Developmental neuropathology [Second edition.] 9781119013082, 1119013089

Table of contents :
fmatter
ch1
ch2
ch3
ch4
ch5
ch6
ch7
ch8
ch9
ch10
ch11
ch12
ch13
ch14
ch15
ch16
ch17
ch18
ch19
ch20
ch21
ch22
ch23
ch24
ch25
ch26
ch27
ch28
ch29
ch30
ch31
ch32
ch33
ch34
ch35
ch36
ch37
ch38
ch39
ch40
ch41
ch42
ch43
index

Citation preview

Developmental Neuropathology

Developmental Neuropathology SECOND EDITION

EDITED BY

HOMA ADLE-BIASSET TE Universit´e Paris Diderot, Paris

BRIAN N. HARDING Children’s Hospital of Philadelphia, Philadelphia

JEFFREY A. GOLDEN Ramzi S. Cotran Professor of Pathology, Boston

SERIES EDITORS

FRANC ¸ OISE GRAY Universit´e Denis Diderot Paris 7, France

CATHERINE (KAT Y) KEOHANE University College Cork, Ireland

This first published 2018 © 2018 John Wiley & Sons Ltd All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, except as permitted by law. Advice on how to obtain permission to reuse material from this title is available at http://www.wiley.com/go/permissions. The right of Homa Adle-Biassette, Brian N. Harding and Jeffrey A. Golden to be identified as the authors of the editorial material in this work has been asserted in accordance with law. Registered Office(s) John Wiley & Sons, Inc., 111 River Street, Hoboken, NJ 07030, USA John Wiley & Sons Ltd, The Atrium, Southern Gate, Chichester, West Sussex, PO19 8SQ, UK Editorial Office

9600 Garsington Road, Oxford, OX4 2DQ, UK

For details of our global editorial offices, customer services, and more information about Wiley products visit us at www.wiley.com. Wiley also publishes its books in a variety of electronic formats and by print-on-demand. Some content that appears in standard print versions of this book may not be available in other formats. Limit of Liability/Disclaimer of Warranty The contents of this work are intended to further general scientific research, understanding, and discussion only and are not intended and should not be relied upon as recommending or promoting scientific method, diagnosis, or treatment by physicians for any particular patient. In view of ongoing research, equipment modifications, changes in governmental regulations, and the constant flow of information relating to the use of medicines, equipment, and devices, the reader is urged to review and evaluate the information provided in the package insert or instructions for each medicine, equipment, or device for, among other things, any changes in the instructions or indication of usage and for added warnings and precautions. While the publisher and authors have used their best efforts in preparing this work, they make no representations or warranties with respect to the accuracy or completeness of the contents of this work and specifically disclaim all warranties, including without limitation any implied warranties of merchantability or fitness for a particular purpose. No warranty may be created or extended by sales representatives, written sales materials or promotional statements for this work. The fact that an organization, website, or product is referred to in this work as a citation and/or potential source of further information does not mean that the publisher and authors endorse the information or services the organization, website, or product may provide or recommendations it may make. This work is sold with the understanding that the publisher is not engaged in rendering professional services. The advice and strategies contained herein may not be suitable for your situation. You should consult with a specialist where appropriate. Further, readers should be aware that websites listed in this work may have changed or disappeared between when this work was written and when it is read. Neither the publisher nor authors shall be liable for any loss of profit or any other commercial damages, including but not limited to special, incidental, consequential, or other damages. Library of Congress Cataloging-in-Publication Data Names: Adle-Biassette, Homa, editor. | Harding, Brian (Brian N.), editor. | Golden, Jeffrey A., 1961– editor. Title: Developmental neuropathology / [edited] by Homa Adle-Biassette, Brian Harding, Jeffrey A. Golden. Description: Second edition. | Hoboken, NJ : Wiley, 2018. | Includes bibliographical references and index. | Identifiers: LCCN 2017050347 (print) | LCCN 2017051636 (ebook) | ISBN 9781119013099 (pdf) | ISBN 9781119013105 (epub) | ISBN 9781119013082 (cloth) Subjects: | MESH: Central Nervous System Diseases–genetics | Central Nervous System Diseases–pathology | Genetic Diseases, Inborn–pathology | Nervous System Malformations–pathology Classification: LCC RC347 (ebook) | LCC RC347 (print) | NLM WL 301 | DDC 616.8/047–dc23 LC record available at https://lccn.loc.gov/2017050347 Cover design by: Wiley Cover images: Main image courtesy of Brian Harding Smaller images from top to bottom: 1st one courtesy of Brian Harding; 2nd one courtesy of Jeff Golden; 3rd one courtesy of Pr. Hans Goebel; 4th one courtesy of Pr. Annie Laquerri`ere; 5th one courtesy of Jeff Golden Set in 9.5/12pt Minion by Aptara Inc., New Delhi, India 10 9 8 7 6 5 4 3 2 1

Contents

List of Contributors, vii Introduction, xiii 1 Central Nervous System Manifestations of Chromosomal Change, 1 Joseph R. Siebert 2 Neural Tube Defects, 13 Andrew J. Copp and Brian N. Harding 3 Midline Patterning Defects, 29 Edwin S. Monuki and Jeffrey A. Golden 4 Microcephaly, 41 Sandrine Passemard, Annie Laquerri`ere, Nathalie Journiac, and Pierre Gressens 5 Hemimegalencephaly and Dysplastic Megalencephaly, 55 Ghayda Mirzaa, Achira Roy, William B. Dobyns, Kathleen Millen, and Robert F. Hevner

12 Chiari Malformations, 133 Homa Adle-Biassette and Jeffrey A. Golden 13 Dandy–Walker Malformation, Mega Cisterna Magna, and Blake’s Pouch Cyst, 141 Robert F. Hevner, Kathleen Millen, and William B. Dobyns 14 Joubert Syndrome, 151 Robert F. Hevner, William B. Dobyns, and Enza Maria Valente 15 Cerebellar Heterotopia and Dysplasia, 159 Marie Rivera-Zengotita and Anthony T. Yachnis 16 Brainstem Malformations, 167 Brian N. Harding 17 Spinal Cord Lesions, 179 Annie Laquerri`ere and Florent Marguet 18 Hydrocephalus, 187 Homa Adle-Biassette

6 Lissencephaly, Type I, 63 Jeffrey A. Golden

19 Antenatal Disruptive Lesions, 199 Brian N. Harding

7 Lissencephaly, Type II (Cobblestone Lissencephaly), 75 Jeffrey A. Golden

20 Hemorrhagic Lesions, 203 Marc R. Del Bigio

8 Polymicrogyria, 85 Jeffrey A. Golden

21 White Matter Lesions in the Perinatal Period, 213 Robin L. Haynes and Rebecca D. Folkerth

9 Cerebral Heterotopia, 91 Edwin S. Monuki and Keith L. Ligon

22 Gray Matter Lesions, 229 Marc R. Del Bigio

10 Hippocampal Sclerosis, Granule Cell Dispersion, and Cortical Dysplasia, 101 Maria Thom 11 Tuberous Sclerosis Complex, 117 Shino D. Magaki and Harry V. Vinters

23 Pediatric Head Injury, 241 Colin Smith, Thomas S. Jacques, and R. Ross Reichard 24 Pediatric Vascular Malformations, 251 Shino D. Magaki, Randy Tashjian, and Harry V. Vinters

v

Contents 25 Sudden Infant Death Syndrome, 269 Hannah C. Kinney, Marco M. Hefti, Richard D. Goldstein, and Robin L. Haynes 26 Kernicterus, 281 Mariarita Santi and Lucy B. Rorke 27 Lesions Induced by Toxins, 285 Mariarita Santi, Lucy B. Rorke, and Catherine Keohane 28 Disorders of Carbohydrate Metabolism, 293 Josefine Radke, Carsten G. B¨onnemann, Werner Stenzel, and Hans-H. Goebel 29 Sphingolipidoses and Related Disorders, 313 Annie Laquerri`ere, Soumeya Bekri, Kinoko Suzuki, and Brian N. Harding 30 The Neuronal Ceroid Lipofuscinoses, 369 Josefine Radke, Krystina E. Wisniewski, Werner Stenzel, and Hans-H. Goebel 31 Peroxisomal Disorders, 381 Phyllis L. Faust 32 Mitochondrial Disorders, 393 Anders Oldfors and Brian N. Harding 33 Disorders of Amino Acid Metabolism and Canavan Disease, 403 Dimitri P. Agamanolis 34 Pelizaeus–Merzbacher Disease, 417 Brian N. Harding

vi

35 Cockayne Syndrome, 427 Karen M. Weidenheim and P. J. Brooks 36 Vanishing White Matter Disease, 437 Marianna Bugiani, James M. Powers, and Marjo S. van der Knaap 37 Alexander Disease, 447 James E. Goldman and Mel B. Feany 38 Neuroaxonal Dystrophy/Neurodegeneration with Brain Iron Accumulation, 455 Abi Li, Sarah Wiethoff, Charles Arber, Henry Houlden, Tamas Revesz, and Janice L. Holton 39 Spinal Muscular Atrophy, 469 Brian N. Harding 40 Autism Spectrum Disorders, 477 Matthew P. Anderson 41 Intrauterine Infections, 497 Catherine Keohane and Homa Adle-Biassette 42 Perinatal and Postnatal Infections, 511 Catherine Keohane 43 Rasmussen Encephalitis, 531 Harry V. Vinters, Shino D. Magaki, and Geoffrey C. Owens

Index, 537

List of Contributors

Lucy B. Rorke Department of Pathology and Laboratory Medicine Children’s Hospital of Philadelphia University of Pennsylvania Perelman School of Medicine Philadelphia, PA USA Homa Adle-Biassette Department of Pathology, APHP, Lariboisi`ere Hospital, Universit´e Paris Diderot Paris France Dimitri P. Agamanolis Department of Pathology Akron Childrens Hospital Akron, OH USA; Department of Pathology Northeast Ohio Medical Universities (NEOMED) Rootstown, OH USA

Matthew P. Anderson Department of Pathology Division of Neuropathology Harvard Medical School Beth Israel Deaconess Medical Center Boston, MA USA Charles Arber Department of Molecular Neuroscience UCL Institute of Neurology University College London London UK

Soumeya Bekri Department of Metabolic Biochemistry University Hospital Rouen France ¨ Carsten G. Bonnemann Neuromuscular and Neurogenetic Disorders of Childhood Section National Institute of Neurological Disorders and Stroke/NIH Porter Neuroscience Research Center Bethesda, MD USA P. J. Brooks Office of Rare Diseases Research and Division of Clinical Innovation National Center for Advancing Translational Sciences National Institutes of Health and Laboratory of Neurogenetics National Institute of Alcohol Abuse and Alcoholism National Institutes of Health Bethesda, MD USA Marianna Bugiani Departments of Child Neurology and Pathology VU University Medical Center, Amsterdam The Netherlands Andrew J. Copp Newlife Birth Defects Centre Institute of Child Health University College London London UK

vii

List of Contributors

Marc R. Del Bigio Department of Pathology University of Manitoba Winnipeg Canada; Diagnostic Services Manitoba Children’s Hospital Research Institute of Manitoba Winnipeg Canada

William B. Dobyns Department of Pediatrics University of Washington Seattle, WA USA; Center for Integrative Brain Research Seattle Children’s Research Institute Seattle, WA USA

Phyllis L. Faust Columbia University Department of Pathology and Cell Biology New York, NY USA Mel B. Feany Department of Pathology Brigham and Women’s Hospital Harvard Medical School Boston, MA USA Rebecca D. Folkerth New York City Office of the Chief Medical Examiner New York University School of Medicine New York, NY USA Hans-H. Goebel Department of Neuropathology Charit´e Universit¨atsmedizin Berlin Berlin Germany Jeffrey A. Golden Department of Pathology Brigham and Women’s Hospital Harvard Medical School Boston, MA USA

viii

James E. Goldman Department of Pathology and Cell Biology Columbia University New York, NY USA Richard D. Goldstein Department of General Pediatrics Children’s Hospital Boston Boston, MA USA; Harvard Medical School Boston, MA USA

Pierre Gressens Paris Diderot University Paris France; Inserm U1141 Robert Debr´e Hospital Paris France; Center for Developing Brain King’s College, St. Thomas’ Campus London UK

Brian N. Harding Department of Pathology and Laboratory Medicine Children’s Hospital of Philadelphia Philadelphia, PA USA Robin L. Haynes Department of Pathology Boston Children’s Hospital Boston, MA USA Marco M. Hefti Department of Pathology Mount Sinai Medical Center New York, NY USA Robert F. Hevner Department of Neurological Surgery University of Washington Seattle, WA USA;

List of Contributors

Center for Integrative Brain Research Seattle Children’s Research Institute Seattle, WA USA

Janice L. Holton Queen Square Brain Bank Reta Lila Weston Institute of Neurological Studies UCL Institute of Neurology University College London London UK Henry Houlden Department of Molecular Neuroscience UCL Institute of Neurology University College London London UK Thomas S. Jacques UCL Institute of Child Health and Great Ormond Street Hospital Great Ormond Street Hospital for Children NHS Foundation Trust London UK Nathalie Journiac Inserm U1141 Robert Debr´e Hospital Paris France Catherine Keohane Department of Pathology and School of Medicine University College Cork Ireland Hannah C. Kinney Department of Pathology Boston Children’s Hospital Boston, MA USA; Harvard Medical School Boston, MA USA

` Annie Laquerriere Department of Pathology Pavillon Jacques Delarue Rouen University Hospital Rouen France;

Region-Inserm Team NeoVasc ERI28 Laboratory of Microvascular Endothelium and Neonatal Brain lesions Institute of Research Innovation in Biomedecine Normandy University Rouen Rouen France

Abi Li Queen Square Brain Bank Reta Lila Weston Institute of Neurological Studies UCL Institute of Neurology University College London London UK Keith L. Ligon Division of Neuropathology Department of Pathology Brigham and Women’s Hospital Boston, MA USA Shino D. Magaki Section of Neuropathology Department of Pathology and Laboratory Medicine David Geffen School of Medicine University of California Los Angeles, CA USA Florent Marguet Department of Pathology Rouen University Hospital Rouen France Kathleen Millen Center for Integrative Brain Research Seattle Children’s Research Institute Seattle, WA USA Ghayda Mirzaa Department of Pediatrics University of Washington Seattle, WA USA; Center for Integrative Brain Research Seattle Children’s Research Institute Seattle, WA USA

ix

List of Contributors

Edwin S. Monuki Department of Pathology and Laboratory Medicine UC Irvine School of Medicine Irvine, CA USA Anders Oldfors Department of Pathology Sahlgrenska University Hospital Gothenburg Sweden Geoffrey C. Owens Department of Neurosurgery David Geffen School of Medicine and Ronald Reagan UCLA Medical Center University of California Los Angeles, CA USA Sandrine Passemard Department of Genetics Robert Debr´e Hospital Paris France; Paris Diderot University Paris France; Inserm U1141 Robert Debr´e Hospital Paris France

James M. Powers Department of Pathology and Laboratory Medicine University Rochester Medical Center Rochester, NY USA Josefine Radke Department of Neuropathology Charit´e Universit¨atsmedizin Berlin Berlin Germany R. Ross Reichard Mayo Clinic Rochester, MN USA

x

Tamas Revesz Queen Square Brain Bank Reta Lila Weston Institute of Neurological Studies UCL Institute of Neurology University College London London UK Marie Rivera-Zengotita Department of Pathology, Immunology, and Laboratory Medicine University of Florida College of Medicine Gainesville, FL USA Achira Roy Center for Integrative Brain Research Seattle Children’s Research Institute Seattle, WA USA Mariarita Santi Department of Pathology and Laboratory Medicine Children’s Hospital of Philadelphia University of Pennsylvania Perelman School of Medicine Philadelphia, PA USA Joseph R. Siebert Department of Laboratories Seattle Children’s Hospital Departments of Pathology and Pediatrics University of Washington Seattle, WA USA Colin Smith Academic Department of Neuropathology Centre for Clinical Brain Sciences University of Edinburgh Edinburgh UK Werner Stenzel Department of Neuropathology Charit´e Universit¨atsmedizin Berlin Berlin Germany

List of Contributors

Kinoko Suzuki Department of Pathology and Laboratory Medicine University of North Carolina Chapel Hill, NC USA

Department of Neurology David Geffen School of Medicine at UCLA and Ronald Reagan UCLA Medical Center Los Angeles, CA USA

Randy Tashjian Section of Neuropathology Department of Pathology and Laboratory Medicine David Geffen School of Medicine University of California Los Angeles, CA USA

Karen M. Weidenheim Department of Pathology (Neuropathology) Montefiore Medical Center Albert Einstein College of Medicine Bronx, NY USA

Maria Thom Departments of Neuropathology and Clinical and Experimental Epilepsy UCL Institute of Neurology London UK Enza Maria Valente Section of Neurosciences Department of Medicine and Surgery University of Salerno Fisciano Italy Marjo S. van den Knaap Department of Child Neurology VU University Medical Center Amsterdam The Netherlands

Sarah Wiethoff Department of Molecular Neuroscience UCL Institute of Neurology University College London London UK; Center for Neurology and Hertie Institute for Clinical Brain Research Eberhard-Karls-University T¨ubingen Germany

Krystina E. Wisniewski Deceased Anthony T. Yachnis Department of Pathology, Immunology, and Laboratory Medicine University of Florida College of Medicine Gainesville, FL USA

Harry V. Vinters Section of Neuropathology Department of Pathology and Laboratory Medicine David Geffen School of Medicine University of California Los Angeles, CA USA;

xi

Introduction

In the preface to the first edition of this book, published in 2004, we remarked that the extraordinary advances in understanding the molecular and cellular basis of neurodevelopment had shepherded in a new era for neuropathology. We considered that to have been an opportune time to advance this book in a series devoted to pathology and genetics. This premise stands even stronger today. Over the past nearly decade and a half, new discoveries in metabolic pathways, molecular genetics, and developmental biology have again leapfrogged our knowledge regarding numerous inherited and acquired disorders of the developing and immature nervous system. We hope this second edition will continue to be a useful guide to practicing neuropathologists, pediatric pathologists, general pathologists, and neurologists in deciphering the pathology and pathogenesis of the usually complex disorders affecting the nervous system of the embryo, fetus, and child. In this era of molecular diagnostics, advanced imaging, and prenatal screening, it is true that the neuropathologist is being asked to evaluate many fewer children with nervous system

abnormalities described in this text. This is precisely why this book is needed; to help to guide those individuals who are asked to evaluate these infrequently seen cases that are so important to understand for families and future decisions they will have to make. We have once again been assisted by an astounding cadre of international experts who have provided succinct and wellorganized chapters. The wealth of knowledge in this field has once again necessitated some selectivity and brevity to maintain short and easily navigable chapters. Given the continued pace of discovery in the fields relevant to this book, we anticipate that new understandings are likely even at the time of this publication. We hope that this book provides a practical guide to diagnosing and understanding disorders affecting this vulnerable population and potentially stimulates further advances in this exciting area.

It has been a great pleasure working on this second edition of the ISN series “Developmental Neuropathology, Pathology and Genetics” which has been expertly planned and executed by our colleagues and friends, Homa Adle-Biassette, Brian Harding and Jeffrey Golden. The contributors from several continents make it truly international, and we wish to thank everyone for their patience with unforeseen delays, and for their generosity in providing their time and sharing their material for this

book. We also acknowledge those authors, some now retired, who contributed to the first edition. Special thanks to Claire Bonnett, Prerna Sanjay, and Atiqah Abdul Manaf of John Wiley, for technical expertise and help with production.

Homa Adle-Biassette Brian Harding Jeffrey Golden

Franc¸oise Gray Catherine (Katy) Keohane ISN Book Series Co-Editors

xiii

1

Central Nervous System Manifestations of Chromosomal Change Joseph R. Siebert Department of Laboratories, Seattle Children’s Hospital and Departments of Pathology and Pediatrics, University of Washington, Seattle, WA, USA

Introduction A wide variety of morphologic and functional abnormalities have been identified in the central nervous systems (CNS) of patients with chromosomal defects. These are reviewed for the more commonly encountered karyotypes, with emphasis given to those aberrations in chromosome number (e.g., trisomy, monosomy) or chromosome morphology (i.e., large deletions and duplications) that affect the CNS. Disorders associated with mosaicism, lesser chromosomal changes (including translocations), or single gene mutations are not included. Chromosomal changes are encountered in early pregnancy loss, but their true incidence is hard to determine. A commonly used estimate is 50%. Alterations like trisomy 16 (estimated to occur in 1% of all conceptuses) are unlikely to come to the attention of neuropathologists because of early fetal demise. The prevalence of chromosomal defects is summarized for common CNS anomalies in Table 1.1. Because of genetic mechanisms (e.g., incomplete penetrance and variable expressivity) and other factors, phenotypes do not always correlate precisely with specific karyotypes. For this reason, the tables in this chapter are limited to general summaries.

The craniofacial complex While this chapter is oriented toward the description of changes in the CNS, the cranium can scarcely be ignored. The embryogenesis of brain and cranium proceeds in tandem and anomalies of one structure are almost always reflected by changes in the other. The pathologist whose study of the CNS is hampered by severe autolysis, commonplace in stillbirth, or delivery by dilatation

and evacuation, does well to examine the cranium to get clues to CNS pathology [1]. Two examples are anencephaly and holoprosencephaly. In the former, the cranial base is markedly flattened, much of the calvaria is absent, and sphenoidal anomalies are common. In specimens altered by holoprosencephaly, the anterior cranial fossa is flattened, cribriform plates are small, obscured, or absent, the crista galli is reduced in size or absent, and the anterior falx cerebri is absent or hypoplastic (Figure 1.1). Axial anomalies are especially common in trisomy 13 and 18. Identifying any of these changes is of great help in achieving a diagnosis, or at least in attempting to corroborate prenatal imaging studies.

Genetic counseling and the neuropathologist Affected individuals and/or their families desire to understand both present and future issues surrounding their condition. Families are concerned about implications for present and future care, as well as prevention, recurrence risk, and family planning. Neuropathologists should be integral members of the team that provides information to patients and their families. Specialists will serve their patients well by providing information that is of direct use to the genetic counselor. In addition to written reports, the pathologist should provide photographs, especially of external phenotypic features (face, head, hands, and so forth), that have diagnostic value. Ethical ramifications are considerable, but beyond the scope of this review.

Autosomal trisomy A trisomic cell is defined by the presence of three homologous chromosomes. The condition is serious and often associated

Developmental Neuropathology, Second Edition. Edited by Homa Adle-Biassette, Brian N. Harding, and Jeffrey A. Golden. © 2018 John Wiley & Sons Ltd. Published 2018 by John Wiley & Sons Ltd.

1

Developmental Neuropathology

Table 1.1 Prevalence of chromosomal abnormalities in common central nervous system (CNS) anomalies.

CNS anomaly

Estimated Prevalence of Abnormal Karyotype (%)

Anencephaly Holoprosencephaly Dandy-Walker malformation Complete vermian agenesis Inferior vermian agenesis Chiari malformation Microcephaly/micrencephaly

9 50 45–55 32 53 Occasional Frequent

Isolated ventriculomegaly Agenesis of corpus callosum Spina bifida/myelomeningocele Coloboma

12 18 17 Often unspecified; 10% in 1993 study 25+

Micro-/anophthalmia

with prenatal demise, or live birth with multiple anomalies, some of which are life-threatening. Females with trisomy 13 or 18 have severe ovarian dysgenesis, resulting germ cell failure, and cannot reproduce, should they survive to reproductive age [2]. Women with trisomy 21 who become pregnant give birth to infants with trisomy 21 in about one-third of cases; affected males are sterile. Trisomic conditions with associated changes of the craniofacial complex and CNS are described below and summarized in Table 1.2.

Trisomy 8 Complete trisomy 8 is observed in first-trimester terminations of pregnancy and rarely thereafter. By contrast, survival to term

Predominant Karyotypic Abnormality

Reference

Variable, including trisomy 13, 18, and triploidy Trisomy 13, 18 Trisomy 9, 13, 18, 21; triploidy; del(6p)

26 27 28 28 28 26 26

Trisomy 13, 18 Trisomy 9, 13, 18, 21; sex chromosome trisomy or monosomy Trisomy 21; 47XXY Trisomy 8, 13, 18; del(11q) Trisomy 13, 18; triploidy Trisomy 13, 18; triploidy; 5p-; 4p-

29 26,30 31 32

Trisomy 13

33

or beyond with mosaic trisomy 8 is more common. The severity of phenotypic change does not appear to depend upon the percentage of trisomic cells, and thus, the phenotypes of complete and mosaic trisomy 8 are similar. The frequency is estimated to be between 1:25 000 and 1:50 000 live individuals; males are five times more commonly affected. Patients may manifest psychomotor restriction, seizures, or personality disorders; they have dysmorphic facies, with prominent forehead, widely spaced and deeply set eyes, broad nasal root, micrognathia, thick lips, and large protuberant ears [3]. Agenesis or hypoplasia of the corpus callosum is the chief alteration of the CNS; spina bifida occulta is observed, as well as a number of less common malformations.

Trisomy 9 Most newborns with trisomy 9 die in the perinatal period. Survivors have mental and motor deficiencies, and fail to thrive. Variable degrees of mosaicism are thought to modulate the severity of changes noted in the condition. The CNS is abnormal most of the time, and most consistently shows a Dandy-Walker malformation, although it has been well characterized morphologically in only a few cases [4]. Craniofacial changes may be nonspecific or those associated with holoprosencephaly.

Figure 1.1 Close view of cranial base, showing anterior and middle cranial fossae of patient with holoprosencephaly and trisomy 13. Note absence of ethmoid derivatives (crista galli, cribriform plates) and falx cerebri. In cases of ocular hypotelorism, the basisphenoid and sella turcica may be narrowed.

2

Trisomy 13 Trisomy 13 is the third most common autosomal trisomy, with a prevalence variably reported as 1:5000 to 1:29 000 live births [5]. Like other aneuploid conditions, the spontaneous death rate is increased dramatically, both prenatally and perinatally; mean postnatal survival is 2.5–4 days. Diploid–aneuploid mosaicism confined to the placenta may affect intrauterine survival, and, inexplicably, mothers often suffer preeclampsia, which may contribute to spontaneous pregnancy loss [6].

Central Nervous System Manifestations of Chromosomal Change Chapter 1

Table 1.2 Craniocerebral findings in selected autosomal aneuploid conditions. Karyotype

Craniofacial Findings

Morphologic and Functional Abnormalities of CNS

Trisomy 8

Scaphocephaly; prominent forehead; ocular hypertelorism; deeply set eyes; bulbous nose; thickened, everted lower lip; high or cleft palate; low set, dysplastic ears; micrognathia

Trisomy 9

Microcephaly, dolichocephaly; widening of cranial sutures; deep-set eyes; small palpebral fissures; bulbous nose, with broad base; low set, anomalous ears; micrognathia

Trisomy 13 (Patau syndrome)

Microcephaly; trigonocephaly; holoprosencephalic facies (e.g., cleft lip/palate; cebocephaly; ethmocephaly; cyclopia); cutaneous scalp defects; dysplastic ears

Trisomy 18 (Edwards syndrome)

Microcephaly; ‘strawberry-shaped’ skull; broad, high forehead; low set, dysplastic ears and temporal bone anomalies; small nose and mouth; choanal atresia; cleft lip/palate; micrognathia

Trisomy 21 (Down syndrome)

Incomplete ossification of calvaria; hypertelorism; cleft lip and/or palate; micrognathia; low set, malformed ears

Triploidy (69,XXX or 69,XXY)

Incomplete ossification of calvaria, with enlarged fontanelles; facial dysmorphism associated with holoprosencephaly; ‘Harlequin’ orbits; low nasal bridge; choanal atresia; preauricular malformations; cleft lip/palate

Tetraploidy (92,XXXX or 92,XXYY)

Microcephaly; prominent, narrow forehead; low set ears; preauricular ear tags; beaked nose; micrognathia; cleft lip/palate

Variable psychomotor restriction; personality disorder/psychosis; agenesis of the corpus callosum; spina bifida occulta; less frequent CNS changes: aqueductal stenosis; abnormal falx cerebri; Dandy–Walker malformation or isolated hypoplastic cerebellum; eye changes (variable): microphthalmia; strabismus; coloboma; corneal/lens opacity; glaucoma Growth restriction; Dandy–Walker malformation; holoprosencephaly; lissencephaly; ventriculomegaly/ hydrocephalus; agenesis of corpus callosum; hypoplasia of septum pellucidum; cortical migration anomalies (including subpial glial nodules, pachygyria); simplified inferior olivary nuclei, abnormal hippocampal formation; germinal matrix cysts; syringomyelia; myelomeningocele; eye changes: anophthalmia; retinal or iridal coloboma; hypoplasia of optic nerves Holoprosencephaly (most common); Chiari malformation; hypoplastic cerebellum; neuronal migration defects; spina bifida; eye changes: microphthalmia; coloboma; retinal dysplasia; aniridia; hypoplastic optic nerve Partial or complete absence of the corpus callosum; Chiari malformation; neural tube defects (myelomeningocele, anencephaly); holoprosencephaly; neuronal migration defects; choroid plexus cysts; eye changes (less common): coloboma; cataract; cloudy cornea; retinal hypopigmentation; microphthalmia; iridal hypoplasia Hydrocephalus, holoprosencephaly, hypoplasia of basal ganglia, cerebellum, occipital lobes, or other focal structure; lumbar myelomeningocele; less often: Chiari or Dandy–Walker malformation, hydranencephaly. Eyes: ocular coloboma, microphthalmia, anophthalmia, corneal clouding, cataract Intrauterine growth restriction; all varieties of holoprosencephaly; hydrocephalus; hydranencephaly; hypoplasia of basal ganglia, occipital lobes, or cerebellum; choroid plexus cysts; cavum septi pellucidi; less common: Chiari or Dandy–Walker malformation; eye changes: micro- or anophthalmia; proptosis; hypoplasia of iris or retina; coloboma; corneal clouding; cataract Severe pre- and postnatal growth restriction; severe mental restriction; hydrocephalus; Chiari malformation; myelomeningocele; arhinencephaly; eye changes: micro- or anophthalmia; cloudy cornea; coloboma

CNS, central nervous system

Phenotypic changes are well recognized (Table 1.2) and involve the CNS (Figures 1.2, 1.3, 1.4), craniofacial complex, axial skeleton, and multiple extracranial tissues. The most common CNS manifestations among this group are holoprosencephaly and anencephaly.

Trisomy 18 The incidence of trisomy 18 is given as 0.3 per 1000 live births, with a female to male ratio of 3:1. Cranial and brain

abnormalities are common, as are eye malformations, axial, hypophyseal, and extracranial anomalies [3].

Trisomy 21 The literature on Down syndrome and descriptions of CNS changes are voluminous, although probably not entirely reliable, in that older cases were not confirmed by karyotyping. Because phenotypic variability is substantial, diagnosis is not always possible by physical examination, especially in the

3

Developmental Neuropathology

(a)

(b)

Figure 1.2 Two term infants with trisomy 13. (a) Superior view of calvaria (after reflection of scalp) altered by trigonocephaly and partial metopic craniosynostosis. (b) Basal view of deformed brain from patient with trigonocephaly; note absence of olfactory tracts (arhinencephaly), asymmetric optic nerves, and dysplastic cerebellar folia.

(a)

(b)

Figure 1.3 Infant with trisomy 13. (a) Basal view of brain with holoprosencephaly. (b) superior view of same brain. Note fusion of frontal lobes and lateral ventricles.

4

Central Nervous System Manifestations of Chromosomal Change Chapter 1

Other autosomal aneuploidies

Figure 1.4 Coloboma is encountered frequently in patients with chromosomal abnormalities. Iridal colobomata are shown from a newborn infant with classic findings of trisomy 13.

prenatal period. The condition is rather common, with estimates generally given at about 1.3 per 1000 live births. About 95% of patients have 47,+21 karyotypes, while the remainder are mosaic or manifest unbalanced translocations (mostly Robertsonian). The cranium is round (brachycephaly) and the brain likewise, with foreshortened frontal poles and a flattened occiput; the superior temporal gyrus is often small and straight (Figure 1.5). Brain weight is usually reduced by 20–25% after the first 2 years. A variety of dendritic abnormalities have been identified. The most consistent findings are reduced complexity and numbers of dendritic branches and spines after 1–2 years of age. Patients suffer early dementia leading to Alzheimer-type changes in the brain.

Figure 1.5 Lateral view of brain of newborn infant with trisomy 21. Note mild blunting of frontal lobe and abnormally small superior temporal gyrus, common findings in this condition. The middle temporal gyrus is enlarged.

Triploidy The most common chromosomal abnormality observed in firsttrimester spontaneous miscarriages is a complete, supernumerary set of chromosomes, or triploidy, occurring in 12% of all such fetuses. Affected individuals (69,XXX or 69,XXY), who may be mosaic or manifest complete trisomy, survive occasionally to term, then die in the immediate postnatal period. Increased nuchal translucency and characteristic placental abnormalities (enlarged placenta with hydatidiform degeneration, or a small placenta in cases of digyny (i.e., a diploid ovum fertilized by a monoploid sperm), are noted by prenatal ultrasound. Intrauterine growth restriction is also common, as are a wide variety of well-known craniofacial, CNS (Figures 1.6 and 1.7), and extracranial anomalies. Syndactyly of the third and fourth fingers, or first and second or third and fourth toes, occurs in 50% and 30% of individuals, respectively, and should compel

Figure 1.6 Median section of cerebellum and brainstem altered by Chiari malformation. Note small size of cerebellum and caudal herniation of the cerebellum and medulla, unrolled posterior medullary vellum, beaking of colliculi, and herniation of brainstem; the fourth ventricle is nearly obliterated.

5

Developmental Neuropathology

(a)

(b)

Figure 1.7 Dandy–Walker malformation. (a) In situ demonstration of absent cerebellar vermis. (b) Horizontal section of cerebellum, with midline space representing absence of vermis.

the pathologist to obtain a karyotype. Tissue other than blood should be cultured, as triploid cells are eliminated selectively from lymphocytes. Flow cytometry is an efficient way of reaching a diagnosis, in that the quantity of DNA measured by the process will be increased by 50%.

Tetraploidy Like triploidy, tetraploidy (92,XXXX or 92,XXYY) is also associated with early loss. Together, the two conditions account for 30% of karyotypically abnormal spontaneous miscarriages. Growth and mental restriction and a wide variety of craniofacial and CNS anomalies are found in affected patients.

Sex chromosome aneuploidy Patients with alterations in the number of X or Y chromosomes oftentimes manifest normal development. However, an increased risk for gonadal dysgenesis and decreased fertility are also recognized, as is developmental delay involving speech, motor abilities, and learning. Phenotype, including behavior, is affected in patients with sex chromosome aneuploidies, but the severity does not correlate with the magnitude of aneuploidy. Mental restriction, schizophrenia, and bipolar disorders have all been associated variably with sex chromosome aneuploidy. It is impossible to assign prognosis with certainty. Although specific neuropathological changes are not well defined, sporadic anomalies are reported and summarized in Table 1.3.

citations. Deletions may appear de novo as isolated anomalies, may result from de novo inversions, or may be inherited as familial translocations or inversions. The magnitude of phenotypic or functional deficit may or may not correlate with the size of deletion. The recurrence risk for deletion syndromes in siblings is generally negligible, unless a parent is a translocation carrier. Offspring of balanced translocation carriers may inherit balanced translocations or deletion (or duplication) syndromes. Carriers of balanced translocations do not, as a rule, have severely altered phenotypes. Affected patients, if able to reproduce, may transmit deletions. The absence of genetic material has, in some instances, compelled researchers to hypothesize haploinsufficiency as a pathogenetic mechanism. Clearly, knowledge of breakpoints and exact genes that are lost is important to understanding genotype/phenotype correlations. The contribution of subtelomeric deletions to CNS development and function is important. Patients with terminal deletions may exhibit phenotypes that differ from those with interstitial phenotypes. General statements regarding the more common deletion syndromes, with anatomic and functional details, are provided in Table 1.4.

Deletion 3pPatients with 3p- deletions are rare and have an equal sex ratio. Growth restriction and developmental delay are major findings, and, like patients with many deletions, survival depends upon the severity of anomalies. The gene MEGAP is lost in 3p- and this gene is thought to play an important role in cognition, learning, and memory, presumably by regulating the cytoskeleton, axonal branching, and neuronal migration [7].

Deletions Deletions arise from a variety of mechanisms. They involve absence of the terminal or interstitial regions of the chromosome, leaving a haploid DNA content for the affected segment, and for this reason were referred to as “monosomy” in older

6

Deletion 4pPatients with deletions of the short arm of chromosome 4 (WolfHirschhorn syndrome) occur infrequently (1:50 000 live births), with a female: male ratio of 2:1. Infants may manifest intrauterine growth restriction and hypotonia at birth; 35% die in the

Central Nervous System Manifestations of Chromosomal Change Chapter 1

Table 1.3 Craniocerebral findings in selected sex chromosome aneuploidies. Karyotype

Craniofacial Findings

Morphologic and Functional Abnormalities of CNS

45, X0 (Turner syndrome or Ullrich-Turner syndrome) 47,XXX (triple-X or trisomy X)

Facies without expression; low posterior hairline; ocular hypertelorism; downslanting palpebral fissures

Pre- and postnatal growth restriction; eye changes (frequent): strabismus; ptosis; learning difficulties, including visual-spatial and perceptive deficits; psychiatric impediments, but no psychopathology Poor motor skills; learning disabilities and language delay in some patients; mental restriction possible, but highly variable

47,XYY

Often phenotypically normal (tall stature common); uncommon dysmorphic features: ocular hypertelorism; epicanthal folds; depressed nasal bridge Often phenotypically normal (tall stature common); increased craniofacial dimensions by cephalometry

47,XXY (Klinefelter syndrome)

Feminization; tall stature

48,XXXX (tetra-X s. or tetrasomy X)

Tall stature; dull, flat facial expression; ocular hypertelorism; flat nasal bridge; epicanthal folds

49,XXXXX (penta-X s. or pentasomy X)

Microcephaly; coarse facies, with epicanthal folds; upslanting palpebral fissures; broad, depressed nasal bridge; ocular hypertelorism; low set, malformed ears; low posterior hairline; cleft palate; micrognathia Microcephaly; plagiocephaly; trigonocephaly; many traits shared with pentasomy X, including: epicanthal folds; upslanting palpebral fissures; ocular hypertelorism; midface hypoplasia; low nasal bridge; cleft soft palate; micrognathia

49,XXXXY

Learning disabilities, including speech delay and autism; behavioral delay is inconsistent; hyperactivity; increased risk for schizophrenia or bipolar disorder; intention tremor; hypotonia; hypoplastic or absent corpus callosum; ventriculomegaly; Dandy–Walker malformation Normal intelligence quotient, but reading deficiency and poor spelling in two-thirds of patients; reduced control of impulses and poor motor coordination; dyslexia, attention-deficit disorder, or psychiatric disorders, including psychosis, in some Intrauterine growth restriction; mild to moderate mental restriction; emotional disturbances; eye changes: strabismus, iridal coloboma; myopia Growth restriction (uncommon); severe psychomotor restriction

Intrauterine and postnatal growth restriction; delayed ossification; mental restriction; psychomotor delay; psychiatric/personality disorders, including cognitive impairment; aggressiveness, self-inflicted injury; severely delayed or absent speech

CNS, central nervous system

first 2 years of life. Survivors have seizures that can be constant and severe psychomotor restriction; they have been described as being without personality. It is possible, however, that the condition is more common than previous estimates. It has probably been misdiagnosed at times (for example, midline scalp defects, facial clefts, and coloboma in a subset of patients may be confused with trisomy 13) and only about one-half are recognized by routine banding techniques [8]. Thus, prognosis may not be as poor as believed previously. Only a few adults have been described, but accurate diagnosis facilitates appropriate interventional care and counseling.

Deletion 5pCommonly known as cri-du-chat (“cat cry”) syndrome, deletion 5p- is one of the most common deletion syndromes, occurring in 1:15 000 to 1:50 000 live births and constituting nearly 1% of all institutionalized patients; a slight female preponderance is recognized, with a male to female ratio of 0.72 [3]. Findings evolve as patients age. Infants exhibit a high-pitched or mewing, cat-like cry, which disappears in childhood as the orientation of

posteriorly approximated vocal cords changes. Similarly, the rounded face becomes thinner and more asymmetric. Speech and language development are delayed, sometimes severely.

Deletion 9pDeletion 9p- is well defined clinically, with over 100 cases published; a female predilection of 2:3 to 3:4 has been reported [3]. Nonspecific facial dysmorphia, hypotonia (or occasional hypertonia), psychomotor and mental restriction are common, and a particular neurobehavioral profile is recognized. The eyes can slant either upward or downward; infants with the former can be confused with patients with trisomy 21. As with del (5p), facial abnormalities become less obvious with age. Deletion 11qPatients with deletions of the long arm of chromosome 11 (Jacobsen syndrome) manifest psychomotor restriction and a variety of nonspecific changes of the craniofacial complex. Some 32 cases have been reported; the condition has a strong female preponderance. Over one-half of patients manifest a variety of

7

Developmental Neuropathology

Table 1.4 Craniocerebral findings in selected deletion syndromes. Deletion

Craniofacial Findings

Morphologic and Functional Abnormalities of CNS

3p-

Microcephaly; dolichocephaly; synophrys; angulated face; ocular hypertelorism; short, broad nose; low set, dysplastic ears; micrognathia Also known as Wolf–Hirschhorn syndrome; microcephaly; dolichocephaly; midline scalp defect; cleft lip and/or palate; ‘Greek helmet’ facies, with ocular hypertelorism; frontal bossing; hypoplastic orbits; epicanthal folds; beaked nose with broad ridge; midline scalp defect; preauricular dimples; small mouth; dental anomalies Microcephaly; abnormal cranial contour; ocular hypertelorism; epicanthal folds; depressed nasal bridge, with anteverted nares; Robin sequence, with cleft lip/palate and retro- or micrognathia; low set, posteriorly rotated, pointed ears Also known as cri du chat syndrome; microcephaly; round face, with ocular hypertelorism and low set, posteriorly rotated ears that evolves after infancy to a thin, asymmetric face; epicanthal folds (more common in young); wide, depressed nasal bridge; micrognathia Trigonocephaly; flattened occiput; depressed temples; ocular hypertelorism, with variably slanted eyes; exophthalmos; microphthalmia; epicanthal folds; low set ears with dysplastic or aplastic lobes; choanal atresia; short nose with flattened bridge; midface hypoplasia, with small mouth; cleft palate; long philtrum Also known as Jacobsen syndrome; trigonocephaly; telecanthus; ocular hypertelorism; broad, depressed nasal bridge; low set, dysplastic ears; micrognathia; palatal anomalies

Pre- and postnatal growth restriction, with severe developmental, psychomotor, and speech delay; ptosis

4p-

4q-

5p-

9p-

11q-

13q-

Microcephaly; high, broad forehead; large, low set ears; depressed nasal bridge; protruding maxilla; micrognathia

18p-

Facial changes may be non-specific (round face, ocular hypertelorism, large ears, broad nasal bridge, upturned nostrils, micrognathia) or those associated with holoprosencephaly (see text); also micro- or brachycephaly; craniosynostosis; choanal stenosis, cleft palate Microcephaly; deeply set eyes; narrow/atretic ear canals; frontal bossing; midface hypoplasia; prominent chin; small ‘carp’ mouth

18q-

21q-

Microcephaly; holoprosencephaly; downward slanting palpebral fissures; prominent nasal bridge; large, low set ears; cleft lip/palate; micrognathia

CNS, central nervous system

8

Postnatal growth restriction; severe psychomotor restriction; seizures; eye changes: ptosis, strabismus, nystagmus, coloboma

Postnatal growth restriction; mild to moderate mental restriction; hypotonia; seizures; sensorineural hearing loss and oculomotor nerve palsy (latter findings in one patient) Severe growth and mental restriction; hypotonia in infancy changes over time to hyperactive reflexes; ventriculomegaly; strabismus; tortuous retinal vessels/optic atrophy in adults

Developmental delay, with variable degree of psychomotor and mental restriction; both hypotonia and hypertonia reported; seizures; autism; hypoplastic corpus callosum; large cisterna magna

Psychomotor restriction, with severe speech impairment; micrencephaly, with white matter deficiency (supratentorial); ventriculomegaly, with or without spina bifida (latter infrequent); cerebral atrophy; agenesis of corpus callosum; eye changes: iris coloboma; strabismus Severe mental restriction and developmental delay; hypotonia; hydrocephalus; hydranencephaly; holoprosencephaly; agenesis of the corpus callosum; neural tube defects (lumbosacral myelomeningocele); eye changes: ptosis; microphthalmia; coloboma (iris or choroid); corneal opacity; cataract; retinoblastoma (in 18% of patients) Variable mental restriction; diabetes insipidus; holoprosencephaly; hypotonia; spinal muscular atrophy; aphasia/dysphasia; eye changes: ptosis, strabismus, nystagmus, microphthalmia, coloboma, cloudy cornea, cataract; hearing loss Hypotonia, with classic ‘froglike’ position of legs; mild to severe mental restriction; behavioral changes (see text) and seizures; ventriculomegaly secondary to volume loss and white matter disease (involving basal ganglia and/or thalami); holoprosencephaly; thinning of corpus callosum; hypotonia; hearing impairment; eye changes: strabismus, nystagmus, glaucoma, tapetoretinal degeneration, bilateral optic atrophy Variable growth and mental restriction; hypo- or hypertonia; sensorineural hearing loss; cortical atrophy; hypoplasia of cerebellum/brain stem; agenesis of corpus callosum; ventriculomegaly; eye changes: Peters anomaly of iris; microphthalmia; cataract

Central Nervous System Manifestations of Chromosomal Change Chapter 1

CNS abnormalities; some are structural, but others appear to represent deficient or delayed white matter formation [9]. A host of extracranial malformations are also recognized.

Deletion 13qDeletion 13q- is often lethal in early gestation. Surviving patients are rare, presenting in an estimated 2 in 100 000 births, with a gender ratio of 1:1 [10]. Growth restriction is common; mental restriction is moderate to severe [11]. Holoprosencephaly (ZIC2, mutated in some cases of holoprosencephaly, maps to 13q), hydranencephaly, and neural tube defects have been reported, and retinoblastoma is recognized commonly [12]. Deletion 18pOver 150 cases of deletion 18p- have been reported, with a male to female ratio of 2:3 [3]. Affected individuals show variable degrees of developmental delay and mental restriction, with consistent facial dysmorphism in infants and adults. Extracranial anomalies are variable as well, and include cardiac defects and endocrine dysfunction [13]. Patients with holoprosencephaly may exhibit any of the associated facial changes, including mild facial changes, cleft lip and palate, cebocephaly, and cyclopia. Of note, TGIF, one of the genes mutated in holoprosencephaly, maps to 18p. Focal or generalized dystonia and hypokinesia are observed, with distal spinal muscular atrophy reported. Deletion 18qDeletion 18q- is one of the most common deletion syndromes (without sex predilection), and is manifest by tapered digits, facial dysmorphism, including deeply set eyes, dysplastic ears, and small, rounded, “carp” mouth. Hypotonia and growth failure are common. Decreased growth hormone production in patients suggests that a gene on 18q is involved in hormone production [14]. Correlated with this is the report of hypoplasia of the anterior pituitary gland [15]. Fifty to eighty percent of patients have sensorineural or conductive hearing loss, associated with malformations of the external and middle ears [16]. Intelligence is mildly to severely deficient, and behavioral problems include hyperactivity, aggressiveness, and temper tantrums; autism occurs in some patients, but probably not with increased incidence. By magnetic resonance imaging, white matter abnormalities involve periventricular and deep white regions (more pronounced in parieto-occipital areas), internal and external capsules, centrum semiovale, corona radiata, and subcortical regions. Changes are thought to result from incomplete myelination, most probably due to a missing copy (haploinsufficiency) of the myelin basic protein gene [17]. Deletion 21qDeletion 21q- has been called the “phenotypic countertype of trisomy 21,” in that studies of affected patients who have reduced amounts of genetic material from chromosome 21 might shed light on those with increased amounts of material (i.e., those

with Down syndrome). Phenotypic variation is considerable, and it is possible that a variety of conditions, including monosomy 21, have been reported inappropriately under this designation [3]. That being said, complete monosomy 21 is very rare, and phenotypic alterations probably arise from absence of the long arm [18]. Many findings are nonspecific [19].

Duplications Pure duplications are rare. More likely, they result from unbalanced translocations, in which case, the duplication of material on one chromosome is accompanied by a deletion on another chromosome. This makes it difficult to attribute a given phenotype solely to the duplication. Selected syndromes are presented below, with specific findings provided in Table 1.5.

Duplication 3q+ Duplication 3q+ is a rare condition, known by about 40 published cases; it can resemble Cornelia de Lange syndrome. The male to female ratio is equal. A wide variety of craniofacial and CNS anomalies and functional alterations are recognized. The fingers can be held in a “trisomy 18 position” which can be confusing to the examiner [20]. The pure dup (3q) syndrome is rare, in part because most patients with the syndrome have unbalanced translocations, the deletion on another chromosome complicating analysis [21]. Affected patients have an extra copy of the KCNMB3 gene, which shows sequence similarities to regulatory subunits of calcium-activated potassium channels; because of the importance of these channels in neuronal function, it is possible that overexpression is related to the seizures and restriction observed in patients [22]. Congenital heart defects are recognized.

Duplication 9p+ Duplication 9p+ is one of the most common partial trisomy syndromes, and is also known as Rethor´e syndrome. Individuals have well-recognized features, of which ventriculomegaly and Dandy-Walker malformation (Figure 1.7) are prominent [23]. In fact, the prevalence of Dandy–Walker malformation in this condition has compelled some to suggest a dosage effect of genes located on chromosome 9 [24]. Abnormalities in neuronal migration are also recognized, including subcortical laminar (band) heterotopia or double cortex. The 100 or so cases known have occurred twice as often in females. A generalized delay in bone maturation is manifest by failure of timely closure of fontanelles and cranial sutures; however, catch-up growth often occurs [3]. Findings from magnetic resonance imaging in the single reported adult consist of underdeveloped white matter, atrophic corpus callosum, and choroidal fissure cyst [25].

9

Developmental Neuropathology

Table 1.5 Craniocerebral findings in selected duplication syndromes. Duplication

Craniofacial Findings

Morphologic and Functional Abnormalities of CNS

3q+

Microcephaly; brachycephaly or dolichocephaly; craniosynostosis or widely separated sutures, with large anterior fontanelle; hypertrichosis/synophrys/bushy eyebrows/long eyelashes; ocular hypertelorism; upslanted palpebral fissures; low set, malformed ears; broad nasal root; cleft palate; prominent philtrum; prognathism; micro- or retrognathia

4p+

Microcephaly; prominent glabella/supraorbital ridge; ocular hypertelorism; bulbous nose, with depressed nasal bridge; ear abnormalities; prominent chin Also known as Rethore´ syndrome; microcephaly; brachycephaly; delayed closure of fontanelles and cranial sutures; frontal bossing; ocular hypertelorism; deep-set, downslanting eyes; broad nose; large, low set, dysplastic ears; short, upper lip; cleft lip/palate; downturned corners of mouth

Severe growth and mental restriction; seizures; micrencephaly; polymicrogyria; hypoplastic olfactory bulbs and tracts or arhinencephaly; cerebellar hypoplasia; hypoplastic corpus callosum; neuronal migrational defects; ventriculomegaly; aqueductal stenosis; hypoplastic pons; hypoplastic pyramidal tracts; gross spinal cord anomalies, including anomalous fissures, narrowing, or shortening; eye changes: strabismus; ptosis; microphthalmia; anophthalmia; cataract; coloboma; glaucoma Psychomotor restriction; postnatal growth restriction; seizures; respiratory and feeding problems; eye changes: anomalies reported (unspecified) Mild to moderate developmental delay, especially language; variable mental restriction; hypotonia; Dandy–Walker malformation; ventriculomegaly/hydrocephalus (sometimes associated with seizures); aqueductal stenosis; polymicrogyria; subcortical band heterotopia (in one adult); underdeveloped white matter; hypoplastic corpus callosum; choroidal fissure cyst; eye changes: microphthalmia; Brushfield spots; keratoconus; strabismus; myopia; cataract; iridal coloboma; ectopic pupil/dislocated lens; optic nerve atrophy Profound psychomotor and developmental delay; pre- and postnatal restriction; eye changes: blepharophimosis; scoliosis

9p+

10q+

15q+

Microcephaly; high forehead; small palpebral fissures; flat, broad nasal bridge; small upturned nose; midface hypoplasia; prominent philtrum; cleft palate; micrognathia; small mouth Microcephaly; dolichocephaly; prominent occiput; downslanting palpebral fissures; large, low set ears; bulbous nose; long philtrum; small mouth, with micro- or retrognathia

Severe developmental delay, with postnatal growth restriction due in part to swallowing difficulties; distal 15q duplication may manifest hydrocephaly rather than microcephaly

CNS, central nervous system

Future perspective, conclusions Clearly, a wide variety of morphologic and functional deficits affect the CNS of patients with chromosomal abnormalities. These changes are recognized in some detail, and continue to be delineated. However, while some embryologic and even genetic mechanisms are beginning to be identified, most conditions are poorly understood, and little is known about genotype/phenotype correlations. It will require the continue efforts of clinical workers and researchers alike to bring understanding and, it is hoped, prevention or cures to patients and their families.

References 1. Siebert JR, Kapur RJ (2000) Congenital anomalies in the fetus: Approaches to examination and diagnosis. Pathol Case Rev 5:3–13 2. Cunniff C, Jones KL, Benirschke K (1991) Ovarian dysgenesis in individuals with chromosomal abnormalities. Hum Genet 86:552–6

10

3. Schinzel A (2001) Catalogue of unbalanced chromosome aberrations in man, 2nd ed., New York, Walter de Gruyter 4. Golden JA, Schoene WC (1993) Central nervous system malformations in trisomy 9. J Neuropathol Exp Neurol 52:71–7 5. Goldstein H, Nielsen KG (1988) Rates and survival of individuals with trisomy 13 and 18. Clin Genet 34:366–72 6. Kalousek DK, Barrett IJ, McGillivray BC (1989) Placental mosaicism and intrauterine survival of trisomies 13 and 18. Am J Hum Genet 44:338–43 7. Heath RJW, Insall RH (2008) Dictyostelium MEGAPs: F-BAR domain proteins that regulate motility and membrane tubulation in contractile vacuoles. J Cell Science 121:1054–64 8. Battaglia A, Carey JC, Wright TJ (2001) Wolf-Hirschhorn (4p) syndrome. Adv Pediatr 48:75–113 9. Mattina T, Perrotta CS, Grossfeld P (2009) Jacobsen syndrome. Orphanet J Rare Dis 4:9 10. Russell LF (1990) In: ML Buyse, ed., Birth Defects Encyclopedia, Dover, Center for Birth Defects Information Services, pp. 366–7 11. Stoll C, Alembik Y (1998) A patient with 13q- syndrome with mild mental retardation and with growth retardation. Ann Genet 41: 209–12

Central Nervous System Manifestations of Chromosomal Change Chapter 1 12. Ballarati L, Rossi E, Bonati MT et al. (2007) 13q deletion and central nervous system anomalies: Further insights from karyotype– phenotype analyses of 14 patients. J Med Genet 44:e60 13. Sebold C, Soileau B, Heard P et al. (2015) Whole arm deletions of 18p: medical and developmental effects. Am J Med Genet A 167A:313–23 14. Ghidoni PD, Hale DE, Cody JD et al. (1997) Growth hormone deficiency associated in the 18q deletion syndrome. Am J Med Genet 69:7–12 15. Bekiesinska-Figatowska M, Walecki J (2001) MRI of the hypophysis in a patient with the 18q-syndrome. Neuroradiology 43:875–6 16. Jayarajan V, Swan IR, Patton MA (2000) Hearing impairment in 18q deletion syndrome. J Laryngol Otology 114:963–6 17. Gay CT, Hardies LJ, Rauch RA et al. (1997) Magnetic resonance imaging demonstrates incomplete myelination in 18q-syndrome: evidence for myelin basic protein haploinsufficiency. Am J Med Genet 74:422–31 18. Curry CJR (1990) Chromosome 21, monosomy 21. In: ML Buyse, ed., Birth Defects Encyclopedia. Dover, MA, Center for Birth Defects Information Services in association with Blackwell Scientific. p. 390 19. Wakui K, Toyoda A, Kubota T et al. (2002) Familial 14-Mb deletion at 21q11.2-q21.3 and variable phenotypic expression. J Hum Genet 47:511–16 20. Steinbach P, Adkins WN, Caspar H et al. (1981) The dup(3q) syndrome: report of eight cases and review of the literature. Am J Med Genet 10:159–77 21. Faas BH, De Vries BB, Van Es-Van Gaal J et al. (2002) A new case of dup(3q) syndrome due to a pure duplication of 3qter. Clin Genet 62:315–20 22. Riazi MA, Brinkman-Mills P, Johnson A et al. (1999) Identification of a putative regulatory subunit of a calcium-activated potassium

23.

24.

25. 26. 27. 28.

29.

30.

31.

32. 33.

channel in the dup (3q) syndrome region and a related sequence on 22q11. Genomics 62:90–4 Chen CP, Chang TY, Shih JC et al. (2002) Prenatal diagnosis of the Dandy-Walker malformation and ventriculomegaly associated with partial trisomy 9p and distal 12p deletion. Prenat Diagn 22: 1063–6 Von Kaisenberg CS, Caliebe A, Krams M et al. (2000) Absence of 9q22-9qter in trisomy 9 does not prevent a Dandy-Walker phenotype. Am J Med Genet 18:425–8 Stern JM (1996) The epilepsy of trisomy 9p. Neurology 47:821–4 Norman MG, McGillivray BC, Kalousek DK et al. (1995) Congenital malformations of the brain. New York, Oxford University Press Siebert JR, Cohen MM, Jr, Sulik KK et al. (1990) Holoprosencephaly: An overview and atlas of cases. New York, Wiley–Liss Chang MC, Russell SA, Callen PW et al. (1994) Sonographic detection of inferior vermian agenesis in Dandy-Walker malformations: Prognostic implications. Radiology 193:765–70 Tomlinson MW, Treadwell MC, Bottoms SF (1997) Isolated mild ventriculomegaly: Associated karyotypic abnormalities and in utero observations. J Matern Fetal Med 6:241–4 Santo S, D’Antonio F, Homfray T et al. (2012) Counseling in fetal medicine: agenesis of the corpus callosum. Ultrasound Obstet Gynecol 40:513–2 Babcook CJ, Goldstein R, Filly R (1995) Prenatally detected fetal myelomeningocele: is karyotype analysis warranted? Radiology 194:491–4 Leppig KA, Pagon RA (1993) Phenotypic correlations of ocular coloboma without known cause. Clin Dysmorphol 2:322–31 Kallen B, Robert E, Harris J (1996) The descriptive epidemiology of anophthalmia and microphthalmia. Int J Epidemiol 25: 1009–16

11

2

Neural Tube Defects Andrew J. Copp1 and Brian N. Harding2 1 2

Newlife Birth Defects Centre, Institute of Child Health, University College London, UK Department of Pathology and Laboratory Medicine, Children’s Hospital of Philadelphia, Philadelphia, PA, USA

Definition Neural tube defects (NTDs) are common congenital malformations of the central nervous system and axial skeleton. They are considered here in three groups. 1. Failure of neural tube closure: Craniorachischisis denotes the almost complete absence of neural tube closure, affecting both brain and spinal cord. Exencephaly, seen only in embryos and early fetuses, describes persistently open and exposed cranial neural folds. Anencephaly results from neuroepithelial degeneration, with absence of the skull vault. Myelomeningocele (syn. open spina bifida) results from failure of closure of the spinal neural tube, most often in the lumbosacral region. The open spinal cord or neural ‘placode’ is often associated with a protrusion of meninges through the open vertebral defect (spina bifida ‘cystica’). Alternatively, the open spinal cord is a flat open lesion, without meningeal sac (myelocele). 2. Primary abnormalities of skeletal development: Encephalocele is a herniation of the brain through an opening in the skull, while protrusion of meninges from the vertebral column is a meningocele. 3. Disorders of secondary neurulation: Spinal dysraphism refers to skin-covered abnormalities of the spinal cord, usually in the low lumbar and sacral regions. These include diplomyelia, a duplicated cord, diastematomyelia, a split cord, and hydromyelia, where the central canal is distended. An associated fatty tissue deposit is termed lipomeningocele (syn. spinal lipoma), often implicated with tethering of the spinal cord.

Normal embryology Neural tube closure Neurulation is conventionally divided into primary and secondary phases. Primary neurulation begins with the induction

of the neural plate, a thickened dorsal midline ectodermal structure. The edges of the neural plate then elevate, beginning at about 17–18 days post-fertilization, defining a longitudinal neural groove that deepens with progressive elevation of the sides of the neural plate. The neural folds converge toward the midline and fuse, forming the neural tube, beginning at the future cervical/occipital boundary (designated Closure 1, Figure 2.1) on day 22. Fusion proceeds in cranial and caudal directions from this level. Studies in the mouse have shown that fusion occurs separately, soon after this initial closure, at two other sites within the developing brain. Closure 2 is situated at the forebrain/ midbrain boundary, and Closure 3 occurs at the extreme rostral end of the neural plate. Closure of the cranial neural tube then proceeds between Closures 1, 2 and 3 to complete brain neurulation. Fusion spreads simultaneously along the future spinal region from the site of Closure 1, and is completed with closure of the posterior neuropore in the upper sacral region around days 26–28 (Figure 2.1). It has been demonstrated in mice that the different types of NTDs arise from failure of the varying components of the neurulation sequence. Disruption of Closure 1 results in craniorachischisis, whereas defective Closures 2 or 3, or failure of closure of the anterior or hindbrain neuropores, leads to the varying types of exencephaly, and subsequently anencephaly. Failure of completion of closure at the posterior neuropore results in an open spina bifida (myelomeningocele). Multi-site neurulation has also been suggested to occur in human embryos, on the basis that it might explain the variation in level of the body axis affected by NTDs in different individuals [1]. However, while direct studies of neurulation-stage human embryos have confirmed the occurrence of events similar to Closures 1 and 3 (Figure 2.1), the balance of evidence suggests there is no human neurulation event corresponding to mouse Closure 2 [2]. In fact, Closure 2 is not obligatory for successful brain neurulation, even in mice: this closure point is absent from

Developmental Neuropathology, Second Edition. Edited by Homa Adle-Biassette, Brian N. Harding, and Jeffrey A. Golden. © 2018 John Wiley & Sons Ltd. Published 2018 by John Wiley & Sons Ltd.

13

Developmental Neuropathology (b)

Anencephaly (a)

Rostral neuropore 0 Craniorachischisis 3

Closure 1 (c)

Spinal dysraphism

Caudal neuropore

(d)

Open spina bifida (myelomeningocele) Figure 2.1 Events of neural tube closure in the human embryo and the main types of neural tube defects arising from disorders of neurulation. Closures 1 and 3 are sites of de novo initiation of fusion of the neural folds. Closure 2, which occurs at the forebrain–midbrain boundary in mouse embryos, is absent from human neurulation. Neural tube closure spreads between these sites with completion of closure at the rostral (or anterior), and caudal (or posterior) neuropores. The tail

bud (green shaded region) contains a bi-potential neuromesodermal precursor cell population that gives rise to the sacral and caudal neural tube, through the process of secondary neurulation. These morphogenetic events occur sequentially between days 22 and 28 post-fertilization but, for purposes of clarity, have been projected onto a drawing of a late neurulation-stage embryo (reproduced with permission from Lancet Neurology 12:799–810 [71]).

the SELH/Bc strain and yet more than 80% of embryos successfully complete brain formation. Hence, there appears to be no fundamental difference between humans and mice in the process of cranial neurulation.

initially undergoing cartilaginous differentiation followed by ossification, to form the entire bony vertebra. In both skull and vertebral column, the skeletal elements become ‘modeled’ around the already formed neural tube. Even the pattern of the skull sutures appears to be dictated by inductive interactions from the underlying dura mater [4]. Hence, if the neural tube remains open, pathologically, the skeletal elements are certain to form abnormally in consequence. This is the reason for the absent cranial vault in anencephaly and the absent dorsal vertebral elements in myelomeningocele. On the other hand, primary defects of skeletal development can also occur and, where the defect leaves an opening in the bony structure, the normally formed brain or spinal cord/meninges can herniate through, as in encephalocele or meningocele. Experimental research has been limited by a lack of models of these defects in mice; for example, there is currently no mouse model of occipital encephalocele, the most common form of congenital brain herniation seen in humans. As a result, the

Development of the skull and vertebral column The skull has a dual embryonic origin. Posterior parts of the vault and base are formed by cranial mesoderm, whereas the more anterior elements have an entirely different origin from a migratory cell population: the cranial neural crest [3]. The skull vault forms directly as ‘membrane’ bone, without the intervening step of cartilage formation, whereas the skull base is first formed as a cartilaginous model that is subsequently replaced by bone. Unlike the skull, spinal vertebral structures are formed entirely from the segmented, mesodermal somites, with no contribution from the neural crest. The sclerotomal component of each somite migrates to surround the recently closed neural tube,

14

Neural Tube Defects Chapter 2

pathogenic mechanisms responsible for brain herniation defects are poorly understood.

and human caudal development are not fundamental, but rather a matter of degree.

Secondary neurulation within the tail bud The caudal tip of the embryo, at the stage when primary neurulation is being completed, is called the ‘tail bud’ or ‘caudal eminence’, and represents the remnant of the primitive streak of the gastrulation stage embryo. This cell population contains a selfrenewing, bi-potential population of progenitor cells that generates both neural and mesodermal tissue derivatives in the lower body. Cell proliferation in the tail bud leads to growth of the sacral and coccygeal body axis. Moreover, as cells are left in the wake of the ‘retreating’ tail bud, they condense into cell masses that subsequently differentiate to form the main structures of the post-lumbar region: the secondary neural tube, notochord and somites [5]. The secondary neural tube is formed from a longitudinal cellular condensation, sometimes called the ‘medullary cord’, located dorsally within the tail bud. This solid neural precursor becomes converted to the hollow secondary neural tube through the process of ‘canalization’. In cell biological terms, this equates to a ‘mesenchyme-to-epithelial transformation’ in which the previously mesenchymal cells of the tail bud are converted to an epithelial structure whose cells have distinct apicobasal polarity with apically located junctional complexes and a basement membrane at the basal surface. As part of this process, a lumen forms in the center of the secondary neural tube primordium, and this lumen rapidly becomes continuous with the cavity of the primary neural tube. Because neural and mesodermal tissues of the low body axis have a common cellular origin from the tail bud, malformations of the sacral and coccygeal regions are often found to embrace several tissue types. Defective separation of neuroepithelial and mesodermal tissues during differentiation of the tail bud in animal models can yield a split cord. Similarly, tethering of the cord within the vertebral canal probably represents the incomplete separation of neural from mesodermal components during tail bud development. The association of low spinal lesions with sacrococcygeal teratoma and lipoma is another manifestation of aberrant differentiation of the tail bud, although why there is a particular tendency for adipose tissue to form in dysraphic conditions is not known. Development of the tail bud continues for a prolonged period in tailed animals such as the mouse. In contrast, tail development is short-lived in humans, and is followed by a resorption process in which the tail structure degenerates and is ‘reabsorbed’ into the embryonic trunk. While the entire human tail structure degenerates, most likely mediated by programmed cell death (i.e. apoptosis), the neural tube and tail gut structures of the mouse tail also degenerate. On the other hand, the notochord persists in the mouse tail, forming nucleation sites for the centra of the caudal vertebrae, which are derived from the tail somites. Therefore, as with cranial neurulation, the differences between rodent

Epidemiology Prevalence NTDs occur on average in 0.5–2 per 1000 established pregnancies (i.e. in births plus therapeutic terminations of pregnancy), with a higher frequency of NTDs among spontaneous miscarriages. Marked variations in prevalence have been recorded over the decades between different geographical and ethnic populations. For example, close to 1% of births were affected by NTDs in Northern Ireland in the 1960s, with a sharply declining prevalence toward the South-East of the UK [6, 7]. Similarly, a fivefold higher NTD prevalence (around 5 per 1000 pregnancies) was observed in a region of northern China in the 1990s, compared with a southern region of China [8]. Although these variations in frequency usually encompass all NTD types, geographical variations are known in the prevalence of particular defects. For example, the same study in China revealed particularly abundant craniorachischisis in the high prevalence region [9], while frontoethmoidal encephalocele is known to be relatively common in South-East Asia [10] but rare in Western countries. Sex distribution A marked skewing of the sex ratio toward a female preponderance (up to three females for each male affected) has been reported in several studies of upper NTDs (lesions above T12, which mainly comprise anencephaly). In contrast, low spinal defects (myelomeningocele below T12) show an even sex ratio or even a slight male preponderance [11]. Upper and lower lesions also differ in the frequency of association with malformations in other body systems: defects above T12 are often part of multimalformation syndromes, whereas low spinal lesions are more often isolated. The female excess among fetuses with high lesions could reflect a disproportionate lethality in utero among males, but this does not seem likely based on evidence from animal studies. In several strains of genetically predisposed mice, an excess of females are also affected by cranial NTDs, and this excess can be traced to the earliest stages of neurulation, when females exhibit a higher frequency of failed cranial neural tube closure [12]. The reason why female embryos are more susceptible to anencephaly than males is unknown, although one hypothesis invokes the large inactive (methylated) X chromosome, present in female but not male cells of neurulation-stage embryos. Maintaining the inactive X is suggested to be a ‘sink’ for methyl groups and associated epigenetic silencing proteins [13], limiting methylation potential and thereby placing female embryos at greater risk of anencephaly. Non-genetic risk factors Both genetic and non-genetic risk factors have been identified for NTDs. Among the non-genetic risk factors are certain

15

Developmental Neuropathology maternal disease states and maternal exposure to environmental teratogens. Poorly controlled maternal diabetes mellitus is associated with an elevated risk of many malformations, including NTDs [14]. Studies in humans have generally supported hyperglycemia as the principal cause of fetal defects in diabetic mothers [15], although animal studies show that high glucose and elevated concentrations of ketone bodies can both cause NTDs [16]. Hence, multiple factors in the diabetic milieu may be teratogenic. The increasing epidemic of obesity is highlighting a likely increased risk of NTDs in obese mothers [17], a link that may be mediated via the association of obesity with type II diabetes. Hyperthermia in early pregnancy (e.g. caused by fever or excessive sauna usage) has also been suggested as a risk factor for NTDs, initially based on animal models, and a meta-analysis of the human data reveals an approximate twofold increased risk of NTD in women exposed to hyperthermia in the first trimester of pregnancy [18]. Many teratogenic agents cause NTDs in rodents, but only a few have been demonstrated to have similar effects in human. The anticonvulsants valproic acid and carbamazepine have recognized associations with human NTDs, specifically myelomeningocele. A 20–30-fold increase in risk of NTD is associated with continuous exposure to valproic acid during the early weeks of pregnancy [19]. Another teratogen whose effects have been demonstrated in humans is the fungal product fumonisin. This was the causative factor in studies of an ‘outbreak’ of NTDs in South Texas in the 1990s, linked to fungal contamination of tortilla flour [20]. A principle that is emerging is the importance of genetic background in determining the effect of a teratogenic agent on neurulation. For example, the position of Closure 2 is affected by genetic background in mouse inbred strains and determines their susceptibility to cranial NTD. Hence, treatment with valproic acid or hyperthermia, or inheritance of the NTD-causing genetic mutation splotch [21], cause cranial NTDs with a frequency dictated by the genetic background of the inbred strain.

Clinical features Clinical presentation The most severe cranial NTDs, craniorachischisis and anencephaly, are incompatible with survival beyond birth and today are observed almost exclusively on ultrasound examination during pregnancy. In contrast, many cases of myelomeningocele (open spinal NTDs) and all spinal dysraphic conditions are compatible with postnatal survival, although in developed countries only a proportion of myelomeningocele cases proceed to birth (around 15% in the UK), as prenatal diagnosis often leads to termination of pregnancy. Survivors with myelomeningocele exhibit motor and sensory deficit below the level of the spinal lesion. The level of defect has been used as a prognostic indicator both in terms of the likely benefit of cesarean delivery before the onset of labor, which has

16

been demonstrated to be beneficial in cases of relatively low, mild lesions [22], and the outcome of surgery to close the defect in the neonatal period. Abnormalities of rectal and bladder innervation can lead to incontinence and urinary tract infections. Curvature of the vertebral column (kyphosis) is a frequent accompaniment to myelomeningocele, as is hydrocephalus, which may require cerebrospinal fluid shunting soon after birth, or even in utero. The Chiari type II (Arnold–Chiari) malformation is frequently observed with myelomeningocele, and may contribute to the hydrocephalus. This malformation includes elongation of the brain stem, and displacement of the cerebellar vermis, into the foramen magnum, together with a variety of supra-tentorial defects [23] (see Chapter 12 for further discussion).

Biochemistry and prenatal diagnosis Methods for prenatal diagnosis of open NTDs were developed in the 1970s, based on the detection of alphafetoprotein in the amniotic fluid and maternal blood, while detection of acetylcholinesterase in amniotic fluid has also been used diagnostically. Alphafetoprotein and acetylcholinesterase are components of the cerebrospinal fluid that leak out of the open brain or spinal cord, and so are only detectable in cases of open lesions such as anencephaly and spina bifida. Closed lesions including encephalocele are not associated with elevated alphafetoprotein concentrations. More recently, ultrasound methods have been established that allow detection of fetuses with NTDs with a high degree of reliability [24]. As a result of these technologies, most fetuses with NTDs are now terminated in developed countries, although rates of pregnancy termination elsewhere are more variable. Differential diagnosis Diagnostic signs have been described to aid in the definitive recognition of cranial NTDs during prenatal ultrasound examination. For example, the cranial (‘lemon’) and cerebellar (‘banana’) signs have gained widespread usage in the ultrasound detection of cranial NTDs [25]. Postnatally, the diagnosis of NTD depends on the type of defect that is present. In the case of severe lesions such as myelomeningocele, there is rarely any doubt about the diagnosis, whereas spinal dysraphism requires specialist investigations such as magnetic resonance imaging to detect the spinal cord tethering that is the hallmark of this condition.

Pathology Neural tube closure disorders Craniorachischisis This is the most severe form of NTD in which the brain and spinal cord are exposed to the surrounding amniotic fluid (Figure 2.2), resulting in degeneration of neural tissue and angioma-like formations. Whereas the neural tube is persistently open from midbrain to low spine, many cases of

Neural Tube Defects Chapter 2

(a)

(b)

Figure 2.2 Craniorachischisis, as viewed from the back (a) and top of the head (b). Note the widely spaced pedicles of the vertebrae (arrows in a), and the area cerebrovasculosa (red arrow in b).

craniorachischisis exhibit relatively well-developed forebrain and optic development. Exencephaly This corresponds to an early stage in the evolution of anencephaly, following failure of cranial neural tube closure. Exencephaly has been described rarely in human fetal pathology, probably because of the rapid degeneration of brain tissue exposed to amniotic fluid, leading to anencephaly. This degeneration phenomenon has been demonstrated directly in the retinoic acid-treated rat where an initially exencephalic appearance is converted to anencephaly by late gestation [26]. Anencephaly This common NTD has been known since Egyptian times, with specimens documented in 1761 by Morgagni. Anencephaly was also described by Geoffroy Saint-Hilaire early in the nineteenth century [27]. The calvarium is hypoplastic or absent, the base of the skull is thick and flattened, and there is a constant anomaly of the sphenoid bone resembling “a bat with folded wings” [28]. The orbits are shallow, causing protrusion of the eyes (Figure 2.3). Attached to the skull base is a dark reddish irregular mass of vascular tissue, the area cerebrovasculosa, with multiple cavities containing cerebrospinal fluid. The mass is

cystic, with a midline dorsal aperture opening to the exterior. No recognizable neural tissue can be found in the anterior and middle fossae except for the trigeminal ganglia and limited lengths of the second to fifth cranial nerves. A hypoplastic anterior pituitary is present in a shallow sella, but intermediate and posterior lobes are missing. The residual amount of brain tissue varies. If the foramen magnum is intact, a considerable proportion of the medulla is visible. Usually, the foramen is deficient, with cervical spina bifida, short neck and deformed pinnae, in which case only the caudal part of the medulla is present, together with the distal parts of the lower cranial nerves. Pons, cerebellum and midbrain are grossly absent, although fragments of cerebellar folia may be seen histologically [29]. Spinal involvement varies from nonfusion of the upper cervical arches without accompanying skull defect to complete craniorachischisis (Figures 2.2 and 2.3) [30]. Histologically, the area cerebrovasculosa includes irregular masses of neural tissue, mainly glia with some neuroblasts or neurons, ependymal, and tufts of choroid plexus and numerous thin-walled blood vessels. A resemblance to forebrain, including ventricular spaces, has been suggested by some authors [29] but denied by others [31]. Covering the cystic area cerebrovasculosa is non-keratinizing squamous epithelium that is laterally continuous with normal skin.

17

Developmental Neuropathology

(a)

(b)

Figure 2.3 Anencephaly, as viewed from the front (a) and back (b). Note the area cerebrovasculosa (red arrow in b).

The medulla may be reasonably well preserved but, as with the spinal cord, there is aplasia of descending tracts. The spinal leptomeninges are excessively vascular and contain islands of heterotopic neuroglial tissue [32]. The absence of neurohypophysis and hypothalamus is associated with a hypoplastic adrenal cortex [33]. Various other visceral abnormalities have been reported in cases of anencephaly [34], a large thymus and hypoplastic lungs being the most frequent. Polyhydramnios is observed in about half of the cases, but its pathogenesis is unclear. Myelomeningocele This is the most severe type of spinal NTD in which a fluctuant mass (spina bifida cystica) herniates through a large vertebral defect, most commonly in the lumbosacral region (Figure 2.4a). The mass consists of a distended meningeal sac filled with cerebrospinal fluid and covered by a thin membrane or by skin. Within the sac, the spinal cord may be closed, floating on the posterior surface of the arachnoid cyst, in which case the central canal is often dilated and the posterior part herniates with the meninges. In other cases, often referred to as myelocele, the malformation appears as a flat open lesion, with cerebrospinal fluid leaking onto the exposed area (Figure 2.4b). The spinal cord at the site of the bony defect forms a flat discoidal, highly vascular mass, the area medullovasculosa, which becomes epithelialized after birth. The posterior surface of the spinal cord is open and the central canal blends into the skin. Peripheral nerves end blindly within the vascular mass that comprises highly vascular connective tissue and islands of central nervous system including neurons, glia and ependyma.

18

Hydrocephalus occurs in many infants with myelomeningocele, usually in association with the Chiari type II malformation. Anomalies in the spinal cord above the myelomeningocele are common and include syringomyelia, hydromyelia, diastematomyelia, a double or multiple central canal and diplomyelia [35].

Axial mesodermal defects with herniation of the neural tube Encephalocele and cranial meningocele affect the occipital region in 75–80% of cases, with frontal and lateral parts of the skull being affected infrequently, although frontal encephaloceles are much commoner in South-East Asia. Occipital encephalocele Brain herniation occurs through the occipital bone, with or without involvement of the foramen magnum. The mass of tissue is often voluminous (Figure 2.5), is attached to one of the cerebral hemispheres by a narrow pedicle, and is partially covered with normal skin and hair. Ulceration of the skin and secondary infection are frequent. Smaller encephaloceles may contain fragments of disorganized central nervous system, glia and ependyma. Larger herniations may include large portions of the cerebral hemispheres with ventricular cavities [36,37] as well as parts of the brain stem and cerebellum. Often asymmetrical and deformed, the hemispheric cortex may be normal or show polymicrogyria [38]. Other features included distortion, displacement and asymmetry of the basal ganglia, anomalies of the hippocampi and commissural system, aberrant neural

Neural Tube Defects Chapter 2

(a)

(b)

Figure 2.4 Myelocele (a) and myelomeningocele (b) as viewed from the back. In myelocele, the lumbosacral spinal cord is open and appears as a flat vascular lesion (a), whereas in myelomeningocele, the lesion comprises a fluctuant, cystic mass protruding through the dorsally open vertebrae (b).

tissue in the cavity of ventricles, distortion of the brain stem and agenesis of cranial nerve nuclei, absence of the vermis and near or complete absence of the cerebellar hemispheres. Of note, there is an extensive plexus of thin-walled sinusoidal vessels in the leptomeninges, suggested to be a persistent fetal vasculature.

Meckel-Gruber syndrome Occipital encephalocele is an important component of the Meckel–Gruber syndrome, a lethal autosomal recessive disorder that is now recognized to be a ‘ciliopathy’. The phenotype comprises sloping forehead, occipital encephalocele, polydactyly, polycystic kidneys, and hepatic fibrosis, with bile duct

(a) (b)

Figure 2.5 Occipital encephalocele, as viewed from behind (a) and in a postmortem dissected specimen (b). The brain and meninges have partially herniated symmetrically through a defect in the occipital skull (a). Cut surface of a large surgical specimen (b) shows two attenuated occipital lobes that have herniated with their ventricular cavities (vc). The cortex varies in thickness and is partly fused (red arrow). Between brain and skin there is vascular meningeal tissue (red arrowhead) and a cystic space (yellow asterisk).

19

Developmental Neuropathology proliferation [39]. The variability of neuropathological findings among the mass of case reports has prompted two extended series in an attempt to define a consistent pattern of abnormality. From observations in 59 cases, and detailed neuropathology in ten, Paetau et al. [40] noted encephalocele in 90%, other prominent features being olfactory aplasia, midline defects, and migration disorders such as polymicrogyria. Another autopsy study of seven fetal or neonatal cases [39] revealed a consistent pattern of malformations: (1) prosencephalic dysgenesis, arhinencephaly– holoprosencephaly, and other midline anomalies; (2) occipital exencephalocele taking the form of extrusion of parts of the rhombic roof through the posterior fontanelle; (3) rhombic dysgenesis, notably supracerebellar cyst, vermal agenesis, stenosis of the aqueduct, and flattening and dysplasia of the brain stem. Parietal encephalocele or meningocele Parietal encephalocele or meningocele occurs only occasionally [41]. Deformities of the brain are usually present and are not confined to the ipsilateral hemisphere. They include asymmetry, distortion of the ventricular walls, agenesis of the corpus callosum and hydrocephalus. Anterior encephalocele Anterior encephalocele (Figure 2.6) is most commonly found at the frontoethmoidal junction. It is usually visible at the bridge of the nose (in 60% of cases) as a bulging subcutaneous

nodule and mild hypertelorism, or as a mass of brain tissue. The encephalocele may expand into the nasal cavity (30% of cases), ethmoidal or sphenoidal air sinuses, pharynx or orbit [42]. The abnormal mass may contain disorganized brain tissue or gliotic cerebral cortex. As with occipital encephalocele, the cerebral hemispheres within the intracranial cavity may be markedly skewed with nonregister of the basal ganglia and commissural anomalies. The clinical diagnosis may be difficult if only the meninges protrude through the cribriform plate of the ethmoid bone. Cerebrospinal fluid passing into the nasal cavity is indicative of a free communication between the subarachnoid space and the encephalocele. Meningocele This is usually classified as a variant of spina bifida cystica, although it seems unlikely to originate from a primary defect of neural tube closure. There is a vertebral opening combined with a cystic lesion of the back, most often in the lumbosacral region. Both the dura and arachnoid herniate through the vertebral defect, with the closed spinal cord remaining in a normal position in the spinal canal, although it may show hydromyelia, syringomyelia, diastematomyelia, or tethering. The cyst is covered by skin, which has atrophic epidermis, and lacks rete pegs and skin appendages. The wall of the cyst contains thin-walled blood vessels and islands of arachnoidal tissue, a narrow channel connecting the cyst with the vertebral canal.

Spinal dysraphism: defects of tail bud development Defects of tail bud development comprise the least severe group of NTDs, which are always skin-covered and typically come to attention due to the clinical consequences of spinal cord tethering. Although the spinal cord abnormality may be a prominent feature, there are often accompanying defects of skeletal (e.g. sacral agenesis), anorectal, and urogenital systems. The spinal cord abnormalities may comprise overdistension of the central canal (hydromyelia; Figure 2.7a), longitudinal duplication or splitting of the spinal cord (diplomyelia, diastematomyelia; Figure 2.7b) and tethering of the lower end of the cord, often in association with lipoma (lipomyelomeningocele). The defects are most often located in the low lumbar and sacral regions, broadly corresponding to the region of secondary neurulation. Since neural folding is not involved at this level, the defects probably result from a disturbance of embryonic tail bud development. Occurrence of split cord at higher levels seems most likely to reflect secondary injury to the closed neural tube from malformed vertebral elements, as evidenced by the frequent association of diastematomyelia with a bony spur. These higher defects should probably be considered a malformation of axial mesodermal differentiation.

Figure 2.6 Anterior encephalocele. The large frontoethmoidal brain herniation has severely deformed the face, with compression of orbits and nose.

20

Tethered cord Defects of the spinal cord can be associated with tethering within the vertebral canal, whatever the level of cord affected as, for

Neural Tube Defects Chapter 2

(a)

(b)

Figure 2.7 Hydromyelia and split cord (diplomyelia): (a) Hydromyelia, as observed in serial slices through the spinal cord. The expansion of the lumen is localised to the lumbar level of the cord (red arrow). (b) Split cord, observed in a haematoxylin and eosin-stained section. As the hemi-cords exist within a single dural sac, this defect is referred to as split cord malformation type II, according to the classification of Pang et al. [72]. Reproduced with permission of Oxford University Press.

example, in cervical myelomeningocele [43]. However, the ‘tethered cord syndrome’ is usually reserved for lumbosacral defects in which there are variable combinations of thickening of the filum terminale, low or dilated conus medullaris, spinal lipoma, dermoid cyst, split cord, hydromyelia, and sacral agenesis. Clinical signs associated with cord tethering include lower limb motor and sensory deficits, and neuropathic bladder. The severity of symptoms increases with age and surgical treatment is now well established.

Genetics The causation of open NTDs involves both genetic predisposition and nongenetic factors. Up to 70% of the variance in NTD prevalence is estimated to be genetic [44], as evidenced by the increased recurrence risk in siblings of index cases and in mothers with a previous affected pregnancy (2–5%) compared with the 0.1% (1 per 1000) risk in the general population. Recurrence risk rises to around 10% among women with two or more previously affected pregnancies [45]. NTDs rarely present as multiple cases in families, as would be expected of a single gene disorder, but rather a sporadic pattern is observed. Taken together with the relatively high prevalence of NTDs worldwide, this suggests a multifactorial causation in which genes interact with each other, and with nongenetic factors.

Human genetic studies The search for the genetic variants that predispose to human NTDs has centered mainly on two groups of genes: those that participate in folate one-carbon metabolism, in light of the demonstrated preventive effect of folic acid on NTDs, and those that have been demonstrated to cause NTDs in mice. Among the many genes that regulate folate metabolism, the best known genetic risk factor is 5,10-methylene tetrahydrofolate reductase (MTHFR), which encodes a key cytosolic

enzyme of folate metabolism (Figure 2.8). MTHFR converts 5,10-methylene tetrahydrofolate to 5-methyl tetrahydrofolate, providing the substrate for homocysteine remethylation. The C677T (rs1801133) polymorphism in MTHFR is associated with an approximately 1.8-fold increased risk of NTDs, although this genetic predisposition is only detected in non-Hispanic populations [46]. Apart from MTHFR, relatively few other positive associations have been reported for genes encoding enzymes of cytosolic folate metabolism and, similarly, there have been no reproducible findings with regard to folate transport genes. On the other hand, genes encoding enzymes that function in mitochondrial folate metabolism have recently been implicated in NTD causation. For example, missense (i.e. amino acidchanging) mutations in two genes of the glycine cleavage system (Figure 2.8), amino methyltransferase (AMT) and glycine decarboxylase (GLDC), have been found in a number of individuals with NTDs, but not in unaffected controls [47]. In the case of GLDC, these variants reduce enzymatic activity indicating a functional effect on folate metabolism. Mice lacking function of Amt, Gldc or Mthfd1l also develop NTDs, strengthening the likelihood that these genetic changes in patients actually are causative of the NTDs. Mitochondrial folate metabolism generates around 70% of the methyl (one-carbon) groups that enter cytosolic folate metabolism as formate (Figure 2.8). This one-carbon metabolism is key to many cellular functions, including methylation of DNA and other macromolecules and nucleotide biosynthesis for cell division [48]. Hence, genes of mitochondrial one-carbon metabolism are emerging as key requirements for neurulation in mammals.

Mouse genetic studies More than 200 different mutant genes cause NTDs in mice [49]. These mutant effects have been discovered either through studies of spontaneously arising mutations that cause NTD phenotypes, or by the introduction of mutations into specific, cloned

21

Developmental Neuropathology

Outside of cell

Cytosol Mitochondrion

Methylation reactions

SAH

SAM

Hcy

Methionine

5,10methylene THF

5,10methenyl THF

GCS THF Formate Dietary folate

5-MTHF MTHFR

Folic acid

5,10methylene THF

THF DHF

Purines & pyrimidines

5,10methenyl THF

genes through mutagenesis or via engineered embryonic stem cells (gene ‘targeting’ or gene ‘trap’ technology). The NTDcausing genes can be grouped according to the level of the body axis affected. For example, a small number of the genes are required for the initial closure event (Closure 1), with failure of closure initiation leading to craniorachischisis. A much larger group of mutants exhibits exencephaly, in which one or more of the events of cranial neurulation are defective. A third group of mutants has open spina bifida affecting the lumbosacral region, as a result of failure of closure of the posterior neuropore. Hence, each element of the neural tube closure sequence has its own distinct genetic requirements [50]. Similarly, a distinct set of genes is required for development of the secondary neural tube, and of the skull and axial skeleton, providing candidates for the causation of human spinal dysraphism, and of skull and vertebral herniation defects (e.g. encephalocele), respectively. At first sight, the existence of so many different mouse genes – each of which causes NTDs – might suggest a similar etiology to human NTDs. However, it should be noted that the great majority of mouse NTD phenotypes occur in individuals homozygous for recessive mutations, whereas mouse heterozygotes generally do not exhibit NTDs. In contrast, the high frequency but non-familial pattern of human NTDs does not suggest recessive inheritance. Rather, this pattern is consistent with multiple heterozygous gene–gene interactions, in which the genes themselves might vary from individual to individual. Gene–gene interactions have also been reported in mice [51], leading to the realization that a particular NTD phenotype (e.g. open spina bifida) can arise from several different heterozygous gene–gene combinations. When combined with environmental influences,

22

10-formyl THF

Figure 2.8 Simplified summary of folate one-carbon metabolism. Cytosolic and mitochondrial reactions move one-carbon units around the pathways, generating purines and pyrimidines for DNA synthesis and methyl groups for methylation reactions (bold dotted arrows and white boxes) as the main outputs. MTHFR (5,10-methylene tetrahydrofolate reductase) and GCS (glycine cleavage system) are two enzyme systems (grey boxes) in which mutations have been identified to increase risk of neural tube defects. Note that dietary folate and folic acid enter at different steps in the pathways (light dotted arrows). DHF, dihydrofolate; Hcy, homocysteine; 5-MTHF, 5-methyl tetrahydrofolate; SAH, s-adenosyl homocysteine; SAM, s-adenosyl methionine; THF, tetrahydrofolate.

this probably provides the best current model of human NTD genetic etiology.

Genetics of occipital encephalocele In contrast to open NTDs, occipital encephalocele is encountered most commonly as part of Meckel–Gruber syndrome. The 2000s have seen a number of causative genes identified for this syndrome, including MKS1, MKS2 (TMEM216), MKS3 (TMEM67), RPGRIP1L and CEP290 [52]. These genes encode proteins that play a key role in the structure and function of primary cilia: protrusions of the cell surface that are rooted in the centrosome and undergo a disassembly and reassembly cycle as the cell proliferates. Primary cilia are essential for signaling pathways, particularly the molecular events downstream of sonic hedgehog (Shh), and a number of rare disorders have been causally associated with genes required for ciliary structure and function, leading to the concept of ‘ciliopathies’ [53]. How encephalocele in Meckel–Gruber syndrome results from disordered ciliary function is currently unknown.

Animal models and pathogenesis The mouse provides a number of models of NTDs, based on either genetic mutations or the action of teratogenic agents. Analysis of these model systems, in comparison with normal neurulation, has revealed several aspects of the pathogenesis of NTDs.

Neural Tube Defects Chapter 2

(a)

Figure 2.9 Mouse fetuses at embryonic day (E) 15.5 illustrating the appearance of (a) craniorachischisis in a Celsr1 mutant and (b) exencephaly and open spina bifida in a curly tail (Grhl3) mutant. In craniorachischisis, the neural tube is open from midbrain to low spine (between the thin arrows in a). Exencephaly appears as an eversion of the cranial neuroepithelium which, in the curly tail fetus is restricted to the midbrain (thin arrow in b). The spina bifida affects the lumbosacral region (arrowhead in b). Note the presence of a curled tail in both fetuses (thick arrows in a and b).

Craniorachischisis In the loop-tail mutant model, neural tube closure fails to be initiated at Closure 1, leading to craniorachischisis. In contrast, the forebrain neural tube closes relatively normally in these embryos. Vangl2 is the gene mutated in the loop-tail mouse, and is known to form part of a molecular complex that mediates the so-called ‘planar cell polarity’ (PCP) signaling pathway, in which Wnt/frizzled signals are transduced by a β-catenin-independent mechanism [54]. Some other gene components of this molecular pathway, Dishevelled, Celsr1 (Figure 2.9a) and Scrb1, have similarly been found to cause craniorachischisis when mutated in the mouse. The PCP pathway is required for normal shaping of the embryo immediately prior to the onset of neurulation. The initially round or oval neural plate is converted to an elongated structure with broad cranial and narrow spinal regions. The main driving force for this neural plate shaping is ‘convergent extension’, a net medially directed movement of cells, with intercalation and rostrocaudal extension in the midline. Convergent extension fails in mice with PCP mutations, leading to short, broad embryos in which the neural folds are spaced widely apart (Figure 2.10), which prevents Closure 1 and causes craniorachischisis. Exencephaly Many mutant genes and a large number of teratogens cause cranial NTDs in the mouse, with the neural tube failing to close in the future brain (Figure 2.9b). Analysis of these genetic models has revealed several critical events in cranial neurulation (Figure 2.11) [55]. The initial elevation of the cranial neural folds coincides with a marked expansion of the cranial mesenchyme, owing to cell proliferation and increase of extracellular space. As a result, the elevating neural folds adopt a bi-convex appearance, particularly in the midbrain. Mice with loss of function of

(b)

Craniorachischisis

Exencephaly and spina bifida

the Twist, Cart1 or Hectd1 genes have cranial NTDs, in which the principal defect is a reduction in the proliferation and expansion of the cranial mesenchyme. Once the bi-convex neural folds have elevated, a second phase of cranial neurulation occurs, in which the dorsolateral aspects of the neural fold bend medially, allowing the folds to adopt a bi-concave morphology and approach the dorsal midline for fusion. This second phase is highly dependent on the actin cytoskeleton, as illustrated by embryos lacking Shroom3, which fails to close their brains. The Shroom3 gene encodes a protein that localizes apically within neuroepithelial cells and is required for the organization of apical actin microfilaments, whose contraction appears essential for this phase of closure. A third requirement for cranial closure may be initiation of cranial neural crest migration, which precedes neural tube closure in the cranial region of mammals. Mice overexpressing connexin 43 exhibit defects of both cranial neural crest emigration and exencephaly, suggesting that the delamination of neural crest cells from the neural fold apices may be required for the neural folds to bend medially. Cranial closure is closely correlated, both spatially and temporally, with programmed cell death (apoptosis), and knockout mice with either increased (e.g. AP-2α, bcl10 and Tulp1) or decreased (e.g. Apaf-1, caspase 9 and p53) apoptosis exhibit cranial NTDs. However, specific inhibition of apoptosis using chemical inhibitors in embryo culture did not prevent cranial neural tube closure in mouse embryos [56], suggesting that programmed cell death is not essential for brain closure. A further factor in cranial neurulation is the balance between cell proliferation and neuronal differentiation within the neural tube. For example, mice with mutations in components of the Notch signaling pathway show premature differentiation of the neuroepithelium and failure of brain closure.

23

Developmental Neuropathology

(a)

(b)

Node at anterior primitive streak

E7.5-8

(c)

Somites

E8.5

(d)

Primitive streak and allantois Convergent extension cell movements

Site of Closure 1

(f)

(e)

(g)

hnf cnf +/+

Lp/Lp

Figure 2.10 Pathogenesis of craniorachischisis, as revealed by studies in the mouse. (a–d) Diagram of normal development in which the neural plate undergoes shaping prior to the onset of neural tube closure, at E8.5 Views from the left side (a, b) and top (c, d) of E7.5–8.0 (a, c) and E8.5 (b, d) embryos. Anterior is to the left, and posterior to the right. At E7.5–8.0, the retreating node (red), the site of origin of midline tissues including the notochord and neural tube floor plate, is prominent at the anterior end of the primitive streak. Anterior to the node, cells are moving medially and intercalating in the midline (red arrows in c). This process, termed ‘convergent extension’, leads to an increase in embryonic length relative to width by the stage of onset of neural tube closure. By E8.5, the primitive streak occupies only the caudal part of the embryo, with a well-defined neural plate

anteriorly, flanked by five pairs of somites. The neural folds at the level of the third somite pair approach each other in the midline, to create the incipient Closure 1 site. (e–g) Scanning electron micrographs of neurulation stage embryos. The red line (e) passes through the Closure 1 site, and shows the plane of section in (f) and (g). Normal neural tube closure at this level involves midline bending to create a V-shaped neural plate (f) whereas this bending is disrupted in an embryo homozygous for the loop-tail (Lp; Vangl2 gene) mutation. A broadened midline, resulting from defective convergent extension, yields a U-shaped neural plate that cannot close owing to the wide spacing of the neural folds (g). Scale bar in (e) represents 0.25 mm (e) and 0.04 mm (f and f); cnf, caudal neural folds; hnf, hindbrain neural folds (modified from Greene et al. [73]).

Myelomeningocele A number of mouse mutants exhibit low spinal neurulation defects leading to open spina bifida (Figure 2.9b). Here, a critical event appears to be regulation of dorsolateral bending of the neural plate, for which Shh and bone morphogenetic protein (BMP) signaling are critical (Figure 2.12). Shh is produced by the notochord underlying the ventral neural plate, and inhibits dorsolateral bending by suppressing noggin production from the dorsal neural plate [57]. Noggin is required to overcome BMPmediated inhibition of dorsolateral bending. In the absence of Shh, for example in the Shh mutant mouse, dorsolateral bending occurs as a default mechanism that ensures spinal closure.

Mutants that exhibit overstimulation of the Shh signaling pathway (e.g. Ptc1, Opb) fail to close their low spinal neural tube owing to absence of dorsolateral bending [58]. In contrast, the curly tail (Grhl3 gene) mutant does not lack dorsolateral bending but apposition of the neural folds is hampered owing to ventral curvature of the caudal body axis, so that a proportion of homozygotes exhibit open spina bifida. Mouse models have aided in resolving a controversy concerning the developmental origin of high and low NTD lesions (i.e. above and below the T12 level), which differ in sex ratio and frequency of association with other malformations. One view was that high and low NTDs arose from defects of neural folding

24

Neural Tube Defects Chapter 2

(a)

Mesodermal expansion

Programmed cell death

Dorso-lateral bending points

(b) Neural crest emigration

Figure 2.11 Key developmental events in mouse cranial neurulation, each of which has been implicated in the development of exencephaly in mouse mutants. (a) Initial elevation of the cranial neural folds, and (b) dorsolateral inward bending of the neural fold tips, to achieve closure. See text for details (reproduced with permission from Copp and Greene [74].

(primary neurulation) and canalization (secondary neurulation) respectively. However, this hypothesis now appears unlikely in the light of finding that the transition from primary to secondary neurulation occurs in the upper sacral region [59]. This suggests that the vast majority of cases of myelomeningocele, both high and low, arise from disturbance of primary neurulation.

Treatment, future directions and conclusions Surgical treatment While cranial NTDs are usually lethal in the newborn or earlier, open spina bifida (myelomeningocele) is compatible with postnatal survival, provided that ascending infection leading to meningitis and encephalitis is avoided. Surgery has long been practiced postnatally, with repair of the spina bifida lesion in the neonatal period. In recent decades, surgery has been performed to cover the spina bifida while the fetus remains in the mother’s uterus [60]. A clinical trial, the Management of Myelomeningocele Study (MOMS) [61], was performed to evaluate the success of this procedure, compared with standard postnatal repair. Pregnancies before 26 weeks of gestation with prenatally diagnosed myelomeningocele were randomly assigned to either in utero surgery or postnatal repair. The trial revealed

Neuroepithelial proliferation

Contraction of sub-apical actin microfilaments

significant short-term benefits of fetal surgery for the child with spina bifida, including a 50% reduction in shunting for hydrocephalus and significantly improved neurological function. On the other hand, the in utero surgery group showed a higher rate of premature birth and maternal complications, such as uterine dehiscence at the operation site. Follow-up to determine longterm outcomes of the surgery is continuing, but the initial success of the MOMS trial is now encouraging centres in other countries to implement in utero surgery. The possible benefits of including additional biomaterials and/or neural stem cells during surgical repair is also being explored in experimental models of spina bifida. Why does surgical closure of open spina bifida in the fetal period lead to a better neurological outcome than postnatal closure? In both human and mouse spina bifida, the persistently open spinal cord undergoes relatively normal neuronal differentiation in the embryonic period, including development of spinal motor and sensory function even below the lesion level [62]. As pregnancy progresses, however, neurons die within the exposed spinal cord, probably as a result of toxicity of the amniotic fluid environment. Axonal connections are interrupted, and function is lost. Hence, early coverage of the lesion, in the fetal period, may arrest or prevent further neurodegeneration, leading to improved neurological function.

25

Developmental Neuropathology

Figure 2.12 Regulation of spinal neural tube closure by the influence of sonic hedgehog (Shh) and BMP diffusible proteins. Shh enhances midline bending (yellow triangles) but inhibits dorsolateral bending (red triangles). At upper spinal levels, Shh influence is strong and Noggin synthesis in the dorsal neural plate is inhibited. Bmp2 signalling from the adjacent surface ectoderm is unopposed by Noggin, and this inhibits dorsolateral bending. At low spinal levels, Shh influence during neurulation is less strong. Noggin expression is no longer inhibited and represses Bmp2 action, leading to dorsolateral bending (modified from Ybot-Gonzalez et al. [57]). Reproduced with kind courtesy of The Company of Biologists Ltd.

Children or adults with symptomatic spinal dysraphism, with neurological or bladder disability, can be treated by surgical untethering of the cord. This usually requires magnetic resonance imaging to monitor cord mobility before and after operation. Follow-up studies to determine the long-term effects of surgery have shown a good outcome in terms of maintained cord mobility, and symptomatic improvement in some cases, with resolution of upper motor neuron signs and enhanced bladder function. A challenge for the future will be to develop biomarkers that can predict which patients with spinal dysraphism are likely to deteriorate, and will benefit from early surgery, as opposed to those who will remain asymptomatic and should be spared unnecessary surgery.

Primary prevention The prospect for primary prevention of NTDs was raised in the 1970s by a non-randomized trial [63] that reported a significant reduction in recurrence of NTDs following periconceptional administration of a folate-containing multivitamin preparation (Pregnavite Forte F). A randomized controlled clinical trial in the UK and Hungary [64] subsequently demonstrated an approximately 70% reduction in NTD recurrence after periconceptional supplementation with 4 mg/day folic acid. There is also evidence that folic acid may exert a preventive effect on the first occurrence of NTDs. These trials have led most countries to recommend the consumption of 0.4 mg/day folic acid

26

by all women planning a pregnancy, beginning before conception and continuing throughout the early months of pregnancy. High-risk women, where a previous pregnancy has been affected by NTD, are recommended to take 4 mg/day folic acid. However, concerns about poor compliance with this voluntary supplementation program has led some countries to fortify the food supply with folic acid. In the United States, from 1999 onwards, it became compulsory for millers to add 0.14 mg folic acid to each 100 g flour. This policy has led to a marked rise in the folate status of women of childbearing age in the United States, and was associated with a 19% decline in the frequency of NTDs [65]. A much larger, 46%, decrease in NTD prevalence followed the fortification program in Canada [66]. Other countries have followed this policy, and have reported variable responses in terms of decline in NTD prevalence. The consensus of opinion today is that, while some NTD cases are preventable with folic acid in all countries, others are nonresponsive, and require alternative therapeutic interventions. Mouse models of NTD fall into two groups with respect to prevention by folic acid. Splotch (Pax3 gene) mice develop NTDs that are preventable by folic acid treatment of pregnant females or by direct application to embryos in vitro and which respond to dietary folate deficiency with an increase in NTD frequency [67]. Similarly, mice with mutations in the Cited2, Cart1 and Cd genes also exhibit NTDs that can be prevented by exogenous folic acid. In contrast, curly tail (Grhl3 gene) mutant mice are resistant to folic acid but low spinal defects in this system can be prevented by another vitamin-like molecule, inositol, administered either in vivo or in vitro [68]. Both myo- and D-chiro-isomeric forms of inositol can prevent NTDs in this mouse model, via a mechanism involving activation of specific isoforms of protein kinase C. Inositol stimulates cell proliferation in the hindgut of curly tail embryos, overcoming the genetically determined defect that is known to lead to NTDs. The Prevention of Neural Tube Defects by Inositol (PONTI) clinical trial has provided preliminary evidence that inositol may be a useful adjunct to folic acid therapy for prevention of human NTDs [69]. A further possible avenue toward preventing folic acid resistant NTDs is to intervene further ‘downstream’ in folate onecarbon metabolism. For example, the Gldc mutant mouse is deficient in formate production from mitochondrial folate metabolism (Figure 2.8), but NTDs can be prevented by providing formate orally to the pregnant females [70]. These studies suggest that possible primary prevention of human NTDs should not be limited to folic acid, but may also benefit from therapy to intervene downstream in folate metabolism, or in completely different pathways, as with inositol.

References 1. Van Allen MI, Kalousek DK, Chernoff GF et al. (1993) Evidence for multi-site closure of the neural tube in humans. Am J Med Genet 47:723–43

Neural Tube Defects Chapter 2 2. O’Rahilly R, M¨uller F (2002) The two sites of fusion of the neural folds and the two neuropores in the human embryo. Teratology 65:162–17. 3. Jiang XB, Iseki S, Maxson RE et al. (2002) Tissue origins and interactions in the mammalian skull vault. Dev Biol 241:106–16 4. Opperman LA, Sweeney TM, Redmon J et al. (1993) Tissue interactions with underlying dura mater inhibit osseous obliteration of developing cranial sutures. Dev Dyn 198:312–22 5. Wilson V, Olivera-Martinez I, Storey KG (2009) Stem cells, signals and vertebrate body axis extension. Development 136:1591–604 6. Carter CO, Evans KA (1973) Spina bifida and anencephalus in Greater London. J Med Genet 10:209–34 7. Elwood JH, Nevin NC (1973) Factors associated with anencephalus and spina bifida in Belfast. Br J Prev Soc Med 27:73–80 8. Berry RJ, Li Z, Erickson JD et al. China–US Collaborative Project Neu. (1999) Prevention of neural-tube defects with folic acid in China. N Engl J Med 341:1485–90 9. Moore CA, Li S, Li Z et al. (1997) Elevated rates of severe neural tube defects in a high-prevalence area in northern China. Am J Med Genet 73:113–18 10. Flatz G, Sukumaran M (1970) Fronto-ethmoidal encephalomeningoceles in the population of northern Thailand. Humangenetik 11:1–9 11. Seller MJ (1987) Neural tube defects and sex ratios. Am J Med Genet 26:699–707 12. Brook FA, Estibeiro JP, Copp AJ (1994) Female predisposition to cranial neural tube defects is not because of a difference between the sexes in rate of embryonic growth or development during neurulation. J Med Genet 31:383–7 13. Juriloff DM, Harris MJ (2012) Hypothesis: the female excess in cranial neural tube defects reflects an epigenetic drag of the inactivating X chromosome on the molecular mechanisms of neural fold elevation. Birth Defects Res A Clin Mol Teratol 94:849–55 14. Garne E, Loane M, Dolk H et al. (2012) Spectrum of congenital anomalies in pregnancies with pregestational diabetes. Birth Defects Res A Clin Mol Teratol 94:134–40 15. Kitzmiller JL, Wallerstein R, Correa A, Kwan S (2010) Preconception care for women with diabetes and prevention of major congenital malformations. Birth Defects Res A Clin Mol Teratol 88:791–803 16. Buchanan TA, Denno KM, Sipos GF, Sadler TW (1994) Diabetic teratogenesis: In vitro evidence for a multifactorial etiology with little contribution from glucose per se. Diabetes 43:656–60 17. Rasmussen SA, Chu SY, Kim SY et al. (2008) Maternal obesity and risk of neural tube defects: a metaanalysis. Am J Obstet Gynecol 198:611–19 18. Moretti ME, Bar-Oz B, Fried S, Koren G (2005) Maternal hyperthermia and the risk for neural tube defects in offspring: systematic review and meta-analysis. Epidemiology 16:216–19 19. Robert E, Rosa F (1983) Valproate and birth defects. Lancet 2:1142 20. Hendricks KA, Simpson JS, Larsen RD (1999) Neural tube defects along the Texas–Mexico border, 1993–1995. Am J Epidemiol 149:1119–27 21. Fleming A, Copp AJ (2000) A genetic risk factor for mouse neural tube defects: defining the embryonic basis. Hum Mol Genet 9:575– 81 22. Luthy DA, Wardinsky T, Shurtleff DB et al. (1991) Cesarean section before the onset of labor and subsequent motor function in infants with meningomyelocele diagnosed antenatally. N Engl J Med 324:662–6

23. Stevenson KL (2004) Chiari type II malformation: past, present, and future. Neurosurg Focus 16:E5 24. Cameron M, Moran P (2009) Prenatal screening and diagnosis of neural tube defects. Prenatal Diag 29:402–11 25. Nicolaides KH, Gabbe SG, Campbell S, Guidetti R (1986) Ultrasound screening for spina bifida: cranial and cerebellar signs. Lancet 2:72–4 26. Wood LR, Smith MT (1984) Generation of anencephaly: 1. Aberrant neurulation and 2. Conversion of exencephaly to anencephaly. J Neuropath exp Neurol 43:620–33 27. Geoffroy Saint–“Hilaire l (1825) Sur de nouveaux anencephales humains confirmant par leur fait d’organisation la derniere theorie sur les monstres. Mem Museum d’Hist Nat 12:233–56 28. Marin-Padilla JM (1965) Study of the skull in human cranioschisis. Acta Anat 62:1–20 29. Bell JE, Green RP (1982) Studies on the area cerebrovasculosa of anencephalic fetuses. J Pathol 137:315–28 30. Barson AJ. (1970) Spina bifida: the significance of the level and extent of the defect to the morphogenesis. Dev Med Child Neurol 12:129–44 31. Giroud A (1960) Causes and morphogenesis of anencephaly. In: Wolstenholme GEW, O’Connor CM, eds. Congenital Malformations. Ciba Foundation Symposium. Churchill, London, pp. 199– 212 32. Bell JE, Gordon A, Maloney AFJ (1981) Abnormalities of the spinal meninges in anencephalic fetuses. J Pathol 133:131–4 33. Angevine DM (1938) Pathologic anatomy of hypophysis and adrenals in anencephaly. Arch Pathol 26:507–18 34. Nakado KK (1973) Anencephaly. Dev Med Child Neurol 15:383– 400 35. Emery JL, Lendon RG (1973) The local cord lesion in neurospinal dysraphism (meningomyelocele). J Pathol 110:83–96 36. Caviness VS, Evrard P (1975) Occipital encephalocele: a pathologic and anatomic analysis. Acta Neuropathol (Berl) 32:245–55 37. Karch SB, Urich H (1972) Occipital encephalocele: a morphological study. J Neurol Sci 15:89–112 38. Harding BN, Golden JA (2015) Malformations. In: S Love, H Budka, JW Ironside, A Perry eds., Greenfields’s Neuropathology, 9th ed., Boca Raton, FL, CRC Press, pp. 270–398 39. Ahdab-Barmada M, Claassen D (1990). A distinctive triad of malformations of the central nervous system in the Meckel–Gruber syndrome. J Neuropath Exp Neurol 49:610–20 40. Paetau A, Salonen R, Haltia M (1985) Brain pathology in the Meckel syndrome: A study of 59 cases. Clin Neuropathol 4:56–62 41. McLaurin RL (1964) Parietal cephaloceles. Neurology 14:764–74 42. Ziter F, Bramwit D (1970) Nasal encephaloceles and gliomas. Br J Radiol 43:136 43. Pang D, Dias MS (1993) Cervical myelomeningoceles. Neurosurgery 33:363–73 44. Leck I (1974) Causation of neural tube defects: Clues from epidemiology. Br Med Bull 30:158–63 45. Rampersaud E, Melvin EC, Speer MC (2006) Nonsyndromic neural tube defects: Genetic basis and genetic investigations. In: Wyszynski DF, ed., Neural Tube Defects: From Origin to Treatment. Oxford, Oxford University Press, pp. 165–75 46. Amorim MR, Lima MA, Castilla EE, Orioli IM. (2007) Non-Latin European descent could be a requirement for association of NTDs and MTHFR variant 677C >T: a meta-analysis. Am J Med Genet A 143A:1726–32

27

Developmental Neuropathology 47. Narisawa A, Komatsuzaki S, Kikuchi A et al. (2012) Mutations in genes encoding the glycine cleavage system predispose to neural tube defects in mice and humans. Hum Mol Genet 21:1496– 503 48. Tibbetts AS, Appling DR (2010) Compartmentalization of mammalian folate-mediated one-carbon metabolism. Annu Rev Nutr 30:57–81 49. Harris MJ, Juriloff DM (2010) An update to the list of mouse mutants with neural tube closure defects and advances toward a complete genetic perspective of neural tube closure. Birth Defects Res A Clin Mol Teratol 88:653–69 50. Copp AJ, Greene NDE, Murdoch JN (2003) The genetic basis of mammalian neurulation. Nat Rev Genet 4:784–93 51. Murdoch JN, Damrau C, Paudyal A et al. (2014) Genetic interactions between planar cell polarity genes cause diverse neural tube defects in mice. Dis Model Mech 7:1153–63 52. Logan CV, Abdel-Hamed Z, Johnson CA (2010) Molecular genetics and pathogenic mechanisms for the severe ciliopathies: Insights into neurodevelopment and pathogenesis of neural tube defects. Mol Neurobiol 43:12–26 53. Waters AM, Beales PL (2011) Ciliopathies: An expanding disease spectrum. Pediatr Nephrol 26:1039–56 54. Wallingford JB (2012) Planar cell polarity and the developmental control of cell behavior in vertebrate embryos. Annu Rev Cell Dev Biol 28:627–53 55. Copp AJ (2005) Neurulation in the cranial region – normal and abnormal. J Anat 207:623–35 56. Massa V, Savery D, Ybot-Gonzalez P et al. (2009) Apoptosis is not required for mammalian neural tube closure. Proc Natl Acad Sci USA 106:8233–8 57. Ybot-Gonzalez P, Gaston-Massuet C, Girdler G et al. (2007) Neural plate morphogenesis during mouse neurulation is regulated by antagonism of BMP signalling. Development 134:3203–11 58. Murdoch JN, Copp AJ (2010) The relationship between Hedgehog signalling, cilia and neural tube defects. Birth Defects Res A Clin Mol Teratol 88:633–52 59. Copp AJ, Brook FA (1989) Does lumbosacral spina bifida arise by failure of neural folding or by defective canalisation? J Med Genet 26:160–6 60. Adzick NS, Sutton LN, Crombleholme TM, Flake AW (1998). Successful fetal surgery for spina bifida. Lancet 352:1675–6 61. Adzick NS, Thom EA, Spong CY et al. (2011). A randomized trial

28

62.

63.

64.

65.

66.

67.

68. 69.

70.

71.

72.

73.

74.

of prenatal versus postnatal repair of myelomeningocele. N Engl J Med 364:993–1004 Stiefel D, Shibata T, Meuli M et al. (2003) Tethering of the spinal cord in mouse fetuses and neonates with spina bifida. J Neurosurg (Spine) 99:206–13 Smithells RW, Nevin NC, Seller MJ et al. (1983) Further experience of vitamin supplementation for prevention of neural tube defect recurrences. Lancet 1:1027–31 MRC Vitamin Study Research Group (1991) Prevention of neural tube defects: Results of the Medical Research Council Vitamin Study. Lancet 338:131–7 Honein MA, Paulozzi LJ, Mathews TJ et al. (2001) Impact of folic acid fortification of the US food supply on the occurrence of neural tube defects. JAMA 285:2981–6 De Wals P, Tairou F, Van Allen MI et al. (2007) Reduction in neuraltube defects after folic acid fortification in Canada. N Engl J Med 357:135–42 Burren KA, Savery D, Massa V et al. (2008) Gene–environment interactions in the causation of neural tube defects: folate deficiency increases susceptibility conferred by loss of Pax3 function. Hum Mol Genet 17:3675–85 Greene NDE, Copp AJ (1997) Inositol prevents folate-resistant neural tube defects in the mouse. Nature Med 3:60–6 Greene NDE, Leung KY, Gay V et al. (2016). Inositol for prevention of neural tube defects: A pilot randomised controlled trial. Br J Nutr 115:974–83 Pai YJ, Leung KY, Savery D et al. (2015) Glycine decarboxylase deficiency causes neural tube defects and features of non-ketotic hyperglycinemia in mice. Nat Commun 6:6388 Copp AJ, Stanier P, Greene ND (2013) Neural tube defects: Recent advances, unsolved questions, and controversies. Lancet Neurol 12:799–810 Pang D, Dias MS, Ahab-Barmada M (1992) Split cord malformation: Part I: A unified theory of embryogenesis for double spinal cord malformations. Neurosurgery 31:451–80 Greene NDE, Gerrelli D, Van Straaten HWM, Copp AJ (1998) Abnormalities of floor plate, notochord and somite differentiation in the loop-tail (Lp) mouse: a model of severe neural tube defects. Mech Dev 73:59–72 Copp AJ, Greene NDE (2016) Neural tube defects. In: Encyclopaedia of Life Sciences. Chichester, John Wiley & Sons Ltd. doi: 10.1002/9780470015902.a0000804.pub3

3

Midline Patterning Defects Edwin S. Monuki1 and Jeffrey A. Golden2 1 2

Department of Pathology and Laboratory Medicine, UC Irvine School of Medicine, Irvine, CA, USA Department of Pathology, Brigham and Women’s Hospital, Harvard Medical School, Boston, MA, USA

Definition of the disorder, major synonyms and historical perspective The midline malformations described in this chapter are holoprosencephaly (HPE), atelencephaly/aprosencephaly, agenesis of the corpus callosum, and septo-optic dysplasia. Advances over the past few decades have greatly clarified the embryologic, molecular, and genetic bases for HPE, and to a lesser extent, the other disorders.

Embryology of forebrain patterning The morphogens involved in forebrain patterning emanate from an interacting network of midline signaling centers in both neural and extraneural sites (Figure 3.1) [1]. Temporally, this network starts with extraneural organizers, which then induce secondary organizers within neural tissue. Important extraneural organizers include the node, notochord, prechordal plate, anterior visceral endoderm, and epidermal ectoderm, which act from gastrulation through neurulation. Later-acting midline organizers within neural tissue include the ventral telencephalic midline (VTM); that is, the hypothalamic, preoptic and entopeduncular regions; rostral telencephalic midline (RTM), including the anterior neural ridge or border; and dorsal telencephalic midline (DTM), including the roof plate (Figure 3.1). Each of these neural organizers depends on earlier organizers and organizer signals for proper induction. In many cases, these inductions are homeogenetic (“like inducing like”); that is, extraneural signals induce their own production in the neural organizers they specify [2,3]. The node, or Spemann organizer in lower vertebrates, is the primary organizer of body axis, gastrulation, and neural induction [4], which centrally involves bone morphogenetic protein

(BMP) antagonists (noggin and chordin) produced by the node [5]. Following induction, early influences on the neural plate include the anterior visceral endoderm and epidermal ectoderm. Epidermal ectoderm produces BMPs and is continuous with the neural plate rostrally and laterally, while the anterior visceral endoderm underlies the rostral neural tube and imparts it with forebrain character by antagonizing caudalizing factors such as Wnts [5]. In the meantime, the node, via Nodal signaling [6], induces two signaling centers within mesendoderm interposed between the neural plate and the anterior visceral endoderm – the notochord and its rostral extension, the prechordal plate. The notochord and prechordal plate directly underlie the midline neural plate, with the prechordal plate underlying the future forebrain. These extraneural signaling centers then induce the aforementioned signaling centers within neural tissue. The prechordal plate induces VTM and “splits” the eye field via Sonic hedgehog (SHH) and Nodal signaling [8,9], while the RTM is probably induced by the anterior visceral endoderm [10]. Epidermal ectoderm induces the DTM via BMP signals [11]. VTM development also involves morphogenetic movements, with the future hypothalamus moving rostrally to split the eye field, and both hypothalamus fate and morphogenesis require Nodal and Shh signals from the prechordal plate [5]. (Note: Since the eye field is initially ventral to the nasal placode, this explains the proboscis occurring above a cyclopic eye in HPE.) Midline neural centers (VTM, RTM, and DTM) then pattern adjacent forebrain tissues via the same signals important for their own induction: VTM inducing hypothalamus/ventral forebrain via Shh, RTM inducing rostral forebrain via Fgfs, and DTM regulating dorsal forebrain via BMPS [11–15]. Interactions within this morphogen signaling network are myriad and supported by many studies in multiple experimental organisms [1]. For example, ventral Nodal from the prechordal plate positively regulates Shh in the prechordal plate and VTM,

Developmental Neuropathology, Second Edition. Edited by Homa Adle-Biassette, Brian N. Harding, and Jeffrey A. Golden. © 2018 John Wiley & Sons Ltd. Published 2018 by John Wiley & Sons Ltd.

29

Developmental Neuropathology

Figure 3.1 The morphogen and holoprosencephaly (HPE) gene network in forebrain development (modified from Monuki [1]) Left, network diagram based on human and experimental model studies, with the four major HPE genes in red. Mutations in FGF8 and NODAL have also been described in human HPE. Right, schematic of the developing forebrain (anterior oblique view) with midline morphogen sources indicated (purple, dorsal; green, rostral; dark blue, ventral). Reproduced with permission of Oxford University Press.

while VTM/Shh and RTM/FGFs engage in a mutual positive regulatory loop. VTM/FGFs and DTM/BMPs have a more complex relationship involving both positive and negative effects – FGF8 genetics (see below) indicate a positive role for FGF8 in DTM fate, possibly by providing DTM competency as seen in the midbrain [16], although DTM inhibition by RTM/Fgf8 is more often seen experimentally. Collectively, these studies define an orderly set of positive midline signaling interactions; that is, Nodal (node, notochord) → Shh (notochord, prechordal plate, VTM) → Fgf8 (RTM) → BMPs (DTM). The spatiotemporal sequence of the positive interactions in this network can explain how ventral signaling extends to dorsal regions within the developing forebrain. Conversely, the negative (inhibitory) interactions between DTM/BMPs and RTM/Fgf8 (and VTM/Shh) prevents the spread of dorsal signaling toward ventral domains [1].

Holoprosencephaly Definition Holoprosencephaly represents a group of forebrain malformations that involve primary defects in midline induction. Classic HPE is divided into lobar, semilobar, and alobar forms, which represent a continuous phenotypic spectrum. In addition, a distinct middle interhemispheric subtype, also known as syntelencephaly, has been established [17,18]. A mild septopreoptic form has been recently proposed [19]. Many patients with HPE also exhibit extra-central nervous system (CNS) malformations of midline craniofacial structures, such as cyclopia, midline clefts, and nasal anomalies; these combinations are referred to as the HPE complex. “Microform” HPE refers to patients with HPEassociated craniofacial anomalies, but without accompanying structural brain defects. Synonyms and historical annotations Holoprosencephaly nosology is of historic interest and based on pathology. Kundrat first designated these malformations as arhinencephalies based on the absence of olfactory bulbs and tracts [20]. Yakovlev pointed out the involvement of the entire

30

telencephalon, and referred to the malformation as “holotelencephaly” [21]. Recognition of diencephalic involvement led DeMyer to coin the term “HPE.” Recognizing a correlation in severity between brain and craniofacial anomalies, DeMyer famously concluded that the “face predicts the brain” [22]. Although generally true, exceptions are well recognized, and Cohen estimated that the face predicts the brain in 80% of cases [23].

Epidemiology Incidence Incidence in liveborn children has been estimated to be 0.48– 0.88 per 10 000 while the rate among human abortuses is much higher (40 per 10 000, or 1 in 250 conceptions), indicating a high rate of fetal loss [24]. Affected patients have been described in many ethnic groups. There is a slight predominance of female cases. Severely affected children often die in infancy or early childhood, although an 11-year old child with alobar HPE has been described [25]. Children with milder semilobar, lobar, or middle interhemispheric forms may survive to adulthood. Risk factors Several environmental and maternal factors have been associated with HPE [26,27]. In humans, the best recognized association is with maternal diabetes, although the underlying mechanism remains unclear. Ethanol can induce HPE-like craniofacial and CNS anomalies in mice and macaque, although findings in humans from large case–control studies are mixed [27]. Another HPE-causing teratogen, cyclopamine, was identified from an outbreak of cyclopia in lambs born to ewes grazing on Veratrum californicum [28]. Cyclopamine and related alkaloids act via the SHH pathway [29]. Cholesterol biosynthesis inhibitors have also been linked to the SHH pathway, a common teratogenic and genetic pathway in human HPE (see below), although retrospective and case review studies of statins have yielded inconclusive results [30]. Despite the inherent challenges of HPE epidemiology, large case–control studies suggest potential associations

Midline Patterning Defects Chapter 3

for salicylates, certain infections, and assisted reproductive technologies [27].

genes in humans and animal models impact morphogenic signaling during forebrain development (see below).

Genetics Our understanding of human HPE genetics has increased dramatically over the first decades of the twenty-first century. Importantly, most cases have a recognizable genetic etiology. About 50% are cytogenetic/chromosomal and 25% are syndromic [31,32]. Among the remaining quarter, three groups each constitute approximately 5%: mutations/deletions of the four “major” HPE genes (SHH, SIX3, ZIC2, and TGIF), cholesterol metabolism gene defects, and submicroscopic chromosomal alterations. Mutations in other HPE genes (around 1%) and unknown/environmental causes (around 9%) constitute the remainder [33]. In the cytogenetic/chromosomal category, trisomy 13 is most common, accounting for up to 75% of cases in this category when cryptic rearrangements are also included. Triploidies account for approximately 20%, while trisomy 18 (1–2%) and other chromosomal anomalies are less common. Conversely, HPE is common in trisomy 13, being seen in about 67% of neuropathologically examined trisomy 13 brains (compared with 17–39% by ultrasound) [31]. Many syndromes are associated with HPE, including pseudotrisomy 13, Smith–Lemli–Opitz, Pallister–Hall, Meckel, velocardiofacial, Genoa, Lambotte, Martin, and Steinfeld syndromes [24]. Smith–Lemli–Opitz syndrome, which is due to DHCR7 mutations and defective cholesterol biosynthesis, is of particular interest owing to the cholesterol dependence of SHH signaling [34]. The HPE loci represent 13 cytogenetically determined autosomal dominant loci, and the four major genes (SIX3, SHH, TGIF, and ZIC2) represent HPE2–HPE5, respectively [32,33]. In addition, HPE7 (PTCH1), HPE9 (GLI2), and HPE11 (CDON) have been molecularly defined [33,35], although the GLI2 mutation phenotype is somewhat distinct from HPE [36]. Beyond the defined HPE loci, at least six other human genes are now established as HPE genes: DISP1, FGF8, FOXH1, GAS1, NODAL, and STIL [33]. With the exception of STIL (autosomal recessive), the HPE genes are autosomal dominant and predict reduced or loss of function, suggesting haploinsufficiency as a causal genetic mechanism. Mutations or polymorphisms in multiple HPE genes can also occur, and may account for phenotypic variation within families [37,38]. In a large European study (645 probands), the most severe HPE phenotypes were associated with SIX3 and ZIC2 mutations, while SHH mutations led to the most microforms, and ZIC2 mutations were more often de novo compared with the other three genes [38]. Much less is known about the genetics underlying the middle interhemispheric variant. Interestingly, however, ZIC2 mutations can cause not only classic HPE, but also the middle interhemispheric variant [39]. A middle interhemispheric phenotype has also been associated with a deletion spanning EYA4, which interacts with SIX3 [40]. Pathogenetically, essentially all HPE

Clinical features including appropriate investigations Clinical presentation Among those that survive the in utero period, severely affected children often die in the neonatal period. Common issues include hydrocephalus, epilepsy, motor and oromotor impairments, hypothalamic and endocrine dysfunction, anosmia, developmental delay, and mental restriction [41]. Endocrine abnormalities can include hypernatremia (due to diabetes insipidus), adrenal hypoplasia, hypothyroidism, and hypogonadism. Like HPE neuropathology, HPE clinical phenotypes vary significantly. Even within autosomal dominant families, phenotypic variability can be striking. For example, within autosomal dominant HPE pedigrees, it has been estimated that 37% of carriers will have HPE, 27% will have a microform, and 36% will have no clinical abnormality with normal intelligence [23]. Consistent with the milder neuropathology, patients with the middle interhemispheric variant tend to have fewer clinical symptoms, including fewer endocrine and movement abnormalities, although motor abnormalities (spasticity) and seizures remain common [42]. Neuroimaging The diagnosis of HPE is generally made or confirmed through brain imaging. Magnetic resonance imaging (MRI) is the study of choice. While severe cases can often be diagnosed by prenatal ultrasound, its sensitivity for HPE is understandably lower than that of direct neuropathological examination [31], and milder cases may not always be detected.

