt, D. R., Snyder, S. H. (1976). Dopamine receptor binding predicts clinical and
pharmacological potencies of antischizophrenic drugs. Science 192: 596–598.
Cullinan, W. E., Herman, J. P., Helmreich, D. L., Watson, Jr., S. J. (1995). A neuroanatomy
of stress. In Neurobiological and Clinical Consequences of Stress: From Normal Adaptation
to PTSD, ed. M. J. Friedman, D. S. Charney, A. Y. Deutch. Philadelphia, PA: Lippincott,
pp. 3–25.
Curtis, A. L., Lechner, S. M., Pavcovich, L. A., Valentino, R. J. (1997). Activation of the locus
coeruleus noradrenergic system by intracoerulear microinfusion of corticotropin-releasing
factor: effects on discharge rate, cortical norepinephrine levels and cortical electroencephalo-
graphic activity. J Pharmacol Exp Ther 281: 163–172.
DeLisi, L. E. (1992). The significance of age of onset for schizophrenia. Schizophr Bull 18: 209–
215.
Diamond, A. (1990). The development and neural bases of memory functions as indexed by the
AB and delayed response tasks in human infants and infant monkeys. Ann NY Acad Sci 608:
267–309. [Discussion 309–317.]
Drevets, W. C. (1999). Prefrontal cortical–amygdalar metabolism in major depression. Ann N Y
Acad Sci 877: 614–637.
Dunn, A. J., Berridge, C. W. (1987). Corticotropin-releasing factor administration elicits a stress-
like activation of cerebral catecholaminergic systems.Pharmacol Biochem Behav 27: 685–691.
Eastwood, S. L., Cairns, N. J., Harrison, P. J. (2000). Synaptophysin gene expression in schizophre-
nia. Investigation of synaptic pathology in the cerebral cortex. Br J Psychiatry 176: 236–
242.
Edwards, H. (1970). The significance of brain damage in persistent oral dyskinesia. Br J Psychiatry
116: 271–275.
Emoto, H., Koga, C., Ishii, H., et al. (1993). A CRF antagonist attenuates stress-induced increases
in NA turnover in extended brain regions in rats. Brain Res 627: 171–176.
Ennis, M., Shipley, M. T., Aston-Jones, G., Williams, J. T. (1998). Afferent control of nucleus locus
ceruleus: differential regulation by “shell” and “core” inputs. Adv Pharmacol 42: 767–771.
Finlay J. M., Jedema H. P., Rabinovic, A. D., et al. (1997). Impact of corticotropin-releasing
hormone on extracellular norepinephrine in prefrontal cortex after chronic cold stress. J
Neurochem 69: 144–150.
Floresco, S. B., Todd, C. L., Grace, A. A. (2001). Glutamatergic afferents from the hippocampus
to the nucleus accumbens regulate activity of ventral tegmental area dopamine neurons.
J Neurosci 21: 4915–4922.
288 A. A. Grace

Floresco, S. B., West, A. R., Ash, B., Moore, H., Grace, A. A. (2003). Afferent modulation of
dopamine neuron firing differentially regulates tonic and phasic dopamine transmission.
Nat Neurosci 6: 968–973.
Garey, L. J., Ong, W. Y., Patel, T. S., et al. (1998). Reduced dendritic spine density on cerebral
cortical pyramidal neurons in schizophrenia. J Neurol Neurosurg Psychiatry 65: 446–453.
Ghajarnia, M., Moore, H., Grace, A. A. (1998). Enhanced behavioral effects of phencyclidine
(PCP) in rats with developmental abnormalities of the temporal lobe. Soc Neurosci Abstr 24:
2177.
Glantz, L. A., Lewis, D. A. (1997). Reduction of synaptophysin immunoreactivity in the pre-
frontal cortex of subjects with schizophrenia. Regional and diagnostic specificity. Arch Gen
Psychiatry 54: 943–952.
Goddard, A. W., Charney, D. S. (1997). Toward an integrated neurobiology of panic disorder. J
Clin Psychiatry 58(Suppl. 2): 4–11.
Goff, D. C., Coyle, J. T. (2001). The emerging role of glutamate in the pathophysiology and
treatment of schizophrenia. Am J Psychiatry 158: 1367–1377.
Goldman, P. S., Rosvold, H. E. (1972). The effects of selective caudate lesions in infant and juvenile
rhesus monkeys. Brain Res 43: 53–66.
Goldman-Rakic, P. S. (1995). Cellular basis of working memory. Neuron 14: 477–485.
(1996). Regional and cellular fractionation of working memory. Proc Natl Acad Sci USA 93:
13473–13480.
(1999). The physiological approach: functional architecture of working memory and disor-
dered cognition in schizophrenia. Biol Psychiatry 46: 650–661.
Grace, A. A. (1991). Phasic versus tonic dopamine release and the modulation of dopamine
system responsivity: a hypothesis for the etiology of schizophrenia. Neuroscience 41: 1–24.
(2000). Gating of information flow within the limbic system and the pathophysiology of
schizophrenia. [In Proceedings from the Nobel Symposium – Schizophrenia: Pathophysi-
ological Mechanisms.] Brain Res Rev 31: 330–361. Online at Brain Research Interactive:
http://www.elsevier.nl/gej-ng/29/19/30/26/43/article.html.
Grace, A. A., Moore, H. (1998). Regulation of information flow in the accumbens: a model for the
pathophysiology of schizophrenia. In Origins and Development of Schizophrenia: Advances
in Experimental Psychopathology, ed. M. F. Lenzenweger, R. H. Dworkin. Washington, DC:
American Psychological Association Press, pp. 123–157.
Grace, A. A., Bunney, B. S., Moore, H., Todd. C. L. (1997). Dopamine cell depolarization block
as a model for the therapeutic actions of antipsychotic drugs. Trends Neurosci 20: 31–37.
Gunne, L. M., Growdon, J., Glaeser, B. (1982). Oral dyskinesia in rats following brain lesions and
neuroleptic drug administration. Psychopharmacology 77: 134–139.
Harden, D. G., King, D., Finlay, J., Grace, A. A. (1998). Depletion of dopamine in the prefrontal
cortex decreases the basal electrophysiological activity of mesolimbic dopamine neurons.
Brain Res 794: 96–102.
Harrison, P. J. (1999). The neuropathology of schizophrenia. A critical review of the data and
their interpretation. Brain 122: 593–624.
Herman, J. P., Cullinan, W. E. (1997). Neurocircuitry of stress: central control of the hypothalamo–
pituitary–adrenocortical axis. Trends Neurosci 20: 78–84.
289 Developmental dysregulation of the dopamine system

Hoschl, C., Hajek, T. (2001). Hippocampal damage mediated by corticosteroids: a neuropsychi-

atric research challenge. Eur Arch Psychiatry Clin Neurosci 251(Suppl. 2): 81–88.
Imperato, A., Puglisi-Allegra, S., Casolini, P., Zocchi, A., Angelucci, L. (1989). Stress-induced
enhancement of dopamine and acetylcholine release in limbic structures: role of cortico-
sterone. Eur J Pharmacol 165: 337–338.
Jedema, H. P., Grace, A. A. (2003). Chronic exposure to cold stress alters electrophysiologi-
cal properties of locus coeruleus neurons recorded in vitro. Neuropsychopharmacology 28:
63–72.
Jentsch, J. D., Roth, R. H. (1999). The neuropsychopharmacology of phencyclidine: from NMDA
receptor hypofunction to the dopamine hypothesis of schizophrenia. Neuropsychopharma-
cology 20: 201–225.
Jones, E. G. (1997). Cortical development and thalamic pathology in schizophrenia. Schizophr
Bull 23: 483–501.
Kalin, N. H. (1985). Behavioral effects of ovine corticotropin-releasing factor administered to
rhesus monkeys. Fed Proc 44: 249–253.
Kelley, A. E., Delfs, J. M. (1994). Excitatory amino acid receptors mediate the orofacial stereo-
typy elicited by dopaminergic stimulation of the ventrolateral striatum. Neuroscience 60:
85–95.
Kelley, A. E., Swanson, C. J. (1997). Feeding induced by blockade of AMPA and kainate receptors
within the ventral striatum: a microinfusion mapping study. Behav Brain Res 89: 107–113.
Kendler, K. (1983). Overview: a current perspective on twin studies of schizophrenia. Am J
Psychiatry 140: 1413–1425.
King, D., Zigmond, M. J., Finlay, J. M. (1997). Effects of dopamine depletion in the medial
prefrontal cortex on the stress-induced increase in extracellular dopamine in the nucleus
accumbens core and shell. Neuroscience 77: 141–153.
Laruelle, M. (2000). The role of endogenous sensitization in the pathophysiology of schizophrenia:
implications from recent brain imaging studies. Brain Res Rev 31: 371–384.
Laruelle, M., Abi-Dargham, A. (1999). Dopamine as the wind of the psychotic fire: new evidence
from brain imaging studies. J Psychopharmacol 13: 358–371.
Lavin, A., Grace, A. A. (1997). Effects of afferent stimulation and DA application on prefrontal
cortical cells recorded intracellularly in vivo: comparisons between intact rats and rats with
pharmacologically-induced disruption of cortical development. Soc Neurosci Abstr 23: 2080.
Lawrie, S. M., Abukmeil, S. S. (1998). Brain abnormality in schizophrenia. A systematic and
quantitative review of volumetric magnetic resonance imaging studies. Br J Psychiatry 172:
110–120.
Lechner, S. M., Curtis, A. L., Brons, R., Valentino, R. J. (1997). Locus coeruleus activation by
colon distention: role of corticotropin-releasing factor and excitatory amino acids. Brain
Res 756: 114–124.
LeDoux, J. E. (2000). Emotion circuits in the brain. Annu Rev of Neurosci 23: 155–184.
Lewis, D. A. (1997). Development of the prefrontal cortex during adolescence: insights into
vulnerable neural circuits in schizophrenia. Neuropsychopharmacology 16: 385–398.
Lewis, D. A., Levitt, P. (2002). Schizophrenia as a disorder of neurodevelopment. Annu Rev
Neurosci 25: 409–432.
290 A. A. Grace

Liddle, P. F., Friston, K. J., Frith, C. D., Frackowiak, R. S. (1992). Cerebral blood flow and mental
processes in schizophrenia. J Roy Soc Med 85: 224–227.
Lipska, B. K., Weinberger, D. R. (2000). To model a psychiatric disorder in animals: schizophrenia
as a reality test. Neuropsychopharmacology 23: 223–239.
Lipska, B. K., Jaskiw, G. E., Weinberger, D. (1993). Postpubertal emergence of hyperresponsiveness
to stress and to amphetamine after neonatal hippocampal damage: a potential animal model
of schizophrenia. Neuropsychopharmacology 9: 67–75.
Mana, M. J., Grace, A. A. (1997). Chronic cold stress alters the basal and evoked electrophysio-
logical activity of rat locus coeruleus neurons. Neuroscience 81: 1055–1064.
Mayer, M. L., Westbrook, G. L., Guthrie, P. B. (1984). Voltage-dependent block by Mg2+ of NMDA
responses in spinal cord neurones. Nature 309: 261–263.
McCarley, R. W., Wible, C. G., Frumin, M., et al. (1999). MRI anatomy of schizophrenia. Biol
Psychiatry 45: 1099–1119.
McDonald, C., Murray, R. M. (2000). Early and late environmental risk factors for schizophrenia.
Brain Res Rev 31: 130–137.
McEwen, B. S. (1998). Protective and damaging effects of stress mediators. N Engl J Med 338:
171–179.
McKernan, M. G., Shinnick-Gallagher, P. (1997). Fear conditioning induces a lasting potentiation
of synaptic currents in vitro. Nature 390: 607–611.
Meyer-Lindenberg, A., Poline, J. B., Kohn, P. D., et al. (2001). Evidence for abnormal cortical
functional connectivity during working memory in schizophrenia. Am J Psychiatry 158:
1809–1817.
Mirnics, K., Middleton, F. A., Marquez. A., Lewis, D. A., Levitt, P. (2000). Molecular characteriza-
tion of schizophrenia viewed by microarray analysis of gene expression in prefrontal cortex.
Neuron 28: 53–67.
Moore, H., Grace, A. A. (1997). Anatomical changes in limbic structures produced by methy-
lazoxymethanol acetate (MAM) during brain development are associated with changes in
physiological interactions among afferents to the nucleus accumbens. Soc Neurosci Abstr 23:
2378.
Moore, H., Ghajarnia, M., Geyer, M., Jentsch, J. D., Grace, A. A. (2001a). Selective disruption of
prefrontal and limbic corticostriatal circuits by prenatal exposure to the DNA methylation
agent methylazoxymethanol acetate (MAM): anatomical, neurophysiological and behavioral
studies. Schizophr Res 49(Suppl. 1–2): 48.
Moore, H., Rose, H. J., Grace, A. A. (2001b). Chronic cold stress reduces the spontaneous
activity of ventral tegmental dopamine neurons. Neuropsychopharmacology 24: 410–
419.
Nagata, Y., Matsumoto, H. (1969). Studies on methylazoxymethanol: methylation of nucleic acids
in the fetal rat brain. Proc Soc Exp Biol Med 132: 383–385.
Ninan, P. T. (1999). The functional anatomy, neurochemistry, and pharmacology of anxiety.
J Clin Psychiatry 60(Suppl. 22): 12–17.
Nirenberg, M. J., Chan, J., Pohorille, A., et al. (1997). The dopamine transporter: comparative
ultrastructure of dopaminergic axons in limbic and motor compartments of the nucleus
accumbens. J Neurosci 17: 6899–6907.
291 Developmental dysregulation of the dopamine system

Norman, R. M., Malla, A. K. (1993a). Stressful life events and schizophrenia: I. A review of the
research. Br J Psychiatry 162: 161–166.
(1993b). Stressful life events and schizophrenia: II. Conceptual and methodological issues.
Br J Psychiatry 162: 166–174.
O’Donnell, P., Grace, A. A. (1995). Synaptic interactions among excitatory afferents to nucleus
accumbens neurons: hippocampal gating of prefrontal cortical input. J Neurosci 15: 3622–
3639.
Olney, J. W., Farber, N. B. (1995). NMDA antagonists as neurotherapeutic drugs, psychotogens,
neurotoxins, and research tools for studying schizophrenia. Neuropsychopharmacology 13:
335–345.
Ono, T., Nishijo, H., Uwano, T. (1995). Amygdala role in conditioned associative learning. Prog
Neurobiol 46: 401–423.
Page, M. E., Abercrombie, E. D. (1999). Discrete local application of corticotropin-releasing factor
increases locus coeruleus discharge and extracellular norepinephrine in rat hippocampus.
Synapse 33: 304–313.
Pantelis, C., Velakoulis, D., McGorry, P. D., et al. (2003). Neuroanatomical abnormalities before
and after onset of psychosis: a cross-sectional and longitudinal MRI comparison. Lancet
361: 281–288.
Pare, D., Collins, D. R. (2000). Neuronal correlates of fear in the lateral amygdala: multiple
extracellular recordings in conscious rats. J Neurosci 20: 2701–2710.
Phillips, L. J., Velakoulis, D., Pantelis, C., et al. (2002). Non-reduction in hippocampal volume is
associated with higher risk of psychosis. Schizophr Res 58: 145–158.
Pilowsky, L. S., Kerwin, R. W., Murray, R. M. (1993). Schizophrenia: a neurodevelopmental
perspective. Neuropsychopharmacology 9: 83–91.
Pitkanen, A., Savander, V., LeDoux, J. E. (1997). Organization of intra-amygdaloid circuitries
in the rat: an emerging framework for understanding functions of the amygdala. Trends
Neurosci 20: 517–523.
Post, R. M., Fink, E., Carpenter, W. T., Goodwin, F. K. (1975). Cerebrospinal fluid amine metabo-
lites in acute schizophrenia. Arch Gen Psychiatry 32: 1063–1069.
Ramsooksingh M. D., Jedema, H. P., Moore, H., Sved, A. F., Grace, A. A. (2001). The effects
of chronic stress on the regulation of locus coeruleus neurons by peripheral, limbic, and
prefrontal cortical inputs. Soc Neurosci Abstr 27: 1961.
Ramsooksingh, M. D., Jedema, H. P., Moore, H., Sved, A. F., Grace, A. A. (2002). Chronic cold
stress leads to persistent changes in the behavioral response to footshock with coincident
changes in evoked spike activity in the locus coeruleus. Soc Neurosci Abstr 28: [Program No.]
669.4.
Richfield, E. K., Penney, J. B., Young, A. B. (1989). Anatomical and affinity state comparisons
between dopamine D1 and D2 receptors in the rat central nervous system. Neuroscience 30:
767–777.
Rogan, M. T., Stauble, U. V., LeDoux, J. E. (1997). Fear conditioning induces associative long-term
potentiation in the amygdala. Nature 390: 604–607.
Rosenkranz, J. A., Grace, A. A. (1999). Modulation of basolateral amygdala neuronal firing and
afferent drive by dopamine receptor activation in vivo. J Neurosci 19: 11027–11039.
292 A. A. Grace

(2001). Dopamine attenuates prefrontal cortical suppression of sensory inputs to the basolat-
eral amygdala of rats. J Neurosci 21: 4090–4103.
(2002a). Cellular mechanisms of infralimbic and prelimbic prefrontal cortical inhibition and
dopaminergic modulation of basolateral amygdala neurons in vivo. J Neurosci 22: 324–
337.
(2002b). Dopamine-mediated modulation of odour-evoked amygdala potentials during Pavlo-
vian conditioning. Nature 417: 282–287.
Sabban, E. L., Kvetnansky, R. (2001). Stress-triggered activation of gene expression in cate-
cholaminergic systems: dynamics of transcriptional events. Trends Neurosci 24: 91–98.
Sapolsky, R. (1992). Stress, The Aging Brain, and the Mechanisms of Neuron Death. Cambridge,
MA: MIT Press.
Sapolsky, R., Krey, L., McEwen, B. (1985). Prolonged glucocorticoid exposure reduces hippocam-
pal neural number: Implications for aging. J Neurosci 5: 1221–1224.
Sapolsky, R. M., Armanini, M. P., Packan, D. R., Sutton, W. S., Plotsky, P. M. (1990). Glucocorticoid
feedback inhibition of adrenocorticotropic hormone secretagogue release. Neuroendocrinol-
ogy 51: 328–336.
Saunders R. C., Kolachana, B. S., Bachevalier, J., Weinberger, D. R. (1998). Neonatal lesions of
the medial temporal lobe disrupt prefrontal cortical regulation of striatal dopamine. Nature
393: 169–171.
Seeman, P. (1987). Dopamine receptors and the dopamine hypothesis of schizophrenia. Synapse
1: 133–152.
Schultz, W. (1998). Predictive reward signal of dopamine neurons. J Neurophysiol 80: 1–27.
Selemon, L. D., Rajkowska, G., Goldman-Rakic, P. S. (1995). Abnormally high neuronal density
in the schizophrenic cortex. A morphometric analysis of prefrontal area 9 and occipital area
17. Arch Gen Psychiatry 52: 805–818.
Serova, L. I., Nankova, B. B., Feng, Z., et al. (1999). Heightened transcription for enzymes
involved in norepinephrine biosynthesis in the rat locus coeruleus by immobilization stress.
Biol Psychiatry 45: 853–862.
Shanks, N., Zalcman, S., Zacharko, R. M., Anisman, H. (1991). Alterations of central nore-
pinephrine, dopamine and serotonin in several strains of mice following acute stressor
exposure. Pharmacol Biochem Behav 38: 69–75.
Sherman, J. E., Kalin, N. H. (1986). ICV–CRH potently affects behavior without altering antinoci-
ceptive responding. Life Sci 39: 433–441.
Smagin, G. N., Swiergiel. A. H., Dunn, A. J. (1995). Corticotropin-releasing factor administered
into the locus coeruleus, but not the parabrachial nucleus, stimulates norepinephrine release
in the prefrontal cortex. Brain Res Bull 36: 71–76.
Smagin, G. N., Harris, R. B., Ryan, D. H. (1996). Corticotropin-releasing factor receptor antag-
onist infused into the locus coeruleus attenuates immobilization stress-induced defensive
withdrawal in rats. Neurosci Lett 220: 167–170.
Smagin, G. N., Zhou, J., Harris, R. B., Ryan, D. H. (1997). CRF receptor antagonist attenuates
immobilization stress-induced norepinephrine release in the prefrontal cortex in rats. Brain
Res Bull 42: 431–434.
293 Developmental dysregulation of the dopamine system

Suddath, R. L., Christison, G. W., Torrey, E. F., Casanova, M. F., Weinberger, D. R. (1990).
Anatomical abnormalities in the brains of monozygotic twins discordant for schizophrenia.
N Engl J Med 322: 789–794.
Susser, E., Neugebauer, R., Hoek, H. W., et al. (1996). Schizophrenia after prenatal famine. Arch
Gen Psychiatry 53: 25–31.
Tamminga, C. A., Thaker, G. K., Buchanan, R., et al. (1992). Limbic system abnormalities iden-
tified in schizophrenia using positron emission tomography with fluorodeoxyglucose and
neocortical alterations with deficit syndrome. Arch Gen Psychiatry 49: 522–530.
Tebartz van Elst, L., Woermann, F. G., Lemieux, L., Trimble, M. R. (1999). Amygdala enlargement
in dysthymia: a volumetric study of patients with temporal lobe epilepsy. Biol Psychiatry 46:
1614–1623.
Thierry, A. M., Javoy. F., Glowinski, J., Kety, S. S. (1968). Effects of stress on the metabolism
of norepinephrine, dopamine and serotonin in the central nervous system of the rat. I.
Modifications of norepinephrine turnover. J Pharmacol Exp Ther 163: 163–171.
Thune, J. J., Pakkenberg, B. (2000). Stereological studies of the schizophrenic brain. Brain Res
Rev 31: 200–204.
Torrey, E. F., Miller, J., Rawlings, R., Yolken R. H. (1997). Seasonality of births in schizophrenia
and bipolar disorder: a review of the literature. Schizophr Res 28: 1–38.
Vale, W., Spiess, J., Rivier, C., Rivier, J. (1981). Characterization of a 41-residue ovine hypothalamic
peptide that stimulates secretion of corticotropin and beta-endorphin. Science 213: 1394–
1397.
Valentino, R. J., Foote, S. L. (1986). Brain noradrenergic neurons, corticotropin-releasing factor,
and stress. In: Neural and Endocrine Peptides and Receptors, ed. T. W. Moody. New York:
Plenum Press, pp. 101–120.
Valentino, R. J., Wehby, R. G. (1988). Corticotropin-releasing factor: evidence for a neurotrans-
mitter role in the locus ceruleus during hemodynamic stress. Neuroendocrinology 48: 674–
677.
Valentino, R. J., Foote S. L., Aston-Jones G (1983). Corticotropin-releasing factor activates nora-
drenergic neurons of the locus coeruleus. Brain Res 270: 363–367.
Valentino, R. J., Page, M. E., Curtis, A. L. (1991). Activation of noradrenergic locus coeruleus
neurons by hemodynamic stress is due to local release of corticotropin-releasing factor.
Brain Res 555: 25–34.
Valentino, R. J., Foote, S. L., Page, M. E. (1993). The locus coeruleus as a site for integrating
corticotropin-releasing factor and noradrenergic mediation of stress responses. Ann NY
Acad Sci 697: 173–188.
Valentino, R. J., Chen, S., Zhu, Y., Aston-Jones, G. (1996). Evidence for divergent projections to
the brain noradrenergic system and the spinal parasympathetic system from Barrington’s
nucleus. Brain Res 732: 1–15.
Van Bockstaele, E. J., Colago, E. E., Valentino, R. J. (1996). Corticotropin-releasing factor-
containing axon terminals synapse onto catecholamine dendrites and may presynaptically
modulate other afferents in the rostral pole of the nucleus locus coeruleus in the rat brain.
J Comp Neurol 364: 523–534.
294 A. A. Grace

(1998). Amygdaloid corticotropin-releasing factor targets locus coeruleus dendrites: substrate

for the co-ordination of emotional and cognitive limbs of the stress response. J Neuroen-
docrinol 10: 743–757.
Van Bockstaele, E. J., Peoples, J., Valentino, R. J. (1999). A. E. Bennett Research Award. Anatomic
basis for differential regulation of the rostrolateral peri-locus coeruleus region by limbic
afferents. Biol Psychiatry 46: 1352–1363.
Van Kammen, D. P., Bok van Kammen, W., Mann, L. S., Seppala, T., Linnoila, M. (1986).
Dopamine metabolism in the cerebrospinal fluid of drug-free schizophrenic patients with
and without cortical atrophy. Arch Gen Psychiatry 43: 978–983.
Waddington, J. L. (1990). Spontaneous orofacial movements induced in rodents by very long-term
neuroleptic drug administration: phenomenology, pathophysiology and putative relation-
ship to tardive dyskinesia. [See comments] Psychopharmacology 101: 431–447.
Walker, E. F., Diforio, D. (1997). Schizophrenia: a neural diathesis-stress model. Psychol Rev 104:
667–685.
Weinberger, D. R. (1996). On the plausibility of “the neurodevelopmental hypothesis” of
schizophrenia. Neuropsychopharmacology 14(Suppl. 3): 1S–11S.
(1999). Cell biology of the hippocampal formation in schizophrenia. Biol Psychiatry 45: 395–
402.
Weinberger, D. R., Lipska, B. K. (1995). Cortical maldevelopment, anti-psychotic drugs, and
schizophrenia: a search for common ground. Schizophr Res 16: 87–110.
Weinberger, D. R., Berman, K. F., Ostrem, J. L., Abi-Dargham, A., Torrey, E. F. (1993). Disorgan-
ization of prefrontal-hippocampal connectivity in schizophrenia: a PET study of discordant
MZ twins. Soc Neurosci Abstr 19: 7.
Wightman, R. M., Robinson, D. L. (2002). Transient changes in mesolimbic dopamine and their
association with ‘reward.’ J Neurochem 82: 721–735.
Yang, C. R., Mogenson, G. J. (1985). An electrophysiological study of the neural projection
from the hippocampus to the ventral pallidum and subpallidal areas by way of the nucleus
accumbens. Neuroscience 15: 1015–1024.
Zhu, Y., Aston-Jones, G. (1996). The medial prefrontal cortex prominently innervates a peri-locus
coeruleus zone in rat. Soc Neurosci Abstr 22: 601.
 16

The development of “mis-wired” limbic

lobe circuitry in schizophrenia
and bipolar disorder
Francine M. Benes
McLean Hospital, Belmont and Harvard Medical School, Boston, USA

Since the early 1980s, neuroscience studies have begun to define the pathophys-
iology of schizophrenia and bipolar disorder by using a combination of highly
selective molecular probes and quantitative microscopy. It is now possible to define
how neural circuits have been altered in psychosis in terms of an evolving set of
neurobiological principles (Benes, 2000). In patients with schizophrenia, and more
recently bipolar disorder, a variety of anomalies have been detected in the limbic
lobe, a phylogenetically old portion of the cerebral cortex that includes the anterior
cingulate region and hippocampal formation (Fig. 16.1). In addition, the basolat-
eral subdivision of the amygdala (not shown), another component of the limbic
lobe, has been implicated in the pathophysiology of psychotic disorders, in part
because it sends important inputs to the cingulate cortex and hippocampus. It now
seems likely that the three major components of the limbic lobe play a central role
in generating disturbances in motivation, attention, emotion, and social interac-
tions in schizophrenia and bipolar disorder (Benes, 1993). The discussion below
describes studies in both human and rodent brains that have contributed to our
understanding of how this complex circuitry may be altered during development,
leading to the appearance of psychotic disorders during late adolescence and early
adulthood. A clear understanding of how the brain is normally changing during
the postnatal period is an essential ingredient in building plausible models for what
may be happening during the prodrome of psychotic disorders.

Postmortem findings in schizophrenia and bipolar disorder

Postmortem studies of the brains of schizophrenia patients have examined many
different neuronal and glial markers and it is virtually impossible to review all of
them here. Instead, this discussion will focus on two particular neurotransmitters,
Neurodevelopment and Schizophrenia, ed. Matcheri S. Keshavan et al. Published by Cambridge University
Press.

C Cambridge University Press 2004.

295
296 F. M. Benes

Rat Cat Monkey

Neocortex

Fig. 16.1. The limbic lobe is shown on the midsagittal surface of the brains of a rat, cat, and monkey.
There is a remarkable degree of preservation of this structure across mammalian species. In
contrast, the surrounding neocortex show a marked expansion in primates when compared
with that in rodents. (Adapted from McLean, 1954.)

gamma-aminobutyric acid (GABA) and dopamine, that have been implicated

in both the pathophysiology of psychosis and its treatment with antipsychotic
medications.

The gamma-aminobutyric acid system

The neurotransmitter GABA has long been considered the most important
inhibitory molecule in the mammalian brain (Jones, 1993) and recent evidence
has consistently pointed to a defect of this system in schizophrenia and bipolar
disorder (Benes, 2000). For example, a preferential decrease in the density of non-
pyramidal cells has been reported in the anterior cingulate and prefrontal cortex
(Benes et al., 1991, 2001), as well as in the hippocampal formation (Benes et al.,
1998). The non-pyramidal cells are local circuit neurons, or interneurons, that are
largely GABAergic in nature. Although this particular finding has been the subject
of considerable controversy (Selemon et al., 2002), a number of studies have sug-
gested that a dysfunction of this system might occur in schizophrenia (Benes et al.,
1992a, 1996a) based on markers such as the GABAA receptor; terminals showing
297 “Mis-wired” limbic lobe circuitry

CONTROL PSYCHOSIS

DA DA

Inhibitory GABA Interneuron

Pyramidal Cell

Dopamine Input (Inhibitory)

Glutamatergic Afferents from Amygdala

Fig. 16.2. A schematic diagram depicting a “mis-wired” circuit within layer II of the anterior cingulate
cortex of a normal control (left) and psychotic (right) subject. A pyramidal neuron in the
psychotic circuit is receiving a reduced inhibition from both dopaminergic (DA) and gamma-
aminobutyric acid (GABA)-ergic inputs, while the GABA cell is receiving increased inputs from
the mutually antagonistic amygdalar (excitatory) and dopaminergic (inhibitory) fibers. The
release of DA might prevent the GABA cell from responding to increased excitatory inflow
from the amygdala. Overall, the circuit associated with psychotic disorders has too much
excitation, too little inhibition, and would result in abnormal information processing and
possibly apoptotic changes. Antipsychotic drugs would ameliorate the effect of stress by
blocking DA receptors on GABA cells and allowing the GABA cells to inhibit the firing of
pyramidal neurons.

immunoreactivity for glutamate decarboxylase (GAD) 65 kDa isoform (GAD65 )

(Todtenkopf and Benes, 1998) mRNA for GAD67 (Guidotti et al., 2000; Heckers
et al., 2002; Volk et al., 2000), and GAD65 (Heckers et al., 2002); the GABA trans-
porter (Pierri et al., 1999; Volk et al., 2001; Woo et al., 1998); and several peptides,
including parvalbumin (Beasley and Reynolds, 1997; Zhang and Reynolds, 2002),
calbindin (Daviss and Lewis, 1995; Iritani et al., 1999), and reelin (Fatemi et al.,
1999; Guidotti et al., 2000; Impagnatiello et al., 1998). Most studies have pointed to
a decrease of GABAergic activity. As depicted in Fig. 16.2, the postmortem changes
detected in the anterior cingulate cortex include a decrease of GABAergic modu-
lation in the anterior cingulate region, a change that could give rise to abnormal
information processing (see below).

The dopamine system

All models of schizophrenia must take into account the dopamine system because
neuroleptic drugs block dopamine receptors. To explore this issue, we evaluated
298 F. M. Benes

the distribution of dopamine projections in the anterior cingulate cortex and

unexpectedly obtained evidence for a “mis-wiring” of these fibers (Benes et al.,
1997). Dopamine inputs to GABAergic interneurons in layer II of the anterior
cingulate cortex may be increased, while those to the main projection cells may
be decreased. Using a careful process of hypothesis generation and testing, it was
found that this pattern was most likely attributable to a trophic shift of fibers away
from pyramidal cells to interneurons (Benes, 1998) (Fig. 16.2). A change of this type
could be related to pre- and postnatal stress, since an increase of dopamine inputs
to interneurons has been induced in adult rats using corticosterone administration
(Benes, 1997). The net effect of such a change would be a decline of GABAergic
activity, as dopamine exerts an inhibitory effect on the interneurons (Penit-Soria
et al., 1989).
Consistent with the above model, chronic treatment of rats with the neuroleptic
drug haloperidol has been associated with a significant sprouting of GABAergic
terminals in the medial prefrontal cortex (equivalent to the anterior cingulate of
human brain), a change that could help to compensate for such circuitry defects
(Vincent et al., 1994). Taken together, there is accumulating evidence suggesting
that a “mis-wiring” of dopaminergic inputs to GABA cells may be a potential target
for antipsychotic medications to exert their therapeutic action. More recent studies
(Benes et al., 2000; Heckers et al., 2002) have suggested that a GABA defect may
be a feature not only of schizophrenia but also of bipolar disorder. This could also
help to explain why neuroleptic drugs and GABA-mimetic mood-stabilizing anti-
convulsants are used to treat both disorders, although the latter are more effective
in patients with bipolar disorder. Therefore the GABAergic neuron, together with
its dopaminergic inputs, may be a final common factor in the pathophysiology of
psychotic disorders and a target for treatment.

The role of the amygdala in limbic lobe changes

It is relevant to consider whether changes in the GABA and dopamine systems may
be related to one another or whether they are induced by a third system that interacts
with both. Several of our studies have pointed to changes in glutamatergic fibers
in the anterior cingulate region playing a role in schizophrenia, which is particu-
larly intriguing because this neurotransmitter has been associated with excitotoxic
injury to neurons (Coyle and Puttfarcken, 1993). In the mid-1980s, we observed an
increase of axons with a vertical, but not horizontal, orientation in superficial layers
of the anterior cingulate cortex in schizophrenia (Benes et al., 1987). This finding
was later replicated using a more specific technique that visualized glutamatergic
axons (Benes et al., 1992b). Based on the orientation and location, it was postu-
lated that these vertical axons might be incoming excitatory fibers from another
region. Although there were several candidate regions, the amygdala seemed to be a
299 “Mis-wired” limbic lobe circuitry

particularly likely source because an increase of glutamatergic axons was also

observed in the entorhinal cortex, which receives a significant projection from the
basolateral subdivision of the amygdala (Longson et al., 1996) (Fig. 16.2). In con-
trast, the dorsolateral prefrontal area, which does not receive an appreciable input
from the amygdala, did not show this axonal change in schizophrenia patients
(Benes et al., 1992b), providing further support for this hypothesis. It is notewor-
thy that the basolateral nuclear complex sends a massive projection to layer II of
the anterior cingulate cortex (Amaral and Price, 1984; Cunningham et al., 2002;
Van Hoesen et al., 1993), as well as to sectors CA2/3 of the hippocampal formation
(Berretta et al., 2001; Pitkanen et al., 2000), the two sites where we have observed
a preponderance of microscopic changes in our postmortem studies. As shown in
Fig. 16.2, the working model predicts that there might be excessive excitatory inputs
to both pyramidal cells and GABAergic interneurons in patients with schizophrenia,
although the former would be vulnerable to excitotoxicity.
In order to test the hypothesis that amygdalar projections may play a role in
a decompensation of GABA cells, we have developed a “partial” rodent model
to mimic a defect of GABAergic transmission. In this model, a local infusion of
the GABAA receptor antagonist picrotoxin is made into the basolateral nucleus
(Fig. 16.3). As a result, a selective reduction of GABAergic terminals was observed
in sectors CA3 and CA2 of the hippocampus (Berretta et al., 2001), a pattern that is
remarkably similar to the one observed in our postmortem studies (Todtenkopf and
Benes, 1998). Overall, these findings suggest that the defect of GABAergic cells seen
in schizophrenia and bipolar disorder may involve, at least in part, an increased flow
of excitatory activity from the amygdala to the hippocampus and anterior cingulate
cortex.

Postnatal maturation of limbic lobe circuitry

A relevant question to ask regarding the model circuit shown in Fig. 16.2 is when
during development might these changes have appeared in schizophrenia: this
disorder has a characteristic age of onset between 16 and 25 years and it is possible
that alterations in layer II of the anterior cingulate cortex or sectors CA3/2 of
the hippocampus might have been present from birth and/or were induced at
the time the illness presents. Since the development of the brain is now known
to continue well into the postnatal period, it is possible that there are normal
maturation changes that may “trigger” the start of the schizophrenia syndrome.

The development of the gamma-aminobutyric acid system

Extensive neurochemical studies of the development of the rodent brain suggest
that the maturation of the cortex continues well into the postnatal period (reviewed
by Johnston, 1988). For example, GABA-accumulating cells show a progressive
300 F. M. Benes

Fig. 16.3. The distribution of gamma-aminobutyric acid (GABA)-ergic cells in the medial prefrontal
cortex of rat brain during the postnatal period. Between day 1 (P1) and day 10 (P10),
the density of cells is much greater than that seen at days 20 to 41 (P20 to P41), the
postweanling period that is considered to be the equivalent in rats of human adolescence.
Although the overall number of cells at day 41 is identical to that found at day 5, the packing
density decreases dramatically as the cortical mantle increases its thickness. The GABAergic
cells are surrounded by neuropil, where axons and dendrites occupy most of the space.

increase in numerical density until postnatal day II (Chronwall and Wolff, 1980).
In contrast, the concentration of GABA and the specific activity of GAD (Coyle and
Enna, 1976), as well as the binding activity of the GABA receptor (Coyle and Enna,
1976; Palacios et al., 1979) and its mRNA (Gambarana et al., 1990), all increase
until the third postnatal week.
As shown in Fig. 16.3, the numerical density of GABA-immunoreactive neurons
in the anterior cingulate cortex of the rat reaches a peak at approximately postnatal
day 5 and then diminishes until day 20, when the thickness of the cortical mantle
is maximal (Vincent et al., 1995). During this same period, the relative amount
of neuropil surrounding all cell bodies in this region is expanding as dendritic
and axonal fibers are increasing. GABAergic cell bodies not only show a gradual
increase in their size, but by postnatal days 20 to 40, primary, secondary, and tertiary
branches of their dendritic tree can also be visualized. The expansion of neuropil in
rat medial prefrontal cortex probably involves an increase of both dendritic branches
301 “Mis-wired” limbic lobe circuitry

A B C

GABA-Immunoreactive Neurons

Dopamine-Immunoreactive Fibers

Fig. 16.4. A series of confocal photomicrographs showing colocalization of gamma-aminobutyric acid

(GABA) neurons (green) and dopamine fibers (yellow) in rat medial prefrontal cortex before
(A) and after (B) weaning and in the early adult period (C). There is an apparent increase
in the number of dopamine fibers and they appear to form increased contacts with GABA
neurons. (See plate section for color version.)

and terminals of GABAergic neurons, a process that continues in the superficial

layers until day 25 (Vincent et al., 1995) when the efficacy of GABAergic synaptic
transmission also becomes optimal (Luhmann and Prince, 1991). In general terms,
the maturation of the GABA system in rats continues for approximately 2–3 weeks
postnatally. Presumably, then, the dendritic branches of GABAergic interneurons
will await ingrowing dopamine fibers to target them for the formation of functional
interactions.

Late maturation of dopamine projections

Dopaminergic projections to rat medial prefrontal cortex increase progressively
beyond the weanling stage until the early adult period (Kalsbeek et al., 1988;
Verney et al., 1982). During the first 2 weeks of postnatal life, the distribution
of dopaminergic varicose fibers in medial prefrontal cortex is quite low but with
the highest density in deeper laminae. During the third postnatal week, how-
ever, the density of such fibers clearly increases and this pattern continues until
adulthood (Fig. 16.4). The increase of fiber density occurs to a proportionate
degree across layers VI–II and does not progress in a distinct “inside-out” man-
ner. For sections in which single immunoperoxidase-processing is combined with
 A
I

II

III

VI
P6 P16 P26 P36 P45 Adult

B
2000
r = 0.873

1500

1000
Fiber Density

500

0
0 50 100 150

Postnatal Day
Fig. 16.5. Amygdala projections to the anterior cingulate cortex after birth in rats. (A) Amygdalar fibers
visualized with anterograde tracing show a marked increase in layers II and V during the
postnatal period. (B) The increase in fiber density shows a highly significant curvilinear rise
between birth (postnatal day 0) and adulthood (postnatal day 120) in layer II. (See plate
section for color version.)
303 “Mis-wired” limbic lobe circuitry

cresyl violet staining, DA-containing varicose fibers can be observed throughout

the neuropil, but very commonly, such fibers course toward neuronal cell bodies.
These postnatal increases in the density of dopaminergic projections to rat
medial prefrontal cortex are paralleled by an increase in the binding activity of
D2 receptors, which begins prenatally (Bruinink et al., 1983) and continues until
the fourth postnatal week (Deskin et al., 1981). Interestingly, administration of
6-hydroxydopamine prevents this increase in D2 receptor binding, an effect that
is associated with dystrophic changes in the basal dendrites of pyramidal neurons
(Kalsbeek et al., 1989). Lesions induced in the prefrontal cortex of adult monkeys
using 6-hydroxydopamine result in impaired performance of the spatial delayed
alternation task (Brozoski et al., 1979) and it seems likely that this functional deficit
could be associated with changes in the D2 receptor on pyramidal neurons. It is
noteworthy that dopamine fibers show non-random interactions with cortical neu-
rons that are probably of functional significance (Benes et al., 1993). As shown in
Fig. 16.4, as the cortical dopamine system actively grows into the anterior cingulate
cortex during the equivalent of adolescence in rats, its fibers contact GABAergic
neurons (Benes et al., 1996b). These contacts continue to form until the early adult
period. At this point, there is an overall increase of 150% compared with the period
before weaning (which would correspond to childhood in humans). As discussed
above, the formation of GABA terminals in the cortex reaches a plateau at approx-
imately the fourth postnatal week and the development of the GABA system seems
to precede that of the dopamine system. Accordingly, GABA cells may act as passive
targets for the reception of functional dopaminergic inputs during the equivalent
of late adolescence and early adulthood. If this interpretation is correct, it could
have far-reaching implications for our understanding of psychotic disorders.

The ingrowth of amygdalar fibers into the anterior cingulate cortex

If dopamine fibers show such late developmental changes, is it possible that there
might be analogous changes occurring in other projection systems, such as that
from the basolateral amygdala? As shown in Fig. 16.5, using anterograde tracing,
amygdalar projections to the anterior cingulate cortex do show significant increases
during the postnatal period, as they form synaptic connections with both the spines
and shafts of dendrites (Cunningham et al., 2002). A significant proportion of
these fibers form increased connections with GABAergic neurons (Cunningham
et al., 2002), an observation that is consistent with electrophysiological studies,
which have suggested that these latter cells may be a pivotal target of amygdalar
activation (Perez-Jananay and Vives, 1991). In addition, a recent study has demon-
strated a marked increase in the frequency with which amygdalar fibers form con-
tacts on GABAergic cells in rat cingulate cortex (Cunningham et al., 2002). Since
the amygdalar projections and dopamine fibers show extensive overlap in layer V
304 F. M. Benes

Fig. 16.6. A set of schematic diagrams depicting the normal development of an intrinsic circuit in
the anterior cingulate cortex during early and late adolescence. Gamma-aminobutyric acid
(GABA) cell development is probably relatively complete by early adolescence while that
of dopamine fibers and amygdalar projections continues into late adolescence and early
adulthood. By late adolescence, the neural response to emotionally stressful situations may
be relatively complete in normal individuals. However, individuals exposed to stress early in
life may be sensitized to the effects of stress later. During late adolescence, the stressed circuit
could theoretically develop excessive dopaminergic and amygdalar inputs. Such changes
may be associated with a decompensation of the circuit if GABAergic cells are unable to
provide sufficient inhibitory modulation.

of the cingulate cortex, it will be relevant to know whether these two systems may
actually interact with one another through a presynaptic mechanism.

A model for how a “mis-wired” circuit might lead to decompensation

The findings described above indicate that dopamine and amygdalar fibers seem
to be increasing in number and converging on individual GABAergic cells at a
stage when schizophrenia is beginning. According to the working model shown in
Fig. 16.6, amygdalar fibers would provide increased excitatory drive (McDonald
et al., 1989) while dopamine fibers would provide increased inhibitory modulation
(Retaux et al., 1991). Under stressful conditions when the release of dopamine is
markedly increased (Roth et al., 1988), the target GABA cell could fail to provide
adequate amounts of inhibitory modulation to pyramidal neurons. In this setting,
these latter cells could fire excessively, particularly since excitatory activity from the
amygdala is probably increasing at the same time. This proposed mechanism could
305 “Mis-wired” limbic lobe circuitry

theoretically result in oxidative stress to pyramidal cells and/or GABA neurons

and lead to further dysfunction or perhaps even decompensation of the circuit.
As observed clinically, neuroleptic drugs produce prompt remission of psychotic
symptoms in the early stages of a psychotic illness, possibly by blocking dopamine
receptors on these GABA cells. The net effect of dopamine receptor blockade would
be the release of GABA cells to fire more optimally.
Overall, the studies and model discussed above have suggested the importance
of viewing psychotic disorders from the standpoint of the development of inte-
grated circuitry involving the amygdala, anterior cingulate cortex, and hippocam-
pus. Changes in the activity level of convergent inputs to GABA cells in one or all of
these regions during adolescence may serve as “triggers” for the onset of psychosis
in susceptible individuals. A more detailed understanding of how such develop-
mental changes could interplay with defective neurons may eventually suggest novel
treatment strategies that can be applied early in the course of the illness to offset
the deterioration in functioning that often occurs in individuals with a chronic
psychotic disorder.

Acknowledgements
This work has been supported by grants from the National Institutes of Health
(MH00423, MH31862, MH31154), the Stanley Foundation, the Carmella and
Menachem Abraham Fund and the Taplin Fund.

REFERENCES

Amaral, D. G., Price, J. L. (1984). Amygdalo-cortical projections in the monkey (macaca

fascicularis). J Comp Neurol 230: 465–496.
Beasley, C. L., Reynolds, G. P. (1997). Parvalbumin-immunoreactive neurons are reduced in the
prefrontal cortex of schizophrenics. Schizophr Res 24: 349–355.
Benes, F. M. (1993). Neurobiological investigations in cingulate cortex of schizophrenic brain.
Schizophr Bull 19: 537–549.
(1997). The role of stress and dopamine–GABA interactions in the vulnerability for
schizophrenia. J Psychiatr Res 31: 257–275.
(1998). Model generation and testing to probe neural circuitry in the cingulate cortex of
postmortem schizophrenic brain. Schizophr Bull 24: 219–230.
(2000). Emerging principles of altered neural circuitry in schizophrenia. Brain Res Brain Res
Rev 31: 251–269.
Benes, F. M., Majocha, R., Bird, E. D., Marotta, C. A. (1987). Increased vertical axon numbers in
cingulate cortex of schizophrenics. Arch Gen Psychiatry 44: 1017–1021.
306 F. M. Benes

Benes, F. M., McSparren, J., Bird, E. D., SanGiovanni, J. P., Vincent, S. L. (1991). Deficits in
small interneurons in prefrontal and cingulate cortices of schizophrenic and schizoaffective
patients. Arch Gen Psychiatry 48: 996–1001.
Benes, F. M., Vincent, S. L., Alsterberg, G., Bird, E. D., SanGiovanni, J. P. (1992a). Increased GABAA
receptor binding in superficial layers of cingulate cortex in schizophrenics. J Neurosci 12:
924–929.
Benes, F. M., Sorensen, I., Vincent, S. L., Bird, E. D., Sathi, M. (1992b). Increased density of
glutamate-immunoreactive vertical processes in superficial laminae in cingulate cortex of
schizophrenic brain. Cereb Cortex 2: 503–512.
Benes, F. M., Vincent, S. L., Molloy, R. (1993). Dopamine-immunoreactive axon varicosities form
nonrandom contacts with GABA-immunoreactive neurons of rat medial prefrontal cortex.
Synapse 15: 285–295.
Benes, F. M., Khan, Y., Vincent, S. L., Wickramasinghe, R. (1996a). Differences in the subregional
and cellular distribution of GABAA receptor binding in the hippocampal formation of
schizophrenic brain. Synapse 22: 338–349.
Benes, F. M., Vincent, S. L., Molloy, R., Khan, Y. (1996b). Increased interaction of dopamine-
immunoreactive varicosities with GABA neurons of rat medial prefrontal cortex occurs
during the postweanling period. Synapse 23: 237–245.
Benes, F. M., Todtenkopf, M. S., Taylor, J. B. (1997). Differential distribution of tyrosine hydroxy-
lase fibers on small and large neurons in layer II of anterior cingulate cortex of schizophrenic
brain. Synapse 25: 80–92.
Benes, F. M., Kwok, E. W., Vincent, S. L., Todtenkopf, M. S. (1998). A reduction of nonpyramidal
cells in sector CA2 of schizophrenics and manic depressives. [See comments] Biol Psychiatry
44: 88–97.
Benes, F. M., Todtenkopf, M. S., Logiotatos, P., Williams, M. (2000). Glutamate decarboxylase(65)-
immunoreactive terminals in cingulate and prefrontal cortices of schizophrenic and bipolar
brain. J Chem Neuroanat 20: 259–269.
Benes, F. M., Vincent, S. L., Todtenkopf, M. S. (2001). The density of pyramidal and nonpyramidal
neurons in anterior cingulate cortex of schizophrenic and bipolar subjects. Biol Psychiatry
50: 395–406.
Berretta, S., Munno, D. W., Benes, F. M. (2001). Amygdalar activation alters the hippocampal
GABA system: “partial” modelling for postmortem changes in schizophrenia. J Comp Neurol
431: 129–138.
Brozoski, T., Brown, R. M., Rosvold, H. E., Goldman, P. S. (1979). Cognitive deficit
caused by depletion of dopamine in prefrontal cortex of rhesus monkey. Science 205:
929–931.
Bruinink, A., Lichtensteiner, W., Schlumpf, M. (1983). Pre- and postnatal ontogeny and char-
acterization of dopaminergic D2, serotonergic S2, and spirodecanone binding sites in rat
forebrain. J Neurochem 40: 1227–1237.
Chronwall, B., Wolff, J. R. (1980). Prenatal and postnatal development of GABA-accumulating
cells in the occipital neocortex of rat. J Comp Neurol 190: 187–208.
Coyle, J. T., Enna, S. (1976). Neurochemical aspects of the ontogenesis of GABAnergic neurons
in the rat brain. Brain Res 111: 119–133.
307 “Mis-wired” limbic lobe circuitry

Coyle, J. T., Puttfarcken, P. (1993). Oxidative stress, glutamate, and neurodegenerative disorders.
[Review] Science 262: 689–695.
Cunningham, M. G., Bhattacharyya, S., Benes, F. M. (2002). Amygdalo-cortical sprouting con-
tinues into early adulthood: Implications for the development of normal and abnormal
function during adolescence. J Comp Neurol 453: 116–130.
Daviss, S. R., Lewis, D. A. (1995). Local circuit neurons of the prefrontal cortex in schizophrenia:
selective increase in the density of calbindin-immunoreactive neurons. Psychiatry Res 59:
81–96.
Deskin, R., Seidler, F. J., Whitmore, W. L., Slotkin, T. A. (1981). Development of α-noradrenergic
and dopaminergic receptor systems depends on maturation of their presynaptic nerve
terminals in the rat brain. J Neurochem 36: 1683–1690.
Fatemi, S. H., Emamian, E. S., Kist, D., et al. (1999). Defective corticogenesis and reduction in
Reelin immunoreactivity in cortex and hippocampus of prenatally infected neonatal mice.
Mol Psychiatry 4: 145–154.
Gambarana, C., Pittman, R., Siegel, R. E. (1990). Development expression of the GABA-A receptor
γ 1 subunit mRNA in the rat brain. J Neurobiol 21: 1169–1179.
Guidotti, A., Auta, J., Davis, J. M., et al. (2000). Decrease in Reelin and glutamate acid
decarboxylase67 (GAD67 ) expression in schizophrenia and bipolar disorder. Arch Gen
Psychiatry 57: 1061–1069.
Heckers, S., Stone, D., Walsh, J., et al. (2002). Differential hippocampal expression of glutamic
acid decarboxylase 65 and 67 messenger RNA in bipolar disorder and schizophrenia. Arch
Gen Psychiatry 59: 521–529.
Impagnatiello, F., Guidotti, A. R., Pesold, C., et al. (1998). A decrease of reelin expression as a
putative vulnerability factor in schizophrenia. Proc Natl Acad Sci USA 95: 15718–15723.
Iritani, S., Kuroki, N., Ikeda, K., Kazamatsuri, H. (1999). Calbindin immunoreactivity in the
hippocampal formation and neocortex of schizophrenics. Prog Neuropsychopharmacol Biol
Psychiatry 23: 409–421.
Johnston, M. V. (1988). Biochemistry of neurotransmitters in cortical development. In Cerebral
Cortex, ed. E. G. Jones. New York: Plenum Press, pp. 211–236.
Jones, E. G. (1993). GABAergic neurons and their role in cortical plasticity in primates. Cereb
Cortex 3: 361–372.
Kalsbeek, A., Voorn, P., Buijs, R. M., Pool, C. W., Uylings, H. B. (1988). Development of the
dopaminergic innervation in the prefrontal cortex of the rat. J Comp Neurol 269: 58–72.
Kalsbeek, A., Matthijssen, M. A., Uylings, H. B. (1989). Morphometric analysis of prefrontal corti-
cal development following neonatal lesioning of the dopaminergic mesocortical projection.
Exp Brain Res 78: 279–289.
Longson, D., Deakin, J. F., Benes, F. M. (1996). Increased density of entorhinal glutamate-
immunoreactive vertical fibers in schizophrenia. J Neur Transm 103: 503–507.
Luhmann, H. J., Prince, D. A. (1991). Postnatal maturation of the GABAergic system in rat
neocortex. J Neurophysiol 65: 247–263.
McDonald, A. J., Beitz, A. J., Larson, A. A., et al. (1989). Co-localization of glutamate and tubulin
in putative excitatory neurons of the hippocampus and amygdala: an immunohistochemical
study using monoclonal antibodies. Neuroscience 30: 405–421.
308 F. M. Benes

McLean, P. D. (1954). Studies on limbic system (visceral brain) and their bearing on psycho-
somatic problems. In Recent Developments in Psychosomatic Medicine, ed. E. Wittkower,
R. Cleghorn. Philadelphia, PA: Lippincott, p. 106.
Palacios, J. M., Niehoff, D. L., Kuhar, M. J. (1979). Ontogeny of GABA and benzodiazepine
receptors: effects of Triton X-100, bromide and muscimol. Brain Res 179: 390–395.
Penit-Soria, J., Retaux, S., Maurin, Y. (1989). Effets de la stimulation des recepteurs D1 et D2
dopaminergiques sur la liberation d’acide γ -[3 H] aminobutyrique induite electriquement
dans le cortex prefrontal du rat. C R Acad Sci Paris 309: 441–446.
Perez-Jananay, J. M., Vives, F. (1991). Electrophysiological study of the response of medical
prefrontal cortex neurons to stimulation of the basolateral nucleus of the amygdala in the
rat. Brain Res 564: 97–101.
Pierri, J. N., Chaudry, A. S., Woo, T. U., Lewis, D. A. (1999). Alterations in chandelier neuron
axon terminals in the prefrontal cortex of schizophrenic subjects. Am J Psychiatry 156:
1709–1719.
Pitkanen, A., Pikkarainen, M., Nurminen, N., Ylinen, A. (2000). Reciprocal connections between
the amygdala and the hippocampal formation, perirhinal cortex, and postrhinal cortex in
rat. A review. Ann N Y Acad Sci 911: 369–391.
Retaux, S., Besson, M. J., Penit-Soria, J. (1991). Opposing effects of dopamine D2 receptor
stimulation on the spontaneous and the electrically evoked release of [3 H] GABA on rat
prefrontal cortex slices. Neuroscience 42: 61–71.
Roth, R. H., Tam, S. Y., Ida, Y., Yang, J. X., Deutch, A. Y. (1988). Stress and the mesocorticolimbic
dopamine systems. Ann N Y Acad Sci 537: 138–147.
Selemon, L. D., Goldman, Rakic, P., Rajkowska, G. (2002). 2D versus 3D cell counting. Biol
Psychiatry 51: 838–840.
Todtenkopf, M. S., Benes, F. M. (1998). Distribution of glutamate decarboxylase65 immunore-
active puncta on pyramidal and nonpyramidal neurons in hippocampus of schizophrenic
brain. Synapse 29: 323–332.
Van Hoesen, G. W., Morecraft, R. J., Vogt, B. A. (1993). Connections of the monkey cingulate
cortex. In Neurobiology of Cingulate Cortex and Limbic Thalamus, ed. M. Gabriel. Boston,
MA: Birkhauser, pp. 249–284.
Verney, C., Berger, B., Adrien, J., Vigny, A., Gay, M. (1982). Development of the dopaminergic
innervation of the rat cerebral cortex. A light microscopic immunocytochemical study using
anti-tyrosine hydroxylase antibodies. Brain Res 281: 41–52.
Vincent, S. L., Adamec, E., Sorensen, I., Benes, F. M. (1994). The effects of chronic haloperidol
administration on GABA-immunoreactive axon terminals in rat medial prefrontal cortex.
Synapse 17: 26–35.
Vincent, S. L., Pabreza, L., Benes, F. M. (1995). Postnatal maturation of GABA-immunoreactive
neurons of rat medial prefrontal cortex. J Comp Neurol 355: 81–92.
Volk, D. W., Austin, M. C., Pierri, J. N., Sampson, A. R., Lewis, D. A. (2000). Decreased glu-
tamic acid decarboxylase67 messenger RNA expression in a subset of prefrontal cortical
gamma-aminobutyric acid neurons in subjects with schizophrenia. Arch Gen Psychiatry 57:
237–245.
309 “Mis-wired” limbic lobe circuitry

(2001). GABA transporter-1 mRNA in the prefrontal cortex in schizophrenia: decreased

expression in a subset of neurons. Am J Psychiatry 158: 256–265.
Woo, T. U., Whitehead, R. E., Melchitzky, D. S., Lewis, D. A. (1998). A subclass of prefrontal
gamma-aminobutyric acid axon terminals are selectively altered in schizophrenia. Proc Natl
Acad Sci USA 95: 5341–5346.
Zhang, Z. J., Reynolds, G. P. (2002). A selective decrease in the relative density of parvalbumin-
immunoreactive neurons in the hippocampus in schizophrenia. Schizophr Res 55: 1–10.
 17

Development of thalamocortical circuitry

and the pathophysiology of schizophrenia
Darlene S. Melchitzky1 and David A. Lewis2
1
Mercyhurst College, Erie, USA
2
University of Pittsburgh, Pittsburgh, USA

Schizophrenia is clearly a multithetic disorder in which the diverse signs and symp-
toms of the illness arise as a result of dysfunction in a number of brain regions.
Of these clinical features, a growing body of literature indicates that core features
of the illness are represented by disturbances in certain cognitive functions, such
as those mediated by neural circuits that include components of the dorsolateral
prefrontal cortex (DLPFC) and the mediodorsal nucleus (MDN) of the thalamus
(Elvevåg and Goldberg, 2000). The specific causal factors that give rise to these
brain abnormalities remain enigmatic, with a number of genes and a range of envi-
ronmental events identified as potential risk factors for schizophrenia (Lewis and
Lieberman, 2000). In addition, developmental processes appear to play a central
role in the translation of these etiological factors into the appearance of the clinical
features of the illness (Lewis and Levitt, 2002). Consequently, understanding the
structural and functional maturation of the intrinsic and extrinsic circuitry of the
DLPFC and MDN is essential for understanding the pathophysiological basis for
the cognitive deficits in schizophrenia.

Function and development of the circuitry of the mediodorsal nucleus

and dorsolateral prefrontal cortex
Individuals with schizophrenia perform poorly on cognitive tasks that require the
use of working memory: the ability to maintain information “on line” in order to
guide behavior (Goldman-Rakic, 1994). For example, subjects with schizophrenia
are impaired on oculomotor delayed-response tasks (Park and Holzman, 1992), a
cognitive paradigm on which non-human primates with structural or reversible
cooling lesions of the DLPFC also perform poorly (Fuster, 1997). Consistent with
these observations, many subjects with schizophrenia fail to show normal activation

Neurodevelopment and Schizophrenia, ed. Matcheri S. Keshavan et al. Published by Cambridge University
Press.

C Cambridge University Press 2004.