Pathology Fundamentally, HPE is caused by defective induction of the forebrain midline. Accordingly, neuropathology is greatest at or toward the forebrain midline in all forms of the condition. Secondarily, midline induction failure leads to the apparent “fusion” of bilateral forebrain structures across the midline. Several methods of grading or classifying the CNS anomalies have been developed, with the most widely used method subdividing HPE into alobar (most severe), semilobar, and lobar categories (least severe). However, these three categories represent a spectrum rather than discrete forms, and individuals within a single family can have phenotypes spanning all three categories [21]. General pathology Together with the neuropathology, specific midline craniofacial abnormalities are associated with HPE and make up the HPE sequence or syndrome [23]. The craniofacial anomalies also vary markedly and include, in order of severity: 1. cyclopia (single orbit with one or a partially divided eye and proboscis, the abnormally formed and placed nose, above the eye)

31

Developmental Neuropathology 2. ethmocephaly (separate but closely spaced orbits with the proboscis in between) 3. cebocephaly (ocular hypotelorism with a single-nostril nose). The range of ophthalmic anomalies also includes anophthalmia, synophthalmia, microphthalmia, and coloboma. In addition to the proboscis, nasal anomalies can include a single blind-ended nares and hypoplasia of the nasal bones resulting in a flatappearing nose. Congenital nasal pyriform aperture stenosis can present like choanal atresia and lead to respiratory distress. Median cleft lip and palate are often present, and lateral cleft lip/palate can also occur in addition to a single midline maxillary incisor. When craniofacial anomalies occur in the absence of forebrain defects, the cases are termed “microforms” and are generally associated with normal cognitive development [43]. However, individuals with microform HPE may be genetic carriers at risk for having affected offspring. The middle interhemispheric variant is notable for the lack of associated craniofacial anomalies [18], which relates to its distinct pathogenesis from classic HPE (see below). Macroscopic neuropathology Diagnosis of HPE is based on macroscopic (gross) neuropathology. We refer readers to Norman et al. [24] for a superb and comprehensive review of HPE neuropathology; here, we focus on classification and principal features. In all forms, the brain is small, averaging less than 100 g at full term (normal 350 g) for alobar and semilobar forms. Classification into alobar, semilobar, and lobar forms is based on the presence and extent of the midline interhemispheric fissure. In alobar HPE, the midline fissure is absent, resulting in cerebral hemispheres that appear entirely “fused” into a single holosphere (Figure 3.2a,b). Sylvian fissures and other distinctions between the cerebral lobes are also lacking. In infants and older individuals, gyri usually develop, but are abnormal in surface anatomy, often coursing as sets of medial and paramedial gyri. The most medial gyri frequently contain hippocampus and entorhinal cortex, and course along the lateral aspect of a cyst-like membrane located more posterior–dorsally. The posterior aspect of the cerebrum is shaped like a horseshoe with the posterior–dorsal rim composed of a cyst-like membrane, which also serves as the roof of the single midline ventricle. The basal ganglia and thalami are immediately under the membrane and vary from distinct bilateral structures to fused midline masses. Basal ganglia may appear completely absent. The corpus callosum and anterior commissure are usually absent, although rudimentary callosal plates may be identified [21]. Of interest, Probst bundles are not present [44], which may reflect the absence of normal cortical projections. Brainstem and cerebellum most often appear normal, although corticospinal tracts are usually hypoplastic or absent. Semilobar HPE represents an intermediate degree of severity, with the interhemispheric fissure being absent anteriorly, but present posteriorly (Figure 3.2c,d). However, even at posterior

32

levels, cerebral cortex is abnormally continuous across the midline and its gyral pattern is disorganized. Lobar HPE, the least severe form of classic HPE, has an interhemispheric fissure that divides nearly the entire forebrain, except for the most rostral and ventral regions of the frontal lobes (Figure 3.2e,f). Upon cross-sectioning, cerebral cortex abnormally spans the midline at most levels, although posterior regions can be quite well formed with complete hemispheric separation and a well-developed splenium. Cerebral lobes and normal gyral patterns may be recognizable. Cross-sections display a single ventricle with basal ganglia fusion commonly restricted to its most rostral portions. The thalamus may be less involved, although an enlarged massa intermedia is common. In addition to representing a phenotypic spectrum, the alobar, semilobar, and lobar forms of classic HPE display a consistent gradient in macroscopic neuropathology; that is, ventral forebrain regions are most often and most severely affected, with progressively less pathology seen more posteriorly. This severity gradient applies across and within cases. Importantly, this neuropathological gradient correlates with the normal spatiotemporal gradients of the morphogen signaling network that underlies forebrain development and HPE pathogenesis (see below). Olfactory bulbs and tracts are absent in virtually all cases of classic HPE. Optic nerves are often hypoplastic, while the remaining cranial nerves are usually normal. The middle and anterior cerebral arteries course abnormally. In severe cases, a disorganized cluster of vessels extends from the internal carotid region anteriorly over the ventral and frontal holoprosencephalic brain. In less severe forms, a single anterior cerebral artery may be present. A variety of other less common anomalies (e.g. encephaloceles, myelomeningoceles, arachnoid cysts, porencephaly) may also co-occur [24,44]. In the middle interhemispheric variant [17,45], the interhemispheric fissure is present ventrally, frontally, and posteriorly, but lacking in the ‘middle’ of the cerebrum; that is, in posterior frontal and parietal regions (Figure 3.3). In a high proportion of cases, Sylvian fissures are upslanted and actually fuse across the dorsal midline [18]. Midsagittal sections reveal corpus callosum pathology that is also greatest in its midportion (i.e. in its body more than rostrum, genu, and splenium). Subcortically, thalamic noncleavage and dorsal cysts occur in a significant fraction of cases. Midbrain noncleavage has been noted in some cases [18]. Importantly, however, basal ganglia involvement is rare, and hypothalamic and craniofacial defects are virtually absent. Thus, in contrast to classic HPE, ventral forebrain and craniofacial structures are preserved in the middle interhemispheric variant, and accordingly, clinical signs and symptoms are typically less severe. Histopathology While gross neuropathology is often clear, HPE histopathology can vary. Yakovlev [21] identified all cytoarchitectural areas of cerebral cortex in his cases, while others have reported marked organizational disruption [24]. Evidence of abnormal radial

Midline Patterning Defects Chapter 3

(a)

(b)

(c)

(d)

(e)

(f)

Figure 3.2 Alobar, semilobar and lobar holoprosencephaly. (a) Frontal view of a 20-week fetus with alobar holoprosencephaly, note there is no fissure separating the two hemispheres. (b) An adult with alobar holoprosencepahly showing continuity of the gray matter across the dorsal midline. This patient had minimal ventral abnormalities which is atypical for most cases of alobar holoprosencephaly. (c) Semilobar holoprosencephaly with a small cleft partially separating the dorsal–caudal cerebral hemispheres (caudal view, 19-week fetus). (d) Full-term

child with semilobar holoprosencephaly. The diencephalic structures are fused and there appears to be a sagittal fissure, however, gray matter is seen crossing the midline and there is no corpus callosum. (e) 20-week fetus with lobar holoprosencepahly. The cerebral hemispheres appear separated in both anterior (top) and posterior (bottom) sections, although the diencephalon is fused (bottom section). (f) Coronal section from the same brain as (e) showing fusion across the midline in the frontal cortex only.

33

Developmental Neuropathology of septal, optic, and hypothalamic/pituitary regions rather than fusions; if fused, these may represent septopreoptic HPE [19]. Distinguishing aprosencephaly/atelencephaly from severe alobar HPE with secondary destructive changes can sometimes be challenging, depending on the quantity and quality of remaining forebrain/telencephalic tissue.

Figure 3.3 Middle interhemispheric variant of holoprosencephaly. A gray matter mass is fused across the midline of the dorsal forebrain, however the ventral structures including the myelinated optic tracts (arrowheads), appear normal. Coronal whole brain section (Luxol fast blue–hematoxylin and eosin).

migration in the neocortex may or may not be present [24]. From our own experience, large periventricular and white matter heterotopia (suggestive of defective radial migration) can be encountered, but are uncommon. Moreover, secondary injury or disruption resulting in extensive glioneuronal rests and “crust” formation (thereby mimicking primary migration defects) can also be seen. Among major HPE genes, neuronal migration defects are most likely to be associated with ZIC2 mutations [38]. The hippocampus is virtually always present, although it may show incomplete or abnormal development. Disorganization of the basal ganglia and thalamus typically varies directly with overall severity of HPE. In severe alobar cases, caudate, putamen, and globus pallidus may be unrecognizable or only superficially identifiable. In less severe forms, all basal ganglia components are present, although the caudate heads may be fused, and septal nuclei may not be identifiable. The cerebellum may show various degrees of cortical dysplasia and/or heterotopia, particularly in cases with trisomy 13 and other cytogenetic abnormalities. Brainstem abnormalities are frequently limited to hypoplastic or absent corticobulbar and corticospinal tracts. Differential diagnosis The primacy of defective midline induction in HPE, with secondary “fusion” of bilateral forebrain structures, assists greatly in differential diagnosis. For example, absence of the septum pellucidum associated with hydrocephalus can be distinguished from HPE by identifying normal midline tissues and separation of cortex and other bilateral forebrain structures. Similarly, other midline anomalies (e.g. arrhinencephaly, agenesis of the corpus callosum), which can occur in isolation or in conjunction with other brain anomalies, do not themselves constitute HPE without accompanying midline and fusion defects. The midline abnormalities seen in septo-optic dysplasia are dysplasias

34

Pathogenesis and experimental models Historically, HPE was thought to represent a failure in rostral forebrain induction or in “cleavage” of the rostral neural tube. More recent evidence suggests a critical period for HPE that begins prior to neurulation [46], thus dating to the fifth gestational week or earlier. The concept of failed midline induction due to defects in prechordal mesendoderm (prechordal plate) was introduced by the late 1980s [47]. Genetic and experimental observations over the first two decades of the twenty-first century have led to further refinement of this model, with HPE now understood as a failure in midline induction stemming from defective morphogenic signaling (production, reception, transduction, and/or interpretation). All of the morphogens relevant to HPE emanate from the interacting signaling network described earlier, and importantly, essentially all human and animal HPE genes and pathways can now be directly linked to this network via genetics (Figure 3.1). The orderly spatiotemporal sequence of positive midline signaling interactions (ventral Nodal → ventral Shh → rostral Fgf8 → dorsal Bmps) can explain how primary defects in ventral signaling extend to dorsal regions in classic HPE, as well as the consistent ventral to dorsal gradient of HPE neuropathology. Roles for SHH from prechordal mesendoderm in midline craniofacial development may then provide linkage between forebrain and craniofacial defects. This interaction network also helps to explain the unique phenotype associated with the middle interhemispheric variant, which was first postulated [18], then confirmed to arise from primary defects in the roof plate and DTM [15]. Unlike the positive interactions enabling ventral to dorsal spread, the negative (inhibitory) interactions between dorsal Bmps and rostral Fgf8 (and ventral Shh) would limit the spreading of DTM defects, resulting in more restricted middle interhemispheric neuropathology [1]. HPE phenotypes have now been observed in animal models for all four major human HPE genes. HPE and cyclopia phenotypes are seen in Shh homozygous mutant mice [8] and in animals exposed to cyclopamine, as mentioned earlier [28]. More recently, loss of function mutations in Six3 [48], Zic2 [49], and Tgif1/Tgif2 together [50] yield HPE-like phenotypes in mice. Unlike their human counterparts, most mouse HPE genes require homozygosity to produce HPE-like phenotypes, although such human-mouse discordance in zygosity and brain phenotypes is not uncommon, and genetic modifiers of HPE (e.g. due to inbred mouse strain differences) can be significant [33]. The major human HPE genes that do not encode morphogens themselves also link directly to the morphogen network model.

Midline Patterning Defects Chapter 3

Tgif mutations disrupt Shh and Nodal signaling [50], while Zic2 mutations result in defective node and prechordal plate development, thereby impacting Shh and Nodal signaling [49]. Six3 is required for Shh transcriptional activation [48]. Similarly, experimental work has established how defects in cholesterol metabolism disrupt the morphogen signaling network and cause HPE via SHH. DHCR7, the last step in cholesterol biosynthesis, regulates SHH ligand processing and tissue responsivity to SHH [30]. More specifically, cholesterol is attached to the carboxyterminus of Shh and is needed for autoproteolytic Shh cleavage and target tissue responses [29,34]. This work has been further bolstered by studies using pharmacologic inhibitors of cholesterol biosynthesis and transport, and cyclopamine, the HPEcausing teratogen discussed earlier [30]. Interestingly, however, cyclopamine does not act at the level of Shh itself, but rather at the level of the Shh receptor Smoothened to inhibit tissue responsivity [51]. Given the number of links within the morphogen network and their partial disruptions in HPE (i.e. reductions rather than complete loss of gene dosage or signaling activity), this network could also explain the variable dorsal/posterior extension seen in classic HPE.

Future directions and therapy Additional HPE genes remain to be defined, as does the nosology of septopreoptic HPE and other subtle midline defects, including primary septal agenesis and septo-optic dysplasia (see below). Our thinking about the relationships between HPE and other midline defects will undoubtedly continue to evolve. Parts of the morphogen network model also remain to be fully substantiated. While Fgf8 regulates the midbrain dorsal midline in a biphasic fashion (promoting competency, then inhibiting differentiation) [16], direct evidence for an early DTM competency role remains lacking. The inferred role for dorsal BMPs in HPE [52] – based on animal model genetics, ablations, and ectopic BMP studies [53] – are not yet supported by human genetics despite directed attempts [54, 55]. From a systems perspective, additional mechanisms that reinforce and compensate network links/edges, on similar and different timescales, would be anticipated to provide for additional network robustness. While partial disruptions and nonlinearities within the morphogen signaling network can theoretically account for phenotypic variability, a true understanding of HPE phenotypic variability will require additional thought and investigation. Similarly, a satisfying understanding of the links between brain, eye, and face remains elusive. Shh, which links the eye and forebrain phenotypes [9,56], may also provide the link to midline craniofacial development [57] via homeogenetic induction of an Shh-expressing craniofacial signaling center, the frontonasal ectodermal zone [58]. The roles of neural crest in forebrain and facial organizers and development also remain to be fully incorporated. Therapy remains supportive and directed toward individual medical issues (e.g. ocular anomalies, cleft palate, seizures, hypopituitarism, developmental delay), which are best handled

by coordinated, multidisciplinary healthcare teams [41]. Given the consistent deficiencies in morphogens and morphogenic signaling associated with HPE, the concept of morphogen supplementation or signaling augmentation has a rational basis, at least in animal models.

Atelencephaly and aprosencephaly Definition and synonyms Atelencephaly (AT) and aprosencephaly (AP) are related malformations defined by the absence of telencephalon with residual diencephalon (AT) or complete loss of telencephalon and diencephalon (AP) [59]. AT and AP are extremely rare malformations that likely represent a continuum of malformation phenotypes. This malformation complex is also known as the Garcia–Lurie syndrome, atelencephalic microcephaly syndrome, XK syndrome (named for the first two patients), and XKaprosencephaly syndrome. Genetics AT/AP is generally sporadic, although familial recurrence has been reported. Defects on chromosome 13, including ring chromosome 13 and 13q duplication, have been described [60,61]. Interestingly, the HPE gene SIX3 has been associated with AT/AP in a family that also included an individual with HPE [62], raising the possibility that AT/AP represents the most severe end of the HPE phenotypic spectrum. Pathology General pathology Anomalies of other organ systems are frequently associated with AT/AP. These include heart (ventricular septal defect, atrial septal defect, patent ductus arteriosus, and coarctation of aorta), limbs (thumb hypoplasia/aplasia, oligodactyly, fanshaped toes, syndactyly, clubfoot, and single palmar creases), eyes (cyclopia, coloboma of the iris or retina, hypertelorism, downslanting palpebral fissures, blepharophimosis, microphthalmia, and microcornea), skeleton, and genitalia. The disorder is frequently lethal, with surviving infants exhibiting severe growth and mental retstriction. Neuropathology Our understanding of AT/AP neuropathology is limited to a few reported cases [24]. Descriptions include a “blunted stub” of forebrain that may include some diencephalic derivatives. Histology of the stub generally reveals disorganized immature tissue. A calcifying cerebral angiopathy has been observed in some cases, which has led to suggestions of a primary destructive process. Differential diagnosis AT/AP is distinguished from anencephaly by the presence of a hypoplastic skull covered by skin. Distinguishing AT/AP from

35

Developmental Neuropathology destructive encephaloclastic processes (e.g. based on skull–brain disproportion and histological features of a destructive process) can be challenging and not always clear.

Pathogenesis Little is known about AT/AP pathogenesis, although the genetic association with SIX3 mentioned earlier raises the possibility of shared pathogenetic mechanisms between AT/AP and HPE.

Agenesis of the corpus callosum Definition and epidemiology While “agenesis” implies absence, agenesis of the corpus callosum (ACC) is often used to include partial, thin, or combination defects of the corpus callosum in addition to complete absence [63]. ACC can be found as an isolated malformation or in association with many other malformations, such as HPE and septooptic dysplasia [24]. Indeed, ACC is usually syndromic, being associated with additional CNS anomalies in 80% of cases and with extra-CNS anomalies in 62% [64]. Over 100 syndromes have ACC as a component, including a growing list of inborn errors of metabolism. Genetics Consistent with its frequent syndromic associations, ACC is often polygenic, and over 100 genes have been associated with the disorder [65]. Among the genes implicated in ACC are some associated with HPE, including FGF8, ZIC2 and SIX3 [65,66]. Clinical features including appropriate investigations Clinical presentation Isolated ACC can be clinically silent or subtle [67], and neurodevelopment in isolated ACC is generally favorable [68]. Abnormal (a)

(b)

Figure 3.4 Partial and complete agenesis of the corpus callosum. (a) A midsagittal section of the brain from a patient with complete agenesis of the corpus callosum. The medial gyri project to the midline without an intervening cingulate gyrus (arrows). This brain also shows a dilated and posteriorly rotated vermis (Dandy-Walker malformation). (b) A midsagittal section of a brain with partial agenesis of the corpus callosum. The partial corpus callosum present (black arrowheads) extends from the lamina terminalis posteriorly. The cingulate gyrus

36

karyotype or cytogenetic abnormalities are poor prognostic indicators for developmental outcome. Findings resembling disconnection syndromes have been described in some patients, but this is by no means a constant or common feature. ACC is associated with a higher incidence of seizures and mental restriction, consistent with the higher incidence of telencephalic migrational abnormalities seen in these patients. Since CNS malformations associated with ACC can dominate the clinical picture, careful clinical assessment for associated anomalies is essential. Neuroimaging Prenatally, ACC can be detected by ultrasonography. Postnatally, isolated ACC can be discovered incidentally upon imaging for other indications, such as trauma, or at autopsy. Although ACC can be diagnosed on computed tomography (high-set third ventricle, parallel configuration of the lateral ventricles, and colpocephaly), MRI is far superior for diagnosing ACC and associated structural abnormalities, such as neuronal migration defects. Partial ACC is best seen by midsagittal MRI.

Pathology ACC pathology is well defined and has been described in detail elsewhere [24,69]. In complete ACC, cingulate gyri are absent, and medial hemispheric sulci radiate downwards toward a high-set roof of the third ventricle (Figure 3.4a). Coronal sections often reveal slightly dilated ventricles with an irregular “batwing” contour. A white matter tract, known as the bundle of Probst, frequently runs in the anterior–posterior direction just above the lateral ventricle where callosal fibers would normally fasciculate (Figure 3.4c). However, Probst bundles are not always present, which can differentiate pathogenesis. Partial ACC is recognized by the relative absence of posterior corpus callosum (c)

(partially obscured by the arrowheads) extends the length of the corpus callosum. Sulci extend from the dorsal surface of the brain to the third ventricle only posterior to the formed corpus callosum (white arrows). The cingulate gyrus and sulcus is found only where the corpus callosum is present (white arrowhead). (c) Section of a brain with agenesis of the corpus callosum (Luxol fast blue–hematoxylin and eosin). The lateral ventricle has taken on the batwing appearance and a bundle of Probst is present.

Midline Patterning Defects Chapter 3

compared with rostrum and genu, thus displaying a consistent posterior to anterior severity gradient (Figure 3.4b). In partial ACC, cingulate gyri are typically identifiable at levels where a corpus callosum is present. The anterior commissure, derived embryologically from lamina terminalis, is often absent in cases of complete ACC, but usually spared in partial ACC.

Differential diagnosis Distinguishing isolated from syndromic ACC is an essential initial step in the diagnostic workup. If syndromic, karyotype analysis and careful examination for signs/symptoms of metabolic disorders associated with ACC (e.g. non-ketotic hyperglycinemia, infantile lactic acidosis associated with pyruvate carboxylase or PDH deficiency, Smith–Lemli–Opitz, and Zellweger syndrome) are warranted. The consistent posterior to anterior severity gradient in partial ACC should be distinguished from ACC associated with classic HPE, which lacks Probst bundles and displays the opposite anterior to posterior gradient. In secondary disruption of the corpus callosum, associated developmental features, such as lack of cingulate gyri and Probst bundles, are not found. Embryology, pathogenesis and experimental models The corpus callosum forms from the commissural plate, a local thickening of the lamina terminalis and the most anterior boundary of the rostral telencephalic midline. The anterior commissure is also derived from the commissural plate. In humans, these commissures can be observed by day 45 of gestation. Differentiation of the corpus callosum is guided by specialized midline glia (glial wedge, glial sling, and others) that guide the crossing of projecting callosal axons [70]. Separation of the corpus callosum from anterior commissure results in a cystic space that is the anlagen to the septum pellucidum. The corpus callosum grows from rostral to caudal, which helps to explain the preferential posterior involvement (caudal body and splenium) in partial ACC. ACC phenotypes are associated with some HPE genes (FGF8, ZIC2, SIX3), which link ACC and HPE pathogeneses despite their differing severity gradients. At the same time, the polygenic nature of ACC [65] reflects its pathogenetic complexity, and many mouse mutants have ACC [71], which can arise from intrinsic defects of callosal neurons and axons, midline glial populations, or other aspects of the midline environment that impact attractive and repulsive cues for callosal axons.

septo-optic dysplasia can be isolated or syndromic. The condition is notoriously variable and heterogeneous at multiple levels, and is not well understood. Within the spectrum, some consider Kallman syndrome and combined pituitary hormone deficiency as the mildest, pituitary-specific end. Despite these complexities and heterogeneity, septo-optic dysplasia remains a useful diagnostic entity for clinicians and pathologists alike.

Genetics A handful of genes have been identified from patients with septo-optic dysplasia, and can be divided into syndromic and non-syndromic forms. The first septo-optic dysplasia gene is HESX1, which has been identified in siblings and individuals with milder phenotypes, and supported by mouse genetics. More recently, SOX2 and SOX3, FGFR1, FGF8, and PROKR2 have been implicated [72,73]. FGF8 involvement in HPE, ACC, and septooptic dysplasia provides a genetic link among these midline disorders. Genes associated with syndromic forms include PAX3 and mitochondrial cytochrome b. It is important to note that vascular and other destructive/disruptive conditions may also result in septo-optic dysplasia phenotypes [74]. Clinical features Optic nerve hypoplasia in septo-optic dysplasia results in visual impairment, while endocrine abnormalities result from the pituitary (and/or hypothalamic) insufficiency. Visual impairment can vary from blindness and amaurotic nystagmoid eye movements to normal vision in some cases. Endocrine insufficiencies can cause hypoglycemia and diabetes insipidus, and growth hormone deficiency may become apparent. These endocrine problems can generally be treated by standard replacement therapies. Not infrequently, seizures occur in the setting of concurrent cortical abnormalities, which can also result in cognitive deficits. Pathology Descriptions of septo-optic dysplasia pathology remain restricted to a limited number of patients (Figure 3.5) [24]. Optic nerves are usually small to absent, although they can appear relatively normal in size. Dysplasias of the retina and lateral geniculate nucleus have also been described, albeit infrequently. The pituitary is usually present, but may be small. The overlying hypothalamus may show primary loss of nuclei or secondary destructive lesions. The cerebral cortex can be normal, but heterotopia, polymicrogyria, and porencephaly have also been described.

Septo-optic dysplasia Definition and synonyms Septo-optic dysplasia is defined by defects in a triad of structures: septum pellucidum (aplasia), optic nerve (aplasia– hypoplasia), and pituitary gland (hypoplasia–dysplasia). The clinical constellation of signs and symptoms is also known as De Morsier syndrome. In addition to being a spectrum disorder,

Differential diagnosis Distinction from isolated defects of the triad structures (septum pellucidum, optic nerve, and pituitary gland) is important, such as primary septal agenesis, isolated optic nerve hypoplasia, Kallmann syndrome, and combined pituitary hormone deficiency. The septum pellucidum can be absent in septopreoptic HPE, but septo-optic dysplasia does not display the midline

37

Developmental Neuropathology likely, arrhinencephaly can result from a number of independent mechanisms. One example of an arrhinencephaly syndrome that is not associated with HPE is Kallmann syndrome. This syndrome is characterized by arrhinencephaly combined with hypogonadotropic hypogonadism resulting from gonadotropinreleasing hormone deficiency [77]. Other neurological features may be associated, and the syndrome can be inherited in an autosomal dominant, recessive, or X-linked recessive pattern. The X-linked KAL1 gene shares homology with neural cell adhesion molecules, and the developmental defect has been linked to the failure of gonadotrophin-releasing hormone-producing neurons to reach the hypothalamus [77]. Due to space limitations, other malformations of the forebrain, such as midline cysts of the corpus callosum and septum, midline lipomas, and other tract crossing defects, are not described here.

Figure 3.5 Septo-optic dysplasia. A coronal section through the brain of a child (two weeks postnatal, full term pregnancy) with septo-optic dysplasia. Note the smooth surface under the corpus callosum where the septum usually inserts. The fornix is free-floating in the ventricle.

septal and preoptic fusion seen in septopreoptic HPE. As a frequent syndromic entity, searches for associated anomalies, such as porencephaly/schizencephaly [75] or other defects of the cerebral cortex, are critical.

Embryology, pathogenesis and experimental models Embryologically, the septum pellucidum, optic nerve, and pituitary gland are linked more by spatial proximity than lineage. Thus, concepts in septo-optic dysplasia pathogenesis have revolved around vascular supply and signaling molecules, processes that can link lineally unrelated structures. For example, hypothalamus-specific loss of SHH in mice can result in septooptic dysplasia, and both Sox2 and Sox3 have been implicated as positive regulators of SHH expression in this domain [76]. Together with FGF8 [73], these findings provide for further links to the midline morphogen signaling network in forebrain development described earlier [1].

Other midline patterning defects Absence of the olfactory bulbs and tracts, termed arrhinencephaly, is often considered a midline defect. Arrhinencephaly can be found as an isolated malformation or in conjunction with a variety of malformation syndromes. Clinically, bilateral arrhinencephaly manifests as anosmia. Although arrhinencephaly is frequently included in the midline dysplasias and can represent a mild manifestation within the HPE spectrum, it would be incorrect to conclude that every isolated case of arrhinencephaly belongs with this spectrum. Moreover, reports of single midline olfactory bulb and tract are not found in the literature. Quite

38

References 1. Monuki ES (2007) The morphogen signaling network in forebrain development and holoprosencephaly. J Neuropathol Exp Neurol 66:566–75 2. Placzek M, Jessell TM, Dodd J (1995) Induction of floor plate differentiation by contact–dependent, homeogenetic signals. Development 117:205–18 3. Liem KF Jr, Tremml G, Roelink H, Jessell TM (1995) Dorsal differentiation of neural plate cells induced by BMP–mediated signals from epidermal ectoderm. Cell 82:969–79 ¨ 4. Spemann H, Mangold H (1924) Uber Induktion von Embryonalanlagen durch Implantation artfremder Organisatoren. Arch f mikr Anat u Entw Med 100:599–638 5. Wilson SW, Houart C (2004) Early steps in the development of the forebrain. Dev Cell 6:167–81 6. Sampath K, Rubinstein AL, Cheng AM, et al. (1998) Induction of the zebrafish ventral brain and floorplate requires cyclops/nodal signaling. Nature 395:185–9 7. Vincent SD, Dunn NR, Hayashi S, et al. (2003) Cell fate decisions within the mouse organizer are governed by graded Nodal signals. Genes Dev 17:1646–62 8. Chiang C, Litingtung Y, Lee E, et al. (1996) Cyclopia and defective axial patterning in mice lacking Sonic hedgehog gene function. Nature 383:407–13 9. Li H, Tierney C, Wen L, et al. (1997) A single morphogenetic field gives rise to two retina primordia under the influence of the prechordal plate. Development 124:603–15 10. Houart C, Caneparo L, Heisenberg C, et al. (2002) Establishment of the telencephalon during gastrulation by local antagonism of Wnt signaling. Neuron 35:255–65 11. Furuta Y, Piston DW, Hogan BL (1997) Bone morphogenetic proteins (BMPs) as regulators of dorsal forebrain development. Development 124:2203–12 12. Ericson J, Muhr 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 81:747–56 13. Shimamura K, Rubenstein JL (1997) Inductive interactions direct early regionalization of the mouse forebrain. Development 124:2709–18

Midline Patterning Defects Chapter 3 14. Fukuchi–Shimogori T, Grove EA (2001) Neocortex patterning by the secreted signaling molecule FGF8. Science 294:1071–4 15. Cheng X, Hsu CM, Currle DS, et al. (2006) Central roles of the roof plate in telencephalic development and holoprosencephaly. J Neurosci 26:7640–9 16. Alexandre P, Bachy I, Marcou M, Wassef M (2006) Positive and negative regulations by FGF8 contribute to midbrain roof plate developmental plasticity. Development 133:2905–13 17. Barkovich AJ, Quint DJ (1993) Middle interhemispheric fusion: an unusual variant of holoprosencephaly. AJNR 14:431–40 18. Simon EM, Hevner RF, Pinter JD, et al. (2002) The middle interhemispheric variant of holoprosencephaly. AJNR 23:151–6 19. Hahn JS, Barnes PD, Clegg NJ, Stashinko EE (2010) Septopreoptic holoprosencephaly: a mild subtype associated with midline craniofacial anomalies. AJNR 31:1596–601 20. Kundrat H (1882) Arhinencephalie als typische Art von Missbildung. Wien Med Bl 5:1395 21. Yakovlev P (1959) Pathoarchitectonic studies of cerebral malformations: III. Arrhinenecephalies (Holoprosencephaly) J Neuropathol Exp Neurol 18:22–55 22. DeMyer W, Zeman W, Palmer C (1964) The face predicts the brain, diagnostic significance of median facial anomalies from holoprosencephaly (arhinencephaly). Pediatrics 34:711–18 23. Cohen MM Jr (1989) Perspectives on holoprosencephaly: Part III. Spectra, distinctions, continuities, and discontinuities. Am J Med Genet 34:271–88 24. Norman M, McGillivray B, Kalousek D, et al. (1995) Congenital Malformations of the Brain: Pathological, Embryological, Clinical, Radiolological and Genetic Aspects. New York, Oxford University Press, pp. 187–221 25. Barr M Jr, Cohen MM Jr (1999) Holoprosencephaly survival and performance. Am J Med Genet 89:116–20 26. Lipinski RJ, Godin EA, O’Leary-Moore SK, et al. (2010) Genesis of teratogen-induced holoprosencephaly in mice. Am J Med Genet C Semin Med Genet 154(C):29–42 27. Johnson CY, Rasmussen SA (2010) Non-genetic risk factors for holoprosencephaly. Am J Med Genet C Semin Med Genet 154(C):73– 85 28. Keeler RF (1978) Cyclopamine and related steroidal alkaloid teratogens: their occurrence, structural relationship, and biologic effects. Lipids 13:708–15 29. Cooper MK, Porter JA, Young KE, Beachy PA (1998) Teratogenmediated inhibition of target tissue response to Shh signaling. Science 280:1603–7 30. Haas D, Muenke M (2010) Abnormal sterol metabolism in holoprosencephaly. Am J Med Genet C Semin Med Genet 154(C): 102–8 31. Solomon BD, Rosenbaum KN, Meck JM, Muenke M (2010) Holoprosencephaly due to numeric chromosome abnormalities. Am J Med Genet C Semin Med Genet 154(C):146–8 32. Bendavid C, Dup´e V, Rochard L, et al. (2010) Holoprosencephaly: An update on cytogenetic abnormalities. Am J Med Genet Part C Semin Med Genet 154(C):86–92 33. Roessler E, Muenke M (2010) The molecular genetics of holoprosencephaly. Am J Med Genet C Semin Med Genet 154(C):52–61 34. Porter JA, Young KE, Beachy PA (1996) Cholesterol modification of hedgehog signaling proteins in animal development. Science 274:255–9

35. Bae GU, Domen´e S, Roessler E, et al. (2011) Mutations in CDON, encoding a hedgehog receptor, result inholoprosencephaly and defective interactions with other hedgehog receptors. Am J Hum Genet 89:231–40 36. Bear KA, Solomon BD, Antonini S, et al. (2014) Pathogenic mutations in GLI2 cause a specific phenotype that is distinct from holoprosencephaly. J Med Genet 51:413–8 37. Ming JE, Muenke M (2002) Multiple hits during early embryonic development: digenic diseases and holoprosencephaly. Am J Hum Genet 71:1017–32 38. Mercier S, Dubourg C, Garcelon N, et al. (2011) New findings for phenotype–genotype correlations in a large European series of holoprosencephaly cases. J Med Genet 48:752–60 39. Brown LY, Odent S, David V, et al. (2001) Holoprosencephaly due to mutations in ZIC2: alanine tract expansion mutations may be caused by parental somatic recombination. Hum Mol Genet 10: 791–6 40. Abe Y, Oka A, Mizuguchi M, et al. (2009) EYA4, deleted in a case with middle interhemispheric variant of holoprosencephaly, interacts with SIX3 both physically and functionally. Hum Mutat 30:E946–55 41. Kauvar EF, Muenke M (2010) Holoprosencephaly: recommendations for diagnosis and management. Curr Opin Pediatr 22:687–95 42. Lewis AJ, Simon EM, Barkovich AJ, et al. (2002) Middle interhemispheric variant of holoprosencephaly: a distinct cliniconeuroradiologic subtype. Neurology 59:1860–5 43. Solomon BD, Pineda-Alvarez DE, Gropman AL, et al. (2012) High intellectual function in individuals with mutation–positive microform holoprosencephaly. Mol Syndromol 3:140–2 44. Jellinger K, Gross H, Kaltenback E, Grisold W (1981) Holoprosencephaly and agenesis of the corpus callosum: frequency of associated malformations. Acta Neuropathol 55:1–10 45. Marcorelles P, Loget P, Fallet-Bianco C, et al. (2002) Unusual variant of holoprosencephaly in monosomy 13q. Pediatr Dev Pathol 5:170–8 46. Shiota K, Yamada S (2010) Early pathogenesis of holoprosencephaly. Am J Med Genet C Semin Med Genet 154(C):22–8 47. Muller F, O’Rahilly R (1989) Mediobasal prosencephalic defects, including holoprosencephaly and cyclopia, in relation to the development of the human forebrain. Am J Anat 185:391–414 48. Geng X, Speirs C, Lagutin O, et al. (2008) Haploinsufficiency of Six3 fails to activate Sonic hedgehog expression in the ventral forebrain and causes holoprosencephaly. Dev Cell 15:236–47 49. Warr N, Powles-Glover N, Chappell A, et al. (2008) Zic2-associated holoprosencephaly is caused by a transient defect in the organizer region during gastrulation. Hum Mol Genet 17:2986–96 50. Taniguchi K, Anderson AE, Sutherland AE, Wotton D (2012) Loss of Tgif function causes holoprosencephaly by disrupting the SHH signaling pathway. PLoS Genet 8:e1002524 51. Incardona JP, Gaffield W, Lange Y, et al. (2000) Cyclopamine inhibition of Sonic hedgehog signal transduction is not mediated through effects on cholesterol transport. Dev Biol 224:440–52 52. Klingensmith J, Matsui M, Yang YP, Anderson RM (2010) Roles of bone morphogenetic protein signaling and its antagonism in holoprosencephaly. Am J Med Genet C Semin Med Genet 154(C): 43–51 53. Golden J, Bracilovic A, McFadden K, et al. (1999) Ectopic bone morphogenetic proteins 5 and 4 in the chicken forebrain leads to

39

Developmental Neuropathology

54.

55.

56.

57.

58.

59. 60. 61.

62.

63.

64.

40

cyclopia and holoprosencephaly. Proc Natl Acad Sci U S A 96:2439– 44 Srivastava K, Hu P, Solomon BD, et al. (2012) Molecular analysis of the Noggin (NOG) gene in holoprosencephaly patients. Mol Genet Metab 106:241–3 Kauvar EF, Hu P, Pineda-Alvarez DE, et al. (2011) Minimal evidence for a direct involvement of twisted gastrulation homolog 1 (TWSG1) gene in human holoprosencephaly. Mol Genet Metab 102:470–80 Ekker S, Ungar A, Greenstein P, et al. (1995) Patterning activities of vertebrate hedgehog proteins in the developing eye and brain. Curr Biol 5:944–55 Hu D, Helms JA (1999) The role of sonic hedgehog in normal and abnormal craniofacial morphogenesis. Development 126:4873– 84 Marcucio R, Hallgrimsson B, Young NM (2015) Facial Morphogenesis: Physical and Molecular Interactions between the Brain and the Face. Curr Top Dev Biol 115:299–320 Martin RA, Carey JG (1982) A review and case report of aprosencephaly and the XK syndrome. Am J Med Genet 11:369–71 Goldsmith CL Tawagi GF, Carpenter BF, et al. (1993) Mosaic r(13) in an infant with aprosencephaly. Am J Med Genet 47:531–3 Guala A, Dellavecchia C, Mannarino S, et al. (1997) Ring chromosome 13 with loss of the region D13S317–D13S285: Phenotype overlap with XK syndrome. Am J Med Genet 72:319–23 Pasquier L, Dubourg C, Gonzales M, et al. (2005) First occurrence of aprosencephaly/atelencephaly and holoprosencephaly in a family with a SIX3 gene mutation and phenotype/genotype correlation in our series of SIX3 mutations. J Med Genet 42:e4 Palmer EE, Mowat D (2014) Agenesis of the corpus callosum: a clinical approach to diagnosis. Am J Med Genet C Semin Med Genet 166(C):184–97 Parrish ML, Roessmann U, Levinsohn MW (1979) Agenesis of the corpus callosum: a study of the frequency of associated malformations. Ann Neurol 6:349–54

65. Edwards TJ, Sherr EH, Barkovich AJ, Richards LJ (2014) Clinical, genetic and imaging findings identify new causes for corpus callosum development syndromes. Brain 137:1579–613 66. McCabe MJ, Gaston-Massuet C, Tziaferi V, et al. (2011) Novel FGF8 mutations associated with recessive holoprosencephaly, craniofacial defects, and hypothalamo–pituitary dysfunction. J Clin Endocrinol Metab 96:E1709–18 67. Shevell MI (2002) Clinical and diagnostic profile of agenesis of the corpus callosum. J Child Neurol 17:896–900 68. Sotiriadis A, Makrydimas G (2012) Neurodevelopment after prenatal diagnosis of isolated agenesis of the corpus callosum: an integrative review. Am J Obstet Gynecol 206:337.e1–5 69. Friede R (1989) Developmental Neuropathology, 2nd edn. SpringerVerlag, Berlin, pp. 307–329 70. Paul LK, Brown WS, Adolphs R, et al. (2007) Agenesis of the corpus callosum: genetic, developmental and functional aspects of connectivity. Nat Rev Neurosci 8:287–99 71. Donahoo AL, Richards LJ (2009) Understanding the mechanisms of callosal development through the use of transgenic mouse models. Semin Pediatr Neurol 16:127–42 72. Kelberman D, Dattani MT (2007) Genetics of septo-optic dysplasia. Pituitary 10:393–407 73. Raivio T, Avbelj M, McCabe MJ, et al. (2012) Genetic overlap in Kallmann syndrome, combined pituitary hormone deficiency, and septo-optic dysplasia. J Clin Endocrinol Metab 97:E694–9 74. Atapattu N, Ainsworth J, Willshaw H, et al. (2012) Septo-optic dysplasia: antenatal risk factors and clinical features in a regional study. Horm Res Paediatr 78:81–7 75. Barkovich AJ, Fram EK, Norman D (1989) Septo-optic dysplasia: MR imaging. Radiology 171:189–92 76. Zhao L, Zevallos SE, Rizzoti K, et al. (2012) Disruption of SoxB1dependent Sonic hedgehog expression in the hypothalamus causes septo-optic dysplasia. Dev Cell 22:585–96 77. Rugarli EI, Ballabio A (1993) Kallmann syndrome. From genetics to neurobiology. JAMA 270:2713–6

4

Microcephaly ` 4,5 Nathalie Journiac,3 and Sandrine Passemard,1,2,3 Annie Laquerriere, Pierre Gressens2,3,6 1

Department of Genetics, Robert Debr´e Hospital, Paris, France Paris Diderot University, Paris, France 3 Inserm U1141, Robert Debr´ e Hospital, Paris, France 4 Department of Pathology, Pavillon Jacques Delarue, Rouen University Hospital, Rouen, France 5 Region-Inserm Team NeoVasc ERI28, Laboratory of Microvascular Endothelium and Neonatal Brain lesions, Institute of Research Innovation in Biomedecine, Normandy University Rouen, Rouen, France 6 Center for Developing Brain, King’s College, St. Thomas’ Campus, London, UK 2

Definitions, major synonyms and historical perspective Microcephaly is one of the most common clinical signs of a constellation of developmental disorders of the brain [1,2]. Clinically, microcephaly is classically defined by a decrease in brain size that is indirectly reflected by a significant decrease in head circumference by more than two standard deviations (SD) for age and sex, and is closely associated with intellectual disability [3]. Before birth, the term micrencephaly, strictly speaking, is more appropriate as it reflects brain volume reduction measured by imaging or brain weight reduction, both below the third centile compared with the corresponding biometric databases. Microcephalies are clinically divided into two subgroups: primary and secondary. Primary microcephaly is a static developmental anomaly and may be isolated, or associated with minor or major visceral malformations. Primary microcephaly reflects a reduction in neuron production during neurogenesis, while secondary microcephaly is a progressive condition, which develops after birth and attests to white matter development abnormalities or neurodegenerative processes. Presently, the etiology and pathogenesis of primary genetic microcephalies are still incompletely characterized. Among these various conditions, autosomal recessive primary microcephaly (MCPH, which stands for “microcephaly primary hereditary”), previously called microcephalia vera, is a very rare disorder with a recessive inheritance.

Normal embryology This chapter focuses on neurogenesis only, corresponding to the period of neuron production [for review, see 4,5]. The neural tube is made up of neuroepithelial cells that form a pseudostratified epithelium. These neuroepithelial cells divide at the ventricular surface in the ventricular zone and acquire properties of apical radial glial cells at the beginning of neurogenesis. Apical radial glial cells are polarized cells with an apical process at the ventricular surface and a basal process at the pial membrane and populate the ventricular zone (Figure 4.1a,b). These cells have specific cellular features, including adherens junction proteins that are crucial for cell adhesion and ventricular attachment, and a primary cilium at the ventricular surface and express Pax6, Sox2 and Glast markers. They divide symmetrically (proliferative divisions) to maintain their self-renewal at the beginning of neurogenesis (Figure 4.1c), or switch to asymmetric division (neurogenic division) giving rise to apical radial glial cells and young neurons which will later migrate towards the cortical plate, or an intermediate progenitor, a transit-amplifying cell expressing Tbr2 which populates the subventricular zone and produces neurons. During fetal life, a second germinal zone appears in the subventricular zone, the outer subventricular zone (Figure 4.1d), populated by basal progenitors that include radial glial cells which retain a basal process and express the same markers as apical radial glial cells (Figure 4.1b) and an intermediate progenitor. The type of division (symmetric versus asymmetric) of neuroepithelial cells and radial glial cells determines

Developmental Neuropathology, Second Edition. Edited by Homa Adle-Biassette, Brian N. Harding, and Jeffrey A. Golden. © 2018 John Wiley & Sons Ltd. Published 2018 by John Wiley & Sons Ltd.

41

Developmental Neuropathology (b) IZ/SP CP

(a)

PIA

iN

bRGC

Progenitor cell types aRGC apical Radial glial cell bRGC basal Radial Glial Cell IPC Intermediate Progenitor Cell Neurons N Neuron iN Interneuron

IPC

VZ

Cortical Plate Subplate Intermediate Zone Outer Subventricular Zone iSZV inner Subventricular Zone Ventricular Zone VZ CP SP IZ oSVZ

iSVZ

VZ

iSVZ

oSVZ

oSVZ

IZ/SP

CP

N

aRGC apical process aRGC Centrosome Cilium (c)

VENTRICLE

Symmetric division

(d)

Asymmetric division

or +

aRGC

2 aRGC

+

aRGC

aRGC

+

IPC

aRGC

N

Figure 4.1 Overview of the different types of neural progenitors and their division modes during neurogenesis. (a) GW25 brain coronal section. (b) Diagram showing the different neural progenitors populating the VZ, i and oSVZ and generated neurons at mid neurogenesis. (c) Symmetric division of neural progenitors. (d) Asymmetric division of neural progenitors and their daughter cells.

the balance between proliferation and differentiation of progenitors, and the resulting final number of neurons. It depends on the spindle pole orientation and the inheritance of the membrane determinants, mother/daughter centriole, cilium and transcription factors between mother and daughter cells (Figure 4.1c,d).

42

Epidemiology Microcephaly has a prevalence of about 2–3% worldwide [6]. The incidence of primary genetic microcephaly is estimated

Microcephaly Chapter 4

between 0.5 and 3 per 100 000 births in Western countries, and is much higher in the Indian subcontinent where consanguinity is high and where most of the published families live [7,8].

Clinical features MCPH patients exhibit microcephaly at birth that may already be detectable by the second trimester of pregnancy [9] since it results from insufficient neurogenesis during fetal development, mainly affecting the cerebral cortex. Clinical features of MCPH are variable from one case to another. Affected babies are usually of normal height and weight. However, in utero growth restriction may be associated with microcephaly, and, in that case, investigations for microcephalic dwarfism such as Seckel syndrome or microcephalic osteodysplastic primordial dwarfism type II should be undertaken. MCPH and microcephalic dwarfism are phenotypically distinguishable; both have in utero and postnatal growth restriction and skeletal anomalies associated with microcephaly. It is now known that MCPH and Seckel syndrome are caused by mutations in the same genes and represent a phenotypic continuum [10]. Patients with primary genetic microcephaly usually display normal motor development, but they often exhibit mild to moderate intellectual disabilities and epilepsy, both features influencing development. The major concern in microcephaly is the intellectual prognosis of these patients, which is critical for genetic counseling. Very early detection of microcephaly is an adverse sign: severe microcephaly during infancy (−6 to −10 SD) is more often associated with intellectual deficiency. The severity of microcephaly and its etiology are the two key points that influence learning abilities. Secondary microcephaly clinically appears after birth and is not explained by neurogenesis disruption, but rather by synaptogenesis and/or myelination defects, or due to various degenerative processes. In the former case, motor and language development will be delayed from birth with real but slow improvement; chromosomal rearrangements or monogenic disorders are frequently responsible. In the latter case, progressive neurologic degradation with motor regression usually precedes microcephaly: features include severe seizures and stereotypic movements in Rett syndrome, blindness and seizures in ceroid lipofuscinosis, seizures and abnormal ocular movements in De Vivo syndrome, acute neurologic degradation with coma, hypotonia and liver/hematologic failure in mitochondrial disorders.

Imaging Primary microcephaly is usually evident on ultrasound examination and magnetic resonance imaging (MRI) during the third trimester of the pregnancy from 26 weeks of gestation when primary fissures, which are normally present, are lacking, with absent or poorly formed Sylvian and central fissures. It may be

detectable from 20 weeks of gestation in early severe microcephaly. In all cases, complementary fetal MRI is required, to enable detection of cortical mantle abnormalities such as polymicrogyria, and abnormalities of other brain structures, in particular corpus callosum, basal ganglia, as well as brainstem and cerebellum. Furthermore, MRI allows for the detection of potential visceral malformations in favor of a syndrome, notably Feingold, Rubinstein–Taybi, Smith–Lemli–Opitz or Cornelia de Lange syndromes, which are constantly associated with primary microcephaly.

Genetics Genetic investigations in fetuses and children with microcephaly are based on the combination of clinical phenotype/ diagnosis, results of biological and imaging investigations and neuropathology findings in cases of termination of pregnancy. Advances in next-generation sequencing have radically changed the genetic approach for these conditions. Next-generation sequencing will very soon become the first genetic test in MCPH and microcephalic dwarfisms, as well as for other diseasecausing genes involved in metabolic diseases or syndromic microcephalies. The main causes of primary and secondary microcephalies are summarized in Tables 4.1–4.4.

Autosomal recessive primary microcephaly To date, mutations in 13 genes have been involved in MCPH (Table 4.1). Mutations in WDR62 gene produce a more severe phenotype than other MCPH genes, resulting in severe epilepsy and intellectual disability due to polymicrogyria or schizencephaly on MRI associated with severe microcephaly. Examples of neuroimaging features of MCPH are shown in Figure 4.2. Dwarfism associated with microcephaly and DNA repair deficiency syndromes Dwarfism associated with microcephaly and DNA repair deficiency syndromes is described in Table 4.2. Chromosomal structural microcephalies Numerous chromosomal anomalies are well-recognized causes of microcephaly (trisomy 13, 18, 21, monosomy 1p36, partial trisomy 12q or tetrasomy 9q), as well as structural chromosomal abnormalities, mainly represented by microdeletional syndromes; the main abnormalities are described in Table 4.3. In some cases, fetal anomalies with both craniofacial dysmorphism and several brain and visceral malformations do not correspond to a known nosological framework (Table 4.3), but to de novo pangenic or subtelomeric rearrangements (deletions, duplications or copy number variants). Monogenic syndromic microcephalies Monogenic syndromic microcephalies are described in Table 4.4.

43

Developmental Neuropathology

Table 4.1 Primary autosomal recessive microcephaly (MCPH). MCPH

Gene

Protein

Locus

OMIM

Brain Development Defects

Ref.

1

MCPH1

Microcephalin

8p23

251200

29

2

WDR62

WD repeat domain 62

19q12

604317

3

CDK5RAP2

9q33.3

608201

4 5

CASC5 ASPM

15q15.1 1q31

604321 605481

CENPJ STIL CEP135 CEP152 ZNF335

13q12.2 1p32 4q12 15q21.1 20q13.12

609279 612703 614673 604321 615095

11 12 13

PHC1 CDK6 CENPE

Polyhomeotic-like 1 protein Cyclin-dependent kinase 6 Centromeric protein E

12p13.31 7q21.2 4q24

615414 616080 616051

Primary microcephaly Primary microcephaly, polymicrogyria, simplified gyration Primary microcephaly Primary microcephaly Primary microcephaly Primary microcephaly Primary microcephaly, brain stem and cerebellar hypolasia Primary microcephaly Primary microcephaly Primary microcephaly

33 34

6 7 8 9 10

Cyclin dependent kinase 5 regulatory associated protein 2 Cancer susceptibility candidate 5 Abnormal spindle-like, microcephaly associated Centromeric protein J SCL/TAL1 interrupting locus Centrosomal protein 135kD Centrosomal protein 152kD Zing Finger protein 335

Primary microcephaly, periventricular heterotopia Primary microcephaly, microlissencephaly, schizencephaly, periventricular heterotopia Primary microcephaly

30, 31

32

32 35 36 37 38 39 40 41

OMIM, Online Mendelian Inheritance in Man number

Metabolic microcephalies For metabolic microcephalies, see Chapters 28–38.

Neuropathology The neuropathological classification of microcephaly parallels the clinical classification to include primary and secondary microcephaly. In primary microcephaly due to MCPH there may be normal gyration or a simplified pattern referred to as microlissencephaly. Although also primary, microcephaly associated with other brain anomalies but with no significant associated visceral malformations, chromosomal and genomic syndromes represent the main differential diagnoses of MCPH. Secondary microcephaly, also called environmental or acquired microcephaly, is related to infections, vascular ischemic hemorrhagic lesions or toxic insults that interfere with normal brain development. It is important that examination of fetal brains should be performed only by specialists with knowledge of normal macroscopic and microscopic appearances of the brain and other organs at different maturational stages and of other parameters, such as skeletal measurements. The macroscopic confirmation of microcephaly is usually easy from 26 weeks of gestation, owing to the presence of all primary fissures, and the development of secondary sulci. The occipito-frontal diameters and brain weights are below the third centile, either with a normal appearance of the primary fissures, secondary and tertiary sulci (microcephalia vera; Figure 4.3a) or with incomplete formation of the convolutions (microcephaly with gyral simplification).

44

Gyration is less developed than expected for the age, either corresponding to a lower gestational age (harmonious state) (Figure 4.3b), or with some missing primary fissures, while some secondary sulci are already in place (disharmonious state; Figure 4.3c). Cerebral hypoplasia often predominates in the frontal lobes (Figure 4.3d), and the corpus callosum may be absent (Figure 4.3e). On cut sections, the Sylvian fissure is usually dysmorphic (Figure 4.3f). So far, histological reports of MCPH are rare. In microcephalia vera, the columnar organization in the germinal matrix is affected, with a lack of columnar expansion, but without any disturbance in the formation of primary fissures and sulci. Cortical lamination is normal but neuronal density is decreased and there is associated early depletion of progenitors in the ventricular and subventricular zones. In microcephaly with simplified gyral pattern, cortical plate thickness is preserved, either with normal lamination (Figures 4.4a,b), with diffuse microcolumnar arrangement (reminiscent of focal cortical dysplasia type 1a; Figure 4.4c,d) [11], or patchy microcolumnar dysplasia predominating in the frontal and temporal lobes. The germinal zone and the ganglionic eminences may disappear prematurely, sometimes containing germinolysis pseudocysts, or abnormally persist. The corpus callosum may be absent or hypoplastic, due to neuron depletion in layers II and III. Layer II is sometimes poorly demarcated from layer I, or forms multiple small protrusions into layer I leading to a festooned appearance (Figure 4.4e). Polymicrogyria, when present, is generally focal (Figure 4.4f). Associated anomalies of gyration consist of nodular or streaky heterotopias located in the outer subventricular zone, or

RAD50 MRE11A LIG4 NHEJ1

FANC A-N

ERCC6/8

ERCC1-5, XPA/C, DDB2 ERCC2/3

Nijmegen breakage syndrome-like disorder

Severe combined immunodeficiency (SCID) with microcephaly, growth restriction and sensitivity to ionizing radiation

Fanconi anemia

Cockayne syndrome

Xeroderma pigmentosum

Skin cancer predisposition

Intellectual disability, hyporeflexia, spasticity, ataxia, choreoathetosis Intellectual disability

Hypogammaglobulinemia

–

Immunodeficiency, pancytopenia, lymphoma, radiosensitivity SCID – recurrent infections with opportunistic organisms, lymphopenia, agammaglobulinemia Anemia, cancer predisposition, myelodysplasia, leukemia

Immunodeficiency, cancer predisposition None

–

Pancytopenia

Haematological and Immunological Defects

Intellectual disability, neurodegeneration

Intellectual disability

developmental delay (intellectual disability) Intellectual disability

Intellectual disability, Moyamoya disease, multiple aneurysms and infarcts normal IQ (or mild intellectual disability) Mild intellectual disability

Mild intellectual disability, seizures, cerebellar vermis hypoplasia, pachygyria, simplified gyri, hyperactivity

Neurological Featuresa

Icthyosis, photosensitivity, cataract, brittle hair and nails

‘Cachectic’ dwarfism, progressive retinitis pigmentosa sensorineural deafness, cutaneous photosensitivity, thin/dry hair, premature aging with progeroid appearance, contractures Icthyosis, photosensitivity, short stature, deafness, cataract

Heart, kidney and limb malformations, dermal pigmentation changes, low birth weight, short stature

Growth delay, facial features similar to NBS, skin anomalies Growth restriction and dystrophy, dysmorphism

Bird-like’ facies

Dwarfism, craniofacial dysmorphism, bone dysplasia, type II diabetes, ´ cafe-au-lait spots Growth restriction, ‘bird-like’ facies

Pre-/postnatal growth restriction, dwarfism, ‘bird-like’ facies

Other Features

54

For review, see 55

611291

227650, 300514, 227645, 605724, 227646, 600901, 603467, 602956, 609053, 609054, 608111, 609644, 610832 216400, 133540

126380, 278700, 610651, 278720, 278730, 278740, 278760, 278780 126340, 133510

606593

51 52 53

50

49

42 43 44 45 46 47 48

Ref.

613078

251260

613676 613823 614728 614851 615807 210720

210600 606744

OMIM

ATR, ataxia–telangiectasia and RAD3-related protein; DDB2, DNA damage-binding protein 2; ERCC, excision-repair cross-complementing protein; FANC A-N, Fanconi anemia complementation group A-N; LIG4, ATP-dependent DNA ligase IV; NBS1, Nibrin; NHEJ1, non-homologous end-joining factor 1; OMIM, Online Mendelian Inheritance in Man number; PCNT, pericentrin 2; XP, xeroderma pigmentosum complementation group.

Photosensitive trichothiodystrophy

LIG4 syndrome

NBS1

CENPJ CEP152 CEP63 NIN DNA2 PCNT

ATR RBBP8

Gene

Nijmegen breakage syndrome (NBS)

Seckel syndrome: 1 2 3 4 5 6 7 8 Microcephalic osteodysplastic primordial dwarfism type 2 (MOPD2)

Syndrome

Table 4.2 Dwarfism associated with microcephaly and DNA repair deficiency syndromes (inheritance: autosomal recessive).

Developmental Neuropathology

Table 4.3 Chromosomal structural microcephalies (inheritance, chromosomal).a Syndrome

Genetic Anomaly

Key Signs

1q43-q44 deletion

Submicroscopic deletion

4p deletion (Wolff-Hirshhorn)

WHSC1 and WHSC2: 2 non-overlapping regions of chromosome 4 which lead to 4p- phenotype Continuous gene deletion syndrome NHAR-mediated deletion. No convincing gene-phenotype established so far

Developmental delay, then cognitive impairment, microcephaly, growth restriction, dysmorphic features (hypertelorism, depressed nasal bridge), congenital heart defects Intrauterine growth restriction; typical dysmorphism: hypertelorism, prominent glabella, giving a Greek-helmet appearance, broad nasal tip, bilateral cleft lip, short philtrum, microcephaly, large ears Hypertelorism, round face; typical high-pitched voice Short stature, hyypertelorism, short upturned nose, long philtrum, everted lower lip, aortic stenosis, specific infantile hypercalcemia, overfriendly personality Cleft palate, congenital heart defect (mainly, but not exclusively, conotruncal), abnormal nose shape (broad base, bulbous tip, parallel edges, velar insufficiency) Microcephaly (2/3), congenital heart abnormality (1/3), ligamentous laxity or joint hypermobility, hypotonia (5/21), seizures (3/21); note: incomplete penetrance – the deletion can be found in normally developed or borderline carrier parent Microcephaly, brachycephaly (18/30 or 60%), frontal bossing, deep-set eyes, straight eyebrows, narrow palpebral fissures, flat nose, and pointed chin; heart defect

5p deletion (cri-du-chat) 7q11 deletion (Williams)

22q11 (DiGeorge/velocardiofacial)

NHAR-mediated deletion

1q21.1 microdeletion

NHAR-mediated deletion

1p36 microdeletion

Deletion of variable size in the sus-subtelomeric 1p region

a Some

emblematic microcephaly syndromes. The syndromes associated with DNA repair defect are listed separately in Table 4.2. NHAR, non-homologous allelic recombination – faulty recombination between highly similar loci (duplicons) located at relatively short distance from one another; common mechanisms for recurrent microrearrangements.

intermingled with the basal ganglia or the internal capsule (Figure 4.5a–c). Purkinje cell heterotopias may be observed in the cerebellar white matter (Figure 4.5d). The neuropathology of molecularly-proven Seckel syndrome remains unknown, a single description in a fetus has been published [12]. The case presented with a bird-like facies, extreme microcephaly, a smooth cerebral mantle with a thin cortical plate with no obvious demarcation from the underlying intermediate zone, hypoplastic corpus callosum, hippocampal dysplasia and cystic germinal matrix. Mutations in the WDR62 and NDE1 genes cause the most severe forms of MCPH. WDR62-mutated patients exhibit extreme microcephaly (≤10 SD) and a wide pattern of cortical malformations ranging from microlissencephaly to pachygyria. The cortical plate has been described as abnormally thin, with absent cortical layers II and III underneath a preserved layer I. Overmigration of neuroglial nests through the pia into the subarachnoid spaces, streaky heterotopia in the intermediate zone and clusters of immature neurons in the outer subventricular zone have also been reported [13]. To date, brain lesions of patients harboring mutations in the NDE1 gene, a Lis1 interacting protein implicated in mitotic progression and orientation, have not been morphologically characterized, but on MRI studies show severe microlissencephaly or microcephaly with extreme simplified gyri (−8 to −10 SD), agenesis of the corpus callosum and hydranencephaly [14].

46

Differential diagnosis The role of environmental factors in microcephaly is well known [15], interfering with neuron production in the germinal zone and ganglionic eminences. Neuronal radial and tangential migration, as well as cellular maturation abnormalities resulting in the disorganization of the different brain structures, is observed in association with destructive lesions. Intrauterine infections responsible for microcephaly include cytomegalovirus, herpes simplex 1 and varicella zoster viruses, toxoplasmosis, syphilis and acquired immunodeficiency syndrome, and, more recently, Zika virus (see Chapter 41). Antiepileptic drugs (carbamazepine, phenytoin, sodium valproate) and alcohol consumption during pregnancy are also well-recognized causes of microcephaly. In fetal alcohol syndrome, there is characteristic craniofacial dysmorphism, intrauterine growth restriction and a wide spectrum of lesions involving supra- and infratentorial structures, including midline anomalies (agenesis of the corpus callosum, holoprosencephaly), neural tube closure defects and neuronal heterotopia. Addictive drugs such as opioids and cocaine are also responsible for microcephaly due to intracranial hemorrhage and infarcts. In maternal phenylketonuria, microcephaly occurs in about 75% of offspring with delayed psychomotor milestones in nearly all cases. Brain lesions include ventricular dilatation, prominent

Microcephaly Chapter 4

Table 4.4 Monogenic syndromic microcephalies. Syndrome

Inheritance

Genomic Anomaly

Key Signs

Renpenning (X-linked microcephaly)

X-linked

PQBP1

Feingold

Autosomal dominant

MYCN

Cohen

Autosomal recessive

COH1

Rubinstein–Taybi

Autosomal dominant or chromosomal

CREBP mutation or microdeletion, EP300

Rett

X-linked, dominant

MECP2

Rett variants with early epilepsy Congenital Rett

X-linked dominant

CDKL5/STK9

Autosomal dominant

FOXG1, MEF2C

Mowat-Wilson

Autosomal dominant

ZEB2 (or ZFHX1B or SIP1)

Smith–Lemli–Opitz

Autosomal recessive

7-dehydro-cholesterol-delta-7 reductase deficiency

Cornelia de Lange

Autosomal dominant

NIPBL

X-linked

SMC1A

Postnatal microcephaly, long narrow face, short stature with lean body build, coloboma and other eye defects, stiff thumbs, small testes; note: PQBP1 mutation is also associated with nonsyndromic mental restriction Microcephaly with cerebral and cerebellar white matter anomalies, often with normal intelligence, multiple digestive tract atresias (esophagus, duodenum), brachymesophalangy of the fifth fingers, 4–5 syndactyly of the toes and short palpebral fissures Microcephaly with large corpus callosum, cerebellar hypoplasia, focal micropolygyria, truncal obesity, mental restriction, poor muscle tone, narrow hands and feet, and distinctive facial features with prominent upper central teeth, retinitis pigmentosa and leukopenia Microcephaly, Chiari type 1 malformation, intellectual disability, short stature, broad deviated thumbs and great toes (not with EP300 mutations), and characteristic facial features (antimongoloid eye slant, hypertelorism, and a convex nose with the columella protruding below the alae nasi). Sometimes agenesis of the corpus callosum Girls with normal pre- and perinatal history, normal development and head circumference up to 6 months of age, subsequent regression of social and motor skills, hand wringing or clapping with frequent mouthing, truncal and gait ataxia, epilepsy, alternating bouts of polypnea and apnea Girls with a severe form of Rett with severe epilepsy onset before 6 months of age Boys and girls with early onset, severe encephalopathy and Rett-like stereotypic movements, epilepsy and cerebral malformations Postnatal microcephaly, agenesis or dysgenesis of the corpus callosum, Hirschsprung disease or constipation, facial dysmorphism. deep-set, large eyes, a broad low nasal bridge, prominent columella, an open-mouthed expression, prominent chin, and large uplifted, fleshy ear lobules. Agenesis of the corpus callosum in some Microcephaly, cerebral midline malformation, ptosis, bi-temporal narrowing, anteverted nostrils, broad nasal tip, micrognathia, cleft palate, visceral anomalies, hypospadias, toe 2–3 syndactyly, postaxial polydactyly. High 7OH cholesterol in serum. May be observed with holoprosencephaly Intrauterine growth restriction, postnatal short stature, microcephaly, limb reduction defect, cardiac defects, hirsutism, facial dysmorphism with synophrys. Cerebral midline anomalies, optic nerve hypoplasia, brainstem and cerebellar hypoplasia, tethered cord, hypomyelination of the frontal lobes. Marked variability with severe and mild cases with normal growth and occipital frontal circumference. X linked variant: milder form, no limb anomalies. Female carriers may express the phenotype. Mild phenotype.

pseudo TORCH

Autosomal dominant Autosomal dominant X-linked Autosomal recessive

SMC3 RAD21 HDAC8 OCLN

Aicardi Goutieres

Autosomal recessive

TREX1

X linked dominant. Female carriers may exhibit a mild phenotype Microcephaly with periventricular band-like calcifications, abnormal gyration and polymicrogyria Microcephaly with intracranial calcifications (basal ganglia and white matter), leukodystrophy, increased CSF interferon

RNASEH2B RNASEH2C RNASEH2A SAMHD1 ADAR IFIH1 CSF: cerebrospinal fluid; PQBP1, polyglutamine tract binding protein 1 gene

47

Developmental Neuropathology

Figure 4.2 T1-weighted images showing the typical features of MCPH1 (a) WDR62 (b) and ASPM (c) compared with an age-matched control (d).

diminution of the white matter, and diffuse neuronal necrosis [16]. Investigation for vascular insufficiency, maternal malnutrition or anemia, maternal coagulopathy is also essential. Microcephaly and microlissencephaly associated with other brain lesions are easily distinguished from MCPH, and are mainly due to mutations in tubulin genes (TUBA1A, TUBB2B, TUBB3, TUBB5 and TUBG1). These genes disrupt several steps of brain development, including proliferation and early neuronal differentiation leading to microcephaly, and defects of migration and integrity and maintenance of the pial membrane. In this group of disorders, in addition to microcephaly, cortical dysplasia such as lissencephaly, pachygyria or micropolygyria is always present, combined with various brain abnormalities (see Chapters 6 and 7). The basal ganglia and hippocampi are dysplastic, the corpus callosum is absent or hypoplastic, as are the corticospinal tracts. Prominent germinative zones, neuronal overmigration through the pial membrane may also be observed. Infratentorial lesions consist of brainstem and cerebellar hypoplasia and/or dysplasia [17]. Microlissencephaly due to mutations in ARFGEF2 gene are associated with periventricular

48

nodular heterotopias [18]. In microcephaly with pontocerebellar hypoplasia due to mutations in TSEN54 gene (pontocerebellar hypoplasia type V), infratentorial signs predominate [19]. Most recently described is an extreme form of autosomal recessive microlissencephaly associated with mutations in citron kinase: the resulting phenotype includes cortical, hippocampal and cerebellar dysplasia and the prominent presence of multinucleated neurons throughout the neuraxis [20]. Chromosomal microcephalies are usually detected using karyotype and comparative genomic hybridization array analyses performed on the basis of fetal ultrasound and brain MRI findings (Table 4.3). To date, over 400 syndromes have been reported in association with microcephaly. They are also suspected on the basis of fetal ultrasound and brain MRI findings. Importantly, except for those reported in Table 4.4, microcephaly is not always a key sign of the syndrome. Finally, microcephaly may be a main feature in some metabolic disorders and has been reported in ceroid lipofuscinosis type 1, congenital disorders of N- and O-glycosylation, which are linked with more than 40 disease-causing genes, mitochondrial encephalopathies,

Microcephaly Chapter 4

(a)

(b)

(c)

(d)

(e)

(f)

Figure 4.3 (a) Microcephalia vera: termination of pregnancy for microcephaly at 35 weeks of gestation; brain weight: 251 g; third centile for term. All primary fissures, secondary and tertiary convolutions are present. (b) Microcephaly with harmonious simplified pattern. Termination of pregnancy for microcephaly at 31 weeks of gestation; brain weight (128 g) and external maturation correspond to 27 weeks of gestation. (c) Microcephaly with dysharmonious simplified gyration, with dysmorphic Sylvian fissure (arrow), presence of superior temporal fissure (normally

present at 27 weeks of gestation), but with no pre- and post-central gyri (normally present at 25 weeks of gestation). Termination of pregnancy at 31 weeks of gestation; brain biometric data correspond to 24 weeks. (d) Marked frontal lobe hypoplasia (same case). (e) Agenesis of the corpus callosum (arrow) (same case). (f) Coronal section through the diencephalon, showing almost no fissures; brain weight: 88 g (< third centile for 22 weeks of gestation). Termination of pregnancy at 28 weeks for isolated microlissencephaly.

49

Developmental Neuropathology

(a)

(b)

(c)

(d)

(e)

(f)

Figure 4.4 (a) Immature cortical plate with accumulation of granular neurons in layer II, other layers being hardly visible (×20). Termination of pregnancy at 26 weeks of gestation for microcephaly with simplified gyration and short corpus callosum. (b) Normal cortical plate in an age-matched control (×20). (c) Diffuse microcolumnar dysplasia of the cerebral cortex (×200). Termination of pregnancy at 23 weeks of gestation, brain measurements correspond to 20 weeks (recurrence

50

of case depicted in d). (d) Marked neuron depletion (×250). Termination of pregnancy at 36 weeks of gestation for microcephaly with simplified gyration, brain measurements correspond to 32 weeks). (e) Small protrusions of external granular layer within layer I (×400), not to be confused with status verrucosus. Termination of pregnancy at 33 weeks of gestation for isolated microcephaly. (f) With focal frontal micropolygyria (arrow) (×10).

Microcephaly Chapter 4

(a)

(b)

(c)

(d)

Figure 4.5 (a) Periventricular neuronal heterotopia (arrow) and premature regression of the ganglionic eminences (×200) (same case as in figure 4.1c). Termination of pregnancy at 31weeks of gestation. (b) Nodular neuronal heterotopia intermingled with the internal capsule (arrow) and the basal ganglia (×200). Termination of pregnancy at 19 weeks of gestation for severe microcephaly; brain weight: 14.3 g, less than third centile. (c) Numerous streaky

neuronal heterotopia in the intermediate zone (×200). Termination of pregnancy at 30 weeks of gestation for microcephaly with simplified gyration; brain weight: 89 g, corresponding to 23 weeks of gestation. (d) Purkinje cell heterotopia (arrow) (×400). Termination of pregnancy at 35 weeks of gestation for microcephaly with simplified gyral pattern.

deficiencies of pyruvate dehydrogenase, mevalonate kinase and Glut1 (De Vivo syndrome) and Neu-Laxova syndrome. In these conditions, careful pathological examination of the fetus or child, including histology and electron microscopy raise suspicion of a metabolic disease, which must be confirmed by biochemical and genetic testing (see Chapters 33 and 35).

regulation, alterations in microtubule dynamics and centrosomal mechanisms involved in the regulation of the number of neuron precursors. Most of the MCPH microcephalic dwarfism genes are also expressed in the ventricular zone during neurogenesis and code for centrosomal proteins that play a role in spindle orientation and segregation. Some of these genes are also expressed by the progenitors of the subventricular zone, which are thought to be involved in cortical expansion in humans. Different mouse models of microcephaly have been produced. Deletion of Mcph1 has been associated with an excess cell death of progenitors [21]. Deletion of Cdk5rap2 leads to a depletion of apical progenitors, increased cell cycle exit, and premature neuronal differentiation [22]. Deletion of WDR62 leads to premature differentiation of progenitors [23]. Studies of Aspm mutants

Animal models and pathogenesis MCPH phenotypes turbance of several the mitotic spindle, signaling pathways

and brain lesions result from the dismechanisms, including disorientation of anomalies of chromosome condensation, involved in DNA repair, transcriptional

51

Developmental Neuropathology have shown that Aspm is crucial for maintaining a cleavage plane orientation that allows symmetric divisions of progenitors [24]. Interestingly, the microcephaly in Aspm knock-out mice is much less severe than in humans with ASPM mutations, suggesting additional roles for the protein have been acquired during evolution of mammals, in particular in humans. Pharmacological blockade of vasoactive intestinal peptide, a factor of maternal origin, leads to microcephaly partially through the reduced expression of Mcph1. This, supports the hypothesis that environmental factors impinge on the intrinsic control of neural progenitor proliferation and therefore modulate the final size of the brain [25]. Microlissencephaly with multinucleated neurons resulting from citron kinase defects has a radically different underlying pathogenesis since citron kinase is a multidomain protein that localizes to the cleavage furrow and midbody of mitotic cells, where it is required for the completion of cytokinesis. Its role in neurogenesis has been worked out both in man and in rodent models, which also show a remarkably similar phenotype [26–28].

Treatment, future perspectives, conclusions The key challenges of microcephalies may be summarized as follows: 1. identification of further genes mutated in primary microcephalies 2. better understanding of the cellular and molecular mechanisms responsible for the deficit of progenitors (type of progenitor affected, symmetric versus asymmetric division, deficit of cell division versus enhanced cell death) 3. identification of the roles acquired during evolution (especially in humans) by some of the genes linked to primary microcephaly (the use of human brain tissues and of human-induced pluripotent stem cells might be critical in this regard) 4. determination of the brain regions that are most affected and those that are relatively spared with better characterization of the neuropsychological profile (strengths and weaknesses) of the patients with different types of microcephaly, potentially allowing for improved rehabilitation strategies 5. understanding of the cross-talk between environmental factors and genetically determined mechanisms of control of the progenitor pool 6. identification of molecular pathways that could be targeted by pharmacological agents designed to improve the function of microcephalic brains.

References 1. Barkovich AJ, Guerrini R, Kuzniecky RI, et al. (2012) A developmental and genetic classification for malformations of cortical development: update 2012. Brain 135:1348–69.

52

2. Barkovich AJ, Kuzniecky RI, Jackson GD, et al. (2005) A developmental and genetic classification for malformations of cortical development. Neurology 65:1873–87. 3. Passemard S, Kaindl AM, Verloes A. (2013) Microcephaly. Handb Clin Neurol 111:129–41. 4. Florio M, Huttner WB. (2014) Neural progenitors, neurogenesis and the evolution of the neocortex. Development 141:2182–94. 5. Taverna E, Gotz M, Huttner WB. (2014) The cell biology of neurogenesis: toward an understanding of the development and evolution of the neocortex. Annu Rev Cell Dev Biol 30:465–502. 6. Van Den Bosch J. (1959) Microcephaly in the Netherlands: a clinical and genetical study. Ann Hum Genet 23:91–116. 7. Szabo N, Pap C, Kobor J, et al. (2010) Primary microcephaly in Hungary: epidemiology and clinical features. Acta Paediatr 99:690– 3. 8. Woods CG, Bond J, Enard W. (2005) Autosomal recessive primary microcephaly (MCPH): a review of clinical, molecular, and evolutionary findings. Am J Hum Genet 76:717–28. 9. Kaindl AM, Passemard S, Kumar P, et al. (2009) Many roads lead to primary autosomal recessive microcephaly. Prog Neurobiol 90:363– 83. 10. Verloes A, Drunat S, Gressens P, Passemard S. (2013) Primary autosomal recessive microcephalies and Seckel syndrome spectrum disorders. In: RA Pagon, MP Adam, HH Ardinger, et al., eds. GeneReviews. Seattle, WA: University of Washington; 2009 [updated October 2013]. 11. Blumcke I, Thom M, Aronica E, et al. (2011) The clinicopathologic spectrum of focal cortical dysplasias: a consensus classification proposed by an ad hoc Task Force of the ILAE Diagnostic Methods Commission. Epilepsia 52:158–74. 12. Fitzgerald B, O’Driscoll M, Chong K, et al. (2012) Neuropathology of fetal stage Seckel syndrome: a case report providing a morphological correlate for the emerging molecular mechanisms. Brain Dev 34:238–43. 13. Yu TW, Mochida GH, Tischfield DJ, et al. (2010) Mutations in WDR62, encoding a centrosome–associated protein, cause microcephaly with simplified gyri and abnormal cortical architecture. Nat Genet 42:1015–20. 14. Alkuraya FS, Cai X, Emery C, et al. (2011) Human mutations in NDE1 cause extreme microcephaly with lissencephaly [corrected]. Am J Hum Genet 88:536–47. 15. Abuelo D. (2007) Microcephaly syndromes. Semin Pediatr Neurol 14:118–27. 16. Koch R, Verma S, Gilles FH. (2008) Neuropathology of a 4-monthold infant born to a woman with phenylketonuria. Dev Med Child Neurol 50:230–3. 17. Fallet-Bianco C, Laquerri`ere A, Poirier K, et al. (2014) Mutations in tubulin genes are frequent causes of various foetal malformations of cortical development including microlissencephaly. Acta Neuropathol Commun 2:69. 18. Tanyalcin I, Verhelst H, Halley DJ, et al. (2013) Elaborating the phenotypic spectrum associated with mutations in ARFGEF2: case study and literature review. Eur J Paediatr Neurol 17:666–70. 19. Namavar Y, Barth PG, Kasher PR, et al. (2011) Clinical, neuroradiological and genetic findings in pontocerebellar hypoplasia. Brain 134:143–56. 20. Harding BN, Moccia A, Drunat S, et al. (2016) Mutations in citron kinase cause recessive microlissencephaly with multinucleated neurons. Am J Hum Genet 99:511–20.

Microcephaly Chapter 4 21. Gruber R, Zhou Z, Sukchev MJ, et al. (2011) MCPH1 regulates the neuroprogenitor division mode by coupling the centrosomal cycle with mitotic entry through the Chk1-Cdc25 pathway. Nat Cell Biol 13:1325–34. 22. Barrera JA, Kao LR, Hammer RE, et al. (2010) CDK5RAP2 regulates centriole engagement and cohesion in mice. Dev Cell 18:913–26. 23. Xu D, Zhang F, Wang Y, et al. (2014) Microcephaly-associated protein WDR62 regulates neurogenesis through JNK1 in the developing neocortex. Cell Rep 6:104–16. 24. Pulvers JN, Bryk J, Fish JL, et al. (2010) Mutations in mouse Aspm (abnormal spindle-like microcephaly associated) cause not only microcephaly but also major defects in the germline. Proc Natl Acad Sci U S A 107:16595–600. 25. Passemard S, El Ghouzzi V, Nasser H, et al. (2011) VIP blockade leads to microcephaly in mice via disruption of Mcph1–Chk1 signaling. J Clin Invest 121:3071–87. 26. Li H, Bielas SL, Zaki MS, et al. (2016) Biallelic mutations in citron kinase link mitotic cytokinesis to human primary microcephaly. Am J Hum Genet 99:501–10. 27. Sarkisian MR, Li W, Di Cunto F, et al. (2002) Citron-kinase, a protein essential to cytokinesis in neuronal progenitors, is deleted in the flathead mutant rat. J Neurosci 22: RC217. 28. Di Cunto F, Imarisio S, Hirsch E, et al. (2000) Defective neurogenesis in citron kinase knockout mice by altered cytokinesis and massive apoptosis. Neuron 28, 115–27. 29. Jackson AP, Eastwood H, Bell SM, et al. (2002) Identification of microcephalin, a protein implicated in determining the size of the human brain. Am J Hum Genet 71:136–42. 30. Bilg¨uvar K, Ozt¨urk AK, Louvi A, et al. (2010) Whole-exome sequencing identifies recessive WDR62 mutations in severe brain malformations. Nature 467:207–10. 31. Nicholas AK, Khurshid M, D´esir J, et al. (2010) WDR62 is associated with the spindle pole and is mutated in human microcephaly. Nat Genet 42:1010–14. 32. Bond J, Roberts E, Springell K et al. (2005) A centrosomal mechanism involving CDK5RAP2 and CENPJ controls brain size. Nat Genet 37:353–5. Erratum in 2005;37:555. 33. Genin A, Desir J, Lambert N, et al. (2012). Kinetochore KMN network gene CASC5 mutated in primary microcephaly. Hum Mol Genet 21:5306–17. 34. Bond J, Roberts E, Mochida GH, et al. (2002) ASPM is a major determinant of cerebral cortical size. Nat Genet 32:316–20. 35. Kumar A, Girimaji SC, Duvvari MR, Blanton SH. (2009) Mutations in STIL, encoding a pericentriolar and centrosomal protein, cause primary microcephaly. Am J Hum Genet 84:286–90. 36. Hussain MS, Baig SM, Neumann S, et al. (2012) A truncating mutation of CEP135 causes primary microcephaly and disturbed centrosomal function. Am J Hum Genet 90:871–8. 37. Guernsey DL, Jiang H, Hussin J, et al. (2010) Mutations in centrosomal protein CEP152 in primary microcephaly families linked to MCPH4. Am J Hum Genet 87:40–51. 38. Yang YJ, Baltus AE, Mathew RS, et al. (2012) Microcephaly gene links trithorax and REST/NRSF to control neural stem cell proliferation and differentiation. Cell 151:1097–112.

39. Awad S, Al-Dosari MS, Al-Yacoub N, et al. (2013) Mutation in PHC1 implicates chromatin remodeling in primary microcephaly pathogenesis. Hum Mol Genet 22:2200–13. 40. Hussain MS, Baig SM, Neumann S, et al. (2013) CDK6 associates with the centrosome during mitosis and is mutated in a large Pakistani family with primary microcephaly. Hum Mol Genet 22:5199– 214. 41. Mirzaa GM, Vitre B, Carpenter G, et al. (2014) Mutations in CENPE define a novel kinetochore–centromeric mechanism for microcephalic primordial dwarfism. Hum Genet 133:1023–39. 42. O’Driscoll M, Ruiz-Perez VL, Woods CG, et al. (2003) A splicing mutation affecting expression of ataxia–telangiectasia and Rad3related protein (ATR) results in Seckel syndrome. Nat Genet 33:497– 501. 43. Qvist P, Huertas P, Jimeno S, et al. (2011) CtIP Mutations Cause Seckel and Jawad Syndromes. PLoS Genet 7:e1002310. 44. Al-Dosari MS, Shaheen R, Colak D, Alkuraya FS. (2010) Novel CENPJ mutation causes Seckel syndrome. J Med Genet 47:411– 14. 45. Kalay E, Yigit G, Aslan Y, et al. (2011) CEP152 is a genome maintenance protein disrupted in Seckel syndrome. Nat Genet 43:23–6. 46. Sir JH, Barr AR, Nicholas AK, et al. (2011) A primary microcephaly protein complex forms a ring around parental centrioles. Nat Genet 43:1147–53. 47. Dauber A, Lafranchi SH, Maliga Z, et al. (2012) Novel microcephalic primordial dwarfism disorder associated with variants in the centrosomal protein ninein. J Clin Endocrinol Metab 97:E2140– 51. 48. Shaheen R, Faqeih E, Ansari S, et al. (2014) Genomic analysis of primordial dwarfism reveals novel disease genes. Genome Res 24:291– 9. 49. Rauch A, Thiel CT, Schindler D, et al. (2008) Mutations in the pericentrin (PCNT) gene cause primordial dwarfism. Science 319:816– 19. 50. Carney JP, Maser RS, Olivares H, et al. (1998) The hMre11/hRad50 protein complex and Nijmegen breakage syndrome: linkage of double–strand break repair to the cellular DNA damage response. Cell 93:477–86. 51. Waltes R, Kalb R, Gatei M, et al. (2009) Human RAD50 deficiency in a Nijmegen breakage syndrome-like disorder. Am J Hum Genet 84:605–16. 52. Matsumoto Y, Miyamoto T, Sakamoto H, et al. (2011) Two unrelated patients with MRE11A mutations and Nijmegen breakage syndrome-like severe microcephaly. DNA Repair (Amst) 10:314– 21. 53. O’Driscoll M, Cerosaletti KM, Girard PM, et al. (2001) DNA ligase IV mutations identified in patients exhibiting developmental delay and immunodeficiency. Mol Cell 8:1175–85. 54. Buck D, Malivert L, de Chasseval R, et al. (2006) Cernunnos, a novel nonhomologous end–joining factor, is mutated in human immunodeficiency with microcephaly. Cell 124:287–99. 55. Schneider M, Chandler K, Tischkowitz M, Meyer S. (2015) Fanconi anaemia: genetics, molecular biology, and cancer – implications for clinical management in children and adults. Clin Genet 88:13–24.

53

5

Hemimegalencephaly and Dysplastic Megalencephaly Ghayda Mirzaa,1,2 Achira Roy,2 William B. Dobyns,1,2 Kathleen Millen,2 and Robert F. Hevner2,3 1

Department of Pediatrics, University of Washington, Seattle, WA, USA Center for Integrative Brain Research, Seattle Children’s Research Institute, Seattle, WA, USA 3 Department of Neurological Surgery, University of Washington, Seattle, WA, USA 2

Definition and synonyms Hemimegalencephaly (HMEG) and dysplastic megalencephaly (DMEG) are congenital brain malformations characterized by marked overgrowth and dysplasia affecting one or both cerebral hemispheres, and, in some cases, subcortical brain regions as well. Some types of focal cortical dysplasia represent the same disease process in smaller areas of cortex. Many cases of DMEG and HMEG are caused by de novo mutations, constitutional or mosaic, resulting in overactivation of the phosphatidylinositol3-kinase (PI3K)-AKT (serine/threonine kinase)–mammalian target of rapamycin (MTOR) signaling pathway. HMEG is also known as unilateral megalencephaly. Conversely, DMEG has sometimes been called bilateral hemimegalencephaly, denoting the presence of dysplasia, which distinguishes DMEG from other forms of generalized megalencephaly. “Total hemimegalencephaly” refers to HMEG with overgrowth of the ipsilateral brainstem and cerebellum [1].

Many, and probably most, cases of DMEG and HMEG result from mutations in PI3K-AKT-MTOR pathway genes [6]. Among the genes in this pathway that have been linked to HMEG or DMEG are PIK3CA, AKT3, PTEN, DEPDC5, and MTOR (Figure 5.1). Different HMEG or DMEG syndromes can result from mutations of the same, or different genes in this pathway (Figure 5.1). For example, MCAP syndrome is caused by PIK3CA mutations; MPPH syndrome by PIK3R2, CCND2 and AKT3 mutations; and Cowden and Bannayan–Riley–Ruvalcaba syndromes by PTEN mutations. Conversely, mutations that lead to decreased signaling in the PI3K–AKT–MTOR pathway (such as AKT3 deletions) have been shown to be associated with decreased brain growth, or microcephaly [7]. Genetic studies have highlighted the importance of de novo mutations and somatic mosaicism in DMEG and HMEG [8]. New zygotic mutations result in all cells carrying the mutation, such that the entire brain is affected, as in diffuse types of DMEG. In contrast, new postzygotic mutations in neural stem cells cause disease in spatially restricted portions of the brain, such as one hemisphere in HMEG, or a smaller region of cortex in some focal cortical dysplasias.

Epidemiology and genetics Clinical features and differential diagnosis HMEG and DMEG are rare, sporadic disorders that appear in isolation or as part of rare syndromes. A few of the syndromes associated with HMEG include tuberous sclerosis complex, Proteus syndrome, and congenital lipomatosis, overgrowth, vascular malformations, and epidermal nevi syndrome (CLOVES), where overgrowth also involves somatic organs. Syndromes associated with DMEG include the megalencephaly capillary malformation (MCAP) and megalencephaly polymicrogyria polydactyly hydrocephalus (MPPH) syndromes [2–5].

Clinical examination DMEG combines overall megalencephaly (usually defined as cerebral mass greater than 2.5 standard deviations above the mean) with bilateral cortical dysplasia, visible on neuroimaging. In contrast, overall head size is not significantly increased in most cases of HMEG, as only one hemisphere is overgrown, and the contralateral hemisphere is generally smaller than normal [9]. Facial asymmetries may also occur in HMEG [10].

Developmental Neuropathology, Second Edition. Edited by Homa Adle-Biassette, Brian N. Harding, and Jeffrey A. Golden. © 2018 John Wiley & Sons Ltd. Published 2018 by John Wiley & Sons Ltd.

55

Developmental Neuropathology

PTEN (constitutional)

RTK

(FGFRs, EFGRs, IGF1R)

PHTS MEG-autism

lgf1 overexpression mouse

Pten conditional knock-out mouse

PTEN

PDK1

T308 P P S473

mTORC2 PIK3CA (mosaic)

GSK3 TSC1/2

Akt3 gain-offunction mouse

SGK1 Rheb

Conditional transgenic mouse with stabilized β-catenin

MPPH

MCAP FCD HMEG

β-catenin NDRG1 Pik3ca (H1047R, E545K) conditional gainof-function mice

mTORC1

S6K CCND2 (constitutional)

PIK3R2 (constitutional) MPPH

AKT

MEG MPPH HMEG

Adaptors

p110 p85

Adaptors

AKT3 (constitutional/ mosaic)

PI3K

PIP2

PIP3

Cell cycle

4E-BP1

Cell growth Protein synthesis

Cell survival

MTOR (constitutional/ mosaic) HMEG FCD

Figure 5.1 Overview of the PI3K–AKT–MTOR intracellular signaling pathway, and genes linked to different dysplastic megalencephaly (DMEG) and hemimegalencephaly (HMEG) phenotypes. The PI3K–AKT–MTOR pathway regulates cell and tissue growth, in response to signals from receptor tyrosine kinase (RTK) binding by growth factors and other molecules [13]. Activation of PI3K, which has catalytic (p110) and regulatory (p85) subunits, leads to increased synthesis of PIP3, a potent signaling molecule that controls multiple downstream cascades. PTEN, a phosphatase, damps PI3K-AKT-MTOR signaling by degrading

PIP3. Many enzymes in this pathway, such as AKT, have multiple isoforms encoded by different genes with tissue-specific expression, such as AKT3 in brain. Genes and mutation status linked to different DMEG or HMEG phenotypes and syndromes are indicated in boxes, and notable mouse models outside the boxes. EGFR, epidermal growth factor receptor; FGFR, fibroblast growth factor receptor; MCAP, megalencephaly capillary malformation syndrome; MEG, megalencephaly (usually DMEG); MPPH, megalencephaly polymicrogyria polydactyly hydrocephalus syndrome; PHTS, PTEN hamartoma tumor syndrome.

Both HMEG and DMEG are associated with seizures, abnormalities in muscle tone, focal neurologic signs, and cognitive and developmental disabilities [10]. Seizures, which occur in more than 90% of HMEG cases, usually manifest first during infancy or early childhood. The seizures are typically partial, and may include infantile spasms, tonic seizures, or electroclinical features of Ohtahara syndrome. The severity of these features correlates with the extent of brain overgrowth, the severity and locations of cortical dysplasia, and the degree of abnormal activity on electroencephalography. All children suspected to have HMEG or DEMG on imaging should undergo a thorough medical assessment to check for stigmata of syndromic causes of HMEG and DMEG including tuberous sclerosis complex, Proteus syndrome, CLOVES, and other forms of somatic overgrowth or vascular malformations.

Neuroimaging In DMEG (Figure 5.2a), overall cerebral volume is increased – sometimes massively so – and the cerebral cortex shows extensive bilateral dysplasia, consisting of abnormal sulcus formation and irregularities of cortical thickness. Myelination is variably increased or decreased. Ventricular size varies among cases, from normal to markedly enlarged (hydrocephalus). In HMEG (Figure 5.2b), one hemisphere is enlarged while the other has reduced volume, so total cerebral volume is usually not increased [9]. Dysplasia is localized mainly in the larger, overgrown hemisphere. However, asymmetry of cortical growth and dysplasia is rarely, if ever, complete: part of the overgrown hemisphere may be spared, and part of the smaller hemisphere may be involved by dysplasia. A severity scale for HMEG has been proposed, based on the extent of dysplasia, ventricular size and

56

Hemimegalencephaly and Dysplastic Megalencephaly Chapter 5

Figure 5.2 Neuroimaging of dysplastic megalencephaly and hemimegalencephaly caused by germline and somatic AKT3 activating mutations, respectively. (a) Axial magnetic resonance imaging (MRI) of a boy with a germline AKT3 mutation (p.R465T) shows generalized brain overgrowth, with bilateral cortical dysplasia resembling polymicrogyria (arrows). (b) Axial MRI of a girl with a functionally severe AKT3 mutation (p.E17K) in the mosaic state, causing left HMEG (3 arrows), plus limited right posterior cortical dysplasia (single arrow).

configuration, and other macroscopic features [10]. The “occipital sign,” consisting of deviation of the larger occipital lobe across the midline, is observed in relatively severe cases of HMEG [10]. Neuroimaging features are similar in syndromic and nonsyndromic forms of HMEG [10]. The appearance of dysplastic cortex in HMEG and DMEG may resemble pachygyria or polymicrogyria, or may show only subtle changes, such as mild dyslamination and blurring of the gray–white junction [9,10]. Subcortical and periventricular heterotopia, and asymmetries of subcortical nuclei, the cerebellum, and the brainstem, may also be present.

In HMEG, overgrowth and dysplasia are distinctly asymmetric. Generally, the dysplastic hemisphere is abnormally large, but the contralateral hemisphere is small for age, leaving overall brain mass unchanged. Regions of dysplasia may cross the midline, and may affect most but not all of one hemisphere [9]. The neuroimaging “occipital sign” corresponds to overgrowth of the dysplastic occipital lobe across the midline, with lateral displacement of the contralateral occipital lobe (Figure 5.3e, arrow). Colpocephaly (posterior ventriculomegaly) is not unusual. Overall, the extent and severity of dysplasia appear to reflect the distribution and abundance of mutant cells, as well as the protein functional consequences of the underlying mutation [5].

Macroscopy Histopathology In DMEG, both hemispheres are overgrown (but generally not symmetrically), and overall brain mass is increased (Figure 5.3a–d). The gyral pattern is abnormal. Gyri appear irregular, firm, and sometimes coarse or thickened (pachygyria). Foci of polymicrogyria may be present. The gray–white interface is often blurred. White matter is usually overall increased, but may exhibit increased, normal, or decreased myelination. Subcortical structures (brainstem and cerebellum) may be enlarged. Lateral ventricles are asymmetric and, in most cases, moderately enlarged.

The histopathology of DMEG and HMEG (Figure 5.3d,g–i) is remarkably diverse among cases [1,9,11]. The degree and extent of cortical dysplasia range from mild dyslamination to severe pachygyria, sometimes mixed with areas of polymicrogyria (layered or unlayered) and leptomeningeal glioneuronal heterotopia. White matter neurons are often increased, and periventricular or subcortical heterotopia may be present. The gray and white matter exhibit moderate to severe gliosis. Aggregates of immature-appearing, neuroblast-like cells

57

Developmental Neuropathology

Figure 5.3 Macroscopic brain abnormalities and histopathology of dysplastic megalencephaly (DMEG) and hemimegalencephaly (HMEG): (a–d) DMEG in a child born at 37 gestational weeks who lived for 11 weeks. (a,b) Both hemispheres were large (brain weight 905 g, expected 506 ± 67 g) and exhibited abnormal gyral patterns. (c) In slices, the cortical gray matter appeared thick and dysplastic (pachygyria-like), and cerebral white matter was prematurely myelinated. (d) Histologic section through lateral temporal cortex, stained by immunohistochemistry for microtubule-associated protein 2 (MAP2), revealed

58

thickening and dyslamination of the cortical ribbon, and numerous small periventricular heterotopia (arrows). (e–i) HMEG in a premature infant born at 34 gestational weeks who lived for 11 days: (e) The left hemisphere was larger, most notably in the occipital region (arrow). (f) The cortex was variably thickened, with abnormal configuration of sulci and gyri. (g) The left cerebellar hemisphere was larger. (h) Dysplasia involved the left hippocampus. (i) Leptomeningeal glioneuronal heterotopia. Scale bar: (d), 2 mm (adapted from Hevner [11]). Reproduced with permission of Elsevier.

Hemimegalencephaly and Dysplastic Megalencephaly Chapter 5

(hamartias or microdysgenesis) are seen in many cases of HMEG [9]. Dystrophic calcifications are often seen. The degrees of cytologic hypertrophy and atypia are quite variable among cases. In some cases, neurons and astrocytes are only slightly hypertrophic. But in others, neurons and astrocytes display disorganized cytoplasm, nuclear pleomorphism, and cytoskeletal abnormalities. “Balloon cells” resembling those in focal cortical dysplasia type IIb or tuberous sclerosis complex are found in a minority of cases. Bi- or multinucleated neurons may be present. Oligodendrocytes appear cytologically unremarkable. Foci of tissue necrosis or cell loss are seen in some cases, presumably a consequence of tissue overgrowth leading to compromised vascular perfusion. Proliferative activity is not elevated, except due to astrocytic reaction or microglial inflammation associated with seizure activity. Histologically, areas of cortex in HMEG and DMEG may appear indistinguishable from tuberous sclerosis complex or focal cortical dysplasia type I, or II.

Immunohistochemistry Gliosis and dyslamination are demonstrated by standard neuropathological markers such as GFAP and NeuN. Cells may show abnormal accumulation of microtubule-associated proteins, neurofilament proteins, nestin, and ubiquitin. Some hypertrophic cells may express both neuronal and glial markers, signifying ambiguous or immature differentiation [1]. Cytoskeletal proteins, such as neurofilament proteins and microtubule-associated proteins, may accumulate excessively in neuronal cell bodies and dendrites, forming cytomegalic neurons. Balloon cells, when present, may be weakly immunoreactive for glial markers, neuronal markers, or both. Immunohistochemical markers of active PI3K-AKT signaling, such as phosphorylated S6 ribosomal protein, are elevated in subsets of neurons and astrocytes in at least some cases [12].

and associated developmental anomalies. To date, the most extensively studied model of megalencephaly due to overactive PI3K–Akt–mTOR signaling is the Pten conditional null mouse [14]. Mice lacking Pten function in brain develop megalencephaly and abnormal cytoarchitecture, along with enlarged cell size, reduced cell death and increased astrocyte proliferation and hypertrophy [15]. Mice with an activating mutation in the kinase domain of Akt3 show enlarged brain size, increased seizure susceptibility, and abnormal hippocampal development [16]. Mice with mutations in Tsc1 or Tsc2 (models of tuberous sclerosis complex, in which MTOR signaling is increased) display enlarged dysplastic neurons and predisposition to epilepsy [17,18]. Recently, mice with Pik3ca gain-of-function mutations were shown to recapitulate many features of DMEG, including ventriculomegaly, cortical dysplasia, and epilepsy [19]. Apart from core components of the PI3K pathway, mutations in some upstream molecules can also lead to megalencephaly. For example, IGF-1 overexpressor mice display brain overgrowth with increased cell number [20,21].

Models of HMEG or focal cortical dysplasia As a model of focal cortical dysplasia type II, in utero electroporation of mutant Mtor in embryonic mouse brain was sufficient to disrupt neuronal migration and cause spontaneous seizures and cytomegalic neurons [22]. Similar observations were obtained from electroporation models involving activating mutation of Akt3 [23], or deletion of Tsc2 [24]. Both the latter models provided evidence that Reelin signaling downstream of MTOR may be instrumental in the generation of cytoarchitectural defects.

Future investigations and therapies

Experimental models

Current therapies Seizures are the most critical cause of morbidity and mortality associated with HMEG and DMEG. The first line of treatment typically consists of antiepileptic drugs, but seizures in HMEG, DMEG, and focal cortical dysplasia are often intractable to such medical treatments. A ketogenic diet may be effective in some cases of focal cortical dysplasia, and there is a strong biochemical rationale for this approach [25]. In focal cortical dysplasia and HMEG, surgical resection of the dysplastic cortex (partial or total lobectomy or hemispherectomy) can reduce seizures, when medical therapy fails. In small children, plasticity of the immature brain permits relatively favorable neurological outcomes after large resections. However, postsurgical recurrence of epilepsy is sometimes a problem, as epileptic foci may arise in unresected cortex. In these cases, reresection of additional cortex is sometimes effective.

Models of DMEG Several mouse models have been developed with mutations in the PI3K–Akt–mTOR pathway, which display megalencephaly

Future investigation and potential therapies Intractable epilepsy remains the most important treatment challenge in DMEG and HMEG. Accordingly, much current

Pathogenesis In the past five years, genetic analyses have highlighted the role of genetic mutations leading to overactive PI3K–AKT–MTOR signaling as prominent, and possibly the major causes of HMEG and DMEG (Figure 5.1). Briefly, this complex pathway transmits signals from activated growth factor receptors to promote cell and tissue growth [13]. Mutations involving this pathway are important not only in development, but also in some cancers, such as glioblastoma.

59

Developmental Neuropathology research is directed to testing new drugs or other approaches that reduce seizures, and to improving our understanding of disease pathogenesis so that other novel approaches may be found. Additional genetic studies in humans, and improved mouse models will be important to reveal if potential new genes and pathways are involved in DMEG and HMEG, and to determine how different mutations cause specific abnormalities. Since many of the mutations in humans are mosaic, future investigations might include regional and single-cell analysis of the genome to understand the molecular genetic basis of asymmetrical and focal brain overgrowth. One interesting question concerns the relation between mutant cell abundance and dysplasia. Current evidence suggests that the abundance of mutant cells may be relatively low (0.1–9%), even in areas of severe dysplasia [5]. These observations suggest the possibility that mutant cells may induce the overgrowth of normal cells by a non-autonomous effect. A ketogenic diet may be a potential “new” treatment approach for DMEG and HMEG. Interestingly, a ketogenic diet inhibits PI3K–AKT–MTOR signaling [26,27], providing a rationale for its use to ameliorate epilepsy caused by overactivation of this pathway in HMEG and DMEG. Indeed, a ketogenic diet is more effective than drugs in focal cortical dysplasia [25,27]. Many drugs that inhibit the PI3K–AKT pathway are already under development for cancer, and in principle might be useful for seizure reduction as well. Rapamycin analogs (inhibitors of MTOR), as well as new PI3K inhibitors (such as BKM120, a panClass I PI3K inhibitor; and MK2206 and ARQ092, both pan-Akt inhibitors) appear to be promising candidates [19,22].

6.

7.

8.

9.

10. 11.

12.

13. 14.

15.

16.

17.

References 18. 1. Flores-Sarnat L, Sarnat HB, D´avila-Guti´errez G, Alvarez A (2003) Hemimegalencephaly: part 2. Neuropathology suggests a disorder of cellular lineage. J Child Neurol 18:776–85. 2. Rivi`ere J-B, Mirzaa GM, O’Roak BJ, et al. (2012) De novo germline and postzygotic mutations in AKT3, PIK3R2 and PIK3CA cause a spectrum of related megalencephaly syndromes. Nat Genet 44: 934–40. 3. Mirzaa GM, Conway RL, Gripp KW, et al. (2012) Megalencephaly– capillary malformation (MCAP) and megalencephaly– polydactyly–polymicrogyria–hydrocephalus (MPPH) syndromes: two closely related disorders of brain overgrowth and abnormal brain and body morphogenesis. Am J Med Genet A 158A: 269–91. 4. Mirzaa GM, Parry DA, Fry AE, et al. (2014) De novo CCND2 mutations leading to stabilization of cyclin D2 cause megalencephalypolymicrogyria-polydactyly-hydrocephalus syndrome. Nat Genet 46:510–15. 5. Mirzaa GM, Campbell CD, Solovieff N, et al. (2016) Association of MTOR mutations with developmental brain disorders,

60

19.

20.

21.

22.

23.

including megalencephaly, focal cortical dysplasia, and pigmentary mosaicism. JAMA Neurol 2016 73(7):836–45. Mirzaa GM, Poduri A. (2014) Megalencephaly and hemimegalencephaly: breakthroughs in molecular etiology. Am J Medical Genetics Part C 166:156–72. Gai D, Haan E, Scholar M, et al. (2015) Phenotypes of AKT3 deletion: a case report and literature review. Am J Med Genet A 167A:174–9. Poduri A, Evrony GD, Cai X, Walsh CA. (2013) Somatic mutation, genomic variation, and neurological disease. Science 341(6141):1237758. Salamon N, Andres M, Chute DJ, et al. (2006) Contralateral hemimicrencephaly and clinical-pathological correlations in children with hemimegalencephaly. Brain 129:352–65. Flores-Sarnat L. (2002) Hemimegalencephaly: part 1. Genetic, clinical, and imaging aspects. J Child Neurol 17:373–84. Hevner RF. (2015) Brain overgrowth in disorders of RTK-PI3KAKT signaling: a mosaic of malformations. Semin Perinatol 39:36– 43. Jansen LA, Mirzaa GM, Ishak GE, et al. (2015) PI3K/AKT pathway mutations cause a spectrum of brain malformations from megalencephaly to focal cortical dysplasia. Brain 138:1613–28. Chalhoub N, Baker SJ. (2009) PTEN and the PI3-kinase pathway in cancer. Annu Rev Pathol 4:127–50. Kwon CH, Zhu X, Zhang J, et al. (2001) Pten regulates neuronal soma size: a mouse model of Lhermitte-Duclos disease. Nat Genet 29:404–11. Fraser MM, Zhu X, Kwon CH, et al. (2004) Pten loss causes hypertrophy and increased proliferation of astrocytes in vivo. Cancer Res 64:7773–9. Tokuda S, Mahaffey CL, Monks B, et al. (2011) A novel Akt3 mutation associated with enhanced kinase activity and seizure susceptibility in mice. Hum Mol Genet 20:988–99. Meikle L, Talos DM, Onda H, et al. (2007) A mouse model of tuberous sclerosis: neuronal loss of Tsc1 causes dysplastic and ectopic neurons, reduced myelination, seizure activity, and limited survival. J Neurosci 27:5546–58. Choi YJ, Di Nardo A, Kramvis I, et al. (2008) Tuberous sclerosis complex proteins control axon formation. Genes Dev 22:2485–95. Roy A, Skibo J, Kalume F, et al. (2015) Mouse models of human PIK3CA-related brain overgrowth have acutely treatable epilepsy. Elife 4 pii: e12703. Popken GJ, Hodge RD, Ye P, et al. (2004) In vivo effects of insulin-like growth factor-I (IGF-I) on prenatal and early postnatal development of the central nervous system. Eur J Neurosci 19: 2056–68. Hodge RD, D’Ercole AJ, O’Kusky JR. (2005) Increased expression of insulin-like growth factor-I (IGF-I) during embryonic development produces neocortical overgrowth with differentially greater effects on specific cytoarchitectonic areas and cortical layers. Brain Res Dev Brain Res 154:227–37. Lim JS, Kim WI, Kang HC, et al. (2015) Brain somatic mutations in MTOR cause focal cortical dysplasia type II leading to intractable epilepsy. Nat Med 21:395–400. Baek ST, Copeland B, Yun EJ, et al. (2015). An AKT3-FOXG1-reelin network underlies defective migration in human focal malformations of cortical development. Nat Med 21:1445–54.

Hemimegalencephaly and Dysplastic Megalencephaly Chapter 5 24. Moon UY, Park JY, Park R, et al. (2015) Impaired Reelin-Dab1 signaling contributes to neuronal migration deficits of tuberous sclerosis complex. Cell Rep 12:965–78. 25. Kang JW, Rhie SK, Yu R, et al. (2013) Seizure outcome of infantile spasms with focal cortical dysplasia. Brain Dev 35:816– 20.

26. Jung DE, Kang HC, Kim HD. (2008) Long-term outcome of the ketogenic diet for intractable childhood epilepsy with focal malformation of cortical development. Pediatrics 122:e330–3. 27. McDaniel SS, Rensing NR, Thio LL, et al. (2011) The ketogenic diet inhibits the mammalian target of rapamycin (mTOR) pathway. Epilepsia 52:e7–11.

61

6

Lissencephaly, Type I Jeffrey A. Golden Department of Pathology, Brigham and Women’s Hospital, Harvard Medical School, Boston, MA, USA

Definition Lissencephaly, derived from the Greek words lissos meaning “smooth” and enkephalos meaning “brain”, is a descriptive term defining a class of human cerebral malformations characterized by an agyric surface of the brain. In fact, most cases of lissencephaly do not have a complete loss of gyri; often, the most ventral and medial gyri are relatively spared, and gyri are present in the more anterior or posterior brain regions, depending on the specific genetic mutation as described below. Lissencephaly is not a single malformation, but a descriptive term applied to many malformations with distinct genetic etiologies. This chapter focuses on the lissencephalies associated with a cell migration defect, and Chapter 7 examines the lissencephalies resulting from an apparent over-migration of neurons. Heterotopia, also a migration disorder, are addressed in Chapter 9; however, it is important to note that some heterotopia malformations are genetically linked to the lissencephalies (e.g. many cases of subcortical band heterotopia). The lissencephalies described in this chapter have also been referred to as type I, or classical, lissencephaly. As of today, only the type I lissencephalies, together with heterotopia (Chapter 9), have been experimentally confirmed as a neuronal migration disorder. Other disorders, such as polymicrogyria, cortical dysplasias, and type II lissencephaly, have been ascribed to a defect in cell migration, although current data are not supportive. Finally, the pathogenesis of these disorders is the result of a slowing or delay in cell migration rather than the absence of movement.

Synonyms and historical annotations Lissencephaly, agyria and pachygyria have, at various times, been used to describe an overlapping spectrum of disorders.

Pachygyria has generally been reserved for cases with multifocal or focal malformations, whereas agyria and lissencephaly are used to describe a diffuse malformation of the cerebral hemispheres. Studies indicate that the same genetic mutation can give a spectrum of disorders ranging from localized lissencephaly or pachygyria to diffuse lissencephaly. An understanding of how the same mutation can give rise to distinct morphological anomalies is only beginning to be understood. However, it is also important to recognize that this may not be the only explanation for pathological heterogeneity.

Epidemiology Incidence and prevalence The incidence of type I lissencephaly is difficult to estimate, owing to the paucity of reports. The best data, coming from the Netherlands, found 22 cases in 11.7 million births, giving an incidence of approximately 1 : 500 000 [1]. As a result of diagnostic difficulties and subtleties in the classification, this incidence is believed to be an underestimate by as much as 10 times. Sex and age distribution Type I lissencephaly is often cited to affect males and females equally; the presence of at least two X-linked forms suggests that males should have a higher incidence. The failure of epidemiologic studies to recognize this is likely a reflection of the overall low incidence and small numbers evaluated. Risk factors There are no known risk factors for type I lissencephaly. All cases where the etiology is defined are clearly genetic in origin. Whether the closely related pachygyria has environmental, toxic, infectious or other associated risk factors remains to be established (see below).

Developmental Neuropathology, Second Edition. Edited by Homa Adle-Biassette, Brian N. Harding, and Jeffrey A. Golden. © 2018 John Wiley & Sons Ltd. Published 2018 by John Wiley & Sons Ltd.

63

Developmental Neuropathology

Embryology A hallmark of nervous system development is that the site to which neural progenitor cells proliferate to generate neurons and glia, differs significantly from the location of the differentiated cells in the mature brain. Experimental studies over the first two decades of the twenty-first century have significantly expanded our understanding of neuronal migration. We now know that, essentially, all the projection neurons (the excitatory neurons) in the cerebral cortex, representing approximately 80% of neurons, are derived from the cortical ventricular zone, where cells migrate radially out to the definitive cortex. This radial pathway of cell migration was first postulated by Ramon y Cajal in the 1890s and experimentally delineated in the 1960s and 1970s [2–4]. More recent work shows that the radial glia are in fact the progenitor cells and that there are multiple types of radial and nonradial progenitor cells in the pallium that together contribute to the number and heterogeneity of projection neurons

(Figure 6.1) [5]. Proliferating cells in the ventricular zone give rise to one of a group of different cell types: 1. A new ventricular zone progenitor cell (radial glial). 2. A new progenitor cell (known as an outer radial glial cell), which resides in the subventricular zone and can be distinguished from a ventricular zone radial glial cell by virtue of its cell body being in the subventricular zone and having its short apically oriented process not extending to the ventricular surface. 3. An intermediate progenitor cell, which divides only once or a few times, giving rise mainly to neurons. 4. A differentiated cell that migrates radially to the definitive cortex. Upon exit from the cell cycle, newly born neuroblasts associate with adjacent radial glial fibers and use them as guides to the cortical plate [6]. Bidirectional signaling occurs between the migrating neuron and the radial glial cell that permits the neuroblast to migrate and provides a signal to maintain the structure of the radial glial fiber scaffold [2]. This process requires

I II III IV V VI

S WM G2

G1 M

E 8.5

E 10

Figure 6.1 Schematic outline of early mouse cortical development. At E8.5, the wall of the neural tube consists of a pseudostratified neuropepithelium (band of blue cells). Cell processes maintain contact with both the inner (bottom of figure) and outer (top of figure) surfaces of the neural tube. However, the nucleus migrates according to the cell cycle with mitoses (M phase) occurring at the ventricular surface and S-phase at the pial surface. At E10, the first cells delaminate from the germinal neuroepithelium, which defines the ventricular zone and form the preplate. The preplate is composed of Cajal–Retzius neurons (light green cells) and

64

E 12-17

Adult

subplate neuros (yellow cells). During the ensuing days, those ventricular zone cells which exit the cell cycle become neuroblasts (dark blue cells) and migrate from the ventricular zone along radial glia (red cells in the ventricular zone) to populate the definitive cortical plate. The neuroblasts that become layers II–VI split the preplate, leaving the subplate neurons adjacent to the ventricular zone while the Cajal-Retzius cells remain in contact with the pial surface. Cells accumulate from inside-to-out beginning with layer VI (purple cells) and ending with layer II neurons (pink cells).

Lissencephaly, Type I Chapter 6

many known receptors and ligands, such as neuregulin and ErbB4 [7,8], cell adhesion molecules [9,10], putative ligands with unknown receptors, such as astrotactin [11–13], extracellular matrix molecules [14], and their cell surface receptors [15]. Blocking any of these components can slow down or prevent radial cell migration and thus are candidate pathways for the development of cell migration disorders. A second site of cerebral cortical neurogenesis is in the ventral ganglionic eminences (medial and caudal). Progenitors in these brain regions give rise to the inhibitory neurons (interneurons) of the cerebral cortex, which make up the remaining approximately 20% of the cortical neurons. These neurons must take a long, circuitous path to the cortex that is similarly under the control of numerous signaling pathways including semaphorin [16,17], neuregulin [18], and Cxcl-12 [19]. While interneuron deficits have been identified in patients with lissencephaly, it remains unclear what their contributions to the structural defects are, though they likely contribute to the neurologic dysfunction observed in these patients [20]. Cells that migrate into the cerebral cortex organize into six layers or laminae. These six layers require an embryonic structure, known as the preplate, to form prior to development of the definitive layers. The preplate is formed by two cell types: an outer layer composed largely of Cajal–Retzius neurons and an inner layer of subplate neurons (Figure 6.1). The Cajal–Retzius neurons are derived from three unique sites: adjacent to the cortical hem, at the pallial–subpallial boundary, and in the preoptic/ septal region. From these areas, they migrate over the surface of the brain [21]. The subplate neurons are derived from the ventricular zone and are the first-born neurons, essentially they do not migrate but simply delaminate. The Cajal–Retzius neurons will remain superficial in the mature cortex in layer 1 along the surface of the brain, while the subplate neurons will end up deep to the definitive cortex at the border between layer 6 and the white matter. Both of these populations largely or completely

die off in the adult (mature) brain. Neurons that ultimately populate the definitive cortical layers, layers 2 through 6, must migrate out as described above, split the preplate, where they generate the cortex in an inside-to-outside sequence [22]. The first cells to arrive and split the preplate will eventually reside in the deepest layer, layer 6. Later-born cells will migrate past the existing cells to reside in progressively more superficial layers (Figure 6.1).

Genetics Type I lissencephaly is well recognized as a genetic disorder. It has known sporadic, autosomal dominant, autosomal recessive, and X-linked inheritance patterns. To date, 13 genes have been causally linked (Table 6.1). The first to be identified was LIS1 on chromosome 17p. LIS1 is deleted in all patients with the Miller– Dieker syndrome (see signs and symptoms below for more on the syndrome) and many patients with isolated lissencephaly sequence. The more severely affected patients are likely a reflection of larger deletions that include other genes and most relevantly 14-3-3𝜀, located telomeric to LIS1 [23–25]. The LIS1 transcript encodes a 410 amino acid protein known as LIS1, or PAFAH1B1 (see below). The LIS interacting protein encoded by NDE1 was identified in 2011 as another lissencephaly gene; these patients also have severe microcephaly [26]. Both NDE1 and LIS1 interact with DYNC1H1, mutations which have been associated with the lissencephaly-related disorder of bilateral frontal pachygyria/agyria [27]. The second lissencephaly gene identified was XLIS, which encodes doublecortin (DCX) [28,29]. DCX is a microtubule associated protein, but its exact function remains uncertain [30]. XLIS maps to Xq22.3-23 and mutations result in lissencephaly in males. Females heterozygous for XLIS show subcortical band heterotopia [31], the presence of the normal overlying cortex and the subcortical band of cortical neurons presumably reflecting

Table 6.1 Genes associated with lissencephaly and related disorders. Gene

Inheritance

Lissencephaly Syndrome

ACTB ACTG1 ARX CDK5 DCX DYNC1HI KIF2A NDE1 PAFAH1B1/LIS1 RELN TUBAIA TUBG1 VLDLR

Autosomal dominant/de novo Autosomal dominant/de novo X-linked Autosomal recessive X-linked Autosomal dominant/de novo Autosomal dominant/de novo Autosomal recessive Autosomal dominant/de novo Autosomal recessive Autosomal dominant/de novo Autosomal dominant/de novo Autosomal recessive

Baraitser–Winter Baraitser–Winter X-linked lissencephaly with ambiguous genitalia with cerebellar hypoplasia X-linked Frontal lobe predominant Posterior predominant Severe microcephaly Isolated and Miller–Dieker With cerebellar hypoplasia With multiple brain anomalies Posterior predominant With cerebellar hypoplasia

Reference 33 33 39 38 29 27 32 26 89 34 90 32 37

65

Developmental Neuropathology two populations of migratory neurons that are distinguished by normal Lyonization (X-chromosome inactivation). In those cells where the X-chromosome encoding the mutant allele is inactivated, only the normal allele is expressed and these cells migrate normally to form the cortex. Cells inactivating the normal allele will only express the mutant allele and will not migrate normally, ultimately stopping and residing in small to medium-sized heterotopia in the subcortical white matter that together represent the subcortical band heterotopia. In addition to the genes described above, which are all involved in molecular motors and the movement of cargo, including the nucleus, along microtubules, a number of cytoskeletal components have themselves been identified as lissencephaly genes. They include two tubulin genes (TUBA1A and TUBG1) and two actin genes (ACTB and ACTG1). The brains of patients with TUBA1A mutations show multiple anomalies, including abnormal white matter bundles, and defects of the cortex, hippocampus and cerebellum [32]. In contrast, mutations in TUBG1 are associated with a posterior predominant, bilateral pachygyria/agyria [32]. Mutations in either of the two actin genes are associated with the lissencephaly syndrome known as Baraitser–Winter syndrome (33). The third gene to be identified was RELN [34]. RELN is required for normal migration into appropriate layers. Reeler mice show an inverted cortex and severe migrational anomalies of the cerebellum [35,36]. An autosomal recessive form of lissencephaly with cerebellar hypoplasia characterizes patients with RELN mutations. Lymphedema and neuromuscular problems are seen in some patients [34]. More recently, mutations in the RELN pathway genes CDK5 and VLDLR (a RELN receptor) have also been associated with lissencephaly with cerebellar hypoplasia (37, 38). ARX represents the only transcription factor associated with lissencephaly (39). Mutations in this gene, which is also located on the X-chromosome, result in lissencephaly and ambiguous genitalia in males. The spectrum of disorders in females is more heterogeneous but does not appear to include lissencephaly.

Clinical features Signs and symptoms The clinical presentation of all type I lissencephaly syndromes show considerable overlap. Most patients exhibit moderate to severe developmental delay and neuromotor impairment. Patients with isolated type I lissencephaly most commonly have profound intellectual disabilities and seizures that are often intractable. Infantile spasms may be present, although they are most common with mutations in ARX [40,41]. Patients with isolated lissencephaly may be less severely affected than those with the Miller–Dieker syndrome. This is likely due to deletions involving 14-3-3𝜀 in addition to LIS1 [25]. Patients with the Baraitser–Winter syndrome also exhibit relatively specific facial features, including hypertelorism, ptosis, high-arched eyebrows,

66

a broad nasal bridge and tip of the nose, a long philtrum, full cheeks, and a pointed chin. They may also have anomalies outside the brain and face, including limb and heart defects. Patients with type I lissencephaly may also exhibit extra central nervous system abnormalities. Patients with isolated lissencephaly typically have acquired microcephaly and mild dysmorphic facial features. In contrast, patients with Miller– Dieker syndrome have bi-temporal hollowing, hypertelorism, frontal bossing, a short-upturned nose, a small jaw and a prominent upper lip with a thin vermillion border. Other anomalies occasionally associated with the syndrome include clinodactyly, polydactyly, cryptorchidism, sacral dimples, and congenital heart defects. Patients with ARX mutations exhibit microphalus (and thus ambiguous genitalia) and small adrenal glands, giving rise to the X-linked lissencephaly with ambiguous genitalia syndrome. RELN mutations result in congenital lymphedema in a subset of patients.

Imaging Magnetic resonance imaging (MRI) is the modality of choice for evaluating the lissencephalic brain for structural anomalies. The loss of sulci and gyri over the surface of the brain combined with the thickened cerebral cortex (1–2 cm in contrast to the normal 0.3–0.5 cm) is characteristic of type I lissencephaly. In addition, heterotopia, subcortical bands, hydrocephalus, and posterior fossa anomalies can be evaluated and may be helpful in guiding genetic testing. Furthermore, the pattern of the structural anomalies of the brain may assist in guiding genetic testing. A loss of posterior gyri (occipital) with relative sparing anteriorly (frontal lobes; a posterior to anterior gradient) favors a LIS1 mutation, whereas an anterior to posterior gradient (more severe in the frontal lobes) suggests a mutation in DCX [42]. An anterior to posterior gradient with severe cerebellar hypoplasia would favor a RELN, CDK5 or VLDLR mutation. As noted above (see also Table 9.1), mutations in several genes may give a more focal pachygyria/agyria that can be anterior (DYNC1H1) or posterior (KIF2A and TUBG1), or can be multifocal in a spectrum with lissencephaly in the case of the Baraister–Winter syndrome. Grading of the lissencephaly is generally based on MRI findings [43]. Laboratory findings There are no specific laboratory findings associated with type I lissencephaly other then genetic analysis.

Macroscopy The external surface of the brain from patients with lissencephaly reveals a marked paucity of gyri and sulci (Figure 6.2). The abnormality need not include the entire brain, and, as described above, relative sparing of the frontal or occipital gyri and sulci usually reflects mutations in distinct genes. Gross sectioning of the brain reveals a markedly thickened

Lissencephaly, Type I Chapter 6

Figure 6.2 External view of brain with lissencephaly. Note the complete absence of gyri and sulci at the surface.

cerebral cortex and significant diminution of the underlying white matter (Figure 6.3). A transition to a more normal thickness of the cortex is seen in areas that are grossly spared. Transition areas can include areas that have a subcortical band heterotopia phenotype. Periventricular heterotopia and white matter heterotopia may also be observed. In older individuals, a white matter band can be seen in the superficial cortex corresponding to layer 3. The ventricles are usually generous in size. The cerebellum may be hypoplastic in some cases, usually corresponding to a specific mutation. Pachygyria (also referred to as regional agyria) is a localized thickening of the cortex, which, in most other ways, resembles lissencephaly. It can involve a single gyrus or a greater region of one or both hemispheres. When extensive and bilateral, it can be difficult to distinguish from lissencephaly, although multifocal pachygyria is usually not bilaterally symmetric.

Histopathology The cerebral cortex of type I lissencephaly is classically described as a four-layered cortex replacing the normal six-layered ribbon (Figure 6.4). The outermost layer is the molecular layer, which contains Cajal–Retzius neurons in most cases (layer 1). Layer 2 is a band of primarily medium to large pyramidal neurons, which may be correctly oriented but which more often show variable degrees of disorganization. The third layer has a decreased density of neurons and contains numerous axons, which in children older than one to two years are myelinated. This layer is also visible by neuroimaging. Finally, layer 4 is composed of a broad band of disorganized neurons [44]. Layer 4 can be of variable thickness, but always contains a mixed population of small and medium sized neurons that have no clear organization or

Figure 6.3 Hemisection of brain from patient with lissencephaly. The cerebral cortical gray matter is markedly thickened and the white matter is diminished. Myelination in layer III can be seen (arrow). There is mild sparing of the temporal lobe where some evidence of gyral formation is present and the hippocampus is preserved.

orientation. With minor variations, the cortex is histologically as described in lissencephaly due to mutations in LIS1 and DCX, although a clear distinction is seen in the white matter. While the white matter, which is severely reduced in volume, occasionally contains individual neurons but is relatively well distinguished from the cortex in brains with a LIS1 mutation, with DCX mutations there are collections of neurons forming heterotopia blurring the gray–white border and extending deep into the white matter [45]. This description corresponds to the areas of classic lissencephaly. Sampling other cortical regions gives a variety of pathological features ranging from a normal cortex to the classical four-layer cortex. Other genetically-defined lissencephalies have distinct histopathologic features including a three-layered cortex in patients with ARX mutations. The threelayer lissencephaly has a cellular molecular layer, a pyramidal cell layer two, similar to the four-layer lissencephaly, and then a thick layer composed of neurons of all types including many pyramidal type neurons [45]. Finally, a number of cases examined with

67

Developmental Neuropathology

Normal Cortex I

Lissencephaly

I

II II

III III

IV

IV

V

VI

Figure 6.4 Neurofilament stained section from patient with lissencephaly and schematic of normal cortical laminar organization and that found in lissencephaly. Neurofilament staining highlights the few cells in layer I, the numerous large and disorganized pyramidal neurons in layer II, the relatively hypocellular layer III which becomes myelinated (see Figure 6.3.), and the thick layer IV composed of medium and small neurons. The schematic highlights the normal organization of the cortex in contrast to that found in classical lisssencephaly. Note the apparent reversal of layers with the exception of Cajal–Retzius neurons in layer I (green).

cerebellar hypoplasia, but unfortunately none of the genetically defined examples, have shown two-layer lissencephaly. An outer molecular layer and a very broad layer 2 with almost no white matter. Layer 2 has a mix of neuronal types, together with pyramidal neurons throughout the depth of this layer [45]. In addition to the cortical changes described here, simplification of the hippocampus, inferior olivary nucleus dysplasia, cerebellar cortical dysplasias, and corticospinal tract anomalies may also be present [44,46]. Pachygyria histologically resembles lissencephaly but is more localized, involving only a single gyrus up to an entire lobe. It can also be multifocal. Pachygyria characteristically shows a loss of the gray–white junction, together with the thickening of the cortex, and the border with normal cortex is difficult to identify. This is in distinction to polymicrogyria (Chapter 11) where the gray–white junction and the border with adjacent normal cortex are usually well-defined.

bodies can be used to localize Cajal–Retzius neurons. Other studies are generally noncontributory, including ultrastructural studies.

Differential diagnosis The differential diagnosis of lissencephaly is limited. When bilaterally symmetric and involving the entire brain, the primary consideration is differentiating the genetic subtype. Features such as the pattern of the lissencephaly by imaging or gross pathology, sex, and coexisting malformations, can all help to guide molecular genetic investigations. In less severe cases, the differential diagnosis may include pachygyria, and this can be difficult to distinguish from polymicrogyria (Chapter 11) and cortical dysplasia (Chapter 13). Although radiographically distinguishable by MRI in some cases, this is not always true, and only pathology can definitively separate out these entities in many cases.

Immunohistochemistry and ultrastructural findings

Experimental models and pathogenesis

Immunohistochemistry with neurofilament antibodies is often helpful in defining the cortical architecture. Anti-RELN anti-

The genes mutated in several human disorders of neuronal migration also provided a basis for linking the cytoskeleton

68

Lissencephaly, Type I Chapter 6

to neuronal migration. The human LIS1 gene was positionally cloned from patients with Miller–Dieker syndrome and isolated lissencephaly sequence, both associated with type I lissencephaly (classical lissencephaly). It was found to encode the 45-kD enzymatic β-subunit of platelet-activating factor acetylhydrolase (PAFAH), PAFAH1B1 [34,47,48]. PAFAH1B1 homozygous mutant mice die at the trophoblast stage [49]. Heterozygote mice are grossly normal, but show a dose-dependent histopathological disorganization of cerebral cortical lamination, cerebellar cortical defects, and disruptions of the pyramidal cell layer in the hippocampus [49]. In vitro analysis of heterozygous cerebellar granule neurons show slowed migration but no alteration of neurite dynamics [49,50]. This reduced rate of migration has been hypothesized to be the basis for the observed pathologic phenotype. The available data do not distinguish between a direct role for PAFAH in transducing guidance signals to the cytoskeletal machinery compared with a loss of component of the migration apparatus. However, the yeast homolog of the human LIS1 gene, NudF [51] has provided some insight into the possible function of LIS1. Proper nuclear migration in the Aspergillus nidulans depends on NudF [51]. NudF associates with NudC in yeast [52]. NudC is a tyrosine kinase, and its binding partner, NudA, is a cytoplasmic dynein [51]. The human homolog of NudC is highly conserved in higher vertebrates, and, in conjunction with the LIS1 gene product, regulates dynein activity during mitosis in vertebrate cells [53,54]. These data suggest that type I lissencephaly, due to LIS1 mutations, is caused by a loss of dynein function, which is required for nuclear translocation. The inhibition or slowing of nuclear movement during cell migration may, in turn, retard the rate of cell migration. Further evidence to support this hypothesis came from four laboratories reporting that LIS1 also interacts with NUDEL and mNudE [55–58]. This interaction again parallels that found in A. nidulans, where NudF interacts with NudE, the homolog of mNudE and NUDEL. Through a series of experiments, they have shown that NUDEL, and possibly mNudE, directly interact with LIS1, and that together they regulate cytoplasmic dynein motor function and location within the cell [59]. Disruption in LIS1 results in abnormal localization of NUDEL and likely a defect in cytoplasmic dynein motor function, leading to a defect in cell migration or at least nuclear migration. It is interesting to note that one feature that separates axon guidance, another form of migration found in the developing nervous system, from cell migration, is nuclear translocation. LIS1 mutations may result in specifically a nuclear movement defect and therefore a preferential defect in cell migration rather than an axon guidance defect. Finally, the LIS1/NUDEL pathway has also been linked to the Reelin/Cdk5/p35 pathway described above. NUDEL is a direct target of Cdk5 [57,58]. The phosphorylation of NUDEL by Cdk5 appears to control its cellular localization and thus is likely to influence dynein motor function (Figure 6.5).