310
311 Development of thalamocortical circuitry

of the DLPFC when attempting to perform tasks that tap working memory
(Weinberger et al., 1986).
The DLPFC maintains reciprocal excitatory connections with specific subdivi-
sions of the MDN, the principal source of thalamic inputs to the DLPFC (Barbas
et al., 1991; Giguere and Goldman-Rakic, 1988; Goldman-Rakic and Porrino, 1985).
Like the DLPFC, the MDN has been implicated as a site of dysfunction in schizo-
phrenia (Lewis, 2000), and it also appears to play a critical role in working memory
functioning. For example, destruction of some or all of the MDN produces deficits
in working memory performance (Isseroff et al., 1982). Furthermore, during the
delay period of delayed-response tasks, neurons in both the DLPFC (Funahashi
et al., 1989) and the MDN (Fuster and Alexander, 1971) exhibit sustained activity,
which, at least in the DLPFC, has been proposed to be the neural substrate for
working memory (Fuster, 1997; Goldman-Rakic, 1995).
The ability to perform delayed-response tasks first appears between 2 and
4 months of age in monkeys (Goldman-Rakic, 1987) and around 1 year of age in
humans (Diamond, 1985). Performance on these tasks then continues to improve
(as evidenced by the ability to withstand longer delay periods) at a slower rate until
adult functional competence is achieved after puberty in both monkeys (Alexander
and Goldman, 1978) and humans (Casey et al., 2000; Diamond, 2002). Although
the contributions of the MDN are less well studied, studies in humans clearly indi-
cate that changes in the frontal cortex between the ages of 12 and 19 years contribute
to the maturation of cognitive control (Bunge et al., 2002).
Studies in monkeys also suggest that DLPFC neural circuitry is increasingly
involved in the mediation of these improvements in cognitive performance during
postnatal development. That is, the emergence of the ability to perform delayed-
response tasks during infancy does not appear to depend upon the integrity of the
DLPFC but appears to be mediated by subcortical structures like the MDN. For
example, ablation of the DLPFC in infant monkeys does not produce the same
degree of impairment on spatial delayed-response tasks that is observed in adult
animals with such lesions (Alexander and Goldman, 1978). In contrast, during the
first year of life, performance of oculomotor delayed-response tasks do appear to
depend upon the integrity of the MDN (Alexander and Goldman, 1978). Further-
more, during postnatal development, the percentage of DLPFC neurons exhibiting
delay period activity doubles between 12 and 36 months of age (Alexander, 1982).
In contrast, the MDN does not exhibit this type of developmental increase in the
number of delay-activated neurons (Alexander, 1982). Together, these findings sug-
gest that postnatal development is associated with an increasing dependence upon
the DLPFC for the mediation of delayed-response behavior, such that the matu-
ration of the DLPFC enhances the function of brain regions like the MDN, which
appear to subserve these tasks earlier in life (Alexander and Goldman, 1978).
312 D. S. Melchitzky and D. A. Lewis

Fig. 17.1. Connections between the human prefrontal cortex and the mediodorsal nucleus (MDN) of
the thalamus. (A) Schematic view of the lateral surface of the human brain. (B) Coronal
section of the prefrontal cortex, taken at the level of vertical line 1 in (A). (C) Expanded
schematic drawing of the box in (A). Pyramidal neurons in layer 6 send projections to the
MDN as well as to the reticular nucleus of the thalamus (Rt). (D) Coronal section through
the thalamus, taken at the level of vertical line 2 in (A). The individual thalamic nuclei are
outlined. (E) Expanded view of the outline of the thalamus shown in (D). Neurons in the
MDN send axons to the prefrontal cortex that arborize primarily in deep layers 3 and 4
but also extend into layer 1 (C). MDN neurons also send axon collaterals to the inhibitory
neurons of the Rt. Large cortically projecting excitatory neurons in the MDN receive inhibitory
input from smaller GABA neurons located in the MDN and Rt.
313 Development of thalamocortical circuitry

Organization of thalamocortical circuitry

In both rodent and primate brains, retrograde tracing studies have demonstrated
that projection neurons in the MDN are topographically organized such that differ-
ent MDN subdivisions project to specific areas of the frontal lobe (Barbas et al., 1991;
Goldman-Rakic and Porrino, 1985). In general, the magnocellular MDN subdivi-
sion is connected primarily with medial and orbital prefrontal regions, whereas the
parvocellular subdivision projects to the DLPFC (Barbas et al., 1991; Giguere and
Goldman-Rakic, 1988; Goldman-Rakic and Porrino, 1985). However, the three-
dimensional topography of these connections is complex and parvocellular MDN
projections are not restricted to the DLPFC. For example, a recent study demon-
strated that portions of the parvocellular subdivision may be more extensively
connected with more caudal areas of frontal cortex, such as areas 8 (frontal eye
fields) and 6 (premotor cortex), than with the DLPFC proper (Erickson and Lewis,
2000).
Within the DLPFC, MDN thalamic axons terminate primarily in deep layers
3 and 4 (Giguere and Goldman-Rakic, 1988), although other cortical layers do
contain thalamic axons (Fig. 17.1C). For example, MDN thalamic axons have been
observed in layers 1, 2, and superficial 3, although these axons are not as branched
and they exhibit a lower density of boutons compared with those in deep layers 3
and 4 (Erickson and Lewis, 2000). Axons arising from a given location within the
MDN also exhibit both global and local discontinuities in their cortical terminal
fields. However, the relationship between clusters of thalamic axon terminals and
the columnar organization of the intrinsic circuitry of the DLPFC (see below) has
not been directly examined.
The majority of the axon terminals arising from the MDN form asymmetric
synaptic contacts onto the dendritic spines of DLPFC pyramidal neurons, at least
in the rodent (Kuroda et al., 1998). In addition, several studies have demonstrated
that the calcium-binding protein parvalbumin (PV) is present in axon terminals
arising from thalamic nuclei (DeFelipe and Jones, 1991). More specifically, in deep
layers 3 and 4 of monkey DLPFC, PV-immunoreactive axon terminals with the mor-
phological features of thalamic axons form asymmetric synapses onto the dendritic
spines of pyramidal neurons (Fig. 17.2A) (Melchitzky et al., 1999). Although most
axon terminals arising from thalamic nuclei contact the dendritic spines of pyrami-
dal neurons, the dendrites of gamma-aminobutyric acid (GABA)-ergic cells are also
targets of thalamic axons. For example, dendritic shafts of PV-containing GABA
neurons in the DLPFC receive asymmetric synapses from PV-immunoreactive axon
terminals, which most likely arise from the MDN (Melchitzky et al., 1999).
In the DLPFC, spines on the basilar dendrites of layer 3 pyramidal neurons are
among the targets of afferents from the MDN. Layer 3 pyramidal neurons play a
central role in the flow of information both between and within cortical regions.
314 D. S. Melchitzky and D. A. Lewis

Fig. 17.2. Axon terminals in the dorsolateral prefrontal cortex (DLPFC). (A) Electron micrograph of a
parvalbumin-immunoreactive thalamic axon terminal (PV-T) in layer 4 of the DLPFC forming
an asymmetric, excitatory synapse onto a dendritic spine(s). (B) Electron micrograph of a
biotinylated dextran amine-labeled intrinsic axon terminal (BDA-T) in DLPFC layer 3 forming
an asymmetric synapse onto a parvalbumin-containing dendritic shaft (PVd ). (Adapted from
Melchitzky et al., 1999 (A) and 2001(B).)
315 Development of thalamocortical circuitry

In addition to furnishing projections through the white matter to other cortical

regions (Fig. 17.3B), these neurons have long-range intrinsic axon collaterals that
travel horizontally through the gray matter and give rise to stripe-like clusters of
axon terminals in layers 1–3 (Fig. 17.3A) (Levitt et al., 1993; Pucak et al., 1996). The
dendritic spines of other pyramidal neurons constitute the postsynaptic targets of
almost all (95%) of both the white matter and the long-range intrinsic projections of
these neurons (Melchitzky et al., 1998). These findings suggest that specific groups
of pyramidal neurons, clustered as stripes that are reciprocally interconnected by
long-range axon collaterals, form functional modules that may serve to recruit
and/or coordinate the activity of specific, spatially segregated populations of DLPFC
pyramidal neurons (Fig. 17.3C, D).
Layer 3 pyramidal neurons also provide local axon collaterals that arborize and
terminate near (within 300
m) the parent cell body, thus contributing to the
regulation of the neuronal activity of a given DLPFC stripe. In contrast to the long-
range axons, the targets of the local axon terminals are equally divided between
dendritic shafts and dendritic spines (Melchitzky et al., 2001). Furthermore, the
PV-containing subclass of GABAergic neurons (Fig. 17.2B) are the primary target
of these local inputs to dendritic shafts (Melchitzky and Lewis, 2003). Given the
potent inhibitory inputs that PV-containing cells provide to the soma and the axon
initial segment (AIS) of pyramidal neurons (Condé et al., 1994), these data suggest,
at least within the middle layers of the monkey DLPFC, that the connectivity of
PV-containing neurons enables them to serve a critical role in feedback inhibition,
constraining the propagation of pyramidal cell excitation both locally and at a
distance.
The dendritic spines of layer 4 pyramidal neurons may also be targets of afferents
from the MDN. Such contacts would result in a complementary activation of the
DLPFC compared with that resulting from the MDN inputs to layer 3 pyramidal
neurons. For example, unlike the horizontally oriented axon collaterals of layer
3 pyramidal neurons (Fig. 17.3C), axons emanating from layer 4 pyramidal cells
have a distinct and restricted vertical orientation, projecting to other cortical layers
in the same column (Levitt et al., 1993). Thus, MDN inputs to layer 4 pyramidal
neurons result in the propagation of excitation within a column of the DLPFC,
whereas those to layer 3 pyramidal cells produces a more widespread excitation to
a spatially distributed network of stripes (see Fig. 17.3D).
The last link in MDN–DLPFC circuitry involves the reciprocal excitatory projec-
tions from DLPFC back to MDN (Fig. 17.1). The majority of MDN-projecting cells
are small pyramidal neurons located in superficial layer 6, with small pyramidal
cells in deep layer 6 and large pyramidal cells in layer 5 making up the remainder
(Erickson and Lewis, 2000; Giguere and Goldman-Rakic, 1988). The corticotha-
lamic axons target both the projection neurons of the MDN and the GABAergic
316 D. S. Melchitzky and D. A. Lewis
317 Development of thalamocortical circuitry

neurons of the reticular thalamic nucleus, which, in turn, provide inhibitory regu-
lation of MDN projection neurons (Fig. 17.1E).
Unfortunately, attempts to explore the functional properties of the circuitry
between the MDN and DLPFC face certain limitations. First, because they lack direct
connections with the periphery, it is very difficult to manipulate the inputs to or
intrinsic components of MDN–DLPFC circuitry; therefore, most of our knowledge
base involves inferences from sensory thalamocortical systems. For example, in con-
trast to the deprivation or manipulation of visual inputs that have been used to study
the functional properties of lateral geniculate thalamus and primary visual cortex
circuitry, such manipulations cannot be made to study the connections between
the MDN and the DLPFC. Second, the majority of studies investigating thalamo-
corticothalamic circuitry have been performed in non-primates, such as rodents
and cats. Given the apparent differences in DLPFC circuitry between rodents and
primates (Lewis, 2002), extrapolation of rodent data to primate systems can only
be made with a number of caveats.

Development of thalamocortical circuitry

Our knowledge of the development of the connections between the MDN and
DLPFC are subject to the same limitations: most studies are performed in rodents
and focus on the sensory thalamic systems. Therefore, the following account of the

Fig. 17.3. Neuron circuits in the dorsolateral prefrontal cortex (DLPFC). (A, B) Darkfield photomicro-
graphs of biotinylated dextran amine (BDA)-labeled axons resulting from an injection site in
layer 3 of monkey DLPFC. Horizontally oriented, long-range intrinsic axon collaterals of layer
3 pyramidal neurons travel through the gray matter (A: arrows) and form a stripe-like cluster
of labeled axons and terminals confined to layers 1–3 (A). The labeled associational axons
from the same injection site travel through the white matter (B: arrows) and form a dense
cluster that extends across all cortical layers (B). PS indicates principal sulcus. (Scale bar =
100 m.) (C) Schematic diagram illustrating the spatial distribution of the three types of axon
terminal furnished by layer 3 pyramidal neurons in monkey DLPFC. BDA-labeled axons of
pyramidal neurons terminate nearby (local) or after traveling horizontally through the gray
matter (intrinsic) or through the white matter (associational) to form stripes of labeled
terminals. (D) Tangential reconstruction of all stripe-like clusters of retrogradely labeled
neurons arising from a single injection in monkey DLPFC. On this unfolded map, the PS is
oriented as a vertical line, and all labeled structures and other sulci are shown relative to
the PS. The location of the injection site is indicated by asterisk. Intrinsic stripes are shaded
in black, and associational stripes are cross-hatched. Note that the associational projections
form several distinct groups of stripes located at varying distances from the intrinsic stripes.
ASs, superior ramus of arcuate sulcus; CS, cingulate sulcus. (Adapted from Melchitzky et al.,
1998 (A, B) and 2001 (C) and Pucak et al., 1996 (D).)
318 D. S. Melchitzky and D. A. Lewis

development of the connections between MDN and DLPFC has been extrapolated
from data from primary sensory regions in primates and rodents. In primates, the
growth of thalamic axons towards the cortical plate (a dense sheet of immature
neurons that, with differentiation, acquire the characteristic laminated appearance
of the mature cortex) occurs relatively early in the gestation period of 165 days for
macaque monkeys. For example, axons from the lateral geniculate nucleus begin to
approach the area of the primary visual cortex by embryonic day 78 (Rakic, 1977).
Although the thalamic axons begin their journey towards the developing cerebral
cortex early, they appear to “wait” for an extended time in the subplate (composed
of loosely packed neurons located below and born prior to those that form the
cortical plate) before invading the cortical plate. During this “waiting period,”
thalamic axons appear to form transient, synapse-like structures with subplate
neurons, and such contacts appear to be critical for the subsequent formation
of proper connections between thalamic axons and cortical neurons. For example,
following lesions of the subplate, axons from the lateral geniculate nucleus continue
to grow past the visual cortex, suggesting that subplate neurons are required for
these thalamic axons to recognize their appropriate cortical target (Ghosh et al.,
1990). After this “waiting period,” thalamic axons make a second growth spurt
to enter the cortical plate, such that, in the monkey, thalamic axons enter the
visual cortex by embryonic day 124. This foray of thalamic axons into the cortical
plate is somewhat diffuse. For example, in monkey visual cortex, the presence of
ocular dominance columns is not apparent initially and segregation of fibers into
these columns begins to appear 3 weeks before birth (Rakic, 1977). Therefore, the
formation of thalamocorticothalamic circuitry could be affected by adverse events
occurring over a broad period of prenatal development.
During postnatal development, substantial changes transpire in the develop-
mental trajectory for synaptic density in monkey DLPFC observed via electron
microscopy (Bourgeois et al., 1994). The number of asymmetric, excitatory synapses
on dendritic spines increases rapidly between birth and 2 months of age; it then
remains at this relatively high level until adolescence, when the number declines to
adult levels (Fig. 17.4A). A similar process of synaptic overproduction and elimi-
nation also occurs in human DLPFC (Huttenlocher, 1979). In addition, imaging
studies of the DLPFC in children and young adults demonstrate a similar pattern of
initial gray matter growth in childhood followed by gray matter reduction starting
in adolescence (Giedd, 1999; Sowell et al., 2001).
Developmental changes in markers of synaptic inputs to the pyramidal neurons
that participate in thalamocorticothalamic connections occur in a similar fashion.
For example, dendritic spines are the primary sites of excitatory synaptic inputs to
pyramidal cells, and changes in spine number appear to reflect parallel alterations in
excitatory inputs to these neurons (Lund and Holbach, 1991). In monkey DLPFC,
319 Development of thalamocortical circuitry

Fig. 17.4. Age-related changes. (A) Changes during development in the densities of asymmetric exci-
tatory (•) and symmetric inhibitory (◦) synapses in area 46 of monkey dorsolateral prefrontal
cortex (DLPFC). (B) Age-related changes in the mean densities of spines on the dendrites of
layer 3 pyramidal neurons () and of parvalbumin (PV)-containing chandelier neuron axon
cartridges (◦) in layer 3 of monkey DLPFC. Note the parallel time course of developmental
changes in these markers of excitatory and inhibitory inputs to layer 3 pyramidal neurons.
(Adapted from Bourgeois et al., 1994 (A) and Anderson et al., 1995 (B).)

the density of dendritic spines on layer 3 pyramidal neurons undergoes substantial

changes during postnatal development (Fig. 17.4B) (Anderson et al., 1995). The
number of spines rapidly increases between birth and 2–3 months of age and this
high density of spines, approximately 50% higher than at birth, is maintained until
about 18 months of age. Layer 3 pyramidal neurons undergo spine attrition, again by
approximately 50%, until adult levels are achieved by 5 years of age. Although spines
exhibit remarkably rapid changes under certain conditions, recent in vivo two-
photon imaging studies in transgenic mice expressing yellow fluorescent protein
demonstrate that the vast majority of dendritic spines are stable in adult animals
(Grutzendler et al., 2002).
As described above, the axons of DLPFC layer 3 pyramidal neurons are a major
source of inputs to the dendritic spines of other layer 3 pyramidal cells. Inter-
estingly, limited data suggest that pruning of the long-range intrinsic collaterals
may make a greater contribution to the adolescence-related reduction in excitatory
inputs to dendritic spines than do associational projections. For example, the den-
sities of varicosities and axonal branch points (measures of the relative number of
synapses and degree of arborization, respectively) decreased by approximately half
for long-range intrinsic collaterals during adolescence (Woo et al., 1997). Interest-
ingly, computational model approaches support the idea that excessive pruning of
such intrinsic connections may contribute to the symptomatology of schizophrenia
(Siekmeier and Hoffman, 2002).
320 D. S. Melchitzky and D. A. Lewis

Although the overall density of DLPFC inhibitory synapses does not change
postnatally (Fig. 17.4A) (Bourgeois et al., 1994), functional markers of DLPFC
GABA circuits exhibit substantial postnatal changes. These changes are particularly
prominent in the chandelier class of GABAergic neurons, whose axon terminals are
vertically aligned as morphologically distinct structures (termed “cartridges”) that
form symmetric, inhibitory synapses on the AIS of pyramidal neurons (DeFelipe
et al., 1985). Thus, chandelier neurons provide potent inhibitory regulation of
pyramidal neuron output. Chandelier neurons express the calcium-binding pro-
tein PV (Lewis and Lund, 1990), and PV-containing axon cartridges of chandelier
neurons located in DLPFC layer 3 undergo substantial changes during postnatal
development. As shown in Figure 17.4B, the density of PV-positive cartridges in
monkey DLPFC increases rapidly during the first 3 months of postnatal devel-
opment, undergoes a plateau period that lasts until at least 1.5 years of age, and
then declines during the peripubertal age range to stable adult values (Anderson et
al., 1995). The close temporal match between the developmental changes in spine
density on layer 3 pyramidal neurons and that of PV-containing chandelier neuron
axon cartridges, reflecting a coordination between excitatory and inhibitory inputs,
could be achieved via the synaptic contacts of the axon collaterals of layer 3 pyra-
midal cells onto the dendrites of PV-containing chandelier neurons (Melchitzky
and Lewis, 2003).
Both pre- and postsynaptic markers of GABA neurotransmission at the AIS of
pyramidal neurons also undergo postnatal developmental changes. Like PV, anti-
bodies against the GABA membrane transporter (GAT-1) label the axon cartridges
of chandelier cells. The density of GAT-1-positive cartridges rapidly increases from
birth to 3 months of age, when maximal levels are achieved (Cruz et al., 2003).
GAT-1-positive cartridge density is maintained at these high levels until adoles-
cence, after which a rapid decline brings the density to adult levels. In contrast,
immunoreactivity for the GABAA α 2 -receptor at postsynaptic AIS is greatest at
birth, after which the density of GABAA α 2 -receptor-immunoreactive AIS steadily
decreases to adult levels (Cruz et al., 2003).
The specificity of these postnatal developmental changes in chandelier neuron
inputs to pyramidal cells is illustrated by comparison with the development of
the axon terminals of another subclass of PV-expressing GABA cells, the wide-
arbor neuron. For example, the density of PV-positive axon terminals arising from
wide-arbor cells is low at birth in monkeys and then increases 10-fold, reaching
adult levels approximately 3–4 years of age (Erickson and Lewis, 2002). Because
changes in activity level are associated with corresponding changes in PV content
(Heizmann, 1984), these findings suggest that chandelier neurons provide intra-
columnar inhibition at early postnatal ages, whereas wide-arbor cells supply inter-
columnar inhibition later during postnatal development (Erickson and Lewis,
321 Development of thalamocortical circuitry

2002). This gradual shift from intra- to intercolumnar inhibition may be related to
the late emergence of mature functional properties in the DLPFC, such as spatially
tuned delay-period activity between adjacent GABA and pyramidal neurons (Rao
et al., 1999).

Dysfunction of thalamocortical circuitry in schizophrenia

Both imaging and postmortem studies have revealed abnormalities in the thala-
mus of subjects with schizophrenia. For example, a recent meta-analysis examining
15 magnetic resonance studies of thalamic volume found significant reductions
(composite effect size of −0.29) in schizophrenia (Konick and Friedman, 2001).
Furthermore, a correlation between thalamic volumes and prefrontal white matter
has been reported in individuals with schizophrenia but not in normal subjects
(Portas et al., 1998), suggesting that reductions in thalamocortical projection neu-
rons may contribute to the decrease in DLPFC volume. In addition, two recent
imaging studies that subdivided the thalamus support a selective reduction in MDN
volume (Byne et al., 2001; Gilbert et al., 2001). Reductions in thalamic volume have
also been reported in postmortem studies (Byne et al., 2002; Pakkenberg, 1990;
Popken et al., 2000), and these changes may be localized to MDN and related asso-
ciation relay nuclei (e.g. anterior nucleus and pulvinar) while sparing the sensory-
and motor-related thalamic nuclei.
These reductions in MDN volume have been associated with a decreased number
of MDN neurons in schizophrenia (Fig. 17.5) (Byne et al., 2002; Pakkenberg, 1990;
Popken et al., 2000; Young et al., 2000). Furthermore, reduced neuronal numbers in
the parvocellular and densocellular portions of the MDN, which furnish projections
to the DLPFC, have been reported, whereas neuronal number was unchanged in the
magnocellular region, which is connected with the medial and orbital prefrontal
regions (Popken et al., 2000). In addition, decreased somal volume in the MDN
has also been reported (Byne et al., 2002). However, these postmortem studies
have generally involved older individuals with schizophrenia, making it difficult
to determine whether the reductions in MDN neuron number are a cause or a
consequence of MDN–DLPFC circuitry dysfunction.
Neuroimaging studies have revealed a 3–5% reduction in cerebral gray matter
in subjects with schizophrenia (Pearlson and Marsh, 1999), with areas such as the
DLPFC having the largest reductions (McCarley et al., 1999). Such reductions in
gray matter may reflect decreases in synaptic connections in the DLPFC (Selemon
and Goldman-Rakic, 1999). This abnormal synaptic connectivity may involve dis-
turbances in afferent projections to the DLPFC as well as in excitatory and/or
inhibitory neurons within the DLPFC.
322 D. S. Melchitzky and D. A. Lewis

Fig. 17.5. Schematic diagram summarizing the findings from postmortem studies showing alterations
in mediodorsal nucleus of the thalamus (MDN) and dorsolateral prefrontal cortex (DLPFC)
connectivity in schizophrenia. (1) The MDN exhibits a 30–35% reduction in neuron number
and (2) axon terminals from the MDN are reduced by 25% in the thalamic recipient layers
in the DLPFC. (3) Dendritic spine density is decreased by 25% on pyramidal neurons in
deep layer 3 of DLPFC, and (4) the somal size of pyramidal neurons is reduced by 10%.
(5) A subpopulation of gamma-aminobutyric acid (GABA)-ergic neurons exhibit decreased
glutamic acid decarboxylase isoform of 67 kDa (GAD67 ), GABA membrane transporter
1 (GAT-1) and parvalbumin (PV) mRNAs, whereas (6) calretinin (CR) mRNA levels are
unchanged. Therefore, the affected subclass of GABA cells is likely to be the PV-containing
chandelier neuron and not the CR-containing double bouquet cell, an interpretation sup-
ported by (7) the decreased density of GAT1-containing axon cartridges and (8) the
increased density of GABAA α 2 -receptor-immunoreactive pyramidal neuron axon initial seg-
ments. (9) In layer 6, the location of the pyramidal neurons that project to MDN, the density
of dopamine (DA)-containing axons is decreased. (Adapted from Lewis and Lieberman,
2000.)

Evidence for abnormalities in MDN to DLPFC projections includes reductions

of both pre- and postsynaptic markers for these axons. For example, the density of
a marker of axonal varicosities, presumably originating from the MDN, is reduced
only in thalamic recipient layers (deep layers 3 and 4) in the DLPFC (Fig. 17.5)
(Lewis et al., 2001). Furthermore, because dendritic spine integrity depends on
intact presynaptic inputs (see above), reductions in MDN axon projections may
323 Development of thalamocortical circuitry

also cause reduced dendritic spine density in the thalamic recipient layers in the
DLPFC. Indeed, in schizophrenia, spine density is decreased on the basilar dendrites
of DLPFC layer 3 pyramidal cells (Glantz and Lewis, 2000). Furthermore, the largest
reduction in spine density was for those pyramidal cells whose basilar dendrites
extend through the thalamic recipient zone. In addition, the somal size of deep layer
3 pyramidal neurons is reduced in the DLPFC of schizophrenia subjects (Fig. 17.5)
(Pierri et al., 2001; Rajkowska et al., 1998), a finding that may reflect deafferentation.
However, only a subpopulation of deep layer 3 pyramidal neurons appear to be
reduced in somal size (Pierri et al., 2003), making it unlikely that a reduction in
thalamic inputs accounts for this observation.
The presence of reciprocal excitatory connections between the DLPFC and MDN
raises the question of whether corticothalamic projections are also compromised.
Although one study reported a decreased neuronal density in layer 6 of the DLPFC
(Benes et al., 1986), the primary site of origin of corticothalamic projections, other
studies of neuronal size and number have not revealed any changes in this layer
(Rajkowska et al., 1998). These findings suggest that abnormalities in DLPFC effer-
ent projection neurons do not contribute significantly to abnormalities in MDN–
DLPFC connectivity. However, the density of dopamine-containing projections to
the DLPFC, which principally target pyramidal cells, is reduced in layer 6 in sub-
jects with schizophrenia (Fig. 17.5) (Akil et al., 1999). Therefore, although layer 6
pyramidal cells appear to be unaffected structurally in schizophrenia, more subtle
disturbances in afferents that could affect the function of these neurons may be
present in schizophrenia.
A number of findings consistent with functional anomalies in DLPFC GABA
interneurons have been reported in schizophrenia, and recent studies have
revealed that a subset of these neurons, the chandelier class, may be preferentially
affected (Volk and Lewis, 2003). Abnormal function of DLPFC chandelier cells in
schizophrenia is suggested by the finding that reductions in the mRNA level of
the 67 kDa isoform of glutamic acid decarboxylase (GAD67 ) (Akbarian et al., 1995;
Guidotti et al., 2000), an enzyme involved in GABA synthesis, are present in a subset
of GABA neurons found in the same cortical layers in which the chandelier cells
are distributed (Fig. 17.5) (Volk et al., 2000, 2001). In addition, decreased GAT-1
mRNA in a subset of GABA neurons is paralleled by an apparent reduction in GAT-1
protein in the axon cartridges of chandelier cells (Fig. 17.5) (Volk et al., 2001).
Finally, GAD67 mRNA appears to be selectively decreased in PV-containing GABA
neurons and not in neurons that contain another calcium-binding protein, calre-
tinin, which is not found in chandelier neurons (Fig. 17.5) (Hashimoto et al., 2003).
If these reductions in the expression of mRNAs and proteins reflect a deficit
in synaptic GABA, one would expect compensatory upregulation of postsynaptic
receptors. As noted above, immunoreactivity for the α 2 -subunit of the GABAA
324 D. S. Melchitzky and D. A. Lewis

receptor is preferentially localized to pyramidal AIS. Accordingly, a recent study

found an increased density of GABAA α 2 -receptor immunoreactive AIS in the
DLPFC of subjects with schizophrenia (Fig. 17.5)(Volk et al., 2002) that was
inversely related to GAT-1-positive cartridges in the same subjects.
Studies of the primate visual system have revealed that decreased input from
thalamic afferents results in reduced markers of cortical GABA neuron activity.
It has been hypothesized that DLPFC GABAergic connectivity may be altered as
a result of changes in afferent input from the MDN (Harrison and Lewis, 2003).
Interestingly, in monkey DLPFC, both the local axon collaterals of layer 3 pyramidal
neurons and a portion of the thalamic axon terminals form synapses onto the
dendrites of PV-containing GABA cells (Melchitzky et al., 2001), which include
the chandelier neuron subclass. Thus, a reduction in the projections from nearby
pyramidal cells and/or the thalamus could produce the observed reductions in
GABA neurotransmission in chandelier cells. However, large excitotoxic lesions of
the MDN in adult rats did not result in a reduction of prefrontal GAD67 mRNA
(Volk and Lewis, 2003). The failure of the MDN lesions to produce alterations in a
marker of GABA neurotransmission may reflect the age of the animal. Given that the
age of onset of schizophrenia is in adolescence, the cortical effects of MDN lesions
may only occur in young animals with immature thalamocortical connections. In
addition, a functional alteration in, rather than a loss of, thalamocortical projections
may be necessary to produce a change in GABA neurotransmission in the cortex.
For example, preventing visual input from reaching the lateral geniculate nucleus
by enucleation results in decreased cortical GAD67 mRNA levels, whereas lesions
of the thalamus do not (Jones et al., 1994). Despite these caveats, no available data
directly support the hypothesis that abnormalities in the MDN produce the DLPFC
GABA abnormalities in schizophrenia.

Conclusions and future directions

The findings reviewed above suggest that alterations in MDN–DLPFC circuitry
play a critical role in the pathophysiology of cognitive dysfunction in schizophre-
nia (see Fig. 17.5). In addition, given the protracted postnatal maturation of this
circuitry in the primate brain, normal development processes, or disturbances in
them, may be essential for these cognitive disturbances to become manifest. How-
ever, there are substantial gaps in our knowledge of the functional architecture
and development of MDN–DLPFC circuitry. For example, as noted above, rel-
atively few studies have examined details of the organization of MDN–DLPFC
circuitry in primates, and much of our understanding of the functional attributes
of this circuitry represents reasonable but speculative extrapolation from studies of
sensory thalamic systems. Furthermore, the findings implicating disturbances in
325 Development of thalamocortical circuitry

MDN–DLPFC circuitry in schizophrenia have not yet adequately addressed a num-

ber of critical issues, including the potential influence of confounds such as phar-
macological treatments, substance abuse, disease duration, etc. Finally, as in other
areas of schizophrenia research, our understanding of the mechanisms that give
rise to these cognitive disturbances awaits the implementation of model systems
that incorporate potential pathogenetic processes in the context of the protracted
maturation of primate MDN–DLPFC circuitry.

REFERENCES

Akbarian, S., Kim, J. J., Potkin, S. G., et al. (1995). Gene expression for glutamic acid decarboxylase
is reduced without loss of neurons in prefrontal cortex of schizophrenics. Arch Gen Psychiatry
52: 258–266.
Akil, M., Pierri, J. N., Whitehead, R. E., et al. (1999). Lamina-specific alteration in the dopamine
innervation of the prefrontal cortex in schizophrenic subjects. Am J Psychiatry 156: 1580–
1589.
Alexander, G. E. (1982). Functional development of frontal association cortex in monkeys:
behavioral and electrophysiological studies. Neurosci Res Prog Bull 20: 471–479.
Alexander, G. E., Goldman, P. S. (1978). Functional development of the dorsolateral prefrontal
cortex: an analysis utilizing reversible cryogenic depression. Brain Res 143: 233–249.
Anderson, S. A., Classey, J. D., Condé, F., Lund, J. S., Lewis, D. A. (1995). Synchronous develop-
ment of pyramidal neuron dendritic spines and parvalbumin-immunoreactive chandelier
neuron axon terminals in layer III of monkey prefrontal cortex. Neuroscience 67: 7–22.
Barbas, H., Haswell Henion, T. H., Dermon, C. R. (1991). Diverse thalamic projections to the
prefrontal cortex in the rhesus monkey. J Comp Neurol 313: 65–94.
Benes, F. M., Davidson, J., Bird, E. D. (1986). Quantitative cytoarchitectural studies of the cerebral
cortex of schizophrenics. Arch Gen Psychiatry 43: 31–35.
Bourgeois, J.-P., Goldman-Rakic, P. S., Rakic, P. (1994). Synaptogenesis in the prefrontal cortex
of rhesus monkeys. Cereb Cortex 4: 78–96.
Bunge, S. A., Dudukovic, N. M., Thomason, M. E., Vaidya, C. J., Gabrieli, J. D. E. (2002). Immature
frontal lobe contributions to cognitive control in children: Evidence from MRI. Neuron 33:
311.
Byne, W., Buchsbaum, M. S., Kemether, E., et al. (2001). Magnetic resonance imaging of the
thalamic mediodorsal nucleus and pulvinar in schizophrenia and schizotypal personality
disorder. Arch Gen Psychiatry 58: 133–140.
Byne, W., Buchsbaum, M. S., Mattiace, L. A., et al. (2002). Postmortem assessment of thalamic
nuclear volumes in subjects with schizophrenia. Am J Psychiatry 159: 59–65.
Casey, B. J., Giedd, J. N., Thomas, K. M. (2000). Structural and functional brain development
and its relation to cognitive development. Biol Psychol 54: 241–257.
Condé, F., Lund, J. S., Jacobowitz, D. M., Baimbridge, K. G., Lewis, D. A. (1994). Local circuit neu-
rons immunoreactive for calretinin, calbindin D-28k, or parvalbumin in monkey prefrontal
cortex: distribution and morphology. J Comp Neurol 341: 95–116.
326 D. S. Melchitzky and D. A. Lewis

Cruz, D. A., Eggan, S. M., Lewis, D. A. (2003). Postnatal development of pre- and post-synaptic
GABA markers at chandelier cell inputs to pyramidal neurons in monkey prefrontal cortex.
J Comp Neurol 465: 385–400.
DeFelipe, J., Jones, E. G. (1991). Parvalbumin immunoreactivity reveals layer IV of monkey
cerebral cortex as a mosaic of microzones of thalamic afferent terminations. Brain Res 562:
39–47.
DeFelipe, J., Hendry, S. H. C., Jones, E. G., Schmechel, D. (1985). Variability in the terminations of
GABAergic chandelier cell axons on initial segments of pyramidal cell axons in the monkey
sensory-motor cortex. J Comp Neurol 231: 364–384.
Diamond, A. (1985). Development of the ability to use recall to guide action, as indicated by
infants’ performances on AB. Child Devel 56: 868–883.
(2002). Normal development of prefrontal cortex from birth to young adulthood: cognitive
functions, anatomy, and biochemistry. In Principles of Frontal Lobe Function, ed. D. T. Stuss,
R. T. Knight. London: Oxford University Press, pp. 466–503.
Elvevåg, B., Goldberg, T. E. (2000). Cognitive impairment in schizophrenia is the core of the
disorder. Crit Rev Neurobiol 14: 1–21.
Erickson, S. L., Lewis, D. A. (2000). Prefrontal cortical inputs to monkey mediodorsal thalamus.
Soc Neurosci Abstr 26: 1237.
(2002). Postnatal development of parvalbumin- and GABA transporter-immunoreactive axon
terminals in monkey prefrontal cortex. J Comp Neurol 448: 186–202.
Funahashi, S., Bruce, C. J., Goldman-Rakic, P. S. (1989). Mnemonic coding of visual space in the
monkey’s dorsolateral prefrontal cortex. J Neurophysiol 61: 331–349.
Fuster, J. M. (1997). The Prefrontal Cortex: Anatomy, Physiology, and Neuropsychology of the Frontal
Lobe. Philadelphia, PA: Lippincott-Raven.
Fuster, J. J., Alexander, G. E. (1971). Neuron activity related to short-term memory. Science 173:
652–654.
Ghosh, A., Antonini, A., McConnell, S. K., Shatz, C. J. (1990). Requirement for subplate neurons
in the formation of thalamocortical connections. Nature 347: 179–181.
Giedd, J. N. (1999). Brain development during childhood and adolescence: a longitudinal MRI
study. Nat Neurosci 2: 861–863.
Giguere, M., Goldman-Rakic, P. S. (1988). Mediodorsal nucleus: areal, laminar, and tangential
distribution of afferents and efferents in the frontal lobe of rhesus monkeys. J Comp Neurol
277: 195–213.
Gilbert, A. R., Rosenberg, D. R., Harenski, K., et al. (2001). Thalamic volumes in patients with
first-episode schizophrenia. Am J Psychiatry 158: 618–624.
Glantz, L. A., Lewis, D. A. (2000). Decreased dendritic spine density on prefrontal cortical pyra-
midal neurons in schizophrenia. Arch Gen Psychiatry 57: 65–73.
Goldman-Rakic, P. S. (1987). Development of cortical circuitry and cognitive function. Child
Devel 58: 601–622.
(1994). Working memory dysfunction in schizophrenia. J Neuropsychiatry 6: 348–357.
(1995). Cellular basis of working memory. Neuron 14: 477–485.
Goldman-Rakic, P. S., Porrino, L. J. (1985). The primate mediodorsal (MD) nucleus and its
projection to the frontal lobe. J Comp Neurol 242: 535–560.
327 Development of thalamocortical circuitry

Grutzendler, J., Kasthuri, N., Gan, W. B. (2002). Long-term dendritic spine stability in the adult
cortex. Nature 420: 812–816.
Guidotti, A., Auta, J., Davis, J. M., et al. (2000). Decrease in reelin and glutamic acid
decarboxylase67 (GAD67 ) expression in schizophrenia and bipolar disorder. Arch Gen
Psychiatry 57: 1061–1069.
Harrison, P. J., Lewis, D. A. (2003). Neuropathology in schizophrenia. In Schizophrenia, 2nd edn,
ed. S. Hirsch, D. R. Weinberger. Oxford: Blackwell Science, Oxford University Press, pp.
310–325.
Hashimoto, T., Volk, D. W., Eggan, S. M., et al. (2003). Gene expression deficits in a subclass
of GABA neurons in the prefrontal cortex of subjects with schizophrenia. J Neurosci 23:
6315–6326.
Heizmann, C. W. (1984). Parvalbumin an intracellular calcium-binding protein. Distribution
properties and possible roles in mammalian cells. Experientia 40: 910–921.
Huttenlocher, P. R. (1979). Synaptic density in human frontal cortex: developmental changes and
effects of aging. Brain Res 163: 195–205.
Isseroff, A., Rosvold, H. E., Galkin, T. W., Goldman-Rakic, P. S. (1982). Spatial memory impair-
ments following damage to the mediodorsal nucleus of the thalamus in rhesus monkeys.
Brain Res 232: 97–113.
Jones, E. G., Hendry, S. H. C., DeFelipe, J., Benson, D. L. (1994). GABA neurons and their role
in activity-dependent plasticity of adult primate visual cortex. In Cerebral Cortex, Vol. 10:
Primary Visual Cortex in Primates, ed. A. Peters, K. S. Rockland. New York: Plenum Press,
pp. 61–140.
Konick, L. C., Friedman, L. (2001). Meta-analysis of thalamic size in schizophrenia. Biol Psychiatry
49: 28–38.
Kuroda, M., Yokofujita, J., Murakami, K. (1998). An ultrastructural study of the neural circuit
between the prefrontal cortex and the mediodorsal nucleus of the thalamus. Prog Neurobiol
54: 417–458.
Levitt, J. B., Lewis, D. A., Yoshioka, T., Lund, J. S. (1993). Topography of pyramidal neuron
intrinsic connections in macaque monkey prefrontal cortex (areas 9 and 46). J Comp Neurol
338: 360–376.
Lewis, D. A. (2000). Is there a neuropathology of schizophrenia? The Neuroscientist 6: 208–218.
(2002). Neural circuitry approaches to understanding the pathophysiology of schizophrenia.
In Neuropsychopharmacology: The Fifth Generation of Progress, ed. K. L. Davis, D. S. Charney,
J. T. Coyle, C. B. Nemeroff. Philadelphia, PA: Lippincott Williams and Wilkins, pp. 729–743.
Lewis, D. A., Levitt, P. (2002). Schizophrenia as a disorder of neurodevelopment. Annu Rev
Neurosci 25: 409–432.
Lewis, D. A., Lieberman, J. A. (2000). Catching up on schizophrenia: natural history and neuro-
biology. Neuron 28: 325–334.
Lewis, D. A., Lund, J. S. (1990). Heterogeneity of chandelier neurons in monkey neocor-
tex: corticotropin-releasing factor and parvalbumin immunoreactive populations. J Comp
Neurol 293: 599–615.
Lewis, D. A., Cruz, D. A., Melchitzky, D. S., Pierri, J. N. (2001). Lamina-specific reductions
in parvalbumin-immunoreactive axon terminals in the prefrontal cortex of subjects with
328 D. S. Melchitzky and D. A. Lewis

schizophrenia: evidence for decreased projections from the thalamus. Am J Psychiatry 158:
1411–1422.
Lund, J. S., Holbach, S. (1991). Postnatal development of thalamic recipient neurons in monkey
striate cortex: I. A comparison of spine acquisition and dendritic growth of layer 4C alpha
and beta spiny stellate neurons. J Comp Neurol 309: 115–128.
McCarley, R. W., Wible, C. G., Frumin, M., et al. (1999). MRI anatomy of schizophrenia. Biol
Psychiatry 45: 1099–1119.
Melchitzky, D. S., Lewis, D. A. (2003). Preferential targeting of parvalbumin interneurons by
local axon terminals of supragranular pyramidal neurons in monkey prefrontal cortex.
Cereb Cortex 13: 452–460.
Melchitzky, D. S., Sesack, S. R., Pucak, M. L., Lewis, D. A. (1998). Synaptic targets of pyramidal
neurons providing intrinsic horizontal connections in monkey prefrontal cortex. J Comp
Neurol 390: 211–224.
Melchitzky, D. S., Sesack, S. R., Lewis, D. A. (1999). Parvalbumin-immunoreactive axon terminals
in monkey and human prefrontal cortex: Laminar, regional and target specificity of type I
and type II synapses. J Comp Neurol 408: 11–22.
Melchitzky, D. S., Gonzalez-Burgos, G., Barrionuevo, G., Lewis, D. A. (2001). Synaptic targets
of the intrinsic axon collaterals of supragranular pyramidal neurons in monkey prefrontal
cortex. J Comp Neurol 430: 209–221.
Pakkenberg, B. (1990). Pronounced reduction of total neuron number in mediodorsal thalamic
nucleus and nucleus accumbens in schizophrenics. Arch Gen Psychiatry 47: 1023–1028.
Park, S., Holzman, P. S. (1992). Schizophrenics show spatial working memory deficits. Arch Gen
Psychiatry 49: 975–982.
Pearlson, G. D., Marsh, L. (1999). Structural brain imaging in schizophrenia: a selective review.
Biol Psychiatry 46: 627–649.
Pierri, J. N., Volk, C. L. E., Auh, S., Sampson, A., Lewis, D. A. (2001). Decreased somal size of
deep layer 3 pyramidal neurons in the prefrontal cortex in subjects with schizophrenia. Arch
Gen Psychiatry 58: 466–473.
(2003). Somal size of prefrontal cortical pyramidal neurons in schizophrenia: differential effects
across neuronal subpopulations. Biol Psychiatry 54: 111–120.
Popken, G. J., Bunney, Jr W. E., Potkin, S. G., Jones, E. G. (2000). Subnucleus-specific loss
of neurons in medial thalamus of schizophrenics. Proc Natl Acad Sci USA 97: 9276–
9280.
Portas, C. M., Goldstein, J. M., Shenton, M. E., et al. (1998). Volumetric evaluation of the
thalamus in schizophrenic male patients using magnetic resonance imaging. Biol Psychiatry
43: 649–659.
Pucak, M. L., Levitt, J. B., Lund, J. S., Lewis, D. A. (1996). Patterns of intrinsic and associational
circuitry in monkey prefrontal cortex. J Comp Neurol 376: 614–630.
Rajkowska, G., Selemon, L. D., Goldman-Rakic, P. S. (1998). Neuronal and glial somal size in
the prefrontal cortex: a postmortem morphometric study of schizophrenia and Huntington
disease. Arch Gen Psychiatry 55: 215–224.
Rakic, P. (1977). Prenatal development of the visual system in rhesus monkey. Philos Trans R Soc
Lond Ser B 278: 245–260.
329 Development of thalamocortical circuitry

Rao, S. G., Williams, G. V., Goldman-Rakic, P. S. (1999). Isodirectional tuning of adjacent

interneurons and pyramidal cells during working memory: evidence for microcolumnar
organization in PFC. J Neurophysiol 81: 1903–1916.
Selemon, L. D., Goldman-Rakic, P. S. (1999). The reduced neuropil hypothesis: a circuit based
model of schizophrenia. Biol Psychiatry 45: 17–25.
Siekmeier, P. J., Hoffman, R. E. (2002). Enhanced semantic priming in schizophrenia: a computer
model based on excessive pruning of local connections in association cortex. Br J Psychiatry
180: 345–350.
Sowell, E. R., Thompson, P. M., Tessner, K. D., Toga, A. W. (2001). Mapping continued brain
growth and gray matter density reduction in dorsal frontal cortex: inverse relationships
during postadolescent brain maturation. J Neurosci 21: 8819–8829.
Volk, D. W., Lewis, D. A. (2003). Effects of a mediodorsal thalamus lesion on prefrontal inhibitory
circuitry: implications for schizophrenia. Biol Psychiatry 53: 385–389.
Volk, D. W., Austin, M. C., Pierri, J. N., Sampson, A. R., Lewis, D. A. (2000). Decreased
GAD67 mRNA expression in a subset of prefrontal cortical GABA neurons in subjects with
schizophrenia. Arch Gen Psychiatry 57: 237–245.
(2001). GABA transporter-1 mRNA in the prefrontal cortex in schizophrenia: decreased expres-
sion in a subset of neurons. Am J Psychiatry 158: 256–265.
Volk, D. W., Pierri, J. N., Fritschy, J.-M., Auh, S., Sampson, A. R., Lewis, D. A. (2002). Reciprocal
alterations in pre- and postsynaptic inhibitory markers at chandelier cell inputs to pyramidal
neurons in schizophrenia. Cereb Cortex 12: 1063–1070.
Weinberger, D. R., Berman, K. F., Zec, R. F. (1986). Physiologic dysfunction of dorsolateral pre-
frontal cortex in schizophrenia. I. Regional cerebral blood flow evidence. Arch Gen Psychiatry
43: 114–124.
Woo, T.-U., Pucak, M. L., Kye, C. H., Matus, C. V., Lewis, D. A. (1997). Peripubertal refinement
of the intrinsic and associational circuitry in monkey prefrontal cortex. Neuroscience 80:
1149–1158.
Young, K. A., Manaye, K. F., Liang, C.-L., Hicks, P. B., German, D. C. (2000). Reduced number of
mediodorsal and anterior thalamic neurons in schizophrenia. Biol Psychiatry 47: 944–953.
 18

X chromosome, estrogen, and brain

development: implications for
schizophrenia
Michael Craig, William Cutter, Ray Norbury, and Declan Murphy
Institute of Psychiatry, King’s College, London, UK

The “neurodevelopmental theory” is now regarded by many psychiatrists as the

dominant explanatory model of schizophrenia (O’Connell et al., 1997). In order
to evaluate this model critically, it is important to understand how the normal
brain develops and changes across the lifespan. Nonetheless, there are relatively few
studies by schizophrenia researchers on the biological factors affecting maturation
of brain regions implicated in psychosis. The determinants of normal neurode-
velopment, and hence brain function, are most likely multifactorial and include
both genetic and environmental factors. For example, recent work has examined
the role of the gene for the catechol-O-methyltransferase (COMT) (Egan et al.,
2001) and a polymorphism of the promoter region of the gene for the serotonin
5-hydroxytryptamine (5-HT) transporter (5-HTTLPR) (Hariri et al., 2002) in nor-
mal brain function (e.g. amygdala function during processing of facial emotion),
and the effect of deletions at q11 chromosome 22 on the anatomy of prefrontal
and medial temporal regions (van Amelsvoort et al., 2001a). Other important fac-
tors, to be reviewed in this chapter, include the sex chromosomes and sex steroids.
We do not suggest that schizophrenia is an X-linked disorder, or that sex steroids
cause schizophrenia (just as schizophrenia is unlikely to be explained solely by
variations in the COMT gene or the serotonin transporter promoter polymor-
phism). Rather, we will offer evidence that sex chromosomes and sex steroids
(in particular estrogen) significantly modulate the structure and function of nor-
mal brain, and this needs to be considered when forming developmental theories
of schizophrenia, when attempting to explain sex differences in schizophrenia,
and when considering the role of sex steroids in the genesis and treatment of
schizophrenia.

Neurodevelopment and Schizophrenia, ed. Matcheri S. Keshavan et al. Published by Cambridge University
Press.

C Cambridge University Press 2004.

330
331 X chromosome, estrogen, and brain development

Sex differences in epidemiology and clinical features of schizophrenia

There is a significantly greater male to female ratio in schizophrenia when “restric-
tive” (e.g. Feighner) diagnostic criteria are used (Castle et al., 1993) and in patients
who have an onset over the age of 40 years (late-onset) (Howard et al., 2000). How-
ever, the male to female ratio approaches unity when less restrictive criteria are used
(e.g. Research Diagnostic Criteria; Spitzer et al., 1978).
There may also be gender differences in age of onset. A later age of onset of
schizophrenia amongst women has been a consistent finding across many studies.
Females are generally reported as having a mean age of onset that is approximately
3–4 years later than males, regardless of the definition applied (Hafner et al., 1993,
1994, 1998a,b). Also, men are reported to have a peak in illness onset between 15
and 25 years of age, with a steady decline after 30 years. In contrast, women have a
broader onset peak between 15 and 30 years, with a second peak between 45 and
49 years, and perhaps a third peak amongst women over 75 years of age. These
findings have been replicated across a variety of cultures, and in both rural and
urban settings (Castle and Murray, 1993; Jablensky et al., 1992).
There is a high degree of overlap in the clinical symptoms between the gen-
ders; however, women are generally reported to present with a less-severe form of
schizophrenia, with more affective symptoms, but fewer negative symptoms and
hospitalizations (Leung and Chue, 2000).

Anatomical differences between men and women with schizophrenia

Studies on sex differences in the brain anatomy of patients with schizophrenia have
reported that males have a significantly smaller temporal lobe volume than females
(Shenton et al., 1992) but a greater ventricular–brain ratio (Andreasen et al., 1986,
1990a,b; Flaum et al., 1990; Harvey et al., 1990; Nopoulos et al., 1997; Shelton et al.,
1988; Williams et al., 1985) and right–left ventricular asymmetry (Haas et al., 1989,
1991). However, most of these studies were relatively small and some have not been
replicated; for example, ventricular dilatation has been reported to be significantly
more common in women than men in some studies (Andreasen et al., 1982; Gur
et al., 1994; Nasrallah et al., 1990; Vazquez-Barquero et al., 1995) but not others
(Lauriello et al., 1997). Moreover, studies of gender differences in neuropathol-
ogy, or symptoms, need to take account of normal gender differences in the brain.
For example, postmortem studies suggested that early onset and negative symp-
toms in schizophrenia are associated with a thickened corpus callosum (Coger and
Serafetinides, 1990). This would predict that women with schizophrenia have a thin-
ner corpus callosum than men with schizophrenia because women generally have
fewer negative symptoms and a later illness onset than men. However, there is no
332 M. Craig, W. Cutter, R. Norbury, and D. Murphy

consistent evidence for gender differences in callosal thickness in schizophrenia

(Salem and Kring, 1998). Moreover, there are significant gender differences in devel-
opment and aging of many brain regions (including hippocampus and frontal lobe;
see below), and this needs to be controlled for in studies of schizophrenia.

Normal brain development and aging

Myelination starts near the end of the second trimester of fetal development and
in some brain regions (e.g. hippocampus) may continue until the sixth decade or
beyond (Benes et al., 1994). Myelination progresses from inferior to superior and
posterior to anterior and the brainstem and cerebellar regions are myelinated prior
to the cerebral hemispheres, and the frontal lobes and hippocampus are myelinated
last. This suggests that humans, unlike other primates, have non-synchronous devel-
opment of the cerebral cortex (i.e. different areas of the cortex develop at different
times). Furthermore, females have been reported to have greater myelination than
males during adolescence in areas such as the superior medullary lamina (Benes
et al., 1994). It has been suggested that earlier myelination in females may con-
tribute to their greater early achievement in language skills and reading. Similarly,
later myelination in males may contribute to their increased vulnerability to various
forms of psychopathology (Benes, 1998).
Non-synchronous development of the cerebral cortex has recently been con-
firmed by longitudinal magnetic resonance imaging (MRI) studies, and there is
preliminary evidence for gender differences in maturation of brain regions impli-
cated in schizophrenia. For example, in healthy populations, the volume of the
frontal and temporal lobes peaks at about 12 years and 16 years, respectively,
whereas the occipital lobe continues to develop into the twenties (Giedd et al.,
1999). Furthermore, during childhood and adolescence, volume of frontal gray
matter peaks approximately 1 year earlier in females than males: the left amygdala
increases significantly only in males but the right hippocampus increases signif-
icantly only in females (Giedd et al., 1996). In the age range 7–17 years, males
also show a significantly greater loss of cerebral gray matter volume and increase
in white matter volume and corpus callosal area compared with females (De Bellis
et al., 2001).
There are also significant gender differences in brain maturation in mid-life
and later years. Studies using MRI have reported that age-related volume loss is
significantly greater in the hippocampus and parietal lobes in women compared
with men, whereas age-related volume loss in men is greater in the frontal and
temporal lobes (Murphy et al., 1996). A positron electron tomography (PET) study
of 120 healthy people reported that women have a greater age-related metabolic
decline in the hippocampus and thalamus than men, and that age-related decline
333 X chromosome, estrogen, and brain development

in brain metabolism is asymmetric in males but symmetric in females (Murphy

et al., 1996). Consequently, there is now substantial evidence for gender differences
in the development and aging of brain regions implicated in schizophrenia (see
Ch. 2 for a further discussion). It has, therefore, been suggested that this may modify
symptom expression (Murphy et al., 1996) and that the earlier age of onset and
increase in severity of schizophrenia in males may be attributable to an accelerated
rate of pruning in male gray matter volume during the adolescent period (De Bellis
et al., 2001).

Development and aging of the serotonergic and dopaminergic systems

In all mammals, the highest functional status of the serotonin system is reached
earliest in development: and adult levels of the system are much lower than in the
younger animal (Goldman-Rakic and Brown, 1982; Lidow et al., 1991). In human
brain, serotonergic neurons are first evident by 5 weeks of gestation (Sundstrom
et al., 1993) and increase rapidly through the tenth week (Levallois et al., 1997).
By 15 weeks of gestation, the typical organization of serotonin cell bodies into the
raphe nucleus can be seen (Takahashi et al., 1986). In humans, there are higher
levels of the serotonin metabolite 5-Hydroxyindolacetic acid in children than in
adults; serotonin levels increase throughout the first 2 years of life and then decline
to adult levels by 5 years of age (Hedner et al., 1986; Seifert et al., 1980).
Serotonin acts as a trophic or differentiation factor in addition to its role as a
neurotransmitter, and alteration of serotonin during brain development alters neur-
onal differentiation (reviewed by Whitaker-Azmitia, 2001). For example, serotonin
plays a role in modulation of synaptogenesis in sensory cortices postnatally in the rat
(Cases et al., 1995; D’Amato et al., 1987), and this may represent transient expression
of the high-affinity serotonin transporter and vesicular monoamine transporter by
glutamatergic thalamocortical neurons (Bennett-Clarke et al., 1996; Lebrand et al.,
1996). Decreased or increased brain serotonin during this period of development
results in disruption of brain development (Bennett-Clarke et al., 1994; Cases et al.,
1995, 1996). For example, depletion of serotonin delays the development (Blue
et al., 1991) and decreases the size (Bennett-Clarke et al., 1994) of barrel fields in
rat somatosensory cortex; it prolongs cell division and increases neuronal cell num-
bers in hippocampus, superior colliculus, and several thalamic nuclei (Lauder and
Krebs, 1978); and it decreases the number of hippocampal dendritic spines (Yan
et al., 1997). In contrast, increased serotonin during development results in
increased tangential arborization of somatosensory axons and blurring of the
boundaries of the cortical barrels (Cases et al., 1996).
Later life is associated with a reduction in serotonergic responsivity to neuroen-
docrine probes (e.g. prolactin response to fenfluramine), and this decline is more
334 M. Craig, W. Cutter, R. Norbury, and D. Murphy

pronounced in women (Lerer et al., 1999). Also there are age-related reductions in
the number of cortical serotonin receptors 5-HT1A , 5-HT1Bδ : and 5-HT2A in the
frontal and occipital lobes and the hippocampus (Meltzer et al., 1998).
Dopaminergic neurons begin to develop by 6–8 weeks of gestation in humans
(Sundstrom et al., 1993), and development begins earlier in females than in
males (Herlenius and Lagercrantz, 2001). Interestingly, with the onset of puberty,
dopaminergic activity (unlike serotoninergic) increases in some cortical regions,
especially the frontal cortex (Spear, 2000). It has been hypothesized that the
increased risk for the onset of psychosis during adolescence/early adulthood may
be linked to the elevation in dopaminergic activity (Walker, 1994).
PET studies of age-related differences in the dopaminergic system reported a
reduction in dopamine D2 receptor concentration of the caudate nuclei and the
frontal lobes in both men and women (Wong et al., 1984). Furthermore, this decline
is more pronounced in women than in men (Wong et al., 1988).
In summary, therefore, there are age and gender differences in the maturation of
brain regions and neurochemical systems implicated in schizophrenia. What then
is the evidence that gender differences in the brain are caused by sex steroids or sex
chromosomes?

The role of estrogen in brain development and aging

Development
There is increasing evidence that estrogen modulates sexual differentiation of the
brain and affects neuronal development both directly and indirectly. For example,
receptors for estrogen and neurotrophins (such as nerve growth factor [NGF]) are
located on the same neurons in rodent basal forebrain, hippocampus, and cerebral
cortex, and this colocalization may be important for the long-term survival of
neurons (Toran-Allerand, 1996).
Traditionally, sexual differentiation of brain has been explained in terms of the
presence or absence of testosterone during the perinatal period (Phoenix et al.,
1959). Development of the female brain was perceived as a passive process occur-
ring by default in the absence of testosterone. However, the current view is that this
process may be more complex for two main reasons. First, in many brain regions
testosterone is converted (by the enzyme aromatase) to estradiol and, therefore,
“masculinization” is primarily attributable to the action of estrogen at estrogen
receptors. Females avoid masculinization in utero from high circulating levels of
maternal estrogen because α-fetoprotein binds to circulating estrogen and prevents
it from entering the neuron. Second, animal studies suggest that ovarian estro-
gen may have an effect independent of testosterone on corpus callosum size once
335 X chromosome, estrogen, and brain development

circulating α-fetoprotein has been “switched off ” (Fitch and Denenberg, 1998).
In vitro studies suggest that filopodial outgrowth (an important part of neu-
rotrophism because filopodia act as an early support for dendritic spine matu-
ration) occurs within minutes of exposure to 17β-estradiol (Brinton, 1993). The
mechanism by which estrogen initiates filopodial outgrowth is most likely direct
activation of the Rac1B G-protein, a member of the Rho family of GTPases involved
in actin organization (Dumontier et al., 2000). Further development of dendritic
spines appears to be mediated by estrogen activation of a Src tyrosine kinase that
phosphorylates the glutamate N-methyl-d-aspartate (NMDA) receptor (Yu et al.,
1997). For example, in cortical and hippocampal neurons without synaptic con-
tacts, 17β-estradiol induces a significant increase in the outgrowth of both macro-
and micromorphological features, which is blocked by a NMDA receptor antago-
nist. Activation of these intracellular signaling cascades appears to be dependent on
the subtype of estrogen receptor present: neuroblastoma cells with the α-subtype,
for example, are dependent on the Rac1B G-protein, whereas mechanisms appear
independent of Rac1B where the β-subtype occurs (Brinton, 2001).
These findings have been confirmed by in vivo studies. In the adult female rat,
the densities of dendritic spines and synapses on hippocampal CA1 pyramidal cells
fluctuate across the menstrual cycle with natural fluctuations in ovarian steroid
estradiol levels. Furthermore, these estradiol-induced dendritic changes are blocked
by treatment with an NMDA receptor antagonist (Woolley and McEwen, 1994).
In addition to direct effects on neurons, estrogen also acts with neurotrophins
(such as NGF) to stimulate nerve cell growth indirectly. NGF is essential for early
neuronal development and influences neuron differentiation and growth. Other
important neurotrophins are brain-derived neurotrophic factor (BDNF) and neu-
rotrophins 3 and 4/5 (NT-3 and NT-4/5). Binding of NGF and neurotrophins
to their cognate receptors results in the activation of mechanisms necessary for
growth and survival of neurites, as well as in stimulation of functions related
to neurotransmitter production and release. Importantly, receptors for estrogen
and neurotrophins are located on the same neurons in rodent basal forebrain,
hippocampus, and cerebral cortex (Toran-Allerand, 1996). This colocalization of
estrogen and neurotrophin receptors suggests that they act synergistically on the
same neuron to regulate expression of specific genes enhancing neuronal survival,
differentiation, and plasticity (Toran-Allerand, 1996).

Cognition and aging

The most robust effect of estrogen on cognitive function is most likely on
verbal memory. Prospective randomized studies of hormone replacement ther-
apy (HRT) versus placebo following total abdominal hysterectomy and bilateral
336 M. Craig, W. Cutter, R. Norbury, and D. Murphy

salpingoophorectomy report a significant positive effect of estrogens on verbal

memory (Phillips and Sherwin, 1992; Sherwin, 1988). Performance in some cogni-
tive tasks also varies as a function of menstrual cycle in healthy premenopausal
women. During the luteal phase (characterized by high levels of estrogen and
progesterone), verbal articulation is improved whereas spatial ability is decreased
(Hampson, 1990), a pattern that is reversed during the follicular phase (when there
is relatively low estrogen and progesterone). A similar pattern of cognitive perform-
ance is also observed if subjects are tested during the preovulatory estradiol surge
(to control for the potential effects of progesterone on cognitive performance),
suggesting that estrogen rather than progesterone is responsible for the observed
cognitive effects.
Functional imaging techniques have also been employed to assess the effects of
estrogen on networks subserving various aspects of cognitive function (reviewed by
Maki and Resnick, 2001). A recent randomized placebo-controlled crossover study
using functional MRI (Shaywitz et al., 1999) found estrogen-induced alterations
in brain activation patterns during encoding and retrieval of both verbal and non-
verbal stimuli. More recently, Maki and Resnick (2000) used PET and oxygen-15 to
examine longitudinal changes in regional cerebral blood flow over a 2 year interval
in women on and off HRT (both with and without adjuvant progesterone therapy).
Significant differences were found in right hippocampus, parahippocampal gyrus,
and left middle temporal gyrus. We recently reported that estrogen reduces age-
related differences in neuronal membrane breakdown (as measured by proton
magnetic resonance spectroscopy) in hippocampus and parietal lobe, and this was
related to memory function (Robertson et al., 2001) (Fig. 18.1). Therefore, there
is increasing evidence from in vivo brain imaging studies that estrogen modu-
lates cognitive function, cerebral blood flow, and membrane breakdown. However,
further prospective randomized studies are required.
Estrogen replacement therapy may also modulate aging of the serotonergic sys-
tem. For example, van Amelsvoort et al. (2001b) reported that postmenopausal
women who use estrogen replacement therapy have a significant reduction in age-
related differences in serotonergic responsivity (as measured by prolactin response
to a fenfluaramine challenge) (Fig. 18.2). Also both PET (Moses et al., 2000) and
single-photon PET studies (Travis et al., 1999) found that a short course of estradiol
increases cortical 5-HT2A receptor density in healthy postmenopausal women.
Estrogen can also modulate dopaminergic function. In animal studies, estro-
gen reduces dopamine concentration in the striatum and modulates the sensitivity
and number of dopamine receptors (Bedard et al., 1984; Foreman and Porter,
1980; Gordon et al., 1980). In human adults, dopamine D2 receptor concentration
reduces with increasing age, especially in frontal and basal ganglia regions (Wong
et al., 1984). Also, women exhibit a linear decline in caudate D2 density, whereas
337 X chromosome, estrogen, and brain development

5.0

Choline (mmol/l) 4.5

4.0

3.5

3.0

2.5
N
young on off

Fig. 18.1. Mean right hippocampal choline as an indication of neuronal membrane breakdown in
postmenopausal women taking estrogen replacement therapy (“on”) or never using such
therapy (“off”) plus that in premenopausal women (“young”). Bars indicate 95% confidence
internal. “Off” values versus “young,” p < 0.03; “off” values versus “on,” p < 0.04. (From
Robertson D. M. W., et al., Neurology 2001, 57(11): 2114–2117.)

men exhibit a decline best represented by a second-degree polynomial regression,

with a much steeper fall between 20 and 40 years of age. It has been suggested that
the later age of onset of schizophrenia in females than in males reflects a loss of
the antidopaminergic effect of estrogen as levels decline in older age. This hypoth-
esis is partially supported by the observation that psychotic symptoms emerge in
some females when estrogen levels are low, for example during the perimenopausal
(Castle and Murray, 1993; Jablensky et al., 1992) and postpartum periods (Kendell
et al., 1987), and during low estrogen phases of the menstrual cycle (Seeman and
Lang, 1990). Furthermore, adjunct estrogen treatment has been reported to improve
response to antipsychotic medication (Kulkarni et al., 2001).
Therefore, there is evidence that estrogen significantly modulates the devel-
opment and aging of brain regions and neurochemical systems implicated in
schizophrenia. However, there is little evidence that physiological differences in
estrogen cause schizophrenia per se; rather, estrogen most likely modifies symp-
tom presentation and/or disease onset in women already at risk for developing
schizophrenia.
338 M. Craig, W. Cutter, R. Norbury, and D. Murphy

500
AUC PRL (mU × h/l)
400

300

200

100

young on off

Fig. 18.2. Serotonergic responsibility as measured by prolactin (PRL) response to fenfluramine in

postmenopausal women taking estrogen replacement therapy (“on”) or never using such
therapy (“off”) plus that in premenopausal women (“young”). AUC, area under the curve.
Kruskal-Wallis, p = 0.02. (van Amelsvoort et al. Psychoneuroendocrinology 2001a, 26:
493–502.)

X chromosome
It has been suggested that a gene(s) for schizophrenia could be X-linked or X–Y
linked, with homologous loci on both X and Y chromosomes (DeLisi et al., 1994),
but support for this view has been limited and there is currently little evidence
that schizophrenia is an X-linked disorder. Nonetheless, it is clear that cognitive
functions which are usually highly lateralized (e.g. language) and have gender differ-
ences in development are affected in schizophrenia. Also it has been suggested that
schizophrenia arises from abnormalities in development of cerebral asymmetry.
Consequently, it is of interest to investigate if/how the X chromosome affects
hemispheric asymmetry and development/function of brain regions implicated
in schizophrenia.
Normally in women, the second X chromosome is randomly inactivated; how-
ever, some genes escape inactivation and so may have significant effects on brain.
Also, males always receive their X chromosome from their mother whereas “normal”
females inherit one X chromosome from each of their parents. Turner syndrome is
a genetic disorder in which all or part of one X chromosome is deleted (45X). Indi-
viduals with Turner syndrome provide a model where the effect of loss of the second
X chromosome and/or mode of inheritance on the brain can be investigated. We
reported that loss of the second X chromosome affects the development of anatomi-
cal, metabolic, and functional cerebral asymmetry; language; and visuospatial skills
(Murphy et al., 1993, 1994, 1997). We suggested that this most likely arises from
339 X chromosome, estrogen, and brain development

loss of genes carried on the X chromosome that normally escape inactivation.

Recently, it has been suggested that social cognition might be “imprinted” from
a genetic locus on the X chromosome (Skuse et al., 1997). The authors exam-
ined the association between the parental origin of the X chromosome in women
with Turner syndrome and their social cognition. Women with paternally derived
X chromosomes were found to be significantly better adjusted, with superior verbal
and executive function skills, than those with maternally derived X chromosomes.
The authors suggested that there is a locus for social cognition that is imprinted
from the paternally derived X chromosome and not expressed from the maternal
one, and that this explains why men are more vulnerable to social communication
disorders than women.
Another way to examine the effect of the X chromosome on the brain is to
study the effect of specific genes and/or gene products, such as trinucleotide triplet
repeats. Expanded trinucleotide repeats are associated with several neurological and
neuropsychiatric conditions, including Huntington’s disease, mytonic dystrophy,
spinal and bulbar muscular atrophy, and fragile X syndrome. It has been proposed
by some that transmission of schizophrenia may be associated with expanded triplet
repeats (Morris et al., 1995; O’Donovan et al., 1995), although this suggestion has
not been supported by others. The current genetic model proposes that fragile
X syndrome results from having more than 200 cytosine–guanine–guanine (CGG)
trinucleotide repeats, with subsequent methylation of the fragile X mental retar-
dation gene (FMR1) and loss of production of its protein product FMRP. It was
originally thought that premutation expansion of CGGs trinucleotide repeats has
no biological effect. FMRP is an RNA-binding protein, required for the normal
process of synaptic maturation. Mice with that Fmr1 gene “knocked out” have
abnormal dendritic spines, hyperactivity, and deficits in learning memory; recent
studies of dfxr mutations in Drosophila have demonstrated synaptic alterations and
impaired circadian rhythms. Overexpression of fmr1 homologs in mice and flies
produces opposing adverse effects, such as lethargy. People with full mutation frag-
ile X syndrome display variable levels of mental retardation with particular deficits
in language, attention, and visuospatial abilities; they also have a high prevalence of
neuropsychiatric abnormalities, including deviant language and communication
patterns, attention deficit disorder, schizotypal personality disorder, and perhaps
psychosis (Freund et al., 1992; Fryns, 1986; Hagerman, 1987). These clinical deficits
become more severe as they pass from one generation to another; this phenomenon
is known as genetic anticipation and is also reported to occur in families with
schizophrenia (McInnis et al., 1993).
However, studies of the biological actions of trinucleotide repeats in people
with full mutation fragile X syndrome are confounded by the presence of learning
340 M. Craig, W. Cutter, R. Norbury, and D. Murphy

disability. Consequently, we examined the effect of premutation expansion of

X chromosome trinucleotide repeats on brain morphometry (using MRI) and
regional metabolism (using PET and [18 F]-2-fluoro-2-deoxy-d-glucose). Carrier
of such a premutation expansion had significant differences in the anatomy and
metabolism of hippocampus and temporal regions plus differences in right–left
asymmetry of the Wernicke and Broca language areas. Therefore, premutation
expansion of X chromosome CGG trinucleotide repeats does affect the devel-
opment of brain regions implicated in schizophrenia and the right–left sym-
metry of language areas. This may be of relevance to the general population, since
approximately 0.5% are premutation carriers of fragile X syndrome; moreover,
intermediate alleles (with CGG repeats over 40) occur in up to 4% of people.