In addition to its role in cell migration, cell proliferation may also be influenced by LIS1. A series of experiments again linked interactions of LIS1 with cytoplasmic dynein and dynactin [60]. Both of these proteins are important in mitotic cell division and cytokinesis. Although a role for NUDEL or mNudE has not been directly investigated in cell division, given the data reviewed above, it is interesting to speculate on such an association. Further evidence that LIS1 may have a role in cell proliferation comes from observations in mutant mice and in humans with lissencephaly. Mice engineered to have incremental decreases in LIS1 show a progressive thinning of the cortex suggesting a cell proliferation defect [49]. The brains of patients with lissencephaly are usually small, again implying a possible proliferation defect [46,61]. Together, these data imply LIS1 may have a role in cell proliferation and in cell migration, both contributing to the human phenotype of lissencephaly. Mutations in the DCX gene located on the X-chromosome are associated with subcortical band heterotopia, also known as “double cortex,” in females [29,62]. Males with DCX mutations usually have classical lissencephaly [31,63–65], but have also presented with subcortical band heterotopia. Both the type of mutation and somatic mosaicism have been associated with this milder phenotype in males [31,66]. In pedigrees where females have subcortical band heterotopia and males lissencephaly, a DCX mutation has so far always been identified. In females with sporadic subcortical band heterotopia, the figures vary [28,67, 68], but in those with bilateral diffuse bands, or thin frontal bands only, the mutation rate appears to be high. This again emphasizes the severity gradient of the cortical malformation in relation to the gene mutation, with a frontal, or more severe frontal, malformation of the brain associated with DCX mutations. Random versus skewed X-inactivation was thought to determine the thickness and extent of the band. However, studies on lymphocytes did not confirm this hypothesis, although they may not reflect the inactivation ratio in the brain. It has become clear that the type of mutation also determines the phenotypic spectrum in females [67,68]. Mutations in either LIS1 or DCX both result in disruptions of cell migration. In the case of LIS1, cell migration appears to be slowed. Cells that must travel the furthest would be predicted to be the most retarded. This appears to be the case; the superficial layer of the cortex, layer 2, is predominately populated by large pyramidal cells that usually reside in the deeper layers of the cortex. These are among the first cells to migrate from the ventricular zone and because the intermediate zone is relatively thin earlier in development, they have the shortest distance to travel. In contrast, the later-born cells normally destined for the more superficial layers must travel the farthest and are the cells found closer to the ventricle in classical lissencephaly, implying they are affected for a longer period of time or more severely affected. Dcx was disrupted in mice using targeted mutagenesis, but these mice did not have a cortical defect, although they did have a hippocampal cell positioning phenotype [69]. In contrast, using a method to inhibit Dcx RNA, both rats and mice exhibited

69

Developmental Neuropathology

CNR

Reelin

microtubules

mDab1

γ–tubulin

mNudE

VLDLR/ ApoER2 α3β1 Integrin

?

c-Abl

Lis1 14-3-3e

Dcx

Dynein

Nudel

Cdk5

P

P

p35

Other pathways Nucleokinesis Microtubule cytoskeleton Migration

Actin cytoskeleton Migration pathways

Figure 6.5 Summary of the LIS1 and RELN signaling pathways for neural migration. Extracellular Reln binds to one of three receptor complexes: cadherin-related neural receptor (Cnr), VldlR/ApoER2, or 3β1 integrin. Receptor stimulation causes activation of mDab1, which, via c-Abl or other pathways, can activate Cdk5/p35. Cdk5 can phosporylate many intracellular targets including regulators of the actin cytoskeleton and Nudel. Phosphorylated Nudel forms a complex with Lis1, mNudE, Dynein and microtubules. This complex is required for nuclear positioning and cell migration. Dcx is also believed to modulate microtubule function. Adapted from Wynshaw-Boris and Gambello [59].

subcortical band heterotopia resembling the pathology observed in humans [70]. Reelin is a large extracellular matrix molecule produced in Cajal–Reitzius neurons and has been found to bind to cadherinrelated neuronal receptors (CNRs) [71], at least two members of the low-density lipoprotein (LDL) receptor family (VLDLR and ApoER2) [72–74], and α3β1-integrin [75]. Upon contact with reelin, CNRs initiate phosphorylation of the cytoplasmic second messenger mDab1, possibly through a CNR-associated tyrosine kinase Fyn [71] but more likely through the LDL receptor [73,76]. The result of this pathway activation is the control of expression and activation of cellular interactions with the extracellular matrix resulting in changes in cell adhesiveness, and, ultimately, the formation of the cortical lamina [77]. This is regulated through activation of integrin α5β1 through the Dab1-Crk/CrkL-C3G-Rap1 pathway that triggers the translocation and activation of N-cadherin. Genetic studies in mice have identified the Scrambler and Yotari mutant mice as mutations in the mDab1 [78]. Scrambler, Yotari and mDab1-/- all exhibit a Reeler phenotype, further supporting the notion that they lie in the same pathway. mDab1 also activates the protooncogene c-Abl. Once activated, c-Abl can phosphorylate Cdk5, a

70

process that is enhanced by Cables, thus activating Cdk5 [79]. Cdk5 and p35 (another activator of Cdk5) have also been implicated in directing neurons to the appropriate location within the cerebral cortex [80–82]. Both are highly expressed in the developing central nervous system and mice engineered to be homozygous mutant for Cdk5 or p35 also show a cortical defect similar, although not identical, to the Reeler phenotype [81]. Cdk5 co-localization with p35, Rac and Pak-1 in neurons [83]. They suggest that a Rac-dependent hyperphosphorylation of Pak-1 results in a dynamic downregulation of actin polymerization and enhancement of new focal complex formation during cell migration and process outgrowth [83]. Activation of Pak has also been shown to result in a loss of stress fibers and focal adhesions [61]. These data indicate that the Rac family of GTPases along with Src family members can regulate cytoskeletal remodeling and therefore transduce guidance signals from the cell membrane to the cytoskeleton. Cdk5 can also phosphorylate 14-3-3ε, thus linking this pathway to the LIS1 pathway. The involvement of microtubule associated proteins such as LIS1 and DCX have clearly been defined in cell migration and proliferation. A role for microtubules has more recently begun to be elucidated. Mutations in various tubulin subunits have

Lissencephaly, Type I Chapter 6

clearly been associated with human cell migration disorders (see above). Experimental studies have now shown that loss of these specific subunits result in defects of migration, cell morphology and proliferation [84]. Of considerable interest is that different tubulin subunits often cannot substitute for each other. ARX is the only transcription factor known to cause lissencephaly. An Arx mutant mouse was generated and shows defects in both radial and nonradial cell migration [39]. These mice also exhibit cortical progenitor proliferation defects. Further studies have identified the transcriptional targets that likely affect the proliferation and cell migration [85,86].

Future directions and therapy The fields of developmental biology, genetics, pathology, neurology and neuroimaging have come together to begin providing detailed explanations for the pathogenesis of several human conditions resulting from cell migration anomalies. The rate of progress in this field of study will undoubtedly provide many advances in the coming years. Finding additional molecules in these pathways along with defining the genetic defects in other families and other syndromes will no doubt provide further insights into cell migration during normal central nervous system development and in the pathogenesis of human malformations. Although malformations like this are largely considered intractable to therapy, two recent studies have begun to challenge this notion. The first came through the recognition LIS1 is a haploinsufficiency disorder in patients. Elucidation of the pathway found that calpain was responsible, at least in part, for the degradation of LIS1. By inhibiting calpain, the half-life of endogenous LIS1 is extended and able to correct some of the deficits [87], which provides some potential opportunity for treating this specific genetic disorder if identified earlier. Studying models of human ARX mutations, estradiol was identified as being able to correct some of the transcriptional dysfunction, improve the number of interneurons in the cortex, and reduce seizure activity in Arx mutant mice if given before puberty in mice [88]. This, too, provides exciting new opportunities to begin treating at least some of the most significant clinical manifestations of these malformation syndromes.

References 1. de Rijk-van Andel JF, Arts WF, Hofman A et al. (1991) Epidemiology of lissencephaly type I. Neuroepidemiology 10:200–4 2. Hatten M E (1999) Central nervous system neuronal migration. Annu Rev Neurosci 22:511–39 3. Rakic P (1990) Principles of neural cell migration Experientia 46:882–91 4. Ramon y Cajal S (1929) Etudes sur la neurogenesis de quelques vertebres. trans. L. Guth, 1960. Springfield, Ill, CC Thomas

5. Rakic P (1972) Mode of cell migration to the superficial layers of fetal monkey neocortex. J Comp Neurol 145:61–83 6. Noctor S C, Flint AC, Weissman TA et al. (2002) Dividing precursor cells of the embryonic cortical ventricular zone have morphological and molecular characteristics of radial glia. J Neurosci 22: 3161–73 7. Anton E S, Marchionni MA, Lee KF, Rakic P (1997) Role of GGF/neuregulin signaling in interactions between migrating neurons and radial glia in the developing cerebral cortex. Development 124:3501–10 8. Rio C, Rieff HI, Qi P et al. (1997) Neuregulin and erbB receptors play a critical role in neuronal migration. Neuron 19:39–50 9. Asou H, Miura M, Kobayashi M et al. (1992) Cell adhesion molecule L1 guides cell migration in primary reaggregation cultures of mouse cerebellar cells. Neurosci Lett 144:221–4 10. Ono K, Toasiewicz H, Magnuson T, Rutishauser U (1994) N-CAM mutation inhibits tangential neuronal migration and is phenocopied by enzymatic removal of polysialic acid. Neuron 13:595–609 11. Fishell G, Hatten M (1991) Astrotactin provides a receptor system for CNS neuronal migration. Development 113:755–65 12. Stitt T, Hatten M (1990). Antibodies that recognize astrotactin block granule neuron binding to astroglia. Neuron 5:639–49 13. Zheng C, Heintz N, Hatten M (1996) CNS gene encoding astrotactin, which supports neuronal migration along glial fibers. Science 272:417–19 14. Husmann K, Faissner A, Schachner M (1992) Tenascin promotes cerebellar granule cell migration and neurite outgrowth by different domains in the fibronectin type III repeats. J Cell Biol 116: 1475–86 15. Galileo D, Majors J, Horwitz A, Sanes J (1992) Retrovirally introduced antisense integrin RNA inhibits neuroblast migration in vivo. Neuron 9:1117–31 16. Marin O, Yaron A, Bagri A et al. (2001) Sorting of striatal and cortical interneurons regulated by semaphorin-neuropilin interactions. Science 293:872–5 17. Tamamaki N, Fujimori K, Nojyo Y et al. (2003) Evidence that Sema3A and Sema3F regulate the migration of GABAergic neurons in the developing neocortex. J Comp Neurol 455:238–48 18. Flames N, Long JE, Garratt AN et al. (2004) Short- and longrange attraction of cortical GABAergic interneurons by neuregulin1. Neuron 44:251–61 19. Lysko D E, Putt M, Golden JA (2014) SDF1 reduces interneuron leading process branching through dual regulation of actin and microtubules. J Neurosci 34:4941–62 20. Marsh ED, Golden JA (2012) Developing Models of Aristalessrelated homeobox mutations. In: Noebels JL, Avoli M, Rogawski MA et al. (eds). Jasper’s Basic Mechanisms of the Epilepsies. 4th ed. Bethesda, MD: National Center for Biotechnology Information 21. Bielle F, Griveau A, Narboux-Neme N et al. (2005) Multiple origins of Cajal-Retzius cells at the borders of the developing pallium. Nat Neurosci 8:1002–12 22. Boulder C (1970) Embryonic vertebrate central nervous system:revised terminology. Anat Rec 166:257–61 23. Cardoso C, Leventer RJ, Ward HL et al. (2003) Refinement of a 400kb critical region allows genotypic differentiation between isolated lissencephaly, Miller-Dieker syndrome, and other phenotypes secondary to deletions of 17p13.3. Am J Hum Genet 72:918–30 24. Kato M, Dobyns WB (2003) Lissencephaly and the molecular basis of neuronal migration. Hum Mol Genet 12(Suppl 1):R89–96

71

Developmental Neuropathology 25. Toyo-oka K, Shionoya A, Gambello MJ et al. (2003) 14-3-3epsilon is important for neuronal migration by binding to NUDEL:a molecular explanation for Miller-Dieker syndrome. Nat Genet 34:274–85 26. Alkuraya F S, Cai X, Emery C et al. (2011) Human mutations in NDE1 cause extreme microcephaly with lissencephaly [corrected]. Am J Hum Genet 88:536–47 27. Willemsen M H, Vissers LE, Willemsen MA et al. (2012) Mutations in DYNC1H1 cause severe intellectual disability with neuronal migration defects. J Med Genet 49:179–83 28. des Portes V, Francis F, Pinard JM et al. (1998) Doublecortin is the major gene causing X-linked subcortical laminar heterotopia (SCLH). Human Molecular Genetics 7:1063–70 29. Gleeson JG, Allen KM, Fox JW et al. (1998) Doublecortin, a brainspecific gene mutated in human X-linked lissencephaly and double cortex syndrome, encodes a putative signaling protein. Cell 92:63– 72 30. Gleeson JG, Lin PT, Flanagan LA, Walsh CA (1999) Doublecortin is a microtubule-associated protein and is expressed widely by migrating neurons. Neuron 23:257–71 31. Gleeson JG, Minnerath S, Kuzniecky RI et al. (2000) Somatic and germline mosaic mutations in the doublecortin gene are associated with variable phenotypes. Am J Hum Genet 67:574–81 32. Poirier K, Lebrun N, Broix L et al. (2013) Mutations in TUBG1, DYNC1H1, KIF5C and KIF2A cause malformations of cortical development and microcephaly. Nat Genet 45:639–47 33. Riviere JB, van Bon BW, Hoischen A et al. (2012) De novo mutations in the actin genes ACTB and ACTG1 cause Baraitser-Winter syndrome. Nat Genet 44:440–4 34. Hong S E, Shugart YY, Huang DT et al. (2000) Autosomal recessive lissencephaly with cerebellar hypoplasia is associated with human RELN mutations. Nat Genet 26:93–6 35. Caviness V (1982) Neocortical histogenesis in normal and reeler mice:a developmental study based upon [3H]thymidine autoradiography. Dev Brain Res 4:293–302 36. Goldowitz D, Cushing RC, Laywell E et al. (1997) Cerebellar disorganization characteristic of reeler in scrambler mutant mice despite presence of reelin. J Neurosci 17:8767–77 37. Lecourtois M, Poirier K, Friocourt G et al. (2010) Human lissencephaly with cerebellar hypoplasia due to mutations in TUBA1A:expansion of the foetal neuropathological phenotype. Acta Neuropathol 119:779–89 38. Magen D, Ofir A, Berger L et al. (2015) Autosomal recessive lissencephaly with cerebellar hypoplasia is associated with a lossof-function mutation in CDK5. Hum Genet 134:305–14 39. Kitamura K, Yanazawa M, Sugiyama N et al. (2002) Mutation of ARX causes abnormal development of forebrain and testes in mice and X-linked lissencephaly with abnormal genitalia in humans. Nat Genet 32:359–69 40. Scheffer I E, Wallace RH, Phillips FL et al. (2002) X-linked myoclonic epilepsy with spasticity and intellectual disability:mutation in the homeobox gene ARX. Neurology 59:348–56 41. Stromme P, Mangelsdorf ME, Shaw MA et al. (2002) Mutations in the human ortholog of Aristaless cause X-linked mental retardation and epilepsy. Nat Genet 30:441–5 42. Hatten M, Mason C (1990) Mechanisms of glial-guided neuronal migration in vitro and in vivo. Experientia 46:907–16 43. Dobyns WB, Truwit CL, Ross ME et al. (1999) Differences in the gyral pattern distinguish chromosome 17-linked and X-linked lissencephaly. Neurology 53:270–7

72

44. Norman M, McGillivray B, Kalousek D et al. (1995). Congenital Malformations of the Brain:Pathological, Embryological, Clinical, Radiolological and Genetic Aspects. New York, NY, Oxford University Press 45. Forman M S, Squier W, Dobyns WB, Golden JA (2005) Genotypically defined lissencephalies show distinct pathologies. J Neuropathol Exp Neurol 64:847–57 46. Friede R (1989) Developmental Neuropathology Berlin, SpringerVerlag 47. Hourihane JO, Bennett CP, Chaudhuri R et al. (1993) A sibship with a neuronal migration defect, cerebellar hypoplasia and congenital lymphedema. Neuropediatrics 24:43–6 48. Howell BW, Gertler FB, Cooper JA (1997) Mouse disabled (mDab1):a Src binding protein implicated in neuronal development. Embo J 16:121–32 49. Hirotsune S, Fleck MW, Gambello MJ et al. (1998) Graded reduction of Pafah1b1 (Lis1) activity results in neuronal migration defects and early embryonic lethality. Nat Genet 19:333–9 50. Bix GJ, Clark GD (1998) Platelet-activating factor receptor stimulation disrupts neuronal migration In vitro. J Neurosci 18:307–18 51. Xiang X, Osmani AH, Osmani SA et al. (1995) NudF, a nuclear migration gene in Aspergillus nidulans, is similar to the human LIS1 gene required for neuronal migration. Mol Biol Cell 6:297–310 52. Morris SM, Albrecht U, Reiner O et al. (1998) The lissencephaly gene product Lis1, a protein involved in neuronal migration, interacts with a nuclear movement protein, NudC. Curr Biol 8:603–6 53. Garces J A, Clark IB, Meyer DI, Vallee RB (1999) Interaction of the p62 subunit of dynactin with Arp1 and the cortical actin cytoskeleton. Curr Biol 9:1497–500 54. Vallee R B, Faulkner NE, Tai C (2000) The role of cytoplasmic dynein in the human brain developmental disease lissencephaly. Biochim Biophys Acta 1496:89–98 55. Feng Y, Olson EC, Stukenberg PT et al. (2000) LIS1 regulates CNS lamination by interacting with mNudE, a central component of the centrosome. Neuron 28:665–79 56. Kitagawa M, Umezu M, Aoki J et al. (2000) Direct association of LIS1, the lissencephaly gene product, with a mammalian homologue of a fungal nuclear distribution protein, rNUDE. FEBS Letters 479:57–62 57. Niethammer M, Smith DS, Ayala R et al. (2000) NUDEL is a novel Cdk5 substrate that associates with LIS1 and cytoplasmic dynein. [see comments]. Neuron 28:697–711 58. Sasaki S, Shionoya A, Ishida M et al. (2000) A LIS/NUDEL/ cytoplasmic dynein heavy chain complex in the developing and adult nervous system. Neuron 28:681–96 59. Wynshaw-Boris A, Gambello MJ (2001) LIS1 and dynein motor function in neuronal migration and development. Genes & Development 15:639–51 60. Faulkner NE, Dujardin DL, Tai CY et al. (2000) A role for the lissencephaly gene LIS1 in mitosis and cytoplasmic dynein function. [see comments]. Nature Cell Biology 2:784–91 61. Manser E, Huang HY, Loo TH et al. (1997) Expression of constitutively active alpha-PAK reveals effects of the kinase on actin and focal complexes. Mol Cell Biol 17:1129–43 62. des Portes V, Pinard JM, Billuart P et al. (1998) A novel CNS gene required for neuronal migration and involved in X- linked subcortical laminar heterotopia and lissencephaly syndrome. Cell 92:51–61 63. Pilz D T, Macha ME, Precht KS et al. (1998) Fluorescence in situ hybridization analysis with LIS1 specific probes reveals a high

Lissencephaly, Type I Chapter 6

64.

65.

66.

67.

68.

69.

70.

71. 72.

73.

74.

75. 76.

77.

deletion mutation rate in isolated lissencephaly sequence. Genetics in Medicine 1:29–33 Pilz DT, Matsumoto N, Minnerath S et al. (1998) LIS1 and XLIS (DCX) mutations cause most classical lissencephaly, but different patterns of malformation. Hum Mol Genet 7:2029–37 Pinard J M, Motte J, Chiron C et al. (1994) Subcortical laminar heterotopia and lissencephaly in two families:a single X linked dominant gene. J Neurol Neurosurg Psychiat 57:914–20 Pilz DT, Kuc J, Matsumoto N et al. (1999) Subcortical band heterotopia in rare affected males can be caused by missense mutations in DCX (XLIS) or LIS1. Hum Mol Genet 8:1757–60 Gleeson JG, Minnerath SR, Fox JW et al. (1999) Characterization of mutations in the gene doublecortin in patients with double cortex syndrome. Ann Neurol 45:146–53 Matsumoto N, Leventer RJ, Kuc JA et al. (2001) Mutation analysis of the DCX gene and genotype/phenotype correlation in subcortical band heterotopia. Eur J Hum Genet 9:5–12 Corbo J C, Deuel TA, Long JM et al. (2002) Doublecortin is required in mice for lamination of the hippocampus but not the neocortex. J Neurosci 22:7548–57 Bai J, Ramos RL, Ackman JB et al. (2003) RNAi reveals doublecortin is required for radial migration in rat neocortex. Nat Neurosci 6:1277–83 SenzakiK, Ogawa M, Yagi T (1999) Proteins of the CNR family are multiple receptors for Reelin. Cell 99:635–47 D’Arcangelo G, Miao GG, Chen SC et al. (1995) A protein related to extracellular matrix proteins deleted in the mouse mutant reeler. Nature 374:719–23 Hiesberger T, Trommsdorff M, Howell BW et al. (1999) Direct binding of Reelin to VLDL receptor and ApoE receptor 2 induces tyrosine phosphorylation of disabled-1 and modulates tau phosphorylation. Neuron 24:481–9 Trommsdorff M, Gotthardt M, Hiesberger T et al. (1999). Reeler/Disabled-like disruption of neuronal migration in knockout mice lacking the VLDL receptor and ApoE receptor 2. Cell 97:689– 701 Dulabon L, Olson EC, Taglienti MG et al. (2000) Reelin binds a3β1integrin and inhibits neuronal migration. Neuron 27:33–44 D’Arcangelo G, Nakajima K, Miyata T et al. (1997) Reelin is a secreted glycoprotein recognized by the CR-50 monoclonal antibody. J Neurosci 17:23–31 Sekine K, Kubo K, Nakajima K (2014) How does Reelin control neuronal migration and layer formation in the developing mammalian neocortex? Neurosci Res 86:50–8

78. Sheldon M, Rice DS, D’Arcangelo G et al. (1997) Scrambler and yotari disrupt the disabled gene and produce a reeler-like phenotype in mice. Nature 389:730–3 79. Zukerberg LR, Patrick GN, Nikolic M et al. (2000) Cables links Cdk5 and c-Abl and facilitates Cdk5 tyrosine phosphorylation, kinase upregulation, and neurite outgrowth. Neuron 26: 633–46 80. Chae T, Kwon YT, Bronson R et al. (1997) Mice lacking p35, a neuronal specific activator of Cdk5, display cortical lamination defects, seizures, and adult lethality. Neuron 18:29–42 81. Gilmore E C, Ohshima T, Goffinet AM et al. (1998) Cyclindependent kinase 5-deficient mice demonstrate novel developmental arrest in cerebral cortex. J Neurosci 18:6370–7 82. Ohshima T, Ward JM, Huh CG et al. (1996) Targeted disruption of the cyclin-dependent kinase 5 gene results in abnormal corticogenesis, neuronal pathology and perinatal death. Proc Natl Acad Sci U S A 93:11173–8 83. Nikolic M, Chou MM, Lu W et al. (1998) The p35/Cdk5 kinase is a neuron-specific Rac effector that inhibits Pak1 activity. Nature 395:194–8 84. Saillour Y, Broix L, Bruel-Jungerman E et al. (2014) Beta tubulin isoforms are not interchangeable for rescuing impaired radial migration due to Tubb3 knockdown. Hum Mol Genet 23: 1516–26 85. Simonet J C, Sunnen CN, Wu J et al. (2015) Conditional Loss of Arx From the Developing Dorsal Telencephalon Results in Behavioral Phenotypes Resembling Mild Human ARX Mutations. Cereb Cortex 25:2939–50 86. Nasrallah MP, Cho G, Simonet JC et al. (2012) Differential effects of a polyalanine tract expansion in Arx on neural development and gene expression. Hum Mol Genet 21:1090–8 87. Yamada M, Yoshida Y, Mori D et al. (2009) Inhibition of calpain increases LIS1 expression and partially rescues in vivo phenotypes in a mouse model of lissencephaly. Nat Med 15:1202–7 88. Olivetti P R, Maheshwari A, Noebels JL (2014) Neonatal estradiol stimulation prevents epilepsy in Arx model of X-linked infantile spasms syndrome. Sci Transl Med 6:220ra12 89. Reiner O, Carrozzo R, Shen Y et al. (1993) Isolation of a Miller– Dieker lissencephaly gene containing G protein beta-subunit-like repeats. Nature 364:717–21 90. Poirier K, Keays DA, Francis F et al. (2007) Large spectrum of lissencephaly and pachygyria phenotypes resulting from de novo missense mutations in tubulin alpha 1A (TUBA1A). Hum Mutat 28:1055–64

73

7

Lissencephaly, Type II (Cobblestone Lissencephaly) Jeffrey A. Golden Department of Pathology, Brigham and Women’s Hospital, Harvard Medical School, Boston, MA, USA

Definition Type II lissencephaly, also known as cobblestone lissencephaly, is genetically, embryologically and pathologically distinct from type I lissencephaly. The loss of gyri and sulci resulting in a smooth surface is the lone common link between the two types of lissencephaly. Even this link is somewhat tenuous as the surface of most brains with type II lissencephaly has an uneven appearance resembling a cobblestone street. While type II lissencephaly shares a thickened cortex with type I, the underlying pathologic process is quite distinct. Many, if not most, cases of type II lissencephaly also have cerebellar and ocular abnormalities in addition to congenital muscular dystrophy. Together, these features make up a series of well-defined syndromes including the Walker–Warburg syndrome (Mendelian Inheritance in Man number, MIM, 236670), muscle–eye–brain disease (MEB of Santovori; MIM 253280) and Fukuyama congenital muscular dystrophy (now called muscular dystrophy dystroglycanopathy; FCMD; MIM 253800), together with a number of recently identified molecular genetic defects with these similar features (Table 7.1). A defect in O-mannosylation appears to underlie the defect in most of these disorders; thus, type II lissencephalies can also be considered disorders of glycosylation. The primary target appears to be α-dystroglycan.

Synonyms Synonyms for the Walker–Warburg syndrome have included cerebro-ocular dysplasia muscular dystrophy syndrome, Warburg syndrome, hydrocephalus, agyria, retinal dysplasia and encephalocele (HARD ± E) syndrome, Chemke syndrome, cerebro-oculo muscular syndrome, Walker type and cerebroocular dysgenesis. Some of these patients may also represent

patients with MEB. The characterization of the molecular genetics for these disorders has allowed a more precise definition.

Epidemiology FCMD is seen almost exclusively in Japan. In contrast, MEB disease is seen primarily in Finland, while Walker–Warburg syndrome is found worldwide. Mutations in fukutin-related protein (FKRP) and acetylglucosaminyltransferase-like protein (LARGE), resulting in MDC1C (now called muscular dystrophy-dystroglycanopathy with or without mental restriction, type B5; MDDGB5) and MDC1D, respectively, are relatively rare, precluding accurate incidence or prevalence data. All of these disorders are quite rare, with the exception of FCMD, which has an incidence of approximately 3/100 000, making it the second most common muscular dystrophy, after Duchenne type, in Japan. Males and females are affected equally in all of these disorders [1].

Genetics Mutations in 16 genes have been associated with the type II lissencephaly spectrum (Table 7.1) and all exhibit autosomal recessive inheritance. Several unique aspects of the genetics deserve comment. FCMD is perhaps the most interesting. The FCMD gene, FUKUTIN, located on chromosome 9, was first identified by positional cloning [2]. The molecular defect was identified as a 3-kb retrotransposon insertion into the 3′ non-coding region (UTR) of the gene. The incidence of FCMD in Japan is approximately 3/100 000, so 1 in 90 persons could be a heterozygote [1]. Eighty-seven percent of FCMD-bearing chromosomes are derived from a single ancestral founder. Estimates suggest that

Developmental Neuropathology, Second Edition. Edited by Homa Adle-Biassette, Brian N. Harding, and Jeffrey A. Golden. © 2018 John Wiley & Sons Ltd. Published 2018 by John Wiley & Sons Ltd.

75

Developmental Neuropathology

Table 7.1 Genes associated with type II lissencephaly (inheritance: autosomal recessive). Gene

Syndrome

Reference

ATP6V0A2

Congenital disorder of glycosylation type II Walker–Warburg syndrome, muscle–eye–brain syndrome Walker–Warburg syndrome, Fukuyama congenital muscular dystrophy

43

FKRP FKTN

GPR56 POMGNT1 POMT1 POMT2 LARGE

LAMC3 LAMB1 ISPD TMEM5 B3GALNT2 B3GNT1 GTDC2 COL4A1

Muscle–eye–brain syndrome Walker–Warburg syndrome, muscle–eye–brain syndrome Walker–Warburg syndrome, muscle–eye–brain syndrome Walker–Warburg syndrome, muscle–eye–brain syndrome, Fukuyama congenital muscular dystrophy

Walker–Warburg syndrome, muscle–eye–brain syndrome Walker–Warburg syndrome, muscle–eye–brain syndrome Walker–Warburg syndrome, muscle–eye–brain syndrome

8, 44 2

45, 46 6 7 47 15

40 41 48 48 49 50 51 42

the mutation was introduced into this population approximately 2500 years ago. This was the first human genetic disorder to be identified as a consequence of a retrotransposon. The presence of this retrotransposon in the 3′ -UTR is believed to destabilize the mRNA and thus to reduce protein levels. The ancestral origin of this retrotransposon accounts for the frequency in Japan and the relative rarity of FCMD outside Japan, although other mutations in FUKUTIN have been identified both within and outside Japan [3,4]. An animal model of fukutin deficiency indicates that this protein is necessary for normal embryonic development and specifically for brain, eye and skeletal muscle development [5]. Interestingly, when mutations that truncate Fukutin are heterozygous with the common retrotransposon mutation, the phenotype is almost always more severe [3]. It is likely that mutations outside Japan will be more common deletions or truncations that are likely be lethal when homozygous, thus explaining the rare incidence outside Japan and the common occurrence within Japan. MEB disease is seen almost exclusively in Finland, although the exact reason remains unclear. MEB disease was mapped

76

to chromosome 1p32 and the gene subsequently identified as POMGnT1, a glycosyltransferase [6]. Numerous mutations have been identified; the majority are nonsense mutations (either deletions or insertions that alter the reading frame of the protein and lead to premature termination). When assayed for catalytic activity, the missense mutation and a mutation caused by an exon-skipping event both resulted in activities that were less than 1% of wild type [6]. Mutations in POMT1, mapping to 9q34, have been identified in some patients with Walker–Warburg syndrome [7]. POMT1 mutations account for 20% or fewer patients with Walker–Warburg syndrome, with a long list of additional genes being more recently identified (see Table 10.1). FKRP was cloned based on homology with FUKUTIN. Like fukutin, it is believed to be a phospholigand transferase [8]. Mutations in FKRP have been identified in a subset of patients with congenital muscular dystrophy now known as MDC1C, and in some mild cases of limb girdle muscular dystrophy 21 (LMGD21). Finally, LARGE encodes bifunctional glycosyltransferase that was first identified as being mutated in the myodystrophy (myd) mouse [9,10]. Analysis of the myd phenotype has shown involvement of the skeletal muscle, heart, retina, peripheral, and central nervous systems (CNS) [11–13]. Neuronal migration abnormalities are present in the brain, particularly the cortex and cerebellum and resemble those seen in the fukutin-deficient mice [5]. LARGE has an N-terminal transmembrane anchor, typical of glycosyltransferases and a coiled coil motif with two putative catalytic regions [10,14]. A screen for mutations in LARGE was subsequently conducted in patients with muscular dystrophy and either mental restriction, structural brain changes or abnormal α-dystroglycan immunolabelling, unlinked to any reported congenital muscular dystrophy loci. A single patient (from 36 tested) with congenital muscular dystrophy, profound mental restriction, white matter changes and subtle structural abnormalities on brain magnetic resonance imaging (MRI) was identified [15]. Her skeletal muscle biopsy showed reduced immunolabelling of α-dystroglycan and a reduced molecular weight of α-dystroglycan by immunoblotting. Subsequently, a number of additional reports have emerged including those reporting the neuropathology [16,17].

Clinical features The hallmark of all of the conditions presented in this chapter is the combination of a congenital muscular dystrophy coupled with CNS involvement. The muscular dystrophy for each of the disorders usually presents in the first year of life but the relative degree of weakness can vary. Walker–Warburg syndrome tends to be the most severe, with children presenting in the neonatal period with hypotonia. Both proximal and distal muscles may be involved and contractures may be present at birth or develop shortly thereafter. Most children with

Lissencephaly, Type II (Cobblestone Lissencephaly) Chapter 7

Walker–Warburg syndrome do not survive beyond one to two years. Patients with FCMD manifest weakness of muscles and general hypotonia that usually appears before nine months of age. Poor sucking and a mildly weak cry may be present in the neonatal period. Joint contractures are not marked at birth, but generally appear in the lower extremities before one year of age. Muscle pseudohypertrophy becomes evident in the cheeks in infancy and in the calves and forearms by early childhood. Functional disability is more severe than in patients with Duchenne muscular dystrophy; most FCMD patients are never able to walk. Patients usually become bedridden before 10 years of age and most of them die by 20 years of age. The presentation of MEB disease is quite similar to FCMD, although it can present even earlier in life and can appear more severe. All infants with MEB are floppy with generalized muscle weakness, including facial and neck muscles, from birth. They also develop contractures, which can be present at birth. Life expectancy for these patients is actually longer than FCMD, as the disorder tends to progress more slowly and patients can often live into adulthood. Mutations in FKRP (MCD1C) and LARGE (MCD1D) have been reported as having congenital muscular dystrophy, however, the condition appears to be milder, presenting at the end of the first decade or in the second decade of life [15,18]. Mental restrcition is another feature characterizing this group of disorders. While the mental capacity of people with Walker– Warburg syndrome is difficult to assess due to the shortened lifespan, severe mental disability and delays in reaching all developmental milestones are observed. Patients with MEB disease and FCMD are profoundly impaired with intelligence quotients ranging from 30 to 50. Seizures are another common feature for patients with congenital muscular dystrophy and structural brain anomalies. Virtually all patients with Walker– Warburg syndrome and MEB disease show abnormal electroencephalograms (EEGs) with seizures. Seizures occur in nearly half of patients with FCMD in association with abnormal EEGs. Patients with Walker–Warburg syndrome, MEB and FCMD usually develop hydrocephalus that can contribute to the clinical presentation and can require shunting. Finally, approximately 20% of patients with Walker–Warburg syndrome will have what appears to be a Dandy–Walker malformation (Chapter 13), while the remaining patients have cerebellar vermal hypoplasia. A somewhat lower percentage (5–10%) will have an occipital encephalocele. All patients, by definition, with MEB disease and Walker– Warburg syndrome have eye anomalies. These include severe congenital myopia, congenital glaucoma, pallor of the optic disks, retinal hypoplasia and retinal dysplasias. The ophthalmologic lesions in FCMD tend to be less severe but are still commonly present. They include myopia, cataract, abnormal eye movement, pale optic disk, and retinal detachment, although the patients are capable of making visual contact.

Imaging MRI is the modality of choice for characterizing the structural brain anomalies in patients with type II lissencephaly. The cerebral cortex is markedly thickened in affected regions with the cortical ribbon averaging 0.8 to 1.4 cm. Imaging studies are usually interpreted as polymicrogyria or pachygyria. Defining where the abnormal cortex ends and where normal cortex begins can be extremely difficult. T2-weighted high intensity in the white matter is frequently present. The high intensity in the white matter is thought to be due to delayed myelination. Other imaging findings can include hypoplasia and flattening of the pons, Dandy-Walker malformation, cerebellar dysplasia (Walker–Warburg syndrome, MEB disease and FCMD), occipital encephaloceles (Walker–Warburg syndrome), cerebellar cysts (with POMGnT1 mutations in particular), and hydrocephalus (19).

Embryology In contrast to type I lissencephaly, which is clearly a defect in cell migration, type II lissencephaly appears to be an overmigration of neurons and glia beyond the glial–pial limitans. Studies of human brains affected with type II lissencephalies, independent of the underlying genetic basis, have similar findings, suggesting a common embryologic origin, although the extent of brain involvement varies significantly. Potentially related to abnormal glycosylation of proteins such as α-dystroglycan (and thus failing to interact with integrins or other proteins), there is a failure in formation (or breakdown) of the glial–pial limitans allowing radial glial processes to extend beyond the definitive cortex (Figure 7.1). As development proceeds, immature neurons and glia migrate past the preplate and early forming cortical plate, producing a thickened malformed cortex. The embryologic bases for the cerebellar and eye anomalies are less well defined but in the case of the cerebellum may represent a similar mechanism.

Macroscopy The examination of the external surfaces of the brain demonstrates numerous abnormalities in type II lissencephaly. A large case series has identified three subtypes which appear to be on a spectrum but have some correlations with the genotype [20]. These changes vary between the different disorders and between cases with mutations in the same gene. Given that the neuropathology has not been reported for patients with mutations, this section focuses primarily on the syndromes without direct reference to the genes, except when noted.

77

Developmental Neuropathology

Model of Cortical Development Type II Lissencephaly Blood vessels Pial boundary Neurone

Cortical plate Intermediate zone Ventricular zone

Radial glial fibers Progenitor cells Postmitotic neuroblaste migrate to cortical plate; populate progressively more superficial positions

Neuronal migration “breaks through” original pial boundary

Inside-to-outside orderly migration

Type II Lissencephaly

Normal Development

Migrating neurons break through pial boundary

Maturation

Maturation Disorganized cortex forms in leptomenigeal space

Large, Internalized vessels

The pathology of these malformations has been described and reviewed many times [21–24]. The brains are almost always small. The most striking abnormality is the loss of sulci and broadening of gyri, giving the brain a lissencephalic appearance. In contrast to the type I lissencephalies, the surfaces of the brain in affected regions of type II lissencephalies have a somewhat irregular, “lumpy bumpy,” appearance reminiscent of a cobblestone street (thus the alternative terminology of cobblestone lissencephaly). In Walker–Warburg syndrome, the entire brain, or nearly the entire brain, is involved, often sparing the medial temporal cortex. Patients with Walker–Warburg syndrome may also show agenesis of the corpus callosum. Patients with MEB disease and, to an even greater extent, FCMD, show variable extents of affected cortex. Malformed regions have been extensively mapped in FCMD, indicating a bias toward the lateral posterior temporoparietal and occipital lobes; however, in

78

Figure 7.1 A hypothesized scheme for the pathogenesis of type II lissencephaly. In normal development, cells migrate out and populate the cortical plate in an inside to outside fashion. In type II lissencephaly, the initial migration to the cortical plate may be normal, but subsequent cells migrate past the cortical plate and into the leptomeninges.

severe cases extensive involvement of the frontal lobes can also be observed [24]. No boundary exists between the abnormal and the normal cortex. Instead, there is a gradient between the cobblestone-like lissencephalic cortex, through broad gyri with shallow sulci, to what appears to be normal gyri. Fusion across the interhemispheric fissure and across the Sylvian fissure can also be observed. This latter finding accounts for the now refuted hypothesis that these disorders resulted from an inutero infection. Occipital encephaloceles, when present in the Walker–Warburg syndrome, involve the occiptal lobes and not the cerebellum or brainstem. Examination of the cerebellum also shows numerous defects. The cerebellum is often small, particularly the vermis, in the Walker–Warburg syndrome, MEB disease, and FCMD. As mentioned above, approximately 15–20% of patients with the Walker–Warburg syndrome have a Dandy–Walker

Lissencephaly, Type II (Cobblestone Lissencephaly) Chapter 7

malformation. In addition, the folia of the cerebellum are often fused giving a matted appearance that varies from focal and usually bilateral, to diffuse. The brainstem is almost always small. Gross sections of the brain show hydrocephalus and relative preservation of the deep gray nuclei. The cerebral white matter is markedly reduced in volume and the cerebral cortex is irregularly thickened. In the most severely affected areas, the cortex can be over 1 cm thick. In less severe areas, the cortex has an irregular appearance reminiscent of polymicrogyria. Unlike type I lissencephaly, the gray–white junction tends to be preserved, although it can have an irregular and nodular appearance, particularly in the most severely affected regions. On cut sections, the midbrain, pons, and medulla are small, primarily reflecting hypoplasia of the corticospinal tracts. The inferior olivary complex is usually normal. The cerebellum shows fused and disorganized folia and may show white matter tracts running over the surface.

Histopathology Microscopic examination of the cerebral cortex reveals a highly disordered structure. Neurons in the most superficial cortex are disorganized and intermingle with thin white matter tracts, and leptomeninges that are clearly fused. The histopathology shares many features with polymicrogyria. The deeper part of the cortex has a suggestion of organization that is disrupted and separated from the more superficial cortex by blood vessels and

(a)

(b)

Figure 7.2 Histopathological features of type II lissencephaly. (a) Normal cortex at 20 weeks of gestation. Note the well defined and uniform molecular layer (black arrowhead) and the well defined thickness of the cortical plate (white line). (b) Twenty-week fetus with Walker–Warburg syndrome. The cortex is markedly thickened (white line, compare to a). It shows marked disorganization with large blood vessels (black arrow) buried deep in the parenchyma. There is a well-defined

leptomeninges (Figure 7.2). The gray matter–white matter junction is usually well defined and smooth, but may show nodules that blend with the overlying cortex. In sum, the cortex is highly disorganized and shows no recognizable lamination as found in type I lissencephaly, and shows no preservation of the molecular layer. In regions that are less severely affected, there continues to be extensive disorganization of the cortex, although less thick and showing features more clearly similar to polymicrogyria with fusion of the molecular layer. However, this is not a fourlayer type of polymicrogyria, but rather an unlayered structure. In the least affected regions, there are small breaches in the pial– glial limitans with the presence of neuroglial tissue extending over the surface of the brain, forming a type of rind. The underlying cortex may be minimally affected in these areas. Finally, it is valuable to note that junctions between normal, or near normal, cortex and malformed cortex may be abrupt and well defined, again in contrast to type I lissencephaly. Examinations of fetuses with type II lissencephaly, either Walker–Warburg syndrome or FCMD, have provided insight into the pathogenesis of this disorder. During development, focal breaches in the pial–glial limitans appear to provide a pathway by which migrating neurons can move past the cortical plate and into the overlying leptomeningeal space (Figure 7.3). We have observed radial glia passing into this space potentially providing the pathway upon which these cells can migrate (Figure 7.3). In doing so, the molecular layer becomes obliterated, and in some areas fused. The original leptomeninges, with its blood vessels, becomes buried within the cortex and the cells that

(c)

white matter but the irregular border with the cortex blurs the gray–white junction. (c) Four-month-old child with Walker–Warburg syndrome. The cortex shows no clear laminar structure with undulating vessels in the parenchyma and irregular arrangement of neurons around these vessels. Although not well appreciated on this section, areas of polymicrogyria-like cortex can be found.

79

Developmental Neuropathology

(a)

(b) Figure 7.4 Cerebellum from a patient with Walker–Warburg syndrome, showing the marked disorganization of both granular neurons and Purkinje cells.

migrate above the original definitive cortical plate fail to form an organized structure, instead populating the region in a haphazard fashion (Figure 7.2). The cerebellum is also a highly disorganized, although there may be relative preservation of the general arrangement of granular neurons and Purkinje cells (Figure 7.4). The general appearance has been labeled cerebellar polymicrogyria. The pathogenesis of the cerebellar defect is less well defined. In addition, abnormal bands of white matter can be seen, fasciculated or diffusely, over the surface of the cerebellar hemispheres. The overall architecture of the brainstem is well preserved, with the exception of the hypoplastic corticospinal tracts as observed grossly. Histopathology of the skeletal muscle shows classical features of congenital muscular dystrophy. Degenerating and regenerating fibers along with marked fibrosis is prominent. Further details of the muscle pathology can be found in a volume on muscle diseases [25]. Finally, the eyes may show retinal dysplasias and cataracts.

Differential diagnosis

Figure 7.3 Section of 21-week fetus with Walker–Warburg syndrome. (a) Compare with Figure 7.1; an apparent cortical plate (CPa) exists below a line of what was likely the original pial–glial limitans (horizontal white arrow at right). There are numerous areas where an apparent breach of the pial–glial limitans has occurred (see vertical white arrows as examples). The subsequent cells migrate past this into the leptomeninges forming a second and highly disorganized cortical region (CPb). (b) Vimentin-labeled radial glia extend past CPa into the area of CPb, possibly providing a migratory route for neurons. The disorganized cortex is again appreciated, as in Figure 7.2.

80

The class of type II lissencephalies have many features in common, at times making them difficult to distinguish, although for some, like FCMD, being of Japanese descent is of an obvious clue to the diagnosis. The development of mass spectrometry based assays from patients’ blood provides an additional method for identifying glycosylation disorders [26]. The first step toward clarifying the diagnosis in these disorders is to determine that they are one of the type II lissencephalies. As outlined above, all of the type II lissencephalies have congenital muscular dystrophy as a prominent component, whereas this is not a feature of any of the known type I lissencephalies. Eye anomalies are also common in this family of disorders. The spectrum of brain anomalies, ranging from severe and diffuse cortical

Lissencephaly, Type II (Cobblestone Lissencephaly) Chapter 7

malformation to small focal anomalies, can make distinguishing one of the type II lissencephaly disorders from congenital muscular dystrophy without brain involvement difficult, but important for patient management and genetic counseling. Focal cortical dysplasia can be extensive and bilateral giving radiographic features similar to those in this class of malformation. However, the absence of the congenital muscular dystrophy and the absence of eye anomalies will distinguish these anomalies.

Pathogenesis The identification of the genetic basis for the various disorders discussed in this chapter has focused the pathogenesis on abnormal glycosylation [14,27]. Glycosylation is a very common post-translation modification of proteins. It is predicated that up to 1% of human genes encode proteins that participate in oligosaccharide synthesis, degradation and function [28]. Glycosylation plays numerous essential roles in development and homeostasis including; enhancing protein stability, mediating ligand–receptor interactions and ensuring the fidelity of proteins targeted for secretion. The correct processing of glycoproteins is also required for normal development of Caenorhabditis elegans [29], Drosophila melanogaster [30] and mice [31]. N-linked glycosylation is the most common form of protein glycosylation. It begins in the endoplasmic reticulum, with the addition of a dolichol pyrophosphate oligosaccharide precursor to an asparagine residue [32]. Numerous human disorders are linked to Nlinked glycosylation defects [27]. O-linked glycosylation is less common and results from the addition of a glycan to the hydroxyl groups on either serine or threonine residues [33]. Several important proteins, including α-dystroglycan, contain large amounts of O-linked glycans. Furthermore, hypoglycosylation of dystroglycan has been observed in FCMD, Walker–Warburg syndrome, MEB disease, MCD1C+ and MCD1D, supporting the

potential role of this protein in the pathogenesis of these disorders [14,27]. Dystroglycans function in skeletal muscle to maintain the sarcolemmal integrity by linking cytoskeletal actin (via dystrophin) to the extracellular matrix through the alpha subunit; α-dystroglycan has been shown to bind laminin, agrin, perlecan and neurexin [34,35]. In MEB disease, FCMD and the Myd mouse, binding of α-dystroglycan to laminin, agrin and neurexin is defective [11,36]. Dystroglycan (Dag1) null mice are embryonic lethal due to very early defects in basement membrane assembly [37]. In a highly informative set of experiments, Moore and colleagues eliminated dystroglycan from the CNS alone and produced a phenotype with striking similarities to Walker–Warburg syndrome, MEB disease, FCMD, and the Myd mouse, with disruption of cortical layering, fusion of cerebral hemispheres, and aberrant migration of granule cells [38]. In combination with the evidence of altered glycosylation of α-dystroglycan, these data implicate an essential role for dystroglycan in both the muscle and brain phenotypes of the glycosylation-deficient muscular dystrophies. POMT1 is a good candidate for the initial attachment of the mannose to dystroglycan, while POMGnT1 is likely to add the GlcNAc residue (Figure 7.5) [6,39]. It has been suggested that fukutin, LARGE and FKRP function in the synthesis of this tetrasaccharide, although this is only conjecture. However, it is unlikely that the mutated proteins in each respective disorder are only involved in post-translational modification of dystroglycan, suggesting that other proteins are also likely to play a role in the pathogenesis of these disorders. Three genes that are causally associated with type II lissencephaly have been identified that are not directly related to glycosylation. All three, including LAMC3 [40], LAMB1 [41], and COL4 [42] do have active roles in the integrity of basement membranes. Given the defective pial–glial membrane that appears to play a role in these disorders, it is reasonable to hypothesize a common pathogenesis with mutations in these genes as well.

Figure 7.5 The glycosylation that occurs on α-dystroglycan in response to the enzymes affected in various congenital disorders of glycosylation (reproduced from Willer et al. [52]).

81

Developmental Neuropathology

Future directions and therapy There is currently no therapy for the CNS manifestations of these disorders. As with most developmental disorders, and particularly those affect the CNS, a difficulty remains to devise interventions extremely early in embryonic development. The concept of using glycosylation enzymes for therapy has been proposed, but specificity and bioavailability remain difficult issues among other problems. Further studies into the enzymes and their substrates will likely help clarify and better define potential strategies. Gene therapy also would seem to hold some promise; however, timing of intervention remains a major hurdle. Prenatal diagnosis is currently the clearest information that can currently be provided to families.

References 1. Toda T, Kobayashi K, Kondo-Iida E et al. (2000) The Fukuyama congenital muscular dystrophy story. Neuromuscul Disord 10:153–9 2. Kobayashi K., Nakahori Y, Miyake M et al. (1998) An ancient retrotransposal insertion causes Fukuyama-type congenital muscular dystrophy. Nature 394:388–92 3. Kondo-Iida E, Kobayashi K, Watanabe M et al. (1999) Novel mutations and genotype–phenotype relationships in 107 families with Fukuyama-type congenital muscular dystrophy (FCMD). Hum Mol Genet 8:2303–9 4. Silan F, Yoshioka M, Kobayashi K et al. (2003) A new mutation of the fukutin gene in a non-Japanese patient. Ann Neurol 53:392–6 5. Takeda S, Kondo M, Sasaki J et al. (2003) Fukutin is required for maintenance of muscle integrity, cortical histiogenesis and normal eye development. Hum Mol Genet 12:1449–59 6. Yoshida A, Kobayashi K, Manya H et al. (2001) Muscular dystrophy and neuronal migration disorder caused by mutations in a glycosyltransferase, POMGnT1. Dev Cell 1:717–24 7. Beltran-Valero de Bernabe D, Currier S, Steinbrecher A et al. (2002) Mutations in the O-mannosyltransferase gene POMT1 give rise to the severe neuronal migration disorder Walker–Warburg syndrome. Am J Hum Genet 71):1033–43 8. Brockington M, Blake DJ, Prandini P et al. (2001) Mutations in the fukutin-related protein gene (FKRP) cause a form of congenital muscular dystrophy with secondary laminin alpha2 deficiency and abnormal glycosylation of alpha-dystroglycan. Am J Hum Genet 69:1198–209 9. Grewal PK, Hewitt JE (2003) Glycosylation defects:a new mechanism for muscular dystrophy? Hum Mol Genet 12 Spec No 2:R259– 64 10. Peyrard M, Seroussi E, Sandberg-Nordqvist AC et al. (1999) The human LARGE gene from 22q12.3-q13.1 is a new, distinct member of the glycosyltransferase gene family. Proc Natl Acad Sci U S A 96:598–603 11. Holzfeind PJ, Grewal PK, Reitsamer HA et al. (2002) Skeletal, cardiac and tongue muscle pathology, defective retinal transmission, and neuronal migration defects in the Large (myd) mouse defines a natural model for glycosylation-deficient muscle–eye–brain disorders. Hum Mol Genet 11:2673–87

82

12. Mathews K D, Rapisarda D, Bailey HL et al. (1995) Phenotypic and pathologic evaluation of the myd mouse. A candidate model for facioscapulohumeral dystrophy. J Neuropathol Exp Neurol 54: 601–6 13. Rayburn HB, Peterson AC (1978) Naked axons in myodystrophic mice. Brain Res 146:380–4 14. Grewal PK, Holzfeind PJ, Bittner RE, Hewitt JE (2001) Mutant glycosyltransferase and altered glycosylation of alpha-dystroglycan in the myodystrophy mouse. Nat Genet 28:151–4 15. Longman C, Brockington M, Torelli S et al. (2003) Mutations in the human LARGE gene cause MDC1D, a novel form of congenital muscular dystrophy with severe mental retardation and abnormal glycosylation of alpha-dystroglycan. Hum Mol Genet 12:2853–61 16. Godfrey C, Clement E, Mein R et al. (2007) Refining genotype phenotype correlations in muscular dystrophies with defective glycosylation of dystroglycan. Brain 130(Pt 10):2725–35 17. Meilleur KG, Zukosky K, Medne L et al. (2014) Clinical, pathologic, and mutational spectrum of dystroglycanopathy caused by LARGE mutations. J Neuropathol Exp Neurol 73:425–41 18. Topaloglu H, Brockington M, Yuva Y et al. (2003) FKRP gene mutations cause congenital muscular dystrophy, mental retardation, and cerebellar cysts. Neurology 60:988–92 19. Clement E, Mercuri E, Godfrey C et al. (2008) Brain involvement in muscular dystrophies with defective dystroglycan glycosylation. Ann Neurol 64:573–82 20. Devisme L, Bouchet C, Gonzal`es M et al. (2012) Cobblestone lissencephaly: neuropathological subtypes and correlations with genes of dystroglycanopathies. Brain 135:469–82 21. Friede R. (1989) In: Developmental Neuropathology. 2nd edition. Berlin, Springer-Verlag, pp. 339–40. 22. Norman M, McGillivray B, Kalousek D et al. (1995). Congenital Malformations of the Brain:Pathological, Embryological, Clinical, Radiolological and Genetic Aspects. New York, NY, Oxford University Press 23. Takada K (1988) Fukuyama congenital muscular dystrophy as a unique disorder of neuronal migration: a neuropathological review and hypothesis. Yonago Acta Medica 31:1–16 24. Takada K, Nakamura H, Tanaka J (1984) Cortical dysplasia in congenital muscular dystrophy with central nervous system involvement (Fukuyama type). J Neuropathol Exp Neurol 43:395–407 25. Karpati G, ed. (2002) Structural and Molecular Basis of Skeletal Muscle Diseases, Basel, ISN Neuropath Press 26. Xia B, Zhang W, Li X et al. (2013) Serum N-glycan and O-glycan analysis by mass spectrometry for diagnosis of congenital disorders of glycosylation. Anal Biochem 442:178–85 27. Martin-Rendon E, Blake DJ (2003) Protein glycosylation in disease: new insights into the congenital muscular dystrophies. Trends Pharmacol Sci 24:178–83 28. Lowe JB, Marth JD (2003) A genetic approach to Mammalian glycan function. Annu Rev Biochem 72:643–91 29. Herman T, Horvitz HR (1999) Three proteins involved in Caenorhabditis elegans vulval invagination are similar to components of a glycosylation pathway. Proc Natl Acad Sci U S A 96:974–9 30. Blair SS (2000) Notch signaling: Fringe really is a glycosyltransferase. Curr Biol 10:R608–12 31. Shafi R, Iyer SP, Ellies LG et al. (2000) The O-GlcNAc transferase gene resides on the X chromosome and is essential for embryonic stem cell viability and mouse ontogeny. Proc Natl Acad Sci U S A 97:5735–9

Lissencephaly, Type II (Cobblestone Lissencephaly) Chapter 7 32. Burda P, Aebi M (1999) The dolichol pathway of N-linked glycosylation. Biochim Biophys Acta 1426:239–57 33. Van den Steen P, Rudd PM, Dwek RA, Opdenakker G (1998) Concepts and principles of O-linked glycosylation. Crit Rev Biochem Mol Biol 33:151–208 34. Henry MD, Campbell KP (1999) Dystroglycan inside and out. Curr Opin Cell Biol 11:602–7 35. Sugita S, Saito F, Tang J et al. (2001) A stoichiometric complex of neurexins and dystroglycan in brain. J Cell Biol 154:435–45 36. Michele DE, Barresi R, Kanagawa M et al. (2002) Post-translational disruption of dystroglycan–ligand interactions in congenital muscular dystrophies. Nature 418:417–22 37. Williamson RA, Henry MD, Daniels KJ et al. (1997) Dystroglycan is essential for early embryonic development: disruption of Reichert’s membrane in Dag1-null mice. Hum Mol Genet 6:831–4. 38. Moore SA, Saito F, Chen J et al. (2002) Deletion of brain dystroglycan recapitulates aspects of congenital muscular dystrophy. Nature 418:422–5 39. Zhang W, Betel D, Schachter H (2002) Cloning and expression of a novel UDP-GlcNAc:alpha-D-mannoside beta1,2-Nacetylglucosaminyltransferase homologous to UDP-GlcNAc:alpha3-D-mannoside beta1,2-N-acetylglucosaminyltransferase I. Biochem J 361(Pt 1):153–62 40. Barak T, Kwan KY, Louvi A et al. (2011) Recessive LAMC3 mutations cause malformations of occipital cortical development. Nat Genet 43:590–4 41. Radmanesh F, Caglayan AO, Silhavy JL et al. (2013) Mutations in LAMB1 cause cobblestone brain malformation without muscular or ocular abnormalities. Am J Hum Genet 92:468–74 42. Labelle-Dumais C, Dilworth DJ, Harrington EP et al. (2011) COL4A1 mutations cause ocular dysgenesis, neuronal localization

43.

44.

45.

46.

47. 48.

49.

50.

51.

52.

defects, and myopathy in mice and Walker–Warburg syndrome in humans. PLoS Genet 7:e1002062 Van Maldergem L, Yuksel-Apak M, Kayserili H et al. (2008) Cobblestone-like brain dysgenesis and altered glycosylation in congenital cutis laxa, Debre type. Neurology 71:1602–8 Mercuri E, Topaloglu H, Brockington M et al. (2006) Spectrum of brain changes in patients with congenital muscular dystrophy and FKRP gene mutations. Arch Neurol 63:251–7 Piao X, Hill RS, Bodell A et al. (2004) G protein-coupled receptordependent development of human frontal cortex. Science 303: 2033–6 Bahi-Buisson N, Poirier K, Boddaert N et al. (2010) GPR56-related bilateral frontoparietal polymicrogyria: further evidence for an overlap with the cobblestone complex. Brain 133:3194–209 van Reeuwijk J, Brunner HG, van Bokhoven H (2005) Glyc-Ogenetics of Walker–Warburg syndrome. Clin Genet 67:281–9 Vuillaumier-Barrot S, Bouchet-Seraphin C, Chelbi M et al. (2012). Identification of mutations in TMEM5 and ISPD as a cause of severe cobblestone lissencephaly. Am J Hum Genet 91:1135–43 Stevens E, Carss KJ, Cirak S et al. (2013) Mutations in B3GALNT2 cause congenital muscular dystrophy and hypoglycosylation of alpha-dystroglycan. Am J Hum Genet 92:354–65 Buysse K, Riemersma M, Powell G et al. (2013) Missense mutations in beta-1,3-N-acetylglucosaminyltransferase 1 (B3GNT1) cause Walker–Warburg syndrome. Hum Mol Genet 22:1746–54 Manzini MC, Tambunan DE, Hill RS et al. (2012) Exome sequencing and functional validation in zebrafish identify GTDC2 mutations as a cause of Walker–Warburg syndrome. Am J Hum Genet 91:541–7 Willer T, Inamori K, Venzke D et al. (2014) Willer (2014), eLife 3:e03941

83

8

Polymicrogyria Jeffrey A. Golden Department of Pathology, Brigham and Women’s Hospital, Harvard Medical School, Boston, MA, USA

Definition Polymicrogyria is a cerebral cortical malformation characterized by an excessively folded cortical ribbon of small, individually thin gyri that are fused across the molecular layer and often closely packed on top of one another. Max Bielschowsky is credited with first using the term and defining the entity [1].

Epidemiology Incidence and prevalence Polymicrogyria has been increasingly recognized since the advent of magnetic resonance imaging (MRI); however, actual incidence and prevalence data are not available. It is considered a rare disease, making the likely prevalence less than 1 in 2000. It is more common than other cortical malformations (i.e. the lissencephalies, personal observation). Sex and age distribution Polymicrogyria is a congenital anomaly of the cerebral cortex, occurring in both males and females. The prevalence is slightly higher in males, suggesting the likely involvement of X-linked genes, although few are known (Table 8.1) [2]. Risk factors A positive family history and intrauterine infections, especially with cytomegalovirus [3], are risk factors. Intrauterine hypoperfusion may also be a risk factor, given that polymicrogyria is associated with severe blood loss during pregnancy, or loss of a twin or triplet [4]. Polymicrogyria has been observed in metabolic disorders, specifically Zellweger syndrome [5]. Individuals with polymicrogyria have also been reported in several multiple congenital anomaly syndromes (22q and 1p36 deletions

syndrome, Adams–Oliver syndrome, and the Niikawa–Kuroki syndrome). The identification of many genes over the 2010s have clearly determined that many cases of polymicrogyria are the result of mutations (see next section).

Genetics Families with autosomal dominant, autosomal recessive and X-linked inheritance patterns of inheritance have been reported, often associated with a specific distribution of the polymicrogyria. In addition, somatic mosaicism has also been identified in a subset of patients. To date, mutations in 20 genes have been found in patients with polymicrogyria (Table 8.1). These include cases where the polymicrogyria is localized to specific lobes or regions of the brain (Table 8.1). Certain chromosomal abnormalities, including the 1p36 and 22q11 deletion syndromes, have also been reported to include polymicrogyria [2,6].

Clinical features Signs and symptoms Individuals born with polymicrogyria may not come to attention until they have a child with the same diagnosis. However, they retrospectively often have some difficulties in their medical or educational history. Polymicrogyria in these individuals will generally be quite localized. More commonly, children present with developmental delay, abnormal tone (often spasticity), or seizures, and are usually microcephalic, although this is not necessarily congenital. About 50% of individuals with polymicrogyria develop epilepsy. Perisylvian polymicrogyria is commonly associated with signs of pseudobulbar palsy, also known as Worster–Drought syndrome [7]. Other associated features may include sensorineural

Developmental Neuropathology, Second Edition. Edited by Homa Adle-Biassette, Brian N. Harding, and Jeffrey A. Golden. © 2018 John Wiley & Sons Ltd. Published 2018 by John Wiley & Sons Ltd.

85

Developmental Neuropathology

Table 8.1 Genes associated with polymicrogyria. Gene

Inheritance

Phenotype

Syndrome

Ref.

AKT3

Autosomal dominant/de novo

Megalencephaly polymicrogyria polydactyly hydrocephalus

27

CCND2

Autosomal dominant/de novo

Megalencephaly, polymicrogyria, hydrocephalus, bilateral perisylvian polymicrogyria Megalencephaly, polymicrogyria

Megalencephaly polymicrogyria polydactyly hydrocephalus

28

FIG4 FGFR3 GPR56

Autosomal recessive Autosomal recessive Autosomal recessive

GPSM2

Autosomal recessive

KIAA1279

Autosomal recessive

OCLN PAX6 PEX1 PI4KA PIK3CA

PIK3R2

Autosomal recessive Autosomal dominant/de novo Autosomal recessive Autosomal recessive Postzygotic/mosaic (most), autosomal dominant/de novo (rare) Autosomal dominant/de novo

RAB3GAP1

Autosomal recessive

RTTN SRPX2 TBR2 TUBA8 TUBB2B TUBB5 WDR62

Autosomal recessive X-linked Autosomal recessive Autosomal recessive Autosomal dominant/de novo Autosomal dominant/de novo Autosomal recessive

Polymicrogyria Medial Temporal lobes Frontal-predominant polymicrogyria, bilateral perisylvian polymicrogyria Frontal-predominant polymicrogyria, polymicrogyria Polymicrogyria, generalized polymicrogyria Generalized polymicrogyria Polymicrogyria Polymicrogyria Polymicrogyria Bilateral perisylvian polymicrogyria

Megalencephaly, polymicrogyria, hydrocephalus, bilateral perisylvian polymicrogyria Polymicrogyria, generalized polymicrogyria, frontal-predominant polymicrogyria Polymicrogyria Polymicrogyria Polymicrogyria Polymicrogyria Polymicrogyria Polymicrogyria Polymicrogyria, generalized polymicrogyria

Thanatophoric dysplasia See also Chapter 10

29 30 31

Chudley–McCullough

32

Goldberg–Shprintzen

33

Pseudo-TORCH

34 35 36 37 27

Zellweger Megalencephaly polymicrogyria polydactyly hydrocephalus Megalencephaly polymicrogyria polydactyly hydrocephalus

27

Warburg micro

38

39 40 41 42 43 44 45

deafness, arthrogryposis [8], and other congenital anomalies (i.e. related to underlying syndromes, such as peroxisomal disorders, thanatophoric dysplasia, or the 22q11 deletion syndrome.

to the distribution of the polymicrogyria, whether unilateral or bilateral, predominantly frontal, perisylvian or parieto-occipital, or diffuse [9].

Imaging Polymicrogyria can generally be diagnosed with confidence on MRI scanning but not necessarily on computed tomography (CT). The features are of apparent cortical thickening, with an irregular border between the gray and the white matter. Usually, there are no additional structural abnormalities of the cerebrum or posterior fossa structures, but a reduction in white matter, or other white matter abnormalities are not uncommon. Intracerebral calcifications associated with a congenital infection may be more readily picked up on CT. Radiological classification relates

Laboratory findings Today, the most important testing is molecular genetic studies to evaluate the known polymicrogyria genes or deep sequencing of the patients’ exome or whole genome. Microarray, fluorescence in situ hybridization, or traditional cytogenetic studies can be valuable adjunct testing to molecular genetic studies to evaluate structural chromosomal abnormalities (e.g. 1p36, 22q11 deletions). Very long chain fatty acids to exclude Zellweger syndrome are appropriate with extreme central nervous system (CNS) involvement and extracranial manifestations

86

Polymicrogyria Chapter 8

(a)

(b)

Figure 8.1 Polymicrogyria. (a) The external surface of the brain exhibits a fine bumpy appearance often likened to morocco leather or cobblestones. (b) Section of the brain demonstrates the hyperconvoluted cortex with a festooned surface. The abnormal cortex is markedly thickened on gross appearance; however, this is an illusion created by the extensively folded microgyria; in fact, the cortex in polymicrogyria is thinned.

supporting the diagnosis. Finally, a TORCH screen (toxoplasmosis, rubella, cytomegalovirus, herpes simplex, and HIV) is recommended when CNS calcifications or other features suggesting that an intrauterine infection are present.

Macroscopy

punctuated by the presence of apparently trapped small and medium-sized blood vessels at what appears to have been the previous pial surface that was likely disrupted during development (see Pathogenesis section). Although previously described as two (unlayered), four or six layers, all three patterns have been observed in the same case [10]. In addition, evaluation of a series of cases with molecular layer markers indicated that independent of the number of layers, the cortex is fundamentally

The cortical surface is irregular, and the convolutions can appear wider than expected, with a “bumpy” surface, like cobblestones or morocco leather (Figure 8.1). On coronal sectioning, the heaped up and fused mini-gyri may produce an apparent thickening of the cortical ribbon (Figure 8.1). The gray–white junction is usually recognizable but frequently shows a festooned architecture. Polymicrogyria can be widespread over one or both hemispheres, symmetric or asymmetric between the hemispheres, in an arterial territory (especially the middle cerebral), restricted to the opercular region or depths of the insula, or be a small focus in any neocortical area except cingulate or striate cortex. Polymicrogyria is almost always found bordering porencephaly or in the preserved temporal lobes of hydranencephaly.

Histopathology The hallmark of polymicrogyria is the presence of a thinned, hyperconvoluted (excessively folded) cortical ribbon with fusion of the molecular layer between adjacent cortical ribbons (Figure 8.2). The fusion across the molecular layers is also

Figure 8.2 Acquired polymicrogyria in varicella zoster embryopathy. The polymicrogyric cortex shows complex folding, fusion and branching.

87

Developmental Neuropathology laminated correctly [10]. While this appears to be true for most cases, the presence of polymicrogyria in patients with a glycosylation disorder (see Chapter 7) clearly individuate polymicrogyria in these cases as completely disorganized [11]. Another characteristic feature is that the gray–white junction is usually very well defined, although it can appear blurred on imaging, due to the festooned morphology at the deep aspect of the malformation. Similarly, the border between normal cortex and polymicrogyria is usually sharp and well defined; this also distinguishes it from pachygyria/agyria.

Differential diagnosis Polymicrogyria must be distinguished from polygyria, a macroscopically hyperconvoluted cortex most often associated with hydrocephalus in early life in which cortical histology and layering is normal. An undulating cortical ribbon may also occur in focal cortical dysplasia, but this is differentiated by the large size and atypia of neurons and glia. Patients with extensive polymicrogyria can be confused with lissencephaly type II, but obliteration of the subarachnoid space, complete disorganization of the laminar pattern into neuronal waves and clusters, and deeply placed nodular heterotopias are notable differences in lissencephaly type II, notwithstanding the presence on occasion of short stretches of thin undulating cortex.

Experimental models The best studied animal model is a rat model of four-layer polymicrogyria produced by contact freezing of the cortical

surface [12]. A 2003 investigation of glutamate receptor subunits in this model found a normal staining pattern and no changes in the tissue surrounding the focal lesion to account for the increased epileptogenicity [13].

Pathogenesis The pathogensis of polymicrogyria is definitively multifaceted. While clearly genetic in many cases, it may also arise as a result of disruptions in embryonic development, usually vascular in nature, and as a consequence of intrauterine infection [14]. Included among the embryonic disruptions are intrauterine ischemia, encephaloclastic lesions, twinning, and intrauterine infection with cytomegalovirus [3,15], toxoplasma [16], varicella zoster virus [17] and syphilis [16]. The genetic basis for polymicrogyria has been described above (Table 8.1). In addition, a variety of metabolic disorders are known to include polymicrogyria, including; Pelizaeus–Merzbacher [18], glutaric acidemia type II [19], thanatophoric dysplasia [20], maple syrup urine disease [21], histidinemia [22], Leigh syndrome, mitochondrial respiratory chain disorder [23], neonatal adrenoleukodystrophy [24] and Zellweger syndrome [25]. Autopsy observations in fetuses 17–22 weeks of gestation suggest the origin to be the early second trimester at the latest, and more likely between 10 and 20 weeks gestation [26]. Given one of the hallmarks of polymicrogyria is the fusion across the molecular layers and the presence in the Oglycosylation disorders (see Chapter 10), it is likely the pathogenesis includes a perturbation in the glial–pial limitans with fusion of adjacent regions (Figure 8.3). The early fusion of the cortical surface, whether genetically, due to loss of glycosylation

Figure 8.3 Schematic diagram of polymicrogyria. The upper image shows the normal development of gyri and sulci with leptomeningeal vessels (red), the pial–glial limitans (lavender) and a cartoon of the laminated cortex. In polymicrogyria (lower image), there is fusion across the pial–glial limitans with entrapment of blood vessels and thinning of the cortex.

88

Polymicrogyria Chapter 8

of α-dysroglycan or perturbation of the brain surface, probably through the disruption of radial glial endplates as the result of infection or ischemic insult, appears to result in the abnormal growth of the brain in a constrained environment. Thus, polymicrogyria appears to be the morphologic manifestation of multiple different pathways with a common underlying disruption in development.

Future directions and therapy Identification of genes associated with polymicrogyria, and their function has begun to provide improved counseling of families, and a better understanding of the underlying mechanisms resulting in this cortical malformation.

References 1. Bielschowsky M (1916) Uber Mikrogyrie. J Psycholog Neurolog 22:1–47 2. Dobyns WB, Mirzaa G, Christian SL, et al. (2008) Consistent chromosome abnormalities identify novel polymicrogyria loci in 1p36.3, 2p16.1-p23.1, 4q21.21-q22.1, 6q26-q27, and 21q2. Am J Med Genet A 146A:1637–54 3. Marques Dias M Harmant-van Rijckevorsel JG, Landrieu P, Lyon G (1984) Prenatal cytomegalovirus disease and cerebral microgyria: evidence for perfusion failure, not disturbance of histogenesis, as the major cause of fetal cytomegalovirus encephalopathy. Neuropediatrics 15:18–24 4. Barkovich A J, Rowley H, Bollen A (1995) Correlation of prenatal events with the development of polymicrogyria. AJNR Am J Neuroradiol 16(4 Suppl):822–7 5. Volpe J.J, Adams RD (1972) Cerebro-hepato-renal syndrome of Zellweger. An inherited disorder of neuronal migration. Acta Neuropathologica 20:175–98 6. Bingham PM, Lynch D, McDonald-McGinn D, Zackai E (1998) Polymicrogyria in chromosome 22 delection syndrome. Neurology 51:1500–2 7. Clark M, Carr L, Reilly S, Neville BG (2000) Worster–Drought syndrome, a mild tetraplegic perisylvian cerebral palsy. Review of 47 cases. Brain 123 (Pt 10):2160–70 8. Baker E M, Khorasgani MG, Gardner-Medwin D et al. (1996) Arthrogryposis multiplex congenita and bilateral parietal polymicrogyria in association with the intrauterine death of a twin. Neuropediatrics 27:54–6 9. Barkovich AJ, Kuzniecky RI, Jackson GD et al. (2001) Classification system for malformations of cortical development: update 2001. Neurology 57:2168–78 10. Judkins AR, Martinez D, Ferreira P et al. (2011) Polymicrogyria includes fusion of the molecular layer and decreased neuronal populations but normal cortical laminar organization. J Neuropathol Exp Neurol 70:438–43 11. Meilleur KG, Zukosky K, Medne L et al. (2014) Clinical, pathologic, and mutational spectrum of dystroglycanopathy caused by LARGE mutations. J Neuropathol Exp Neurol 73:425–41

12. Dvorak K, Feit J, Jurankova Z (1978) Experimentally induced focal microgyria and status verrucosus deformis in rats: pathogenesis and interrelation. Histological and autoradiographical study. Acta Neuropathol 44:121–9 13. Hagemann G, Kluska MM, Redecker C et al. (2003) Distribution of glutamate receptor subunits in experimentally induced cortical malformations. Neuroscience 117:991–1002 14. Golden JA, Harding BN (2010) Cortical malformations: unfolding polymicrogyria. Nat Rev Neurol 6:471–2 15. Friede RL, Mikolasek J (1978) Postencephalitic porencephaly, hydranencephaly or polymicrogyria. A review. Acta Neuropathol 43:161–8 16. Evrard P, de Saint-Georges P, Kadhim HJ Gadisseux JF (1989) Pathology of prenatal encephalopathies. In: JH French, S Hard, P Caser (eds) Child Neurology and Developmental Disabilities, Baltimore, MA, Paul H Brookes, pp. 153–78 17. Harding B, Baumer JA (1988) Congenital varicella zoster. A serologically proven case with necrotizing encephalitis and malformation. Acta Neuropathol 76:311–5 18. Pamphlett R, Silberstein P (1986) Pelizaeus–Merzbacher disease in a brother and sister. Acta Neuropathol 69:343–6 19. Colevas AD, Edwards JL, Hruban RH et al. (1988) Glutaric acidemia type II. Comparison of pathologic features in two infants. Arch Pathol Lab Med 112:1133–9 20. Ho KL, Chang CH, Yang SS, Chason JL (1984) Neuropathologic findings in thanatophoric dysplasia. Acta Neuropathol 63:218–28 21. Martin JK, Norman RM (1967) Maple syrup urine disease in an infant with microgyria. Dev Med Child Neurol 9:152–9 22. Corner BD, Holton JB, Norman RM, Williams PM (1968) A case of histidinemia controlled with a low histidine diet. Pediatrics 41:1074–81 23. Samson JF, Barth PG, de Vries JI et al. (1994) Familial mitochondrial encephalopathy with fetal ultrasonographic ventriculomegaly and intracerebral calcifications. Eur J Pediatr 153:510–6 24. Powers JM (1985) Adreno-leukodystrophy (adreno-testiculoleukomyelo-neuropathic-complex). Clin Neuropathol 4:181–99 25. Evrard P, Caviness VS Jr., Prats-Vinas J, Lyon G (1978) The mechanism of arrest of neuronal migration in the Zellweger malformation: an hypothesis bases upon cytoarchitectonic analysis. Acta Neuropathol 41:109–17 26. Norman MG (1980) Bilateral encephaloclastic lesions in a 26 week gestation fetus: effect on neuroblast migration. Can J Neurol Sci 7:191–4 27. Riviere JB, van Bon BW, Hoischen A et al. (2012) De novo mutations in the actin genes ACTB and ACTG1 cause Baraitser–Winter syndrome. Nat Genet 44:440–4, S441–2 28. Mirzaa GM, Parry DA, Fry AE et al. (2014) De novo CCND2 mutations leading to stabilization of cyclin D2 cause megalencephalypolymicrogyria-polydactyly-hydrocephalus syndrome. Nat Genet 46:510–15 29. Baulac S, Lenk GM, Dufresnois B et al. (2014) Role of the phosphoinositide phosphatase FIG4 gene in familial epilepsy with polymicrogyria. Neurology 82:1068–75 30. Tavormina PL, Shiang R, Thompson LM et al. (1995) Thanatophoric dysplasia (types I and II) caused by distinct mutations in fibroblast growth factor receptor 3. Nat Genet 9:321–8 31. Piao X, Hill RS, Bodell A et al. (2004) G protein-coupled receptordependent development of human frontal cortex. Science 303: 2033–6

89

Developmental Neuropathology 32. Doherty D, Chudley AE, Coghlan G et al. (2012) GPSM2 mutations cause the brain malformations and hearing loss in Chudley–McCullough syndrome. Am J Hum Genet 90:1088– 93 33. Brooks AS, Bertoli-Avella AM, Burzynski GM et al. (2005) Homozygous nonsense mutations in KIAA1279 are associated with malformations of the central and enteric nervous systems. Am J Hum Genet 77:120–6 34. O’Driscoll MC, Daly SB, Urquhart JE et al. (2010) Recessive mutations in the gene encoding the tight junction protein occludin cause band-like calcification with simplified gyration and polymicrogyria. Am J Hum Genet 87:354–64 35. Glaser T, Jepeal L, Edwards JG et al. (1994) PAX6 gene dosage effect in a family with congenital cataracts, aniridia, anophthalmia and central nervous system defects. Nat Genet 7: 463–71 36. Portsteffen H, Beyer A, Becker E et al. (1997) Human PEX1 is mutated in complementation group 1 of the peroxisome biogenesis disorders. Nat Genet 17:449–52 37. Pagnamenta AT, Howard MF, Wisniewski E et al. (2015). Germline recessive mutations in PI4KA are associated with perisylvian polymicrogyria, cerebellar hypoplasia and arthrogryposis. Hum Mol Genet 24:3732–41

90

38. Aligianis I., Johnson CA, Gissen P et al. (2005) Mutations of the catalytic subunit of RAB3GAP cause Warburg Micro syndrome. Nat Genet 37:221–3 39. Kheradmand Kia S, Verbeek E, Engelen E et al. (2012) RTTN mutations link primary cilia function to organization of the human cerebral cortex. Am J Hum Genet 91:533–40 40. Roll P, Rudolf G, Pereira S et al. (2006) SRPX2 mutations in disorders of language cortex and cognition. Hum Mol Genet 15:1195–207 41. Baala L, Briault S, Etchevers HC et al. (2007) Homozygous silencing of T-box transcription factor EOMES leads to microcephaly with polymicrogyria and corpus callosum agenesis. Nat Genet 39:454–6 42. Abdollahi MR, Morrison E, Sirey T et al. (2009) Mutation of the variant alpha-tubulin TUBA8 results in polymicrogyria with optic nerve hypoplasia. Am J Hum Genet 85:737–44 43. Jaglin XH, Poirier K, Saillour Y et al. (2009) Mutations in the betatubulin gene TUBB2B result in asymmetrical polymicrogyria. Nat Genet 41:746–52 44. Breuss M, Heng JI, Poirier K et al. (2012) Mutations in the betatubulin gene TUBB5 cause microcephaly with structural brain abnormalities. Cell Rep 2:1554–62 45. Bilguvar K, Ozturk AK, Louvi A et al. (2010) Whole-exome sequencing identifies recessive WDR62 mutations in severe brain malformations. Nature 467:207–10

9

Cerebral Heterotopia Edwin S. Monuki1 and Keith L. Ligon2 1

Departments of Pathology and Laboratory Medicine, and Developmental and Cell Biology, University of California Irvine, Irvine, CA, USA 2 Division of Neuropathology, Department of Pathology, Brigham and Women’s Hospital, Children’s Hospital Boston, and Department of Oncologic Pathology, Dana-Farber Cancer Institute, and Harvard Medical School, Boston, MA, USA

Definition, major synonyms and historical perspective A heterotopion (Greek heteros, “other” and topos, “place”; plural heterotopia) in the brain refers to a malformative lesion comprising normal cells in the wrong place. Unlike other parts of the body, cerebral heterotopia are not easily distinguished by cell type; essentially, all contain neurons and glia. Thus, a nomenclature based on the location of heterotopia within the cerebrum has historically been more useful than one based on its cellular constituents. This chapter considers three major classes of cerebral heterotopia: leptomeningeal, subcortical and periventricular. Related entities can be found in Chapters 10, 11 and 15. Leptomeningeal heterotopia Leptomeningeal heterotopia (synonyms: leptomeningeal glioneuronal heterotopia, meningeal heterotopia; no Mendelian Inheritance in Man, MIM, entry) consists of focal or diffuse collections of glioneuronal tissue within the leptomeninges (Figure 9.1). It is generally contiguous with the underlying marginal zone or brain parenchyma, although this continuity is not always apparent. The malformation may therefore be related to nodular cortical dysplasia and possibly to marginal zone heterotopia, except that these entities classically do not involve the meninges. Leptomeningeal heterotopia is most commonly multiple, and is associated with numerous other types of central nervous system (CNS) malformations, although it may rarely occur in isolation [1,2]. Subcortical-band heterotopia Subcortical-band heterotopia (synonyms: double cortex syndrome, band heterotopia, ribbon-like heterotopia, subcortical

laminar heterotopia; MIM 300067, 600348) consists of nodular to confluent linear masses of gray matter between and well separated from the ventricular surface and overlying cortex, often forming a contiguous “band” within the subcortical white matter (Figure 9.2). Subcortical-band heterotopia is occasionally seen together with pachygyria (“subcortical-band heterotopiapachygyria”), suggesting that it and pachygyria form a continuous spectrum of malformation phenotypes [3–5]. Subcorticalband heterotopia is generally seen in pure form with no other malformative lesions. Periventricular heterotopia Periventricular heterotopia (synonyms: periventricular nodular heterotopia, bilateral periventricular nodular heterotopia; MIM 300049) consists of nodular masses of gray matter near the ventricular surface (Figure 9.3). Like leptomeningeal heterotopia, periventricular heterotopia can occur alone, but most often occurs with other CNS malformations, such as microcephaly, agenesis of the corpus callosum, septo-optic dysplasia, thanatophoric dysplasia and various chromosomal anomalies, or as a part of defined multiorgan syndromes (OMIM 300049) [6–9].

Normal embryology The normal placement of neurons and glia in the architecture of the brain is a highly coordinated and complex process dependent on proliferation, differentiation and migration. Aberrations of each of these normal embryological processes have been reported to contribute to heterotopia formation. Relevant embryological details are described in other chapters of this book.

Developmental Neuropathology, Second Edition. Edited by Homa Adle-Biassette, Brian N. Harding, and Jeffrey A. Golden. © 2018 John Wiley & Sons Ltd. Published 2018 by John Wiley & Sons Ltd.

91

Developmental Neuropathology (a)

(b)

(c)

(d)

Figure 9.1 Leptomeningeal heterotopia. (a) Focal leptomeningeal heterotopia in a nine-month-old infant demonstrating the “eruption” of cortical neurons and marginal zone tissue through the pial–glial barrier and into the overlying leptomeninges (hematoxylin and eosin, H&E). (b) Immunohistochemistry for NeuN shows mature pyramidal neurons present within a leptomeningeal heterotopia. Synaptic connections appear to exist between these heterotopic neurons and those within the underlying brain (diaminobenzidine reaction). (c) Immunohistochemistry

for glial fibrillary acidic protein shows a disorganized collection of enlarged reactive astrocytes within a leptomeningeal heterotopia. Focal discontinuity of the normally uniform layer of subpial astrocytes that form the glia limitans is associated with the lesion (alkaline phosphatase reaction). (d) Diffuse leptomeningeal heterotopia overlying the brainstem in a fetal case of lissencephaly. The lesion contained classic confluent eruptions of glioneuronal tissue into the leptomeninges with encasement of cranial nerves and large vessels (H&E).

Epidemiology

usually seen without other malformative lesions and is the rarest of the three heterotopia.

Incidence and prevalence Only rough estimates can be made, but, collectively, heterotopia is likely to represent the most common type of malformation, estimated to be present in more than 30% of patients with cortical malformations. Both leptomeningeal and periventricular heterotopias are commonly associated with other malformations, and much less frequently occur in isolation. Leptomeningeal heterotopia may be one of the most common congenital malformations of the brain. In a series of 129 infant autopsies performed for congenital anomalies, 31% of cases had leptomeningeal heterotopia [1]. Subcortical-band heterotopia is

Sex and age distribution There is no known sex predilection for leptomeningeal heterotopia. Subcortical-band heterotopia has a strong female predominance, due to X-linkage of the major causative gene, DCX (see Genetics section). Genetic forms of periventricular heterotopia also have a female predominance, due to X-linkage of the causative gene FLNA and perinatal lethality in males (see Genetics section). However, the association with other anomalies and epigenetic etiologies diminishes the female predominance of this malformation overall.

92

Cerebral Heterotopia Chapter 9

(a)

Figure 9.2 Macroscopic appearance of subcortical-band heterotopia demonstrating the classic features of a well-delineated band of gray matter located in the centrum semiovale. Note the presence of well-myelinated white matter superficial and deep to the heterotopic gray matter, and the lack of involvement of intragyral white matter. The overlying cortex in this case had a normal appearance, but can be pachygyric in others. Subcortical-band heterotopia may appear as a solid band of heterotopic tissue or as numerous islands of radially oriented gray matter separated by white matter, as in this case. (Courtesy of Joe Gleeson, UC San Diego.)

Risk factors Genetic risk factors are involved in all three heterotopia types (see Genetics section). In addition, epigenetic insults, such as hemorrhages, hypoxic ischemic injury and toxins, can result in leptomeningeal and periventricular heterotopias in the absence of genetic risk factors [10,11]. Risk factors that predispose the fetus to epigenetic insults (e.g. uteroplacental insufficiency) therefore increase the risk of leptomeningeal and periventricular heterotopias. It remains unclear whether epigenetic insults have a significant role in subcortical-band heterotopia.

(b)

(c)

Clinical features Signs and symptoms Although the clinical features of isolated leptomeningeal heterotopia are not known, these lesions are frequently seen in syndromes associated with severe mental restriction and seizures. Leptomeningeal heterotopia lesions generally contain cortical neurons and are well situated for involvement in seizures, although this remains unproven. Diffuse leptomeningeal heterotopia may also be associated with hydrocephalus, due to obliteration of the subarachnoid space. The high incidence of focal leptomeningeal heterotopia, coupled with the inability to detect small lesions by current imaging techniques makes leptomeningeal heterotopia attractive as a potential cause for many cases of structurally “silent,” intractable epilepsy.

Figure 9.3 Periventricular heterotopia. (a) Macroscopic appearance of periventricular heterotopia, consisting of multiple confluent nodules lining the lateral ventricles and distorting the ventricular surface. (b) Macroscopic appearance of nodular heterotopia in a diffuse distribution. Note the presence of nodules in periventricular and subcortical locations, as well as the involvement of normal gray matter structures. Such florid cases can make identification of normal gray matter and the distinction of periventricular compared with subcortical-band heterotopia difficult. (Courtesy of Phil Boyer, UT Southwestern.) (c) Microscopic appearance demonstrating multiple discrete nodules of gray matter separated by strands of myelinated white matter (hematoxylin and eosin/Luxol fast blue).

93

Developmental Neuropathology Females with diffuse subcortical-band heterotopia may have only mild to moderate clinical symptoms, while males tend to occupy the extremes of clinical severity (i.e. both more normal and more severe). The most common clinical findings are epilepsy and mental restriction, but these findings are highly variable. When present, epilepsy often takes the form of partial complex seizures with secondary generalization or atypical absence epilepsy with regional or focal onset. Despite the clinical variability, there is a reasonable correlation between clinical severity and extent of subcortical-band heterotopia; for example, bilaterality and thickness of the band [4,12]. The contribution of the heterotopic band to the epilepsy remains to be defined, although the band is synaptically active and participates in at least some normal cognitive processes [13,14]. Like subcortical-band heterotopia, X-linked females with periventricular heterotopia can have epilepsy and cognitive impairment, but these clinical findings may be absent. In families with periventricular heterotopia, epilepsy can be strikingly variable, but often begins in the second decade of life and is drug-resistant. Like subcortical-band heterotopia, partial complex and atypical absence epilepsy are common epilepsy phenotypes [15]. Although the relationship between periventricular heterotopia and epilepsy remains uncertain, the nodules have been reported to be intrinsically epileptogenic. Males with periventricular heterotopia may have more severe symptoms than females, owing to a greater likelihood for males to have other structural abnormalities, such as cortical dysplasia. Specific clinical associations of syndactyly, microcephaly, cleft palate and callosal abnormalities are associated with genetic subtypes.

Imaging T1-weighted magnetic resonance imaging (MRI) has had a tremendous impact on the detection of heterotopia as well as other CNS malformations, and remains the imaging modality of choice. By definition, heterotopia have imaging features similar to normal gray matter, although radiographic features can be unusual. Heterotopia and other malformations are also being increasingly recognized in utero by ultrasonography and MRI. Focal leptomeningeal heterotopia lesions are generally undetectable by current imaging techniques. Diffuse leptomeningeal heterotopia may sometimes be evident on imaging as thickening of the leptomeninges or obliteration of the subpial/ subarachnoidal space [16]. Functional imaging of subcorticalband heterotopia has revealed its participation in some, but not all, normal cognitive functions, while tractography imaging and positron emission tomography studies have determined that subcortical-band heterotopia is innervated by myelinated fiber tracts [17], is synaptically active and shows normal to elevated glucose uptake [18]. There is some evidence that spectroscopy may distinguish subcortical-band heterotopia from other heterotopia, cortical dysplasias and normal cortex [19]. Functional imaging suggests that periventricular heterotopia is similar to normal gray matter in terms of glucose metabolism and

94

perfusion [20], and spectroscopy often reveals abnormalities in the cortex overlying periventricular heterotopia nodules [21].

Macroscopy The microscopic size of leptomeningeal heterotopia lesions (Figure 9.1) normally precludes their gross identification, except in cases of diffuse leptomeningeal heterotopia associated with the type II lissencephalies. These extensive lesions may be recognized as irregularities on the surface of the brain and are most common in frontal regions, at the base of the brain and in the brainstem [2]. Although subcortical-band heterotopia classically has a laminar appearance, its morphology can vary from curvilinear ribbons to nodules [15] and can vary in extent (e.g. bilateral or unilateral, diffuse or focal). Regional differences in band thickness may suggest the underlying genetic defect, since DCX mutations tends to cause diffuse or anterior-biased subcortical-band heterotopia [22] while LIS1 mutations, although extremely rare, may result in posterior-biased lesions [4]. Typically, subcorticalband heterotopia has the appearance of multiple conglomerated nodules, often with radially oriented major axes, which are wholly or partially separated by myelinated tracts (Figure 9.2). Nodules may merge with subcortical gray nuclei, or even with the overlying cortex in combined subcortical-band heterotopia pachygyria. The overlying cortex appears normal in classic cases (e.g. heterozygous DCX females). Periventricular heterotopia can be highly variable in extent, ranging from isolated single nodules to confluent bilateral lesions (Figure 9.3). Nodules often distort the lateral ventricles, forming visible bumps along the ventricular surface commonly near the angle of the lateral ventricles and around the temporal horns. The cortex overlying periventricular heterotopia often shows macroscopic abnormalities, most commonly polymicrogyria [5]. Posterior periventricular heterotopia has been associated with FAT4 and DCHS1 mutations [23].

Histopathology, immunohistochemistry and electron microscopy Cerebral heterotopia are not readily distinguished by their cellular composition. All can contain pyramidal (glutamatergic) and non-pyramidal (GABAergic) neurons that are randomly oriented and without obvious lamination. Other elements characteristic of normal cortex (astrocytes, oligodendrocytes, vessels, neuropil, axons and scant myelin) can be seen to a varying degree. Classic leptomeningeal heterotopia is a focal protrusion of glioneuronal tissue from the surface of the brain into the leptomeninges and subarachnoid space (Figure 9.1). When carefully evaluated by serial sectioning, it is invariably connected to the underlying marginal zone via “bridges” of glioneuronal

Cerebral Heterotopia Chapter 9

tissue [24]. Leptomeningeal heterotopia is often polypoid, with a central blood vessel and a core of radially oriented fibers similar to nodular cortical dysplasias. The underlying cortex generally exhibits features of cortical dysplasia, and the leptomeninges can be fibrotic. Identification can be aided by immunohistochemistry for glial fibrillary acidic protein (GFAP) or synaptophysin, which are not normally present in the meninges. Interestingly, ultrastructural findings suggestive of a primary defect in the glia limitans have been described in leptomeningeal heterotopia associated with a fetal case of Fukuyama-type congenital muscular dystrophy (FCMD) [25]. Subcortical-band heterotopia possesses the general histologic features described above, but lacks distinctive histopathology. The confluent nodules of subcortical-band heterotopia are often surrounded by well-myelinated tracts. Pyramidal cells in the heterotopic band may appear smaller than normal compared with those in the overlying cortex. Unlike leptomeningeal and periventricular heterotopias, the cortex overlying subcorticalband heterotopia is typically normal, although it would be expected to show features of pachygyria (e.g. four-layered) in subcortical-band heterotopia pachygyria. Interestingly, nodules of periventricular heterotopia can sometimes show some rudimentary lamination, and its neurons show changes in size consistent with developmental age. A variety of neuronal types are present, including small neurons that express parvalbumin, calretinin, or neuropeptide Y [26,27]. GFAP is remarkably slight to absent [28]. Decreased expression of N-methyl-D-aspartate receptor and its signaling pathway has been reported in periventricular heterotopia nodules and overlying dysplastic cortex. The neocortex overlying the malformation is frequently dysplastic, often showing typical histologic features of polymicrogyria [5].

Differential diagnosis Categorization into leptomeningeal, subcortical-band or periventricular heterotopia can often be problematic for individual cases. Heterotopia in different locations are often seen together and can merge with overlying cortex or subcortical gray matter. While these features may be understandable based on genetics and pathogenesis, they nonetheless lead to diagnostic confusion. Consideration of all available information, including the preponderant location of lesions, clinical history, family history and molecular genetic testing, should provide diagnostic guidance. The identification of ectopic neuroglial tissue within the leptomeninges does not generally present a diagnostic difficulty. Leptomeningeal heterotopia may sometimes appear to be marginal zone only heterotopia, or nodular cortical dysplasias if serial sectioning is not performed to identify the presence of leptomeningeal involvement. A classical four-layered lissencephalic cortex, when wellmyelinated in layer 3, can sometimes be confused with

subcortical-band heterotopia. Typical subcortical-band heterotopia should be well-separated, not only from the ventricular surface, but also from overlying cortex (i.e. the heterotopic band generally does not involve intragyral white matter). In addition, the cortex overlying the malformation is generally normal. Subcortical-band heterotopia can sometimes merge with a pachygyric cortex (subcortical-band heterotopia pachygyria), but does not do so in a diffuse fashion [3]. Periventricular heterotopia can be confused with the subependymal nodules of tuberous sclerosis [29]. In addition to the lack of other tuberous sclerosis stigmata, the lack of nodular calcification in periventricular heterotopia distinguishes it from tuberous sclerosis. Care must also be taken not to confuse the malformation with gray matter that normally occupies a periventricular location (caudate nucleus, thalamus and hypothalamus). Importantly, distinction between epigenetic and genetic causes of periventricular heterotopia can be suggested by neuropathologic findings. Focality of lesions, lesion location in watershed regions (e.g. paratrigonal region), and macroscopic or microscopic evidence of an epigenetic insult [28] would support an epigenetic etiology.

Genetics Leptomeningeal heterotopia Genetic syndromes associated with leptomeningeal heterotopia are numerous, and include trisomy 13, holoprosencephaly, and three forms of type II lissencephaly: Walker–Warburg syndrome (MIM 236670), FCMD (MIM 253800), and muscle–eye– brain disease (MIM 253280). The genes and pathways responsible for the more common focal forms of leptomeningeal heterotopia are not clear. However, single-gene defects have now been identified in the three type II lissencephalies, which are associated with the diffuse form of leptomeningeal heterotopia. These diseases exhibit autosomal recessive inheritance and share the findings of abnormal cortical lamination and architecture (“cobblestone cortex”), diffuse leptomeningeal heterotopia and congenital muscular dystrophy. Walker–Warburg syndrome is associated with mutations in the protein Omannosyltransferase-1 gene (POMT1, MIM 607423) at 9q34.1 [30]. Muscle–eye–brain disease maps to chromosome 1p34-p33 and involves mutations in the protein O-mannose beta-1,2-Nacetylglucosaminyltransferase gene (POMGNT1, MIM 606822) [31]. FCMD is caused by mutations in the Fukutin gene (FCMD, OMIM #607440) on chromosome 9q31 [32]. The genetics of type II lissencephalies are covered in more detail in Chapter 7. Subcortical-band heterotopia Genetic syndromes associated with subcortical-band heterotopia are doublecortin (DCX; MIM 300121), LIS1 (MIM 601545), and echinoderm microtubule-associated protein-like 1 (EML1; MIM 602033). Somatic and familial subcorticalband heterotopia is most commonly seen in females, due to

95

Developmental Neuropathology mutations in the DCX gene, located at Xq22.3-q23 [33,34]. Families harboring DCX mutations generally show X-linked dominant inheritance, with subcortical-band heterotopia in heterozygous females and lissencephaly in hemizygous males. This is consistent with obligatory mosaicism in females, due to random X-inactivation. In addition to accounting for almost 100% of familial subcortical-band heterotopia, DCX mutations also account for most sporadic cases in females. DCX mutations are most commonly heterozygous loss of function mutations or missense mutations that cluster in two internal tandem repeats. DCX remains the only recurrent causative gene in females with subcortical-band heterotopia. Analysis of rare males with subcortical-band heterotopia has provided a number of illuminating insights. Male patients also possess rare DCX mutations, but these mutations appear to be mild hypomorphic alleles resulting in functional mosaicism. Somatic mosaicism represents another mechanism for generating subcortical-band heterotopia in males [35,36]. Importantly, the apparent lack of genotype–phenotype correlation between DCX mutations and subcortical-band heterotopia phenotypes may also be accounted for by hypomorphic alleles, germline or somatic mosaicism, and skewed X-inactivation. The classical lissencephaly gene, LIS1 (17p13.3), has also been associated with subcortical-band heterotopia in males, with somatic mosaicism resulting in a milder phenotype and posterior involvement. This correlates with the posterior-bias of LIS1-dependent lissencephaly, and contrasts with the anteriorbias and expression of DCX-associated subcortical-band heterotopia [37]. The low frequency of DCX or LIS1 mutations in males with subcortical-band heterotopia has suggested the involvement of other genetic or epigenetic factors. Compound heterozygous and homozygous missense mutations were identified in two families with subcortical-band heterotopia in 2014 [38]. The mutations are proposed to affect the association of EML1 with microtubules [38]. Furthermore, single cases of heterozygous germline mutations in the microtubule-associated proteins KIF2A, TUBA1A, TUBG1 have been reported in patients with subcortical-band heterotopia and are presumed to act as dominant negative mutations.

Periventricular heterotopia Genetic syndromes associated with periventricular heterotopia include Filamin alpha also (Filamin 1; FLNA; MIM 300017), autosomal recessive periventricular nodular heterotopia with microcephaly (ARFGEF2; MIM 608097), ERMARD (MIM 615532), FAT4 (MIM 612411), DCHS1 (MIM 603057), NEDD4L (MIM 606384). The genetics of periventricular heterotopia are complex, owing to the number of genetic syndromes in which the malformation is found; more than 13 types of periventricular heterotopia have been proposed. In most of these syndromes, the causative gene(s) remains unknown. The most well-established and pure form of periventricular heterotopia is

96

X-linked and involves the FLNA gene. Families carrying FLNA mutations show X-linked dominant inheritance with prenatal male lethality [39], and most of the periventricular heterotopia pedigrees studied have FLNA mutations [40]. Familial and sporadic FLNA mutations are generally loss of function mutations. However, milder missense mutations have been identified in sporadic and familial cases, which result in females with only a few nodules and male survival [26,40]. Thus, like DCX mutations in subcortical-band heterotopia, functional mosaicism due to hypomorphic FLNA alleles can give rise to periventricular heterotopia in males. FLNA mutations have recently been associated with four other multiorgan syndromes: otopalatodigital syndrome type 1 and type 2 (MIM #11300 and 304120, respectively), frontometaphyseal dysplasia (MIM 305620) and Melnick–Needles syndrome (MIM 309350). In contrast to the loss of function FLNA mutations seen in patients with periventricular heterotopia, these four syndromes appear to involve gain of function FLNA mutations. FLNA mutations account for only a small minority of cases of sporadic periventricular heterotopia, suggesting the presence of other periventricular heterotopia genes. Additional genes involved in periventricular heterotopia have begun to be described. Rare recessive mutations in ARFGET2 have been reported in two families with periventricular heterotopia [41]. Other families with nonsense and missense mutations in the protocadherin FAT4 (FAT atypical cadherin 4) and its ligand DCHS1 (Dachsous cadherin-related 1) exhibit a unique posterior periventricular heterotopia, which is partially penetrant and is associated with multisystem disorders outside the brain [23]. The periventricular heterotopia (PVNH6) seen in a family with 6q27 deletion syndrome has been shown to be due to loss of function of ERMARD (endoplasmic reticulum membrane associated RNA degradation protein; aka C6orf70) and missense mutations in the same gene have been suggested to also result in periventricular heterotopia [42]. Sporadic heterozygous missense mutations in the E3 ubiquitin ligase NEDD4L were noted in six patients with periventricular heterotopia (PVNH7) and toe syndactyly, adding further diversity to the means by which periventricular heterotopia can develop [43].

Animal models Animal models of cerebral heterotopia have provided mechanistic insights into their human disease counterparts but frequently have also had challenges in recapitulating the human disease. One potential reason for the disparity between humans and mice in terms of modeling neuronal migration disorders is the distance that neurons must migrate in the cerebrum of humans compared with mice. The far greater migratory distance in humans would be expected to amplify subtle defects in migration, and, in turn, could explain why haploinsufficiency is such

Cerebral Heterotopia Chapter 9

a common genetic mechanism in human developmental brain disorders.

Leptomeningeal heterotopia A number of genetic and non-genetic rodent leptomeningeal heterotopia models have been described. Well-characterized genetic defects in the transcription factor genes Lmx1a (the dreher mutant mouse) [44] and Emx2 [45] result in leptomeningeal heterotopia associated with other features of cortical dysplasia. Many other genetic models of leptomeningeal heterotopia have presumed effects on the pial–glial barrier: targeted deletion of the nidogen binding site of laminin γ1 [46], glial-specific deletion of brain dystroglycan via Crelox recombination [47], deficiency of presenilin-1 [48], loss of myristoylated alanine-rich C kinase substrate (MARCKS) [49,50]. Non-genetic insults can produce remarkably similar lesions to those caused by genetic defects. These insults include puncture wounds [51,52], prenatal ethanol exposure [53], and prenatal methylmercury exposure [54]. Subcortical-band heterotopia Despite their clear roles in the human condition, Dcx and Lis1 mouse models are somewhat disappointing models of human subcortical-band heterotopia [55]. Hemizygous Dcx male mice die perinatally, while heterozygous females and hemizygous males possess lamination and functional defects of the hippocampus, but not of the cortex. Lis1 deficient mice, on the other hand, possess a complex dose-dependent neuronal migration phenotype in the neocortex. Mice heterozygous for a null Lis1 allele show delay in neuronal migration, and compound heterozygotes with less than half the normal Lis1 protein levels have even more severe cortical migration defects, indicating a dosedependence on Lis1 [56,57]. This dose dependency is reminiscent of human hypomorphic LIS1 alleles. A fascinating rodent model of uncertain significance to human subcortical-band heterotopia is the tish rat [58]. Tish (“telencephalic internal structural heterotopia”) rats possess subcortical laminar heterotopia that are transmitted in an autosomal recessive pattern and associated with seizures. The spontaneous HeCo mouse mutant was discovered to harbor autosomal recessive mutations in the Eml1 gene, resulting in bilateral subcortical-band heterotopia of direct relevance to the human EML1 mutations. In contrast to typical migration defects seen in other models, the HeCo mice have abnormal proliferation and spindle dynamics of neural progenitor cells [38]. Periventricular heterotopia Attempts to genetically model periventricular heterotopia in mice by FLNA knockout do not produce periventricular heterotopia but knockdown in utero of the gene in rats can produce periventricular heterotopia lesions. Mice with mutations in vesicular pathway components including MEKK4 (engineered) and Napa (spontaneous) have periventricular

heterotopia lesions. However, these do not appear to be mutated in humans with periventricular heterotopia. Excellent animal models of periventricular and other heterotopia can be induced by non-genetic means, such as radiation of the embryo during the neuronal migratory period. Interestingly, radiation to the developing rat brain may have a selective destructive effect on radial glia [59], consistent with the idea that disruption of radial glial–guided migration is a major epigenetic cause of periventricular heterotopia.

Pathogenesis Genetic mechanisms in heterotopia: mosaicism in subcortical-band and periventricular heterotopias The genetic studies of cerebral heterotopia have revealed important insights into mosaicism in human genetic disorders, for which subcortical-band and periventricular heterotopias represent excellent prototypes. The X-linkage of DCX and FLNA result in the well-understood obligatory mosaicism in heterozygous females, due to X-inactivation, while analysis of unusual sporadic and familial cases implicates germline and somatic mosaicism as contributors to the mosaic phenotype. In addition, the identification of hypomorphic DCX, LIS1 and FLNA alleles suggest that hypomorphism is a bona fide mechanism for generating mosaic brain phenotypes, the subcortical-band heterotopia pachygria spectrum of disorders, and individual isolated heterotopia. Cellular mechanisms in heterotopia: defects at distinct stages in radial glial-guided neuronal migration The human and mouse studies suggest that radial migration involves several distinct steps: 1. Onset of radial migration out of the precursor field (germinal matrix or ventricular zone/subventricular zone) 2. Ongoing migration through the future white matter (intermediate zone) 3. Penetration through the subplate 4. Progression through the cortical plate [22]. A fifth step would be stopping at the marginal zone and pial–glial barrier. The three human cerebral heterotopia types fall neatly into this scheme, with periventricular heterotopia as a defect in the onset of migration (step 1), subcortical-band heterotopia as a defect in continuing migration (step 2), and leptomeningeal heterotopia as a defect in stopping (step 5). Defects in steps 3 and 4 would not give rise to heterotopia of themselves, but rather to lamination defects or other cortical dysplasias. Additional human and mouse studies need to be carried out, however, to establish several aspects of this working model. A potentially novel cellular mechanism suggested from animal studies, but not yet shown in humans, is the displacement of neuronal precursor cells. Both the tish model of subcortical-band heterotopia and

97

Developmental Neuropathology radiation or Eml1 induced periventricular heterotopia provides evidence for displaced progenitor cells as a cause [60,61].

Molecular mechanisms and pathways regulating heterotopia formation The genes mutated in diffuse leptomeningeal heterotopia associated with type II lissencephalies (POMT1 in Walker–Warburg syndrome, POMGNT1 in muscle–eye–brain disease, and FCMD in FCMD) have all been implicated in maintenance of membrane integrity and defective glycosylation of skeletal muscle proteins. The marginal zone and subplate structures are proteoglycan-rich, and thus disordered glycosylation in these regions has been postulated as a candidate mechanism critical to leptomeningeal heterotopia formation. DCX and LIS1 both are involved in microtubule dynamics [62], thus implicating microtubules in neuronal migration and subcortical-band heterotopia formation. Doublecortin is preferentially found at the leading edge of processes and acts as a microtubule-associated protein (MAP) to stimulate microtubule polymerization. Human mutations cluster in two internal tandem repeats (“doublecortin repeats”) that mediate tubulin or microtubule binding. Lis1 interacts with components of the centrosome [63], a structure important for microtubule-dependent translocation of the nucleus during cell migration. Doublecortin and Lis1 have been shown to directly interact, although the significance of this interaction in migrating neurons remains uncertain [62]. Filamin A is a well-studied actin-binding protein that is cell-autonomously required for migration of many cell types and the key mechanism implicated to periventricular heterotopia. Filamin A anchors a multitude of membrane proteins, including some implicated in neuronal migration, to the actin cytoskeleton, thus transmitting ligand-receptor signaling into actin reorganization [64]. A cell-autonomous role for Filamin A in neuronal migration is supported by its expression in cortical precursors [39,65] and its interaction with Filamin A-interacting protein [65]. Further studies of NEDD4L and ERMARD function in rodents have suggested that AKT-mTOR signaling and microvesicle trafficking are key regulatory processes that may be altered in a subset of periventricular heterotopia. Non-genetic mechanisms in heterotopia While genetic analysis has provided significant insight into heterotopia pathogenesis, epigenetic insults probably represent the most common cause of leptomeningeal and periventricular heterotopias, as highlighted by discordant monozygotic twin studies [66–68]. It remains unclear whether epigenetic mechanisms have a significant role in subcortical-band heterotopia. A host of epigenetic insults, such as germinal matrix or subpial hemorrhages [10], white matter damage (periventricular leukomalacia) [11], and several other forms of hypoxic–ischemic, physical, radiation, chemical or thermal-induced injury, have been shown to induce heterotopia that are remarkably similar to geneticallyinduced lesions.

98

Treatment, future perspectives, and conclusions While our understanding of heterotopia pathogenesis has advanced greatly, more work remains to understand how to best translate this knowledge into more effective therapies. Diagnosis is a critical step toward this goal, and the advent of nextgeneration sequencing methods has aided diagnosis of these disorders. High read-depth targeted amplicon sequencing of peripheral blood leukocytes (greater than 200 times mean coverage) applied to patients with imaging evidence of cerebral malformations was able to detect specific somatic and germline events related to the disease in at least 17% of patients (27/158) including heterotopia subjects [69]. Future studies are needed to better understand the genetics of these disorders and the means by which they cause symptoms of epilepsy and cognitive effects. The specific neurophysiology and chemistry of subcortical-band and periventricular heterotopias (e.g. relation to MAPK-mTOR signaling) will become increasingly important as future therapies are designed and tested, since surgical treatment of heterotopia for seizure control is not very effective and medical treatments are increasingly discovered for other indications such as cancer. Owing to its more superficial nature, leptomeningeal heterotopia may actually have potential for surgical cure or treatment if methods for ablation and imaging are further developed.

References 1. Hirano S, Houdou S, Hasegawa M et al. (1992) Clinicopathologic studies on leptomeningeal glioneuronal heterotopia in congenital anomalies. Pediatr Neurol 8:441–4 2. Iida K, Hirano S, Takashima S, Miyahara S (1994) Developmental study of leptomeningeal glioneuronal heterotopia. Pediatr Neurol 10:295–8 3. Pilz DT, Kuc J, Matsumoto N et al. (1999) Subcortical band heterotopia in rare affected males can be caused by missense mutations in DCX (XLIS) or LIS1. Hum Mol Genet 8:1757–60 4. D’Agostino MD, Bernasconi A, Das S et al. (2002) Subcortical band heterotopia (SBH) in males: clinical, imaging and genetic findings in comparison with females. Brain 125:2507–22 5. Friede RL (1989) Developmental Neuropathology. Berlin, SpringerVerlag, pp. 330–346 6. Roelens FA, Barth PG, van der Harten JJ (1999) Subependymal nodular heterotopia in patients with encephalocele. Eur J Paediatr Neurol 3:59–63 7. Jan MM (1999) Outcome of bilateral periventricular nodular heterotopia in monozygotic twins with megalencephaly. Dev Med Child Neurol 41:486–8 8. Slaney SF, Chong WK, Winter RM (1999) A new syndrome of short stature, distinctive facial features and periventricular grey matter heterotopia. Clin Dysmorphol 8:5–9 9. Dobyns WB, Guerrini R, Czapansky-Beilman DK et al. (1997) Bilateral periventricular nodular heterotopia with mental retardation

Cerebral Heterotopia Chapter 9

10.

11.

12.

13. 14. 15. 16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

26.

27. 28.

and syndactyly in boys: a new X-linked mental retardation syndrome. Neurology 49:1042–7 Marin-Padilla M (1996) Developmental neuropathology and impact of perinatal brain damage. I: Hemorrhagic lesions of neocortex. J Neuropathol Exp Neurol 55:758–73 Marin-Padilla M (1997) Developmental neuropathology and impact of perinatal brain damage. II: white matter lesions of the neocortex. J Neuropathol Exp Neurol 56:219–35 Barkovich AJ, Guerrini R, Battaglia G et al. (1994) Band heterotopia: correlation of outcome with magnetic resonance imaging parameters. Ann Neurol 36:609–17 Pinard J, Feydy A, Carlier R et al. (2000) Functional MRI in double cortex: functionality of heterotopia. Neurology 54:1531–3 Spreer J, Martin P, Greenlee MW et al. (2001) Functional MRI in patients with band heterotopia. Neuroimage 14:357–65 Barkovich AJ (2000) Morphologic characteristics of subcortical heterotopia: MR imaging study. AJNR Am J Neuroradiol 21:290–5 Toda T, Watanabe T, Matsumura K et al. (1995) Three-dimensional MR imaging of brain surface anomalies in Fukuyama- type congenital muscular dystrophy. Muscle Nerve 18:508–17 Eriksson SH, Symms MR, Rugg-Gunn FJ et al. (2002) Exploring white matter tracts in band heterotopia using diffusion tractography. Ann Neurol 52:327–34 De Volder AG, Gadisseux JF, Michel CJ et al. (1994) Brain glucose utilization in band heterotopia: synaptic activity of “double cortex”. Pediatr Neurol 11:290–4 Munakata M, Haginoya K, Soga T et al. (2003) Metabolic properties of band heterotopia differ from those of other cortical dysplasias: a proton magnetic resonance spectroscopy study. Epilepsia 44:366–71 Morioka T, Nishio S, Sasaki M et al. (1999) Functional imaging in periventricular nodular heterotopia with the use of FDG-PET and HMPAO-SPECT. Neurosurg Rev 22:41–4 Kaminaga T, Kobayashi M, Abe T (2001) Proton magnetic resonance spectroscopy in disturbances of cortical development. Neuroradiology 43:575–80 Gleeson JG, Luo RF, Grant PE et al. (2000) Genetic and neuroradiological heterogeneity of double cortex syndrome. Ann Neurol 47:265–9 Cappello S, Gray MJ, Badouel C et al. (2013) Mutations in genes encoding the cadherin receptor-ligand pair DCHS1 and FAT4 disrupt cerebral cortical development. Nat Genet 45:1300–8. Choi BH, Matthias SC (1987) Cortical dysplasia associated with massive ectopia of neurons and glial cells within the subarachnoid space. Acta Neuropathol 73:105–9 Nakano I, Funahashi M, Takada K, Toda T (1996) Are breaches in the glia limitans the primary cause of the micropolygyria in Fukuyama-type congenital muscular dystrophy (FCMD)? Pathological study of the cerebral cortex of an FCMD fetus. Acta Neuropathol 91:313–21 Kakita A, Hayashi S, Moro F et al. (2002) Bilateral periventricular nodular heterotopia due to filamin 1 gene mutation: widespread glomeruloid microvascular anomaly and dysplastic cytoarchitecture in the cerebral cortex. Acta Neuropathol 104:649–57 Hannan AJ, Servotte S, Katsnelson A et al. (1999) Characterization of nodular neuronal heterotopia in children. Brain 122:219–38 Santi MR, Golden JA (2001) Periventricular heterotopia may result from radial glial fiber disruption. J Neuropathol Exp Neurol 60: 856–62

29. Jardine PE, Clarke MA, Super M (1996) Familial bilateral periventricular nodular heterotopia mimics tuberous sclerosis. Arch Dis Child 74:244–6 30. Beltran-Valero de Bernabe D, Currier S, Streinbrecher A et al. (2002) Mutations in the O-mannosyltransferase gene POMT1 give rise to the severe neuronal migration disorder Walker-Warburg syndrome. Am J Hum Genet 71:1033–43 31. Yoshida A, Kobayashi K, Manya H et al. (2001) Muscular dystrophy and neuronal migration disorder caused by mutations in a glycosyltransferase, POMGnT1. Dev Cell 1:717–24 32. Kobayashi K, Nakahori Y, Miyake M et al. (1998) An ancient retrotransposal insertion causes Fukuyama-type congenital muscular dystrophy. Nature 394:388–92 33. Gleeson JG, Allen KM, Fox JW et al. (1998) Doublecortin, a brainspecific gene mutated in human X-linked lissencephaly and double cortex syndrome, encodes a putative signaling protein. Cell 92:63– 72 34. des Portes V, Francis F, Pinard JM et al. (1998) Doublecortin is the major gene causing X-linked subcortical laminar heterotopia (SCLH). Hum Mol Genet 7:1063–70 35. Kato M, Kanai M, Soma O et al. (2001) Mutation of the doublecortin gene in male patients with double cortex syndrome: somatic mosaicism detected by hair root analysis. Ann Neurol 50: 547–51 36. Poolos NP, Das S, Clark GD et al. (2002) Males with epilepsy, complete subcortical band heterotopia, and somatic mosaicism for DCX. Neurology 58:1559–62 37. Sossey-Alaoui K, Hartung AJ, Guerrini R et al. (1998) Human doublecortin (DCX) and the homologous gene in mouse encode a putative Ca2+-dependent signaling protein which is mutated in human X-linked neuronal migration defects. Hum Mol Genet 7: 1327–32 38. Kielar M, Tuy FP, Bizzotto S et al. (2014) Mutations in Eml1 lead to ectopic progenitors and neuronal heterotopia in mouse and human. Nat Neurosci 17: 923–33 39. Fox JW, Lamperti ED, Eksioglu YZ et al. (1998) Mutations in filamin 1 prevent migration of cerebral cortical neurons in human periventricular heterotopia. Neuron 21:1315–25 40. Sheen VL, Dixon PH, Fox JW et al. (2001) Mutations in the X-linked filamin 1 gene cause periventricular nodular heterotopia in males as well as in females. Hum Mol Genet 10:1775–83 41. Sheen VL, Ganesh VS, Topcu M et al. (2004) Mutations in ARFGEF2 implicate vesicle trafficking in neural progenitor proliferation and migration in the human cerebral cortex. Nat Genet 36:69– 76 42. Conti V, Carabalona A, Pallesi-Pocachard E et al. (2013) Periventricular heterotopia in 6q terminal deletion syndrome: role of the C6orf70 gene. Brain 136:3378–94 43. Broix L, Jagline H, Ivanova L et al. (2016) Mutations in the HECT domain of NEDD4L lead to AKT-mTOR pathway deregulation and cause periventricular nodular heterotopia. Nat Genet 48: 1349–58 44. Costa C, Harding B, Copp AJ (2001) Neuronal migration defects in the Dreher (Lmx1a) mutant mouse: role of disorders of the glial limiting membrane. Cereb Cortex 11:498–505 45. Ligon KL, Echelard Y, Assimacopoulos S et al. (2003) Loss of Emx2 function leads to ectopic expression of Wnt1 in the developing telencephalon and cortical dysplasia. Development 130:2275–87

99

Developmental Neuropathology 46. Halfter W, Dong S, Yip YP et al. (2002) A critical function of the pial basement membrane in cortical histogenesis. J Neurosci 22:6029–40 47. Moore SA, Saito F, Chen J et al. (2002) Deletion of brain dystroglycan recapitulates aspects of congenital muscular dystrophy. Nature 418:422–5 48. Hartmann D, De Strooper B, Saftig P (1999) Presenilin-1 deficiency leads to loss of Cajal-Retzius neurons and cortical dysplasia similar to human type 2 lissencephaly. Curr Biol 9:719–27 49. Stumpo DJ, Bock CB, Tuttle JS, Blackshear PJ (1995) MARCKS deficiency in mice leads to abnormal brain development and perinatal death. Proc Natl Acad Sci U S A 92:944–8 50. Blackshear PJ, Silver J, Nairn AC et al. (1997) Widespread neuronal ectopia associated with secondary defects in cerebrocortical chondroitin sulfate proteoglycans and basal lamina in MARCKSdeficient mice. Exp Neurol 145:46–61 51. Brunstrom JE, Gray-Swain MR, Osborne PA, Pearlman AL (1997) Neuronal heterotopias in the developing cerebral cortex produced by neurotrophin-4. Neuron 18:505–17 52. Rosen GD, Sherman GF, Richman JM Sherman GF, Rosen GD, Stone LV et al. (1992) Induction of molecular layer ectopias by puncture wounds in newborn rats and mice. Brain Res Dev Brain Res 67:285–91 53. Komatsu S, Sakata-Haga H, Sawada K et al. (2001) Prenatal exposure to ethanol induces leptomeningeal heterotopia in the cerebral cortex of the rat fetus. Acta Neuropathol 101:22–6 54. Kakita A, Wakabayashi K, Su M et al. (2001) Experimentally induced leptomeningeal glioneuronal heterotopia and underlying cortical dysplasia of the lateral limbic area in rats treated transplacentally with methylmercury. J Neuropathol Exp Neurol 60:768–77 55. Corbo JC, Deuel TA, Long JM et al. (2002) Doublecortin is required in mice for lamination of the hippocampus but not the neocortex. J Neurosci 22:7548–57 56. Hirotsune S, Fleck MW, Gambello MJ et al. (1998) Graded reduction of Pafah1b1 (Lis1) activity results in neuronal migration defects and early embryonic lethality. Nat Genet 19:333–9

100

57. Gambello MJ, Darling DL, Yingling J et al. (2003) Multiple dosedependent effects of Lis1 on cerebral cortical development. J Neurosci 23:1719–29 58. Lee KS, Schottler F, Collins JL et al. (1997) A genetic animal model of human neocortical heterotopia associated with seizures. J Neurosci 17:6236–42 59. Roper SN (1998) In utero irradiation of rats as a model of human cerebrocortical dysgenesis: a review. Epilepsy Res 32:63–74 60. Lee KS, Collins JL, Anzivino MJ et al. (1998) Heterotopic neurogenesis in a rat with cortical heterotopia. J Neurosci 18:9365–75 61. Ferrer I, Santamaria J, Alcantara S et al. (1993) Neuronal ectopic masses induced by prenatal irradiation in the rat. Virchows Arch A Pathol Anat Histopathol 422:1–6 62. Feng Y, Walsh CA (2001) Protein-protein interactions, cytoskeletal regulation and neuronal migration. Nat Rev Neurosci 2:408–16 63. Feng Y, Olson EC, Stukenberg PT et al. (2000) LIS1 regulates CNS lamination by interacting with mNudE, a central component of the centrosome. Neuron 28: 665–79 64. Gleeson JG, Walsh CA (2000) Neuronal migration disorders: from genetic diseases to developmental mechanisms. Trends Neurosci 23:352–9 65. Nagano T, Yoneda T, Hatanaka Y et al. (2002) Filamin A-interacting protein (FILIP) regulates cortical cell migration out of the ventricular zone. Nat Cell Biol 4:495–501 66. Briellmann RS, Jackson GD, Torn-Broers Y, Berkovic SF (2001) Causes of epilepsies: insights from discordant monozygous twins. Ann Neurol 49:45–52 67. Kuzniecky R, Gilliam F, Faught E (1995) Discordant occurrence of cerebral unilateral heterotopia and epilepsy in monozygotic twins. Epilepsia 36:1155–7 68. Sisodiya SM, Marques W Jr., Everitt A, Sander JW (1999) Male monozygotic twins discordant for periventricular nodular heterotopia and epilepsy. Epilepsia 40:248–50 69. Jamuar SS, Lam AT, Kircher M et al. (2014) Somatic mutations in cerebral cortical malformations. N Engl J Med 371:733–43

10

Hippocampal Sclerosis, Granule Cell Dispersion, and Cortical Dysplasia Maria Thom Departments of Neuropathology and Clinical and Experimental Epilepsy, UCL Institute of Neurology, London, UK

Part 1: Hippocampal sclerosis and granule cell dispersion Definition, major synonyms and historical perspective Hippocampal sclerosis is an acquired pathology characterized by neuronal loss and gliosis involving specific subfields of the hippocampus, which results in volume reduction and hardening of this structure (sclerosis). There are several causes and associations. In childhood and young adults, it is encountered in epilepsy syndromes (particularly temporal-lobe epilepsy) and synonyms include “Ammon’s horn sclerosis” and “mesial temporal sclerosis.” In older adults and the elderly, hippocampal sclerosis can arise in the context of vascular/ischemic brain damage and neurodegenerative conditions/dementia [1], which is not discussed further in this chapter. Hippocampal sclerosis has been recognized in patients with epilepsy for nearly 200 years, with first reports appearing in 1825 and more detailed descriptions of the pathology in a series of 90 postmortem cases, published by Sommer in 1880 (as reviewed in [2]). Systematic pathological descriptions of some of the earliest patients undergoing temporal-lobe surgery for epilepsy treatments were then published in the 1950s by Cavanagh and Meyer [3]. Granule cell dispersion (also termed “granule cell dysplasia”) is commonly observed in the context of epilepsy-associated hippocampal sclerosis. Although there is no agreed definition for granule cell dispersion, a granule cell layer thickness greater than 10 cells or 120 μm has been proposed. It was first described by Carolyn Hasuer in 1990 [4]. Normal embryology The hippocampal formation comprises a group of cortical regions located in the medial temporal lobe that includes the

dentate gyrus, hippocampus, subiculum, presubiculum, parasubiculum, and entorhinal cortex. The human hippocampus begins to form in the second quarter of gestation and, by 15– 19 weeks of gestation, the hippocampus flexes over the parahippocampal gyrus to form the hippocampal fissure, and the distinction of neurons in cornu ammonis subfields CA1–4 and the three layers of the dentate gyrus become evident. The dentate gyrus has an outside-in lamination, with earliest neurons destined for outer layers and the supragranular blade being formed before the infragranular blade. The cytoarchitecture of the hippocampus is complete at birth, although granule cell proliferation continues into the first weeks of postnatal life, with an additional 30% of granule cell neurons added in the first three months. There is also significant postnatal maturation of the mossy fiber pathway A low rate of functional neurogenesis continues in the dentate gyrus into adulthood, which may be influenced by perinatal events.

Epidemiology The incidence of hippocampal sclerosis is 33.6% in large series of patients undergoing surgery for treatment of refractory epilepsy [5]. In post mortem series of patients with epilepsy the incidence is between 30% to 45% [6,7]. There are no reliable prevalence data as the diagnosis of hippocampal sclerosis is based on magnetic resonance imaging (MRI) or pathology diagnosis. In a hospital-based study of 2200 adult outpatients, 25% of patients with temporal-lobe epilepsy were reported to have findings on MRI indicative of hippocampal volume loss [8]. The onset of habitual seizures in hippocampal sclerosis/temporallobe epilepsy is typically in the first or second decades, with some evidence of older age of onset in patients with atypical hippocampal sclerosis [5]. Granule cell dispersion is present in 40–50% of all cases of hippocampal sclerosis/temporal-lobe epilepsy.

Developmental Neuropathology, Second Edition. Edited by Homa Adle-Biassette, Brian N. Harding, and Jeffrey A. Golden. © 2018 John Wiley & Sons Ltd. Published 2018 by John Wiley & Sons Ltd.

101

Developmental Neuropathology

Clinical features Hippocampal sclerosis is clinically associated with the syndrome of mesial temporal-lobe epilepsy and is frequently associated with drug resistance (failure to control seizures with two or more appropriate anti-epileptic drugs). Although benign clinical forms are recognized, temporal-lobe epilepsy/hippocampal sclerosis has a characteristic natural history and seizure type, with dyscognitive seizures and associated aura (the old term for which is “complex partial seizures”) as well as generalized convulsions. Clinical history often discloses an initial precipitating injury occurring in the first five years of life, such as a complex febrile convulsion, head trauma, episode of hypoxia, or intracranial infection. This is typically followed by a latent interval of several years before the emergence of habitual seizures. The diagnosis of hippocampal sclerosis is usually established by high-resolution MRI (including thin contiguous slices appropriately oriented through the hippocampus), which is highly sensitive at demonstrating an atrophic hippocampus with increased T2-weighted signal. Scalp electroencephalogram (EEG) patterns typically show ipsilateral anterior temporal interictal discharges. Concordant ictal discharges and video EEG studies and neuropsychometry are also carried out prior to any surgical treatments such as anterior temporal lobectomy or hippocampectomy. The main clinical differential diagnoses include structural causes of temporal-lobe epilepsy, such as low-grade tumors, as well as inflammatory conditions such as autoimmune limbic encephalitis. Pathology Hippocampal sclerosis Hippocampal sclerosis can be macroscopically evident at postmortem examination, with unilateral (or bilateral) atrophy of the hippocampus and corresponding expansion of the temporal horn of the lateral ventricle (Figure 10.1). In a typical surgical temporal lobectomy specimen carried out for epilepsy treatment, the hippocampus appears small on coronal sectioning (Figure 10.1) and firm in texture. The pathological diagnosis of hippocampal sclerosis is usually straightforward in haemotoxylin and eosin stained sections and based on the identification of pyramidal neuron loss and gliosis affecting specific subfields, primarily CA1, CA4 and CA3. The depletion of pyramidal cells in these regions may be complete, with only sparse pyramidal and horizontal, interneuronaltype cells remaining. Severe neuronal loss is accompanied by the presence of a dense, scar-like fibrous gliosis, contracting the stratum pyrimidale, attesting to a longstanding process. In some cases, less complete pyramidal cell loss may be apparent with a more cellular, astroglial reaction. Nevertheless, there is typically a sharp cut-off between the preserved subiculum and gliotic CA1. The granule cell layer and CA2 are regarded as more “resistant” regions but patchy neuronal depletion in these areas is often evident. Conventional stains such as Luxol fast blue/cresyl

102

violet and immunohistochemistry for MAP2, NeuN, GFAP (Figure 10.1) may all assist in the appreciation of neuronal loss and gliosis and are particularly useful tools when dealing with a fragmented or poorly oriented surgical specimen. Many quantitative studies have aimed to correlate the severity and distribution of hippocampal damage with both clinical and imaging features [9]. In addition, various grading schemes, including the Wyler score, have been proposed based on regional patterns of damage. The International League Against Epilepsy (ILAE) scheme was introduced in 2013, and recognizes three distinct types of hippocampal sclerosis (Figure 10.2, Table 10.1). The utility of this practical system remains to be confirmed, but early clinical and pathological correlations have shown that type 1 hippocampal sclerosis is associated with a good postoperative outcome following surgical resection [10], but in type 2 hippocampal sclerosis there are even better long-term outcomes and preserved preoperative memory function [11]. Additional pathological features that may accompany hippocampal sclerosis include hypertrophy and neurofilament accumulation in residual hilar neurons, increased corpora amylacea, and prominence of the microvasculature. Tracts from prior depth electrode studies may be present, eliciting a local inflammatory and glial reaction, and mild focal chronic inflammation may be present, but extensive inflammation requires exclusion of underlying limbic encephalitis including viral causes. In addition to these degenerative features, regenerative changes, including axonal sprouting and alteration of interneuronal populations, occur in hippocampal sclerosis in a relatively stereotypical fashion, and may be more relevant to the process of epileptogenesis [12]. These changes include alteration of interneurone number cell hypertrophy, abnormal dendritic projections [13] and axonal sprouting. Axonal sprouting in hippocampal sclerosis is also exemplified by the mossy fiber pathway, the axons of the granule cells. Mossy fiber sprouting is relatively specific to epilepsy-related hippocampal sclerosis and is readily demonstrated in sections with immunohistochemistry to dynorphin or Zinc-transporter 3 (Figure 10.1). Although the extent of mossy fiber sprouting can vary between cases, it is considered to be triggered early following seizure activity, although the exact mechanism and its clinical significance are uncertain. In addition to sclerosis of the hippocampus, accompanying gliosis and neuronal loss are often identified in the temporal pole, neocortex, parahippocampal gyrus, white matter, amygdala, and thalamus. Granule cell dispersion In granule cell dispersion, the typical pathological features are blurring of the outer boundary of the cell layer with the molecular layer, as well as a less well-defined basal cell layer in some cases (Figure 10.2). Increased distances between individual granule cells are observed, with neuropil visible between cells and an overall increased width of the cell layer up to 400 microns (compared with the normal thickness of around 120 microns). In some cases, clusters of granule cells in the

Hippocampal Sclerosis, Granule Cell Dispersion, and Cortical Dysplasia Chapter 10

(a)

(b)

(d)

(c)

(g)

CA1 CA4

(h)

(e)

(f)

(i)

Figure 10.1 Hippocampal sclerosis. (a) Postmortem from a patient with sudden death in epilepsy who had not undergone magnetic resonance imaging during life; bilateral hippocampal sclerosis was confirmed at postmortem. (b) A surgical resection of the hippocampus in a patient with temporal lobe epilepsy, sliced coronally, showing visible atrophy of CA1 sector (arrowed). (c) and (d) MAP2 staining can aid in the confirmation of hippocampal sclerosis with preservation of neurons and dendrites in the intact CA1 and CA4 of a normal hippocampus (c)

compared with reduction of labeling in these regions in hippocampal sclerosis (d). (e) GFAP confirmed intense labeling in CA1 and CA4 compared with the subiculum in hippocampal sclerosis. (f, g, h, i) Mossy fiber sprouting in the molecular layer is a common finding and can be confirmed with zinc transporter 3 (ZnT3) immunolabeling (f); mossy fiber sprouting also shown in higher magnification in (g) Timms stain, (h) dynorphin immunohistochemistry and (i) ZnT3. Original magnification (b, c, d, e, f) ×1; (g, h, i) ×20.

103

Developmental Neuropathology

(a)

(b)

(c)

(d)

(e)

(f)

(g)

Figure 10.2 Granule cell dispersion. (a, b) Shown at higher magnification with haemotoxylin and eosin stain (a) and cresyl violet (b), with increased separation of the cells, intervening neuropil, and elongated and fusiform appearing neurons. (c–f) NeuN showing a relatively compact and normal granule cell layer in hippocampal sclerosis (c) contrasting with regions showing clusters of dispersed cells in the

molecular layer, or a focal bilaminar pattern (d), poor definition of the basal cell layer in addition to the outer layer (e), and “migration” of single neurons into the molecular layer (f). Original magnification: ×20. (g) Low-power view of entire dentate gyrus with dispersion of the granule cells imparting a bilaminar pattern (NeuN).

molecular layer and hypertrophied cells with a more fusiform shaped cell body [14] are observed (Figure 10.2). A “bi-layer” pattern may be present in some cases and granule cell dispersion may alternate with stretches of neuronal loss along the dentate gyrus. Extensive granule cell dispersion is virtually pathognomonic of seizure-induced hippocampal changes and is virtually always seen in the context of hippocampal neuronal loss, particularly of CA4, suggesting that it is more common in ILAE type I and 3 hippocampal sclerosis. Loss of calbindin expression, particularly from the basal cells in granule cell dispersion, has been reported [15].

Genetics and pathogenesis The relationship between seizures and hippocampal sclerosis is a complex, reciprocal cause and effect model. Meyer’s hypothesis of the 1950s, that an initiating insult primed an immature or “susceptible” hippocampus for the subsequent development of hippocampal sclerosis, remains the most plausible explanation i.e. a “two hit” model with a plethora of potential initiating hits including febrile seizures. With its extensive cortical connections, the hippocampus may represent the vulnerable “fuse box” in an immature, seizing brain. Seizures have the potential to damage the hippocampus, particularly prolonged seizures

104

Table 10.1 The International League Against Epilepsy (ILAE) 2013 scheme for the classification of hippocampal sclerosis (HS) in epilepsy. Semi-quantitative microscopic examination is based on formalin-fixed, paraffin-embedded surgical specimen (of 4–7 μm section thickness), with hematoxylin andeosin staining, cresyl violet combined with Luxol-fast blue staining, GFAP and NeuN immunohistochemistry. The scoring system refers to neuronal cell loss best evaluated on NeuN and is defined for cornu ammonis subfields CA1–CA4: 0 = no obvious neuronal loss or moderate astrogliosis only; 1 = moderate neuronal loss (which corresponds to 30–40% of neuronal loss) and gliosis; 2 = severe neuronal loss (majority of neurons appear lost) and fibrillary astrogliosis. Scores for the dentate gyrus: granule cell layer is normal (score 0), dispersed (score 1; can be focal), or shows severe granule cell loss (score 2; can be focal). In the diagrams, neurones are represented in red, granule cells in blue and astrocytes in green. Modified from Thom [12]. Type 3

No HS

CA1 predominant; atypical HS

End folium sclerosis; atypical HS

End folium gliosis

1–2 0–1 0–1 0–1 0–1 5–25

0–1 0–1 0–1 1–2 0–2 1–7.5

0 0 0 0 0–1 20

Type 2

CA2

Z

GCL

Subiculum

a Relative

frequency of hippocampal sclerosis types in surgical series

CA1

CA1

SG

SG

CA4

Z SGZ

GCL

Subiculum

1 CA

CA2

1

1

GC L CA3

CA

1

GC L

No HS 1 CA

CA

1 CA

Type 3 1 CA

1 CA

GC L CA3

GCL

Subiculum

CA

Synonyms Classical; total HS; typical Semi-quantitative assessment of neuronal loss; total HS: CA1 2 CA2 0–2 CA3 0–2 CA4 2 Dentate gyrus 0–2 Relative frequency (%)a 60–80 Diagram Type 1

Type 2

CA2

Z

SG SG Z

CA1

Type 1

CA1

ILAE (2013)

CA4

GC L CA3

SGZ

GCL

Subiculum

CA2

Developmental Neuropathology and status epilepticus. The FEBSTAT (hippocampal sclerosis after febrile status epilepticus) prospective study has shown that febrile status epilepticus may evolve into hippocampal sclerosis, as confirmed on MRI [16]. Hippocampal sclerosis can also arise in association with a second “epileptogenic” pathology such as a malformation or tumor (“dual pathology”). Hippocampal sclerosis is typically a sporadic condition with a complex, polygenic background; genetic susceptibility determinants are not, as yet, well defined [17]. Rare pedigrees provide opportunities to explore the genetic basis. The ApoEε4 genotype has been associated with increased risk of bilateral hippocampal sclerosis [18], and sporadic hippocampal sclerosis and febrile seizures are linked by common genetic variation around SCN1A gene [19]. In addition, epigenetic factors may contribute, and specific DNA methylation profiles have been defined in hippocampal sclerosis [20], as well as abnormal expression of miR218 and miR-204 [21]. Differential miRNA expression is also reported in hippocampal sclerosis with or without granule cell dispersion [22]. There have been several reports of hippocampal developmental abnormalities predisposing to hippocampal sclerosis. An MRI study of asymptomatic family members of patients with temporal-lobe epilepsy and hippocampal sclerosis confirmed smaller, asymmetrical hippocampal volumes [23]. Genomewide studies have linked common genetic variants with hippocampal volume [24,25]. Granule cell dispersion was itself initially considered a developmental defect, and it has been reported in the context of generalized malformations, such as polymicrogyria, and also reported as a potential developmental vulnerability factor in sudden infant death syndrome [26]. In the context of hippocampal sclerosis, current evidence favors an acquired process. Granule cell dispersion is associated with early onset of epilepsy and febrile seizures (less than four years of age) and a longer duration of epilepsy [14]. There are two main theories of granule cell dispersion: (i) it represents neuronal ectopia of newly generated neurons following aberrant neurogenesis as an effect of seizure activity; (ii) it results from abnormal migration of mature neurons influenced by seizures [27]. Local deficiency of reelin protein and reelin-expressing cells, affecting the radial glial scaffold, has been implicated as orchestrating the process of granule cell dispersion [28]. Loss of regenerative capacity in the dentate gyrus has been linked to memory impairment in temporal-lobe epilepsy [29] highlighting the important contribution of granule cell pathology to comorbidities in epilepsy.

Treatment, future perspective, conclusions Surgery is currently regarded as the treatment of choice in patients with drug-resistant focal epilepsy [35,36]. In a 2015 meta-analysis of 1735 surgical patients with confirmed hippocampal sclerosis, 74% achieved a good outcome at one year [36]; in long-term studies, over 50% remain seizure-free at fiveyear follow-up [37]. There are conflicting data as to whether the presence of granule cell dispersion in hippocampal sclerosis signifies a good outcome following surgery [14,38,39]. Future research, including tissue-based studies, is likely to focus on the heterogenous causes and effects of hippocampal sclerosis and granule cell dispersion, predisposing genetic factors, epigenetics, identification of prognostic biomarkers, the extra-hippocampal networks involved, as well as novel treatment targets and early interventions to prevent hippocampal sclerosis.

Animal models There are several well-studied animal models of hippocampal sclerosis/temporal-lobe epilepsy that recapitulate the human condition [30]; many are based on systemic chemoconvulsant administration [31]. Experimental models have enabled study of the processes of epileptogenesis during the “latent interval” between insult and onset of habitual seizures [32], neuroprotection, drug resistance and new therapeutic targets [33] and involvement of wider hippocampal networks [34].

Normal embryology In the developing neocortex, divisions of progenitor cells (radial glial stem cells) and intermediate progenitor cells in the ventricular and subventricular zones, give rise to the precursors of excitatory pyramidal neurons that migrate radially through the subplate to form the cortical plate. A six-layered cortex is formed in an inside-out order by 18 weeks of gestation, with organization of neurons into radial columnar units or modules [42]. This complex process is dependent upon both

106

Part 2: Cortical dysplasia Definition, major synonyms and historical perspective The focal cortical dysplasias (FCD) represent a spectrum of localized brain abnormalities of the cortical cytoarchitecture and myeloarchitecture, considered to represent malformations of cortical development. They are highly associated with epilepsy. The original description was by Taylor et al. in 1971 [40], of a specific cortical abnormality bearing “balloon cells”. In subsequent decades, with increased availability of neuroimaging and surgical treatments for focal epilepsies, the term “focal dysplasia” became generically used for various malformations from heterotopia and polymicrogyria to the more subtle “microdysgenesis” of the cortex. Additional terms were then introduced, including “mild dysplasia”, “non-Taylor dysplasia”, “cytoarchitectural dysplasia” and “glioneuronal hamartia”. The ILAE classification of FCD in 2011 [41] was devised by pathologists, radiologists, and clinicians to provide a simple, universally agreed evidence-based system that could be consistently applied and that was useful for clinical diagnosis and prognosis (Table 10.2). Some key features were the segregation of “balloon” cell and isolated dysplasia from dysplasias associated with other lesions. Microscopic malformations lacking cortical laminar abnormalities (previously termed microdysgenesis) remained separate from FCDs under the umbrella term “mild malformations of cortical development” (MCD).

Hippocampal Sclerosis, Granule Cell Dispersion, and Cortical Dysplasia Chapter 10

Table 10.2 The International League Against Epilepsy classification of focal cortical dysplasias, 2011. In this system, it also notes that “FCD type III (not otherwise specified, NOS)” can be used if a clinically/radiologically suspected principal lesion is not available for microscopic inspection. The rare association between FCD types IIa and IIb with hippocampal sclerosis, tumors, or vascular malformations, should not be classified as FCD type III variant but FCD type II and the second lesion (i.e. a double pathology). Subtype

FCD type I

FCD type II

FCD type III

Definition

Isolated dysplasias that lack dysmorphic cells

Isolated dysplasias with dysmorphic cells

Combined dysplasias with other pathology

Dysmorphic neurons and abnormal lamination Dysmorphic neurons and balloon Cells and abnormal lamination –

Cortical lamination abnormalities in the temporal lobe associated with hippocampal sclerosis Cortical lamination abnormalities adjacent to a glial or glioneuronal tumor Cortical lamination abnormalities adjacent to vascular malformation Cortical lamination abnormalities adjacent to other lesion acquired during early life; e.g., trauma, ischemic injury, encephalitis

Defining cytoarchitectural pathology: A Abnormal radial cortical organisation B

Abnormal tangential cortical lamination

C

Abnormal radial organisation and tangential cortical lamination –

D

–

intrinsic cellular and extrinsic molecular clues, including from the early Cajal–Retzius neurons in the marginal zone. Subpopulations of tangentially migrating GABAergic cortical interneurons synchronize with the radial migration, integrating into cortical networks. Cortical gliogenesis mainly follows neurogenesis. Neuronal specification and fate is determined at cell birth, and the maturing cortex develops regional patterns of organization with distinct patterns of laminar architecture [43]. The radial columnar architecture becomes less prominent in the second half of gestation, but persists postnatally in some regions [44]. Although neurogenesis, migration, and cortical architecture formation are mainly completed by birth, the processes of gliogenesis, synaptogenesis, and myelination continue postnatally. FCDs are considered to represent malformations due to abnormalities arising during post-migrational brain development and maturation.

Epidemiology FCD is the most common malformation in large epilepsy surgical series. It represents the third most common lesion in adult epilepsy surgical series (20% of diagnoses; hippocampal sclerosis and tumors being more common), but is the most common abnormality in pediatric epilepsy series (40–50% of diagnoses) [45–47]. The relative incidence of subtypes varies [48], but, based on larger surgical series of isolated FCD, type IA represents 46%, type IB 19%, type IIA 15% and type IIB 39%. The incidence of type IIIA varies between 11% and 46% in temporallobe epilepsy series [49–51] and type IIIB is more variable [52], with no firm data available for types IIIC and IIID. Seizures associated with isolated FCD usually begin in the first decade of life. They often present in infancy, but can present in adolescence or even into adulthood. There is no clear sex predilection or geographical clustering.

Isolated FCD is mainly a sporadic condition, but genetic risk factors are now being identified, and familial cases have been reported [53–55], accounting for a minority of cases of FCDII (see FCD type II section). It remains possible that external events acting during cortical development and maturation (head injury, ischemia, infection) and epigenetic factors also contribute and that FCD is multifactorial [56]. Intrauterine human papilloma virus was identified as a potential causative pathogen in FCDIIB [57] but not confirmed in subsequent studies [58,59]. Specific risk factors for FCDIII, over and above those for the main associated lesion, remain to be fully validated; for example, more febrile seizures were noted in cases of hippocampal sclerosis with FCDIIIA compared with those without FCDIIIA [49], and higher seizure frequency and male predominance in FCDIIIB compared with tumors without FCDIIIB [60].

Clinical features There is a strong association between FCD and intractable, often disabling, epilepsy [61], which is the primary clinical feature. In FCD types I and II, the seizure semiology and the presence of other neurology is dependent on the localization and extent of the dysplasia [41,62]. In FCDIII, there is no clear evidence that the clinical presentation is significantly different from the main associated lesion [49,60]. Surgical treatment is often indicated for seizure control in FCD, owing to a poor response to appropriate anti-epileptic medications [36]. Imaging and electrophysiology Presurgical evaluation includes combinations of high-resolution MRI, electrophysiological studies as EEG and functional studies (functional MRI, positron emission tomography, singlephoton emission computed tomography), with increasing use of quantitative MRI, magnetoencephalography and integrated

107

Developmental Neuropathology multimodal analysis. The ultimate aim is to define the structural abnormality anatomically, and to map the electrical abnormal region, to individually tailor surgery for complete resection and best outcome in terms of seizure alleviation, but to minimize resection of functional cortex and deficits [61,63]. MRI abnormalities are more evident in FCD type II than other types; typical features include cortical thickening or gray– white matter blurring on T1-weighted images and abnormal signal intensity in both the cortex and white matter on T2 and fluid attenuation inversion recovery sequences. The white matter signal changes frequently taper toward the ventricle (the transmantle sign). In FCDI, MRI may demonstrate hypoplasia and, occasionally, abnormal signal intensity [56]. Conventional MRI can be normal in proven FCD, but detection rates improve with quantitative MRI. EEG studies support the FCD lesion as the generator of ictal activity and typical findings include repetitive subcontinuous, spikes, spike and waves, polyspikes, or bursts of fast rhythms. In surgical series, FCD type II is more often extratemporal, involving the frontal lobe (55%), central sulcus (16%), parietal lobe (10%), occipital lobe (7%), temporal lobe (7%), insular cortex (5%), and being multilobar in 22% of cases. FCDIII is congruent with the lobar localization of the main lesion; its presence may not be suspected preoperatively on MRI, or other investigations and surgical management is dictated by the main lesion. A study of pathology-confirmed FCDIIIA did not detect this on quantitative MRI [64], but gray matter MRI changes have been reported in other series [50]. “Blurring” of the gray–white matter noted on MRI in the temporal lobe adjacent to hippocampal sclerosis may correlate with a reduction of myelin and axons rather than a cortical developmental abnormality.

Important differential diagnosis The main histological differential diagnosis of FCD are other malformative lesions. For FCDII cortical tubers in tuberous sclerosis, some forms of hemimegalencephaly and tumors harboring dysmorphic neuronal cells, such as gangliogliomas or nodular vacuolated neuronal tumors, are the main differentials; distinction in small samples is aided by clinical–radiological correlation. For FCD types I and III, “dyslamination” needs to be distinguished from normal variability in laminar and radial cortical architecture appropriate for the anatomical site [44]; use of layer-specific neuronal markers may help this distinction. Laminar sclerosis (neuronal loss and gliosis) due to a previous insult (including status epilepticus) should not be misinterpreted as FCDIB. In FCDIIIB, bona fide dysplasia should be distinguished from the cortical infiltration zone of the tumor, using tumor markers like CD34. In mild MCD, practical problems arise in the distinction of “pathological” from “normal” numbers of layer I or interstitial white matter neurons, which may necessitate quantitative evaluation [65].

108

Pathology Macroscopic features In tissue resections, FCDII may appear normal to the naked eye, but, in some cases, broadening of the cortex with poor demarcation from the underlying white matter is perceived (Figure 10.3) and the tissue may appear firmer. The size of the abnormality varies, typically being a few centimeters across and involving adjacent gyri, but it can be localized to the depth of one sulcus, or more extensive throughout one hemisphere or multifocal. In FCDIII, the primary lesion itself (e.g. a tumor or cavernoma), is typically more apparent at the macroscopic level (Figures 10.4 and 10.5). Microscopic features FCD type I Abnormal cortical architecture defines the pathology of FCD type I (Figure 10.3). Three subtypes are recognized (Table 10.2). In FCDIA, an exaggerated radial or “microcolumnar” cortex is observed, with rows of small neurons arranged perpendicularly in the cortex [66], often more apparent in midlayers. In FCDIB, the normal six-layered cortical architecture appropriate for the anatomical site is lacking (dyslamination); this may involve all layers (“alaminar” cortex) or specific layers [41]. FCDIC has the combined features of FCDIA and IB. FCDI may be accompanied by blurring of the gray–white matter boundary, and occasional hypertrophic or immature neurons, as well as increased numbers of single, ectopic white matter neurons in the underlying white matter [67]. There is no diagnostic immunohistochemistry biomarker for FCDI; NeuN is helpful in assessment of cortical architecture; a relative lack of mTOR pathway activation has been noted and abnormal expression of developmental markers has been described including SOX2, Otx1, DCX and Map1b. FCD type II The hallmarks of FCD type II (Figure 10.3) are the presence of abnormal neuroglial cell types in addition to disorganization of the cortical architecture [41]. Hypertrophic pyramidal neurons and abnormal neurons with irregular, globoid shapes, orientation, and dendritic patterns (dysmorphic neurons) populate the full thickness of the cortex and subcortical zone, or are scattered through laminae with intervening normal neurons; more rarely, they can predominate in one lamina. Nissl staining of dysmorphic neurons is coarse with elliptical “thickening” of the nuclear membrane. The normal six-layered horizontal lamination of the cortex in these regions is lost, as well as radial organization, when compared with adjacent normal cortex. The normal myeloarchitecture of the cortex is also typically disrupted in FCDII, and the white matter is hypomyelinated, more often in FCDIIB than in IIA. Cortical layer I typically remains hypocellular and distinct from deeper laminae. Balloon cells are also present and define FCDIIB. These enlarged, round cells with glassy pink cytoplasm and eccentric, often multiple nuclei on haemotoxylin and eosin staining, are typically located in the superficial layers,

Hippocampal Sclerosis, Granule Cell Dispersion, and Cortical Dysplasia Chapter 10

(a)

(b)

NeuN

NeuN

(d)

MPB (g)

(c)

(e)

Nestin (h)

NF-NP (f)

CD34

Cx43

(i)

(j)

Figure 10.3 Focal cortical dysplasia type IIA. (a) With region of cortex showing scattered dysmorphic neurons on hematoxylin and eosin (H&E) stain; (inset) these neurons are highlighted with neurofilament immunohistochemistry. (b) FCDIIB on macroscopic examination shows regions of poor definition between the white and gray matter, and ‘thickening’ (arrow) compared with adjacent slice on the right side, with normal cortex. (c) Balloon cells in FCDIIB on H&E with dysmorphic neuron shown at higher magnification (inset) with coarse Nissl substance. (d) The abnormalities of lamination can be appreciated on NeuN staining, which, in this case, shows a transition from laminated cortex to the right of the arrows compared with a dyslaminar cortex to the left of the arrows in the region of dysplasia. (e) Neurons in FCDIIB from case shown in (d) at higher magnification on NeuN, with dysmorphic and hypertrophic neurons mingled with normal-appearing and

-sized neurons. (f) Neurofilament immunohistochemistry (nonphosphorylated, as shown here, as well as phosphorylated neurofilament) highlight the dysmorphic and hypertrophic pyramidal neuronal cells, as well as thick dendrites and axons. (g) Myelin basic protein immunohistochemistry shows paucity of axon labeling in the white matter in the zone of FCDIIB compared with (inset) normal white matter myelination. (h, i, j) Balloon cells. Immunohistochemsitry for intermediate filaments such as nestin shown here (h), as well as vimentin, GFAP and GFAP-delta, highlight balloon-cell populations, which can also show focal membranous labeling with stem-cell markers as CD34 (i) and strong expression of gap junction proteins (connexin 43) (j), among others. Original magnification: (a) ×10, (c, e, f, g, h) ×20, (d) ×2.5, (i, j) ×40.

as well as the subcortical regions around vessels, and in regions of hypomyelination, trailing in the white matter toward the ventricle. The pathology can be diagnosed on conventional stains but immunohistochemistry is helpful for confirmation of FCDII in small biopsies, or where representation of abnormal cell types is limited. NeuN and neurofilament antibodies can aid in the

identification of dysmorphic neurons and abnormal lamination whereas balloon cells are highlighted with intermediate filament antibodies (vimentin > GFAP-delta > nestin > GFAP). Retained abnormal expression of immature or developmental regulated proteins, including CD34, DCX, and β1-integrin, among others [68,69] has been shown in FCDII, as well as upregulation of proteins of potential relevance to epileptogenicity, including mTOR

109

Developmental Neuropathology

(a)

(b)

(c)

CD34

GFAP (e)

(d)

(f)

Figure 10.4 Focal cortical dysplasia type IIIA and IIIB. (a) FCDIIIA (dysplasia adjacent to hippocampal sclerosis in temporal lobe epilepsy) with clustering of neurons in the most superficial aspect of layer II and loss of neurons from the deeper parts of layer II and layer III pyramidal cells. (b) FCDIIIA architectural abnormalities are always accompanied by gliosis of the superficial cortex on GFAP in the regions of cell loss. (c, d) FCDIIIB is cortical dysplasia adjacent to low-grade glioneuronal tumors. In this case, apparent abnormal lamination in NeuN (c) was a

result of infiltrating CD34-positive tumor cells at the margins of a ganglioglioma, highlighted in (d) with CD34 labelling. (e) Clustering of small layer II neurons in FCDIIIA shown at high magnification through a thick section and abnormal orientation compared with (f) normal temporal lobe cortex. Original magnification: (a, b) ×10 and (c, d) ×2.5, (e, f) ×40; all figures are NeuN apart from those indicated.

pathway activation markers, connexins and inflammatory mediators.

this group from isolated FCD type I [41]. In type IIIA, the temporal neocortex adjacent to hippocampal sclerosis in patients with temporal-lobe epilepsy can show FCDI-like dyslaminar abnormalities, small “lentiform” neuronal heterotopias, or a more specific cortical abnormality also termed “temporallobe sclerosis” [49], characterized by clustering of neurons in

FCD type III The ILAE classification of FCD introduced type III (dysplasias adjacent to epileptogenic lesions) primarily to separate

110

Hippocampal Sclerosis, Granule Cell Dispersion, and Cortical Dysplasia Chapter 10

(a)

(b)

gangliogliomas and dysembryoplastic neuroepithelial tumors. FCDIIIC is cortical dyslamination, or abnormal cytoarchitectural composition (hypertrophic neurons) adjacent to vascular malformations as cavernomas, meningioangiomatosis and so on (Figure 10.4). FCDIIID encompasses the spectrum of alterations of the cortical architecture (cortical dyslamination, hypoplasia), and marked myeloarchitectural cortical patterns reported adjacent to other perinatally acquired lesions or injuries, including infarcts, traumatic and inflammatory lesions as Rasmussen’s encephalitis (Figure 10.5). FCDIIID encompases pathologies previously termed etat cribl´e or post-ischemic marbled cortex.

(c)

(d)

Figure 10.5 Focal cortical dysplasia type IIIC. (a, b, c, d) Occipital lobe resections (a) for an adult with Sturge–Weber syndrome seen in (b) on coronal slicing. Abnormal leptomeningeal vessels with calcification and atrophy of the underlying cortex on haemotoxylin and eosin (c). Dyslamination and cortical thinning, as well as some mineralization, extended beyond the boundaries of the vascular anomaly and into adjacent gyri a shown with NeuN (d), amounting to FCD IIIC (FCD adjacent to a vascular malformation). Original magnifications in (c) and (d) ×10.

cortical layer II, accompanied by neuronal loss and gliosis in the superficial cortex (Figure 10.6). In FCDIIIB (dysplasias adjacent to low-grade epilepsyassociated tumors), FCDI-like dyslaminar abnormalities and hypertrophic neurons have been variably described (Figure 10.6), and are more commonly reported adjacent to

Mild malformations of cortical development Mild MCD are more often reported in the lateral temporal lobe in association with hippocampal sclerosis, but also as a more widespread cortical alteration. Histopathology features include excess of heterotopic and other neurons in layer I, including Cajal–Retzius cells (mild MCD type I), although precise pathological criteria for this condition remain poorly defined [41]. Mild MCD type II encompasses an excess of single or clusters of neurons of normal morphology in the white matter, deep to the immediate subcortical region. Several quantitative methods have confirmed a mean excess of white matter neurons in patients with temporal-lobe epilepsy compared with control groups (as reviewed in [65]). In a single case based on NeuN or MAP2 staining, densities of >8 neurons per high power field (or >100 neurons/mm2 ) represent definite mild MCDII [65], and >30 neurons/mm2 possible mild MCDII [41,65]. Other pathological features, such as neuronal clustering or excess oligodendroglia in the white matter, are not currently included in definitions of mild MCD.

Genetics and pathogenesis The molecular genetic bases of FCD subtypes are likely to be different. Most FCD occur sporadically and familial cases are uncommon [53–55]. Genetic conditions, such as tuberous sclerosis and hemimegalencephaly, share histological features with FCDII. Mutations of TSC1 and TSC2 in tuberous sclerosis and in AKT3, DEPDC5, MTOR, PIK3CA or PTEN in hemimegalencephaly [70–73] involve the phosphatidylinositol 3-kinase (PI3K)/AKT3/target of rapamycin mTOR signaling pathway. Evidence for activation of this pathway in the dysmorphic neurons and balloon cells of FCDII are also evident [74,75]. Deep sequencing studies have identified somatic activating mutations in MTOR in 15.6–46% of cases of FCDIIA/B [76–78]. PIK3CA mutation in FCDIIA [73], AKT3 duplication [79], and DEPDC5 mutations of FCDII and I [53–55,72], upstream regulators of the MTOR pathway, have also been reported. It seems likely that as yet unidentified somatic, as well as germline, mutations in mTOR pathway genes may be involved in the pathogenesis of FCD [78] and that FCD, tuberous sclerosis and hemimegalencephaly represent a disease spectrum [73] of brain “overgrowth” disorders [80]. FCD can also coexist with other

111

Developmental Neuropathology

(a)

(b) (c)

(d)

MAP2

(e)

GFAP

(f)

NeuN

(g)

NeuN

Figure 10.6 Focal cortical dysplasia type IIID. (a) Cortical resection in an adult with epilepsy with a known hypoxic ischemic cortical damage occurring in the perinatal period. (b) Coronal slices confirm a “ulegyric” pattern of atrophy to some gyri. (c) High field magnetic resonance imaging (MRI) with a 9.4T MRI shows the lamination of the cortex in more detail than conventional MRI, as well as regions with disrupted lamination (arrow). (d) Neuronal markers as MAP2 can highlight the abnormal cortex adjacent to the main lesion/infarct; in this case, illustrating the

thick cortical layer with a corrugated interface between gray and white matter – ´ or post-ischemic marbled cortex. (e) GFAP stain also the so-called etat crible, highlights patchy islands of gliosis in the cortex at this region and islands of gray matter. (f, g) NeuN showing a regions of more normal laminar cortex in the margins (f) compared with abnormal regions harboring FCDIIID in (g). Original magnifications (d, e) ×1 and (f, g) ×2.5.

common epilepsy gene mutations, like the SCN1A gene, which may represent “double-hit” or susceptibility factors [81]. The pathogenesis of FCDIII is considered to be different from isolated FCD, being acquired rather than primarily genetic, and arising at later developmental stages, and affecting postnatal cortical maturation, regeneration and reorganization. Its etiology is linked with that of the associated main pathology, and factors such as prematurity, hypoxic ischemic insult, febrile seizures, head trauma, and infection are potential pathogenetic causes for types IIIA and D [56]. The microcolumnar patterns in some FCDI and III may represent arrested maturation with persistence of developmental, vertically arranged pyramidal

cells and their bundled projections of axons and dendrites. In FCDIIIB, a preexisting cortical dysplasia on which a low-grade tumor arises still remains an unproven concept.

112

Animal models Cortical injury experimental models in the developing cortex (freeze injury, ionizing radiation, neurotoxins including methylazomethanol, and other exogenous insults) producing focal malformations with some similarities to human cortical dysplasia, have been traditionally used to study epileptogenesis. Currently, transgenic models incorporating identified gene mutations in human FCD and related lesions such as tuberous

Hippocampal Sclerosis, Granule Cell Dispersion, and Cortical Dysplasia Chapter 10

sclerosis, can recapitulate the clinical, pathological and molecular abnormalities, and are likely to be increasingly used and developed to explore epileptogenic mechanisms, as well as efficacy of novel treatment targets [82].

Surgical treatments and outcome A recent meta-analysis of 1190 operated cases with FCD/MCD of all types showed 56% achieved a good outcome following surgery [36]. Studies of operated FCDIIB report seizure-free outcomes as high as 75% [47] and 87.5% [83]. Postsurgical outcome in terms of seizure control is likely to be influenced by many factors, including the dysplasia type, the extent or completeness of resection (of the epileptogenic zone and/or radiological lesion), the presence of a second pathology (e.g., hippocampal sclerosis) and the length of follow-up. Completeness of excision (as assessed by postoperative MRI or histological margins) is associated with a better outcome with excision of the cortical component more critical than the white matter. Clinical improvements have been reported for MRI-negative FCD [84], with equally good outcomes for surgery carried out in adulthood and childhood. The relative contribution of FCDIII compared with the main lesion to epileptogenicity and outcome following resection still remains to be clarified. There are studies that report worse outcome in patients with hippocampal sclerosis and FCDIIIA compared with those without [50,51], but not in all series [49]. In a 2013 study of tumor-associated FCDIIIB, the clinical outcome was not significantly different from that of isolated tumors [60].The significance of mild MCD variants in terms of independent epileptogenicity is unclear [41], but there is some evidence for better seizure-free outcomes when this pathology is present [65]. Future perspectives and conclusions The next decade is likely to see significant advances of the genetic basis and understanding of FCD types and the development of targeted medical treatments, including mTOR pathway inhibitors [56,82]. In addition, improved preoperative diagnosis of FCDI, III, and mild MCD (e.g., through high-field or postprocessing MRI technologies; Figure 10.5), as well as development of biomarkers, and a greater understanding of any cellular mechanisms contributing to epileptogenicity, will enable appropriate surgical management to maximize best outcomes [63].

References 1. Nelson PT, Smith CD, Abner EL, et al. (2013) Hippocampal sclerosis of aging, a prevalent and high-morbidity brain disease. Acta Neuropathol 126:161–77 2. Thom M (2009) Hippocampal sclerosis: progress since Sommer. Brain pathol 19:565–72

3. Cavanagh JB, Meyer A (1956) Aetiological aspects of Ammon’s horn sclerosis associated with temporal lobe epilepsy. Br Med J 2:1403–7 4. Houser CR (1990) Granule cell dispersion in the dentate gyrus of humans with temporal lobe epilepsy. Brain Res 535:195–204 5. Blumcke I, Thom M, Aronica E et al. (2013) International consensus classification of hippocampal sclerosis in temporal lobe epilepsy: A Task Force report from the ILAE Commission on Diagnostic Methods. Epilepsia 54:1315–29 6. Meencke HJ, Veith G, Lund S (1996) Bilateral hippocampal sclerosis and secondary epileptogenesis. Epilepsy Res Suppl 12:335–4 7. Novy J, Belluzo M, Caboclo LO et al. (2013) The lifelong course of chronic epilepsy: the Chalfont experience. Brain 136:3187–99 8. Semah F, Picot MC, Adam C et al. (1998) Is the underlying cause of epilepsy a major prognostic factor for recurrence? Neurology 51:1256–62 9. Steve TA, Jirsch JD, Gross DW (2014) Quantification of subfield pathology in hippocampal sclerosis: a systematic review and metaanalysis. Epilepsy Res 108:1279–85 10. Na M, Ge H, Shi C et al. (2015) Long-term seizure outcome for international consensus classification of hippocampal sclerosis: a survival analysis. Seizure 25:141–6 11. Coras R, Pauli J, Li J et al. (2014) Differential influence of hippocampal subfields to memory formation: insights from patients with temporal lobe epilepsy. Brain 137:1945–57 12. Thom M (2014) Review: Hippocampal sclerosis in epilepsy: a neuropathology review. Neuropathol Appl Neurobiol 40:520–43 13. Magloczky Z, Freund TF (2005) Impaired and repaired inhibitory circuits in the epileptic human hippocampus. Trends Neurosci 28:334–40 14. Blumcke I, Kistner I, Clusmann H et al. (2009) Towards a clinicopathological classification of granule cell dispersion in human mesial temporal lobe epilepsies. Acta Neuropathol 117:535–44 15. Martinian L, Catarino CB, Thompson P et al. (2012) Calbindin D28K expression in relation to granule cell dispersion, mossy fibre sprouting and memory impairment in hippocampal sclerosis: a surgical and post mortem series. Epilepsy Res 98:14–24 16. Lewis DV, Shinnar S, Hesdorffer DC et al. (2014) Hippocampal sclerosis after febrile status epilepticus: the FEBSTAT study. Ann Neurol 75:178–85 17. Salzmann A, Malafosse A (2012) Genetics of temporal lobe epilepsy: a review. Epilepsy Res Treat 2012: 863702 18. Busch RM, Floden D, lineweaver TT et al. (2011) Effect of apolipoprotein epsilon4 allele on hippocampal and brain volume in intractable temporal lobe epilepsy. Epilepsy Behav 21:88–90 19. Kasperaviciute D, Catarino CB, matarin M et al. (2013) Epilepsy, hippocampal sclerosis and febrile seizures linked by common genetic variation around SCN1A. Brain 136:3140–50 20. Miller-Delaney SF, Bryan K, Das S et al. (2015) Differential DNA methylation profiles of coding and non-coding genes define hippocampal sclerosis in human temporal lobe epilepsy. Brain 138:616–31 21. Kaalund SS, Veno MT, Bak M et al. (2014) Aberrant expression of miR-218 and miR-204 in human mesial temporal lobe epilepsy and hippocampal sclerosis-convergence on axonal guidance. Epilepsia 55:2017–27 22. Zucchini S, Marucci G, Paradiso B et al. (2014) Identification of miRNAs differentially expressed in human epilepsy with or without granule cell pathology. PLoS One 9:e105521

113

Developmental Neuropathology 23. Tsai MH, Pardoe HR, Perchyonok Y et al. (2013) Etiology of hippocampal sclerosis: evidence for a predisposing familial morphologic anomaly. Neurology 81:144–9 24. Bis JC, DeCarli C, Smith AV et al. (2012) Common variants at 12q14 and 12q24 are associated with hippocampal volume. Nat Genet 44:545–51 25. Stein JL, Medland SE, Vasquez AA et al. (2012) Identification of common variants associated with human hippocampal and intracranial volumes. Nat Genet 44:552–61 26. Kinney HC, Cryan JB, Haynes RL et al. (2015) Dentate gyrus abnormalities in sudden unexplained death in infants: morphological marker of underlying brain vulnerability. Acta Neuropathol 129:65– 80 27. Hester MS, Danzer SC (2013) Accumulation of abnormal adultgenerated hippocampal granule cells predicts seizure frequency and severity. J Neurosci 33:8926–36 28. Haas CA, Frotscher, M (2010) Reelin deficiency causes granule cell dispersion in epilepsy. Exp Brain Res 200:141–9 29. Coras R, Siebzehnrubl FA, Pauli E et al. (2010) Low proliferation and differentiation capacities of adult hippocampal stem cells correlate with memory dysfunction in humans. Brain 133: 3359–72 30. Grone BP, Baraban SC (2015) Animal models in epilepsy research: legacies and new directions. Nat Neurosci 18:339–43 31. Levesque M, Avoli M, Bernard C (2016) Animal models of temporal lobe epilepsy following systemic chemoconvulsant administration. J Neurosci Methods 260:45–52 32. Sloviter RS, Bumanglag AV (2013) Defining “epileptogenesis” and identifying “antiepileptogenic targets” in animal models of acquired temporal lobe epilepsy is not as simple as it might seem. Neuropharmacology 69:3–15 33. Kovacs R, Heinemann U (2014) Models in research of pharmacoresistant epilepsy: present and future in development of antiepileptic drugs. Curr Med Chem 21:689–703 34. Bertram EH (2014) Extratemporal lobe circuits in temporal lobe epilepsy. Epilepsy Behav 38: 13–18 35. Jobst BC, Cascino GD (2015) Resective epilepsy surgery for drugresistant focal epilepsy: a review. JAMA 313:285–93 36. West S, Nolan SJ, Cotton J et al. (2015) Surgery for epilepsy. Cochrane Database Syst Rev (7):CD010541. doi: 10.1002/14651858.CD010541.pub2 37. de Tisi J, Bell GS, Peacock JL et al. (2011) The long-term outcome of adult epilepsy surgery, patterns of seizure remission, and relapse: a cohort study. Lancet 378:1388–95 38. Thom M, Liagkouras I, Elliot KJ et al. (2010) Reliability of patterns of hippocampal sclerosis as predictors of postsurgical outcome. Epilepsia 51:1801–8 39. da Costa Neves RS, Jardim AP, Caboclo LO et al. (2013) Granule cell dispersion is not a predictor of surgical outcome in temporal lobe epilepsy with mesial temporal sclerosis. Clin Neuropathol 32: 24–30 40. Taylor DC, Falconer MA, bruton CJ, Corselis JA (1971) Focal dysplasia of the cerebral cortex in epilepsy. J Neurol Neurosurg Psychiatry 34:369–87 41. Blumcke I, Thom M, Aronica E et al. (2011) The clinicopathologic spectrum of focal cortical dysplasias: a consensus classification proposed by an ad hoc Task Force of the ILAE Diagnostic Methods Commission. Epilepsia 52:158–74

114

42. Geschwind DH, Rakic P (2013) Cortical evolution: judge the brain by its cover. Neuron 80:633–47 43. Toma K, Hanashima C (2015) Switching modes in corticogenesis: mechanisms of neuronal subtype transitions and integration in the cerebral cortex. Front Neurosci 9:274 44. Sarnat HB, Flores-Sarnat L (2013) Radial microcolumnar cortical architecture: maturational arrest or cortical dysplasia? Pediatr Neurol 48:259–70 45. Blumcke I. (2009) Neuropathology of focal epilepsies: a critical review. Epilepsy Behav 15: 34–9 46. Harvey AS, Cross JH, Shinnar S et al. (2008) Defining the spectrum of international practice in pediatric epilepsy surgery patients. Epilepsia 49:146–55 47. Krsek P, Maton B, Korman B et al. (2008) Different features of histopathological subtypes of pediatric focal cortical dysplasia. Ann Neurol 63:758–69 48. Chamberlain WA, Cohen ML, Gyure KA et al. (2009) Interobserver and intraobserver reproducibility in focal cortical dysplasia (malformations of cortical development). Epilepsia 50:2593–8 49. Thom M, Eriksson S, Martinian L et al. (2009) Temporal lobe sclerosis associated with hippocampal sclerosis in temporal lobe epilepsy: neuropathological features. J Neuropathol Exp Neurol 68:928–38 50. Johnson AM, Sugo E, Barreto D et al. (2014) Clinicopathological associations in temporal lobe epilepsy patients utilising the current ILAE focal cortical dysplasia classification. Epilepsy Res 108:1345– 51 51. Deleo F, Garbelli R, Milesi G et al. (2016) Short- and long-term surgical outcomes of temporal lobe epilepsy associated with hippocampal sclerosis: Relationships with neuropathology. Epilepsia 57:306– 15 52. Thom M, Blumcke I, Aronica E (2012) Long-term epilepsyassociated tumors. Brain Pathol 22:350–79 53. Baulac S, Ishida S, Marsan E et al. (2015) Familial focal epilepsy with focal cortical dysplasia due to DEPDC5 mutations. Ann Neurol 77:675–83 54. Scerri T, Riseley JR, Gillies G et al. (2015) Familial cortical dysplasia type IIA caused by a germline mutation in DEPDC5. Ann Clin Transl Neurol 2:575–80 55. Scheffer IE, Heron SE, Regan BM et al. (2014) Mutations in mammalian target of rapamycin regulator DEPDC5 cause focal epilepsy with brain malformations. Ann Neurol 75:782–7 56. Marin-Valencia I, Guerrini R, Gleeson JG (2014) Pathogenetic mechanisms of focal cortical dysplasia. Epilepsia 55:970–8 57. Chen J, Tsai V, Parker WE et al. (2012) Detection of human papillomavirus in human focal cortical dysplasia type IIB. Ann Neurol 72:881–92 58. Coras R, Korn K, Bien CG et al. (2015) No evidence for human papillomavirus infection in focal cortical dysplasia IIb. Ann Neurol 77:312–19 59. Shapiro KA, McGuone D, Deshpande V et al. (2015) Failure to detect human papillomavirus in focal cortical dysplasia type IIb. Ann Neurol 78:63–7 60. Cossu M, Fuschillo D, Bramerio M et al. (2013) Epilepsy surgery of focal cortical dysplasia-associated tumors. Epilepsia 54 Suppl 9:115– 22 61. Pestana Knight EM, Gonzalez-Martinez J, Gupta A (2015) Preoperative evaluation in pediatric patients with cortical dysplasia. Childs Nerv Syst 31:2225–33

Hippocampal Sclerosis, Granule Cell Dispersion, and Cortical Dysplasia Chapter 10 62. Holthausen H, Pieper T, Winkler P et al. (2014) Electro-clinicalpathological correlations in focal cortical dysplasia (FCD) at young ages. Childs Nerv Syst 30:2015–26 63. Guerrini R, Duchowny M, Jayakar P et al. (2015) Diagnostic methods and treatment options for focal cortical dysplasia. Epilepsia 56:1669–86 64. Eriksson SH, Thom M, Symms MR et al. (2009) Cortical neuronal loss and hippocampal sclerosis are not detected by voxel-based morphometry in individual epilepsy surgery patients. Hum Brain Mapp 30: 3351–60 65. Liu JY, Ellis M, Brooke-Ball H et al. (2014) High-throughput, automated quantification of white matter neurons in mild malformation of cortical development in epilepsy. Acta Neuropathol Commun 2:72 66. Hildebrandt M, Pieper T, Winkler P et al. (2005) Neuropathological spectrum of cortical dysplasia in children with severe focal epilepsies. Acta Neuropathol 110:1–11 67. Krsek P, Pieper T, Karlmeier A et al. (2009) Different presurgical characteristics and seizure outcomes in children with focal cortical dysplasia type I or II. Epilepsia 50:125–37 68. Orlova KA, Tsai V, Baybis M et al. (2010) Early progenitor cell marker expression distinguishes type II from type I focal cortical dysplasias. J Neuropathol Exp Neurol 69:850–63 69. Yasin SA, Latak K, Becherini F et al. (2010) Balloon cells in human cortical dysplasia and tuberous sclerosis: isolation of a pathological progenitor-like cell. Acta Neuropathol 120:85–96 70. Riviere JB, Mirzaa GM, O’Roak BJ et al. (2012) De novo germline and postzygotic mutations in AKT3, PIK3R2 and PIK3CA cause a spectrum of related megalencephaly syndromes. Nat Genet 44:934– 40 71. Lee JH, Huynh M, Silhavy JL et al. (2012) De novo somatic mutations in components of the PI3K-AKT3-mTOR pathway cause hemimegalencephaly. Nat Genet 44:941–5 72. D’Gama AM, Geng Y, Couto JA et al. (2015) Mammalian target of rapamycin pathway mutations cause hemimegalencephaly and focal cortical dysplasia. Ann Neurol 77:720–5

73. Jansen LA, Mirzaa GM, Ishak GE et al. (2015) PI3K/AKT pathway mutations cause a spectrum of brain malformations from megalencephaly to focal cortical dysplasia. Brain 138:1613–28 74. Baybis M, Yu J, Lee A et al. (2004) mTOR cascade activation distinguishes tubers from focal cortical dysplasia. Ann Neurol 56:478–87 75. Crino PB (2011) mTOR: A pathogenic signaling pathway in developmental brain malformations. Trends Mol Med 17:734–42 76. Lim JS, Kim WI, Kang HC et al. (2015) Brain somatic mutations in MTOR cause focal cortical dysplasia type II leading to intractable epilepsy. Nat Med 21:395–400 77. Leventer RJ, Scerri, Marsh AP et al. (2015) Hemispheric cortical dysplasia secondary to a mosaic somatic mutation in MTOR. Neurology 84:2029–32 78. Nakashima M, Saitsu H, Takei N et al. (2015) Somatic Mutations in the MTOR gene cause focal cortical dysplasia type IIb. Ann Neurol 78:375–86 79. Conti V, Pantaleo M, Barba C et al. (2015) Focal dysplasia of the cerebral cortex and infantile spasms associated with somatic 1q21.1-q44 duplication including the AKT3 gene. Clin Genet 88:241–7 80. Hevner RF (2015) Brain overgrowth in disorders of RTK-PI3KAKT signaling: a mosaic of malformations. Semin Perinatol 39:36– 43 81. Barba C, Parrini E, Coras R et al. (2014) Co-occurring malformations of cortical development and SCN1A gene mutations. Epilepsia 55:1009–19 82. Wong M, Roper SN (2016) Genetic animal models of malformations of cortical development and epilepsy. J Neurosci Methods 260:73–82 83. Kim DW, Lee SK, Chu K et al. (2009) Predictors of surgical outcome and pathologic considerations in focal cortical dysplasia. Neurology 72:211–16 84. Chapman K, Wyllie E, Najm I et al. (2005) Seizure outcome after epilepsy surgery in patients with normal preoperative MRI. J Neurol Neurosurg Psychiatry 76:710–13

115

11

Tuberous Sclerosis Complex Shino D. Magaki1 and Harry V. Vinters1,2 1

Section of Neuropathology, Department of Pathology and Laboratory Medicine, David Geffen School of Medicine at UCLA, Los Angeles, CA, USA 2 Department of Neurology, David Geffen School of Medicine at UCLA and Ronald Reagan UCLA Medical Center, Los Angeles, CA, USA

Definition

Epidemiology

Tuberous sclerosis, more appropriately described as tuberous sclerosis complex (abbreviated as TSC rather than TS, the latter usually referring to Tourette syndrome) is an autosomal dominant, multisystem disorder or group of disorders in which the central nervous system (CNS), eyes, lungs, kidneys, skin, and heart are most commonly affected by malformative, hamartomatous or neoplastic lesions [1–10]. The characteristic brain abnormalities include neocortical tubers (which may be epileptogenic), subependymal nodules and subependymal giant cell astrocytomas (SEGAs), which may obstruct cerebrospinal fluid pathways or behave as space-occupying lesions.

Incidence and prevalence Tuberous sclerosis is estimated to occur in 1 in 6000 to 10 000 births [1,2,5]. Within the United States, 25 000–40 000 individuals are affected; worldwide, the number is between one and two million. Accurate estimates are difficult to ascertain because of the markedly variable phenotype and penetrance of the disorder (i.e., it may go unrecognized for many years or even decades), and cases with nonpenetrance have been described [12]. Even within families in which an affected individual is found, others in the same families may have mutations/deletions in the genes associated with tuberous sclerosis (TSC1, TSC2), but they may be asymptomatic or may show only subtle evidence of neuropsychiatric disease, such as a learning disability [13,14]. Investigating asymptomatic individuals in any family with a confirmed tuberous sclerosis proband may be revealing. Because the causal genes are now known, mutations within them can be looked for in individuals suspected of carrying the diagnosis, but this remains an extremely time-consuming, labor-intensive and expensive process available in relatively few centers. Precise estimates of incidence and prevalence of the disorder have also been difficult to obtain. Based upon studies of institutionalized patients, the incidence of tuberous sclerosis has been estimated to be 1 in 100–300, and prevalence of the disorder in the general population from 1 in 20 000 to 1 in 50 000 [5]. More accurate estimates of incidence and prevalence postdate the advent of modern neuroimaging techniques, first available in the mid- to late-1970s with the advent of computed tomography, which has allowed for the discovery of asymptomatic SEGAs and tubers in patients clinically suspected of having tuberous sclerosis or belonging to a family with tuberous sclerosis.

Synonyms and historical annotations Tuberous sclerosis was first described in 1880 and is still sometimes referred to using the name of the first investigator who definitively described its clinicopathologic features, Bourneville; within the past 25 years, especially since recognition of the two tuberous sclerosis genes, mutations in which determine the phenotype, there has been an explosion in our understanding of its cellular pathogenesis [1,2,5,11]. Indeed, its multiorgan involvement and polymorphous clinicopathologic manifestations have piqued the interest of a range of investigators, from neurobiologists intrigued by “excitable/epileptogenic” brain tissue to cancer researchers and cell biologists interested in the molecular mechanisms that control cell proliferation and differentiation.

Developmental Neuropathology, Second Edition. Edited by Homa Adle-Biassette, Brian N. Harding, and Jeffrey A. Golden. © 2018 John Wiley & Sons Ltd. Published 2018 by John Wiley & Sons Ltd.

117

Developmental Neuropathology

Sex and age distribution Tuberous sclerosis usually manifests in infancy (e.g., as infantile spasms) or childhood but may not present until adult life. It is an autosomal dominant disorder with no major sex predominance. Some manifestations of tuberous sclerosis, however, have a pronounced sex preponderance: lymphangioleiomyomatosis occurs almost exclusively in females [7]. Risk factors The only known risk factor for developing tuberous sclerosis is carrying a mutation on one of the two causal genes, TSC1 (on chromosome 9q34) or TSC2 (chromosome 16p13.3). Factors modifying expression of these genes are beginning to be understood (e.g. in carefully constructed animal models).

Embryology As pertains to the brain, the cortical tubers resemble severe sporadic cortical dysplasia, specifically cortical dysplasia designated in the International League Against Epilepsy (ILAE) classification, as type IIb [10,15–19]. Both disorders are assumed to represent malformations of cortical development (MCD) or “neuronal migration disorders.” These disorders reflect profound derangement in the orderly process that leads to population of the six-layered neocortex by neurons that have previously migrated, as neuronal precursors, from the germinal matrix during the early part of gestation [20]. A number of pathogenetic theories of cortical dysplasia have been proposed; Cepeda et al. [21] put forth the dysmature cerebral developmental hypothesis, which summarizes the embryologic and neurodevelopmental underpinnings of sporadic cortical dysplasia, many of which may apply to the evolution of cortical tubers. However, the neuropathologic features of a tuber suggest a much more fundamental lesion of neuroglial differentiation, growth, maturation, and development. SEGAs may evolve from subependymal nodules, which can be found within the tuberous sclerosis brain as early as 20 weeks of gestation [22], based upon serial neuroimaging studies [23].

Genetics The first years of the twenty-first century have witnessed an exponential growth in our understanding of the molecular genetics of tuberous sclerosis. This began with cloning the TSC2 gene on chromosome 16p13.3 in 1993 [24], followed approximately four years later by cloning TSC1 on chromosome 9q34 [25], both tasks accomplished by large multidisciplinary international groups working in collaboration and using shared tissue and blood resources. If another tuberous sclerosis gene exists, it accounts for a very small proportion of cases [11]. The TSC1 gene contains 23 exons (of which 21 carry coding information)

118

and generates an 8.6 kilobase (kb) transcript that encodes a protein, hamartin, containing 1164 amino acids. The TSC2 gene has 43 kb of genomic DNA containing 40 exons; it yields a 5.5 kb transcript that encodes a 1807 amino acid protein, tuberin, with molecular weight of approximately 180 kDa. However, up to two-thirds of patients with tuberous sclerosis have no family history of the disorder, and probably represent spontaneous mutations [1,5]. Families with tuberous sclerosis may show linkage to either of the two identified genes. TSC1 mutations (usually small deletions and nonsense mutations) account for a minority of mutations identified; they are slightly less common in patients with sporadic tuberous sclerosis, and are more common in familial cases (13–50%) [11]. If no pathogenic variant is found in either gene, somatic mosaicism for a pathogenic variant needs to be considered [8]. Soon after the tuberous sclerosis genes were isolated, probes were constructed to analyze localization of their transcripts within human tissues, and antibodies generated to synthetic peptides representing portions of tuberin and hamartin, then used in Western blot and immunohistochemical studies to show widespread expression of the genes in the brain and viscera [26–29]. Mutations within the TSC2 gene appear to be distributed throughout the gene, although possibly with an excess of missense mutations in the GTPase activating protein-related domain [1]. Mutations in TSC1 appear, for the most part, to cause premature protein truncation. Phenotype–genotype correlations have not been well established. TSC2 mutations appear to be more common than TSC1 mutations among patients with sporadic tuberous sclerosis. The uncommon occurrence of germline mosaicism appears to explain how asymptomatic parents may have multiple offspring with symptomatic tuberous sclerosis.

Clinical features Signs and symptoms The clinical presentation of an individual with tuberous sclerosis may be dramatic, in the form of infantile spasms, autism or mental restriction, although extra-CNS manifestations may also be the presenting feature [2,5]. The syndrome can present with cutaneous or visceral manifestations, before any neurologic problems have occurred or such lesions may come to prominence during investigation of a new-onset seizure disorder or the recognition of mental restriction. Approximately 85% of all patients with tuberous sclerosis who come to medical attention have experienced a seizure at some point during their course. Learning disabilities and cognitive impairment are common in patients with tuberous sclerosis [8]. Brain tumors are the leading cause of morbidity and mortality, while renal disease is the second leading cause of death. Of interest, the clinical manifestations of tuberous sclerosis arise in different time frames or “waves”; manifestations of cardiac rhabdomyomas, cortical tubers and subependymal nodules/SEGAs are likely to present

Tuberous Sclerosis Complex Chapter 11

Box 11.1 Diagnostic criteria for clinical diagnosis of tuberous sclerosis. Major features r Three or more hypomelanotic macules 5 mm or greater in diameter r Three or more angiofibromas or fibrous cephalic plaque r Two or more ungual fibromas r Shagreen patch r Multiple retinal hamartomas r Cortical dysplasias (including classical “tubers” and cerebral white matter “radial migration lines” by neuroimaging) r Subependymal nodules r Subependymal giant cell astrocytoma r Cardiac rhabdomyoma(s) r Lymphangioleiomyomatosis r Two or more angiomyolipomas (kidney) Minor features r Confetti skin lesions r Four or more dental enamel pits r Two or more intraoral fibromas r Retinal achromic patch r Multiple renal cysts r Non-renal hamartomas Definitive diagnosis requires two major features or one major feature with two or more minor features. A heterozygous pathogenic variant in either TSC1 or TSC2 by molecular genetic testing also provides “definitive diagnosis” r Possible diagnosis requires one major feature or two or more minor features

within the first weeks or months of life (even at birth), whereas facial lesions are often detected around the time of puberty, and renal cysts and/or angiomyolipomas commonly manifest at around the age of 20 years [9]. Lymphangioleiomyomatosis is a progressive pulmonary degeneration in which proliferation of abnormal smooth muscle-like cells occlude airways and lymphatics resulting in multiple cysts, most commonly in females and (in the absence of pulmonary transplantation) may be a cause of death in a given patient [7]. Because genetic analysis to confirm the diagnosis of tuberous sclerosis remains unavailable to most physicians caring for suspected families or family members, “diagnostic criteria” have been enunciated and have undergone significant modification over decades [30,31]. The current major and minor clinical criteria used to support the diagnosis of tuberous sclerosis are summarized in Box 11.1. These include primary or major features of the disease: facial angiofibromas, multiple subungual fibromas, one or more cortical tubers identified by characteristic neuroradiographic features, (histologically confirmed) subependymal nodules or SEGA, (radiographically confirmed) multiple calcified subependymal noduless protruding into the ventricular cavity, lymphangioleiomyomatosis, cardiac rhabdomymoma(s) and

multiple retinal hamartomas; minor (formerly called secondary) features include multiple renal cysts, intraoral fibroma(s), and “confetti skin lesions” [30]. A diagnosis of definite tuberous sclerosis is made when an individual has either one major feature with two or more minor features as listed, or two major features, or genetic evidence of a pathogenic mutation in either TSC1 or TSC2; possible tuberous sclerosis is diagnosed in the presence of either one major feature or two or more minor features. It is interesting to note that, with the refinement of neuroradiographic techniques, confirmatory histopathologic diagnosis of some criteria is no longer required.

Neuroimaging Neuroimaging has revolutionized our understanding of tuberous sclerosis to almost as great an extent as the discovery of the previously described tuberous sclerosis-associated genes. While ultrasound is of limited value in defining intraventricular lesions (subependymal nodules, SEGAs), they (and tubers) are most easily identified using computed tomography (CT) and magnetic resonance imaging (MRI) [5], both of which may further identify the ventriculomegaly and overall “paucity” of white matter that is commonly noted in patients with tuberous sclerosis. Calcified subependymal nodules and SEGAs are easily demonstrable by CT, but are also visible by MRI, some even show enhancement with gadolinium (Figure 11.1). T1and T2-weighted MRI images are especially useful for highlighting the presence of cortical and subcortical tubers, which

Figure 11.1 Subependymal giant cell astrocytoma. Coronal post-contrast T1-weighted magnetic resonance image demonstrates a large exophytic mass with avid enhancement in the right lateral ventricle near the foramen of Monro. Images courtesy of Dr. Noriko Salamon, Department of Radiological Sciences, UCLA Medical Center, California.

119

Developmental Neuropathology (a)

(b)

Figure 11.2 Multiple cortical tubers and subependymal nodules. Axial magnetic resonance imaging demonstrates multiple areas of subcortical hypointensity on T1-weighted image (a) and corresponding hyperintensity on T2/fluid attenuated inversion recovery sequences (b) including bilateral posterior parietal lobes (arrows). Multiple small subependymal nodules are also present (arrowheads). Images courtesy of Dr. Whitney B Pope, Department of Radiological Sciences, UCLA Medical Center, California.

are often not detectable with CT unless they are calcified; relatively low T1 signal is found in the subcortical region of some tubers detected by MRI (Figure 11.2). Fluid-attenuated inversion recovery imaging is of great help in discovering multiple small tubers. An observation that would appear to be intuitively obvious has been rigorously proven: the numbers of cortical tubers, assessed by MRI, appear to be a biomarker that quite accurately predicts severity of cerebral impairment in patients with tuberous sclerosis [32]. Sophisticated approaches to coregistering fluorodeoxyglucose positron emission tomography with MRI improves detection of cortical dysplasia in children with intractable epilepsy; this same coregistration used in conjunction with diffusion tensor imaging can distinguish epileptogenic tubers in the cerebral cortex of patients with tuberous sclerosis [33–35].

approaches [5]. An interesting neurobiologic issue in tuberous sclerosis is the question of precisely what features render a given neocortical tuber “epileptogenic.” Cardiac rhabdomyomas are often detected by echocardiography. Various renal lesions in patients with tuberous sclerosis, including cysts and angiomyolipomas, may occasionally lead to renal failure, with the expected concomitant abnormal biochemical parameters in blood. “Screening” for germline TSC1/TSC2 mutations is not currently feasible on a large scale, given the extreme variability of their location on either gene [5,11]. Guidelines for monitoring progression of tuberous sclerosis-associated lesions have been provided [8].

Laboratory findings As in any multisystem disorder, a “lesion” within a given tissue may lead to biochemical abnormalities reflective of injury to that tissue, or metabolic dysfunction within it. With respect to abnormal brain function, the most frequent presenting feature of tuberous sclerosis, electroencephalography and the newer technique of magnetoencephalography may be of value in characterizing the epileptogenic foci present in the brains of patients with tuberous sclerosis, as are high-resolution neuroimaging

The superficial cortical tubers characteristic of tuberous sclerosis are quite easily appreciated in the intact or fixed brain (of the rather infrequent cases of tuberous sclerosis that come to necropsy), often by palpation rather than inspection. When one gently runs a hand over the brain of an affected individual, irregularly spaced firm nodule-like regions are appreciated in the cortex. Inspection of the brain may show large, malformed and somewhat dysmorphic (sometimes “mushroom-shaped” gyri), often with umbilication [5,6,36]. Cut sections of the fixed brain

120

Macroscopy

Tuberous Sclerosis Complex Chapter 11

(a)

(b)

Figure 11.3 Cortical tuber and subependymal nodules. (a) Coronal section of right cerebral hemisphere from an autopsy of a patient with tuberous sclerosis showing blurring of the cortical–white matter junction (arrow) indicative of a tuber. (b) Multiple subependymal nodules with calcification are seen in both lateral ventricles (arrowheads), of the same patient.

show cortical tubers, subependymal nodules, and often SEGAs, well-demarcated tumors that protrude into the lateral ventricles. Tubers manifest as enlarged gyri in which the cortex– white matter junction has become blurred; again, the similarity to sporadic, severe cortical dysplasia is emphasized (Figure 11.3) [16,18]. A tuber, in particular its white matter component, may show poorly demarcated regions of encephalomalacia and/or gritty calcification, a feature which, in the authors’ experience, is almost never seen in sporadic cortical dysplasia. Subependymal nodules appear as smooth-surfaced, firm (because of calcification), round to oval or sausage-shaped elevations that project into the ventricular cavities, but may be quite superficially “planted” within the substance of the thalamic or caudate nuclei. Multiple such lesions may be present, and are sometimes described as “candle gutterings” on neuroimaging. Most such subependymal nodules are found along the thalamostriate sulcus of the lateral ventricle, although rarely they are discovered in the third or fourth ventricles (Figure 11.3). Large subependymal nodules, especially in close proximity to the foramina of Monro, may cause hydrocephalus [5,36]. Hemorrhage into a subependymal nodule is documented, although rare. Subependymal nodules often persist as small, asymptomatic nodules, but by sequential imaging it is clear that some evolve into SEGAs, lesions with a much more substantive growth potential and the possibility of behaving like more aggressive malignancies [5,23]. Most SEGAs are one centimeter or greater in size, frequently attached to the lateral ventricular wall at or near the foramen of Monro, and present as smooth surfaced, solitary, well-delineated masses, similar to subependymal nodules, only larger and more likely to become symptomatic. They do not diffusely infiltrate underlying brain, and have a much better prognosis than histologically similar lesions that infiltrate the deep white matter in patients without tuberous sclerosis.

Histopathology Cortical tubers have the histologic appearance of fantastically disorganized neocortex, in which bizarre neuroglial cells fight for prominence with gemistocyte-like balloon cells (Figure 11.4) [37]. This disorganization usually involves the subcortical white matter obscuring the border between gray and white matter. A tail of the tuber may traverse the cortical mantle, extending to the ventricular surface. Many cells within a tuber appear to have features of both neurons and astrocytes, suggesting a failure of commitment in neuroglial differentiation (Figure 11.5). While architectural disarray is a defining feature of a tuber, cellularity of a lesion may be extremely variable, although when high cell density is noted, a tuber may resemble a ganglioglioma [10,36]. Proliferative potential of these lesions, however, appears low, as judged by immunohistochemistry using primary antibodies to Ki-67. Tubers contain a rich variety of dysmorphic, markedly enlarged neurons, often with bizarre ramified processes (Figure 11.6). These processes, and frequently associated cytoskeletal disorganization and filamentous accumulations within the “neuronal” cytoplasm, can be highlighted by Nissl stains or silver impregnation techniques (e.g., the modified Bielschowsky technique), or immunohistochemistry using primary antibodies to various cytoskeletal components. Large, bizarre astrocytes can be dramatically highlighted using the Golgi–Cox technique [38]. Balloon cells resembling gemistocytic astrocytes have eccentric nuclei containing relatively coarse chromatin, and glassy eosinophilic cytoplasm which is labeled with variable intensity (even among cells within a tuber) using primary antibodies to glial fibrillary acidic protein (GFAP; Figure 11.5) [5,10]. Balloon cells may cluster together and be prominent within subcortical white matter, or be scattered among the dysmorphic neurons. Tubers may show prominent punctate calcification within their center or at their periphery. Cells with vacuolated cytoplasm may

121

Developmental Neuropathology

(a)

(b)

(c)

(d)

Figure 11.4 Cortical tuber. (a) Neuronal disorganization can be appreciated at low magnification. (b) Highlighted with NeuN immunohistochemistry. (c) Enlarged cells with abundant glassy eosinophilic cytoplasm, “balloon cells” (arrow), admixed with bizarre dysmorphic appearing neurons (arrowheads), which show variable staining with NeuN (d).

be present in or adjacent to tubers (Figure 11.7). Although tubers are thought to be inherently “epileptogenic,” careful autopsy studies [39] have shown that the brain tissue between identifiable tubers often shows architectural disorganization and cytologic abnormalities similar to those noted within tubers; almost certainly, these significant cytoarchitectural abnormalities contribute to seizures. Subependymal nodules, which may be densely calcified, include plump cells with eosinophilic cytoplasm in a fibrillary vascular stroma [5,6,36]. The fibrillary stroma may be of variable density, while microvessels are often hyalinized. The overlying ependyma is usually intact. Giant neuroglial cells, similar to those found in tubers and sometimes containing large, multiple or convoluted nuclei, may be noted within subependymal nodules, often at their centers. SEGAs, some of which may arise from subependymal nodules, also resemble them histologically. Neuroradiologists

122

sometimes use a size criterion to differentiate subependymal nodules from SEGAs, the latter being defined as larger than one centimeter in maximal dimension. SEGAs often contain spindle-shaped cells, sometimes apparently “sweeping” from microvessels in a fashion resembling pseudorosettes, and even “true” perivascular pseudorosettes may be noted; other elements within SEGA simply resemble gemistocytic astrocytes (Figure 11.8) [5,40]. These heterogeneous tumors may also contain ganglion-like cells with prominent nucleoli, although lacking Nissl substance (note the similarity to “undifferentiated neuroglial cells” in tubers), and collections of epithelioid cells. Mitoses are variable within SEGAs; although often completely absent, they do not seem to impart a more malignant “phenotype” or worse prognosis. The same applies to the presence of vascular endothelial hyperplasia and necrosis, which are encountered in no more than 5–15% of SEGAs. A review of 15 cases of surgically resected SEGAs [41], operated upon at ages

Tuberous Sclerosis Complex Chapter 11

(a)

(b)

(c)

(d)

Figure 11.5 Cortical tuber from a young child with tuberous sclerosis and infantile spasms. (a) Numerous balloon cells with eccentric nuclei, coarse chromatin, occasional prominent nucleoli, and eosinophilic cytoplasm that are variably positive for glial fibrillary acidic protein (b). Dysmorphic neurons with abnormal processes highlighted by neurofilament staining (arrows, c) and membranous halo of synaptophysin immunoreactivity (d).

Figure 11.6 Cortical tuber. Dysmorphic, markedly enlarged neurons, with abnormally dispersed Nissl substance and bizarre ramified processes.

123

Developmental Neuropathology (a)

(b)

Figure 11.7 Cortical tuber from a young child with tuberous sclerosis, intractable epilepsy and developmental delay. (a,b) Balloon cells with vacuolated cytoplasm, focally in small clusters (a). (a)

(b)

(c)

(d)

Figure 11.8 Surgically resected SEGA. (a) SEGAs are well demarcated from adjacent brain (arrows) and are composed of large cells and often perivascular pseudorosettes (inset). (b) The polygonal cells resemble gemistocytic astrocytes, as well as ganglion cells with their coarse chromatin, prominent nucleoli, and glassy eosinophilic cytoplasm. (c) Calcifications, prominent in this case can also be seen. (d) The tumour cells are variably positive for glial fibrilaary acidic protein.

124

Tuberous Sclerosis Complex Chapter 11

55–228 months, found that many of the patients had a long seizure history (probably not caused by the SEGAs), and variable size of the tumors (8–46 mm). The Ki-67 labeling index in these neoplasms was quite variable (2.1–14.1%), reflecting our experience. In addition to tubers, subependymal nodules and SEGAs, tuberous sclerosis brains may contain heterotopic neuroglial elements within the deep subcortical white matter, frequently as small clusters of balloon cells, which can be easily missed on low power examination of a brain section.

Immunohistochemistry findings Immunohistochemical study of tuberous sclerosis-associated brain lesions has taken one of two major approaches: i) examination of cytoskeletal and differentiation markers, in an attempt to evaluate whether component cells are predominantly neuronal, glial or of “mixed/indeterminate” lineage; and ii) since 1993, studies using primary antibodies directed against different amino acid sequences or “segments” of the tuberin and hamartin proteins. Not surprisingly, since neuroglial elements within tuberous sclerosis brain lesions have neuronal, astrocytic and indeterminate features, cells within both tubers and SEGAs can be effectively immunolabeled with primary antibodies to both astrocytic and neuronal (including cytoskeletal) epitopes (Figure 11.8) [10,36,42]. In some studies, component cells of SEGAs show variable immunoreactivity for GFAP but more consistent positivity for S-100 protein [42], while others find a complete absence of GFAP immunoreactivity within SEGAs, but brisk GFAP immunoreactivity within subependymal nodules [43], a somewhat puzzling observation if indeed the latter lesion gives rise to the former. When astrocytic markers are detected within SEGAs, smaller spindle-shaped cells within the tumor are more likely to be GFAP immunoreactive than its large epithelioid elements [5,44]. Tubers often show prominent GFAP immunoreactivity, indicating an abundance of astrocytes, but balloon cells that so closely resemble gemistocytes are themselves variably positive (Figure 11.5). The neuron-specific enolase protein (14-3-2) shows prominent localization within giant cells of tubers, rare immunoreactivity within subependymal nodules and prominent immunolabelling within SEGAs [45]. The presence of a major neuronal component within SEGAs is further supported by the presence within them of cells strongly immunoreactive for neurofilament and synaptophysin [44]. Whereas cortical tubers demonstrate reduced immunoreactivity for the synaptic protein synapsin 1 [46], giant cells within tuberous sclerosis tubers show dramatic membranous halos of synaptophysin immunoreactivity resembling those noted in gangliogliomas, as well as strong immunostaining with antibodies to the microtubule-associated protein 2 [47]. Alpha B-crystallin, a member of the heatshock protein family of peptides, is found in abundance within “dysgenetic” cells

of tubers, and within both SEGAs and subependymal nodules [48]. Within months of the discovery of the TSC2 then TSC1 genes, probes to look for TSC1/2 transcripts had been developed, as had many antibodies to segments of the predicted proteins tuberin and hamartin. These reagents were used to confirm site-specific localization of the peptides within cultured cells, normal viscera and CNS in both animal and human tissue, as well as tuberous sclerosis lesions, using both Western blotting and immunohistochemical protocols. Using in situ hybridization with a digoxigenin-labeled cDNA probe, TSC2 mRNA was found to be widely expressed in various cell types throughout the body, including epithelia, lymphocytes and endocrine organs; within the CNS, it was prominently and selectively expressed within neurons, especially motor neurons, including cortical, brainstem and spinal cord neurons [49]. Widespread expression of the TSC2 gene within developing and adult nervous system was noted in another study using reverse transcription polymerase chain reaction, Northern blot and in situ hybridization analysis [50]. The results of a study in mice showed that tuberin localized to the perinuclear region of cerebellar Purkinje cells, whereas hamartin was distributed along neuronal or astrocytic processes [51]. In vitro, hamartin was highly expressed in astrocytes of mixed glial–neuronal cultures. Screening a large “library” of widely used cell culture lines, Catania et al. [52] found robust hamartin expression in virtually all cell lines examined, including those derived from primary brain neoplasms. Using co-immnoprecipitation, this group confirmed physical interaction of tuberin and hamartin in a diverse group of human and rat cell types in culture. This observation, together with the often repeated finding that hamartin and tuberin are expressed in similar cell types throughout the body, supported the notion that they may interact in normal cellular processes, and furthermore that impairment of this interaction may be of importance in TSC pathogenesis (53). Making use of human autopsy and biopsy material, TSC2 mRNA and tuberin were found in abundance in many CNS cell types, including neurons and ependymal cells [28]. Dysmorphic cytomegalic neurons express high levels of tuberin, as do individual cells within SEGAs, subependymal nodules and sporadic cortical dysplasia. By immunohistochemistry, TSC1 and TSC2 gene products co-localize within tubers and sometimes within individual dysmorphic cells of patients with tuberous sclerosis [26], a curious observation given that, by definition, affected patients have a mutation in one or the other gene. Slightly divergent patterns of hamartin and tuberin expression are noted in some organs and tissues (e.g., distal nephron of the kidney, endocrine pancreas), suggesting that hamartin has a discrete and specialized function in some cell types [27]. In a developmental time frame, tuberin appears to be present in most neuronal populations of the CNS from at least 20 weeks of gestation, with an apparent upregulation of its expression after 40 weeks of gestation [10]. Hamartin was found, albeit with a weaker signal,

125

Developmental Neuropathology in the same cell types during CNS development. Unfortunately, hamartin and tuberin antibodies cannot be used to “confirm” the diagnosis of tuberous sclerosis using a tissue specimen obtained from a patient. Immunohistochemistry has also been used to examine potential patterns of interaction between hamartin and tuberin, and other molecules that may be of major importance in tuberous sclerosis pathogenesis. For example, the ERM (ezrin, radixin, moesin) group of proteins belong to the band-4.1 superfamily of membrane-cytoskeleton-linking proteins, which bind to the actin cytoskeleton, and furthermore link to ERM-binding membrane proteins. Abnormal interactions between hamartin and ERM proteins may play a role in tuberous sclerosis pathogenesis, especially because inhibition of hamartin function in cells containing focal adhesions causes loss of adhesion to the cell substrate; by contrast, overexpression of hamartin in cells lacking focal adhesions results in activation of the GTP-binding protein Rho, assembly of actin stress fibers and formation of focal adhesions [54]. These observations suggest a plausible mechanism whereby disruption of adhesion to the cell matrix mediated by hamartin loss may contribute to the formation of hamartomas. In tuberous sclerosis tubers, both ezrin and moesin appear to be upregulated and are co-localized within a population of abnormal neuroglial cells [55]. Tissue microarray methodology has been used to screen for abnormal expression of molecules in the insulin-signaling pathway, which may explain some of the cytopathology characteristic of tuberous sclerosis tubers, as well as sporadic cortical dysplasia [56]. Analysis of surgical and autopsy specimens of tuberous sclerosis cortical tubers by deep sequencing of TSC1, TSC2 and KRAS has demonstrated that small “second-hit” mutations in these genes are rare events [57].

Ultrastructural findings Ultrastructural features of tuberous sclerosis lesions are interesting, but not especially informative in terms of providing clues to disease pathogenesis. Electron microscopic study is not essential to confirming the diagnosis of either SEGA or a tuber, and is rarely performed except to satisfy curiosity about possibly novel ultrastructural features within these lesions. Balloon cells within SEGAs show an abundance of cytoplasmic intermediate filaments and dense bodies resembling lysosomes [58]. Large collections of mitochondria may almost “fill” the cytoplasm of cells within a SEGA [59]. Crystalloids with a lamellar substructure, possibly originating from lysosomes, may be seen within tubers. An abundance of lysosomes is also noted in cells within SEGAs, as are swollen mitochondria, Golgi complexes, rough and smooth endoplasmic reticulum, and sparse intermediate filaments [42]. Megamitochondria have been noted within some glial processes, as have rare junctional complexes suggestive of aberrant synapse formation [60].

126

Differential diagnosis Tuberous sclerosis represents a distinctive “neurocutaneous syndrome” with characteristic clinical and genetic features [5,11]. It must be in the differential diagnosis in any infant or child who presents with infantile spasms, new-onset epilepsy, or mental restriction with a learning disability. The major neuropathologic “mimic” of tuberous sclerosis tubers is sporadic cortical dysplasia, a lesion that may, in its most extreme form, appear identical to a tuber in virtually all respects [16,18,20]. Conversely, any patient in whom severe cortical dysplasia is discovered in a cortical resection specimen must be carefully examined for the possible diagnosis of tuberous sclerosis; given the high rate of spontaneous mutation, absence of a family history does not exclude the diagnosis.

Experimental models Several animal models for tuberous sclerosis demonstrate varying degrees of neuropathologic abnormality and epilepsy. The Eker rat carries a spontaneous germline heterozygous Tsc2 mutation resulting from the insertion of an approximately 5-kb DNA fragment into the Tsc2 gene, with aberrant RNA expression from the mutant allele [61–63]. This model develops renal tumors with high frequency, but demonstrates a very mild neurologic phenotype with autistic-like behavior, no spontaneous seizures, and, only rarely, lesions resembling cortical tubers, subcortical hamartomas, subependymal hamartomas or anaplastic gangliogliomas [61,63–65]. All of these CNS lesions show variable immunoreactivity for nonphosphorylated neurofilament, neuron-specific enolase, and GFAP, and an absence of immunohistochemically demonstrable tuberin [64]. Homozygous mutation of Tsc2 in the Eker rat results in embryonic lethality in midgestation, characterized by markedly disrupted neuroepithelial development, with dysraphism and papillary overgrowth of the neuroepithelium [66]. Heterozygous Tsc1 ± and Tsc2 ± mice demonstrate neurocognitive defects but no seizures, except only during early postnatal life in Tsc1 ± mice, and no detectable neuropathologic lesions [61,63,67]. Although conventional homozygous Tsc knockout mice are embryonic lethal, homozygous inactivation of Tsc1 or Tsc2 in specific neural cell populations such as neurons, astrocytes, and neuroglial progenitor cells results in mice with neurologic abnormalities and epilepsy (63). Many of these TSC mice models demonstrate megalencephaly, cellular hypertrophy, and proliferation due to hyperactivation of the mammalian target of rapamycin (mTOR) pathway (Figure 11.9), and mTOR inhibitors have been shown to reverse some pathologic changes as well as prevent seizures (63). Loss of the Tsc1 gene in subventricular zone neural stem cells leads to development of subependymal nodule- and SEGA-like structural abnormalities

Tuberous Sclerosis Complex Chapter 11 (a)

(b)

(c)

(d)

Figure 11.9 Tuberous sclerosis mouse model. Coronal haemotoxylin and eosin stained section of brain from a tuberous sclerosis mouse model demonstrating megalencephaly (a) and neuronal disorganization, highlighted with NeuN immunohistochemistry (c), compared with a control mouse (b, d).

in the lateral ventricle, although seizures do not occur [61,68]. Mice with genetically-altered neural progenitor cells that give rise to neurons and astrocytes demonstrate lesions resembling cortical tubers, with vacuolated giant cells; they exhibit epilepsy, suggesting that not only neurons but astrocytes play a significant role in seizures associated with tuberous sclerosis [61,69,70]. Conditional inactivation of Tsc1 and Tsc2 in GFAP-positive cells in mice shows that Tsc2 inactivation causes more severe epilepsy and neuropathologic abnormalities including megalencephaly, neuronal disorganization, and astrocytic proliferation as well as greater mTOR hyperactivation compared to Tsc1 inactivation [71]. These models have helped elucidate the complexity of tuberous sclerosis pathogenesis, but a significant limitation is that most fail to recapitulate focal lesions of cortical tubers, likely due to gene inactivation of extensive areas of the brain in conventional knockout mice and the rodent brain being naturally lissencephalic and not conducive to the development of localized lesions [63].

Pathogenesis The plethora of neurologic and extra-CNS abnormalities in patients with tuberous sclerosis suggest that the TSC1/2 gene products have diverse effects on cellular physiology, replicative ability, growth, and differentiation; experimental results have tended to confirm this suspicion, and continue to challenge investigators who seek to explain the myriad clinicopathologic manifestations of TSC1/2 mutations. Given the defining hamartomas and low-grade neoplasms in patients with tuberous sclerosis, hamartin and tuberin have been hypothesized to be growth suppressors. Within the cell cytoplasm, hamartin and tuberin form a complex; mutations in either TSC1 or TSC2 lead to dysfunction of this complex. The complex of tuberin and hamartin is a “signaling node” that integrates growth factor and stress response signals from the upstream P13K/AKT pathway and then transmits signals downstream to coordinate

127

Developmental Neuropathology cellular processes such as cell size and proliferation [6,72]. Tuberin and hamartin have important regulatory effects on the mTOR signaling pathway, and this has led directly to therapies that modulate mTOR signaling [6,8,73,74]. The lesions that characterize the disease are thought to result from a “two-hit” mechanism, whereby one defective tuberous sclerosis gene is inherited, and the remaining normal allele lost in the course of evolution of CNS tumors/hamartomas, a pathway initially proposed by Knudson in 1971 based upon his studies in the Eker rat [11,62,75–77]. Western blot and immunohistochemical studies of tissues from some patients with tuberous sclerosis suggest loss of both hamartin and tuberin from cerebrum, kidney and myocardium [78]. Tuberin is variably reported to be absent from SEGAs, or present within a subset of their component cells [28,75,79,80]. Tuberin and hamartin have been found to associate physically in vivo, and it has been suggested that tuberin is a cytosolic chaperone for hamartin, preventing hamartin self-aggregation [81,82]. Influences of both hamartin and tuberin on molecules that affect the cell cycle continue to be a major focus of investigation. Immunohistochemical markers of cellular proliferation (e.g., collapsin response mediator protein 4, CRMP4, doublecortin, DCX) are expressed within giant cells of tubers and SEGAs (human-derived material) and subependymal nodules of Eker rats, suggesting that they may represent newly-generated cells that have migrated into tubers from the subventricular zone [83]. Previous studies looking at single cells within tuberous sclerosis lesions suggested that the presence of immature phenotypic markers (protein and mRNAs) within tubers indicated disruption of cell-cycle regulation and neuronal maturation within the tuberous sclerosis brain during development of the neocortex [84]. Drosophila experiments indicate that coexpression of Tsc1 and Tsc2 restricts tissue growth and reduces cell size and proliferation, whereas inactivation of either gene causes similar phenotypes, consisting of enhanced tissue growth and increased cell size with no change in ploidy [85]. A fruit fly homolog of the TSC2 gene, gigas, appears to regulate the cell cycle [86]. In other experiments in the fly, Drosophila Tsc1 was shown to function with Tsc2 to antagonize insulin signaling in the regulation of cell growth, proliferation, and organ size [87,88]. Mention is made above of how tissue microarrays constructed from surgically resected human tuberous sclerosis lesions (especially tubers) are being used to screen for tuberin/hamartin-mediated “lesions” in the insulin signaling pathway, which is critical to cell growth and function [56,72,73,89,90]. Hamartin appears to negatively regulate cell proliferation, and interacts with tuberin at every phase of the cell cycle in HeLa cells [91]. TSC2-negative fibroblasts show inactivation of the cyclindependent kinase inhibitor p27 [92]. Tissue culture experiments in various cell types, using both confocal microscopy and coimmunoprecipitation, show that both hamartin and tuberin interact with the G2/M cyclin-dependent kinase CDK1 [93]. It has further been suggested that hamartin and tuberin have separable functions in mammalian cell cycle regulation [94].

128

Biochemically, these diverse effects of hamartin/tuberin on the cell cycle may be explained by the observation that tuberin contains a hydrophobic N-terminal domain and a conserved C-terminal region that exhibits homology to the catalytic domain of a GTPase-activating protein for Rap1, which in turn may regulate DNA synthesis and cell cycle transition [1,5]. Interesting observations of possible relevance to the pathogenesis of CNS lesions of tuberous sclerosis continue to emerge from morphoanatomical observations. Park et al. [22] have shown, in a 20-week gestation fetus, that lesions suggestive of early cortical tubers and subependymal nodules are clearly identifiable even at this early stage of brain development. Subependymal radial glia expressed both vimentin and GFAP by immunohistochemistry, but subpial radial glia failed to express these markers. This suggests that abnormalities of these “guide wires,” by which germinal matrix cells normally migrate to the neocortex, may explain some of the neuronal migration anomalies noted in the tuberous sclerosis brain. Neuroinflammation (defined broadly to include the activation of microglia and astrocytes) has been suggested to have a role in the evolution of “epileptogenic” tubers; however, it has been difficult to separate intrinsic structural abnormalities from the part seizures themselves may play in activating these cells [95,96]. Novel microRNAs have been observed within tubers [97].

Future directions and therapy The treatment of the myriad complications of tuberous sclerosis is now largely symptomatic, but a major emerging theme is the use of mTOR inhibitors (e.g., everolimus), which effectively decrease the size of tuberous sclerosis-associated lesions (e.g., SEGAs, renal tumors); when treatment is discontinued, these lesions increase in size [8,74,98]. Seizures are treated with antiepileptic medications. Vigabatrin, an inhibitor of gammaaminobutyric acid, is especially effective for treating infantile spasms caused by tuberous sclerosis [5]. Increasingly, intractable seizures that can be attributed to a specific cortical tuber are amenable to treatment by “lesionectomy” or “tuberectomy,” just as corticectomy can be used to remove foci of cortical dysplasia in the treatment of pediatric seizures [17]. This affords the neuropathologist an opportunity to help in providing unique insights into “structure–function” correlations within epileptogenic brain tissue, especially if a skilled electrophysiologist is also interested in studying the resected brain fragment [17,99,100]. For example, Cepeda et al. [101] have shown subtle differences in spontaneous synaptic activity between cortical dysplasia (ILAE type II) and tuberous sclerosis tubers, when these are examined immediately after surgical removal. SEGAs frequently come to the attention of a neurosurgeon when hydrocephalus results from occlusion of the foramen of Monro. Cardiac and renal complications of tuberous sclerosis are treated surgically or medically.

Tuberous Sclerosis Complex Chapter 11

Is gene therapy a likely treatment for tuberous sclerosis in the future? Many of the neurobiologic events important in formation of the cerebral neocortex are complete by the third trimester of pregnancy. Arguably, therefore, many of the symptomatic lesions of tuberous sclerosis within the CNS have already formed in utero, rendering gene therapy ineffectual, unless it could theoretically be applied to the fertilized egg or the very early embryo.

References 1. Crino PB, Henske EP (1999) New developments in the neurobiology of the tuberous sclerosis complex. Neurology 53:1384–90 2. Crino PB, Nathanson KL, Henske EP (2006) The tuberous sclerosis complex. N Engl J Med 355:1345–56. 3. di Mario FJ Jr, Sahin M, Ebrahimi-Fakhari D (2015) Tuberous sclerosis complex. Pediatr Clin N Am 62:633–48 4. Kwiatkowski DJ, Short MP (1994) Tuberous sclerosis. Arch Dermatol 130:348–54 5. Kwiatkowski DJ, Whittemore VH, Thiele EA, eds (2010) In: Tuberous Sclerosis Complex: Genes, Clinical Features, and Therapeutics, Weinheim, Germany, Wiley-VCH Verlag, p. 409 6. Lopes MBS, Wiestler OD, Stemmer-Rachamimov AO et al. (2016) Tuberous sclerosis. In: Louis DN, Ohgaki H, Wiestler OD, Cavenee WK, eds, WHO Classification of Tumours of the Central Nervous System, rev. 4th ed., Lyon, France, International Agency for Research on Cancer, pp. 306–9 7. Moir LM (2016) Lymphangioleiomyomatosis: current understanding and potential treatments. Pharmacol Ther 158:114–24 8. Northrup H, Koenig MK, Pearson DA, Au K-S (1999, updated 2015) Tuberous sclerosis complex. In: Pagon RA, Adam MP, Ardinger HH, et al., eds, GeneReviews Seattle, WA, University of Washington 9. Short MP, Richardson EP Jr, Haines JL, Kwiatkowski DJ (1995) Clinical, neuropathological and genetic aspects of the tuberous sclerosis complex. Brain Pathol 5:173–9 10. Vinters HV, Park SH, Johnson MW et al. (1999) Cortical dysplasia, genetic abnormalities and neurocutaneous syndromes. Dev Neurosci 21:248–59 11. MacCollin M, Kwiatkowski D (2001) Molecular genetic aspects of the phakomatoses: tuberous sclerosis complex and neurofibromatosis 1. Curr Opin Neurol 14:163–9 12. Webb DW, Osborne JP (1991) Non-penetrance in tuberous sclerosis. J Med Genet 28:417–9 13. Northrup H, Wheless JW, Bertin TK, Lewis RA (1993) Variability of expression in tuberous sclerosis. J Med Genet 30:41–3 14. Smalley SL, Burger F, Smith M (1994) Phenotypic variation of tuberous sclerosis in a single extended kindred. J Med Genet 31:761–5 15. Blumcke I, Thom M, Aronica E et al. (2011) The clinicopathologic spectrum of focal cortical dysplasias: a consensus classification proposed by an ad hoc Task Force of the ILAE Diagnostic Methods Commission. Epilepsia 52:158–74 16. Crino PB, Miyata H, Vinters HV (2002) Neurodevelopmental disorders as a cause of seizures: Neuropathologic, genetic, and mechanistic considerations. Brain Pathol 12:212–33

17. Mathern GW, Giza CC, Yudovin S et al. (1999) Postoperative seizure control and antiepileptic drug use in pediatric epilepsy surgery patients: the UCLA experience, 1986–1997. Epilepsia 40:1740–9 18. Mischel PS, Nguyen LP, Vinters HV (1995) Cerebral cortical dysplasia associated with pediatric epilepsy. Review of neuropathologic features and proposal for a grading system. J Neuropathol Exp Neurol 54:137–53 19. Vinters HV (2002) Histopathology of brain tissue from patients with infantile spasms. Int Rev Neurobiol 49:63–76. 20. Desikan RS, Barkovich AJ (2016) Malformations of cortical development. Ann Neurol 80:797–810 21. Cepeda C, Andre VM, Levine MS et al. (2006) Epileptogenesis in pediatric cortical dysplasia: the dysmature cerebral developmental hypothesis. Epilepsy Behav 9:219–35 22. Park S-H, Pepkowitz SH, Kerfoot C et al. (1997) Tuberous sclerosis in a 20-week gestation fetus: immunohistochemical study. Acta Neuropathol 94:180–6 23. Fujiwara S, Takaki T, Hikita T, Nishio S (1989) Subependymal giant-cell astrocytoma associated with tuberous sclerosis. Do subependymal nodules grow? Childs Nerv Syst 5:43–4 24. European Chromosome 16 Tuberous Sclerosis Consortium (1993) Identification and characterization of the tuberous sclerosis gene on chromosome 16. Cell 75:1305–15 25. van Slegtenhorst M, de Hoogt R, Hermans C et al. (1997) Identification of the tuberous sclerosis gene TSC1 on chromosome 9q34. Science 277:805–8 26. Johnson MW, Emelin JK, Park S-H, Vinters HV (1999) Colocalization of TSC1 and TSC2 gene products in tubers of patients with tuberous sclerosis. Brain Pathol 9:45–54 27. Johnson MW, Kerfoot C, Bushnell T et al. (2001) Hamartin and tuberin expression in human tissues. Mod Pathol 14: 202–10 28. Kerfoot C, Wienecke R, Menchine M et al. (1996) Localization of tuberous sclerosis 2 mRNA and its protein product tuberin in normal human brain and in cerebral lesions of patients with tuberous sclerosis. Brain Pathol 6:367–77 29. Vinters HV, Kerfoot C, Catania M et al. (1998) Tuberous sclerosisrelated gene expression in normal and dysplastic brain. Epilepsy Res 32:12–23 30. Northrup H, Krueger DA, on behalf of the International Tuberous Sclerosis Complex Consensus Group (2013) Tuberous Sclerosis Complex diagnostic criteria update: recommendations of the 2012 International Tuberous Sclerosis Complex Conensus Conference. Pediatr Neurol 49:243–54 31. Roach ES, Smith M, Huttenlocher P et al. (1992) Report of the Diagnostic Criteria Committee of the National Tuberous Sclerosis Association. J Child Neurol 7:221–4 32. Goodman M, Lamm SH, Engel A et al. (1997) Cortical tuber count: a biomarker indicating neurologic severity of tuberous sclerosis complex. J Child Neurol 12:85–90 33. Chandra PS, Salamon N, Huang J et al. (2006) FDG–PET/MRI coregistration and diffusion-tensor imaging distinguish epileptogenic tubers and cortex in patients with tuberous sclerosis complex: a preliminary report. Epilepsia 47:1543–9 34. Pellock JM, Nordli DR Jr, Sankar R, Wheless JW, eds (2017) Pellock’s Pediatric Epilepsy. Diagnosis and Therapy. 4th ed., New York, NY, Demos Medical, p. 1228

129

Developmental Neuropathology 35. Salamon N, Kung J, Shaw SJ et al. (2008). FDG-PET/MRI coregistration improves detection of cortical dysplasia in patients with epilepsy. Neurology 71:1594–601 36. Richardson EP Jr (1991) Pathology of tuberous sclerosis: neuropathologic aspects. Ann NY Acad Sci 615:128–39 37. Vinters HV, Miyata H (2006) Neuropathologic features of tuberous sclerosis. In: McLendon RE, Rosenblum MK, Bigner DD, eds Russell and Rubinstein’s Pathology of Tumors of the Nervous System, 7th ed., London, UK: Hodder Arnold, pp. 955–69 38. Nardelli E, De Benedictis G, La Stilla G, Nicolardi G (1986) Tuberous sclerosis: a neuropathological and immunohistochemical (PAP) study. Clin Neuropathol 5:261–6 39. Marcotte L, Aronica E, Baybis M, Crino PB (2012) Cytoarchitectural alterations are widespread in cerebral cortex in tuberous sclerosis complex. Acta Neuropathol 123:685–93 40. Shepherd CW, Scheithauer BW, Gomez MR et al. (1991) Subependymal giant cell astrocytoma: a clinical, pathological, and flow cytometric study. Neurosurgery 28:864–8 41. Cuccia V, Zuccaro G, Sosa F et al. (2003) Subependymal giant cell astrocytoma in children with tuberous sclerosis. Childs Nerv Syst 19:232–43 42. Nakamura Y, Becker LE (1983) Subependymal giant-cell tumor: astrocytic or neuronal? Acta Neuropathol 60:271–7 43. Stefansson K, Wollmann R (1980) Distribution of glial fibrillary acidic protein in central nervous system lesions of tuberous sclerosis. Acta Neuropathol 52:135–40 44. Kimura N, Watanabe M, Date F et al. (1997) HMB-45 and tuberin in hamartomas associated with tuberous sclerosis. Mod Pathol 10:952–9 45. Stefansson K, Wollmann R (1981) Distribution of the neuronal specific protein 14-3-2, in central nervous system lesions of tuberous sclerosis. Acta Neuropathol 53:113–17 46. Lippa CF, Pearson D, Smith TW (1993) Cortical tubers demonstrate reduced immunoreactivity for synapsin I. Acta Neuropathol 85:449–51 47. Yamanouchi H, Ho M, Jay V, Becker LE (1997) Giant cells in cortical tubers in tuberous sclerosis showing synaptophysin– immunoreactive halos. Brain & Development 19:21–4 48. Iwaki T, Tateishi J (1991) Immunohistochemical demonstration of AlphaB-crystallin in hamartomas of tuberous sclerosis. Am J Pathol 139:1303–8 49. Menchine M, Emelin JK, Mischel PS et al. (1996) Tissue and cell type-specific expression of the tuberous sclerosis-related gene, TSC2, in human tissues. Mod Pathol 9:1071–80 50. Geist RT, Gutmann DH (1995) The tuberous sclerosis 2 gene is expressed at high levels in the cerebellum and developing spinal cord. Cell Growth Differ 6:1477–83 51. Gutmann DH, Zhang Y, Hasbani J et al. (2000) Expression of the tuberous sclerosis complex gene products, hamartin and tuberin, in central nervous system tissues. Acta Neuropathol 99: 223–30 52. Catania MG, Johnson MW, Liau LM et al. (2001) Hamartin expression and interaction with tuberin in tumor cell lines and primary cultures. J Neurosci Res 63:276–83 53. Plank TL, Logginidou H, Klein-Szanto A, Henske EP (1999) The expression of hamartin, the product of the TSC1 gene, in normal human tissues and in TSC1- and TSC2-linked angiomyolipomas. Mod Pathol 12:539–45

130

54. Lamb RF, Roy C, Diefenbach TJ et al. (2000) The TSC1 tumor suppressor hamartin regulates cell adhesion through ERM proteins and the GTPase Rho. Nat Cell Biol 2:281–7 55. Johnson MW, Miyata H, Vinters HV (2002) Ezrin and moesin expression within the developing human cerebrum and tuberous sclerosis-associated cortical tubers. Acta Neuropathol 104:188–96 56. Miyata H, Chiang AC, Vinters HV (2004) Insulin signaling pathways in cortical dysplasia and TSC–tubers: tissue microarray analysis. Ann Neurol 56:510–19 57. Qin W, Chan JA, Vinters HV et al. (2010) Analysis of TSC cortical tubers by deep sequencing of TSC1, TSC2 and KRAS demonstrates that small second-hit mutations in these genes are rare events. Brain Pathol 20:1096–105 58. Crino PB, Mehta R, Vinters HV (2010) Pathogenesis of TSC in the brain. In: Kwiatkowski DJ, Whittemore VH, Thiele EA, eds, Tuberous Sclerosis Complex. Genes, Clinical Features, and Therapeutics, Weinheim, Germany, Wiley-Blackwell, pp. 161–85 59. de Chadarevian J-P, Hollenberg RD (1979) Subependymal giant cell tumor of tuberose sclerosis. A light and ultrastructural study. J Neuropathol Exp Neurol 38:419–33 60. Trombley IK, Mirra SS (1981) Ultrastructure of tuberous sclerosis: cortical tuber and subependymal tumor. Ann Neurol 9:174–81 61. Kirschstein T, K¨ohling R (2016) Animal models of tumourassociated epilepsy. J Neurosci Methods 260:109–17 62. Kobayashi T, Hirayama Y, Kobayashi E et al. (1995) A germline insertion in the tuberous sclerosis (Tsc2) gene gives rise to the Eker rat model of dominantly inherited cancer. Nat Genet 9:70–4 63. Wong M, Roper SN (2016) Genetic animal models of malformations of cortical development and epilepsy. J Neurosci Methods 260:73–82 64. Mizuguchi M, Takashima S, Yamanouchi H et al. (2000) Novel cerebral lesions in the Eker rat model of tuberous sclerosis: cortical tuber and anaplastic ganglioglioma. J Neuropathol Exp Neurol 59:188–96 65. Yeung RS, Katsetos CD, Klein-Szanto A (1997) Subependymal astrocytic hamartomas in the Eker rat model of tuberous sclerosis. Am J Pathol 151:1477–86 66. Rennebeck G, Kleymenova EV, Anderson R et al. (1998) Loss of function of the tuberous sclerosis 2 tumor suppressor gene results in embryonic lethality characterized by disrupted neuroepithelial growth and development. Proc Natl Acad Sci U S A 95:15629–34 67. Lozovaya N, Gataullina S, Tsintsadze T et al. (2014) Selective suppression of excessive GluN2C expression rescues early epilepsy in a tuberous sclerosis murine model. Nat Commun 5:4563 68. Zhou J, Shrikhande G, Xu J et al. (2011) Tsc1 mutant neural stem/progenitor cells exhibit migration deficits and give rise to subependymal lesions in the lateral ventricle. Genes Dev 25:1595– 600 69. Goto J, Talos DM, Klein P et al. (2011) Regulable neural progenitor–specific Tsc1 loss yields giant cells with organellar dysfunction in a model of tuberous sclerosis complex. Proc Natl Acad Sci U S A 108:E1070–9 70. Wong M, Crino PB (2012) Tuberous sclerosis and epilepsy: role of astrocytes. Glia 60:1244–50. 71. Zeng LH, Rensing NR, Zhang B et al. (2011) Tsc2 gene inactivation causes a more severe epilepsy phenotype than Tsc1 inactivation in a mouse model of Tuberous Sclerosis Complex. Hum Mol Genet 20:445–54

Tuberous Sclerosis Complex Chapter 11 72. Potter CJ, Pedraza LG, Xu T (2002) Akt regulates growth by directly phosphorylating Tsc2. Nat Cell Biol 4:658–65 73. Inoki K, Li Y, Zhu T et al. (2002) TSC2 is phosphorylated and inhibited by Akt and suppresses mTOR signalling. Nat Cell Biol 4:648–57 74. MacKeigan JP, Krueger DA (2015) Differentiating the mTOR inhibitors everolimus and sirolimus in the treatment of tuberous sclerosis complex. Neuro Oncol 17:1550–9 75. Henske EP, Wessner LL, Golden J et al. (1997) Loss of tuberin in both subependymal giant cell astrocytomas and angiomyolipomas supports a two-hit model for the pathogenesis of tuberous sclerosis tumors. Am J Pathol 151:1639–47 76. Hino O, Kobayashi T (2016) Mourning Dr. Alfred G. Knudson: the two-hit hypothesis, tumor suppressor genes, and the tuberous sclerosis complex. Cancer Sci 108:5–11. 77. Soucek T, Holzl G, Bernaschek G, Hengstschlager M (1998) A role of the tuberous sclerosis gene-2 product during neuronal differentiation. Oncogene 16:2197–204 78. Mizuguchi M, Ikeda K, Takashima S (2000) Simultaneous loss of hamartin and tuberin from the cerebrum, kidney and heart with tuberous sclerosis. Acta Neuropathol 99:503–10 79. Arai T, Ackerley CA, Becker LE (1999) Loss of the TSC2 product tuberin in subependymal giant-cell tumors. Acta Neuropathol 98:233–9 80. Mizuguchi M, Kato M, Yamanouchi H et al. (1996) Loss of tuberin from cerebral tissues with tuberous sclerosis and astrocytoma. Ann Neurol 40:941–4 81. Nellist M, van Slegtenhorst M, Goedblood M et al. (1999) Characterization of the cytosolic tuberin–hamartin complex. Tuberin is a cytosolic chaperone for hamartin. J Biol Chem 274:35647–52 82. van Slegtenhorst M, Nellist M, Nagelkerken B et al. (1998) Interaction between hamartin and tuberin, the TSC1 and TSC2 gene products. Hum Mol Genet 7:1053–7 83. Lee A, Maldonado M, Baybis M et al. (2003) Markers of cellular proliferation are expressed in cortical tubers. Ann Neurol 53:668– 73 84. Crino PB, Trojanowski JQ, Dichter MA, Eberwine J (1996) Embryonic neuronal markers in tuberous sclerosis: single-cell molecular pathology. Proc Natl Acad Sci U S A 93:14152–7 85. Tapon N, Ito N, Dickson BJ et al. (2001) The Drosophila tuberous sclerosis complex gene homologs restrict cell growth and cell proliferation. Cell 105:345–55 86. Ito N, Rubin GM (1999) gigas, a Drosophila homolog of tuberous sclerosis gene product-2, regulates the cell cycle. Cell 96: 529–39 87. Gao X, Pan D (2001) TSC1 and TSC2 tumor suppressors antagonize insulin signaling in cell growth. Genes Dev 15:1383–92

88. Potter CJ, Huang H, Xu T (2001) Drosophila Tsc1 functions with Tsc2 to antagonize insulin signaling in regulating cell growth, cell proliferation, and organ size. Cell 105:357–68 89. Gao X, Zhang Y, Arrazola P et al. (2002) Tsc tumour suppressor proteins antagonize amino-acid–TOR signaling. Nature Cell Biol 4:699–704 90. Tee AR, Fingar DC, Manning BD et al. (2002) Tuberous sclerosis complex-1 and -2 gene products function together to inhibit mammalian target of rapamycin (mTOR)-mediated downstream signaling. Proc Natl Acad Sci U S A 99:13571–6 91. Miloloza A, Rosner M, Nellist M et al. (2000) The TSC1 gene product, hamartin, negatively regulates cell proliferation. Hum Mol Genet 9:1721–7 92. Soucek T, Yeung RS, Hengstschlager M (1998) Inactivation of the cyclin-dependent kinase inhibitor p27 upon loss of the tuberous sclerosis complex gene-2. Proc Natl Acad Sci U S A 95:15653–8 93. Catania MG, Mischel PS, Vinters HV (2001) Hamartin and tuberin interaction with the G2/M cyclin-dependent kinase CDK1 and its regulatory cyclins A and B. J Neuropathol Exp Neurol 60:711–23 94. Miloloza A, Kubista M, Rosner M, Hengstschlager M (2002) Evidence for separable functions of tuberous sclerosis gene products in mammalian cell cycle regulation. J Neuropathol Exp Neurol 61:154–63 95. Vezzani A, Aronica E, Mazarati A, Pittman QJ (2013) Epilepsy and brain inflammation. Experimental Neurology 244 (Special Issue):11–21 96. Wirenfeldt M, Clare R, Tung S et al. (2009) Increased activation of Iba1+ microglia in pediatric epilepsy patients with Rasmussen’s encephalitis compared with cortical dysplasia and tuberous sclerosis complex. Neurobiol Dis 34:432–40 97. van Scheppingen J, Iyer AM, Prabowo AS et al. (2016) Expression of microRNAs miR21, miR146a, and miR155 in tuberous sclerosis complex cortical tubers and their regulation in human astrocytes and SEGA-derived cell cultures. Glia 64:1066–82 98. Citraro R, Leo A, Constanti A et al. (2016) mTOR pathway inhibition as a new therapeutic strategy in epilepsy and epileptogenesis. Pharmacological Research 107:333–43 99. Cepeda C, Hurst RS, Flores-Hernandez J et al. (2003) Morphological and electrophysiological characterization of abnormal cell types in pediatric cortical dysplasia. J Neurosci Res 72:472–86 100. Jeong A, Wong M (2016) Tuberous sclerosis complex as a model disease for developing new therapeutics for epilepsy. Expert Rev of Neurotherapeutics 16:437–47 101. Cepeda C, Andre VM, Yamazaki I et al. (2010) Comparative study of cellular and synaptic abnormalities in brain tissue samples from pediatric tuberous sclerosis complex and cortical dysplasia type II. Epilepsia 51 (Suppl 3):160–5

131

12

Chiari Malformations Homa Adle-Biassette1 and Jeffrey A. Golden2 1 2

Department of Pathology, APHP, Lariboisi`ere Hospital, Universit´e Paris Diderot, Paris, France Department of Pathology, Brigham and Women’s Hospital, Harvard Medical School, Boston, MA, USA

Definition, major synonyms and historical perspective The history of Chiari malformation was reviewed in 2013 [1]. The earliest cases of hindbrain herniation (Chiari type II) associated with spina bifida were reported by the Dutch anatomist Nicholas Tulp in the seventeenth century, and illustrated by Jean Cruveilhier at the beginning of the nineteenth century. Chiari type I malformation was first described by Theodor Langhans. At the end of the nineteenth century, the Austrian pathologist Hans Chiari [2] reported a 17-year-old woman with Chiari type I malformation and subsequently classified a series of hindbrain deformities associated with hydrocephalus into four types that have come to be known as Chiari malformations [3,4]. Common to all the Chiari malformations is the herniation of the cerebellum, and, in some cases, parts of the brainstem outside the posterior fossa. The term Chiari type I malformation (Mendelian Inheritance in Man, MIM, number 118420) refers to the presence of the cerebellar tonsils extending into the upper cervical canal. The Chiari type II malformation (MIM 207950) represents the displacement of the cerebellar vermis into the upper cervical canal with accompanying anomalies of the midbrain, pons, and medulla. Chiari type III is displacement of the cerebellum into a cervical or low occipital encephalocele [5]. A fourth Chiari malformation (Chiari type IV) described the absence of the cerebellum, a term no longer in general usage as it referred to what was probably a secondary destruction of the cerebellum [6]. More recently, there has been some refinement and expansion of the classification of Chiari malformations. A Chiari type 0 malformation was described in individuals without a hindbrain hernia, or with a malformation that is minimal, but having crowding of the craniocervical junction and a syrinx that develops as a consequence of the lack of free and easy cerebrospinal fluid movement across

this area. Another, referred to as the Chiari type 1.5 malformation, represents a Chiari type I malformation in combination with brainstem herniation through the foramen magnum. Chiari type I malformations were previously equated with downward herniation of the cerebellum resulting from supratentorial mass effect, or, less commonly, a mass in the anterior cerebellum. Herniation syndromes are now considered a separate entity, although chronic herniation of the cerebellum is still referred to as a Chiari type I in some settings. Chiari type I malformations can be seen without any downward herniation, and the pathology in the cerebellum is distinct. The Chiari type II malformation is synonymous with the Arnold–Chiari malformations. Arnold’s contribution consists of a single infant with a myelomeningocele and herniation of the cerebellum and the fourth ventricle into the spinal canal, which was incompletely described; however, his students later attached the label Arnold and named their features as “Arnold– Chiari malformation”. Unfortunately, Arnold’s work completely ignores the first description by John Cleland [7] for which Chiari gives credit. Thus, it would be most appropriate to refer to this malformation as the Cleland–Chiari anomaly. Short of this, it seems best to describe this malformation as the Chiari type II.

Embryology The embryologic basis of the Chiari malformations has not been conclusively established [8,9]. Although originally considered a herniation syndrome, Chiari type I malformation is now recognized as a primary malformation where the cerebellar tonsils develop low within the foramen magnum. It is thought to develop in utero prior to the full development of the cerebellar tonsils. The embryogenesis of the Chiari type II malformation has been explored by many investigators without a consensus; these theories are briefly described in the section on

Developmental Neuropathology, Second Edition. Edited by Homa Adle-Biassette, Brian N. Harding, and Jeffrey A. Golden. © 2018 John Wiley & Sons Ltd. Published 2018 by John Wiley & Sons Ltd.

133

Developmental Neuropathology pathogenesis. Nonetheless, it is important to recognize that an embryologic basis for the Chiari type II malformation has not been determined. Chiari type III malformations are a specialized type of encephalocele; a discussion of their embryogenesis can be found in Chapter 5.

Epidemiology Incidence and prevalence The incidence of Chiari type I malformations is not well defined. Until the advent of magnetic resonance imaging (MRI), this anomaly was rarely identified in life. However, with the current use of MRI, the incidence and prevalence have risen precipitously, particularly in patients with scoliosis. Despite the increased recognition, the incidence and prevalence for a randomly selected population remains poorly defined. No variation in frequency has been noted between males or females. The incidence of the Chiari type II malformation exactly parallels that for lumbosacral myelomeningoceles. Myelomeningoceles at other levels, such as the cervical, thoracic, and lower sacral, have a much lower rate of association. This correlation, first documented in the 1930s [8], appears to reflect the incidence of neural tube closure defects and does not reflect variation among different genetic populations. A slight female predominance exists for the Chiari type II malformations, paralleling the incidence of myelomeningoceles. The Chiari type III malformations are extremely rare, no definitive incidence data are available. Risk factors Other than the association of myelomeningocele, and therefore of its risk factors (Chapter 2) with Chiari type II malformations, there are no defined risk factors for any type of Chiari malformation.

Clinical features The clinical spectrum associated with the Chiari type I malformation has broadened in recent years with the increasing use of MRI. The malformation usually becomes symptomatic in teenage to early adult years, although initial presentation can also be in an infant or older adult. The initial complaint is often neck pain, at times presenting with torticollis or retrocollis. If cervical spinal cord syringomyelia coexists, symmetric, asymmetric or even totally unilateral arm pain can be the initial complaint. Neurological symptoms and signs also are dependent on the neural structures involved. Predominant impairment of the brainstem and cerebellar tonsils by either compression at the level of the foramen magnum or, by extension of the syrinx into the brainstem (syringobulbia), can lead to a variety of brainstem and cerebellar signs.

134

In the very young age group, sleep apnea – including near-miss sudden infant death syndrome, stridor and feeding difficulties – may be the presenting symptoms. Eye movement abnormalities are relatively common. Oscillopsia is reported by the patient, and a variety of forms of nystagmus can be seen on examination. Downbeating nystagmus, especially when accentuated by lateral downgaze, should alert the examiner to a potential problem at the craniocervical–pontomedullary junction. Torsional nystagmus may also be present, especially in the presence of syringomyelia/syringobulbia. Lower cranial nerve impairment can lead to dysphagia and aspiration, which may first present as late as the eighth decade. When complicated by syringobulbia, asymmetrical cranial nerve involvement up to cranial nerve V may be present. Imbalance and vertigo with truncal ataxia may indicate impairment of the vestibulo- and spinocerebellar systems. With the coexistence of syringohydromyelia, this clinical picture becomes even more complicated. Forty to seventy-five percent of cases of Chiari type I have an associated syringomyelia, depending on the series and method of ascertainment [10]. When the patient presents with symptoms referable to the spinal cord, syringomyelia can be demonstrated in over 90% of cases. The most common location of maximal expansion is C4 to C6, although the cavity can expand up and down the entire cord. The cavity often is centrally or slightly asymmetrically located, thus first impairing crossing fibers in the anterior spinal commissure. Since these fibers carry sensory information predominantly for pain and temperature, and the posterior columns can be relatively spared, a dissociated sensory loss may result; thus, pain and temperature sensation may be disturbed in a cape-like distribution over the shoulders and down the arms, while light touch and joint position remain relatively unaffected. Compromise of the anterior horn cells may result in patchy weakness, amyotrophy, and loss of reflexes in the arms. These findings are often asymmetric or even unilateral. With expansion of the cavity additional symptoms, including long tract signs and symptoms can become apparent. Another important manifestation of syringomyelia, especially in childhood, is progressive scoliosis, sometimes the earliest sign of a syringomyelia. The clinical features referable to Chiari type II malformations are often overshadowed by the coexistent myelomeningocele and the frequent hydrocephalus. Lower cranial nerve defects, particularly related to swallowing and respiration may be apparent. Dysphagia leading to feeding difficulties, drooling, nasal regurgitation, stridor, vocal cord paralysis, and life-threatening apnea spells may all occur. Cyanotic episodes are ominous, carrying a considerable mortality. Nystagmus, retrocollis and opisthotonus can also be seen. Later presentation of the Chiari type II malformation may in addition include loss of head control, newly arising weakness in the arms, and increasing spasticity leading to quadriparesis. In some cases, direct compression by the cerebellar vermis may also contribute to this clinical phenotype. The major symptoms are usually referable to the hydrocephalus (Chapter 18) that occurs in greater than 90% of cases,

Chiari Malformations Chapter 12

and lesions occurring in the spinal cord (Chapter 17), or the primary myelomeningocele (Chapter 2). Chiari type III malformations generally have a poor prognosis. The children tend to live only days to weeks. Clinical symptoms are generally referable to both the hydrocephalus and brainstem abnormalities similar to the Chiari type II malformations, although the brainstem abnormalities are generally more severe.

Radiology Diagnosis of the Chiari type I malformation, as well as of syringomyelia, relies on clinical suspicion and adequate imaging. The imaging modality of choice is MRI, especially sagittal images. The now routine inclusion of these images in any MRI study of the head has led to the discovery of a number of asymptomatic Chiari type I malformations and asymptomatic cases of syringomyelia. It has also become apparent that the clinical picture is broader and more varied then the classical clinical syndromes described. The question frequently arises about the significance of lowlying tonsils that are discovered incidentally on a cranial MRI examination. The tonsils normally retract upwards with age, so their location must be interpreted in an age-dependent context. The cut-off in the first 10 years of life appears to be 6 mm below the level of the foramen magnum; from 10 to 30 years it would be 5 mm, and after that 4 mm. Herniation greater than 12 mm is almost invariably symptomatic; on the other hand, about 30% of individuals with significant displacements of 5–10 mm are asymptomatic [11]. Similarly, syrinx cavities can be quite large, with minimal clinical symptomatology. Other findings consist of a small posterior fossa or having an abnormal shape [12–15] containing the normally developed nervous structures that overcrowd it and tend to herniate downward with narrowing or obliteration of the retrocerebellar cerebrospinal fluid spaces, compression of the fourth ventricle, retroflexed odontoid process that may compress the anterior aspect of the medulla, hydrocephalus (mild or moderate), and empty sella [16]. Evaluation of the remainder of the brain is also important in patients with Chiari malformations, looking in particular for coexisting anomalies outside the posterior fossa, especially with type II malformations, where numerous associated malformations have been reported [17]. The principal associated abnormalities are lacunar defects in the skull and scalloping of the bones of the posterior fossa, abnormality of the tectal plate (tectal beaking), interhemispheric cysts, abnormal gyral pattern (stenogyria), and callosal dysgenesis, some of which may be diagnosed by fetal sonogram around midgestation [18–21].

Macroscopy and histopathology Chiari type I and II Chiari type I malformations are characterized and actually defined by the presence of the cerebellar tonsils below the level of

Figure 12.1 Midline section of adult with a Chiari type I malformation. Note that the cerebellar vermis is completely intact and the fourth ventricle has a normal morphology. The cerebellar tonsils extend down over the cervical spinal cord.

the foramen magnum (Figure 12.1). The distance below the foramen magnum that defines a Chiari type I malformation is not uniformly agreed upon (see Imaging section), but one centimeter or more is generally accepted. The cerebellar tissue located in the cervical cord can be firm and sclerotic. Sections will often show sclerotic and degenerate cerebellum in the protruded tonsils. Acute changes of softening and necrosis are rarely present, unlike those found with cerebellar herniation syndrome. Surgical resection of the tonsils will also show sclerotic cerebellar tissue. The remainder of the cerebellum is normal in those cases where it has been studied. Accompanying Chiari type I malformations, in approximately 40% of cases, is syringobulbia, hydromyelia or syringomyelia (Chapter 17). As described above, many of the clinical features associated with Chiari type I malformations are actually referable to these later abnormalities. Hydrocephalus may also occur with Chiari type I malformations, although somewhat less frequently. The pathogenesis of syringomyelia, hydromyelia, syringobulbia, and the hydrocephalus associated with Chiari type I malformations remains unknown. The Chiari type II malformation is seen in greater than 95% of children with myelomeningoceles. Although a broad spectrum of findings are part of the Chiari type II malformation [17], the change found in virtually all cases includes displacement of the cerebellar vermis (and to a lesser extent, the inferior lateral cerebellar hemispheres) over the dorsal aspect of the cervical spinal cord (Figure 12.2) within the upper vertebral column. The fourth ventricle, pons, and medulla are similarly elongated and partially located within in the spinal canal. The lower medulla may be kinked. The posterior fossa is always small with scalloping and erosion of the posteromedial part of the petrous pyramids. The cerebellum, confined

135

Developmental Neuropathology

(a)

(b)

Figure 12.2 A Chiari type II malformation in a 20-week fetus with a lumbosacral myelomeningocele. (a) Dorsal view with the midline cerebellar vermis extending down over much of the cervical cord (arrow). The fused inferior colliculus can also be appreciated from this view. (b) Lateral view from the same case with the cerebellar vermis (arrow) lifted up exposing the underlying cord. The elongated fourth ventricle, abnormal medulla and pons are not as obvious.

to the small posterior fossa, can be found around the lateral aspect of the brainstem and may even extend to the ventral midline. Forking of the aqueduct, aqueductal stenosis, and/or aqueductal atresia may be present (Chapter 18). Less common findings include fusion (beaking) of the inferior tectum and anomalies of cranial nerve nuclei. Hydrocephalus is a common complication, and may manifest ante- or postnatally (Chapter 18). The etiology of the hydrocephalus is not always clear, although the aqueductal changes likely account for at least 70% of cases. Supratentorial lesions consisting of dysgenesis of corpus callosum, absence of septum pellucidum, enlarged thalamic massa intermedia and cerebral periventricular heterotopia, polymicrogyria or pachygyria may be present [17,22]. Finally, although not specific, lacunar lucencies in the skull (Luckensch¨adel) are seen in 85% of cases [23]. In addition to the dysraphism, other malformations may be concurrently found in the spinal cord (Chapter 17) [17]. The two most frequent abnormalities of the spinal cord are hydromyelia and syringomelia. They may either occur in isolation or they may be found together (hydrosyringomyelia). These disorders may be found localized to a short segment of the spinal cord, or along great distances. Partial (diastematomyelia) and complete (diplomyelia) duplications of the spinal cord are occasionally present in association with myelomeningocele. These potentially associated anomalies are also not specific for dysraphisms

136

of the spinal cord, and each may be found in isolation or with other central nervous system malformations.

Chiari type III The Chiari type III is a specific subtype of encephalocele (Chapter 2). Like those found along the remainder of the neural axis, there is an overlying bony defect and protrusion of the meninges and brain tissue through the defect; in the case of Chiari type III malformations the brain defect must include the cerebellum (Figure 12.3). The bones involved always include the upper cervical vertebrae (the atlas and axis), and usually the basal component of the occipital bone that is presumably axially and not neural crest-derived squamous bone. In contrast, encephaloceles involving the squamoid portions of the occipital bone tend to involve the occipital lobes of the cerebral hemispheres with or without the cerebellum. It is the latter that are found in syndromes that include occipital encephalocele, in contrast to the Chiari type III malformation, which tends to be sporadic and extremely rare. The cerebellum is often distorted and small. The brainstem is frequently distorted and may partially be located within the neural tube closure defect. The resulting abnormal morphology of the brainstem and cervical spinal cord compromises the course of cranial and cervical nerve roots, the posterior fossa vasculature, and intrinsic structures of the brainstem and cerebellum accounting for the clinical symptoms described

Chiari Malformations Chapter 12

defining its contents, usually by MRI, will readily clarify the diagnosis.

Genetics

Figure 12.3 A newborn child with a Chiari type III malformation. Note the cervical/low occipital origin of the encephalocele, which contained cerebellum (courtesy of Dr. R Hakim).

above. Histopathologically, the cerebellum usually shows dysplastic as well as degenerative changes. The brainstem is distorted but often anatomically intact. Concurrent pathologies such as syringobulbia, syringomyelia, and hydrocephalus may coexist, similar to the Chiari type I and II malformations.

Differential diagnosis The differential diagnosis in the Chiari type I malformation with syringomyelia is broad, given the wide range of the possible clinical symptomatology. Multiple sclerosis, spinal muscular atrophy, amyotrophic lateral sclerosis, spinocerebellar ataxias, mononeuropathy multiplex, cervical disc degenerative disease, and a variety of other disorders of the spinal cord and cerebellum can all be confused with the condition on clinical grounds. Imaging will resolve the majority of these differential diagnoses. The situation can be more difficult in patients with prominent neurological symptoms, but only borderline abnormal findings on imaging. Alternative diagnoses have to be considered in that scenario. A cavity in the spinal cord can also have a traumatic (hematomyelia), inflammatory (necrotizing myelopathy), metabolic (Leigh disease), or neoplastic (astrocytoma and ependymoma) basis. Of the neurocutaneous disorders neurofibromatosis type 1 and von Hippel–Lindau disease can be associated with a syrinx. Demonstration of a flow void within the cavity by MRI can be helpful to demonstrate continuity with the cerebrospinal fluid spaces. The diagnoses of Chiari type II and III malformations have little differential. The consistent association of Chiari type II with myelomeningoceles makes it essential for every patient with myelomeningocele to be screened for the presence or absence of a Chiari type II malformation. For Chiari type III malformations, the encephalocele is usually obvious,

A possible genetic contribution to some Chiari type I is suggested by familial aggregation and twin studies. At least 20 case reports and small series can be found in the literature linking Chiari type I malformations with a variety of syndromes. For type I, recent genome-wide linkage analysis identified two candidate loci on chromosome 9q22.31 and 15q21.1-q22.3 containing FIBRILLIN1, which is a major gene in Marfan syndrome and has been linked to Shprintzen–Goldberg syndrome, of which Chiari type I is a distinguishing characteristic [24]. A case– control association study of single-nucleotide polymorphism across 58 candidate genes involved in early paraxial mesoderm development identified four intronic variants for classical Chiari type I in ALDH1A2, CDX1 and FLT1 [25]. ALDH1A2, CDX1 are directly or indirectly related with retinoic acid signaling during somitogenesis. A study of whole genome expression profiles has identified some biological candidates, but further investigations and larger series are still necessary [26]. In patients with Chiari malformation type II, screening of genes critical for neurulation and neural tube closure disclosed some risk factors but no major causative gene [27]. Numerous syndromes, for some of which the gene has been identified, include an occipital encephalocele; however, these tend to not be true Chiari type III malformations, but occipital encephaloceles involving the occipital lobes and not the posterior fossa.

Experimental models Few experimental models exist to help define the pathogenesis or genetics of any of the Chiari malformations [28]. A breed of dogs has been identified with heritable features similar to the Chiari type I malformation, including having syringomyelia and syringobulbia [29,30]; however, it is not clear whether the clinical features are correlated with these malformations [29]. Administration of a single dose of vitamin A to pregnant hamsters, early during the morning of their eighth day of gestation, induces types I and II Chiari malformations, showing an underdevelopment of the occipital bone and overcrowding of the cerebellum in the small posterior fossa [31]. The spinal open neural tube defects caused by retinoic acid in rats are confined to the lumbosacral region and are similar to human cases in the pathology and clinical features of talipes, neurogenic bladder, and the Chiari-like lesion. Mice exposed to all transretinoic acid develop spinal dysraphism and a hindbrain malformation; however, the exact relationship of the hindbrain malformation to Chiari type II malformations remains uncertain [32]. Similarly, 10 calves

137

Developmental Neuropathology have been described with Chiari type II malformations, but no follow-up studies have been reported [33]. Surgical models of in utero laminectomy and creation of an open spinal dysraphism induces Chiari type II malformation [28,34,35]. In 2005, Matsuoka et al. [36] mapped a cryptic neural crest–mesoderm boundary inside the neck and shoulder girdle skeleton and identified a neural crest-derived skeleton that is specifically affected in human Klippel–Feil syndrome (MIM 118100), Sprengel deformity (MIM184400), and Chiari type II malformation.

Pathogenesis The theories for the pathogenesis of the Chiari malformations may be grouped into the hindbrain dysgenesis and developmental arrest theory, the caudal traction theory, the hydrocephalus and hydrodynamic theory of Gardner, the small posterior fossa/hindbrain overgrowth theory, and the lack of embryological ventricular distention theory [37]. No single hypothesis can explain the full malformation spectrum for Chiari types I or II [9,37]. The pathogenesis of the Chiari type I malformation is enigmatic, but is now recognized as a developmental anomaly, suggesting that herniation is not the cause. The pathogenesis of the associated hydrocephalus, syringomyelia, syringobulbia, and hydromyelia is also unknown. Some have postulated, without proof, that changes in cerebrospinal fluid dynamics occur, possibly even transiently or intermittently as the cerebellar tonsils press on the brainstem and cervical spinal cord with changes in venous pressure. No fewer than six etiologies for the Chiari type II malformation have been described since the malformation was first defined [38]. The hydrodynamic theory of Gardner contends that increased intracranial pressure during development pushes the cerebellum down. Inconsistent with this theory, hydrocephalus is almost always preceded by the Chiari type II malformation. A second theory suggests that tethering of the cord is the primary defect; unfortunately, the course of nerve roots and experimental studies do not support this model. A primary defect in neurulation represents the third hypothesis, one that has been neither proven nor disproved at this point. The fourth hypothesis suggests that the primary defect is in the mesenchyme surrounding the posterior fossa; the abnormal growth of the posterior fossa forcing the inferior cerebellum to develop in a caudal location. The small posterior fossa in parallel to hindbrain overgrowth theory has also been proposed. The fifth theory is the lack of embryological ventricular distention due to the leakage of the myelomeningocele as the cause of the malformation. The lack of distention of the fourth ventricle prevents enlargement of the subsequent bony confines of the posterior fossa. The volume of the posterior fossa would become inadequate to the growth of cerebellum. The sixth hypothesis, somewhat related to the first, is that abnormal fluid dynamics between the rapid egress of cerebrospinal fluid at the site

138

of an open myelomeningocele, and slower resorption above the posterior fossa results in the downward displacement of the cerebellum during development. This hypothesis has gained support from data acquired after fetal surgery to repair myelomeningoceles in utero. Clearly documented Chiari type II malformations, seen before the development of hydrocephalus, have completely resolved after the closure of the caudal defect [39,40]. If proven correct in additional studies, it would seem likely that Chiari type II malformations would represent a deformation secondary to the myelomeningocele, as supported by surgical models of open neural tube defects (see Animal models section). The pathogenesis of Chiari type III malformations centers on the primary neural tube closure defect (Chapter 2). The Chiari type III malformation may be unique in that it involves a specific region of the body that rarely shows a neural tube closure defect.

Future directions and therapy For each of the three Chiari malformations, the most important and immediate requirement is to further define the pathogenesis and genetics and clarify the pathogenesis. Treatment of the Chiari type I complex mainly consists of decompressing the structures trapped in the foramen magnum. Surgical approaches for pediatric [41] and adult [42] patients are comparable. Most often, a suboccipital craniectomy, with or without dural patch grafting, and cervical laminectomies are performed, but anterior approaches may become necessary in selected cases to achieve satisfactory decompression. Occasionally, large syrinx cavities require drainage by fenestration or shunting into the subarachnoid space. In general, the decompressive surgery at the foramen magnum is the more successful of the two [42], and a coexisting syrinx cavities may even shrink or resolve postoperatively without further intervention. Signs and symptoms referable to brainstem compression respond better to surgery then symptoms related to the spinal cord, even when syrinx decompression is included in the operation. Major preexisting sensory or motor deficits are poor prognosticators for functional recovery [41]. Surgical management of the Chiari type II malformation itself may become necessary when prominent brainstem signs persist despite adequate treatment of the hydrocephalus [43]. In these selected cases, decompression via suboccipital craniectomy and cervical laminectomy can be beneficial. Urological management also plays a prominent role in the care for these children [44].