Conclusions
Both estrogen and the X chromosome affect the development of brain regions
and neurochemical systems that are both crucial to higher cognitive function and
implicated in neuropsychiatric disorders such as schizophrenia.
There are gender differences in schizophrenia and because estrogen has puta-
tive antidopaminergic/antipsychotic actions, it has been suggested that estrogen
may be responsible in women for the delayed onset of the first incidence peak
and that the second peak seen in women may result from the decline in estro-
gen levels at menopause. There is evidence that estrogen protects the nigrostriatal
dopaminergic system against the neurotoxic effects of 1-methyl-a-phenyl-2, 2, 3,
6-tetrahydropyridine in rats (Dluzen et al., 1996). An antipsychotic action of estro-
gens is supported by clinical studies reporting that women have increased admission
rates for psychosis around the menses (Hallonquist et al., 1993) and that psychotic
symptomatology varies with the phase of the menstrual cycle (Lindamer et al.,
1997). Also, women with schizophrenia may have reduced estradiol levels compared
with non-schizophrenia controls (Riecher-Rossler et al., 1994). However, there is
currently little evidence to support the therapeutic use of estrogen in schizophre-
nia. For example, when women with schizophrenia are treated with adjunctive
estrogen there is a slight increase in speed of recovery, but no improvement overall
compared with antipsychotic medication alone (Kulkarni et al., 2001). Despite the
lack of clear evidence for the efficacy of estrogen as an antipsychotic treatment,
it remains plausible that estrogen replacement therapy might protect against late-
onset schizophrenia in postmenopausal women by reducing age-related changes in
brain structure and neurochemistry (e.g. in the hippocampus).
We do not propose that estrogen and the X chromosome are crucial to the
development of schizophrenia in most people. Rather, we suggest that they play a
modulatory role in brain maturation, and their mechanism of action needs to be
341 X chromosome, estrogen, and brain development

understood in order to place the “neurodevelopmental theory” of schizophrenia in

context.

REFERENCES

Andreasen, N. C., Smith, M. R., Jacoby, C. G., Dennert, J. W., Olsen, S. A. (1982). Ventricular
enlargement in schizophrenia: definition and prevalence. Am J Psychiatry 139: 292–296.
Andreasen, N. C., Nasrallah, H. A., Dunn, et al. (1986). Structural abnormalities in the frontal
system in schizophrenia: a magnetic resonance imaging study. Arch Gen Psychiatry 43: 136–
144.
Andreasen, N. C., Ehrhardt, J. C., Swayze, III, V. W., et al. (1990a). Magnetic resonance imaging of
the brain in schizophrenia. The pathophysiologic significance of structural abnormalities.
Arch Gen Psychiatry 47: 35–44.
Andreasen, N. C., Swayze, III, V. W., Flaum, M., et al. (1990b). Ventricular enlargement in
schizophrenia evaluated with computed tomographic scanning. Effects of gender, age, and
stage of illness. Arch Gen Psychiatry 47: 1008–1015.
Bedard, P. J., Boucher, R., Daigle, M., di Paolo, T. (1984). Similar effect of estradiol and haloperidol
on experimental tardive dyskinesia in monkeys. Psychoneuroendocrinology 9: 375–379.
Benes, F. M. (1998). Brain development, VII. Human brain growth spans decades. Am J Psychiatry
155: 1489.
Benes, F. M., Turtle, M., Khan, Y., Farol, P. (1994). Myelination of a key relay zone in the hippocam-
pal formation occurs in the human brain during childhood, adolescence, and adulthood.
Arch Gen Psychiatry 51: 477–484.
Bennett-Clarke, C. A., Leslie, M. J., Lane, R. D., Rhoades, R. W. (1994). Effect of serotonin
depletion on vibrissa-related patterns of thalamic afferents in the rat’s somatosensory cortex.
J Neurosci 14: 7594–7607.
Bennett-Clarke, C. A., Chiaia, N. L., Rhoades, R. W. (1996). Thalamocortical afferents in rat
transiently express high-affinity serotonin uptake sites. Brain Res 733: 301–306.
Blue, M. E., Erzurumlu, R. S., Jhaveri, S. (1991). A comparison of pattern formation by thalam-
ocortical and serotonergic afferents in the rat barrel field cortex. Cereb Cortex 1: 380–389.
Brinton, R. D. (1993). 17-Estradiol induction of filopodial growth in cultured hippocampal
neurons within minutes of exposure. Mol Cell Neurosci 4: 36–46.
(2001). Cellular and molecular mechanisms of estrogen regulation of memory function and
neuroprotection against Alzheimer’s disease: recent insights and remaining challenges. Learn
Mem 8: 121–133.
Cases, O., Seif, I., Grimsby, J., Gaspar, P., et al. (1995). Aggressive behavior and altered amounts
of brain serotonin and norepinephrine in mice lacking MAOA. Science 268: 1763–1766.
Cases, O., Vitalis, T., Seif, I., et al. (1996). Lack of barrels in the somatosensory cortex of
monoamine oxidase A-deficient mice: role of a serotonin excess during the critical period.
Neuron 16: 297–307.
Castle, D. J., Murray, R. M. (1993). The epidemiology of late-onset schizophrenia. Schizophr Bull
19: 691–700.
342 M. Craig, W. Cutter, R. Norbury, and D. Murphy

Castle, D. J., Wessely, S., Murray, R. M. (1993). Sex and schizophrenia: effects of diag-
nostic stringency, and associations with and premorbid variables. Br J Psychiatry 162:
658–664.
Coger, R. W., Serafetinides, E. A. (1990). Schizophrenia, corpus callosum, and interhemispheric
communication: a review. Psychiatr Res 34: 163–184.
D’Amato, R. J., Blue, M. E., Largent, B. L., et al. (1987). Ontogeny of the serotonergic projection
to rat neocortex: transient expression of a dense innervation to primary sensory areas. Proc
Natl Acad Sci USA 84: 4322–4326.
De Bellis, M. D., Keshavan, M. S., Beers, S. R., et al. (2001). Sex differences in brain maturation
during childhood and adolescence. Cereb Cortex 11: 552–557.
DeLisi, L. E., Devoto, M., Lofthouse, R., et al. (1994). Search for linkage to schizophrenia on the
X and Y chromosomes. Am J Med Genet 54: 113–121.
Dluzen, D. E., McDermott, J. L., Liu, B. (1996). Estrogen alters MPTP-induced neurotoxicity
in female mice: effects on striatal dopamine concentrations and release. J Neurochem 66:
658–666.
Dumontier, M., Hocht, P., Mintert, U., Faix, J. (2000). Rac1 GTPases control filopodia formation,
cell motility, endocytosis, cytokinesis and development in Dictyostelium. J Cell Sci 113:
2253–2265.
Egan, M. F., Goldberg, T. E., Kolachana, B. S., et al. (2001). Effect of COMT Val108/158 Met
genotype on frontal lobe function and risk for schizophrenia. Proc Natl Acad Sci USA 98:
6917–6922.
Fitch, R. H., Denenberg, V. H. (1998). A role for ovarian hormones in sexual differentiation of
the brain. Behav Brain Sci 21: 311–327; discussion 327–352.
Flaum, M., Arndt, S., Andreasen, N. C. (1990). The role of gender in studies of ventricle enlarge-
ment in schizophrenia: a predominantly male effect. Am J Psychiatry 147: 1327–1332.
Foreman, M. M., Porter, J. C. (1980). Effects of catechol estrogens and catecholamines on hypotha-
lamic and corpus striatal tyrosine hydroxylase activity. J Neurochem 34: 1175–1183.
Freund, L. S., Reiss, A. L., Hagerman, R., Vinogradov, S. (1992). Chromosome fragility and
psychopathology in obligate female carriers of the fragile X syndrome. Arch Gen Psychiatry
49: 54–60.
Fryns, J. P. (1986). The female and the fragile X A study of 144 obligate female carriers. Am J Med
Genet 23: 157–169.
Giedd, J. N., Vaituzis, A. C., Hamburger, S. D., et al. (1996). Quantitative MRI of the temporal
lobe, amygdala, and hippocampus in normal human development: ages 4–18 years. J Comp
Neurol 366: 223–230.
Giedd, J. N., Blumenthal, J., Jeffries, N. O., et al. (1999). Brain development during childhood
and adolescence: a longitudinal MRI study. Nat Neurosci 2: 861–863.
Goldman-Rakic, P. S., Brown, R. M. (1982). Postnatal development of monoamine content and
synthesis in the cerebral cortex of rhesus monkeys. Brain Res 256: 339–349.
Gordon, J. H., Borison, R. L., Diamond, B. I. (1980). Modulation of dopamine receptor sensitivity
by estrogen. Biol Psychiatry 15: 389–396.
Gur, R. E., Mozley, P. D., Shtasel, D. L., et al. (1994). Clinical subtypes of schizophrenia: differences
in brain and CSF volume. Am J Psychiatry 151: 343–350.
343 X chromosome, estrogen, and brain development

Haas, G. L., Hien, D. A., Waked, W., et al. (1989). Sex differences in schizophrenia. Schizophr Res
2: 11.
Haas, G. L., Sweeney, J. A., Hien, D. A., Goldman, D., Deck, M. (1991). Gender differences in
schizophrenia. Schizophr Res 4: 277.
Hafner, H., Riecher-Rossler, A., Maurer, K., Fatkenheuer, B., Loffler, W. (1993). Generating and
testing a causal explanation of the gender difference in age at first onset of schizophrenia.
Psychol Med 23: 925–940.
Hafner, H., Maurer, K., Loffler, W., et al. (1994). The epidemiology of early schizophrenia. Influ-
ence of age and gender on onset and early course. Br J Psychiatry Suppl 23: 29–38.
Hafner, H., An der Heiden, W., Behrens, S., et al. (1998a). Causes and consequences of the gender
difference in age at onset of schizophrenia. Schizophr Bull 24: 99–113.
Hafner, H., Maurer, K., Loffler, W., et al. (1998b). The ABC Schizophrenia Study: a preliminary
overview of the results. Soc Psychiatry Psychiatr Epidemiol 33: 380–386.
Hagerman, R. J. (1987). Fragile X syndrome. Curr Probl Pediatr 17: 627–674.
Hallonquist, J. D., Seeman, M. V., Lang, M., Rector, N. A. (1993). Variation in symptom severity
over the menstrual cycle of schizophrenics. Biol Psychiatry 33: 207–209.
Hampson, E. (1990). Variations in sex-related cognitive abilities across the menstrual cycle. Brain
Cogn 14: 26–43.
Hariri, A. R., Mattay, V. S., Tessitore, A., et al. (2002). Serotonin transporter genetic variation and
the response of the human amygdala. Science 297: 400–403.
Harvey, I., Williams, M., Toone, B. K., et al. (1990). The ventricular–brain ratio (VBR) in func-
tional psychoses: the relationship of lateral ventricular and total intracranial area. Psychol
Med 20: 55–62.
Hedner, J., Lundell, K. H., Breese, G. R., Mueller, R. A., Hedner, T. (1986). Developmental
variations in CSF monoamine metabolites during childhood. Biol Neonate 49: 190–197.
Herlenius, E., Lagercrantz, H. (2001). Neurotransmitters and neuromodulators during early
human development. Early Hum Dev 65: 21–37.
Howard, R., Rabins, P. V., Seeman, M. V., Jeste, D. V., et al. (2000). Late-onset schizophrenia and
very-late-onset schizophrenia-like psychosis: an international consensus. Am J Psychiatry
157: 172–178.
Jablensky, A., Sartorius, N., Ernberg, G. (1992). Schizophrenia: manifestations, incidence and
course in different cultures. A World Health Organization Ten-Country Study. Psychol Med
Monograph 20.
Kendell, R. E., Chalmers, J. C., Platz, C. (1987). Epidemiology of puerperal psychoses. Br J
Psychiatry 150: 662–673.
Kulkarni, J., Riedel, A., de Castella, A. R., et al. (2001). Estrogen: a potential treatment for
schizophrenia. Schizophr Res 48: 137–144.
Lauder, J. M., Krebs, H. (1978). Serotonin as a differentiation signal in early neurogenesis. Dev
Neurosci 1: 15–30.
Lauriello, J., Hoff, A., Wieneke, M. H., et al. (1997). Similar extent of brain dysmorphology in
severely ill women and men with schizophrenia. Am J Psychiatry 154: 819–825.
Lebrand, C., Cases, O., Adelbrecht, C., et al. (1996). Transient uptake and storage of serotonin in
developing thalamic neurons. Neuron 17: 823–835.
344 M. Craig, W. Cutter, R. Norbury, and D. Murphy

Lerer, B., Gelfin, Y., Shapira, B. (1999). Neuroendocrine evidence for age-related decline in central
serotonergic function. Neuropsychopharmacology 21: 321–322.
Leung, A., Chue, P. (2000). Sex differences in schizophrenia, a review of the literature. Acta
Psychiatr Scand Suppl 401: 3–38.
Levallois, C., Valence, C., Baldet, P., Privat, A. (1997). Morphological and morphometric anal-
ysis of serotonin-containing neurons in primary dissociated cultures of human rhomben-
cephalon: a study of development. Brain Res Dev Brain Res 99: 243–252.
Lidow, M. S., Goldman-Rakic, P. S., Rakic, P. (1991). Synchronized overproduction of neuro-
transmitter receptors in diverse regions of the primate cerebral cortex. Proc Natl Acad Sci
USA 88: 10218–10221.
Lindamer, L. A., Lohr, J. B., Harris, M. J., Jeste, D. V. (1997). Gender, estrogen, and schizophrenia.
Psychopharmacol Bull 33: 221–228.
Maki, P. M., Resnick, S. M. (2000). Longitudinal effects of estrogen replacement therapy on PET
cerebral blood flow and cognition. Neurobiol Aging 21: 373–383.
(2001). Effects of estrogen on patterns of brain activity at rest and during cognitive activity: A
review of neuroimaging studies. Neuroimage 14: 789–801.
McInnis, M. G., McMahon, F. J., Chase, G. A., et al. (1993). Anticipation in bipolar affective
disorder. Am J Hum Genet 53: 385–390.
Meltzer, C. C., Smith, G., DeKosky, S. T., et al. (1998). Serotonin in aging, late-life depression, and
Alzheimer’s disease: the emerging role of functional imaging. Neuropsychopharmacology 18:
407–430.
Morris, A. G., Gaitonde, E., McKenna, P. J., Mollon, J. D., Hunt, D. M. (1995). CAG repeat
expansions and schizophrenia: association with disease in females and with early age-at-
onset. Hum Mol Genet 4: 1957–1961.
Moses, E. L., Drevets, W. C., Smith, G., et al. (2000). Effects of estradiol and progesterone
administration on human serotonin 2A receptor binding: a PET study. Biol Psychiatry 48:
854–860.
Murphy, D. G., DeCarli, C., Daly, E., et al. (1993). X-chromosome effects on female brain: a
magnetic resonance imaging study of Turner’s syndrome. Lancet 342: 1197–1200.
Murphy, D. G., Allen, G., Haxby, J. V., et al. (1994). The effects of sex steroids, and the X chromo-
some, on female brain function: a study of the neuropsychology of adult Turner syndrome.
Neuropsychologia 32: 1309–1323.
Murphy, D. G., DeCarli, C., McIntosh, A. R., et al. (1996). Sex differences in human brain
morphometry and metabolism: an in vivo quantitative magnetic resonance imaging and
positron emission tomography study on the effect of aging. Arch Gen Psychiatry 53: 585–594.
Murphy, D. G., Mentis, M. J., Pietrini, P., et al. (1997). A PET study of Turner’s syndrome: effects
of sex steroids and the X chromosome on brain. Biol Psychiatry 41: 285–298.
Nasrallah, H. A., Schwarzkopf, S. B., Olson, S. C., Coffman, J. A. (1990). Gender differences in
schizophrenia on MRI brain scans. Schizophr Bull 16: 205–210.
Nopoulos, P., Flaum, M., Andreasen, N. C. (1997). Sex differences in brain morphology in
schizophrenia. Am J Psychiatry 154: 1648–1654.
O’Connell, P., Woodruff, P. W. R., Wright, I., Jones, P., Murray, R. M. (1997). Developmental
insanity or dementia praecox: was the wrong concept adopted? Schizophr Res 23: 97–106.
345 X chromosome, estrogen, and brain development

O’Donovan, M. C., Guy, C., Craddock, N., et al. (1995). Expanded CAG repeats in schizophrenia
and bipolar disorder. Nat Genet 10: 380–381.
Phillips, S. M., Sherwin, B. B. (1992). Effects of estrogen on memory function in surgically
menopausal women. Psychoneuroendocrinology 17: 485–495.
Phoenix, C. H., Goy, R. W., Gerall, A. A., Young, W. C. (1959). Organizing action of prenatally
administered testosterone proprionate on the tissues mediating behavior in the guinea pig.
Endocrinology 65: 369–382.
Riecher-Rossler, A., Hafner, H., Dutsch-Strobel, A., et al. (1994). Further evidence for a specific
role of estradiol in schizophrenia? Biol Psychiatry 36: 492–494.
Robertson, D. M., van Amelsvoort, T., Daly, E., et al. (2001). Effects of estrogen replacement
therapy on human brain aging: an in vivo 1 H MRS study. Neurology 57: 2114–2117.
Salem, J. E., Kring, A. M. (1998). The role of gender differences in the reduction of etiologic
heterogeneity in schizophrenia. Clin Psychol Rev 18: 795–819.
Seeman, M. V., Lang, M. (1990). The role of estrogens in schizophrenia gender differences.
Schizophr Bull 16: 185–194.
Seifert, Jr., W. E., Foxx, J. L., Butler, I. J. (1980). Age effect on dopamine and serotonin metabolite
levels in cerebrospinal fluid. Ann Neurol 8: 38–42.
Shaywitz, S. E., Shaywitz, B. A., Pugh, K. R., et al. (1999). Effect of estrogen on brain activation
patterns in postmenopausal women during working memory tasks. J Am Med Assoc 281:
1197–1202.
Shelton, R. C., Karson, C. N., Doran, A. R., et al. (1988). Cerebral structural pathology in
schizophrenia: evidence for a selective prefrontal cortical defect. Am J Psychiatry 145: 154–
163.
Shenton, M. E., Kikinis, R., Jolesz, F. A., et al. (1992). Abnormalities of the left temporal lobe
and thought disorder in schizophrenia. A quantitative magnetic resonance imaging study.
N Engl J Med 327: 604–612.
Sherwin, B. B. (1988). Estrogen and/or androgen replacement therapy and cognitive functioning
in surgically menopausal women. Psychoneuroendocrinology 13: 345–357.
Skuse, D. H., James, R. S., Bishop, D. V., et al. (1997). Evidence from Turner’s syndrome of an
imprinted X-linked locus affecting cognitive function. Nature 387: 705–708.
Spear, L. P. (2000). The adolescent brain and age-related behavioral manifestations. Neurosci
Biobehav Rev 24: 417–463.
Spitzer, R., Endicott, J., Robins, E. (1978). Research Diagnostic Criteria (RDC): rationale and
reliability. Arch Gen Psychiatry 35: 773–782.
Sundstrom, E., Kolare, S., Souverbie, F., et al. (1993). Neurochemical differentiation of
human bulbospinal monoaminergic neurons during the first trimester. Dev Brain Res 75:
1–12.
Takahashi, H., Nakashima, S., Ohama, E., Takeda, S., Ikuta, F. (1986). Distribution of serotonin-
containing cell bodies in the brainstem of the human fetus determined with immunohisto-
chemistry using antiserotonin serum. Brain Dev 8: 355–365.
Toran-Allerand, C. D. (1996). The estrogen/neurotrophin connection during neural develop-
ment: is co-localization of estrogen receptors with the neurotrophins and their receptors
biologically relevant? Dev Neurosci 18: 36–48.
346 M. Craig, W. Cutter, R. Norbury, and D. Murphy

Travis, M. J., Mulligan, O., Mulligan, R. S., et al. (1999). Preliminary investigation of the effect
of oestradiol treatment on cortical 5–HT2A receptor binding: a single photon emission
tomography (SPET) study using 123 I -5-I-R91150. Neuroimage 9: S672.
van Amelsvoort, T. A. M. J., Daly, E. Robertson, D. M. R., et al. (2001a). Structural brain abnor-
malities associated with deletion of chromosome 22qaa: quantitative neuroimaging study
of adults with velo-cardio-facial syndrome. Br J Psychiatry 178: 412–419.
van Amelsvoort, T. A. M. J., Abel, K. M., Robertson, D. M. R., et al. (2001b). Prolactin response to
d-fenfluramine in postmenopausal women on and off ERT: comparison with young women.
Psychoneuroendocrinology 26: 493–502.
Vazquez-Barquero, J. L., Cuesta Nunez, M. J., Quintana Pando, F., et al. (1995). Structural abnor-
malities of the brain in schizophrenia: sex differences in the Cantabria First Episode of
Schizophrenia Study. Psychol Med 25: 1247–1257.
Walker, E. F. (1994). Developmentally moderated expressions of the neuropathology underlying
schizophrenia. Schizophr Bull 20: 453–480.
Whitaker-Azmitia, P. M. (2001). Serotonin and brain development: role in human developmental
diseases. Brain Res Bull 56: 479–485.
Williams, A. O., Reveley, M. A., Kolakowska, T., Ardern, M., Mandelbrote, B. M. (1985).
Schizophrenia with good and poor outcome. II: Cerebral ventricular size and its clinical
significance. Br J Psychiatry 146: 239–246.
Wong, D. F., Wagner, Jr., H. N., Dannals, R. F., et al. (1984). Effects of age on dopamine and
serotonin receptors measured by positron tomography in the living human brain. Science
226: 1393–1396.
Wong, D. F., Broussolle, E. P., Wand, G., et al. (1988). In vivo measurement of dopamine receptors
in human brain by positron emission tomography. Age and sex differences. Ann N Y Acad
Sci 515: 203–214.
Woolley, C. S., McEwen, B. S. (1994). Estradiol regulates hippocampal dendritic spine density
via an N-methyl-d-aspartate receptor-dependent mechanism. J Neurosci 14: 7680–7687.
Yan, W., Wilson, C. C., Haring, J. H. (1997). Effects of neonatal serotonin depletion on the
development of rat dentate granule cells. Brain Res Dev Brain Res 98: 177–184.
Yu, X.-M., Askalan, R., Keil, II, G. J., Salter, M. W. (1997). NMDA channel regulation by channel-
associated protein tyrosine kinase Src. Science 275: 674–678.
 19

Premorbid structural abnormalities

in schizophrenia
Stephen M. Lawrie
Royal Edinburgh Hospital, Edinburgh, UK

Hundreds of structural brain imaging studies in groups of patients have demon-

strated that there is a neuroanatomy of schizophrenia. When the various abnormal-
ities are first evident, however, remains unknown. This is not merely an academic
issue. If all or even some differences are present from early life in people who go on
to develop schizophrenia, then very early detection is a possibility; if some findings
are only manifest later in development, or around the time of onset, therapeutic
opportunities may arise.
Identifying the time course of the structural abnormalities in schizophrenia is
a difficult task. Ideally, prospective studies would follow people from early fetal
life to the onset of schizophrenia and beyond, but this is impractical for a host of
technological, clinical, and epidemiological reasons. The best available compro-
mise is to conduct repeated longitudinal examinations of populations at high risk
of developing schizophrenia a relatively short time before onset. Even these studies
are expensive, vulnerable to changes in technology, and potentially biased by partic-
ipant drop-out. Case–control studies are most practical, but generally less reliable;
these seek to relate early exposures (e.g. family history, adverse obstetric compli-
cations [OCs]) and/or developmental indices (e.g. motor milestones, premorbid
adjustment) to imaging findings. These measurements are usually retrospective
and often imprecise.
This chapter attempts to review all of the evidence, from these types of imag-
ing study, for premorbid abnormalities in patients with schizophrenia and related
populations. Before considering these, it is necessary to highlight the anatomical
abnormalities that require explanation.

The neuroanatomy of schizophrenia

Table 19.1 lists the main findings from quantitative systematic reviews of struc-
tural imaging studies comparing patients with schizophrenia and healthy controls.

Neurodevelopment and Schizophrenia, ed. Matcheri S. Keshavan et al. Published by Cambridge University
Press.

C Cambridge University Press 2004.

347
348 S. M. Lawrie

Table 19.1. Systematic reviews of structural brain abnormalities in schizophrenia versus controls

Reference Type of study Measures Main findings

Raz and Raz, 1990 CT Lateral ventricles Enlarged (ES, 0.7)

CT Third ventricle Enlarged (ES, 0.66)
CT Cortical sulci/fissures Widened (ES, 0.35)
Van Horn and CT/sMRI Ventricular to brain ratio Enlarged
McManus (1992)
Woodruff et al. (1995) sMRI Corpus callosum Reduced area non-significant
after control for brain area
Ward et al. (1996) sMRI/CT/PM Brain size Reduced (ES, 0.31)
sMRI/CT Intracranial size Reduced (ES, 0.16)
Nelson et al. (1998) sMRI Hippocampus Reduced (ES, 0.4) by about 4%
bilaterally
Amygdala/hippocampus Reduced (ES, 0.7) by about 8%
bilaterally
Amygdala Reduced (ES, 0.3) by about 4%
bilaterally
Lawrie and Abukmeil sMRI Most regions measured in Include reductions in whole
(1998) (amended in previous studies brain (∼3%), frontal and
Lawrie et al., 2000) temporal lobes (∼5%), and
parahippocampal gyrus
(10%), and increases in CSF
(∼15%)
Shapleske et al. (1999) sMRI Planum temporale Possibly reversed asymmetry
Wright et al. (2000) sMRI All regions measured in Include reductions in gray and
previous studies white matter (∼3%),
superior temporal gyrus
(∼3%) and increases in
globus pallidus (∼20%)
Konick and Friedman sMRI/PM Thalamus Reduced (ES, 0.4)
(2001)
Sommer et al. (2001) sMRI All studies of cerebral Reduced asymmetry of
asymmetry posterior superior temporal
gyrus (and possibly of
Sylvian fissure)

CT, computed tomography; sMRI, structural magnetic resonance imaging; PM, postmortem; ES, effect size;
CSF, cerebrospinal fluid.
349 Premorbid structural abnormalities in schizophrenia

Computed tomography (CT) demonstrated ventricular enlargement and a gener-

alized loss of brain tissue (Raz and Raz, 1990; van Horn and McManus, 1992), albeit
in one composite area measure, the ventricular to brain ratio (VBR), which may
have conflated separate disease processes. Magnetic resonance imaging, which now
requires a structural prefix (sMRI), has replicated these findings and convincingly
shown additional volume deficits in the prefrontal and temporal lobes, as well as
further decrements in the medial and superior temporal lobe. The largest percent-
age differences are in the body of the lateral ventricle, which is approximately 50%
enlarged, and the parahippocampal gyrus, which is approximately 10% reduced in
volume (Lawrie and Abukmeil, 1998; Wright et al., 2000), although the largest effect
size reported is for increases in the globus pallidus, which are probably related to
therapeutic dopamine blockade (Wright et al., 2000). The thalamus is also reduced
in volume and there are strong suggestions that schizophrenia patients may have
reduced or even reversed asymmetry of the usually left-lateralized temporal cortex.
None of these reviews have, however, been able to relate these abnormalities to any
specific period of brain development or risk factor for schizophrenia.
Most sMRI studies have taken a region of interest (ROI) approach, which requires
laborious hand tracings facilitated by varying degrees of semi-automated tissue
classification and boundary detection. The alternative is to use entirely automated
procedures, such as voxel-based morphometry (VBM), to construct statistical prob-
ability maps of the likelihood that a particular brain region has a different tissue
density in two or more subject groups. This is more reliable and quicker than ROI
tracing but cannot identify global effects, requires correction for multiple hypoth-
esis testing at each image voxel, and is likely to be more able to detect changes at
the boundaries of structures than within them. Various VBM studies comparing
patients with schizophrenia and healthy controls have replicated the main ROI
findings and, in addition, repeatedly suggested that the insula and the medial
frontal lobe, including the anterior cingulate, have reduced grey matter density
in schizophrenia (e.g. Job et al., 2002; Paillere-Martinot et al., 2001; Sigmundsson
et al., 2001).
The consistent demonstration of such structural abnormalities is in striking
contrast to the lack of associations found with almost any clinical measure in the
patients. By far the best replicated finding is an apparent absence of any relation-
ship with illness duration (Lawrie and Abukmeil, 1998; Lewis, 1990). Together
with repeated demonstrations that abnormalities are evident in patients with first-
episode schizophrenia (FES) (seemingly to a similar extent as in patients with
chronic schizophrenia), this provides strong support for neurodevelopmental mod-
els of schizophrenia and suggests that at least some of the findings predate the
onset of psychosis. Theories of an essentially “static encephalopathy” are, however,
seriously challenged by increasing evidence from prospective studies that some
350 S. M. Lawrie

abnormalities may be progressive in the first few years after onset (see Ch. 20).
They are also challenged by the preliminary findings from prospective high-risk
studies.

Prospective studies of high-risk populations

The first high-risk study to include brain structure measurements was the Copen-
hagen High Risk Project, but this used CT and the scans were not repeated over
time. Indeed, as yet, there are published data from only two longitudinal studies of
pre- and perimorbid structural differences in people at elevated risk of schizophre-
nia. These two studies, which recruited high-risk subjects in entirely different ways,
will now be discussed in detail.

Edinburgh High Risk Study

In the Edinburgh High Risk Study, we have examined subjects with at least two close
relatives with schizophrenia. Potential participants were identified by reviewing the
psychiatric case notes of patients with schizophrenia whose clinicians indicated were
likely to have family histories. If the diagnosis and family history were confirmed,
we then asked permission to approach their healthy relatives aged 16–24. A total
of 229 suitable high-risk subjects were identified: 162 provided some data and 150
had one or more sMRI scans between 1994 and 1999. Groups of similarly aged
healthy individuals (controls) and patients with FES were also examined. Most of
the high-risk subjects and controls have returned for at least one further scan since
1999 and are likely to do so again before 2004.
Our preliminary ROI findings, in the first 100 high-risk subjects (Lawrie et al.,
1999), were that the amygdala–hippocampal complex (AHC) was significantly
smaller (by about 4%, with an effect size [ES] of 0.3) than in controls, but about
4% larger (ES, 0.3) than in FES. Reductions in the volume of the thalamus were
only evident in comparison of the high-risk group versus the controls. An increase
in third ventricular volume in the FES group was not significant after control-
ling for familial clustering (reduced variance) in the full high-risk sample (Lawrie
et al., 2001). The main findings have been replicated by VBM analyses of the same
data sets (Job et al., 2002, 2003), which extended them by showing anterior cin-
gulate, medial prefrontal, and parahippocampal reductions in gray matter density
with the greatest reductions in FES, then in the high-risk group, and then the
controls.
Our finding that the volumes of the AHCs in high-risk subjects were midway
between those of the controls and FES (Lawrie et al., 1999, 2001) has important
implications. Based on their family histories, we expect that only 10–20% of our
high-risk subjects will develop schizophrenia. As there is no obvious subgroup of
351 Premorbid structural abnormalities in schizophrenia

high-risk subjects with particularly small AHCs (and assuming that those who
develop schizophrenia will ultimately have AHCs that approximate to those of
FES), the reduced AHC volume may be a trait marker. Prefrontal lobe and thalami
volumes, but not those of the AHC, were associated with measures of genetic
liability (Lawrie et al., 2001). The apparent absence of a relationship between AHC
volume and genetic liability was unexpected and may be attributable to a relatively
narrow range of liability in the Edinburgh High Risk Study or that hippocampal
volumes particularly are related to OCs (see below). If, however, there are further
AHC reductions in those who develop schizophrenia, this might reflect genetic
effects on later neurodevelopment (but could also reflect environmental effects or
a peri-onset effect on plasticity).
In those with psychotic symptoms at any point in the first 5 years of the study,
the whole brain volume at study entry was reduced (after controlling for age,
sex, paternal social class, height, and handedness) compared with those did not
have psychotic symptoms over this time (Lawrie et al., 2001). Those with psy-
chotic symptoms and two or more sMRI scans had non-significant reductions in
whole brain and (left) AHC volume over almost 2 years, but did have significant
reductions in the (right) temporal lobes over time (Lawrie et al., 2002). Very pre-
liminary results suggest that relatively large brains (and relatively small thalami in
women) predict the early onset of psychosis in high-risk subjects (Johnstone et al.,
2002).
At the time of writing, the twentieth person in the Edinburgh High Risk Study
has just developed schizophrenia or a related psychosis. We are currently analyzing
the possible sMRI predictors of schizophrenia and changes as the illness develops,
with both ROI and VBM. Future analyses will include relating pre- and perimorbid
brain structure to other risk factors for schizophrenia.

Melbourne High Risk Study

Researchers in the Melbourne High Risk Study have adopted a different but com-
plementary approach. They have examined groups of people at “ultra high risk”
(UHR) of psychosis, those with first-episode psychosis (FEP; including about 50%
with schizophrenia/schizophreniform psychosis), and a large group of healthy con-
trols. The UHR group consists of people aged 14–30 years and referred with “atten-
uated” partial (positive) psychotic symptoms several times a week for up to 5 years;
transient symptoms of less than 1 week’s duration in the past year; and/or both trait
and state risk factors for psychosis (a family history and a worsening in mental state
or general functioning in the past year). Preliminary ROI results suggested non-
significant hippocampal reductions in the UHR group and the subgroup of these
patients that developed psychosis (Copolov et al., 2000). An early VBM study of
nine who developed psychosis and 12 who remained well did identify significant
352 S. M. Lawrie

volume reductions in the hippocampus, entorhinal cortex, and inferior frontal and
fusiform gyri during the transition to psychosis (Pantelis et al., 2000).
Phillips et al. (2002) examined the traced whole brain and hippocampal volumes
as possible predictors of psychosis. Of the eligible referrals, 75% were scanned
between 1995 and 1998; 5% of scans were lost through movement artifact, leaving
60 subjects. Twenty (33%) of those with usable scans developed an acute psy-
chosis (defined as an increase of one or more points in psychotic symptom severity,
several times a week, for more than 1 week) within a year. A structured clinical
interview identified diagnoses of schizophrenia spectrum disorder (11), affective
psychosis (3), bipolar disorder (3), and others (3). The UHR groups who did not
develop psychosis included four patients with depression, and 11 with an anxiety
state.
The scans were conducted at two sites and there was a weak tendency for the scan-
ners to deliver different whole brain volumes and for more of those who developed
psychosis (and of one sex) to be scanned at one site (Phillips et al., 2002). Control-
ling for a slightly smaller whole brain, the 60 UHR patients had significantly smaller
hippocampi (by about 11%; ES, 0.9) than the controls but did not differ from the
FEP group. However, the UHR patients were on average 10 years younger and
11 points lower in premorbid intelligence quotient (IQ) scores than the normals.
The 20 who became psychotic were much more closely similar in demographics to
the 40 who remained non-psychotic but, rather counterintuitively, the (left) hip-
pocampus was larger in those who went on to develop a psychosis. These confusing
results may be attributable to scanner effects, selection bias, and/or confounding,
for example by other diagnoses or gender.
Pantelis et al. (2003) have recently reported on their VBM findings in a slightly
extended sample. In 75 people with prodromal symptoms, defined as above, 23
(31%) developed a psychotic disorder (with roughly equal numbers of schizophre-
nia and affective psychoses). Compared with those who did not develop a psy-
chosis, they had less grey matter in (right) medial temporal, lateral temporal,
and inferior frontal cortex, and in bilateral cingulate cortex. Twenty-one sub-
jects had a repeat scan after 1–2 years, 10 of whom had become psychotic (five
with schizophrenia). They showed reductions over time in (left) parahippocam-
pal, fusiform, orbitofrontal, and cerebellar grey matter, while those who did not
become psychotic only exhibited cerebellar reductions. These are intriguing results
that suggest grey matter reductions in the run up to psychosis and immediately
after it. There are, however, some important issues outstanding; in particular,
whether some of those in the UHR group actually had a psychosis before or at
study entry, and why the ROI and VBM studies appear to have such different
findings.
353 Premorbid structural abnormalities in schizophrenia

Summary
These studies illustrate well the difficulties in conducting this type of research:
acquiring a representative sample who will continue to participate in longitudinal
studies, maintaining technical control of the scanner(s) over lengthy time inter-
vals, and distinguishing psychotic symptoms from psychosis. The two recruitment
strategies used so far have different and complementary strengths. Sampling rela-
tives rather than referrals may provide a population who are more committed to the
research, may be more likely to develop schizophrenia than other psychoses, and
gives the potential to study gene–environment interactions. The main limitation
is the time taken to accumulate enough subjects who develop psychosis, which is
likely to be longer than in studies of people presenting with psychotic symptoms.
The latter studies also have the greater ability to compare diagnoses. Regardless, the
two main studies so far both report reduced medial temporal lobe structures as a
trait marker of schizophrenia and suggest further volume losses around the time
of onset. Such prospective studies are the only way to establish temporal patterns
with any certainty, although very important information can also be obtained from
case-control studies that examine the associations between imaging abnormalities
and known risk factors for the disorder.

Structural imaging and risk factors for schizophrenia

The strongest known risk factor for schizophrenia is family history, which increases
the risk of the disorder by approximately 5–50 times depending on the degree of
genetic association (Cannon and Jones, 1996). Other risk factors are comparatively
weak, with odds ratios of slightly more than 1–3. Of these, OCs have been most
commonly related to imaging abnormalities, but there are also some studies of
the structural associations of premorbid adjustment, season of birth, and fetal
undernutrition/infection.

Family history
Family history is, of course, a proxy for genetic effects and these might only be
expressed later in development. Given, however, the relative importance of genetic
as opposed to environmental causation of schizophrenia, and that brain growth is
maximal prenatally and for 1–2 years thereafter, it is likely that familial structural
abnormalities are at least partly evident at or shortly after birth. The relationship
between an elevated genetic risk of schizophrenia and structural abnormalities has
been examined by scanning relatives (twins, siblings, offspring, parents), comparing
known familial with presumed sporadic cases, and by examining people with genetic
variants such as schizotypal personality disorder (SPD).
354 S. M. Lawrie

Computed tomography
In the first CT study, Weinberger et al. (1981) compared the VBRs of 10 inpatients
with schizophrenia; 12 of their siblings (from seven sibships), and 17 controls.
The VBR appeared to be, at least partially, under genetic control (as did some
local horn enlargements) and differences were evident: largest in the schizophrenia
patients, then in the siblings, and then the controls. Reveley et al. (1982) obtained
a similar pattern of results in discordant monozygotic (MZ) twins and controls, as
have other sibling studies (DeLisi et al., 1986; Mourot et al., 1997; Silverman et al.,
1998). Mourot et al. (1997) also found the same effect for third ventricle width,
and Honer et al. (1995) reported larger temporal sulci and sylvian fissures (but not
temporal horns) in patients compared with unaffected siblings.
The Copenhagen High Risk Project initially examined 34 offspring of mothers
with schizophrenia, 10 of whose fathers also had a spectrum disorder. Widened
fissures and sulci were related to genetic risk, while an increased VBR appeared
to relate to a gene–environment interaction (see below; Cannon et al., 1989). The
results were internally consistent on substantially enlarging the sample (Cannon
et al., 1993). Widening the sample to include offspring with schizophrenia, SPD,
or other psychiatric disorders demonstrated that global effects were evident in
schizophrenia and SPD but only the schizophrenia patients had enlarged ventricles
(Cannon et al., 1994). However, a subsequent comparison of 16 “discordant sibling
pairs” found that the differences were almost as marked for the sulcal area or the
sulcal to brain ratio as for ventricular area or VBR (Zorrilla et al., 1997).
Studies of the VBR in “sporadic versus familial” schizophrenia provide less con-
sensus, as one of the originators of the distinction noted in a review of the early
studies (Lewis, 1990). Patients without a family history of schizophrenia have been
shown repeatedly to have higher VBRs and more marked ventriculomegaly than
patients with a positive family history (e.g. Vita et al., 1994; reviewed by DeQuardo
et al., 1996). Some notable large studies (n > 100) have failed to find an associ-
ation between family history and the VBR (Johnstone et al., 1989; Jones et al.,
1994), yet several studies do report such an association (e.g. Owens et al., 1985),
including those which go some way to supporting the distinction (e.g. Vita et al.,
1994; Silverman et al., 1998). It may be that schizophrenia patients without an
obvious genetic loading are more likely to have had environmental triggers likely
to increase the VBR (see below). However, determining sporadic status is at best
difficult, when many healthy relatives are likely to be carrying the gene(s), and
the limited resolution of CT increases measurement error. Indeed, sMRI studies
that have adopted similar approaches have all reported similar or greater abnor-
malities in familial cases (Dauphinais et al., 1990; Falkai et al., 2002; Harris et al.,
2002; Roy et al., 1994; Sanderson et al., 2001; Schwarzkopf et al., 1991; Seidman
et al., 2002), with the exception of McDonald et al. (2002), and some suggest
355 Premorbid structural abnormalities in schizophrenia

specific genetic effects in frontal and temporal lobes, the ventricles, and the basal
ganglia.

Structural magnetic resonance studies of twins

Twin studies have the ability to distinguish genetic from environmental effects,
although this depends upon including both healthy and discordant MZ and dizy-
gotic (DZ) twin pairs. A genetic role is likely when MZ patients and their twins
have similar volumes, but differ from healthy MZ twins, and the finding is more
pronounced in MZ than DZ twins. Even then, however, the discordance may still
be attributable to differences in gene expression or gene–environment interactions.
Suddath et al. (1990) found that the affected individuals in 15 pairs of discordant
MZ twins had reduced left temporal lobe grey matter and bilaterally smaller hip-
pocampal volumes, as well as larger ventricular structures, with no such differences
in the frontal lobe or in white matter. They did not, however, include DZ pairs
or healthy controls. Subsequent reports on 8–13 of these MZ pairs discordant for
schizophrenia and 5–10 healthy MZ twin pairs have found no difference in sylvian
fissure or planum temporale asymmetry (Bartley et al., 1993); whole brain volume
reductions of apparent genetic and environmental cause (Noga et al., 1996); but, in
a smaller study, no apparent differences in whole brain or thalamus (Bridle et al.,
2002).
In the largest ROI twin study so far, Baare et al. (2001) examined 15 MZ and
14 DZ twin pairs discordant for schizophrenia and 29 healthy twin pairs. They found
within-pair similarity to be greater in MZ twins for intracranial, whole brain, frontal
lobe, and hippocampal volumes (and evidence of a high degree of genetic control on
the size of the lateral ventricles in healthy twins only). By comparison the intraclass
correlation coefficient for the parahippocampal gyrus and third ventricle was lower
in discordant MZ than healthy MZ pairs, suggesting primarily environmental
effects. The affected twins had smaller brains, more extracerebral cerebrospinal
fluid (CSF) and larger lateral ventricles. Therefore, sMRI ROI (and CT) twin studies
suggest that most volumetric reductions represent both genetic and environmen-
tal effects, but reductions in whole brain and frontal lobes may be more genetic
(cf. Lawrie et al., 2001) and increases in the ventricles (in schizophrenia) more
environmental. They disagree about the attribution for hippocampal decrements.

Automated structural magnetic resonance imaging studies in relatives

Cannon et al. (2002a) constructed probabilistic cortical surface maps in MZ and
DZ discordant twins and control twins. Twin group effects (as MZ > DZ > control
twin differences) and intraclass correlation coefficient maps (MZ > DZ) highlighted
apparently genetic deficits in the poles of the frontal and temporal lobes, the dor-
solateral prefrontal cortex, and Broca’s and Wernicke’s areas. MZ affected status
356 S. M. Lawrie

contrasts, and findings of higher intraclass correlation coefficients in healthy than

in diseased pairs, suggested environmental effects in the dorsolateral prefrontal cor-
tex and parts of the temporal lobe (including Heschl’s gyrus), which were correlated
with symptom severity (especially negative) and cognitive dysfunction (especially
on memory tasks). Narr et al. (2002) used a similar technique to map hippocam-
pal surface morphology in the same twin groups and found both schizophrenia
and genetic liability effects. Overall, these studies are compatible with our own
findings of effects in parts of the frontal and temporal lobes that were greatest in
schizophrenia patients and intermediate in high-risk subjects, compared with con-
trols (Job et al., 2003), with genetic effects being most evident medially. Ananth
et al. (2002) have also shown that a family history of schizophrenia is negatively
correlated with medial prefrontal gray matter. Meyer-Lindenberg et al. (2001) have
published preliminary results showing dorsolateral prefrontal cortex differences
that were greatest in schizophrenia patients, intermediate in their siblings, and least
apparant in controls. They also showed anterior hippocampal reductions in gray
matter density in patients compared with their siblings. The emergent picture is,
therefore, of familial, probably genetic, effects that are generalized but perhaps most
evident frontally and in the medial temporal lobe; these effects are likely to be at
least partially evident early in development. Phenotypic and possibly later effects
may be relatively greater in the temporal cortex.

Structural magnetic resonance imaging studies of siblings, parents, and offspring

The underlying rationale of examining the first-degree relatives of patients with
schizophrenia is that they share approximately 50% of their genome and that com-
mon differences versus controls probably reflect genetic factors, while differences
between unaffected and affected relatives presumably represent phenotypic effects.
It is clear that these studies cannot in themselves distinguish genetic and environ-
mental causation, but as schizophrenia is usually found to reflect mainly genetic
factors, with a relatively small unique environmental effect and almost no familial
environment involvement (e.g. Cannon et al., 1998a; McGuffin et al., 1994), the
commonalities between patients and their relatives are probably genetic.
Most of these relatives studies have examined the volumes of the lateral ventricles
and/or the AHCs. Only one study has reported significantly enlarged lateral ven-
tricles in relatives compared with controls (Sharma et al., 1998), although most of
the studies in siblings (Cannon et al., 1998b; Seidman et al., 1999; Staal et al., 2000)
or offspring (Keshavan et al., 1997, 2002; Lawrie et al., 1999, 2001; Schreiber et al.,
1999) indicate such enlargement. The few comparisons of patients and siblings
are universally significant (Cannon et al., 1998b; McDonald et al., 2002; Sharma
et al., 1998; Staal et al., 2000), as are the twin studies, suggesting stronger envi-
ronmental and phenotypic effects. Indeed, in older “obligate gene carriers,” lateral
357 Premorbid structural abnormalities in schizophrenia

ventriculomegaly (McDonald et al., 2002; Sharma et al., 1998) has not yet been
externally replicated (Steel et al., 2002). The increased VBR seen by CT in rela-
tives may, therefore, primarily reflect a reduction in brain volume, although the
results on sMRI are equivocal. While some studies find the brain is smaller in rel-
atives than in controls (Cannon et al., 1998b; Keshavan et al., 1997, 2002), and
no patient–relative differences (Cannon et al., 1998b; McDonald et al., 2002; Seid-
man et al., 2002), many find no differences (McDonald et al., 2002; Seidman et al.,
1999, 2002; Sharma et al., 1998; Staal et al., 1998) and one small study did actually
find differences between “obligate carriers” and their affected siblings (Steel et al.,
2002).
The evidence from studies of relatives is also inconclusive for most other brain
regions, in some cases because of insufficient studies and in others because of low
power. There are, for example, isolated reports of abnormal cerebral torque (Sharma
et al., 1999), but the most consistent abnormalities in patients (Table 19.1) have
not been found in relatives (Bartley et al., 1993; Frangou et al., 1997); findings
of abnormal sylvian fissure (Honer et al., 1995) and AHC asymmetry (Schreiber
et al., 1999) have yet to be replicated (Bartley et al., 1993). There is, however,
a fair degree of agreement with the twin and automated studies reviewed above
for frontotemporal differences with a pattern of most effect seen in schizophrenia
patients compared with controls and intermediate effects in relatives. The main
support for this from ROI studies in relatives is a trend from a large study using
sophisticated segmentation algorithms (Cannon et al., 1998b), and from studies
contrasting patients with their unaffected relatives (McDonald et al., 2002; Staal
et al., 2000; Steel et al., 2002). Small (Keshavan et al., 1997, 2002; Schreiber et al.,
1999) and large (Lawrie et al., 2001) studies of high-risk offspring have not found
such differences, and neither have medium-sized studies that did not control for
family membership (Sharma et al., 1998). It appears that the relatively small effects
can also be detected with automated approaches and/or control for between-family
variance.
The importance of statistical power is again illustrated in studies of the third ven-
tricle and the thalamus (the increase in the former probably relating to reductions
in the latter and other surrounding structures). The two studies that did not find
changes in the third ventricle (Schreiber et al., 1999) and the thalamus (Keshavan
et al., 1997) were the two smallest were offspring studies (15 and 11 patients, respect-
ively). Keshavan et al. (2002) found thalamus reductions when they increased the
sample size from 11 to 19. Differences have, however, been found with samples as
small as six (Seidman et al., 1997) and all of the available literature suggests that
third ventricle increases and/or thalamus reductions occur in relatives compared
with controls (Keshavan et al., 1997, 2002; Lawrie et al., 2001; McDonald et al.,
2002; Seidman et al., 1997, 1999; Staal et al., 1998, 2000). Very few studies have
358 S. M. Lawrie

looked at patient–relative differences. Staal et al. (1998) found that the thalamus
was smaller in patients than in their siblings, while thalamus reductions were related
to genetic liability but not psychotic symptoms in the Edinburgh High Risk Study
(Lawrie et al., 2001). Third ventricle volumes are consistently reported as trends in
patient–relative differences (Lawrie et al., 2001; McDonald et al., 2002; Staal et al.,
1998), suggesting that environmental influences on third ventricle increases may
be related to disease expression.
The studies of relatives do make clear that reductions in the AHC or its component
structures are both genetic and phenotypic. In terms of relative differences, patients
are generally found to have smaller AHCs than their unaffected relatives, and rela-
tives are generally found to have smaller AHCs than healthy controls (Table 19.2). In
terms of statistical significance, relatives have smaller AHCs than controls (Keshavan
et al., 1997, 2002; Lawrie et al., 1999, 2001; Schreiber et al., 1999; Seidman et al.,
1997, 1999), and schizophrenia patients have smaller AHCs than relatives (Lawrie
et al., 1999, 2001; O’Driscoll et al., 2001; Steel et al., 2002), although there are some
negative studies (Staal et al., 2000). There are suggestions that the reductions may
be more marked anteriorly (Keshavan et al., 2002; O’Driscoll et al., 2001) and on the
left side (Keshavan et al., 2002; Lawrie et al., 2001; but see Schreiber et al., 1999).
There is, however, good evidence for hippocampal differences as well (Seidman
et al., 2002; Waldo et al., 1994; see also van Erp et al., 2002).

Schizotypal personality disorder

The SPD is widely regarded as a genetic variant of, and a risk factor for, schizophre-
nia. Imaging abnormalities in SPD could, therefore, identify genetic and premorbid
effects, although there are, as yet, comparatively few studies and even fewer repli-
cated results. Two CT studies, from the same research group, found larger VBRs
in those with SPD than in those with other personality disorders (Siever et al.,
1995) or indeed in unaffected siblings (Silverman et al., 1998). However, studies
distinguishing brain and ventricles in SPD suggest that the genetic similarities may
arise in reduced whole brain grey matter and increased CSF (Cannon et al., 1994;
Dickey et al., 2000), and that only patients with schizophrenia have enlarged lateral
ventricles (Buchsbaum et al., 1997; Dickey et al., 2000; Levitt et al., 2002). There
are replicated accounts of common reductions in the superior temporal gyrus,
which might be related to common features such as “disordered thinking” (Dickey
et al., 1999; Downhill et al., 2001), but these studies disagree as to whether other
reductions in the temporal lobe, and particularly in medial structures, are evident.

Imaging specific gene effects

By far the best indication of genetic effects is, of course, to relate gene frequency or
expression directly to specific regional volumes. This approach is obviously limited
359 Premorbid structural abnormalities in schizophrenia

Table 19.2. Region of interest studies of the amygdala–hippocampal complex in patients with
schizophrenia, their relatives and/or controls

Reference Study populations Differences (%)

Relatives versus Relatives versus

schizophrenia controls

Suddath et al. (1990) 15 pairs discordant Left Hipp. +11% –

MZ twins Right Hipp. +9% –
Waldo et al. (1994) 11 schizophrenia, Left Amyg. −2% Left Amyg. −13%
20 siblings, Right Amyg. −2% Rightt Amyg. +2%
13 controls Left Hipp. +7% Left Hipp. +6%
Right Hipp. +5% Right Hipp. +2%
Schreiber et al. (1999) 15 adolescent offspring, – Both AHC −11%
15 controls
Staal et al. (2000) 16 schizophrenia, Both Amyg. +15% Both Amyg. no change
16 siblings, Both Hipp. +6% Both Hipp. +1%
32 controls
Baare et al. (2001) 15 MZ discordant twin pairs Left AHC +4% Left AHC −8%
(15 DZ discordant twin Right AHC +2% Right AHC −6%
pairs, 15 controls (matched
to well co-twin)
Lawrie et al. (2001) 34 First-episode schizophrenia, Left AHC +4% Left AHC −4%
147 high risk, 36 controls Right AHC +2% Right AHC −2%
O’Driscoll et al. (2001) 20 mixed relatives, 14 controls – Left Ant. AHC −8%
– Right Ant. AHC −9%
Left Post. AHC +4%
Right Post. AHC –3%
Harris et al. (2002) 6 schizophrenia, 12 parents, Both Hipp. −2% Both Hipp. +40%
6 controls
Keshavan et al. (2002) 17 offspring, 22 controls – Left Ant. AHC −25%
– Right Ant. AHC −12%
Left Post. AHC −12%
Right Post. AHC −10%
Seidman et al. (2002) 18 schizophrenia, 18 relatives, Left Hipp. −1% Left Hipp. −12%
48 controls Right Hipp +4% Right Hipp. +2%
Steel et al. (2002) 6 schizophrenia, 6 obligate, Both AHC +8% Both AHC –5%
6 control siblings Left AHC +8% Left AHC –5%
Right AHC +10% Right AHC –4%

MZ, monozygotic; DZ, dizygotic; Hipp., hippocampus; Amyg., amygdala; AHC, amygdala–hippocampal
complex; Ant., anterior; Post., posterior.
360 S. M. Lawrie

by our knowledge of the genes that increase the liability to schizophrenia. That
being said, a small number of studies have directly examined the associations of
some putative genes. With probably the greatest justification, Kunugi et al. (1999)
examined the relationship between hippocampal volume and the gene for neu-
rotrophin 3 (NT-3) (A3 allele) in a small number of patients with schizophrenia
and bipolar disorder. NT-3 A3 allelic status was associated with smaller volumes
only in schizophrenia patients: the absence of an effect in those with bipolar disorder
suggesting a possible interaction with other schizophrenia genes or environmen-
tal risk factors. Any association between hippocampal volume and apolipoprotein
E genotype in schizophrenia is at best unreplicated (Fernandez et al., 1999; Hata
et al., 2002). Finally, increased VBR and enlarged CSF spaces have been found to
be associated with a linkage marker on chromosome 5p in one large pedigree.
This is of note as the imaging findings (presumably reflecting generally reduced
brain volumes) were more strongly associated with the gene than with the disor-
der (Shihabuddin et al., 1996). There is, fortunately, rather stronger evidence for
specific environmental effects: particularly with arguably the best replicated risk
factor, namely OCs.

Obstetric complications
There is actually quite good evidence that people with family histories of schizophre-
nia do not suffer a greater number of or any more severe OCs. Rather, they appear
to be more sensitive to their effects – perhaps because they already have abnormal
brains. There is also an extensive literature on the clinical and imaging abnormalities
associated with OCs in neonates; a wide array of adverse events have the potential to
cause ventriculomegaly, periventricular leucomalacia, and/or hippocampal dam-
age, with a range of neurodevelopmental outcomes. What specificity there is in any
relationship between OCs and brain damage in schizophrenia will, therefore, arise
from gene–environment interactions. It is on this basis that a number of studies
have examined the association between OCs and enlarged ventricles or reduced
hippocampi in schizophrenia patients and related populations.

Ventricles
As with family history, studies of the associations between OCs and CT indices in
schizophrenia have produced inconsistent results. The largest CT studies of patients
tended to find no association (Johnstone et al., 1989; Jones et al., 1994; Owens et al.,
1985). Of 10 studies reviewed by Lewis (1990), five reported an association between
OCs and ventriculomegaly and two of these were studies of relatives. Further,
although Reveley et al. (1982) found an apparent environmental effect on the VBR
in discordant MZ twins and controls, which they attributed to common delivery
361 Premorbid structural abnormalities in schizophrenia

complications with more severe perinatal brain damage in the schizophrenia twin,
they actually found a stronger association between OCs and ventricular size in the
controls (Reveley et al., 1984). DeLisi et al. (1986) also suggested part genetic and
part environmental explanations for ventriculomegaly in their sample and found
such an effect for the frontal horns; however, both OCs and a history of head
injury in childhood were related to ventricular body enlargement, with no effect
attributable to a later diagnosis of schizophrenia.
Data from the Copenhagen High Risk Project strongly suggest a gene–
environment interaction. In the 27 high-risk subjects who were scanned initially and
whose midwife records were available, birth weight was inversely correlated (0.6)
with the VBR. There were weaker correlations with duration of labor and length
at birth (Schulsinger et al., 1984). Those with relatively lower weight (and those
with signs of prematurity) were more likely to have VBRs above the group median,
with stronger correlations in those who had subsequently developed schizophrenia
(0.77) than in those who had not (Silverton et al., 1985). On stepwise regression,
birth weight accounted for 22% of the variance in VBR in the high-risk group, but
64% in the UHR group with schizotypal fathers (Silverton et al., 1988a). There
was no apparent effect of genetic risk alone, maternal illness, or maternal neglect
(Silverton et al., 1988b). More sophisticated analysis revealed that an increased
VBR (and third ventricle width) was particularly related to an interaction between
the level of genetic risk and the severity of delivery complications, with a smaller
effect of low birth weight and no significant effect of total pregnancy complications
(Cannon et al., 1989). Increasing the sample to 157 subjects, with none, one, or two
affected parents, replicated these findings (Cannon et al., 1993).
A history of OCs has also been related to enlarged ventricles on sMRI in
schizophrenia patients (Alvir et al., 1999) and in relatives. Following Suddath et al.
(1990), McNeil et al. (2000) examined an extended sample of 22 discordant MZ
twin pairs and found that those with schizophrenia and large (right lateral and
total) ventricle volumes were more likely to have had labor and neonatal problems,
in addition to a prolonged labor, but no more total adverse events throughout
the pregnancy. More recently, Cannon et al. (2002b) compared 64 patients, 51 of
their unaffected siblings, and 54 controls with obstetric records of hypoxia, pre-
maturity, being small for dates and any perinatal complication. Hypoxic compli-
cations interacted with genetic risk to reduce whole-brain gray matter volumes,
especially in the temporal lobe and in schizophrenia patients, with tendencies
to do so for sulcal and ventricular CSF. Prematurity, small for gestational age
status, and prenatal infection had similar effects but were not associated with
adult schizophrenia and mainly rendered the brain more sensitive to the effects of
hypoxia.
362 S. M. Lawrie

Hippocampi
There is also some good evidence that OCs are related to small hippocampi in
schizophrenia, possibly through gene–environment interaction. A small study of
patients with a positive family history found no differences in limbic areas between
those with and without a history of birth complications (DeLisi et al., 1988), while
Stefanis and colleagues (1999) found smaller (left) hippocampi in patients with a
history of pregnancy and birth complications, but not in patients from multiply
affected families without OCs. McNeil et al. (2000) reported that intrapair dif-
ferences in hippocampal volumes between discordant MZ twins were related to
a prolonged labor (rather than neonatal complications). Using original hospital
records, and focusing on hypoxic events and “blue babies,” van Erp et al. (2002)
found that 72 patients had smaller hippocampi than 58 siblings, who, in turn, had
smaller volumes than healthy controls, but only schizophrenia patients with docu-
mented OCs had smaller hippocampi (siblings with OCs did not). The fact that OC
frequency was equal across the groups argues against gene–environment covaria-
tion, and the relatively small effect size for hypoxia in the probands (0.24) suggests a
sensitivity to the effects of OCs in those destined to develop schizophrenia. Further,
the intraclass correlation coefficients for healthy sibling pairs were higher than for
discordant pairs, suggesting that genetic influences on hippocampal volume are
larger in health than in schizophrenia, and that large genetic variation and/or
unique environmental events influence hippocampal volume in schizophrenia
patients.

Other obstetric complications

One possible non-shared environmental factor is exposure to infection in utero.
This is, however, rare and difficult to verify. Small and so far unreplicated stud-
ies have reported associations between reduced intracranial volume in patients
with schizophrenia and first-trimester famine (Hulshoff Pol et al., 2000) or con-
genital rubella (Lim et al., 1995). Sanderson et al. (2001) found that a small
number of patients with a history of early meningitis had reduced AHCs. Sea-
son of birth was a widely examined proxy for possible in utero infection in pre-
vious years, but most studies find no association between the VBR and time
of birth (e.g. DeQuardo et al., 1996; Silverton et al., 1988b). By far the largest
study did, however, find that patients born between December and April had
greater VBRs than those born in other months (Sacchetti et al., 1992). This
and the Lim et al. (1995) study are of additional interest as neither found any
apparent increases in CSF, as would be expected of an early disruption in brain
growth but contrary to what is usually found in schizophrenia (see Table 19.1 and
below).
363 Premorbid structural abnormalities in schizophrenia

Summary
There is fairly persuasive evidence that OCs, particularly labor complications and
proven hypoxia in neonates, interact with genetic risk for schizophrenia to produce
larger lateral ventricles and smaller hippocampi. It may be that a genetically small,
abnormal brain and other OCs increase the susceptibility to these effects. These
effects are likely to be observable at birth, although there is no direct evidence
for this. There are, however, indications that animals exposed experimentally to
hypoxemia in utero have these types of abnormality around term (e.g. Rees et al.,
1999). As during early development the enlarging brain drives intracranial vol-
ume expansion, an early developmental disruption should not be associated with
increases in cortical CSF (Woods, 1998); yet cortical CSF is known to be enlarged in
schizophrenia and appears to be related to genetic effects (see above). The implica-
tion is that a small brain in schizophrenia is, at least in part, attributable to postnatal
events.

Neurodevelopment and premorbid adjustment

Various neurodevelopmental abnormalities and social difficulties are known both
to predate and to increase the risk of schizophrenia (Cannon and Jones, 1996).
Although there are a few studies reporting an association between left handedness or
neurological “soft signs” and an increased VBR on CT, most of the evidence and the
largest studies do not find relationships with early development, academic record,
or premorbid social functioning (Johnstone et al., 1989; Jones et al., 1994; Lewis,
1990; Owens et al., 1985), and some studies have even reported better function and
adjustment in those with larger VBRs. Associations between these retrospective
measures and ROI volumes on sMRI are no more consistent. Correlations have,
for example, been reported between premorbid adjustment and overall gray matter
volume in women (Gur et al., 1999), hippocampal volume in men (Gur et al., 2000),
and ventral frontal cortex volume, but not whole brain, in both sexes (Chemerinski
et al., 2002). Fannon et al. (2000) reported an association between the severity of
various developmental problems and ventricular volumes. There is no knowing
how many times similar associations have been examined, found to be negative,
and not published.
Prospectively acquired data are likely to be much more reliable and there are
two reports of interest. In the Copenhagen High Risk Project, social work ratings
of scholastic and occupational adjustment at ages 15, 20, and 25 were worse in
those with enlarged ventricles on CT (Erel et al., 1991). Childhood home video
ratings of early neuromotor deficits and negative affect were associated with
ventricular enlargement on sMRI, while reported externalizing behavioral prob-
lems were related to smaller brains (Walker et al., 1996). The lack of consistent
364 S. M. Lawrie

correlations with structural measures on both CT and sMRI may, however, suggest
that function is more heavily determined by other factors and/or that there are
structural changes nearer to the time of psychosis. As the evidence for the latter
is increasing, one might expect that likely precipitants for schizophrenia – such as
stressful life events and cannabis consumption – have been examined in relation to
these findings. This does not appear to have happened as yet, at least in relation to
schizophrenia. While there are some studies of the structural effects of cannabis in
non-psychotic subjects, they are far from conclusive. There are a number of studies
in anxiety and depression suggesting that predisposing and precipitating stressors
are associated with reduced volumes of the hippocampus (reviewed by Sapolsky,
2000). Such research is now surely a priority, both to build up a coherent picture
of the pathophysiology of schizophrenia and because it may well have therapeutic
implications.

Conclusions
The results of sMRI studies of twins and of relatives of patients with schizophrenia
strongly suggest that at least some of the neuroanatomical features of the disorder
are present before onset and probably for some time. It is likely that reductions in the
volumes of the whole brain and frontal lobes are primarily genetically mediated and
that this partly accounts for reductions in the amygdala and hippocampus. Increases
in ventricular size and further reductions in the hippocampus are likely to occur
around the time of labor complications. This does not preclude later developmental
changes (some of which could be genetic) or degenerative effects. Relatively little is
known about structural brain development in childhood and adolescence, in health
and schizophrenia, but inference from the degrees of abnormality found in the above
studies suggests that there are further changes in frontal and especially temporal
lobes, perhaps particularly in medial structures, shortly before and possibly after
the onset of psychosis.
More of the same studies can be justified on the grounds that no finding is as
yet conclusive. However, ROI studies should probably have minimum group sizes
of about 60 – unless variance is reduced by recruiting within families – if sub-
tle differences in the brain and ventricles are to be examined, and particularly if
gene–environment interactions are to be assessed. More specific regional effects are
probably best assessed with automated approaches. It should be clear that case–
control and prospective cohort studies of those at elevated risk, for both genetic
and other reasons, are complementary. There may also be important information
gained from studies of early-onset schizophrenia and related neurodevelopmen-
tal disorders. Other analyses of sMRI data, such as calculating the gyral folding
365 Premorbid structural abnormalities in schizophrenia

index, and novel uses of MRI, such as proton spectroscopy and diffusion tensor
imaging, could give additional insights (but have not been described in this chap-
ter as there are as yet only interesting suggestions and no replicated findings in
relatives).
Perhaps the greatest gains in knowledge will arise from more novel approaches.
For example, ultrasound images of fetal and neonatal brains could be used to study
the time course of the effects of genetic and environmental risk factors relevant
to schizophrenia: although studying demonstrable effects rather than even well-
documented risks may be most relevant. Levels of psychosocial stress and cannabis
consumption need to be related to brain structure in subjects at high risk. Most
importantly, specific genetic abnormalities in schizophrenia (once clearly identi-
fied) may have regionally and temporally specific effects. The search for these genes
is, of course, informed by the foregoing studies, and both will be informed by
advances in the knowledge of the genetics and imaging of normal neurodevelop-
ment.
It will take some time before these developments lead to a sophisticated under-
standing of the time course of subregional anatomical differences in schizophrenia,
but it is not too early to begin to think about therapeutic implications. There is good
evidence, for example, that hypothermic treatment of hypoxic brains may reduce
the number of adverse neurodevelopmental outcomes. Once specified, structural
predictors of psychosis would provide a means of identifying subjects at very high
risk of conversion. This might justify early interventions to reduce stress or cannabis
use, and perhaps to prescribe antipsychotic or neuroprotective drugs, which might
ameliorate or even prevent the devastating effects of schizophrenia.

REFERENCES

Alvir, J. M. J., Woerner, M. G., Gunduz, H., Degreef, G., Lieberman, J. A. (1999). Obstretric
complications predict treatment response in first-episode schizophrenia. Psychol Med 29:
621–627.
Ananth, H., Popescu, I., Critchley, H. D., et al. (2002). Cortical and subcortical gray matter
abnormalities in schizophrenia determined through structural magnetic resonance imaging
with optimized volumetric voxel-based morphometry. Am J Psychiatry 159: 1497–1505.
Baare, W. F., van Oel, C. J., Hulshoff Pol, H. E., et al. (2001). Volumes of brain structures in twins
discordant for schizophrenia. Arch Gen Psychiatry 58: 33–40.
Bartley, A. J., Jones, D. W., Torrey, E. F., Zigun, J. R., Weinberger, D. R. (1993). Sylvian fissure
asymmetries in monozygotic twins: a test of laterality in schizophrenia. Biol Psychiatry 34:
853–863.
366 S. M. Lawrie

Bridle, N., Pantelis, C., Wood, S. J., et al. (2002). Thalamic and caudate volumes in monozygotic
twins discordant for schizophrenia. Aust N Z J Psychiatry 36: 347–354.
Buchsbaum, M. S., Yang, S., Hazlett, E., et al. (1997). Ventricular volume and asymmetry in
schizotypal personality disorder and schizophrenia assessed with magnetic resonance imag-
ing. Schizophr Res 27: 45–53.
Cannon, M., Jones, P. (1996). Schizophrenia. J Neurol Neurosurg Psychiatry 61: 604–613.
Cannon, T. D., Mednick, S. A., Parnas, J. (1989). Genetic and perinatal determinants of structural
brain deficits in schizophrenia. Arch Gen Psychiatry 46: 883–889.
Cannon, T. D., Mednick, S. A., Parnas, J., et al. (1993). Developmental brain abnormalities in the
offspring of schizophrenic mothers. I. Contributions of genetic and perinatal factors. Arch
Gen Psychiatry 50: 551–564.
(1994). Developmental brain abnormalities in the offspring of schizophrenic mothers. II.
Structural brain characteristics of schizophrenia and schizotypal personality disorder. Arch
Gen Psychiatry 51: 955–962.
Cannon, T. D., Kaprio, J., Lönnqvist, J., Huttunen, M., Koskenvuo, M. (1998a). The genetic
epidemiology of schizophrenia in a Finnish twin cohort. Arch Gen Psychiatry 55: 67–74.
Cannon, T. D., van Erp, T. G., Huttunen, M., et al. (1998b). Regional gray matter, white matter,
and cerebrospinal fluid distributions in schizophrenic patients, their siblings, and controls.
Arch Gen Psychiatry 55: 1084–1091.
Cannon, T. D., Thompson, P. M., van Erp, T. G., et al. (2002a). Cortex mapping reveals regionally
specific patterns of genetic and disease-specific gray-matter deficits in twins discordant for
schizophrenia. Proc Natl Acad Sci USA 99: 3228–3233.
Cannon, T. D., van Erp, T. G., Rosso, I. M., et al. (2002b). Fetal hypoxia and structural brain
abnormalities in schizophrenic patients, their siblings, and controls. Arch Gen Psychiatry 59:
35–41.
Chemerinski, E., Nopoulos, P. C., Crespo-Facorro, B., Andreasen, N. C., Magnotta, V. (2002).
Morphology of the ventral frontal cortex in schizophrenia: relationship with social dysfunc-
tion. Biol Psychiatry 52: 1–8.
Copolov, D., Velakoulis, D., McGorry, P., et al. (2000). Neurobiological findings in early phase
schizophrenia. Brain Res Rev 31: 157–165.
Dauphinais, I. D., DeLisi, L. E., Crow, T. J., et al. (1990). Reduction in temporal lobe size in siblings
with schizophrenia: a magnetic resonance imaging study. Psychiatry Res 35: 137–147.
DeLisi, L. E., Goldin, L. R., Hamovit, J. R., et al. (1986). A family study of the association of
increased ventricular size with schizophrenia. Arch Gen Psychiatry 43: 148–153.
DeLisi, L. E., Dauphinais, I. D., Gershon, E. S. (1988). Perinatal complications and reduced size
of brain limbic structures in familial schizophrenia. Schizophr Bull 14: 185–191.
DeQuardo, J. R., Goldman, M., Tandon, R. (1996). VBR in schizophrenia: relationship to family
history of psychosis and season of birth. Schizophr Res 20: 275–285.
Dickey, C. C., McCarley, R. W., Voglmaier, M. M., et al. (1999). Schizotypal personality disorder
and MRI abnormalities of temporal lobe gray matter. Biol Psychiatry 45: 1393–1402.
Dickey, C. C., Shenton, M. E., Hirayasu, Y., et al. (2000). Large CSF volume not attributable to
ventricular volume in schizotypal personality disorder. Am J Psychiatry 157: 48–54.
367 Premorbid structural abnormalities in schizophrenia

Downhill, J. E., Buchsbaum, M. S., Hazlett, E. A., et al. (2001). Temporal lobe volume deter-
mined by magnetic resonance imaging in schizotypal personality disorder and schizophre-
nia. Schizophr Res 48: 187–199.
Erel, O., Cannon, T. D., Hollister, J. M., Mednick, S. A., Parnas, J. (1991). Ventricular enlargement
and premorbid deficits in school-occupational attainment in a high risk sample. Schizophr
Res 4: 49–52.
Falkai, P., Honer, W. G., Alfter, D., et al. (2002). The temporal lobe in schizophrenia from uni-
and multiply affected families. Neurosci Lett 325: 25–28.
Fannon, D., Tennakoon, L., Sumich, A., et al. (2000). Third ventricle enlargement and devel-
opmental delay in first-episode psychosis: preliminary findings. Br J Psychiatry 177: 354–
359.
Fernandez, T., Yan, W. L., Hamburger, S., et al. (1999). Apolipoprotine E alleles in childhood-onset
schizophrenia. Am J Med Genet 88: 211–213.
Frangou, S., Sharma, T., Sigmudsson, T., et al. (1997). The Maudsley Family Study 4. Nor-
mal planum temporale asymmetry in familial schizophrenia. Br J Psychiatry 170: 328–
333.
Gur, R. E., Turetsky, B. I., Bilker, W. B., Gur, R. C. (1999). Reduced gray matter volume in
schizophrenia. Arch Gen Psychiatry 56: 905–911.
Gur, R. E., Turetsky, B. I., Cowell, P. E., et al. (2000). Temporolimbic volume reductions in
schizophrenia. Arch Gen Psychiatry 57: 769–775.
Harris, J. G., Young, D. A., Rojas, D. C., et al. (2002). Increased hippocampal volume
in schizophrenics’ parents with ancestral history of schizophrenia. Schizophr Res 55:
11–17.
Hata, T., Kunugi, H., Nanko, S., Fukuda, R., Kaminaga, T. (2002). Possible effect of the APOE
e4 allele on the hippocampal volume and asymmetry in schizophrenia. Am J Med Genet
Neuropsychiatr Genet 114: 641–642.
Honer, W. G., Bassett, A. S., Squires-Wheeler, E., et al. (1995). The temporal lobes, reversed
asymmetry and the genetics of schizophrenia. Neuroreport 7: 221–224.
Hulshoff Pol, H. E., Hoek, H. W., Susser, E., et al. (2000). Prenatal exposure to famine and brain
morphology in schizophrenia. Am J Psychiatry 157: 1170–1172.
Job, D. E., Whalley, H. C., McConnell, S., et al. (2002). Structural gray matter differences between
first-episode schizophrenics and normal controls on voxel-based morphometry. Neuroimage
17: 880–889.
Job, D. E., Whalley, H. C., Glabus, M., Johnstone E. C., Lawrie, S. M. (2003). Voxel-based
morphometry of grey matter reductions in subjects at high risk of schizophrenia. Schizophr
Res 64: 1–13.
Johnstone, E. C., Owens, D. G., Bydder, G. M., et al. (1989). The spectrum of structural brain
changes in schizophrenia: age of onset as a predictor of cognitive and clinical impairments
and their cerebral correlates. Psychol Med 19: 91–103.
Johnstone, E. C., Cosway, R., Lawrie, S. M. (2002). Distinguishing characteristics of subjects
with good and poor early outcome in the Edinburgh High Risk Study. Br J Psychiatry 181:
S26–S29.
368 S. M. Lawrie

Jones, P. B., Harvey, I., Lewis, S. W., et al. (1994). Cerebral ventricle dimensions as risk factors
for schizophrenia and affective psychosis: an epidemiological approach to analysis. Psychol
Med 24: 995–1011.
Keshavan, M. S., Montrose, D. M., Pierri, J. N., et al. (1997). Magnetic resonance imaging and
spectroscopy in offspring at risk for schizophrenia: preliminary studies. Prog Neuropsy-
chopharmacol Biol Psychiatry 21: 1285–1295.
Keshavan, M. S., Dick, E., Mankowski, I., et al. (2002). Decreased left amygddala and hippocampal
volumes in young offspring at risk for schizophrenia. Schizophr Res 58: 173–183.
Konick, L. C., Friedman, L. (2001). Meta-analysis of thalamic size in schizophrenia. Biol Psychiatry
49: 28–38.
Kunugi, H., Hattori, M., Nanko, S., et al. (1999). Dinucleotide repeat polymorphism in the
neurotrophin-3 gene and hippocampal volume in psychoses. Schizophr Res 37: 271–
273.
Lawrie, S. M., Abukmeil, S. S. (1998). Brain abnormality in schizophrenia. A systematic and
quantitative review of volumetric magnetic resonance imaging studies. Br J Psychiatry 172:
110–120.
Lawrie, S. M., Whalley, H., Kestelman, J. N., et al. (1999). Magnetic resonance imaging of brain
in people at high risk of developing schizophrenia. Lancet 353: 30–33.
Lawrie, S. M., Adams, C. E., Thornley, B., Joy, C. (2000). Comprehensiveness of systematic review:
update. Br J Psychiatry 176: 396–401.
Lawrie, S. M., Whalley, H. C., Abukmeil, S. S., et al. (2001). Brain structure, genetic liability and
psychotic symptoms in subjects at high risk of developing schizophrenia. Biol Psychiatry 49:
811–823.
(2002). Temporal lobe volume changes in subjects at high risk of schizophrenia with psychotic
symptoms. Br J Psychiatry 181: 138–143.
Levitt, J. J., McCarley, R. W., Dickey, C. C., et al. (2002). MRI study of caudate nucleus volume and
its cognitive correlates in neuroleptic-naive patients with schizotypal personality disorder.
Am J Psychiatry 159: 1190–1197.
Lewis, S. W. (1990). Computerised tomography in schizophrenia 15 years on. Br J Psychiatry 157:
16–24.
Lim, K. O., Beal, D. M., Harvey, Jr., R. L., et al. (1995). Brain dysmorphology in adults with
congenital rubella plus schizophrenia-like symptoms. Biol Psychiatry 37: 764–776.
McDonald, C., Grech, A., Toulopoulou, T., et al. (2002). Brain volumes in familial and non-familial
schizophrenic probands and their unaffected relatives. Am J Med Genet (Neuropsychiatric
Genetics) 114: 616–625.
McGuffin, P., Asherson, P., Owen, M., Farmer, A. (1994). The strength of the genetic effect. Is
there room for an environmental influence in the aetiology of schizophrenia? Br J Psychiatry
164: 593–599.
McNeil, T. F., Cantor-Graae, E., Weinberger, D. R. (2000). Relationship of obstetric complica-
tions and differences in size of brain structures in monozygotic twin pairs discordant for
schizophrenia. Am J Psychiatry 157: 203–212.
369 Premorbid structural abnormalities in schizophrenia

Meyer-Lindenberg, A., Japee, S., Verchinski, B., et al. (2001). Structural MRI abnormalities in
schizophrenic patients and their unaffected siblings: voxel-based morphometry. Neuroimage
13: S217.
Mourot, A., d’Amato, T., Rochet, T., et al. (1997). Cerebral investigation of healthy siblings of
schizophrenics. Eur Psychiatry 12: 273–278.
Narr, K. L., van Erp, T. G., Cannon, T. D., et al. (2002). A twin study of genetic contributions to
hippocampal morphology in schizophrenia. Neurobiol Dis 11: 83–95.
Nelson, M. D., Saykin, A. J., Flashman, L. A., Riordan, H. J. (1998). Hippocampal volume reduc-
tion in schizophrenia as assessed by magnetic resonance imaging: a meta-analytic study.
Arch Gen Psychiatry 55: 433–440.
Noga, J. T., Bartley, A. J., Jones, D. W., Torrey, E. F., Weinberger, D. R. (1996). Cortical gyral
anatomy and gross brain dimensions in monozygotic twins discordant for schizophrenia.
Schizophr Res 22: 27–40.
O’Driscoll, G. A., Florencio, P. S., Gagnon, D., et al. (2001). Amygdala–hippocampal volume and
verbal memory in first-degree relatives of schizophrenic patients. Psychiatry Res Neuroimag-
ing 107: 75–85.
Owens, D. G., Johnstone, E. C., Crow, T. J., et al. (1985). Lateral ventricular size in schizophrenia:
relationship to the disease process and its clinical manifestations. Psychol Med 15: 27–41.
Paillere-Martinot, M., Caclin, A., Artiges, E., et al. (2001). Cerebral gray and white matter reduc-
tions and clinical correlates in patients with early onset schizophrenia. Schizophr Res 50:
19–26.
Pantelis, C., Velakoulis, D., Suckling, J., et al. (2000). Left medial temporal volume reduction
occurs during the transition from high-risk to first-episode psychosis. Schizophr Res 41: 35.
Pantelis, C., Velakoulis, D., McGorry, P. D., et al. (2003). Neuroanatomical abnormalities before
and after onset of psychosis: a cross-sectional and longitudinal MRI comparison. Lancet
361: 281–288.
Phillips, L. J., Velakoulis, D., Pantelis, C., et al. (2002). Non-reduction in hippocampal volume is
associated with higher risk of psychosis. Schizophr Res 58: 145–158.
Raz, S., Raz, N. (1990). Structural brain abnormalities in the major psychoses: a quantitative
review of the evidence from computerized imaging. Psychol Bull 108: 93–108.
Rees, S., Breen, S., Loeliger, M., McCrabb, G., Harding, R. (1999). Hypoxemia near mid-gestation
has long-term effects on fetal brain development. J Neuropathol Exp Neurol 58: 932–945.
Reveley, A. M., Reveley, M. A., Clifford, C. A., Murray, R. M. (1982). Cerebral ventricular size in
twins discordant for schizophrenia. Lancet i: 540–541.
Reveley, A. M., Reveley, M. A., Murray, R. M. (1984). Cerebral ventricular enlargement in non-
genetic schizophrenia: a controlled twin study. Br J Psychiatry 144: 89–93.
Roy, M. A., Flaum, M. A., Arndt, S. V., Crowe, R. R., Andreasen, N. C. (1994). Magnetic resonance
imaging in familial versus sporadic cases of schizophrenia. Psychiatry Res 54: 25–36.
Sacchetti, E., Calzeroni, A., Vita, A., et al. (1992). The brain damage hypothesis of the season-
ality of births in schizophrenia and major affective disorders: evidence from computerised
tomography. Br J Psychiatry 160: 390–397.
370 S. M. Lawrie

Sanderson, T. L., Doody, G. A., Best, J., Owens, D. G. C., Johnstone, E. C. (2001). Correla-
tions between clinical and historical variables and cerebral structural variables in peo-
ple with mild intellectual disability and schizophrenia. J Intellect Disabil Res 45: 89–
98.
Sapolsky, R. M. (2000). Glucocorticoids and hippocampal atrophy in neuropsychiatric disorders.
Arch Gen Psychiatry 57: 925–935.
Schreiber, H., Baur-Seack, K., Kornhuber, H. H., et al. (1999). Brain morphology in adolescents
at genetic risk for schizophrenia assessed by qualitative and quantitative magnetic resonance
imaging. Schizophr Res 40: 81–84.
Schulsinger, F., Parnas, J., Petersen, E. T., et al. (1984). Cerebral ventricular size in the offspring
of schizophrenic mothers. Arch Gen Psychiatry 41: 602–606.
Schwarzkopf, S. B., Nasrallah, H. A., Olson, S. C., Bogerts, B., McLaughlin, J. A., Mitra, T. (1991).
Family history and brain morphology in schizophrenia: an MRI study. Psychiatry Res 40:
49–60.
Seidman, L. J., Faraone, S. V., Goldstein, J. M., et al. (1997). Reduced subcortical brain volumes in
nonpsychotic siblings of schizophrenic patients: a pilot magnetic resonance imaging study.
Am J Med Genet 74: 507–514.
(1999). Thalamic and amygdala-hippocampal volume reductions in first-degree relatives of
patients with schizophrenia: an MRI-based morphometric analysis. Biol Psychiatry 46: 941–
954.
(2002). Left hippocampal volume as a vulnerability indicator for schizophrenia. Arch Gen
Psychiatry 59: 839–849.
Shapleske, J., Rossell, S. L., Woodruff, P. W. R., David, A. S. (1999). The planum temporale: a
systematic, quantitative review of its structural, functional and clinical significance. Brain
Res Rev 29: 26–49.
Sharma, T., Lancaster, E., Lee, D., et al. (1998). Brain changes in schizophrenia. Volumetric MRI
study of families multiply affected with schizophrenia: the Maudsley Family Study 5. Br J
Psychiatry 173: 132–138.
Sharma, T., Lancaster, E., Sigmundsson, T., et al. (1999). Lack of normal pattern of cerebral
asymmetry in familial schizophrenic patients and their relatives: The Maudsley Family
Study. Schizophr Res 40: 111–120.
Shihabuddin, L., Silverman, J. M., Buchsbaum, M. S., et al. (1996). Ventricular enlargement
associated with linkage marker for schizophrenia-related disorders in one pedigree. Mol
Psychiatry 1: 215–222.
Siever, L. J., Rotter, M., Losonczy, M., et al. (1995). Lateral ventricular enlargement in schizotypal
personality disorder. Psychiatry Res 57: 109–118.
Sigmundsson, T., Suckling, J., Maier, M., et al. (2001). Structural abnormalities in frontal, tem-
poral, and limbic regions and interconnecting white matter tracts in schizophrenic patients
with prominent negative symptoms. Am J Psychiatry 158: 234–243.
Silverman, J. M., Smith, C. J., Guo, S. L., et al. (1998). Lateral ventricular enlargement in
schizophrenic probands and their siblings with schizophrenia-related disorders. Biol Psy-
chiatry 43: 97–106.
371 Premorbid structural abnormalities in schizophrenia

Silverton, L., Finello, K. M., Schulsinger, F., Mednick, S. A. (1985). Low birth weight and ven-
tricular enlargement in a high-risk sample. J Abnorm Psychol 94: 405–409.
Silverton, L., Mednick, S. A., Schulsinger, F., Parnas, J., Harrington, M. E. (1988a). Genetic risk
for schizophrenia, birthweight, and cerebral ventricular enlargement. J Abnorm Psychol 97:
496–498.
Silverton, L., Mednick, S. A., Harrington, M. E. (1988b). Birthweight, schizophrenia and ventric-
ular enlargement in a high-risk sample. Psychiatry 51: 272–280.
Sommer, I., Aleman, A., Ramsey, N., Bouma, A., Kahn, R. (2001). Handedness, language later-
alisation and anatomical asymmetry in schizophrenia. Meta-analysis. Br J Psychiatry 178:
344–351.
Staal, W. G., Hulshoff, Pol, H. E., Schnack, H., van der Schot, A. C., Kahn, R. S. (1998). Partial
volume decrease of the thalamus in relatives of patients with schizophrenia. Am J Psychiatry
155: 1784–1786.
Staal, W. G., Hulshoff Pol, H. E., Schnack, H. G., et al. (2000). Structural brain abnormalities in
patients with schizophrenia and their healthy siblings. Am J Psychiatry 157: 416–421.
Steel, R., Whalley, H., Miller, P., et al. (2002). Structural MRI of the brain in presumed carriers of
genes for schizophrenia, their affected and unaffected siblings. J Neurol Neurosurg Psychiatry
72: 455–458.
Stefanis, N., Frangou, S., Yakeley, J., et al. (1999). Hippocampal volume reduction in schizophre-
nia: effects of genetic risk and pregnancy and birth complications. Biol Psychiatry 46: 697–
702.
Suddath, R. L., Christison, G. W., Torrey, E. F., Casanova, M. F., Weinberger, D. R. (1990).
Anatomical abnormalities in the brains of monozygotic twins discordant for schizophrenia.
N Engl J Med 322: 789–794.
van Erp, T. G. M., Saleh, P. A., Rosso, I. M., et al. (2002). Contributions of genetic risk and
fetal hypoxia to hippocampal volume in patients with schizophrenia or schizoaffective dis-
order, their unaffected siblings, and healthy unrelated volunteers. Am J Psychiatry 159:
1514–1520.
van Horn, J. D., McManus, I. C. (1992). Ventricular enlargement in schizophrenia: a meta-analysis
of studies of the ventricle: brain ratio (VBR). Br J Psychiatry 160: 687–697.
Vita, A., Dieci, M., Giobbio, G. M., et al. (1994). A reconsideration of the relationship between
cerebral structural abnormalities and family history of schizophrenia. Psychiatry Res 53:
41–55.
Waldo, M. C., Cawthra, E., Adler, L. E., et al. (1994). Auditory sensory gating, hippocampal
volume, and catecholamine metabolism in schizophrenics and their siblings. Schizophr Res
12: 93–106.
Walker, E. F., Lewine, R. R., Neumann, C. (1996). Childhood behavioral characteristics and adult
brain morphology in schizophrenia. Schizophr Res 22: 93–101.
Ward, K. E., Friedman, L., Wise, A., Schulz, S. C. (1996). Meta-analysis of brain and cranial size
in schizophrenia. Schizophr Res 22: 197–213.
Weinberger, D. R., DeLisi, L. E., Neophytides, A. N., Wyatt, R. J. (1981). Familial aspects of CT
scan abnormalities in chronic schizophrenic patients. Psychiatry Res 4: 65–71.
372 S. M. Lawrie

Woodruff, P. W. R., McManus, I. C., David, A. S. (1995). Meta-analysis of corpus callosum size
in schizophrenia. J Neurol Neurosurg Psychiatry 58: 457–461.
Woods, B. T. (1998). Is schizophrenia a progressive neurodevelopmental disorder? Toward a
unitary pathogenetic mechanism. Am J Psychiatry 155: 1661–1670.
Wright, I. C., Rabe-Hesketh, S., Woodruff, P. W., et al. (2000). Meta-analysis of regional brain
volumes in schizophrenia. Am J Psychiatry 157: 16–25.
Zorrilla, L. T. E., Cannon, T. D., Kronenberg, S., et al. (1997). Structural brain abnormalities in
schizophrenia: a family study. Biol Psychiatry 42: 1080–1986.
 20

Neurodegenerative models of
schizophrenia
L. Fredrik Jarskog, John H. Gilmore, and Jeffrey A. Lieberman
University of North Carolina School of Medicine, Chapel Hill, USA

The concept of schizophrenia as a neurodegenerative disorder has a long and some-

what controversial past. Neurodegenerative aspects of schizophrenia were recog-
nized early in the twentieth century by Kraeplin (1919), who described “dementia
praecox” as a progressive illness with a deteriorating clinical course. This hypoth-
esis was originally supported both by Alzheimer, who found evidence of cortical
neuron loss, and by Southard, who described cerebral atrophy; however, inves-
tigators during the remainder of the twentieth century were largely unable to
demonstrate consistent and replicable evidence of neurodegeneration (Pinals and
Breier, 1997). The reframing of schizophrenia as a disorder of neurodevelopment
was met with considerable interest, given the absence of a clear neuropathological
phenotype to support a neurodegenerative hypothesis (Murray and Lewis, 1987;
Weinberger, 1987). The neurodevelopmental hypothesis postulates that neurobi-
ological insults can adversely influence the course of early neurodevelopment to
yield permanent brain deficits; these ultimately manifest as psychosis during the
second or third decades of life. Furthermore, it suggests that the illness runs a het-
erogeneous course following the onset of clinical symptoms but that no further
pathophysiological consequences are incurred beyond those associated with nor-
mal maturation, aging, and treatment-related factors (Weinberger, 1987). Clearly,
the neurodevelopmental hypothesis has substantially advanced our understanding
of schizophrenia vis-à-vis the important role that developmental factors play in
the etiology and pathophysiology of the disorder. Nevertheless, the neurodevelop-
mental perspective fails to account adequately for a number of cardinal features of
schizophrenia, including the protracted period of symptomatic dormancy between
the putative insult and the emergence of clinical symptoms, the progressive clinical
deterioration that affects a significant subgroup of patients, recent evidence for
progressive changes in ventricular and cortical brain structures, and the apparent
ability of antipsychotic treatment to modify the course of the illness.

Neurodevelopment and Schizophrenia, ed. Matcheri S. Keshavan et al. Published by Cambridge University
Press.

C Cambridge University Press 2004.

373
374 L. F. Jarskog, J. H. Gilmore, and J. A. Lieberman

Phase of Illness Premorbid Prodromal Recent-Onset Chronic and Residual

Psychosis
Onset of
Psychosis
Normal
LIMITED
Psychopathology NEURODEGENERATION
and
Level of Function NEURODEVELOPMENTAL
PROCESSES
Severe

Fig. 20.1. Model of schizophrenia as a limited neurodegenerative disorder with neurodevelopmen-

tal antecedents. The line represents the functional and psychopathological course of
schizophrenia superimposed over the proposed intersection of developmental and degener-
ative processes. As depicted, the degenerative pathophysiology is most active in the interval
surrounding the onset of psychosis and then attenuates with time.

A neurodegenerative disorder is, by definition, a progressive disease of the ner-

vous system that is initiated by specific biochemical changes which ultimately lead
to a distinct histopathology with associated clinical manifestations (Hardy and
Gwinn-Hardy, 1998). The disorder often has an underlying genetic basis. The tim-
ing of the onset of clinical symptoms is usually influenced by the interaction of
the disease process with normal developmental and maturational processes, as well
as with the individual’s capacity for adaptive neuroplastic responses. Thus, neu-
rodegenerative disorders may remain clinically dormant until a threshold level of
pathophysiological damage has occurred. The absence of a histopathological phe-
notype in schizophrenia has been cited as evidence against a neurodegenerative
hypothesis (Weinberger and McClure, 2002). However, studies of schizophrenia
increasingly demonstrate subtle yet consistent histopathological deficits in addi-
tion to evidence of progressive clinical and neuroimaging findings. We believe
that schizophrenia can be considered as a limited neurodegenerative disorder with
neurodevelopmental antecedents (Fig. 20.1). Studies from clinical, neurocognitive,
neuroimaging, and neuropathological domains will be reviewed in a critical analysis
of this hypothesis.

Clinical psychopathology and neurocognition

As reviewed elsewhere in this volume, there is substantial evidence that patients in
the premorbid phase of schizophrenia have subtle abnormalities of motor, cognitive,
and social functioning. From this baseline of subtle impairment, most patients with
schizophrenia experience deterioration in overall level of functioning following the
formal onset of illness. This has been convincingly documented for a century by
numerous investigators (Huber et al., 1980; Kraeplin, 1919; Pfohl and Winokur,
1982). Functional deterioration appears to occur mainly in the first 5 years following
375 Neurodegenerative models of schizophrenia

the onset of psychosis and then reaches a plateau of stability (McGlashan, 1988).
A more recent study suggests that the deterioration is confined to the first year of
illness: Mason et al. (1996) found that social adjustment deteriorated in the first
year of illness, though remained relatively stable over the next 13 years.
While there is general agreement that overall functioning declines in schizophre-
nia, it is not clear what factors contribute to this functional deterioration. The most
parsimonious explanation for this deterioration may be that it is a consequence of
the symptoms of schizophrenia itself: positive and negative symptoms, and cogni-
tive dysfunction. Untreated or partially treated positive symptoms would obviously
interfere with overall functioning and quality of life, and degeneration or progres-
sion of illness would not necessarily be involved. Both Eaton et al. (1995) and
Mason et al. (1996) found that positive symptoms tend to remain stable after the
first year of illness. There is evidence that primary negative symptoms or the deficit
syndrome does progress in the early years of the illness. In a retrospective chart
review study, Fenton and McGlashan (1994) found that primary negative symp-
toms progressed in severity over the first 5 years of the illness and that the deficit
syndrome was associated with poor outcome and long-term disability. Mayerhoff
et al. (1994) found that first-episode patients had lower rates of the deficit syn-
drome than more chronic populations, and Eaton et al. (1995) found that negative
symptoms were stable after the first 2 years of illness. These studies suggest that
progression of negative symptoms may be limited to the first few years of ill-
ness, though clarification of this issue awaits additional prospective longitudinal
studies.
While initial studies reported progressive cognitive impairment in schizophrenia
(Abrahamson, 1983; Bilder et al., 1992; Smith, 1964), most of the more recent studies
do not indicate a progression of cognitive dysfunction (reviewed by Rund, 1998).
While cognitive decline has been described in geriatric, chronically institutionalized
patients (Friedman et al., 2002; Harvey et al., 1999), a study of older outpatients
found no differences in age-related cognitive decline (Zorrilla et al., 2000). A more
recent longitudinal study of neuropsychological deficits in older patients (mean age
47.6 years) with schizophrenia found them to be stable (Heaton et al., 2001). These
studies suggest that, if progressive deterioration of cognitive functioning occurs
in older patients with schizophrenia, this deterioration is limited to a subset of
very severely ill, chronically institutionalized patients. Cross-sectional comparisons
of cognitive functioning in patients with first-episode and chronic schizophrenia
have found similar degrees of impairment, suggesting that cognitive deficits do not
progress (Addington and Addington, 2002; Moritz et al., 2002; Saykin et al., 1994).
Longitudinal studies in patients with first-episode schizophrenia found either no
progression (Hoff et al., 1999) or even improvement of cognitive impairments
(Nopoulos et al., 1994).
376 L. F. Jarskog, J. H. Gilmore, and J. A. Lieberman

Although available evidence suggests that neurocognitive deficits do not progress

after the first episode of psychosis, recent data suggest that neurocognitive deteri-
oration may occur during the premorbid and prodromal phases. Children who go
on to develop schizophrenia have a variety of cognitive, motor, and social abnor-
malities in the premorbid stage of their illness (Cannon et al., 2002; Done et al.,
1994; Jones et al., 1994; Kremen et al., 1994). In general, the impairment of patients
in the premorbid and prodromal phases of the disease appears to be less severe than
that documented after the formal onset of symptoms, suggesting that some deteri-
oration has occurred. This is supported by a small study reporting that worsening
neuropsychological deficits in prodromal patients tended to predict a transition to
psychosis (Hambrecht et al., 2002). The strength of this conclusion is limited by
the few studies that have followed prodromal patient cohorts through the transi-
tion to psychosis. Nevertheless, the available data suggest that progressive cognitive
impairment may occur prior to the first episode. This idea is supported by a recent
study showing that children who go on to develop schizophrenia appear to have
deterioration on a standardized scholastic test between grades 8 and 11 (Fuller
et al., 2002). This deterioration was not related to the age of onset, indicating that
the deterioration preceded the first episode. Deterioration late in the premorbid
phase, prior to formal onset of the illness, is also suggested in a study by Cannon
et al. (1999), who found that children who went on to develop schizophrenia often
did not attend or complete high school, even though they had normal academic
performance in elementary school.
The weight of the evidence indicates that progression of cognitive deficits does
occur at some time during the late premorbid, prodromal, or early first-episode
phases of schizophrenia. Neurocognitive functioning appears to remain stable sub-
sequent to the first episode, and there is no significant age-related decline in func-
tioning except in very ill, chronically institutionalized patients. Primary negative
symptoms appear to worsen during the first few years of the illness and remain
stable thereafter. In a similar fashion, overall functional impairment worsens in the
first few years of the illness and then stabilizes. If we infer neurodegeneration from
the progression of these clinical components of schizophrenia, then the degenera-
tion is limited to the late premorbid/prodromal phase and the first episode of the
illness (Fig. 20.1). Thus, a neurodegenerative component would occur in the con-
text of concomitant neurodevelopmental processes that normally operate during
adolescence.

Postmortem neuropathology
During the twentieth century, investigations of the postmortem neuropathology
of schizophrenia have been most notable for demonstrating the absence of a clear
377 Neurodegenerative models of schizophrenia

histopathological phenotype (Harrison, 1999). Although Stevens (1982) reported

increased periventricular and subependymal gliosis in schizophrenia, subsequent
studies have found no evidence of excess gliosis or increased glial fibrillary acidic
protein staining (Arnold et al., 1998; Benes et al., 1991; Purohit et al., 1998;
Roberts et al., 1986, 1987). The absence of glial proliferation clearly distinguishes
the pathophysiology of schizophrenia from that of classic neurodegeneration (e.g.
Alzheimer’s disease), where gliosis is prominent.
Neuronal cell loss is another potential marker of neurodegeneration. Alzheimer’s
disease and other classic neurodegenerative disorders are characterized by large-
scale neuronal loss, especially during their later stages. In schizophrenia, many
studies have not identified reduced numbers of cortical neurons (Akbarian et al.,
1995; Pakkenberg, 1993; Selemon et al., 1995, 1998). However, several studies
have demonstrated significant yet relatively small layer-specific reductions in the
density of pyramidal and non-pyramidal neurons in anterior cingulate cortex in
schizophrenia, interpreted to reflect reductions in neuronal number (Benes et al.,
1991, 2001). Interestingly, a recent study also suggested that glial density may be
reduced in anterior cingulate cortex (Cotter et al., 2001). Furthermore, substantial
reductions of neurons have been identified in several subcortical regions including
nucleus accumbens (Young et al., 2000) and mediodorsal thalamus (Pakkenberg,
1990). Therefore, while large-scale cortical cell loss is clearly absent in schizophre-
nia, available data indicate that limited neuronal and/or glial reductions may occur
with layer and regional specificity, especially in subcortical areas.
While neuronal cell loss may not be a primary cytoarchitectural deficit in
schizophrenia, there is increasing evidence that the cortical neuropathology is
characterized by limited atrophy. Several studies indicate that neuronal density
is increased in a setting of slight cortical thinning, suggesting that cortical neuropil
is reduced (Selemon et al., 1995, 1998). Other studies have found reduced dendritic
spines and total dendritic length (Black et al., 2002; Garey et al., 1998; Glantz and
Lewis, 2000) as well as reductions of certain synaptic marker proteins (Eastwood
and Harrison, 1995; Glantz and Lewis, 1997; Karson et al., 1999; Thompson et al.,
1998). Small reductions in somal volume of neurons in prefrontal cortex have also
been reported in schizophrenia (Pierri et al., 2001; Rajkowska et al., 1998). Because
neuronal size appears to be related to the extent of dendritic and axonal arboriza-
tion (Harrison and Eastwood, 2001), reduced somal size is consistent with reduced
dendrites and synaptic markers in schizophrenia.
In efforts to identify alternative pathophysiological mechanisms that could con-
tribute to the neuropathology of schizophrenia, a role for apoptosis has been increas-
ingly hypothesized (Lewis and Lieberman, 2000; Margolis et al., 1994; Woods, 1998).
Although best recognized for its association with rapid cell death, emerging data
indicate that apoptosis can also produce non-lethal reductions in cellular viability
378 L. F. Jarskog, J. H. Gilmore, and J. A. Lieberman

Pre-degenerative Onset Post-degenerative Onset

Thicker cortical gray matter Thinner cortical gray matter

Smaller ventricles Larger ventricles

Larger soma Smaller soma

- shorter dendrites
- longer dendrites - less complex branching
- more complex branching

Fig. 20.2. Model of limited neurodegeneration on structural and histological levels. Longitudinal
neuroimaging studies have demonstrated progressive cortical gray matter loss and ven-
tricular enlargement in schizophrenia, especially in childhood-onset and adult new-onset
populations. The limited neurodegeneration model proposes that cytopathological corre-
lates of gray matter loss include progressive neuronal atrophy and neuropil loss.

(Ona et al., 1999) and underlies pathological elimination of synapses in a process

termed synaptic apoptosis (Mattson et al., 1998). Interestingly, some studies have
indicated abnormalities in apoptotic mechanisms, including pro- and antiapop-
totic proteins (Jarskog et al., 2000, 2001) and DNA fragmentation profiles (Benes
et al., 2003) in postmortem cortical tissue, suggesting that apoptotic pathways may
indeed underlie certain neuropathological deficits in schizophrenia.
Therefore, while there is no evidence of classic neurodegenerative lesions in
schizophrenia, it is clear that the disorder is characterized by a number of more
subtle neuropathological deficits. Unfortunately, these findings in postmortem tis-
sue do not provide a longitudinal perspective and, therefore, offer little insight
as to the timing of the events. In other words, limited reductions in glia, den-
drites, or neuronal soma could be a consequence of early developmental agenesis
or of an atrophic process later in life. Furthermore, these findings must be con-
sidered in the context of a number of potential confounding variables, including
diagnostic uncertainty, postmortem interval, medication effects, drug abuse, and
non-specific effects of chronic illness, to name a few. Nevertheless, we suggest that
the emerging picture of subtle dendritic and synaptic deficits in schizophrenia in
the context of structural neuroimaging findings and clinical deterioration repre-
sents a limited form of neurodegeneration that will become increasingly defined as
molecular techniques and genomic/proteomic analyses are applied to the disorder
(Fig. 20.2).
379 Neurodegenerative models of schizophrenia

Neuroimaging studies
Structural studies
Cross-sectional structural neuroimaging studies have demonstrated a number of
consistent brain abnormalities in schizophrenia primarily involving enlargement of
the lateral and third ventricles and reduction in cortical gray matter volume, both
globally (Gur et al., 1998; Hulshoff Pol et al., 2002; Zipursky et al., 1992) and in
cortical subregions including prefrontal (Gur et al., 1998; Hulshoff Pol et al., 2002),
temporal (Gur et al., 1998; Schlaepfer et al., 1994), and parietal (Schlaepfer et al.,
1994) areas. Since neuropil is a primary component of gray matter, these structural
imaging findings provide in vivo support for the postmortem findings of reduced
cortical neuropil (Selemon et al., 1995, 1998).
One of the limitations of cross-sectional studies is that they do not adequately
address whether a change in brain volume is active or antecedent to the scan. For this
reason, a number of longitudinal structural magnetic resonance imaging (sMRI)
studies have been conducted to address this issue. Interestingly, progressive brain
volume changes have been identified in several different patient cohorts, includ-
ing in patients with prodromal signs of psychosis, in childhood-onset schizophre-
nia, in new-onset schizophrenia, and in certain chronic patient populations. In a
single small study of prodromal patients judged at high risk of transition to psy-
chosis, a progressive loss of gray matter was found in certain hippocampal and
frontal regions only in those patients that developed psychosis (Pantelis et al.,
2002). In childhood-onset disease, several progressive volume changes have been
reported compared with normal controls, including ventricular enlargement
(Rapoport et al., 1997), cortical gray matter loss in frontal and temporal areas
(Jacobsen et al., 1998; Rapoport et al., 1999), and cerebellar volume loss (Keller
et al., 2003). In new-onset schizophrenia, progressive ventricular enlargement has
been identified by several groups (Cahn et al., 2002; DeLisi et al., 1997; Lieber-
man et al., 2001a) but not others (Gur et al., 1998). Progressive volume loss has
been reported in first-episode schizophrenia in cerebral hemispheres bilaterally
(DeLisi et al., 1997), total cerebral gray matter (Cahn et al., 2002), frontal cortex
(Gur et al., 1998), and in superior temporal gyrus (Kasai et al., 2003), while Lieber-
man et al. (2001a) did not detect cortical volume differences. In adults with chronic
schizophrenia, progressive ventricular enlargement has been reported in male
patients (Mathalon et al., 2001) and in patients with poor outcome (Davis et al.,
1998), also suggested by Nair et al. (1997). Mathalon et al. (2001) also reported
accelerated progression of frontotemporal cortical gray matter loss in males with
schizophrenia.
Although the interpretation that progressive brain changes reflect a neurodegen-
erative process in schizophrenia has been challenged (Weinberger and McClure,
380 L. F. Jarskog, J. H. Gilmore, and J. A. Lieberman

2002), the increasing number of studies that have identified accelerated corti-
cal tissue loss and ventricular enlargement suggest that the pathophysiology of
schizophrenia is a dynamic and progressive process, particularly around the onset
of psychosis (Lewis and Lieberman, 2000; Fig. 20.2). Longitudinal studies that
employ higher-resolution neuroimaging techniques should help to define this pro-
cess more precisely.

Functional and spectroscopic studies

Reduced metabolic activity in the prefrontal cortex (hypofrontality) in schizophre-
nia has been demonstrated using positron emission tomography (PET) and
functional MRI (fMRI) studies (Andreasen et al., 1997; Weinberger et al., 1986).
Since these studies did not follow patients longitudinally, it is not possible to
determine whether this metabolic hypofrontality represents a degenerative pro-
cess. Somewhat more revealing are studies of phosphorus-31 magnetic resonance
spectroscopy (MRS) in the prefrontal cortex, which demonstrated decreased phos-
phomonoesters (Hinsberger et al., 1997; Pettegrew et al., 1991; Stanley et al., 1995)
and increased phosphodiesters (Pettegrew et al., 1991; Stanley et al., 1995) in first-
episode schizophrenia. A profile of reduced phosphomonoesters and increased
phosphodiesters is thought to reflect an accelerated breakdown of membrane phos-
pholipids (Pettegrew et al., 1993). This could indicate an active neurodegenerative
process whereby neuronal and/or glial components are undergoing degradation. In
addition, several studies using proton MRS have reported reduced N-acetylaspartate
levels in temporal and frontal cortex in schizophrenia (Bertolino et al., 1998, Cecil
et al., 1999). N-Acetylaspartate is generally considered a measure of neuronal
metabolism and reduced levels may, therefore, reflect reduced neuronal viabil-
ity. Taken together, these functional and spectroscopic studies are only moderately
indicative of a limited neurodegenerative process in schizophrenia. Applying longi-
tudinal designs to future functional neuroimaging studies will provide important
information regarding progression.

Correlation between neuroimaging data and clinical outcome

There is evidence that a longer duration of untreated initial psychosis is associated
with more severe symptoms, poorer treatment response, and more impaired overall
level of functioning (reviewed by Lieberman et al., 2001b). This clinical observation
has led to the idea that untreated psychosis is neurotoxic (Loebel et al., 1992; Wyatt,
1991) and may cause progression of the underlying neuropathology. However,
studies to date have not found correlations between the duration of untreated
psychosis and neurostructural abnormalities (Fannon et al., 2000; Ho et al., 2003;
Hoff et al., 2000). These studies indicated that the structural abnormalities of the
brain observed in first-episode patients precede the onset of formal symptoms and,
381 Neurodegenerative models of schizophrenia

if they represent progression of the underlying neuropathology, this progression

has occurred premorbidly.
There has also been conflicting evidence concerning whether the progression of
neurostructural abnormalities observed after the onset of symptoms is associated
with worsening clinical symptoms or neurocognitive functioning. In first-episode
schizophrenia, greater progression of cerebral gray matter loss and ventricular
enlargement appears to correlate with worse functional outcome (Cahn et al., 2002;
Lieberman et al., 2001a); frontal cortex loss has been associated with less improve-
ment in negative symptoms but greater improvement in delusions and thought
disorder (Gur et al., 1998), while Kasai et al. (2003) reported no association between
progressive loss of superior temporal gyrus and clinical measures. Gur et al. (1998)
found cortical volume reductions associated with greater neurocognitive decline in
previously treated patients. In males with chronic schizophrenia, Mathalon et al.
(2001) reported greater clinical severity in patients with accelerated cortical gray
matter loss and ventricular enlargement, and Davis et al. (1998) found ventric-
ular enlargement in patients with poor outcome. Similarly, in childhood-onset
schizophrenia, Jacobsen et al. (1998) found progressive decline in right superior
temporal gyrus associated with worse clinical outcome.
Therefore, studies to date demonstrate some evidence of correlation between
neurostructural progression and clinical/neurocognitive function but these corre-
lations are not found in all studies, across the same brain regions, or in the same
clinical symptoms or cognitive domains. One explanation is that the brain has the
finite capacity partially to maintain its function in the face of neurostructural pro-
gression, and that a number of factors may influence whether a given individual
manifests a cognitive decline. Another important variable is the potentially con-
founding effect of antipsychotic medication, which could explain why cognitive
deterioration may be less frequent today than in Kraeplin’s era with no neuroleptic
drugs. Nevertheless, given that relatively few studies have conducted correlational
analyses between structural, clinical, and cognitive variables in carefully defined
patient subgroups, it is somewhat difficult to reconcile findings from a number of
recent imaging studies reporting structural progression with other studies reporting
overall stability of clinical symptoms and cognitive functioning. Further investiga-
tion into this important issue is clearly warranted.

Potential mechanisms underlying neurodegeneration

What process underlies the neuropathological progression of schizophrenia?
A variety of pathogenic mechanisms have been proposed. These specifically
implicate glutamate N-methyl-d-aspartate (NMDA) receptor hypofunction and
excitotoxicity (Onley and Farber, 1995), antagonism of NMDA receptors by
382 L. F. Jarskog, J. H. Gilmore, and J. A. Lieberman

N-acetylaspartylglutamate and consequent oxidative stress (Coyle and Puttfarcken,

1993), reduction in gamma-aminobutyric acid (GABA) interneuron-mediated
inhibition of pyramidal neurons in the cingulate cortex (Benes, 1995), dopamine-
mediated neurochemical sensitization (Lieberman et al., 1997) and neurotoxicity
(Wyatt, 1995). Stress has emerged as a potential factor underlying neurostructural
changes observed in affective and anxiety disorders. Because glucocorticoids may
also play a role in schizophrenia (Arango et al., 2001; Church et al., 2002; see also
Ch. 6), stress may be implicated in the progressive elements of schizophrenia, which
is generally considered a stress-sensitive illness.
As briefly discussed with the postmortem studies above, abnormal regulation
of apoptosis is emerging as a potential mechanism in the pathophysiology of
schizophrenia, and one that could be influenced by glutamatergic, GABAergic, and
dopaminergic dysfunction. Apoptotic mechanisms can underlie both cell death
and non-lethal synaptic apoptosis (Mattson et al., 1998), setting the stage for a
process that could potentially produce the type of subtle reduction in neuronal
viability seen in schizophrenia. Reduced levels of the antiapoptotic protein Bcl-2
have been shown in postmortem temporal cortex in schizophrenia (Jarskog et al.,
2000) as well as elevated ratios of the proapoptotic Bax to Bcl-2 (Jarskog et al.,
2001). These findings suggest an increased vulnerability to apoptotic activation in
schizophrenic cortex. Yet, caspase 3 (a marker of active apoptosis) was unchanged
in schizophrenia versus normal controls, which is consistent with no increase in the
rate of single-stranded DNA fragmentation in schizophrenia (Benes et al., 2003).
This suggests that apoptosis was not active in schizophrenia at the time of death,
although it does not preclude apoptotic activity at other stages of the illness such
as the prodromal phase or the first episode of psychosis.

Conclusions
Although the etiology of schizophrenia remains unknown, the current review
highlights the complexity of its underlying pathophysiology, appearing to involve
both neurodevelopmental and neurodegenerative elements. The histopathology
of schizophrenia remains without a defined phenotype; however, increasing evi-
dence is pointing to subtle neuronal atrophy and synaptodendritic reductions in
postmortem cortex, which may reflect a limited neurodegenerative process. This
conclusion is increasingly supported by neuroimaging studies that find progres-
sive neurostructural changes, especially in gray matter content and ventricle size,
and, less convincingly, studies that report limited progression of clinical symptoms
and neurocognitive function. Taken as a whole, progressive clinical, neurocog-
nitive, and neuroimaging findings appear most evident in the prodromal phase
and the first few years of psychosis, although certain subgroups of chronically ill
383 Neurodegenerative models of schizophrenia

patients with severe and treatment-refractory schizophrenia may also experience

a progression of the illness. However, the longitudinal prodromal cohort find-
ings to date must be considered preliminary given the relatively few studies and
the few subjects involved. Future studies utilizing high-resolution neuroimaging
and sophisticated neuropsychological testing techniques will undoubtedly provide
greater insight on the timing, regionality, and degree of progression in the early
stages of schizophrenia.
Given that the strongest evidence for neuroprogressive events in schizophre-
nia is emerging in the prodromal and first-episode populations, it suggests that
any neurodegenerative processes would intersect with normal neuromaturational
processes, which include myelination and synaptic pruning during adolescence
and early adulthood. Although a number of mechanistic possibilities could be
envisaged, this scenario offers the intriguing possibility that apoptotic mech-
anisms give rise to limited neurodegeneration via perturbations of develop-
mentally regulated synaptic elimination during the prodromal and first-episode
phases of schizophrenia. In order to elucidate the mechanisms that underlie the
neuropathology of schizophrenia, it will be necessary to gain a much clearer
understanding of the mechanisms that guide normal human brain development
than is currently available. Such studies will complement those that examine
high-risk premorbid and prodromal cohorts to shed more light on the com-
plex interactions of pathophysiology and normal development that underlie
schizophrenia.

Acknowledgements
This work was supported in part by the UNC Schizophrenia Research Center,
an NIMH Silvio O. Conte Center for the Neuroscience of Mental Disorders
(MH064065, JAL), NIMH grant K08 MH-01752 (LFJ), NIMH grant MH-60352
(JHG), and the Foundation of Hope of Raleigh, North Carolina (JAL).

REFERENCES

Abrahamson, D. (1983). Schizophrenia deterioration. Br J Psychiatry 143: 82–83.

Addington, J., Addington, D. (2002). Cognitive functioning in first-episode schizophrenia.
J Psychiatry Neurosci 27: 188–192.
Akbarian, S., Kim, J. J., Potkin, S. G., et al. (1995). Gene expression for glutamic acid decarboxylase
is reduced without loss of neurons in prefrontal cortex of schizophrenics. Arch Gen Psychiatry
52: 258–266.
Andreasen, N. C., O’Leary, D. S., Flaum, M., et al. (1997). Hypofrontality in schizophrenia:
distributed dysfunctional circuits in neuroleptic-naı̈ve patients. Lancet 349: 1730–1734.
384 L. F. Jarskog, J. H. Gilmore, and J. A. Lieberman

Arango, C., Kirkpatrick, B., Koenig, J. (2001). At issue: stress, hippocampal neuronal turnover,
and neuropsychiatric disorders. Schizophr Bull 27: 477–480.
Arnold, S. E., Trojanowski, J. Q., Gur, R. E., et al. (1998). Absence of neurodegeneration and
neural injury in the cerebral cortex in a sample of elderly patients with schizophrenia. Arch
Gen Psychiatry 55: 225–232.
Benes, F. M. (1995). Is there a neuroanatomic basis for schizophrenia? An old question revisited.
Neuroscientist 1: 104–115.
Benes, F. M., McSparren, J., Bird, E. D., SanGiovanni, J. P., Vincent, S. L. (1991). Deficits in
small interneurons in prefrontal and cingulate cortices of schizophrenic and schizoaffective
patients. Arch Gen Psychiatry 48: 996–1001.
Benes, F. M., Vincent, S. L., Todtenkopf, M. (2001). The density of pyramidal and nonpyramidal
neurons in anterior cingulate cortex of schizophrenic and bipolar subjects. Biol Psychiatry
50: 395–406.
Benes, F. M., Walsh, J., Bhattacharyya, S., Sheth, A., Berretta, S. (2003). DNA fragmentation
is decreased in schizophrenia but not in bipolar disorder. Arch Gen Psychiatry 60: 359–
364.
Bertolino, A., Callicott, J. H., Nawroz, S., et al. (1998). Reproducibility of proton magnetic
resonance spectroscopic imaging in patients with schizophrenia. Neuropsychopharmacology
18: 1–9.
Bilder, R. M., Lipschutz-Broch, R. G., Geisler, S. H., Mayerhoff, D. I., Lieberman, J. A. (1992).
Intellectual deficits in first-episode schizophrenia: evidence for progressive deterioration.
Schizophr Bull 18: 437–488.
Black, J. E., Klintsova, A. Y., Kodish, I. M., Greenough, W. T., Uranova, N. A. (2002). What is
missing in the reduced neuropil of schizophrenic prefrontal cortex? A combined golgi and
electron microscopy study. Biol Psychiatry 51: S54.
Cahn, W., Hulshoff Pol, H. E., Lems, E. B. T. E., et al. (2002). Brain volume changes in first-episode
schizophrenia. Arch Gen Psychiatry 59: 1002–1010.
Cannon, M., Jones, P., Huttunen, M. O., et al. (1999). School performance in Finnish children
and later development of schizophrenia. Arch Gen Psychiatry 56: 457–463.
Cannon, M., Caspi, A., Moffitt, T. E., et al. (2002). Evidence for early-childhood, pan-
developmental impairment specific to schizophreniform disorder. Results from a longi-
tudinal birth cohort. Arch Gen Psychiatry 59: 449–456.
Cecil, K. M., Lenkinski, R. E., Gur, R. E., Gur, R. C. (1999). Proton magnetic resonance spec-
troscopy in the frontal and temporal lobes of neuroleptic naive patients with schizophrenia.
Neuropsychopharmacology 20: 131–140.
Church, S. M., Cotter, D., Bramon, E., Murray, R. M. (2002). Does schizophrenia result from
developmental or degenerative processes? J Neural Transm Suppl 63: 129–147.
Cotter, D. R., Pariante, C. M., Everall, I. P. (2001). Glial cell abnormalities in major psychiatric
disorders: the evidence and implications. Brain Res Bull 55: 585–595.
Coyle, J. T., Puttfarcken, P. (1993). Oxidative stress, glutamate, and neurodegenerative disorders.
Science 262: 689–695.
Davis, K. L., Buchsbaum, M. S., Shihabuddin, L., et al. (1998). Ventricular enlargement in poor-
outcome schizophrenia. Biol Psychiatry 43: 783–793.
385 Neurodegenerative models of schizophrenia

DeLisi, L. E., Sakuma, M., Tew, W., et al. (1997). Schizophrenia as a chronic active brain process:
a study of progressive brain structural change subsequent to the onset of schizophrenia.
Psychiatry Res Neuroimaging 74: 129–140.
Done, D. J., Crow, T. J., Johnstone, E. C., Sacker, A. (1994). Childhood antecedents of schizophre-
nia and affective illness: social adjustment at ages 7 and 11. Br Med J 309: 699–703.
Eastwood, S. L., Harrison, P. J. (1995). Decreased synaptophysin in the medial temporal lobe
in schizophrenia demonstrated using immunoautoradiography. Neuroscience 69: 339–
343.
Eaton, W. W., Thara, R., Federman, B., Melton, B., Liang, K.-Y. (1995). Structure and course of
positive and negative symptoms in schizophrenia, Arch Gen Psychiatry 52: 127–134.
Fannon, D., Chitnis, X., Doku, V., et al. (2000). Features of structural brain abnormality in
first-episode psychosis. Am J Psychiatry 157: 1829–1834.
Fenton, W. S., McGlashan, T. H. (1994). Antecedents, symptom progression, and long-term
outcome of the deficit syndrome in schizophrenia. Am J Psychiatry 151: 351–356.
Friedman, J. I., Harvey, P. D., McGurk, S. R., et al. (2002). Correlates of change in functional
status of institutionalized geriatric schizophrenic patients: focus on medical comorbidity.
Am J Psychiatry 159: 1388–1394.
Fuller, R., Nopoulos, P., Arndt, S., et al. (2002). Longitudinal assessment of premorbid cognitive
functioning in patients with schizophrenia through examination of standardized scholastic
test performance. Am J Psychiatry 159: 1183–1189.
Garey, L. J., Ong, W. Y., Patel, T. S., Kanani, M., Davis, A., Mortimer, A. M., Barnes, T. R. E.,
Hirsch, S. R. (1998). Reduced dendritic spine density on cerebral cortical pyramidal neurons
in schizophrenia. J Neurol Neurosurg Psychiatry 65: 446–453.
Glantz, L. A., Lewis, D. A. (1997). Reduction of synaptophysin immunoreactivity in the prefrontal
cortex of subjects with schizophrenia: regional and diagnostic specificity. Arch Gen Psychiatry
54: 943–952.
(2000). Decreased dendritic spine density on prefrontal cortical pyramidal neurons in
schizophrenia. Arch Gen Psychiatry 57: 65–73.
Gur, R. E., Cowell, P., Turetsky, B. I., et al. (1998). A follow-up magnetic resonance imaging study
of schizophrenia. Relationship of neuroanatomical changes to clinical and neurobehavioral
measures. Arch Gen Psychiatry 55: 145–152.
Hambrecht, M., Lammertink, M., Klosterkötter, J., Matuschek, E., Pukrop, R. (2002). Subjec-
tive and objective neuropsychological abnormalities in a psychosis prodrome clinic. Br J
Psychiatry 181(Suppl. 43): S30–S37.
Hardy, J., Gwinn-Hardy, K. (1998). Genetic classification of primary neurodegenerative disease.
Science 282: 1075–1079.
Harrison, P. J. (1999). The neuropathology of schizophrenia: a critical review of the data and
their interpretation. Brain 122: 593–624.
Harrison, P. J., Eastwood, S. L. (2001). Neuropathological studies of synaptic connectivity in the
hippocampal formation in schizophrenia. Hippocampus 11: 508–519.
Harvey, P. D., Silverman, J. M., Mohs, R. C., et al. (1999). Cognitive decline in late-life schizophre-
nia: a longitudinal study of geriatric chronically hospitalized patients. Biol Psychiatry 45:
32–40.
386 L. F. Jarskog, J. H. Gilmore, and J. A. Lieberman

Heaton, R. K., Gladsjo, J. A., Palmer, B. W., et al. (2001). Stability and course of neuropsychological
deficits in schizophrenia. Am J Psychiatry 58: 24–32.
Hinsberger, A. D., Williamson, P. C., Carr, T. J., Stanley, J. A., Drost, D. J., Densmore, M., MacFabe,
G. C., Montemurro, D. G. (1997). Magnetic resonance imaging volumetric and phosphorus
31 magnetic resonance spectroscopy measurements in schizophrenia. J Psychiatry Neurosci
22: 111–117.
Ho, B.-C., Alicata, D., Ward, J., et al. (2003). Untreated initial psychosis: relation to cognitive
deficits and brain morphology in first-episode schizophrenia. Am J Psychiatry 160: 142–
148.
Hoff, A. L., Sakuma, M., Wieneke, M., et al. (1999). Longitudinal neuropsychological follow-up
study of patients with first-episode schizophrenia. Am J Psychiatry 156: 1336–1341.
Hoff, A. L., Sakuma, M., Razi, K., et al. (2000). Lack of association between duration of untreated
illness and severity of cognitive and structural brain deficits at the first episode of schizophre-
nia. Am J Psychiatry 157: 1824–1828.
Huber, G., Gross, G., Shuttler, R., Linz, M. (1980). Longitudinal studies of schizophrenic patients.
Schizophr Bull 6: 592–605.
Hulshoff Pol, H. E., Schnack, H. G., Bertens, M. G. B. C., et al. (2002). Volume changes in gray
matter in patients with schizophrenia. Am J Psychiatry 159: 244–250.
Jacobsen, L. K., Giedd, J., Castellanos, F. X., et al. (1998). Progressive reduction of temporal lobe
structures in childhood-onset schizophrenia. Am J Psychiatry 155: 678–685.
Jarskog, L. F., Gilmore, J. H., Selinger, E. S., Lieberman, J. A. (2000). Cortical Bcl-2 protein
expression and apoptotic regulation in schizophrenia. Biol Psychiatry 48: 641–650.
Jarskog, L. F., Selinger, E. S., Lieberman, J. A., Gilmore, J. H. (2001). Apoptotic regulatory proteins
afford reduced neuroprotection in schizophrenia. Schizophr Res 49: 53S.
Jones, P., Rodgers, B., Murray, R., Marmot, M. (1994). Child development risk factors for adult
schizophrenia in the British 1946 birth cohort. Lancet 344: 1398–1402.
Karson, C. N., Mrak, R. E., Schluterman, K. O., et al. (1999). Alterations in synaptic proteins
and their encoding mRNAs in prefrontal cortex in schizophrenia: a possible neurochemical
basis for “hypofrontality”. Mol Psychiatry 4: 39–45.
Kasai, K., Shenton, M. E., Salisbury, D. F., et al. (2003). Progressive decrease of left superior tem-
poral gyrus gray matter volume in patients with first-episode schizophrenia. Am J Psychiatry
160: 156–164.
Keller, A., Castellanos, F. X., Vaituzis, A. C., et al. (2003). Progressive loss of cerebellar volume in
childhood-onset schizophrenia. Am J Psychiatry 160: 128–133.
Kraeplin E. (1919). Dementia Praecox and Paraphrenia, Edinburgh, E and S Livingstone.
Kremen, W. S., Seidman, L. J., Pepple, J. R., et al. (1994). Neuropsychological risk indicators for
schizophrenia: a review of family studies. Schizophr Bull 20: 103–119.
Lewis, D. A., Lieberman, J. A. (2000). Catching up on schizophrenia: natural history and neuro-
biology. Neuron 28: 325–334.
Lieberman, J. A., Sheitman, B. B., Kinon, B. J. (1997). Neurochemical sensitization in the patho-
physiology of schizophrenia: deficits and dysfunction in neuronal regulation and plasticity.
Neuropsychopharmacology 17: 205–229.
387 Neurodegenerative models of schizophrenia