References 1. Tubbs RS, Oakes JW (2013) The Chiari Malformations: A Historical Context. The Chiari Malformations. New York: Springer:5–11 2. Tubbs RS, Cohen-Gadol AA (2010) Hans Chiari (1851–1916). J Neurol 257:1218–20 3. Chiari H (1896) Ver¨anderungen des Kleinhirns, des Pons und der Medulla oblongata in Folge von congenitaler Hydrocephalie des Grosshirns. Denkschrift Akad Wiss Wien 63:71–116

Chiari Malformations Chapter 12 ¨ 4. Chiari H (1891) Uber Ver¨anderungen des Kleinhirns infolge von Hydrocephalie des Grosshirns. Dtsch Med Wochenschr 17:1172–5 5. Ivashchuk G, Loukas M, Blount JP et al. (2015) Chiari III malformation: a comprehensive review of this enigmatic anomaly. Childs Nerv Syst 31:2035–40 6. Tubbs RS, Demerdash A, Vahedi P et al. (2015) Chiari IV malformation: correcting an over one century long historical error. Childs Nerv Syst 2015 Jun 7. [Epub ahead of print] 7. Cleland J (1883) Contribution to the Study of Spina Bifida, Encephalocele, and Anencephalus. J Anat Physiol 17:257–92 8. Russel DS, Donald C (1935) The mechanism of internal hydrocephalus in spina bifida. Brain 58:203–15. 9. Shoja MM, Tubbs RS, Oakes JW (2013) Embryology of the Craniocervical Junction and Posterior Cranial Fossa. The Chiari Malformations. New York: Springer:13–54 10. Menezes AH (1991) Chiari I malformations and hydromyelia: complications. Pediatr Neurosurg 17:146–54 11. Elster AD, Chen MY (1992) Chiari I malformations: clinical and radiologic reappraisal. Radiology 183:347–53 12. Aydin S, Hanimoglu H, Tanriverdi T et al. (2005) Chiari type I malformations in adults: a morphometric analysis of the posterior cranial fossa. Surgical neurology 64:237–41; discussion 41. 13. Badie B, Mendoza D, Batzdorf U (1995) Posterior fossa volume and response to suboccipital decompression in patients with Chiari I malformation. Neurosurgery 37:214–8. 14. Sgouros S, Kountouri M, Natarajan K (2006) Posterior fossa volume in children with Chiari malformation Type I. J Neurosurg 105:101–6 15. Sgouros S, Kountouri M, Natarajan K (2007) Skull base growth in children with Chiari malformation Type I. J Neurosurg 107:188–92 16. Chiapparini L, Saletti V, Solero CL et al. (2011) Neuroradiological diagnosis of Chiari malformations. Neurol Sci 32 Suppl 3:S283–6 17. Gilbert JN, Jones KL, Rorke LB et al. (1986) Central nervous system anomalies associated with meningomyelocele, hydrocephalus, and the Arnold–Chiari malformation: reappraisal of theories regarding the pathogenesis of posterior neural tube closure defects. Neurosurgery 18:559–64 18. Callen AL, Filly RA (2008) Supratentorial abnormalities in the Chiari II malformation, I: the ventricular “point.” J Ultrasound Med 27:33–8 19. Callen AL, Stengel JW, Filly RA (2009) Supratentorial abnormalities in the Chiari II malformation, II: tectal morphologic changes. J Ultrasound Med 28:29–35 20. Filly MR, Filly RA, Barkovich AJ, et al. (2010) Supratentorial abnormalities in the Chiari II malformation, IV: the too–far–back ventricle. J Ultrasound Med 29:243–8 21. Wong SK, Barkovich AJ, Callen AL, et al. (2009) Supratentorial abnormalities in the Chiari II malformation, III: The interhemispheric cyst. J Ultrasound Med 28:999–1006. 22. Naidich TP, Pudlowski RM, Naidich JB (1980) Computed tomographic signs of the Chiari II malformation. III: Ventricles and cisterns. Radiology 134:657–63. 23. Peach B (1965) The Arnold–Chiari Malformation; Morphogenesis. Arch Neurol 12:527–35 24. Boyles AL, Enterline DS, Hammock PH et al. (2006) Phenotypic definition of Chiari type I malformation coupled with highdensity SNP genome screen shows significant evidence for linkage

25.

26.

27. 28.

29.

30.

31.

32.

33. 34.

35.

36. 37. 38.

39.

40. 41. 42. 43. 44.

to regions on chromosomes 9 and 15. Am J Med Genet A 140:2776– 85 Urbizu A, Toma C, Poca MA et al. (2013) Chiari malformation type I: a case–control association study of 58 developmental genes. PloS one 8:e57241 Markunas CA, Lock E, Soldano K et al. (2014) Identification of Chiari Type I Malformation subtypes using whole genome expression profiles and cranial base morphometrics. BMC medical genomics 7:39 Kibar Z, Capra V, Gros P (2007) Toward understanding the genetic basis of neural tube defects. Clin Genetics 71:295–310 Sim KB, Lee JY, Phi JH, et al. (2013) Experimental models of spinal open neural tube defect and Chiari type II malformation. Childs Nerv Syst 29:1435–49 Lu D, Lamb CR, Pfeiffer DU et al. (2003) Neurological signs and results of magnetic resonance imaging in 40 cavalier King Charles spaniels with Chiari type 1-like malformations. Vet Rec 153: 260–3 Rusbridge C, Knowler SP (2004) Inheritance of occipital bone hypoplasia (Chiari type I malformation) in Cavalier King Charles Spaniels. J Vet Intern Med 18:673–8 Marin-Padilla M, Marin-Padilla TM (1981) Morphogenesis of experimentally induced Arnold–Chiari malformation. J Neurol Sci 50:29–55 Alles AJ, Sulik KK (1992) Pathogenesis of retinoid–induced hindbrain malformations in an experimental model. Clin Dysmorphol 1:187–200 Cho DY, Leipold HW (1977) Arnold–Chiari malformation and associated anomalies in calves. Acta Neuropathol 39:129–33 Corti G, Manzur T, Nagle C et al. (2010) Etiopathology of Arnold– Chiari malformation: a fetal rat model of dysraphism. Fetal Diagn Ther 28:28–33. Weber Guimaraes Barreto M, Ferro MM, Guimaraes Bittencourt D et al. (2005) Arnold–Chiari in a fetal rat model of dysraphism. Fetal Diagn Ther 28:28–33 Matsuoka T, Ahlberg PE, Kessaris N et al. (2005) Neural crest origins of the neck and shoulder. Nature 436:347–55 Tubbs RS, Lyerly MJ, Loukas M et al. (2007) The pediatric Chiari I malformation: a review. Childs Nerv Syst 23:1239–50 Jennings MT, Clarren SK, Kokich VG, et al. (1982) Neuroanatomic examination of spina bifida aperta and the Arnold–Chiari malformation in a 130-day human fetus. J Neurol Sci 54:325–38 Adzick NS, Walsh DS (2003) Myelomeningocele: prenatal diagnosis, pathophysiology and management. Semin Pediatr Surg 12:168– 74 Walsh DS, Adzick NS, Sutton LN et al. (2001) The Rationale for in utero repair of myelomeningocele. Fetal Diagn Ther 16:312–22 Nagib MG (1994) An approach to symptomatic children (ages 4–14 years) with Chiari type I malformation. Pediatr Neurosurg 21:31–5 Bindal AK, Dunsker SB, Tew JM Jr. (1995) Chiari I malformation: classification and management. Neurosurgery 37:1069–74 Rauzzino M, Oakes WJ (1995) Chiari II malformation and syringomyelia. Neurosurgery clinics of North America 6:293–309. Hernandez RD, Hurwitz RS, Foote JE et al. (1994) Nonsurgical management of threatened upper urinary tracts and incontinence in children with myelomeningocele. J Urol 152:1582–5

139

13

Dandy–Walker Malformation, Mega Cisterna Magna, and Blake’s Pouch Cyst Robert F. Hevner,1,2 Kathleen Millen,2 and William B. Dobyns2,3 1

Department of Neurological Surgery, University of Washington, Seattle, WA, USA Center for Integrative Brain Research, Seattle Children’s Research Institute, Seattle, WA, USA 3 Department of Pediatrics, University of Washington, Seattle, WA, USA 2

Definitions

Synonyms and historical annotations

Dandy–Walker malformation (DWM) is a congenital or postnatal malformation of the posterior fossa, defined by three principal anomalies: fourth ventricular cyst, enlarged posterior fossa, and hypoplastic cerebellum [1–3]. Other typical features include elevation of tentorium and sinuses, superior rotation of the vermis, closed basal foramina, and hydrocephalus [4–7]. The classic case of Dandy and Blackfan [8] exhibited all of these features (Figure 13.1a,b), but some cases do not. Mega-cisterna magna (MCM), once a generic descriptor, now defines a specific malformation consisting of enlarged cisterna magna and continuous fourth ventricle, without cerebellar dysgenesis [1,6,9–11]. Morphologically, MCM forms a spectrum with DWM (Figure 13.2). These malformations also share origins from a fourth ventricular cyst, which involutes in MCM but persists in DWM. Indeed, MCM is a misnomer, because the enlarged retrocerebellar space initially belongs to the fourth ventricle, not the cisterna magna [3]. Blake’s pouch cyst (BPC) consists of an intact fourth ventricular cyst that may expand the posterior fossa and cause hydrocephalus, without significant vermian dysgenesis [9–12]. Thus, BPC occupies an intermediate position on the DWM–BPC– MCM spectrum. Some classifications have included isolated cerebellar vermis hypoplasia on the DWM spectrum [1]. This concept has gained some support from studies that have revealed signaling and genetic interactions between the meninges and the cerebellum [13]. On the other hand, most forms of cerebellar hypoplasia seem unrelated to fourth ventricular cystic malformations.

The century-long history of DWM has been paralleled by immense progress in neurosurgery, neuroimaging, and neuroembryology [2,11,14,15]. During this time, classifications and definitions have evolved, understanding of pathogenesis has improved, and new diagnostic tools and treatments have been developed. The label “Dandy–Walker syndrome” was first used by Benda [16], “to match this condition with the Arnold–Chiari syndrome” and honor previous authors [8,17]. Since DWM is defined by structural anomalies and not clinical features, the name DWM is preferred over “Dandy-Walker syndrome” today. The Dandy–Walker complex, or continuum, is a classification system that was introduced to recognize that DWM is heterogenous, and forms a spectrum with related disorders [1]. The Dandy–Walker complex was proposed to encompass cystic malformations of the posterior fossa, as well as isolated cerebellar vermis hypoplasia. However, with rare exceptions [13], most forms of cerebellar vermis hypoplasia have unique genetic etiologies (e.g., mutations in OPHN1 or CASK) that are unrelated to DWM (13). Currently, the Dandy–Walker complex framework is controversial [3]. The term “Dandy–Walker variant” has been applied to describe various disorders, exhibiting some but not all features of DWM. Isolated cerebellar vermis hypoplasia, MCM, and BPC are among the anomalies that have been called Dandy–Walker variants [3,11,15]. Since it is nonspecific, use of this term is now discouraged [5,15]. Persisting Blake’s pouch [9] is a synonym for BPC.

Developmental Neuropathology, Second Edition. Edited by Homa Adle-Biassette, Brian N. Harding, and Jeffrey A. Golden. © 2018 John Wiley & Sons Ltd. Published 2018 by John Wiley & Sons Ltd.

141

Developmental Neuropathology

(a)

(c)

(b)

(d)

Figure 13.1 Classic examples of Dandy–Walker malformation (DWM). (a, b) ¨ Case of Dandy and Blackfan [8], drawings by Brodel. The girl was well until she had meningitis and seizures at 4 postnatal months, then had progressive hydrocephalus and death at 13 months. Ventral view (a) shows the distended fourth ventricle cyst, and laterally splayed hemispheres. Sagittal view (b) shows the hypoplastic,

Clinical features DWM typically presents with signs and symptoms of hydrocephalus at birth, or during early childhood [1,2,8,14,17,18]. Other presenting findings may include macrocephaly, dolichocephaly, occipital protrusion, seizures, poor motor control, spasticity, opisthotonos, breathing abnormalities, cranial nerve palsies, nystagmus, ataxia, and autism. Clinical severity is worse in cases with associated supratentorial malformations [1]. About

142

upwardly rotated cerebellum. The cyst displaced tentorium superiorly, and contained choroid plexus. (c, d) DWM in a four-year-old girl [40]. Midsagittal slice (c) of the brainstem and cerebellum reveals upward rotation of the vermis, and splayed hemispheres. Histology (d) reveals flayed appearance of the nodulus (labeled “N”), separated through white matter, with cerebellar cortex in anterior cyst wall.

half of DWM patients have significant intellectual disability or motor problems, such as cerebral palsy, and about 25% have somatic malformations, such as polydactyly or cardiac defects [4,19,20]. A small fraction of DWM cases are clinically silent, and are found incidentally [21]. Like DWM, BPC typically presents with hydrocephalus. MCM is thought to be benign and asymptomatic [6,9]. DWM, BPC, and MCM are increasingly diagnosed in fetuses on the basis of neuroimaging, without knowledge of their clinical impact. However, the significance of fetal neuroimaging

Dandy–Walker Malformation, Mega Cisterna Magna, and Blake’s Pouch Cyst Chapter 13

Figure 13.2 The Dandy–Walker malformation (DWM) spectrum of cystic fourth ventricle malformations includes mega-cisterna magna (MCM), Blake’s pouch cyst (BPC), and DWM. In a previous nosology, the “Dandy–Walker complex” was proposed to include MCM, BPC, DWM, and cerebellar vermis hypoplasia (CVH) [1], but genetic links between CVH and DWM are few [13]. Diagrams represent sagittal magnetic resonance images (anterior left, superior up). Key: brainstem,

gray; vermis, medium blue; hemisphere, lavender; tela choroidea or cyst membrane, light green; ependyma, dark green; choroid plexus, red; torcula, dark blue; pia, yellow; arachnoid, light blue; dura, orange; bone, brown. Arachnoid trabeculae (not shown) traverse the subarachnoid space and contact the pial surface, including tela choroidea.

diagnoses remains uncertain [7]. For example, fetal BPC may persist and worsen with continued cyst expansion, or may undergo spontaneous fenestration to resolve as MCM, or there may be a normal outcome [22].

(Figure 13.3). Sagittal planes enhance the evaluation of anomalies in DWM [1]. Evaluation of the ventricles has paramount clinical importance, as hydrocephalus may require urgent treatment [10,11,23,24]. All malformations in this group display an enlarged retrocerebellar space, either within a membranous cyst continuous with the fourth ventricle (DWM and BPC), or appearing as cisterna magna continuous with the fourth ventricle (MCM). Since cyst membranes are difficult to assess, and may be attenuated or fused with adjacent dura, many neuroimaging interpretations represent “best estimates” of cyst structure. Enlargement of the posterior fossa (present in DWM, and some BPC and MCM) is frequently accompanied by elevation of the torcula and tentorium, scalloping of the internal table of occipital squamous bones, attenuation of the falx cerebelli, and compression of brainstem against the clivus.

Neuroimaging Advanced neuroimaging methods such as ultrasound, computed tomography, and magnetic resonance imaging (MRI), have had a transformative impact on the diagnosis of DWM and related malformations.

Children and adults Neuroimaging is essential to the diagnosis of fourth ventricular cystic malformations. MRI is the modality of choice

(a)

(b)

Figure 13.3 Magnetic resonance images of Dandy–Walker malformation. The vermis (v) is small with reduced folia, and is rotated far upward (a) so the fourth ventricle (4v) communicates widely with a large posterior fossa cyst (cy) (a, b). Cerebellar hemispheres (he) are moderately splayed (a, b), and appear less prominent in sagittal view (a). The posterior fossa is enlarged, and the torcula (t) is elevated.

143

Developmental Neuropathology Significant cerebellar hypoplasia or dysgenesis (present in DWM, but not BPC or MCM) is seen as decreased size or absence of one or more vermian lobules. However, the distinction between hypoplasia and atrophy (sometimes present in MCM and BPC) is not always clear. “Anticlockwise” rotation of the vermis is typical of DWM, but is seen in some BPC and MCM cases as well. (The rotation is “anticlockwise” in standard sagittal MRI orientations, with anterior left and head up.) Lateral splaying of the cerebellar hemispheres manifests in sagittal planes as reduction in size, or absence of hemisphere lobules from medial slices. MRI can reveal DWM in very young patients, including neonates [24].

Fetal neuroimaging Both ultrasound and, increasingly, MRI play important roles in the prenatal diagnosis of DWM, BPC, and MCM. These diagnoses may prompt termination of pregnancy [3,6,24]. Currently, however, the sensitivity and specificity of prenatal neuroimaging to detect DWM are unknown [7]. Of concern, some studies have found low concordance (approximately 50%) between DWM diagnoses by fetal ultrasound and postmortem examination [25]. If possible, DWM-like anomalies on fetal ultrasound should also be examined by MRI [26]. On prenatal MRI, DWM anomalies can be seen as early as 19 weeks of gestation [6,26]. Fetal neuroimaging includes the measurement of standard angles and distances, such as the anteroposterior dimension (depth) of the cisterna magna, for which normative data are available [25–27]. Measurements outside the normal range may suggest a DWM spectrum anomaly. For example, fetal cisterna magna depth greater than 10 mm raises the possibility of a malformation involving the posterior fossa [6]. Some anomalies seen on fetal ultrasound or MRI have been proposed as “signs” of abnormal development in DWM. These include “keyhole” and “trapezoid” signs (describing fourth ventricle structure) on axial images [6,25,28]; and the “tail sign” (describing posterior vermis structure) on sagittal MRI [26]. These signs are indicators, but none is specific for DWM.

Epidemiology and genetics DWM typically presents in newborns and infants, but can present at any age, from mid-gestational fetus to adult. A female preponderance of DWM was found in some early studies, but larger cohorts have detected no statistically significant difference of DWM incidence between sexes [19,20]. The incidence of DWM is widely estimated between 1 in 2500 to 1 in 30 000 live births [5,7,20]. DWM accounts for up to 4% of pediatric hydrocephalus [5]. MCM may be the most common malformation of the posterior fossa, while BPC is rare [11,20]. DWM has been associated with a variety of sporadic and heritable syndromes, among them Joubert, Ritscher–Schinzel,

144

Ellis–van Creveld, Walker–Warburg, Meckel–Gruber, Coffin– Siris, Fraser (cryptophthalmos), Cornelia de Lange, Goldenhar, and Aicardi syndromes [6,19,29,30]. However, the frequency of DWM in these syndromes is generally low. Many studies have suggested that DWM risk may be increased by exposures during fetal or postnatal life [2,7,8,27,31]. Prenatal risk factors may include teratogens (coumadin, alcohol, isotretinoin, clomiphene), infections (rubella, cytomegalovirus, toxoplasma), gestational diabetes, twinning, and cerebellar hemorrhage. Postnatal risk factors may include meningitis and subarachnoid hemorrhage. Embryonic mice and rats exposed to certain teratogens develop a DWM-like malformation [32,44]. Most cases of DWM are sporadic, although many familial cases have also been reported [4]. The recurrence risk is estimated at 1–5%, indicating low heritability [5,7]. Non-inherited mutations may also be important in DWM, but none has been found as yet. Many chromosomal anomalies have been associated with DWM, including deletions (3q24, 6p25), duplications (5p, 8p, and 8q), trisomies (9, 13, and 18), and triploidy [6,7,20,33]. Some subchromosomal deletions have defined critical regions containing specific transcription factor genes, which regulate development of the cerebellum and posterior fossa. These include ZIC1 and ZIC4 on chromosome 3q24 [34], and FOXC1 on 6p25.3 [13]. Other genes implicated in DWM include FGF17, LAMC1, and NID1 [23].

Pathology Macroscopy Macroscopic findings are key to the pathologic diagnosis of DWM, BPC, and MCM (Figures 13.1a–c, 13.4a). The posterior fossa should be assessed radiographically before opening the calvaria. Posterior dissection is optimal in fetuses and infants, to demonstrate cysts in situ [24,27]. At brain cutting, midline sagittal slices are usually best for demonstrating DWM, BPC, or MCM. Enlarged or cystic retrocerebellar space DWM or BPC cysts can rarely be kept intact during brain removal, as cyst walls usually tear and collapse. Immersion of the brain and cyst in fixative or water facilitates examination of the reflated cyst (Figure 13.4a). In DWM and BPC, the cyst fills the retrocerebellar space (virtually eliminating the subarachnoid space), and communicates freely with the fourth ventricle. Cyst walls vary from thin and membranous to thick and fibrotic, and may adhere to surrounding dura, brain, and other tissues. Cyst walls attach to posterior surfaces of the cerebellum, pons, and medulla. Cysts may extend superiorly through the tentorial notch and inferiorly through the foramen magnum, and may be symmetric or asymmetric. If the cyst does not communicate with the fourth ventricle, other cyst types, such as arachnoid cyst, should be considered.

Dandy–Walker Malformation, Mega Cisterna Magna, and Blake’s Pouch Cyst Chapter 13

(a)

scalloped inner tables. The brainstem may be compressed against the clivus. Cerebellar hypoplasia By definition, the cerebellar vermis is hypoplastic in DWM. Hypoplasia is usually signified by greatly decreased size or absence of some folia, especially in posterior lobes of vermis. Complete vermian agenesis is reported in approximately 25% of cases of DWM. Typically, the vermis is also rotated “upward” (superiorly and anteriorly) around its attachment to the superior cerebellar peduncles (corresponding to “anticlockwise” on sagittal MRI). The cerebellar hemispheres may also be hypoplastic, and are often splayed laterally away from the midline. In BPC and MCM, the cerebellum does not appear hypoplastic, but may be atrophic due to cyst pressure. The vermis may be rotated, and the hemispheres splayed [3,9]. In practice, cerebellar hypoplasia, like other features of DWM, may form a continuum of severity, adding to the difficulty of distinguishing between DWM, BPC, and MCM [6,20].

(b)

Figure 13.4 Dandy–Walker malformation in 27 weeks of gestation fetus. (a) Cerebellum and brainstem, posterior view into the fourth ventricle. Immersed under water, the delicate cyst membranes are visible surrounding the cerebellum. The small vermis is rotated upward, and the hemispheres are splayed apart. (b) Cyst membrane, interior down (hematoxylin and eosin). Note trilaminar appearance of flattened ependyma, disorganized neuronal and glial elements, and leptomeninges. Scale bar: 100 μm.

In MCM, no intact cyst is discernible, but the cisterna magna communicates freely with the fourth ventricle through a widened foramen of Magendie, without intervening inferior medullary velum (tela choroidea of the posterior fourth ventricle). If the velum appears intact and no cyst is present, the enlarged retrocerebellar space may be designated “prominent cisterna magna,” to distinguish it from MCM [1]. Enlarged posterior fossa The posterior fossa is enlarged in DWM, and may be in BPC and MCM as well. The tentorium and associated venous sinuses (straight sinus, transverse sinus, and torcula) are elevated. The tentorium and falx cerebelli may be attenuated. Squamous parts of occipital bones may protrude posteriorly, and may have

Hydrocephalus and the foramina of Luschka and Magendie In the classic case of Dandy and Blackfan [8], “the basal foramina were occluded by adhesions resulting from an old meningitic process.” This foraminal occlusion led to “absence of communication between the ventricles and the subarachnoid space” (i.e., non-communicating hydrocephalus). Subsequent pathological studies found that hydrocephalus and foraminal occlusion are frequent, but not universal findings in DWM [14,16–18, 35]. Current estimates are that 70–90% of patients with DWM develop hydrocephalus [6]. Moreover, “hydrocephalus is often not evident at birth; it develops postnatally and presents by three months of age in 75% of patients” [20]. Hydrocephalus is also common in BPC, but does not occur in MCM [9,11]. Choroid plexus In up to 40% of cases of DWM, no choroid plexus is found in the fourth ventricle [20]. When present, the choroid plexus is located in lateral recesses of the fourth ventricle, and along posterior cerebellar surfaces near cyst wall contacts. Morphometry In fetal or infant cases, measurements of cerebellar dimensions and cisterna magna depth may facilitate comparisons to neuroimaging, and to normal development [25,27,36]. Other central nervous system anomalies Approximately half of DWM cases have additional central nervous system malformations [5,6,19,20]. Agenesis of the corpus callosum is most frequent, in 10-40% of DWM cases. Less frequent associations include schizencephaly, polymicrogyria, pachygyria, occipital encephalocele, meningomyelocele, nodular subependymal heterotopia, and other malformations. BPC

145

Developmental Neuropathology and MCM are less frequently associated with other brain malformations [6,11]. Associated somatic malformations Diverse types of somatic malformations have been associated with DWM, but none is consistent, and only around 25% of patients with DWM are affected with somatic anomalies [19, 20]. Cardiac malformations, polydactyly, capillary angioma, and ophthalmic defects (e.g., glaucoma) are among the anomalies found with DWM.

Histopathology Cyst walls The histology of fourth ventricular cysts reflects their origins mainly from the roof of the fourth ventricle (tela choroidea). In addition, 50% or more of DWM cysts incorporate elements of cerebellar cortex, and the histology of these cyst walls varies regionally [14,20]. Regions of cyst wall derived from tela choroidea are developmentally trilaminar, with an inner layer of ependyma, a thin intermediate layer of glia limitans, and an outer layer of leptomeninges with vascular elements (Figure 13.4b). The outer layer may contain pia and arachnoid, as both are normal components of tela choroidea. Expansion causes attenuation of cyst walls, gaps in the ependyma, and loss of arachnoid and glial cells, often leaving a largely fibrous membrane, sometimes containing microcalcifications. Neural tissue, seen in DWM cyst walls near the cerebellum, ranges from laminated cerebellar cortex (Figure 13.1d), to disorganized clusters of neurons and glia. Progenitor cells of the external germinal layer are seen in fetal and infant cases. Cyst floors are formed of brainstem and cerebellar deep white matter. Cyst walls in BPC exhibit similar histopathology as DWM, without elements of cerebellar cortex. In MCM, the fourth ventricular cyst wall degenerates, leaving the fourth ventricle lined by ependyma, and the subarachnoid space lined by arachnoid. Cerebellum and brainstem In DWM, the cerebellar vermis appears hypoplastic, with reduced size and foliation, typically affecting posterior lobules most severely, in fetuses as well as adults [36]. The remaining cerebellar cortex may appear relatively normal or disorganized. Pressure atrophy is frequently noted. Cerebellar heterotopia have been described in DWM [20], but are also common in normal development, not necessarily disease-related [37]. Cerebellar hemispheres often exhibit similar changes as vermis. Brainstem nuclei may show atrophy, less often dysplasia. Crossing defects of the corticospinal tracts have been described [20]. In BPC and MCM, the cerebellum and brainstem may show atrophic changes, but the cerebellum should not exhibit significant hypoplasia or dysgenesis. Distinctions between hypoplasia and atrophy in a brain structure that is still actively growing can be difficult, however.

146

Differential diagnosis The differential diagnosis of enlarged retrocerebellar space, with or without cerebellar vermis hypoplasia, includes disorders of cerebellar development, and cystic abnormalities of the posterior fossa. Disorders of cerebellar development DWM must be distinguished from primary cerebellar hypoplasias without posterior fossa cyst or enlargement. Congenital cerebellar hypoplasias, pontocerebellar hypoplasias, Joubert syndrome, and generalized brain hypoplasias (e.g., Smith–Lemli–Opitz syndrome; Figure 13.5) can all exhibit cerebellar hypoplasia with prominent cisterna magna. However, the prominent cisterna magna is caused by cerebellar deficiency, not cystic dilatation of the fourth ventricle, and the roof of the fourth ventricle is normal. Also, some cerebellar hypoplasias exhibit specific features not associated with DWM; for example, the “molar tooth sign” on axial neuroimaging in Joubert syndrome [38]. Interestingly, DWM can occur in Joubert syndrome, and may obscure the diagnosis [19,29]. Mitochondrial disorders, cobblestone malformations (e.g., Walker–Warburg), and other varieties of cerebellar dysgenesis may also enter the differential diagnosis. They are distinguished from DWM, BPC, and MCM by the absence of posterior fossa cyst or enlargement. Cystic abnormalities of the posterior fossa DWM and BPC must be distinguished from other cyst types, such as arachnoid cysts, epidermoid cysts, and brain tumor cysts. Among these, arachnoid cysts are most similar to fourth ventricular cysts, and present the most challenging differential: DWM and arachnoid cysts both contain cerebrospinal fluid, may cause posterior fossa enlargement and bone scalloping, and may have walls that contain leptomeningeal tissue. However, arachnoid cysts typically compress the cerebellum downward against brainstem, in contrast to DWM cysts, which rotate the vermis upward. Arachnoid cysts are not continuous with the fourth ventricle, and their walls lack ependymal, glial, or neuronal cells.

Embryology Development of the cerebellum and fourth ventricle By 16 weeks of gestation, the major primary lobules of the cerebellar vermis and hemispheres have begun to form, and the foramina of Magendie and Luschka have opened. Thereafter, the cerebellar folia continue to branch and grow throughout gestation [36]. Indeed, genesis of granule neurons continues in the external germinal layer for one to two postnatal years. So, cerebellar growth and morphogenesis are continuous processes during the peak ages of DWM incidence. The fourth ventricle at 16 weeks of gestation is covered by the cerebellum anteriorly, and by the delicate tela choroidea posteriorly. A central segment of tela choroidea gives rise to choroid

Dandy–Walker Malformation, Mega Cisterna Magna, and Blake’s Pouch Cyst Chapter 13 (a)

(b)

Blake’s pouch Neuroanatomical studies have sometimes described a “pouch” or evagination of tela choroidea, continuous with the fourth ventricle and filling the retrocerebellar space of fetuses around midgestation [2,11,39]. The original “Blake’s pouch” was illustrated in sagittal sections from a fetus of 18.5 weeks of gestation with normal vermis [39]. Like DWM cysts, Blake’s pouch is derived from tela choroidea that expands due to atresia or occlusion of the foramina. Previous thoughts that Blake’s pouch may be a normal variant have yielded to current views that Blake’s pouch is abnormal after 16 weeks of gestation. Anterior and posterior membranous areas of embryonic tela choroidea Early in fourth ventricle development (8–10 weeks of gestation), choroid plexus arises from middle portions of the tela choroidea, dividing it into anterior and posterior membranous areas. Early attenuation of the anterior membranous area or “area membranacea anterior of Weed” has been proposed to cause cerebellar hypoplasia in DWM [1,2,15,40]. Perforation of the posterior membranous area leads to opening of the basal foramina. Excessive proliferation of choroid plexus and meninges over the posterior membranous area are hypothesized to contribute to hydrocephalus and foraminal atresia, respectively [7,34,41]. “Fusion” of the vermis Hypoplasias of the vermis, as in DWM, have sometimes been attributed to defective fusion at the cerebellar midline [2]. However, modern studies have shown that the vermis grows by expansion of the medial primordium, not by fusion of bilateral elements [42].

Figure 13.5 Prominent cisterna magna in an 18-day-old infant, born prematurely at 34 weeks of gestation. Transcranial ultrasound indicated mega-cisterna magna, possible Dandy–Walker malformation. Additional studies were diagnostic of Smith–Lemli–Opitz syndrome with microcephaly. (a) Cerebellum and brainstem, posterior view into the fourth ventricle. A thin, gray, glistening, translucent membrane (tela choroidea) covered the posterior fourth ventricle (arrows). The brain was photographed in air so the membrane appears partially collapsed. (b) The vermis appeared hypoplastic. Enlargement of the retrocerebellar space reflected cerebellar hypoplasia without cystic dilatation of the fourth ventricle, so the pathologic diagnosis was prominent cisterna magna (tela choroidea collapsed during slicing).

plexus, which grows into the fourth ventricle and extrudes through the foramina of Luschka and Magendie, into cisterna magna. The extruding choroid plexus is initially enclosed in an evaginating sac of tela choroidea. Spontaneous perforation of the sacs between 8–16 weeks of gestation opens the foramina, allowing cerebrospinal fluid to flow from the fourth ventricle to subarachnoid space of the cisterna magna, and sac walls involute. The presence of an enlarged retrocerebellar space or cyst after 16 weeks of gestation is therefore considered abnormal [6,26].

Pathogenesis and etiology Studies in humans and mice indicate that DWM has multifactorial etiology, linked to genetic mutations and environmental factors at vulnerable ages (Figure 13.6).

Pathogenesis of cystic fourth ventricle Dandy and Blackfan [8] theorized that the cystic fourth ventricle was caused by obstruction of the foramina of Luschka and Magendie; in their case, due to meningitis at age four months, followed by hydrocephalus and death at 13 months. In the ensuing decades, DWM has been linked to impaired cerebrospinal fluid flow caused by diverse mechanisms of foraminal atresia and occlusion, from excessive proliferation of meninges, to occlusion by inflammation or hemorrhage. The form of the cyst and associated defects (DWM, BPC, or MCM) may change depending on the timing and pace of cyst growth, fenestration, and involution. Just as aqueductal stenosis leads to supratentorial hydrocephalus, occlusion of the foramina in fetuses and infants evidently leads to DWM, BPC, or MCM, in at least a subset of vulnerable individuals.

147

Developmental Neuropathology

Figure 13.6 Multifactorial pathogenesis of Dandy–Walker malformation (DWM) and related disorders. The continuum of mega-cisterna magna– Blake’s pouch cyst –DWM malformations is a product of diverse etiologies, pathogenesis, and modifying factors. Important factors in fourth ventricular cyst expansion include timing, severity, and duration of foraminal occlusion. Cerebellar dysgenesis in DWM may be caused by a combination of direct effects of factors such as mutations and teratogens, as well as indirect effects of altered meningeal signaling and cyst expansion and pressure. In some cases, isolated cerebellar vermis hypoplasia may be caused by the same underlying etiology; for example, mutations in genes that are also linked to DWM (e.g., FOXC1 mutations).

Pathogenesis of enlarged posterior fossa Posterior fossa expansion, tentorial elevation, and scalloping of bones may be secondary effects of cyst expansion. In addition, primary defects of posterior fossa meninges have been implicated by genetic studies, indicating that some DWM-associated genes (FOXC1, ZIC1) are important for meningeal development. Since posterior fossa meninges produce factors that regulate cerebellar development, meningeal abnormalities may secondarily impact cerebellar growth in DWM [7]. Pathogenesis of cerebellar hypoplasia in DWM Primary defects of cerebellar growth, as well as secondary effects of cyst expansion and meningeal signaling, may all contribute to cerebellar hypoplasia and atrophy. Primary cerebellar defects have been implicated by studies showing that DWM-linked genes ZIC1 and ZIC4 are highly expressed in the developing cerebellum, and are necessary for cerebellar growth [34]. The impact of DWM cyst growth on cerebellar atrophy is clear [4,43]. In embryos, cyst growth might cause attenuation of the anterior membranous area adjacent to the cerebellar anlage. The latter effect is hypothesized to “produce an extension of the cerebellar plate and impede the cellular migration” [40].

Experimental models DWM has been modeled in rodents by teratogen exposure, and by genetic mutations.

Teratogen exposure Administration of galactoflavin (a teratogen and anticancer agent that depletes and antagonizes riboflavin) to embryonic mice causes a congenital DWM-like malformation [32]. Features include “extreme dilatation of the fourth ventricle, with frequent

148

total absence of the cerebellum,” expanded posterior fossa, and hydrocephalus. The malformation was attributed to “a disturbance of cerebrospinal fluid circulation and abnormal development of the membranous areas of the roof of the fourth ventricle.” However, the precise mechanism of cerebrospinal fluid flow disturbance (foraminal atresia or occlusion) was unknown, and direct effects on cerebellar growth were not excluded. Similarly, in rats, embryonic administration of 6-aminonicotinamide (a niacinamide antagonist) caused a congenital DWM-like malformation with hydrocephalus [44].

Genetic models Several lines of mice with high incidence of hydrocephalus were bred in the last century, some of which exhibited DWM-like anomalies of the cerebellum and fourth ventricle [32]. In hy1 mice, the first evident anomaly in embryos was a stretched anterior membranous area. In hy-3, hydrocephalus appeared postnatally, after CSF block caused by meningeal degeneration. In dreher, some mutants exhibit a cystic fourth ventricle, cerebellar hypoplasia, and blocked foramina. Subsequently, dreher was mapped to Lmx1a, a transcription factor gene expressed in both roof plate (precursor of tela choroidea) and cerebellar anlage [41]. To model a specific human genetic DWM (del chr 3q24), mice with homozygous deletion of both Zic1 and Zic4 were produced [34]. These mice displayed DWM-like cerebellar hypoplasia, attributed to defects in the cerebellum (Zic1 and Zic4 expression) and possibly, the meninges as well (Zic1 expression). Cerebellar hypoplasia in these mice is caused by defective postnatal proliferation of cerebellar granule neurons [45]. In another model of a human genetic DWM (del chr 6p25), Foxc1 mutant mice likewise exhibited cerebellar hypoplasia [7]. Interestingly, Foxc1 is a transcription factor gene that is normally expressed only in embryonic meninges, not cerebellum.

Dandy–Walker Malformation, Mega Cisterna Magna, and Blake’s Pouch Cyst Chapter 13

Thus, signaling interactions between meninges (which secrete factors such as chemokine SDF1α) and cerebellum are implicated in DWM [46]. Moreover, the finding that some humans with FOXC1 mutations have cerebellar vermis hypoplasia but no cystic malformation, supports the view that isolated vermian hypoplasia is part of the DWM spectrum [7].

Treatment The principal goal of treatment is to counter hydrocephalus, a major complication of DWM and BPC, but not MCM (which requires no treatment). Cyst fenestration (“membrane excision”) is less effective than cyst or ventriculoperitoneal shunting [11, 18,20,47]. Endoscopic third ventriculostomy with or without choroid plexus cauterization can also be effective [48]. After treatment, cysts sometimes collapse and ventricles may return to normal size. Epidemiologic data from 2015 suggest that many DWM patients do not receive adequate care because of poor access to neurosurgery [49].

Future investigations Finding additional DWM causative genes is a major focus of continuing research, as currently known genes account for less than 5% of cases. The possibility that some DWM arises from de novo mutations also remains to be explored. To truly appreciate the causes of DWM, the early developmental changes that lead to this phenotype must be identified. Better understanding through animal models, and greater progress in characterizing gene expression patterns and functions, are important goals. Pathologists can contribute by publishing additional and more detailed characterizations of DWM, BPC, and MCM in fetuses and children.

References 1. Barkovich AJ, Kjos BO, Norman D, Edwards MS (1989) Revised classification of posterior fossa cysts and cystlike malformations based on the results of multiplanar MR imaging. AJR Am J Roentgenol 153:1289–300. 2. Kollias SS, Ball WS Jr, Prenger EC (1993) Cystic malformations of the posterior fossa: differential diagnosis clarified through embryologic analysis. Radiographics 13:1211–31 3. Robinson AJ, Ederies MA (2016) Diagnostic imaging of posterior fossa anomalies in the fetus. Semin Fetal Neonatal Med 39(1):1–8 4. Golden JA, Rorke LB, Bruce DA (1987) Dandy–Walker syndrome and associated anomalies. Pediatr Neurosci 13:38–44 5. Parisi MA, Dobyns WB (2003) Human malformations of the midbrain and hindbrain: review and proposed classification scheme. Mol Genet Metab 80:36–53

6. Chapman T, Mahalingam S, Ishak GE et al. (2015) Diagnostic imaging of posterior fossa anomalies in the fetus and neonate: part 2, Posterior fossa disorders. Clin Imaging 39:167–75 7. Aldinger KA, Doherty D (2016) The genetics of cerebellar malformations. Semin Fetal Neonatal Med 21(5):321–32 8. Dandy WE, Blackfan KD (1914) Internal hydrocephalus: an experimental, clinical and pathological study. Am J Dis Child 8:406–82 9. Tortori-Donati P, Fondelli MP, Rossi A, Carini S (1996) Cystic malformations of the posterior cranial fossa originating from a defect of the posterior membranous area. Mega cisterna magna and persisting Blake’s pouch: two separate entities. Childs Nerv Syst 12:303–8 10. Raybaud C, Levrier O, Brunel H et al. (2003) MR imaging of fetal brain malformations. Childs Nerv Syst 19:455–70 11. Azab WA, Shohoud SA, Elmansoury TM et al. (2014) Blake’s pouch cyst. Surg Neurol Int 5:112 12. Strand RD, Barnes PD, Poussaint TY et al. (1993) Cystic retrocerebellar malformations: unification of the Dandy-Walker complex and the Blake’s pouch cyst. Pediatr Radiol 23:258–60 13. Aldinger KA, Lehmann OJ, Hudgins L et al. (2009) FOXC1 is required for normal cerebellar development and is a major contributor to chromosome 6p25.3 Dandy–Walker malformation. Nat Genet 41:1037–42 14. Hart MN, Malamud N, Ellis WG (1972) The Dandy–Walker syndrome. A clinicopathological study based on 28 cases. Neurology 22:771–80 15. Nelson MD Jr, Maher K, Gilles FH (2004) A different approach to cysts of the posterior fossa. Pediatr Radiol 34:720–32 16. Benda CE (1954) The Dandy-Walker syndrome or the so-called atresia of the foramen Magendie. J Neuropathol Exp Neurol 13:14– 29 17. Taggart JK, Walker AE (1942) Congenital atresia of the foramens of Luschka and Magendie. Arch Neurol Psych 48:583–612 18. Hirsch JF, Pierre-Kahn A, Renier D et al. (1984) The Dandy–Walker malformation. A review of 40 cases. J Neurosurg 61:515–22 19. Pascual-Castroviejo I, Velez A, Pascual-Pascual SI et al. (1991) Dandy–Walker malformation: analysis of 38 cases. Childs Nerv Syst 7:88–97 20. Altman NR, Naidich TP, Braffman BH (1992) Posterior fossa malformations. AJNR Am J Neuroradiol 13:691–724 21. Jha VC, Kumar R, Srivastav AK et al. (2012) A case series of 12 patients with incidental asymptomatic Dandy-Walker syndrome and management. Childs Nerv Syst 28:861–7 22. Gandolfi Colleoni G, Contro E, Carletti A et al. (2012) Prenatal diagnosis and outcome of fetal posterior fossa fluid collections. Ultrasound Obstet Gynecol 39:625–31 23. Poretti A, Boltshauser E, Doherty D (2014) Cerebellar hypoplasia: differential diagnosis and diagnostic approach. Am J Med Genet C Semin Med Genet 166C:211–26 24. Chapman T, Mahalingam S, Ishak GE et al. (2015) Diagnostic imaging of posterior fossa anomalies in the fetus and neonate: part 1, normal anatomy and classification of anomalies. Clin Imaging 39:1–8 25. Phillips JJ, Mahony BS, Siebert JR et al. (2006) Dandy–Walker malformation complex: correlation between ultrasonographic diagnosis and postmortem neuropathology. Obstet Gynecol 107:685–93 26. Bernardo S, Vinci V, Saldari M et al. (2015) Dandy–Walker Malformation: is the ‘tail sign’ the key sign? Prenat Diagn 35:1358–64 27. Siebert JR (2006) A pathological approach to anomalies of the posterior fossa. Birth Defects Res A Clin Mol Teratol 76:674–84

149

Developmental Neuropathology 28. Shekdar K (2011) Posterior fossa malformations. Semin Ultrasound CT MR 32:228–41 29. Sartori S, Ludwig K, Fortuna M et al. (2010) Dandy-Walker malformation masking the molar tooth sign: an illustrative case with magnetic resonance imaging follow-up. J Child Neurol 25: 1419–22 30. Klein JL, Lemmon ME, Northington FJ et al. (2016) Clinical and neuroimaging features as diagnostic guides in neonatal neurology diseases with cerebellar involvement. Cerebellum Ataxias 3:1 31. Benke PJ (1984) The isotretinoin teratogen syndrome. JAMA 251:3267–9 32. Kalter H (1963) Experimental mammalian teratogenesis, a study of galactoflavin-induced hydrocephalus in mice. J Morphol 112:303– 17 33. Golden JA, Schoene WC (1993) Central nervous system malformations in trisomy 9. J Neuropathol Exp Neurol 52:71–7 34. Grinberg I, Northrup H, Ardinger H et al. (2004) Heterozygous deletion of the linked genes ZIC1 and ZIC4 is involved in Dandy– Walker malformation. Nat Genet 36:1053–5 35. D.Agostino AN, Kernohan JW, Brown JR (1963) The DandyWalker syndrome. J Neuropathol Exp Neurol 22:450–70 36. Kapur RP, Mahony BS, Finch L, Siebert JR (2009) Normal and abnormal anatomy of the cerebellar vermis in midgestational human fetuses. Birth Defects Res A Clin Mol Teratol 85:700–9 37. Rorke LB, Riggs HE, Fogelson MH (1968) Cerebellar heterotopia in infancy. J Neuropathol Exp Neurol 27:140–1 38. Romani M, Micalizzi A, Valente EM (2013) Joubert syndrome: congenital cerebellar ataxia with the molar tooth. Lancet Neurol 12:894– 905 39. Blake JA (1900) The roof and lateral recesses of the fourth ventricle, considered morphologically and embryologically. J Comp Neurol 10:79–108

150

40. Brodal A, Hauglie-Hanssen E (1959) Congenital hydrocephalus with defective development of the cerebellar vermis (Dandy– Walker syndrome): clinical and anatomical findings in two cases with particular reference to the so-called atresia of the foramina of Magendie and Luschka. J Neurol Neurosurg Psychiatry 22:99–108 41. Millonig JH, Millen KJ, Hatten ME (2000) The mouse Dreher gene Lmx1a controls formation of the roof plate in the vertebrate CNS. Nature 403:764–9 42. Sgaier SK, Millet S, Villanueva MP et al. (2005) Morphogenetic and cellular movements that shape the mouse cerebellum; insights from genetic fate mapping. Neuron 45:27–40 43. Gibson JB (1955) Congenital hydrocephalus due to atresia of the foramen of Magendie. J Neuropathol Exp Neurol 14:244–62 44. Yamada H, Oi S, Tamaki N et al. (1991) Congenital hydrocephalus mimicking Dandy–Walker syndrome induced by 6aminonicotinamide injection in pregnant rat. Neurol Med Chir (Tokyo) 31:326–9 45. Blank MC, Grinberg I, Aryee E et al. (2011) Multiple developmental programs are altered by loss of Zic1 and Zic4 to cause Dandy–Walker malformation cerebellar pathogenesis. Development 138:1207–16 46. Haldipur P, Gillies GS, Janson OK et al. (2014) Foxc1 dependent mesenchymal signalling drives embryonic cerebellar growth. Elife 16:3 47. Alexiou GA, Sfakianos G, Prodromou N et al. (2010) Dandy-Walker malformation: analysis of 19 cases. J Child Neurol 25:188–91 48. Nishiyama K, Yoshimura J, Fujii Y (2015) Limitations of neuroendoscopic treatment for pediatric hydrocephalus and considerations from future perspectives. Neurol Med Chir (Tokyo) 55:611–16 49. McClelland S 3rd, Ukwuoma OI, Lunos S, Okuyemi KS (2015) The natural history of Dandy–Walker syndrome in the United States: A population-based analysis. J Neurosci Rural Pract 6:23–6

14

Joubert Syndrome Robert F. Hevner,1,2 William B. Dobyns,2,3 and Enza Maria Valente4 1

Department of Neurological Surgery, University of Washington, Seattle, WA, USA Center for Integrative Brain Research, Seattle Children’s Research Institute, Seattle, WA, USA 3 Department of Pediatrics, University of Washington, Seattle, WA, USA 4 Section of Neurosciences, Department of Medicine and Surgery, University of Salerno, Fisciano, Italy 2

Definition Joubert syndrome is a genetic ciliopathy defined by characteristic neurologic features and brain malformations. The core features of the syndrome are hypotonia, ataxia, intellectual disability, and the so-called “molar tooth sign” (MTS) on neuroimaging [1,2]. Other classic features include abnormal breathing and eye movements. Brain malformations include cerebellar vermis hypoplasia, abnormal corticospinal tracts, and diverse others. More than 30 Joubert syndrome genes have been identified. All are linked to biogenesis, structure, or function of the primary (nonmotile) cilium, an antenna-like cellular organelle that regulates intercellular and intracellular signal transduction. Accordingly, it is classified as a ciliopathy, together with related disorders such as Meckel (or Meckel–Gruber) syndrome, with which Joubert syndrome overlaps genetically. The term “Joubert syndrome-related disorders” is “applied to any individual who displays MTS irrespective of the additional non-neurological features” [2]. The disorders thus include not only “pure” neurological Joubert syndrome but also “Joubert syndrome plus” disorders that involve other organs. For instance, Meckel–Gruber syndrome is a Joubert syndromerelated disorder; patients with features of Meckel–Gruber syndrome are diagnosed with Joubert syndrome (or Joubert syndrome-related disorders) if the molar tooth sign is present, or as pure Meckel–Gruber syndrome if it is not seen. The Joubert syndrome-related disorders are ciliopathies.

Epidemiology The prevalence of Joubert syndrome is unknown, but has been estimated at 1 in 100 000 [3]. This figure is probably an underestimate [3,4]. Boys and girls are affected about equally, except

in the small minority of X-linked families in which boys are affected and girls are carriers. Joubert syndrome occurs worldwide. Among some isolated populations, such as the Ashkenazi Jews or the Hutterites, Joubert syndrome prevalence is increased due to founder effects [5,6].

Clinical features Neurologic presentation Typically, Joubert syndrome manifests at birth or infancy with neurologic problems including hypotonia, ataxia, abnormal ocular movements (oculomotor apraxia or nystagmus), and, in about 50% of patients, altered breathing patterns (episodic apneas and/or hyperpneas), which can be life-threatening but usually improve with age. All patients experience a global developmental delay, and most will manifest intellectual disability of variable degree, with major difficulties in expressive language. Ambulant patients develop gait ataxia. Seizures develop in approximately 10% of cases [7]. Non-neurologic features and subgroups Core neurological manifestations of Joubert syndrome may be associated with defects of other organs, defining six main subgroups: 1. isolated or “pure” Joubert syndrome (about one-third) 2. Joubert syndrome plus retina 3. Joubert syndrome plus kidney 4. cerebello-oculorenal syndromes 5. Joubert syndrome plus liver, including COACH syndrome (cerebellar vermis hypo/aplasia, oligophrenia, congenital ataxia, coloboma and hepatic fibrosis) 6. oral facial digital syndrome type VI (OFD-VI), also called Varadi–Papp syndrome [2,8].

Developmental Neuropathology, Second Edition. Edited by Homa Adle-Biassette, Brian N. Harding, and Jeffrey A. Golden. © 2018 John Wiley & Sons Ltd. Published 2018 by John Wiley & Sons Ltd.

151

Developmental Neuropathology Analyses of large Joubert syndrome cohorts support this classification, but also demonstrate heterogeneity and overlap among subgroups [7]. Thus, Joubert syndrome has sometimes been characterized as a spectrum disorder [2]. Retinal defects (found in around 30% of cases) range from Leber congenital amaurosis (a Joubert syndrome-related disorder), to milder retinopathies with partially conserved vision, to chorioretinal colobomas [1,9,10]. Cystic kidney disease (25–30%) represents one of the most severe complications, manifesting as juvenile (or, rarely, infantile) onset nephronophthisis (medullary cystic kidneys). This slowly progressive, tubulointerstitial fibrosis may remain paucisymptomatic for years, before progressing to chronic or acute renal failure. Hepatic involvement in Joubert syndrome (around 18%) was underreported in the past [3], but 2015 data revealed that close to one-fifth of patients with Joubert syndrome have liver dysfunction [7]. Congenital liver fibrosis is the most severe presentation; more often, disease at birth is represented only by elevation of liver enzymes. In OFD-VI, Joubert syndrome neurological features are associated with mesoaxial polydactyly, and/or oral–facial defects (tongue hamartomas, additional frenula, or notched upper lip), and/or hypothalamic hamartoma. Preaxial polydactyly is also common in OFD-VI, while postaxial polydactyly can be found in all Joubert syndrome subgroups [11]. Additional clinical problems associated with Joubert syndrome include strabismus, ptosis, facial dysmorphisms, hearing impairment, scoliosis, cleft palate, cardiac defects, endocrine

(a)

abnormalities, altered laterality, and neuropsychiatric disorders [10,37]. Life expectancy is reduced. An estimated 60% of patients with Joubert syndrome survive to five years [12], and more than 50% survive to adulthood [3].

Neuroimaging The first neuroimaging abnormality demonstrated in Joubert syndrome was agenesis of the cerebellar vermis, shown by pneumoencephalography [13]. Cerebellar vermis agenesis (or hypoplasia) remains a key finding, but is only part of the picture.

The “molar tooth” sign With the advent of magnetic resonance imaging (MRI), additional hallmarks of Joubert syndrome were observed, culminating in identification of the molar tooth sign [14,15]. The sign is a tooth-like morphologic pattern seen in axial (transverse) images through the upper pons and ventral midbrain, known as the isthmus region (Figure 14.1). The morphology is produced by peculiar malformations of the brainstem and cerebellum that, taken together, are characteristic of Joubert syndrome: hypoplastic cerebellar vermis; thickened, rotated (“horizontalized”), elongated superior cerebellar peduncles (SCPs); and deep interpeduncular fossa with widely separated corticospinal tracts. In axial slices, these anomalies together generate the appearance of a molar tooth, with crown (the basis

(b)

Figure 14.1 Neuroimaging of Joubert syndrome: the molar tooth sign. (a) Sagittal plane: the vermis is very small, but found in a normal position (yellow arrowheads). The cerebellar tissue filling the posterior fossa behind the vermis is hemisphere tissue, since the imaging plane was not perfectly parallel to midline. The fourth ventricle (white asterisk) is enlarged. (b) Axial image shows thick and long superior cerebellar peduncles (yellow arrowheads), forming the roots of the so-called “molar tooth” sign. The basis pontis and ventral midbrain form the tooth crown (blue arrowheads). Note the fluid-filled midline space (black asterisk), resembling a Dandy–Walker type cyst.

152

Joubert Syndrome Chapter 14

pontis or ventral midbrain) and two roots (the superior cerebellar peduncles); hence, the name molar tooth sign. The sign is essentially pathognomonic of Joubert syndrome, even in utero [2,3,10,15–17]. Interestingly, molar tooth sign morphology varies among individuals with Joubert syndrome; the asymmetric “decaying molar tooth” is one variant [18]. Additional variables that may influence morphology include the degree of nondecussation of superior cerebellar peduncles and separated corticospinal tracts, as shown by diffusion tensor imaging and tractography [19,20], the amount of fourth ventricular enlargement and deformation, with “bat-wing” appearance on axial cuts [21], and the extent of isthmus narrowing [16].

Other brain malformations seen by neuroimaging In at least 30% of cases of Joubert syndrome, neuroimaging will detect other associated brain malformations, in addition to cerebellar vermis hypoplasia and the molar tooth sign [7,18]. Ventriculomegaly is among the most frequent associations (approximately 10%). Similarly, around 10% have retrocerebellar fluid collections resembling Dandy–Walker cysts [16,22]. Less frequent associations include encephalocele or meningocele in the occipital or foramen magnum region (around 8%), callosal agenesis (3–9%), cerebrocortical heterotopia (around 3%), polymicrogyria (around 1%), hypothalamic hamartoma (in OFD-VI), heterotopia in the dorsal caudal medulla, and dysplastic, enlarged, or hypoplastic cerebellar hemispheres [7,15,17,18,23]. Since not all patients with Joubert (a)

(b)

Figure 14.2 Macroscopic features of the brainstem and cerebellum in Joubert syndrome. The patient was 13 months old at death, and had TCTN2 mutations (subject 4 [17]). (a) Ventral view (anterior up) shows pyramids separated by a deep cleft (blue arrowheads), with little or no decussation. (b) Inferior view (dorsal up) shows enlarged cerebellar hemispheres wrapping almost circumferentially around the medulla. The fourth ventricle (white asterisk) was enlarged and communicated

syndrome have comprehensive neuroimaging, brain malformations may be more prevalent than indicated by current data.

Fetal neuroimaging The molar tooth sign may be seen as early as 17–18 weeks of gestation by fetal MRI [24], although not all fetal brains manifest the sign so early [17]. Fetal ultrasound is less reliable for detecting it, and may show only cerebellar vermis hypoplasia. Cases of cerebellar hypoplasia diagnosed by fetal ultrasound should be referred for MRI to complete the neuroimaging work-up [25].

Neuropathology The neuropathology of Joubert syndrome includes several “core” malformations found in the cerebellum and brainstem of all cases, plus additional diverse anomalies that each appear in a fraction of individuals with the syndrome, producing tremendous pathologic heterogeneity. While the cerebellum and brainstem are most consistently affected, Joubert syndrome can affect all levels of the central nervous system from spinal cord to cerebral cortex. Joubert syndrome neuropathology has been studied in fetuses as young as 20 weeks of gestation [17].

Cerebellum By definition, all Joubert syndrome brains show agenesis or hypoplasia (usually marked or moderate, rarely mild) of the cerebellar vermis (Figure 14.2). The cerebellar hemispheres may (c)

with a fluid-filled space covered by a thin membrane, reminiscent of a Dandy–Walker cyst. (c) Transverse slice at the level of the upper pons shows only a tiny rudiment of vermis in the anterior midline; almost all of the cerebellar tissue is hemispheres. The fourth ventricle exhibits a “bat-wing” or “inverted umbrella” appearance. Inset shows a superior view (dorsal up), with enlarged superior cerebellar peduncles (yellow arrowheads) connecting to the midbrain.

153

Developmental Neuropathology be dysplastic, atrophic, or enlarged [17,26,27]. If enlarged, the hemispheres may grow anteriorly around the lateral brainstem (Figure 14.2b). Histologically, the cerebellar cortex varies from normal to atrophic. Heterotopia containing Purkinje cells, granule neurons, or deep nuclei neurons may be present in the subcortical white matter [13,17,26]. Fragmentation of the dentate nuclei into islands (Figure 14.3a), rather than the normal undulating ribbons, is a consistent finding in Joubert syndrome [17].

(a)

Brainstem Elongated separated corticospinal tracts (CSTs) and wide interpeduncular fossa are molar tooth sign anomalies present in all cases of Joubert syndrome. Indeed, the sign can be demonstrated at brain cutting, in axial slices through the isthmus [27]. Other consistent brainstem anomalies include fragmentation of spinal trigeminal nuclei; hypoplasia or dysplasia of pontine, inferior olivary, and solitary nuclei; and nondecussation

(b)

(c)

Figure 14.3 Histologic anomalies in the Joubert syndrome brainstem and cerebellum. (a) The dentate nuclei of the cerebellum, rather than forming a ribbon, are broken up into small islands. Neurofilament immunohistochemistry. (b) Cross section through the cervicomedullary junction of a patient with posterior medullary protuberance [17,18]. The dorsal region (normal location of dorsal column nuclei) is extremely disorganized (blue arrowheads), and no posterior median sulcus is present. Boxed area is the normal location of the pyramidal decussation, where

154

very few crossing fibers were seen. Myelin basic protein immunohistochemistry. (c) Sagittal section (anterior left) through brainstem and cerebellum of a fetus at 22 weeks of gestation with Joubert syndrome and posterior medullary protuberance (arrow). The corticospinal tracts (asterisk) was tightly bundled and did not decussate or shift to dorsolateral funiculi. Calretinin immunohistochemistry (adapted from Juric-Sekhar et al. [17]).

Joubert Syndrome Chapter 14

(complete or partial) of CSTs at the cervicomedullary junction. Brainstem findings reported in some, but not all Joubert syndrome cases include: variable nondecussation of SCPs; enlarged arcuate nuclei; hypoplastic reticular formation; and hypoplastic medial lemnisci. The inferior olives and pyramids are also sometimes poorly demarcated, and gray-matter heterotopia may be present in the cerebral peduncle. The posterior fossa may be enlarged, and some cases (especially Meckel–Gruber syndromelike) may have a posterior (occipital or foramen magnum) meningo- or encephalocele.

Dorsal medullary protuberance Dysplasia of the caudal medulla at the level of the cervicomedullary junction, involving dorsal column nuclei, is an intriguing malformation described in more than 50% of Joubert syndrome brains with detailed neuropathologic study [17,26,27]. Posterior medullary structures are severely disorganized, and the posterior median sulcus is absent (Figures 14.3b,c). This anomaly correlates with a “posterior medullary protuberance” on sagittal MRI [17]. By neuroimaging, the posterior protuberance is reported in only 3% of all cases of Joubert syndrome [18], but neuropathology suggests this anomaly occurs more frequently. The fourth ventricle is slightly to moderately enlarged, and, in slices through pontine levels, has the shape of a bat’s wing [21] or inverted umbrella [27]. The fourth ventricle may be continuous with a large midline fluid collection, mimicking the Dandy–Walker malformation (Figure 14.1b). The cerebral aqueduct may be enlarged. Spinal cord Nondecussation of separated corticospinal tracts, and other tract abnormalities, may result in small dorsolateral funiculi, enlarged anterior funiculi, and aberrant centrally located fiber bundles in the cord [17,22]. In addition, the dorsal spinal cord may be dysplastic and lack a posterior median sulcus at some levels. Cerebral hemispheres and diencephalon Supratentorial anomalies are found in small minorities of cases, but can be severe [18,22]. Reported anomalies include ventriculomegaly (around 10%), agenesis or dysgenesis of the corpus callosum (3–9%), occipital meningocele or cephalocele (around 8%), hippocampal malrotation, polymicrogyria, periventricular nodular heterotopia, and (in OFD-VI) hypothalamic hamartoma. The thalamus may also be fused or dysplastic, and basal ganglia may contain heterotopia [17,18]. Supratentorial malformations may be more common and severe in the OFD-VI subgroup of Joubert syndrome [11,28,29].

Differential diagnosis While the clinical presentation may be relatively nonspecific and requires differentiation from other congenital nonprogressive ataxias, the diagnosis of Joubert syndrome is unequivocally

made on axial neuroimaging, showing the pathognomonic molar tooth sign. Most patients can be assigned to a subgroup on the basis of other clinical findings. Neuropathology remains important not only to confirm the diagnosis (especially in fetal cases), but also to fully evaluate brain and spinal cord malformations, given the extreme heterogeneity of Joubert syndrome, and the need for better genotype–phenotype correlations. In some cases, a Dandy–Walker malformation may be suspected, on the basis of dilatation of the fourth ventricle, atresia of the basal foramina, or enlarged posterior fossa, in conjunction with vermian hypoplasia [22,26]. Indeed, Dandy–Walker malformations (combining vermian hypoplasia, cystic fourth ventricle, and enlarged posterior fossa) can occur in Joubert syndrome, and may obscure the molar tooth sign [30].

Genetics It has long been recognized that Joubert syndrome is familial, and typically shows autosomal recessive inheritance [13]. However, genetic studies have found that the Joubert syndrome phenotype can be caused by mutations in many different genes (over 30). Thus, Joubert syndrome has become “a model for untangling recessive disorders with extreme genetic heterogeneity” [7].

Inheritance and genes Most Joubert syndrome inheritance is autosomal recessive (affecting boys and girls), but it can also be caused by X-linked genes, such as OFD1 (affecting boys only). Most cases are familial, but de novo mutations may contribute to sporadic Joubert syndrome [31]. Using targeted exome sequencing approaches, current DNA methods can identify causative mutations in 62% of cases [7]. All varieties of mutations are found, although nonexonic (regulatory sequence) mutations require further study. Consistent with the extreme genetic heterogeneity of Joubert syndrome, no single gene is associated with more than 10% of cases [7]. The top five causative genes across all subgroups are C5ORF42 (approximately 9% of cases), CC2D2A (around 8%), CEP290 (around 7%), AHI1 (around 7%), and TMEM67 (around 6%). Importantly, all Joubert syndrome genes are involved in primary cilium biology. Genotype–phenotype correlations Only relatively weak genotype–phenotype correlations have been detected so far. Each Joubert syndrome gene seems able to cause multiple subgroup phenotypes, as well as related-disorder phenotypes without the molar tooth sign (such as Meckel– Gruber syndrome). For example, C5ORF42 mutations have been linked to pure Joubert syndrome, and to the OFD-VI subgroup [7]. Some genes do exhibit a tendency to cause certain Joubert syndrome phenotypes, such as TMEM216 and cerebellooculorenal syndromes [2]. Similarly, TMEM67 mutations are associated with greater likelihood of liver disease, especially

155

Developmental Neuropathology COACH syndrome [1]. Interestingly, CC2D2A mutations are associated with increased probability of ventriculomegaly and seizures [7]. The dysplastic posterior medullary protuberance is not gene specific, as it was observed in Joubert syndrome caused by OFD1 and TCTN2 mutations [17].

Allelic Joubert syndrome-related disorders Some Joubert syndrome-related disorders (without molar tooth sign) can be allelic with Joubert syndrome, and can occur in the same families, even in siblings. For example, CEP290 mutations may cause Joubert syndrome, Meckel–Gruber syndrome, or Leber congenital amaurosis [1,2]. The Joubert syndromerelated disorders are a large group (more than a dozen) with diverse clinical phenotypes and overlapping genetics [2,3]. Most Joubert syndrome-related disorders exhibit autosomal recessive inheritance.

Pathogenesis Primary cilia Joubert syndrome is caused by defective biogenesis, structure, or function of primary cilia due to genetic mutations. The primary cilium is nonmotile, and has a 9 + 0 axoneme structure; in contrast, motile cilia (such as those on ependymal cells) have 9 + 2 axoneme structure. Primary cilia can be highly specialized in different organs; for example, photoreceptor outer segments are a type of primary cilia. Primary cilia generally function as antenna-like signaling receptor organelles on most cell types, including neural progenitors and neurons [32,33]. Specifically, primary cilia mediate polarized transduction of physical (e.g., light) and molecular (e.g., Wnt) signals that vary by cell and tissue type; thus, primary cilia serve to integrate cell polarity and extracellular signaling. Reflecting their evolutionary relationship to flagella, primary cilia contain no vesicles or vesicular transport apparatus. Instead, primary cilia rely on unique mechanisms, related to flagellar biogenesis, intraflagellar transport, and membrane turnover. Ciliopathy genes encode proteins that not only localize to primary cilia, but also mediate their specialized functions. In the developing brain, primary cilia are especially important for Sonic hedgehog (Shh) and canonical Wnt signaling. Since cerebellar growth is driven in part by Wnt and Shh signaling [34], cerebellar vermis hypoplasia is one logical outcome of ciliopathy. However, primary cilia transduce other signals as well, and mediate diverse processes in neurodevelopment, including apicobasal patterning, dorsoventral patterning, neurogenesis, cell migration, and axon guidance. For some phenotypes, such as nondecussation of corticospinal tracts, the links to primary cilia signaling are not yet complete, but important steps in understanding Joubert syndrome pathogenesis have been made, most recently through studies in animals.

156

Animal models Research in animal models (especially mice) has been crucial to understanding the dynamics and functions of primary cilia in developing brain. Several knockout mutant mice deficient in Joubert syndrome genes have been studied, and functions of primary cilia in different contexts have been elucidated. In mice with Ahi1 or Cep290 mutations, which perturb intraciliary transport, mild cerebellar vermis hypoplasia was observed [35]. Impaired growth at the cerebellar midline was linked to deficient Wnt signaling. Compensatory expansion of the fourth ventricular roof plate was also noted. A different line of Cep290 homozygous mutant mice died of hydrocephalus [36]. Mice lacking Arl13b, a Joubert syndrome gene important for intraciliary transport and signal transduction, exhibited severe anomalies of the cerebral cortex [37]. The apicobasal polarity of radial glia (neural stem cells), which have their primary cilia at the ventricular surface, was severely perturbed, and neurons differentiated ectopically. Targeted studies of interneurons showed that Arl13b was also necessary for correct directionality of interneuron migration [32]. In contrast, expression of Arl13b in projection neurons was important for axon outgrowth and connections, but not migration. Ofd1 mutant mice exhibited aberrant dorsoventral forebrain patterning, and defective corticogenesis [38]. Interestingly, Shh signaling was aberrantly increased in cerebral cortex, due to dysregulation of Shh target genes.

Treatment Currently, there are no effective specific therapies for Joubert syndrome. Management is focused on motor, cognitive and neurovisual rehabilitation and on prevention or treatment of more severe complications, such as prolonged apneas in the neonatal period, or kidney failure in late childhood. The improved knowledge of primary cilia pathophysiology should help develop more directed, and personalized, therapeutic approaches for Joubert syndrome patients. For example, Shh and Wnt agonists might have a role, and modulation of intracellular signaling pathways might facilitate more balanced signaling [33]. For patients with known mutations, treatment approaches may be tailored to each patient for optimal effects.

References 1. Doherty D (2009) Joubert syndrome: insights into brain development, cilium biology, and complex disease. Semin Pediatr Neurol 16:143–54 2. Sattar S, Gleeson JG (2011) The ciliopathies in neuronal development: a clinical approach to investigation of Joubert syndrome and Joubert syndrome-related disorders. Dev Med Child Neurol 53: 793–8

Joubert Syndrome Chapter 14 3. Parisi MA (2009) Clinical and molecular features of Joubert syndrome and related disorders. Am J Med Genet C Semin Med Genet 151C:326–40 4. Valente EM, Brancati F, Boltshauser E, Dallapiccola B (2013) Clinical utility gene card for: Joubert syndrome: update 2013. Eur J Hum Genet 21(10) doi: 10.1038/ejhg.2013.10 5. Valente EM, Logan CV, Mougou-Zerelli S et al. (2010) Mutations in TMEM216 perturb ciliogenesis and cause Joubert, Meckel and related syndromes. Nat Genet 42:619–25 6. Huang L, Szymanska K, Jensen VL et al. (2011) TMEM237 is mutated in individuals with a Joubert syndrome related disorder and expands the role of the TMEM family at the ciliary transition zone. Am J Hum Genet 89:713–30 7. Bachmann-Gagescu R, Dempsey JC, Phelps IG et al. (2015) Joubert syndrome: a model for untangling recessive disorders with extreme genetic heterogeneity. J Med Genet 52:514–22 8. Brancati F, Dallapiccola B, Valente EM (2010) Joubert syndrome and related disorders. Orphanet J Rare Dis 5:20 9. Romani M, Micalizzi A, Valente EM (2013) Joubert syndrome: congenital cerebellar ataxia with the molar tooth. Lancet Neurol 12:894– 905 10. Poretti A, Boltshauser E, Valente EM (2014) The molar tooth sign is pathognomonic for Joubert syndrome! Pediatr Neurol 50:e15–16 11. Poretti A, Vitiello G, Hennekam RC et al. (2012) Delineation and diagnostic criteria of Oral-Facial-Digital Syndrome type VI. Orphanet J Rare Dis 7:4 12. Saraiva JM, Baraitser M (1992) Joubert syndrome: a review. Am J Med Genet 43:726–31 13. Joubert M, Eisenring JJ, Robb JP, Andermann F (1969) Familial agenesis of the cerebellar vermis. A syndrome of episodic hyperpnea, abnormal eye movements, ataxia, and retardation. Neurology 19:813–25 14. Maria BL, Hoang KB, Tusa RJ et al. (1997) “Joubert syndrome” revisited: key ocular motor signs with magnetic resonance imaging correlation. J Child Neurol 12:423–30 15. Maria BL, Quisling RG, Rosainz LC et al. (1999) Yachnis AT, Gitten J, Dede D, Fennell E. Molar tooth sign in Joubert syndrome: clinical, radiologic, and pathologic significance. J Child Neurol 14:368–76 16. Maria BL, Bozorgmanesh A, Kimmel KN et al. (2001) Quantitative assessment of brainstem development in Joubert syndrome and Dandy–Walker syndrome. J Child Neurol 16:751–8 17. Juric-Sekhar G, Adkins J, Doherty D, Hevner RF (2012) Joubert syndrome: brain and spinal cord malformations in genotyped cases and implications for neurodevelopmental functions of primary cilia. Acta Neuropathol 123:695–709 18. Poretti A, Huisman TA, Scheer I, Boltshauser E (2011) Joubert syndrome and related disorders: spectrum of neuroimaging findings in 75 patients. AJNR Am J Neuroradiol 32:1459–63 19. Poretti A, Boltshauser E, Loenneker T et al. (2007) Diffusion tensor imaging in Joubert syndrome. AJNR Am J Neuroradiol 28:1929–33 20. Hsu CC, Kwan GN, Bhuta S (2015) High-resolution diffusion tensor imaging and tractography in Joubert syndrome: beyond molar tooth sign. Pediatr Neurol 53:47–52

21. McGraw P (2003) The molar tooth sign. Radiology 229:671–2 22. ten Donkelaar HJ, Hoevenaars F, Wesseling P (2000) A case of Joubert’s syndrome with extensive cerebral malformations. Clin Neuropathol 19:85–93 23. Doherty D, Millen KJ, Barkovich AJ (2013) Midbrain and hindbrain malformations: advances in clinical diagnosis, imaging, and genetics. Lancet Neurol 12:381–93 24. Saleem SN, Zaki MS, Soliman NA, Momtaz M (2011) Prenatal magnetic resonance imaging diagnosis of molar tooth sign at 17 to 18 weeks of gestation in two fetuses at risk for Joubert syndrome and related cerebellar disorders. Neuropediatrics 42:35–8 25. Chapman T, Mahalingam S, Ishak GE et al. (2015) Diagnostic imaging of posterior fossa anomalies in the fetus and neonate: part 2, posterior fossa disorders. Clin Imaging 39:167–75 26. Friede RL, Boltshauser E (1978) Uncommon syndromes of cerebellar vermis aplasia. I: Joubert syndrome. Dev Med Child Neurol 20:758–63 27. Yachnis AT, Rorke LB (1999) Neuropathology of Joubert syndrome. J Child Neurol 14:655–9 28. Poretti A, Brehmer U, Scheer I et al. (2008) Prenatal and neonatal MR imaging findings in oral-facial-digital syndrome type VI. AJNR Am J Neuroradiol 29:1090–1 29. Takanashi J, Tada H, Ozaki H, Barkovich AJ (2009) Malformations of cerebral cortical development in oral–facial–digital syndrome type VI. AJNR Am J Neuroradiol 30:E22–3 30. Sartori S, Ludwig K, Fortuna M et al. (2010) Dandy–Walker malformation masking the molar tooth sign: an illustrative case with magnetic resonance imaging follow-up. J Child Neurol 25: 1419–22 31. Srour M, Hamdan FF, McKnight D et al. (2015) Joubert syndrome in French Canadians and identification of mutations in CEP104. Am J Hum Genet 97:744–53 32. Higginbotham H, Eom TY, Mariani LE et al. (2012) Arl13b in primary cilia regulates the migration and placement of interneurons in the developing cerebral cortex. Dev Cell 23:925–38 33. Guemez-Gamboa A, Coufal NG, Gleeson JG (2014) Primary cilia in the developing and mature brain. Neuron 82:511–21 34. Corrales JD, Blaess S, Mahoney EM, Joyner AL (2006) The level of sonic hedgehog signaling regulates the complexity of cerebellar foliation. Development 133:1811–21 35. Lancaster MA, Gopal DJ, Kim J et al. (2011) Defective Wntdependent cerebellar midline fusion in a mouse model of Joubert syndrome. Nat Med 17:726–31 36. Rachel RA, Yamamoto EA, Dewanjee MK et al. (2015) CEP290 alleles in mice disrupt tissue-specific cilia biogenesis and recapitulate features of syndromic ciliopathies. Hum Mol Genet 24: 3775–91 37. Higginbotham H, Guo J, Yokota Y et al. (2013) Arl13b-regulated cilia activities are essential for polarized radial glial scaffold formation. Nat Neurosci 16:1000–7 38. D’Angelo A, De Angelis A, Avallone B et al. (2012) Ofd1 controls dorso-ventral patterning and axoneme elongation during embryonic brain development. PLoS One 7:e52937

157

15

Cerebellar Heterotopia and Dysplasia Marie Rivera-Zengotita and Anthony T. Yachnis Department of Pathology, Immunology, and Laboratory Medicine, University of Florida College of Medicine, Gainesville, FL, USA

Definitions

Cerebellar heterotopia of infancy

This chapter covers a spectrum of developmental disturbances, which at one extreme includes commonly encountered collections of mature and immature cerebellar cells in abnormal areas, and at the other includes more striking but much less common malformations involving the cerebellar hemispheres, vermis, or the entire structure. A rigid definition of this morphologically heterogeneous group of abnormalities is problematic. The most extensive cerebellar anomalies are usually encountered in the setting of more generalized cerebral and systemic malformations; the latter are emphasized because of their more obvious relationship to clinical symptoms. Most attempts to classify cerebellar malformations have been based on magnetic resonance imaging [1,2] and this has provided a practical approach to clinicoradiologic correlation. However, while the genetic basis for cerebellar maldevelopment is being actively studied at the molecular level, relatively few detailed neuropathologic studies are available to define the morphologic spectrum of disease. In this chapter, we consider cerebellar heterotopia of infancy, dysplasias associated with more extensive cerebral malformations, and rhombencephalosynapsis, a syndrome of vermian aplasia with apparent fusion of the cerebellar hemispheres. The term “heterotopia” will refer to abnormal location of gray matter, typically within the white matter. In infants, cerebellar heterotopias may contain collections of immature neuroepithelial cells that resemble cells of the external granular layer, in addition to more mature neurons and glia, reflecting the active histogenesis of this structure late in gestation and into early postnatal life. The term “dysplasia” will be used in a broader sense, referring to abnormalities of tissue development and organization.

Misplaced and/or disorganized collections of mature neurons, glial cells or immature neuroepithelial cells in various combinations are relatively common in the cerebellum [3–5]. Generally referred to as “heterotopia,” these developmental cell rests may be identified in over 50% of otherwise normal infants and are much more common than cerebral heterotopia (Chapter 7). To some extent, such changes disappear or regress, although their occasional persistence into adulthood is well-documented [4,6]. While small cell rests probably represent normal variations of development, heterotopias appear to be more prominent in association with genetic syndromes such as trisomy 13, trisomy 18, cerebellar hypoplasias, and other migration disorders suggesting that some of these heterotopias may be malformative. Four basic histopathologically distinct types of cerebellar heterotopia have been described [7]: compact groups of mature neurons, immature granule cell collections (focal and perivascular), poorly organized mixed cell rests containing mature neurons and immature neuroepithelial cells (“heterotaxia” of Brun), and mixed cell rests with normal relationships of the cerebellar cortex represented (“heterotopia” of Brun). Examples of these histologic changes are shown in Figure 15.1. Foci of mature neurons are most often encountered within or adjacent to the dentate nucleus or in the white matter. Granule cell collections are most frequent in the dentate and roof nuclei. The latter tend to accumulate in the parenchyma around vessels; a location that may represent an extension of the subpial external granule cell layer into the cerebellum along the path of penetrating blood vessels [6,7]. Mixed cell rests occur most frequently near the midline, as lateral projections of the nodulus of the vermis (Figure 15.1), roof nuclei, and flocculonodular lobe. Heterotopic collections of mature neurons often include a

Developmental Neuropathology, Second Edition. Edited by Homa Adle-Biassette, Brian N. Harding, and Jeffrey A. Golden. © 2018 John Wiley & Sons Ltd. Published 2018 by John Wiley & Sons Ltd.

159

Developmental Neuropathology

(a)

(b)

(c)

(d)

(e)

(f)

Figure 15.1 Cerebellar dysplasia/heterotopia. (a) Low-magnification view showing a localized area of disorganized cerebellar cytoarchitecture just lateral to the nodulus of the vermis. (b) Immature-appearing granular cell collections arranged in perivascular configuration. (c) Poorly organized mixed cell rest composed of mature neurons and immature granular cell collections. (d) Mixed cell rest with perivascular granular cells separated from Purkinje-like neurons (bottom

160

of figure) by a cell poor molecular-like layer (“heterotopia” of Brun). (e) Large disorganized mixed cell rest (“heterotaxia” of Brun) in the floculus of a full-term infant (photograph contributed by Dr. Lucy B. Rorke, Children’s Hospital of Philadelphia). (f) Dysplasia of dentate nucleus in a full-term infant with trisomy 18. The structure is fragmented into several islands of gray matter.

Cerebellar Heterotopia and Dysplasia Chapter 15

peripherally placed population of astrocytes with cell processes directed toward the center of the nodule (Figure 15.2). This observation suggests that misplaced clusters of mature neurons may be induced to aggregate in the vicinity of abnormal collections of astrocytes (Figure 15.2c). Abnormal organization of the dentate nucleus may take the form of a simplified (a)

(b)

(non-undulating) C-shaped structure in a fetus of more advanced gestational age or may present as multiple detached islands of gray matter (Figure 15.1f). Such changes can be associated with migrational abnormalities elsewhere in the brain or may occasionally occur in isolation (Table 15.1) [3,8]. The frequency with which minor cerebellar heterotopias are encountered and regression of these rests, especially those composed of granule cells [5,6,8], suggests minimal clinical significance. However, the occasional persistence of immature (or possibly progenitor) cells beyond infancy and early childhood, especially in regions along the midline roof of the fourth ventricle, have led some to suggest a possible etiologic relationship with primitive neuroectodermal tumors (cerebellar medulloblastoma) [6,9]. No clear animal models exist for studying these common anomalies of the cerebellum, although similar lesions have been produced in the chick cerebellum by disrupting the bone morphogenetic protein signaling pathway [10]. BMPs participate in cerebellar granule cell differentiation [11]; the premature differentiation of granule cells is believed to account for the development of the heterotopia in at least one model system [10], possibly reflecting the etiology in some patients.

Cerebellar dysplasias associated with neurodevelopmental syndromes

(c)

Figure 15.2 Nodular cerebellar heterotopia. (a) Nodular collection of mature neurons within the cerebellar white matter (hematoxylin and eosin). (b) Synaptophysin immunostained section of a heterotopic collection of mature neurons. (c) Glial fibrillary acidic protein immunostained section revealing astrocyte-like cells at the periphery of a heterotopic collection of mature neurons in the deep cerebellar white matter. Glial fibrillary acidic protein immunoreactive cell processes extend toward the center of the lesion.

More extensive cerebellar dysplasias involving the cerebellar hemispheres, vermis, or the entire structure are typically associated with more generalized malformations of the cerebral hemispheres [12]. These disorders are characterized according to the cerebral abnormality, which relates more directly to clinical features such as epilepsy and cognitive defects [2]. Because of this, the pathologic spectrum of associated cerebellar defects is not as well characterized. Table 15.1 lists examples of neurodevelopmental syndromes (not intended to be comprehensive) that are associated with cerebellar dysplasias. Only those syndromes that most typically involve the cerebellum are considered here; a discussion of these and other conditions can also be found elsewhere in this book. Cerebellar dysplasias are most pronounced in the cerebroocular dysplasias that are associated with type II lissencephaly (“cobblestone” lissencephaly; Chapter 7) [13–15]. Three clinically severe disorders in this group include the congenital muscular dystrophies: Walker–Warburg syndrome, cerebroocular muscular syndrome, and Fukuyama congenital muscular dystrophy; the last has a less severe cerebral defect. Cerebellar abnormalities are required for the diagnosis of Walker– Warburg syndrome and may consist of extensive cortical dysplasia, white matter heterotopia, simplification or fragmentation of the dentate nucleus, vermian agenesis or hypoplasia, and, in some cases, the Dandy–Walker malformation [3,16]. The cerebro-ocular dysplasias have been associated with defects in glycosylation [17,18]. The genetic basis for a form of type

161

Developmental Neuropathology

Table 15.1 Examples of central nervous system malformation syndromes associated with cerebellar dysplasia. Syndrome Type II lissencephaly Walker–Warburg Cerebro-ocular muscular Fukuyama congenital muscular dystrophy Lissencephaly with cerebellar hypoplasia

Syndromes with occipital encephalocele Meckel–Gruber

Tectocerebellar dysraphia Zellweger Vermian agenesis/hypoplasia Dandy–Walker malformation Joubert Rhombencephalosynapsis

I lissencephaly with striking cerebellar hypoplasia has been identified as a mutation in RELN [19], although other genes are likely to be identified [20]. RELN is a large extracellular matrix molecule required for appropriate laminar positioning in the mouse cortex [21]. Mutations in the mouse result in an inverted cerebral cortex and disorganized cerebellum. Cerebellar dysplasias have been described in association with occipital encephalocele. In some cases, the cerebellar defect may represent more of a disruption than a true malformation [22]. In a rare sporadic condition known as “tectocerebellar dysraphia,” infants have an occipital encephalocele that contains cerebellar cortex [23]. This condition has extreme tectal deformity and midbrain “beaking,” such that this malformation overlaps with the spectrum of Chiari type II malformations (Chapter 12). The Walker–Warburg [16] and Meckel–Gruber [22] syndromes may also have occipital encephalocele with cerebellar dysplasia/hypoplasia. Extensive migrational abnormalities are a regular feature of Zellweger syndrome, which is caused by defects in several genes related to peroxisome metabolism [24]. An animal model of Zellweger syndrome [25] also has extensive cerebellar dysplasia. Septo-optic dysplasia has also rarely been associated with cerebellar hypoplasia in the Cornelia de Lange syndrome [12] and with rhombencephalosynapsis [26].

Rhombencephalosynapsis Definition Rhombencephalosynapsis (RES) is a rare malformation characterized by vermian aplasia or hypoplasia with fusion of the

162

Gene Defect

Discussed Further Chapter 7

POMT1 (9q34.1) POMGNT1 (1p34) FCMD (9q31) LIS-1 (17p13) DCX (Xq22.3-q23) RELN (7q22)

Chapter 6

Chapters 2, 3 MKS1 (17q) MKS2 (11q) MKS3 (8q24) Unknown Peroxisomal genes Heterogeneous 9q34 17p11.2 Unknown

Chapter 31 Chapter 12 Chapter 13 Chapter 14

cerebellar hemispheres [3,5,8,27,28]. It may occur in isolation or together with a variety of other malformations within or outside the central nervous system. Other cerebellar malformations with vermian agenesis or hypoplasia include the Dandy– Walker malformation (Chapter 13), Joubert syndrome (Chapter 14), rhombencephalosynapsis, tectocerebellar dysraphia, and other less well-defined entities [3,5,8].

Normal embryology Although the etiology of RES is unknown, one hypothesis is defective dorsal–ventral patterning [29,30] that affects the “isthmic organizer” at the embryonic mesencephalic–metencephalic junction. This critical organizing center for cerebellar development is defined by the caudal and rostral limits of expression of the homeobox-containing genes Otx2 and Gx2, respectively [31]. In addition, signaling molecules encoded by Fgf-8 and Wnt1 are expressed within the isthmic organizer and these genes are necessary to maintain expression of En1, En2, and Pax2 [31,32]. The transcription factor Lmx1b maintains isthmic Wnt1 expression [33]. Additional molecules that specify dorsal patterning in the nervous system include members of the TGFβ family of secreted proteins (such as the bone morphogenetic proteins) [34]. Epidemiology RES is a rare disorder, with an estimated frequency of 0.13% and unknown prevalence. More than 90 individuals have been reported in the literature, with a number of pediatric and adult cases reported by magnetic resonance imaging [23,35–39].

Cerebellar Heterotopia and Dysplasia Chapter 15

(a)

(b)

(c)

(d)

Figure 15.3 Rhombencephalosynapsis. (a) Gross section at the pontomedullary junction showing the midline location of the dentate nuclei (arrow), and small, diamond-shaped fourth ventricle (arrowhead). A vermis is not present. (b) Synaptophysin immunostained histologic section at the same level shown in (a). The closely apposed dentate nuclei form the shape of an inverted “U” (arrow) above the small fourth ventricle (arrowhead). (c) Gross section at the

mid-medullary level showing the closely apposed dentate nuclei (arrow). The vermis is absent. A nodulus-like structure (fused paramedian lobule?) fills the fourth ventricle (arrowhead). (d) Synaptophysin immunostained whole mount preparation corresponding to (c). The dentate nuclei (arrow) are closely apposed but more completely formed than in (b).

163

Developmental Neuropathology

Clinical and imaging features Clinical features and prognosis reflect the wide spectrum of severity of the disorder. Patients may have a normal intelligence quotient, a mild or more severe intellectual deficit with ataxia, dysarthria, strabismus and/or nystagmus, behavioral problems (such as obsessive–compulsive and autoaggressive behavior) [27]. A comprehensive neuroimaging study [29] of 42 cases of RES revealed a spectrum of abnormalities ranging from partial absence of the vermis, including the nodulus, anterior, and posterior vermis, to complete absence of the vermis including the nodulus. Cerebellar abnormalities of intermediate severity were also described, as well as a spectrum of associated midbrain and forebrain anomalies, including aqueductal stenosis, with fusion of the tectum and absence of olfactory bulbs and tracts, dysgenesis of the corpus callosum, absent septum pelucidum, and rarely, atypical forms of holoprosencephaly [29]. Pathology Of the few cases described in the pathology literature, associated defects may include fusion of the thalami [40] and/or inferior colliculi [41] and septo-optic dysplasia [26]. Some cases have been associated with facial or noncranial anomalies, hypoplasia of the temporal lobes, fornix, and anterior commissure, and incomplete agenesis of the corpus callosum [35]. Grossly, the cerebellum is small and pear-shaped. Cut section reveals aplasia of the vermis with apparent fusion of the cerebellar hemispheres and a diamond-shaped fourth ventricle that is constricted but not obstructed by fusion of the middle cerebellar peduncles (Figure 15.3). The dentate nuclei are closely apposed and, at more rostral levels, have the appearance of a midline inverted U-shaped structure (Figure 15.3). More complete formation of the dentate nuclei is usually present at more caudal levels. Hydrocephalus may be absent or, in cases also having aqueductal atresia, quite massive [28]. Associated abnormalities include absence of the dorsal olivary nuclei, fusion of the inferior colliculi, and rarely, septo-optic dysplasia [26,35]. A neuropathological study of 40 fetal cases of rhombencephalosynapsis [37] revealed other associated anomalies, including Purkinje cell heterotopias, agenesis of the corpus callosum, lobar holoprosencephaly, and neural tube defects. Atresia of the aqueduct and third ventricle are associated with fusion of the thalamus in some cases. Genetics As noted above, RES often occurs in combination with other central nervous system and even extra-central nervous system malformations. The most common neurodevelopmental disorder associated with RES is Gomez–Lopez–Hernandez syndrome (scalp alopecia, abnormal head shape, low-set posteriorly angulated ears, and trigeminal anesthesia) in a subset of patients [42]. Other clinical genetic categories that have been proposed for RES include a VACTERL-like (vertebral, anal, cardiac, tracheoesophageal, renal, and limb) syndrome and RES associated

164

with holoprosencephaly [37]. A study using array comparative genomic hybridization suggested genetic heterogeneity in RES [43].While not a surprising result considering the clinicopathological spectrum of this disorder, it identified micro rearrangements in a minority of cases. These included a 16p11.2 deletion, a 14q12q21.2 deletion, and unbalanced translocation t(2p;10q), and a 16p13.11 microdeletion containing two candidate genes (NDE1 and C16orf45).

Animal models No definitive animal models of RES are currently available, but one candidate molecule has been suggested in the neurological mutant mouse dreher. This mouse has a homozygous mutation of Lmx1a, which results in agenesis of the vermis with apparent fusion of the cerebellar hemispheres and inferior colliculi [44]. Lmx1a was also shown to regulate formation of the roof plate and specification of dorsal cell fates in the spinal cord and developing vertebrae [44]. The midline abnormalities of the cerebellum and caudal midbrain in this mouse mutant resemble the changes observed in patients with RES. Other genes associated with mouse cerebellar hypoplasia include Zic1, Zic4, and Foxc1 [29]. Defects in fibroblast growth factor and bone morphogenetic protein pathways which regulate dorsal midline signaling that result in holoprosencephaly could also be involved in RES. Further study of such dorsalizing factors may reveal other candidate genes that might be altered in RES and other human syndromes of vermian agenesis. Treatment Treatment is primarily supportive. Although most patients die early in life, survival into the third decade has been documented [36,45].

References 1. Demaerel P (2002) Abnormalities of cerebellar foliation and fissuration: classification, neurogenetics and clinicoradiological correlations. Neuroradiology 44:639–46 2. Patel S, Barkovich AJ (2002) Analysis and classification of cerebellar malformations. Am J Neuroradiol 23:1074–87. 3. Harding BN, Copp AJ (2008) Malformations. In: S Love, DN Louis, DW Ellison, eds., Greenfield’s Neuropathology, 8th ed., London, Arnold, pp. 335–479 4. Larroche JC, Razavi-Encha F (1987) Cytoarchitectonic abnormalities. In: NC Myrianthopoulos, ed., Handbook of Clinical Neurology, Vol. 6 (50) Malformations, Amsterdam, Elsevier Science, pp. 245–66 5. Norman MG, McGillivray BC, Kalousek DK et al. (1995) Congenital Malformations of the Brain: Pathological, Embryological, Clinical, Radiological and Genetic Aspects. New York, NY, Oxford University Press, pp. 342–50 6. Yachnis AT, Rorke LB, Trojanowski JQ (1994) Cerebellar dysplasias in humans: development and possible relationship to glial and primitive neuroectodermal tumors of the cerebellar vermis. J Neuropathol Exp Neurol 63:61–71

Cerebellar Heterotopia and Dysplasia Chapter 15 7. Rorke LB, Fogelson MH, Riggs HE (1968) Cerebellar heterotopia in infancy. Dev Med Child Neurol 10:644–50 8. Friede RL (1989) Developmental Neuropathology, 2nd ed., Berlin, Springer, pp. 347–71 9. Jay V (1996) Coexistence of cerebellar primitive neuroectodermal tumor and cerebellar dysplasia: case report. Pediatr Pathol Lab Med 16:837–43 10. Ming JE, Elkan M, Tang K, Golden JA (2002) Type I bone morphogenetic protein receptors are expressed on cerebellar granular neurons and a constitutively active form of the type IA receptor induces cerebellar abnormalities. Neuroscience 114:849–57 11. Alder J, Lee KJ, Jessell TM, Hatten ME (1999) Generation of cerebellar granule neurons in vivo by transplantation of BMP-treated neural progenitor cells. Nat Neurosci 2: 535–40 12. Hayashi M, Sakomoto K, Kurata K et al. (1996) Septo-optic dysplasia with cerebellar hypoplasia in Cornelia de Lange syndrome. Acta Neuropathol 92:625–30 13. Devisme L, Bouchet C, Gonzales M et al. (2012) Cobblestone lissencephaly: neuropathological subtypes and correlations with genes of dystroglycanopathies. Brain 135:469–82 14. Moore SA, Saito F, Chen J et al. (2002) Deletion of brain dystroglycan recapitulates aspects of congenital muscular dystrophy. Nature 418:422–5 15. Myshrall TD, Moore SA, Ostendorf AP et al. (2012) Dystroglycan on radial glia end feet is required for pial basement membrane integrity and columnar organization of the developing cerebral cortex. J Neuropathol Exp Neurol 71:1047–63 16. Volpe JJ (2001) Neurology of the Newborn, 4th ed., Philadelphia, PA, WB Saunders, pp. 66–7 17. Beltran-Valero de Bernabe D, Currier S, Steinbrecher A et al. (2002) Mutations in the O-mannosyltransferase gene POMT1 give rise to the severe neuronal migration disorder Walker–Warburg syndrome. Am J Hum Genet 71:1033–43 18. Grewal PK, Hewitt JE (2003) Glycosylation defects: a new mechanism for muscular dystrophy? Hum Mol Genet 12:R259–64 19. Hong SE, Shugart YY, Huang DT et al. (2000) Autosomal recessive lissencephaly with cerebellar hypoplasia is associated with human RELN mutations. Nat Genet 26:93–6 20. Ross ME, Swanson K, Dobyns WB (2001) Lissencephaly with cerebellar hypoplasia (LCH): a heterogeneous group of cortical malformations. Neuropediatrics 32:256–63 21. Rice DS, Curran T (2001) Role of the reelin signaling pathway in central nervous system development. Annu Rev Neurosci 24:1005– 39 22. Salonen R, Paavola P (1998) Meckel syndrome. J Med Genet 35:497– 501 23. Guyot LL, Kazmierczak CD, Michael DB (2000) Adult rhombencephalosynapsis. Case report. J Neurosurg 93:323–5 24. Baumgartner MR, Saudubray JM (2002) Peroxisomal disorders. Semin Neonatol 7: 85–94 25. Faust PL (2003) Abnormal cerebellar histogenesis in PEX2 Zellweger mice reflects multiple neuronal defects induced by peroxisome deficiency. J Comp Neurol 461:394–413 26. Michaud J Mizrahi EM, Urich H (1982) Agenesis of the vermis with fusion of the cerebellar hemispheres, septo-optic dysplasia, and associated abnormalities. Acta Neuropathol 56:161–6

27. Rhombencephalosynapsis. Orphanet. Available at http://www .orpha.net//consor/cgi-bin/OC_Exp.php?Lng=GB&Expert=59315 (accessed October 22, 2017) 28. Yachnis AT (2002) Rhombencephalosynapsis with massive hydrocephalus: case report and pathogenetic considerations. Acta Neuropathol 103:301–4 29. Ishak GE, Dempsey JC, Shaw DWW et al. (2012) Rhombencephalosynapsis: a hindbrain malformation with associated incomplete separation of midbrain and forebrain, hydrocephalus and a broad spectrum of severity. Brain 135:1370–86 30. Sarnat HB (2000) Molecular genetic classification of central nervous system malformations. J Child Neurol 15:675–87 31. Butts T, Greene MJ, Wingate RJT (2014) Development of the cerebellum: simple steps to make a ‘little brain.’ Development 141:4031– 41 32. Hatten ME, Heintz N (1995) Mechanisms of neural patterning and specification in developing cerebellum. Annu Rev Neurosci 18:385– 408 33. Adams KA, Maida JM, Golden JA, Riddle RD (2000) The transcription factor Lmx1b maintains Wnt1 expression within the isthmic organizer. Development 127:1857–67 34. Lee KJ, Jessell TM (1999) The specification of dorsal cell fates in the vertebrate central nervous system. Annu Rev Neurosci 22:261–94 35. Jellinger KA (2002) Rhombencephalosynapsis. Acta Neuropathol 103:305–6 36. Montull C, Mercader JM, Peri J et al. (2000) Neuroradiological and clinical findings in rhombencephalosynapsis. Neuroradiology 42:272–4 37. Pasquier L, Marcorelles P, Loget P et al. (2009) Rhombencephalosynapsis and related anomalies: a neuropathological study of 40 fetal cases. Acta Neuropathol 117:185–200 38. Takanashi J, Katsuo S, Barkovitch AJ et al. (1999) Partial midline fusion of the cerebellar hemispheres with vertical folia: a new cerebellar malformation? Am J Neuroradiol 20:1151–3 39. Utsunomya H, Takano K, Ogasawara T et al. (1998) Rhombencephalosynapsis: Cerebellar Embryogenesis. Am J Neuroradiol 19:547–9 40. Kepes JJ, Clough C, Villanueva A (1969) Congenital fusion of the thalami (atresia of the third ventricle) and associated anomalies in a six month old infant. Acta Neuropathol 13:97–104 41. Schachenmayr W, Friede RL (1982) Rhombencencephalosynapsis: a Viennese malformation? Dev Med Child Neurol 24: 178–82 42. Sukhudyan B, Jaladyan V, Melikyan G et al. (2010) Gomez–Lopez– Hernandez syndrome: reappraisal of the diagnostic criteria. Eur J Pediatr 169:1523–8 43. Demurger F, Pasquier L, Dubourg C et al. (2013) Array-CGH analysis suggests genetic heterogeneity in rhombencephalosynapsis. Mol Syndromol 4:267–72 44. Millonig JH, Millen KJ, Hatten ME (2000) The mouse Dreher gene Lmx1a controls formation of the roof plate in the vertebrate CNS. Nature 403:764–9 45. Issac M, Best P (1987) Two cases of agenesis of the vermis of cerebellum with fusion of the dentate gyri and cerebellar hemispheres. Acta Neuropathol 74:278–80

165

16

Brainstem Malformations Brian N. Harding Department of Pathology and Laboratory Medicine, Children’s Hospital of Philadelphia, Philadelphia, PA, USA

Introduction Although individually rare, the disorders described in this chapter are becoming increasingly recognized by advanced neuroimaging and consequent genetic analysis, while morphologic studies remain infrequent and incomplete. As the embryologic origins for many brainstem and cerebellar neuronal populations are shared, this chapter encompasses both brainstem and some related cerebellar malformations.

Olivary heterotopia Definition A minor or major part of the inferior olivary nucleus can be displaced dorsolaterally from its normal position in the ventral medulla as a consequence of faulty migration. Epidemiology, clinical features and genetics Olivary heterotopias are rare microscopic abnormalities with no known distinctive semiology, and usually occur in combination with other malformations and their particular clinical features and genetic background. Thus, olivary heterotopia is associated with pachygyria, in particular Miller–Dieker syndrome [1–6] resulting from deletions of the LIS1 gene on chromosome 17p13.3 [7], X- linked neonatal pyruvate dehydrogenase deficiency and cerebral lactic acidosis [8], Dandy–Walker syndrome [9], megalencephaly [10] and trisomy 13 [4]. Imaging and laboratory findings At present there are no reported abnormalities.

Macroscopy Olivary heterotopias are usually too small to be detected by naked eye examination, but dysplastic changes in inferior olives may be visible. Histopathology Olivary heterotopias are mostly situated laterally in the rostral medulla, near the inferior cerebellar peduncle, but occasionally are more medial, close to the outflow tracts of the hypoglossal nucleus (Figure 16.1). Mature olivary neurons and surrounding neuropil within the ectopias form nodules or serpiginous structures with a myelin fiber capsule, somewhat reminiscent of the normal olivary ribbon. They may be single or form multiple islands of gray matter scattered in an archipelago dorsoventrally across the medulla. What remains of completely migrated cells within the normally positioned inferior olive is variably dysplastic, poorly folded, thickened, or fragmented and smaller than normal. Embryology and pathogenesis Neural precursors destined for the inferior olive originate in the rhombic lip before the third gestational month [11]. Most precursors arise laterally in the alar plate, although some form medially in the basal plate [12]. It is postulated that arrested migration during the first trimester allows neuroblasts stranded in ectopic sites to differentiate into distorted fragments of olivary ribbon. Such an early determination date accords with the association with pachygyria, rather than with polymicrogyria.

Dysplasias of the dentate and olivary nuclei Definition Dysplasias of the dentate and olivary nuclei (synonym dentate–olivary dysplasia) comprise a variety of congenital

Developmental Neuropathology, Second Edition. Edited by Homa Adle-Biassette, Brian N. Harding, and Jeffrey A. Golden. © 2018 John Wiley & Sons Ltd. Published 2018 by John Wiley & Sons Ltd.

167

Developmental Neuropathology

(a)

(b)

hindbrain during the first trimester. Early in the fourth month, the immature olive is a hook-shaped plate, while the dentate forms an unstructured ovoid cell mass, both being transformed into their mature state by seven months [16]. It is presumed that interference with either the late stages of neuroblast migration or the differentiation of these precursor cells is responsible for the failure of normal organization. For Zellweger syndrome, there is evidence for migration failure from 14 weeks of gestation onward, which is thought to result from the complex biochemical disorder; there are also several knockout mouse models. Nothing is known in relation to other types of dentate–olivary dysplasia.

Macroscopy In the very young, structural alterations in these nuclei may be hard to discern, but in older children and adults, close observation in well-fixed brains will detect changes in size, shape, folding or continuity. Histopathology Excessive folding of both dentate and olivary nuclei is observed in children with thanatophoric dysplasia [17,18]. In cerebellar aplasia or hypoplasia, the olive can be poorly convoluted, even a simple C-shape in Joubert syndrome [19], while the dentate is simplified or fragmented. In relation to trisomy 13, the olive forms a dorsally thickened C-shape [20], as it does in Zellweger syndrome in addition to loss of folding, peripheral margination of neurons and discontinuities in the ribbon [14,15].

Dentato–olivary dysplasia with intractable seizures in infancy Figure 16.1 Olivary heterotopia. (a) Small islands of olivary tissue stranded along their line of migration in the medullary tegmentum (arrows) (luxol fast blue–cresyl violet). (b) In the medial tegmentum on the left side, there is a small S-shaped ectopic island of olivary tissue (arrow) (luxol fast blue–cresyl violet).

morphological irregularities affect these ontogenetically closely related nuclei, usually though not inevitably involving both olive and dentate in similar fashion.

Epidemiology, clinical features and genetics Dysplasias of the dentate and olivary nuclei are rarely recorded and may be overlooked by pathologists. They are too rare for epidemiologic study. Generally, they are a minor part of more extensive developmental anomalies, displaying the clinical consequences of the disorder as a whole; for example, as part of callosal agenesis [13], Zellweger syndrome [14,15], or in relation to cerebellar hypoplasias [4]. Embryology and pathogenesis The dentate nucleus and inferior olive have a common ancestry in the rhombic lip, their precursor cells migrating across the fetal

168

Definition Dentato–olivary dysplasia with intractable seizures in infancy (synonyms, early epileptic encephalopathy with suppression bursts and olivary–dentate dysplasia) is a distinctive form of combined dentate and inferior olivary malformation that is associated with intractable neonatal seizures. Electroencephalograms (EEGs) show burst suppression, and poor prognosis. Historical annotation After two brief communications [21,22], the syndrome was formally proposed by Harding and Boyd in 1991 [23] supported by clinicopathologic evidence from five unrelated patients. The malformation itself had been briefly described in two earlier German publications [16,24], and more recent case reports confirm these findings [25–27]. Epidemiology There is no information regarding incidence for this very rare disorder. Both sexes can be affected, and all patients present in the neonatal period. No extrinsic risk factors have been identified, but siblings have been affected in one instance [25].

Brainstem Malformations Chapter 16

Genetics All published cases have been sporadic, with the exception of one report of brother and sister with the identical clinical syndrome and pathologic confirmation in one of them [25], suggesting autosomal recessive inheritance. Clinical features Presentation includes hypotonia, poor feeding, and the onset of frequent seizures in the first few days of life. Clinical examination is essentially normal at this stage. From their commencement, the seizures are intractable to medical treatment, and of various types, with tonic seizures predominant. There is gross developmental delay, but not clinical microcephaly. Some degree of visual responsiveness is attained, although the children are profoundly handicapped, and do not usually survive beyond the third year. The terminal illness is usually pneumonia.

(a)

(b)

Imaging Computed tomography and magnetic resonance imaging (MRI) studies are normal, except where there is additional ischemic damage [23]. With prolonged acquisition times examining very thin cuts, it is possible to recognize the dentate aberration with MRI in the postmortem brain (Chong and Harding personal observation). Laboratory findings Initial interictal EEGs show a burst suppression pattern, while the EEG during a typical tonic seizure shows indefinite onset, runs of initially lateralized rhythmic activity lasting up to 50 seconds at a time, which are associated with stiffening of the body, facial flushing, and fluttering of the eyelids. In more prolonged attacks, the EEG could remain isoelectric for up to 80 seconds, to be followed in some patients by bursts of sharp waves, mixed with slower components associated with brief generalized flexor and extensor movements resembling infantile spasms. Electroretinography, visual evoked potentials and brainstem auditory evoked potentials are usually normal. Extensive biochemical and metabolic investigations have been essentially unremarkable. In one patient, electromyography was myopathic, and a muscle biopsy showed lipid storage. Macroscopy The brain is small in half the patients. Careful examination under optimal conditions of fixation may reveal loss of the normal crenated outline of both inferior olive and dentate nucleus, and, for the latter, a solid globose appearance. Histopathology A minority show cerebral cortical atrophy or more extensive ischemic lesions. All cases demonstrate a solid ovoid, teardrop or club-shaped dentate nucleus with indistinct hilum, the dentate neurons arranged as interconnected small gray islands separated by thin strands of myelin fibers (Figure 16.2 ). Gliosis is minimal,

(c)

Figure 16.2 Dentato-olivary dysplasia with intractable seizure in infancy. (a) Characteristic globose shaped dentate (arrow). (b) Same at higher magnification is composed of meandering islands of grey matter. (c) Coarse C-shaped inferior olive (arrows) (luxol fast blue–cresyl violet).

while the cerebellar cortex, superior cerebellar peduncles, midbrain and pons are unremarkable. In the medulla only the principal inferior olive is pathological: a coarse hook-shaped structure virtually without undulation, neurons haphazardly mixed with radially arranged bundles of myelin fibers. The hilum and outflow tract are normal. In a minority of cases the inferior olive is only partially affected or fragmented [23].

169

Developmental Neuropathology

Differential diagnosis With present inability to confirm the diagnosis by neuroimaging, reliance is placed on the electroclinical features, which have to be distinguished from early myoclonic epilepsy [28] by a lack of myoclonus, and from Ohtahara syndrome [29], by the pattern and duration of the tonic seizures, which in addition are present from the outset; that is, much earlier in the disease process than occurs when such seizures evolve from infantile spasms in late infancy. Premature infants with gross hemispheric damage may also present with tonic seizures in early infancy, but they do not show the peculiar conformational changes of dentate and olivary nuclei only recognizable by histologic examination, which appear confined to this electroclinical syndrome, and do not occur in relation to known metabolic disorders or the well-recognized cerebellar hypoplasias (such as pontocerebellar hypoplasia, Joubert syndrome, Dandy–Walker syndrome, and Granule cell aplasia). Pathogenesis, animal models, therapy and future directions The rarity of this disorder, and present lack of any animal model, renders pathogenetic argument speculative. Therapy is purely supportive, while we await further advances in the resolutional power of MRI, which may allow some further genetic analysis when diagnosis is more readily obtainable in vivo.

¨ Mobius syndrome

restriction only in a minority. Occasionally, the syndrome is unilateral. Some or all of the IXth, Xth, XIth and XIIth cranial nerves may be involved, and ophthalmoplegia can be partial or complete. Skeletal abnormality is present in one-third of patients, including talipes, syndactyly, arthrogryposis, small limbs, and Poland anomaly (absent pectoralis major and symbrachydactyly). Feeding problems may be severe in infancy, intensified by micrognathia in some cases. If severe, respiratory and bulbar problems may be fatal in the neonatal period.

Laboratory findings Laboratory findings, including imaging studies, are noncontributory. Macroscopy Brainstem tegmental damage may be too subtle to discern. Histopathology There are scattered case reports. The largest pathological review by Tow [32] concluded that the morphological changes were heterogeneous, with four types of pathology: myopathy, primary peripheral nerve involvement, aplasia/hypoplasia of cranial nerve nuclei associated with other brainstem anomalies such as olivary dysplasia, and lastly focal necrosis and calcification in brainstem nuclei (Figure 16.3). Sudarshan and Goldie (a)

Definition M¨obius described the syndrome in 1888 [30]. It is characterized by facial diplegia usually associated with bilateral abducens palsy, occasionally with more widespread involvement of cranial nerves, especially the lower cranial nerves with notable involvement of the tongue. There is some clinical heterogeneity among published cases, and this is to some extent reflected in the various pathologic substrates that have been demonstrated. Epidemiology There are insufficient data for comment. One possible risk factor is in utero exposure to cocaine [31].

(b)

Genetics While most cases are sporadic, familial occurrence with autosomal dominant, recessive and X-linked modes of inheritance have been reported. Linkage studies have delineated three loci, mapping to chromosomes 13q12.2-12 (MBS1; Mendelian Inheritance in Man number, MIM, 157900), 3q21 (MBS2; MIM 601471) and 10q21 (MBS3; MIM 604185). Clinical features The clinical picture is of a mask-like expressionless face, with bilateral internal strabismus due to involvement of both abducens nerves. There is also drooling and speech difficulty, but nonetheless, intelligence is usually normal, with mental

170

¨ Figure 16.3 Mobius syndrome. Old scars in the (a) medullary and (b) pontine tegmentum (luxol fast blue–cresyl violet).

Brainstem Malformations Chapter 16

report two further autopsies, one showing brainstem necrosis, the other brainstem hypoplasia [33]; while, in four personally examined examples two show aplasia of the VIIth and IXth cranial nerve nuclei, and two show destruction of the brainstem tegmentum (Figure 16.3).

Differential diagnosis Clinically, M¨obius syndrome must be distinguished from congenital myopathies, particularly dystrophia mytonica, fascioscapulohumeral dystrophy and myasthenia, from acquired facial palsy traumatic due to birth trauma, and from congenital suprabulbar paresis. Experimental models Targeted disruptions of various mouse Hox genes responsible for ordered brainstem differentiation have induced brainstem dysgenesis [34,35]. Whether this is related to M¨obius syndrome remains to be determined. Pathogenesis It is thought that the midline and paramedian zones of the developing brainstem tegmentum are poorly vascularized relative to more lateral areas, and thus selectively vulnerable to vascular injury such as thrombosis ischemia or hemorrhage occurring in early fetal life [36]. The occurrence of M¨obius syndrome in an infant exposed to cocaine and alcohol in utero might also be explained by an ischemic injury. Different clinical syndromes could arise, depending on which vascular territory is involved. Thus, it has been proposed that M¨obius syndrome, and regional malformations such as Poland syndrome, are the result of disrupted blood supply within the territory of the subclavian artery [37], while vascular disruption involving the basilar artery was suggested in a patient with M¨obius syndrome combined with cerebellar hypoplasia [38]. Brainstem malformations resembling M¨obius syndrome have also been reported in association with defects of homeobox genes occurring both in man and in experimental animals [34,35,39]. Treatment Problems of maternal bonding and family care may need to be addressed, although facial mobility may improve with age.

Pontine tegmental cap dysplasia Definition Pontine tegmental cap dysplasia was defined by Barth et al. [40] using MRI with diffusion tensor imaging and tractography. It is characterized by hypoplasia of the ventral pons with a dorsal cap overlying the pontine tegmentum, protruding into the fourth ventricle. Brain imaging, and a 2015 anatomical study [41] have shown the cap to be an aberrant transverse fiber bundle.

Embryology The neuronal progenitors of the basis pontis arise in the rhombic lip, and migrate toward the ventral midline as the “anterior extramural migratory stream” of Altman and Bayer [42]. While most neuronal cell bodies remain uncrossed, their axonal growth cones cross the ventral midline [43]. The immature pontine neurons migrate toward the ventricle along radial glia [44], and their axonal growth cones move into the middle cerebellar peduncle. Anatomists can discern the inferior and superior cerebellar peduncles in fetuses as early as eight weeks of gestation [45,46] but crossing pontine fibers and a distinct middle cerebellar peduncle are only detected around 11 weeks [46]. Epidemiology There are only 30 published sporadic cases with equal sex incidence (see [41]); one has consanguineous parents. Clinical features and investigation Children present with mild to severe developmental delay, feeding disorders and impaired swallowing, horizontal gaze palsy, and ataxia. There is variable involvement of cranial nerves V VI VII and IX, while the VIIIth nerve is always affected. Extracranial abnormalities include vertebral anomalies and scoliosis most frequently, and in 25% significant cardiac malformations: atrial septal defect, aortic arch hypoplasia and tetralogy of Fallot [41]. MRI demonstrates flattening of the ventral pons, a dorsal hump, or cap, encroaching on the fourth ventricle. In a few cases, the angulation is sharper with a beak-like protrusion and a sharp bend at the very narrowed caudal pons producing a Zshape in sagittal views. Diffusion tensor imaging and tractography studies indicate absence of normal transverse pontine fibers but an ectopic transverse tract in the cap [40,41,47–49]. In some cases, the shape of the superior cerebellar peduncles and vermal hypoplasia mimic the molar tooth arrangement of Joubert syndrome [40]. Pathology There is a single morphologic study [41] of a patient with tetralogy of Fallot. Macroscopic findings were microcephaly, absence of cranial nerves VII and VIII, a hypoplastic pons distorted into a Z-shape, and a band of tissue, the cap, in the floor of the fourth ventricle (Figure 16.4.). Histologically, the pons was disorganized without recognizable tegmentum or transverse fibers, irregular myelinated fiber bundles running dorsoventrally to coalesce with transversely arranged fibers in the cap. Pontine nuclei were dispersed, but more apparent caudally in the attenuated pons. Descending corticospinal tracts were displaced laterally. Facial and acoustic nuclei were undetectable, and VI was not discrete. Superior cerebellar peduncles were thin and splayed with reduced decussation. The very small middle peduncles formed narrow bands around the pons, while the inferior peduncles were hypoplastic. Dentate and inferior olives

171

Developmental Neuropathology

Figure 16.4 Pontine tegmental cap dysplasia. Horizontal section at level of fourth ventricle and pons. The cap is an aberrant white tract extending across the floor of the fourth ventricle. Note the disorganized tegmentum, through which myelinated fibers run towards the cap.

were dysplastic, and there was evidence of cerebellar cortical atrophy.

Genetics and pathogenesis Etiology at present is undetermined. In one patient, 2q13del has been implicated [50]. This includes the gene NPHP1 which is associated with Joubert syndrome type 4. There are no animal models. Extrapolating from tractography data, Barth et al. [40] argued that pontine tegmental cap dysplasia was an axonal guidance disorder, but an additional contribution from defective neuronal migration has also been suggested [47], and this is supported by the anatomic evidence [41].

Pontocerebellar hypoplasia Definition and synonyms Pontocerebellar hypoplasia (PCH; synonyms, pontoneocerebellar hypoplasia, olivopontocerebellar hypoplasia) denotes a group of rare, inherited, progressive disorders of prenatal onset, characterized by hypoplasia/atrophy of the cerebellum (predominantly the neocerebellum) and severe involvement of other cell groups originating from the rhombic lip: pontine nuclei, dentate nuclei and inferior olives. Historical perspective First mentioned by Brun [51], the first clinical description was by Krause [52]. Norman [53] and later Gouti`eres [54] and Pfeiffer [55] described a variant with additional involvement of the anterior horn cells. A proposed classification of PCH by Barth et al. into two subtypes [56] has expanded dramatically with increasing genetic information [57].

172

Embryology Two germinal matrix zones generate the cellular components of the cerebellum. In the ventricular matrix, dentate and Purkinje cell precursors arise during the second and third gestational months: first, the dentate anlage, but as they begin their tangential migration, the Purkinje cell precursors accumulate in the ventricular matrix before migrating radially into the cortex where they begin to appear at 10 weeks. At this time, migration begins from a second matrix zone in the metencephalic rhombic lip, with neuroblast migration rostrally around the outer part of the marginal zone to form the external granular layer. This layer continues to be mitotically active well into postnatal life, generating huge numbers of granule cells, perhaps 85% of all neurons in the human brain, the postmitotic cells migrating inwards passing the Purkinje cells to form the internal granular layer. This is possibly the main driver toward the enormous extraventricular expansion of the cerebellum and the complex transverse folding of lobules and folia occurring between 12 and 18 weeks of gestation, and which greatly increases its surface area. A longitudinal zonal organization adds a further level of complexity as sectors of Purkinje cells are associated with corresponding anatomically linked sectors of the deep nuclei [58–61]. Epidemiology Assessment is problematic in such a rare disorder. The most common form, type 2 with TSEN54 mutation, is a rare disease clustered in isolated communities, particularly Dutch and German, where the carrier frequency is estimated as 0.004 [62]. Clinical features Severe intellectual deficit, swallowing problems and seizures occur in all subtypes. In PCH1, severe hypotonia, motor weakness, central visual failure, dysphagia, respiratory insufficiency, and psychomotor restriction are chief features, with death in the first year. Extrapyramidal dyskinesia and dystonia are notable in PCH2, pure spasticity only in a minority, while impairment of swallowing from birth, neonatal irritability, visual failure and seizures are also significant. Survival varies, some patients attain adulthood, but most do not reach puberty. MRI in all cases demonstrates pontocerebellar hypoplasia, with atrophy of the cerebellum and ventral pons and some cerebral atrophy, but while in PCH1 the hemispheric cerebellar involvement is variable, in PCH2 there is a dragonfly appearance with small flat hemispheres and relative sparing of the vermis [63]. Differential diagnosis includes congenital disorder of glycosylation type 1, congenital mitochondrial disorders and progressive encephalopathy with edema, hypsarrhythmia and optic atrophy, and PCH associated with lissencephaly and alphadystroglycanopathies. Pathology The morphology (Figure 16.5.) has been most intensively studied in PCH2 [56]. Not all the other subtypes have detailed

Brainstem Malformations Chapter 16

(a)

(b)

(c)

(d)

Figure 16.5 Pontocerebellar hypoplasia, type 2 (PCH2). (a) In PCH2, there is a tiny cerebellum with plate-like hemispheres. (b) The folia show variable depletion of Purkinje and granule cells (hematoxylin and eosin). (c) The dentate nucleus is formed of disconnected islands (microtubule-associated protein 2). (d) In another case with severe hypoplasia of the cerebellar hemispheres, there is evidence of segmental cortical loss, fragmented dentate nucleus and dysplastic olives. (a,b,c, Courtesy of Dr. Eleonora Aronica. d, From Ellison et al. [78]).

pathology, although in many respects where descriptions are available they are quite similar. Microcephaly is evident and can be profound. The hindbrain is disproportionately small, often less than 3% of total brain weight, as the brainstem, and especially the pons, is slender, and the cerebellum extremely hypoplastic, likened to a butterfly or bat’s wing, with narrow flattened lateral lobes, coarsely convoluted or smooth, contrasting with normally developed vermis and flocculonodular lobe. Both hypoplastic and atrophic features are present histologically. The former is manifested in the hemispheres as short stubby folia with few branches, and a very shallow pons. Arguably, regressive changes are segmental loss of the hemispheric cortex, or more extensive stretches of completely denuded cortex, as well as a generalized Purkinje and granule cell depletion. The last, to a lesser degree, is also present in the vermis. Typically, the dentate nucleus is greatly reduced to small islands or nests of large vacuolated neurons surrounded by neuropil, the ventral pons lacks

nuclei pontis and transverse fibers while descending tracts are preserved, and the inferior olives are dysplastic, poorly folded and fragmented. The main distinguishing feature of PCH1 is coexistent anterior horn cell degeneration, leading to neurogenic muscle atrophy.

Genetics Mutations in the four genes of the tRNA splicing endonuclease (TSEN) complex involved in the removal of introns, are the cause of most types of PCH. Other important genes are VRK1 in PCH1 and RARS2 in PCH6 (see Table 16.1. for a fuller list of genes). These genes have essential roles in protein synthesis and transfer RNA processing. Pathogenesis Disparity in the involvement of neo- and paleocerebellum and analysis of the developmental stage of residual vermis has

173

Developmental Neuropathology

Table 16.1 Gene mutations in the different types of pontocerebellar hypoplasia. Phenotype Pontocerebellar hypoplasia: Type 9 Type 2B Type 2D Type 6 ? type 3 Type 1B Epileptic encephalopathy, early infantile, 14 Pontocerebellar hypoplasia: Type 10 Type 1A Type 8 Type 2E Type 2A Type 4 ? Type 5 Type 2C

Location ▴

MIM

Gene/Locus

1p13.3 3p25.2 4p15.2 6q15 7q21.11 9p13.2 9q34.3

615809 612389 613811 611523 608027 614678 614959

AMPD2, SPG63, PCH9 TSEN2, SEN2, PCH2B SEPSECS, SLA, LP, PCH2D RARS2, RARSL, PCH6 PCLO, PCH3 EXOSC3, RRP40, PCH1B KCNT1, KIAA1422, EIEE14, ENFL5

11q12.1 14q32.2 16q24.3 17p13.3 17q25.1 17q25.1 17q25.1 19q13.42

615803 607596 614961 615851 277470 225753 610204 612390

CLP1, HEAB, PCH10 VRK1, PCH1A CHMP1A, PCOLN3, PRSM1, PCH8 VPS53, HCCS1, PCH2E TSEN54, SEN54, PCH2A, PCH4, PCH5 TSEN54, SEN54, PCH2A, PCH4, PCH5 TSEN54, SEN54, PCH2A, PCH4, PCH5 TSEN34, PCH2C, LENG5, SEN34

MIM, Mendelian Inheritance in Man number.

suggested a component of developmental arrest in the second trimester [57], the hypoplastic folia suggesting failure of folial outgrowth, which begins around 15 weeks of gestation. But most authors also argue for an important regressive or degenerative contribution to the pathologic picture, citing diffuse loss of cerebellar cortical neurons, and nuclei pontis, and as Barth et al. [56] demonstrate in a series of PCH2 (the most common form) a superimposed segmental degeneration of the cerebellar cortex along with segmental losses in dentate nuclei and inferior olives. These neuronal groups are functionally linked by the climbing fiber pathway, topographically arranged as sagittal strips or sectors [61].

Animal models. In zebrafish, knockdown of homologous genes, tsen54 and rars2 [64], results in a neurodegenerative phenotype with abnormalities in mid-hindbrain, developmental delay and excessive cell death. Treatment Management at present is symptomatic, including nutritional and respiratory support; there is no cure. Cot death, sleep apnea and malignant hypothermia are life-threatening complications.

Granule cell aplasia Definition and synonym Granule cell aplasia (synonym, granule cell hypoplasia) is distinguished by a predominant reduction in the granule cell

174

population associated with aberrant granule cell migration and Purkinje cell deafferentation [65,66].

Embryology Post-mitotic granule cell precursors in the external granular layer become bipolar, forming two processes aligned parallel to the long axis of the folium. A central process now develops, into which the nucleus moves, the soma becoming spindle shaped and radially oriented, then migrating radially down through the marginal layer trailing the central process behind it. It will pass through the Purkinje cell layer into the internal granular layer, while the bipolar processes become the parallel fibers innervating the spines of the Purkinje cell dendritic tree within the marginal, now molecular layer. Epidemiology and clinical features This rare disorder cannot be precisely identified on imaging and therefore there is no epidemiologic data. Mental restriction and cerebellar ataxia are present early, and are not progressive, but survival to adulthood is known. While granule cell aplasia may be an isolated lesion, it is also reported in association with GM2 gangliosidosis [67] and Pelizaeus–Merzbacher disease [68], in Menkes disease [69] and, in 2015, in siblings with a nephrocerebellar syndrome in the Mowat–Galloway spectrum [70]. Pathology There is a variable degree of microcephaly but no overt gyral anomaly. The cerebellum is very much reduced, with thinned

Brainstem Malformations Chapter 16

(a)

(b)

Figure 16.6 Granule cell aplasia. (a) Atrophic cerebellar folia showing massive depletion of the internal granular layer. There is also some Purkinje cell fall out and ectopic granule cells are present in the molecular layer. (hematoxylin and eosin). (b) Neurofilament immunostain of cerebellar cortex. There are numerous Purkinje cell dendritic expansions or cactus bodies, and some axonal torpedoes.

Genetics Both sporadic and familial cases are known. There are no encompassing genetic studies. However, granule cell aplasia was a major part of the pathology in the nephrocerebellar disorder mentioned above, presenting in a cohort of Amish kindred and mapped to 700 kB on chromosome 15 that contained a single novel frameshift variant (WDR73 c.888delT) homozygous in all affected children [70]. Pathogenesis and animal models Granule cell aplasia occurred following maternal irradiation for cancer between five and six months of gestation [71]. Consistent with this, in various experimental animal models the external granular layer can be destroyed and granule cell ectopia produced by multiple doses of X-irradiation, or using anti-mitotic agents, or with intrauterine infection [72]. The best genetic animal model of granule cell aplasia is the weaver mouse mutant wv, in which there is an abnormality of the Bergmann glia and failure of granule cell migration [73]. Granule cell aplasia was first described in a litter of cats [74], and this has now been ascribed to parvovirus infection, specifically feline panleukopenia virus, which has been shown to be directly cytolytic to granule cells [75]. Experimental neonatal inoculation of kittens produces a phenotype remarkably similar to the human granule cell aplasia: loss of granule cells, ectopic granule cells, and Purkinje cell abnormalities, including dendritic swellings [76]. While many authors have considered Purkinje cell abnormalities to be secondary to the depletion of granule cells, allowing excessive migration (Purkinje cell dislocation) into the molecular layer, and excessive dendritic growth during the period of most rapid Purkinje cell development after 36 weeks of gestation (cactus bodies and somatic sprouts) [77], in the infected cats, at least, there is evidence of direct viral invasion of Purkinje cells [76].

References and sclerotic folia although it has a normal convolutional pattern, and the brainstem shows little change. Histological findings are largely confined to the cerebellar cortex (Figure 16.6.). The molecular layer is thin, the internal granular layer greatly or completely lacking, and the Purkinje layer often crowded. The molecular layer may contain variable numbers of ectopic granule cells, and occasional dislocated Purkinje cells. Other notable Purkinje cell abnormalities are weeping-willow arborizations and cactus-like expansions of terminal dendrites, and occasionally spiky processes sprouting directly from cell somata. In some cases where Purkinje cell degeneration is also evident, there are axonal torpedo swellings. There is a diffuse fibrillary gliosis in the folia and white matter, but minimal myelin depletion or involvement of dentate nucleus or the precerebellar nuclei, olives, and nuclei pontis.