Lieberman. J. A., Chakos, M., Wu, H., et al. (2001a). Longitudinal study of brain morphology in
first episode schizophrenia. Biol Psychiatry 49: 487–499.
Lieberman, J. A., Perkins, D., Belger, A., et al. (2001b). The early stages of schizophrenia: spec-
ulations on pathogenesis, pathophysiology, and therapeutic approaches. Biol Psychiatry 50:
884–897.
Loebel, A. D., Lieberman, J. A., Alvir, J. M. J., et al. (1992). Duration of psychosis and outcome
in first-episode schizophrenia. Am J Psychiatry 149: 1183–1188.
Margolis, R. L., Chuang, D. M., Post, R. M. (1994). Programmed cell death: implications for
neuropsychiatric disorders. Biol Psychiatry 35: 946–956.
Mason, P., Harrison, G., Glazebrook, C., Medley, I., Croudance, T. (1996). The course of
schizophrenia over 13 years. A report from the International Study on Schizophrenia (IsoS)
coordinated by the World Health Organization. Br J Psychiatry 169: 580–586.
Mathalon, D. H., Sullivan, E. V., Lim, K. O., Pfefferbaum, A. (2001). Progressive brain volume
changes and the clinical course of schizophrenia in men: a longitudinal magnetic resonance
imaging study. Arch Gen Psychiatry 58: 148–157.
Mattson, M. P., Keller, J. N., Begley, J. G. (1998). Evidence for synaptic apoptosis. Exp Neurol 153:
35–48.
Mayerhoff, D. I., Loebel, A. D., Alvir, J. M. J., et al. (1994). The deficit state in first-episode
schizophrenia. Am J Psychiatry 151: 1417–1422.
McGlashan, T. H. (1988). A selective review of recent North American long-term follow-up
studies of schizophrenia. Schizophr Bull 14: 515–542.
Moritz, S., Andresen, B., Perro, C., for the PERSIST Study Group (2002). Neurocognitive perfor-
mance in first-episode and chronic schizophrenic patients. Eur Arch Psychiatry Clin Neurosci
252: 33–37.
Murray, R. M., Lewis, S. W. (1987). Is schizophrenia a neurodevelopmental disorder? Br Med J
296: 681–682.
Nair, T. R., Christensen, J. D., Kingsbury, S. J., et al. (1997). Progression of cerebroventricular
enlargement and the subtyping of schizophrenia. Psychiatry Res Neuroimaging 74: 141–
150.
Nopoulos, P., Flashman, L., Flaum, M., Arndt, S., Andreasen, N. (1994). Stability of cognitive
functioning early in the course of schizophrenia. Schizophr Res 14: 29–37.
Ona, V. O., Li, M., Vonsattel, J. P. G., et al. (1999). Inhibition of caspase-1 slows disease progression
in a mouse model of Huntington’s disease. Nature 399: 263–267.
Onley, J. W., Farber, N. B. (1995). Glutamate receptor dysfunction and schizophrenia. Arch Gen
Psychiatry 52: 998–1007.
Pakkenberg, B. (1990). Pronounced reduction of total neuron number in mediodorsal thalamic
nucleus and nucleus accumbens in schizophrenics. Arch Gen Psychiatry 47: 1023–1028.
(1993). Total nerve cell number in neocortex in chronic schizophrenics and controls estimated
using optical disectors. Biol Psychiatry 34: 768–772.
Pantelis, C., Velakoulis, D., McGorry, P. D., et al. (2002). Neuroanatomical abnormalities before
and after onset of psychosis: a cross-sectional and longitudinal MRI comparison. Lancet
Accessed 10 December 2002: http://image.thelancet.com/extras/01art9092web.pdf.
388 L. F. Jarskog, J. H. Gilmore, and J. A. Lieberman

Pettegrew, J. W., Keshavan, M. S., Panchalingam, K., et al. (1991). Alterations in brain high-
energy phosphate and membrane phospholipid metabolism in first-episode, drug-naive
schizophrenics: a pilot study of the dorsal prefrontal cortex by in vivo phosphorus 31 nuclear
magnetic resonance spectroscopy. Arch Gen Psychiatry 48: 563–568.
Pettegrew, J. W., Keshavan, M. S., Minshew, N. J. (1993). 31 P Nuclear magnetic resonance spec-
troscopy: neurodevelopment and schizophrenia. Schizophr Bull 19: 35–53.
Pfohl, B., Winokur, G. (1982). The evolution of symptoms in institutionialized hebephrenic/
catatonic schizophrenics. Br J Psychiatry 141: 567–572.
Pierri, J. N., Volk, C. L., Auh, S., Sampson, A., Lewis, D. A. (2001). Decreased somal size of deep
layer 3 pyramidal neurons in the prefrontal cortex of subjects with schizophrenia. Arch Gen
Psychiatry 58: 466–473.
Pinals, D. A., Breier, A. (1997). Schizophrenia. In Psychiatry, ed. A. Tasman, J. Kay, J. A. Lieberman.
Philadelphia, PA: W. B. Saunders, pp. 927–965.
Purohit, D. P., Perl, D. P., Haroutunian, V., et al. (1998). Alzheimer disease and related neu-
rodegenerative diseases in elderly patients with schizophrenia. Arch Gen Psychiatry 55: 205–
211.
Rajkowska, G., Selemon, L. D., Goldman-Rakic, P. S. (1998). Neuronal and glial somal size in
the prefrontal cortex: a postmortem morphometric study of schizophrenia and Huntington
disease. Arch Gen Psychiatry 55: 215–224.
Rapoport, J. L., Giedd, J., Kumra, S., et al. (1997). Childhood-onset schizophrenia: progressive
ventricular change during adolescence. Arch Gen Psychiatry 54: 897–903.
Rapoport, J. L., Giedd, J. N., Blumenthal, J., et al. (1999). Progressive cortical change during
adolescence in childhood-onset schizophrenia: a longitudinal magnetic resonance imaging
study. Arch Gen Psychiatry 56: 649–654.
Roberts, G. W., Colter, N., Lofthouse, R., et al. (1986). Gliosis in schizophrenia: a survey. Biol
Psychiatry 21: 1043–1050.
Roberts, G. W., Colter, N., Lofthouse, R., Johnstone, E. C., Crow, T. J. (1987). Is there
gliosis in schizophrenia? Investigation of the temporal lobe. Biol Psychiatry 22: 1459–
1468.
Rund, B. R. (1998). A review of the longitudinal studies of cognitive decline in schizophrenia
patients. Schizophr Bull 24: 425–435.
Saykin, A. J., Shtasel, D. L., Gur, R. E., et al. (1994). Neuropsychological deficits in neuroleptic
naive patients with first episode schizophrenia. Arch Gen Psychiatry 51: 124–131.
Schlaepfer, T. E., Harris, G. J., Tien, A. Y., et al. (1994). Decreased regional cortical gray matter
volume in schizophrenia. Am J Psychiatry 151: 842–848.
Selemon, L. D., Rajkowska, G., Goldman-Rakic, P. S. (1995). Abnormally high neuronal density
in the schizophrenic cortex: a morphometric analysis of prefrontal area 9 and occipital area
17. Arch Gen Psychiatry 52: 805–818.
(1998). Elevated neuronal density in prefrontal area 46 in brains from schizophrenic patients:
application of a three-dimensional, stereologic counting method. J Comp Neurol 392: 402–
412.
Smith, A. (1964). Mental deterioration in chronic schizophrenia. J Nerv Ment Dis 139: 479–487.
389 Neurodegenerative models of schizophrenia

Stanley, J. A., Williamson, P. C., Drost, D. J., et al. (1995). An in vivo study of the prefrontal cortex
of schizophrenic patients at different stages of illness via phosphorus magnetic resonance
spectroscopy. Arch Gen Psychiatry 52: 399–406.
Stevens, J. R. (1982). Neuropathology of schizophrenia. Arch Gen Psychiatry 39: 1131–1139.
Thompson, P. M., Sower, A. C., Perrone-Bizzozero, N. I. (1998). Altered levels of the synaptosomal
associated protein SNAP-25 in schizophrenia. Biol Psychiatry 43: 239–243.
Weinberger, D. R. (1987). Implications of normal brain development for the pathogenesis of
schizophrenia. Arch Gen Psychiatry 44: 660–669.
Weinberger, D. R., McClure, R. K. (2002). Neurotoxicity, neuroplasticity, and magnetic resonance
imaging morphometry: what is happening in the schizophrenic brain? Arch Gen Psychiatry
59: 553–558.
Weinberger, D. R., Berman, K. F., Zec, R. F. (1986). Physiologic dysfunction of dorsolateral pre-
frontal cortex in schizophrenia. I. Regional cerebral blood flow evidence. Arch Gen Psychiatry
43: 114–124.
Woods, B. T. (1998). Is schizophrenia a progressive neurodevelopmental disorder? Toward a
unitary pathogenetic mechanism. Am J Psychiatry 155: 1661–1670.
Wyatt, R. J. (1991). Neuroleptics and the natural course of schizophrenia. Schizophr Bull 17:
325–351.
(1995). Early intervention for schizophrenia: can the course of the illness be altered? Biol
Psychiatry 38: 1–3.
Young, K. A., Manaye, K. F., Liang, C. H., Hicks, P. B., German, D. C. (2000). Reduced number of
mediodorsal and anterior thalamic neurons in schizophrenia. Biol Psychiatry 47: 944–953.
Zipursky, R. B., Lim, K. O., Sullivan, E. V., Brown, B. W., Pfefferbaum, A. (1992). Widespread
cerebral gray matter volume deficits in schizophrenia. Arch Gen Psychiatry 49: 195–205.
Zorrilla, L. T. E., Heaton, R. K, McAdams, L. A., et al. (2000). Cross-sectional study of older
outpatients with schizophrenia and health comparison subjects: no differences in age-related
decline. Am J Psychiatry 157: 1324–1326.
 21

Does disordered brain development occur

across diagnostic boundaries?
Christian W. Kreipke1 , David R. Rosenberg1 , and Matcheri S. Keshavan1,2
1
Wayne State University School of Medicine, Detroit, USA
2
University of Pittsburgh School of Medicine, Pittsburgh, USA

The view that major psychiatric disorders such as schizophrenia and bipolar dis-
orders (BPD) are distinct clinical entities is being increasingly challenged (Taylor,
1992). Several authors have proposed the concept of an expanded psychiatric con-
tinuum between affective disorders and schizophrenia and several lines of evidence
support this notion. First, the cross-sectional and longitudinal co-occurrence of
affective and schizophrenic symptoms is well known, leading to that diagnostic
conundrum called schizoaffective disorder (Adler and Strakowski, 2003). Second,
both affective disorder and schizophrenia demonstrate a high degree of genetic sus-
ceptibility; some recent gene mapping studies show common susceptibility genes for
the two disorders (Berrettini, 2000). Third, BPD and schizophrenia also demon-
strate some similarities in neurotransmitter and neurophysiological dysfunction
(Moller, 2003). Finally, many newer atypical antipsychotic agents such as olanzap-
ine and quetiapine, initially approved for the treatment of schizophrenia, are also
being increasingly used in BPD, suggesting a therapeutic continuum (Moller, 2003).
Similar boundary issues exist between schizophrenia and other psychiatric dis-
orders. Although obsessive–compulsive disorder (OCD) and schizophrenia are
generally considered to be separate psychiatric disorders, comorbidities of the
two illnesses are frequent, leading to the occasional diagnosis of schizo-obsessive
disorder (Gross-Isseroff et al., 2003). As will be reviewed below, neuroimaging
studies show overlapping findings in these two illnesses, in particular abnormali-
ties in the frontostriatal circuits. Attention-deficit hyperactivity disorder (ADHD)
also frequently coexists with schizophrenia (Bellak et al., 1987) and BPD (Kim and
Miklowitz, 2002). Many children with ADHD have features of thought disorder
similar to children with schizophrenia (Caplan et al., 2002). Individuals at genetic
risk for schizophrenia (Keshavan et al., 2002a) and BPD (Chang et al., 2000) also
have an increased frequency of ADHD.
Neurodevelopment and Schizophrenia, ed. Matcheri S. Keshavan et al. Published by Cambridge University
Press.

C Cambridge University Press 2004.

390
391 Disordered brain development in psychiatric disorders

OCD

Schizo-
ADHD
phrenia

BPD/MDD

Age (years) 5 10 15 20+

Fig. 21.1. Developmental trajectories of diverse childhood/adolescent psychiatric disorders. OCD,
obsessive–compulsive disorder; ADHD, attention-deficit hyperactivity disorder; BPD, bipolar
disorder; MDD, major depressive disorder.

Not surprisingly, the pathophysiological characterization of the neurodevelop-

mental basis of these psychiatric disorders has been limited by comorbidity and
symptom commonality. Distinct clinical phenotypes of illness expression have often
been elusive. Nevertheless, technical advances in areas such as neuroimaging and
genetics are capable of enhancing our understanding of the developmental neu-
robiological underpinnings across childhood/adolescent-onset neuropsychiatric
disorders. The convergence of findings across major neuropsychiatric disorders in
neurodiagnostic and treatment research presents a unique opportunity to deepen
our understanding of the etiopathogenesis of childhood-onset neuropsychiatric dis-
orders in the context of the clinically relevant question, “Which treatments for which
child with which set of subgrouping characteristics?” Specifically, the combination
of biological and behavioral/symptomatic predictor and outcome variables enables
translational approaches to childhood-onset neuropsychiatric disorders and, in
turn, may lead to new diagnostic and treatment modalities. Such an approach may
also help to identify developmental biomarkers that are applicable across diagnostic
boundaries as well as those that are specific to a particular condition, resulting in
improved diagnosis and interventions in complex neurodevelopmental disorders.
In this chapter, we will focus on the neurodevelopmental basis of diagnostic
overlap and symptom commonality by addressing similarities and differences
in the neuroanatomical and functional neurochemical basis of three common
childhood/adolescent-onset neuropsychiatric disorders: ADHD, OCD, and mood
disorders, including major depressive disorder (MDD) and BPD. We chose these
disorders because of the clinical commonalities with schizophrenia, as discussed
above, and because of the known or presumed developmental origins of these dis-
orders. Schizophrenia, BPD, and OCD have their onset in adolescence or early
392 C. W. Kreipke, D. R. Rosenberg, and M. S. Keshavan

adulthood, and they may be preceded by symptoms similar to those in develop-

mental disorders such as ADHD (Fig. 21.1). Each of the disorders is discussed in
comparison with schizophrenia since the latter represents the main focus of this
volume. We will only briefly describe the relevant findings in schizophrenia and
will refer the reader to the appropriate chapter(s) for details. We will limit our
pathophysiological discussion to the neuroimaging findings, which are the focus of
our work, but will briefly consider common genetic and environmental etiologic
factors that may cut across these disorders.

Neuroimaging and the pathological trajectory of psychiatric disorders

Recent advances in neuroimaging techniques have afforded clinicians and
researchers the opportunity to study alterations in neurodevelopment as they relate
to psychiatric disorders. Much of the intrinsic neural circuitry abnormalities under-
lying the pathological trajectory of psychiatric disorders may be elucidated. Further,
this information may be used to explore the relationship between psychiatric disor-
ders and commonalities that may exist. The following will outline current findings
in structural abnormalities inherent to several psychiatric disorders as they relate
to schizophrenia, in order to address the complexities of diagnostic overlap (see
Table 21.1 for summary).

Attention-deficit hyperactivity disorder

ADHD is the most common childhood psychiatric disorder (Tannock, 1998). Con-
sequently, it is not surprising that it has been studied with various brain imaging
techniques to elucidate its developmental neurobiological mechanisms. A large lit-
erature has now accumulated and several brain structural alterations have been
described.

Total cerebral volume

Zametkin et al. (1990) reported decreased cerebral metabolism throughout the
brain in adult patients with ADHD. One of the most replicated findings in
schizophrenia is that total brain volume is decreased (reviewed by Sowell et al.,
2000); a study comparing 152 children and adolescents with ADHD (5–18 years of
age) with 139 age- and sex-matched controls (Castellanos et al., 2002) demonstrated
significantly smaller total cerebral volumes in children with ADHD persisting across
the age range studied (Fig. 21.2). Total cerebral volume was comparably decreased
in medicated and non-medicated patients with ADHD. Comparable reductions
in total cerebral volume were observed in males and females. Decrease in cere-
bral volume was most marked in the white matter of unmedicated patients with
ADHD. Specifically, significantly smaller total white matter volumes were observed
393 Disordered brain development in psychiatric disorders

Table 21.1. Brief review of neuroanatomical abnormalities across

several psychiatric disorders

Schizophrenia ADHD OCD MDD BPD

Total brain

Corpus callosum

Prefrontal cortex

Hippocampus

Amygdala

Basal ganglia

Thalamus

Cingulate

ADHD, attention-deficit hyperactivity disorder; OCD, obsessive–

computive disorder; MDD, major depressive disorder; BPD, bipolar
disorder.

in unmedicated children with ADHD, compared with both medicated children with
ADHD and controls. Medicated patients with ADHD did not differ from controls
in total white matter volumes. This underscores the importance of controlling for
potentially confounding factors such as medication status and age. Reductions in
total brain volume is similar to that observed in schizophrenia (see Ch. 19).

Frontostriatal circuitry
As the executive, “decision-making” center of the brain, the prefrontal cortex and
its striatal target fields have been considered key regions of interest in ADHD.
Similar to reports of decreased prefrontal lobe activity (Andreasen et al., 1992)
and decreased striatal volumes (Keshavan et al., 1998; Shihabuddin et al., 1998)
in schizophrenia, converging lines of evidence support reduced prefrontal and
striatal volumes and hypoactivation of the prefrontal cortex and striatum in
patients with ADHD (Aylward et al., 1996; Bush et al., 1999; Casey et al., 1997;
Castellanos et al., 1994, 2001, 2002; Filipek et al., 1997; Hynd et al., 1990; Rubia
et al., 1999; Teicher et al., 2000). Reversed asymmetry of the caudate nucleus has
also been found in pediatric patients with ADHD (Hynd et al., 1993). Reduced cau-
date volumes appear to be most pronounced in younger patients with ADHD, and
394 C. W. Kreipke, D. R. Rosenberg, and M. S. Keshavan

Controls ADHD
Mean Mean
95% Confidence Interval 95% Confidence Interval

Males Females
1150
Volume (ml)

1050

950

850
5 10 15 20 5 10 15 20
Age (years) Age (years)

Fig. 21.2. Predicted unadjusted longitudinal growth curves for total cerebral volumes for patients with
attention-deficit hyperactivity disorder (ADHD) and controls. (Taken with permission from
Castellanos, F. X., Lee, P. P., Sharp, W., et al., 2002. Developmental trajectories of brain vol-
ume abnormalities in children and adolescents with attention-deficit/hyperactivity disorder.
JAMA 288: 1740–1748.)

the difference from controls disappears in adolescence as caudate volumes decrease

with age (Castellanos et al., 2002; Durston et al., 2001; Fig. 21.3).
Proton magnetic resonance spectroscopy (MRS) investigations allow direct, in
vivo, and non-invasive measurement of brain chemistry, including the putative
neuronal marker N-acetyl aspartate (NAA). Consistent with volumetric reduction
and hypoactivation in the striatum in patients with ADHD, decreased NAA levels
and increased levels of choline, a critical membrane component believed to play
a critical role in brain signal transduction, have been found in the left and right
striatum in pediatric patients with ADHD (Birken and Oldendorf, 1989; Hesslinger
et al., 2001). This is consistent with reports of reductions in NAA and related
gray matter volumes in frontal cortex and anterior cingulate cortex in childhood-
onset schizophrenia and in unaffected offspring at increased risk for developing
schizophrenia (Keshavan et al., 1997; Sowell et al., 2000). MacMaster et al. (2003)
found increased frontostriatal glutamatergic levels in medication-free children with
ADHD, aged 7–16 years, compared with case-matched controls. Reductions in NAA
levels may result from excess glutamatergic excitotoxicity (Bartha et al., 1998).
Vaidya et al. (1998) used functional magnetic resonance imaging (fMRI) to exam-
ine the effect of pharmacologic challenge with methylphenidate, a first-line treat-
ment for ADHD (Rosenberg et al., 2002a), in children with ADHD and healthy
controls. Specifically, methylphenidate challenge resulted in distinct alterations in
frontostriatal activity in the children with ADHD compared with the controls.
Differences in methylphenidate-related activation were hypothesized to reflect
395 Disordered brain development in psychiatric disorders

Controls ADHD
Mean Mean
95% Confidence Interval 95% Confidence Interval

Total Caudate Volume Total Cerebellar Volume

11 135
Volume (ml)

10 125

9 115
5 10 15 20 5 10 15 20
Age (years) Age (years)

Fig. 21.3. Predicted unadjusted longitudinal growth curves for total caudate and cerebellar volume for
patients with attention-deficit hyperactivity disorder (ADHD) and controls. (Taken with per-
mission from Castellanos, F. X., Lee, P. P., Sharp, W., et al., 2002. Developmental trajectories of
brain volume abnormalities in children and adolescents with attention-deficit/hyperactivity
disorder. JAMA 288: 1740–1748.)

alterations in frontostriatal dopamine systems in ADHD. A positron emission

tomography investigation in children with ADHD that found high midbrain accu-
mulation of [18 F]-labeled dihydroxyphenylalanine, a putative marker of dopamine
synthesis, provides additional support for dopamine dysfunction in ADHD (Ernst
et al., 1999). A shift in dopaminergic dysfunction in ADHD from the midbrain
to frontal cortex during adolescence has been proposed (Ernst et al., 1998, 1999;
Schweitzer et al., 2000).
Studies examining the corpus callosum have generally shown reductions in its
size in ADHD (Castellanos, 1994; Hill et al., 2003), similar to the corpus callosal
reductions observed in schizophrenia (Shenton et al., 2001). Increased attentional
problems have been associated with greater reductions in corpus callosal size (Kayl
et al., 2000).

Obsessive–compulsive disorder
Neurobiological studies in various laboratories using diverse techniques have con-
sistently found alterations in ventral prefrontostriatothalamic function in OCD
associated with symptom severity and response to treatment (Rosenberg et al., 2001;
Fig. 21.4). Recent investigation in treatment-naive pediatric patients with OCD has
suggested a neural network dysplasia, with reduced striatal volumes and increased
anterior cingulate and thalamic volumes. Rosenberg and colleagues (2000a, 2001)
have proposed that a reversible thalamocorticostriatal dysfunction in glutamatergic
396 C. W. Kreipke, D. R. Rosenberg, and M. S. Keshavan

Fig. 21.4. Brain regions implicated in the pathophysiology of obsessive–compulsive disorder.

(Reprinted with permission from Rosenberg, D. R., MacMillan, S. N., Moore, G. J. Brain
anatomy and chemistry may predict treatment response in paediatric obsessive–compulsive
disorder. Int J Neuropsychopharmacol 4: 179, 2001.) (See plate section for color version.)

neurotransmission may be critically involved in the pathogenesis and treatment

response of pediatric patients with OCD (Fig. 21.5).

Total brain volume

In contrast to findings of reduced total cerebral volume in patients with ADHD
and in those with schizophrenia, repeated investigation has revealed no signifi-
cant differences in intracranial volume between treatment-naive pediatric patients
with OCD and normal controls (Gilbert et al., 2000; Rosenberg and Keshavan,
1998; Rosenberg et al., 1997a). No significant changes in intracranial volume have
been observed with treatment with a selective serotonin reuptake inhibitor or with
cognitive behavioral therapy in pediatric patients with OCD (Gilbert et al., 2000;
Rosenberg et al., 2000b).

Prefrontal cortex
Volumetric MRI investigation in treatment-naive pediatric patients with OCD
compared with age- and sex-matched controls has demonstrated increased anterior
397 Disordered brain development in psychiatric disorders

5-HT = Serotonin
GABA = Gamma-aminobutyrate
Glutamate
DA = dopamine

Globus Pallidus
Ventral Prefrontal
Cortex Ventral
Striatum

Thalamus

Temporal Lobe
Cortex

Substantia Nigra/
Ventral Tegmentum
Raphe
Nucleus

Fig. 21.5. Schematic diagram showing selected aspects of corticostriatal connections in the neurode-
velopment of obsessive–compulsive disorder. Arrows show the actions of different neuro-
transmitters. (Adapted by permission from D. R. Rosenberg and M. S. Keshavan. Toward
a neurodevelopmental model of obsessive–compulsive disorder. Biol Psychiatry, 43: 623,
1998. Permission to reprint adaptation also from Farchione, T. R., MacMillan, S. N., Rosen-
berg, D. R. 2003. Obsessive–compulsive disorder as a fronto–striatal–thalamic dysfunction.
In Mental and Behavioral Dysfunction in Movement Disorders. Ed. Bedard, M. A., Agid, Y.,
Chouinard, S., et al. Totowa, NJ: Humana Press.)

cingulate volume in the former (Rosenberg and Keshavan, 1998; Fig. 21.6). These
observations contrast with cingulate volume reductions seen in schizophrenia
(Noga et al., 1995; Takahashi et al., 2003). The age-related increase in anterior
cingulate volume in healthy children appears to be absent in children with OCD.
Increased anterior cingulate volumes were positively correlated with increased OCD
symptom severity and inversely correlated with reduced striatal volumes. No sig-
nificant differences were observed between children with OCD and controls in
dorsolateral prefrontal cortex, posterior cingulate, whole temporal lobe, amygdala,
hippocampus, or superior temporal gyrus volumes. By contrast, volumetric reduc-
tions have been seen in these structures in schizophrenia (reviewed by Shenton
et al., 2001). It should be noted that the increases in anterior cingulate volumes seen
in OCD have not been reported in pediatric patients with ADHD. Finally, reduced
NAA levels suggestive of neuronal dysfunction have been reported in the anterior
398 C. W. Kreipke, D. R. Rosenberg, and M. S. Keshavan

Anterior Cingulate Volume (ml) 7

2
6 8 10 12 14 16 18 20
Age (years)
Controls
OCD Patients
Fig. 21.6. Anterior cingulate volume as a function of age for pediatric patients with obsessive–
compulsive disorder and normal controls. Opposite neurodevelopmental trajectories in the
patients compared with the healthy subjects may reflect an absence of early neuronal prun-
ing in children with OCD. (Reprinted with permission from Rosenberg, D. R. Keshavan,
M. S. Toward a neurodevelopmental model of obsessive–compulsive disorder. Biol Psychi-
atry 43: 623, 1998.)

cingulate of pediatric patients with OCD (Ebert et al., 1997). Reduced anter-
ior cingulate glutamatergic concentrations have also been reported in pediatric
patients with OCD, which increase to levels not significantly different from controls
after effective treatment with the selective serotonin reuptake inhibitor paroxetine
(Rosenberg et al., 2002b). We have suggested that increased caudate glutamater-
gic concentrations coupled with reduced anterior cingulate glutamatergic concen-
trations in treatment-naive pediatric patients with OCD suggest a tonic–phasic
dysregulation of the glutamatergic system involving the corticostriatal circuits in
OCD (Rosenberg and Keshavan, 1998). Tonic cortical glutamatergic activity has an
inhibitory influence on phasic stress-related glutamate release from the striatum,
particularly the ventral striatum. Therefore, reduced tonic glutamatergic activity in
the anterior cingulate might predispose to phasic glutamatergic overactivity in the
striatum. The fact that caudate glutamatergic increase is reversible with paroxetine
treatment further supports this idea.
399 Disordered brain development in psychiatric disorders

Thalamus
The thalamus serves as the final subcortical relay station to various cortical
structures, including the frontal cortex. There is evidence that thalamic volumes
are reduced in first-episode neuroleptic drug-naive patients with schizophrenia
(Gilbert et al., 2001), whereas thalamic volumes are significantly increased in
treatment-naive pediatric patients with OCD compared with controls. After 12
weeks of monodrug therapy with paroxetine, thalamic volumes decreased signif-
icantly to levels not significantly different from controls. Reduction in thalamic
volumes in pediatric patients with OCD was positively correlated with reduction
in OCD symptom severity, with higher pretreatment thalamic volumes predicting
enhanced response to paroxetine. In contrast, thalamic volumes did not change
significantly in treatment-naive pediatric patients with OCD treated for 12 weeks
with cognitive behavioral therapy (Rosenberg et al., 2000b). Consequently, reduc-
tion in thalamic volumes may be specific to treatment with a selective serotonin
reuptake inhibitor rather than linked to a spontaneous resolution of symptoms or
some more generalized treatment response.
Fitzgerald et al. (2000) used proton MRS to evaluate medial and lateral subregions
of the thalamus since medial thalamic regions have been particularly implicated in
OCD (Modell et al., 1989). Reduced levels of NAA were found in medial, but not
lateral, thalamus in pediatric patients with OCD. Increased thalamic volumes in
pediatric patients with OCD were associated with lower levels of NAA, suggestive
of neuronal dysfunction.

Basal ganglia
Similar to findings in patients with ADHD and schizophrenia, reduced striatal vol-
umes have been observed in child and adolescent patients with OCD compared with
controls, using both quantitative computed tomography (Luxenberg et al., 1988)
and volumetric MRI (Rosenberg et al., 1997b). Reductions in striatal volumes
were associated with increased OCD symptom severity, but not illness duration
(Rosenberg et al., 1997b). Consistent with reduced striatal volumes, increased ven-
tricle to brain ratios and increased third ventricular volumes have been observed in
children and adolescents with OCD compared with healthy controls (Behar et al.,
1984; Rosenberg et al., 1997b).
Also, similar to reports in ADHD, reductions in NAA have been reported in the
striatum in patients with OCD compared with controls (Bartha et al., 1998; Ebert
et al., 1997). Bartha et al. (1998) proposed that this reduction in neuronal viabil-
ity may be mediated by glutamatergic excitotoxicity. Therefore, Rosenberg et al.
(2000a) measured caudate glutamatergic concentrations in treatment-naive pedi-
atric patients with OCD before and after monodrug therapy with paroxetine and
compared these with case-matched pediatric controls. The occipital cortex was also
400 C. W. Kreipke, D. R. Rosenberg, and M. S. Keshavan

studied as a control region not implicated in the pathogenesis of OCD. Localized

increased glutamatergic concentrations were observed in treatment-naive pedi-
atric patients with OCD compared with controls in the caudate nucleus but not in
the occipital cortex. Remarkably, after effective treatment with paroxetine, caudate
glutamatergic concentrations decreased to levels not significantly different from
those observed in controls. Reduction in caudate glutamatergic concentrations was
positively correlated with reduction in OCD symptom severity, as measured by the
Children’s Yale Brown Obsessive Compulsive Scale. Increased caudate glutamatergic
levels before treatment predicted better response to paroxetine, whereas decreased
glutamatergic levels in the caudate pretreatment predicted a poorer response. There
were no significant changes in occipital glutamatergic concentrations after parox-
etine treatment. Reduction in caudate glutamatergic concentrations also appears
to persist even after medication discontinuation if remission of OCD symptoms is
maintained (Bolton et al., 2001).
It should be noted that no significant changes in caudate glutamatergic con-
centrations were observed in 21 psychotropic drug-naive pediatric patients with
OCD treated for 12 weeks with cognitive behavioral therapy (Benazon et al., 2001).
Therefore, the reduction in caudate glutamatergic concentrations may be specific to
treatment with a selective serotonin reuptake inhibitor rather than reflecting a more
generalized treatment response or spontaneous resolution of symptoms. It should
be noted that increased striatal glutamate has been reported in medication-free
pediatric patients with both OCD and ADHD (MacMaster et al., 2003). This find-
ing has to be viewed with caution, however, because the glutamine and glutamate
resonances are not clearly resolved in the in vivo spectra.
The corpus callosum appears to be bigger in OCD than in normal controls
(Rosenberg et al., 1997a), which contrasts with the reductions in this structure seen
in schizophrenia and ADHD. The mean signal intensity was also altered in patients
with OCD, suggesting possibly an increase in myelination. However, not all studies
show corpus callosum changes in OCD (Kellner et al., 1991).

Major depressive disorder and bipolar disorder

Total brain volume
Steingard et al. (2002) observed smaller total brain volumes in adolescent patients
with MDD compared with healthy adolescents. This is consistent with an MRI
investigation that found reduced intracranial volumes in medication-free adult
patients with MDD compared with controls (Brambilla et al., 2001). No significant
differences were observed in intracranial volume between patients with BPD and
controls. Therefore, reduced total brain volumes appear to cut across diagnostic
boundaries of ADHD and MDD but not OCD and BPD. A review of the litera-
ture by Soares and Mann (1997) suggested that left hemispheric lesions are more
401 Disordered brain development in psychiatric disorders

commonly associated with depression, whereas right-sided lesions appear to be

more commonly associated with mania.

Frontal cortex
Reduced frontal lobe volumes have been consistently reported in adult and pedi-
atric patients with MDD (Steingard et al., 2002). These authors also noted reduced
frontal white matter volumes and significantly larger frontal gray matter volumes
in adolescent patients with MDD compared with healthy adolescents. It appears
that patients with ADHD or MDD have reduced frontal white matter volumes com-
pared with controls (Castellannos et al., 2002), but the former also have reduced gray
matter volumes. Striking left, but not right, volumetric reductions in the subgenual
region of the prefrontal cortex, which are associated with reduced regional cerebral
blood flow in this region, have been observed in adult patients with familial MDD
or BPD (at least one first-degree relative with MDD or BPD) compared with that
in patients with non-familial MDD or BPD (no evident family history) (Drevets,
2000). Botteron et al. (2002) also reported comparable left but not right volumetric
reductions of the subgenual region of the prefrontal cortex in 30 young women aged
18–23 years with early-onset MDD and 18 women aged 24–52 years with recur-
rent MDD compared with matched controls. It should be noted that no significant
differences from controls in left or right subgenual prefrontal cortex in patients
with MDD or BPD were seen in other studies (Brambilla et al., 2002; Bremner
et al., 2002). A recent investigation by Nolan et al. (2002) in 22 psychotropic drug-
naive pediatric patients with MDD compared with 22 case-matched controls iden-
tified distinct neuroanatomic differences in left but not right prefrontal cortical vol-
umes in familial but not non-familial MDD. Reduced left prefrontal cortical volume
was correlated with increased depressive symptom severity and longer illness dura-
tion in familial but not non-familial MDD. Left-sided prefrontal cortical volume
reduction in patients with familial MDD may represent a progressive rather than a
static developmental effect and familial/genetic factors may be involved in patho-
genesis given. However, since all investigations to date have been cross-sectional
rather than longitudinal studies, definitive conclusions are premature.

Medial temporal cortex

MacMillan and colleagues (2003) have noted that volumetric abnormalities in the
amygdala in pediatric patients with MDD may be more related to severity of anx-
iety than the severity of depression. Specifically, they observed increased bilateral
amygdala hippocampal volumes in 23 treatment-naive pediatric patients with MDD
compared with case-matched controls. Robust positive correlations were observed
between increased amygdala–hippocampal volumes and increased severity of anx-
iety, but not with severity of depression or duration of illness. Amygdala volumes
402 C. W. Kreipke, D. R. Rosenberg, and M. S. Keshavan

were larger in pediatric patients with MDD than in controls, while hippocampal
volumes tended to be smaller in the MDD patients. Pine et al. (2001) has also found
that excess fear in adolescence is a robust predictor for development of MDD in
early adulthood. Moreover, in a recent fMRI study conducted by Thomas et al.
(2001), measuring amygdala response to fearful faces in children with generalized
anxiety disorder and panic disorder, children with MDD, and healthy controls, the
magnitude of signal change in the amygdala was most associated with the severity
of the child’s anxiety in patients with MDD and generalized anxiety disorder.
When considering whether disordered brain development cuts across diagnostic
boundaries, it is important to point out that increased amygdala volume is one of
the more consistent findings observed in volumetric neuroimaging studies of adult
patients with BPD (Altshuler et al., 2000; Frodl et al., 2002; Strakowski et al., 1999).
Increased amygdala activation in response to fearful facial affect has also been
observed in fMRI studies conducted in patients with BPD (Thomas et al., 2001).
Pediatric MDD carries an increased risk for becoming BPD in adulthood (Geller
et al., 2001; Harrington et al., 1990; Weissman et al., 1999), and increased amygdala
and reduced hippocampal volumes are seen in pediatric patients with MDD and in
adults with first-episode MDD (Frodl et al., 2002; MacMillan et al., 2003). These
findings contrast with those reporting decreased amygdala volumes in patients
with schizophrenia (reviewed by Shenton et al., 2001), and in those at risk for this
disorder (Keshavan et al., 2002b; Ch. 19). Longitudinal MRI studies are necessary
to determine whether anatomic and functional alterations in the amygdala may
serve as potential biomarkers of risk for development of BPD.
Only one available study has examined the corpus callosum in unipolar depres-
sion and this found an increase in size (Wu et al., 1993). The corpus callosum also
appears to be larger in BPD (Hauser et al., 1989), and the striatum appears larger in
first-episode BPD (Strakowski et al., 2002). However, basal ganglia volumes appear
to be reduced in depression (reviewed by Soares and Mann, 1997).

Other factors that may occur across psychiatric disorders

While advances in neuroimaging have afforded a new understanding of the devel-
opmental abnormalities associated with psychiatric disorders, genetics and other
etiological factors, such as obstetric complications (OCs), contribute to our under-
standing of psychiatric disorders and further help to identify the inherent diagnostic
overlap associated with psychiatric phenomena.

Genetic component
There is a substantial genetic component to both schizophrenia and ADHD.
Early family, twin, and adoptee studies have shown a strong likelihood of genetic
403 Disordered brain development in psychiatric disorders

susceptibility in both ADHD and schizophrenia. Combined reports suggest that the
mean concordance for schizophrenia within monozygotic twins is approximately
46% and within dizygotic pairs is approximately 14% (Gottesman, 1991; Moldin
and Gottesman, 1997). The rate of ADHD in first-degree relatives of children diag-
nosed as having ADHD ranges from 16% to 25% (Biederman et al., 1992, 1991).
Molecular genetic findings suggest that alterations in the dopaminergic system
may underlie both ADHD and schizophrenia. Several groups have reported alter-
ations in the gene encoding the dopamine D4 receptor in children diagnosed with
ADHD (Benjamin et al., 1996; Ebstein et al., 1996; LaHoste et al., 1996). Cook
et al. (1995) and Gill et al. (1997) have both reported an association between
alterations in the gene encoding the dopamine transporter and ADHD. Alter-
ations in the gene encoding catechol-O-methyltransferase (COMT), an enzyme that
degrades dopamine, have been associated with schizophrenia (reviewed by Sawa and
Snyder, 2002) and with BPD (Badner and Gershon, 2002). In animals with a tar-
geted deletion of the COMT gene, decreased levels of dopamine are seen only in
the prefrontal cortex (Gogos et al., 1998). Therefore, changes affecting dopamine
metabolism could lead to a decrease in prefrontal dopamine levels, and such deficits
may underlie the executive function deficits seen in both schizophrenia and BPD.
A heritability component has been suggested in both OCD and BPD. However,
by contrast with schizophrenia where replicable findings are beginning to emerge
(O’Donovan et al., 2003; Ch. 1), few replicated genetic susceptibility loci have been
identified for OCD and BPD. One finding suggests linkage of BPD to a location on
chromosome 18 (Berrettini et al., 1994; Stine et al., 1995). However, the large-scale
National Institute of Mental Health Genetics Bipolar Group study (1997) did not
find an association between BPD and chromosome 18.

Obstetric complications
While accumulating evidence points to the role of OCs in vulnerability for
schizophrenia (Ch. 11), the diagnostic specificity and the etiological significance of
this association remain unclear. Other severe psychiatric disorders, such as ADHD
and psychotic affective disorder, have been linked with OCs; low birth weight,
neonatal complications, and excessive bleeding have been associated with ADHD
(Milberger et al., 1997; Sprich-Buckminster et al., 1993). OCs have been observed
sporadically in BPD (reviewed by Buka and Fan, 1999) but could not be reliably
replicated. Complications within these studies include small sample sizes and data
obtained primarily in retrospect from maternal recall. A recent comprehensive
study showed no excess in OCs in mania (Browne et al., 2000).
Overall, OCs may increase the vulnerability for a range of severe mental disorders
in a relatively non-specific way and may interact with genetic liability and later
environmental risk factors (Verdoux and Sutter, 2002).
404 C. W. Kreipke, D. R. Rosenberg, and M. S. Keshavan

Conclusions
The studies described in this chapter indicate the impressive commonalities in neu-
rodevelopmentally mediated pathophysiological alterations across several major
psychiatric disorders. The similarities in pathophysiology are consistent with
the observations of frequent comorbidity and diagnostic overlap. The similar-
ity between ADHD and schizophrenia in abnormalities of the overall brain vol-
ume, corpus callosum size, and frontostriatal circuits is striking and is consis-
tent with the increasing understanding of attentional impairment as a core feature
of schizophrenia (Ch. 4). Additionally, observations of attentional problems are
among the most robust premorbid predictors of later schizophrenia (Ch. 23). There
is good evidence that ADHD also predicts the later emergence of BPD (Kim and
Miklowitz, 2002). Interestingly, an association has also been proposed between
ADHD and childhood OCD (Geller et al., 2002). Therefore, attentional impair-
ments in childhood may serve as the bridge connecting the commonalities between
diverse psychiatric presentations in adulthood, such as OCD, BPD, and schizophre-
nia (Fig. 21.1). Considering the earlier onset of ADHD compared with the other
disorders, it may even be suggested that attentional impairments, perhaps related
to distributed cortical gray and white matter changes, especially in the prefrontal
cortex, may represent a non-specific precursor for the later development psychiatric
disorders, at least in some patients.
The comparisons between overlapping disorders also reveal some intriguing
differences. For example, while schizophrenia and OCD share similar reductions
in striatal volumes, OCD shows increased cingulate volumes while schizophre-
nia shows decreased volumes. Similarly, patients with BPD and possibly those
with OCD show larger amygdala volumes, whereas smaller amygdala volumes are
observed in schizophrenia. Patients with BPD may also show larger basal gan-
glia, unlike the schizophrenia patients. This suggests that different developmen-
tal trajectories may emerge during late childhood and adolescence from com-
mon neurodevelopmental precursors in childhood. These developmental trajec-
tories may be related to differential pathoplastic responses of the subcortical
brain structures, such as the cingulate and amygdala, and may mediate the emer-
gence of distinct symptomatic presentations. It is useful to recall here that the
amygdala and the cingulate are critically involved in affect regulation and con-
flict monitoring, respectively (Carter et al., 1999; Ledoux, 2003). Consequently,
the amygdala enlargements in BPD might represent a pathological hyperplasia
in that region related to repeated activation in the context of highly valenced
emotional states, such as the manic and depressive episodes. Likewise, the cin-
gulate enlargement in OCD might reflect a hyperresponsive error-monitoring
system. By comparison, failure of optimum functioning of the cingulate and
405 Disordered brain development in psychiatric disorders

amygdala may underlie the deficits in self-monitoring that lead to disorga-

nized thinking and diminished affective responsivity, leading to blunted affect in
schizophrenia.
These predictions can best be examined by an integrated approach to longitudi-
nal follow-up studies of children at risk for a variety of major psychiatric disorders
using the non-invasive neuroimaging techniques that are available. Such studies
also allow us to clarify the common and distinct neurodevelopmental alterations
across diverse disorders and help us to unravel the shared and distinct vulnerability
factors of genetic or environmental origin. Additionally, mapping the develop-
mental trajectories of adult psychiatric disorders can help in designing prevention
strategies. Several longitudinal–epidemiological studies suggest that most adult psy-
chiatric problems can be traced back to childhood disorders, either of the same type
(i.e. homotypic continuity) or a different one (i.e. sequential comorbidity) (Kim-
Cohen et al., 2003). Improved knowledge of the common developmental basis of
psychiatric disorders will clearly promote a life-course developmental approach to
psychopathology and eventually improve treatment and prevention.

REFERENCES

Adler, C. M., Strakowski, S. M. (2003). Boundaries of schizophrenia. Psychiatr Clin North Am 26:
1–23.
Altshuler, L. L., Bartzokis, G., Grieder, T., et al. (2000). An MRI study of temporal lobe structures
in men with bipolar disorder or schizophrenia. Biol Psychiatry 48: 147–162.
Andreasen, N. C., Rezai, K., Alliger, R., et al. (1992). Hypofrontality in neuroleptic-naive patients
and in patients with chronic schizophrenia. Assessment with xenon 133 single-photon
emission computed tomography and the Tower of London. Arch Gen Psychiatry 49: 943–
958.
Aylward, E. H., Reiss, A. L., Reader, M. J., et al. (1996). Basal ganglia volumes in children with
attention-deficit hyperactivity disorder. J Child Neurol 11: 112–115.
Badner, J. A., Gershon, E. S. (2002). Meta-analysis of whole-genome scans of bipolar disorder
and schizophrenia. Mol Psychiatry 7: 405–411.
Bartha, R., Stein, M. B., Williamson, P. C., et al. (1998). A short echo 1 H spectroscopy and
volumetric MRI study of the corpus striatum in patients with obsessive–compulsive disorder
and comparison subjects. Am J Psychiatry 155: 1584–1591.
Behar, D., Rapoport, J. L., Berg, C. J., et al. (1984). Computerized tomography and neuropsy-
chological test measures in adolescents with obsessive–compulsive disorder. Am J Psychiatry
141: 363–369.
Bellak, L., Kay, S. R., Opler, L. A. (1987). Attention deficit disorder psychosis as a diagnostic
category. Psychiatr Dev 5: 239–263.
406 C. W. Kreipke, D. R. Rosenberg, and M. S. Keshavan

Benazon, N. R., Ager, J., Rosenberg, D. R. (2001). Cognitive behavior therapy in treatment-naive
children and adolescents with obsessive–compulsive disorder: an open trial. Behav Res Ther
40: 529–539.
Benjamin, J., Li, L., Patterson, C., et al. (1996). Population and familial association between the
D4 dopamine receptor gene and measures of novelty seeking. Nat Genet 12: 81–84.
Berrettini, W. (2000). Review of bipolar molecular linkage and association studies. Curr Psychiatry
Rep 4: 124–129.
Berrettini, W. H., Ferraro, T. N., Alexander, R. C., Buchberg, A. M., Vogel, W. H. (1994). Quanti-
tative trait loci mapping of three loci controlling morphine preference using inbred mouse
strains. Nat Genet 7: 54–58.
Biederman, J., Faraone, S. V., Keenan, V., Tsuang, M. T. (1991). Evidence of familial association
between attention deficit disorders and major affective disorder. Arch Gen Psychiatry 78:
633–672.
Biederman, J., Faraone, S. V., Keenan, K., et al. (1992). Further evidence for familial-genetic risk
factors in ADHD: patterns of comorbidity in probands and relatives in psychiatrically and
pediatrically referred samples. Arch Gen Psychiatry 49: 728–738.
Birken, D. L., Oldendorf, W. H. (1989). N-Acetyl-L-aspartic acid: a literature review of a com-
pound prominent in 1 H-NMR spectroscopic studies of brain. Neurosci Biobehav Rev 13:
23–31.
Bolton, J., Moore, G. J., MacMillan, S., Stewart, C. M., Rosenberg, D. R. (2001). Case study:
caudate glutamatergic changes with paroxetine persist after medication discontinuation in
pediatric OCD. J Am Acad Child Adolesc Psychiatry 40: 903–906.
Botteron, K. N., Raichle, M. E., Drevets, W. C., Heath, A. C., Todd, R. D. (2002). Volumetric
reduction in left subgenual prefrontal cortex in early onset depression. Biol Psychiatry 51:
342–344.
Brambilla, P., Harenski, K., Nicoletti, M. A., et al. (2001). Anatomical MRI study of basal ganglia
in bipolar disorder patients. Psychiatry Res 106: 65–80.
Brambilla, P., Nicoletti, M. A., Harenski, K., et al. (2002). Anatomical MRI study of subgenual
prefrontal cortex in bipolar and unipolar subjects. Neuropsychopharmacology 27: 792–799.
Bremner, J. D., Vythilingam, M., Vermetten, E., et al. (2002). Reduced volume of orbitofrontal
cortex in major depression. Biol Psychiatry 51: 273–279.
Browne, R., Byrne, M., Mulryan, N., et al. (2000). Labor and delivery complications at birth and
later mania. Br J Psychiatry 176: 369–372.
Buka, S., Fan, A. (1999). Association of prenatal and perinatal complications with subsequent
bipolar disorder and schizophrenia. Schizophr Res 39: 113–119.
Bush, G., Frazier, J. A., Rauch, S. L., et al. (1999). Anterior cingulate cortex dysfunction in
attention-deficit/hyperactivity disorder revealed by fMRI and the Counting Stroop. Biol
Psychiatry 45: 1542–1552.
Caplan, R., Guthrie, D., Tang, B., Nuechterlein, K. H., Asarnow, R. E. (2002). Thought disorder
in attention-deficit hyperactivity disorder. Int Rev Neurobiol 49: 269–284.
Carter, C. S., Botvinick, M. M., Cohen, J. D. (1999). The contribution of the anterior cingulate
cortex to executive process in cognition. Rev Neurosci 10: 49–57.
407 Disordered brain development in psychiatric disorders

Casey, B. J., Castellanos, F. X., Giedd, J. N., et al. (1997). Implication of right frontostriatal circuitry
in response inhibition and attention-deficit/hyperactivity disorder. J Am Acad Child Adolesc
Psychiatry 36: 374–383.
Castellanos, F. X., Giedd, J. N., Eckburg, P., et al. (1994). Quantitative morphology of the caudate
nucleus in attention deficit hyperactivity disorder. Am J Psychiatry 151: 1791–1796.
Castellanos, F. X., Giedd, J. N., Berquin, P. C., et al. (2001). Quantitative brain magnetic resonance
imaging in girls with attention-deficit/hyperactivity disorder. Arch Gen Psychiatry 58: 289–
295.
Castellanos, F. X., Lee, P. P., Sharp, W., et al. (2002). Developmental trajectories of brain volume
abnormalities in children and adolescents with attention-deficit/hyperactivity disorder. J
Am Mod Assoc 288: 1740–1748.
Chang, K. D., Steiner, H., Ketter, T. A. (2000). Psychiatric phenomenology of child and adolescent
bipolar offspring. J Am Acad Child Adolesc Psychiatry 39: 453–460.
Cook, E. H., Sterin, M. A., Krasowski, M. D., et al. (1995). Association of attention-deficit disorder
and the dopamine transporter gene. Am J Hum Genet 56: 993–998.
Drevets, W. C. (2000). Neuroimaging studies of mood disorders. Biol Psychiatry 48: 813–829.
Durston, S., Hulshoff Pol, H. E., Casey, B. J., et al. (2001). Anatomical MRI of the developing
human brain: what have we learned? J Am Acad Child Adolesc Psychiatry 40: 1012–1020.
Ebert, D., Speck, O., Konig, A., et al. (1997). 1 H-magnetic resonance spectroscopy in obsessive–
compulsive disorder: evidence for neuronal loss in the cingulate gyrus and the right striatum.
Psychiatry Res 74: 173–176.
Ebstein, R. P., Novick, O., Umansky, R., et al. (1996). Dopamine D4 receptor (D4DR) exon III
polymorphism associated with the human personality trait of novelty seeking. Nat Genet
12: 78–80.
Ernst, M., Zametkin, A. J., Matochik, J. A., Jons, P. J., Cohen, R. M. (1998). DOPA decarboxylase
activity in attention deficit hyperactivity disorder adults. A [fluorine-18]fluorodopa positron
emission tomographic study. J Neurosci 18: 5901–5907.
Ernst, M., Zametkin, A. J., Matochik, J. A., et al. (1999). High midbrain [18 F]DOPA accumu-
lation in children with attention deficit hyperactivity disorder. Am J Psychiatry 156: 1209–
1215.
Farchione, T. R., MacMillan, S. N., Rosenberg, D. R. (2003). Obsessive–compulsive disorder as
a fronto–striatal–thalamic dysfunction. In Mental and Behavioral Dysfunction in Movement
Disorders, ed. M. A. Bedord, Y. Agid, S. Chouinard, et al. Totowa, NJ: Humana Press, pp.
477–488.
Filipek, P. A., Semrud-Clikeman, M., Steingard, R. J., et al. (1997). Volumetric MRI analysis
comparing attention-deficit hyperactivity disorder and normal controls. Neurology 48: 589–
601.
Fitzgerald, K. D., Moore, G. J., Paulson, L. D., Stewart, C. M., Rosenberg, D. R. (2000). Proton
spectroscopic imaging of the thalamus in treatment-naive pediatric obsessive–compulsive
disorder. Biol Psychiatry 48: 55–72.
Frodl, T., Meisenzahl, E., Zetzsche, T., et al. (2002). Enlargement of the amygdala in patients with
a first episode of major depression. Biol Psychiatry 51: 708–714.
408 C. W. Kreipke, D. R. Rosenberg, and M. S. Keshavan

Geller, B., Zimerman, B., Williams, M., Bolhofner, K., Craney, J. L. (2001). Bipolar disorder
at prospective follow-up of adults who had prepubertal major depressive disorder. Am J
Psychiatry 158: 125–127.
Geller, D. A., Biederman, J., Faraone, S. V., et al. (2002). Attention-deficit/hyperactivity disorder
in children and adolescents with obsessive–compulsive disorder: fact or artifact? J Am Acad
Child Adolesc Psychiatry 41: 52–58.
Gilbert, A. R., Moore, G. J., Keshavan, M. S., et al. (2000). Decrease in thalamic volumes of
pediatric patients with obsessive–compulsive disorder who are taking paroxetine. Arch Gen
Psychiatry 57: 449–456.
Gilbert, A. R., Rosenberg, D. R., Harenski, K., et al. (2001). Thalamic volumes in patients with
first-episode schizophrenia. Am J Psychiatry 158: 618–624.
Gill, M., Daly, G., Heron, S., Hawi, Z., Fitzgerald, M. (1997). Confirmation of association between
attention deficit hyperactivity disorder and a dopamine transporter polymorphism. Biol
Psychiatry 2: 311–313.
Gogos, J. A., Morgan, M., Luine, V., et al. (1998). Catechol-O-methyltransferase-deficient mice
exhibit sexually dimorphic changes in catecholamine levels and behavior. Proc Natl Acad Sci
USA 95: 9991–9996.
Gottesman, I. L. (1991). Schizophrenia Genesis: The Origins of Madness. New York: Freeman.
Gross-Isseroff, R., Hermesh, H., Zohar, J., Weizman, A. (2003). Neuroimaging communality
between schizophrenia and obsessive–compulsive disorder: a putative basis for schizo-
obsessive disorder? World J Biol Psychiatry 4: 129–134.
Harrington, R., Fudge, H., Rutter, M., Pickles, A., Hill, J. (1990). Adult outcomes of childhood
and adolescent depression: psychiatric status. Arch Gen Psychiatry 47: 465–473.
Hauser, P., Dauphinais, I. D., Berrettini, W., et al. (1989). Corpus callosum dimensions mea-
sured by magnetic resonance imaging in bipolar affective disorder and schizophrenia. Biol
Psychiatry 26: 659–668.
Hesslinger, B., Thiel, T., Tebartz van Elst, L., Hennig, J., Ebert, D. (2001). Attention-deficit disorder
in adults with or without hyperactivity: where is the difference? A study in humans using
short echo 1 H-magnetic resonance spectroscopy. Neurosci Lett 304: 117–119.
Hill, D. E., Yeo, R. A., et al. (2003). Magnetic resonance imaging correlates of attention-
deficit/hyperactivity disorder in children. Neuropsychology 17: 496–506.
Hynd, G. W., Semrud-Clikeman, M., Lorys, A. R., Novey, E. S., Eliopulos, D. (1990). Brain
morphology in developmental dyslexia and attention deficit disorder/hyperactivity. Arch
Neurol 47: 919–926.
Hynd, G. W., Hern, K. L., Novey, E. S., et al. (1993). Attention deficit hyperactivity disorder and
asymmetry of the caudate nucleus. J Child Neurol 8: 339–347.
Kayl, A. E., Moore, III, B. D., Slopis, J. M., Jackson, E. F., Leeds, N. E. (2000). Quantitative
morphology of the corpus callosum in children with neurofibromatosis and attention-deficit
hyperactivity disorder. J Child Neurol 15: 90–96.
Keshavan, M. S., Montrose, D. M., Pierri, J. N., et al. (1997). Magnetic resonance imaging and
spectroscopy in offspring at risk for schizophrenia: preliminary studies. Prog Neuropsy-
chopharmacol Biol Psychiatry 21: 1285–1295.
409 Disordered brain development in psychiatric disorders

Keshavan, M. S., Rosenberg, D., Sweeney, J. A., Pettegrew, J. W. (1998). Decreased caudate volume
in neuroleptic-naive psychotic patients. Am J Psychiatry 155: 774–778.
Keshavan, M. S., Sujata, M., Mehra, A., Montrose, D. M., Sweeney, J. A. (2002a). Psychosis
proneness and ADHD in young relatives at risk for schizophrenia. Schizophr Res 59: 85–92.
Keshavan, M. S., Dick, E., Mankowski, I., et al. (2002b). Decreased left amygdala and hippocampal
volumes in young offspring at risk for schizophrenia. Schizophr Res 58: 173–183.
Kellner, C. H., Jolley, R. R., Holgate, R. C., et al. (1991). Brain MRI in obsessive–compulsive
disorder. Psychiatry Res 36: 45–49.
Kim-Cohen, J., Caspi, A., Moffitt, T. E., et al. (2003). Prior juvenile diagnoses in adults with
mental disorder: developmental follow-back of a prospective-longitudinal cohort. Arch Gen
Psychiatry 60: 709–717.
Kim, E. Y., Miklowitz, D. J. (2002). Childhood mania, attention deficit hyperactivity disorder and
conduct disorder: a critical review of diagnostic dilemmas. Bipolar Disord 4: 215–225.
LaHoste, G. J., Swanson, J. M., Wigal, S. B., et al. (1996). Dopamine D4 receptor gene polymor-
phism is associated with attention-deficit hyperactivity disorder. Mol Psychiatry 1: 121–124.
Ledoux, J. (2003). The emotional brain, fear, and the amygdala. Cell Mol Neurobiol 23: 727–738.
Luxenberg, J. S., Swedo, S. E., Flament, M. F., et al. (1988). Neuroanatomical abnormalities in
obsessive–compulsive disorder detected with quantitative X-ray computed tomography. Am
J Psychiatry 145: 1089–1093.
MacMaster, F. P., Carrey, N., Sparkes, S., Khan, S., Kusumakar, V. (2003). Proton spectroscopy in
medication free pediatric attention deficit hyperactivity disorder. Biol Psychiatry 53: 184–
187.
MacMillan, S. N., Szeszko, P. R., Moore, G. J., et al. (2003). Increased amygdala:hippocampal
volumes associated with severity of anxiety in pediatric major depression. J Child Adol
Psychpharm 13: 65–73.
Milberger, S., Biederman, J., Faraone, S. X., Guite, J., Tsuang, M. T. (1997). Pregnancy, deliv-
ery, and infancy complications and attention deficit hyperactivity disorder: issues of gene–
environment interaction. Biol Psychiatry 41: 65–75.
Modell, J. G., Mountz, J. M., Curtis, G. C., Greden, J. F. (1989). Neurophysiologic dysfunction
in basal ganglia/limbic striatal and thalamocortical circuits as a pathogenetic mechanism of
obsessive–compulsive disorder. J Neuropsychiatry Clin Neurosci 1: 27–36.
Moldin, S. O., Gottesman, I. I. (1997). At issue: genes, experiments and chance in schizophrenia-
positioning for the 21st century. Schizoph Bull 23: 547–561.
Moller, H. J. (2003). Bipolar disorder and schizophrenia: distinct illnesses or a continuum? J Clin
Psychiatry 64: 23–27.
National Institute of Mental Health Genetics Initiative Bipolar Group (1997). Genomic survey
of bipolar illness in the NIMH Genetics Initiative pedigrees: a preliminary report. Am J Med
Gen Neuropsychiatr Genet 74: 227–237.
Noga, J. T., Aylward, E., Barta, P. E., Pearlson, G. D. (1995). Cingulate gyrus in schizophrenic
patients and normal volunteers. Psychiatry Res 61: 201–208.
Nolan, C. L., Moore, G. J., Madden, R., et al. (2002). Prefrontal cortical volume in childhood-onset
major depression: preliminary findings. Arch Gen Psychiatry 59: 173–179.
410 C. W. Kreipke, D. R. Rosenberg, and M. S. Keshavan

O’Donovan, M. C., Williams, N. M., Owen, M. J. (2003). Recent advances in the genetics of
schizophrenia. Hum Mol Genet 12(Suppl 2): R125–R133.
Pine, D. S., Cohen, P., Brook, J. (2001). Adolescent fears as predictors of depression. Biol Psychiatry
50: 721–724.
Rosenberg, D. R., Keshavan, M. S., Dick, E. L., et al. (1997a). Corpus callosal morphology in
treatment-naive pediatric obsessive–compulsive disorder. Prog Neuropsychopharmacol Biol
Psychiatry 21: 1269–1283.
Rosenberg, D. R., Keshavan, M. S., O’Hearn, K. M., et al. (1997b). Frontostriatal measurement
in treatment-naive children with obsessive–compulsive disorder. Arch Gen Psychiatry 54:
824–830.
Rosenberg, D. R., Keshavan, M. S. (1998). A. E. Bennett Research Award. Toward a neurodevel-
opmental model of obsessive–compulsive disorder. Biol Psychiatry 43: 623–640.
Rosenberg, D. R., MacMaster, F. P., Keshavan, M. S., et al. (2000a). Decrease in caudate glutamater-
gic concentrations in pediatric obsessive–compulsive disorder patients taking paroxetine. J
Am Acad Child Adolesc Psychiatry 39: 1096–1103.
Rosenberg, D. R., Benazon, N. R., Gilbert, A., Sullivan, A., Moore, G. J. (2000b). Thalamic volume
in pediatric obsessive–compulsive disorder patients before and after cognitive behavioral
therapy. Biol Psychiatry 48: 294–300.
Rosenberg, D. R., MacMillan, S. N., Moore, G. J. (2001). Brain anatomy and chemistry may predict
treatment response in pediatric obsessive–compulsive disorder. Int J Neuropsychopharmacol
4: 179–190.
Rosenberg, D. R., Davanzo, P. A., Gershon, S. (2002a). Pharmacotherapy for Child and Adolescent
Psychiatric Disorders, 2nd edn (revised and expanded). New York: Marcel Dekker.
Rosenberg, D. R., Madden, R., Tang, J., et al. (2002b). Anterior cingulate glutamate and treat-
ment response in pediatric OCD. In Scientific Proceedings of the 49th Annual Meeting of the
American Academy of Child and Adolescent Psychiatry. Washington DC: American Academy
of Child and Adolescent Psychiatry, Vol. D18, p. 118.
Rubia, K., Overmeyer, S., Taylor, E., et al. (1999). Hypofrontality in attention deficit hyperactivity
disorder during higher-order motor control: a study with functional MRI. Am J Psychiatry
156: 891–896.
Sawa, A., Snyder, S. (2002). Schizophrenia: diverse approaches to a complex disease. Science 296:
692–695.
Schweitzer, J. B., Anderson, C., Ernst, M. (2000). Attention-deficit hyperactivity disorder: neu-
roimaging and behavioral/cognitive probes. In Functional Neuroimaging in Child Psychiatry,
ed. M. Ernst, J. Rumsey. Cambridge, UK: Cambridge University Press, pp. 278–297.
Shenton, M. E., Dickey, C. C., Frumin, M., McCarley, R. W. (2001). A review of MRI findings in
schizophrenia. Schizophr Res 49: 1–52.
Shihabuddin, L., Buchsbaum, M. S., Hazlett, E. A., et al. (1998). Dorsal striatal size, shape, and
metabolic rate in never-medicated and previously medicated schizophrenics performing a
verbal learning task. Arch Gen Psychiatry 55: 235–243.
Soares, J. C., Mann, J. J. (1997). The anatomy of mood disorders: review of structural neuroimag-
ing studies. Biol Psychiatry 41: 86–106.
411 Disordered brain development in psychiatric disorders

Sowell, E. R., Toga, A. W., Asarnow, R. (2000). Brain abnormalities observed in childhood-onset
schizophrenia: a review of the structural magnetic resonance imaging literature. Ment Retard
Dev Disabil Res Rev 6: 180–185.
Sprich-Buckminster, S., Biederman, J., Milberger, S., Faraone, S. W., Lehman, B. K. (1993). Are
perinatal complications relevant to the manifestation of ADD? Issues of comorbidity and
familiality. J Am Acad Child Adolesc Psychiatry 32: 1032–1037.
Steingard, R. J., Renshaw, P. F., Hennen, J., et al. (2002). Smaller frontal lobe white matter volumes
in depressed adolescents. Biol Psychiatry 52: 413–417.
Stine, O. C., Xu, J., Koskela, R., et al. (1995). Evidence for linkage of bipolar disorder to chromo-
some 18 with a parent-of-origin effect. Am J Hum Genet 57: 1384–1394.
Strakowski, S. M., DelBello, M. P., Sax, K. W., et al. (1999). Brain magnetic resonance imaging
of structural abnormalities in bipolar disorder. Arch Gen Psychiatry 56: 254–260.
Strakowski, S. M., DelBello, M. P., Zimmerman, M. E., et al. (2002). Ventricular and periventric-
ular structural volumes in first- versus multiple-episode bipolar disorder. Am J Psychiatry
159: 1841–1847.
Takahashi, T., Suzuki, M., Kawasaki, Y., et al. (2003). Perigenual cingulate gyrus volume in patients
with schizophrenia: a magnetic resonance imaging study. Biol Psychiatry 53: 593–600.
Tannock, R. (1998). Attention deficit hyperactivity disorder: advances in cognitive, neurobiolog-
ical, and genetic research. J Child Psychol Psychiatry 39: 65–99.
Taylor, M. A. (1992). Are schizophrenia and affective disorder related? A selective literature review.
Am J Psychiatry 149: 22–32.
Teicher, M. H., Anderson, C. M., Polcari, A., et al. (2000). Functional deficits in basal ganglia
of children with attention-deficit/hyperactivity disorder shown with functional magnetic
resonance imaging relaxometry. Nat Med 6: 470–473.
Thomas, K. M., Drevets, W. C., Dahl, R. E., et al. (2001). Amygdala response to fearful faces in
anxious and depressed children. Arch Gen Psychiatry 58: 1057–1063.
Vaidya, C. J., Austin, G., Kirkorian, G., et al. (1998). Selective effects of methylphenidate in
attention deficit hyperactivity disorder: a functional magnetic resonance study. Proc Natl
Acad Sci USA 95: 14494–14499.
Verdoux, H., Sutter, A. L. (2002). Perinatal risk factors for schizophrenia: diagnostic specificity
and relationships with maternal psychopathology. Am J Med Genet 114: 898–905.
Weissman, M. M., Wolk, S., Wickramaratne, P., et al. (1999). Children with prepubertal-onset
major depressive disorder and anxiety grown up. Arch Gen Psychiatry 56: 794–801.
Wu, J. C., Buchsbaum, M. S., Johnson, J. C., et al. (1993). Magnetic resonance and positron
emission tomography imaging of the corpus callosum: size, shape and metabolic rate in
unipolar depression. J Affect Disord 28: 15–25.
Zametkin, A. J., Nordahl, T. E., Gross, M., et al. (1990). Cerebral glucose metabolism in adults
with hyperactivity of childhood onset. N Engl J Med 323: 1361–1366.
Part IV

Clinical implications
 22

Can one identify preschizophrenia children?

Eugenia Kravariti, Paola Dazzan, Paul Fearon, and Robin M. Murray
Institute of Psychiatry, King’s College, London, UK

The 1980s and 1990s saw a surge of neurodevelopmental theories of schizophrenia,

which challenged the traditional notion of the disorder as an early form of demen-
tia or other degenerative condition (Kraepelin, 1919). Some early versions of the
neurodevelopmental hypothesis – sometimes unkindly termed “doomed from the
womb” theories – saw schizophrenia as the inevitable consequence of an early brain
lesion (Murray and Lewis, 1987; Weinberger, 1987). More recently, however, it has
become increasingly accepted that the likelihood of schizophrenia developing is
probabilistic rather than deterministic (Fearon et al., 2001).
The postulated aberration in neurodevelopment was initially thought to be silent
until late adolescence or early adulthood. It is now hypothesised to give rise to subtle
brain abnormalities, which determine the underlying susceptibility to schizophre-
nia, and translate into subtle neuro-behavioral deficits that are evident throughout
development (Murray, 1994). Therefore, identifying and characterizing the early-
emerging markers of the schizophrenia diathesis offers the theoretical possibility of
predicting the disorder. However, childhood antecedents of schizophrenia may vari-
ably reflect (i) integral components of the schizophrenia diathesis, (ii) non-specific
factors that potentiate this predisposition, or (iii) early manifestations of the dis-
order itself (Dworkin and Cornblatt, 1995). While all three types of antecedent are
relevant to estimating schizophrenia risk, only the third one is directly relevant to
establishing preschizophrenic status, and, hence, to predicting with certainty which
individuals will express the disorder.
In this chapter, we will explore the nature of the childhood antecedents of
schizophrenia, addressing the following questions. Are children who develop
schizophrenia in adulthood distinguishable from control children? How strong
is the association between the distinguishing features of preschizophrenia children
as a group and the later development of the disorder? How accurately can we identify

Neurodevelopment and Schizophrenia, ed. Matcheri S. Keshavan et al. Published by Cambridge University
Press.

C Cambridge University Press 2004.

415
Table 22.1. Predicting schizophrenia: neurointegrative risk factors in population-based, high-risk and
archival–observational studies

Risk factor OR/(RR) SEN (%) SPE (%) PPV (%) NPV (%)

Population-based studies
Delayed developmental milestones at 2 yrs 4.8 7 98 3 99
(Jones et al., 1994)a
Learned to stand at or after 12 months (1.4) – – – –
(Isohanni et al., 2001)b
Excess of neurological signs at 3 yrs 4.6 – – – –
(Cannon et al., 2002)c
Unusual movements at 7 yrs (Rosso et al., 2000)d 4.4 15 96 3 99
Motor coordination dysfunction at 7 yrs (Rosso 2.4 11 96 2 99
et al., 2000)d
High-risk studies
Deviance on a composite index of attentional, – 89 64 22 98
perceptual and motor dysfunctioning at
8–15 yrs (Marcus et al., 1987)e
Gross motor skills at 7–12 yrs – 75 73 33 94
(Erlenmeyer-Kimling et al., 2000)f
Archival-observational studies
Postural hand abnormalities in the first 2 yrs of 7.9g 63 82 – –
life (Walker et al., 1994)g
Choreoathetoid movements in the first 2 yrs of 8.2g 23 96 – –
life (Walker et al., 1994)g

OR, odds ratio; RR, risk ratio (the ratio between the observed and the expected cases); SEN, sensitivity; SPE,
specificity; PPV, positive predictive value; NPV, negative predictive value.
a The British MRC National Survey of Health and Development of 1946 followed prospectively a stratified

random sample (4746 alive at 16 years) of 13 687 births in Britain during the week 3–9 March 1946. Thirty
subjects developed schizophrenia at ages 16–43 and were compared with the remaining 4716 subjects.
b The Northern Finland Birth Cohort Study (Isohanni et al., 1998, 1999, 2001) followed prospectively 96%

of all births in Northern Finland in 1966 (12 058 liveborn children). Psychiatric outcome was established at
16–28 years (58 subjects had developed schizophrenia), and the comparison group included individuals with
no hospital-treated psychiatric outcome (Isohanni et al., 1998, 1999).
c The Dunedin Birth Cohort Study followed prospectively 91% (1037) of all children born between April

1972 and March 1973 in Dunedin (New Zealand). At age 26, 36 had developed schizophreniform disorder,
20 mania, and 278 anxiety/depression. These diagnostic groups were each compared with the remainder of
the cohort (healthy control group).
d The Philadelphia birth cohort study (Bearden et al., 2000; Rosso et al., 2000) followed prospectively over 90%

(9236) of all deliveries at two inner city hospital obstetric wards in Philadelphia between 1959 and 1966. The
cohort included 72 subjects who developed schizophrenia/schizoaffective disorder at ages 19–36, 63 healthy
siblings, and 7941 non-psychiatric controls.
e The National Institute of Mental Health (NIMH) Israeli Kibbutz–City High-Risk Study followed prospectively

50 offspring of schizophrenia parents and 50 offspring of psychiatrically normal parents. Diagnostic outcome
417 Can one identify preschizophrenia children?

preschizophrenia children on the basis of these characteristics? Are there early risk
factors that, in isolation or in combination, flag the preschizophrenia status? These
questions, including the putative neurobiological predictions of the schizophrenia
illness (Ch. 23), are also addressed elsewhere in this volume.
Where they are available from the primary sources or computable through infor-
mation derived from the text, the following association indexes and predictive
parameters will be reported: the odds ratio (OR) is the likelihood of developing
schizophrenia in the children who test positive for the predictor relative to those
who do not. Sensitivity is the percentage of children with adult schizophrenia out-
comes who test positive for the predictor. Specificity is the percentage of children
with normal adult outcomes who test negative for the predictor. Positive predictive
value is the probability of developing adult schizophrenia given the presence of the
predictor in childhood. Negative predictive value is the probability of escaping adult
schizophrenia given the absence of the predictor in childhood.
We will place particular emphasis on prospective population-based, high-risk,
and follow-back studies. All three types of study provide relatively unbiased infor-
mation, as both researchers and informants do not know which subjects will develop
schizophrenia, and the data collected are contemporaneous rather than retrospec-
tive. In addition, general population cohorts are less confounded by the selection
and information biases usually associated with case–control studies.

Neurointegrative predictors
Population-based studies
Unusual movements and coordination problems in mid-childhood were signif-
icantly associated with adult schizophrenia in the Philadelphia 1959–1966 Birth

was established at 26–32 years (nine high-risk subjects had developed schizophrenia-spectrum disorders). The
predictive validity of neurobehavioral deviance was estimated in a subgroup of 46 high-risk and 44 control
offspring.
f The New York High-Risk Project (Cornblatt et al., 1999; Erlenmeyer-Kimling and Cornblatt 1992; Erlenmeyer-

Kimling et al., 2000; Ott et al., 1998) recruited offspring of schizophrenia, affectively ill, and psychiatrically
normal parents in 1971–2 (sample A: 206, mean age 9.5 years) and 1977–9 (sample B: 150, mean age 9.0 years)
and followed them prospectively. Erlenmeyer-Kimling et al. (2000) assessed the predictive validity of gross
motor skills in a sample of 79 offspring of schizophrenia parents. Diagnostic outcome was established at a
mean age of 30.7 years (12 offspring had developed schizophrenia-related psychoses).
g Walker et al. (1994) observed childhood home videos of 30 schizophrenia patients and their 28 healthy siblings,

19 affective disorder patients and their 14 siblings, and 21 subjects from families with no mental illness. The
OR, sensitivity, specificity, positive and negative predictive values reported here refer to the comparison of the
preschizophrenia subjects with their healthy siblings.
418 E. Kravariti, P. Dazzan, P. Fearon, and R. M. Murray

Cohort Study (Rosso et al., 2000). The positive predictive value of these indices
remained low, however (Table 22.1). An excess of motor coordination dysfunction
was also observed in the unaffected siblings of subjects with adult schizophrenia,
suggesting that coordination problems might be related to a genetic vulnerability
to develop the disorder.
Individuals who developed schizophrenia in the 1946 British birth cohort study
(MRC National Survey of Health andDevelopment; Jones et al.,1994)were 4.8times
more likely to have delayed milestones (sitting, standing, walking, and talking) at
age 2 (Table 22.1). However, the probability of developing schizophrenia was only
3% (Table 22.1). The authors indicated that “. . . no child destined to develop
schizophrenia could be singled out as a late walker.” Subjects with a diagnosis of
childhood affective disorder (but not those with an adult-onset affective disorder)
according to an anxiety–depression rating also attained later motor milestones and
had an excess of grimaces and twitches at age 15 (van Os et al., 1997). Therefore,
although there were some similarities between the antecedents of schizophrenia and
those of affective disorders, the effect on the latter was smaller and more evident
in those with an early onset. Age at learning to stand, walk, or become potty-
trained in the Northern Finland 1966 Birth Cohort Study (Isohanni et al., 2001)
showed a linear association not only with schizophrenia (Table 22.1) but also with
other psychoses. By contrast, motor problems (more than 0.3 standard deviation
[SD]) and an excess of neurological signs (OR, 4.6) in the Dunedin study (Cannon
et al., 2002) were present in children who developed adult schizophreniform, but
not manic or anxiety/depression, outcomes. Similarly, children who developed
adult schizophrenia in the 1946 British cohort and in the 1958 cohort (National
Child Development Study of 1958; Done et al., 1994) studies showed more “hand
control and speech problems” than either children who remained well or those who
developed affective psychosis (Crow et al., 1995; Done et al., 1994; Leask et al., 2002).

High-risk studies
Fish (Fish, 1977; Fish et al., 1992) created the term “pandevelopmental retarda-
tion” to refer to abnormalities in physical growth, gross motor development and
visual–motor development that are frequently observed in subjects who develop
schizophrenia (Fish, 1987). Hyperkinesis, poor coordination, motor dysfunction,
and perceptual signs were present in eight of the nine children who later devel-
oped schizophrenia or schizoid personality disorder in the Israeli High Risk Study
(Marcus et al., 1987). Similarly, impairment in gross motor skills in the New York
High Risk Study was predictive of a future diagnosis of schizophrenia with a rea-
sonably high sensitivity and specificity (Erlenmeyer-Kimling et al., 2000). However,
even children with grossly abnormal motor skills only had a 33% probability of
developing schizophrenia (Table 22.1).
419 Can one identify preschizophrenia children?

Archival–observational studies
In a study of sibships filmed from infancy through at least the first 5 years of life, each
including one offspring who later developed schizophrenia and normal offspring,
viewers blind to the subjects’ diagnosis were able to identify the preschizophrenia
siblings on the basis of neuromotor abnormalities (Walker et al., 1994). Those who
would develop schizophrenia were 7.9 times more likely to show postural hand
abnormalities, and 8.2 times more likely to show choreo-athetoid movements than
their control siblings. These group differences were only significant until 2 years of
age, after which they tended to disappear. It is possible that preschizophrenia chil-
dren have an impaired cortical modulation on subcortical structures, as supported
by evidence of disrupted connectivity in patients with schizophrenia (Rosso et al.,
2000).
Such neurodevelopmental deficits could be the consequence of pregnancy and
delivery complications (for example, fetal hypoxia), which have been associated
with both schizophrenia (Geddes and Lawrie, 1995; McGrath and Murray, 1995;
McNeil, 1995) and unusual movements (Rosso et al., 2000). Alternatively, obstetric
complications may just play a non-specific, precipitating, or facilitating role in
subjects with a genetic liability to schizophrenia (Cannon et al., 2000a).

Neuropsychological predictors
Population-based studies
Prospective investigations of population-based samples have reported persistent
intellectual deficits, language pathology, and educational failures in preschizophre-
nia children and adolescents (Bearden et al., 2000; Cannon et al., 1999, 2000b;
Isohanni et al., 1998, 1999; Jones and Done, 1997; Jones et al., 1994), with the
respective functional abnormalities being relatively stable across developmental
periods (Cannon et al., 2000b; Jones and Done, 1997).
Individuals destined to develop schizophrenia in the Philadelphia Birth Cohort
Study showed intellectual abnormalities as early as 4 and 7 years of age (Cannon
et al., 2000b); intelligence quotient (IQ) category at either age emerged as a signif-
icant predictor of adult schizophrenia (Cannon et al., 2000b). These findings were
supplemented by the 1946 and 1958 British birth cohort studies (Jones and Done,
1997), which reported deficits in intellectual capacity, verbal and non-verbal rea-
soning, reading comprehension, and mathematics at various age points between
7 and 16 years of age in preschizophrenia individuals. Repeating a school grade
or receiving special education at age 14 (mainly owing to low IQ) also predicted
future schizophrenia in the Northern Finland Birth Cohort Study (Isohanni et al.,
1998). Similarly, poor intellectual functioning in a national sample of 18-year-
old male conscripts to the Swedish army emerged as a strong predictor of future
420 E. Kravariti, P. Dazzan, P. Fearon, and R. M. Murray

schizophrenia (David et al., 1997), a finding replicated in the national Israeli cohort
of males aged 16–17 years (Davidson et al., 1999).
A common finding in nearly all of the above studies is a linear association between
intellectual functioning and the risk for schizophrenia: the latter appears to be a
function of performance over the entire range of population scores, increasing
progressively as ability declines (Cannon et al., 1999, 2000b; David et al., 1997;
Davidson et al., 1999; Jones and Done, 1997; Jones et al., 1994). The Philadelphia
Birth Cohort Study, for example, reported a 30–60% increase in schizophrenia risk
per unit decrease in ability category (divided into five performance levels), such
that individuals scoring in the deficient range were five to six times more likely
to become schizophrenia than those scoring in the high-average to superior range
(Cannon et al., 1999). Similarly, David et al. (1997) reported a nine-fold increase
in schizophrenia risk among conscripts scoring in the lowest IQ band compared
with those falling within the highest one.
Despite the significant increase in schizophrenia risk in the lower end of the IQ
distribution, in no study did the disorder arise solely from a population subgroup
with low IQ, nor was there evidence of a subgroup with very low scores (Cannon
et al., 1999, 2000b; David et al., 1997; Davidson et al., 1999; Jones and Done, 1997;
Jones et al., 1994). Rather, the IQ distribution of the schizophrenia population
as a whole is shifted downward in a systematic fashion, most likely reflecting an
effect on each individual. This effect is rather subtle: IQ data at age 11 examined
together for the 1946 and 1958 birth cohorts revealed a shift of the schizophrenia
population mean by < 0.5 SD (Jones and Done, 1997). No single child destined to
develop schizophrenia in the former cohort could be singled out as having learning
difficulties (Jones et al., 1994).
Contrary to the prevailing view that schizophrenia risk increases linearly with
decreasing intellectual capacity, the Northern Finland Birth Cohort Study (Isohanni
et al., 1999) failed to confirm linearity in the association between educational
attainment and schizophrenia outcome, raising the possibility that it is distance
from the cognitive norm – in either direction – that increases the odds for the
disorder. Excellent school performance among 16-year-old males in the latter cohort
was associated with nearly a four-fold increase in schizophrenia risk, with 11% of the
preschizophrenia boys compared with only 3% of the comparison group (with no
hospital-treated psychiatric outcome) obtaining excellent school marks. In accord
with this finding, the proportion of preschizophrenia cases falling within the highest
IQ category among the Israeli males aged 16–17 years was six times higher than
that of the comparison subjects (with no hospital-treated psychiatric outcome)
(Davidson et al., 1999). It remains to be seen whether these surprising findings will
be confirmed in other populations.
421 Can one identify preschizophrenia children?

Premorbid language dysfunction is one of the most potent predictors of future

schizophrenia. Abnormal summary ratings of speech intelligibility at age 7 were
associated with a greater than 12-fold increase in risk for adult schizophrenia in
the Philadelphia cohort (Bearden et al., 2000). Similarly, the evidence for abnor-
mal speech development prior to schizophrenia is strong in the two British birth
cohorts (Jones and Done, 1997). In the 1958 birth cohort, the relevant abnormal-
ities are more than reflections of developmental timing: oral ability and quality of
speech at 7 and 11 years were more frequently rated as qualitatively abnormal in the
preschizophrenia children; speech defects (not caused by structural problems) per-
sisted throughout childhood and adolescence in the 1946 cohort, being associated
with a three-fold increase in schizophrenia risk.
Although the large sample sizes (ranging from approximately 5000 to 50 000
individuals) in these epidemiological studies provide sufficient statistical evidence
for the predictive potential of neuropsychological abnormalities, it would be mis-
leading to conclude that the latter can be used as effective population screens for
the identification of individuals at true risk for schizophrenia. A particular limita-
tion is the extent to which premorbid neuropsychological deficits can distinguish
between schizophrenia and other psychiatric disorders. Impairments in general
ability have also been recorded in the early biographies of individuals with other
psychoses and anxiety or depressive disorders, with any distinction between the
respective groups simply being one of magnitude: preschizophrenia individuals
usually perform worse than the other patient groups, which also perform below
the norm (David et al., 1997; Davidson et al., 1999; Jones and Done, 1997). Indeed,
the association between receiving special education or repeating a school grade
and the diagnostic category of “other psychoses” in the Finnish cohort was even
stronger than that with schizophrenia (Isohanni et al., 1998). A possible exception
to this rather undifferentiated pattern of associations is the one between speech
difficulties and future schizophrenia. The only school-based assessment that dis-
tinguished the preschizophrenia children from the other patient groups in the
British 1958 cohort was the qualitative evaluation of “speech difficulties” at both 7
and 11 years (Jones and Done, 1997).
A second problem is the significant impact on a population level of the false-
negative properties carried by even highly sensitive and specific screens, when used
for low-prevalence disorders such as schizophrenia (O’Toole, 2000). Table 22.2
presents the predictive profiles of the neuropsychological and educational risk fac-
tors examined in the various studies. Although the specificity of the various predict-
ors is moderate to high (53–99%), their sensitivity is less satisfactory (11–55%),
and only a small proportion (1–17%) of individuals falling within the “impaired”
range on the different measures eventually go on to develop schizophrenia. For
422 E. Kravariti, P. Dazzan, P. Fearon, and R. M. Murray

Table 22.2. Predicting schizophrenia: neuropsychological and educational risk factors in

population-based studies

Risk factor OR SEN (%) SPE (%) PPV (%) NPV (%)

IQ at 4 yrs (Cannon et al., 1999)a 1.3b 30 85 2 99

IQ at 7 yrs (Cannon et al., 1999)a 1.6b 33 83 2 99
IQ at 11 and 15 yrs (Jones et al., 1994)c 0.6 and 0.5b – – – –
IQ at 16–17 yrs (Davidson et al., 1999)d 1.6 54 53 – –
IQ < 96 at 18 yrs (David et al., 1997)e 3.5–8.6 55 67 1 100
Class below age level or special education 2.8 17 93 3 99
at 14 yrs (Isohanni et al., 1998)f
Excellent school marks at 16 yrs (Isohanni 3.8 11 97 4 99
et al., 1999)f
Abnormal speech at 7 yrs (Bearden et al., 12.7 14 99 17 99
2000)a
Abnormal speech at 2–15 yrs (Jones et al., 2.8 – – – –
1994)c
Expressive language at 7 yrs (Bearden et al., 0.7 – – – –
2000)a

OR, odds ratio; SEN, sensitivity; SPE, specificity; PPV, positive predictive value; NPV, negative predictive
value; IQ, intelligence quotient.
a See footnote d in Table 22.1.

b Odds ratio for linear trend, i.e., that associated with moving a category (e.g. tertile or quintile) of score

distribution.
c See footnote a in Table 22.1.

d The National Israeli sample of 16–17-year-old males included 509 subjects who developed schizophrenia

by the age of 26, and 9215 controls who did not appear in the national psychiatric registry by the same age
(Davidson et al., 1999).
e A national sample of 49 968 male conscripts to the Swedish army (David et al., 1997): 195 subjects developed

schizophrenia by the age of 33–34 and were compared with 49 773 individuals with no or other psychotic
outcomes. OR values (unadjusted for confounders) were estimated separately for each IQ band (< 74, 74–81,
82–89, 90–95) by David et al. (1997); SEN, SPE, PPV, and NPV were estimated for all the IQ (< 96) bands
together by the present authors.
f See footnote b, in Table 22.1.

example, receiving special education or repeating a school grade (mainly owing to

low IQ) at 14 years of age in the Northern Finland Birth Cohort Study (Isohanni
et al., 1998) correctly predicted 10 of the 58 cases of schizophrenia (sensitivity, 17%)
and 4986 of the 5351 non-cases (specificity, 93%). However, of the 375 individuals
repeating a class or receiving special education at age 14, only 10 went on to develop
schizophrenia (positive predictive value, 3%). It is, therefore, unrealistic to target
early identification and prevention programs for schizophrenia on individuals with
423 Can one identify preschizophrenia children?

deviant cognitive functioning or poor educational achievement. This observation

applies to all the remaining risk factors identified in the population-based studies
reviewed in this section (Table 22.2).

High-risk studies
The “high-risk” strategy reduces the chances of false positives in the face of low
prevalence disorders by studying individuals whose risk for developing schizophre-
nia is high, thus increasing the observed prevalence of the disorder in the sample
studied: children of one schizophrenia parent can be up to 15 times more likely to
develop the disorder than members of the general population. In addition, unlike
population-based studies, which usually study general aspects of development,
research in high-risk groups is a longitudinal quest for variables that predict the
disorder (Erlenmeyer-Kimling et al., 2000). As such, the high-risk paradigm usually
includes measurements of particular relevance to schizophrenia.
Impaired attention is the most extensively investigated and validated neuropsy-
chological predictor in the high-risk literature. For example, the preschizophrenia
offspring of affected parents in the Copenhagen High Risk Study displayed poor
concentration from infancy (Parnas et al., 1982). Similarly, childhood attentional
deficits in the Israeli high-risk sample successfully predicted the development of
schizophrenia-spectrum disorders (Mirsky et al., 1995). In addition, in the New York
High Risk Project, attentional deficits predicted schizophrenia-related psychoses as
opposed to other psychiatric outcomes (Erlenmeyer-Kimling et al., 2000), were
more prevalent among children of schizophrenia parents than offspring of affect-
ively ill patients (Cornblatt and Erlenmeyer-Kimling, 1985; Cornblatt et al., 1992;
Erlenmeyer-Kimling et al., 2000; Freedman et al., 1998), and displayed the greatest
predictive power within the group at genetic risk for schizophrenia (Cornblatt and
Erlenmeyer-Kimling, 1985; Cornblatt et al., 1992). These findings attest both to
the predictive validity and the diagnostic specificity of attentional dysfunction in
high-risk studies (Table 22.3).
Although other cognitive functions have been less extensively investigated, mem-
ory dysfunction, IQ change, and subtest scatter in standard intelligence tests may
provide sensitive indicators of the preschizophrenia state (Erlenmeyer-Kimling
et al., 2000; Ott et al., 1998). The use of intelligence scores from the Revised Wech-
sler Intelligence Scale for Children (WISC-R) or the Revised Adult Scale (WAIS-R)
in the New York High Risk Project achieved one of the best predictive profiles
reported in the high-risk literature to date, with a false-positive rate (equal to
1 minus specificity) of just 2–5% (Ott et al., 1998) (Table 22.3).
However, the focus on single neuropsychological constructs is less promising
than using clusters of cognitive and neurobehavioral variables. Combining an atten-
tional screen with behavioral ratings in the New York High Risk Project enhanced
424 E. Kravariti, P. Dazzan, P. Fearon, and R. M. Murray

Table 22.3. Predicting schizophrenia: neuropsychological risk factors in high-risk studiesa

Risk factor SEN (%) SPE (%) PPV (%) NPV (%)

Attention deficits at 12 yrs (Cornblatt et al., 1999)b 67 79 19 97

Attention deficits at 12 yrs and behavior ratings at 83 90 38.5 99
12–17 yrs (Cornblatt et al., 1999)b
Attention deviance at 7–12 yrs (Erlenmeyer-Kimling 58 82 37 92
et al., 2000)c
Memory deficits at 7–12 yrs (Erlenmeyer-Kimling et al., 83 72 34.5 96
2000)c
Attention deviance, memory and gross motor skills at 50 90 46 91
7–12 yrs (Erlenmeyer-Kimling et al., 2000)c
WISC(-R) or WAIS(-R) subtest scatter at 15 yrs, change 54–85 95–98 61–70 96–99
in Full Scale IQ and Vocabulary between 9 and 15 yrs,
Picture Arrangement minus Vocabulary score at 9 yrs
and parental risk (Ott et al., 1998)d
WISC(-R) or WAIS(-R) subtest scatter at 15 yrs and 56 90 62.5 87
Picture Arrangement minus Vocabulary score at 9 yrs
(Ott et al., 1998)e

SEN, sensitivity; SPE, specificity; PPV, positive predictive value; NPV, negative predictive value.
a Data from the New York High Risk Project (see footnote f, Table 22.1).

b The predictive validity of sustained attention – alone or in conjunction with behavioral ratings – was assessed

in a mixed sample of 21 offspring of schizophrenia parents, 26 children of affectively ill parents, and 40 offspring
of psychiatrically normal parents (total, 87). Diagnostic outcome was last updated when the subjects were in
their early thirties.
c The predictive validity of attention, memory, and motor skills (each alone or in conjunction with the other

predictors) was assessed in a sample of 79 offspring of schizophrenia parents. Diagnostic outcome was estab-
lished at a mean age of 30.7 years (12 offspring had developed schizophrenia-related psychoses).
d The accuracy of a logistic regression model (including variables from Wechsler Intelligence Scales for Children

and Adults) was examined for prediction of schizophrenia-related psychoses (versus no diagnosis) in a mixed
sample of 157 high- and low-risk children (offspring of schizophrenia, affectively ill, and psychiatrically normal
parents). The diagnostic outcome was established at the mean ages of 30.17 years (sample A) and 22.09 years
(sample B). The ranges of SEN, SPE, PPV and NPV resulted from varying the predicted cut-off probability.
e The predictive accuracy of WISC(-R)/WAIS(-R)-derived variables was examined in a sample of 39 offspring

of schizophrenia parents derived from the sample described in footnote d of this table.

measurably the overall accuracy of the attentional model, reducing false positives
by half (Cornblatt et al., 1999). Similarly, combining deficits in attention, verbal
memory, and gross motor skills in the above study achieved higher precision than
any of the contributing variables alone (Erlenmeyer-Kimling et al., 2000).
Table 22.3 presents the predictive profiles of selected neuropsychological predict-
ors used in high-risk studies. As expected from measures specifically selected for
425 Can one identify preschizophrenia children?

their relevance to schizophrenia, and from the increased prevalence of the disorder
in the relatively small samples screened, the examined predictors show moderate-
to-high sensitivity, with 50–85% of future spectrum cases being correctly identi-
fied. Of all subjects predicted to develop schizophrenia-related psychoses, 19–70%
went on to develop the outcome of interest, and, almost invariably, over 90% of
“non-deviant” performers did not develop schizophrenia or related psychoses.
However, the false-positive rates can sometimes be substantial. Although in some
models (Erlenmeyer-Kimling et al., 2000; Ott et al., 1998), only 2–10% of non-cases
are falsely predicted to develop spectrum disorders, about a third of non-cases are
false positives in others (Erlenmeyer-Kimling et al., 2000) (Table 22.3). In addition,
the predictive performance of the reviewed models cannot be extrapolated to the
general population, since only about 10–15% of future schizophrenia patients have
an ill parent, and the normal control samples in the existing high-risk studies are
relatively small (Erlenmeyer-Kimling et al., 2000).

Behavioral predictors
Population-based studies
Teacher-rated behavioral deviance emerged as a potent predictor of adult
schizophrenia in the 1946 and 1958 British birth cohort studies. In the 1946 study,
social anxiety at age 15 showed a significant linear association with schizophre-
nia, such that individuals scoring within the highest tertile were six times more
likely to develop the disorder relative to those scoring in the lowest tertile (Jones
et al., 1994). Teacher-rated anxiety was also a prominent characteristic of the
preschizophrenia 7-year-old males of the 1958 study (Crow et al., 1995), as were
hostility and inconsequential behavior: 50% of the preschizophrenia males com-
pared with only 10% of normal controls received abnormal global ratings of
over-reactive behavior. A clear gender effect was noted in the 1958 study: the
7-year-old girls who later developed schizophrenia were hardly distinguishable
from normal controls. Even though they were finally rated as more withdrawn,
depressed, and likely to dismiss adult values in later childhood (age 11), they
showed no evidence of over-reactive behavior at any time point, with the excep-
tion of a trend towards adult-directed hostility at age 11. By then, depression and
a trend for dismissal of adult values were also added to the behavioral profile
of the preschizophrenia males. In contrast to the preschizophrenia children, the
preaffective children were unremarkable in their profiles, with the exception of
hostility at age 7 (boys) and restlessness at either 7 (boys) or 11 (girls) years of age.
The behavioral abnormalities of the preneurotic individuals were similar to those
of the preschizophrenia group, although they emerged at a slightly later stage of
development and were present in the females rather than the males (Crow et al.,
1995).
426 E. Kravariti, P. Dazzan, P. Fearon, and R. M. Murray

Table 22.4. Predicting schizophrenia: behavioral risk factors in population-based and high-risk studies

Risk factor OR SEN (%) SPE (%) PPV (%) NPV (%)

Population-based studies
Global social maladjustment at 7 yrs 2.54 29 86 2 99
(Bearden et al., 2000)a
Focal behavioral deviance at 4 yrs (Bearden 1.68 33 81 1 99
et al., 2000)a
Focal behavioral deviance at 7 yrs (Bearden 1.65 31 81 2 99
et al., 2000)a
Poor social functioning at 16–17 yrs (males) 4.37 48 88
(Davidson et al., 1999)b
Poor organizational ability at 16–17 yrs 2.03 30 88
(males) (Davidson et al., 1999)b
High-risk studies
Behavior problems at 15 yrs (males) (Olin
and Mednick, 1996)c
a. Disciplinary problems 39 96 88 71
b. Teacher-predicted future emotional or 43 88 67 73
psychotic problems
c. Repeated a grade 29 96 80 73
d. Disturbs class with inappropriate 50 82 64 72
behavior
e. Emotional reaction persists; high-strung 40 85 60 72
f. Lonely and rejected by peers 44 82 58 71
g. Treated by psychologist for problem 36 86 57 73
h. Easily excited or irritated 41 82 58 69
Behavior problems at 15 yrs (females) (Olin
and Mednick, 1996)c
a. Teacher-predicted future emotional or 30 97 75 82
psychotic problems
b. Nervousness 46 82 45 82

OR, odds ratio; SEN, sensitivity; SPE, specificity; PPV, positive predictive value; NPV, negative predictive
value.
a See footnote d in Table 22.1.

b See footnote d in Table 22.2.

c The Copenhagen High Risk Project (Olin and Mednick, 1996) recruited 207 offspring of mothers with severe

schizophrenia and 104 matched control offspring of parents and grandparents with no history of psychiatric
hospitalization (mean age 15.1 yrs). At a mean age of 42 years, 33 subjects had developed schizophrenia,
10 schizophrenia-associated psychosis, and 46 schizotypal or paranoid personality disorder. The predictive
parameters refer to the comparison of the preschizophrenia group with those with non-psychiatric outcomes.
427 Can one identify preschizophrenia children?

Global social maladjustment at 7 years of age was a potent predictor of later

schizophrenia in the Philadelphia Birth Cohort Study (Cannon et al., 1999)
(Table 22.4). In addition, signs of behavioral deviance such as meaningless laugh-
ter, excessive crying, echolalia, thumb-sucking, and nail biting (focal deviance) at
ages 4 and 7 were predictive of both schizophrenia (Table 22.4) and unaffected
sibling status (Bearden et al., 2000). The authors commented that focal behavioral
deviance may reflect the underlying genetic susceptibility to schizophrenia, whereas
global social maladjustment is likely to predate the onset of the clinical phenotype
(Bearden et al., 2000).
Behavioral assessments of males aged 16–17 years in the national Israeli cohort
indicated significant differences in measures of social functioning, individual auton-
omy, and organizational ability between subjects who later developed schizophrenia
and healthy controls (Davidson et al., 1999) (Table 22.4). Using the social function-
ing scale alone with a cut-off point of the lowest two quintiles accurately predicted
group membership in 43% of future cases and 93% of controls.

High-risk studies
In the seminal study of Barbara Fish (1987), half of the preschizophrenia children
with chronic disturbance from childhood had been abnormally quiet infants with
flat affect, who developed withdrawn or explosive and non-compliant behavior by
the age of 3 years.
Eight of the nine adults who received schizophrenia-spectrum diagnoses in the
National Institute of Mental Health (NIMH) Israeli Kibbutz–City High Risk Study
had shown antisocial and/or withdrawn behavior in childhood (Marcus et al.,
1987). These abnormalities were mainly evident in the males. According to teachers
and parents, females had shown adequate interpersonal adjustment and less overt
dysfunctioning, even though the girls themselves had reported acute feelings of
rejection and not belonging.
Teacher ratings of social dysfunction and behavioral deviance among 15-year-old
males were among the best predictors of future schizophrenia in the Copenhagen
High Risk Study (Olin and Mednick, 1996) (Table 22.4). Compared with normal
controls, preschizophrenia males were more likely to pose discipline problems and
disturb the class, to be emotionally reactive and easily excited, to have been treated
by a school psychologist or repeated a grade, and to be lonely and rejected by
their peers. While both males and females with a schizophrenia outcome were
significantly more likely than normal controls to receive a prediction of future
psychosis by their teachers, the only other item to have predictive value for females
was nervousness. These behavioral indices predicted future schizophrenia with
low to moderate sensitivity, high specificity, and moderate to high positive and
negative predictive values (Table 22.4). Compared with males with other diagnostic
428 E. Kravariti, P. Dazzan, P. Fearon, and R. M. Murray

outcomes, preschizophrenia boys showed more externalized signs of abnormality,

educational failures, and interpersonal difficulties, and received more predictions
of future psychosis. Preschizophrenia girls were clearly distinguishable only from
those with non-psychotic psychiatric outcomes, showing more internalized signs
of abnormality and being more prone to psychotic and emotional problems (Olin
and Mednick, 1996).

Archival–observational studies
In a preliminary study of sibships filmed from infancy through at least the first 5
years of life, each including a preschizophrenia child (Walker and Lewine, 1990),
viewers blind to the subjects’ adult psychiatric outcome were able to identify the
future cases at levels well above chance: 25 of their 32 judgements were correct
(P < 0.05). Furthermore, the distinguishing behavioral features of the target chil-
dren compared with their control sibling(s) (i.e. less responsiveness, eye contact,
and positive affect) were apparent at a gross level of analysis, as the viewers were
not instructed to use specific criteria. Similarly, on extending the above study to
a larger sample, Walker et al. (1993) observed lower rates of joy expressions and
increased negative affect in the future cases compared with same-gender siblings.
The emotional blunting of infants, children, and adolescents destined to develop
schizophrenia (Fish, 1987; Walker et al., 1993) is consistent with the lack of
hedonic capacity observed in affected adults, suggesting continuity between some
schizophrenia antecedents and the clinical syndrome. Other evidence in support
of this notion comes from a retrospective study. Baum and Walker (1995) found
positive correlations between withdrawal in childhood and adolescence and psy-
chomotor poverty in adult schizophrenia. Taken together, these findings suggest
that certain primary negative or deficit symptoms may be enduring traits, mani-
fested across the lifespan of schizophrenia patients.

Conclusions
Using a statistical analogy, the null hypothesis that preschizophrenia children as a
group cannot be identified on the basis of early developmental, cognitive, social,
and behavioral characteristics can be rejected. Subtle, yet detectable, abnormalities
in all these functional domains have been reliably reported in preschizophrenia
children, achieving a statistically significant predictive status in epidemiological,
high-risk and archival–observational studies. However, the predictive profile of
these characteristics on a population level advises against their use as schizophre-
nia screens in unselected populations. To date, no single predictor or composite
model has shown sufficient sensitivity, specificity, and positive/negative predictive
values. In the various studies reviewed, 11–93% of individuals with adult-onset
429 Can one identify preschizophrenia children?

schizophrenia were missed on the basis of their childhood characteristics; 1–47%

were falsely predicted to develop the disorder; as few as 12% or as many as 99% of
youths with a given deviant characteristic did not develop schizophrenia as adults;
and as many as 31% of children with no detectable deviance did so. On a popula-
tion level, the misclassification rate can be substantial. Despite the promise of the
various predictors as tools for estimating schizophrenia risk, no single or composite
marker of the preschizophrenia status has been detected so far. The lack of complete
specificity to schizophrenia and the overlapping distributions of the predictors in
normal and preschizophrenia populations suggest that predicting the disorder on
an individual basis is still not feasible.
However, it is becoming increasingly clear that combining risk indicators from
different functional domains can impressively enhance the accuracy of predictive
models. In addition, the measures investigated by the high-risk paradigm can aid
in the identification of truly vulnerable individuals within the high-risk popula-
tion. Although the misclassification rates are still sufficiently high to prevent the
implementation of invasive protocols for preventive intervention, this subgroup
may potentially form a target for cognitive intervention treatments.

REFERENCES

Baum, K. M., Walker, E. F. (1995). Childhood behavioral precursors of adult symptom dimensions
in schizophrenia. Schizophr Res 16: 111–120.
Bearden, C. E., Rosso, I. M., Hollister, J. M., et al. (2000). A prospective cohort study of child-
hood behavioral deviance and language abnormalities as predictors of adult schizophrenia.
Schizophr Bull 26: 395–410.
Cannon, T. D., Rosso, I. M., Bearden, C. E., Sanchez, L. E., Hadley, T. (1999). A prospective cohort
study of neurodevelopmental processes in the genesis and epigenesis of schizophrenia. Dev
Psychopathol 11: 467–485.
Cannon, T. D., Rosso, I. M., Hollister, J. M., et al. (2000a). A prospective cohort study of genetic
and perinatal influences in the etiology of schizophrenia. Schizophr Bull 26: 351–366.
Cannon, T. D., Bearden, C. E., Hollister, J. M., et al. (2000b). Childhood cognitive functioning in
schizophrenia patients and their unaffected siblings: a prospective cohort study. Schizophr
Bull 26: 379–393.
Cannon, M., Caspi, A., Moffitt, T. E., et al. (2002). Evidence for early-childhood, pan-
developmental impairment specific to schizophreniform disorder: results from a longi-
tudinal birth cohort. Arch Gen Psychiatry 59: 449–456.
Cornblatt, B. A., Erlenmeyer-Kimling, L. (1985). Global attentional deviance as a marker of risk
for schizophrenia: specificity and predictive validity. J Abnorm Psychol 94: 470–486.
Cornblatt, B. A., Lenzenweger, M. F., Dworkin, R. H., Erlenmeyer-Kimling, L. (1992). Childhood
attentional dysfunctions predict social deficits in unaffected adults at risk for schizophrenia.
Br J Psychiatry 18(Suppl.): 59–64.
430 E. Kravariti, P. Dazzan, P. Fearon, and R. M. Murray

Cornblatt, B. A., Obuchowski, M., Roberts, S., Pollack, S., Erlenmeyer-Kimling, L. (1999). Cog-
nitive and behavioral precursors of schizophrenia. Dev Psychopathol 11: 487–508.
Crow, T. J., Done, D. J., Sacker, A. (1995). Childhood precursors of psychosis as clues to its
evolutionary origins. Eur Arch Psychiatry Clin Neurosci 245: 61–69.
David, A. S., Malmberg, A., Brandt, L., Allebeck, P., Lewis, G. (1997). I. Q. and risk for schi-
zophrenia: a population-based cohort study. Psychol Med 27: 1311–1323.
Davidson, M., Reichenberg, A., Rabinowitz, J., et al. (1999). Behavioral and intellectual markers
for schizophrenia in apparently healthy male adolescents. Am J Psychiatry 156: 1328–1335.
Done, D. J., Crow, T. J., Johnstone, E. C., Sacker, A. (1994). Childhood antecedents of schizophre-
nia and affective illness: social adjustment at ages 7 and 11. Br Med J 309: 699–703.
Dworkin, R. H., Cornblatt, B. A. (1995). Predicting schizophrenia. Lancet 345: 139–140.
Erlenmeyer-Kimling, L., Cornblatt, B. A. (1992). A summary of attentional findings in the New
York High-Risk Project. J Psychiatr Res 26: 405–426.
Erlenmeyer-Kimling, L., Rock, D., Roberts, S. A., et al. (2000). Attention, memory, and motor
skills as childhood predictors of schizophrenia-related psychoses: the New York High-Risk
Project. Am J Psychiatry 157: 1416–1422.
Fearon, P., Cannon, M., Murray, R. M. (2001). A critique of the idea and science of risk-factor
research in schizophrenia. Int J Ment Health 30: 82–90.
Fish, B. (1977). Neurobiologic antecedents of schizophrenia in children. Evidence for an inherited,
congenital neurointegrative defect. Arch Gen Psychiatry 34: 1297–1313.
(1987). Infant predictors of the longitudinal course of schizophrenia development. Schizophr
Bull 13: 395–409.
Fish, B., Marcus, J., Hans, S. L., Auerbach, J. G., Perdue, S. (1992). Infants at risk for schizophrenia:
sequelae of a genetic neurointegrative defect. A review and replication analysis of pandys-
maturation in the Jerusalem infant development study. Arch Gen Psychiatry 49: 221–235.
Freedman, L. R., Rock, D., Roberts, S. A., Cornblatt, B. A., Erlenmeyer-Kimling, L. (1998). The
New York High-Risk Project: attention, anhedonia and social outcome. Schizophr Res 30:
1–9.
Geddes, J. R., Lawrie, S. M. (1995). Obstetric complications and schizophrenia: a meta-analysis.
Br J Psychiatry 167: 786–793.
Isohanni, I., Jarvelin, M.-R., Nieminen, P., et al. (1998). School performance as a predictor of
psychiatric hospitalization in adult life. A 28-year follow-up in the northern finland 1966
birth cohort. Psychol Med 28: 967–974.
Isohanni, I., Jarvelin, M. R., Jones, P., Jokelainen, J., Isohanni, M. (1999). Can excellent school
performance be a precursor of schizophrenia? A 28-year follow-up in the Northern Finland
1966 birth cohort. Acta Psychiatr Scand 100: 17–26.
Isohanni, M., Jones, P. B., Moilanen, K., et al. (2001). Early developmental milestones in adult
schizophrenia and other psychoses. A 31-year follow-up of the Northern Finland 1966 Birth
Cohort. Schizophr Res 52: 1–19.
Jones, P., Done, D. J. (1997). From birth to onset: a developmental perspective of schizophrenia
in two national birth cohorts. In Neurodevelopment and Adult Psychopathology, ed. M. S.
Keshavan, R. M. Murray. Cambridge, UK: Cambridge University Press, pp. 119–136.
431 Can one identify preschizophrenia children?

Jones, P., Rodgers, B., Murray, R., Marmot, M. (1994). Child developmental risk factors for adult
schizophrenia in the British 1946 birth cohort. Lancet 344: 1398–1402.
Kraepelin, E. (1919). Dementia Praecox and Paraphrenia. Edinburgh: Livingston.
Leask, S. J., Done, D. J., Crow, T. J. (2002). Adult psychosis, common childhood infections and
neurological soft signs in a national birth cohort. Br J Psychiatry 181: 387–392.
Marcus, J., Hans, S. L., Nagler, S., et al. (1987). Review of the NIMH Israeli Kibbutz-City Study
and the Jerusalem Infant Development Study. Schizophr Bull 13: 425–438.
McGrath, J., Murray, R. M. (1995). Risk factors for schizophrenia: from conception to birth. In
Schizophrenia, ed. S. Hirsh, D. Weinberger. Oxford: Blackwell, pp. 187–205.
McNeil, T. F. (1995). Perinatal risk factors and schizophrenia: selective review and methodological
concerns. Epidemiol Rev 17: 107–112.
Mirsky, A. F., Ingraham, L. J., Kugelmass, S. (1995). Neuropsychological assessment of attention
and its pathology in the Israeli cohort. Schizophr Bull 21: 193–204.
Murray, R. M. (1994). Neurodevelopmental schizophrenia: the rediscovery of dementia praecox.
Br J Psychiatry 165: 6–12.
Murray, R. M., Lewis, S. W. (1987). Is schizophrenia a neurodevelopmental disorder? Br Med J
(Clin Res Ed) 295: 681–682.
Olin, S. C., Mednick, S. A. (1996). Risk factors of psychosis: identifying vulnerable populations
premorbidly. Schizophr Bull 22: 223–240.
O’Toole, B. I. (2000). Screening for low prevalence disorders. Aust N Z J Psychiatry 34(Suppl):
S39–S46.
Ott, S. L., Spinelli, S., Rock, D., et al. (1998). The New York High-Risk Project: social and general
intelligence in children at risk for schizophrenia. Schizophr Res 31: 1–11.
Parnas, J., Schulsinger, F., Schulsinger, H., Mednick, S. A., Teasdale, T. W. (1982). Behavioral
precursors of schizophrenia spectrum. A prospective study. Arch Gen Psychiatry 39: 658–
664.
Rosso, I. M., Bearden, C. E., Hollister, J. M., et al. (2000). Childhood neuromotor dysfunction in
schizophrenia patients and their unaffected siblings: a prospective cohort study. Schizophr
Bull 26: 367–378.
van Os, J., Jones, P. B., Lewis, S., Wadsworth, M., Murray, R. M. (1997). Developmental precursors
of affective illness in a general population birth cohort. Arch Gen Psychiatry 54: 625–631.
Walker, E., Lewine, R. J. (1990). Prediction of adult-onset schizophrenia from childhood home
movies of the patients. Am J Psychiatry 147: 1052–1056.
Walker, E. F., Grimes, K. E., Davis, D. M., Smith, A. J. (1993). Childhood precursors of schizophre-
nia: facial expressions of emotion. Am J Psychiatry 150: 1654–1660.
Walker, E. F., Savoie, T., Davis, D. (1994). Neuromotor precursors of schizophrenia. Schizophr
Bull 20: 441–451.
Weinberger, D. R. (1987). Implications of normal brain development for the pathogenesis of
schizophrenia. Arch Gen Psychiatry 44: 660–669.
 23

High-risk studies, brain development,

and schizophrenia
Matcheri S. Keshavan
University of Pittsburgh School of Medicine, Pittsburgh and
Wayne State University School of Medicine, Detroit, USA

Research on schizophrenia since the 1970s is conspicuous for its “shift to the left,”
emphasizing earlier phases of schizophrenia. This shift toward first-episode studies
replaced earlier cross-sectionally designed neurobiological studies of mostly those
with chronic schizophrenia, thus minimizing the confounds of chronicity and treat-
ment effects (Keshavan and Schooler, 1992). During the 1990s, attention shifted
further left, underscoring the importance of studying the prodromal phase of
schizophrenia with a glimmer of hope for secondary, and possibly primary preven-
tion (McGorry, 1998). Neurodevelopmental models of schizophrenia (Murray and
Lewis, 1987; Weinberger, 1987) have spurred the characterization of the premorbid
phase of schizophrenia. Studies of the premorbid phase help us to identify neu-
rodevelopmentally mediated vulnerability factors, examine the evolution of the
early phase of schizophrenia, unravel the premorbid pathophysiology without the
potential confounds of illness or treatment effects, and facilitate early diagnosis and
preventive intervention (Cornblatt and Obuchowski, 1997; Ch. 24).
This chapter will review the several approaches to investigate premorbid risk
for schizophrenia. What follows will be a critical appraisal of the existing studies
focusing on the populations at risk for schizophrenia, the issues surrounding study
design, predictive and outcome factors identified so far, and the timing of the
studies. Conceptual and methodological issues and future directions will also be
discussed. Specifically, we will address the following questions. Who are the most
likely individuals that, when studied, will reveal useful insight about the nature
of vulnerability to schizophrenia? How should such studies be designed, and what
are the methodological issues that need to be addressed in these studies? What
neurodevelopmental predictors, and/outcome measures, if studied, are likely to
yield the best insights? When (i.e. during which critical periods of development)
is it most fruitful to assess the risk factors? Focused, hypothesis-driven studies

Neurodevelopment and Schizophrenia, ed. Matcheri S. Keshavan et al. Published by Cambridge University
Press.

C Cambridge University Press 2004.

432
433 High-risk studies, brain development, and schizophrenia

mindful of the above questions and deploying state of the art neurobiological tools
of assessment are likely to provide fresh impetus to this important, rather neglected
field of research. They are reviewed here.

At-risk populations
Studies of premorbid risk refer to the investigation of individuals who are consid-
ered more vulnerable to develop schizophrenia than are individuals in the general
population. Studies of risk for schizophrenia can be retrospective or “follow-back”
investigations (Walker et al., 1993) or prospective. Prospective studies could involve
longitudinal investigation of large cohorts of unselected general populations (such
as birth cohorts) or individuals selected for one or other index of high-risk (HR).
These studies have utilized genetic propensity, neurobehavioral markers, or psy-
chopathology to identify the risk status. In this section, we will outline the potential
merits and disadvantages of these strategies, review the lessons learned from the
early first HR studies and present a rationale for more focused next-generation
studies to examine premorbid risk.

General population cohort studies

Longitudinal studies of general population samples have been used to investigate
developmental antecedents of schizophrenia. Two British birth cohorts, the MRC
National Survey of Health and Development of 1946 (Jones et al., 1994) and the
National Child Development Study of 1958 (Done et al., 1994), in which the par-
ticipants have lived through most of the age of risk, showed that children destined
to develop schizophrenia could be differentiated based on motor and cognitive
dysfunction throughout childhood. David et al. (1997) showed a significant asso-
ciation between low intelligence quotient (IQ) score and later schizophrenia in a
cohort study of male Swedish conscripts. Population-based cohort studies have
also revealed valuable information about potential etiological variables; the North
Finland Birth Cohort (Rantakallio et al., 1997) and the UK National Child Develop-
ment Study (Leask et al., 2002) revealed associations between childhood infections
and later schizophrenia. Risk for schizophrenia has been associated in birth cohort
studies with urban place of birth (Harrison and Owen, 2003), migration (Cantor-
Graae et al., 2003), paternal age (Brown et al., 2002), birth order (Kemppainen
et al., 2001), exposure to prenatal rubella (Brown et al., 2001), and low maternal
and birth weights (Wahlbeck et al., 2001).
The main advantage with general population cohort studies is that they are an
unbiased estimate of the at-risk population; hence the findings, however limited,
are generalizable and can be specifically applied to this illness. However, there are
limitations. First, large population samples are needed to yield a sufficient number
434 M. S. Keshavan

of cases, and the number of predictive variables is often inadequate to address the
specific questions. Second, birth cohort studies involve prolonged follow-up per-
iods and are very expensive, leading to a delay of several decades before gaining
predictive knowledge. However, prospective studies of young adult and adolescent
cohorts (e.g. Davidson et al., 1999) may have an advantage of requiring a shorter
follow-up. Third, birth cohort studies are limited by “cohort effects,” which refer
to variation in illness characteristics (such as age of onset) over time among indi-
viduals defined by shared temporal features such as year of birth. This could lead
to variable findings between studies, and bias in interpretation of data (Di Maggio
et al., 2001). Finally, since large populations are involved, and most cohort studies
were not begun with schizophrenia-related research questions in mind, the data
collected at the outset are not fine-grained enough to ascertain neurodevelopmen-
tal antecedents of schizophrenia with any degree of specificity (Jones and Tarrant,
1999).

Follow-back and retrospective studies

Follow-back studies examine precursors of adult-onset psychopathology by exam-
ining medical or scholastic archives of individuals with known outcome in
adulthood. Using this approach (archival–observational approach), Walker and
colleagues (Walker and Lewine, 1990; Walker et al., 1993) collected old home movies
from families of schizophrenia patients. Children who later developed schizophre-
nia were characterized by subtle but significant neuromotor abnormalities com-
pared with their unaffected siblings. These provocative findings support the view
that dysmaturation of motor systems may occur in the future schizophrenia patient
even in early childhood. This approach is relatively free from recall bias (because of
the archival source of information), and a representative sample of the schizophre-
nia population is obtained; hence the findings are fairly generalizable to both genetic
and non-genetic forms of schizophrenia, unlike the genetic HR studies. However,
the information available for such studies is severely limited and may or may not
be related to pathophysiological questions of the premorbid diathesis.
One needs to make a distinction between “follow-back” and retrospective stud-
ies; the former utilize archival sources of information that are relatively unbiased
whereas the latter are limited by the problems of recall bias. Retrospective studies
rely on chart-review data and other historical sources of information. The high
prevalence of behavioral and cognitive difficulties in the general population can
make them likely to be selectively attributed to the illness. Further, the informa-
tion obtained in any retrospective analysis is likely to be biased; can be variable,
depending on the doctor’s record-keeping practices; and often is too sketchy to
allow examination of specific neurodevelopmental hypotheses.
435 High-risk studies, brain development, and schizophrenia

Genetic high-risk studies

Genetic factors are among the best-established etiological factors in schizophrenia.
Prospectively studying relatives of schizophrenia patients with enhanced genetic
risk should, therefore, be instructive in our search of markers that may predict the
onset of the illness. Pearson and Kley (1957) first recognized the benefits of this HR
approach. The risk of schizophrenia increases relative to the general population
in proportion to the proximity of the relationship and the number of affected
relatives. Offspring of parents with schizophrenia represent an attractive HR popu-
lation for study since having one schizophrenia parent entails about 13% risk of
developing the illness, and having two schizophrenia parents increases the risk to
about 40% (Gottesman and Shields, 1982). In terms of the risk, having a first-
degree relative with schizophrenia increases the risk by five times in parents, eight
times in siblings, 13 times in children with one schizophrenia parent and 40 times
in children of two schizophrenia parents compared with the general population
(where it is about 1%). Heritability of schizophrenia, calculated from correlations
of liability in monozygotic and dizygotic twins and presuming that schizophrenia
is multifactorial–polygenic in inheritance, is estimated to be 60–90% (Gottesman,
1991; Kendler, 2002; McGuffin et al., 1984).
Mednick and several other investigators initiated HR studies in the early 1960s
and 1970s and some of these “first-generation” studies have continued to date
(Table 23.1). These studies typically involved follow-up of offspring of schizophre-
nia parents, though younger siblings (Maier et al., 1992; Saitoh et al., 1984;
Weinberger et al., 1981) and discordant monozygotic twins have also been studied
as at-risk populations (Reveley et al., 1982; Suddath et al., 1990). An advantage
of the twin approach is that this is an elegant way to tease apart genetic from
non-genetic developmental risk factors in the vulnerability to schizophrenia. How-
ever, twin studies are limited by difficulties in generalizability of twin data to the
non-twin population, increased frequency of obstetric complications associated
with twin births, and problems in determining zygosity.
Three HR studies, the New York Infant Study (Fish et al., 1992), the Swedish
High Risk Study (McNeil et al., 1993), and the Israeli Infant Study (Marcus et al.,
1993) followed the offspring from birth onwards. The New York High Risk Project
(NYHRP) (Erlenmeyer-Kimling et al., 1995) and the Israeli Kibbutz High Risk
Study (Mirsky et al., 1995) involved offspring from elementary school ages; the
Copenhagen High Risk Project (CHRP) (Mednick et al., 1987) and the Edinburgh
High Risk Study (Johnstone et al., 2003) followed the subjects from adolescence
onwards. Some, but not all of these studies have followed subjects through the risk
period and have provided data on risk for schizophrenia and related disorders.
Rates of axis I schizophrenia and related psychotic disorders among the offspring
Table 23.1. High-risk studies in schizophreniaa

Study Sample Beginning year/duration Main predictors Outcome

of follow-up

New York Infant High Risk 12 HR offspring; 12 controls 1952/> 30 years follow-up Pandysmaturation, predicted outcome Scz, 8%; spectrum PD, 50%
Study (Fish et al., 1992) from infancy
Boston Providence High 118 HR Scz; 126 HR Aff 1959/7 years Low IQ at age 7 associated with risk
Risk Study (Goldstein for Scz
et al., 2000)
Copenhagen High Risk 207 HR offspring; 104 1962/30+ years Poor affective control; psychoticism on Psychosis, 20.8%; spectrum
Project (Parnas et al., controls; mean age 15 years MMPI; institutional care; unstable PD, 21.9%
1993) early rearing
Israel High Risk Study 25 HR; 25 controls; all raised 1964/27 years Poor neurobehavioral functioning; Scz, 8%; Aff, 24%
(Mirsky et al., 1995) in kibbutz; age 8–14 years anxiety; undesirable behavior
Rochester High Risk Study HR Scz 20; HR Aff 38 1972/10+ years follow-up Parental psychopathology and family 52% of offspring with
(Wynne et al., 1987) from infancy variables predict school functioning psychopathology
in the offspring
Jerusalem Infant Study HR Scz 29; other 30; 1973/20+ years Poor neurobehavioral functioning 4.1% Scz among HR Scz;
(Marcus et al., 1987) controls 27 0% in controls
New York High Risk Project 63 HR offspring; 100 1971/20+ years Attentional impairment; poor verbal Psychosis, 18.6%; spectrum
(Erlenmeyer-Kimling controls; 43 psychiatric memory; motor abnormalities; PD, 18.1%
et al., 1995) controls behavioral problems; psychoticism on
MMPI
Finnish Adoptive Family 180 offspring of Scz mothers 1974/20+ years Communication deviance in parents Scz, 5.2%; psychosis, 7.8%
Study (Tienari et al., adopted away; 200 associated with schizophrenia risk
1994) offspring of mentally
well mothers
Edinburgh High Risk Study 162 HR Scz; 36 controls 1994/5+ years followed up Childhood behavioral abnormalities; self 18 of 162 (11%) developed Scz
(Johnstone et al., 2003) from adolescence/early and interviewer rated schizotypy; drug
adulthood abuse; major life events; temporal lobe
volume reductions

HR, high risk; Scz, schizophrenia; PD, personality disorder; Aff, affective disorder; MMPI, Minnesota Multiphasic Personality Inventory.
a Includes only studies with published data pertaining to psychopathological outcome at follow-up.
437 High-risk studies, brain development, and schizophrenia

Table 23.2. Relative advantages and limitations of prospective research designs to assess
developmental premorbid risk

General population Genetic Neurobehavioral Clinical “Enhanced”

cohort studies HR studies HR studies HR studies Clinical/genetic
HR studies

Statistical power/cost − ± ± + +
Feasible number of ± ± ± + +
predictors
Generalizability to + − − ± ±
schizophrenia
Specificity to + + ± − +
schizophrenia

HR, high risk; −, weakness; ±, questionable; +, strength.

of schizophrenia patients have ranged from 8% (NYHRP studies) to 21% (CHRP

study), and these risks have been substantially higher than in control offspring.
Offspring of schizophrenia parents also have significantly elevated risk for cluster
A personality disorders (Erlenmeyer-Kimling et al., 1995).
The prospective HR approaches are free from recall bias. Since subjects with a
genetic propensity for a known disorder are studied, it is feasible to examine puta-
tive, disease-specific predictors. The main disadvantage of the earlier HR studies,
however, is the lack of statistical power, and the relatively modest cost effectiveness
of the studies. Large samples of subjects will need to be studied, potentially lim-
iting the number of feasible predictive variables. Another disadvantage is that the
findings may be generalizable to only patients with familial schizophrenia. Finally,
the low reproductive rates of schizophrenia patients make it hard to recruit ade-
quate numbers of offspring. Despite these shortcomings, studies of HR offspring
have indicated that some individuals with genetic risk show evidence of increased
liability to schizophrenia. However, the findings are highly variable across studies
and often lack specificity (Gooding and Iacono, 1995; Niemi et al., 2003; Sarfati
and Hardy-Bayle, 2002; reviewed by Cornblatt and Obuchowski, 1997; Erlenmeyer-
Kimling, 2000) (Table 23.2).

“Neurobehavioral” high-risk strategies

The genetic HR strategy is likely to overlook the majority of schizophrenia patients
who do not have an affected relative. Nearly 63% of persons with schizophrenia
have neither a first- nor a second-degree relative with this illness (Gottesman and
Erlenmeyer-Kimling, 2001). Because of this, an alternative strategy has been uti-
lized: using one or other neurobehavioral characteristics to define risk. For this
438 M. S. Keshavan

purpose, the putative neurobehavioral marker(s) has to be a valid and relatively

specific characteristic of the schizophrenia diathesis. Both population-based and
clinic-based samples have been used. Some studies have used psychometric indices
of deviation (e.g. high scores on the Minnesota Multiphasic Personality Inventory
[MMPI]; Carter et al., 1999) or schizotypal personality features (Lenzenweger, 1994;
Siever, 1994). Performance on questionnaires such as the Magical Ideation Scale
and the Perceptual Aberration Scale (Chapman et al., 1976, 1978) have also been
used to identify schizotypal subjects for this purpose. Using MMPI scores obtained
during the ninth grade in 15 000 adolescents, Hanson et al. (1990) observed that
those who later developed schizophrenia were characterized by elevations in the
schizophrenia and the psychopathic deviate scales.
Other studies have used psychophysiological measures. Venables et al. (1978)
observed that electrodermal hyper-responsiveness in early childhood was predic-
tive of deviant smooth pursuit eye movements (SPEM) later, though it was not
predictive of a schizophrenia outcome. Josiassen et al. (1985) have shown reduced
amplitudes on late positive evoked response potentials in college students with
extreme scores on Chapman’s psychosis proneness scales. Lencz et al. (1993) have
observed an association between deviant SPEM and schizotypal personality traits.
Collectively, these observations suggest that certain personality characteristics such
as schizotypy may be related to physiological indices such as SPEM, evoked response
potentials, and electrodermal responsiveness. However, none of these measures
have been shown in prospective studies to predict schizophrenia outcome reliably
at follow-up (Cornblatt and Obuchowski, 1997).
Like the genetic HR strategies, the neurobehavioral HR strategy is free from
recall bias. However, its utility will depend on the choice of the markers used; in a
longitudinal study, the initial choice of predictive measures will be guided by the
state of understanding of the illness at that time, which is a “moving target” that
surely changes over time. The longer the duration of follow-up, the more likely it
is that advances in the field would have made prior predictive markers increasingly
out of date. Further, normative data for developmental changes are often lacking;
thus, it is difficult to be certain by what age these measures have “stabilized.” For this
reason, it is unwise to conduct long-term follow-up studies using this strategy. The
biobehavioral measures often lack specificity for schizophrenia; the significance of
these markers for the pathophysiology of schizophrenia is also often unclear (i.e.
construct validity). Finally, this strategy also suffers from a lack of statistical power,
and is ineffective in cost terms (Table 23.2).

Clinical high-risk strategies

The onset of schizophrenia in adolescence or early adulthood is preceded by a period
of non-specific, subthreshold psychotic-like and negative symptoms for weeks to
439 High-risk studies, brain development, and schizophrenia

years. This prodromal phase is essentially a retrospective concept. However, in

recent years, there has been an increasing interest in prospectively characterizing
the prodromal signs and symptoms of schizophrenia, with the hope of potentially
diagnosing, and even intervening, early in the illness (McGlashan, 1998). The goal in
these studies is to identify potential clinical predictors of “conversion” to psychosis
and schizophrenia during follow-up. This approach seeks to identify individuals at
risk for psychosis in a clinically defined sample, hence the term clinical HR strategy
(Lencz et al., 2001). To pursue this goal, an attempt has been made to operationalize
criteria for prospective characterization of the prodrome (Yung et al., 1998). Recent
data suggest that early pharmacotherapeutic intervention in the prodromal phase
may potentially reduce the rate of “conversion” to psychotic disorders (McGorry
et al., 2002). However, the difficulty with this approach is that many persons who
may not have converted to schizophrenia could receive unnecessary treatment.
The definition of the prodrome, therefore, needs to be stringent and to include
trait-related (family history, personality traits) and state-related psychopathological
criteria (subthreshold positive and negative symptoms) and functional impairment
(Yung et al., 1998).
Using a clinical HR strategy, the frequency of conversion to psychosis in prospect-
ively characterized prodromal samples has been reported to be approximately 41%
over 12 months in the Melbourne PACE clinic study (Yung et al., 2003). In the
Melbourne study, highly significant predictors of psychosis included prolonged
prodromal symptoms, poor functioning at intake, low-grade psychotic symptoms,
depression, and disorganization. Combining some predictive measures yielded
good sensitivity (86%), specificity (91%), positive predictive value (80%), and neg-
ative predictive value (94%) for the emergence of psychosis within 6 months. One
of the advantages of the clinical HR strategy is its generalizability to schizophrenia
but a relative lack of specificity. Another advantage is that given the temporal prox-
imity of prodromal symptoms to schizophrenia, the follow-up durations needed
are shorter.

“Enhanced” high-risk strategies

One way to enhance the power of HR studies is to combine the genetic and psy-
chobiological and/or clinical HR strategies, rather than use any one approach alone.
In this way, subjects could be selected initially for the presence of an affected rel-
ative, and then only those that also showed one or other psychobiological mark-
ers (such as a personality deviation, SPEM impairment, attentional deviation, or
evoked response potential abnormality) or those with prodromal symptomatol-
ogy (using one or other operational definitions) would be finally included (Yung
et al., 1998). Such an approach is supported by the observation that, among off-
spring of schizophrenia parents, those with a deviant MMPI score had a higher
440 M. S. Keshavan

risk of manifesting a schizophrenia outcome (Moldin et al., 1991) and that global
attentional deviance was predictive of the subsequent emergence of schizophrenia
among the HR offspring (Cornblatt and Erlenmeyer-Kimling, 1989). Such an
“enhanced HR strategy” suffers from the difficulty that findings from this design
cannot be generalized to all of schizophrenia because a narrowly defined popula-
tion is studied. However, this approach is likely to have more statistical power to
detect differences from controls and can, therefore, be used to conduct the newer
“second-generation” HR studies involving more expensive neurobiological studies
(Table 23.2).