1. Daube JR, Chou SM (1966) Lissencephaly. Two cases. Neurology 16:179–91 2. Dieker H, Edwards RH, ZuRhein G et al. (1969) The lissencephaly syndrome. Birth Defects 5:53–64 3. Dobyns WB, Stratton RF, Greenberg F (1984) Syndromes with lissencephaly: I Miller-Dieker and Norman-Roberts syndromes. Am J Med Genet 18:509–26 4. Golden JA, Harding B (2015) Malformations. In: S Love, H Budka, JW Ironside, A Perry, eds, Greenfield’s Neuropathology, 9th ed., London, Arnold, pp. 357–9 5. Norman MG, Becker LE, Sirois J, Tremblay LJM (1976) Lissencephaly. Can J Neurol Sci 3:39–46 6. Sharief N, Craze J, Summers D et al. (1991) Miller–Dieker syndrome with ring chromsome 17. Arch Dis Child 66:710–12 7. Ledbetter SA, Kuwano A, Dobyns WB, Ledbetter DH (1992) Microdeletions of chromosome 17p13 as a cause of isolated lissencephaly. Am J Hum Genet 50:182–9

175

Developmental Neuropathology 8. Chow CH, Anderson RM, Kelly GCT (1987) Neuropathology in cerebral lactic acidosis. Acta Neuropathol 74:393–6 9. Hanaway J, Netsky M (1968) Heterotopias of the inferior olive: relation to Dandy–Walker malformation and correlation with experimental data. J Neuropath Exp Neurol 30:380–9 10. Bailey A, Luthert P, Dean A et al. (1988) A clinicopathological study of autism. Brain 121:889–905 11. Essick CR (1912) The development of the nuclei pontis and nucleus arcuatus in man. Am J Anat 13:25–54 12. Ellenberger C Jr, Hanaway J, Netsky MG (1969) Embryogenesis of the inferior olivary nucleus in the rat: a radiographic study and a reevaluation of the rhombic lip. J Comp Neurol 137:71–87 13. Jellinger K, Gross H, Kaltenb¨ack E, Grisold W (1981) Holoprosencephaly and agenesis of the corpus callosum; frequency of associated malformations. Acta Neuropathol 55:1–10 14. Powers JM, Moser HW, Moser AB et al. (1985) Fetal cerebrohepatorenal (Zellweger) syndrome: dysmorphic, radiologic, biochemical, and pathologic findings in four affected fetuses. Hum Pathol 16:610–20 15. Volpe JJ, Adams RD (1972) Cerebro-hepato-renal syndrome of Zellweger: an inherited disorder of neuronal migration. Acta Neuropathol 20:175–98 16. Murofushi K (1974) Normalentwicklung und Dysgenesien von Dentatum und Oliva inferior. Acta Neuropathol 27:317–28 17. Ho KL, Chang CH, Yang SS, Chason JL (1984) Neuropathologic findings in thanatophoric dysplasia. Acta Neuropathol 63:218–28 18. Wongmongkolrit T, Bush M, Roessmann U (1983) Neuropathological findings in thanatophoric dysplasia. Arch Pathol Lab Med 107:132–5 19. Friede RL, Boltshauser E (1978) Uncommon syndromes of cerebellar vermis aplasia I. Joubert syndrome. Dev Med Child Neurol 20:758–63 20. Sumi SM (1970) Brain malformations in the trisomy 18 syndrome. Brain 93:821–7 21. Boyd SG, Harding BN (1991) Intractable seizures, intermittent EEG activity and dentato-olivary dysplasia. Electroenceph Clin Neurophys 78:85P 22. Harding BN, Boyd SG (1991) Dentato-olivary dysplasia associated with intractable seizures in infancy. Neuropathol Appl Neurobiol 17:518 23. Harding BN, Boyd S (1991) Intractable seizures from infancy can be associated with dentato-olivary dysplasia. J Neurol Sci 104:157–65 ¨ 24. Gross H, Kaltenback F, Seitelberger F (1962) Uber eine systemisierte Fehlbildungdes Rautenhirns. Wien Klin Wochenschr 41705–8 25. Martland T, Harding BN, Morton RE, Young I (1997) Dentatoolivary dysplasia in sibs: an autosomal recessive disorder? J Med Genetics 3:41021–3 26. Miller SP, Shevell M, Rosenblatt B et al. (1998) Congenital bilateral perisylvian polymicrogyria presenting as congenital hemiplegia. Neurology 50:1866–9 27. Robain O, Dulac O (1992) Early epileptic encephalopathy with suppression bursts and olivary–dentate dysplasia. Neuropaediatrics 23:162–4 28. Aicardi J (1986) Myoclonic epilepsies of infancy and childhood. Adv Neurol 43:11–31 29. Ohtahara S, Ohtsuka Y, Yamatogi Y, Oka E (1987) The earlyinfantile epileptic encephalopathy with suppression-burst: developmental aspects. Brain Dev 9:371–6

176

¨ 30. M¨obius PJ (1888) Uber angeborene doppelseitige AbducensFacialis-L¨ahmung. M¨unchner Med Wochensch 35:108–11 31. Kankirawatana P, Tennison MB, D’Cruz O, Greenwood RS (1993) Mobius syndrome in infant exposed to cocaine in utero. Pediatr Neurol 9:71–2 32. Tow J, Marks K, Palmer E, Vannucci R (1979) Moebius syndrome. Neuropathologic observations. Acta Neuropathol 48:11–17 33. Sudarshan A, Goldie WD (1985) The spectrum of congenital facial diplegia (Moebius syndrome). Pediatric Neurol 1:180–4 34. Barrow JR, Capecchi MR (1996) Targeted disruption of the Hoxb2 locus in mice interferes with expression of Hoxb-1 and Hoxb-4. Development 122:3817–28 35. Chisaka O, Musci TS, Capecchi MR (1992) Developmental defects of the ear, cranial nerves and hindbrain resulting from targeted disruption of the mouse homeobox gene Hox-1.6. Nature 355:516–20 36. Leong S, Ashwell KW (1997) Is there a zone of vascular vulnerability in the fetal brain stem? Neurotoxicol Teratol 19:265–75 37. St-Charles S, DiMario F-JJ, Grunnet ML (1993) Mobius sequence: further in vivo support for the subclavian artery supply disruption sequence. Am J Med Genet 47:289–93 38. Harbord MG, Finn JP, Hall-Craggs MA et al. (1989) Moebius syndrome with unilateral cerebellar hypoplasia. J Med Genet 26: 579–82 39. Lufkin T, Dierich A, LeMeur M et al. (1991) Disruption of the Hox1.6 homeobox gene results in defects in a region corresponding to its rostral domain of expression. Cell 66:1105–19 40. Barth PG, Majoie CB, Caan MWA et al. (2007) Pontine tegmental cap dysplasia: a novel brain malformation with a defect in axonal guidance. Brain 130: 2258–66 41. Harding B, Vossough A, Goldberg E, Santi M. (2016) Pontine tegmental cap dysplasia: neuropathological confirmation of a rare clinical/radiological syndrome. Neuropathol Appl Neurobiol 42(3):301–6 42. Altman J, Bayer SA (1997) Development of the Cerebellar System, Boca Raton, FL, CRC Press 43. Marillat V, Sabatier C, Failli V, et al. (2004) The slit receptor Rig1/Robo3 controls midline crossing by hindbrain precerebellar neurons and axons. Neuron. 43:69–79 44. Kawauchi D, Taniguchi H, Watanabe H, et al. (2006) Direct visualization of nucleo-genesis by precerebellar neurons: involvement of ventricle-directed, radial fibre-associated migration. Development 133:1113–23 45. Muller F, O’Rahily R (1990) The human brain at stages 21–23 with particular reference to the cerebral cortical plate and to the development of the cerebellum. Anat Embryol 182: 375–400 46. Bayer SA, Altman J (2005) The Human Brain During the Second Trimester, Boca Raton, FL, Taylor and Francis 47. Jissendi-Tchofo P, Doherty D, McGillivray G et al. (2009) Pontine tegmental cap dysplasia: MR imaging and diffusion tensor imaging features of impaired axonal navigation. AJNR 30:113–19 48. Briguglio M, Pinelli L, Giordano L et al. (2011) Pontine tegmental cap dysplasia: developmental and cognitive outcome in three adolescent patients. Orphanet J Rare Dis 6:36 49. Caan MW, Barth PG, Niermeijer JM et al. (2014) Ectopic peripontine arcuate fibres, a novel finding in pontine tegmental cap dysplasia. Eur J Paediatr Neurol 18:434–8 50. McFerran KM, Buchmann RF, Ramakrishnaiah R et al. (2010) Pontine tegmental cap dysplasia with a 2q13 microdeletion involving the NPHP1 gene. Insights into malformations of the mid-hindbrain. Semin Pediatr Neurol 17:69–74

Brainstem Malformations Chapter 16 51. Brun R (1917) Zur Kenntnis der Bildungsfehler des Kleinhirns. Epikritische Bemerkungen zur Entwicklungspathologie, Morphologie und Klinik der umschriebenen Entwicklungshemmungen des Neozerebellums. Schweiz Arch Neurol Psychiatr 1:48– 105 ¨ 52. Krause F (1928) Uber einen Bildungsfehler des Kleinhirns und einige faseranatomische Beziehungen des Organs. Zeitschr Gesam Neurol Psychiatr 119:788–815 53. Norman RM (1961) Cerebellar hypoplasia in Werdnig-Hoffmann disease. Arch Dis Child 36:96–101 54. Gouti`eres F, Aicardi J, Farkas E (1977) Anterior horn cell disease associated with pontocerebellar hypoplasia in infants. J Neurol Neurosurg Psychiatry 40:370–8 55. Peiffer J, Pfeiffer RA (1977) Hypoplasia ponto-neocerebellaris. J Neurol 215:241–51 56. Barth PG, Aronica E, de Vries L et al. (2007) Pontocerebellar hypoplasia type 2: a neuropathological update. Acta Neuropathol 114:373–86 57. Joseph JT, Innes AM, Smith AC et al. (2014) Neuropathologic features of pontocerebellar hypoplasia type 6. J Neuropathol Exp Neurol 73:1009–25 58. Larsell O (1947) The development of the cerebellum in man in relation to its comparative anatomy J Comp Neurol 87:85–129 59. Rakic P, Sidman RL (1970) Histogenesis of cortical layers in human cerebellum, particularly the lamina dissecans. J Comp Neurol 139:473–500 60. Rakic P (1990) Principles of neural cell migration. Experientia 46:882–91 61. Voogd J (1992) The morphology of the cerebellum in the last 25 years. Eur J Morphol 30:81–96 62. Budde BS, Namavar Y, Barth PG et al. (2008) tRNA splicing endonuclease mutations cause pontocerebellar hypoplasia Nat Genet 40:1113–18 63. Namavar Y, Barth PG, Poll-The BT, Baas F. (2011) Classification, diagnosis and potential mechanisms in pontocerebellar hypoplasia. Orphanet J Rare Dis 6:50 64. Kasher PR, Namavar Y, van Tinj P et al. (2011) Impairment of the tRNA-splicing endonuclease subunit 54 (tsen54) gene causes

65.

66. 67.

68. 69. 70.

71. 72.

73.

74. 75. 76.

77. 78.

neurological abnormalities and larval death in zebrafish models of pontocerebellar hypoplasia. Hum Mol Genet 20:1574–84 Norman RM (1940) Primary Degeneration of the granular layer of the cerebellum: an unusual form of familial cerebellar atrophy occurring in early life. Brain 63:365–79 Sarnat HB, Alcala H (1980) Human cerebellar hypoplasia: a syndrome of diverse causes. Arch Neurol 37: 300–5 Friede RL (1964) Arrested cerebellar development: a type of cerebellar degeneration in amaurotic idiocy. J Neurol Neurosurg Psychiatr 27:41–5 Scheffer IE, Baraitser M, Wilson J et al. (1991) PelizaeusMerzbacher Disease: classical or connatal? Neuropediatrics 22:71–8 Barnard RO, Best PV, Erdohazi M (1978) Neuropathology of Menkes disease. Develop Med Child Neurol 20: 586–97 Jinks RN, Puffenberger EG, Baple E et al. (2015 Recessive nephrocerebellar syndrome on the Galloway–Mowat syndrome spectrum is caused by homozygous protein-truncating mutations of WDR73 Brain 138:2173–90 Bogaert L van, Radermecker MA (1955) Une dysg´en´esie c´er´ebelleuse chez un enfant du radium. Rev Neurol 93:65–82 Margolis G, Kilham L (1968) In pursuit of an ataxic hamster, or virus induced cerebellar hypoplasia. In: OT Bailey and DE Smith, eds, The Central Nervous System, IAP Monographs of Pathology, Baltimore, MD, Williams and Wilkins, pp. 157–83 Rakic P, Sidman RL (1973) Organization of cerebellar cortex secondary to deficit of granule cells in weaver mutant mice. J Comp Neurol 152:133–62 Herringham WP, Andrewes F (1888) Two cases of cerebellar disease in cats, with staggering. St Bartholomew Hosp Rep 24:241–8 Margolis G, Kilham L (1968) Virus-induced cerebellar hypoplasia. Res Publ Assoc Res Nerv Ment Dis 44:113–46 Poncelet L, H´eraud C, Springinsfeld M et al. (2013) Identification of feline panleukopenia virus proteins expressed in Purkinje cell nuclei of cats with cerebellar hypoplasia. Vet J 196:381–7 Friede RL (1975) Developmental Neuropathology, New York, Springer Verlag, pp. 332–5 Ellison D, Love S, Chimelli L et al. (2013) Neuropathology. A Reference Text of CNS Pathology, 3rd ed. London, Mosby

177

17

Spinal Cord Lesions ` 1,2 and Florent Marguet1 Annie Laquerriere 1 Department

of Pathology, Rouen University Hospital, Rouen, France Team NeoVasc ERI28, Laboratory of Microvascular Endothelium and Neonatal Brain lesions, Institute of Research Innovation in Biomedecine, Normandy University Rouen, Rouen, France

2 Region-Inserm

Introduction Malformations of the spinal cord, previously named myelodysplasias [1] may be symptomatic and isolated, or asymptomatic, and be an incidental finding on systematic histological examination. They may be a cause of fetal akinesia sequence, or constitute a minor or a major component of multiple malformations of chromosomal or genetic etiology. They encompass a wide range of disorders, but most are associated with neural tube defects, in particular myelomeningocele and/or Chiari malformation (Chapter 12).

Definitions, synonyms and epidemiology Tethered spinal cord Tethered spinal cord is responsible for neurological dysfunction by traction; it may be congenital or acquired secondary to trauma. Congenital cases are almost always associated with spinal dysraphism. The filum terminale of the spinal cord is firmly attached to the dura mater by dense fibrous tissue or by lipomatous hamartomas of the conus terminale. This condition was initially called “filum terminale syndrome” [1]. Hydromyelia Hydromyelia consists of an overdistension of the central canal of the spinal cord. It usually occurs at the lumbosacral level. Most often, hydromyelia is associated with spinal dysraphism (Figure 17.1a,b). Myelocystocele Myelocystocele is an extreme degree of dilatation of the central canal in which the cyst cavity replaces the spinal cord, and the nerve roots originate from the outer surfaces of the cyst wall [1].

Syringomyelia Syringomyelia is similar to hydromyelia in its congenital form. Literally, syringomyelia means a cavity within the spinal cord. This chronic progressive condition is idiopathic in less than 1% of cases. The prevalence varies from 1.94 in 100 000 in Japan to 8.4 in 100 000 in Western countries. It is often associated with Chiari type I malformation, or with tethered cord, or is secondary to acquired conditions such as infections, tumors or trauma [2]. “Split notochord syndrome” “Split notochord syndrome” is rare; it encompasses three pathological conditions: dimyelia, diplomyelia, and diastematomyelia. Split notochord syndrome may present with spina bifida and is also observed with vertebral dysgenesis [3] or with neurenteric cysts. The latter are rare congenital lesions that may be intraor extraspinal in location (usually located in the thorax or abdomen), extending into the spinal canal and causing compression of the anterior spinal cord. In their spinal location, they are commonly extramedullary, intradural and ventrally located [4]. Dimyelia is a very uncommon malformation consisting of a complete duplication of spinal cord, each cord having its own set of roots (Figure 17.1c). A fibrocartilaginous or bony septum may separate both cords (Figure 17.1d). Diastematomyelia is frequently associated with other spinal cord and vertebral malformations. It consists of two hemicords, either contained in a single dural sac (Figure 17.2a) or separated by vascular and connective tissue, or contained in two distinct dural sacs between which a bony septum may exist. The most frequent location is lumbar. Diplomyelia, which also belongs to the split cord syndrome, is a side by side or anteroposterior duplication of splitting of the spinal cord (Figures 17.3a,b). The spinal cord is duplicated but in contrast to dimyelia, one part is accessory and does not supply any nerve roots.

Developmental Neuropathology, Second Edition. Edited by Homa Adle-Biassette, Brian N. Harding, and Jeffrey A. Golden. © 2018 John Wiley & Sons Ltd. Published 2018 by John Wiley & Sons Ltd.

179

Developmental Neuropathology

(a)

(b)

(c)

(d)

Figure 17.1 (a) Dilatation of the central canal in hydromyelia, with no structural anomalies of the spinal cord; termination of pregnancy at 19 weeks of gestation for lumbosacral myelomeningocele (hematoxylin and eosin, H&E, ×40). (b) At higher magnification, note an intact ependymal lining (H&E, ×200). (c) Dimyelia associated with open neural tube defect, each cord having its own set of roots; termination of pregnancy at 23 weeks of gestation for myelomeningocele (H&E, ×10). (d) At higher magnification, both spinal cords are separated by an incomplete fibrocartilaginous septum (H&E, ×40).

Holomyelia Holomyelia is defined by the fusion of the anterior horns, and, although exceedingly rare, it may occur in holoprosencephaly spectrum [5] and in sacral agenesis [6]. Hemimegamyelia Hemimegamyelia is an overgrowth of a hemicord and is always associated with unilateral overgrowth of the cerebral hemispheres, brainstem and cerebellum (Figure 17.2b). Brain overgrowth phenotypes range from very localized lesions to more diffuse, often multifocal forms. The complete form of hemimegalencephaly involves all brain structures including the spinal cord [7]. Duplication of the central canal of the spinal cord Duplication of the central canal of the spinal cord is an incidental finding on systematic examination and is considered as an anatomic variant when located in the vicinity of the

180

filum terminale. It may also be observed in association with myelomeningocele.

Malformative atresia–forking of the central canal of the spinal cord Malformative atresia–forking of the central canal of the spinal cord is an extremely rare lesion in which the central canal of the spinal cord is replaced by one or more small channels lined by ependyma. It usually occurs in the upper cervical cord, at the level of the decussation of the pyramids (Figures 17.2c,d).

Risk factors The risk factors implicated in the above-defined lesions depend on their etiology. In the context of neural tube defects, antiepileptic medication, in particular antenatal use of valproic acid and carbamazepine, results in a cluster of abnormalities including neural tube defects, facial dysmorphism, mental

Spinal Cord Lesions Chapter 17

(a)

(b)

(c)

(d)

Figure 17.2 (a) Diastematomyelia with two hemicords contained in a single dural sac (hematoxylin and eosin, H&E, ×10). (b) Hemimegamyelia with overgrowth of the left hemicord, displacing the ascending tracts (arrow); termination of pregnancy at 35 weeks of gestation for total hemimegalencephaly (Luxol fast blue stain, ×40). (c) Atresia– forking of the central canal, at the level of the pyramidal decussation; termination of pregnancy at 23 weeks of gestation for Roberts syndrome (H&E, ×40). (d) These consist of three small channels lined by normal ependyma (H&E, ×100).

restriction, and congenital malformations such as congenital heart disease (fetal valproate syndrome; MIM 609442). Other risk factors include hyperthermia, periconceptional use of retinoic acid, androgenic hormones, exposure to organic solvents or anesthetic agents and pesticides [8]. Maternal diabetes mellitus and obesity, also called “the metabolic syndrome Met-Syn” are well-recognized risk factors for spinal cord lesions with neural tube defects [9]. Additionally, in over 15% of cases, maternal diabetes mellitus is responsible for caudal regression syndrome in which spinal lesions are constantly present. Its incidence has been estimated at 0.05 in 1000 births. It should be kept in mind that spinal cord lesions, in particular in neonates and infants, may be secondary to traumatic, ischemic or hemorrhagic insults, as well as to spinal tumors or pseudotumors, which must be excluded. As an example, syringomyelia can occur in association with spinal cord tumors, inflammatory arachnoiditis, or following trauma. Finally, the most important risk and prognostic factors for spinal cord malformations are the presence of associated brain and visceral malformations [10].

Genetics When spinal cord lesions are combined with neural tube defects, the responsible disease causing and predisposing genes are those for neural tube defects (Chapter 2). Spinal cord lesions with spinal dysraphism have been described in the pentalogy of Cantrell (MIM 313850; cytogenetic location on Xq25-q26.1), a rare congenital syndrome consisting of anterior diaphragm and supra- or periumbilical (omphalocele) abdominal wall deficiency and various congenital cardiac anomalies including cardiac ectopia [11]. Hydromyelia has been described in a single case of monosomy 1p36, in association with brainstem, cerebellar and cortical malformations [12]. Tethered cord with corpus callosum anomalies and periventricular cysts have been reported in Wolf Hirschorn (MIM 194190; 4p- syndrome), Cornelia de Lange syndromes and in RASopathies. It has also been reported in Currarino syndrome (MIM 176450), in which partial sacral agenesis, presacral mass, anorectal malformations, and motoneuron anomalies occur

181

Developmental Neuropathology

(a)

(b)

(c)

(d)

Figure 17.3 (a) Diplomyelia, one of the two hemicords being assessory; termination of pregnancy at 23 weeks of gestation for myelomeningocele (hematoxylin and eosin, H&E, ×10). (b) Duplication of the central canal in the same case (Luxol fast blue stain, ×100). (c) Marked depletion of the anterior horns; termination of pregnancy at 17 weeks of gestation for recurrent cerebro-oculofacioskeletal syndrome (Luxol fast blue stain, ×40). (d) Numerous pyknotic motor neurons (arrow); termination of pregnancy at 27 weeks of gestation for lethal multiple pterygium syndrome (H&E, ×200).

[13]. This autosomal dominant condition is due to mutations in the homeobox MNX1 or HLXB9 genes on chromosome 7q36.3. Caudal regression syndrome (MIM 600145) is considered as part of a spectrum including imperforate anus, sacral agenesis, and sirenomelia, in which the lower limbs are fused [14]. It is most often sporadic, it may also be due to a 7p.36 deletion [15] or caused by mutations in VANGL1 gene, on chromosome 1p13.1. Syringomyelia has been reported in association with Lhermitte–Duclos disease, one of the key features of Cowden disease 1 (MIM 158350), due to mutations in the PTEN gene, located on chromosome 10q23.31 (16) and in other RASopathies. Diastematomyelia or diplomyelia may be associated with vertebral abnormalities in Klippel–Feil syndrome (KFS), a genetic heterogeneous condition characterized by a defect

182

or segmentation of cervical vertebrae, resulting in a fused appearance. Associated lesions include low posterior hairline and short neck, ear lesions, and ocular abnormalities. Respiratory problems due to incomplete paralysis of the fifth root have also been reported [17]. Two forms are inherited as autosomal dominant traits: KFS1 (MIM 118100), related to mutations in the GDF6 gene located on chromosome 8q22.1, and KFS3 (MIM 606522), due to mutations in FDF3 gene on chromosome 12p13.1. The autosomal recessive form named KFS2 (MIM 214300) is due to mutations in MEOX1 gene, on chromosome 17q21.31. Holomyelia has been reported in Smith–Lemli–Opitz type II (MIM 270400), an autosomal recessive inherited metabolic disease with multiple malformations, including malformations of the cerebral midline, genital abnormalities, polydactyly or

Spinal Cord Lesions Chapter 17

syndactyly, together with various other visceral malformations. This syndrome is due to mutations in the DHCR7 gene, located on chromosome 11q13.4 [18]. In partial or total hemimegalencephaly, asymmetry of the lesions is the rule. The underlying molecular basis is due to mosaic postzygotic mutations in core components of key cellular pathways, such as the phosphatidylinositol 3-kinase (PI3K)akt murine thymoma viral oncogene homolog (AKT)-mTOR pathway [19].

Clinical features In tethered spinal cord syndrome, patients present with foot deformities, back pain and scoliosis, gait abnormalities, and thereafter with progressive lower limb weakness and neurogenic bladder. In split notochord syndrome (dimyelia, diplomyelia and diastematomyelia), clinical symptoms depend on the location and the extent of the malformation. Patients with diastematomyelia present with congenital scoliosis, sometimes with cutaneous lesions such as a hairy patch, dimple or subcutaneous mass, often in the lumbosacral region. They subsequently develop a progressive myelopathy with foot deformities and kyphosis. Clinical features associated with split cord anomalies are often difficult to characterize owing to the frequency of comorbid pathologies of the cord [10]. Patients with syringomyelia may be asymptomatic, and the lesion is discovered as an incidental finding on spinal cord imaging, or may exhibit mild symptoms such as cape-like loss of pain and temperature sensation. If syringomyelia is associated with neural tube defect, the clinical course is different: patients present with muscle weakness with neurological deficits or stiffness/spasticity, chronic neuropathic pain, autonomic symptoms (bladder and bowel dysfunction) and scoliosis. Split cord malformations resulting from enterogenic cysts are revealed by signs of spinal compression and vertebral anomalies are associated in a number of cases. In neonates, isolated hydromyelia is usually asymptomatic but children generally develop spastic paraparesis. The symptoms correlate with the level of the lesion and are caused by spinal cord compression. In caudal regression syndrome with spinal anomalies, the level of bone and cord involvement corresponds first with the level of weakness, followed by a slowly progressive deterioration in neurologic function [20].

Imaging Diastematomyelia may be detected prenatally by ultrasonography, but must be confirmed by magnetic resonance imaging (MRI), which is the preferred imaging tool for the detection of spinal cord lesions. Before and after birth, the combination of computed tomography and MRI is necessary to appropriately

delineate the extent of the lesions and explore the neighboring structures.

Macroscopy In the tethered spinal cord syndrome, the filum is attached to the dura mater or an extradural band may anchor the cord. The cord may also be bound down by lipomas located at the distal end of the conus and attached or intermixed with the rootlets. In hydromyelia, the central dilatation is symmetric, angulated or oval-shaped. The central cavity may extend over many segments, from the fourth ventricle to the conus medullaris, where it can be associated with lipoma of the conus [21]. Myelocystoceles form a cystic cavity replacing the spinal cord. Syringomyelic cavities are asymmetric, and of various shape. Most are located between C2 and T9, but can extend upwards into the brainstem (syringobulbia). In this location, they are associated with abnormality of the foramen magnum, Chiari malformation type I, basilar impression and Dandy–Walker malformation. In diplomyelia, two cords may be macroscopically identified. Diastematomyelia may be suspected if a cartilaginous or bony septum divides the spinal cord. Holomyelia consists of a round-shaped spinal cord devoid of anterior and posterior sulci. In hemimegamyelia, the spinal cord is asymmetric. Duplication or atresia-forking of the central canal, and anomalies of the ascending and descending tracts, are diagnosed on microscopic examination.

Histopathology In hydromyelia, the central dilatation is surrounded by an intact or flattened ependymal lining or by ruptured ependyma replaced by gliosis. Syringomyelia is a fluid-filled cavity, lined by ependyma or gliosis, and may form a diverticulum of hydromyelia or may be totally separated from the spinal canal. The cavity may be distinct from, or merged with, the central canal. It can involve gray matter, white matter or both, and result in loss of neurons and white matter [10]. Associated spinal arachnoid cysts may be observed [22]. In the split cord syndrome associated with neurenteric cysts, the cysts are lined by cuboidal or columnar epithelium, usually of gastrointestinal or respiratory type. In diplomyelia, two cords with two central canals are usually present, each being surrounded by white and gray matter arranged in a normal pattern. The cord is duplicated, but one cord is accessory and does not supply nerve roots, in contrast to dimyelia, in which each cord has its own set of roots. Both cords are often completely caudally fused but remain separated to the tip of the conus medullaris. An incomplete bony septum may be present. Diplomyelia may be observed in association with intramedullary teratomas [23]. In diastematomyelia, a midline septum divides the spinal cord into two equal or unequal portions, similarly to diplomyelia and dimyelia. The septum may span the entire width of the spinal

183

Developmental Neuropathology

(a)

(b)

(c)

(d)

(e)

(f)

Figure 17.4 (a) Symmetric hypoplasia of the ascending tracts (arrow); termination of pregnancy at 31 weeks of gestation for partial trisomy 1 (hematoxylin and eosin, H&E, ×40). (b) Age-matched control case (H&E, ×20). (c) Abnormally shaped pyramidal tracts associated with asymmetrical ascending tracts; termination of pregnancy at 25 weeks of gestation for caudal regression syndrome (H&E, ×10). (d) Fragmentation of the pyramidal tracts into several bundles in the same case (H&E, ×20). (e) Unilateral agenesis of the pyramidal

184

tracts (arrow); termination of pregnancy at 32 weeks of gestation for arthrogryposis multiplex congenita (Luxol fast blue stain, ×20). (f) Glomeruloid proliferative vasculopathy (arrow), with hemorrhage and calcifications (black triangle), hydromyelia and dilatation of the subarachnoid spaces; termination of pregnancy at 25 weeks of gestation for Fowler syndrome (hematoxylin eosin saffron stain, ×20).

Spinal Cord Lesions Chapter 17

cord and is ventrally anchored to the dura mater. It is composed of fibrous tissue, cartilage or mature bone. Both hemicords are either contained in a single dural sac, or separated by vascular and connective tissue, or contained in two dural sacs with a bony septum in between. The cleft between both hemicords is of variable length. One or both hemicords may be hypoplastic. Each faces the other, and is medially rotated with its own spinal artery, allowing distinction from diplomyelia. In caudal regression syndrome, the cord is truncated and tethered. The dural sac is stenosed, and diastematomyelia, syringomyelia, lipomas and dermoid cysts are usually observed. Other lesions of the spinal cord consist of anomalies of the anterior horns, such as paucity of motorneurons (Figures 17.3c,d) or fusion of the anterior horns (i.e. holomyelia; for iconography, see [6,18]). Asymmetry and hypoplasia of ascending sensory tracts (Figures 17.4a–c) can be isolated or associated with spinal cord dysraphism, total hemimegalencephaly, or fetal akinesia sequence. Abnormalities of the descending pyramidal tracts consist of misshapen tracts (Figure 17.4c) with anomalies of their decussation. They are often fragmented into several small or large bundles (Figure 17.4d). Unilateral asymmetry or agenesis may be observed alone or in association with other spinal cord lesions (Figure 17.4e).

Differential diagnosis All malformative lesions described above may be distinguished from acquired lesions due to trauma (Chapter 23), inflammation or infection (Chapter 44), hydrocephalus (Chapter 21), or spinal cord tumors. In cases with hemorrhagic and calcified lesions of the spinal cord with hydrocephalus, it is important to consider Fowler syndrome (MIM 225790), a proliferative vasculopathy with massive destruction of the parenchyma involving the brain and spinal cord (Figure 17.4f) caused by mutations in the FLVCR2 gene on chromosome 14 [24].

Pathogenesis Isolated hydromyelia is generally regarded as a true malformation, in particular when it involves the entire spinal cord, and due to a persistent embryonic–fetal state. Localized hydromyelia is usually associated with neural tube defects. Some authors believe congenital syringomyelia results from a defect of primary neurulation and they classify two forms, embryonic and fetal. In the embryonic form, mesenchymal tissue is almost absent between the spinal cord and the ectoderm, causing a gap in the posterior vertebral arches (i.e. an open neural tube defect). In the fetal form, the vertebral column is intact and mesenchyme is not affected. Others contend that syringomyelia is a destructive lesion related to a disturbance of cerebrospinal fluid circulation linked to other malformations, such as malformations of the craniocervical junction

or of the brainstem (atresia of the foramen of Magendie and Chiari type II malformation), or occurs in the setting of tethered cord. These malformations are responsible for severe disorders of cerebrospinal fluid flow and pressure in the spinal cord, the subarachnoid spaces and cisterna magna, which in turn lead to formation of the syrinx [10]. Diastematomyelia is considered by some authors as an extreme dilatation of the central canal of the spinal cord [1], although this lesion is mainly observed with failure of neural tube closure. Neurenteric cysts associated with split cord syndrome are considered as resulting from a disturbance in the early interaction between the notochord, the neural plate, endoderm and mesoderm during the third week of gestation. The etiology of caudal regression syndrome remains uncertain, but it is thought to result from an ischemic insult to the developing spine [9]. The current hypothesis of the pathogenesis of the pentalogy of Cantrell is that it results from a developmental failure of the lateral mesoderm and subsequent septum tranversum during the first month of embryogenesis [11]. Atresia– forking of the brain cavities has essentially been described at the level of the aqueduct of Sylvius and the third ventricle. They may be isolated or associated with other malformations, such as rhombencephalosynapsis, agenesis of the corpus callosum, holoprosencephaly, and neural tube defects [25]. Presently, neither recurrent chromosomal rearrangements nor diseasecausing specific genes have been identified in nonsyndromic forms [26].

Treatment Preventive measures to reduce lesions of the spinal cord described in this chapter in the context of neural tube defects include avoidance of teratogens, periconceptional supplementation of folic acid, and careful control of glycemia before and during pregnancy. Medical therapies for all these conditions consist in the management of the neurologic deficits. The natural history of split cord syndrome supports early surgical treatment directed toward untethering the cord from its most caudal region, and restoring spinal anatomy with reestablishment of unobstructed cerebrospinal fluid flow in the subarachnoid spaces [27]. If the cord can be completely untethered, various surgical treatments to decompress the syrinx are available [28].

References 1. Friede RL (1989) Spina bifida and related lesions. In Developmental Neuropathology, 2nd ed., Heidelberg, Springer, pp. 248–62 2. Vandertop WP (2014) Syringomyelia. Neuropediatrics 45:3–9 3. Andro C, Pecquery R, De Vries P et al. (2009) Split cervical spinal cord malformation and vertebral dysgenesis. Orthop Traumatol Surg Res 95:547–50

185

Developmental Neuropathology 4. Ebisu T, Odake G, Fujimoto M et al. (1990) Neurenteric cysts with meningomyelocele or meningocele. Split notochord syndrome. Childs Nerv Syst 8:465–7 5. Marcorelles P, Laquerriere A (2010) Neuropathology of holoprosencephaly. Am J Med Genet C Semin Med Genet 154C:109–19 6. Friede RL (1989) Dysplasias of the brainstem and spinal cord. In: Developmental Neuropathology, 2nd ed., Heidelberg, Springer, pp. 372–86 7. Flores-Sarnat L, Sarnat HB, D´avila-Guti´errez G, Alvarez A (2003) Hemimegalencephaly: part 2. Neuropathology suggests a disorder of cellular lineage. J Child Neurol 11:776–85 8. Padmanabhan R (2006) Etiology, pathogenesis and prevention of neural tube defects. Congenit Anom (Kyoto) 46:55–67 9. Ray JG, Thompson MD, Vermeulen MJ et al. (2007) Metabolic syndrome features and risk of neural tube defects. BMC Pregnancy Childbirth 7:21 10. Simonati A (2006) Spinal cord lesions. In: JA Golden, BN Harding, eds, Pathology and Genetics. Developmental Neuropathology, Basel, Neuropath Press, pp. 109–13 11. Kachare MB, Patki VK, Saboo SS et al. (2013) Pentalogy of Cantrell associated with exencephaly and spinal dysraphism: antenatal ultrasonographic diagnosis. Case report. Med Ultrason 3:237–9 12. Shiba N, Daza RA, Shaffer LG et al. (2013) Neuropathology of brain and spinal malformations in a case of monosomy 1p36. Acta Neuropathol Commun 1:45 13. Kole MJ, Fridley JS, Jea A, Bollo RJ (2014) Currarino syndrome and spinal dysraphism. J Neurosurg Pediatr 6:685–9 14. Duesterhoeft SM, Ernst LM, Siebert JR, Kapur RP (2007) Five cases of caudal regression with an aberrant abdominal umbilical artery: Further support for a caudal regression-sirenomelia spectrum. Am J Med Genet A 143A:3175–84 15. Al Kaissi A, Klaushofer K, Grill F (2008) Caudal regression syndrome and popliteal webbing in connection with maternal diabetes mellitus: a case report and literature review. Cases J 1:407 16. Nayil K, Wani M, Ramzan A et al. (2011) Lhermitte–Duclos disease with syrinx: case report and literature review. Turk Neurosurg 4:651–4

186

17. Cece J, Aboharb F, Rezzadeh KS et al. (2015) Klippel–Feil syndrome and unilateral diaphragmatic paralysis. Eplasty.15:ic10.e 18. Qu´elin C, Loget P, Verloes A et al. (2012) Phenotypic spectrum of foetal Smith–Lemli–Opitz syndrome. Eur J Med Genet 55: 81–90 19. Mirzaa GM, Poduri A (2014) Megalencephaly and hemimegalencephaly: breakthroughs in molecular etiology. Am J Med Genet C Semin Med Genet 166C:156–72 20. Ashwall S (2001) Congenital defects of the brain. In: MI Levene, FA ChervenakMJ Whittle, eds, Foetal and Neonatal Neurology and Neurosurgery, 3rd ed., London, Churchill Livingstone, pp. 199– 236 21. Faggin R, Drigo P, Denaro L et al. (2010) Hydromyelia associated with spinal lipoma of the conus: case report. Spine (Phila Pa 1976) 35:E1069–71 22. Tucer B, Yilmaz MB, Ekici MA et al. (2014) Spinal arachnoid cysts associated with syringomyelia: a review of the literature and report of a case. Turk Neurosurg 24:606–12 23. Mut M, Shaffrey ME, Bourne TD et al. (2007) Unusual presentation of an adult intramedullary spinal teratoma with diplomyelia. Surg Neurol 2:190–4 24. Thomas S, Encha-Razavi F, Devisme L et al. (2010) Highthroughput sequencing of a 4.1 Mb linkage interval reveals FLVCR2 deletions and mutations in lethal cerebral vasculopathy. Hum Mutat 10:1134–41 25. Pasquier L, Marcorelles P, Loget P et al. (2001) Rhombencephalosynapsis and related anomalies: a neuropathological study of 40 foetal cases. Acta Neuropathol 117:185–200 26. D´emurger F, Pasquier L, Dubourg C et al. (2013) Array-CGH analysis suggests genetic heterogeneity in rhombencephalosynapsis. Mol Syndromol 4:267–72 27. Tsitouras V, Sgouros S (2013) Syringomyelia and tethered cord in children. Childs Nerv Syst 9:1625–34 28. Roy AK, Slimack NP, Ganju A (2011) Idiopathic syringomyelia: retrospective case series, comprehensive review, and update on management. Neurosurg Focus 31:E15

18

Hydrocephalus Homa Adle-Biassette Department of Pathology, APHP, Lariboisi`ere Hospital, Universit´e Paris Diderot, Paris, France

Definition, major synonyms and historical perspective

Normal embryology and pathways of CSF production, circulation and resorption

Hydrocephalus derives from Greek, hydro meaning “water,” and kephalos “head,” defining a net accumulation of cerebrospinal fluid (CSF) within the central nervous system, especially within the ventricles. Ventriculomegaly is the condition of enlarged cerebral ventricles [1]. Hydrocephalus is the consequence of an imbalance between the production, circulation and resorption of CSF. The pathophysiology of hydrocephalus is not well understood, and thus its classification has long been a matter of debate. It is a complex disorder, encompassing a variety of clinicopathological conditions. By definition, obstructive or noncommunicating hydrocephalus indicates impairment of CSF flow within the ventricular system, whereas in communicating hydrocephalus, the obstruction is distal to the ventricle, mostly in the subarachnoid spaces or due to sagittal sinus thrombosis [2]. Diffuse villous hyperplasia of the choroid plexus and choroid plexus papillomas, including rare congenital examples, may give rise to hydrocephalus due to an overproduction of CSF [3]. The actual consensus classification is based on the location of obstruction to the flow of CSF (foramina of Monro, the aqueduct of Sylvius, the basal cisterns, the arachnoid granulations, and outflow of venous blood from the dural venous sinuses), the etiology of the inciting condition, the chronicity or rapidity of onset, and the age of the patients or experimental animals [4]. Other items such as intracranial pressure dynamics, underlying lesions, symptomatology, and treatment have also been considered [5]. Hydrocephalus can also result from a loss of brain parenchyma, so-called hydrocephalus ex vacuo, occurring in the setting of disruptive and degenerative changes and is not further considered in this chapter.

In 1664, Willis proposed that the choroid plexuses work as a secretory gland for the production of CSF. Luschka later supported this theory in 1855, following microscopic examination of the Plexuses [6]. Dandy defined the choroid plexuses as the source of the production of CSF in adults. Classically, CSF is generated by filtration or secretion from ventricular choroid plexuses or blood vessels, and, to a lesser extent, from ependyma and the parenchyma, and is absorbed in venous sinus via arachnoid granulations (pacchionian bodies) or blood vessels in subarachnoid spaces [7,8]. However, the sources of CSF remain partly unknown [6,9,10]. The role of various molecules involved in the transport of water and ions is being extensively studied [11]. CSF flows out of the ventricular system via the foramina of Luschka and Magendie into the subarachnoid spaces, then flows around the spinal cord and the cerebral convexity. It is primarily absorbed by the arachnoid granulations and arachnoid villi of the superior sagittal sinus [12]. A microscopic flow of CSF has been described through the brain interstitial spaces [13]. The developmental pattern of CSF circulation is thought to be different from the generally believed pattern during adulthood. In higher vertebrates, CSF is formed within the neural tube before the choroid plexus appears. The choroid plexuses start to develop around the seventh week of gestation, the epithelium derives from the neural tube epithelium, the endothelial cells from the mesoderm, and mesenchymal cells from the cephalic mesoderm and mesencephalic neural crest, which gives rise to mesectodermal cells [14]. The arachnoid villi are apparent by the 35th week of gestation and arachnoid granulations are observed after the 39th week [15]. The subcommissural organ belongs to the circumventricular organs and is likely involved in the secretion of CSF-soluble

Developmental Neuropathology, Second Edition. Edited by Homa Adle-Biassette, Brian N. Harding, and Jeffrey A. Golden. © 2018 John Wiley & Sons Ltd. Published 2018 by John Wiley & Sons Ltd.

187

Developmental Neuropathology proteins or circulation of CSF, but its precise functions are unknown [16,17]. The human subcommissural organ is a small gland characterized by a highly differentiated ependyma located in the epithalamus, caudal to the pineal gland at the entrance to the aqueduct of Sylvius [16]. It is the first secretory structure of the brain that differentiates during the second month in embryos, becoming an active secretory structure of the brain at midgestation and starting to regress during the first year of postnatal life [18]. Changes in the subcommissural organ have been described in all species developing congenital hydrocephalus and in human fetuses [19]. Several animal models have been used to explore the function the subcommissural organ in the pathophysiology of hydrocephalus [20] (and see below). Animal studies suggest the subcommissural organ is essential for CSF circulation through the narrow aqueduct of Sylvius. Dysfunction of the subcommissural organ could influence the patency of the cerebral aqueduct [7,21]. In mammalians, the subcommissural organ secretes glycoproteins into the CSF that assemble to form the Reissner’s fiber, which extends along the length of the CSF tract. In mice, the lack of Reissner’s fiber is followed by aqueductal stenosis [22], then provoking chronic hydrocephalus, which, in turn, induces additional alterations of the subcommissural organ [7,22]. The subcommissural organ could also be involved in osmoregulation, detoxification of the CSF or mechanoreception. However, none of these hypotheses has been convincingly substantiated. The function of the human subcommissural organ is not well known; it secretes soluble glycoproteins that do not aggregate [16,18] so that Reissner’s fiber has not been detected in humans [18].

Epidemiology Hydrocephalus is among the most common central nervous system (CNS) abnormalities. It can be seen in isolation or as one feature of a malformation complex [23] either restricted to the brain or involving several organs and/or the skeleton (see below). The prevalence and demographics of congenital hydrocephalus remain poorly defined, in part because the definition of “congenital hydrocephalus” varies between studies and the type of population studied. A large population-based investigation of idiopathic infantile hydrocephalus in Denmark over a 30-year period included liveborn children diagnosed before one year of age with no known extrinsic cause or spina bifida. The prevalence of isolated congenital hydrocephalus was 0.062 per 1000 infants [24]. Data from four European registries of congenital malformations (EUROCAT) included information about live births, fetal deaths at gestational age over 20 weeks and termination of pregnancies for fetal anomaly, and excluded cases with associated neural tube defects. The overall prevalence was 4.65 per 10,000 births. There were 47% live births, 5% fetal deaths and 48% terminations of pregnancy for fetal anomaly [25]. In the study from the California Office of Statewide Health Planning

188

and Development, the overall prevalence of congenital hydrocephalus during hospitalization for delivery was 5.9 per 10,000 live births [26]. The prevalence is higher in developing countries, such as Brazil, reaching rates of 3.16 per 1000 newborns [27]. The main associated conditions were malformations including spina bifida, chromosomal anomalies, facial clefts, congenital heart disease, polycystic kidneys, and acquired conditions such as intracranial hemorrhage or infections (meningitis, cytomegalovirus, toxoplasmosis and rubella) [25,26,28]. Risk factors included lack of prenatal care and follow-up, multiparity, maternal diabetes, maternal chronic hypertension or gestational hypertension, intake of antidepressants, and alcohol during pregnancy [24,28,29]. Of patients with congenital hydrocephalus, 12.1% had an additional family member also diagnosed with hydrocephalus [29]. Recurrence risk is dependent upon the etiology of hydrocephalus [30].

Clinical features Clinical evaluation of hydrocephalus requires a multidisciplinary approach that should take into account the detailed medical history, including family history, physical examination, and imaging data [4,30]. Further investigations for TORCH (toxoplasmosis, rubella, cytomegalovirus, herpes simplex, and HIV) or genetic diagnoses may be required. Clinical presentation varies with age [11]. In newborns and children up to two years, the most important and consistent clinical finding is macrocephaly. In children, before cranial suture fusion, hydrocephalus causes skull enlargement and widened fontanelles, ocular disturbances including paralysis of upward gaze, and sluggish pupillary reaction, together with progressive weakness and spasticity of the lower limbs. Children older than two years tend to present with neurological symptoms resulting from increased intracranial pressure, or with focal deficits referable to the primary lesion [31]. Motor, sensory, visual, and memory systems may be disturbed through involvement of the long projection axons [32] and hypothalamic function may be impaired. Prenatal diagnosis of ventriculomegaly is based on ultrasound measurement of the atrial width of the lateral ventricles [33,34] and is classified as mild (10–12 mm), moderate (12–15 mm) or severe (>15 mm). However, false positive and negative diagnoses are possible [33]. The combination of computed tomography and magnetic resonance imaging (MRI) are the most sensitive tools for the detection of hydrocephalus (Figure 18.1) [35].

Pathology and physiopathology Pathophysiology and morphological consequences of hydrocephalus and ventricomegaly The pathophysiology of congenital hydrocephalus has been reviewed [7,36]. It includes primary abnormalities that may

Hydrocephalus Chapter 18

(a)

(b)

(c)

Figure 18.1 Brain computed tomography (a) and T1 weighted MRI (b, c) of three patients with X-linked hydrocephalus. (The cases from which images b and c were taken have been confirmed genetically to have mutations in the L1CAM gene. Courtesy of Dr. M Yamasaki, Department of Neurosurgery, Osaka National Hospital, Japan.) (a) An axial plane image of a boy taken on the day of birth, demonstrating extensive hydrocephalus with marked thinning of the cerebral mantle. (b) An axial image of another boy taken at 18 months of life, following a shunt operation, showing marked decompression of the ventricular system and thin white matter. (c) A sagittal image of a four-year-old boy discloses hypoplasia of the corpus callosum, hypoplasia of the anterior portion of the cerebellar vermis, and an abnormally large interthalamic structure.

affect short or long-term outcome, and secondary injury mechanisms that occur mainly as a result of expanding ventricles and/or altered CSF physiology. The morphological consequences of hydrocephalus vary considerably and are influenced by age, causes, duration, severity, and onset. Cerebral mantle disruption, such as absence of the septum pellucidum, and severe thinning or absence of the posteromedial cerebral mantle [37], various degrees of “hypoplasia” or atrophy of corpus callosum, fimbria/fornix, basal ganglia, and disappearance of the sulci or polymicrogyria may be observed. Obliteration of the third ventricle with fusion of the thalami is seen under various circumstances [38–40]. Figure 18.2 depicts an example of congenital hydrocephalus in an eight-month-old child. It is clear that several pathological processes ensue when the brain is compressed and distorted by ventriculomegaly. In the developing brain, ventricle expansion and increased CSF pressure enlarge the skull. The most acute mechanisms include compression and stretch of periventricular tissue that cause “ependymal” tearing and denudation with protrusion of the underlying periventricular/subependymal neuropil, formation of ependymal pseudorosettes in the subventricular zone and gliosis [41,42]. Ependymal cells do not appear to regenerate [42]. Whether the ependymal loss is the first pathologic event prior to the onset of hydrocephalus needs to be clarified [43]. However, these lesions may be observed in fetuses at all ages, even in case of transient ventriculomegaly, which can be masked by cerebral edema during perinatal death. Progressively, gliosis begins to develop in the subventricular zone, indicating chronicity, and glial nodules can bulge forming the so-called granular ependymitis. Glial scar formation is a permanent feature in hydrocephalic brains, even in those that have been successfully shunted [7]. According to the age of gestation, loss of germinal cells in ventricular/subventricular zone will impact neurogenesis, gliogenesis and myelination. Hydrocephalus is also responsible for ischemia–hypoxia and impairment of the blood–brain barrier, with neuroinflammation, periventricular edema, demyelination, axonal degeneration, and slowed axoplasmic transport, with secondary metabolic anomalies, dendritic, and synaptic deterioration resulting in altered connectivity and cell death [7,36].

Etiology There is a constellation of causes of hydrocephalus [23,44], involving a large variety of gross brain malformations, such as neural tube defects [45] with Chiari type II malformation, Dandy–Walker syndrome, brain patterning defects such as holoprosencephaly, and rhombencephalosynapsis, neuronal migration disorders, and tectal plate lesions [46], including aqueductal glioneuronal hamartoma [47] and cystic lesions. Agenesis of, or deficient, arachnoidal granulations have also been implicated [23,44]. Hydrocephalus may occur in some individuals with primary ciliary dyskinesia, sometimes associated with situs inversus [43,48–51].

189

Developmental Neuropathology (a)

(b)

Figure 18.2 The brain of an eight-week-old boy with marked hydrocephalus. (a) A coronal section of the cerebral hemispheres, demonstrating enlargement of the lateral and third ventricles. (b) Section through the thalamus and hippocampus showing marked volume loss of the white matter (arrows) around the inferior horn of the lateral ventricle. Note that the cortical ribbon of the inferior gyri of the temporal lobe is thin (Luxol fast blue stain).

Secondary causes are more frequent. Infection including meningitis or TORCH, aqueduct stenosis related to toxoplasmosis [52], or mumps have been reported [53]; the latter and subacute sclerosing panencephalitis have also been implicated in late-onset hydrocephalus [53,54] (Chapter 42). Intraventricular hemorrhage (Chapter 20), vascular malformations (Chapter 24), trauma (Chapter 23), tumors, neonatal lupus erythematosus [55], teratogens, and nutritional disorders including vitamin A [56] have also been reported. Membranous obstruction of the foramen of Monro may have a developmental origin or may be secondary [23,57].

Pathology Morphological changes of the aqueduct The most common cause of congenital hydrocephalus is aqueduct stenosis, although, clinically, the precise mechanism responsible for the stenosis remains unclear. The aqueduct of Sylvius is the narrowest part of the ventricular system. Narrowing generally occurs at the level of either the middle of the superior colliculus or the intercollicular sulcus (Figure 18.3a) [58]. The CSF flows through the aqueduct in a pulsatile fashion. The shape and caliber of the aqueduct of Sylvius changes along the anatomic levels and at various developmental stages. During normal development, the lumen of the aqueduct decreases in size from the second month of fetal life until birth. This narrowing appears to be due to growth pressure on the

190

aqueduct by neighboring mesencephalic structures. Neuropathological examination allows classification into three main categories [23,44,59]: r gliosis r aqueductal stenosis r atresia and/or forking. Gliosis The term “aqueduct gliosis” is used when a contour of the aqueduct lumen remains recognizable as an interrupted ring of ependymal cells, rosettes, and tubules, associated with marked surrounding gliosis [23]. Most often, the aqueduct lumen is filled with iron-laden macrophages, and sometimes by hematoma, since aqueduct gliosis is often related to hemorrhage. Inflammation or infectious diseases may also be apparent on microscopic examination [44]. However, absence of gliosis does not exclude infection, as shown in animal models [60]. Aqueductal stenosis Pathologically, the aqueduct shows focal reduction in size, lined by a normal ependyma, without histological abnormality in the adjacent neuropil (Figure 18.3b). Emery and Staschack [61] consider that an abnormal stenosed aqueduct should have a diameter less than 0.5 mm2 in a child of any age. Thus, a definitive diagnosis may require serial sections through the midbrain. Aqueduct stenosis may precede hydrocephalus [62–64], it may

Hydrocephalus Chapter 18

(a)

(b)

(d)

(c)

(e)

Figure 18.3 Histopathological features of aqueductal stenosis. (a) Section through the inferior colliculus of the midbrain. The aqueduct is not visible. (b) Light micrograph of the mid-sagittal portion of the section depicted in (a), showing marked narrowing of the qaqueductal lumen. Note that there is no gliosis or inflammatory cell infiltration in the surrounding tissue (Luxol fast blue stain). (c)

Light micrograph of another case with aqueductal atresia, showing a small tubule and nearby scattered small ependymal canals (forking) in the midline of the midbrain tegmentum (hematoxylin and eosin). (d,e) Atresia and forking at the level of the medulla in a seven-month-old child presenting with arthrogryposis, who died suddenly.

also be a result of the lateral compression exerted by the expanding hydrocephalic ventricles [65–67]. There are many underlying causes, but several are linked to autosomal and X-linked genetic disorders [30,68]. It is a cardinal feature of L1 and L1like syndromes, but is also observed in various isolated CNS malformations or polymalformation syndromes involving the CNS [38].

a channel ventral to the aqueduct), other normal anatomical variations or secondary lesions with gliosis [23,44]. The aqueduct of Sylvius may show complex forking: two or more distinct permeable channels, surrounded by several ectopic ependymal tubes, separated from each other by normal brain parenchyma. Atresia consists of a completely impermeable channel replaced by several small tubules lined by ependymal cells [23,44]. The aqueduct of Sylvius can be atretic for its entire length, or for a short segment, or can be present in other locations, such as the fourth ventricle foramina [44] or the central canal of the medulla (Figure 18.3d,e). The most frequent cause of aqueduct atresia and/or forking is its association with Chiari type II malformation (Chapter 12) [44]. These lesions also characterize the rhombencephalosynapsis spectrum, or mutation in MPDZ gene [40].

Atresia and/or forking Both “atresia” and “forking” have been used to describe the lesion in which groups of ependymal canals are irregularly arranged in the expected location of the aqueduct, lying in loose glial tissue (Figure 18.3c). This lesion should not be confused with simple forking, which is an anatomic variant (presence of

191

Developmental Neuropathology

Genetics of congenital hydrocephalus (and pathogenesis) A possible genetic etiology (cytogenetic abnormalities both numeric or structural, monogenic, or complex inherited conditions and multifactorial hereditary disorders) is present in about 40% of patients with congenital hydrocephalus. More than 600 syndromic causes have been reported so far and cannot be detailed here. The most important syndromes to be recognized include mutations in L1CAM gene, autosomal VACTERL-H association (vertebral anomalies, anal atresia, congenital cardiac disease, tracheoesophageal fistula, renal anomalies, radial dysplasia, and other limb defects plus hydrocephalus), and the X-linked forms (Mendelian Inheritance in Man number, MIM, 314390 related to ZIC3 gene and MIM 300514 related to FANCB gene) as well as secondary α-dystroglycanopathies. Other metabolic disorders associated with congenital hydrocephalus include mucopolysaccharidoses, alpha-mannosidosis, and Smith–Lemli–Opitz syndrome. In a study by Verhagen et al. [30], confined to primary forms, syndromes accounted for twothirds of cases, while nonsyndromic hydrocephalus accounted for one-third. In 30% of the syndromic group, a known cause was found. Imaging studies in both groups showed aqueduct stenosis, corpus callosum agenesis, septum pellucidum aplasia, secondary cerebral atrophy, and cerebellar hypoplasia, in addition to the presence of corticospinal hypoplasia in the syndromic group. In both groups, familial cases without identifiable genetic mutations were present. These authors proposed a flowchart based on clinical and molecular diagnostic parameters [30].

X-linked hydrocephalus (L1 syndrome and related disorders) L1 syndrome is the most frequent genetic cause of congenital hydrocephalus caused by mutations in the L1CAM gene at Xq28. L1 is a transmembrane glycoprotein belonging to the immunoglobulin superfamily of cell adhesion molecules. There is wide clinical inter- and intrafamilial variability, and the condition may manifest as a syndromic or nonsyndromic hydrocephalus in males [69]. Females may manifest minor features such as adducted thumbs and/or subnormal intelligence, hydrocephalus or rarely the complete phenotype. To date, more than 200 different L1CAM mutations have been reported [38,70]. Clinical presentation includes X-linked hydrocephalus with stenosis of the aqueduct of Sylvius (MIM 307000) [71], MASA syndrome (mental restriction, aphasia, spastic paraplegia and adducted thumbs) (MIM 303350) [72], X-linked complicated hereditary spastic paraplegia type 1 and X-linked complicated corpus callosum agenesis [73]. In hydrocephalus with stenosis of the aqueduct of Sylvius, fetuses and patients displaying a patent aqueduct have been reported [38,66,74]. Imaging studies reveal hydrocephalus with or without stenosis of the aqueduct of Sylvius, with corpus callosum agenesis/

192

hypogenesis and/or cerebellar hypoplasia, small brain stem, agenesis or hypoplasia of the corticospinal tracts [74] with absent decussation of the pyramids. In a neuropathological review of 138 fetuses genetically tested for X-linked hydrocephalus, of the six cardinal criteria (male gender, hydrocephalus, adducted thumbs, pyramidal tract agenesis/hypoplasia, stenosis of the aqueduct of Sylvius and agenesis/hypoplasia of the corpus callosum), the most specific criteria were adducted thumbs (71.93%), stenosis of the aqueduct of Sylvius (64.7%), and corticospinal tract abnormalities (62.07%; Figure 18.4a). Adducted thumbs had the highest predictive value, confirming postnatal clinical studies [75]. Abnormal thalami (Figure 18.4b) and mild or moderate infratentorial abnormalities, such as cerebellar hypoplasia, were also present in some cases. Only a few patients with aqueduct atresia spectrum, CNS malformations or polymalformation syndromes displayed five or more cardinal signs. Thus, when hydrocephalus is associated with at least three or more other cardinal signs in fetuses, screening of the L1CAM gene is recommended. Two main differential diagnoses should be considered: fetal alcohol syndrome and aqueduct atresia spectrum [38]. Two types of related disorders without a mutation in L1CAM have been reported. An X-linked factor other than L1 [30] and L1-like syndrome consist of phenocopies with no mutations in the L1CAM gene and in whom family history strongly suggests an autosomal recessive mode of transmission [38].

Autosomal recessive hydrocephalus Several reports describe familial cases of multiple female or mixed sex siblings having congenital hydrocephalus within a context of autosomal recessive inheritance [68,76]. Two responsible genes were been identified in 2013: MPDZ and CCDC88C. MPDZ gene (MIM 603785) is located on chromosome 9p, encodes a tight junction protein and corresponds to nonsyndromic hydrocephalus-2 (HYC2; hydrocephalus, nonsyndromic, autosomal recessive 2) [76]. It manifests with seizures, mildly decreased intelligence quotient, with simplified gyral pattern, corpus callosum abnormalities; one case had coloboma and congenital heart disease. Ependymal rosettes are found within the aqueduct, third and fourth ventricles. The other genetic cause is the homozygous mutation in the CCDC88C gene (MIM 611204) located on chromosome 14 [77,78] in autosomal recessive nonsyndromic hydrocephalus-1 (HYC1). CCDC88C encodes DAPLE (HkRP2), a Hook-related protein with a binding domain for the central Wnt signaling pathway protein Dishevelled. Three families have been reported. In living members, it manifests as congenital hydrocephalus with seizures, without any other malformation, dysmorphism, or neurologic anomalies. Motor and congnitive defects are variable. On MRI, ventricular dilatation may be associated with midline or extra-axial parietal cystic structures, a small vermis, an enlarged posterior fossa and biparietal polymicrogyria. Fetal pathology in one case at 21 weeks of gestation showed lateral

Hydrocephalus Chapter 18

(a)

(b)

Figure 18.4 L1CAM mutation in fetus aged 24 postovulatory weeks [38], presenting all the cardinal signs. (a) Section through the medulla showing agenesis of the pyramids. (b) Note the presence of hydrocephalus, corpus callosum agenesis, atrophy of the germinal zone, and fusion of the thalami. Reproduced with permission of Taylor and Francis.

ventricular dilatation with a thin six-layered cortex, a permeable aqueduct of Sylvius, and lung lymphangiectasias.

Animal models Although multiple animal models of congenital hydrocephalus have been developed (a complete list of all the models is not possible), the precise mechanisms of hydrocephalus remain unknown in several of them [7,20,79]. Acquired hydrocephalus is mainly induced by injections of kaolin [80] to induce congenital obstructive or communicating hydrocephalus in mice, rats, rabbits, cats, dogs, and sheep. Many other substances, such as trypan blue, salicylates, and cuprizone, administered to the mother were able to induce aqueduct stenosis. Hypovitaminosis A [56], hypovitaminosis B12 and deficiency of folic acid [21,81] were also reported to induce congenital hydrocephalus. Hydrocephalic Texas (H-Tx) rat strain and LEW/Jms rat strain are the result of spontaneous mutations and show a similar phenotype. H-Tx rat model has an incomplete formation of subcommissural organ, develops a congenital obstructive hydrocephalus following aqueduct stenosis in the late fetal and perinatal periods; while a few offspring develop mild ventriculomegaly and may survive [63,82]. Inbred strains of Wistar-Lewis (LEW/Jms) rats develop aqueduct stenosis before birth [83]. Several mouse models have shown the role of subcommissural organ impairment, alteration in ventricular ependymal cells and ciliary function in the development of congenital hydrocephalus

and/or in aqueduct stenosis. Pax6 mutant mouse [84], a mutation affecting the dorsal patterning such as Msx1 mutants [85], transgenic mice expressing ectopically Engrailed 1 or mutant wnt1 mouse [86] do not develop a subcommissural organ and show ependymal differentiation defects [85,86]. Subcommissural organ agenesis also affects mice expressing a transcript variant of RFX4 [87] and RFX3 mutants, who additionally have a defect in ependymal cell development. These two transcription factors play a role in brain patterning and ciliary function [87–89]. Hyh mice harbor a point mutation in alpha-SNAP, a component of the protein machinery responsible for diverse types of membrane fusion events [90]. They are born with moderate hydrocephalus and develop severe or slowly progressive congenital hydrocephalus with ependymal denudation and a dysfunctional subcommissural organ that precedes aqueduct stenosis [22,91–94]. The subcommissural organ showed signs of increased secretory activity; it released a material that aggregated but did not form a Reissner’s fiber [95]. Hy3 mice possessing homozygous-recessive Hydin gene mutations have primary ciliary dyskinesia and develop aqueduct stenosis during the progression of congenital hydrocephalus [67,96–98]. In axonemal dynein Mdnh5-mutant mice, the lack of ependymal flow due to immotile ependymal cilia causes closure of the aqueduct and subsequent formation of hydrocephalus during early postnatal brain development [99,100]. Several models of L1 deficiency have defects in the guidance of axons of the corticospinal tract, ventriculomegaly or hypoplastic vermis as shown in humans [101–104]. Interestingly, aqueduct stenosis was not present prior to the development of the congenital hydrocephalus in some models; the aqueduct, although

193

Developmental Neuropathology abnormally shaped, was normal in size [101,103] suggesting that the stenosis could be secondary to the deformation of the brain by massively enlarged ventricles [105]. Other models include Ro1 transgenic mice having communicating congenital hydrocephalus with subsequent ependymal denudation [106]; Mf1 forkhead winged helix [107] mutant, in which congenital hydrocephalus appears to be related to a primary defect in the development of the meninges, TGF beta1 overexpression [108,109], aquaporin deficiency [80], collagen deficiencies [110], nuclear factor I-A (Nfia) gene disruption [111].

7. 8.

9.

10.

11.

Treatment, future perspective, conclusions 12.

The etiology, the age at the time of onset, the severity of the ventriculomegaly, the degree of the damage to the brain, and the duration of the increased intracranial pressure significantly influence the clinical consequences of congenital hydrocephalus [7,25]. Ventriculomegaly may regress during fetal life [33]. Therapeutic options should be individualized to the child; some, but not all, changes may be preventable. CSF shunting remains the standard form of treatment. Endoscopic third ventriculostomy combined with choroid plexus cauterization seems to provide encouraging results [11]. Various pharmacological interventions are under investigation, aimed at reducing CSF production, enhancing CSF flow and absorption, decreasing neuroinflammation to provide neuroprotection, and improving recovery and regeneration of damaged tissues.

13.

14.

15.

16.

17.

18.

References 19. 1. McKechnie L, Vasudevan C, Levene M (2012) Neonatal outcome of congenital ventriculomegaly. Semin Fetal Neonatal Med 17: 301–7 2. Rekate HL (2011) A consensus on the classification of hydrocephalus: its utility in the assessment of abnormalities of cerebrospinal fluid dynamics. Childs Nerv Syst 27:1535–41 3. Fujimoto Y, Matsushita H, Plese JP, Marino R Jr (2004) Hydrocephalus due to diffuse villous hyperplasia of the choroid plexus. Case report and review of the literature. Pediatr Neurosurg 40: 32–6 4. Rekate HL, Aygok GA, Kouzelis K et al. (2012) Fifth International Hydrocephalus Workshop, Crete, Greece, May 20-23, 2010: themes and highlights. Acta Neurochir Suppl 113:1–7 5. Oi S (2011) Classification of hydrocephalus: critical analysis of classification categories and advantages of “Multi-categorical Hydrocephalus Classification” (Mc HC). Childs Nerv Syst 27:1523– 33 6. Mortazavi MM, Griessenauer CJ, Adeeb N et al. (2014) The choroid plexus: a comprehensive review of its history, anatomy,

194

20.

21.

22.

23.

24.

function, histology, embryology, and surgical considerations. Childs Nerv Syst 30:205–14 McAllister J.P 2nd (2012) Pathophysiology of congenital and neonatal hydrocephalus. Semin Fetal Neonatal Med 17:285–94 Oi S, Inagaki T, Shinoda M et al. (2011) Guideline for management and treatment of fetal and congenital hydrocephalus: Center of Excellence. Fetal and congenital hydrocephalus top 10 Japan guideline 2011. Childs Nerv Syst 27:1563–70 Milhorat TH, Hammock MK, Fenstermacher JD, Levin VA (1971) Cerebrospinal fluid production by the choroid plexus and brain. Science 173:330–2 Oreskovic D, Klarica M (2010) The formation of cerebrospinal fluid: nearly a hundred years of interpretations and misinterpretations. Brain Res Rev 64:241–62 Kahle KT, Kulkarni AV, Limbrick DD Jr, Warf BC (2015) Hydrocephalus in children. Lancet 387(10020):788–99 Weller RO (2005) Microscopic morphology and histology of the human meninges. Morphologie 89:22–34 Iliff JJ, Wang M, Liao Y et al. (2012) A paravascular pathway facilitates CSF flow through the brain parenchyma and the clearance of interstitial solutes, including amyloid beta. Sci Transl Med 4:147ra111 Catala M (2012) Development of the cerebrospinal fluid pathways during embryonic and fetal life in humans. In: G Cinalli, C Sainte-Rose and W Maixner, eds, Pediatric Hydrocephalus. Milan, Springer Nature, pp. 19–45 Gomez DG, DiBenedetto AT, Pavese AM et al. (1982) Development of arachnoid villi and granulations in man. Acta Anat (Basel) 111:247–58 Galarza M (2002) Evidence of the subcommissural organ in humans and its association with hydrocephalus. Neurosurg Rev 25:205–15 Perez-Figares JM, Jimenez AJ, Rodriguez EM (2001) Subcommissural organ, cerebrospinal fluid circulation, and hydrocephalus. Microsc Res Tech 52:591–607 Rodriguez EM, Oksche A, Montecinos H (2001) Human subcommissural organ, with particular emphasis on its secretory activity during the fetal life. Microsc Res Tech 52:573–90 Castaneyra-Perdomo A, Meyer G, Carmona-Calero E, et al. (1994) Alterations of the subcommissural organ in the hydrocephalic human fetal brain. Brain Res Dev Brain Res 79:316–20 Huh MS, Todd MA, Picketts DJ (2009) SCO-ping out the mechanisms underlying the etiology of hydrocephalus. Physiology (Bethesda) 24:117–26 Overholser MD, Whitley JR, O’Dell BL, Hogan AG (1954) The ventricular system in hydrocephalic rat brains produced by a deficiency of vitamin B12 or of folic acid in the maternal diet. Anat Rec 120:917–33 Vio K, Rodriguez S, Navarrete EH et al. (2000) Hydrocephalus induced by immunological blockage of the subcommissural organ-Reissner’s fiber (RF) complex by maternal transfer of antiRF antibodies. Exp Brain Res 135:41–52 Harding B, Golden JA (2014) Malformations. In S Love, A Perry, J Ironside H Budka, eds., Greenfield’s Neuropathology, 9th ed, Vol. 1. Boca Raton, FL, CRC Press, pp. 270–398 Munch T., Rasmussen ML, Wohlfahrt J et al. (2014) Risk factors for congenital hydrocephalus: a nationwide, register-based, cohort study. J Neurol Neurosurg Psychiatry 85:1253–9

Hydrocephalus Chapter 18 25. Garne E. Loane M, Addor MC et al. (2010) Congenital hydrocephalus: prevalence, prenatal diagnosis and outcome of pregnancy in four European regions. Eur J Paediatr Neurol 14:150–5 26. Jeng S, Gupta N, Wrensch M et al. (2011) Prevalence of congenital hydrocephalus in California, 1991–2000. Pediatr Neurol 45:67–71 27. Melo JR, de Melo EN, de Vasconcellos AG, Pacheco P (2013) Congenital hydrocephalus in the northeast of Brazil: epidemiological aspects, prenatal diagnosis, and treatment. Childs Nerv Syst 29:1899–903 28. Tully HM, Capote RT, Saltzman BS (2015) Maternal and infant factors associated with infancy-onset hydrocephalus in Washington State. Pediatr Neurol 52:320–5 29. Van Landingham M, Nguyen TV, Roberts A et al. (2009) Risk factors of congenital hydrocephalus: a 10 year retrospective study. J Neurol Neurosurg Psychiatry 80:213–17 30. Verhagen JM, Schrander-Stumpel CT, Krapels IP et al. (2011) Congenital hydrocephalus in clinical practice: a genetic diagnostic approach. Eur J Med Genet 54:e542–7 31. Pattisapu JV (2001) Etiology and clinical course of hydrocephalus. Neurosurg Clin N Am 12: 651–9, vii 32. Del Bigio MR (2010) Neuropathology and structural changes in hydrocephalus. Dev Disabil Res Rev 16:16–22 33. Gaglioti P, Oberto M, Todros T (2009) The significance of fetal ventriculomegaly: etiology, short- and long-term outcomes. Prenat Diagn 29:381–8 34. International Society of Ultrasound in Obstetrics and Gynecology Education Committee (2007) Sonographic examination of the fetal central nervous system: guidelines for performing the ‘basic examination’ and the ‘fetal neurosonogram’. Ultrasound Obstet Gynecol 29:109–16 35. Brunelle F (2012) Modern imaging of pediatric hydroceplus In G. Cinalli, C. Sainte-Rose, and W. Maixner (eds) Pediatric Hydrocephalus. Milan, Springer Nature, pp. 79–94 36. Vinters HV, Kleinschmidt-DeMasters BK (2014) General pathology of the central nervous system. In S. Love, A Perry, J Ironside H Budka, et al., eds., Greenfield’s Neuropathology, Boca Raton, FL, CRC Press, pp. 1–58 37. Humphreys P, Muzumdar DP, Sly LE, Michaud J (2007) Focal cerebral mantle disruption in fetal hydrocephalus. Pediatr Neurol 36:236–43 38. Adle-Biassette H, Saugier-Veber P, Fallet-Bianco C et al. (2013) Neuropathological review of 138 cases genetically tested for Xlinked hydrocephalus: evidence for closely related clinical entities of unknown molecular bases. Acta Neuropathol 126:427–42 39. Ludwig K, Salmaso R, Manara R et al. (2012) Apert syndrome with fused thalami. Fetal Pediatr Pathol 31:410–14 40. Pasquier L, Marcorelles P, Loget P et al. (2009) Rhombencephalosynapsis and related anomalies: a neuropathological study of 40 fetal cases. Acta Neuropathol 117:185–200 41. Dominguez-Pinos MD, Paez P, Jimenez AJ et al. (2005) Ependymal denudation and alterations of the subventricular zone occur in human fetuses with a moderate communicating hydrocephalus. J Neuropathol Exp Neurol 64:595–604 42. Sarnat HB (1995) Ependymal reactions to injury: a review. J Neuropathol Exp Neurol 54:1–15 43. Del Bigio MR (2010) Ependymal cells: biology and pathology. Acta Neuropathol 119:55–73 44. Friede RL (1989) Hydrocephalus-Special Pathology. In Developmental Neuropathology. Springer Berlin Heidelberg. pp. 231–46

45. Jeelani Y, McComb JG (2011) Congenital hydrocephalus associated with myeloschisis. Childs Nerv Syst 27:1585–8 46. Ternier J, Wray A, Puget S et al. (2006) Tectal plate lesions in children. J Neurosurg 104(6 Suppl): 369–76 47. Marcorelles P, Fallet-Bianco C, Oury JF et al. (2005) Fetal aqueductal glioneuronal hamartoma: a clinicopathological and physiopathological study of three cases. Clin Neuropathol 24: 155–62 48. Fliegauf M, Benzing T, Omran H (2007) When cilia go bad: cilia defects and ciliopathies. Nat Rev Mol Cell Biol 8:880–93 49. Jimenez AJ, Dominguez-Pinos MD, Guerra MM et al. (2014) Structure and function of the ependymal barrier and diseases associated with ependyma disruption. Tissue Barriers 2: e28426 50. Kosaki K, Ikeda K, Miyakoshi K et al. (2004) Absent inner dynein arms in a fetus with familial hydrocephalus-situs abnormality. Am J Med Genet A 129A:308–11 51. Lee L (2013) Riding the wave of ependymal cilia: genetic susceptibility to hydrocephalus in primary ciliary dyskinesia. J Neurosci Res 91:1117–32 52. Hirsch JF, Hirsch E, Sainte Rose C et al. (1986) Stenosis of the aqueduct of Sylvius. Etiology and treatment. J Neurosurg Sci 30:29– 39 53. Cinalli G, Spennato P, Ruggiero C et al. (2004) Aqueductal stenosis 9 years after mumps meningoencephalitis: treatment by endoscopic third ventriculostomy. Childs Nerv Syst 20:61–4 54. Shah N, Kale K (1999) Subacute sclerosing panencephalitis in a case of aqueductal stenosis. J Indian Med Assoc 97:501–3 55. Boros CA, Spence D, Blaser S, Silverman ED (2007) Hydrocephalus and macrocephaly: new manifestations of neonatal lupus erythematosus. Arthritis Rheum 57:261–6 56. No authors listed (1955) Vitamin A and hydrocephalus. Br Med J 1(4915):715–16 57. Pfeiffer G, Friede RL (1984) Unilateral hydrocephalus from early developmental occlusion of one foramen of Monro. Acta Neuropathol 64:75–7 58. Woollam DH, Millen JW (1953) Anatomical considerations in the pathology of stenosis of the cerebral aqueduct. Brain 76(1): 104–12 59. Beckett RS, Netsky MG, Zimmerman HM (1950) Developmental stenosis of the aqueduct of Sylvius. Am J Pathol 26:755–87 60. Timmons GD, Johnson KP (1970) Aqueductal stenosis and hydrocephalus after mumps encephalitis. N Engl J Med 283:1505–7 61. Emery JL, Staschak MC (1972) The size and form of the cerebral aqueduct in children. Brain 95:591–83 62. Bruni JE, Del Bigio MR, Cardoso ER, Persaud TV (1988) Hereditary hydrocephalus in laboratory animals and humans. Exp Pathol 35:239–46 63. Jones HC, Bucknall RM (1988) Inherited prenatal hydrocephalus in the H-Tx rat: a morphological study. Neuropathol Appl Neurobiol 14:263–74 64. Jones HC, Dack S, Ellis C (1987) Morphological aspects of the development of hydrocephalus in a mouse mutant (SUMS/NP). Acta Neuropathol 72:268–76 65. Borit A, Sidman RL (1972) New mutant mouse with communicating hydrocephalus and secondary aqueductal stenosis. Acta Neuropathol 21:316–31 66. Landrieu P, Ninane J, Ferriere G, Lyon G (1979) Aqueductal stenosis in X-linked hydrocephalus: a secondary phenomenon? Dev Med Child Neurol 21:637–42

195

Developmental Neuropathology 67. Raimondi AJ, Clark SJ, McLone DG (1976) Pathogenesis of aqueductal occlusion in congenital murine hydrocephalus. J Neurosurg, 45:66–77 68. Zhang J, Williams MA, Rigamonti D (2006) Genetics of human hydrocephalus. J Neurol 253:1255–66 69. Schrander-Stumpel C, Howeler C, Jones M et al. (1995) Spectrum of X-linked hydrocephalus (HSAS), MASA syndrome, and complicated spastic paraplegia (SPG1): clinical review with six additional families. Am J Med Genet 57:107–16 70. Vos YJ, Hofstra RM (2010) An updated and upgraded L1CAM mutation database. Hum Mutat, 31:E1102–9 71. Bickers DS, Adams RD (1949) Hereditary stenosis of the aqueduct of Sylvius as a cause of congenital hydrocephalus. Brain 72: 246–62 72. Bianchine JW, Lewis RC Jr (1974) The MASA syndrome: a new heritable mental retardation syndrome. Clin Genet 5:298–306 73. Stumpel C, Vos YJ (2004 updated 2014) L1 syndrome. In RA Pagon, Ardinger HH, Pagon RA et al., eds, GeneReviews, Seattle WA, University of Washington. Available at: Available from: https://www.ncbi.nlm.nih.gov/books/NBK1484 (accessed October 23, 2017) 74. Yamasaki M, Arita N, Hiraga S et al. (1995) A clinical and neuroradiological study of X-linked hydrocephalus in Japan. J Neurosurg 83:50–5 75. Vos YJ, de Walle HE, Bos KK et al. (2010) Genotype-phenotype correlations in L1 syndrome: a guide for genetic counselling and mutation analysis. J Med Genet 47:169–75 76. Al-Dosari, MS, Al-Owain M, Tulbah M et al. (2013) Mutation in MPDZ causes severe congenital hydrocephalus. J Med Genet 50:54–8 77. Drielsma A, Jalas C, Simonis N et al. (2012) Two novel CCDC88C mutations confirm the role of DAPLE in autosomal recessive congenital hydrocephalus. J Med Genet 49:708–12 78. Ekici AB, Hilfinger D, Jatzwauk M et al. (2010) Disturbed Wnt signalling due to a mutation in CCDC88C causes an autosomal recessive non-syndromic hydrocephalus with medial diverticulum. Mol Syndromol 1:99–112 79. Johanson C, Del Bigio M, Kinsman S et al. (2001) New models for analysing hydrocephalus and disorders of CSF volume transmission. Br J Neurosurg 15:281–3 80. Bloch O, Auguste KI, Manley GT, Verkman AS (2006) Accelerated progression of kaolin-induced hydrocephalus in aquaporin4-deficient mice. J Cereb Blood Flow Metab 26:1527–37 81. Newberne PM (1962) The subcommissural organ of the vitamin B12–deficient rat. J Nutr 76:393–413 82. Kohn DF, Chinookoswong N, Chou SM (1981) A new model of congenital hydrocephalus in the rat. Acta Neuropathol 54:211–18 83. Sasaki S, Goto H, Nagano H et al. (1983) Congenital hydrocephalus revealed in the inbred rat, LEW/Jms. Neurosurgery 13:548–54 84. Estivill-Torrus G, Vitalis T, Fernandez-Llebrez P, Price DJ (2001) The transcription factor Pax6 is required for development of the diencephalic dorsal midline secretory radial glia that form the subcommissural organ. Mech Dev 109:215–24 85. Fernandez-Llebrez P, Grondona JM, Perez J et al. (2004) Msx1deficient mice fail to form prosomere 1 derivatives, subcommissural organ, and posterior commissure and develop hydrocephalus. J Neuropathol Exp Neurol 63:574–86

196

86. Louvi A, Wassef M (2000) Ectopic engrailed 1 expression in the dorsal midline causes cell death, abnormal differentiation of circumventricular organs and errors in axonal pathfinding. Development 127:4061–71 87. Blackshear PJ, Graves JP, Stumpo DJ et al. (2003) Graded phenotypic response to partial and complete deficiency of a brainspecific transcript variant of the winged helix transcription factor RFX4. Development 130:4539–52 88. Baas D, Meiniel A, Benadiba C et al. (2006) A deficiency in RFX3 causes hydrocephalus associated with abnormal differentiation of ependymal cells. Eur J Neurosci 24:1020–30 89. Bonnafe E, Touka M, AitLounis A et al. (2004) The transcription factor RFX3 directs nodal cilium development and left-right asymmetry specification. Mol Cell Biol 24:4417–27 90. Hong HK, Chakravarti A, Takahashi JS (2004) The gene for soluble N-ethylmaleimide sensitive factor attachment protein alpha is mutated in hydrocephaly with hop gait (hyh) mice. Proc Natl Acad Sci U S A 101:1748–53 91. Jimenez AJ, Garcia-Verdugo JM, Gonzalez CA et al. (2009) Disruption of the neurogenic niche in the subventricular zone of postnatal hydrocephalic hyh mice. J Neuropathol Exp Neurol 68:1006– 20 92. Jimenez AJ, Tome M, Paez P et al. (2001) A programmed ependymal denudation precedes congenital hydrocephalus in the hyh mutant mouse. J Neuropathol Exp Neurol 60:1105–19 93. Roales-Bujan R, Paez P, Guerra M et al. (2012) Astrocytes acquire morphological and functional characteristics of ependymal cells following disruption of ependyma in hydrocephalus. Acta Neuropathol 124:531–46 94. Wagner C, Batiz LF, Rodriguez S et al. (2003) Cellular mechanisms involved in the stenosis and obliteration of the cerebral aqueduct of hyh mutant mice developing congenital hydrocephalus. J Neuropathol Exp Neurol 62:1019–40 95. Perez-Figares JM, Jimenez AJ, Perez-Martin M et al. (1998) Spontaneous congenital hydrocephalus in the mutant mouse hyh. Changes in the ventricular system and the subcommissural organ. J Neuropathol Exp Neurol 57:188–202 96. Davy BE, Robinson ML (2003) Congenital hydrocephalus in hy3 mice is caused by a frameshift mutation in Hydin, a large novel gene. Hum Mol Genet 12:1163–70 97. Dawe HR, Shaw MK, Farr H, Gull K (2007) The hydrocephalus inducing gene product, Hydin, positions axonemal central pair microtubules. BMC Biol 5:33 98. Lechtreck KF, Delmotte P, Robinson ML et al. (2008) Mutations in Hydin impair ciliary motility in mice. J Cell Biol 180:633–43 99. Ibanez-Tallon I, Pagenstecher A, Fliegauf M et al. (2004) Dysfunction of axonemal dynein heavy chain Mdnah5 inhibits ependymal flow and reveals a novel mechanism for hydrocephalus formation. Hum Mol Genet 13:2133–41 100. Mitchison HM, Schmidts M, Loges NT et al. (2012) Mutations in axonemal dynein assembly factor DNAAF3 cause primary ciliary dyskinesia. Nat Genet 44:381–9, S1–2 101. Cohen NR, Taylor JS, Scott LB et al. (1998) Errors in corticospinal axon guidance in mice lacking the neural cell adhesion molecule L1. Curr Biol 8:26–33 102. Dahme M, Bartsch U, Martini R et al. (1997) Disruption of the mouse L1 gene leads to malformations of the nervous system. Nat Genet 17:346–9

Hydrocephalus Chapter 18 103. Fransen E, D’Hooge R, Van Camp G et al. (1998) L1 knockout mice show dilated ventricles, vermis hypoplasia and impaired exploration patterns. Hum Mol Genet 7:999–1009 104. Kamiguchi H, Hlavin ML, Lemmon V (1998) Role of L1 in neural development: what the knockouts tell us. Mol Cell Neurosci 12:48– 55 105. Rolf B, Kutsche M, Bartsch U (2001) Severe hydrocephalus in L1deficient mice. Brain Res 891:247–52 106. McMullen AB, Baidwan GS, McCarthy KD (2012) Morphological and behavioral changes in the pathogenesis of a novel mouse model of communicating hydrocephalus. PLoS One 7:e30159 107. Kume T, Deng KY, Winfrey V et al. (1998) The forkhead/winged helix gene Mf1 is disrupted in the pleiotropic mouse mutation congenital hydrocephalus. Cell 93:985–96

108. Crews L, Wyss-Coray T, Masliah E (2004) Insights into the pathogenesis of hydrocephalus from transgenic and experimental animal models. Brain Pathol 14:312–6 109. Galbreath E, Kim SJ, Park K et al. (1995) Overexpression of TGFbeta 1 in the central nervous system of transgenic mice results in hydrocephalus. J Neuropathol Exp Neurol 54:339–49 110. Utriainen A, Sormunen R, Kettunen M et al. (2004) Structurally altered basement membranes and hydrocephalus in a type XVIII collagen deficient mouse line. Hum Mol Genet 13: 2089–99 111. das Neves L, Duchala CS, Tolentino-Silva F et al. (1999) Disruption of the murine nuclear factor I-A gene (Nfia) results in perinatal lethality, hydrocephalus, and agenesis of the corpus callosum. Proc Natl Acad Sci U S A 96:11946–51

197

19

Antenatal Disruptive Lesions Brian N. Harding Department of Pathology and Laboratory Medicine, Children’s Hospital of Philadelphia, Philadelphia, PA, USA

Definition, major synonyms and historical perspective Encephaloclastic lesions of the second and third trimesters disrupt normal development and are to be distinguished from primary malformations, which result from an intrinsically abnormal developmental process [1]. There are two categories. Smooth-walled lesions originating in mid-gestation and associated with surrounding cortical disorganization include hydranencephaly (synonym, bubble brain), where a thin, translucent, membrane replaces much of the cerebral mantle (Figure 19.1); porencephaly, exhibiting one or more circumscribed defects in the cerebral wall, variably communicating with the ventricle (Figure 19.2); and an intermediate lesion of extensive bilateral defects, but with residual intact parasagittal cortex, imaginatively labelled basket brain. By contrast, multicystic encephalopathy refers to ragged defects of hemispheric gray and white matter unaccompanied by cortical malformation (Figure 19.3), with its origin in the third trimester or, rarely, postnatally. Early descriptions date from the nineteenth century [2,3]. The term “schizencephaly,” used by some authors synonymously with porencephaly, and particularly by neuroradiologists, confusingly, for clefts not communicating with the ventricle or lined by abnormal gray matter, is considered by most neuropathologists to be outdated, and inconsistent with modern pathogenetic theories and morphologic studies. Any apparent (on imaging) distinction is undermined by evidence that the same gene mutation is related to lesions given both these labels.

(a)

(b)

Figure 19.1 Hydranencephaly, 22 weeks of gestation. Twin, with co-twin deceased at 17 weeks. (a) Diaphenous bubble-like hemispheres photographed in water, vertex view. (b) When removed from water the membranes collapse showing the shape; demarcated areas of destruction involving internal carotid artery territory and surrounded by ridged polymicrogyric cortex. Reproduced with permission from Ellison et al. [15] and Elsevier.

Developmental Neuropathology, Second Edition. Edited by Homa Adle-Biassette, Brian N. Harding, and Jeffrey A. Golden. © 2018 John Wiley & Sons Ltd. Published 2018 by John Wiley & Sons Ltd.

199

Developmental Neuropathology (a)

(b)

Figure 19.2 Porencephaly (a) A full-thickness defect in the occipital lobe communicating with the ventricle. Note the radial pattern of the surrounding convolutions, the cortex extending over the edge of the defect into the cleft; (b) bilateral porencephaly in 32-weeks of gestation twin (co-twin deceased at 19 weeks of gestation). Macrophages and disorganized immature neural tissue fill the center of the defects, which are fringed by polymicrogyria. Reproduced with permission from Ellison et al. [15] and Elsevier.

(a)

Normal embryology The most relevant embryology is the maturational state of macrophages and astrocytes at the time of the initiating insult. Phagocytosis by macrophages appears to occur earlier in development than recognizable astrocytic responses; thus, necrotic tissue may be resorbed without trace of glial repair. Macrophages appear as early as 11 weeks of gestation [4] and are demonstrable by immunohistochemistry early in the second trimester. But astrocytosis only begins to be observed between 20 and 23 weeks of gestation.

(b)

Epidemiology Overall incidence of the porencephaly–hydranencephaly spectrum estimated from imaging studies is 8.3 per 100 000 live births [5], but this may be an overestimate, given the tendency to misinterpret cystic lesions confined to the white matter as porencephaly. There are no comparable estimates for multicystic encephalopathy.

Clinical features, investigations and important differential diagnosis

Figure 19.3 Multicystic encephalopathy. (a) Vertex view of cystic convolutions. (b) Coronal section. A “spider’s web” of fine gliovascular bands replaces deep cortex, white matter and basal ganglia.

200

In porencephaly, although survival to adulthood is possible, clinical manifestations are more usually severe mental restriction and blindness, seizures (less often), tetraplegia, and decerebrate rigidity. Generally, the involvement of the basal ganglia and hypothalamus in hydranencephaly results in impaired

Antenatal Disruptive Lesions Chapter 19

thermoregulation, disordered sleep, sucking and swallowing, and therefore early demise. If the basal ganglia are preserved, survival may be prolonged for a few years, but spasticity, epilepsy, and minimal psychomotor development are the rule. The head circumference, usually normal at birth, may increase over the first few months, the hydrocephalus thought to result from obstruction at the aqueduct or foramen of Monro. Ultrasound and magnetic resonance imaging are the most useful investigations, and are regularly employed for intrauterine evaluation. When these lesions are suspected, possible antecedents should be sought in the clinical history. Differential diagnosis of hydranencephaly includes severe obstructive hydrocephalus and Fowler’s familial hydranencephaly syndrome (Mendelian Inheritance in Man number, MIM, 2255790; mapped to 14q24.3), which, although indistinguishable macroscopically and ultrasonographically, is easily recognizable postmortem, given the pathognomonic glomerular vasculopathy present throughout the central nervous system.

Pathology Macroscopically, porencephalic defects are smooth walled and surrounded by disorganized convolutions, radiating or irregular (Figure 19.2a,b). The defect may be superficial or deep, gaping and wide, or a narrow cleft, but the typical porus breeches the full thickness of the hemisphere, communicates with the ventricle, and is covered externally by a thin membrane. Defects are often bilateral and centered around the Sylvian fissures or central sulci. If the defect is unilateral, further gyral abnormality, usually polymicrogyria, may be present in a similar part of the contralateral hemisphere. Occasional defects are orbital, parasagittal or occipital. Microscopically, the abnormal cortex surrounding the defect may show polymicrogyria or irregular islands of gray matter with frequent calcifications, and this tissue extends over the edge of the porus into the cleft, where it meets tissue from the ventricular wall extending upwards into the cleft. The ventricular covering is largely gliotic and lacking ependyma, and this glial tissue covers the contiguous cortex before contributing to the inner part of the covering membrane, the outside of which is arachnoid. Other associated findings include periventricular nodular gray heterotopias, further areas of polymicrogyria, deficiency of the septum pellucidum and thalamic atrophy or hypoplasia. In reality, the histologic features of basket brain are similar, but the bilateral defects are much larger involving much of both middle cerebral vascular territories while sparing parasagittal cortex (supplied by the anterior cerebral arteries). At the severe end of this spectrum of mid-gestational cerebral disruptions, in hydranencephaly (Figure 19.1a,b), almost the whole cerebral mantle is replaced by a translucent membrane (transillumination of the infant’s head can be dramatic) though inferior temporal and occipital lobes may be spared. The

membrane has an outer layer of connective tissue and an inner gliotic layer including occasional neurons, mineralized debris and hemosiderin-laden macrophages. This glial tissue fuses with the molecular layer of any surrounding surviving cortex and covers it for a short distance. The surrounding cortex may show polymicrogyria. Multicystic encephalopathy, by contrast, presents as a sponge or net-like transformation of the white matter and deeper layers of the cortex involving large parts of the cerebral hemispheres (Figure 19.3a,b). This multicystic appearance is produced by numerous gliovascular strands enmeshing collections of lipid-laden macrophages. No gyral maldevelopment is present. Basal ganglia may be normal, but often show bilateral cystic necrosis.

Genetics and pathogenesis Scattered familial cases have been reported, and evidence for mutations in genes encoding type IV collagen alpha-1 and alpha2 (COL4A1 and 2) chains has been demonstrated in patients radiologically classified as either porencephaly or schizencephaly [6,7]. Most cases are sporadic, however. The most likely final common pathogenetic pathway is infarction due to hypoxic ischemic encephalopathy; the symmetry and topography of the lesions suggesting perfusion failure of the carotids or middle and anterior cerebral arteries since vascular occlusion is not demonstrated [8,9]. The presence of polymicrogyria in many examples of porencephaly and hydranencephaly suggests that the insult occurs in mid-gestation. It has been argued that the porus represents the ischemic epicentre [10,11], with a penumbra of less intense ischemia where polymicrogyria results. Such a postulate is supported in some fetal morphological studies [12] and the cases of twins illustrated in Figures 19.1 and 19.2. Twinning is a well-recognized antecedent for these encephaloclastic lesions, together with maternal poisoning, attempted suicide, attempted abortion and anaphylactic shock. Fetal infection, especially toxoplasmosis and cytomegalovirus are other antecedents [13].

Animal models and pathogenesis Experimental ligation of both carotids in a macaque has produced hydranencephaly [14].

Treatment, future perspective, conclusions Anti-epileptic measures and physical therapy are the mainstays of treatment. Increasing hydrocephalus may require surgical shunting. It is likely that further genetic causes will be uncovered.

201

Developmental Neuropathology

References 1. Spranger J, Benirschke K, Hall JG et al. (1982) Errors of morphogenesis: concepts and terms. Recommendations of an international working group. J Pediatr 100:160–5 2. Heschl R (1861) Ein neuer Fall von Porencephalie. Vierteljahrschrift fur ̏ praktikale Heilkunde Prague 72:102–4 3. Cruveilhier J (1829) Anatomie-pathologique du corps humain. Paris, Bailli`ere 4. Kershman J (1939) Genesis of microglia in the human brain. Arch Neurol Psychiatry 41:24-50 5. Hino-Fukuyo N, Togashi N, Takahashi R et al. (2015) Neuroepidemiology of porencephaly, schizencephaly, and hydranencephaly in Miyagi Prefecture, Japan. Pediatr Neurol. 54:39–42.e1 6. Yoneda Y, Haginoya K, Kato M et al. (2015) Phenotypic spectrum of COL4A1 mutations: porencephaly to schizencephaly. Ann. Neurol 73: 48–57 7. Yoneda Y, Haginoya K, Arai H et al. (2012) De novo and inherited mutations in COL4A2, encoding the type IV collagen alpha-2 chain cause porencephaly. Am J Hum Genet 90:86–90

202

8. Dekaban A (1965) Large defects in cerebral hemispheres associated with cortical dysgenesis. J Neuropath Exp Neurol 24:512–30 9. Lange-Cosack H (1944) Die Hydranencephalie (Blasenhirn) als Sonderform der Grosshirnlosigkeit. Arch Psychiatr Nervenkrankheit 117:1–51 10. Levine DN, Fisher MA, Caviness VS Jr (1974) Porencephaly with microgyria: a pathologic study. Acta Neuropathol 29:99–113 11. Lyon G, Robain O (1967) Comparative study of prenatal and perinatal circulatory encephalopathies (hydranencephalies, porencephalies) and cystic encephalomalacias of the white matter. Acta Neuropathol 9:79–98 12. Norman MG (1989) Bilateral encephaloclastic lesions in a 36 week gestation fetus: effect on neuroblast migration. Can J Neurol Sci 7:191–4 13. Harding BN, Golden JA (2015) Malformations In: Love S, Perry A, Ironside JW, Budka H, eds., Greenfields Neuropathology, 9th ed., Boca Raton, FL, CRC Press, pp. 270–398 14. Myers RE (1969) Brain pathology following fetal vascular occlusion: an experimental study. Invest Ophthalmol 8:41–50 15. Ellison D, Love S, Chimelli L et al. (2013) Neuropathology. A Reference Text of CNS Pathology, 3rd ed. London, Mosby

20

Hemorrhagic Lesions Marc R. Del Bigio1,2 1 Department 2 Diagnostic

of Pathology, University of Manitoba, Winnipeg, Canada Services Manitoba, Children’s Hospital Research Institute of Manitoba, Winnipeg, Canada

Definition, major synonyms and historical perspective Intracranial bleeding is best described according to the anatomical location and the circumstance. It may be due to physical trauma, intravascular hypertension, thrombosis, or breakdown of vessel integrity following infarction. The location and type of hemorrhage are dictated, at least in part, by the stage of development. Periventricular hemorrhage (PVH) refers to bleeding adjacent to the lateral ventricles, typically but not exclusively from the germinal matrix. Roughly equivalent terms include ganglionic eminence hemorrhage, subependymal hemorrhage, and germinal matrix hemorrhage. Intraventricular hemorrhage (IVH) refers to blood collections within the ventricles. These collections usually arise from PVH or choroid plexus. Brain hemorrhage has been recognized as a cause of perinatal death since the late 1800s, but systematic clinical investigation did not begin until the 1960s [1]. This chapter confines itself to the discussion of hemorrhagic lesions that occur in utero or in the perinatal period. There is some overlap between this chapter and those concerning hydrocephalus (Chapter 18), antenatal disruptive lesions (Chapter 19), white matter lesions (Chapter 21), and gray matter lesions (Chapter 22). Brain trauma, which can result in hemorrhage, is covered in pediatric head injury (Chapter 23).

Normal development This section focuses on regions of immature brain prone to hemorrhage and the brain vasculature. Smooth muscle accumulation

around artery walls begins in deep cerebral structures at about 30 weeks of gestation and progresses outward toward the brain surface [2]. This correlates with the progressive improvement in cerebral blood flow autoregulation [3]. The density of the microvascular (including capillary) bed is roughly stable in the second half of gestation; highest densities are in the germinal regions and cerebellum [4]. Blood vessels in the germinal tissue may be fragile because the surrounding tissues are immature and structurally unstable [5]. The reader is referred to general reviews that cover human fetal blood vessel development [6]. Because brain hemorrhage in the fetus or premature infant often arises from the periventricular germinal tissue, a good understanding of the developing brain layers is important [7]. The ventricular zone immediately adjacent to the ventricle gives rise to most glutamatergic excitatory neurons by around 18 weeks of gestation, with some generated as late as 28 weeks [8]. The adjacent subventricular zone is mitotically active throughout the brain until 18–20 weeks of gestation; thereafter it becomes more restricted to the region overlying the caudate nucleus (Figure 20.1a). Subventricular zone volume peaks at 23–25 weeks of gestation [9] and cell proliferation peaks at 20–26 weeks, reaching low levels by 30–32 weeks [10]. The subventricular zone site over the caudate nucleus head is called the ganglionic eminence. Most neocortical GABAergic inhibitory interneurons arise from the ventral ganglionic eminences [11]; in the neocortex, this continues to around 27 weeks [12] and in the thalamus to around 34 weeks [13]. The gangliothalamic migratory pathway [14] can be damaged by PVH. During the second half of gestation, the ganglionic eminence gives rise to glial precursors for the cerebrum [15]. The choroid plexus is the other common source of IVH. Between 17 and 29 weeks of gestation, stromal mesenchyme surrounding the vasculature is weak [16].

Developmental Neuropathology, Second Edition. Edited by Homa Adle-Biassette, Brian N. Harding, and Jeffrey A. Golden. © 2018 John Wiley & Sons Ltd. Published 2018 by John Wiley & Sons Ltd.

203

Developmental Neuropathology (a)

(b) 90

70

3000

60

IVH incidence

GE thickness (μm)

80 3500

2500 2000 1500

40 30

1000

20

500

10

0

0 15

20

25

30

35

40

Post-conception age (weeks) Figure 20.1 (a) Scattergram showing the maximum thickness of the ganglionic eminence (GE) over the head of the caudate nucleus as a function of gestational age (weeks). The third order polynomial regression curve shows that germinal tissue in this region is maximal from around 20–26 weeks, and involutes shortly before or around full-term gestation [10,18]. (b) Scattergram showing the

Epidemiology Incidence and prevalence Identification of hemorrhage depends on in vivo imaging or autopsy examination. Autopsy studies are biased because they evaluate only cases with the worst outcome. However, they offer the best opportunity to search for small hemorrhages. In utero bleeding related to fetal clotting disorders or trauma to the maternal abdomen is rare. The largest studies of perinatal autopsies and imaging, spanning the 1950s to the 1990s [17–19], in addition to more recent imaging studies [20,21], show that small subdural, subarachnoid, and intracerebral hemorrhages are common [20–40% of live births), even in apparently asymptomatic term infants. Approximately 16% of all births are premature (less than 37 weeks of gestation) and around 2% are extremely premature (less than 31 weeksof gestation) [22]. The incidence and magnitude of detectable PVH/IVH is an inverse function of gestational age at the time of birth (Figure 20.1b). It occurs in 40–50% of infants born at less than 26 weeks [23]. The incidence and severity of PVH/IVH has declined, likely a consequence of better intensive care management of premature infants [24]. However, because smaller infants are now salvaged, PVH/IVH is still frequent [25]. Clinically significant brain hemorrhage in term infants is rare (less than 3%); low platelet count and assisted delivery are the main risk factors [26,27]. Sex and age distribution Prospective ultrasound studies show that approximately 60% of hemorrhages occur within the first day and almost all have occurred by three to four days after premature birth [28,29]. Across studies, there is no consistently identified significant sex or gender influence.

204

50

45

24

26

28

30

32

34

36

Post-conception age at birth (weeks) incidence of intraventricular hemorrhage (IVH) as a function of gestational age (in weeks) at the time of birth. The data were compiled from four large, prospective cranial ultrasound studies of premature infants conducted in the early to mid-1990s [24,39,82,83]. The second order polynomial regression curve shows that the risk of IVH becomes low after around 31 weeks of gestation.

Risk factors The main risk factors for IVH are young gestational age and low weight. Additional risk factors include delivery method, premature rupture of membranes, patent ductus arteriosus, mechanical ventilation, and hyperglycemia [30]. Cesarean section and prenatal steroids are associated with reduced incidence of IVH in premature infants [31]. Many other factors secondary to prematurity and respiratory compromise are involved, but are not independent determinants.

Clinical features Signs and symptoms Most intracranial hemorrhage is clinically occult and may be detected only by imaging [32]. Infants who suffer significant intracranial hemorrhage may exhibit irritability, decreased level of consciousness, tense fontanel from an elevated intracranial pressure, lowered hematocrit due to blood extravasation, or seizures. Severe IVH carries a high risk of abnormal neurologic outcome. Among those with grade 4 IVH (grades explained in Imaging section), death occurs in approximately half; the worst outcome is in infants born at less than 26 weeks of gestational age [33]. Follow-up of premature infants to three years of age showed that those with grade 1 or 2 IVH had neurosensory impairment (22%), developmental delay (7.8%), cerebral palsy (10.4%), and deafness (6.0%), even in the absence of white matter injury. Infants with grade 3 or 4 IVH had higher rates of developmental delay (17.5%), cerebral palsy (30%), deafness (8.6%), and blindness (2.2%) [34,35]. Subdural hematoma may have grave consequences if the intracranial pressure is elevated or if it is located in the posterior fossa. Subgaleal hemorrhage can be associated with

Hemorrhagic Lesions Chapter 20

considerable loss of circulating blood volume in around 0.05% of births [36].

Imaging Imaging studies are the primary means for diagnosis of brain hemorrhage in the fetus or infant [37]. The first grading system for peri-/intraventricular hemorrhage was based on computed tomography (CT) scans [38]. The grading system, which takes into account the extent of hemorrhage in the ganglionic eminence and quantity of blood in the lateral ventricles, has been adapted for ultrasound: grade 1, isolated PVH; grade 2, PVH with IVH but no ventricle enlargement; grade 3, IVH with ventricles expanded; grade 4, PVH and IVH with extension into parenchyma (Figure 20.2a). Other grading systems have been suggested [39,40]. Magnetic resonance imaging (MRI) is more sensitive for detecting hemorrhage in white matter and yields much better anatomical detail (Figure 20.2b). Studies from 2013 and 2015 show that ultrasonography detects only 60% of the grade 1/2 PVH/IVH detected by MRI [41,42]. Among fetuses with ventriculomegaly detected by ultrasound, approximately 14 in 1000 will have IVH demonstrable by in utero MRI [43]. Sequential ultrasonography of premature infants seldom misses lesions detected by late MRI at term-equivalent age [44]. Laboratory findings Few tests beyond imaging are specifically relevant to the diagnosis and management of PVH/IVH in infants. Hemoglobin levels

(a)

and clotting parameters must be assessed. Maternal antiplatelet antibodies may cause neonatal alloimmune thrombocytopenia [45]. The cerebrospinal fluid may be bloody, but protein markers of brain damage are not yet useful [46].

Differential diagnosis The diagnosis of hemorrhage is readily made by imaging studies or by gross examination. At autopsy, the most challenging aspect for the pathologist is to identify the cause of spontaneous in utero hemorrhage, some of which can carry risk of recurrence.

Pathology Macroscopy Subdural and subarachnoid blood collections are generally thin (Figure 20.3) and rarely of clinical significance [47]. Large subdural hematomas can result when the sinuses of the falx cerebri or tentorium are torn during traumatic birth with excessive head molding [48]. PVH is typically located in the anterior ganglionic eminence over the caudate nucleus. It is often multifocal and asymmetric, and is often associated with IVH (Figure 20.4a) [49,50]. IVH is seldom due to extension of white matter hemorrhage. Blood that enters the ventricles can obstruct the aqueduct of Sylvius, and pass through the fourth ventricle to fill the subarachnoid compartment in the posterior fossa [50,51]. Large intraventricular clots degrade very slowly. Although the ventricles enlarge transiently in infants with severe IVH, progressive

(b)

Figure 20.2 (a) Sonogram in the coronal plane through the anterior fontanel showing the brain of a 28-weeks of gestation premature infant imaged at one day of age. There is a large, echogenic (bright; arrow), unilateral intraventricular hematoma. (b) Axial T2-weighted magnetic resonance image from a premature infant born at 30 weeks of gestation and imaged at 34 weeks. Three separate foci of residual hemosiderin at sites of periventricular hemorrhage are evident (dark areas; arrows).

205

Developmental Neuropathology

Figure 20.3 Photograph showing the medial surface of a hemisphere from a premature infant who was born alive at 22 weeks and died immediately after birth. Extensive subarachnoid hemorrhage is evident on the brain surface.

hydrocephalus requiring treatment only occurs in around 20% [45]. Lateral extension of germinal matrix hemorrhage is associated with destruction of the maturing brain parenchyma. Following resolution of the hematoma (typically a few months), this leaves a focal expansion or a smooth-walled cyst at the lateral angle of the lateral ventricle (Figure 20.4b). Less often, the cerebellum may be grossly hemorrhagic (Figure 20.5); this can represent primary hemorrhage or conversion of hypoxic–ischemic damage [52]. (a)

Figure 20.5 Photograph showing the ventral surface of the brain from an infant who was born at 36 weeks of gestation, sustained hypoxic–ischemic brain damage, developed secondary hemorrhagic infarction in the cerebellum, and died two days later.

Histopathology Microscopic diagnosis is of greatest value in the acute stages. Blood cells become intermingled with germinal matrix cells, which are not cohesive. With care, microscopic analysis might demonstrate a specific site of vein disruption; while it offers understanding of the process, microscopic analysis does not contribute significantly to the diagnosis. If a site of bleeding cannot be demonstrated in the periventricular tissue, microscopic (b)

Figure 20.4 (a) Photograph showing a coronal slice of the brain from a 26-weeks of gestation infant, who died three days after birth. There is bilateral ganglionic eminence hemorrhage with lateral extension into the brain parenchyma on the right and near complete filling of the lateral ventricles, which are moderately enlarged. (b) Photograph showing a coronal brain slice from a three-year-old child who was born prematurely, suffered a germinal matrix hemorrhage, and survived with hemiplegia and a seizure disorder (“cerebral palsy”). The site of the hemorrhage is marked by focal expansion of the frontal horn of the right lateral ventricle.

206

Hemorrhagic Lesions Chapter 20

hemorrhage site along the ventricle wall (Figure 20.8b) [53]. This initiates a much different investigation and counseling process than demonstration of a primary developmental abnormality. Microscopic examination of infant brains after PVH/IVH shows that erythrocytes begin to degenerate by one day. Hemosiderin is evident in macrophages by three days [54]. Neutrophilic and lymphocytic inflammation in the adjacent tissue is mild. The surrounding tissue may become necrotic if the hematoma is large and secondary infarction ensues [55]. Large clots degrade slowly; after one month only the outer 1–2 mm may show signs of degradation and organization. Following clot resolution, focal disruption of the ependymal surface is apparent; residual hemosiderin and mineralization may be detected along the ventricle wall for many months (Figure 20.8b). Residual reactive astroglial and microglial changes are typically subtle.

Figure 20.6 Photomicrograph of choroid plexus hemorrhage and intraventricular hematoma. Tissue is from an infant born at 25 weeks of gestational age who lived a few hours. (Hematoxylin and eosin stain; original magnification × 200).

examination of the choroid plexus surrounded by intraventricular blood clot may reveal vascular disruption in the plexus (Figure 20.6). Hemorrhagic conversion of ischemic white matter lesions is often bilateral, symmetric, and located in the posterior white matter. It consists of innumerable petechial hemorrhages with minimal space occupying effect (Figure 20.7). Microscopic examination is extremely valuable in cases where in utero bleeding is suspected but not grossly obvious. Fetuses aborted because of hydrocephalus may have blood debris obstructing the cerebral aqueduct (Figure 20.8a), with only microscopic foci of residual

Figure 20.7 Low magnification photomicrograph showing multiple petechial hemorrhages in the parietal white matter of an infant who was born at 28 weeks of gestation and who survived for 18 hours (hematoxylin and eosin stain; original magnification × 40).

Immunohistochemistry and ultrastructure Although specialized microscopic studies have contributed to the understanding of the pathogenesis of germinal matrix hemorrhage and to the understanding of reactions in damaged brain, they are not necessary for diagnosis.

Pathogenesis and genetics The pathogenesis differs for the various types and locations of intracranial bleeding, although all ultimately depends on a mechanical disruption of blood vessel walls due to external (deformation) or internal (pressure) mechanical stress. In utero brain hemorrhage often has no obvious cause; in some cases, maternal trauma, cocaine exposure, or coagulation defect can be documented [56]. Following premature birth, PVH appears to arise from large diameter, thin-walled veins, which lack significant adventitia or support from surrounding immature brain cells [57]. The veins are prone to rupture during or shortly after the distortions that occur at the time of the birth. They may also rupture during fluctuations of arterial, venous, and intracranial pressures, which occur in premature infants on hemodynamic and ventilatory support [32]. Premature infants lack normal autoregulation so blood flow in the brain passively follows blood pressure [58]. Ventricular enlargement can occur acutely because the cerebral aqueduct or fourth ventricle outlets are obstructed by clotted blood or debris. Although this may resolve spontaneously, persistent progressive ventricular enlargement can be the consequence of scarring in the cerebrospinal fluid pathways. Secondary hemorrhage at sites of periventricular leukomalacia could be related to arterial reperfusion into capillaries damaged by ischemia [32]. Hemorrhagic and ischemic conditions often coexist indicating that one may predispose to the other or that the underlying risk factors are the same [59]. Clinically significant subdural blood collections are caused by tearing of the falx cerebri, tentorium, or venous sinuses during the severe head distortions that occur at the time

207

Developmental Neuropathology

(a)

(b)

Figure 20.8 (a) Photomicrograph showing hemosiderin-containing macrophages (stained blue) obstructing the cerebral aqueduct of a 20-week fetus that was aborted following ultrasonographic demonstration of hydrocephalus (Perls Prussian blue stain; × 200). (b) Photomicrograph showing an extrusion of macrophages and mineralized debris arising from the ependymal surface of an infant who was born at 25 weeks of gestation, suffered a large intraventricular hemorrhage, and died four weeks later (hematoxylin and eosin stain; original magnification × 400).

of birth [60]. However, extreme head distortion at the time of delivery is not a prerequisite for mild degrees of hemorrhage or for germinal matrix hemorrhage. Late consequences of PVH commonly include severe neurodevelopmental delay [34]. In addition to direct tissue destruction, brain development may be impaired because proliferation in the germinal tissue is suppressed, with reduced generation of neurons and glial cells [10]. Genetic factors are only minor modifiers of the risk of PVH/IVH. Rare gene mutations related to coagulation (factor V Leiden, methylenetetrahydrofolate reductase, prothrombin), inflammation (interleukins 1beta and 6, tumor necrosis factor alpha), and vascular integrity (collagen type 4A1) were reviewed recently [61].

and rats, which are at a developmental stage comparable to 24– 26 week of gestation in humans, allows investigation of the cellular and molecular processes that contribute to brain damage after hemorrhage [63,64]. These studies show that several plasma proteins (thrombin, plasmin, et al.) are injurious to the immature brain [65,66]. Hemoglobin breakdown, immune system activation, and hypoxia prior to intravascular hypertension can exacerbate the damage [67–69]. Post-hemorrhagic hydrocephalus can be modeled by injection of blood into the ventricles of immature rats [70] or piglets [71]. Overexpression of vascular endothelial growth factor in glial fibrillary acidic proteinexpressing cells of mouse brain is associated with PVH prior to birth [72]. Genetic knockout of plasminogen activator inhibitor 1 in mice is associated with multifocal brain hemorrhage in the first week of life [73].

Animal models Using rabbits, dogs, cats, and sheep, the physiologic mechanisms of PVH/IVH have been elucidated. Some models involve preterm delivery, others involve manipulation of physiologic variables to create hemorrhage in periventricular tissue. These studies show that fluctuations in arterial and venous blood pressure, but not asphyxia alone, can cause brain hemorrhage [62]. Blood or collagenase injection into the brains of newborn mice

208

Treatment, future perspective, conclusions Improved understanding of the physiologic determinants of PVH/IVH has been used to develop better obstetrical and perinatal care for premature infants; however, PVH/IVH remains a major clinical problem [74]. Attempts to prevent IVH in preterm infants using phenobarbital, vitamin K, vitamin E,

Hemorrhagic Lesions Chapter 20

indomethacin, and ethamsylate have not been effective [32,45, 75]. Limited prenatal administration of glucocorticoids may be associated with lower risk for PVH, probably through accelerated maturation of lungs [32]. A retrospective analysis suggested that maintenance of platelet levels greater than 200 × 109 /l is associated with reduced incidence of IVH [76]. Careful study of available hemostatic agents has been recommended for IVH prevention [77]. Prospective clinical trials show that intraventricular blood clot lysis with proteolytic agents (e.g. streptokinase, urokinase, and tissue plasminogen activator) do not prevent post-IVH hydrocephalus [78,79]. For post-hemorrhagic ventricular enlargement in premature infants, the best time for intervention by shunting remains unclear [80]. In a neonatal rabbit model of IVH, supplementation with thyroxine improves the behavioral outcome and cerebral myelin production [81]. Prevention of brain hemorrhage through the avoidance of preterm birth (including predisposing factors such as amniotic infection) remains the major goal. Further understanding of the mechanisms of cellular brain damage after hemorrhage through study of animal models and human autopsy material is necessary to expand treatment options if PVH occurs.

References 1. Takashima S, Itoh M, Oka A (2009) A history of our understanding of cerebral vascular development and pathogenesis of perinatal brain damage over the past 30 years. Semin Pediatr Neurol 16:226– 36 2. Kuban KC, Gilles FH (1985) Human telencephalic angiogenesis. Ann Neurol 17:539–48 3. Rhee CJ, Fraser CD, 3rd, Kibler K et al. (2014) The ontogeny of cerebrovascular pressure autoregulation in premature infants. J Perinatol 34:926–31 4. Chang H, Cho KH, Hayashi S et al. (2014) Site- and stage-dependent differences in vascular density of the human fetal brain. Childs Nerv Syst 30:399–409 5. Ballabh P (2010) Intraventricular hemorrhage in premature infants: mechanism of disease. Pediatr Res 67:1–8 6. Marin-Padilla M (2012) The human brain intracerebral microvascular system: development and structure. Front Neuroanat 6:38 7. Bystron I, Blakemore C, Rakic P (2008) Development of the human cerebral cortex: Boulder Committee revisited. Nat Rev Neurosci 9:110–22 8. Malik S, Vinukonda G, Vose LR et al. (2013) Neurogenesis continues in the third trimester of pregnancy and is suppressed by premature birth. J Neurosci 33:411–23 9. Kinoshita Y, Okudera T, Tsuru E, Yokota A (2001) Volumetric analysis of the germinal matrix and lateral ventricles performed using MR images of postmortem fetuses. Am J Neuroradiol 22:382–8 10. Del Bigio MR (2011) Cell proliferation in human ganglionic eminence and suppression after prematurity-associated haemorrhage. Brain 134:1344–61 11. Ma T, Wang C, Wang L et al. (2013) Subcortical origins of human and monkey neocortical interneurons. Nat Neurosci 16:1588–97

12. Zecevic N, Chen Y, Filipovic R (2005) Contributions of cortical subventricular zone to the development of the human cerebral cortex. J Comp Neurol 491:109–22 13. Rakic P, Sidman RL (1969) Telencephalic origin of pulvinar neurons in the fetal human brain. Z Anat Entwicklungsgesch 129:53–82 14. Letinic K, Kostovic I (1997) Transient fetal structure, the gangliothalamic body, connects telencephalic germinal zone with all thalamic regions in the developing human brain. J Comp Neurol. 384(3):373–95. 15. Brazel CY, Romanko MJ, Rothstein RP, Levison SW (2003) Roles of the mammalian subventricular zone in brain development. Prog Neurobiol 69:49–69 16. Shuangshoti S, Netsky MG (1966) Histogenesis of choroid plexus in man. Am J Anat 118:283–316 17. De Noronha L, Medeiros F, Nones RB et al. (1999) Inj´uria hip´oxicoisquˆemica de padrao hemorr´agico em enc´efalos de neomortos do Hospital de Cl´ınicas de Curitiba. An´alise de 1028 casos de necr´opsia entre 1960 e 1995. Arq Neuropsiquiatr 57:950–8 18. Jammes JL, Gilles FH (1983) Telencephalic development: matrix volume and isocortex and allocortex surface areas. In: FH Gilles, A Leviton, EC Dooling, eds., The Developing Human Brain Growth and Epidemiologic Neuropathology, Boston, MA, John Wright PSG, pp. 87–93 19. Terplan KL (1967) Histopathologic brain changes in 1152 cases of the perinatal and early infancy period. Biol Neonat 11:348–66 20. Chen HJ, Wei KL, Zhou CL et al. (2013) Incidence of brain injuries in premature infants with gestational age ≤34 weeks in ten urban hospitals in China. World J Pediatr 9:17–24 21. Sirgiovanni I, Avignone S, Groppo M et al. (2014) Intracranial haemorrhage: an incidental finding at magnetic resonance imaging in a cohort of late preterm and term infants. Pediatr Radiol 44:289– 96 22. McCormick MC, Litt JS, Smith VC, Zupancic JA (2011) Prematurity: an overview and public health implications. Annu Rev Public Health 32:367–79 23. Berger R, Bender S, Sefkow S et al. (1997) Peri/intraventricular haemorrhage: a cranial ultrasound study on 5286 neonates. Eur J Obstet Gynecol Reprod Biol 75:191–203 24. Sheth RD (1998) Trends in incidence and severity of intraventricular hemorrhage. J Child Neurol 13:261–4 25. Groenendaal F, Termote J, van der Heide-Jalving M et al. (2010) Complications affecting preterm neonates from 1991 to 2006: what have we gained? Acta Paediatr 99:354–8 26. Brouwer AJ, Groenendaal F, Koopman C et al. (2010) Intracranial hemorrhage in full-term newborns: a hospital-based cohort study. Neuroradiology 52:567–76 27. Gradnitzer E, Urlesberger B, Maurer U et al. (2002) Hirnblutung beim reifen Neugeborenen: eine Analyse von 10 Jahren (1989– 1999). Wien Med Wochenschr 152:9–13 28. Dolfin T, Skidmore MB, Fong KW et al. (1983) Incidence, severity, and timing of subependymal and intraventricular hemorrhages in preterm infants born in a perinatal unit as detected by serial realtime ultrasound. Pediatrics 71:541–6 29. Paneth N, Pinto-Martin J, Gardiner J et al. (1993) Incidence and timing of germinal matrix/intraventricular hemorrhage in low birth weight infants. Am J Epidemiol 137:1167–76 30. Kim KR, Jung SW, Kim DW (2014) Risk factors associated with germinal matrix-intraventricular hemorrhage in preterm neonates. J Korean Neurosurg Soc 56:334–7

209

Developmental Neuropathology 31. Rong Z, Liu H, Xia S, Chang L (2012) Risk and protective factors of intraventricular hemorrhage in preterm babies in Wuhan, China. Childs Nerv Syst 28:2077–84 32. Shalak L, Perlman JM (2002) Hemorrhagic–ischemic cerebral injury in the preterm infant: current concepts. Clin Perinatol 29:745–63 33. Goldstein RF, Cotten CM, Shankaran S et al. (2013) Influence of gestational age on death and neurodevelopmental outcome in premature infants with severe intracranial hemorrhage. J Perinatol 33:25–32 34. Bolisetty S, Dhawan A, Abdel-Latif M et al. (2014) Intraventricular hemorrhage and neurodevelopmental outcomes in extreme preterm infants. Pediatrics 133:55–62 35. Fumagalli M, Bassi L, Sirgiovanni I et al. (2015) From germinal matrix to cerebellar haemorrhage. J Matern Fetal Neonatal Med 28 Suppl 1:2280–5 36. Swanson AE, Veldman A, Wallace EM, Malhotra A (2012) Subgaleal hemorrhage: risk factors and outcomes. Acta Obstet Gynecol Scand 91:260–3 37. Ment LR, Bada HS, Barnes P et al. (2002) Practice parameter: neuroimaging of the neonate: report of the Quality Standards Subcommittee of the American Academy of Neurology and the Practice Committee of the Child Neurology Society. Neurology 58:1726–38 38. Papile LA, Burstein J, Burstein R, Koffler H (1978) Incidence and evolution of subependymal and intraventricular hemorrhage: a study of infants with birth weights less than 1,500 gm. J Pediatr 92:529–34 39. Hesser U, Katz-Salamon M, Mortensson W et al. (1997) Diagnosis of intracranial lesions in very-low-birthweight infants by ultrasound: incidence and association with potential risk factors. Acta Paediatr Suppl 419:16–26 40. Levene MI, Fawer CL, Lamont RF (1982) Risk factors in the development of intraventricular haemorrhage in the preterm neonate. Arch Dis Child 57:410–7 41. Intrapiromkul J, Northington F, Huisman TA et al. (2013) Accuracy of head ultrasound for the detection of intracranial hemorrhage in preterm neonates: comparison with brain MRI and susceptibilityweighted imaging. J Neuroradiol 40:81–8 42. Parodi A, Morana G, Severino MS et al. (2015) Low-grade intraventricular hemorrhage: is ultrasound good enough? J Matern Fetal Neonatal Med 28 Suppl1:2261–4 43. Kutuk MS, Yikilmaz A, Ozgun MT et al. (2014) Prenatal diagnosis and postnatal outcome of fetal intracranial hemorrhage. Childs Nerv Syst 30:411–8 44. Leijser LM, de Bruine FT, Steggerda SJ et al. (2009) Brain imaging findings in very preterm infants throughout the neonatal period: part I. Incidences and evolution of lesions, comparison between ultrasound and MRI. Early Hum Dev 85:101–9 45. hankaran S (2003) Hemorrhagic lesions of the central nervous system. In: DK Stevenson, WE Benitz, P Sunshine, eds., Fetal and Neonatal Brain Injury, 3rd ed., Cambridge, UK, Cambridge University Press, pp. 175–88 46. Morales DM, Holubkov R, Inder TE et al. (2015) Cerebrospinal fluid levels of amyloid precursor protein are associated with ventricular size in post-hemorrhagic hydrocephalus of prematurity. PLoS One 10:e0115045 47. Leviton A, Gilles FH, Dooling EC (1983) The epidemiology of subarachnoid hemorrhages. In: Gilles FH, Leviton A, Dooling EC,

210

48. 49.

50.

51. 52.

53.

54. 55.

56. 57. 58.

59.

60. 61.

62. 63.

64.

65.

66.

67.

eds., The Developing Human Brain Growth and Epidemiologic Neuropathology, Boston, MA, John Wright Inc, pp. 217–26 Towbin A (1970) Central nervous system damage in the human fetus and newborn infant. Am J Dis Child 119:529–42 Kuban KC, Allred EN, Dammann O et al. (2001) Topography of cerebral white-matter disease of prematurity studied prospectively in 1607 very-low-birthweight infants. J Child Neurol 16:401–8 Larroche JC (1964) H´emorragies c´er´ebrales intra-ventriculaires chez le pr´ematur´e. Ie partie: anatomie at physiopathologie. Biol Neonat 91:26–56 Rorke LB (1982) Pathology of Perinatal Brain Injury, New York, NY, Raven Press Haines KM, Wang W, Pierson CR (2013) Cerebellar hemorrhagic injury in premature infants occurs during a vulnerable developmental period and is associated with wider neuropathology. Acta Neuropathologica communications 1:69 Lategan B, Chodirker BN, Del Bigio MR (2010) Fetal hydrocephalus caused by cryptic intraventricular hemorrhage. Brain Pathol 20:391–8 Leech RW, Kohnen P (1974) Subependymal and intraventricular hemorrhages in the newborn. Am J Pathol 77:465–75 Towbin A (1968) Cerebral intraventricular hemorrhage and subependymal matrix infarction in the fetus and premature newborn. Am J Pathol 52:121–40 Kapur RP, Shaw CM, Shepard TH (1991) Brain hemorrhages in cocaine-exposed human fetuses. Teratology 44:11–8 Ballabh P (2014) Pathogenesis and prevention of intraventricular hemorrhage. Clin Perinatol 41:47–67 Alderliesten T, Lemmers PM, Smarius JJ et al. (2013) Cerebral oxygenation, extraction, and autoregulation in very preterm infants who develop peri-intraventricular hemorrhage. J Pediatr 162:698– 704 Kusters CD, Chen ML, Follett PL, Dammann O (2009) “Intraventricular” hemorrhage and cystic periventricular leukomalacia in preterm infants: how are they related? J Child Neurol 24:1158–70 Wigglesworth JS, Pape KE (1980) Pathophysiology of intracranial haemorrhage in the newborn. J Perinat Med 8:119–33 Ment LR, Aden U, Lin A et al. (2014) Gene–environment interactions in severe intraventricular hemorrhage of preterm neonates. Pediatr Res 75:241–50 Balasubramaniam J, Del Bigio MR (2006) Animal models of germinal matrix hemorrhage. J Child Neurol 21:365–71 Lekic T, Manaenko A, Rolland W et al. (2012) Rodent neonatal germinal matrix hemorrhage mimics the human brain injury, neurological consequences, and post-hemorrhagic hydrocephalus. Exp Neurol 236:69–78 Xue M, Balasubramaniam J, Buist RJ et al. (2003) Periventricular/intraventricular hemorrhage in neonatal mouse cerebrum. J Neuropathol Exp Neurol 62:1154–65 Xue M, Balasubramaniam J, Parsons KA et al. (2005) Does thrombin play a role in the pathogenesis of brain damage after periventricular hemorrhage? Brain Pathol 15:241–9 Xue M, Del Bigio MR (2005) Injections of blood, thrombin, and plasminogen more severely damage neonatal mouse brain than mature mouse brain. Brain Pathol 15:273–80 Gram M, Sveinsdottir S, Ruscher K et al. (2013) Hemoglobin induces inflammation after preterm intraventricular hemorrhage by methemoglobin formation. J Neuroinflammation 10:100

Hemorrhagic Lesions Chapter 20 68. Pahlavan PS, Sutton W, Buist RJ, Del Bigio MR (2012) Multifocal haemorrhagic brain damage following hypoxia and blood pressure lability: case report and rat model. Neuropathol Appl Neurobiol 38:723–33 69. Xue M, Del Bigio MR (2005) Immune pre-activation exacerbates hemorrhagic brain injury in immature mouse brain. J Neuroimmunol 165:75–82 70. Cherian SS, Love S, Silver IA et al. (2003) Posthemorrhagic ventricular dilation in the neonate: development and characterization of a rat model. J Neuropathol Exp Neurol 62:292–303 71. Aquilina K, Hobbs C, Cherian S et al. (2007) A neonatal piglet model of intraventricular hemorrhage and posthemorrhagic ventricular dilation. J Neurosurg 107:126–36 72. Yang D, Baumann JM, Sun YY et al. (2013) Overexpression of vascular endothelial growth factor in the germinal matrix induces neurovascular proteases and intraventricular hemorrhage. Sci Transl Med 5:193ra90 73. Leroux P, Omouendze PL, Roy V et al. (2014) Age-dependent neonatal intracerebral hemorrhage in plasminogen activator inhibitor 1 knockout mice. J Neuropathol Exp Neurol 73:387–402 74. Whitelaw A (2012) Periventricular hemorrhage: a problem still today. Early Hum Dev 88:965–9 75. Whitelaw A, Odd D (2007) Postnatal phenobarbital for the prevention of intraventricular hemorrhage in preterm infants. Cochrane Database Syst Rev 4:CD001691

76. Coen RW (2013) Preventing germinal matrix layer rupture and intraventricular hemorrhage. Front Pediatr 1:22 77. Kuperman AA, Brenner B, Kenet G (2013) Intraventricular hemorrhage in preterm infants and coagulation–ambivalent perspectives? Thromb Res 131 Suppl 1:S35–S8 78. Whitelaw A, Evans D, Carter M et al. (2007) Randomized clinical trial of prevention of hydrocephalus after intraventricular hemorrhage in preterm infants: brain-washing versus tapping fluid. Pediatrics 119:e1071–8 79. Whitelaw A, Odd D (2007) Intraventricular streptokinase after intraventricular hemorrhage in newborn infants. Cochrane Database Syst Rev 4:CD000498 80. de Vries LS, Brouwer AJ, Groenendaal F (2013) Posthaemorrhagic ventricular dilatation: when should we intervene? Arch Dis Child Fetal Neonatal Ed 98:F284–5 81. Vose LR, Vinukonda G, Jo S et al. (2013) Treatment with thyroxine restores myelination and clinical recovery after intraventricular hemorrhage. J Neurosci 33:17232–46 82. Claris O, Besnier S, Lapillonne A et al. (1996) Incidence of ischemic–hemorrhagic cerebral lesions in premature infants of gestational age < or = 28 weeks: a prospective ultrasound study. Biol Neonate 70:29–34 83. Gleissner M, Jorch G, Avenarius S (2000) Risk factors for intraventricular hemorrhage in a birth cohort of 3721 premature infants. J Perinat Med 28:104–10

211

21

White Matter Lesions in the Perinatal Period Robin L. Haynes1 and Rebecca D. Folkerth2 1 2

Department of Pathology, Boston Children’s Hospital, Boston, MA, USA New York City Office of the Chief Medical Examiner, and New York University School of Medicine, New York, NY, USA

Introduction The major cerebral white matter lesion in the perinatal period is diffuse white matter injury (DWMI), encompassing the previously used terms periventricular leukomalacia (PVL), defined as focal white matter necrosis, and diffuse cerebral white matter gliosis (DWMG), with which PVL is virtually always associated. This chapter considers these two patterns as points along a spectrum of severity, rather than as separate pathologic entities. A distinct entity affecting the white matter is periventricular hemorrhagic infarction, which is pathogenetically secondary to massive germinal matrix hemorrhage (Chapter 20). DWMI and periventricular hemorrhagic infarction frequently accompany gray matter lesions comprising the “encephalopathy of prematurity” (Chapter 22).