Methodological issues
As discussed above, all the strategies described seeking to elucidate premorbid
neurodevelopmental risk factors have methodological limitations. Lack of statistical
power, increased cost owing to prolonged follow-up and the use of ill-established
and “dated” predictive measures characterize the HR and birth cohort studies;
the problems of recall bias and the inadequacy of retrospective information are
limitations of retrospective studies. Studies of genetic “ultra-HR” populations such
as discordant monozygotic twins and children of two parents with schizophrenia
are limited by the fact that these are rare populations to find. Other methodological
issues that have plagued the previous HR studies are worth considering.

Reliability of parental diagnoses

Early HR studies were initiated at a time when explicit diagnostic criteria for psychi-
atric disorders had not yet been developed, and structured psychiatric diagnostic
interviews were not used for obtaining reliable information for the diagnosis. There-
fore, many patients diagnosed as schizophrenic in these studies may not fulfill the
current DSM-IV (American Psychiatric Association, 1994) criteria for this illness.
Diagnostic criteria are bound to change over time, but availability of comprehensive
clinical information at baseline is critical to enable the researcher to apply future
diagnostic schema to such information.

Specificity to schizophrenia
Several of the early HR studies lacked psychiatric control groups (i.e. offspring of
parents with non-schizophrenic disorders) and so failed to examine the issue of
specificity. It is known that individuals at genetic risk for schizophrenia are at an
increased risk for non-schizophrenic psychiatric disorders (Amminger et al., 1999).
It remains unclear whether this is from a transmitted risk for a broad range of psy-
chopathology or whether non-schizophrenic psychopathology is a precursor, or a
milder manifestation, of schizophrenia. Follow-back studies and prospective birth
441 High-risk studies, brain development, and schizophrenia

cohort studies have shown that individuals with schizophrenia in adult life fre-
quently have prior histories of childhood internalizing and externalizing disorders
(Kim-Cohen et al., 2003). These disorders may also precede other psychiatric dis-
orders. The issue of diagnostic specificity can best be addressed in HR studies by
having appropriate psychiatric control groups.

Generalizability to schizophrenia
The HR studies have often suffered from difficulties in generalizability to
schizophrenia. Children of schizophrenia patients are fewer and paternity may often
be uncertain. For these reasons, only children of female schizophrenia patients were
included in many studies (Fish et al., 1992; Parnas et al., 1993; Tienari et al., 1994).
The etiological factors among offspring of schizophrenia mothers may involve a
higher risk of non-genetic factors (such as alcohol/drug use during pregnancy and
increased perinatal complications) interacting with the genetic factors; therefore,
the causative risk factors may differ depending on which parent had the illness
(Verdoux and Sutter, 2002). Some studies excluded offspring because of lack of
intact families (e.g. Erlenmeyer-Kimling et al., 1995). There are practical diffi-
culties in conducting longitudinal studies of offspring from non-intact families.
Recruiting non-representative samples of HR subjects could lead to loss of critical
information on risk factors, leading to the problem of throwing the “baby out with
the bathwater.”

Predictive and outcome measures

The emerging neurodevelopmental models of schizophrenia have led to a search
for the possible developmentally mediated and trait-related alterations that may
precede schizophrenia. A trait marker, to be useful in studies of familial vulnerabil-
ity to schizophrenia, should (i) robustly distinguish the individuals with the illness
from control populations; (ii) be stable over time; (iii) have greater prevalence in
family members of patients identified schizophrenia patients than in the general
population and be associated with psychotic spectrum disorder in family members;
(iv) be correlated with subsequent development of psychotic spectrum illness in
HR children and precede the development of clinical manifestations of psychotic
spectrum disease; and (v) be relatively non-invasive and reliable (Garver, 1987;
Sarfati and Hardy-Bayle, 2002). While none of the biobehavioral markers so far
studied in HR studies fulfills all of these criteria (Table 23.3), some are relatively
more promising. Nevertheless, biological and neurobehavioral variables that may
not strictly meet the above criteria may still be of considerable interest for pre-
diction and prevention purposes. Nuechterlein et al. (1992) have described medi-
ating vulnerability factors that may be intervening variables between risk factors
442 M. S. Keshavan

Table 23.3. Relative advantages and disadvantages of putative neurobehavioral and biological markers
for use in prospective high-risk studies

Sensitivity Specificity Heritability Stability Predictive power

Cognitive
Attentional impairment + ± + + +

Psychophysiological
Smooth pursuit eye movements + ± + ? ?
P300a − − ± ± ?
P50a + ± ? ? ?
Neurobiological
MRI + ± ± + ±
1
H MRS (NAA) + ± + ± ?
31
P MRS (PME) + ± ? ? ?

MRI, magnetic resonance imaging; MRS, magnetic resonance spectroscopy; NAA, N-acetyl aspartate; PME,
phosphomonoesters; +, good; ± , inconsistent; −, unsatisfactory; ?, data unavailable.
a Components of cognitive evoked response potentials.

and symptoms. An example is social cognitive deficits, which may mediate the
relation between cognitive impairment and psychopathology in predisposition to
schizophrenia (Keshavan and Hogarty, 1999). In this review, we confine ourselves
to a discussion of risk markers in the conventional sense. Further details are also
provided on this issue in Ch. 22.

Predictive measures
Cognitive deficits
The strongest evidence of impairment in relatives of schizophrenia patients appears
to be in sustained attention, abstract thinking, and perceptual motor speed (Kremen
et al., 1994). Among the various neuropsychological measures, the continuous per-
formance test appears to be consistently associated with liability to schizophrenia
(Cornblatt and Keilp, 1994). In the NYHRP, attentional impairment in childhood
predicted 58% of the HR subjects who developed schizophrenia spectrum disor-
ders in adulthood (Erlenmeyer-Kimling, 2000). Attentional impairment is trait
related, stable over time, and related to genetic vulnerability (Michie et al., 2000).
Gross motor skills were also abnormal in 75% of offspring, while false-positive
rates were 27%. Short-term verbal memory was impaired in 83% of offspring who
later developed schizophrenia (Erlenmeyer-Kimling, 2000), showing a high sen-
sitivity but with relatively high false-positive rates (28%). By contrast, attentional
impairments had lower sensitivity (58%) and also lower false-positive rates (18%).
443 High-risk studies, brain development, and schizophrenia

In summary, therefore, attentional impairments may be among the most useful

neurobehavioral measures for prediction of outcome in HR offspring at risk for
schizophrenia.

Neuromotor and minor physical anomalies

Fish (1984) originally observed neuromotor deviations in about a half of the off-
spring subjects in her infant and children study, a pattern she termed “pandysmat-
uration.” Similar neuromotor dysfunctions have been observed in other studies,
including the NYHRP (Erlenmeyer-Kimling, 2000), and may predict affective flat-
tening in adolescence (Dworkin et al., 1993).
Minor physical anomalies such as malformations of the ear or palate and facial
dysmorphology may provide important clues to understanding schizophrenia spec-
trum disorders from a neurodevelopmental perspective. Offspring of schizophre-
nia parents with a high number of minor physical anomalies have been found to
develop schizophrenia spectrum disorders more often than other psychopathology
(Dworkin et al., 1993). In the Edinburgh High Risk Study (EHRS), minor physical
anomalies were elevated in the HR subjects compared with controls but did not
predict psychotic symptoms (Lawrie et al., 2001).

Electrophysiological measures
A physiological measure that has received attention in HR studies is eye track-
ing abnormality (Levy et al., 1994), seen in approximately 50% of adult relatives;
studies of SPEMs in adolescent HR subjects have shown significant dysfunction
compared with healthy comparison subjects (Ross, 2003). However, this measure
has not been investigated as a predictor of schizophrenia risk in prospective stud-
ies. Cognitive evoked response potentials have also been proposed as measures of
liability; prolonged latency and reduced amplitude of N100, P300, and P50 compo-
nents have been observed among relatives (Friedman and Squires-Wheeler, 1994).
Abnormal auditory event potentials (Schreiber et al., 1989) and electrodermal hypo-
or hyper-responsiveness (Dykes et al., 1992; Hollister et al., 1994) have also been
demonstrated, albeit less consistently.

Neurobiological measures
The advent of in vivo structural and physiological neuroimaging studies since the
early 1980s has raised the possibility that altered brain structure and function could
be detected in the premorbid phase of schizophrenia. New in vivo approaches to
examine the brain biology of abnormal neurodevelopment are being developed.
Several studies, including our own (Keshavan et al., 2002a; Lawrie et al., 1999;
Schreiber et al., 1999; Seidman et al., 2002), have shown evidence of structural
brain abnormalities in young relatives of schizophrenia parents. In recent years, two
444 M. S. Keshavan

prospective HR follow-up studies have been initiated, the EHRS and the Pittsburgh
Risk Evaluation Program (PREP). The EHRS has made important observations
regarding the neurobehavioral and neurobiological predictors of the emergence
of psychosis among young HR relatives: these include self- and observer-rated
schizotypy, school behavior abnormalities, and medial temporal volume reduc-
tions (reviewed by Johnstone et al., 2003; Miller et al., 2002). Cross-sectional data
from the PREP have provided preliminary magnetic resonance spectroscopy (MRS)
data suggesting that reductions in N-acetylaspartate (an in vivo marker of neu-
ronal integrity) can be detected in offspring at risk for schizophrenia (Keshavan
et al., 1997). Similar reductions have been observed in adult relatives of schizophre-
nia patients (Callicott et al., 1998). Blood oxygenation level-dependent and contrast
functional magnetic resonance imaging enables abnormal regional brain activation
to be studied in adolescent HR subjects. In a preliminary study, reduced activation
in prefrontal brain regions was seen in HR adolescents during a spatial working
memory task (Keshavan et al., 2002b). In vivo phosphorus MRS studies in HR sub-
jects have shown alterations in membrane phospholipid metabolism (decreased
phosphomonoesters, which are precursors of membrane phospholipids) similar to
those seen in first-episode schizophrenia (Keshavan et al., 2003); similar results have
been observed by Klemm et al. (2001). Overall, data regarding the predictive value
of neuroimaging measures for the development of psychotic symptoms are still
sparse, since HR studies incorporating such measures have started only recently,
and only limited follow-up data are available (Johnstone et al., 2003).

Outcome measures
The choice of outcome measures is also critical in designing HR studies. Devel-
opmental pathways can vary widely between individuals; one specific etiological
factor can lead to diverse psychopathological outcomes (multi-finality), and sev-
eral etiological variables and several developmental pathways can lead to a common
clinical outcome (equi-finality) (Von Bertalanffy, 1968). The failure to identify reli-
able predictors of future illness in previous HR studies may have been related to
the use of narrowly defined schizophrenia as the gold standard in assessing out-
come. The likelihood of the adolescent relatives at risk for developing schizophrenia
actually being diagnosed with schizophrenia during the span of follow-up in these
studies was relatively small; in HR follow-up studies, the schizophrenia-related psy-
chotic disorders (schizophrenia, schizoaffective disorders, and schizophreniform
disorders) occurred in 16–18% of offspring of subjects followed up into adult-
hood (Erlenmeyer-Kimling et al., 1995). However, retrospective analyses as well
as follow-up studies of young at-risk individuals suggest that many individuals
develop behavioral and emotional disturbances that may represent features of a
445 High-risk studies, brain development, and schizophrenia

“broad” schizophrenia spectrum of psychopathology. These features form three

groups.
The non-schizophrenia psychiatric disorders (non-schizophrenia psychotic disor-
ders, mood and anxiety disorders) Prospective studies have shown that adol-
escents with non-psychotic psychiatric diagnoses have a significantly increased
frequency of subsequent emergence of schizophrenia (Amminger et al., 1999;
Hafner, 1990; Weiser et al., 2001).
Schizotypy This represents a set of personality dimensions that may underlie the
predisposition to schizophrenia. Approximately 16–20% of high risk offspring
develop cluster A personality disorders (Erlenmeyer-Kimling et al., 1995); adol-
escents with schizotypal personality traits appear to be at a particularly higher
risk for future psychosis (Chapman et al., 1994; Kwapil, 1998; Kwapil et al.,
1997).
Attenuated psychotic and negative symptoms Several studies have shown that the
emergence of subthreshold or attenuated psychotic symptoms during childhood
or adolescence, and brief intermittent psychotic symptoms, have a high predic-
tive value for the future development of schizophrenia (Klosterkotter et al., 2001;
Poulton et al., 2000) or of full-blown psychotic episodes (Yung et al., 2003).
Each of these categories of broad-spectrum psychopathology can be operationally
defined; taken together, one or other of these manifestations may be seen in at least
one third of adolescent relatives at familial risk for schizophrenia. The schizophre-
nia spectrum of psychopathology can, therefore, provide us with a point of entry
for identifying individuals at risk for schizophrenia; characterization of the biobe-
havioral predictors of such broad-spectrum psychopathology is likely to help us
eventually to predict the emergence of schizophrenia by a “close-in” strategy. This
approach may be of advantage from a public health point of view, since such data
may allow more than one outcome to be identified, as well as prevented (Jones and
Tarrant, 1999).

Critical periods of development

Certain periods during development may be particularly sensitive for the impact
of adverse biological or environmental influences. Such “sensitive” periods include
the fetal, neonatal, and early childhood periods, as well as adolescence. Studies of
adolescent individuals at risk for schizophrenia are advantageous in studies of pre-
morbid risk for schizophrenia for several reasons. First, adolescence is characterized
by continuing experience-dependent reorganization of neural structures. Neuro-
biological phenomena that undergo major changes during such critical periods
(e.g. delta sleep, neuronal membrane synthesis, and cortical gray matter struc-
ture) are also abnormal in schizophrenia, suggesting that an abnormality in the
446 M. S. Keshavan

peri-adolescent brain maturational processes may underlie pathogenesis of

schizophrenia (Feinberg, 1982; Keshavan and Hogarty, 1999). Second, adolescence
represents a developmental period most proximal to the illness onset; follow-up
of adolescent HR subjects may, therefore, be cost effective since a shorter dura-
tion of follow-up may be needed to determine outcome. Third, genetic HR studies
(reviewed by Gooding and Iacono, 1995) suggest that at least a subgroup of HR sub-
jects may become increasingly deviant with age during childhood and adolescence.
Finally, recent data support the predictive ability of neurobehavioral indices in
adolescence for schizophrenia-related outcome. General population cohort stud-
ies (Gunnell et al., 2002) and “historical” prospective (or follow-back) studies
(Davidson et al., 1999) have shown that schizophrenia could be predicted by deficits
in social and intellectual functioning that were evident during adolescence. Neu-
robehavioral and neurobiological characterization of adolescent HR subjects is,
therefore, likely to be fruitful; however, it is important to ensure that the variables
being examined have temporally stabilized, because of the enormous influence of
age on several biological parameters of interest.

Conclusions
Studies of HR subjects have revealed impaired attentional and neurobehavioral
abnormalities in individuals at genetic risk for schizophrenia. These observations
are consistent with the neurodevelopmental basis of schizophrenia and may have
some predictive value. Further, these findings provide a good foundation for future
studies; the goal of future HR studies is to elucidate the premorbid neurobiolog-
ical risk for schizophrenia and utilize such knowledge for early identification and
intervention in this illness. The following caveats are worth keeping in mind while
designing such studies.
First, previous HR studies were limited by their expense, as they involved pro-
longed follow-up periods with the likelihood of only a small proportion of the
individuals developing the illness. Table 23.2 outlines the merits and disadvantages
of the various approaches to studying premorbid developmental risk. It is proposed
in this review that an “enhanced” HR strategy that involves a genetically at-risk
population and the use of putative neurobehavioral/psychopathological markers
of risk for further selection of participants may increase the likelihood of iden-
tifying premorbid risk indicators in schizophrenia. Using a narrow approach of
identifying only individuals with a genetic predisposition to schizophrenia is likely
to limit generalizability of HR studies to schizophrenia; an alternative is to define
risk broadly, using both vulnerability indicators (family history, biological markers)
as well as prodromal psychopathological indicators (e.g. “subthreshold” psychotic
and negative symptoms) (Yung et al., 1998). Such individuals are difficult to
447 High-risk studies, brain development, and schizophrenia

identify and characterize reliably, however. Carefully coordinated, multicenter stud-

ies of broadly defined HR subjects are a valuable approach toward enhancing sta-
tistical power.
Second, the first-generation HR studies have often been explorative, at least in
part, because of the lack of adequate theoretical models. Since the 1980s, newer
conceptualizations such as the neurodevelopmental models of schizophrenia have
generated many testable predictions; the advent of non-invasive methodologies
such as structural and functional neuroimaging allow testing of such predictions.
Future HR studies are, therefore, likely to be more hypothesis driven. The choice of
predictive measures in future HR studies should be driven by the emerging neurobi-
ological hypotheses of schizophrenia. Among the promising predictive markers are
neurobehavioral indices such as attentional impairment and neuroimaging mea-
sures reflecting the integrity of frontotemporal and subcortical brain structures. The
follow-up studies should also include a broad range of outcome measures since pre-
vious studies have suggested that risk is increased not only for schizophrenia but
also for a broader spectrum of schizophrenia-related psychopathology. Issues of
sensitivity, specificity, and generalizability should also be carefully considered.
Third, the question of when might it be most fruitful to study premorbid risk
for schizophrenia should also draw upon ongoing newer scientific research in
schizophrenia. Adolescence is increasingly viewed as a critical period of develop-
ment during which the schizophrenia diathesis might unfold. Studies of adolescent
HR subjects also offer the advantage of proximity to the onset of schizophrenic
illness and are, therefore, likely to be cost efficient.
Fourth, schizophrenia is a complex multifactorial and polygenic disorder con-
sidered to result from an interaction between genetic and environmental factors.
Genetic linkage and candidate gene association studies have implicated various
loci and genes in predisposition to schizophrenia. These include polymorphisms in
the genes encoding catechol-O-methyltransferase, regulatory G-protein signaling
(RGS4), neuregulin, and dysbindin, among others (Harrison and Owen, 2003).
Large-scale association studies that examine many gene polymorphisms simulta-
neously are required to predict genetic risk for schizophrenia. The known ethnic
divergence of gene polymorphisms makes it important to construct a database
of polymorphisms related to schizophrenia in each ethnic group. Few studies so
far have examined the value of specific genetic polymorphisms in combination
with environmental risk factors for predicting the risk for psychoses among HR
subjects.
Finally, while this review has focused on premorbid neurodevelopmental
antecedents of schizophrenia, the approaches discussed herein are applicable to
other neuropsychiatrisc developmental disorders as well. Bipolar disorder, which
has been considered to have neurodevelopmental origins (Nasrallah and Tolbert,
448 M. S. Keshavan

1997), also has an onset in late adolescence or early adulthood; obsessive compulsive
disorder has also been recently viewed as a neurodevelopmental disorder (Ch. 21).
Recent advances in developmental neurobiology and neuroscience make it rea-
sonable to expect a paradigm shift in research on schizophrenia. Studies of vul-
nerability as well as protective factors and studies of nature as well as nurture are
needed. It is critical that such research is linked meaningfully to our efforts for early
detection and intervention of this debilitating disorder. It is hoped that the third
millennium will usher in a new generation of research studies on HR populations
and move us closer to piecing together the puzzle of schizophrenia.

Acknowledgements
This work was supported in part by NIMH grants MH 01180, MH 64023, and
MH 45156, and by a NARSAD grant.

REFERENCES

American Psychiatric Association (1994). Diagnostic and Statistical Manual of Mental Disorders,
4th edn. Washington, DC: American Psychiatric Press.
Amminger, G. P., Pape, S., Rock, D., et al. (1999). Relationship between childhood behavioral
disturbance and later schizophrenia in the New York High-risk Project. Am J Psychiatry 156:
525–530.
Brown, A. S., Schaefer, C. A., Wyatt, R. J., et al. (2002). Paternal age and risk of schizophrenia in
adult offspring. Am J Psychiatry 159: 1528–1533.
Brown, R. T., Freeman, W. S., Perrin, J. M., et al. (2001). Prevalence and assessment of attention-
deficit/hyperactivity disorder in primary care settings. Pediatrics 107: E43.
Callicott, J. H., Egan, M. F., Bertolino, A., et al. (1998). Hippocampal N-acetyl aspartate in
unaffected siblings of patients with schizophrenia: a possible intermediate neurobiological
phenotype. Biol Psychiatry 44: 941–950. [Erratum in Biol Psychiatry (1999). 45: 244.]
Cantor-Graae, E., Pedersen, C. B., McNeil, T. F., Mortensen, P. B. (2003). Migration as a risk factor
for schizophrenia: a Danish population-based cohort study. Br J Psychiatry 182: 117–122.
Carter, J. W., Parnas, J., Cannon, T. D., Schulsinger, F., Mednick, S. A. (1999). MMPI variables
predictive of schizophrenia in the Copenhagen High-risk Project: a 25-year follow-up. Acta
Psychiatr Scand 99: 432–440.
Chapman, L. J., Chapman, J. P., Raulin, M. L. (1976). Scales for physical and social anhedonia.
J Abnorm Psychol 85: 374–382.
(1978). Body-image aberration in schizophrenia. J Abnorm Psychol 87: 399–407.
Chapman, L. J., Chapman, J. P., Kwapil, T. R., Eckblad, M., Zinser, M. C. (1994). Putatively
psychosis-prone subjects 10 years later. J Abnorm Psychol 103: 171–183.
Cornblatt, B., Erlenmeyer-Kimling, L. (1989). Attention and schizophrenia. Schizophr Res 2: 58.
449 High-risk studies, brain development, and schizophrenia

Cornblatt, B. A., Keilp, J. G. (1994). Impaired attention, genetics, and the pathophysiology of
schizophrenia. Schizophr Bull 20: 31–46.
Cornblatt, B., Obuchowski, M. (1997). Update of high risk research: 1987–1997. Int Rev Psychiatry
9: 437–447.
David, A. S., Malmberg, A., Brandt, L., Allebeck, P., Lewis, G. (1997). IQ and risk for schizophrenia:
a population-based cohort study. Psychol Med 27: 1311–1323.
Davidson, M., Reichenberg, A., Rabinowitz, J., et al. (1999). Behavioral and intellectual markers
for schizophrenia in apparently healthy male adolescents. Am J Psychiatry 156: 1328–1335.
Di Maggio, C., Martinez, M., Menard, J. F., Petit, M., Thibaut, F. (2001). Evidence of a cohort
effect for age at onset of schizophrenia. Am J Psychiatry 158: 489–492.
Done, D. J., Crow, T. J., Johnstone, E. C., Sacker, A. (1994). Childhood antecedents of schizophre-
nia and affective illness: social adjustments at ages 7 and 11. Br Med J 309: 699–703.
Dworkin, R. H., Cornblatt, B. A., Friedmann, R., et al. (1993). Childhood precursors of affective
vs. social deficits in adolescents at risk for schizophrenia. Schizophr Bull 19: 563–577.
Dykes, K. L., Mednick, S. A., Machon, R. A., Praestholm, J., Parnas, J. (1992). Adult third ventricle
width and infant behavioral arousal in groups at high and low risk for schizophrenia.
Schizophr Res 7: 13–18.
Erlenmeyer-Kimling, L. (2000). Neurobehavioral deficits in offspring of schizophrenic parents:
liability indicators and predictors of illness. Am J Med Genet 97: 65–71.
Erlenmeyer-Kimling, L., Squires-Wheeler, E., Hilldoff-Adamo, U. H., et al. (1995). The New York
High-risk Project. Psychoses and cluster A personality disorders in offspring of schizophrenic
parents at 23 years of follow-up. Arch Gen Psychiatry 52: 857–865.
Feinberg, I. (1982). Schizophrenia and late maturational brain changes in man. Psychopharmacol
Bull 18: 29–31.
Fish, B. (1984). Characteristics and sequelae of the neurointegrative disorder in infants at risk for
schizophrenia: 1952–1982. In Children at Risk for Schizophrenia: A Longitudinal Perspective,
ed. N. F. Watt, E. J. Anthony, L. C. Wynne, J. E. Rolf. New York: Cambridge University Press,
pp. 423–439.
Fish, B., Marcus, J., Hans, S. L., Auerbach, J. G., Perdue, S. (1992). Infants at risk for schizophrenia:
sequelae of a genetic neurointegrative defect. A review and replication analysis of pandys-
maturation in the Jerusalem infant development study. Arch Gen Psychiatry 49: 221–
235.
Friedman, D., Squires-Wheeler, E. (1994). Event-related potentials (ERPs) as indicators for risk
for schizophrenia. Schizophr Bull 20: 63–74.
Garver, D. L. (1987). Methodological issues facing the interpretation of high-risk studies:
biological heterogeneity. Schizophr Bull 13: 525–529.
Goldstein, J. M., Seidman, L. J., Buka, S. L., et al. (2000). Impact of genetic vulnerability and
hypoxia on overall intelligence by age 7 in offspring at high risk for schizophrenia compared
with affective psychoses. Schizophr Bull 26: 323–334.
Gooding, D. C., Iacono, W. G. (1995). Schizophrenia through the lens of a developmental
psychopathology perspective. In Developmental Psychopathology: Risk, Disorder, and
Adaptation, Vol. 2, ed. D. Cicchetti, D. J. Cohen. New York: Wiley, pp. 535–580.
Gottesman, I. I. (1991). Schizophrenia Genesis: The Origins of Madness. New York: W. H. Freeman.
450 M. S. Keshavan

Gottesman, I. I., Erlenmeyer-Kimling, L. (2001). Family and twin strategies as a head start in defin-
ing prodromes and endophenotypes for hypothetical early-interventions in schizophrenia.
Schizophr Res 51: 93–102.
Gottesman, I. I., Shields, J. (1982). Schizophrenia: The Epigenetic Puzzle. New York: Cambridge
University Press.
Gunnell, D., Harrison, G., Rasmussen, F., Fouskakis, D., Tynelius, P. (2002). Associations between
premorbid intellectual performance, early-life exposures and early-onset schizophrenia.
Cohort study. Br J Psychiatry 181: 298–305.
Hafner, H. (1990). New perspectives in the epidemiology of schizophrenia. In Search for the
Causes of Schizophrenia, Vol. II, ed. H. Hafner, W. F. Gattaz. Berlin: Springer-Verlag,
pp. 408–431.
Hanson, D. R., Gottesman, I. I., Heston, L. L. (1990). Long-range schizophrenia forecasting: many
a slip twixt cup and lip. In Risk and Protective Factors in the Development of Psychopathology,
ed. J. Rolf, A. Masten, D. Cicchetti, K. Nuechterlein, S. Weintraub. New York: Cambridge
University Press.
Harrison, P. J., Owen, M. J. (2003). Genes for schizophrenia? Recent findings and their patho-
physiological implications. Lancet 361: 417–419.
Hollister, J. M., Mednick, S. A., Brennan, P., Cannon, T. D. (1994). Impaired autonomic ner-
vous system-habituation in those at genetic risk for schizophrenia. Arch Gen Psychiatry 51:
552–558.
Johnstone, E. C., Russell, K. D., Harrison, L. K., Lawrie, S. M. J. (2003). The Edinburgh High
Risk Study: current status and future prospects. World Psychiatry 2: 45–49.
Jones, P. B., Tarrant, C. J. (1999). Specificity of developmental precursors to schizophrenia and
affective disorders. Schizophr Res 39: 121–125.
Jones, P., Rodgers, B., Murray, R., Marmot, M. (1994). Child developmental risk factors for adult
schizophrenia in the British 1946 birth cohort. Lancet 344: 1398–1402.
Josiassen, R. C., Shagass, C., Roemer, R. A., Straumanis, J. J. (1985). Attention-related effects
on somatosensory evoked potentials in college students at high risk for psychopathology.
J Abnorm Psychiatry 94: 507–518.
Kemppainen, L., Veijola, J., Jokelainen, J., et al. (2001). Birth order and risk for schizophrenia:
a 31-year follow-up of the Northern Finland 1966 Birth Cohort. Acta Psychiatr Scand 104:
148–152.
Kendler, K. S. (2002). Hierarchy and heritability: the role of diagnosis and modeling in psychiatric
genetics. Am J Psychiatry 159: 515–518.
Keshavan, M. S., Hogarty, G. E. (1999). Brain maturational processes and delayed onset in
schizophrenia. Dev Psychopathol 11: 525–543.
Keshavan, M. S., Schooler, N. R. (1992). First-episode studies of schizophrenia: criteria and
characterization. Schizophr Bull 18: 491–513.
Keshavan, M. S., Montrose, D. M., Pierri, J. N., et al. (1997). Magnetic resonance imaging and
spectroscopy in offspring at risk for schizophrenia: preliminary studies. Prog Neuropsy-
chopharmacol Biol Psychiatry 21: 1285–1295.
Keshavan, M. S., Dick, E., Mankowski, I., et al. (2002a). Decreased left amygdala and hippocampal
volumes in young offspring at risk for schizophrenia. Schizophr Res 58: 173–183.
451 High-risk studies, brain development, and schizophrenia

Keshavan, M. S., Diwadkar, V. A., Spencer, S. M., et al. (2002b). A preliminary functional magnetic
resonance imaging study in offspring of schizophrenic parents. Prog Neuropsychopharmacol
Biol Psychiatry 26: 1143–1149.
Keshavan, M. S., Stanley, J. A., Montrose, D. M., Minshew, N. J., Pettegrew, J. W. (2003). Pre-
frontal membrane phospholipid metabolism of child and adolescent offspring at risk for
schizophrenia or schizoaffective disorder: an in vivo (31)P MRS study. Mol Psychiatry 8:
316–323.
Kim-Cohen, J., Caspi, A., Moffitt, T. E., et al. (2003). Prior juvenile diagnoses in adults with
mental disorder: developmental follow-back of a prospective-longitudinal cohort. Arch Gen
Psychiatry 60: 709–717.
Klemm, S., Rzanny, R., Riehemann, S., et al. (2001). Cerebral phosphate metabolism in first-
degree relatives of patients with schizophrenia. Am J Psychiatry 158: 958–960.
Klosterkotter, J., Hellmich, M., Steinmeyer, E. M., Schultze-Lutter, F. (2001). Diagnosing
schizophrenia in the initial prodromal phase. Arch Gen Psychiatry 58: 158–164.
Kremen, W. S., Seidman, L. J., Pepple, J. R., et al. (1994). Neuropsychological risk indicators for
schizophrenia: a review of family studies. Schizophr Bull 20: 103–119.
Kwapil, T. R. (1998). Social anhedonia as a predictor of the development of schizophrenia-
spectrum disorders. J Abnorm Psychol 107: 558–565.
Kwapil, T. R., Miller, M. B., Zinser, M. C., Chapman, J., Chapman, L. J. (1997). Magical ideation
and social anhedonia as predictors of psychosis proneness: a partial replication. J Abnorm
Psychol 106: 491–495.
Lawrie, S. M., Whalley, H., Kestelman, J. N., et al. (1999). Magnetic resonance imaging of
brain in people at high risk of developing schizophrenia. [See comments] Lancet 353:
30–33.
Lawrie, S. M., Byrne, M., Miller, P., et al. (2001). Neurodevelopmental indices and the develop-
ment of psychotic symptoms in subjects at high risk of schizophrenia. Br J Psychiatry 178:
524–530.
Leask, S. J., Done, D. J., Crow, T. J. (2002). Adult psychosis, common childhood infections and
neurological soft signs in a national birth cohort. Br J Psychiatry 181: 387–392.
Lencz, T., Raine, A., Scerbo, A., et al. (1993). Impaired eye tracking in undergraduates with
schizotypal personality disorder. Am J Psychiatry 150: 152–154.
Lencz, T., Cornblatt, B., Bilder, R. M. (2001). Neurodevelopmental models of schizophrenia:
pathophysiologic synthesis and directions for intervention research. Psychopharmacol Bull
35: 95–125.
Lenzenweger, M. F. (1994). Psychometric high-risk paradigm, perceptual aberrations, and
schizotypy: an update. Schizophr Bull 20: 121–135.
Levy, D. L., Holzman, P. S., Matthysse, S., Mendell, N. R. (1994). Eye tracking and schizophrenia:
a selective review. Schizophr Bull 20: 47–62.
Maier, W., Franke, P., Hain, C., Kipp, B., Rist, F. (1992). Neuropsychological indicators of
the vulnerability to schizophrenia. Prog Neuropsychopharmacol Biol Psychiatry 16: 703–
715.
Marcus, J., Hans, S. L., Nagler, S., et al. (1987). Review of the NIMH Israeli Kibbutz-City Study
and the Jerusalem Infant Development Study. Schizophr Bull 13: 425–438.
452 M. S. Keshavan

Marcus, J., Hans, S. L., Auerbach, J. G., Auerbach, A. G. (1993). Children at risk for schizophrenia:
the Jerusalem Infant Development Study. II. Neurobehavioral deficits at school age. Arch
Gen Psychiatry 50: 797–809.
McGlashan, T. H. (1998). Early detection and intervention of schizophrenia: rationale and
research. Br J Psychiatry 172: 3–6.
McGorry, P. D. (1998). “A stitch in time” . . . the scope for preventive strategies in early psychosis.
Eur Arch Psychiatry Clin Neurosci 248: 22–31.
McGorry, P. D., Yung, A. R., Phillips, L. J., et al. (2002). Randomized controlled trial of inter-
ventions designed to reduce the risk of progression to first-episode psychosis in a clinical
sample with subthreshold symptoms. Arch Gen Psychiatry 59: 921–928.
McGuffin, P., Farmer, A. E., Gottesman, I. I., Murray, R. M., Reveley, A. M. (1984). Twin concor-
dance for operationally defined schizophrenia. Confirmation of familiality and heritability.
Arch Gen Psychiatry 41: 541–545.
McNeil, T. F., Harty, B., Blennow, G., Cantor-Graae, E. (1993). Neuromotor deviation in offspring
of psychotic mothers: a selective developmental deficiency in two groups of children at
heightened psychiatric risk? J Psychiatr Res 27: 39–54.
Mednick, S. A., Parnas, J., Schulsinger, F. (1987). The Copenhagen High-risk Project, 1962–1986.
Schizophr Bull 13: 485–495.
Michie, P. T., Kent, A., Stienstra, R., et al. (2000). Phenotypic markers as risk factors in
schizophrenia: neurocognitive functions. Aust NZ J Psychiatry 34(Suppl): S74–S85.
Miller, P., Byrne, M., Hodges, A., et al. (2002). Schizotypal components in people at high risk of
developing schizophrenia: early findings from the Edinburgh High Risk Study. Br J Psychiatry
180: 179–184.
Mirsky, A. F., Kugelmass, S., Ingraham, L. J., Frenkel, E., Nathan, M. (1995). Overview
and summary: twenty-five year follow-up of high-risk children. Schizophr Bull 21: 227–
239.
Moldin, S. O., Gottesman, I. I., Rice, J., Erlenmeyer-Kimling, L. (1991). Replicated psychometric
correlates of schizophrenia. Am J Psychiatry 148: 762–767.
Murray, R. M., Lewis, S. W. (1987). Is schizophrenia a neurodevelopmental disorder? [Editorial]
Br Med J (Clin Res Ed) 295: 681–682.
Nasrallah, H. A., Tolbert, H. A. (1997). Neurobiology and neuroplasticity in schizophrenia.
Continuity across the life cycle. [Comment] Arch Gen Psychiatry 54: 913–914.
Niemi, L. T., Suvisaari, J. M., Tuulio-Henriksson, A., Lonnqvist, J. K. (2003). Childhood develop-
mental abnormalities in schizophrenia: evidence from high-risk studies. Schizophr Res 60:
239–258.
Nuechterlein, K. H., Dawson, M. E., Gitlin, M., et al. (1992). Developmental processes in
schizophrenic disorders: longitudinal studies of vulnerability and stress. Schizophr Bull 18:
387–425.
Parnas, J., Cannon, T. D., Jacobsen, B., et al. (1993). Lifetime DSM-III-R diagnostic outcomes
in the offspring of schizophrenic mothers. Results from the Copenhagen High-risk Study.
Arch Gen Psychiatry 50: 707–714.
Pearson, J. S., Kley, I. B. (1957). On the application of genetic expectancies as age specific base
rates in the study of human behavior disorders. Psychol Bull 54: 406–420.
453 High-risk studies, brain development, and schizophrenia

Poulton, R. A. C., Moffitt, T. E., Cannon, M., Murray, R., Harrington, H. (2000). Children’s self-
reported psychotic symptoms and adult schizophreniform disorder: a 15-year longitudinal
study. Arch Gen Psychiatry 57: 1053–1058.
Rantakallio, P., Jones, P. B., Moring, J., von Wendt, L. (1997). Association between central nervous
system infections during childhood and adult onset schizophrenia and other psychoses:
a 28-year follow-up. Int J Epidemiol 26: 837–843.
Reveley, A. M., Reveley, M. A., Clifford, C. A., Murray, R. M. (1982). Cerebral ventricular size in
twins discordant for schizophrenia. Lancet i: 540–541.
Ross, R. G. (2003). Early expression of a pathophysiological feature of schizophrenia: saccadic
intrusions into smooth-pursuit eye movements in school-age children vulnerable to
schizophrenia. J Am Acad Child Adolesc Psychiatry 42: 468–476.
Saitoh, O., Niwa, S., Hiramatsu, K., et al. (1984). Abnormalities in late positive components of
event-related potentials may reflect genetic predisposition to schizophrenia. Biol Psychiatry
19: 293–303.
Sarfati, Y., Hardy-Bayle, M. C. (2002). Could cognitive vulnerability identify high-risk subjects
for schizophrenia? Am J Med Genet 114: 893–897.
Schreiber, H., Stolz, G., Rothmeier, J., Kornhuber, H. H., Born, J. (1989). Prolonged latencies of
the N2 and P3 of the auditory event-related potential in children at risk for schizophrenia.
A preliminary report. Eur Arch Psychiatry Neurol Sci 238: 185–188.
Schreiber, H., Baur-Seack, K., Kornhuber, H. H., et al. (1999). Brain morphology in adolescents
at genetic risk for schizophrenia assessed by qualitative and quantitative magnetic resonance
imaging. Schizophr Res 40: 81–84.
Seidman, L. J., Faraone, S. V., Goldstein, J. M., et al. (2002). Left hippocampal volume as a
vulnerability indicator for schizophrenia: a magnetic resonance imaging morphometric
study of nonpsychotic first-degree relatives. Arch Gen Psychiatry 59: 839–849.
Siever, L. J. (1994). Biologic factors in schizotypal personal disorders. Acta Psychiatr Scand 384:
45–50.
Suddath, R. L., Christison, G. W., Torrey, E. F., Casanova, M. F., Weinberger, D. R. (1990).
Anatomical abnormalities in the brains of monozygotic twins discordant for schizophrenia.
N Engl J Med 322: 789–794.
Tienari, P., Wynne, L. C., Moring, J., et al. (1994). The Finnish adoptive family study of
schizophrenia: implications for family research. Br J Psychiatry 164: 20–26.
Venables, P. H., Mednick, S. A., Schulsinger, S. F., et al. (1978). Screening for risk in mental
illness. In Cognitive Defects in the Development of Mental Illness, ed. G. M. Serban. New York:
Brunner/Mazel, pp. 273–303.
Verdoux, H., Sutter, A. L. (2002). Perinatal risk factors for schizophrenia: diagnostic specificity
and relationships with maternal psychopathology. Am J Med Genet 114: 898–905.
Von Bertalanffy, K. (1968). General Systems Theory. New York: Brazilier.
Wahlbeck, K., Forsen, T., Osmond, C., Barker, D. J., Eriksson, J. G. (2001). Association of
schizophrenia with low maternal body mass index, small size at birth, and thinness during
childhood. Arch Gen Psychiatry 58: 48–52.
Walker, E., Lewine, R. (1990). Prediction of adult onset schizophrenia from childhood movies of
patients. Am J Psychiatry 147: 1052–1056.
454 M. S. Keshavan

Walker, E., Grimes, K., Davis, D., Smith, A. (1993). Childhood precursors of schizophrenia: facial
expressions of emotion. Am J Psychiatry 150: 1654–1660.
Weinberger, D. R. (1987). Implications of normal brain development for the pathogenesis of
schizophrenia. Arch Gen Psychiatry 44: 660–669.
Weinberger, D. R., DeLisi, L. E., Neophytides, A. N., Wyatt, R. J. (1981). Familial aspects of CT
scan abnormalities in chronic schizophrenic patients. Psychiatry Res 4: 65–71.
Weiser, M., Reichenberg, A., Rabinowitz, J., et al. (2001). Association between nonpsychotic
psychiatric diagnoses in adolescent males and subsequent onset of schizophrenia. Arch Gen
Psychiatry 58: 959–964.
Wynne, L. C., Cole, R. E., Perkins, P. (1987). University of Rochester child and family study: risk
research in progress. Schizophr Bull 13: 463–476.
Yung, A. R., Phillips, L. J., McGorry, P. D., et al. (1998). Prediction of psychosis. A step towards
indicated prevention of schizophrenia. Br J Psychiatry Suppl 172: 14–20.
Yung, A. R., Phillips, L. J., Yuen, H. P., et al. (2003). Psychosis prediction: 12-month follow up of
a high-risk (“prodromal”) group. Schizophr Res 60: 21–32.
 24

Developmental models and

hypothesis-driven early interventions
in schizophrenia
Matcheri S. Keshavan1 and Barbara A. Cornblatt2
1
University of Pittsburgh School of Medicine, Pittsburgh and
Wayne State University School of Medicine, Detroit, USA
2
Recognition and Prevention Program, Lake Success, New York, USA

As evident by several chapters in this volume, schizophrenia research in recent

decades has been dominated by neurodevelopmental theories. During recent years,
at least three types of pathophysiological models have been proposed, those that
posit altered pre- or perinatal brain development (Murray and Lewis, 1987; Wein-
berger, 1987), those proposing peri-adolescent developmental abnormalities (Fein-
berg, 1982–83; Keshavan et al., 1994; McGlashan and Hoffman, 2000), and those
that argue for neuronal degenerative processes after illness onset (Lieberman et al.,
1997; Ch. 20). There is clearly a need for a unifying hypothesis. In this chapter,
the lines of evidence from both clinical observations and neurobiological research
leading to the three seemingly conflicting models are reviewed, and an alternative
model that potentially integrates all three is outlined. We will also discuss the pos-
sible remediative and preventive treatment options suggested by the three current
pathophysiological models, and review the data that have emerged so far from the
prevention programs generated by these models (Table 24.1). We end by describing
the way in which the unifying model newly proposed here provides an integrated
theoretical foundation for prevention and early intervention in schizophrenia.

Clinical “facts” about the early course of schizophrenia

From the clinical perspective, consistent observations about the developmental
course of schizophrenia can be viewed to support, at least in part, each of three
currently competing pathophysiological models.

The premorbid phase is characterized by cognitive impairments

Clinical, genetic high-risk, and epidemiological studies report premorbid cognitive,
academic, and social deficits that are typically quite subtle, yet, in many cases,
Neurodevelopment and Schizophrenia, ed. Matcheri S. Keshavan et al. Published by Cambridge University
Press.

C Cambridge University Press 2004.

455
456 M. S. Keshavan and B. A. Cornblatt

Table 24.1. Phases of early schizophrenia outlining the clinical features, pathophysiology, etiology, and
hypothesis-based preventive interventions

Phase of illness Premorbid phase Prodromal phase Post-illness onset

deterioration

Characteristic Impairments in cognitive Gradual progression of Further functional decline

clinical features and neuromotor cognitive, affective, and possible
functions and social social, and educational progression in
skills difficulties, and cognitive impairments
subthreshold and negative symptoms
psychotic-like
symptoms
Proposed Neuronal proliferation or Neuronal pruning/ Excitoxicity (possibly
pathophysiology migration? myelination processes; glutamatergically
Glutamatergic tonic glutamate/ mediated) and
hypofunction dopamine oxidative stress
hypofunction (early
prodrome) and phasic
excess (late prodrome)
Likely etiologic Genetic factors; early Psychosocial stress; Relapses; psychosocial
factors brain adversity hormonal changes; stress; substance
(pre/perinatal drug use and misuse misuse
complications, viruses,
malnutrition, maternal
drug use), abuse/
neglect, parental age
Strategies for Primary (universal or Secondary (indicated) Tertiary prevention:
preventive selective) prevention: prevention: cognitive relapse prevention;
intervention early detection, remediation, Omega-3 adherence promotion;
counseling re risk fatty acids (early cognitive remediation;
factors for high-risk prodrome); CBT, Omega-3 fatty acid
parents; improved low-dose supplementation;
pre/prenatal care antipsychotics, glutamatergic
antianxiety, modulators,
antidepressant, mood (ampakenes, glycine,
stabilizer of memantine)
antipsychotic
medications; glutamate
agonists (late
prodrome)

CBT, cognitive behavioral therapy.

457 Developmental models and early interventions

date back to early development (Done et al., 1994; Jones and Cannon, 1998). In
particular, impaired cognition in areas such as attention and working memory
appear to be early enduring core characteristics of schizophrenia, which support
the involvement of pre- or perinatal brain abnormalities (Cornblatt and Malhotra,
2001; Elvevag and Goldberg, 2000; Erlenmeyer-Kimling et al., 1995). These early
deficits have been increasingly viewed as signs of an underlying vulnerability to
illness that is dormant clinically (in terms of psychosis) but, nevertheless, suggests
a fulminating process that will eventually lead to schizophrenia. This is one of
the primary justifications for early intervention, since it is currently believed that
preemptive treatment may stop, or at least reduce, the emergence of psychosis.

The prodromal phase is characterized by emergence of subthreshold symptoms

When they do begin to emerge in schizophrenia, psychotic symptoms have a charac-
teristic onset in late adolescence or early adulthood (Hafner et al., 1993). However,
recent research has demonstrated that, for many patients, there is a developmen-
tal period between the premorbid and psychotic phases, currently widely referred
to as schizophrenia “prodrome.” During this phase, which can last from months
to years, some premorbid deficits appear to progress (e.g. school and social dif-
ficulties) while, at the same time, new clinical symptoms begin to emerge, most
notably subthreshold (or “attenuated”) positive symptoms. This pattern of evolving
difficulties is consistent with a peri-adolescent pathogenesis. At present, premor-
bid deficits cannot be identified with sufficient accuracy to attempt preventive
intervention. However, based on the early intervention findings reported since the
mid-1990s, it is now widely believed that prodromal signs and symptoms can be
reliably identified and treated effectively, with the expectation that this will prevent,
or at minimum contain, the progression into florid schizophrenia.

Early psychosis may be associated with further functional decline

There appears to be some deterioration during the early course of schizophrenia, at
least in a subgroup of patients (McGlashan and Fenton, 1993). While the positive
symptoms appear to stabilize after the first year of illness (Eaton et al., 1995),
primary negative symptoms appear to progress during the first 5 years of the illness
(Fenton and McGlashan, 1994). Overall functional ability also appears to worsen
during the first few years of the illness (Ch. 20).

Clinical observations can inform us about pathophysiology

The “early” developmental model and premorbid alterations
The well-known premorbid abnormalities in schizophrenia (reviewed by Haas
and Sweeney, 1992) have suggested that this illness may have its roots early in
458 M. S. Keshavan and B. A. Cornblatt

development. The observed brain abnormalities in schizophrenia have been

thought to stem primarily from one or other causal factors intra- or perinatally,
perhaps during the second half of gestation. In these early neurodevelopmental
models (Murray and Lewis, 1987; Weinberger, 1987), a fixed lesion from early
life interacts with normal brain maturation occurring later. Consistent with this
view, neuropathological observations suggest altered cytoarchitecture indicative
of possible early developmental errors in neural genesis or migration (Akbarian
et al., 1993; Kovelman and Scheibel, 1984; Weinberger, 1995). High-risk studies
have shown neurointegrative and cognitive deficits in preschizophrenia children
dating back to early childhood (Erlenmeyer-Kimling and Cornblatt, 1992; Fish,
1987). Additionally, birth cohort studies have found that children and adolescents
who developed schizophrenia at a later date had several characteristic abnormali-
ties such as poorer intellectual and social functioning and language deficits while
young. Regional brain structural (Ch. 19) and metabolic abnormalities as shown
by magnetic resonance spectroscopy (MRS) studies (Keshavan et al., 2003) have
also been seen in young relatives of schizophrenia patients.

The “late” developmental model and the peri-adolescent beginning of illness

Based on the typical adolescent onset of schizophrenic illness, an alternative view
has been proposed that the pathophysiology of this disorder has its onset in post-
natal life. Based on data that indicate substantial changes in brain biology during
adolescence, Feinberg (1982–83, 1990) initially proposed that schizophrenia may
result from an abnormality in peri-adolescent synaptic pruning, though this model
did not clarify whether too much, too little, or the wrong synapses were being
pruned.
Adolescence is characterized by substantive changes in several in vivo neuro-
biological measures that may indirectly reflect changes in synapse density. Delta
sleep, which represents the summed postsynaptic potentials in large assemblies of
cortical and subcortical axons and dendrites, dramatically decreases during adoles-
cence (Feinberg, 1982–83). Peri-adolescent reductions are also seen in synthesis of
membrane phospholipids, as measured by phosphorus-3 magnetic resonance MRS
studies (Pettegrew et al., 1991), cortical gray matter volume (Jernigan and Tallal,
1990), and regional prefrontal metabolism (Chugani et al., 1987). In schizophre-
nia, similar, but more pronounced decrements are seen, relative to healthy con-
trols, in delta sleep (Keshavan et al., 1998a), membrane synthesis (Pettegrew et al.,
1991), gray matter volume (Zipursky, 1992), and prefrontal metabolism (Andreasen
et al., 1992). These observations indirectly suggest the possibility in schizophre-
nia of an exaggeration of the normative process of synaptic pruning in certain
brain regions such as prefrontal cortex (Pettegrew et al., 1997; Keshavan et al.,
1994).
459 Developmental models and early interventions

Neural network modeling studies also support the view that schizophrenia
may be associated with synaptic or axonal pruning, but not neuronal cell death
(Hoffman and McGlashan, 1997; McGlashan and Hoffman, 2000). Neuropatholog-
ical studies show reduction in the synapse-rich neuropil and a consequent increase
in cortical neuron density (Selemon et al., 1995). Reductions have also been reported
in the expression of synaptophysin, a synaptic marker (Eastwood and Harrison,
1995; Glantz and Lewis, 1997), and in the density of dendritic spines (Garey et al.,
1998; Ch. 17). These observations suggest an overall reduction of cortical synapse-
rich neuropil in schizophrenia. This may lead to reduced neuronal plasticity. The
net effect of reduced neuropil may, therefore, be a compromised neuronal capacity
to modulate the normal academic, familial, and interpersonal demands of adoles-
cence. When a critical threshold of such neuropil loss is exceeded, symptoms and
signs of schizophrenia may appear.

The neurodegenerative model and deterioration after the onset of illness

Kraepelin (1919) suggested that schizophrenia may be associated with acquired
neurodegenerative changes beginning in early adulthood, a view that led to the
term dementia praecox. This view continues to receive support from several lines
of clinical evidence (see also Ch. 20). First, some, though not all, schizophrenia
patients may have a deteriorating rather than static course, at least during the first
few years of their illness (Loebel et al., 1992; McGlashan and Fenton, 1993). More-
over, patients typically appear to take longer to recover and show less-complete
recovery after each successive episode of illness (Lieberman et al., 1996). Second,
early treatment with antipsychotic drugs may arrest disease progression in some
schizophrenia patients (Wyatt, 1991). Third, prolonged untreated illness predicts a
poorer outcome, suggesting a possible “neurotoxic” effect of psychosis (Lieberman,
1993). Fourth, progressive structural alterations may be seen in the brains of some
schizophrenia patients in some studies (DeLisi, 1995; Thompson et al., 2001),
though others show no change (Jaskiw et al., 1994) and still others show increases
in temporal lobe structures (Keshavan et al., 1998b; Lieberman et al., 2001). Alter-
ations in neurophysiological (P300 component of the cognitive evoked response
potentials) indices are also consistent with ongoing cerebral degeneration in a
significant subgroup of schizophrenia patients (Mathalon et al., 2000).

Pathophysiology may be explained by an integrative neurochemical model

The three models discussed above may seem to conflict on the surface but can be
reconciled. Studies of neurochemical and molecular mechanisms that regulate early
brain development, postnatal brain plasticity, and brain degeneration can poten-
tially help to understand the many complexities of schizophrenia. The glutamatergic
460 M. S. Keshavan and B. A. Cornblatt

system, which involves excitatory amino acid neurotransmitters, has recently

attracted considerable interest as a candidate mechanism that unifies the multi-
ple underlying pathophysiological processes. The glutamatergic hypothesis is sup-
ported by several lines of converging evidence, including (i) observations of similar-
ities between the clinical manifestations of schizophrenia and psychosis caused by
phencyclidine, a N-methyl-d-aspartate (NMDA) receptor antagonist (Coyle, 1996;
Javitt and Zukin, 1991; Tamminga, 1998); (ii) reductions in cerebrospinal fluid glu-
tamate levels in schizophrenia (Kim et al., 1980); (iii) altered glutamate metabolism
in schizophrenia (Tsai et al., 1995); and (iv) altered gene expression for NMDA
receptor subunits in affected individuals (Akbarian et al., 1996). This hypothe-
sis can be formulated in a “three-hit” model that incorporates all three phases
discussed above: the early, late developmental, and neurodegenerative processes of
illness. Furthermore, glutamatergic dysfunction ties together several known “facts”
about schizophrenia: the premorbid deficits seen in many patients, the onset during
adolescence, the relatively enduring cognitive impairments and negative symptoms,
the “phasic” features of the illness such as psychoses, the role of dopamine, and the
deterioration in the early course of illness seen in some patients (Keshavan, 1999).
We discuss these aspects of the model below.
Glutamate is the predominant excitatory neurotransmitter in the mammalian
brain. Glutamate is now known to be critical for neuronal migration and neuronal
survival during early development, for neuronal plasticity during adolescence, and
for neuronal excitability and viability throughout life. Therefore, acting via NMDA
receptors, glutamate is likely to be essential for normal neuronal migration in early
development (Behar et al., 1999). The immature brain may be selectively vulnerable
to NMDA-mediated excitotoxicity occurring during hypoxia–ischemia and other
insults. Such insults during a critical developmental period (such as the second or
the third trimester of pregnancy) can lead to glutamatergic neuronal loss or receptor
hypofunction (Olney and Farber, 1995); this may generate dysplastic neural net-
works manifesting as premorbid brain abnormalities. In addition, sustained activity
of NMDA receptors is needed for synaptic survival. Absence of activity-dependent
synaptic change as a result of persistent hypofunction of NMDA receptors may
lead to excessive peri-adolescent synaptic pruning, hypothesized to underlie the
adolescent onset of schizophrenia (Keshavan, 1999).
Glutamate, especially the NMDA receptors, may also be critical for cognition
and working memory mechanisms (Goff and Coyle, 2001). Therefore, persist-
ent or “tonic” glutamatergic hypofunction may lead to the enduring cognitive
deficits characterizing the schizophrenic illness. Corticostriatal glutamatergic neu-
rons exert excitatory control over dopaminergic inputs to the prefrontal cortex and
tonically inhibit mesolimbic dopamine neurons. Glutamatergic hypotransmission
461 Developmental models and early interventions

could, therefore, decrease tonic prefrontal dopamine release, and increase phasic
stress-induced mesolimbic dopamine release (Moghaddam, 2002). A dysregula-
tion of the tonic–phasic dopamine system may account for the positive and negative
symptoms of schizophrenia (Grace, 1993).
Excess phasic release of glutamate during recurrent psychotic exacerbations can
cause excitotoxic cell or dendrite loss via NMDA receptor-mediated calcium influx
into cells, or oxidative stress (Goff and Coyle, 2001). This may explain the neurode-
teriorative changes that may occur in some patients with untreated schizophrenic
illness.
It may be seen from the above discussion that glutamate may play a role in
each of the characteristic components of schizophrenic illness: premorbid cogni-
tive deficits, adolescent onset of the illness, and early deterioration. Glutamater-
gic dysfunction may serve as a unitary explanation that links these observations
together. The schizophrenia syndrome may result from the cumulative effect of a
possibly genetically mediated hypoplasia in early and late maturational processes
of brain development interacting with adverse psychosocial factors during adoles-
cence and early adulthood and possible neurotoxic effects of untreated psychosis
early in the schizophrenic illness. The premorbid vulnerability to schizophrenia
may be caused by multiple genetic and environmental factors interacting to affect
early brain development. The adolescent onset of the disorder may be determined
by late brain maturational processes, as well as the stresses unique to adolescence.
It is likely that early brain developmental defects (e.g. neuromigrational errors or
defective neuronal proliferation) may predispose to excessive postnatal synaptic
pruning and/or neuronal apoptosis, leading to the schizophrenic endophenotype.
The development of a deficit state and of persistent psychopathology after illness
onset may be determined by the possible neurotoxic effect of continuing psychotic
illness.
Excitotoxic neuronal loss, interacting with genetic factors, may result from early
brain adversity (viruses, malnutrition, or perinatal trauma) and lead to selective
loss of NMDA receptor-bearing glutamatergic neurons in cortical and subcortical
structures. This would account for the premorbid neurocognitive abnormalities
seen among children and adolescents at risk for schizophrenia. The normative
synaptic pruning process that sets in around late childhood and adolescence may
interact with the state of reduced tonic glutamatergic neurotransmission, perhaps
acting via NMDA receptors, to result in a net excess of synapse elimination in
the cortical and subcortical structures. Reduced activity in the corticostriatal gluta-
matergic neurons would lead to diminished tonic cortical dopamine release because
of reduced mesocortical dopamine release, which comes “on line” during adoles-
cence. Recent data (Scott et al., 2002) have suggested that NMDA agonists may shift
462 M. S. Keshavan and B. A. Cornblatt

the balance of dopamine signaling toward the D1 receptor and away from the D2
receptors, by recruiting D1 receptors to the plasma membrane; this may account for
the therapeutic effects of glutamatergic agonists such as glycine and d-cycloserine
in improving negative symptoms and cognitive deficits in schizophrenia. The inter-
personal and cognitive deficits could lead to an inability to handle the psychosocial
stresses unique to adolescence and early adulthood (Keshavan and Hogarty, 1999).
Such stresses, in turn, may lead to an upregulation of the subcortical dopaminergic
neurons; the behavioral response to stress-induced phasic glutamate and dopamine
release is, therefore, likely to be exaggerated, leading to positive psychotic symptoms.
If untreated, persistent dopaminergic, and consequent phasic glutamatergic, excess
could lead to further excitotoxic brain damage, perhaps by increasing oxidative
stress.
Other similar, integrative models view schizophrenia as a lifetime disorder of
brain development, plasticity, and aging (DeLisi, 1997; Lieberman et al., 1997; Olney
and Farber, 1995). In this chapter, we attempt to extend these models by proposing
that the successive premorbid, morbid, and post-onset pathology of schizophrenia
results from a glutamatergic dysfunction at critical windows of vulnerability during
early and late developmental phases, and the early course of this illness. Each of the
steps of dysfunction in glutamatergic functioning in this model could increase the
risk for the subsequent step, leading to a pathophysiological cascade.

Toward an understanding of etiological factors

Both biological and psychosocial factors might be involved in the causation of
the proposed developmental brain abnormalities. In regard to early premorbid
alterations, epidemiological data suggest associations between intrauterine or peri-
natal brain insult (Ch. 11), exposure to viral infection (Mednick et al., 1988),
early malnutrition (Susser and Lin, 1992; see Ch. 9, this volume), and subsequent
emergence of schizophrenia. The etiologic factors underlying peri-adolescent brain
developmental deviations are less clear. Genetic factors may clearly be involved, per-
haps involving genes regulating such maturational processes as myelination (Ch. 1
and the discussion of synaptic/neuronal pruning earlier in this chapter). In addition
to glutamate, other neurotransmitters such as gamma-aminobutyric acid (GABA)
or neurotrophic factors such as brain-derived neurotrophic factor may be involved
in disordered pruning processes. Pathology involving one or more of such genes
might lead to excessive or disordered pruning, or a premature loss of neuronal
plasticity. The psychosocial challenges initiated by the interpersonal and academic
demands of adolescence and young adulthood might interact with cognitive limi-
tations (especially social cognition) leading to the emergence of psychopathology
(Keshavan and Hogarty, 1999). The hormonal changes that characterize adolescence
463 Developmental models and early interventions

also take their toll. Finally, adolescence is the period when substance use becomes
increasingly prevalent and this may interact with pre-existing risk to lead to psy-
chopathology.
At least a subgroup of schizophrenia patients may have progressive brain deteri-
orative processes following illness onset. Proposed mechanisms for such neurode-
generation include oxidative stress (Coyle, 1996) or neurochemical sensitization
from repeated exposure to neurochemical stressors (Lieberman et al., 1997). How-
ever, one needs also to entertain the possibility that the observed neurobiological
changes during the course of the illness may be plastic adaptations of the nervous
system to being psychotic or cognitively impoverished (Weinberger and McClure,
2002).

Strategies for preventive intervention suggested by current

pathophysiological models
Therapeutic interventions in schizophrenia, which have historically been derived
from serendipitous observations, are being increasingly guided by recent patho-
physiological models such as those proposed here. Several specific predictions, as
outlined below, can be generated from these models and have already begun to yield
promising findings in the literature. We will organize preventive interventions into
the classic domains of primary, secondary, and tertiary preventions. We will also
consider preventive strategies under the three categories proposed by the widely
cited Institute of Medicine report: universal (targeting the general population),
selective (targeting subject groups selected by the presence of one or more risk
factors) and indicated (preventing the illness in at-risk persons with symptoms but
not yet meeting diagnostic criteria) (Mrazek and Haggarty, 1994).

Premorbid phase: primary (universal or selective) prevention efforts

It is clear from the above discussion that, while schizophrenia begins in late adoles-
cence or early adulthood, the seeds of schizophrenia are planted early in a long-term
neurodevelopmental process eventually leading to deviant brain functioning. It is
also evident that multiple and sequential etiological factors may interactively and
additively contribute to the emergence of the illness. This view suggests that pre-
ventive treatments may be tailored to the stage of evolution of the disease processes
in individuals predisposed to the disorder. However, prevention studies must first
define their target population. The identification of risk factors is critical for accu-
rately selecting the at-risk individuals most appropriate for preventive treatment.
Considerable ongoing research is currently devoted to the identification and vali-
dation of such risk factors (Chs. 9–11).
464 M. S. Keshavan and B. A. Cornblatt

The role of universal (population-based) interventions designed to pre-

vent schizophrenia remains speculative. As reviewed elsewhere in this volume
(Chs. 9–11), the widely reported risk factors for schizophrenia include genetic
factors, season and place of birth, pregnancy and birth complications, antenatal
exposure to viruses, and nutritional anomalies. Vaccinations and interventions
related to improving prenatal nutrition and antenatal care appear to be plausible
options for universal prevention. However, such interventions are not cost effective
(Cuijpers, 2003; McGrath, 2000).
Selected preventive intervention targets a population or population subgroup
identified to be at risk for illness based on biological, psychological, or social diffi-
culties. At present, primary prevention initiated during the premorbid stage most
feasibly will target individuals at genetic risk for schizophrenia. This is because
the identification of such at-risk individuals is initially based on family history,
so can begin at any developmental stage and can precede the emergence of clini-
cal symptoms. Interventions for the families of genetically at-risk individuals are
possible at present and include genetic counseling, family planning, and prenatal
care for schizophrenia mothers. In the future, however, it will hopefully be possi-
ble to design interventions that are specific to those genetically at-risk individuals
who display neurobehavioral and neurobiological precursors found to be accurate
predictors of later illness by ongoing studies (e.g. Johnstone et al., 2003; Keshavan
et al., 2003). For example, a program for an adolescent at genetic risk for schizophre-
nia who also displays cognitive and social deficits might include improvement of
the home environment, prevention of abuse and neglect, a structured school envi-
ronment combined with cognitive remediation, and possibly social skills training
(Keshavan and Hogarty, 1999). While we do not have concrete evidence about the
effectiveness of these interventions to actually prevent schizophrenia, most have
been established to reduce suffering at an individual level (Christodolou, 1991)
and, therefore, have an intrinsic therapeutic value. Cuijpers (2003) has suggested
strategies that can potentially increase statistical power in such prevention studies:
(i) placing increased emphasis on high-incidence groups, those with multiple risk
factors, and those with multiple disorders; (ii) strengthening the effects of pre-
vention programs; and (iii) increased use of cumulative meta-analyses to improve
utilization of data from a large number of trials.

Prodromal phase: secondary (indicated) prevention strategies

Because the presence of attenuated positive symptoms is a primary criterion for
identifying the prodromal phase of schizophrenia, it is assumed that the process
leading to psychosis has already begun and secondary prevention will be the desir-
able approach. In the Mrazek and Haggarty (1994) classification, such approaches
465 Developmental models and early interventions

would be considered to represent an indicated prevention, since already evident

but not yet diagnosable symptoms are being addressed.