Definition, synonyms and historical annotations DWMI is a lesion of immature cerebral white matter in the perinatal period. It is defined by the presence of DWMG, with or without focal necrosis (sometimes referred to as “cystic PVL”). Any hypothesis concerning the underlying cause(s) and pathogenesis of DWMI must account for both of these distinct but interrelated pathologic patterns. Historically, DWMI, especially including focal necrosis, has been assumed to be the primary neuropathologic substrate underlying the variable motor deficits comprising “cerebral palsy.” It is now recognized that the white matter lesions are only a contributing factor to a larger, more global injury pattern involving gray and white matter sites. This pattern, now termed the “encephalopathy of prematurity” involves multiple cell types, including oligodendrocytes of the white matter as well as neurons of the thalamus, basal ganglia, cerebral cortex, brainstem, and cerebellum [1]. In this chapter,

we focus primarily on the white matter lesions, while the gray matter injury is described in Chapter 22. In the mid-nineteenth century, Little described spastic diplegia in children, and noted that the majority were born prematurely and required resuscitation for asphyxia at birth [2]; hence, cerebral palsy became known as “Little disease.” The first neuropathologic description of this clinical syndrome is generally considered to be the mid-nineteenth century report of Virchow [3]. Around this same time, Parrot described foci of steatosis (i.e., foamy macrophage infiltration), infarction, and hemorrhage in the periventricular white matter, and suggested that the periventricular white matter was vulnerable because it was in the distal field of vascular perfusion; he stressed the relationship of the pathology to prematurity [4]. In 1962, Banker and Larroche published the classic neuropathologic description of what they termed “periventricular leukomalacia” [5]. These investigators were among the first to systematically examine the predisposing pre- and neonatal factors, particularly anoxia, and to characterize the sequential evolution of the necrotic lesions. Like Parrot, they emphasized focal necrotic periventricular white matter lesions as the hallmark of the disease, and suggested their relationship to vascular border zones. In the 1970s and 1980s, Gilles drew attention to the association between periventricular necrotic foci and “hypertrophic” or reactive astrocytes in the surrounding white matter [6]. He also described “acutely damaged glia” and amphophilic perivascular globules (putative markers of vascular injury resulting from serum protein extravasation, and occasionally undergoing mineralization) as additional key features of perinatal white matter damage [6,7]. With his colleagues, Gilles suggested that focal periventricular necrosis, hypertrophic astrocytes, acutely damaged glia, and globules are all manifestations of the same entity, which he termed “acquired perinatal telencephalic leukoencephalopathy”, because of the similarities in their risk-factor profiles. Additional descriptions emphasized lipid-laden cells in the

Developmental Neuropathology, Second Edition. Edited by Homa Adle-Biassette, Brian N. Harding, and Jeffrey A. Golden. © 2018 John Wiley & Sons Ltd. Published 2018 by John Wiley & Sons Ltd.

213

Developmental Neuropathology white matter, and the term “fatty metamorphosis” was applied [8]. Debate then centered upon whether the diffuse “fatty transformation” of glia in the periventricular white matter represented a normal phenomenon associated with myelination and increased lipid metabolism, or whether it represented a disease process, particularly a metabolic derangement of myelinating glia [9]. With the introduction of glial fibrillary acidic protein (GFAP) immunocytochemistry, the phenotype of these cells was established as astrocytic. In the twenty-first century, clinical attention has focused upon the ultrasonographic finding of focal periventricular echolucency (i.e., cysts) as the distinguishing feature of “cystic PVL”. It is now recognized that the large foci of necrosis which define cystic PVL are no longer the principle feature of white matter injury [10]. Rather, smaller or microscopic foci of necrosis are more common, seen in association with a widespread diffuse injury, referred to by the interchangeable terms perinatal telencephalic leukoencephalopathy or DWMG, use of which varies depending on the author. The variability from cystic to noncystic diffuse injury possibly represents a spectrum of pathology that changes with time [11,12]. This evolving pattern of disease likely reflects advances in the management of respiratory failure, hemodynamic instability, and/or infection in modern neonatal intensive care units. Despite major advances, the incidence of cerebral palsy in premature infants is increasing, presumably due to increased survival of extremely low birthweight infants [13]. This increasing prevalence of cerebral palsy along with a decreasing incidence of cystic PVL supports the importance of the noncystic component in the central white matter, as well as the gray matter injury of the encephalopathy of prematurity, in producing long-term disability.

Epidemiology Incidence and prevalence In the United States, approximately 55 000 infants are born very prematurely (weighing < 1500 g, i.e., “very low birth weight”) each year, and nearly 80% survive [14], reflecting the extraordinary advances in neonatal intensive care. Approximately 50% of very low birthweight infants show white matter injury of varying severity [15]. At the severe end of the spectrum of injury, the incidence of PVL with foci of cystic (necrotic) lesions, detectable by cranial ultrasound, has decreased to less than 3% [15]. This may be an underestimate, depending on the timing of imaging and whether sequential imaging is used [16]. A less severe form of necrotic PVL is detectable primarily at autopsy and is characterized by small microcystic lesions (35 years), obesity, tobacco smoking, diabetes, suboptimal antenatal care, poverty, maternal sleeping position, and amniotic infection [24]; most of these factors can compromise the placenta, leading to reduced oxygen and nutrient delivery to the fetus. Because these risk factors are persistent, pregnancies subsequent to stillbirth may also be associated with adverse neurological outcome [25]. Approximately 16% of all births are premature (G - p.(Asn409Ser); .c.1448T>C - p.(Leu483Pro); c.84dupG - p.(Leu29AlafsTer18), c.115+1G>A -p.(?)] are very common (90% and 60% of variants in Ashkenazi Jewish and non-Jewish patients respectively); p.(Asn409Ser) variant is associated with late onset nonneuronopathic forms of Gaucher disease [131]. Animal models and pathogenesis In 1982, Nilsson and Svennerholm reported an accumulation of glucosylsphingosine in the brain of infantile and juvenile cases

[132] that was later confirmed by other authors. Glucosylsphingosine does not accumulate in the brain in type 1, in which little pathology is present, but the highest accumulation of glucosylsphingosine was detected in fetal hydrops, followed by type 2 and then type 3 brains, paralleling the extent of pathology [114], suggesting that it may contribute to neuronal degeneration. Glucosylsphingosine levels are also elevated in visceral organs, particularly in the spleen. Inflammatory responses and factors released by Gaucher cells including proinflammatory cytokines and chemokines, as well as abnormal calcium pooling in the neural compartments in type 2 and 3 induced by glucosylsphingosine, have been suggested as a mechanistic link between lysosomal storage and the clinical manifestation of the disease [111]. Using targeted gene disruption, Tybulewicz et al. [133] first generated a knockout murine model of Gaucher disease in 1992. The homozygous mutant mice died within 24 hours of birth. Significant infiltration of Gaucher cells was not observed in organs, although tubular inclusions similar to those of human Gaucher disease were found in macrophages at the ultrastructural level. Skin abnormalities were similar to the severe type 2 Gaucher disease [133]. Knock-in mice generated by inserting known human mutant genes resulted in similar consequences [35,134]. In 2012, in a neuronopathic-form knockout mouse model of Gaucher disease, significant increase in brain mRNA levels of inflammatory mediators was observed, including interleukin1-β, tumor necrosis factor-α (TNF- α) and its receptor, transforming growth factor beta, chemokines CCL1, CCL2 and CCL3, and nitrotyrosine (a biomarker of oxidative damage), strongly suggesting a cytotoxic role for activated microglia leading to neuronal death [135]. Another mechanism by which proand anti-inflammatory molecules could be activated is by abnormal folding of mutant proteins in the endoplasmic reticulum which can trigger apoptotic and/or inflammatory pathways. Since the generation of the first knockout mouse model, several viable models have been developed for the study of neuronopathic forms, in particular the Nestin-flox/flox mouse in which glucocerebrosidase deficiency was restricted to neurons and astrocytes. Generation of mice with particular point mutations and using transplantation of hematopoietic stem cells from glucocerebrosidase-null mice have allowed for the study of Gaucher disease type1 [38]. Using a neuronopathic Gaucher disease mouse model, Xu et al. [136] reproduced human Gaucher disease with parkinsonism. These mice had widespread aggregates of β-amyloid, α-synuclein and β-amyloid precursor protein in the cortex, hippocampus, and substantia nigra. Moreover, β-amyloid precursor protein colocalized with mitochondrial markers, autophagy and lysosomal proteins, arguing for Gaucher disease as a risk factor for mitochondrial dysfunction resulting in the co-occurrence of the disease with parkinsonism. In a naturally occurring ovine model of acute neuronopathic Gaucher disease, Karageorgos and colleagues [137] found a very high increase in glucosylceramide and also in gangliosides GM1, 2 and 3, provoking an imbalance or a disruption of cell

331

Developmental Neuropathology membrane homeostasis that would be expected to impair neuronal function.

Treatment, future perspective and conclusions Enzyme replacement therapy Enzyme replacement therapy is the first line of recommended treatment for Gaucher disease. Currently, there are three different commercially available enzyme replacement therapies: imiglucerase (Cerezyme® , Genzyme-Sanofi) expressed in Chinese hamster ovary cells, taliglucerase alfa (Elelyso® , Pfizer) expressed in carrot cells, and velaglucerase alfa (VPRIV® , Shire) expressed in a continuous human cell line. Enzyme replacement therapy has significant effects on systemic manifestations, but the enzyme does not cross the blood–brain barrier and has no effect on Parkinsonian manifestations [138]. Substrate reduction therapy Miglustat (N-butyl-deoxynojirimycin; Zavesca® , Actelion), is an iminosugar that inhibits the ceramide glucosyltransferase and reduces the glucosylceramide overload. Miglustat was the first oral substrate reduction therapy approved for mild-to-moderate type 1 Gaucher disease in adults. Miglustat can cross the blood– brain barrier, but its effect on neurologic manifestations of Gaucher disease is not established [139]. Furthermore, tremors have been associated with miglustat treatment over and above those of the Parkinsonian manifestations of Gaucher disease. A ceramide analogue, eliglustat tartrate (Cerdelga® , GenzymeSanofi) is available for treating type 1 Gaucher disease. The short- and long-term benefits are comparable to those of enzyme replacement therapy. However, eliglustat does not cross the blood–brain barrier [140]. Pharmacologic chaperone therapy The molecules investigated as chaperones for Gaucher disease treatment belong to two main categories, carbohydrate mimetics, such as iminosugar, azasugar, or carbasugars, chemically of similar structure to the glycoside of glucosylceramide and noncarbohydrate compounds. Several molecules have been patented, however, two seem promising: isofagomine (AT2101, afegostat tartrate) and ambroxol. Isofagomine is an azasugar that binds to the enzyme active site and enables enzyme folding and trafficking. Preclinical studies showed its efficacy on both main common variants (1226A>G p.(Asn409Ser); c.1448T>C - p.(Leu483Pro) and its effects on CNS alterations by reducing the neuroinflammation and delaying neurological deterioration [141]. Ambroxol, a noncarbohydrate compound known as an expectorant, has a pH-dependent affinity for β-glucosidase leading to stabilization of the enzyme. A pilot study showed that, at a high oral dose, ambroxol crosses the blood–brain barrier and leads to neurological improvement in neuronopathic GD [142]. Gene therapy strategies for the treatment of Gaucher disease are under investigation [143].

332

Metachromatic leukodystrophy Definition, dynonyms and historical perspective Metachromatic leukodystrophy (MLD), also named Scholz disease (MIM 250100) is a lysosomal storage disease due to deficient activity of enzyme arylsulfatase A, leading to accumulation of sulfated glycolipids (sulfatides; cerebroside sulfates) in various tissues including the central and peripheral nervous systems. The first report of MLD was an adult type in 1910 by Alzheimer [144], who identified diffuse demyelination of the white matter in a 51-year-old woman, with intense metachromatic deposits in the spinal white matter on toluidine blue stain. In 1925, Scholz [145] reported juvenile cases as progressive leukodystrophy in three children from one family. Diagnosis of MLD in these cases was established later by the demonstration of a brown metachromasia using acidic cresyl violet stain by Hirsch and Peiffer [146]. An infantile form was reported by Greenfield [147] and Brain and Greenfield [148]. Jatzkewitz and Austin, independently, in 1958 [149] and 1959 [150], found an excessive accumulation of sulfatides in the tissues from a patient with MLD. In 1963, Austin et al. [151] identified arylsulfatase A as the deficient enzyme in MLD. Arylsulfatase A was later identified as the heat-labile component of sulfatide sulfatase. Sulfatide sulfatase and aryl sulfatase A are both deficient in patients with MLD. Mehl and Jatzkewitz [152] discovered the heat-stable factor that increases the activity of sulfatide sulfatase several-fold. This factor is now known as saposin B (or SAP-1). In rare patients, saposin B deficiency is the cause of MLD rather than aryl sulfatase A [153,154]. Epidemiology Since MLD is transmitted as a Mendelian autosomal-recessive disorder, both sexes are affected equally. Although it occurs panethnically, several ethnic groups have a higher incidence, in particular the Habbanite Jewish community in Israel, Arabs living in lower Galilee in Israel, Inuit, and Navajo Native Americans. The overall incidence varies considerably among countries, ranging from 1 in 40 000 in Sweden to 1 in 170 000 in Germany [154]. Saposin B deficiency appears to be the most frequent of the sphingolipid activator deficiencies. Genetics The ARSA gene has been mapped on chromosome 22q13. This gene spans approximately 5.4 kb, contains 8 exons and encodes for a protein of 509 amino acids; 212 pathogenic variants have been reported in the HGMD database. MLD is caused by mutations of the ARSA locus, except for the few cases with saposin B deficiency. Three variants are the most frequent: c.465+1G>A is linked to a severe phenotype, while c.1283C>T - p.(Pro428Leu) and c.542T>G - p.(Ile181Ser) variants have been identified in patients with a mild phenotype. Attention should be given to arylsulfatase A pseudodeficiency, since a high proportion (up to 15%) of individuals display a

Sphingolipidoses and Related Disorders Chapter 29

substantial in vitro reduction of arylsulfatase A activity without any clinical or biological alterations. Two variants, p.(Gly311Ser) and p.(Glu314Asp), have mainly been associated with the socalled pseudodeficiency. Of note, some patients presented with a pseudodeficiency allele and a pathogenic allele.

Clinical features Signs and symptoms The initial signs of MLD may appear at any age. According to the recent review by Fluharty [155], MLD is divided into three subtypes depending on the age of onset: late infantile, juvenile, and an adult form, representing 50–60%, 20–30%, and 15–20% of the cases, respectively. In the late-infantile form, clinical symptoms are usually noticeable between one and two years of age. Typical signs include weakness, hypotonia, and gait disturbance. The other key signs appear later and include pyramidal signs, increased and painful muscle tone, gradual mental regression, speech deterioration, generalized or partial seizures, and peripheral neuropathy. Deafness and optic atrophy follow and the child becomes totally blind with a terminal decerebrate posture. In juvenile cases, onset usually ranges from 4 to 14 years. Gradual decline in school performance with emergence of behavioral disturbance, slurred speech, and clumsiness are common presenting signs and symptoms. Within one year, the child becomes unable to walk, with spastic paresis and cerebellar ataxia. Seizures may develop. This group may be subdivided into early-onset (4–6 years) and late-onset (6–16 years) groups. The former group tends to resemble late-infantile, presenting with early motor dysfunction. In the latter, behavioral abnormalities and poor school performances are the initial symptoms, and motor disturbance follows. Progression of the disease is similar to the late-infantile form, although slower. Adult MLD has been reported at any age after sexual maturity, and sometimes not until the fifth decade. Two distinct types are identified in the adult form. Signs of motor system involvement such as pyramidal, cerebellar signs and dystonia, and peripheral neuropathy characterize the first type. Some patients may just present with peripheral neuropathy [156]. In the large majority of patients, motor nerve conduction velocity is decreased and sensory nerve action potentials have diminished amplitude with prolonged peak latency. The other type initially demonstrates personality changes, with emotional disorder and bizarre behavior and gradually develops progressive mental deterioration. Because of behavioral abnormalities, this second type of patient may be confused with having schizophrenia [157]. Two other allelic forms have been recognized, namely partial cerebroside sulfate deficiency and pseudoarylsulfatase A. Two nonallelic forms have also been described: metachromatic leukodystrophy due to saposin B deficiency (see Sphingolipid activator protein deficiency section) and multiple sulfatase deficiency (see Niemann–Pick disease section).

Neuroimaging MRI is a very useful tool in establishing the diagnosis since it reveals bilateral symmetric abnormal T2-hyperintensities starting in the corpus callosum and extending into the periventricular areas. In the late-infantile form, the lesions are first observed in the splenium and parieto-occipital areas, while, in late adult forms, the rostrum and the frontal white matter are initially affected. A typical, although nonspecific morphology (also observed in Krabbe disease and infantile GM1 gangliosidosis,) is the “tigroid pattern” of radiating normal myelin strips related to the relative preservation of myelin in the perivascular areas. The subcortical U fibers are also generally spared. In severe forms, projection fibers are involved, whereas thalami, basal ganglia and cerebellar white matter display decreased intensity on T2weighted images. Proton MRS in patients with late-infantile and juvenile forms reveals marked reduction of N-acetylaspartate in both gray and white matter and high myoinositol levels, attributed to neuron loss and reactive gliosis, respectively. Increased lactate/creatinine ration may account for oligodendrocyte injury [for review see 158].

Pathology, histochemistry and ultrastructural findings Grossly, the brain shows variable atrophy. Atrophy of optic nerve, brainstem and cerebellum is marked in the late-infantile type. The white matter is firm to touch and may have a dull chalky white appearance with a clear demarcation from the cortex. The demyelinated cerebral white matter shows slight gray discoloration with relative preservation of myelin in subcortical U-fibers (Figure 29.7a). These changes are marked in lateinfantile type and mild in the juvenile form. In adult MLD, the gross appearance of the brain is mostly normal. Viscera show no gross abnormalities, except for the gall bladder that may be small and fibrotic or may exhibit papillomatous changes [156]. Histologically, MLD is characterized by demyelination of the central (Figure 29.7) and peripheral nervous systems and the presence of numerous macrophages that contain eosinophilic and PAS-positive globular deposits on paraffin sections (Figure 29.8a,b). The demyelination in the cerebral white matter typically excludes subcortical U-fibers (Figure 29.7b). Unlike in Krabbe disease, macrophages in MLD are diffusely scattered and not clustered around vessels (Figure 29.8a). On frozen sections, the PAS-positive deposits show pink metachromasia with toluidine blue (Figure 29.8b) and characteristic brown metachromasia with acidic cresyl violet stain (Figure 29.8c). Metachromatic deposits are seen in the nervous system as well as visceral organs (Figure 29.8d). Chemical analysis has shown cholesterol, phospholipids, and galactolipids (the majority of which are sulfatides) with the molar ratio of 1 : 1 : 1 as the constituents of these metachromatic deposits [35]. Electron microscopic study reveals various types of inclusions in macrophages, glial cells, and neurons (Figure 29.9a). Some

333

Developmental Neuropathology

(a)

(b)

(c)

Figure 29.7 White matter changes in metachromatic leukodystrophy (MLD). (a) Coronal section of the brain of an infantile patient with MLD, showing diffusely gray discolored cerebral white matter indicating demyelination. The subcortical U-fibers are relatively well preserved and remain white in color. (b) Diffuse white matter demyelination with preserved subcortical U-fibers is well demonstrated. Arrows indicate the location of U fibers. Myelin stain. (c) On higher magnification, many fragmented myelin sheaths are recognized in the demyelinated white matter. Myelin stain.

334

inclusions show characteristic herringbone and honeycomb patterns (Figures 29.9b,c). These MLD inclusions can be reproduced in vivo and in vitro by overloading sulfatides. In adult type MLD, inclusions frequently form composites with lipofuscin. With progression of the demyelination, the number of oligodendrocytes diminishes and reactive astrocytes increase. A variable degree of axonal degeneration is apparent in demyelinated areas. Neuronal accumulation of metachromatic materials has been described in the dentate nucleus of the cerebellum, in globus pallidus, thalamus, cranial nerve nuclei of the brainstem and spinal gray matter (Figure 29.8e). The neuronal inclusions in the spinal anterior horn cells show brown metachromasia (Figure 29.8f) but their ultrastructural morphology closely resembles membranous cytoplasmic bodies, as in gangliosidosis (Figure 29.9d). The cerebral cortical neurons and cerebellar Purkinje cells contain minimal metachromatic material in their perikarya, although focal swelling of dendrites and axonal torpedo formation are often recognized. Large pyramidal neurons such as Betz cells may contain some metachromatic material. This absence of the cortical neuronal storage in MLD contrasts with cortical neuronal storage in multiple sulfatase deficiency (MSD). Retinal ganglion cells also accumulate metachromatic materials [35,154,160,161). Peripheral nerve analysis reveals reduction of myelin sheath thickness in all types of MLD with segmental demyelination. Metachromatic deposits are seen in Schwann cells and endoneural macrophages. Metachromatic deposits can also be demonstrated in liver, gallbladder, islets of Langerhans in the pancreas, lymph nodes, kidney, adrenal, and ovary. In the kidney, metachromatic deposits are present in the thin limb of the loop of Henle, the distal convoluted tubules, and collecting tubules (Figure 29.8d).

Biochemistry Storage compound The major accumulated compounds in MLD are sulfatides or 3-O-sulfogalactosyl ceramides. Sulfatides are abundant in the brain and are major components of myelin. These compounds are also found in extraneural tissues such as gall bladder, respiratory and gastric mucosa [162]. Due to pseudodeficiency, urinary sulfatide assessment is a convenient tool for diagnosing patients with MLD. Saposin B is a coactivator of the alpha galactosidase A and arylsulfatase and its deficiency combines MLD and Fabry storage patterns, thus these patients excrete not only sulfatides, but also globotriaosylceramide. Enzyme deficiency The primary enzyme defect in MLD is arylsulfatase A (arylsulfatase A-sulfatidase or cerebroside 3-sulfatase). This enzyme requires a coactivator, saposin B, to be active. The assessment of enzyme activity is informative when the residual activity is absent or very low. As discussed above, urinary sulfatide assessment is useful in such a situation.

Sphingolipidoses and Related Disorders Chapter 29

(a)

(b)

(c)

(d)

(e)

(f)

Figure 29.8 Histological features in metachromatic leukodystrophy (MLD). (a) Scattered astrocytes and macrophages in the white matter containing deeply eosinophilic material in their cytoplasm. Hematoxylin and eosin. (b) Storage material in glial cells and macrophages stain metachromatically pink with toluidine blue. (c) Cerebral white matter showing characteristic brown metachromasia in the

demyelinating white matter. Acidic cresyl violet. (d) Renal tubules contain characteristic storage material with brown metachromasia. Acidic cresyl violet. (e) Swollen storage neurons in CN III nucleus. (f) Neuronal storage material demonstrates brown metachromasia in the spinal anterior horn. Acidic cresyl violet.

Pathogenesis and animal models Sulfatide component accumulation in the nervous system and viscera, which are responsible for the cellular metachromasia, is highly toxic. Suggestive mechanisms include oligodendroglial dysfunction due to excessive storage of sulfated lipids, unstable and easily broken-down myelin due to excessive sulfatide contents and lysosulfatide inhibition of the activity of protein kinase C, cytochrome oxidase C.

There is no naturally occurring animal model of MLD. With targeted disruption of aryl sulfatase A (Arsa) gene, Arsa knockout mice have been generated. These mice are unable to degrade sulfatides and lipids are stored in the visceral organs and within glia as well as some neurons in the brain. However, these mice have a normal life span and do not develop widespread demyelination, unlike human MLD [163,164]. Apart from this model, no other animal models have been useful to gain knowledge of

335

Developmental Neuropathology

(a)

(b)

(c)

(d)

Figure 29.9 Ultrastructural features in metachromatic leukodystrophy (MLD). (a) Electron micrograph of pleomorphic inclusions in a glial cell. (b) Electron micrograph of an inclusion showing the typical herringbone pattern. (c) Electron micrograph of an inclusion showing a honeycomb or cobblestone appearance. (d) Electron micrograph of neuronal inclusions (storage material), closely resembling membranous cytoplasmic bodies of gangliosidosis.

the pathophysiology, but have been developed to explore treatment strategies of MLD, and have been recently reviewed by Van Rappard et al. [158] and Zigdon et al. [38].

Differential diagnosis At the initial stages of the disease, MLD can mimic Guillain– Barre syndrome or chronic inflammatory demyelinating polyneuropathy. Thereafter, MLD must be suspected in patients with progressive neurological signs and demyelination on imaging. Other leukodystrophies, such as adrenoleukodystrophy, Krabbe disease, as well as mitochondrial disease and vanishing white matter disease, may be considered. But the demonstration of characteristic brown metachromasia and/or unique ultrastructural features of storage inclusion in Schwann cells on skin biopsy, for example, provides a good morphological basis for diagnosis. Furthermore, low arylsulfatase A activities may be observed in asymptomatic patients with pseudodeficiency alleles (10–15% of the total population). In contrast, normal arylsulfatase A activities may be observed in nonallelic MLD, such as sphingolipid activator protein deficiency [155]. These diagnostic difficulties warrant systematic molecular analysis.

336

Treatment, future perspective and conclusions There is currently no specific therapy for MLD. Hematopoietic stem cell transplantation improves the outcome when it is performed in presymptomatic individuals or at the very beginning of the clinical course [165]. A gene therapy clinical trial using an adeno-associated virus vector is in progress at the time of writing [166]. This open-label, single arm, monocentric, phase I/II clinical study included five patients with early onset form of MLD. Enzyme replacement therapy using infusion has been investigated in late-onset forms [167] and intrathecally in the early-onset form [168]. Another therapeutic approach has been explored; since vitamin K has an essential role in the biosynthesis of sulfatides and other sphingolipids in the brain, the administration of warfarin, a vitamin K antagonist, may reduce brain sulfatide concentration. However, this drug had no beneficial effects in four advanced cases of MLD [169].

Multiple sulfatase deficiency Definition, synonyms and historical perspective MSD (MIM 272200), also named Austin disease [35,154,161,170,171] or mucosulfatidosis is a rare inherited autosomal-recessive disorder caused by homozygous or

Sphingolipidoses and Related Disorders Chapter 29

compound heterozygous mutations in the SUMF1 gene, which encodes an enzyme, formylglycine-generating enzyme, involved in the post-translational modification of sulfatases resulting in a combination of features found in other diseases due to deficiencies of the individual sulfatases [172,173]. The first case of multiple sulfatase deficiency was reported by James Austin as an atypical form of metachromatic leukodystrophy (MLD) in two siblings who had clinical phenotypes of combined late infantile MLD and skeletal disorders suggestive of mucopolysaccharidoses [170]. In these patients, the activities of arylsulfatases A, B and C were deficient and, in addition to sulfatides that accumulate in MLD, mucopolysaccharide sulfates, steroid sulfates, and gangliosides accumulated in the various tissues. MSD is now defined as a single genetic entity resulting from deficient cellular sulfatases, explaining the clinical heterogeneity.

Epidemiology MSD is an extremely rare disorder with a prevalence of about 1 in 1.4 million births. Around 50 patients have been reported to date [172,173]. Clinical features Signs and symptoms According to the classification of Eto and colleagues [174], which takes into account age of onset and extent of the disease severity, MSD is subdivided into several forms including neonatal, late infantile, either with severe or mild presentation (0– 2 years) and juvenile variants (2–4 years). The infantile form is the most severe and represents the classical example by exhibiting clinical features of combined MLD with mild features of mucopolysaccharidosis, such as coarse facies, skeletal abnormalities, heart involvement and hepatosplenomegaly, and even visceromegaly. Additional features sometimes observed consist of deafness, corneal clouding, ocular albinism, hypertrichosis, and ichthyosis, all these signs are not necessarily present from birth [175]. A variant with severe bony abnormalities and mild degree of mental restriction has been reported from Saudi Arabia [154, 161]. Neurological signs are common to all forms, and particularly include particularly hypotonia, psychomotor restriction or regression. Neuroimaging. On MRI, several abnormalities may be observed, such as bilateral symmetric white matter hyperintensities generally sparing the subcortical U-fibers [176]. Microcephaly resulting from brain and cerebellar atrophy, together with subcortical white matter hyperintensities, enlargement of sulci and subdural spaces, corpus callosum hypoplasia with dysmorphic hippocampi have also been reported [175,177]. In two juvenile cases, MRI showed diffuse cortical–subcortical atrophy, pontocerebellar atrophy, dilated lateral ventricles with an enlarged cisterna magna and high intensity signal in the periventricular white matter [178].

Biochemistry Storage compound Sulfatides, sulfated glycosaminoglycans, sphingolipids and steroid sulfates accumulate as a result of the deficient activity of several sulfatases. Enzyme deficiency Formylglycine-generating enzyme oxidizes a cysteine residue to generate a formylglycine. The targeted cysteine is located within a consensus CysXProXArg sequence common to all sulfatases. This cysteine oxidation is mandatory for sulfatase activity. The diagnosis is made by assessing the enzyme activity of several sulfatases in leukocytes or cultured fibroblasts: arylsulfatase A (metachromatic leukodystrophy), arylsulfatase B (MPS type VI), iduronate sulfatase (MPS type II), galactose 6 sulfatase (MPS type IVA), heparin sulfatase (MPS type IIIA), Nacetylglucosamine-6-sulfatase (MPS type IIID).

Differential diagnosis In relation to white matter abnormalities, MLD is a consideration. Regarding other clinical features, MPS II, IIIA, IVA and VI, nonrhizomelic chondrodysplasia punctata and X-linked ichthyosis are the major diseases to be differentiated from MSD [166,170].

Pathology Macroscopy Gross pathology displays overlapping feature of classical MLD and MPS. A fibrous thickening of the leptomeninges combined with cerebral atrophy results in hydrocephalus and features similar to mucopolysaccharidosis [35] are conspicuous. On cut sections, the brain may be severely atrophic with thin cerebral cortex and firm and gray discolored white matter, a feature similar to MLD. In some cases, mucopolysaccharidosis-like features may be more prominent [179]. Histopathology, histochemistry and ultrastructural findings White matter degeneration with an accumulation of metachromatic materials is superimposed on neuronal storage, which is particularly evident in the deeper layers of the cortex. Metachromatic deposits are observed in visceral organs as seen in MLD cases. In addition, Alder–Reilly granules, a typical feature of mucopolysaccharidosis, are detected in the bone marrow and peripheral blood leukocytes [154]. The predominant feature of a child reported by Macaulay et al. [179] was that of mucopolysaccharidosis with conspicuous dilatation of perivascular spaces, very limited demyelination, and brown metachromasia. Electron microscopy of skin fibroblasts reveals the presence of pleomorphic lysosomal inclusions in Schwann cells, and this may be used as a simple morphological diagnostic tool [179].

337

Developmental Neuropathology

Genetics The gene mutated in MSD, SUFM1, was identified by Cosma et al. [180] and Dierk et al. [181] in 2003. This gene maps to chromosome 3p26, spans 106 kb, contains 9 exons and encodes for a glycoprotein of 374 amino acids, the formylglycine-generating enzyme. Fifty-one pathogenic variants have been reported on the HGMD database. So far, most were missense mutations that affect stability and residual enzyme activity. Genotype– phenotype correlations may roughly predict the prognosis. Missense and non-sense mutations affecting key functional and structural residues are predicted to cause severe forms, whereas missense mutations not suppressing the functional conformation of the protein lead to less severe forms [175]. Pathogenesis and animal models Seventeen sulfatases are encoded by the human genome, eight of which are associated with diseases resulting from single sulfatase deficiency. Sulfatases are necessary for nonredundant sulfatation of glycosaminoglycans and sulfatides in lysosomes and steroid sulfates in the endoplasmic reticulum, thus playing a role in multiple organs and in the central nervous system. The phenotype depends on the residual amount of formylglycinegenerating enzyme activity and stability [176], and patients with drastic reduction of functional enzyme represent the most severe forms. Three mouse models have been generated to study the neurodegenerative process. A mouse line carrying a null mutation was generated more than 10 years ago. These animals displayed early lethality or congenital growth restriction with cell vacuolization and lysosomal storage of glycosaminoglycans. They also displayed activated microglia in layers II–IV of the cerebral cortex and a strong increase in inflammatory cytokines and apoptosis [183]. The authors then generated conditioned mouse models with Sufm1 deletion, either in both neurons and astrocytes or in astrocytes only [184], and demonstrated that lysosomal storage in astrocytes alone was able to cause degeneration in the mouse cortex and to elicit neuroinflammation (chemokines Mip1-α and -β and the cytokine TNF-α), a nonspecific mechanism involved in a variety of neurodegenerative diseases. Furthermore, they performed a panel of behavioral tests whose findings recapitulate those of human disease (i.e., motor and behavioral impairment consisting of early anxiety later followed by hyperactivity). Treatment, future perspective and conclusions Whatever the initial presentation, the prognosis remains poor. The neonatal form is usually fatal within one year of life. Gene therapy A 2011 study investigated the effects of combined systemic and cerebral intraventricular administration of an rAAV9 vector encoding SUMF1 gene to MSD mice. This combined treatment resulted in a significant activation of sulfatases with subsequent clearance of sulfated compound storage [185].

338

MicroRNA A novel microRNA (miR-95) acting as a regulator of sulfate metabolism has been identified. This miRNA suppresses sulfatase activity. Targeting miR-95 in cells from patients with MSD led to increased residual formylglycine-generating enzyme and enhanced sulfatase activity [186]. Understanding regulatory mechanisms may pave the way to future targeted approaches.

Globoid-cell leukodystrophy (Krabbe disease) Definition, synonyms and historical perspective Krabbe disease (MIM 245200) is an autosomal-recessive, rapidly progressive neurodegenerative disease affecting young infants in more than 85% of the cases; later onset cases including adult cases, represent 10–15% of patients [187]. This disease is also known as galactosyl cerebrosidase deficiency, galactoceramide beta-galactosidase deficiency, globoid-cell leukodystrophy or leukoencephalopathy, owing to the presence of unique globoid cells in the demyelinating lesions within the white matter. Krabbe disease is due to deleterious variants in the galactocerebrosidase gene that encodes a key enzyme for central and peripheral system myelination. This condition was first described by Krabbe in 1916, as an “acute infantile familial diffuse sclerosis of the brain” [188]. It was only in 1970 that Suzuki and Suzuki identified the deficiency of beta-galactocerebrosidase as the cause of human and canine globoid-cell leukodystrophy [189]. In 1993, the complementary DNA was cloned and, in 1994, the disease-causing gene located on 14q13 encoding galactocerebrosidase was characterized by two different teams [190–192]. Epidemiology Globoid-cell leukodystrophy is a pan-ethnic disease. Although clinical presentation is mainly infantile, the incidence of adultonset cases is probably underestimated. The overall incidence is approximately 1 in 100 000–200 000 births in the United States and Europe. The carrier frequency in cases without family history is 1 in 150. No cases have been identified in individuals of Jewish ancestry. Conversely, a high incidence of infantile Krabbe disease has also been reported in the Druse, with a carrier frequency of one in six. The incidence of the infantile type appears to be higher in Scandinavian countries, whereas late-onset cases appear to be more common in southern Europe, especially in Italy and Sicily. No late-onset cases have been reported in Scandinavian countries [187]. Clinical features Signs and symptoms According to age of onset, Krabbe disease has been divided into two main groups: infantile and later onset cases. The infantile form evolves in three stages. In stage I, infants present during the first months of life with extreme hyperirritability, episodic fever, and hyperesthesia to auditory, tactile and visual stimuli.

Sphingolipidoses and Related Disorders Chapter 29

They also have feeding difficulties and vomiting. Seizures and peripheral neuropathy occur in some of them. Stage II is characterized by rapid and severe psychomotor development regression and deterioration, hypertonicity, occasionally seizures and optic atrophy. At the terminal stage III, infants become decerebrate and blind, death occurring within two years. Later-onset cases have usually been clinically divided according to age of onset into late-infantile (6 months to 3 years), juvenile (3–15 years), and adolescent–adult (10–35 years). Some patients may develop neurological symptoms after the age of 40 years. The first signs in the late-infantile form resemble the classical infantile form. In the juvenile group, children develop vision loss, hemiparesis, gait disturbances (spastic paraparesis or ataxia), with or without peripheral neuropathy. These patients’ clinical course is generally protracted with initial rapid deterioration and then more gradual progression lasting for years. Adolescent and adult patients may have peripheral neuropathy with burning paresthesia and weakness without intellectual deterioration as initial features. More often, they show clinical features resembling the hereditary spastic paraplegias. Importantly, the onset of symptoms may be variable among siblings, particularly in adult onset forms, despite the same genotype [187].

Neuroimaging In typical infantile cases, computed tomography can be initially normal while diffuse cerebral atrophy with white matter hypodensity develops later. Discrete hyperdense areas may be found in the deep gray matter. MRI abnormalities were reviewed in 2014 in a large cohort of 64 patients [193]. In the infantile form, the most common abnormalities consisted of increased T2 intensity in the periventricular cerebral white matter and centrum semiovale that correlated with areas of demyelination and globoid cell accumulation, together with involvement of the dentate hilum and cerebellar white matter (Figure 29.10a). Brain atrophy and tigroid appearance of the white matter were observed in a minority of cases, but involvement of the motor cortex, posterior corpus callosum, and posterior arm of the internal capsule were found in 25% of the cases. Corticospinal tract abnormalities were not observed. In late-infantile-onset cases, abnormal T2 intensities were noted in all patients and in the majority of them, decreased intensity in the thalami. Involvement of the corticospinal tract was observed in 50% of the cases. Early involvement of the cerebellum correlated with shorter survival. In later onset cases, the cerebellum was never affected, and T2-weighted images revealed more localized demyelinated areas in the posterior deep white matter, and sometimes restricted to the corticospinal tracts. Proton MRS shows increased choline and myoinositol in the affected white matter [194]. Motor nerve conduction velocity is consistently low, although electromyography and nerve conduction studies have been reported normal in an adult patient. CSF protein is almost invariably elevated in infantile cases, but inconsistently in late onset forms [187].

Biochemistry Storage compound Galactosylceramidase deficiency impairs the hydrolysis of galactosphingolipid including β-galactocerebroside, a myelin component and galactosphingosine (also named psychosine), a cytotoxic component that underlies cell death and demyelination in Krabbe disease. Intriguingly, galactosylceramidase deficiency results mainly in the accumulation of psychosine in the macrophages (globoid cells) and neural cells, especially in oligodendrocytes and there is no significant storage of galactoceramide which may be due to the rapid loss of oligodendrocytes [195]. Enzyme deficiency Galactosylceramidase activity is deficient in Krabbe disease. The residual activity is low (less than 10% of the mean of the reference range) or absent. However, a higher residual activity ( C. In the juvenile variant, Arg447His in one allele and a splice site junction mutation (IVS5-1G > C) or non-sense mutation (Arg208X) on the other allele have been identified [10]. Pathogenesis The cause of extensive neuronal loss in CLN2 subjects is unknown. The contribution of massive accumulation of SCMAS to neuronal death needs further studies. Imbalance and/or lysosomal accumulation of undigested neuropeptides, TPP I substrates in vitro, in triggering neurodegeneration was proposed in 2002 [23]. Animal models A naturally occurring animal model has been described in longhaired dachshunds, and a knock-out mouse model exists [8]. Future directions and therapy Only supportive treatment is available. Bone marrow transplantation was ineffective in preventing clinical deterioration [14]. Gene therapy is in the preclinical phase.

CLN3: Juvenile NCL with mutations in the CLN3 gene Definition The juvenile form of NCL with mutations in the CLN3 gene (MIM 204200) is a slowly progressive neurodegenerative disease caused by mutations in the CLN3 gene encoding a multitransmembrane lysosomal protein of unknown function. There is little clinical variability, and only a few patients with a more benign, delayed, or protracted course have been reported. Fingerprint profiles represent the characteristic component of the storage material on ultrastructure. Synonyms and historical annotations CLN3 was traditionally called juvenile NCL, Batten disease, or Spielmeyer–Vogt–Sj¨ogren disease, after the authors of the first descriptions of this entity. The first NCL-recorded patients apparently had the CLN3 form [24]. Epidemiology CLN3 occurs worldwide, and other than CLN2, is the most common form of NCL. The highest incidence, 7.0 per 100,000 live births, was reported in Iceland [10].

376

Clinical features Signs and symptoms Progressive visual loss at the age of 4 to 7 years represents the first and, for the next few years, the only clinical sign of the disease. Children become blind within 2 to 10 years. Fundoscopy shows macular and retinal degeneration with pigment aggregations and optic nerve atrophy. Speech disturbances, a slow decline in cognitive functions, and epilepsy follow. Behavioral problems, Parkinsonism, myoclonic jerks, and sleep disturbances complete the clinical picture. EEG shows disorganization and spikeand-slow-wave complexes. ERG and visual evoked potentials are abnormal early in the disease process. In the protracted form, visual loss may be the only clinical sign for decades, and neurological manifestations other than blindness (cognitive/motor dysfunction, epilepsy) may appear after the age of 40 years [25]. Structural imaging CT and MRI reveal cerebral and, to a lesser degree, cerebellar atrophy in the later stages of the disease (usually after the age of 15 years). PET demonstrates decreased glucose utilization, with the earliest changes in the calcarine area. Laboratory findings Cytoplasmic vacuoles are present in a large proportion of lymphocytes (Figure 30.2 E), which can be easily demonstrated on blood films and are not encountered in other NCLs.

Macroscopy The brain weighs 800 to 1 000 g (Figure 30.1 C) and shows moderate, generalized atrophy also involving the cerebellar hemispheres. On coronal sections, the ventricular system is enlarged. The cortical ribbon is slightly thinned and brownish. The white matter is unchanged. Substantia nigra may be pale. Histopathology Mild to moderate neuronal loss with gliosis and accumulation of autofluorescent, Luxol fast blue, PAS- and SBB-positive material, that is strongly immunopositive for SCMAS are typical pathological features of CLN3. In the neocortex, neuronal loss is found mostly in cortical layers II and V. Neurons with proximal axonal segments distended by the storage granules (meganeurites) are seen most often in cortical layer III. Neuronal loss may be prominent in the CA2 hippocampal sectors, whereas pyramidal neurons in the CA1 sector do not appear reduced in number. A moderate decrease in neurons is visible in the corpus striatum, the amygdala, and the substantia nigra. The cerebellar cortex may show moderate to severe reduction of both granule cells and Purkinje cells. Proliferation of astrocytes is most prominent in deep cortical layers, accompanied by small clusters of activated microglia. The white matter shows no distinct abnormalities. Storage material is most abundant in neuronal cytoplasm, although it also accumulates in other cell types in the brain. The storage material in visceral organs is less often

The Neuronal Ceroid Lipofuscinoses Chapter 30

immunopositive for SCMAS than in CLN2 patients [26] and is also weakly immunoreactive for SAPs and amyloid-β. The retina shows severe atrophy with gliosis and migration of retinal pigmentary epithelial cells into the atrophic retina.

Ultrastructural findings FP, lysosomal inclusions typically observed in CLN3 tissues, are lamellar structures visualized as paired parallel lines with a central lucent line. The same storage body may contain fingerprint profiles (Figure 30.3 B) and curvilinear/rectilinear complexes, especially in the secretory sweat gland epithelial cells, smooth muscle cells, and endothelial cells. Fingerprint profiles may form solid inclusions, or they may be embedded in large, electronlucent vacuoles. For diagnostic purposes, Ficoll-isolated blood lymphocytes and skin, conjunctiva, and rectal biopsies are useful. Lysosomal inclusions in chorionic villi and aborted CLN3 fetuses display membranous, not fingerprint profiles [27]. Biochemistry CLN3 is a lysosomal glycoprotein of still unknown function with multi-transmembrane topology. SCMAS constitutes around 20% of the protein content of the brain storage; however, the cause of its accumulation is unclear, given the increased activity of numerous lysosomal enzymes, including TPP I and cathepsin D, which are potentially implicated in lysosomal degradation of SCMAS. Other identified components of the storage include dolichyl pyrophosphoryl oligosaccharides, SAPs, beta amyloid and amyloid precursor protein. Diagnosis A typical clinical presentation, the presence of vacuolated lymphocytes, and fingerprint profiles strongly suggest a diagnosis of CLN3. DNA testing is necessary to confirm the diagnosis, because biochemical assays are not available. Retinitis pigmentosa, hyperornithinemia, and peroxisomal, mitochondrial and lysosomal disorders associated with retinopathy should be excluded. Genetics The CLN3 gene is located on chromosome 16p11.2-12.1 (MIM#607042) and encodes a protein of 438-amino acid residues. Up to autumn of 2017, 67 different disease-associated mutations have been described [1]. The most common is a 1-kb deletion accounting for around 85% of mutations. Patients heterozygous for this mutation and either Leu170Pro or Glu295Lys have a delayed or protracted clinical course [27]. Pathogenesis The pathogenesis is unknown. It was proposed that CLN3 protein exerts a neurotrophic and anti-apoptotic action [28] and may be involved in regulation of lysosomal pH in mammalian cells [29]; thus, lack of these activities in the mutated protein could explain both the storage and neuronal death.

Animal models Four different mouse models have been generated by using either knock-out or knock-in strategies [8]. All these animals recapitulate numerous clinicopathologic features of the human disease, except blindness. Zebrafish and Drosophila models also exist. Future directions and therapy Only supportive and symptomatic treatment is available. Bone marrow transplantation was ineffective [14]. Gene therapy is at the experimental stage in animals.

Rare Forms of Neuronal Ceroid Lipofuscinoses Many years after the original classification of four clinical forms of NCL (infantile to adult), results of gene identification, initially by linkage studies, expanded the genetic and nosographic spectrum to CLN8, although numerically there were fewer patients and families than those with CLN1-3.

CLN4: Adult NCL CLN4 is now considered the only dominant form of NCL, not further discussed here. CLN5: Finnish variant late infantile NCL CLN5 represents a rare, late infantile variant of NCLs that is predominant in Finland (MIM#256731). The disease is caused by mutations in the CLN5 gene located on chromosome 13q21q32 encoding a soluble lysosomal glycoprotein of unknown function. Up to autumn 2017, 36 different mutations have been described. The major Finnish mutation is a 2-bp deletion (c.1175delAT) identified in 94% of Finnish diseased chromosomes [30]. The CLN5 phenotype resembles the CLN2 disease process in many aspects. However, the disease onset is later, usually at the age of 4.5 to 7 years, most often with motor clumsiness or concentration problems. The clinical course is slower than in CLN2 patients, but clinical symptoms are similar, with progressive intellectual decline, visual failure, ataxia, myoclonus, and epilepsy [11]. Early involvement of the cerebellum is observed by both MRI and SPECT. Death occurs usually in the second or third decade. At autopsy, the brain is severely atrophic and weighs from 450 to 660 g. A striking and typical feature is extreme cerebellar atrophy. Histopathological study shows similar morphological changes to those seen in CLN2 patients [31]. Neuronal loss is most severe in the neocortex and the cerebellum. A laminar pattern of neuronal loss, most prominent in cortical layers III and V, the presence of meganeurites in the remaining neurons of cortical layer III, and extensive gliosis are typical. The cerebellar cortex shows almost complete loss of Purkinje cells and granule

377

Developmental Neuropathology cells with gliosis. Storage material in the CNS and in extraneural sites reveals strong immunoreactivity for SCMAS and weak immunoreactivity for SAPs. At the ultrastructural level, the pleomorphic character of the storage cytosomes is another typical feature of CLN5. fingerprint, curvilinear, and rectilinear profiles and their modified forms may be observed in various proportions in the tissues examined. Rectilinear profiles are stacks of short oligolamellar mostly straight structures, which are seen either as classic forms or as numerous variants; so the term “rectilinear complex” describes their multifaceted ultrastructure [32]. Vacuolated lymphocytes are absent. Biochemical diagnostic assays are not yet available; thus, diagnosis is based on clinicopathologic findings and DNA study. The pathogenesis is unknown. Treatment is symptomatic. Spontaneous animal models exist in borderdale sheep, Devon cattle, border collie species, and as a knockout murine model.

CLN6: Late infantile/early juvenile variant NCL CLN6 is a rare form of NCL known as a variant late infantile/early juvenile NCL or Lake–Cavanagh disease (MIM 601780). Clinical features are similar to those observed in classic late infantile NCL, but the pathology resembles juvenile NCL [33]. The CLN6 gene is located on chromosome 15q2l-23 (MIM 606725). The predicted product of the CLN6 gene is a 311-amino acid protein of multi-transmembrane topology, localized to the endoplasmic reticulum. Up to autumn 2017, 71 different mutations had been identified in the CLN6 gene in families from various countries [1]. The most common mutation, found in 20 Costa-Rican families, is a non-sense mutation Glu72X [34]. As in other NCLs, the age at onset and the rate of progression of clinical symptoms may vary. The disease may start as early as 18 months and as late as eight years, usually with visual loss and seizures. Progressive visual failure, loss of speech, intellectual decline, epilepsy, ataxia, myoclonus, and irritability are typical. Loss of all motor skills is observed between 4 and 10 years, and children die usually in the second or third decade of life. CT and MRI show severe atrophy of the brain and cerebellum. At autopsy, the brain shows generalized atrophy and weighs from 600–900 g. Neuronal loss is generalized, most severe in the Vth layer of the neocortex. In the cerebellum, there is prominent loss of granule cells but Purkinje cells are better preserved. The storage granules in the cytoplasm of all cell types in the CNS and in extraneural tissues show the same staining properties and distribution as in CLN2. SCMAS immunoreactivity is strong in the brain tissue, but in contrast to CLN2, is absent in the liver, adrenals, and endocrine pancreas [26]. Lymphocytes are not vacuolated. At the ultrastructural level, rectilinear profile complexes and fingerprint profiles are found in the storage cytosomes in the brain; pure fingerprint profile inclusions in intestinal neurons; and fingerprint, curvilinear, and rectilinear profile complexes in visceral cells.

378

Treatment is symptomatic. Diagnosis is based on clinicopathologic findings and DNA study. Several naturally occurring animal models for CLN6 exist, namely the nclf mouse model, the South Hampshire sheep model, ovine neuronal ceroidlipofuscinoses, the merino sheep and the Australian shepherd dog.

CLN7: Turkish variant late infantile NCL Originally, several children of Turkish origin demonstrating a late infantile onset of seizures, poor mobility, visual impairment, and cognitive and motor deterioration with fingerprint, rectilinear, or curvilinear profile inclusions in the biopsy specimens were assigned to NCL locus CLN7, but now, patients have been identified worldwide [1]. The CLN7/MFSD8 gene is located on chromosome 4q28.2. The gene product is a 518-aminoacid lysosomal membrane protein. Up to autumn 2017, 38 mutations had been identified [1]. No spontaneous or genetically engineered animal models have been reported, except in the zebrafish [8]. CLN8: Turkish variant late infantile and northern epilepsy The Turkish variant late infantile and northern epilepsy is a newly identified form of NCL caused by mutations in the CLN8 gene located on chromosome 8p23 (18) (MIM 607837). The product of CLN8 gene is a 33-kDa protein of predicted multitransmembrane topology localized to the endoplasmic reticulum. CLN8 mutations may produce two different clinical phenotypes: so-called Turkish variant late infantile, and progressive epilepsy with mental restriction also known as northern epilepsy (MIM 600143). Up to autumn 2017, 31 different mutations from across the globe have been identified in patients with the Turkish variant late infantile [1], while all patients with northern epilepsy are homozygous for an Arg24Gly missense mutation. Turkish variant late infantile The Turkish variant late infantile disease starts at the age of 3– 7.5 years [30]. The clinical picture includes progressive visual loss, speech delay, seizures, intellectual decline, myoclonus, and ataxia. Storage material found in biopsy specimens demonstrates autofluorescence and staining characteristics typically observed in NCLs. Ultrastructurally, condensed fingerprint, rectilinear, and curvilinear profile inclusions are found. Autopsy studies have not been published. Northern epilepsy So far, the northern epilepsy form of NCL has been identified in several patients, all of whom are from northeastern Finland [35]. The disease starts with epilepsy at 5–10 years of age followed by slight motor dysfunction and slowly progressive dementia. Visual acuity may be reduced, but retinal degeneration is absent. The patients may survive to the sixth decade of life. CT and MRI show mild generalized brain atrophy in patients older than 30 years of age.

The Neuronal Ceroid Lipofuscinoses Chapter 30

At autopsy, the brain weighs 1040–1530 g and may appear normal or be atrophic. Histopathology shows slight loss of neurons in the Vth cortical layer and CA2 hippocampal sector. Storage demonstrates staining characteristics typical for all NCLs. The most prominent accretion of the storage material is found in the deep part of cortical layer III and in hippocampal sectors CA2-CA4, whereas Purkinje cells, substantia nigra, and locus coeruleus are almost unaffected. Neurons located in the superficial parts of lamina III demonstrate meganeurites filled with the storage material. Storage shows strong immunoreactivity for SCMAS, SAP D, and amyloid beta and is also observed in internal organs. By electron microscopy, structures resembling curvilinear profiles with some admixture of a granular component were identified [35]. The pathogenesis is unknown. Treatment is symptomatic. The sporadic English setter model has been found equivalent to CLN8 [36]. The mnd mouse represents a spontaneous mouse model of CLN8 [30].

CLN10: Congenital NCL CLN10 is a congenital form of NCL, and postmortem findings have been published in the premolecular era [37]. The CLN10/CTSD gene encodes cathepsin D, a lysosomal protease. Spontaneous animal models occur in Swedish Landrace sheep and in the American bulldog. Genetically engineered models also exist in mice, zebra fish and drosophila [8]. Human late infantile and juvenile forms have also been reported [1]. Additional recent forms of neuronal ceroid lipofuscinosis Thus far, a further six genetic forms have been identified by exome sequencing (Table 30.1) [7]. These include new adult forms (CLN11, CLN13) and exome sequencing has also clarified CLN4 as a separate adult autosomal-dominant type of NCL. There are few patients or families and mutations (CLN10:7; CLN11:2; CLN12:1; CLN13:5; CLN14:1) in each type up to autumn 2017 [1], but several adult families and patients, described in the premolecular era have later been assigned to new genes (CLN4, CLN11). Publications are sparse and findings are still incomplete.

References 1. Mole SE. (2017) MRC Laboratory for Molecular Cell Biology, University College London. NCL Mutation and Patient Database. Available at: http://www.ucl.ac.uk/ncl/mutation.shtml (accessed November 30, 2017) 2. Haltia M, Goebel H (2013) The neuronal ceroid-lipofuscinoses: a historical introduction. Biochim Biophys Acta 1832:1795–800 3. Anderson GW, Goebel HH, Simonati A (2013) Human pathology in NCL. Biochim Biophys Acta 1832:1807–26 4. Schulz A, Kohlsch¨utter A, Mink J et al. (2013) NCL diseases: clinical perspectives. Biochim Biophys Acta 1832:1801–6

5. Roosing S, van den Born LI, Sangermano R et al. (2015) Mutations in MFSD8, encoding a lysosomal membrane protein, are associated with nonsyndromic autosomal recessive macular dystrophy. Ophthalmology 122:170–9 6. Sun Y, Almomani R, Breedveld GJ et al. (2013) Autosomal recessive spinocerebellar ataxia 7 (SCAR7) is caused by variants in TPP1, the gene involved in classic late-infantile neuronal ceroid lipofuscinosis 2 disease (CLN2 disease). Hum Mutat 34:706–13 7. Warrier V, Vieira M, Mole SE (2013) Genetic basis and phenotypic correlations of the neuronal ceroid lipofusinoses. Biochim Biophys Acta 1832:1827–30 8. Bond M, Holthaus SM, Tammen I et al. (2013) Use of model organisms for the study of neuronal ceroid lipofuscinosis. Biochim Biophys Acta 1832:1842–65 9. Santavuori P, Gottlob I, Haltia M et al. (1999) Infantile and other types of NCL with GROD. In: HH Goebel, SE Mole, BD Lake, eds, The Neuronal Ceroid Lipofuscinoses (Batten Disease). Amsterdam, IOS Press, pp. 16–36 10. Wisniewski KE, Kida E, Golabek AA et al. (2001) Neuronal ceroid lipofuscinoses: classification and diagnosis. Adv Genet 45: 1–34 11. Santavuori P, Lauronen L, Kirveskari E et al. (2000) Neuronal ceroid lipofuscinoses in childhood. Neurol Sci 21:35–41 12. van Diggelen OP, Thobois S, Tilikete C et al. (2001) Adult neuronal ceroid lipofuscinosis with palmitoyl-protein thioesterase deficiency: first adult-onset patients of a childhood disease. Ann Neurol 50:269–72 13. Haltia M (2003) The neuronal ceroid-lipofuscinoses. J Neuropathol Exp Neurol 62:1–13 14. Lake BD, Steward CG, Oakhill A et al. (1997) Bone marrow transplantation in late infantile Batten disease and juvenile Batten disease. Neuropediatrics 28:80–1 15. Hofmann S, Das AK, Lu JY, Soyombo AA (2001) Positional candidate gene cloning of CLN1. In: KE Wisniewski, N Zhong, eds, Batten Disease: Diagnosis, Treatment And Research. San Diego, CA, Academic Press, pp. 69–92 16. Lukacs Z, Santavuori P, Keil A et al. (2003) Rapid and simple assay for the determination of tripeptidyl peptidase and palmitoyl protein thioesterase activities in dried blood spots. Clin Chem 49: 509–11 17. Gupta P, Soyombo AA, Atashband A et al. (2001) Disruption of PPT1 or PPT2 causes neuronal ceroid lipofuscinosis in knockout mice. Proc Nat Acad Sci U S A 98:13566–71 18. Lonnqvist T, Vanhanen SL, Vettenranta K et al. (2001) Hematopoietic stem cell transplantation in infantile neuronal ceroid lipofuscinosis. Neurology 57:1411–16 19. Sleat DE, Donnelly RJ, Lackland H et al. (1997) Association of mutations in a lysosomal protein with classical late-infantile neuronal ceroid lipofuscinosis. Science 277:1802–5 20. Wheeler RB, Schlie M, Kominami E et al. (2001) Neuronal ceroid lipofuscinosis: late infantile or Jansky Bielschowsky type– re-revisited. Acta Neuropathol 102:485–8 21. Williams RE, Gottlob BD, Lake BD et al. (1999) CLN2. Classic late infanile NCL. In: HH Goebel, SE Mole, BD Lake, eds, The Neuronal Ceroid Lipofuscinoses (Batten Disease). Amsterdam, IOS Press, pp. 37–54 22. Kida E, Golabek AA, Wisniewski KE (2001) Cellular pathology and pathogenic aspects of neuronal ceroid lipofuscinoses. Adv Genet 45:35–68

379

Developmental Neuropathology 23. Bernardini F, Warburton MJ (2002) Lysosomal degradation of cholecystokinin-(29-33)-amide in mouse brain is dependent on tripeptidyl peptidase-I: implications for the degradation and storage of peptides in classical late-infantile neuronal ceroid lipofuscinosis. Biochem J 366:521–9 24. Stengel C (1982) Account of a singular illness among four siblings in the vicinity of Røraas. In: D Armstrong, N Koppang, JA Rider, eds, Ceroid-Lipofuscinosis (Batten’s Disease). Amsterdam, Elsevier Biomedical Press. 25. Wisniewski KE, Zhong N, Kaczmarski W et al. (1998) Compound heterozygous genotype is associated with protracted juvenile neuronal ceroid lipofuscinosis. Ann Neurol 43:106–10 26. Elleder M, Sokolova J, Hrebicek M (1997) Follow-up study of subunit c of mitochondrial ATP synthase (SCMAS) in Batten disease and in unrelated lysosomal disorders. Acta Neuropathol 93: 379–90 27. Hofman I, Kohlsch¨utter A, Santavuori P et al. (1999) CLN3. Juvenile NCL. In: HH Goebel, SE Mole, BD Lake, eds, The Neuronal Ceroid Lipofuscinoses (Batten Disease), Amsterdam, IOS Press, pp. 55–76 28. Puranam KL, Guo WX, Qian WH et al. (1999) CLN3 defines a novel antiapoptotic pathway operative in neurodegeneration and mediated by ceramide. Mol Genet Metabolism 66:294–308 29. Golabek AA, Kida E, Walus M et al. (2000) CLN3 protein regulates lysosomal pH and alters intracellular processing of Alzheimer’s amyloid-beta protein precursor and cathepsin D in human cells. Mol Genet Metab 70:203–13 30. Ranta S, Savukoski M, Santavuori P, Haltia M (2001) Studies of homogenous populations: CLN5 and CLN8 In: KE Wisniewski, N Zhong, eds, Batten Disease: Diagnosis, Treatment and Research. San Diego, CA, Academic Press, pp. 123–40

380

31. Tyynela J, Suopanki J, Santavuori P et al. (1997) Variant late infantile neuronal ceroid-lipofuscinosis: pathology and biochemistry. J Neuropathol Exp Neurol 56:369–75 32. Elleder M, Lake BD, Goebel HH et al. (1999) Definitions of the ultrastructural patterns found in NCL. In: HH Goebel, SE Mole, BD Lake, eds, The Neuronal Ceroid Lipofuscinoses (Batten Disease). Amsterdam, IOS Press, pp. 5–15 33. Williams RE, Lake BD, Elleder M, Sharp J. (1999) CLN6 Variant late infantile/early juvenile NCL. In: HH Goebel, SE Mole, BD Lake, eds, The Neuronal Ceroid Lipofuscinoses (Batten Disease). Amsterdam, IOS Press, pp. 103–7 34. Teixeira CA, Espinola J, Huo L et al. (2003) Novel mutations in the CLN6 gene causing a variant late infantile neuronal ceroid lipofuscinosis. Hum Mutat 21:502–8 35. Herva R, Tyynela J, Hirvasniemi A et al. (2000) Northern epilepsy: a novel form of neuronal ceroid-lipofuscinosis. Brain Pathol 10:215–22 36. Katz ML, Khan S, Awano T et al. (2005) A mutation in the CLN8 gene in English Setter dogs with neuronal ceroid-lipofuscinosis. Biochem Biophys Res Commun 327:541–7 37. Siintola E, Partanen S, Stromme P et al. (2006) Cathepsin D deficiency underlies congenital human neuronal ceroid-lipofuscinosis. Brain 129:1438–45

NCL websites Batten Disease Support and Research Association (BDSRA): http://www.bdsra.org NCL Resources: http://ucl.ac.uk/ncl NCL Mutation Database: http://www.ucl.ac.uk/ncl/mutation.shtml

31

Peroxisomal Disorders Phyllis L. Faust Columbia University Department of Pathology and Cell Biology, New York, NY, USA

Definition, major synonyms and historical perspective Peroxisomal disorders are genetically determined metabolic diseases due to either an abnormality in the biogenesis of the peroxisome, termed peroxisomal biogenesis disorders (PBD), or a single peroxisomal enzyme or transporter deficiency. The major organ systems affected are the adrenal cortex, ears, eyes, kidney, liver, skeleton, and particularly the central nervous system. Those with prominent involvement of the nervous system are listed in Box 31.1, but only those lesions affecting the nervous system in children are discussed in this chapter. The systemic and other findings are described in several reviews [1–3]. Of the single peroxisomal protein deficiencies, only the most common pseudo-PBD, D-bifunctional protein deficiency (BPD), and the far more prevalent childhood form of X-linked adrenoleukodystrophy (X-ALD) are considered. Details about the other rarer types can be found elsewhere [1]. PBD includes four major entities: Zellweger (cerebrohepatorenal) syndrome, neonatal adrenoleukodystrophy, infantile Refsum or phytanic acid storage disease, and rhizomelic chondrodysplasia punctate (RCDP), type I or classical. Zellweger syndrome is the most common of the PBD and was the first human disease recognized as peroxisomal in 1973 [4]. Zellweger syndrome, neonatal adrenoleukodystrophy, and infantile Refsum disease are included under the rubric of the Zellweger spectrum disorders, representing a disease spectrum ranging from severe (Zellweger syndrome), to intermediate (neonatal adrenoleukodystrophy), and mild (infantile Refsum disease) phenotypes [5,6].

Box 31.1 Neuroperoxisomal disorders. Peroxisomal biogenesis disorders (multiple peroxisomal protein deficiencies, morphologically abnormal peroxisomes) r Zellweger syndrome r Neonatal adrenoleukodystrophy r Infantile Refsum disease r Rhizomelic chondrodysplasia punctata, type 1, classical Single protein deficiencies (morphologically intact peroxisomes) r Pseudoperoxisomal biogenesis disorders: r Acyl-CoA oxidase deficiency (pseudo-neonatal adrenoleukodystrophy) r Bifunctional protein deficiency r Rhizomelic chondrodysplasia punctata type II (dihydroxyacetone phosphate acyltransferase deficiency) r Rhizomelic chondrodysplasia punctata type III (alkyl dihydroxyacetone phosphate synthase deficiency) r Zellweger-like syndrome r Adrenoleukodystrophy/adrenomyeloneuropathy r Refsum’s disease, classical, adult r Miscellaneous: glutaric aciduria, type III (glutaryl-CoA oxidase deficiency)

X-ALD has two major phenotypes: juvenile (childhood cerebral) adrenoleukodystrophy and its adult variant adrenomyeloneuropathy [3]. Adrenoleukodystrophy is the major disease with a single protein deficiency presenting in childhood. It had been referred to as “Schilder disease” and was classified as a sudanophilic leukodystrophy until the uniqueness of

Developmental Neuropathology, Second Edition. Edited by Homa Adle-Biassette, Brian N. Harding, and Jeffrey A. Golden. © 2018 John Wiley & Sons Ltd. Published 2018 by John Wiley & Sons Ltd.

381

Developmental Neuropathology PBD and most other peroxisomal diseases, including the pseudo-PBD, are neonatal–infantile diseases in which males and females are equally affected. Adrenoleukodystrophy primarily affects boys (5–10 years of age); adrenomyeloneuropathy has onset in adults between 20–40 years of age. No risk factors have been identified for PBD. Head trauma has been reported as an inciting factor in a small number of adrenoleukodystrophy patients [3].

Clinical features

Figure 31.1 Diagnostic ballooned striated adrenocortical cells from signal case of adrenoleukodystrophy (hematoxylin and eosin stain).

its adrenal lesion (Figure 31.1) [7] was recognized at about the same time and in the same department (Pathology, Albert Einstein College of Medicine, New York, NY) as the discovery of Zellweger syndrome. Adrenoleukodystrophy was not recognized to be a peroxisomal disease, however, until a decade later [3].

Normal embryology The peroxisomes in the brain are smaller (micro-peroxisomes) than those in the liver and kidney and are present in all neural cell types, but show biochemical and functional diversity between cell types and brain regions [8,9]. In the developing and mature mammalian brain, peroxisomes are abundant in oligodendrocytes, where they become concentrated in paranodal loops and play important roles in myelin formation and maintenance. Peroxisomes are also abundant at the termini of developing neurons. In the mature brain, peroxisomes remain abundant in astrocytes. In human fetuses, catalase-positive neurons are detected in the basal ganglia, thalamus and cerebellum at 27–28 weeks and in the frontal cortex at 35 weeks of gestation. In humans and other mammals, the prominence of peroxisomes in neurons decreases with postnatal age. Catalase-positive glia are identified in deep white matter at 31–32 weeks of gestation, and throughout the remainder of gestation they appear to shift from deep to superficial white matter [10].

Signs and symptoms Zellweger syndrome is the most severe form and accounts for about one-half of patients with PBD, while neonatal adrenoleukodystrophy accounts for about one-third, and infantile Refsum disease about one-tenth of cases. The PBD and pseudo-PBD are characterized by dysmorphic facies, hypotonia, seizures, and psychomotor restriction, usually recognized in the neonatal period. Some may have calcific stippling (e.g., in the patella), and shortened forelimbs (rhizomelia, most prominent in RCDP). Atypical pigmentary retinopathy, cataracts, sensorineural hearing deficits, and hepatomegaly with eventual cirrhosis and steatosis are common. Death usually occurs in the first year of life for patients with Zellweger syndrome. The clinical course of patients with neonatal adrenoleukodystrophy and infantile Refsum disease is more variable, with survival into teens for neonatal adrenoleukodystrophy and even adulthood in infantile Refsum disease. Clinical phenotypes in classical RCDP or isolated deficiency of plasmalogen biosynthesis enzymes in RCDP type 2 or RCDP type 3 are indistinguishable, and include severe shortening and stippled calcification of the humerus and femur at birth, psychomotor and growth restriction, and microcephaly. The clinical presentation of adrenoleukodystrophy is markedly different and may consist of either Addison’s disease, affecting 80% of males before adulthood [11,12], or progressive neurological impairment. In adrenoleukodystrophy (around 35% of patients with X-ALD), this includes behavioral abnormalities (emotional lability, hyperactive behavior, school difficulties), spasticity, ataxia, aphasia and visual or auditory agnosia. Untreated, the disease is rapidly progressive, typically leading to death within three years.

Imaging Epidemiology Incidence and prevalence The estimated incidence of the PBD is 1 in 50 000 births [6]. In adrenoleukodystrophy, the minimum frequency of hemizygotes in the United States is 1 in 42 000 and of combined hemizygotes plus heterozygotes is 1 in 16 800; worldwide studies suggest a similar incidence rate around 1 in 30 000 in males [11].

382

The neuronal migration defects of the PBD and certain pseudoPBD, which are particularly prominent in Zellweger syndrome, are well appreciated on magnetic resonance imaging (MRI) (13, 14). Germinal matrix or subventricular cysts in the caudothalamic groove are common. Thickened cortical mantles containing many little dots on T2-weighted images reflect polymicrogyria, often in a perisylvian location. Pachygyria is seen as broad gyri

Peroxisomal Disorders Chapter 31

due to thickened cortex, characteristically more perirolandic where the sulcus can be exceptionally deep. Heterotopic gray matter deposits may be seen in subcortical and periventricular areas. The white matter in Zellweger syndrome (and RCDP) is reduced in volume on CT and MRI and in Zellweger syndrome shows signal abnormalities in T1- and T2-weighted images, often including the posterior limbs, signifying hypomyelination or dysmyelination [13,14]. On the other hand, neonatal adrenoleukodystrophy exhibits progressive white matter hypodensities, indicative of myelin breakdown. While cerebral abnormalities were previously believed to be absent in infantile Refsum disease, some patients develop clinical and radiological evidence of a progressive leukoencephalopathy involving cerebral and cerebellar white matter or MRI evidence of parasylvian dysplasia [1]. Adrenoleukodystrophy shows diffuse bilateral symmetrical changes in cerebral white matter, especially the parieto-occipital areas, on both computed tomography (CT, hypodensity) and T2-weighted MRI (high signal) images (Figure 31.2). Contrast enhancement at the edges of the white matter lesions is so characteristic as to be almost diagnostic. Arcuate fiber sparing is best seen on T1-weighted images [3]. Proton magnetic resonance spectroscopy can anticipate future MRI abnormalities in

Figure 31.2 T2-weighted magnetic resonance image exhibiting characteristic bilaterally symmetrical parieto-occipital high signal confluent lesions of adrenoleukodystrophy (light regions). Image kindly provided by Marjo van der Knaap.

white matter of patients with adrenoleukodystrophy while it still appears normal on conventional MRI or diffusion tensor imaging: specifically, there is increased choline and decreased N-acetylaspartate, presumed to reflect biochemical alterations in myelin sheaths and axonal dysfunction, respectively. Even in the normal appearing white matter of patients only with the Addisonian phenotype, N-acetylaspartate is lower than controls without this phenotype [15].

Biochemistry and laboratory findings Peroxisomes are involved in many facets of lipid metabolism, in addition to their role in neutralizing hydrogen peroxide and reactive oxygen species, as executed by over 50 matrix enzymes. Laboratory tests to confirm clinical diagnoses begin with targeted analyses of metabolite levels in plasma. Virtually all peroxisomal disorders are characterized by elevations of very long chain fatty acids (VLCFA), particularly C26:0, due to defective peroxisomal β-oxidation. PBD also display diagnostic abnormalities reflecting loss of multiple peroxisomal functions: elevations of di- and trihydroxycholestanoic (bile acid precursors) and pipecolic acids, but also phytanic/pristanic acids if enough has been ingested through the diet. Decreased levels of ether phospholipids, such as plasmalogens, and polyunsaturated fatty acids, such as docosahexaenoic acid, are typical [6]. The pseudo-PBD and adrenoleukodystrophy, which selectively interrupt the peroxisomal β-oxidation system, exhibit only VLCFA abnormalities; BPD also has elevations in the cholestanoic acids [3,6,11]. Diagnosis may require functional enzyme assays of fatty acid β-oxidation and phytanic acid α-oxidation, de novo plasmalogen biosynthesis, measurement of enzyme activities (e.g., acyl-CoA oxidase, D-bifunctional protein, dihydroxyacetone phosphate [DHAP]-acyltransferase) and immunostaining for peroxisomal matrix/membrane proteins in fibroblast cultures. These biochemical tests will characterize several of the single enzyme deficiencies, which may be difficult to clinically distinguish from PBD, in particular for BPD and the different RCDPs. Of note, patients with milder peroxisomal defects may have normal plasma metabolites, which may not necessarily reflect the level of accumulation in tissues, and their recognition as peroxisomal disorders requires more detailed testing in fibroblasts and/or genetic analyses (see Genetics section of this chapter). VLCFA abnormalities also can be identified in brain white matter of active demyelinating lesions and in the adrenal cortex of patients with adrenoleukodystrophy, as well as in the brains of fetal Zellweger syndrome. If diagnosis is not made antemortem and a peroxisomal disorder is suspected, samples of plasma, brain, liver and adrenal gland should be frozen at autopsy for biochemical analysis, and a sample of skin removed for fibroblast culture. It is of practical importance to note that assay of VLCFA and phytanic acid is possible on formalin-fixed tissue due to their great insolubility in aqueous solutions. Newborn screening methods have been developed, implemented in New York State to detect adrenoleukodystrophy and

383

Developmental Neuropathology other peroxisomal β-oxidation disorders [16]. In March 2016, the US Health and Human Services added this test to the Recommended Uniform Screening Panel for newborns.

Differential diagnosis For the PBD, a variety of dysgenetic syndromes characterized by facial dysmorphism must be eliminated by appropriate laboratory studies, as mentioned above. In view of the multisystemic involvement, several mitochondrial disorders also need to be differentiated from the PBD and pseudo-PBD. The differential diagnosis for adrenoleukodystrophy is more limited. Firstly, the Addisonian phenotype must be distinguished from true Addison disease due to primary immunemediated adrenocortical atrophy. The central nervous system (CNS) component needs to be distinguished from other disorders of white matter, in particular other leukodystrophies. The childhood–juvenile age of onset in males is highly suggestive of adrenoleukodystrophy, particularly when combined with the imaging findings discussed above. Confirmation of the appropriate elevation of VLCFA in blood and genetic testing makes a specific diagnosis possible.

Pathology Macroscopy The most impressive gross findings in PBD and certain pseudoPBD, particularly Zellweger syndrome and BPD, are the cerebral neocortical abnormalities of pachygyria or polymicrogyria due to disordered neuronal migrations (Figure 31.3) [17,18]. The white matter is often discolored in Zellweger syndrome and neonatal adrenoleukodystrophy or frankly reduced in volume in Zellweger syndrome and RCDP. In Zellweger syndrome, there is a unique combination of bilaterally symmetric gyral abnormalities: centrosylvian or parasylvian pachygyria (medial) and polymicrogyria (laterally). Other areas of polymicrogyria and pachygyria are variably seen, and there may be an abnormal vertical tilt to the sylvian fissure (Figure 31.3). Coronal sections of the cerebrum exhibit a thickened cortex with either numerous superficial indentations and/or prominent subcortical heterotopias. Subventricular or germinal matrix cysts adjacent to the caudate nuclei are common [1]. The deep gray matter, cerebellum, brainstem, and spinal cord are usually within normal limits, except for dysplasia of inferior olive and often cerebellar dentate nuclei. The CNS in neonatal adrenoleukodystrophy shows greater involvement of white matter than disrupted neuronal migration, although polymicrogyria may be diffuse, focal or multifocal and associated with subcortical heterotopias; pachygyria has not been reported. A wide spectrum of changes is reported in BPD, ranging from changes similar to that seen in Zellweger syndrome or neonatal adrenoleukodystrophy [19]. Atrophy of the cerebellar cortex is prominent in chronic RCDP, and also reported in infantile Refsum disease [1,20].

384

Figure 31.3 Overview of cerebrum of Zellweger syndrome showing pachygyria along the medial frontal lobes (black arrow), polymicrogyria more laterally and prominent Rolandic sulcus (white arrow).

Adrenoleukodystrophy displays preservation of cortical gray matter, but profound alterations in cerebral and less commonly or severely in cerebellar white matter. Both parieto-occipital regions are usually the most severely and earliest involved by a symmetrical gray/tan discoloration of white matter that is firm to rubbery on palpation (Figure 31.4). As confirmed by imaging studies, these lesions progress, often asymmetrically, toward the frontal lobes. The edge between the lesion and “normal” white matter can be distinct or blurred and U-fibers are usually spared. Atrophy of the corpus callosum and optic nerves are typical, as is the early and prominent involvement of the splenium and posterior limbs (12, 21). Secondary corticospinal tract degeneration extending down through the peduncles, basis pontis, medullary pyramids and spinal cord is characteristic, and the basis pontis may also display demyelinated foci [1,21].

Histopathology, immunohistochemistry and ultrastructure Gray matter In the PBD and pseudo-PBD, areas of polymicrogyria exhibit fusions of the molecular layer and abnormal cortical laminations (Figure 31.5). Medium to large pyramidal cells typically destined for deep cortical layers are admixed with reduced numbers of layer II and III neurons in the outer cortex, while many layer II and III neurons are found within the deep cortical layers and subcortical heterotopias. Pachygyric cortex lacks the superficial fusions and has larger subcortical heterotopias often

Peroxisomal Disorders Chapter 31

Figure 31.6 Periodic acid–Schiff stain-positive macrophages with lipid cytosomes in Zellweger syndrome (periodic acid–Schiff stain with diastase).

Figure 31.4 Coronal section of posterior cerebral hemisphere of adrenoleukodystrophy demonstrating confluent demyelination with some arcuate sparing (top left).

containing a higher percentage of larger pyramidal neurons. Classical studies of Zellweger syndrome suggest that all neuronal classes are affected, but especially the outer cortical group [17]. Striated and globose periodic acid–Schiff (PAS)-positive macrophages have been noted in several gray and white matter areas of Zellweger syndrome (Figure 31.6); at the ultrastructural level, abnormal lipid cytosomes are observed (Figure 31.7). Neonatal adrenoleukodystrophy has fewer cortical abnormalities, usually foci of polymicrogyria, while infantile Refsum disease and RCDP usually have little to none [1]. In cerebellum of Zellweger syndrome, neonatal adrenoleukodystrophy and BPD, heterotopic Purkinje cells or

Figure 31.5 Polymicrogyria of Zellweger syndrome (hematoxylin and eosin stain).

combinations of Purkinje cells and granule cells (heterotaxias) are found in the white matter; classical cerebellar cortical heterotaxias are most conspicuous in the nodulus of Zellweger syndrome [1,17,18]. A marked loss of cerebellar granule cells occurs in chronic RCDP and infantile Refsum disease; Purkinje cells were also deficient in RCDP, and reportedly heterotopic in the molecular layer in infantile Refsum disease and BPD [1,19,20]. Microscopic neuronal organizational deficits (dysplasias) are most prominent in the principal nuclei of the inferior medullary olives of Zellweger syndrome, RCDP, infantile Refsum disease, and BPD, where the normal serpiginous pattern is replaced by focal discontinuities to a simplification, or peripheral palisading of its neurons (Figure 31.8) [17]. Dysplastic claustra and dentate nuclei have been reported in Zellweger syndrome, dysplastic dentate in BPD [19], and neuronal loss without dysplasias in the dentate and olivary nuclei of neonatal adrenoleukodystrophy [1].

Figure 31.7 Pale membranous cytoplasmic bodies in Zellweger syndrome macrophage. (Uranyl acetate-lead citrate).

385

Developmental Neuropathology

Figure 31.8 Dysplasia of the inferior olive in Zellweger syndrome with a discontinuous and simplified morphology and peripheral palisading of neurons (hematoxylin and eosin stain).

Immunohistochemical decreases in multiple peroxisomal matrix proteins have been demonstrated in Zellweger syndrome [22,23]. In many PBD, peroxisomal membrane protein antibodies detect remnant peroxisomes present as “membrane ghosts”, representing enlarged vesicles lacking matrix proteins due to the peroxisomal protein import defect [1,2]. In BPD, peroxisomes may be reduced in number, enlarged or undetectable [19]. Zellweger syndrome fetuses have revealed neocortical lesions as early as 14 weeks estimated gestational age, including micropolygyric ripples, subtle subcortical heterotopias and thin cortical plates, all of which are more obvious at an estimated gestational age of 22–24 weeks [24]. PAS-positive macrophages, astrocytes, neuroblasts, immature neurons, radial glia, and meningeal cells display pleomorphic lipid cytosomes on electron microscopy, including electron-opaque membranous cytoplasmic bodies. Dysplastic olives and dentate nuclei were also present in the Zellweger syndrome fetuses (24). After completion of neuronal migrations, neurons in peroxisomal disorders may be affected by degeneration or storage processes. Sensorineural hearing loss has been observed in the Zellweger spectrum and RCDP; atypical pigmentary retinopathy also is typical of the Zellweger spectrum. Few pathologic studies are available, but neurodegenerative atrophy of the inner ear sensory epithelium and loss of ganglion and photoreceptor cells in the retina occurs. There is also some evidence of storage in the form of spicular inclusions (typical of PBD and some adrenoleukodystrophy/adrenomyelneuropathy macrophages) in retinal macrophages and pale membranous cytoplasmic bodies in ganglion cells [1]. Last, there is a limited neuronal storage restricted to the dorsal nucleus of Clarke and lateral cuneate nuclei, seen as striated perikarya due to accumulations of lamellar lipid clefts, similar to those of adrenoleukodystrophy, in both the perikarya and neighboring axonal spheroids [25].

386

White matter All of the PBD and pseudo-PBD with neurological signs or symptoms demonstrate lesions in CNS white matter, particularly cerebral, which for the most part appear to reflect developmental problems (i.e., hypomyelination or dysmyelination). Demyelination may also play a role particularly in longer-surviving patients. Myelin pallor, some oligodendrocyte loss, mild reactive astrocytosis, and a few sudanophilic/PAS-positive macrophages may be the sole microscopic abnormality, even when the white matter is grossly reduced in volume. As expected, this change can be most severe in Zellweger syndrome, but classical inflammatory cells (lymphocytes, plasma cells) are not observed in Zellweger syndrome. On the other hand, inflammatory demyelinative lesions are typical of neonatal adrenoleukodystrophy, which resemble those of adrenoleukodystrophy but on a reduced scale. Comparable inflammatory demyelination also has been noted in an older (14-month) child with BPD, suggesting that the age of patient rather than the specific peroxisomal defect may determine whether the white matter lesion is inflammatory or not [1,25]. Adrenoleukodystrophy, however, is best known for its severe inflammatory CNS white matter disease, with confluent areas of myelin and oligodendrocyte loss (Figure 31.9) that are replaced by fibrillary astrogliosis. The earliest morphologic lesions are interpreted as dysmyelinative, with myelin vacuolation and splitting, mild reactive astrocytosis and scant microglial infiltration (Figure 31.10), seen just beyond or even far from the active edge of inflammatory demyelination. These early lesions are believed to reflect spontaneous myelin breakdown (myelinolysis) due to a destabilizing accumulation of VLCFA in proteolipid protein and complex myelin lipids, such as gangliosides, phosphatidylcholine and sulfatides [12]. The most distinctive hallmark of the adrenoleukodystrophy white matter lesion is marked chronic inflammation with many T-lymphocytes, particularly cytotoxic CD8+ cells

Figure 31.9 Inflammatory demyelination of cerebral white matter in adrenoleukodystrophy with perivascular lymphocytes (far right) and macrophages (far left) separated by demyelinated white matter containing many reactive astrocytes and few oligodendrocytes (hematoxylin and eosin stain).

Peroxisomal Disorders Chapter 31

Figure 31.10 Dysmyelinative lesion of adrenoleukodystrophy. Myelin pallor with scattered interstitial and perivascular periodic acid–Schiff-positive macrophages in “normal” white matter. (Luxol fast blue–periodic acid–Schiff stain).

in the early stage, and macrophages [21,26,27]. This resembles multiple sclerosis, but differs from it by the paucity of B-cells/plasma cells and the localization of the lymphocytes at the lesion’s edge in multiple sclerosis and within the demyelinative edge in adrenoleukodystrophy. Macrophages present in adrenoleukodystrophy demyelinative lesions are variably sudanophilic and PAS-positive, reflecting myelin lipid catabolism to cholesterol esters, typical of biochemically normal myelin. However, many of the macrophages are also birefringent in polarized light, due to accumulation of cholesterol esterified with predominantly saturated VLCFA. At the ultrastructural level these latter macrophages may contain either membranebound spicules, presumably lysosomal, or lamellar lipid profiles lying free in the cytoplasm [28]. Lamellar lipid profiles are the most characteristic ultrastructural inclusion of adrenoleukodystrophy, representing linear to gently curved electron dense lipid leaflets separated by a variable electron-lucent space. This severe neuroinflammation probably causes the loss of oligodendrocytes, which die by cytolysis rather than by apoptosis [26]. Many effector molecules are increased in lesions, including proinflammatory cytokines (tumor necrosis factor-alpha, interleukins (IL)-1, -2, -6 and -12, interferon-γ), chemokines, nitric oxide and CD1 [27,29,30]. As in most leukodystrophies, axonal loss is severe; secondary neuronal loss in deep gray matter can occur and secondary tract (Wallerian-like) degeneration is commonly noted in the pyramidal system. Even here, a mild lymphocytic response is uniquely noted, while none is seen in other examples of tract degeneration.

Genetics and pathogenesis PBD are inherited in an autosomal recessive pattern and caused by mutations in one of at least 14 genes (PEX genes) encoding

proteins called peroxins (PEXp) that control peroxisome assembly and division [31]. Most human peroxins are involved in matrix protein import (PEX1, 2, 5, 6, 7, 10, 12, 13, 14), whereas PEX3, 16, and 19 regulate targeting of integral membrane proteins, and PEX11β is primarily involved in peroxisomal proliferation. Defects in these genes lead to variable loss of matrix protein import and peroxisomal membrane biogenesis. Matrix protein import, which is ATP-dependent, employs two peroxisomal targeting signals (PTS):1 and 2. The vast majority of matrix proteins are sorted by the PTS1 receptor PEX5p, which recognizes a carboxyl-terminal serine–lysine–leucine sequence, followed by translocation across the membrane through complex import machinery. The PTS2 receptor PEX7p recognizes aminoterminal sequences on thiolase, alkyl-DHAP synthase and phytanoyl-CoA hydroxylase, with proteolytic cleavage of the PTS2 signal after peroxisomal import. Peroxisomes import fully folded, cofactor-bound and even oligomeric matrix proteins. The targeting of integral membrane proteins is less well understood and controversial; mechanisms involving either direct import from the cytosol by PEX19p recognition of internally located membrane targeting signals or initial trafficking through the endoplasmic reticulum induced by PEX3p are proposed [31]. Genotype–phenotype correlations are poor because defects in the same gene can produce different phenotypes and defects in different genes can produce the same phenotype. Defects in PEX1 are causal in about 60–70% of patients with PBD and manifest as Zellweger syndrome, neonatal adrenoleukodystrophy, or infantile Refsum disease: the Zellweger spectrum. The severity of the various clinical phenotypes has been linked to the amount of residual protein/enzyme activity resulting from the specific genetic defect [2]. For instance, the PEX1 2097insT frameshift mutation has no import activity and a Zellweger syndrome phenotype, whereas the PEX1 G843D missense mutation has approximately 15% residual matrix protein import and the mildest infantile Refsum disease phenotype [32]. Even when matrix protein import is severely compromised, the peroxisomal membrane can still be recognizable as distorted or empty “ghosts” with the use of antibodies specific for peroxisomal integral membrane proteins, such as PMP70/ABCD3 or PEX14. With deficiency of PEX3p, PEX16p or PEX19p, peroxisomal membranes do not form and a Zellweger syndrome phenotype occurs. Over 90% of RCDP patients have defects in PEX7, with L292ter causing a severe classical phenotype in almost 50% of patients with RCDP due to lack of residual activity in the mutant PEX7p [2]. Patients with PEX7 mutations are grouped into PBD because more than one peroxisomal pathway is affected, although the peroxisomal structure remains intact. The genetic deficiencies in both BPD and adrenoleukodystrophy affect peroxisomal β-oxidation, but the impact of the mutational defect in BPD is much clearer than in adrenoleukodystrophy. For example, the G46A missense mutation in BPD disrupts the NAD+ binding site of the dehydrogenase-coding region of D-bifunctional protein and, hence, will block

387

Developmental Neuropathology peroxisomal β-oxidation. The adrenoleukodystrophy gene encodes an ATP binding cassette (ABC) half-transporter, originally designated ALDP but now ABCD1; it is a peroxisomal integral membrane protein that is predominantly a homodimer, although heterodimers with related ABCD2 and ABCD3 are possible [12]. ABCD1 transports VLCFA-coenzyme A across the peroxisomal membrane [12]. Mutation analysis from several laboratories reveals more than 643 different ABCD1 mutations (summarized in the Adrenoleukodystrophy Database [33]), many of which are private, affecting any of the 10 exons with exon 5 being a “hot spot.” Approximately 54% are missense, 25% are frameshift, 10% are non-sense mutations and 7% are deletions. Two-thirds of the fibroblasts of patients with adrenoleukodystrophy do not express detectable levels of ALDP/ABCD1. Of great importance, no genotype– phenotype correlations have been identified and the same molecular defect can produce any of the X-ALD phenotypes from isolated Addisonian hypocortisolism to juvenile cerebral adrenoleukodystrophy to adrenomyelneuropathy [11]. This has generated the hypothesis that a modifier gene(s) and/or environmental factors (e.g., possible head trauma) are responsible for the phenotypic variation.

Animal models and pathogenesis Several knockout mouse models have been developed for Zellweger syndrome through targeted disruption of Pex5 [34], Pex2 [35] and Pex13 [36], and to some degree Pex11β [37]. These models have recapitulated many of the biochemical and pathologic features of Zellweger syndrome, particularly a neocortical neuronal migration defect, severe hypotonia and early neonatal lethality. CNS intrinsic defects are modeled by Nestin-Cre mediated deletion of Pex5 in all neural cell types, where neuronal migration defects, delayed cerebellar development, myelination defects, axon loss and neuroinflammation throughout the brain occur [38]. However, defective hepatic and CNS peroxisomes both contribute to neuronal migration defects, and systemic metabolic abnormalities, particularly due to liver dysfunction, may impact on neuronal migration (reviewed in Berger et al. [39]). A knock-in mouse model carrying a missense mutation in the Pex1 gene (Pex1-G844D) recapitulates the most frequent mutation in the human Zellweger spectrum disorders with a milder pathology [40]. A Pex7 knockout mouse mimicking RCDP has been developed [41]. The neuronal migration defect of Zellweger syndrome, based on human postnatal and fetal studies correlating neuropathologic and biochemical abnormalities, was postulated to be due to the highly insoluble VLCFA being incorporated into cell membranes or acylated adhesion molecules and acting in concert with local axon-radial glia tissue constraints [24]. Mouse models have provided insights but also raised questions about the relationship between particular biochemical abnormalities and CNS defects. Mice with defective peroxisomal β-oxidation

388

alone [42], in contrast to humans with BPD, do not have neuronal migration defects, although they do develop late-onset cerebellar aberrations and axonal loss, motor impairment and lethargy, similar to Nes-Pex5-/-. In Pex7 knockout mice, neocortical lamination abnormalities were detected, but so were elevations in VLCFA (deficient Pex7p-mediated import of thiolase) that are not seen in human RCDP. Both the Pex7 knockout mouse, and more severe early defects in other mice with combined VLCFA and plasmalogen deficiencies, suggests a synergistic effect of these metabolic abnormalities [39]. The Pex11β knockout mouse [37] has neurodevelopmental defects but no accumulation of VLCFA and only a slight decrease in plasmalogen levels; the disease pathogenesis remains unknown, although an effect on neuroinflammatory pathways is proposed. Peroxisome deficiency in mice is associated with early activation of the innate immune system that evolves into chronic microglial activation, myelin abnormalities and axon loss [43]; however, the microglial signature in peroxisomal β-oxidation deficient mice has unique “non-neurodegenerative” features and does not lead to overt phagocytic or neurotoxic activity [44]. The Pex5, Pex2 and Pex13 knockouts have displayed severe biochemical and structural abnormalities in mitochondria (reviewed in Crane [5], Berger et al. [39]), comparable to the original study of human Zellweger syndrome [4]. Increasing evidence strongly implicates pathogenic roles for oxidative stress/ damage and concomitant mitochondrial dysfunction in PBD. More recently, recognition of aggregation of α-synuclein in peroxisome-deficient mouse brain and peroxisomal functions in cell signaling, including innate immunity, structural disturbance to membrane-raft domains, tuberous sclerosis complex regulation of mammalian target of rapamycin, and potential accumulation of signaling compounds such as prostaglandins and leukotrienes (normally degraded in peroxisomes) that mediate inflammatory pathways, raise new ideas for disease pathogenesis (reviewed in Crane [5]). The insolubility and membrane perturbing capacity of VLCFA also have been invoked to play a major role in the destabilization of the myelin membrane in adrenoleukodystrophy [25]. The human neuropathologic data have long suggested a two-stage pathogenesis in the white matter lesion of adrenoleukodystrophy: dysmyelination due to a biochemicallyinduced destabilization of the myelin sheaths followed by a fulminant immune inflammatory demyelination [7] that may be initiated by cytokines, particularly TNF-α, CD1 molecules and cytotoxic CD8 T cells [26]. The profound loss of oligodendrocytes appears to be mediated by cytotoxic CD8 T cells and perhaps nitric oxide and is most consistent with a cytolytic death [26]. The presence of CD1 molecules indicates that lipid antigen presentation may be significant. The vehemence of the inflammatory reaction and the rapidity of the clinical progression suggest a role for antigenic determinant spreading and/or superantigen presentation. Accumulation of VLCFA in macrophages and microglia in adrenoleukodystrophy is proposed to affect their ability to support normal immunological brain function [12,39].

Peroxisomal Disorders Chapter 31

Progressive inflammatory demyelination in adrenoleukodystrophy also coincides with dysfunction of brain endothelial cells, contributing to increased trafficking of leukocytes across the blood brain barrier [45]. Several ABCD1 knockout mouse models have been developed for adrenoleukodystrophy [46,47). Notably, none have yet demonstrated the inflammatory demyelinative lesions of human adrenoleukodystrophy, and most of the neuropathologic lesions in these murine models have consisted of axonal degeneration more reminiscent of adrenomyelneuropathy [48]. Further studies in ABCD1 knockout mice have strongly supported a pathogenic effect of saturated VLCFA and a role for oxidative stress, mitochondrial dysfunction, and proteasome and autophagy regulation in the pathogenesis of X-ALD [49]. Importantly, it has provided another link to a concomitant mitochondrial abnormality in adrenoleukodystrophy, first demonstrated in adrenocortical cells in the 1970s [7]. VLCFA can disturb calcium homeostasis and cause mitochondrial dysfunction in neuronal cell cultures as well as toxicity to oligodendrocytes [39]. Increased microsomal chain elongation by ELOVL fatty acid elongase 1 may also explain elevations in VLCFA [50]. Mice with oligodendroglia selective peroxisome deficiency, using CNPase-Cre mediated Pex5 deletion, can be considered as a phenocopy model for the inflammatory form of adrenoleukodystrophy, recapitulating progressive symmetric subcortical demyelination, severe axonal loss, VLCFA accumulation and neuroinflammation, including infiltrating CD8 T cells and cytokine production [51]. It is postulated that ABCD1deficient peroxisomes gradually accumulate secondary peroxisomal changes that ignite the inflammatory response in brain; for instance, a concomitant depletion of plasmalogens is noted in human adrenoleukodystrophy demyelinated brain tissue. The concentration of peroxisomes at paranodal loops in oligodendrocytes may also provide trophic support to axons independent of myelin.

mildly impaired adrenoleukodystrophy have responded to bone marrow transplantation, which has been its most effective therapy thus far [56]. Gene therapy in autologous hematopoietic stem cells is becoming a realistic possibility [57]. Newborn screening methods are a major advancement for early identification of adrenoleukodystrophy. Immunomodulatory and immunosuppressive drugs have failed to prevent progression of cerebral neuroinflammation in adrenoleukodystrophy. Dietary restriction and treatment with glyceryl trierucate/glyceryl trioleate (Lorenzo’s oil), despite its ability to reduce plasma VLCFA, has not had a significant effect on the progression of adrenoleukodystrophy. It may, however, help to delay the onset of cerebral disease in asymptomatic boys with normal MRI findings [58]. Modulation of microsomal elongases have the potential to reduce VLCFA levels (12,50). In view of the mitochondrial defects and presence of reactive species and oxidative damage in human adrenoleukodystrophy and mouse ABCD1-/- tissues, antioxidant and mitochondrial-enhancing therapies are under consideration [49]. Much has been learned about the morphological, biochemical, cellular, and molecular intricacies of peroxisomes and their disorders. However, insight into how these metabolic defects cause the various developmental and degenerative brain phenotypes remains limited and debated. The increasing recognition of pathogenic roles for oxidative stress/damage and concomitant mitochondrial dysfunction and activation of neuroinflammatory responses in both groups of peroxisomal disorders may open new avenues for therapeutic intervention. Peroxisomes are no longer considered as isolated organelles, and both physical connections and metabolic exchanges with mitochondria, endoplasmic reticulum and the lysosomal–endosomal– autophagosomal system are increasingly recognized and offer further avenues for investigations of disease pathogenesis.

References Treatment, future perspective, conclusions At present, there is no effective therapy for PBD, and supportive treatments are employed [6]. Dietary restriction can reduce phytanic acid levels. Docosahexaenoic acid has been used in the treatment of PBD with reported success, but a double-blind trial did not show efficacy [52]. Small molecules with chaperone-like properties have enhanced residual protein activity in cells from PBD patients with intermediate and milder phenotypes [53]; a clinical trial for patients with Pex1-G843D mutations has been initiated [54]. Ether lipid precursors will rescue plasmalogen levels and non-CNS organ pathology in Pex7 knockout mice [55] but efficacy in patients with RCDP remains to be determined. Adrenocortical insufficiency in adrenoleukodystrophy can and must be controlled with glucocorticoid replacement therapy; however, this has had no impact on the neurological disease. Children who are neurologically intact or who have very

1. Faust PL, Powers JM (2015) Peroxisomal disorders. In: S Love, H Budka, JW Ironside, A Perry, eds, Greenfield’s Neuropathology, 9th ed. Boca Raton, FL, CRC Press. pp. 562–88 2. Gould SJ, Raymond GV, Valle D (2001) The peroxisome biogenesis disorders. In: CR Scriver, AL Beaudet, WS Sly, D Valle, eds, The Metabolic and Molecular Bases of Inherited Disease. New York, NY, McGraw-Hill. pp. 3181–218 3. Moser HW, Smith KD, Watkins PA et al. (2001) X-linked adrenoleukodystrophy. In: The Metabolic and Molecular Bases of Inherited Disease, CR Scriver, AL Beaudet, WS Sly, D Valle, eds. New York, NY, McGraw-Hill. pp. 3257–301 4. Goldfischer S, Moore CL, Johnson AB et al. (1973) Peroxisomal and mitochondrial defects in the cerebro-hepato-renal syndrome. Science 182:62–4 5. Crane DI (2014) Revisiting the neuropathogenesis of Zellweger syndrome. Neurochem Int 69:1–8 6. Braverman NE, Raymond GV, Rizzo WB et al. (2016) Peroxisome biogenesis disorders in the Zellweger spectrum: An overview of

389

Developmental Neuropathology

7.

8.

9.

10.

11.

12. 13. 14. 15.

16.

17.

18.

19.

20.

21.

22.

23.

24.

390

current diagnosis, clinical manifestations, and treatment guidelines. Mol Genet Metab 117:313–21 Powers JM, Schaumburg HH (1974) Adreno-leukodystrophy (sexlinked Schilder’s disease). A pathogenetic hypothesis based on ultrastructural lesions in the adrenal cortex, peripheral nerve and testis. Am J Pathol 76:481–92 Ahlemeyer B, Neubert I, Kovacs WJ, Baumgart-Vogt E (2007) Differential expression of peroxisomal matrix and membrane proteins during postnatal development of mouse brain. J Comp Neurol 505:1–17 Roels F, Depreter M, Espeel M et al. (2003) Peroxisomes during development and in distinct cell types. In: Roels F, Baes M, De Bie S (eds.) Peroxisomal Disorders and Regulation of Genes. New York, NY, Kluwer Academic/Plenum. pp 39–54 Houdou S, Kuruta H, Hasegawa M (1991) Developmental immunohistochemistry of catalase in the human brain. Brain Res 556: 267–70 Wiesinger C, Eichler FS, Berger J (2015) The genetic landscape of X-linked adrenoleukodystrophy: inheritance, mutations, modifier genes, and diagnosis. Appl Clin Genet 8:109–21 Berger J, Forss-Petter S, Eichler FS (2014) Pathophysiology of X-linked adrenoleukodystrophy. Biochimie 98:135–42 Barkovich J, Peck WW (1997) MR of Zellweger syndrome. Am J Neuroradiol 18:1163–70 van der Knaap MS, Valk J (1995) Magnetic Resonance of Myelin, Myelination, and Myelin Disorders. Berlin, Springer-Verlag Eichler FS, Itoh R, Baker PB et al. (2002) Proton MR spectroscopic and diffusion tensor brain MR imaging in X-linked adrenoleukodystrophy: initial experience. Radiology 225:245–52 Turgeon CT, Moser AB, Morkrid L et al. (2015) Streamlined determination of lysophosphatidylcholines in dried blood spots for newborn screening of X-linked adrenoleukodystrophy. Mol Genet Metab 114:46–50 Evrard P, Caviness VS, Prats-Vinas J, Lyon G (1978) The mechanism of arrest of neuronal migration in the Zellweger malformation: An hypothesis based upon cytoarchitectonic analysis. Acta Neuropathol 41:109–17 Volpe JJ, Adams RD (1972) Cerebro-hepato-renal syndrome of Zellweger. An inherited disorder of neuronal migration. Acta Neuropathol 20:175–98 Ferdinandusse S, Denis S, Mooyer PA et al. (2006) Clinical and biochemical spectrum of D-bifunctional protein deficiency. Ann Neurol 59:92–104 De Munter S, Verheijden S, Regal L, Baes M (2015) Peroxisomal Disorders: A Review on Cerebellar Pathologies. Brain Pathol 25:663–78 Schaumburg HH, Powers JM, Raine CS, Suzuki K (1975) Adrenoleukodystrophy. A clinical and pathological study of 17 cases. Arch Neurol 33:577–91 Itoh M, Suzuki Y, Akaboshi S et al. (2000) Developmental and pathological expression of peroxisomal enzymes: their relationship of D-bifunctional protein deficiency and Zellweger syndrome. Brain Res 858:40–7 Takashima S, Chan F, Becke LE et al. (1991) Cortical cytoarchitectural and immunohistochemical studies on Zellweger syndrome. Brain Dev 13:158–62 Powers JM, Tummons RC, Caviness VS et al. (1989) Structural and chemical alterations in the cerebral maldevelopment of fetal

25.

26.

27.

28.

29.

30.

31. 32.

33.

34. 35.

36.

37.

38.

39.

40.

41.

cerebro-hepato-renal (Zellweger) syndrome. J Neuropathol Exp Neurol 48:270–89 Powers JM, Moser HW (1998) Peroxisomal disorders: Genotype, phenotype, major neuropathologic lesions, and pathogenesis. Brain Pathol 8:101–20 Ito M, Blumberg BM, Mock, DJ et al. (2001) Potential environmental and host participants in the early white matter lesion of adrenoleukodystrophy: morphologic evidence for CD8 cytotoxic t cells, cytolysis of oligodendrocytes and CD1-mediated lipid antigen presentation. J Neuropathol Exp Neurol 60:1004–19 Powers JM, Liu Y, Moser AB, Moser HW (1992) The inflammatory myelinopathy of adreno-leukodystrophy: Cells, effector molecules and pathogenetic implications. J Neuropathol Exp Neurol 51:630–43 Schaumburg HH, Powers JM, Suzuki K, Raine CS (1974) Adrenoleukodystrophy (sex-linked Schilder disease). Ultrastructural demonstration of specific cytoplasmic inclusions in the central nervous system. Arch Neurol 31:210–13 Paintlia AS, Gilg AG, Khan M et al. (2003) Correlation of very long chain fatty acid accumulation and inflammatory disease progression in childhood X-ALD: implications for potential therapies. Neurobiol Dis 14:425–39 Powers JM, Pei Z, Heinzer AK et al. (2005) Adreno-leukodystrophy: oxidative stress of mice and men. J Neuropathol Exp Neurol 64:1067–79 Fujiki Y, Okumoto K, Mukai S et al. (2014) Peroxisome biogenesis in mammalian cells. Front Physiol 5:307 Crane DI, Maxwell MA, Paton BC (2005) PEX1 mutations in the Zellweger spectrum of the peroxisome biogenesis disorders. Hum Mutat 26:167–75 Kennedy Krieger Institute, Laboratory Genetic Metabolic Diseases at the Academic Medical Center/Emma Children’s Hospital (2017) Adrenoleukodystrophy Database. Available at http://www.x--ald.nl (accessed November 6, 2017) Baes M, Gressens P, Baumgart E et al. (1977) A mouse model for Zellweger syndrome. Nat Genet 17:49–57 Faust PL, Hatten ME (1997) Targeted deletion of the PEX 2 peroxisome assembly gene in mice provides a model for Zellweger syndrome, a human neuronal migration disorder. J Cell Biol 59: 1063–9 Maxwell M, Bjorkman J, Nguyen T et al. (2003) Pex13 inactivation in the mouse disrupts peroxisome biogenesis and leads to a Zellweger syndrome phenotype. Mol Cell Biol 23:5947–57 Li X, Baumgart E, Morrell JC et al. (2002) PEX11β deficiency is lethal and impairs neuronal migration but does not abrogate peroxisome function. Mol Cell Biol 22:4358–65 Hulshagen L, Krysko O, Bottelbergs A et al. (2008) Absence of functional peroxisomes from mouse CNS causes dysmyelination and axon degeneration. J Neurosci 28:4015–27 Berger J, Dorninger F, Forss-Petter S, Kunze M (2016) Peroxisomes in brain development and function. Biochim Biophys Acta 1863:934–55 Hiebler S, Masuda T, Hacia JG et al. (2014) The Pex1-G844D mouse: a model for mild human Zellweger spectrum disorder. Mol Genet Metab 111:522–32 Brites P, Motely AM, Gressens P et al. (2003) Impaired neuronal migration and endochronal ossification in Pex7 knockout mice: a model for rhizomelic chondrodysplasia punctata. Hum Mol Genet 12:2255–67

Peroxisomal Disorders Chapter 31 42. Baes M, Gressens P, Huyghe S et al. (2002) The neuronal migration defect in mice with Zellweger syndrome (Pex5 knockout) is not caused by the inactivity of peroxisomal β-oxidation. J Neuropathol Exp Neurol 61:368–74 43. Bottelbergs A, Verheijden S, Van Veldhoven PP et al. (2012) Peroxisome deficiency but not the defect in ether lipid synthesis causes activation of the innate immune system and axonal loss in the central nervous system. J Neuroinflammation 9:61 44. Verheijden S, Beckers L, Casazza A et al. (2015) Identification of a chronic non-neurodegenerative microglia activation state in a mouse model of peroxisomal beta-oxidation deficiency. Glia 63:1606–20 45. Musolino PL, Gong Y, Snyder JM et al. (2015) Brain endothelial dysfunction in cerebral adrenoleukodystrophy. Brain 138:3206–20 46. Kobayashi T, Shinnoh N, Kondo A, Yamada T (1995) Adrenoleukodystrophy protein deficient mice represent abnormality of very long chain fatty acid metabolism. Biochem Biophys Res Commun 232:631–6 47. Lu J-F, Lawler AM, Watkins PA et al. (1997) A mouse model for Xlinked adrenoleukodystrophy. Proc Natl Acad Sci U S A 94:9366–71 48. Pujol A, Hindelang C, Callizot N et al. (2002) Late onset neurological phenotype of the X-ALD gene inactivation in mice: a mouse model for adreno-myeloneuropathy. Hum Mol Genet 11:499– 505 49. Fourcade S, Ferrer I, Pujol A (2015) Oxidative stress, mitochondrial and proteostasis malfunction in adrenoleukodystrophy: a paradigm for axonal degeneration. Free Radic Biol Med 88:18–29 50. Ofman R, Dijkstra IM, van Roermund CW et al. (2010) The role of ELOVL1 in very long-chain fatty acid homeostasis and X-linked adrenoleukodystrophy. EMBO Mol Med 2:90–7

51. Kassmann CM, Lappe-Siefke C, Baes M et al. (2007) Axonal loss and neuroinflammation caused by peroxisome-deficient oligodendrocytes. Nat Genet 39:969–76 52. Paker AM, Sunness JS, Brereton NH et al. (2010) Docosahexaenoic acid therapy in peroxisomal diseases: results of a double-blind, randomized trial. Neurology 75:826–30 53. Zhang R, Chen L, Jiralerspong S et al. (2010) Recovery of PEX1-Gly843Asp peroxisome dysfunction by small-molecule compounds. Proc Natl Acad Sci U S A 107:5569–74 54. McGill University Health Center (2016) Betaine and Peroxisome Biogenesis Disorders. NCT01838941. Available at: https://clinical trials.gov/ct2/show/NCT01838941 (accessed November 6, 2017). 55. Brites P, Ferreira AS, da Silva TF et al. (2011) Alkyl-glycerol rescues plasmalogen levels and pathology of ether-phospholipid deficient mice. PLoS One 6:e2853 56. Shapiro E, Krivit W, Lockman L et al. (2000) Long-term effect of bone-marrow transplantation for childhood-onset cerebral X-inked adrenoleukodystrophy. Lancet 356:713–18 57. Eichler F, Duncan C, Musolino PL et al. (2017) Hematopoietic stem-cell gene therapy for cerebral adrenoleukodystrophy. N Engl J Med 377:1630–1638 58. Moser HW, Raymond GV, Lu SE et al. (2005) Follow–up of 89 asymptomatic patients with adrenoleukodystrophy treated with Lorenzo’s oil. Arch Neurol 62:1073–80

Online resources Peroxisome database: http://www.peroxisomedb.org/home.jsp Adrenoleukodystrophy database: http://www.x-ald.nl Global Foundation for Peroxisomal disorders: http://www.thegfpd.org

391

32

Mitochondrial Disorders Anders Oldfors1 and Brian N. Harding2 1 Department 2 Department

of Pathology, Sahlgrenska University Hospital, Gothenburg, Sweden of Pathology and Laboratory Medicine, Children’s Hospital of Philadelphia, Philadelphia, PA, USA

Definition, major synonyms and historical perspective The mitochondrion is the power plant of the cell. Mitochondrial disorders are due to defective oxidative phosphorylation (OXPHOS), which is of major importance for ATP production. The OXPHOS system is made up of five enzyme complexes (I–V), which are embedded in the inner mitochondrial membrane. Individuals with mitochondrial disorder can present with symptoms from almost every organ or tissue of the body. The nervous system and skeletal muscle are especially vulnerable. Thus, the most common forms of mitochondrial disorders are encephalomyopathies [1]. All age groups can be affected, but this chapter focuses on pediatric forms. Mitochondrial diseases are also called disorders of the respiratory or electron transport chain or diseases of oxidative phosphorylation, since other diseases of the mitochondria such as beta-oxidation defects are not included in this entity. The history of mitochondrial disorders goes back 40 years, when a patient with defective coupling of OXPHOS was first described (Luft disease). During the 1960s, several patients with multisystem disorders and morphologically abnormal mitochondria were identified. By means of biochemical and enzyme histochemical techniques developed in the 1970s, mitochondrial diseases could be more accurately characterized. Several syndromes such as Leigh syndrome, Alpers syndrome and Kearns–Sayre syndrome were demonstrated to be mitochondrial disorders. The first gene mutations associated with mitochondrial diseases were described in 1988. Since the 1990s, mitochondrial disorders have emerged as a major clinical entity and one of the most common hereditary neurodegenerative disorders.

Embryology Prenatal manifestations of mitochondrial disorders are not frequent but may occur and also involve the central nervous system [2].

Epidemiology In a study from western Sweden, the incidence of mitochondrial encephalomyopathies in preschool children (below six years of age) was estimated to be 8.9 in 100 000 [3]. An epidemiological study in south Australia based upon referrals to the Melbourne Children’s Hospital over a 10-year period indicated a minimum birth prevalence of 6.2 in 100 000 [4]. In a study from the north-east of England, the minimum prevalence rate for mtDNA mutations was 1 in 5000, and the prevalence of adult mitochondrial disease was estimated to be approximately 1 in 4300, including mutations in mitochondrial as well as nuclear DNA [5].

Biochemistry The five complexes of OXPHOS are complex I (NADH: ubiquinone oxidoreductase), complex II (succinate:ubiquinone oxidoreductase, succinate dehydrogenase), complex III (ubiquinone: cytochrome c oxidoreductase), complex IV (cytochrome c oxidase), and complex V (ATP synthase). Deficiencies of each of the five complexes have been described, although the most common deficiencies affect complex I or IV. Combined deficiencies are frequent.

Developmental Neuropathology, Second Edition. Edited by Homa Adle-Biassette, Brian N. Harding, and Jeffrey A. Golden. © 2018 John Wiley & Sons Ltd. Published 2018 by John Wiley & Sons Ltd.

393

Developmental Neuropathology is a suspicion of a mitochondrial encephalomyopathy, investigation of mitochondria in muscle tissue has become an important tool, so muscle biopsy is often performed. Common techniques to study the mitochondria in muscle tissue include enzyme histochemistry, electron microscopy, and biochemical investigation of isolated fresh mitochondria by oximetry after the addition of various substrates and spectrophotometric analysis of respiratory chain enzymes and mtDNA analysis.

Neuropathology

Figure 32.1 Schematic illustration of the importance of mitochondrial and nuclear genes for the oxidative phosphorylation system. Mitochondrial DNA (mtDNA) encodes for 13 of the approximately 89 polypeptides of the oxidative phosphorylation system (OXPHOS), which is composed of five enzyme complexes (I–V). The mtDNA-encoded subunits constitute parts of complex I, III, IV, and V but not complex II (succinate dehydrogenase). The 13 polypeptides encoded by mtDNA are synthesized within the mitochondria. tRNA and rRNA genes necessary for this synthesis are encoded by mtDNA. Nuclear genes encode for approximately 76 of the subunits of OXPHOS as well as proteins that are important for assembly of the complexes of OXPHOS. These are synthesized in the cytoplasm and imported into the mitochondria. Nuclear genes also encode proteins that are important for mtDNA replication, transcription and translation.

Various methods have been applied to define the biochemical defect. In most instances, skeletal muscle mitochondria are isolated and subjected to measurements of oxygen consumption or ATP production after the addition of different substrates and ADP. The different respiratory chain enzymes may also be studied by spectrophotometric analysis. Lactate is the product of anaerobic glucose metabolism and accumulates in the body fluids when aerobic metabolism in the mitochondria is impaired. Hyperlactatemia is therefore a marker of mitochondrial disease. Hyperlactatemia also accompanies several other metabolic disorders (e.g. glycogen storage disease and disorders of the fatty acid and amino acid metabolism). It may also be the consequence of impaired circulation, hypoxia, or hepatic or renal failure.

Clinical features The most common symptoms and signs, neuroradiological features and laboratory findings of the main mitochondrial encephalomyopathic syndromes are summarized in Table 32.1. However, there is often overlap and many children with mitochondrial diseases present with other forms of encephalopathies and/or myopathies. These may be associated with other organ manifestations such as cardiomyopathy, renal tubulopathy, and liver failure. In the diagnostic work-up in patients in whom there

394

Central nervous system The neuropathological features of mitochondrial encephalomyopathies are highly variable and, with few exceptions, there are no lesions specific for mitochondrial diseases. Ultrastructural changes of mitochondria are only occasionally found. The diagnosis in most cases rests on combined clinical, biochemical, physiological, morphological and genetic investigations. Although the relatively well-characterized mitochondrial encephalomyopathic syndromes such as Kearns–Sayre syndrome, mitochondrial encephalomyopathy, lactic acidosis and stroke-like episodes (MELAS), myoclonus epilepsy and ragged red fibers (MERRF), and Leigh syndrome, show characteristic neuropathological aspects, there is a great overlap [6]. Leigh syndrome In Leigh syndrome, the brain shows multiple focal lesions of necrosis or spongiform vacuolation (Figures 32.2–32.4) [7–9]. The lesions are frequently symmetric and typically affect the basal ganglia, thalamus, midbrain, brainstem, cerebellar nuclei, and spinal cord. The cerebral cortex is usually not involved. The lesions may show complete or partial destruction of the neuropil followed by gliosis. Old lesions may resemble old infarctions, while more recent lesions typically are cellular and show partial sparing of neurons. In such lesions, there are frequent foam cells and hypertrophic astrocytes. There is also an apparent high density of small vessels, which may partly be due to shrinkage of the tissue in addition to vascular congestion. White matter changes with loss of myelin and/or spongy vacuolation are common in Leigh syndrome and predominantly affect the optic nerve, centrum semiovale, and cerebellum. Kearns–Sayre syndrome In Kearns–Sayre syndrome, the most typical and consistent change is spongiform vacuolation of the white matter (Figures 32.5, 32.6) [6,10]. The distribution and extent of the spongiosis varies and may affect the cerebrum, cerebellum, brain stem, and spinal cord. The cerebrum and midbrain are most frequently affected. The myelin changes are often widespread but may affect only a few regions. The corpus callosum and internal capsule are often spared. The vacuolation may appear as mild spongy myelin vacuolation due to splitting of myelin sheaths or as coarse vacuolation of the white matter. The cellular

Mitochondrial Disorders Chapter 32

Table 32.1 Clinical syndromes of oxidative phosphorylation disorders. Syndrome

Major clinical features

Age at presentation

Brain MRIa

Muscle pathology

Genes

GRACILE

Fetal growth restriction, amino-aciduria, cholestasis, iron overload and lactic acidosis Hypotonia, ataxia, ophthalmoplegia, sensorineural hearing loss, neuropathy, epileptic encephalopathy

Neonatal

Normal

BCS1L

Normal

C10ORF2

Leigh syndrome

Failure to thrive, hypotonia, psychomotor regression, brain stem dysfunction, ataxia, dystonia, optic atrophy

Infancy, sometimes later

Normal or subtle changes with cerebral atrophy ± periventricular white matter changes Cerebellar cortical atrophy, brain stem atrophy, High T2-intensity lesions of cerebellar white matter, the middle cerebellar peduncles and dentate nucleus Increased T2-intensity lesions in basal ganglia, mid brain and brainstem

Usually normal or COX deficiency without RRF

Alpers syndrome

Psychomotor deterioration, generalized seizures, infantile spasms, progressive microcephaly, spasticity, cortical blindness Progressive encephalopathy, complex refractory seizures, epilepsia partialis continua, liver failure Muscle weakness, exercise intolerance, lactic acidosis anemia Muscle hypotonia, weakness and wasting, exercise intolerance, episodic rhabdomyolysis

Infancy

Cortical atrophy, sometimes also involvement of basal ganglia or white matter but to a lesser extent

Usually normal

MTATP6 mt ND genes nuclear ND genes SURF1 SDHA PDHA1 SUCLA2, SUCLG1 NARS2 PARS2 FARS2

Infancy and early childhood or young adulthood

Cortical atrophy, stroke-like cortical T2-hyperintense lesions

Usually normal or few COX deficient fibers

POLG

Childhood or adolescence

Normal

Mitochondrial myopathy with COX-deficient RRF

PUS1 YARS2

Infancy, childhood, adolescence or adulthood

Normal

Mitochondrial myopathy with COX or SDH deficiency

Childhood, adolescence or adulthood

Stroke-like cortical parieto-occipital T2 hyperintense lesions, cerebellar and cerebral atrophy, basal ganglia calcifications

Mitochondrial myopathy with COX positive RRF

MTCYB mt COX genes mt tRNA genes ISCU TK2 RRM2B mt tRNA genes mt ND genes

Childhood, adolescence or adulthood

Normal or unspecific with cortical/subcortical atrophy ± white matter changes

Mitochondrial myopathy with COX deficient RRF

IOSCA

Alpers–Huttenlocher syndrome

MLASA/MSA

Isolated myopathy

MELAS

MERRF

Short stature, migraine-like headache, generalized seizures, stroke-like episodes, dementia, ataxia, sensorineural hearing loss, muscle weakness, diabetes mellitus Myoclonic and generalized seizures, ataxia, muscle weakness, sensorineural hearing loss

Infancy or early childhood

mt tRNA genes

(continued)

395

Developmental Neuropathology

Table 32.1 (Continued) Syndrome

Major clinical features

Age at presentation

Brain MRIa

Muscle pathology

Genes

Kearns–Sayre syndrome

PEO, myopathy, retinitis pigmentosa, cardiac conduction defects, short stature, sensorineural hearing loss, cerebellar ataxia, dementia Severe gastrointestinal dysmotility, cachexia, ptosis, ophthalmoparesis, peripheral neuropathy, leukoencephalopathy Ataxia, sensory neuropathy. Additional features may include PEO, migraine, severe epileptic seizures, myopathy, hearing loss

Childhood, adolescence or adulthood

Increased T2-intensity in subcortical white matter, globus pallidus and thalamus

Mitochondrial myopathy with COX deficient RRF

mtDNA deletions RRM2B

Childhood, adolescence or adulthood

Diffuse T2-hyperintensity in cerebral and cerebellar white matter with sparing of U-fibers and corpus callosum Occasionally stroke-like T2-hyperintense lesions ± cerebral and cerebellar atrophy

Mitochondrial myopathy with COX deficient RRF. Neurogenic atrophy

TYMP RRM2B

Mitochondrial myopathy with COX deficient RRF. Sometimes normal

POLG

MNGIE

SANDO/MSCAE/ MIRAS

Childhood, adolescence or adulthood

a Brain

MRI changes are variable depending on time point during disease course and other factors. COX, cytochrome c oxidase; GRACILE, growth restriction, amino-aciduria, cholestasis, iron overload, lactic acidosis and early death; IOSCA, infantile onset spinocerebellar ataxia; MELAS, mitochondrial encephalomyopathy, lactic acidosis and strokelike episodes; MERRF, myoclonus epilepsy and ragged red fibers; KSS, Kearns–Sayre syndrome; MIRAS, mitochondrial recessive ataxia syndrome; M(LA)SA, mitochondrial myopathy, (lactic acidosis) and sideroblastic anemia; MNGIE, mitochondrial neurogastrointestinal encephalopathy; MSCAE, mitochondrial spinocerebellar ataxia and epilepsy; mt, mitochondrial; ND, NADH dehydrogenase. OXPHOS, oxidative phosphorylation; PEO, progressive external ophthalmoplegia; RRF, ragged red fibers; SANDO, sensory ataxic neuropathy with ophthalmoparesis; SDH, succinate dehydrogenase.

response is often inconspicuous. In addition, there may be nerve cell degeneration and astrogliosis, especially in the brain stem and also in the cerebellum, with particular loss of Purkinje cells. Degeneration with calcifications and iron deposition has been

described in the basal ganglia and substantia nigra in several cases.

Figure 32.2 Section through the brain in a case of Leigh syndrome caused by a mutation in the mtDNA ATPase6 gene. Discoloration and cystic changes are present in the basal ganglia.

Figure 32.3 Section of the central part of the brain in the same case as in Figure 32.2. Cystic changes are present in the basal ganglia and around the third ventricle. Luxol–cresyl violet.

396

MELAS The neuropathology of MELAS syndrome is variable and includes calcification of basal ganglia, small and large necrotic

Mitochondrial Disorders Chapter 32

Figure 32.4 Higher magnification of a brain lesion in the same case as in figures 32.2 and 32.3. These are typical for Leigh syndrome, with degeneration of neuropil, increased vascular density and partial sparing of neurons. Luxol–cresyl violet.

Figure 32.6 Higher magnification of Figure 32.5 showing spongiotic change of the white matter, which is typical for Kearns–Sayre syndrome. Luxol–cresyl violet.

foci, in addition to laminar necrosis, spongy degeneration, capillary proliferation and gliosis in the neocortex [11,12]. The necrotic lesions resemble infarctions, in acute as well as in late stages, but do not respect vascular territories. Edema and an apparent increase of vascular density are frequent findings in association with the infarct-like lesions, which are typically multifocal and asymmetrical and preferentially involve the posterior part of the cerebrum. The basal ganglia and brainstem are rarely involved in the infarct-like lesions. Although mitochondrial angiopathy has been reported in leptomeningeal vessels in MELAS, there is no evidence for a causal relationship between vascular changes and infarct-like lesions. In addition to the large infarct-like lesions, small necrotic lesions may also appear in the neocortex. Mineralization in the basal ganglia is a typical finding in MELAS, and may also be disseminated and involve other brain regions. Widespread white matter gliosis and cerebellar degeneration has also been described.

Figure 32.5 Section through the right half of the brain in a case of Kearns–Sayre syndrome, caused by a single large-scale deletion of mtDNA. There is a typical pallor of the white matter. Luxol–cresyl violet.

MERRF The neuropathology of patients with MERRF syndrome is typically a systemic degeneration involving the globus pallidus, substantia nigra, red, dentate, and inferior olivary nuclei, cerebellar cortex, spinocerebellar tracts, dorsal columns, and Clarke’s column, which show degeneration and gliosis [13,14]. Macroscopic examination of the brain usually does not reveal any marked changes. Microscopically the dentate nuclei show marked neuronal loss and gliosis (Figure 32.7). Similarly, the inferior olivary nuclei are affected by neuronal loss to some extent (Figure 32.8). The red nuclei and substantia nigra are less severely affected. The superior cerebellar peduncles show atrophy and gliosis. In the cerebellar cortex, there is minor loss of Purkinje cells and there may be atrophy of the vermis. The neocortex usually does not show major pathological changes. In spite of the regional

397

Developmental Neuropathology

Figure 32.7 Neuronal loss in nucleus dentatus in a case of myoclonus epilepsy and ragged red fibers syndrome caused by a mutation in the tRNALys gene of mtDNA. Luxol–cresyl violet.

Figure 32.8 Neuronal loss in nucleus olivarius inferior in the same case of myoclonus epilepsy and ragged red fibers syndrome as in Figure 32.7. Luxol–cresyl violet.

variability, all brain regions show a very high load of mutant mtDNA [13].

astrocytosis may also involve the thalamus, lateral geniculate body, amygdala, substantia nigra and dentate nucleus. Hippocampal sclerosis and cerebellar cortical scars are common, while neuronal loss in the brainstem and tract degeneration in the spinal cord occur occasionally. The liver pathology is an important feature of Alpers disease and may vary from fatty change to severe pathological changes and cirrhosis (Figure 32.10) [17]. Exposure to valproic acid may accelerate the liver disease. Myocerebrohepatopathy is a manifestation of POLG mutations with secondary mitochondrial DNA depletion that is distinct from Alpers–Huttenlocher syndrome and presents at infancy with liver failure [18]. Alpers disease was described by Alpers as a poliodystrophy [19]. Later, Huttenlocher [20] described the association with liver disease defining the Alpers–Huttenlocher syndrome, more recently demonstrated to be caused by POLG mutations. Other

Alpers–Huttenlocher disease In Alpers–Huttenlocher disease, there are often patches of thin granular or discolored cortex and occasionally laminar destruction that typically affect the medial occipital lobe (Figure 32.9) [15,16]. Microscopic pathology is much more widespread, although patchy, and may be quite unequal between the hemispheres. It shows a graded intensification and extension through the depth of the cortex, ranging from mild superficial astrocytosis, through increasing neuronal loss, parenchymal vacuolation and astrocytosis spreading deeper within laminae, to complete replacement of the cortical ribbon by hypertrophic astrocytes and their processes associated with a prominent capillary network. Finally, there is a gliomesodermal scar containing considerable neutral fat. Neuronal loss, spongy change, and

Figure 32.9 Alpers’ disease: subtle changes in the right medial occipital cortex of an 18-year-old girl.

398

Mitochondrial Disorders Chapter 32

Figure 32.10 Alpers’ disease: normal hepatic architecture is completely effaced by numerous bile ducts. Residual hepatocytes show fatty change and there is patchy inflammation. Hematoxylin and eosin.

cases of Alpers disease without the typical severe liver association and without POLG mutations differ somewhat from Alpers– Huttenlocher syndrome, with an earlier onset, more protracted disease course and more pronounced brain atrophy [21]. Such cases have been shown to be caused by mutations in mitochondrial aminoacyl-tRNA synthetases [22,23].

Peripheral nervous system Sural nerve biopsies of patients with progressive external ophthalmoplegia, Kearns–Sayre syndrome, MELAS and MERRF syndromes, including light- and electron-microscopic analyses, may show neuropathy with axonal degeneration and loss as the predominant change [24]. Unspecific ultrastructural mitochondrial changes have been observed in Schwann cells, in addition to increased number of mitochondria in endothelial cells and pericytes of vasa nervorum. Studies on Leigh syndrome with peripheral neuropathy disclosed hypomyelination and occasionally demyelination [9]. Skeletal muscle Mitochondrial myopathy with ragged red fibers is a typical finding in diseases due to mtDNA rearrangements or tRNA point mutations. The ragged red fibers are usually cytochrome c oxidase (COX) deficient and show accumulation of abnormal mitochondria (Figures 32.11, 32.12). The fibers become COX deficient when there is accumulation of mutant mtDNA and the threshold for expression of the mutation is reached. However, in MELAS syndrome, the ragged red fibers are usually not COX deficient. Mutations in mtDNA-encoded polypeptides of complex I and V usually do not show mitochondrial myopathy, in contrast to mutations in subunits of complex III and IV. There are usually no major degenerative changes or interstitial fibrosis in muscle tissue, although DNA mutations may rarely cause recurrent rhabdomyolysis and occasionally changes suggestive of muscular dystrophy.

Figure 32.11 Section of muscle showing ragged red muscle fibers in a case of mitochondrial myopathy caused by a mutation in a mtDNA-encoded subunit of cytochrome c oxidase. Modified Gomori trichrome.

Mutations of nuclear genes causing OXPHOS deficiency are in some instances associated with mitochondrial myopathy with ragged red fibers. This is seen in the various myopathies associated with multiple mtDNA deletions and/or mtDNA depletion secondary to mutations in nuclear genes encoding for proteins that are essential for mtDNA replication and maintenance. On the other hand, mutations in nuclear genes encoding for subunits or assembly factors of the different complexes of the respiratory chain frequently do not show typical features of mitochondrial myopathy. If there is complex II (succinate dehydrogenase) or complex IV (COX) deficiency this may be identified by enzyme histochemistry.

Figure 32.12 Enzyme histochemical staining of muscle from the same case as in Figure 35.11. Combined staining of cytochrome c oxidase (COX) and succinate dehydrogenase shows frequent COX deficient (blue) muscle fibers.