Low-dose atypical antipsychotic drugs

The view that the attenuated psychotic-like symptoms characterizing the prodrome
might be mediated by phasic dopaminergic excess has encouraged early treatment
with low doses of dopamine-blocking drugs. In a single blind design study in
Melbourne, Australia, McGorry et al. (2002) compared low-dose risperidone plus
cognitive behavioral therapy with a needs-based intervention (i.e. counseling and
case management) for 6 months. This was followed by a 6-month observation
period of all patients on needs-based therapy only. The combined specific inter-
vention led to a significant preventive treatment effect in a modest sized population
of 60 subjects. However, it was difficult to determine the relative contribution of
the pharmacological and non-pharmacological interventions for the prevention of
progression into frank psychosis. The differential effect of risperidone and cognitive
behavioral therapy for psychosis prevention is currently being investigated by the
Melbourne group in a controlled three-cell trial comparing risperidone, cognitive
behavioral therapy, and needs-based treatment.
There have been other, smaller trials that are beginning to yield data supportive
of early interventions with low-dose antipsychotic drugs. Using a double-blind,
multicenter approach McGlashan et al. (2003a) compared the efficacy of olanzapine
plus supportive and family therapy versus supportive and family therapy alone.
The study involved 1 year of active treatment (McGlashan, 2003b) followed by
another 12 months of observation. Focus was on change in prodromal and other
psychopathology ratings and rates of conversion to psychosis. Preliminary results
(Woods et al., 2003) suggest a modest improvement in symptoms associated with
olanzapine treatment, although final results have not yet been reported and the
potential for the prevention of psychosis remains unknown.
The German Schizophrenia Network has similarly used a 12 month active treat-
ment and 12 month observation phase. Using a randomized, open-label design,
amisulpiride was compared with psychological management for “late” prodromal
symptoms suggested by attenuated psychotic symptoms (Ruhrmann et al., 2002).
Cannon et al. (2002) used low-dose risperidone in a small open-label trial for
3 months with four prodromal patients, and six with first-episode schizophrenia
and have reported some preliminary positive results.

Naturalistic treatment studies

A different intervention strategy has been adopted by the Zucker Hillside Recogni-
tion and Prevention program, which has been ongoing in New York since 1998. In
this program, adolescents and young adults between the ages of 12 and 22 who are
466 M. S. Keshavan and B. A. Cornblatt

considered to be in the prodromal stage of schizophrenia participate in both treat-

ment and research studies (Cornblatt et al., 2002). Treatment is provided within
a naturalistic framework, in that clinicians (who are independent of the research
team) prescribe medication according to best-care standards. This has provided
essential observational data about the types and efficacy of medication currently
prescribed to treat prodromal symptoms, data that have otherwise not been avail-
able. Preliminary findings generated by this initial research strategy are consistent
with other prodromal studies in indicating that antipsychotic drugs, used either
alone or in conjunction with other psychotropic medications, are most effective
for late prodromal symptoms. However, unlike other studies, the data also indi-
cate that antidepressants may be an effective alternative to antipsychotic drugs for
individuals in the earlier prodromal stages, when attenuated positive symptoms are
more moderate. The non-randomized naturalistic design of the Recognition and
Prevention longitudinal study introduces many confounds. However, the results of
such observations suggest the need for more rigorous controlled studies. A formal
clinical trial is currently underway in this program to determine whether antide-
pressants do have a protective effect if initiated sufficiently early in the illness process
(Cornblatt et al., 2003).

Non-antipsychotic medications
Neurodevelopmental models, discussed above, have also suggested therapeutic
trials of drugs with alternative putative (non-dopamine-related) mechanisms of
action. Berger et al. (2004) have been conducting a 1 year trial with low-dose
lithium, the definitive results of which are expected soon. Using predictions
derived from the glutamatergic and membrane models of schizophrenia, respect-
ively, Woods et al. (2002) have conducted a 2 month trial of glycine and a 3 month
study with ethyl eicosopentaenoic acid, an omega-3 fatty acid. Outcome measures
have included psychopathology ratings and rates of conversion into psychosis, cog-
nitive assessments, neuroimaging, and in vitro measures of structural cortical and
cell membrane integrity. The results are awaited.

Psychotherapeutic treatments
It has been suggested (Ch. 20) that cognitive decline has largely occurred in
schizophrenia by the time of the first psychotic presentation. Therefore, a window
of opportunity exists during the early prodromal period, which is characterized
by non-specific behavioral signs and precedes the more schizophrenia-like fea-
tures of the late prodromal phase. The German Schizophrenia Network has been
conducting a 12 month non-pharmacological intervention of up to 26 sessions of
cognitive behavioral therapy over 6 months plus monitoring compared with moni-
toring alone in a group of prodromal patients (Morrison et al., 2002). Bechdolf and
467 Developmental models and early interventions

Wagner (2003) compared a multimodal psychotherapeutic intervention with clini-

cal management for symptoms suggestive of an “early” prodrome (genetic/obstetric
risk in combination plus functional decline as evidenced by a Global Assessment
of Functioning assessment fall of > 30 points) or “basic symptoms” deemed to be
below the level of attenuated positive symptomatology. Results of these trials are
awaited.

Strategies after onset of schizophrenia: toward tertiary prevention

As discussed above, treatments that modulate glutamatergic neurotransmission
may have a protective effect on the progression of the schizophrenia illness. Recent
studies have suggested the potential benefit of drugs reducing glutamate release,
such as lamotrigene; NMDA receptor antagonists such as memantine; and agents
that reduce the glutamate α-amino-3-hydroxy-5-methyl-4-isoxazole propionate
(AMPA) receptor activity, such as topiramate. Such potential treatments are worth
further exploration (Deutsch et al., 2001). The beneficial effects of atypical antipsy-
chotic drugs such as clozapine have been linked to their ability to modulate gluta-
matergic neurotransmission (Arnt and Skarsfeldt, 1998; Meltzer, 1994). Treatments
that seek to enhance glutamatergic neurotransmission, such as glycine, l-serine,
and d-cycloserine have also been found to be promising in the treatment of negative
and cognitive symptoms of schizophrenia (Heresco-Levy and Javitt, 1998).
Another unanswered question is whether early introduction of antioxidant treat-
ments can potentially mitigate the neurotoxic effects of oxidative stress in early
schizophrenia. Cognitive remediation studies have begun to provide encouraging
results in regard to improvements in functional outcome when introduced early
in the course of schizophrenia (Hogarty et al., unpublished data). Well-designed
and carefully thought out clinical trials are needed to investigate these questions
further.
In summary, the rationale for preventive interventions in schizophrenia goes
beyond face validity alone and is rapidly gaining acceptance in the field. However,
as our knowledge matures, it will be necessary to replicate the findings of the handful
of early prevention studies more extensively, validate the criteria currently defining
the prodrome more conclusively, extend the range of potential interventions further
to include treatments other than antipsychotic drugs, and target specific prodromal
and possibly premorbid phases of illness. Evidence-based demonstration of cost
effectiveness of such interventions are critically needed if this field of research is to
sustain and solidify this paradigm shift.

Acknowledgements
This work was supported in part by NIMH grants MH 45156 and MH 64023.
468 M. S. Keshavan and B. A. Cornblatt

REFERENCES

Akbarian, S., Bunney, W. E., Potkin, S. G., et al. (1993). Altered distribution of nicotamine-
adenine dinucleotide phosphate-diaphorase cells in frontal lobe of schizophrenics implies
disturbances of cortical development. Arch Gen Psychiatry 50: 169–177.
Akbarian, S., Sucher, N., Bradley, D., et al. (1996). Selective alterations in gene expression for
NMDA receptor subunits in prefrontal cortex of schizophrenics. J Neurosci 16: 19–30.
Andreasen, N., Rezai, K., Alliger, R., et al. (1992). Hypofrontality in neuroleptic-naive patients
with chronic schizophrenia. Assessment with xenon 133 single-photon emission computed
tomography and the Tower of London. Arch Gen Psychiatry 49: 943–958.
Arnt, J., Skarsfeldt, T. (1998). Do novel antipsychotics have similar pharmacological characteris-
tics? A review of the evidence. Neuropsychopharmacology 18: 63–101.
Bechdolf, A., Wagner, M. (2003). Cognitive behavioral therapy in the prepsychotic phase: an
exploratory study. Schizophr Res 60(Suppl.): 319.
Behar, T. N., Scott, C. A., Greene, C. L., et al. (1999). Glutamate acting at NMDA receptors
stimulates embryonic cortical neuronal migration. J Neurosci 19: 4449–4461.
Berger, G. E., Profitt, T. M., McConchie, M. A., et al. (2004). Ethyl-eicosapentaenoic acid (E-EPA)
supplementation in early psychosis, Schizophr Res 67: 7.
Cannon, T. D., Huttunen, M. O., Dahlstrom, M., et al. (2002). Antipsychotic drug treatment in
the prodromal phase of schizophrenia. Am J Psychiatry 159: 1230–1232.
Christodolou, G. N. (1991). Prevention of psychopathology with early interventions. Psychother
Psychosom 55: 201–207.
Chugani, H. T., Phelps, M. E., Mazziotta, J. C. (1987). Positron-emission tomography study of
human brain functional development. Ann Neurol 22: 487–497.
Cornblatt, B. A., Malhotra, A. K. (2001). Impaired attention as an endophenotype for molecular
genetic studies of schizophrenia. Am J Med Genet 105: 11–15.
Cornblatt, B., Lencz, T., Correll, C., Author, A., Smith, C. (2002). Treating the prodrome natu-
ralistic findings from the RAP Program. Acta Psychiatr Scand 106(S413): 44.
Cornblatt, B. A., Lencz, T., Smith, C. W., et al. (2003). The schizophrenia program revisited: a
neurodevelopmental perspective. Schizophr Bull 29: 633–651.
Coyle, J. T. (1996). The glutamatergic dysfunction hypothesis for schizophrenia. Harvard Rev
Psychiatry 3: 241–253.
Cuijpers, P. (2003). Examining the effects of prevention programs on the incidence of new cases
of mental disorders: the lack of statistical power. Am J Psychiatry 160: 1385–1391.
DeLisi, L. E. (1995). A prospective follow-up study of brain morphology and cognition in first-
episode schizophrenic patients: preliminary findings. Biol Psychiatry 38: 349–360.
(1997). Is schizophrenia a lifetime disorder of brain plasticity, growth, and aging? Schizophr
Res 23: 119–129.
Deutsch, S. I., Rosse, R. B., Schwartz, B. L., Mastropaolo, J. (2001). A revised excitotoxic hypothesis
of schizophrenia: therapeutic implications. Clin Neuropharmacol 24: 43–49.
Done, D. J., Crow, T. J., Johnstone, E. C., Sacker, A. (1994). Childhood antecedents of
schizophrenia and affective illness: social adjustment at ages 7 and 11. Br Med J 309: 699–
703.
469 Developmental models and early interventions

Eastwood, S. L., Harrison, P. J. (1995). Decreased synaptophysin in the medical temporal lobe in
schizophrenia demonstrated using immunoautoradiography. Neuroscience 69: 339–343.
Eaton, W. W., Thara, R., Federman, B., Melton, B., Liang, K. Y. (1995). Structure and course of
positive and negative symptoms in schizophrenia. Arch Gen Psychiatry 52: 127–134.
Elvevag, B., Goldberg, T. E. (2000). Cognitive impairment in schizophrenia is the core of the
disorder. Crit Rev Neurobiol 14: 1–21.
Erlenmeyer-Kimling, L., Cornblatt, B. A. (1992). A summary of attentional findings in the New
York High-risk Project. J Psychiatr Res 26: 405–426.
Erlenmeyer-Kimling, L., Squires-Wheeler, E., Hilldoff-Adamo, U. H., et al. (1995). The New York
High-risk Project. Psychoses and cluster A personality disorders in offspring of schizophrenic
parents at 23 years of follow-up. Arch Gen Psychiatry 52: 857–865.
Feinberg, I. (1982–83). Schizophrenia: caused by a fault in programmed synaptic elimination
during adolescence? J Psychiatr Res 17: 319–334.
(1990). Cortical pruning and the development of schizophrenia. Schizophr Bull 16: 567–570.
Fenton, W. S., McGlashan, T. H. (1994). Antecedents, symptom progression, and long-term
outcome of the deficit syndrome in schizophrenia. Am J Psychiatry 151: 351–356.
Fish, B. (1987). Infant predictors of the longitudinal course of schizophrenic development.
Schizophr Bull 13: 395–409.
Garey, L. J., Ong, W. Y., Patel, T. S., et al. (1998). Reduced dendritic spines density on cerebral
cortical pyramidal cells in schizophrenia. J Neurol Neurosurg Psychiatry 65: 446–453.
Glantz, L. A., Lewis, D. A. (1997). Reduction of synaptophysin immunoreactivity in the pre-
frontal cortex of subjects with schizophrenia: Regional and diagnostic specificity. Arch Gen
Psychiatry 54: 943–952.
Goff, D. C., Coyle, J. T. (2001). The emerging role of glutamate in the pathophysiology and
treatment of schizophrenia. Am J Psychiatry 158: 1367–1377.
Grace, A. (1993). Cortical regulation of subcortical dopamine systems and its possible relevance
to schizophrenia. J Neur Transm (Gen Sect) 91: 111–134.
Haas, G., Sweeney, J. (1992). Premorbid and onset features of first-episode schizophrenia.
Schizophr Bull 18: 373–386.
Hafner, H., Maurer, K., Loffler, W., Riecher-Rossler, A. (1993). The influence of age and sex on
the early course of schizophrenia. Br J Psychiatry 162: 80–86.
Heresco-Levy, U., Javitt, D. (1998). The role of N-methyl-d-aspartate (NMDA) receptor-mediated
neurotransmission in the pathophysiology and therapeutics of psychiatric syndromes. Eur
Neuropsychopharmacol 8: 141–152.
Hoffman, R., McGlashan, T. (1997). Synaptic elimination, neurodevelopment, and the mecha-
nism of hallucinated “voices” in schizophrenia. Am J Psychiatry 154: 1683–1689.
Jaskiw, G. E., Juliano, D. M., Goldberg,T. E., et al. (1994). Cerebral ventricular enlargement in
schizophreniform disorder does not progress. A seven year follow-up study. Schizophr Res
14: 23–28.
Javitt, D., Zukin, S. (1991). Recent advances in the phencyclidine model of schizophrenia. Am J
Psychiatry 148: 1301–1308.
Jernigan, T. L., Tallal, P. (1990). Late childhood changes in brain morphology observable with
MRI. Dev Med Child Neurol 32: 379–385.
470 M. S. Keshavan and B. A. Cornblatt

Johnstone, E. C., Russell, K. D., Harrison, L. K., Lawrie, S. M. J. (2003). The Edinburgh High
Risk Study: current status and future prospects. World Psychiatry 2: 45–49.
Jones, P., Cannon, M. (1998). The new epidemiology of schizophrenia. Psychiatr Clin North Am
21: 1–25.
Keshavan, M., Anderson, S., Pettegrew, J. (1994). Is schizophrenia due to excesive synaptic
pruning in the prefrontal cortex? The Feinberg hypothesis revisited. J Psychiatr Res 28:
239–265.
Keshavan, M. S. Development, disease and degeneration in schizophrenia: a unitary pathophys-
iological model. (1999). J Psychiatr Res 33: 513–521.
Keshavan, M. S., Hogarty, G. E. (1999). Brain maturational processes and delayed onset in
schizophrenia. Dev Psychopathol 11: 525–543.
Keshavan, M. S., Reynolds, C. F., Miewald, M. J., et al. (1998a). Delta sleep deficit in schizophrenia:
evidence from automated analyses of sleep data. Arch Gen Psychiatry 55: 443–448.
Keshavan, M., Haas, G., Kahn, C., et al. (1998b). Superior temporal gyrus and the course of early
schizophrenia: progressive, static, or reversible? J Psychiatr Res 32: 60–65.
Keshavan, M. S., Stanley, J. A., Montrose, D. M., Minshew, N. J., Pettegrew, J. W. (2003). Pre-
frontal membrane phospholipid metabolism of child and adolescent offspring at risk for
schizophrenia or schizoaffective disorder: an in vivo (31)P MRS study. Mol Psychiatry 8:
316–323.
Kim, J., Kornhuber, H., Schmid-Burgk, W., Holzmuller, B. (1980). Low cerebrospinal fluid glu-
tamate in schizophrenic patients and a new hypothesis on schizophrenia. Neurosci Lett 20:
379–382.
Kovelman, J., Scheibel, A. (1984). A neurohistological correlate of schizophrenia. Biol Psychiatry
19: 1601–1621.
Kraepelin, E. (1919). Dementia Praecox and Paraphrenia. Edinburgh: E. and S. Livingstone.
Lieberman, J. (1993). Prediction of outcome in first-episode schizophrenia. J Clin Psychiatry
54(Suppl.): 13–17.
Lieberman, J., Koreen, A., Chakos, M., et al. (1996). Factors influencing treatment response and
outcome of first-episode schizophrenia: implications for understanding the pathophysiology
of schizophrenia. J Clin Psychiatry 9: 5–9.
Lieberman, J., Sheitman, B., Kinon, B. (1997). Neurochemical sensitization in the pathophys-
iology of schiozphrenia: deficits and dysfunction in neuronal regulation and plasticity.
Neuropsychopharmacology 17: 205–229.
Lieberman, J., Chakos, M., Wu, H., et al. (2001). Longitudinal study of brain morphology in first
episode schizophrenia. Biol Psychiatry 49: 487–499.
Loebel, A. D., Liberman, J. A., Alvir, J. M. J. (1992). Duration of psychosis and outcome in
first-episode schiozphrenia. Am J Psychiatry 149: 1183–1188.
Mathalon, D. H., Ford, J. M., Rosenbloom, M., Pfefferbaum, A. (2000). P300 reduction and
prolongation with illness duration in schizophrenia. Biol Psychiatry 47: 413–427.
McGlashan, T., Fenton, W. (1993). Subtype progression and pathophysiologic deterioration in
early schizophrenia. Schizophr Bull 19: 71–84.
McGlashan, T. H., Hoffman, R. E. (2000). Schizophrenia as a disorder of developmentally reduced
synaptic connectivity. Arch Gen Psychiatry 57: 637–648.
471 Developmental models and early interventions

McGlashan, T. H., Zipursky R. B., Perkins, D., et al. (2003a). The PRIME North America ran-
domized double-blind clinical trial of olanzapine versus placebo in patients at risk of being
prodromally symptomatic for psychosis. I. Study rationale and design. Schizophr Res 61:
7–18.
(2003b). Olanzapine versus placebo treatment of the schizophrenia prodrome: one year results.
Schizophr Res 60(Suppl.): 295.
McGorry, P. D., Yung, A. R., Phillips, L. J., et al. (2002). Randomized controlled trial of inter-
ventions designed to reduce the risk of progression to first-episode psychosis in a clinical
sample with subthreshold symptoms. Arch Gen Psychiatry 59: 921–928.
McGrath, J. (2000). Universal interventions for the primary prevention of schizophrenia. Aust N
Z J Psychiatry 34(Suppl.): S58–S64.
Meltzer, H. (1994). An overview of the mechanism of action of clozapine. J Clin Psychiatry 55:
47–52.
Mednick, S. A., Machon, R. A., Huttunen, M. O. (1988). Adult schizophrenia following prenatal
exposure to an influenza epidemic. Arch Gen Psychiatry 45: 171–176.
Moghaddam, B. (2002). Stress activation of glutamate neurotransmission in the prefrontal cor-
tex: implications for dopamine-associated psychiatric disorders. Biol Psychiatry 51: 775–
787.
Morrison, A. P., Bentall, R. P., French, P., et al. (2002). Randomised controlled trial of early
detection and cognitive therapy for preventing transition to psychosis in high-risk individ-
uals. Study design and interim analysis of transition rate and psychological risk factors. Br
J Psychiatry Suppl 43: S78–S84.
Mrazek, P. J., Haggerty R. J. (1994). Reducing Risks for Mental Disorders: Frontiers for Preventive
Intervention Research. Washington, DC: National Academy Press.
Murray, R. M., Lewis, S. W. (1987). Is schizophrenia a neurodevelopmental disorder? [Editorial]
Br Med J (Clin Res Ed) 295: 681–682.
Olney, J., Farber, N. (1995). Glutamate receptor dysfunction and schizophrenia. Arch Gen Psy-
chiatry 52: 998–1007.
Pettegrew, J. W., Keshavan, M. S., Panchalingam, K., et al. (1991). Alterations in brain high-
energy phosphate and membrane phospholipid metabolism in first-episode, drug-naive
schizophrenics. Arch Gen Psychiatry 48: 563–568.
Pettegrew, J. W., McClure, R., Keshavan, M., et al. (1997). 31 P magnetic resonance spectroscopy
studies of developing brain. In Neurodevelopment and Adult Psychopathology, ed. M. Kesha-
van, R. Murray. New York: Cambridge University Press, pp. 71–92.
Ruhrmann, S., Wagner, M., Wieneke, A., et al. (2002). Intervention in the initial phase of psychosis.
Acta Psychiatr Scand 106(S413): 19.
Scott, L., Kruse, M. S., Forssberg, H., et al. (2002). Selective up-regulation of dopamine D1
receptors in dendritic spines by NMDA receptor activation. Proc Natl Acad Sci USA 99:
1661–1664.
Selemon, J., Rajkowska, G., Goldman-Rakic, P. (1995). Abnormally high neuronal density in the
schizophrenic cortex. Arch Gen Psychiatry, 52: 805–818.
Susser, E. S., Lin, S. P. (1992). Schizophrenia after prenatal exposure to the Dutch Hunger Winter
of 1944–1945. Arch Gen Psychiatry 49: 983–988.
472 M. S. Keshavan and B. A. Cornblatt

Tamminga, C. A. (1998). Schizophrenia and glutamatergic transmission. Crit Rev Neurobiol 12:
21–36.
Thompson, P. M., Vidal, C., Giedd, J. N., et al. (2001). Mapping adolescent brain change reveals
dynamic wave of accelerated gray matter loss in very early-onset schizophrenia. Proc Natl
Acad Sci USA 98: 11650–11655.
Tsai, G., Passani, L., Slusher, B., et al. (1995). Abnormal excitatory neurotransmitter metabolism
in schizophrenia brains. Arch Gen Psychiatry 52: 829–836.
Weinberger, D. R. (1987). Implications of normal brain development for the pathogenesis of
schizophrenia. Arch Gen Psychiatry 44: 660–669.
(1995). From neuropathology to neurodevelopment. Lancet 346: 552–557.
Weinberger, D. R., McClure, R. K. (2002). Neurotoxicity, neuroplasticity, and magnetic resonance
imaging morphometry: what is happening in the schizophrenic brain? Arch Gen Psychiatry
59: 553–558.
Woods, S. W., D’Souza, D. C., Wexler, B. E., Hoffman, R. E., McGlashan, T. H. (2002). Novel
early interventions for prodromal states. Acta Psychiatr Scand 106(S413): 12.
Woods, S. W., Breier, A., Zipursky, R. B., et al. (2003). Randomized trial of olanzapine versus
placebo in the symptomatic acute treatment of the schizophrenic prodrome. Biol Psychiatry
54: 453–464.
Wyatt, R. (1991). Neuroleptics and the natural course of schizophrenia. Schizophr Bull 17: 325–
351.
Zipursky, R. O. (1992). Widespread cerebral gray matter volume deficits in schizophrenia. Arch
Gen Psychiatry 49: 195–205.
 Index

N-acetyl aspartate amygdala–hippocampal complex (AHC) volume

in ADHD 394 changes 359
in obsessive–compulsive disorder 397–398, in high-risk population 350–351
399 in unaffected relatives 358
in schizophrenia 380 see also amygdala; hippocampus
predictive value 444 Angelman syndrome 184
ADHD see attention-deficit hyperactivity disorder antecedents of schizophrenia see predictors of
adolescence 70, 445–446, 447, 458 schizophrenia
at-risk population studies 445–446, 447 anterior cingulate 304
brain development 70–75 amygdalar fiber ingrowth 302, 303–304
myelination 71, 332 dopamine input distribution 297–298
synapse elimination 70–71, 458, 462 excitatory fiber changes 297, 298–299
cognitive development 69, 72–75, 82–83 in obsessive–compulsive disorder 396, 398,
executive functions 73–74 404–405
etiological factors 462–463 anticipation, genetic 339
sequelae of preterm birth 94–96 antidepressants 466
affective disorders 390, 391 antioxidants 158–159
see also bipolar disorder; depression dietary supplementation 163–164, 165, 467
African-Caribbeans in the UK 234–235 when and what to treat 166
unemployment 236 antipsychotic drugs
age of onset 274 adverse effects 159
possible origins of delayed onset 279–281 anti/pro-oxidant properties 164
sex differences 331 efficacy 159
alpha-linolenic acid 160 secondary prevention strategies 465
see also essential polyunsaturated fatty acids antisaccade task performance 53–55
d-amino acid oxidase (DAOO) gene 24 apoL lipoprotein 219–220
amisulpiride 465 apoptosis 162
amphetamine psychosis 250–251 neuropathological role 377–378, 382
genetic determinants 254 synaptic 377–378
sensitization 260–261, 273 see also synaptic pruning
amygdala 404–405 arachidonic acid (AA) 160
in bipolar disorder 112, 402, 404–405 in schizophrenia 160, 162
in major depressive disorder 401–402 role in membrane signal transduction
role in limbic lobe changes 298–299 161–162
fiber ingrowth into anterior cingulate cortex role in neurodevelopment 160–161
302, 303–304 see also essential polyunsaturated fatty
“mis-wired” circuit model 297, 304–305 acids
sex differences 41 arborization 39–40
stress and emotional reactivity and 275, 276 arcuate fasciculus 38

473
474 Index

at-risk populations see high-risk population studies postmortem findings 295–298

attention-deficit hyperactivity disorder (ADHD) dopamine system 297–298
40, 98, 390, 391–398 GABA system 296–297
common neuropathology 391–398, 402–405 boxcar design 48–49
genetic basis 402–403 brain
neuroimaging studies 57, 392–395 aging 332–333
frontostriatal circuitry 393–395 sex differences 40–41, 331–333
total cerebral volume 392–393, 394 structural abnormalities
nutritional factors 166 as potential endophenotypes 139
role of basal ganglia 98 epigenetic model and 187
with velo-cardio-facial syndrome 124 in ADHD 392–395
attentional processes in obsessive–compulsive disorder 395–400
development 75 in schizophrenia 141–142, 174, 347–348, 350,
attentional set-shifting ability 74–75, 80–81 415
in schizophrenia 80–81 in unaffected relatives 143–148
structural correlates 76–77 in velo-cardio-facial syndrome 129–131
dysfunction as predictor of schizophrenia 423, molecular genetics 148–149
442–443 predictive value 443–444
see also cognitive development; cognitive progressive structural changes 142, 379, 381,
dysfunction 459
attenuated psychotic symptoms 445, 457 state versus trait 142–143
see also prodromal phase see also premorbid structural abnormalities;
automatic rerouting 100–101 specific abnormalities
structural heritability 139–141
Barrington’s nucleus 276 structure and function relationship 97–98
basal ganglia remapping following brain injury 100
in ADHD 98 see also plasticity
in obsessive–compulsive disorder 399–400 total brain size 35–36, 40
in schizophrenia 141 in bipolar disorder 400–401
sex differences 40–41 in high-risk population 351–352
BDNF see brain-derived neurotrophic factor in major depressive disorder 400–401
behavior in obsessive–compulsive disorder 396
conduct disorder 256–257 in unaffected relatives 145
consequences of preterm birth 97 in velo-cardio-facial syndrome 129
drug abuse and 255–257 progressive decline 379–380
in velo-cardio-facial syndrome 124–125 see also brain development; brain injury; central
predictors of schizophrenia 425–426, 428, nervous system
437–438 brain-derived neurotrophic factor (BDNF) 13, 15,
archival–observational studies 428 216, 258, 335
high-risk studies 427 BDNF gene 17
population-based studies 425–427 brain morphometry and 149
see also obsessive–compulsive disorder changes in schizophrenia 16
bipolar disorder (BPD) 390, 391–398 expression 16
common neuropathology 112–113, 390, 391, brain development 5–6, 39, 330, 332–335
403, 404–405 aberrations 77–82, 142–143
genetic basis 403 animal models 278–279
neuroimaging studies 400–402 childhood injury 78
frontal cortex 401 developmental lesions 78–79
medial temporal cortex 401–402 epigenetic model and 180–181
total brain volume 400–401 etiological factors 462–463
475 Index

impact of 77–79 at the neural network level 99–100

in schizophrenia 72, 79–82, 174–175, 278 at the neuronal level 99
predictive value 443–444 Kennard principle 91
sequelae of preterm birth 94–96 limits of 101
see also premorbid structural abnormalities remapping of functions 100
brain function 45 rerouting 100–101
electroencephalography studies 47 see also obstetric complications
fMRI studies 47–55, 56–57 breast-feeding 162
language development 56–57, 89–90 Brn-1 transcription factor 10
PET studies 47 Brn-2 transcription factor 10
cell adhesion 10–14 Broca, Paul 91
cognitive development relationships 75–77, 83 Broca’s area 38
see also cognitive development language function 56, 91
corpus callosum 38
critical periods 90, 445–446 cadherins 10–12
dopaminergic system 334 cadherin motif 10–11
during adolescence 70–75 cadherin neuronal-related receptors (CNRs)
myelination 71, 332 11–12
synapse elimination 70–71, 458, 462 seven-pass transmembrane cadherins 11
estrogen role 334–335 Cajal–Retzius neurons 14
genetic basis 6–8 Cambridge Neuropsychological Test Automated
epigenetic factors 185 Battery (CANTAB) 73
see also transcriptomes cannabis-induced psychosis 251–253, 257
gray matter 39–40 sensitization 260–261
MDN–DLPFC circuitry 311, 317–321 CATCH 22 (cardiac anomalies, abnormal facies,
nutrition role 158 thymic hypoplasia, cleft palate, and
essential polyunsaturated fatty acids 160–161 hypocalcemia) 123
schizophrenia as model 79–82 see also velo-cardio-facial syndrome
segmentation 8–10 catechol-O-methyltransferase 128
serotonergic system 333–334 COMT gene 127, 128, 403
sex differences 332–333 in bipolar disorder 403
aging 332–333 methamphetamine abuse and 259
estrogen role 334–335 polymorphisms 128
myelination 332 caudate nuclei
social factors 239 ADHD and 98, 395
animal models 224–226 in obsessive–compulsive disorder 399–400
see also limbic lobe; neurodevelopmental model; cell adhesion 10–14
plasticity cell adhesion molecules (CAMs) 10–13
brain injury cadherin superfamily 10–12
childhood injury 78 immunoglobulin superfamily (IgSF) 10, 12–13
following preterm birth 92 NCAM 12–13
consequences of persisting abnormality 96–97 cell differentiation 180
patterns of 93–94 embryonic cells 181, 182
prevalence 94 central nervous system (CNS) development 3
sequelae in adolescents 94–96 cell adhesion 10–14
University College Hospital London Study genetic mechanisms 3, 6–8
92–93 neuronal survival 14–18
perinatal, drug abuse and 260 neurulation 5–8
plasticity after 91–92, 99–101 segmentation 8–10
at the connectivity level 100–101 see also brain development
476 Index

cerebellar cognitive–affective syndrome 98 during adolescence 69, 72–75, 82–83

cerebellum executive functions 73–74
in ADHD 395 early brain lesion effects 79
in velo-cardio-facial syndrome 129, 130 pediatric fMRI studies 47–55
sequelae of preterm birth 95 schizophrenia as model 79–82
consequences 98 voluntary response suppression 51–55
cerebral atrophy, following preterm birth 94 working memory 49–51, 73
cerebral blood flow (CBF), estrogen effect 336 in schizophrenia 80–81
cerebral cortex, asynchronous development structural correlates 76
332 see also attentional processes; executive
see also brain development functions
cerebral volume 37 cognitive dysfunction 310
heritability 140 glutamate role 460–461
in ADHD 392–393, 394 in velo-cardio-facial syndrome 123–124
in schizophrenia 141 predictive value 419–425, 442–443, 458
in unaffected relatives 145–146 high-risk studies 423–425
IQ relationship 36 population-based studies 419–423
sex differences 40 premorbid phase 376, 455–457
cerebrospinal fluid (CSF), sulcal spaces 145–146, progressive impairment 375–376, 381
355 social factors 238
chandelier neurons 320–321 see also working memory
abnormalities in schizophrenia 323–324 cohort studies 433–434
childhood abuse 228–229, 239 communal upbringing 227–228
childhood antecedents see predictors of communication deviance in parents 227
schizophrenia complexins 22
childhood behavior see behavior complexin I 22
childhood neuroimaging see pediatric complexin II 22
neuroimaging COMT see catechol-O-methyl transferase
chromosome 22q11.2 deletion 126–127 conduct disorder 256–257
see also velo-cardio-facial syndrome conotruncal anomaly face syndrome 123
ciliary neurotrophic factor (CNTF) 18 see also velo-cardio-facial syndrome
cingulate 404–405 coordination problems 417–418
see also anterior cingulate Copenhagen High-Risk Project 350, 354, 361, 363
city of birth effect 191, 230 corpus callosum (CC) 36–38
city of upbringing effect 230 development 38
civilization 157 function 38
clinical high-risk studies 438–439 in ADHD 395
see also high-risk population studies in bipolar disorder 402
cloning studies 184 in depression 402
clozapine 467 in obsessive–compulsive disorder 400
cocaine-induced psychosis 259 in unaffected relatives 147
sensitization 260 sex differences 331–332
cognition 72 thinning following preterm birth 95
estrogen effect 335–336, 337, 338 consequences 98
social cognition 339 cortical dysgenesis 129
see also cognitive development; cognitive cortical volume reduction 111–112
dysfunction animal model 279
cognitive behavioral therapy (CBT) 465, 466 corticotropin-releasing factor (CRF) 113–114,
cognitive development 46–47, 69–70, 82, 311 115
cerebral development relationships 75–77, 83 stress response 275–276
477 Index

cortisol 116 DNA methylation 175–177

in major depression 113, 116 during embryogenesis 182–183
in schizophrenia 115 cloned embryos 184
critical periods of development 90, 445–446 knockout models 184
for learning 90 maintenance 180–181
cultural differences see also epigenetics
ethnic minority status 233–234 DNA microarray studies 210–212
methodological issues 234 cDNA microarrays 211–212
nutrition 157 future directions 220
schizophrenia outcome 157–158 synthetic oligonucleotide probe arrays 212
transcriptome assessment 210
DA see dopamine data analysis and interpretation 214–215
DAOO gene 24 experimental design 212–213
deficit syndrome 375 gene expression changes in schizophrenia
deletion 22q11.2 see chromosome 22q11.2 215–220
deletion DNA methyltransferases 175
Delta gene 6–7 docosahexaenoic acid 160
delta sleep 458 dietary supplementation 163–165
dementia praecox 373 in schizophrenia 160, 162
depression role in membrane signal transduction
common neuropathology 112–113 161–162
cortisol role 116 role in neurodevelopment 160–161
glucocorticoid-related neurotoxicity 113 see also essential polyunsaturated fatty acids
hypothalamic–pituitary–adrenal axis dopamine (DA) system 281–285
abnormalities 113–114, 115–116 D1 receptor 461–462
see also major depression D2 receptor (DRD2) gene 258
deprivation 237 D3 receptor (DRD3) gene 258
development D4 receptor (DRD4) gene 258–259
CNS 3 development 301, 304, 334
critical periods 90, 445–446 late maturation of dopamine projections
epigenetic factors 179–187 301–303
see also brain development; estrogen effect 336–337
neurodevelopmental model in ADHD 395, 403
developmental compensation see plasticity in schizophrenia 273, 281–285, 402–403
developmental delays glutamatergic effects 460–462
in velo-cardio-facial syndrome 123–124, 174 sensitization 261, 273
predictive value 418 phasic dopamine levels 283
diencephalon 6 postmortem findings 297–298
diet see nutrition regulation 281–285, 460–462
dietary supplementation see nutritional strategies sensitization 260–261
diffusion tensor imaging (DTI) 36, 61–62 in schizophrenia 261, 273
DiGeorge syndrome 123 stress response 277, 281
see also velo-cardio-facial syndrome tonic dopamine levels 283
discrimination 235–236, 239 dorsolateral prefrontal cortex (DLPFC) see
dissociative identity disorder 229 MDN–DLPFC
distributed neural systems 36 circuitry; prefrontal cortex
dizygotic (DZ) twin studies see twin studies drug abuse
Dlx1 gene 9 as cause of psychosis 250–254
Dlx2 gene 9 amphetamine-induced psychosis 250–251
DNA demethylase 175 cannabis-induced psychosis 251–253
478 Index

drug abuse (cont.) epigenetics 175–177, 180, 181

individual differences in liability 254 misregulation 185–187
LSD-induced psychosis 253–254 relevance to schizophrenia 178–179, 185–187
phencyclidine-induced psychosis 253 role in development 179–187
association with psychosis 248–249 see also DNA methylation; histone modification
childhood development and personality ES2 gene 127
factors 255–257 essential fatty acids (EFAs) 158–159
familial predisposition 257–258 essential polyunsaturated fatty acids (EPUFAs)
direction of association 249–250 159, 160
common factor hypothesis 249–250 dietary intake 162–163
self-medication 249 dietary supplementation
vulnerability hypothesis 250 benefits 163–164, 166
dopamine sensitization 260–261 dose and quality of EPUFA 165
molecular genetic studies 258–260 duration of treatment 165
dopamine D2 receptor gene 258 placebo control 165
dopamine D3 receptor gene 258 preferred combination 165
dopamine D4 receptor gene 258–259 type of EPUFA 164–165
perinatal brain damage, animal models 260 when and what to treat 166
dysbindin gene 24 in schizophrenia 160, 162
dysfunctional family environment 227 role in membrane signal transduction
dyslexia, neuroimaging studies 57–58 161–162
role in neurodevelopment 160–161
early parental loss 228 oxidative breakdown 163–164
early rearing environment 200 antioxidant benefits 163–164
economic deprivation 237 estrogen 334–337, 340–341
Edinburgh High Risk Study (EHRS) 350–351 effect on cognitive function 335–336, 337, 338
EF motif see cadherin motif effect on dopaminergic function 336–337
eicosapentaenoic acid (EPA) 160 effect on serotonergic system 336
dietary supplementation 163–165 role in brain development 334–335
see also essential polyunsaturated fatty therapeutic use 340
acids estrogen receptors 335
electroencephalography (EEG), developmental ethnic minority status see minority groups
studies 47 ethyleicosopentaenoic acid 466
embryogenesis 182–183, 184–185 N-ethylmaleimide-sensitive factor (NSF) 216
pluripotent cell differentiation 181, 182 evoked response potentials 438, 443
emotional reactivity 275–277 executive functions 72
Emx1 gene 9 development 46, 72–74
Emx2 gene 9 attentional set-shifting ability 74–75, 80–81
endophenotypes 138–139 frontal lobe development and 75–76
cerebral volume 145–146 implications for neural development 80–81
frontal lobe 147 in adolescence 73–74
subcortical structures 146 see also cognitive development
temporal lobe 146 see also working memory
ventricular enlargement 144–145 expressed emotion 228
enhanced high-risk strategies 439–440, 446
environmental risk factors see risk factors family factors see social factors
epigenesis 179 family history 353–360
epigenetic inheritance system 177 CT studies 354–355
epigenetic landscape 180, 181, 185, 186 structural MRI studies 355–356
epigenetic metastability 177, 178, 179 famine see malnutrition
479 Index

α-fetoprotein 334–335 genetic factors

first-episode studies 432 ADHD 402–403
FMR1 gene 184, 339 bipolar disorder 403
fold changes 214 obsessive–compulsive disorder 403
follow-back studies 434 velo-cardio-facial syndrome 126–129
forebrain development 5–6 see also genetics of schizophrenia
see also brain development genetic high-risk populations 435–437
fragile X syndrome 184, 339–340 preventive intervention strategies 464
frontal lobes see also high-risk population studies
cognitive development and 72, 75–76 genetics of schizophrenia 138–139, 273–274,
in bipolar disorder 401 447
in major depressive disorder 401 brain structure 148–149
in schizophrenia 349 epigenetic theory 178–179, 185–187
in unaffected relatives 147 family history 353–360
in velo-cardio-facial syndrome 129–130 gene expression changes 215–220
outcome of frontal lesions 78–79 see also DNA microarray studies
see also brain; brain development heritability 138, 435
functional magnetic resonance imaging (fMRI) 47, imaging specific gene effects 358–360
48 multifactorial susceptibility model 138
abnormal populations 57–59 neurodevelopmental mechanisms 3, 4, 24
ADHD 57, 394–395 epigenetic factors 175, 179–185
dyslexia 57–58 susceptibility gene identification 138, 217–218
pediatric schizophrenia 58–59 see also epigenetics; X chromosome
block design 48–49 germinal matrix haemorrhage 93
cognitive development studies 47–55 glial cell loss 112
voluntary response suppression 51–55, 77 glial-cell-derived neurotrophic factor (GDNF)
working memory 49–51 17–18
event-related designs 49 gliosis 377
language development studies 56–57 β-globin gene regulation 182
pediatric studies 49, 57–59, 63 glucocorticoids
challenges 59–61 corticotropin-releasing factor regulation 114
see also magnetic resonance imaging (MRI); glucocorticoid receptor (GR) 114, 116
pediatric neuroimaging mineralocorticoid receptor (MR) 114
neurotoxicity 113, 116
G72 gene 24 stress response 280–281
GABA membrane transporter (GAT1) 320 see also cortisol
GABA system glutamate
abnormalities in schizophrenia 298, 299, cognitive role 460–461
323–324 developmental role 460
“mis-wired” circuit model 297, 304–305 neurodegeneration and 461
development 299–301, 304 transmission 273
gene expression changes 216 gene expression changes 216
postmortem findings 296–297 see also glutamatergic system
gamma-aminobutyric acid see GABA glutamate receptor 2 (AMPA2) 216, 219
gametogenesis 182 glutamatergic system 459–462
Gbx2 9 glutamatergic fiber changes 298–299
gender differences see sex differences in obsessive–compulsive disorder 398,
gene–environment interaction 237–238, 447 399–400
general population cohort studies 433–434 therapeutic strategies 466, 467
genetic anticipation 339 see also glutamate
480 Index

glutamic acid decarboxylase (GAD) 296–297, 300, sex differences 41

323 stress effects on 279–281
microarray studies 216, 219 volume abnormalities 111, 112, 277
glycine 466 genetic basis 148–149
glycogen synthase kinase 3 β 8 in high-risk population 351–352
gray matter 39–40 in major depressive disorder 401–402
developmental changes 39–40 in unaffected relatives 146
during adolescence 71, 332 neurotrophin 3 gene relationship 360
in major depressive disorder 401 obstetric complications and 358, 362
in schizophrenia 130, 131, 349, 379 see also amygdala–hippocampal complex
progressive loss 379, 381 histone modification 176–177, 181–182
in ultra-high-risk population 352 valproic acid effects 184–185
in velo-cardio-facial syndrome 131 see also epigenetics
GSCL (goosecoid-like) gene 127 homeobox (Hox) genes 9
Hoxb1 9
haloperidol 298 homeodomain 9
Hensen’s node 5 hormone replacement therapy (HRT) see estrogen
heritability hydrocephalus 94
of brain structure 139–141 hyperactivity 20, 98
of schizophrenia 138, 435 hypofrontality 380
see also genetics of schizophrenia hypothalamic–pituitary–adrenal (HPA) axis
high-risk population studies 350–353, 432–436, abnormalities in major depression 113–114,
437, 440, 446–448 115–116
adolescents 445–446, 447 abnormalities in schizophrenia 115–116
behavioral predictors 427, 437–438 amygdala influence 275
clinical high-risk strategies 438–439 hypothalamus, corticotropin-releasing factor
Copenhagen High Risk Project 350 (CRF)
Edinburgh High Risk Study 350–351 hypersecretion 113–114
enhanced high-risk strategies 439–440, 446 hypoxia see obstetric complications
follow-back studies 434
general population cohort studies 433–434 immunoglobulin superfamily (IgSF) 10, 11, 12–13
genetic high-risk studies 435–437 NCAM 12–13
Melbourne High Risk Study 351–353 inequality 237
methodological issues 440–441 infections, as risk factors
generalizability to schizophrenia 441 neonatal and childhood 194
reliability of parental diagnoses 440 prenatal 193–194, 362
specificity to schizophrenia 440–441 influenza 192–194
neurobehavioral high-risk strategies 437–438 influenza, prenatal, as risk factor 192–194
neurointegrative predictors 418 inheritance see genetics of schizophrenia;
neuropsychological predictors 423–425 heritability
retrospective studies 434 injury see brain injury
structural imaging 353 integrative neurochemical model 459–462
family history 353–360 intellectual capacity as predictor 419–420
working memory 58–59, 274–275 intelligence quotient see IQ
hindbrain development 5–6 interleukin-1β gene 149
see also brain development intervention strategies 463–467
hippocampus after disease onset 467
possible origins of delayed disease onset premorbid phase 463–464
279–281 at-risk populations 464
sequelae of preterm birth 95 universal interventions 463–464
481 Index

prodromal phase 464–467 locus coeruleus

low-dose atypical antipsychotic drugs 465 corticotropin-releasing hormone effects 276
naturalistic treatment studies 465–466 stress response 275, 276
non-antipsychotic medications 466 LSD psychosis 229
psychotherapeutic treatments 466–467
intraventricular haemorrhage, preterm birth and magnetic resonance imaging (MRI) 35, 48
93 region of interest (ROI) approach 349
IQ risk factor studies
brain size relationship 35–36 family history 355–356
decline during adolescence 82 velo-cardio-facial syndrome 129–131
following preterm birth 96–97 voxel-based morphometry (VBM) 349
in velo-cardio-facial syndrome 123, 124 see also functional magnetic resonance
predictive value 419–420 imaging
isolation of rearing 225 magnetic resonance spectroscopy (MRS) 62
see also social factors phosphorus-31 62, 380
major depressive disorder (MDD) 391–398
Kennard principle 91 common neuropathology 112–113
exceptions to 91–92 hypothalamic–pituitary–adrenal axis
Krox20 gene 8–9 abnormalities 113–114, 115–116
neuroimaging studies 400–402
labor complications see obstetric frontal cortex 401
complications medial temporal cortex 401–402
lamotrigene 467 total brain volume 400–401
language development 56–57, 89–90 see also depression
critical periods 89–90 malate dehydrogenase (MAD1) 217
developmental compensation following brain malnutrition 156, 158
injury 100–101 prenatal, as risk factor 194–195
remapping 100 see also nutrition
rerouting 100–101 Mash1 gene 6
in velo-cardio-facial syndrome 123, 124 MDN–DLPFC (mediodorsal nucleus–dorsolateral
predictive value 421 prefrontal cortex) circuitry 310, 312
learning, critical periods 90 development 311, 317–321
learning disabilities dysfunction in schizophrenia 321–324
in velo-cardio-facial syndrome 124 function 310–311
psychiatric disorders and 125 future directions 324–325
see also language development organization 313–317
life events 236–237 see also mediodorsal nucleus; prefrontal cortex;
limbic lobe 295, 296 thalamus
amygdala role in changes 298–299 MECP2 mutations 184
“mis-wired” circuit model 304–305 mediodorsal nucleus (MDN) 310
postnatal maturation of circuitry 299–304 volume reduction in schizophrenia 321, 322
amygdalar fiber ingrowth into anterior see also MDN–DLPFC circuitry; thalamus
cingulate cortex 303–304 Melbourne High Risk Study (MHRS) 351–353
GABA system 299–301 memantine 467
late maturation of dopamine projections membrane signal transduction 161–162
301–303 memory, estrogen effects 335–336
linolenic acid 160 see also working memory
see also essential polyunsaturated fatty mental retardation
acids in fragile X syndrome 339
lithium 466 in velo-cardio-facial syndrome 123, 124
482 Index

mental retardation (cont.) negative symptoms 445

psychiatric disorders and 125 progression 375, 457
see also IQ neonatal risk factors 200
mesencephalon development 5–6 nerve growth factor (NGF) 15, 335
meta-cognition 69 expression 16
metabolic changes 216–217, 380 neural ectoderm 5
metencephalon 6 neural plate 5
methamphetamine psychosis 250–251 neural tube 5
familial predisposition to schizophrenia and neurobehavioral high-risk studies 437–438
257–258 see also high-risk population studies
genetic studies 259 NeuroD 7
premorbid characteristics 255–257 neurodegenerative disorders 374
sensitization 261 neurodegenerative models 373, 382–383, 459
N-methyl-d-aspartate receptors see NMDA clinical psychopathology and cognition 374–376
receptors integrative neurochemical model 459–462
methyl azoxymethanol acetate (MAM) 278–279 neuroimaging studies 379–381
methylphenidate 394–395 correlation between imaging data and
microarray studies see DNA microarray studies outcome 380–381
midbrain development 5–6 functional and spectroscopic studies 380
see also brain development structural studies 379–380
migration 233–234 postmortem findings 376–378
see also minority groups potential mechanisms of neurodegeneration
mineralocorticoid receptor (MR) 114 381–382, 463
minor physical anomalies as risk factors 195–196, versus neurodevelopmental model 81–82,
443 111–112
minority groups 233–234 neurodevelopmental model 3, 79–82, 111,
African-Caribbeans in the UK 234–235 174–175, 373, 415
discrimination 235–236, 239 animal models 277–279
methodological issues 234 early developmental model 457–458
second generation 235 genetic mechanisms 3, 4, 24
monozygotic (MZ) twin studies see twin studies epigenetic factors 175, 179–187
mood disorders see bipolar disorder; depression integrative neurochemical model 459–462
mother–child relationship 226 late developmental model 458–459
motor coordination dysfunction 417–418 limitations of 373
multifactorial susceptibility model 138 pediatric fMRI studies 58–59
myelination 36, 47, 332 support for 121
diffusion tensor imaging studies 61–62 unifying model 455
during adolescence 71, 332 velo-cardio-facial syndrome as model for
gene expression changes 219 schizophrenia 122
rerouting plasticity and 101 versus atrophy/neurodegeneration 81–82,
sex differences 332 111–112
see also brain development
NAA (N-acetylaspartate) neurogenins 7
in ADHD 394 Ngn1 7
in obsessive–compulsive disorder 397–398, Ngn2 7
399 neuromeres 6
in schizophrenia 380 neuromotor dysfunction 417–418, 443
predictive value 444 neuron 37
NCAM see neuronal cell adhesion molecule neuronal cell adhesion molecule (NCAM) 12–13
NCAM1 gene 12, 13 gene NCAM1 12, 13
483 Index

function 12–13 obsessive–compulsive disorder (OCD) 40–41, 390,

isoforms 12 391–398
role in schizophrenia 13 common neuropathology 391, 404–405
neuronal cell loss 377 genetic basis 403
neuronal survival 14–18 neurodevelopment 397, 398
neuroregulin 1 gene 24 neuroimaging studies 395–400
neurotransmission 18–23 basal ganglia 399–400
neurotrophins (NT) 14–17, 335 prefrontal cortex 396
NT-3 15, 335 thalamus 399
A3 allele 360 total brain volume 396
expression 18 with velo-cardio-facial syndrome 125
NT-4/5 15, 335 obstetric complications (OCs) 143, 175, 197–199,
expression 16 200–202, 403
NT-6 15 imaging abnormalities and 360–363
neurulation 5–8 hippocampi 358, 362
genetic basis 6–8 ventricles 360–361
primary 5 neurodevelopmental deficits and 419
secondary 5 OCD see obsessive–compulsive disorder
NMDA receptors 282, 460, 461–462 oculomotor delayed response (ODR) task 58–59,
estrogen effect 335 310–311
phencyclidine effect 253 see also working memory
role in neurodegeneration 381–382 odds ratio 417
role in neurodevelopment 460 olanzapine 465
Notch genes 6–7 onset of disease 274, 457
NOTCH4 7–8 possible origins of delayed onset
NSF (N-ethylmaleimide-sensitive factor) 216 279–281
NT-3 gene 17 sex differences 331
see also neurotrophins oogenesis 182
nutrition 156 Otx2 gene 9
role in brain and behavioral development 158 outcome measures 441–442, 444–445
role in health 156
role in schizophrenia 156–157, 158–159 p75NTR receptor 15
undernutrition 156, 158 pandevelopmental retardation 418
prenatal malnutrition as risk factor 194–195 parent-of-origin effects 178–179
see also essential polyunsaturated fatty acids parents
nutritional strategies communication deviance 227
dietary supplementation 159, 163–164 early parental loss 228
antioxidants 163–164 mother–child relationship 226
essential polyunsaturated fatty acids 163–165 reliability of parental diagnoses 440
preferred combination 165 parietal lobe, in velo-cardio-facial syndrome 129,
issues 164–165 130
adjunctive medication 164 parvalbumin 313, 314, 315
age and duration of illness 164 chandelier neurons 320
dose and quality of essential polyunsaturated wide-arbor neurons 320–321
fatty acids 165 Pax6 10
duration of treatment 165 pediatric neuroimaging 49, 57–59, 63
placebo control 165 abnormal populations
type of essential polyunsaturated fatty acids ADHD 57
164–165 dyslexia 57–58
when and what to treat 166 schizophrenia 58–59
484 Index

pediatric neuroimaging (cont.) electrophysiological measures 443

challenges to pediatric neuroimaging 59–61 minor physical anomalies 443
differences in performance 60–61 negative predictive value 417
head movement 59–60 neuroimaging measures 443–444
respiratory and heart rate cycles 60 neurointegrative predictors 416, 417–419, 443
stress 59 archival–observational studies 419
structural differences between child and adult high-risk studies 418
brain 60 population-based studies 417–418
future directions 61–63 neuropsychological predictors 419–425
see also specific types of imaging cognitive deficits 442–443, 458
perigigantocellularis region 276 high-risk studies 423–424, 425
periventricular hemorrhage (PVH) 94 population-based studies 419–422, 423
periventricular hemorrhagic infarction (PHI) 93 positive predictive value 417
periventricular leukomalacia (PVL) 93 prodromal symptoms 438–439
PET see positron emission tomography sensitivity 417, 421–423
phencyclidine (PCP) psychosis 253 specificity 417, 421–423
phenotype 138, 147–148 see also high-risk population studies; premorbid
behavioral phase; risk factors
childhood developmental delays 123–124 prefrontal cortex
childhood velo-cardio-facial syndrome dorsolateral (DLPFC) 310, 314, 319
124–125 structural abnormalities in schizophrenia
see also endophenotypes 321–323
phosphodiesters (PDEs) 62, 380 see also MDN–DLPFC circuitry
phospholipid breakdown 380 in ADHD 57, 393–395
phosphomonoesters (PMEs) 62, 380 in bipolar disorder 401
PIK4CA gene 127 in major depressive disorder 401
plasticity 89–90 in obsessive–compulsive disorder 396
following brain injury 91–92, 99–101 in schizophrenia 55, 274–275
at the connectivity level 100–101 metabolic changes 216–217, 380
at the neural network level 99–100 possible origins of delayed onset 280–281
at the neuronal level 99 maturation 46
Kennard principle 91 microarray studies 215–220
remapping of functions 100 role in voluntary response suppression 52–53
rerouting 100–101 role in working memory 49–51, 310–311
following preterm birth 92, 96–97, 99–101 stress and emotional reactivity relationships
limits of 101 275–277
see also brain development premorbid phase 415, 456, 457–458, 460
polysialic acid (PSA) 12–13 cognitive impairments 455–457
positive symptoms 375 progressive impairment 376
attenuated prodromal symptoms 457 etiological factors 462
positron emission tomography (PET), premorbid adjustment 363–364
developmental studies 47 premorbid personality and drug abuse 255–257
POU domain family 9–10 primary prevention strategies 463–464
Prader–Willi syndrome 184 see also high-risk population studies; predictors
predictors of schizophrenia 415–417, 428–429, of schizophrenia
441–444, 447 premorbid structural abnormalities 347, 364–365
behavioral predictors 425–426, 428, 437–438 high-risk population studies 350–353
archival–observational studies 428 Edinburgh High Risk Study 350–351
high-risk studies 427 Melbourne High Risk Study 351–353
population-based studies 425–427 premorbid adjustment and 363–364
485 Index

structural imaging and risk factors 353 onset 457

family history 353–360 see also schizophrenia
see also premorbid phase psychotherapeutic treatments 466–467
prenatal infections 193–194, 362 PSYN group genes 215–216, 218, 219
influenza 192–194 puberty 70
prenatal risk factors see risk factors
preterm birth 92, 93 racial discrimination 235–236, 239
consequences of persisting brain abnormality Reelin 13–14
96–97 RELN gene 14
behavior 97 polymorphisms 14
neurology 96 role in schizophrenia 14
neuropsychology 96–97 regulator of G-protein signaling 4 (RGS4)
limits of brain plasticity 101 217–218
perinatal brain injuries 93–94 remapping plasticity 100
link between brain structure and function rerouting plasticity 100–101
97–98 response suppression 51–55
prevalence 94 developmental studies 51–55, 77
sequelae in adolescent brains 94–96 prefrontal cortex role 52–53
possible plasticity mechanisms 99–101 retrospective studies 434
at the connectivity level 100–101 Rett syndrome 184
at the neural network level 99–100 rhesus incompatibility as risk factor 195
at the neuronal level 99 rhombencephalon development 5–6
remapping of functions 100 risk factors 191, 353
rerouting 100–101 family history 353–360
University College Hospital London study CT studies 354–355
92–93 structural MRI studies 355–356
follow-up 93 neonatal risk factors 200
prevention see intervention strategies neonatal and early childhood infections 194
prion protein (PRNP) gene 149 obstetric complications 143, 175, 197–199,
probes see DNA microarray studies 200–202, 403
prodromal phase 432, 438–439, 456, 457 imaging abnormalities and 360–363
clinical high-risk strategies 438–439 prenatal risk factors 191–197, 201, 274
intervention strategies 464–467 malnutrition 194–195
low-dose atypical antipsychotic drugs 465 minor physical anomalies 195–196
naturalistic treatment studies 465–466 prenatal infection 192–194, 362
non-antipsychotic medications 466 rhesus incompatibility 195
psychotherapeutic treatments 466–467 stress exposure 196–197
progressive cognitive impairment 375–376 time and place of birth 191–192, 362
subthreshold symptoms 445, 457 schizotypal personality disorder 358
proline dehydrogenase (PDH) 127 see also drug abuse; high-risk population
proneural genes 6–7 studies; predictors of schizophrenia; social
prosencephalon development 5–6 factors; urban effect
prostaglandins 162 risperidone 465
proteolipid protein 1 (PLP1) 219 rubella, prenatal, as risk factor 193
pruning see synaptic pruning
psychosis schizophrenia 79–82
drug-induced 250–254 brain structure 141–142, 174, 347–348, 350
early functional decline 457 molecular genetics 148–149
neurotoxicity of untreated psychosis 380–381, prefrontal cortex 55
459 state versus trait 142–143
486 Index

schizophrenia (cont.) schizophrenia 331

chemical changes clinical features 331
BDNF 16 epidemiology 331
NCAM 13 sex steroids 330
SNAP-25 20 see also estrogen
common neuropathology 112–113, 390–392, smooth pursuit eye movements (SPEM) 438, 443
402–403, 404–405 SNAP-25 18, 19–20
early course of disease 456, 457 changes in schizophrenia 20
see also neurodegenerative models; SNARE proteins 18–19
premorbid phase; prodromal phase see also SNAP-25; syntaxins
onset of disease 274, 457 social cognition 339
possible origins of delayed onset social drift 230–231
279–281 social factors 224, 225, 239–240
sex differences 331 animal research on brain development 224–226
outcome variability 157–158 family factors 226–229
postmortem findings 295–298, 376–378 childhood abuse 228–229, 239
dopamine system 297–298 communal upbringing 227–228
GABA system 296–297 dysfunctional family environment 227
see also genetics of schizophrenia; early parental loss 228
neurodegenerative models; expressed emotion 228
neurodevelopmental model family communication deviance 227
schizophrenia spectrum of psychopathology mother–child relationship 226
444–445 unwantedness 226–227
schizotypal personality disorder 358 interaction with other factors 237–238
schizotypy 255, 445 cognitive processing 238
drug abuse association 255–256 gene–environment interaction 237–238
school change in adolescence 232 life events 236–237
season of birth, as risk factor 191–192, social causation versus social selection 238–239
362 social isolation 232–236
segmentation 8–10 African-Caribbeans in the UK 234–235
self-medication 249 at time of onset 233
sensitive periods see critical periods discrimination 235–236, 239
separation anxiety disorder 125 during childhood 232
serotonergic system in young adult life 232–233
development 333–334 methodological issues 234
estrogen effect 336 migration and ethnic minority status 233–234
serotonin 333 moving schools in adolescence 232
seven-pass transmembrane cadherins 11 second generation 235
sex chromosomes 330 unemployment 236
see also X chromosome socioeconomic factors 237
sex differences impact on outcome 157–158
brain development 332–333 see also urban effect
aging 332–333 social maladjustment as predictor 425, 427
estrogen role 334–335 see also behavior
myelination 332 social residue theory 231
brain structure 40–41, 331–332 sodium valproate, teratogenic effects 184–185
amygdala 41 Sonic hedgehog (Shh) gene 7
basal ganglia 40–41 speech difficulties as predictor 421
cerebral volume 40 see also language development
hippocampus 41 spermatogenesis 182
487 Index

strategic rerouting 101 in obsessive–compulsive disorder 399

stress 277, 281, 285 in schizophrenia 146, 321, 349
as risk factor 236–237 see also mediodorsal nucleus
prenatal exposure 196–197 topiramate 467
hippocampal damage 279–281 total brain size see brain
pediatric neuroimaging and 59 Tourette’s syndrome 40
possible origins of delayed disease onset transcription factors 10, 176
279–281 see also specific factors
prefrontal cortex and 275–277 transcriptomes 210
role in neurodegeneration 382 assessment methods 210–212
structural magnetic resonance imaging (sMRI) see experimental design in microarray studies
magnetic resonance imaging 212–213
substance abuse see drug abuse future directions 220
subthreshold symptoms 445, 457 gene expression changes in schizophrenia
susceptibility genes 215–220
identification of 138, 217–218 microarray data analysis and interpretation
multifactorial susceptibility model 138 214–215
see also endophenotypes trinucleotide repeats 339–340
Suzuki method 90 Trk (tyrosine kinase) receptors 15–16
synapsins 20–22 TrkA 15, 16
synapsin I 21 TrkB 13, 15
synapsin II 21–22, 216 TrkC 15, 16
synapsin III 21, 22 tumor necrosis factor receptor II (TNF-RII) gene
synaptic apoptosis 377–378 149
see also synaptic pruning Turner syndrome 338–339
synaptic marker reduction 377 twin studies 435
synaptic pruning 39–40, 47, 130, 218, 458–459, 461 brain structure 139–141, 143, 355–356
DLPFC 318–319 cerebral volume 145
during adolescence 70–71, 458, 462 corpus callosum 147
magnetic resonance spectroscopy studies 62 frontal lobe 147
plasticity following brain injury 99 discordance in monozygotic twins 179
synaptic transmission 18–23 obstetric complications 360–361
impaired synaptic function 218 tyrosine kinase receptors see Trk receptors
synaptobrevins 18
syntaxins 18, 22 UFDIL gene 127
ultra-high-risk population study 351–353
TBX1 gene 127–128 see also high-risk population studies
telencephalon 6 undernutrition 156, 158
temporal lobe see also nutrition
in unaffected relatives 146 unemployment 236
in velo-cardio-facial syndrome 129–130 universal intervention 463–464
teratogenic effects on epigenetic regulation University College Hospital London preterm birth
184–185 study 92–93
testosterone role in brain development 334 see also preterm birth
thalamocortical projections see MDN–DLPFC unwantedness 196, 226–227
circuitry urban effect 230–232
thalamus city of birth 191, 230
axon development 318 city of upbringing 230
volume changes possible explanations 231–232
in high-risk population 350, 357 social drift and social residue theories 230–231
488 Index

valproic acid, teratogenic effects 184–185 voluntary response suppression 51–55

velo-cardio-facial syndrome (VCFS) 122 developmental studies 51–55, 77
as neurodevelopmental model for schizophrenia prefrontal cortex role 52–53
122 voxel-based morphometry (VBM) 349
childhood behavioral phenotype 124–125
childhood developmental delays 123–124 Wernicke’s area 38
genetics of 126–129 white matter 36–38
medical aspects 122–123 changes during adolescence 71, 332
neuroimaging 129–131 changes in major depressive disorder 401
psychiatric presentation in adulthood sequelae of preterm birth 95
125 wide-arbor neurons 320–321
schizophrenia risk and 122, 125, Wnt signaling pathway 7
128–129 abnormalities 8
ventricular dilatation 111 working memory
following obstetric complications development 49–51, 73
360–361 in schizophrenia 80–81
following preterm birth 94 structural correlates 76
sequelae in adolescent brains 95 DLPFC role 310–311
genetic marker association 360 dysfunction in schizophrenia 310–311
in high-risk population 350 see also cognitive dysfunction
family history 354–355 in high-risk group 58–59, 274–275
in schizophrenia 141, 349 oculomotor delayed response task 58–59
common neuropathology 112 pediatric studies 50–51
progressive change 379–380, 381 prefrontal cortex role 49–51
in unaffected relatives 144–145, thalamic mediolateral nucleus role 310–311
356–357
ventricular volume, heritability 139–140 X chromosome 338–341
vesicle-SNAREs (v-SNAREs) 18 trinucleotide repeats 339–340
vesicular ATPase 216 Xash3 gene 6
vitamin C supplementation 165
vitamin D hypothesis 159 Yemenite Jews 235
vitamin E supplementation 163,
165 Zucker Hillside Recognition and Prevention (RAP)
vitamin requirements 158–159 program, New York 465–466