399

Developmental Neuropathology

Genetics Mitochondrial disorders are usually due to genetic defects and involve the nuclear as well as the mitochondrial genomes [25]. Gene defects associated with different syndromes are indicated in Table 32.1. Of some 89 polypeptides that build up the 5 complexes of the OXPHOS system, only 13 are encoded by mtDNA [26]. The other subunits, as well as proteins important for assembly of the complexes and proteins involved in mtDNA replication, transcription, translation, and maintenance are encoded by nuclear genes (Figure 32.1). Thus, mitochondrial disorders may show maternal as well as Mendelian inheritance. Mutations in mtDNA may be of various types [27]. Largescale deletions affect tRNA/rRNA genes as well as polypeptide genes and are usually sporadic. mtDNA point mutations of tRNA genes cause defective translation of the thirteen mtDNAencoded subunits and are usually maternally inherited. mtDNA mutations in polypeptide genes are often sporadic but may show maternal inheritance. Diseases associated with mtDNA mutations include the more well-known syndromes such as Kearns–Sayre syndrome, MELAS, MERRF and Leigh syndrome (Table 32.1). Defective nuclear-encoded proteins involved in mtDNA replication and maintenance have been associated with reduced mtDNA copy number (depletion) or with multiple mtDNA deletions. In such cases, the respiratory chain deficiency is caused by mtDNA deletions or depletion but the disorders show Mendelian inheritance. Alpers–Huttenlocher syndrome (POLG mutations), MNGIE (TYMP mutations), and isolated myopathy (TK2 mutations) are typically associated with mtDNA depletion in affected tissues [28]. Multiple mtDNA deletions secondary to nuclear gene mutations are especially frequent in encephalomyopathic syndromes of adults such as autosomal dominant or recessive progressive external ophthalmoplegia associated with mutations in POLG, or C10orf2. The list of genes associated with multiple mtDNA deletions and/or depletion includes at least twelve genes [29]. During the past few years, the number of nuclear gene mutations demonstrated to cause mitochondrial disease has rapidly increased, usually representing rare disorders with various types of encephalopathies. These genes encode for a variety of mitochondrial proteins that directly or indirectly are involved in the function of the oxidative phosphorylation system [30].

Pathogenesis The pathophysiology of OXPHOS disorders has not yet been clarified, although various mechanisms have been discussed [31,32]. Impaired ATP production as a primary pathogenic event is supported by the fact that tissues with high energy demand such as the nervous system, skeletal muscle, and myocardium are frequently involved. Cells with fluctuating

400

energy demands such as neurons, muscle cells, and pancreatic beta-cells may be more vulnerable, since the function as well as the survival of cells may be affected only during periods of high ATP consumption. In diseases due to mtDNA mutations there is usually heteroplasmy (i.e., there is a mixture of mutated and wild-type mtDNA). The phenotypic expression of such mutations is related to the mutant load in an affected individual. In addition, the mutant load frequently varies between different tissues and also between single cells within a tissue. Thus, the distribution of mutated mtDNA is a factor that influences the adverse effects caused by the mutations. Calcium ion homeostasis may be perturbed by defective mitochondria and affect calcium ion signaling pathways, including muscle contraction, hormonal and neuronal signaling, and apoptosis. Reactive oxygen species are continuously produced during respiratory chain activity. The role of reactive oxygen species in mitochondrial disorders has not been clarified, but the induction of the mitochondrial manganese superoxide dismutase in complex I deficiency may be due to increased reactive oxygen species production [33]. Protein turnover is markedly changed in mitochondrial disorders. Different point mutations in protein encoding genes as well as in tRNA genes will lead to formation of truncated or defective mitochondrial protein which may affect cell function. Neuronal degeneration and cell death is common in mitochondrial diseases. Cell death may occur by necrosis or apoptosis. The importance of apoptosis in mitochondrial disease is unclear [34,35]. Apoptosis may be triggered by several mechanisms such as induction of reactive oxygen species, elevation of free Ca2+, and ATP depletion, all of which may be present in OXPHOS deficiency. Insight into the pathogenesis of human mitochondrial encephalopathies has been obtained by combined immunohistochemical analyses of OXPHOS, molecular genetic analyses at single-cell level and neuropathological investigations, and by correlating these data with clinical manifestations [36]. The angiopathy that has typically been associated with MELAS syndrome may be a more generalized part of the pathogenesis than previously anticipated [37].

Animal models Since hundreds of genes are involved in different mitochondrial diseases there is no single model for mitochondrial diseases. A large number of mouse models have been established for diseases due to primary mtDNA mutations, secondary mtDNA alterations due to nuclear gene mutations affecting mtDNA replication and maintenance and diseases due to mutations affecting the respiratory chain subunits or assembly factors [38,39]. These models mimic the corresponding diseases in humans to a variable extent. Naturally occurring mutations such

Mitochondrial Disorders Chapter 32

as the heteroplasmic mtDNA tRNATyr mutation identified in a large pedigree of golden retriever dogs with sensory ataxic neuropathy may provide possibilities to study many aspects of mitochondrial diseases [40].

Future perspectives and therapy To date, the majority of mitochondrial diseases that have been genetically defined involve mutations of mtDNA. However, an increasing number of nuclear gene mutations affecting the OXPHOS system have been identified. Considering that nuclear gene products comprise the majority of polypeptides that are involved in the OXPHOS system, it is probable that many more nuclear gene defects associated with mitochondrial encephalomyopathies will be discovered. Accurate genetic diagnosis is of great importance for genetic counseling and prognosis. Mitochondrial replacement at the oocyte or zygote stage may be an option in the future to prohibit the transmission of pathogenic mtDNA mutations [41,42]. There is currently no cure available for patients with mitochondrial disease [30]. Adequate caloric intake is important, especially in infants and young children. Treatment with coenzyme Q10 and other cofactors and vitamins is given with the aim to improve ATP production. However, improvement has been documented only in a few cases. Specific treatment of lactic acidosis in children with mitochondrial diseases and metabolic crisis may be of importance for the clinical outcome [43]. Research on targeted treatment of mtDNA diseases aims at shifting the level of heteroplasmy to reduce the level of mutant mtDNA below its threshold level for expression [44].

References 1. Oldfors A, Tulinius M (2007) Mitochondrial encephalomyopathies. Handb Clin Neurol 86:125–65 2. Tulinius M, Oldfors A (2011) Neonatal muscular manifestations in mitochondrial disorders. Semin Fetal Neonatal Med 16:229–35 3. Darin N, Oldfors A, Moslemi AR et al. (2001) The incidence of mitochondrial encephalomyopathies in childhood: clinical features and morphological, biochemical, and DNA anbormalities. Ann Neurol 49:377–83 4. Skladal D, Halliday J, Thorburn DR (2003) Minimum birth prevalence of mitochondrial respiratory chain disorders in children. Brain 126:1905–12 5. Gorman GS, Schaefer AM, Ng Y et al. (2015) Prevalence of nuclear and mtDNA mutations related to adult mitochondrial disease. Ann Neurol 77:753–9 6. Sparaco M, Bonilla E, DiMauro S, Powers JM (1993) Neuropathology of mitochondrial encephalomyopathies due to mitochondrial DNA defects. J Neuropathol Exp Neurol 52:1–10 7. Agapitos E, Pavlopoulos PM, Patsouris E, Davaris P (1997) Subacute necrotizing encephalomyelopathy (Leigh’s disease): a clinicopathologic study of ten cases. Genet Diagn Pathol 142:335–41

8. Cavanagh JB, Harding BN (1994) Pathogenic factors underlying the lesions in Leigh’s disease. Tissue responses to cellular energy deprivation and their clinico-pathological consequences. Brain 117 (Pt 6):1357–76 9. Jacobs JM, Harding BN, Lake BD et al. (1990) Peripheral neuropathy in Leigh’s disease. Brain 113(Pt 2):447–62. 10. Oldfors A, Fyhr IM, Holme E et al. (1990) Neuropathology in Kearns–Sayre syndrome. Acta Neuropathol 80:541–6 11. Hirano M, Pavlakis SG (1994) Mitochondrial myopathy, encephalopathy, lactic acidosis, and strokelike episodes (MELAS): current concepts. J Child Neurol 9:4–13 12. Prayson RA, Wang N (1998) Mitochondrial myopathy, encephalopathy, lactic acidosis, and strokelike episodes (MELAS) syndrome: an autopsy report. Arch Pathol Lab Med 122:978–81 13. Oldfors A, Holme E, Tulinius M, Larsson NG (1995) Tissue distribution and disease manifestations of the tRNA(Lys) A>G(8344) mitochondrial DNA mutation in a case of myoclonus epilepsy and ragged red fibres. Acta Neuropathol 90:328–33 14. Sparaco M, Schon EA, DiMauro S, Bonilla E (1995) Myoclonic epilepsy with ragged-red fibers (MERRF): an immunohistochemical study of the brain. Brain Pathol 5:125–33 15. Harding BN, Alsanjari N, Smith SJ et al. (1995) Progressive neuronal degeneration of childhood with liver disease (Alpers’ disease) presenting in young adults. J Neurol Neurosurg Psychiatry 58: 320–5 16. Harding BN (1990) Progressive neuronal degeneration of childhood with liver disease (Alpers–Huttenlocher syndrome): a personal review. J Child Neurol 5:273–87 17. Saneto RP, Cohen BH, Copeland WC, Naviaux RK (2013) Alpers– Huttenlocher syndrome. Pediatr Neurol 48:167–78 18. Montassir H, Maegaki Y, Murayama K et al. (2014) Myocerebrohepatopathy spectrum disorder due to POLG mutations: a clinicopathological report. Brain Dev 37:719–24 19. Alpers BJ (1931) Diffuse progressive degeneration of the grey matter of the cerebrum. Arch Neurol Psychiatr 25:469–505 20. Huttenlocher PR, Solitare GB, Adams G (1976) Infantile diffuse cerebral degeneration with hepatic cirrhosis. Arch Neurol 33:186– 92 21. Sofou K, Moslemi AR, Kollberg G et al. (2012) Phenotypic and genotypic variability in Alpers syndrome. Eur J Paediatr Neurol 16:379–89 22. Elo JM, Yadavalli SS, Euro L et al. (2012) Mitochondrial phenylalanyl-tRNA synthetase mutations underlie fatal infantile Alpers encephalopathy. Hum Mol Genet 21:4521–9 23. Sofou K, Kollberg G, Holmstrom M et al. (2015) Whole exome sequencing reveals mutations in NARS2 and PARS2, encoding the mitochondrial asparaginyl-tRNA synthetase and prolyl-tRNA synthetase, in patients with Alpers syndrome. Mol Genet Genom Med 3:59–68 24. Schr¨oder JM (1993) Neuropathy associated with mitochondrial disorders. Brain Pathol 3:177–90 25. Ylikallio E, Suomalainen A (2012) Mechanisms of mitochondrial diseases. Ann Med 44:41–59 26. Schon EA, DiMauro S, Hirano M (2012) Human mitochondrial DNA: roles of inherited and somatic mutations. Nat Rev Genet 13:878–90 27. Greaves LC, Reeve AK, Taylor RW, Turnbull DM (2012) Mitochondrial DNA and disease. J Pathol 226:274–86

401

Developmental Neuropathology 28. Suomalainen A, Isohanni P (2010) Mitochondrial DNA depletion syndromes: many genes, common mechanisms. Neuromuscul Disord 20:429–37 29. Copeland WC (2012) Defects in mitochondrial DNA replication and human disease. Crit Rev Biochem Mol Biol 47:64–74 30. Koopman WJ, Willems PH, Smeitink JA (2012) Monogenic mitochondrial disorders. N Engl J Med 366:1132–41 31. Lombes A, Aure K, Bellanne-Chantelot C et al. (2014) Unsolved issues related to human mitochondrial diseases. Biochimie 100: 171–6 32. James AM, Murphy MP (2002) How mitochondrial damage affects cell function. J Biomed Sci 9:475–87 33. Pitkanen S, Robinson BH (1996) Mitochondrial complex I deficiency leads to increased production of superoxide radicals and induction of superoxide dismutase. J Clin Invest 98:345–51 34. Di Giovanni S, Mirabella M, Papacci M et al. (2001) Apoptosis and ROS detoxification enzymes correlate with cytochrome c oxidase deficiency in mitochondrial encephalomyopathies. Mol Cell Neurosci 17:696–705 35. Sciacco M, Fagiolari G, Lamperti C et al. (2001) Lack of apoptosis in mitochondrial encephalomyopathies. Neurology 56:1070–4 36. Lax NZ, Hepplewhite PD, Reeve AK et al. (2012) Cerebellar ataxia in patients with mitochondrial DNA disease: a molecular clinicopathological study. J Neuropathol Exp Neurol 71:148–61

402

37. Lax NZ, Pienaar IS, Reeve AK et al. (2012) Microangiopathy in the cerebellum of patients with mitochondrial DNA disease. Brain 135:1736–50 38. Torraco A, Peralta S, Iommarini L, Diaz F (2015) Mitochondrial diseases. Part I: Mouse models of OXPHOS deficiencies caused by defects in respiratory complex subunits or assembly factors. Mitochondrion 21C:76–91 39. Iommarini L, Peralta S, Torraco A, Diaz F (2015) Mitochondrial diseases. Part II: Mouse models of OXPHOS deficiencies caused by defects in regulatory factors and other components required for mitochondrial function. Mitochondrion 22:96–118 40. Baranowska I, Jaderlund KH, Nennesmo I et al. (2009) Sensory ataxic neuropathy in golden retriever dogs is caused by a deletion in the mitochondrial tRNATyr gene. PLoS Genet 5:e1000499 41. Craven L, Tuppen HA, Greggains GD et al. (2010) Pronuclear transfer in human embryos to prevent transmission of mitochondrial DNA disease. Nature 465:82–5 42. Chinnery PF, Craven L, Mitalipov S et al. (2014) The challenges of mitochondrial replacement. PLoS Genet 10:e1004315 43. Danhauser K, Smeitink JA, Freisinger P et al. (2015) Treatment options for lactic acidosis and metabolic crisis in children with mitochondrial disease. J Inherit Metab Dis 38:467–75 44. Russell O, Turnbull D (2014) Mitochondrial DNA disease: molecular insights and potential routes to a cure. Exp Cell Res 325:38–43

33

Disorders of Amino Acid Metabolism and Canavan Disease Dimitri P. Agamanolis1,2 1 Department 2 Department

of Pathology, Akron Children’s Hospital, Akron, Ohio, USA of Pathology, Northeast Ohio Medical Universities (NEOMED), Rootstown, Ohio, USA

Introduction The catabolism of amino acids begins with the removal of their amino groups by transaminases and their conversion to ketoacids. Ammonia generated from the amino group is converted to urea by the urea cycle enzymes. The carbon skeletons enter the tricarboxylic acid cycle, lipid metabolism, or glycolytic pathways directly or after being further processed. The amino acid disorders are caused by inherited defects involving the degradation of amino acids and disposal of their amino groups. Many organic acidemias are also caused by defects in the catabolism of branched chain amino acids after the initial transamination step. With the exception of ornithine transcarbamylase deficiency which is X-linked, all amino acid disorders are autosomal recessive. Multiple mutations of the affected genes occur, accounting for variations in phenotype and necessitating biochemical rather than DNA-based screening methods. The amino acid disorders and organic acidemias have been the subject of numerous reviews [1–13]. Although individually rare, they are important in clinical practice because they mimic perinatal asphyxia and sepsis and are suspected in older children with encephalopathy and metabolic acidosis. In pediatric departments, more laboratory tests are ordered for the diagnosis of amino acid disorders and organic acidemias than for all other inherited metabolic diseases combined. Most amino acid disorders are associated with severe neurological manifestations which dominate their clinical picture. A few (phenylketonuria, homocystinuria) have an insidious onset and a chronic course. Most cause severe or fatal illness in the neonatal period and intellectual disability in survivors. The initial injury is compounded by subsequent neurotoxic episodes, triggered by high protein intake, infections, or metabolic stress. The clinical picture in older children resembles a static encephalopathy. Milder clinical phenotypes, caused

by less severe mutations, present later in life with episodes of metabolic decompensation, developmental delay, and intellectual disability which may be treatable [14], seizures, and ataxia. The neurological dysfunction is caused by the toxic effects of the accumulating amino acids and their intermediates, build up of harmful nitrogenous substances [5], impairment of energy and synthetic pathways, and defective synthesis of neurotransmitters. Because the neonatal illness often causes respiratory depression and seizures, the primary effects of these disorders are compounded by hypoxic and ischemic encephalopathy and excitotoxic central nervous system (CNS) injury. The “signature” neuropathological lesion of amino acid and organic acid disorders is spongy (vacuolating) myelinopathy [15–18]. This poorly understood abnormality is characterized by fluid-filled vacuoles in myelinated white matter, which, in infants, is present mainly in the brainstem, cerebellum, and spinal cord (Figure 33.1). Spongy myelinopathy involves central myelin. In cranial and spinal roots, only the proximal (CNS) portion of the root is affected. Only exceptionally has spongy myelinopathy been observed in peripheral myelin [17]. The vacuoles develop by splitting of myelin lamellae along the extracellular plane (intraperiod lines) suggesting lack of cohesion of the plasma membranes [19,20]. Accumulation of fluid in myelin corresponds to increased T2 and fluid-attenuated inversion recovery signal and restricted diffusion on diffusionweighted imaging apparent diffusion coefficient (DWI-ADC) seen on magnetic resonance imaging (MRI) [21]. The pathogenesis of spongy myelinopathy is unknown. It is conceivable that impairment of synthetic pathways in amino acid disorders and organic acidemias disturbs the lipid composition of cell membranes resulting in an unstable myelin. The natural course of spongy myelinopathy is unclear. It is not accompanied by myelin breakdown, macrophage reaction, or gliosis. It is not progressive [22]. There is no evidence that it impairs conduction or that it causes acute neonatal illness, which is adequately

Developmental Neuropathology, Second Edition. Edited by Homa Adle-Biassette, Brian N. Harding, and Jeffrey A. Golden. © 2018 John Wiley & Sons Ltd. Published 2018 by John Wiley & Sons Ltd.

403

Developmental Neuropathology explained by biochemical toxicity. Still, spongy myelinopathy is easy to recognize and is helpful as a pathological marker of these disorders. An identical myelin lesion is seen in Canavan disease, the Kearns–Sayre syndrome and other mitochondrial diseases, galactosemia, hexachlorophene encephalopathy, and in experimental triethyltin, isoniazid, cuprizone, and hexachlorophene intoxication [15]. The latter also affects peripheral myelin [15]. Distinctive pathology is seen in the urea cycle disorders (Alzheimer type II astrocytes), and homocystinuria (vascular lesions). Basal ganglia changes and cortical neuronal loss, seen in some disorders, are difficult to distinguish from hypoxic and ischemic pathology. Several amino acid disorders and organic acidurias (nonketotic hyperglycinemia, fumaric aciduria, maple syrup urine disease, histidinemia, glutaric aciduria type 2) are associated with polymicrogyria and other malformations of cortical development [23,24]. Organic acidurias (nonketotic hyperglycinemia, fumaric aciduria, glutaric aciduria type 2, 2-3 hydroxybutyric aciduria) are associated with agenesis of the corpus callosum [25]. The association of amino acid and organic acid disorders with brain malformations suggests that the metabolic abnormality is present during fetal life and affects brain development, possibly by depleting essential metabolites or by accumulation of toxic intermediates. The first step in the detection of amino acid and organic acid disorders is newborn screening of blood spots using tandem mass spectroscopy. A diagnosis can be established by detecting characteristic organic acid profiles in urine by gas chromatography/mass spectroscopy. The results may trigger additional reflex testing, such as amino acid analysis or enzyme assays in cultured fibroblasts and other cells. Prenatal diagnosis can be accomplished by detection of abnormal metabolites in amniotic fluid and by measuring enzyme activity in cultured amniocytes or chorionic villus samples. When the actual mutation is known, DNA analysis can be used for prenatal diagnosis and carrier detection. Many individual and groups of amino acid disorders and related organic acidemias have been described. The entities that are presented in this review are among the most frequent and illustrate the core features and diversity of neuropathology of this group of disorders.

Phenylketonuria Definition and chemistry The hyperphenylalaninemias (HPAs) are hereditary disorders of phenylalanine hydroxylation characterized by severe neurocognitive impairment. Phenylalanine is hydroxylated in the liver to tyrosine by phenylalanine hydroxylase (PAH). Classic phenylketonuria (PKU) is autosomal recessive and is caused by complete deficiency of PAH. These patients are intolerant of phenylalanine, have phenylalanine concentrations greater than 1200 μmol/l and, untreated, develop severe neurocognitive

404

impairment. Patients with milder forms of HPA have lower levels of phenylalanine on a normal diet and may not develop neurocognitive deficits. Tetrahydrobiopterin (TH4) is a co-factor of PAH. Rare forms of HPA arise from defects of 6 pyruvoyl tetrahydrobiopterin synthase and other enzymes of biopterin biosynthesis (“malignant” or “atypical” PKU). These HPAs are refractory to dietary therapy.

Epidemiology and genetics PKU is the most common amino acid disorder. Its incidence is highest in Turkey. In the United States, it is approximately 1 in 15 000, and is least common in African Americans [2,3]. The PAH gene is located on 12q24.1. Over 500 mutations have been shown to cause PKU. Some mutations abolish PAH activity completely and others leave some activity. Blood phenylalanine levels correlate with intelligence quotient except in rare patients who have high blood phenylalanine but normal intelligence. These patients have low brain phenylalanine, possibly because of defective transport of phenylalanine across the blood brain barrier or by other mechanisms [2,3,26]. Clinical features Protected by the maternal liver, fetuses with PKU are normal at birth. Vomiting may appear in the neonatal period. Neurological symptoms appear insidiously during the first year of life. The full-blown clinical picture of spasticity, irritability, tremors, seizures, severe intellectual disability, and microcephaly, develops gradually over 10 years [2,3,8,13,27]. Patients with PKU have a pale complexion, blond hair, and blue eyes. MRI shows increased signal in the white matter. Heterozygous females can have elevated plasma phenylalanine without clinical abnormalities. Children born to mothers with PKU that is not controlled by diet during pregnancy are dysmorphic and have intellectual disability, microcephaly, atrophy or partial agenesis of the corpus callosum, and failure to thrive [2,3,28]. Congenital heart disease is seen in 20% of these children. Fetal damage can be prevented if dietary therapy is started before conception. The clinical picture in rare forms of HPA that are caused by defects of TH4 synthesis also includes dystonia and Parkinsonian manifestations. These symptoms persist despite normalization of phenylalanine levels and require replacement therapy with L-dopa-carbidopa and 5-hydroxytryptophan in addition to dietary restriction of phenylalanine. Laboratory findings Untreated patients with PAH deficiency have phenylalanine concentrations above 1200 μmol/l and an increased phenylalanine/tyrosine ratio. Many newborns have a transient mild HPA that does not require treatment. Pathological findings PKU causes white matter pathology. In young patients, the white matter is normal or shows pallor of myelin staining.

Disorders of Amino Acid Metabolism and Canavan Disease Chapter 33

Older patients have spongy myelinopathy and patchy reduction of myelin without myelin breakdown. In advanced stages, myelin loss is more severe and is accompanied by gliosis and macrophage reaction. These changes affect the centrum semiovale, optic tracts, and fiber tracts of the brainstem [6,8,15,29]. In severe untreated PKU, the brain also shows cortical atrophy due to neuronal loss and gliosis, possibly secondary to seizures.

Pathogenesis The pathogenesis of intellectual disability and the white matter pathology in PKU is not clear. Phenylalanine is probably neurotoxic [2,3]. Amino acid imbalance, especially reduced tyrosine levels, and other poorly understood metabolic abnormalities secondary to HPA, probably cause myelin instability and impair neurotransmission [2,3,6,8]. Intellectual disability and white matter pathology can be prevented by dietary restriction of phenylalanine and can worsen or appear anew if diet is relaxed. TH4 is an essential coenzyme of tyrosine and tryptophan synthesis and patients with disorders of TH4 biosynthesis have defective synthesis of dopamine, norepinephrine, and serotonin, in addition to HPA. Treatment The basis of treatment of PKU is dietary restriction of phenylalanine starting as soon as possible after birth and continuing into adolescence and later. The goal of the diet is to normalize the concentration of phenylalanine and tyrosine, thus preventing intellectual deterioration, while giving other amino acids and nutrients to maintain adequate nutrition. Frequent monitoring of phenylalanine levels is needed to achieve desired targets of plasma phenylalanine.

Nonketotic hyperglycinemia (glycine encephalopathy) Definition and chemistry Nonketotic hyperglycinemia is an autosomal recessive inborn error of glycine degradation. It is caused by failure of the glycine cleavage system and is characterized by severe neonatal illness and increased glycine concentration in tissues, including the brain, and body fluids, including cerebrospinal fluid. Nonketotic hyperglycinemia should be distinguished from ketotic hyperglycinemia (hyperglycinemia, ketosis, and acidosis) which is due to a secondary impairment of the glycine cleavage system in the liver. Ketotic hyperglycinemia occurs in organic acidemias and is not associated with elevation of glycine in cerebrospinal fluid. Epidemiology and genetics Nonketotic hyperglycinemia is autosomal recessive and is a fairly frequent neonatal disorder. Its incidence in British Columbia is 1 in 63 000 in newborns. The highest birth incidence, 1 in 55 000, has been reported in northern Finland. The glycine cleavage system is an intramitochondrial enzyme complex consisting of four

proteins: P, H, T, and L. Mutations of the first three of these proteins have been described. More than 80% of nonketotic hyperglycinemia are caused by autosomal recessive mutations of the GLDC gene on 9p24 which encodes the P protein, and 10–15% of cases are caused by mutations of the AMT gene on 3p21.

Clinical features The most common phenotype of nonketotic hyperglycinemia is neonatal glycine encephalopathy, which, in most cases, is caused by mutations of the P protein [8,11–13]. Patients are normal at birth, but, within hours or days, develop poor feeding, lethargy, hypotonia, seizures, myoclonus, hiccups, and apnea. Acute hydrocephalus with atrophy of the white matter and corpus callosum has been reported [30]. Many patients die in a few days. Survivors have profound psychomotor restriction and seizures. Less commonly, nonketotic hyperglycinemia presents later in childhood with psychomotor restriction, choreoathetosis, and seizures, or with progressive spastic diplegia and optic atrophy. Rare cases of neonatal glycine encephalopathy subside within a few weeks without the high mortality and neurologic sequelae of nonketotic hyperglycinemia [31]. These cases are thought to be caused by transient failure of the glycine cleavage system by an unknown mechanism. Hyperglycinemia also develops in D-glyceric acidemia and following administration of valproate. Imaging findings MRI in the early phase of nonketotic hyperglycinemia shows symmetric high DWI-ADC signal in myelinated white mater, corresponding to areas of spongy myelinopathy. Agenesis of the corpus callosum and cortical dysplasia is seen in some cases. Delayed myelination and white matter loss is seen in older patients. Laboratory findings The diagnosis of nonketotic hyperglycinemia is based on elevation of glycine in CSF above 80 μmol/l (normal < 10 μmol/l) and elevation of the cerebrospinal fluid/plasma glycine ratio above 0.08 (normal less than 0.02) [8]. There is moderate elevation of glycine in plasma. The diagnosis can be made by genetic testing focusing on GLDC, AMT, and GCSH genes, and by measuring glycine cleavage system activity in the liver. The majority of patients have no detectable activity. Prenatal diagnosis can be made by genetic testing or determination of glycine cleavage system activity in chorionic villus samples. There is no reliable heterozygote or prenatal detection by glycine determination. Pathological findings Dobyns reported agenesis of the corpus callosum, abnormalities of the gyral pattern, and cerebellar hypoplasia in 6 of 15 cases of nonketotic hyperglycinemia [32]. In the majority of

405

Developmental Neuropathology

Figure 33.1 Nonketotic hyperglycinemia. Spongy myelinopathy in the cerebellum. Luxol fast blue, × 200.

reported cases the brain was grossly normal. Microscopic examination reveals spongy myelinopathy, most severe in the cerebellar white matter, corticospinal tracts, and optic tracts (Figure 33.1) [17,19,20]. There is no gliosis or macrophage reaction. In an untreated patient with neonatal glycine encephalopathy who died at 17 years of age, myelinopathy was comparable to that seen in neonatal patients [22]. That patient had a mild reduction in white matter mass with normal overall myelination, mild gliosis, and no macrophage reaction. These findings suggest that spongy myelinopathy is not progressive. The same patient also had cerebellar atrophy with extensive deposition of oxalate crystals, probably from degradation of glycine (Figure 33.2).

Pathogenesis The glycine cleavage system is active in the brain, liver, and other tissues. The most important aspect of nonketotic

Figure 33.2 Nonketotic hyperglycinemia. Oxalate crystals in the cerebellum in a 17-year-old patient. Hematoxylin and eosin, × 200.

406

hyperglycinemia is failure of the glycine cleavage system in brain, and increased concentrations of glycine in the CNS [8]. Glycine is an inhibitory neurotransmitter in the spinal cord and brainstem but has an excitatory action in the cerebral cortex as an agonist of the N-methyl-D-aspartate (NMDA) receptor channel. The glycine receptor in the spinal cord and brainstem is excitatory early in development and becomes inhibitory in mature individuals. These functions of glycine explain the combined inhibitory (hypotonia, apnea) and excitatory (seizures) manifestations of NKH [11,12,19,22]. Potentiation of the NMDA receptor of glutamate by glycine explains its neurotoxic effects. Glycine is the smallest and most ubiquitous amino acid. It accounts for a large proportion of the composition of many proteins and donates single carbon units to several synthetic pathways. It has been suggested that impairment of protein and lipid synthesis due to amino acid imbalance and deficiency of one carbon units accounts for the myelin abnormality [6,12]. The agenesis of the corpus callosum and other developmental abnormalities suggest that NKH affects the developing brain between 11 and 20 weeks of gestation.

Treatment Management of nonketotic hyperglycinemia consists of sodium benzoate to reduce glycine levels, NMDA antagonists to block glycinergic receptors, and control of seizures. There is no effective treatment, and prognosis of neonatal glycine encephalopathy is poor.

Homocystinuria and disorders of sulfur amino acids Definition and chemistry Homocystinuria is an inherited disorder of sulfur amino acids characterized by dislocation of the lens, osteoporosis, excessive height and long limbs, mental restriction, and thromboembolic phenomena. Homocystine is the oxidized form of homocysteine. Homocysteine is either catabolized to cystathionine or recycled to methionine. Catabolism to cystathionine is catalyzed by the pyridoxine-dependent enzyme cystathionine β-synthase (CBS). Most cases of homocystinuria are caused by mutations of CBS (classic homocystinuria). There are two variants of classic homocystinuria, pyridoxine (vitamin B6)-non-responsive, and the milder, B6-responsive form. Several other genetic disorders that block the conversion of homocysteine to methionine also cause homocystinuria. This conversion is catalyzed by methionine synthase and requires 5-methyltetrahydrofolic acid as a methyl donor and methylcobalamin as a co-factor. 5-10 methylenetetrahydrofolate reductase (MTHFR) is also needed in order to form 5-methyltetrahydrofolic acid. Mutations of methionine synthase and MTHFR and deficiency of methylcobalamin can cause hyperhomocysteinemia and homocystinuria. Two enzymes use cobalamin: methionine synthase, which requires methylcobalamin, and methylmalonyl coenzyme A

Disorders of Amino Acid Metabolism and Canavan Disease Chapter 33

(CoA) mutase, which uses adenosylcobalamin. Defects of cobalamin absorption (including rarely nutritional vitamin B12 deficiency), biosynthesis, and transport that affect methylcobalamin cause homocystinuria; defects that affect methylcobalamin and adenosylcobalamin cause homocystinuria and methylmalonic acidemia.

Epidemiology and genetics The incidence of homocystinuria varies from 1 in 50 000 to 1 in 400 000, being highest in Ireland and Italy [33]. The gene for CBS is located on 21q22.3. Over 150 mutations have been reported. Most mutations interfere with the activation of the enzyme and cause total loss of function that does not respond to pyridoxine. A few mutations affect the catalytic domain of the gene and leave some residual activity which can be further enhanced by administration of pyridoxine [9,13]. Clinical features Patients with homocystinuria are normal at birth but develop multisystem pathology involving the eye, skeleton, vascular system, CNS, and other organs [8,9,13]. The full-blown phenotype takes several years to evolve and pathological changes develop earlier in pyridoxine non-responsive patients. Upward dislocation of the lens and myopia appear as early as two years of age and affect the majority of the patients. The skeletal abnormalities include marfanoid habitus, osteoporosis, biconcave vertebrae, kyphoscoliosis, pectus excavatum or carinatum, and arachnodactyly. A high incidence of venous and arterial thrombosis is seen at all ages, including infancy [34]. Peripheral vein thrombosis with pulmonary embolism, ischemic stroke, and myocardial infarction are common. The risk of thromboembolic events increases with age and is higher if other risk factors such as factor V Leiden are present. Heterozygotes develop premature atherosclerosis. The neurological abnormalities of homocystinuria include developmental delay, intellectual disability, personality and behavior disorders, seizures, and focal neurological deficits.

lesions resembling subacute combined degeneration have been reported in a case of MTHFR deficiency (Figure 33.3b) [6]. Diffuse pallor of myelin staining in areas not involved by infarcts was present in a personally examined case.

Pathogenesis Dislocation of the lens is due to degeneration of the zonular fibers which are composed of the cysteine-rich protein fibrillin. The skeletal abnormalities are probably caused by

(a)

(b)

Laboratory findings and diagnosis Patients with classic homocystinuria have elevated plasma homocystine and total homocysteine, elevated urine homocysteine, elevated plasma methionine, and low CBS enzyme activity. Patients with other forms of homocystinuria have low or normal methionine. Pyridoxine administration lowers homocystine, homocysteine, and methionine in pyridoxine-responsive patients. CBS mutations can be detected by molecular genetic testing. Pathological findings The brain in homocystinuria shows arterial and venous infarcts of varying ages. Intimal fibrosis and other vascular abnormalities, presumably due to endothelial injury and organized thrombi, have been reported (Figure 33.3a). Perivascular demyelinative foci in the centrum semiovale and spinal cord

Figure 33.3 Homocystinuria. (a) Homocystinuria in a 21-year-old patient; thrombosed leptomeningeal vein. Hematoxylin and eosin, × 80. (b) Subacute combined degeneration of the spinal cord in a case of 5, 10 methylenetetrahydrofolate reductase deficiency. Luxol fast blue-Nissl (image kindly provided by Dr. B. Harding).

407

Developmental Neuropathology Urea Ornithine

Fumarate (5) Arginine

NH3 + Bicarbonate (1) Carbamoyl phosphate + Ornithine (2) Citrulline

(4)

UREA CYCLE

Mitochondrion

Argininosuccinate

(3) Citrulline

defective collagen crosslinking [9]. Thromboembolic events are due to endothelial damage and hypercoagulability, which arise by a variety of mechanisms [9,34]. The focal neurologic deficits and seizures are due to cerebral infarcts, while mental restriction and psychiatric abnormalities are probably due to a poorly understood neurotoxic action of homocysteine or to the cumulative effects of multiple ischemic lesions [9].

Treatment Treatment for B6-responsive patients consists of large doses of vitamin B6. Patients not responding to B6 are treated with a low protein and low methionine diet supplemented with folate, vitamin B12 and betaine, which promote conversion of homocysteine to methionine, thus reducing homocysteine levels.

Urea cycle disorders Definition and chemistry The urea cycle disorders are a group of inherited disorders characterized by neonatal hyperammonemia and encephalopathy. The urea cycle is illustrated in Figure 33.4. The first two reactions occur in the mitochondria and the argininosuccinate step overlaps with the tricarboxylic acid cycle. Urea synthesis occurs virtually exclusively in the liver. In addition to ridding the body of nitrogenous products, the urea cycle plays a key role in the synthesis and degradation of arginine. So, in urea cycle disorders (except arginase deficiency), arginine becomes an essential amino acid. The urea cycle enzymes and their associated disorders are shown in Table 33.1. All enzymes are expressed in the liver and kidney and argininosuccinate synthetase, argininosuccinate lyase, and arginase are also expressed in the brain. Epidemiology and genetics The urea cycle disorders are among the most common inherited metabolic disorders. Their combined incidence has been reported to be 1 in 35 000 births and that of ornithine transcarbamylase (OTC) deficiency, the most common disorder in the

408

TCA CYCLE

Aspartate

Figure 33.4 The urea cycle. 1: carbamoyl phosphate synthetase; 2: ornithine transcarbamylase; 3: argininosuccinate synthetase; 4: argininosuccinate lyase; 5: arginase. TCA, tricarboxylic acid.

Table 33.1 The urea cycle enzymes, associated disorders, and corresponding plasma citrulline levels. Enzyme and disorder

Plasma citrulline

Carbamoyl phosphate synthetase (CPS1) deficiency Ornithine transcarbamylase deficiency Argininosuccinate synthetase deficiency (citrullinemia) Argininosuccinate lyase argininosuccinase deficiency (argininosuccinic aciduria) Arginase deficiency (argininemia) Co-factor producing N-acetyl glutamate synthetase deficiency

Absent or trace Absent or trace Markedly elevated Moderately elevated Normal or reduced Absent or trace, similar to CPS1 deficiency

group, is 1 in 56 000 [1]. OTC deficiency is X-linked. All other enzyme deficiencies are autosomal recessive.

Clinical features Severe deficiencies of carbamoyl phosphate synthetase 1 (CPS1), OTC, argininosuccinate synthase, and argininosuccinate lyase present with a syndrome of neonatal hyperammonemia characterized by hypotonia, irritability, poor feeding, hyperventilation, seizures, lethargy, coma, and often death [1,8,13,35]. Surviving patients, even if promptly diagnosed and treated, have intellectual disability. Patients with milder enzyme deficiencies are often cognitively impaired and are beset by recurrent episodes of hyperammonemia characterized by nausea, vomiting, lethargy, seizures, and other nonspecific (neurological manifestations) signs and symptoms. These episodes, which may be mistaken for the Reye syndrome, appear at any time in life [36], and are triggered by a change in formula, high protein intake, infections, the postpartum state, or other metabolic stress. Some female OTC carriers are asymptomatic and others have an aversion to protein, cyclic vomiting, lethargy, ataxia, and seizures. The severity of these symptoms depends on the proportion of hepatocytes that carry the mutant gene on the active X chromosome. Rarely, female OTC carriers have a severe phenotype, similar to that of males [6,13,37]. Valproate may precipitate hyperammonemia

Disorders of Amino Acid Metabolism and Canavan Disease Chapter 33

in patients and previously asymptomatic carriers. Arginase deficiency has an indolent nonspecific onset and causes progressive spastic tetraplegia beginning in the lower extremities, seizures, and intellectual disability [1,8,13]. Hyperammonemia also occurs in organic acidemias because the accumulating organic acids are toxic to mitochondrial enzymes of urea synthesis. Transient severe hyperammonemia with seizures, respiratory distress, and coma occurs within 24 hours in preterm infants and is more frequent than genetic urea cycle defects [38]. This disorder is probably due to immaturity of urea cycle enzymes.

Laboratory findings The laboratory findings in the urea cycle disorders are marked hyperammonemia, respiratory alkalosis, elevated plasma glutamine, and decreased urea nitrogen [1,8,12,13]. The first step in distinguishing the urea cycle disorders from one another and differentiating them from other causes of hyperammonemia is the determination of plasma citrulline levels (Table 33.1). Citrulline, the product of CPS1 and OTC, is low or undetectable in their deficiency and normal in the organic acidemias. Urinary orotic acid, which results from diversion of excess carbamoyl phosphate to pyrimidine synthesis, is elevated in OTC deficiency and normal or low in CPS1 deficiency. Arginase deficiency is characterized by modest hyperammonemia (three to four times normal), marked elevation of arginine, and orotic aciduria. Molecular genetic testing and assays of enzyme activity can also be used for diagnosis. Pathological findings Patients dying in the newborn period show mainly brain swelling and Alzheimer type II astrocytes [6,12,15]. Additional findings such as symmetrical cortical infarcts, severe cortical atrophy, neuronal loss, gliosis, and calcification of the basal ganglia, spongy myelinopathy, poor myelination, ventriculomegaly, and atrophy of the cerebellar granular layer have been reported (Figure 33.5) [6,12,15,39]. These changes are most likely the result of hypoxic and ischemic insults. Bilateral symmetric old cystic infarcts were reported in a newborn with OTC deficiency [40] and cerebellar heterotopias have been described in two cases [37]. These reports suggest that the brain, in some OTC deficiency cases, is damaged in utero. Portal fibrosis and abnormal hepatic mitochondria have been reported in argininosuccinate lyase deficiency [1]. Mitochondrial abnormalities have also been described in patients with CPS and OTC deficiency when ammonia is elevated [1]. Pathogenesis Brain damage in the urea cycle disorders is due to hyperammonemia. Hyperammonemia alters amino acid metabolism, neurotransmitter patterns, cerebral energy metabolism, and other processes, with wide-ranging effects, especially on the developing nervous system [41]. Excess ammonia is taken up

Figure 33.5 Ornithine carbamyl transferase deficiency. Walnut brain with cystic and knife-edged gyri (image kindly provided by Dr. B. Harding).

by astrocytes and converted to glutamine by glutamine synthetase. Osmotic attraction of water by glutamine causes astrocytic swelling and cerebral edema [1,13,42]. The encephalopathy of urea cycle disorders is due to the combined effects of increased intracranial pressure and impairment of the metabolic function of astrocytes. Encephalopathy, astrocytic swelling, and cerebral edema have been reproduced in primates with experimental hyperammonemia [43]. Some ammonia is also converted to glutamate with the expected excitotoxic effects on neurons and oligodendroglia.

Treatment The most urgent objective is to lower plasma ammonia using hemodialysis or extracorporeal membrane oxygenation, when available. Intravenous sodium benzoate and sodium phenylacetate is used to scavenge ammonia. These agents are combined with L-arginine to prevent arginine deficiency. Protein intake is initially restricted. After plasma ammonia is lowered, enteral or intravenous feeding with glucose, fats, and amino acids resumes. Treatment is also tailored to the specific defect.

Maple syrup urine disease Definition and chemistry Maple syrup urine disease (MSUD) encompasses a group of inherited disorders characterized by neonatal encephalopathy and acidosis and accumulation of branched chain amino acids (BCAAs) and their ketoacids. One of these, methylvaleric acid, imparts a characteristic odor to the urine and skin. The first step in the catabolism of the essential BCAAs valine, leucine, and isoleucine is removal of the amino group, in the cytosol, by the BCAA amino transferase. The branched chain ketoacids thus derived, are transported into the mitochondria where

409

Developmental Neuropathology they are oxidized by a thiamine-dependent, intramitochondrial, multi-enzyme complex, the branched chain ketoacid dehydrogenase (BCKAD). Deficiency of this enzyme complex causes MSUD.

and myelination. The abnormality of myelin is probably due to defective biosynthesis of lipids and proteins resulting from amino acid imbalance [45]. There is evidence that oxidative stress also contributes to brain damage in MSUD [46].

Epidemiology and genetics The incidence of MSUD has been estimated to be 1 in 185 000 to 1 in 220 000 [10]. The BCKAD complex has four subunits: E1α and E1β , E2 , and E3 . Homozygous mutations of any of the four subunits causes MSUD.

Treatment Treatment of MSUD should be left in the hands of metabolic experts. Special BCAA-free diets are used for long term control of plasma BCAA concentration. Prompt supportive treatment during infection, fever, and other conditions that induce metabolic stress combined with intravenous BCAA-free amino acids, glucose, and insulin can prevent or reverse encephalopathy. The enzymes that are involved in BCAA metabolism are primarily active in the liver. Cure of MSUD can be achieved with liver transplantation.

Clinical features There are four clinical types of MSUD, classic, intermediate, intermittent, and thiamine-responsive [8,10,13,35]. Patients with classic MSUD have absent or minimal BCKAD activity. In patients with other forms, BCKAD activity is partially preserved. The classic infantile form accounts for 75% of cases and these patients are normal at birth, but within a few days, develop severe acidosis, hypoglycemia, poor feeding, vomiting, lethargy, bulging fontanelles, opisthotonos, spasticity, ophthalmoplegia, and seizures [8,10,13]. Maple syrup odor is detected in cerumen soon after birth and in the urine within a few days. Untreated, most patients lapse into coma and die during the newborn period [8,10,13]. Survivors have severe neurological damage. Patients with milder forms of MSUD are cognitively impaired and have intermittent ataxia, seizures, and acidosis triggered by excessive protein intake or infections. Imaging studies of classic MSUD show severe diffuse cerebral edema, most pronounced in the cerebellum and brainstem tegmentum. Edema intensifies in the first few weeks of life and then subsides leaving periventricular white matter hypodensity consistent with hypomyelination. Laboratory findings A biochemical diagnosis can be made by finding increased levels of leucine and alloisoleucine in plasma, cerebrospinal fluid, or urine by quantitative amino acid analysis. BCKAD activity can be measured in cultured fibroblasts. Molecular genetic testing of the E1α , E1β , and E2 subunits is also available. Pathological findings The main neuropathological abnormality in patients who die of severe neonatal MSUD or during attacks of MSUD later in life, is cerebral edema [13,15]. The pathogenesis of cerebral edema is not clear. Patients surviving acute MSUD episodes have spongy myelinopathy, deficient myelination, reduced oligodendrocytes, and variable gliosis. Myelin breakdown or macrophage activity is not seen. Defective dendritic development of the cerebral cortex, and neuronal loss in the substantia nigra and pontine nuclei have also been reported [44]. Cerebral atrophy is also seen. Pathogenesis The elevated BCAAs and their ketoacids, especially leucine and its ketoacid alpha-ketoisocaproic acid, are toxic and impair neurotransmitter synthesis, energy metabolism, protein synthesis,

410

Propionic and methylmalonic acidemia Definition and chemistry Propionic and methylmalonic acidemias are inherited disorders of propionic acid catabolism characterized by neonatal metabolic ketoacidosis. Fifty percent of propionate in blood is derived from catabolism of isoleucine, methionine, threonine, and valine, and the rest from β-oxidation of odd chain fatty acids and from bacterial activity in the gut [4,7]. Propionate is converted to methylmalonate in hepatic mitochondria by propionyl CoA carboxylase. Then, methylmalonyl CoA mutase converts methylmalonyl CoA to succinyl CoA which enters the tricarboxylic acid cycle. Propionic acidemia results from deficiency of the biotin-dependent propionyl CoA carboxylase, which has two subunits, PCCA, encoded by a gene on 13q32.3 and PCCB, encoded by a gene on 3q22.3, and methylmalonic acidemia from deficiency of the vitamin B12-dependent methylmalonyl CoA mutase, encoded by the MUT gene on 6p12.3. Methylmalonic acidemia is also caused by abnormalities of cobalamin synthesis that affect adenosylcobalamin, a co-factor of methylmalonyl CoA mutase, and methylmalonyl CoA epimerase. Epidemiology and genetics Propionic and methylmalonic acidemia account for 40% of all organic acidemias [13]. Propionyl CoA carboxylase consists of two alpha subunits coded by a gene on chromosome 13 and two beta subunits coded by a gene on chromosome 3. Mutations of four genes are known to cause methylmalonic academia. The proteins encoded by these genes reside in mitochondria. The gene for methylmalonyl CoA mutase is located on 6p21. Clinical features and imaging Most patients with propionic acidemia and methylmalonic acidemia present in the newborn period with the syndrome of ketotic hyperglycinemia [13,45]. This syndrome consists of severe or fatal metabolic ketoacidosis, vomiting, hypotonia,

Disorders of Amino Acid Metabolism and Canavan Disease Chapter 33

Figure 33.6 Methylmalonic academia. Hemorrhage in the striatum (image kindly provided by Dr. B. Harding).

lethargy, seizures, hyperglycinemia, hyperammonemia, hypoglycemia, thrombocytopenia, and neutropenia. Lethal cerebellar hemorrhage has been reported in association with thrombocytopenia (Figure 33.6) [47,48]. Survivors have cognitive impairment, chorea, and dystonia. Relapses of ketoacidosis, later in life, are triggered by high protein intake and infections [49]. Imaging studies reveal basal ganglia abnormalities in both disorders [13,47,50–52]. Cardiomyopathy is a complication of propionic acidemia. Pancreatitis and optic atrophy have been reported with both. There are milder forms with later onset that depend on residual enzyme activity and other factors.

Figure 33.7 Propionic acidemia. Spongy change in the cerebellum. Hematoxylin and eosin, × 500. The patient had similar changes in the cerebral cortex as well as spongy myelinopathy.

loss of germinal matrix cells and cerebellar external granular neurons have been reported in methylmalonic academia [54]. A striking spongy change of the cerebral and cerebellar cortex (Figure 33.7) was present in a personally examined patient with propionic acidemia who died at four months of age. This patient also had spongy myelinopathy, diffuse loss of white matter mass with enlarged ventricles (Figure 33.8), unilateral anophthalmia, and postaxial polydactyly. A peculiar cerebral angiopathy (intimal thickening, adventitial proliferation, fibrin exudation) and ischemic infarcts were present in a personally examined infant with methylmalonic acidemia (Figure 33.9).

Laboratory findings The diagnosis of propionic and methylmalonic acidemia is considered on the basis of an abnormal urine organic acid screen along with the findings of ketotic hyperglycinemia and can be confirmed by plasma acylcarnitine and urine acylglycine profiles. The gas chromatographic peak caused by propionic acid can be misidentified as ethylene glycol. This error has led to a wrongful conviction of a woman who was wrongly accused of murdering her son with antifreeze [53]. Molecular genetic testing is also available. Pathological findings The neuropathological changes of propionic acidemia and methylmalonic acidemia are similar and consist primarily of spongy myelinopathy involving actively myelinating tracts [6,13,15]. Spinal roots were involved in one case of propionic acidemia [17]. There is no myelin breakdown, gliosis, or myelin deficiency. Neuronal loss and gliosis in the basal ganglia, probably due to hypoxic and ischemic injury, has been reported in propionic academia [47,51]. Basal ganglia hemorrhages [6] and

Figure 33.8 Propionic academia. Loss of white matter and atrophy of the corpus callosum.

411

Developmental Neuropathology aspartoacylase, which hydrolyzes N-acetyl L-aspartate (NAA) into aspartate and acetate in the brain. Deficiency of aspartoacylase results in buildup of NAA in neurons and oligodendroglial precursors, which leads to neuronal dysfunction and myelin deficiency. ASPA mutations are most common in Ashkenazi Jews, among whom the carrier rate is 1 in 40 to 1 in 58 but have also been detected in many other ethnic groups. Two mutations account for the vast majority of Canavan disease in Ashkenazi Jewish patients.

Figure 33.9 Methylmalonic acidemia. Angiopathy with endothelial proliferation and fibrin exudation. The changes suggest endothelial injury and increased vascular permeability. Hematoxylin and eosin, × 500.

That patient had a combined defect of methylcobalamin and adenosylcobalamin synthesis. The angiopathy was probably caused by elevated homocysteine.

Pathogenesis The wide-ranging manifestations of the ketotic hyperglycinemia syndrome have been attributed to acidosis and the toxic effects of accumulating amino acids and their intermediates. These effects include bone marrow suppression, inhibition of mitochondrial enzymes of urea synthesis and glycine cleavage, and impairment of multiple other metabolic pathways [4,7]. Treatment The treatment of propionic and methylmalonic academia consists of dietary restriction of protein and propiogenic amino acids, carnitine to prevent secondary carnitine deficiency, and fluids and electrolytes to manage episodes of ketoacidosis. Methylmalonyl CoA mutase is active in the liver, and liver transplantation can cure methylmalonic acidemia. Forms of methylmalonic acidemia caused by abnormalities of cobalamin synthesis are treated with vitamin B12.

Canavan disease (spongy leukodystrophy) Definition Canavan disease is a genetic leukodystrophy characterized by spongy degeneration of the white matter. It was described by Canavan in 1931 and identified as a distinct disease entity by van Bogaert in 1949 [55]. Epidemiology and genetics Canavan disease is autosomal recessive and is caused by mutations of ASPA, on 17p13.2. ASPA encodes the enzyme

412

Clinical features, diagnosis, and imaging findings Neonatal/infantile Canavan disease, which is the most common form, presents usually in the first few months of life with macrocephaly, lack of head control, hypotonia (and later spasticity,) developmental delay, and seizures. Neurological development is arrested and patients die in their teens. The less frequent mild/juvenile variant presents at four to five years with mild delay in development of verbal and motor skills that may be overlooked. Patients with mild/juvenile Canavan disease may have a normal lifespan. There are rare severe congenital cases that present with lethargy, irritability, hypotonia, and sucking and swallowing difficulty in the first few days of life and cause death in a few weeks [56–58]. The diagnosis of Canavan disease can be made by measuring NAA levels in urine and by molecular genetic testing [59]. Carrier testing is available for individuals of Ashkenazi Jewish descent. Head computed tomography shows diffuse white matter hypodensity. Brain MRI shows diffuse, symmetrical white matter abnormality with T1-hypointensity and T2-hyperintensity, most severe in the subcortical areas. Magnetic resonance spectroscopy shows marked elevation of NAA in the brain, which is unlike other neurodegenerative disorders in which NAA is generally decreased [33]. Pathology In early stages, the brain is significantly larger than normal and there is no myelin (Figure 33.10). As the disease advances, brain weight is reduced and white matter atrophy develops. On microscopic examination, numerous vacuoles are seen in the deeper cortical layers, cortex–white matter junction, and white matter, imparting a spongy appearance to the affected areas, which gave the disease its name (Figure 33.11). Unlike most other leukodystrophies, the subcortical U-fibers are affected more severely than the deep white matter. Spongy change also involves the corpus callosum, internal capsule, and cerebellum, especially the plane between the molecular and granular layers. Canavan disease is similar to the spongy myelinopathy seen in amino acid and organic acid disorders. The vacuoles appear to develop between myelin lamellae. Oligodendroglial cells and axons are preserved. In advanced Canavan disease, myelin is absent. There is no macrophage reaction. Fibrous gliosis is not present but there are Alzheimer type 2 astrocytes in the cortex, where electron microscopy shows these swollen astrocytes

Disorders of Amino Acid Metabolism and Canavan Disease Chapter 33

to have markedly elongated mitochondria. There is no pathology in peripheral nerves.

Pathogenesis NAA is synthesized by neurons, transported to the extracellular space, and taken up by oligodendroglial precursors. It is a marker of brain metabolic activity but its function in myelin formation and brain development is unknown. It may serve as a water pump, and its accumulation may lead to astrocytic swelling. Alternatively, hydrolyzation of NAA accumulated in oligodendroglial cells may impair myelin synthesis [59]. Treatment There is no treatment for Canavan disease other than nutritional support, physical therapy, antiepileptic drugs, and other supportive measures. Figure 33.10 Canavan disease. Most of the white matter in this field has a gray appearance, indicating lack of myelin (reproduced from Neuropathology, D.P. Agamanolis [60], with permission of the author).

Figure 33.11 Canavan disease. The top portion of the picture is cortex and the bottom white matter. Spongy changes are present in both (reproduced from Neuropathology, D.P. Agamanolis [60], with permission of the author).

References 1. Ah Mew N, Lanpher BC, Gropman A et al. (2003 updated 2017) Urea cycle disorders overview. In: RA Pagon, MP Adam, HH Ardinger, et al., eds, GeneReviews® . Seattle, WA, University of Washington. Available from: http://www.ncbi.nlm.nih .gov/books/NBK1217 (accessed November 6, 2017) 2. Blau N, van Spronsen FJ, Levy HL (2010) Phenylketonuria. Lancet 376:1417–27 3. Blau N, Shen N, Carducci C (2014) Molecular genetics and diagnosis of phenylketonuria: state of the art. Expert Rev Mol Diagn 14:655–71 4. Carrillo-Carrasco N, Venditti C (2012 updated 2016) Propionic acidemia. In: RA Pagon, MP Adam, HH Ardinger et al., eds, GeneReviews® . Seattle, WA, University of Washington, Seattle. Available from: https://www.ncbi.nlm.nih.gov/books/NBK92946 (accessed November 6, 2017) 5. Gropman AL (2012) Patterns of brain injury in inborn errors of metabolism. Pediatr Neurol 19:203–10 6. Jacques TS, Harding BN (2015) Metabolic and neurodegenerative diseases of childhood. In: S Love, H Budka, JW Ironside, A Perry, eds, Greenfield’s Neuropathology, 9th ed., Boca Raton, FL, CRC Press. pp. 399–438 7. Manoli I, Venditti CP (2005 updated 2016) Methylmalonic acidemia. In: RA Pagon, MP Adam, HH Ardinger et al., eds, GeneReviews® . Seattle, WA, University of Washington. Available from: http://www.ncbi.nlm.nih.gov/books/NBK1231 (accessed November 6, 2017) 8. Menkes JH (2006) Metabolic diseases of the nervous system. In: JH Menkes, HB Sarnat, eds, Child Neurology, 7th ed. Philadelphia, PA, Lippincott, Williams and Wilkins. pp. 29–141 9. Picker JD, Levy HL (2004 updated 2017) Homocystinuria caused by cystathionine beta-synthase deficiency. In: RA Pagon, MP Adam, HH Ardinger et al., eds, GeneReviews® . Seattle, WA, University of Washington. Available from: http://www.ncbi.nlm.nih .gov/books/NBK1524 (accessed November 6, 2017) 10. Strauss KA, Puffenberger EG, Morton DH (2006 updated 2013) Maple syrup urine disease. In: RA Pagon, MP Adam, HH Ardinger

413

Developmental Neuropathology

11.

12. 13.

14.

15. 16.

17.

18.

19.

20.

21.

22.

23. 24.

25.

26.

27.

414

et al., eds, GeneReviews® . Seattle, WA, University of Washington. Available from: http://www.ncbi.nlm.nih.gov/books/NBK1319 (accessed November 6, 2017) Van Hove J, Coughlin C II, Scharer G (2002 updated 2013) Glycine encephalopathy. In: RA Pagon, MP Adam, HH Ardinger et al., eds, GeneReviews® . Seattle, WA, University of Washington. Available from: http://www.ncbi.nlm.nih.gov/books/NBK1357 (accessed November 6, 2017) Volpe JJ (2008) Neurology of the Newborn 5th ed., Philadelphia, WB Saunders-Elsevier, pp. 652, 686 Wilcox SI, Sederbaum SD (2007) Amino acid metabolism. In: DL Rimoin, JM Connor, RE Pyeritz, BR Corf, eds, Emery and Rimoin’s Principles and Practice of Medical Genetics, 5th ed., London, Churchill Livingstone-Elsevier. pp. 2159–90 Van Karnebeek CDM, Stockler S (2012) Treatable inborn errors of metabolism causing intellectual disability: A systematic literature review. Mol Genet Metab 105:368–81 Friede RL (1989) Developmental Neuropathology, 2nd ed., New York, Springer-Verlag, p. 524 Martin JJ, Schlote W (1972) Central nervous system lesions in disorders of amino-acid metabolism: a neuropathological study. J Neurol Sci 15:49–76 Shuman RM, Leech RW, Scott CR (1978) The neuropathology of the nonketotic and ketotic hyperglycinemias: three cases. Neurology 28:139–46 Valk J, van der Knaap MS (2005) Magnetic resonance of myelin, myelination, and myelin disorders. In: Valk J, van der Knaap MS (eds) Nonketotic Hyperglycinemia. Springer, Berlin, pp. 306–10 Agamanolis DP, Potter JL, Herrick MK et al. (1982) The neuropathology of glycine encephalopathy: a report of five cases with immunohistochemical and ultrastructural observations. Neurology 32:975–85 Brun A, Borjeson M, Hultberg B (1979) Neonatal nonketotic hyperglycinemia: a clinical biochemical and neuropathological study including electron microscopic findings. Neuropaediatrie 10:195– 205 Kanekar S, Byler D (2013) Characteristic MRI findings in neonatal nonketotic hyperglycinemia due to sequence changes in GLDC gene encoding the enzyme glycine decarboxylase. Metab Brain Dis 28:717–20 Agamanolis DP, Potter JL, Lundgren DW (1993) Neonatal glycine encephalopathy: biochemical and neuropathologic findings. Pediatr Neurol 9:140–3 Guerrini R, Dobyns WB (2014) Malformations of cortical development: clinical features and genetic causes. Lancet Neurol 13:710–26 Stutterd CA, Leventer RJ (2014) Polymicrogyria: a common and heterogeneous malformation of cortical development. Am J Med Genet C Semin Med Genet 166C:227–39 Prasad AN, Bunzeluk K, Prasad C, Chodirker BN, Magnus KG, Greenberg CR (2007) Agenesis of the corpus callosum and cerebral anomalies in inborn errors of metabolism. Congenital Anomalies 47:125–135 DiSilvestre D, Koch R, Groffen J (1991) Different clinical manifestations of hyperphenylalaninemia in three siblings with identical phenylalanine hydroxylase genes. Am J Hum Genet 48:1014–16 Camp CM, Parisi MA, Acosta, PB et al. (2014) Phenylketonuria Scientific Review Conference: State of the science and future research needs. Mol Genet Metab 112: 87–122

28. Levy H, Lobbregt D, Barnes PD, Poussaint TY (1996) Maternal phenylketonuria: magnetic resonance imaging of the brain in offspring. J Pediatr 128:770–5 29. Crome L (1962) The association of phenylketonuria with leucodystrophy. J Neurol Neurosurg Psychiatr 25:149–53 30. VanHove JL, Kishnani PS, Demaerel P et al. (2000) Acute hydrocephalus in nonketotic hyperglycinemia. Neurology 54:754–6 31. Luder AL, Davidson A, Goodman SI, Greene CL (1989) Transient nonketotic hyperglycinemia in neonates. J Pediatr 114:1013–15 32. Dobyns WB (1989) Agenesis of the corpus callosum and gyral malformations are frequent manifestations of nonketotic hyperglycemia. Neurology 39:817–20 33. Perlman SJ, Mar S (2012) Leukodystrophies. Adv Exp Med Biol 724:154–71 34. Welch GN, Loscalzo J (1998) Homocysteine and atherothrombosis. N Engl J Med 338:1042–50 35. Hoffmann GF, K¨olker S (2013) Defects in amino acid catabolism and the urea cycle. Handb Clin Neurol 113:1755–73 36. Ahrens–Nicklas RC, Slap G, Ficicioglu C (2015) Adolescent presentations of inborn errors of metabolism. J Adolesc Health 56: 477–82 37. Harding BN, Leonard JV, Erdohazi M (1984) Ornithine carbamoyl transferase deficiency: a neuropathological study. Eur J Pediatr 141:215–20 38. Ballard RA, Vinocur B, Reynolds JW (1978) Transient hyperammonemia of the preterm infant. N Engl J Med 299:920–5 39. Solitare GB, Shih VE, Nelligan DJ et al. (1969) Argininosuccinic aciduria: clinical, biochemical, anatomical and neuropathological observations. J Ment Defic Res 13:153–70 40. Filloux F, Townsend JJ, Leonard C (1986) Ornithine transcarbamylase deficiency: neuropathologic changes acquired in utero. J Pediatr 108:943–5 41. Braissant O, McLin VA, Cudalbu C (2012) Ammonia toxicity to the brain. J Inherit Metab Dis 36:595–612 42. Hawkins RA, Jessy J (1991) Hyperammonaemia does not impair brain function in the absence of net glutamine synthesis. Biochem J 277:697–703 43. Voorhies TM, Ehrlich MC, Duffy TE et al. (1983) Acute hyperammonemia in the young primate: physiologic and neuropathologic correlates. Pediatr Res 17:97–100 44. Kamei A, Takashima S, Chan F et al. (1992) Abnormal dendritic development in maple syrup urine disease. Pediatr Neurol 8:145–7 45. Menkes JH (1967) The pathogenesis of mental retardation in phenylketonuria and other inborn errors of amino-acid metabolism. Pediatrics 39:297–308 46. Sitta A, Ribas GS, Mescka CP et al. (2014) Neurological damage in MSUD: the role of oxidative stress. Cell Mol Neurobiol 34:157–65 47. Hamilton RL, Haas RH, Nyhan WL et al. (1995) Neuropathology of propionic acidemia: a report of two patients with basal ganglia lesions. J Child Neurol 10:25–30 48. Dave P, Curless RG, Steinman L (1984) Cerebellar hemorrhage complicating methylmalonic and propionic acidemia. Arch Neurol 41:1293–6 49. Surtees RAH, Matthews EE, Leonard JV (1992) Neurologic outcome of propionic acidemia. Pediatr Neurol 8:333–7 50. Brismar J, Ozand PT (1994) CT and MR of the brain in disorders of the propionate and methylmalonate metabolism. Am J Neuroradiol 15:1459–73

Disorders of Amino Acid Metabolism and Canavan Disease Chapter 33 51. Harding BN, Leonard JV, Erdohazi M (1991) Propionic acidemia: a neuropathological study of two patients presenting in infancy. Neuropathol Appl Neurobiol 17:133–8 52. Perez-Cerda C, Merinero B, Marti M et al. (1988) An unusual late– onset case of propionic acidaemia: biochemical investigations, neuroradiological findings and mutation analysis. Eur J Pediatr 157: 50–2 53. Hoffman M (1991) Scientific sleuths solve a murder mystery. Science 254:931 54. Sum JM, Twiss JL, Horoupian DS (1993) Selective death of immature neurons in methylmalonic acidemia of the neonate: a case report. Acta Neuropathol. 85:217–21 55. Banker BQ, Robertson JT, Victor M (1964) Spongy degeneration of the central nervous system in infancy. Neurology 14:981– 1001

56. Rodriguez D (2013) Leukodystrophies with astrocytic dysfunction. Handb Clin Neurol 113:1619–28 57. Traeger EC, Rapin I (1998) The clinical course of Canavan disease. Pediatr Neurol 18:207–12 58. Matalon R, Michals-Matalon K (1999 updated 2011) Canavan disease. In: RA Pagon, MP Adam, HH Ardinger et al., eds, GeneReviews® . Seattle, WA, University of Washington. Available from: http://www.ncbi.nlm.nih.gov/books/NBK1234 (accessed November 6, 2017) 59. Hoshino H, Kubota M (2014) Canavan disease: clinical features and recent advances in research. Pediatr Int 56:477–83 60. Agamanolis DP (2011) Neuropathology: An Illustrated Interactive Course for Medical Students and Residents. Northeast Ohio Medical University. Available at: http://neuropathology-web.org/ index.html (accessed November 6, 2017)

415

34

Pelizaeus–Merzbacher Disease Brian N. Harding Department of Pathology and Laboratory Medicine, Children’s Hospital of Philadelphia, Philadelphia, PA, USA

Definition

Genetics

Pelizaeus–Merzbacher disease (PMD) is an X-linked dysmyelinating disorder of the central nervous system (CNS) caused by mutations in the proteolipid protein gene (PLP1) gene located at Xq22. The milder disorder X-linked spastic paraplegia type 2 (SPG2) is allelic, although PMD and SPG2 are listed separately as Mendelian Inheritance in Man (MIM) numbers 31208 and 312920, respectively.

PMD/SPG2 represents a spectrum of CNS myelin disorders caused by dosage effects and mutations of the proteolipid protein (PLP1) gene (Figure 34.1). Increased dosage of PLP1 is the major cause of PMD, with approximately 60–70% of patients having a submicroscopic duplication of chromosome Xq22 including the entire PLP1 gene [5,6]. A few patients have a deletion of PLP1 [7] or a null mutation that gives rise to loss of protein [8,9]. Approximately 15–20% of cases have PLP1 sequence changes that produce abnormal PLP1/DM20 proteins. These are mainly missense point mutations within the gene but may also include nonsense or small deletions and insertions that cause frameshifts (Figure 34.2) [10]. An up-to-date list of the mutation spectrum can be found in Hobson and Kamholz [11]. Splice-site [12] or noncoding region mutations [13] have also been described that cause abnormal expression of PLP1. The remaining 10– 20% of patients have clinical features of PMD but do not have a detectable abnormality, suggesting that mutations in regulatory regions or other gene loci can cause PMD.

Historical annotation Pelizaeus gave the first clinical description of PMD in 1855 [1], to which Merzbacher added further clinical details from the same family and a neuropathological report in 1910 [2]. Postmortem examination also confirmed the disorder in an affected sister [3].

Epidemiology, sex distribution PMD is rare: in Germany amounting to only 6.5% of all leukodystrophies [4]. It principally affects hemizygous males, while females are generally asymptomatic carriers. However, the rate of occurrence in females exceeds Duchenne muscular dystrophy, and is more common in families with milder forms of PMD or SPG2. Manifesting females with a molecular defect in the PLP1 gene mainly have point mutations, there have been only two female patients with PLP1 gene duplications described to date.

The PLP1 gene PLP1 is localized to Xq22 [14] and encodes the proteolipid protein (PLP1) and its smaller alternatively spliced isoform DM20 [15]. PLP1 is the major myelin component in the CNS, constituting approximately 50% of total protein. PLP1 and DM20 are hydrophobic proteins of 276 and 241 amino acids respectively and they are highly conserved across species, being identical in man, mouse and rat. Predicted protein structures include four transmembrane domains with cytoplasmic amino and carboxyl terminals (Figure 34.2) [16]. PLP1 is composed of 7 exons

Developmental Neuropathology, Second Edition. Edited by Homa Adle-Biassette, Brian N. Harding, and Jeffrey A. Golden. © 2018 John Wiley & Sons Ltd. Published 2018 by John Wiley & Sons Ltd.

417

Developmental Neuropathology

PLP1 duplication PLP1 mutation PLP1 deletion/null

extending over approximately 17 kb. Exon 3 contains an internal splice donor site that creates an alternatively spliced transcript DM20 [17] and leads to an internal deletion of 35 amino acids. The expression of PLP1 and DM20 is spatially and temporally regulated during development [18] with expression mainly within oligodendrocytes, the myelinating cells of the CNS, but also at a lower level in Schwann cells within the peripheral nervous system and heart, spleen, thymus, and lymph nodes.

Unknown

Figure 34.1 Genetic mechanisms for disease in patients with Pelizaeus– Merzbacher disease/X-linked spastic paraplegia type 2. The majority of patients, approximatively 60–70%, have a PLP1 gene duplication; approximatively 15–20% have a mutation in the PLP1 gene; a minority have a loss of PLP1 caused by a deletion or a null allele; and approximatively 10–20% have a molecular defect that is currently unknown.

Deletion Frameshift

PLP specific region Silent mutation

Detection and characterization of PLP1 duplications Since duplications are the most frequent cause of PMD, duplication analysis should be performed prior to mutation screening. Dosage detection techniques such as interphase fluorescence in situ hybridization (FISH) (Figure 34.3) [19,20], quantitative multiplex polymerase chain reaction [21,22] and Southern blotting [23] have been successfully used for diagnosis in the patient, the female carrier, and in some cases, prenatal detection. The techniques of multiplex amplifiable probe

Intraperiod line extracellular 195

Polymorphism

P S K T S A S I G S L 200 C F A N/H/V/E/G 45 Premature D A I S L L N S A L I E T R M C termination K R/P G Y S/V I T YC P P/I C V Y 186 E F C G K S K 223 50 Y 42 T Very Severe S G T F V D Q I P S K 208 A G C L H S/A/L F A Severe N E N T 215 P L T N 233 F Y W Moderate P Q L 181 T Q R P M 56 Mild WC 38 A T D V 211 T S T Y E F No clinical description H 239 V/E 58 E C N 178 36 H Y/R Y L L F I F I G V/L Y A P C A I H I V N EG V F L F C G V P V Y C A A F T E F A V A Q F A A A R Y S C A R G F T P G Y I V L S V L G L C F L L V F T A R L T T A S F R W V V E G F M I V L T Y L F F A N Y T F I LP F A S T Y A L G A L A A A I L G V V L G 267 F A L C C C C C C L V E K G A R E I K L H V 86 G L 9 L K 150 Y R F T N/N D L M G Y G R E G P T G K T L M L 147 H Y G K T G L W 144 S S T Y G K D H A A G A F V T K 276 F R O PLP specific region 116 V I Q H Y T 127 Q G G G Cytoplasm R Q S K Major dense line G R G Exon junction

Figure 34.2 PLP1 topology showing four putative transmembrane domains, two extracellular loops and one cytoplasmic loop. The alternatively spliced region present in PLP1 and absent in DM20 and the known mutations are shown (figure kindly provided by Dr. James Garbern).

418

Pelizaeus–Merzbacher Disease Chapter 34 Normal metaphase chromosome and interphase nucleus with one copy of the PLP1 gene at Xq22

Figure 34.3 Representation of PLP1 gene duplication on chromosome X. Xp is shown with green shading and the PLP1 gene is shown in red. The typical tandem duplications are detected as a doublet by interphase nuclei fluorescence in situ hybridization (FISH) using a genomic probe containing the PLP1 gene. Atypical cases where the additional copy of PLP1 is transposed to another site can be detected by metaphase FISH, as the extra copies have been reported within XCp22, Xq26 or Xp11,4 (associated with an inversion. Adapted from Woodward et al. [65]. Reproduced with permission of Future Medicine Ltd.

Typical PLP1 duplication. Tandem orientation in Xq22, not resolved on metaphase chromosome but visible as “doublet” in interphase nucleus

Atypical duplications with additional copies of PLP1 in Xp22, Xq26 and Xp11.4 respectively

hybridization [24] and multiplex ligation-dependent probe amplification [6] are currently being evaluated. Molecular analysis of the duplications in patients with PMD has revealed great variation in size (less than 300 kb to more than 4.6 Mb) and position of the breakpoints on either side of PLP1 [19,25]. Most duplications are arranged in a tandem orientation and have an intrachromosomal origin, suggesting that they probably arise during male meiosis [5,25,26]. The consequence of a grand paternal origin in spermatogenesis is that the mothers are carriers. Three atypical cases with noncontiguous duplication are of particular note, as the additional copy of the PLP1 gene integrated into different regions of the X chromosome (Xp22, Xp11.4 and Xq26) in an apparent transposition event occurring at a submicroscopic level (Figure 34.3) [27]. These unusual duplication events have been termed submicroscopic transposon cases. The advantage to using FISH for molecular genetic diagnosis of PMD is that it is a visual technique and these unusual cases where the PLP1 gene has duplicated and then moved to another site in the genome can easily be identified. The disadvantage is that small duplications, less than 50 kb, could be missed and, consequently, a combination of FISH and another higher-resolution dosage technique is ideal.

Clinical features including appropriate investigations Clinical features There is considerable variation, but common characteristics include nystagmus, stridor, ataxia, psychomotor developmental delay, spasticity and onset within the first year of life [8,10,28,29]. Disease severity ranges from severe connatal PMD through an intermediate classical form with a slowly progressive course, to

mild PMD/SPG2. Symptoms of connatal PMD develop shortly after birth, motor and intellectual development are severely delayed and patients often have seizures and lack head control. Small head circumference and optic atrophy are also common. Death occurs in early childhood to the third decade of life. Classical PMD is the most common form of the disease; symptoms usually present in the first year of life and patients often survive into their sixth decade. SPG2 manifests as progressive weakness and spasticity of the lower extremities with or without CNS involvement. SPG2 has a later onset of one to five years and a milder phenotype with patients often able to walk, talk and have a normal lifespan [30,31].

Imaging Computed tomography shows nonspecific changes, but magnetic resonance imaging demonstrates failure to myelinate in comparison with age-matched controls [32]. T1-weighted images lack the high signal observed in myelinating white matter, while high intensity persists in T2-weighted images due to lack of myelination and high-water content. Magnetic resonance spectroscopy has given conflicting results [33]. Laboratory findings Cerebrospinal fluid findings and routine biochemical examination are normal. Electroencephalographic changes [29] will assist in the differential diagnosis of PMD- elated seizures. Abnormalities in brainstem auditory evoked potential, abnormal waveform, or prolonged latency, are useful in early diagnosis if the only other indication is eye movement disorder or an affected sibling. Visual evoked potential may be abnormal in the face of a normal electroretinogram indicating malfunction of the visual pathways. Nerve conduction studies are usually normal, as the peripheral nervous system is mostly spared.

419

Developmental Neuropathology subcortical white matter, together with the commissures and fornix, are gray and sunken with a gelatinous or firm consistency. U-fibers may be patchily spared, and the centrum semiovale can show white streaks. The optic nerves and chiasm are thin and gray, but other cranial nerves are normal. The cerebellar white matter, brainstem and spinal cord tracts are shrunken and gray, in striking contrast to the normal plump, white cranial and spinal nerve roots.

Figure 34.4 Coronal section through the basal ganglia in a 15-year-old male with slowly progressive disease.

Exceptions include some patients with loss of PLP1 function that have a mild demyelinating peripheral neuropathy and slower nerve conduction velocities that are not uniformly distributed along the nerve [34].

Pathology Macroscopy In connatal cases with early death, the brain is normal in size and the white matter is unremarkable on coronal slicing. In the classic form, with prolonged survival, the brain is usually two-thirds the expected weight. On sectioning, the gray matter appears unremarkable, while ventricular dilatation depends on the degree of white matter loss (Figure 34.4). The central and (a)

(b)

Histopathology The cerebral cortex is usually not affected, but there are rare reports of polymicrogyria [35]. There is often cerebellar cortical degeneration with either predominant granule or Purkinje cell degeneration. With conventional myelin stains, the hemispheric white matter varies: in connatal cases with death in infancy there is almost complete lack of myelin, while in patients with a slower tempo of disease one encounters the classical tigroid or discontinuous pattern with preservation of myelin islets around blood vessels (Figure 34.5). Oligodendrocytes are markedly reduced or absent, especially where myelin is completely lacking. Axons are preserved (Figure 34.6). There is astrocytosis and fibrillary gliosis. In unmyelinated areas microglia are not increased, and there is usually very sparse sudanophilic lipid in perivascular macrophages. Cranial and spinal nerve roots that have a different myelin structural protein (PMP-22) are normally myelinated (Figure 34.7) and it may be possible to follow individual CNS unmyelinated axons in continuity through the root transition zone into the peripheral nervous system, where they appear normally myelinated. One report of the neuropathology of SPG2 demonstrated mild myelin loss in the centrum semiovale compared with severe loss in the spinal cord [36]. Immunohistochemical and ultrastructural findings Lack of PLP1 in PMD was first demonstrated immunohistochemically by Koeppen et al. [37]. Other myelin proteins (c)

Figure 34.5 Little stainable myelin remains (a) Loyez, (b, c) Luxol fast blue–cresyl violet. At higher magnification (c), residual perivascular myelin islets can be seen.

420

Pelizaeus–Merzbacher Disease Chapter 34 (a)

(b)

Figure 34.6 In demyelinated areas (a; Luxol fast blue–cresyl violet) there is relative preservation of axons (b, Glees silver method).

(myelin basic protein, myelin associated glycoprotein, 2′ ,3′ Cyclic nucleotide 3′ -phosphodiesterase) are variably reduced [8,37]. Ultrastructural data in humans are limited and predate discovery of the gene defect. Of two case reports from the same laboratory, one showed no compact myelin, the other regular myelin periodicity and a normal intraperiod line. This was confirmed in a personally examined biopsy [38] where, among many naked axons, there was some preservation of thin but compact myelin discontinuously between internodes, a feature that has been described in some animal models [39].

Biochemistry There is loss of galactolipids specific to the myelin sheath [40]. Loss of cerebroside is variable and nonspecific as a function of deficient myelination. Differential diagnosis PMD is readily distinguishable morphologically from those leukodystrophies with specific features such as globoid cells (Krabbe disease), metachromasia (metachromatic leukodystrophy), systemic involvement, and trilamellar inclusions (adrenoleukodystrophy), spongiosis and abnormal mitochondria (Canavan disease), massive cavitation (childhood ataxia with central hypomyelination), or pigmented macrophages (Nyssen–van Bogaert syndrome). Extensive mineralization is not a feature of PMD, as it is in Aicardi–Gouti`eres leukoencephalopathy, where there is also diffuse lack of myelin staining or fragmentation of myelin fibers. Discontinuous demyelination and calcification are also prominent findings in Cockayne syndrome, which is autosomal-recessive and demonstrates striking dysmorphology, microcephaly, and atrophy of the cortex brainstem and cerebellum.

Experimental models Many spontaneous point mutations have been described [41]. In mice, they include jimpy (Plpjp), myelin synthesis-deficient jp mouse (Plpjp-msd), jimpy-4 J mouse (Plpjp-4j), and rumpshaker (Plpjp-rsh). There is also the myelin deficient rat (Plpmd), the shaking pup (Plpsh), and the rabbit with paralytic tremor (Plppt). The jimpy mouse has a point mutation, while the others have missense mutations causing single amino-acid substitutions of the Plp gene. Several mirror the genetic defects found in clinical human disease. Transgenic mouse models having additional copies of the Plp gene have been generated and show myelination defects, astrogliosis, and seizures, indicating that precise regulation of Plp is required and modeling the common human mutation of duplications [42]. Figure 34.7 Myelin staining is completely absent from the spinal cord but well preserved in spinal roots (Luxol fast blue–cresyl violet).

Pathogenesis The molecular basis for the phenotype variability is not completely understood but probably reflects the distinct cellular

421

Developmental Neuropathology

Clinical severity Severe PMD

Classical PMD

Mild PMD/SPG2

PLP1 duplication Plp1 deletion or null PLP1 point mutation

Oligodendrocyte death Figure 34.8 Generalized association between Pelizaeus–Merzbacher disease (PMD)/X-linked spastic paraplegia type 2 (SPG2) disease severity and dosage effects and mutations of the PLP1 gene. Patients with PLP1 gene duplications mostly have a classical PMD phenotype but symptoms at the severe connatal and mild/SPG2 end of the spectrum are also possible. Patients with a point mutation in the PLP1 gene have a wide range of phenotypes but are often more severely affected than duplication patients. Patients with loss of PLP1 function caused by a deletion or a null mutation have mild PMD/SPG2 and peripheral nervous system involvement.

effects of the different genetic mechanisms involved. There is some general association with disease severity in that loss of PLP1/DM20 gives rise to mild disease [7–9]. Severe connatal PMD is mainly caused by missense point mutations in highly conserved regions of PLP1 [8] and duplications are most often found in patients with classical PMD having a moderate phenotype (Figure 34.8) [43,44]. However, heterogeneity in severity exists even within families [25], suggesting the influence of modifier genes and/or genetic background. Understanding how the different mutational mechanisms underlie pathogenesis is becoming clearer by comparing the patients with animal models.

Distinct cellular defects underlie the pathogenesis in each form of PMD PLP1 mutations Splice-site or noncoding PLP1 mutations can cause abnormal expression and a reduction in message and protein that has been reported to affect myelin stability and axonal integrity [45]. More commonly, the mutations are missense and the clinical phenotype may depend on the position of the altered amino acid. Mutations within the PLP1 specific region (exon 3B) that affect transport of PLP1 but not DM20 give rise to a less severe phenotype and do not usually cause oligodendrocyte death [31,46]. Aberrant or truncated proteins generated by PLP1 mutations are predicted to result in misfolded PLP1 that accumulates in the rough endoplasmic reticulum and fails to be transported to the oligodendrocyte cell membrane [47]. The misfolded proteins are often associated with premature oligodendrocyte death and may gain a novel function that is deleterious to the cell [46].

422

Activation of the unfolded protein response has been reported to modulate disease severity in oligodendrocytes expressing the mutant protein through a signaling cascade that coordinates the accumulation of mutant PLP1 in the endoplasmic reticulum with changes in gene expression, protein synthesis, and possibly apoptosis [48]. PLP1 deletion Patients with a deletion of PLP1 [7] or a loss of function mutation [8,9] have mild disease and widespread demyelination in the central and peripheral nervous systems that is not associated with the other mechanisms of disease. These patients demonstrate that, although PLP1 is expressed at low levels in the peripheral nervous system compared with the CNS, PLP1 or DM20 are required for peripheral myelin function. Lengthdependent axonal degeneration has been found in humans and in knockout mice with a functionally null Plp gene [49]. The degeneration is not associated with significant demyelination and is similar to that observed in the peripheral nervous system in patients with Charcot–Marie–Tooth disease type I who have an inherited demyelinating neuropathy. The disruption in PLP1-mediated axon-oligodendrocyte interactions is probably responsible for this axon degeneration [49]. The Plp mutant mice [50,51] do not develop classic signs of PLP1-related disease and have a normal number of oligodendrocytes [52]. The mice have only ultrastructural abnormalities that include swellings in small diameter axons (six to eight weeks) and late-onset axonal degeneration [15]. They demonstrate that neither PLP1 nor DM20 is necessary for myelin assembly but that they are both required for myelin and axon maintenance probably being needed for axon–glial interaction. PLP1 duplication Patients with increased dosage of PLP1 have a variable phenotype that can range from severe connatal [53] to mild PMD/SPG2 but typically have the classical form [42,44]. At least one severe case is probably due to the patient having a further additional copy of PLP1, as dosage analysis suggests a triplication rather than a duplication [26,53]. The range in clinical severities within the group of patients with PMD who have increased dosage of PLP1 may also be explained by the different duplication structures. This may cause the juxtaposition of different regulatory sequences, which could influence the expression of the PLP1 gene. The effect of modifier genes either within the duplicated region or elsewhere in the genome may also affect PLP1 expression and consequently disease pathogenesis. Three lines of transgenic mice with autosomal copies of wildtype murine Plp have been generated as animal models of PMD caused by PLP1 gene duplication [42,54]. These mice show hypomyelination and, generally, their age of onset and severity of disease is proportional to Plp gene dosage [55]. Mice with high Plp dosage have severe early onset dysmyelination, increased oligodendrocyte apoptosis and premature death. Pathogenesis is not understood but autophagic vacuoles have been observed

Pelizaeus–Merzbacher Disease Chapter 34

in the oligodendrocytes suggesting there is accelerated protein degradation. There is evidence of perturbations of protein trafficking but this may not be the same mechanism as that of the missense mutations. Overexpression of PLP1 has been reported to accumulate in the late endosome/lysosome with cholesterol and to be involved in aberrant trafficking and assembly of myelin components [56]. Such accumulation is suggested to interfere with myelination and reduce the viability of the oligodendrocyte [56]. In contrast, mice with lower transgene copy number have normal development with no clinical signs until later in life. Pathological features include late-onset demyelination and axonal swelling and degeneration, suggesting that the oligodendrocytes are unable to maintain their myelin sheaths. The pathology is similar to that of the Plp knockout mouse, indicating that changes in gene dosage, either increased or decreased, may be causing the axonal changes.

random pattern of X-inactivation [58]. Therefore, half of the cells in point mutation carriers have an active X chromosome with a PLP1 change that expresses the mutant phenotype. The severity of the mutation will determine cell fate. Only oligodendrocytes expressing a mild phenotype will be able to survive and this can lead to late-onset neurological problems [57,59,60]. There have been only two cases reported of affected females with PLP1 duplications [61]. Both unrelated girls had early onset mild PMD or SPG2 followed by progressive clinical improvement. They did not show skewed X-inactivation and their recovery was probably due to increased myelin production from the oligodendrocytes expressing the normal X chromosome with only one copy of PLP1.

Hypomyelinating leukodystrophies Expression of the disease in females As PMD is a recessive X-linked disorder it almost exclusively affects males. Small numbers of manifesting females have been described with PLP1 point mutations and these have been associated with milder disease in males [57]. This observation has been described by patterns of X-inactivation. Female carriers with a duplication have heavily skewed X-inactivation with the X chromosome bearing the duplication being preferentially inactivated. Consequently, these female carriers are mainly asymptomatic. In contrast, point mutation carriers show a

There are several reports of genetic disorders distinct from PMD, but with similar clinical and imaging features, and these are designated Pelizaeus–Merzbacher-like (PMLD) diseases (Table 34.1). The best characterized is PMLD1 or hypomyelinating leukodystrophy 2 [26,62] caused by mutations of the gamma 2 gap junction protein gene (MIM 6008894), which encodes a member of the connexin family of proteins. As with PMD, there is a milder allelic disorder, spastic paraplegia type 44. These autosomal-recessive disorders await pathologic examination.

Table 34.1 Hypomyelinating leukodystrophies. Disorder

Synonym

MIM Phenotype

Location

Gene

Protein

HLD1

Pelizaeus–Merzbacher disease/spastic paraplegia 2 Pelizaeus–Merzbacher-like disease 1/ spastic paraplegia 44

312080

Xp22.2

PLP1

PLP1

608804

1q42.13

GJC2/GJA12

Gamma-2 GAP junction protein

260600

4q24

AIMP1

612233 610532

2q33.1 7p15.3

HSPD1 FAM126A

612438

19p13.3

TUBBA4

Aminoacyl-tRNA synthetase complex interacting multifunctional protein 1 Heat shock protein 60 Family with sequence similarity 126, member A Beta tubulin class IVA

607694 614381 616140 616420 616494

10q22.3 12q23.3 5q34 1q42.2 6p21.1

POLR3A POLR3B RARS PYCR2 POLR1C

RNA polymerase III, subunit A RNA polymerase III, subunit B Arginyl-tRNA synthetase 1 Pyrroline-5-carboxylate reductase RNA polymerase I, subunit C

HLD2 HLD3

HLD4 HLD5 HLD6 HLD7 HLD8 HLD9 HLD10 HLD11

Mitochondrial HPS60 chaperonopathy Hypomyelination and congenital cataract HLD with atrophy of basal ganglia and cerebellum 4H syndromea Similar to HLD8 Microcephaly and hypomyelination

a Hypomyelination, hypodontia

and hypogonadotropic hypogonadism. HLD, hypomyelinative leukodystrophy; MIM, Mendelian Inheritance in Man.

423

Developmental Neuropathology

Future direction and therapy Although there is currently no specific treatment available for PMD/SPG2, the wide range of animal models analogous to the different molecular mechanisms of disease will facilitate the potential for therapy. Possible options for restoring myelin function include either gene therapy or somatic cell transplantation. However, there are great problems with gene therapy due to the sensitivity of PLP1 gene dosage and the gain of function mutations. Gene delivery may be difficult but it will be an even greater challenge to obtain PLP1 expression at the correct level for normal myelination and maintenance. A reduction of PLP1 expression may be more successful, for example by antisense gene therapy, as loss of PLP1 gives a less severe phenotype as shown by patients with PMD with a PLP1 deletion or null mutation. Alternatively, somatic cell therapy may be an easier option and transplantation of oligodendrocyte precursors into the CNS has shown potential for animal models [63,64]. The sustained clinical improvement of two female PLP1 duplication carriers also supports this approach, demonstrating that certain oligodendrocyte lineages may proliferate and compensate for dysmyelination even years after birth [61]. Other therapies aimed at maintaining the axon integrity may also be helpful.

References ¨ 1. Pelizaeus, F (1885) Uber eine eigenth¨umliche Form Spastischer L¨ahmung mit Cerebraler Schrinungen auf heredit¨arer Grundlage (Multiple Sklerose). Arch Psychiatr Nervenkr 16:698–710 2. Merzbacher, L (1910) Eine eigenarige famili¨ar-hereditare Erkrankungsform [Aplasia axialis extra-corticalis congenita]. Z ges Neurol Psych 3:1–138 3. Liebers M (1928) Histopathologie des zweiten Falles von PelizaeusMerzbacherscher Krankheit. Z ges Neurol Psych 115:487–509 4. Heim P, Claussen M, Hoffmann B et al. (1997) Leukodystrophy incidence in Germany. Am J Med Genet 71:475–78 5. Mimault C, Giraud G, Courtois V et al. (1999) Proteolipoprotein gene analysis in 82 patients with sporadic Pelizaeus–Merzbacher disease: duplications, the major cause of the disease, originate more frequently in male germ cells, but point mutations do not. Clinical European Network on Brain Dysmyelinating Disease. Am J Hum Genet 65:360–9 6. Schouten JP, McElgunn CJ, Waaijer R et al. (2002) Relative quantification of 40 nucleic acid sequences by multiplex ligation-dependent probe amplification. Nucleic Acids Res 30:e57 7. Raskind WH, Williams CA, Hudson LD, Bird TD (1991) Complete deletion of the proteolipid protein gene (PLP) in a family with Xlinked Pelizaeus–Merzbacher disease. Am J Hum Genet 49:1355–60 8. Garbern JY, Cambi F, Tang XM et al. (1997) Proteolipid protein is necessary in peripheral as well as central myelin. Neuron 19:205–18 9. Sistermans EA, De-Wijs IJ, de-Coo RF, Smit LM et al. (1996) A (G-to-A) mutation in the initiation codon of the proteolipid protein gene causing a relatively mild form of Pelizaeus–Merzbacher disease in a Dutch family. Hum Genet 97:337–9

424

10. Hodes ME, Pratt VM, Dlouhy SR (1993) Genetics of Pelizaeus– Merzbacher disease. Dev Neurosci 15:383–94 11. Hobson GM, Kamholz J (1999 updated 2013) PLP1-related disorders. In: GeneReviews® . Seattle, WA, University of Washington. Available at: https://www.ncbi.nlm.nih.gov/books/NBK1182 (accessed November 7, 2017) 12. Strautnieks S, Malcolm S (1993) A G to T mutation at a splice site in a case of Pelizaeus–Merzbacher disease. Hum Mol Genet 2:2191– 212 13. Hobson GM, Davis AP, Stowell NC et al. (2000) Mutations in noncoding regions of the proteolipid protein gene in Pelizaeus– Merzbacher disease. Neurology 55:1089–96 14. Mattei MG, Alliel PM, Dautigny A et al. (1986) The gene encoding for the major brain proteolipid (PLP) maps on the q-22 band of the human X chromosome. Hum Genet 72:352–3 15. Griffiths I, Klugmann M, Anderson T et al. (1998) Current concepts of PLP and its role in the nervous system. Microsc Res Tech 41:344– 58 16. Weimbs T, Stoffel W (1992) Proteolipid protein (PLP) of CNS myelin: positions of free, disulfide-bonded, and fatty acid thioesterlinked cysteine residues and implications for the membrane topology of PLP. Biochemistry 31:12289–96 17. Nave KA, Lai C, Bloom FE, Milner RJ (1987) Splice site selection in the proteolipid protein (PLP) gene transcript and primary structure of the DM-20 protein of central nervous system myelin. Proc Natl Acad Sci U S A 84:5665–9 18. Campagnoni AT, Skoff RP (2001) The pathobiology of myelin mutants reveal novel biological functions of the MBP and PLP genes. Brain Pathol 11:74–91 19. Woodward K, Kendall E, Vetrie D, Malcolm S (1998) Pelizaeus– Merzbacher disease: identification of Xq22 proteolipid-protein duplications and characterization of breakpoints by interphase FISH. Am J Hum Genet 63:207–17 20. Woodward K, Palmer R, Rao K, Malcolm S (1999) Prenatal diagnosis by FISH in a family with Pelizaeus–Merzbacher disease caused by duplication of PLP gene. Prenat Diagn 19:266–8 21. Inoue K, Osaka H, Sugiyama N et al. (1996) A duplicated PLP gene causing Pelizaeus–Merzbacher disease detected by comparative multiplex PCR. Am J Hum Genet 59:32–9 22. Regis S, Filocamo M, Mazzotti R et al. (2001) Prenatal diagnosis of Pelizaeus–Merzbacher disease: detection of proteolipid protein gene duplication by quantitative fluorescent multiplex PCR. Prenat Diagn 21:668–71 23. Ellis D, Malcolm S (1994) Proteolipid protein gene dosage effect in Pelizaeus–Merzbacher disease. Nat Genet 6:333–4 24. Armour JA, Barton DE, Cockburn DJ, Taylor GR (2002) The detection of large deletions or duplications in genomic DNA. Hum Mutat 20:325–37 25. Inoue K, Osaka H, Imaizumi K et al. (1999) Proteolipid protein gene duplications causing Pelizaeus–Merzbacher disease: molecular mechanism and phenotypic manifestations. Ann Neurol 45: 624–32 26. Wolf NI, Cundall M, Rutland P et al. (2007) Frameshift mutation in GJA12 leading to nystagmus, spastic ataxia and CNS dys-/ demyelination. Neurogenetics 8:39–44 27. Hodes ME, Woodward K, Spinner NB et al. (2000) Additional copies of the proteolipid protein gene causing Pelizaeus– Merzbacher disease arise by separate integration into the X chromosome. Am J Hum Genet 67:14–22

Pelizaeus–Merzbacher Disease Chapter 34 28. Boulloche J, Aicardi J (1986) Pelizaeus–Merzbacher disease: clinical and nosological study. J Child Neurol 1:233–9 29. Scheffer IE, Baraitser M, Wilson J et al. (1991) Pelizaeus– Merzbacher Disease: classical or connatal? Neuropediatrics 22: 71–8 30. Johnston AW, McKusick VA (1962) A sex-linked recessive form of spastic paraplegia. Am J Hum Genet 14:83–94 31. Saugier–Veber P, Munnich A, Bonneau D et al. (1994) X-linked spastic paraplegia and Pelizaeus–Merzbacher disease are allelic disorders at the proteolipid protein locus. Nat Genet 6:257–62 32. Barkovich AJ (2000) Concepts of myelin and myelination in neuroradiology. Am J Neuroradiol 21:1099–109 33. Takanashi J, Inoue K, Tomita M et al. (2002) Brain N-acetylaspartate is elevated in Pelizaeus–Merzbacher disease with PLP1 duplication. Neurology 58:237–41 34. Garbern JY, Cambi F, Lewis R et al. (1999). Peripheral neuropathy caused by proteolipid protein gene mutations. Ann NY Acad Sci 883:351–65 35. Pamphlett R, Silberstein P (1986) Pelizaeus–Merzbacher disease in a brother and sister. Acta Neuropathol 69:343–6. 36. Bond C, Si X, Crisp M et al. (1997) Family with Pelizaeus– Merzbacher disease/X-linked spastic paraplegia and a nonsense mutation in exon 6 of the proteolipid protein gene. Am J Med Genet 71:357–60 37. Koeppen AH, Ronca NA, Greenfield EA, Hans MB (1987) Defective biosynthesis of proteolipid protein in Pelizaeus–Merzbacher disease. Ann Neurol 21:159–70 38. Harding B, Surtees R (2002) Metabolic and neurodegenerative diseases of childhood. In: DI Graham, PL Lantos, eds, Greenfield’s Neuropathology, 7th ed. London, Arnold. pp. 485–518 39. Billings-Gagliardi S, Kirschner DA, Nadon NL et al. (1995) Jimpy 4J: a new X-linked mouse mutation producing severe CNS hypomyelination. Dev Neurosci 17:300–10 40. Koeppen AH, Robitaille Y (2002) Pelizaeus–Merzbacher disease. J Neuropathol Exp Neurol 61:747–59 41. Yool DA, Edgar JM, Montague P, Malcolm S (2000) The proteolipid protein gene and myelin disorders in man and animal models. Hum Mol Genet 9:987–92 42. Readhead C, Schneider A, Griffiths I, Nave KA (1994) Premature arrest of myelin formation in transgenic mice with increased proteolipid protein gene dosage. Neuron 12:583–95 43. Cailloux F, Gauthier-Barichard F, Mimault C et al. (2000) Genotype–phenotype correlation in inherited brain myelination defects due to proteolipid protein gene mutations. Clinical European Network on Brain Dysmyelinating Disease. Eur J Hum Genet 8:837–45 44. Sistermans EA, de-Coo RF, De-Wijs IJ, van-Oost BA (1998) Duplication of the proteolipid protein gene is the major cause of Pelizaeus–Merzbacher disease. Neurology 50:1749–54 45. Hobson GM, Huang Z, Sperle K et al. (2002) A PLP splicing abnormality is associated with an unusual presentation of PMD. Ann Neurol 52:477–88 46. Gow A, Southwood CM, Lazzarini RA (1998) Disrupted proteolipid protein trafficking results in oligodendrocyte apoptosis in an animal model of Pelizaeus–Merzbacher disease. J Cell Biol 140: 925–34 47. Gow A, Friedrich V-LJ, Lazzarini RA (1994) Intracellular transport and sorting of the oligodendrocyte transmembrane proteolipid protein. J Neurosci Res 37:563–73

48. Southwood CM, Garbern J, Jiang W, Gow A (2002) The unfolded protein response modulates disease severity in Pelizaeus– Merzbacher disease. Neuron 36:585–96 49. Garbern JY, Yool DA, Moore GJ et al. (2002) Patients lacking the major CNS myelin protein, proteolipid protein 1, develop lengthdependent axonal degeneration in the absence of demyelination and inflammation. Brain 125:551–61 50. Boison D, Stoffel W (1994) Disruption of the compacted myelin sheath of axons of the central nervous system in proteolipid proteindeficient mice. Proc Natl Acad Sci U S A 91:11709–13 51. Klugmann M, Schwab MH, Puhlhofer A et al. (1997) Assembly of CNS myelin in the absence of proteolipid protein. Neuron 18:59–70 52. Yool DA, Klugmann M, McLaughlin M et al. (2001) Myelin proteolipid proteins promote the interaction of oligodendrocytes and axons. J Neurosci Res 63:151–64 53. Harding BN, Malcolm S, Ellis D, Wilson J (1995) A case of Pelizaeus–Merzbacher disease showing increased dosage of the proteolipid protein gene. Neuropathol Appl Neurobiol 21:111–15 54. Kagawa T, Nakao J, Yamada M et al. (1994) Fate of jimpy-type oligodendrocytes in jimpy heterozygote. J Neurochem 62:1887–93 55. Anderson TJ, Schneider A, Barrie JA et al. (1998) Late-onset neurodegeneration in mice with increased dosage of the proteolipid protein gene. J Comp Neurol 394:506–19 56. Simons M, Kramer EM, Macchi P et al. (2002) Overexpression of the myelin proteolipid protein leads to accumulation of cholesterol and proteolipid protein in endosomes/lysosomes: implications for Pelizaeus–Merzbacher disease. J Cell Biol 157:327–36 57. Hodes ME, Blank CA, Pratt VM et al. (1997) Nonsense mutation in exon 3 of the proteolipid protein gene (PLP) in a family with an unusual form of Pelizaeus–Merzbacher disease. Am J Med Genet 69:121–5 58. Woodward K, Kirtland K, Dlouhy S et al. (2000) X inactivation phenotype in carriers of Pelizaeus–Merzbacher disease: skewed in carriers of a duplication and random in carriers of point mutations. Eur J Hum Genet 8:449–54 59. Nance MA, Boyadjiev S, Pratt VM et al. (1996) Adult-onset neurodegenerative disorder due to proteolipid protein gene mutation in the mother of a man with Pelizaeus–Merzbacher disease. Neurology 47:1333–5 60. Sivakumar K, Sambuughin N, Selenge B et al. (1999) Novel exon 3B proteolipid protein gene mutation causing late-onset spastic paraplegia type 2 with variable penetrance in female family members. Ann Neurol 45:680–3 61. Inoue K, Tanaka H, Scaglia F et al. (2001). Compensating for central nervous system dysmyelination: females with a proteolipid protein gene duplication and sustained clinical improvement. Ann Neurol 50:747–54 62. Uhlenburg B, Schuelke M, R¨uschendorf F et al. (2004) Mutations in the gene encoding gap junction protein alpha 12 (connexin 46.6.) cause Pelizaeus–Merzbacher-like disease. Am J Hum Genet 75: 251– 60 63. Duncan ID, Grever WE, Zhang SC (1997) Repair of myelin disease: strategies and progress in animal models. Mol Med Today 3:554–61 64. Learish RD, Brustle O, Zhang SC, Duncan ID (1999) Intraventricular transplantation of oligodendrocyte progenitors into a fetal myelin mutant results in widespread formation of myelin. Ann Neurol 46:716–22 65. Woodward K, Malcolm S (2001) CNS myelination and PLP gene dosage. Pharmacogenomics 3:263–72

425

35

Cockayne Syndrome Karen M. Weidenheim1 and P. J. Brooks2 1 Departments

of Pathology (Neuropathology), Montefiore Medical Center and Albert Einstein College of Medicine, Bronx, NY, USA 2 Office of Rare Diseases Research and Division of Clinical Innovation, National Center for Advancing Translational Sciences (NCATS), National Institutes of Health and Laboratory of Neurogenetics, National Institute of Alcohol Abuse and Alcoholism, National Institutes of Health, Bethesda, MD, USA

Definition, major synonyms and historical perspective Cockayne syndrome is one of a family of multisystem, multigenic, multiallelic, autosomal-recessive disorders due to defects in DNA repair. This family of diseases includes Cockayne syndrome, xeroderma pigmentosum, cerebro-oculofacioskeletal syndrome (COFS) and trichothiodystrophy. Neurological involvement is greatest in Cockayne syndrome, where there is postnatal failure of brain growth. In addition, these individuals exhibit marked somatic growth failure with dwarfism, premature senescence and may show extreme sensitivity to ultraviolet irradiation. Cockayne syndrome was first described in 1936 by Edward Cockayne, who reported two siblings with small heads, sunken eyes, short trunk with long legs and large hands and feet, optic atrophy with retinal pigmentary changes, and deafness, whose clinical course was further reported ten years later. In 1978, Brumback et al. [1] reported the clinical features in detail as: 1. cachectic dwarfism with prominent loss of adipose tissue 2. distinctive facies with prognathism, prominent large malformed ears, enophthalmos, prominent beaked nose, and microcephaly resembling an aged person (the term progeria has been applied to these patients) 3. multiple skeletal abnormalities, including abnormal size, shape, and density of bones 4. endocrine abnormalities 5. photosensitivity to sunlight 6. neurological signs including optic atrophy, retinal pigmentary degeneration, sensorineural deafness, cerebellar signs, choreoathetosis, peripheral neuropathy, and cognitive abnormalities. In a subsequent paper, Nance and Berry [2] identified the major criteria of Cockayne syndrome as developmental delay

and growth failure, with minor criteria including cutaneous photosensitivity, retinopathy/cataracts, sensorineural hearing loss, dental caries, and dwarfism. In 2013, Laugel [3] modified the classification. Major criteria include developmental delay, progressive growth failure, and progressive microcephaly. Minor criteria include cutaneous photosensitivity, progressive retinopathy/cataracts, progressive sensorineural hearing loss, dental enamel hypoplasia, and enopthalmia. A 2016 natural history of Cockayne syndrome emphasized postnatal growth failure and microcephaly as cardinal features of the disease [4]. Some cases of the genetically heterogeneous autosomalrecessive disorder cerebro-oculofacioskeletal syndrome have been found to have genetic defects in the ERCC gene similar to those in Cockayne syndrome, and these cases are classified as Cockayne syndrome type II [3,5]. Cockayne syndrome II is best viewed as the most severe variant of the syndrome, distinguished from type I by arthrogryposis, indicative of very early onset neurologic abnormalities [3]. Cockayne syndrome is divided into three types, depending upon age of onset and rate of progression [2–5]. Because of genetic heterogeneity, this classification is imperfect and a given patient may not fit precisely into one of these categories. In type I, the classical type, a previously normal infant shows developmental delay by one year of age. In type II, the patient manifests growth failure at birth and there is little to no postnatal neurological development. In type III, there is later onset, fewer symptoms, and a slower rate of progression, with a lifespan of 40–50 years.

Incidence and prevalence Cockayne syndrome has a frequency of 2–3 cases per million births [3,5].

Developmental Neuropathology, Second Edition. Edited by Homa Adle-Biassette, Brian N. Harding, and Jeffrey A. Golden. © 2018 John Wiley & Sons Ltd. Published 2018 by John Wiley & Sons Ltd.

427

Developmental Neuropathology

(a)

(b)

(c)

(d)

Figure 35.1 Patient with Cockayne syndrome at ages 9 (a), 13 (b), 17 (c), and 28 years (d). Note the cheerful expression and that the characteristic Cockayne facies (sunken eyes, sharp nose, jutting chin, emaciation) was not striking even at age 17 years. Reprinted with permission from Rapin et al. [15] and SAGE Publications.

Clinical features Clinical presentation After a normal birth and neonatal period, Cockayne syndrome I manifests in the first year of life as sensitivity to sunlight. Skin erythema after minimal sun exposure may be the presenting symptom. Within three months, somatic and cephalic growth failure leading to microcephaly and dwarfism are manifest. There is lack of adipose tissue, and delayed motor development, delayed and minimal language acquisition, and cognitive difficulties. Many children display a happy, outgoing, social personality despite their other difficulties. The characteristic facies includes sunken eyes, small, poorly reactive pupils, sharp nose,

428

large malformed ears, and carious teeth (Figure 35.1) [see also 4]. Ataxia and spasticity interfere with motor function, and slowed nerve conduction velocities indicate demyelinating peripheral neuropathy [6]. Seizures may occur. There are corneal opacities, cataracts, retinitis pigmentosa and visual loss, and progressive sensorineural hearing loss. Skeletal anomalies include abnormal size, shape and density of bones, and the extremities are disproportionately long. Wilson et al. [4] also noted the high prevalence of cold extremities in patients with Cockayne syndrome. Some individuals may develop systemic hypertension despite their young age, and hypertensive renal disease and cerebrovascular accidents have been documented [7]. A slowly progressive course is marked by progressive cachexia and dementia, with a lifespan of 10–20 years.

Cockayne Syndrome Chapter 35

Infants with Cockayne syndrome II show symptoms and signs at or soon after birth, with intrauterine growth restriction, microcephaly, microphthalmia, bilateral cataracts, beaked nose, small jaw, photosensitivity, arthrogryposis, and growth failure [8]. Lifespan is approximately seven years. Cockayne syndrome III shows later onset and longer survival; these atypical features may cause diagnostic difficulty [9].

Biochemistry Few data are available regarding laboratory values in Cockayne syndrome. Endocrine abnormalities include abnormally low urinary 17-hydroxy-steroids and 17-ketosteroids, abnormal glucose responses with excessive insulin and growth hormone releases to stress. Parameters of renal function reflect progressive renal failure. Despite the occurrence of systemic hypertension, serum lipids have been normal [10]. Imaging Neuroimaging of 19 cases revealed hypomyelination, calcifications and diffuse atrophy in the cerebral white matter, cerebellum, and brainstem (Figure 35.2) [11]. Calcification was most common in putamen, neocortex and dentate nuclei, and occasionally occurred in the remaining basal ganglia, cerebral (a)

white matter and thalami. Some patients with Cockayne syndrome II also showed evidence of unilateral calcification of large leptomeningeal blood vessels [10]. Proton magnetic resonance spectroscopy performed in 9 of the 19 patients revealed normal creatine levels. N-acetyl aspartate–creatine ratios were decreased in gray and white matter in all patients, together with decreased choline–creatine ratio in the white matter, and elevated lactate in all cases. The choline–creatine ratios were normal or decreased in the gray matter [11].

Differential diagnosis Cockayne syndrome must be differentiated from other disorders characterized by growth failure and developmental delay, including effects of intrauterine infections, Seckel syndrome, and Hutchinson–Guilford progeria, as well as xeroderma pigmentosum and trichothiodystrophy [12]. Clinical features that may be useful for the differential diagnosis include the postnatal nature of the growth failure, presence of cachexia, progressive neurological deterioration, and characteristic dysmorphic facies of Cockayne syndrome. Imaging studies that show white matter changes and calcifications, and the presence of normal hair and lack of malignant tumors may also be helpful. The radiographic differential diagnosis includes congenital (b)

Figure 35.2 Computed tomographic scans showing (a) thickening of the calvarium, severe panventricular enlargement, and calcification of the basal ganglia and (b) extreme thinning of the white matter, widened sulci, and thin bilateral subdural collections. Reprinted with permission from Rapin et al. [15] and SAGE Publications.

429

Developmental Neuropathology cytomegalovirus infection, Aicardi–Gouti`eres syndrome, Pelizaeus–Merzbacher disease, and mitochondrial disorders.

Pathology Systemic pathology Dwarfism with cachexia, without major malformations, is present in almost all cases of Cockayne syndrome. The hair is sparse and there is premature baldness, which, together with atrophy of the skin, produces the appearance of premature aging. Atherosclerosis and arteriolosclerosis, out of proportion to the age of the patient, is well documented and may produce hypertension, renal disease, and strokes [7]. Gonadal maldevelopment is seen in most but not all males, there are atrophic or undescended testes, with or without a small phallus. Female patients show small breasts with normal nipples and areolae, and have menstrual periods, and occasional successful pregnancies have been reported [9]. Neuropathology Brain weights are markedly reduced and weights between 350 g and 600 g are found in individuals with end-stage disease [6]. Despite the small size, the brain is miniature, but the sulci are not visibly enlarged. Large vessel atherosclerosis is present in all cases, and the optic nerves are atrophic. Cut sections show ventricular enlargement (Figure 35.3). There is discoloration of the white matter, which may be patchy, and pancerebellar atrophy. Microscopic examination reveals patchy loss of myelin in the white matter (Figure 35.4a), usually accompanied by hyperchromatic oligodendrocytes, astrocytic gliosis and calcospherites (Figures 35.4b,c). There is atrophy of the deep gray matter, with gliosis and calcospherites. The cortical ribbon is much better preserved, but astrocyte proliferation likely reflects some degree of neuronal loss, and ferruginated neurons may be observed. The pancerebellar atrophy is characterized by loss of Purkinje and granular neurons, with ferrugination along the Purkinje cell interface, axonal torpedoes and neuronal loss in the dentate nucleus (Figure 35.4d). Glia often contain bizarre, hyperchromatic nuclei (Figure 35.4). Very rarely, neurofibrillary tangles have been reported [6]. There is leptomeningeal fibrosis. The spinal cord may show Wallerian degeneration of the corticospinal tract. Peripheral nerves show segmental demyelination with axonal degeneration [13]. Skeletal muscle shows neurogenic atrophy and may show hyperchromatic nuclei in the remaining fibers (Figures 35.4e,f). Ocular findings include small eyes, with cataract formation, pigmentary retinopathy, atrophy of the ganglion cell layers, and atrophy of the pigment epithelium of the choroid [6,9,13]. Findings in the auditory system include marked loss of hair cells, supporting cells, and stria vascularis in the cochlear end organs, loss of neurons in the cochlear nuclei with atrophy of

430

the cochlear nerve, and atrophy of the vestibular nerve and the vestibular end organs [6,9].

Genetics Cockayne syndrome is often considered in association with xeroderma pigmentosum and trichothiodystrophy, since all three disorders have defects in nucleotide excision repair (NER) the DNA repair pathways that removes DNA damage resulting from ultraviolet light [14]. Each phenotypic syndrome may arise on a genetic background usually considered to cause one of the other syndromes. However, there are important clinical differences between these three diseases (13). While xeroderma pigmentosum neurologic disease primarily affects gray matter, Cockayne syndrome neurological disease primarily impacts the white matter. In addition, Cockayne syndrome neurological disease includes many other features, such as calcifications and vasculopathy, that are not seen in xeroderma pigmentosum. Trichothiodystrophy neurologic disease is characterized by abnormal brain development and myelin defects, making it more similar to Cockayne syndrome than xeroderma pigmentosum. Each of the genes in which causative mutations in Cockayne syndrome that have been detected is autosomal, producing an autosomal-recessive mode of inheritance. From the standpoint of both the genetics and biochemistry, it is helpful to consider Cockayne syndrome in relation to another rare sun sensitivity disorder, xeroderma pigmentosum. Like patients with Cockayne syndrome, those with xeroderma pigmentosum show severe sun sensitivity. However, patients with xeroderma pigmentosum also have a greatly increased risk of skin cancer on sun-exposed areas of the body, whereas, as a rule, those with Cockayne syndrome never develop skin cancer. Xeroderma pigmentosum can result from mutations in one of eight xeroderma pigmentosum genes, denoted XPA through XPG and XPV. Patients in xeroderma pigmentosum complementation groups A through G have a defect in NER, the DNA repair process that is responsible for the removal of thymine dimers as well as other ultraviolet light-induced DNA lesions. Pure Cockayne syndrome (i.e., without the associated features of other diseases, such as xeroderma pigmentosum) can result from mutations in either of two genes, CSA or CSB, and the majority of patients described in the literature have mutations in one of these two genes. In addition to pure Cockayne syndrome, some patients have been identified with a combination of the cutaneous features of xeroderma pigmentosum as well as the neurological abnormalities typical of Cockayne syndrome [13,14]. This combination is referred to as the xeroderma pigmentosum–Cockayne syndrome complex. Patients with xeroderma pigmentosum–Cockayne syndrome complex have mutations in one of three genes, either XPB, XPD, or XPG. Mutations in CSB, XPD, and XPG can also result in cerebrooculofacioskeletal syndrome (Cockayne syndrome II in Laugel’s classification scheme) [3].

Cockayne Syndrome Chapter 35

(a)

(b)

Figure 35.3 Gross examination shows a very small brain that may be considered a “minibrain” because its exterior conformation shows only minimal atrophy, (a) despite very low weights in the range of infants, despite the age of the patient. Coronal sections show ventriculomegaly with preservation of the cortical ribbon and decrease in size of the white matter, which is firm and bright white (b, reprinted with permission from Rapin et al. [15]) and SAGE Publications.

A 2013 study described two patients with the combined features of Cockayne syndrome and xeroderma pigmentosum, and one patient also had clinical features of Fanconi anemia [16]. These patients had mutations in the genes encoding XPF or ERCC1, whose protein products act together as a nuclease which participates in both NER, and in the repair of DNA intrastrand crosslinks.

Cellular and molecular biology Cockayne syndrome cells are hypersensitive to killing by ultraviolet light, as measured by decreased cell survival after ultraviolet irradiation. There are two different types of NER, the DNA repair pathway that removes ultraviolet light damage.

431

Developmental Neuropathology

(a)

(b)

(c)

(d)

(e)

(f)

Figure 35.4 (a) Low-power view of white matter stained with Luxol fast blue–periodic acid–Schiff shows hypercellularity and is pale blue, indicative of demyelination (×347). (b) High-power view of white matter stained with Luxol fast blue–periodic acid–Schiff shows pleomorphic oligodendrocytes (arrows) (original magnification, ×1375). C. Calcification (arrows) in the basal ganglia (hematoxylin-eosin stain; ×602). (d) Collagen IV immunostains demonstrates

strong vessels (arrow) (×400). (e) Skeletal muscle showing small group atrophy (arrows) consistent with neurogenic atrophy (hematoxylin and eosin stain; ×641). (f) Toluidine blue-stained 1 micron-thick section of skeletal muscle showing marked nuclear atypia (arrow) (×1280). Reprinted with permission from Rapin et al. [15] and SAGE Publications.

The global genome repair pathway is active throughout the genome, while a more rapid repair pathway, transcriptioncoupled nucleotide excision repair (TC-NER), is responsible for more rapid removal of transcription-blocking DNA damage on the transcribed strand of active genes. Cells from patients with Cockayne syndrome have a specific defect in TC-NER that corresponds to the defect in RNA synthesis following ultraviolet

light that reflects TC-NER [17]. In recovery of RNA synthesis experiments, cells are treated with ultraviolet light to cause damage to DNA, then the recovery of RNA synthesis is measured by the incorporation of tritium labeled uridine, or, more recently, fluorescent uridine analogs. Cells from patients with Cockayne syndrome show a clear defect in this assay, which is used as the diagnostic test of choice for Cockayne syndrome.

432

Cockayne Syndrome Chapter 35

Linking the defect in recovery of RNA synthesis to neurologic disease in Cockayne syndrome is complicated by the fact that cells from patients with ultraviolet-sensitive syndrome also show defective recovery of RNA synthesis. Patients with ultravioletsensitive syndrome are sun-sensitive, but do not exhibit any neurologic abnormalities, in contrast to those with Cockayne syndrome. In the late 1990s, a series of papers by Leadon and colleagues claimed that cells from Cockayne syndrome patients, including those from XP-G patients with Cockayne syndrome, had a specific defect in transcription-coupled repair of oxidative DNA damage. These papers have been retracted due to findings of scientific misconduct. We note this information here because references to these papers may still be found in older review articles. Multiple studies have shown that Cockayne syndrome cells are hypersensitive to a variety of DNA-damaging agents. However, Cockayne syndrome cells are also hypersensitive to other cellular stressors, such as hypoxia, that do not directly cause DNA damage, and may have an abnormal transcriptional response to hypoxia [18]. Transcriptional abnormalities may be present in cells from patients with Cockayne syndrome in the absence of exposure to exogenous DNA damage. When transcriptional activity of CSBnull cells was compared with that in normal cells in a 2006 study, CSB cells appeared to be under chronic inflammatory stress [19]. More recent studies from 2012 [20] have focused on the possible role of a transposable element (found only in primates, not rodents) that forms an alternative exon in the CSB gene, and its potential role in the transcriptional abnormalities in Cockayne syndrome. Multiple abnormalities in genes and proteins related to neuronal development in cells from patients with Cockayne syndrome have been identified [21,22]. In particular, downregulation of the brain-derived neurotrophic factor (BDNF) gene was observed in postmortem brain tissues, an observation with potential clinical implications. Similar studies in normal and Csb-deficient mouse cells revealed that only one of eight genes that were differentially expressed in human cells were similarly regulated in mouse cells [22]. These observations may explain why mouse models of Cockayne syndrome do not accurately reflect the neuropathology of the human syndrome. More work remains to be done to understand the nature of transcriptional abnormalities in Cockayne syndrome cells. Studies from 2013 have focused on a possible role for mitochondrial abnormalities in Cockayne syndrome [23,24]. However, the mitochondrial hypothesis cannot readily explain how mutations in XPB, XPD, or XPG result in Cockayne syndrome, yet XPG mutations clearly result in Cockayne syndrome neurologic disease [13]. The neuropathology of mitochondrial disorders includes infarct-like lesions with neuronal cell death in several neuroanatomic locations, including periventricular and periaqueductal gray matter, tectum, hippocampus and neocortex, which also show capillary proliferation, hemorrhage and

gliosis. White matter pathology is not a significant finding in mitochondrial disorders, and the neuropathology of mitochondrial disorders is therefore quite different from the findings observed in Cockayne syndrome.

Pathogenesis Understanding the pathogenesis of Cockayne syndrome neurologic disease and its systemic manifestations is challenging in view of the multiple pathologies observed (demyelination and dysmyelination, calcification, cerebellar atrophy and dysplasia, cerebral vasculopathy). There may be multiple mechanisms underlying different pathologies [22,25]. Another challenge is to determine which pathologies are cell-autonomous versus which are secondary to loss or dysfunction in other cell types. In addition, any explanation for the Cockayne syndrome neurologic disease must account for the qualitatively similar (although temporally variable) neurologic abnormalities in patients with mutations in CSA, CSB, XPB, XPD, and XPG. Defective TC-NER can explain the sun sensitivity common to Cockayne syndrome and ultraviolet-sensitive syndrome, but is not relevant for Cockayne syndrome neurologic disease [15,26]. While it remains possible that other DNA repair defects are involved in some aspects of Cockayne syndrome neurologic disease, many of the supporting studies involve exposing fibroblasts of patients with Cockayne syndrome to very high levels of DNA-damaging agents. The relevance of such massive exposures for cells in patients with Cockayne syndrome in vivo remains unclear. An expanded version of the transcription syndrome hypothesis can potentially explain several features of Cockayne syndrome, including many aspects of neurologic disease [26]. There is at least some evidence for a role of CSA, CSB, XPB, XPD, and XPG in one or more aspects of transcriptional regulation. Indeed, even the mitochondrial abnormalities reported in Cockayne syndrome might be secondary to transcriptional abnormalities.

Animal models Mutant mice lacking either the Csa or Csb genes have been generated, but show only minor neurologic abnormalities, in contrast to the profound defects seen in humans. However, when Csa or Csb mice are crossed with Xpa-/- mice, which are completely deficient in NER, the resulting animals are more similar to human patients with Cockayne syndrome. They are runted and show evidence of cerebellar degeneration including hypoplastic Purkinje neurons, and impaired cerebellar foliation; these animals do not survive beyond weaning [27]. The similar Csa-/- Xpa-/- mice also die before weaning, but careful feeding during the pre-weaning period enables them to survive past weaning, with a progressive lipodystrophy, as well as

433

Developmental Neuropathology neurologic abnormalities including cerebellar ataxia, hind limb dystonia and paralysis [28]. Related knockout mice, lacking one or more of the relevant gene products, have been generated by several groups, but not all of them recapitulate the human disorder. In addition to whole body gene knockouts, additional mice have been produced in which CSB and one or more NER genes have been deleted in specific subsets of neurons in the mouse brain. The resulting progressive degeneration of neurons lacking both proteins produces a phenotype similar to the neurologic disease shown in some patients with xeroderma pigmentosum, but is different from Cockayne syndrome neurologic disease [29,30]. In view of the fundamental differences between mice and humans lacking Cockayne syndrome genes, future studies of human brain cells and tissues from patients with Cockayne syndrome are likely to provide the most useful pathophysiological insights into Cockayne syndrome neurologic disease, as well as avenues for treatments and therapies.

Treatment, future perspective, conclusions Treatment Symptomatic, supportive treatment is available for some aspects of the disease, including cochlear implants for hearing loss, and Botox injections and surgical interventions for joint contractures. Some patients with Cockayne syndrome (type 1 or type III) develop a Parkinsonian-like tremor, perhaps due to basal ganglia calcification. One study demonstrated clinical benefit of carbidopa levodopa (Sinemet® ) for such patients [31]. Animal studies suggest that manipulation of cellular NAD+ levels could have clinical benefit for the somatic features of the disease [32]. Whether such an intervention could impact those features of Cockayne syndrome neurologic disease that do not occur in mice is less clear. In addition, decreased brain-derived neurotrophic factor levels in human Cockayne syndrome brain tissues and cells raises the possibility of augmenting these levels as a therapeutic strategy [22]. At present, such therapies remain experimental. Future perspectives For many years, Cockayne syndrome was considered a DNA repair disorder. More recent studies challenge this view, with increasing experimental support for the concept of Cockayne syndrome as a disease of abnormal transcription, perhaps in addition to defective DNA repair. This distinction has important implications for therapeutics development. Conclusions Cockayne syndrome is a uniquely human disorder with multiple genetic causes and whose major manifestations occur in the central nervous system. Laboratory investigation of Cockayne syndrome has augmented basic knowledge into DNA repair and abnormal transcription, as well as mitochondrial function.

434

These studies provide promising new pathways that may elucidate the pathogenesis of the neuropathology. Animal models with the genetic features of human Cockayne syndrome do not necessarily recapitulate the clinicopathological features of the human disorder. Future investigations will require analysis of human tissues and human cells to elucidate the mechanisms of human disease, and provide targeted treatments for individuals with Cockayne syndrome.

References 1. Brumback RA, Yoder FW, Andrews AD et al. (1978) Normal pressure hydrocephalus. Recognition and relationship to neurological abnormalities in Cockayne’s syndrome. Arch Neurol 35:337–45 2. Nance MA, Berry SA (1992) Cockayne syndrome: review of 140 cases. Am J Med Genet 42:68–84 3. Laugel V (2013) Cockayne syndrome: the expanding clinical and mutational spectrum. Mech Ageing Dev 134:161–70 4. Wilson BT, Stark Z, Sutton RE et al. (2016). The Cockayne Syndrome Natural History (CoSyNH) study: clinical findings in 102 individuals and recommendations for care. Genet Med 18:483–93 5. Lanzafame M, Vaz B, Nardo T et al. (2013) From laboratory tests to functional characterization of Cockayne syndrome. Mech Ageing Dev 134:171–9 6. Weidenheim KM, Dickson DW, Rapin I (2009) Neuropathology of Cockayne syndrome: Evidence for impaired development, premature aging, and neurodegeneration. Mech Ageing Dev 130:619–36 7. Mizuguchi M, Itoh M (2005) A 35-year-old female with growth and developmental retardation, progressive ataxia, dementia and visual loss Neuropathology 25:103–6 8. Meira LB, Graham JM, Greenberg CR et al. (2000) Manitoba aboriginal kindred with original cerebro-oculo-facio-skeletal syndrome has a mutation in the Cockayne syndrome Group B (CSB) gene. Am J Hum Genet 66:1221–8 9. Rapin I (2013) Disorders of nucleotide excision repair. In: O Dulac, M Lassonde, HB Sarnat, eds, Handbook of Clinical Neurology, Vol 113 (3rd series). Pediatric Neurology Part III. Amsterdam, Elsevier. pp. 1637–50 10. Sugarman GI, Landing BH, Reed WB (1977) Cockayne syndrome: clinical study of two patients and neuropathologic findings in one. Clin Pediatr 16:225–32 11. Koob M, Laugel V, Durand M et al. (2010) Neuroimaging in Cockayne syndrome. Am J Neuroradiol 31:1623–30 12. Rapin I, Lindenbaum Y, Dickson DW et al. (2000) Cockayne syndrome and xeroderma pigmentosum. DNA repair disorders with overlaps and paradoxes. Neurology 55:1442–9 13. Lindenbaum Y, Dickson D, Rosenbaum P et al. (2001) Xeroderma pigmentosum/Cockayne syndrome complex: first neuropathological study and review of eight other cases. Eur J Paediatr Neurol 5:225–42 14. van der Horst GT, van Steeg H, Berg RJ et al. (1997). Defective transcription-coupled repair in Cockayne syndrome B mice is associated with skin cancer predisposition. Cell 89:425–435 15. Rapin I, Weidenheim K, Lindenbaum Y et al. (2006) Cockayne syndrome in adults: review with clinical and pathologic study of a new case. J Child Neurol 21:991–1006

Cockayne Syndrome Chapter 35 16. Kamenisch Y, Berneburg M (2013) Mitochondrial CSA and CSB: protein interactions and protection from aging associated DNA mutations. Mech Ageing Dev 134:270–4 17. Lagerwerf S, Vrouwe MG, Overmeer RM et al. (2011) DNA damage response and transcription. DNA Repair 10:743–50 18. Proietti-DeSantis L, Drane P, Egly JM (2006) Cockayne syndrome B protein regulates the transcriptional program after UV irradiation. EMBO J 25:1915–23 19. Newman JC, Bailey AD, Weiner AM (2006) Cockayne syndrome group B protein (CSB) plays a general role in chromatin maintenance and remodeling. Proc Natl Acad Sci U S A 103:9613–8 20. Bailey AD, Gray LT, Pavelitz T et al. (2012) The conserved Cockayne syndrome B-piggyBac fusion protein (CSB-PGBD3) affects DNA repair and induces both interferon-like and innate antiviral responses in CSB-null cells. DNA Repair 11:488–501 21. Ciaffardini F, Nicolai S, Caputo M et al. (2014) The Cockayne syndrome B protein is essential for neuronal differentiation and neuritogenesis. Cell Death and Disease 5:e1263 22. Wang Y, Chakravarty P, Ranes M et al. (2014) Dysregulation of gene expression as a cause of Cockayne syndrome neurological disease. Proc Natl Acad Sci U S A 111:14454–9 23. Kashiyama K, Nakazawa Y, Pilz DT et al. (2013) Malfunction of nuclease ERCC1-XPF results in diverse clinical manifestations and causes Cockayne syndrome, xeroderma pigmentosum and Fanconi anemia. Am J Human Genet 92:807–19 24. Scheibye-Knudsen M, Croteau DL, Bohr VA (2013) Mitochondrial deficiency in Cockayne syndrome. Mech Ageing Dev 134: 275–83

25. Brooks PJ (2013) Blinded by the UV light: how the focus on transcription-coupled NER has distracted from understanding the mechanisms of Cockayne syndrome neurologic disease. DNA Repair 12:656–71 26. Brooks PJ, Cheng TF, Cooper L (2008) Do all of the neurologic diseases in patients with DNA repair gene mutations result from the accumulation of DNA damage? DNA Repair 7:834–48 27. Murai M, Enokido Y, Inamura N et al. (2001) Early postnatal ataxia and abnormal cerebellar development in mice lacking Xeroderma pigmentosum Group A and Cockayne syndrome Group B DNA repair genes. Proc Natl Acad Sci U S A 98:13379–84 28. Brace LE, Vose SC, Vargas DF et al. (2013) Lifespan extension by dietary intervention in a mouse model of Cockayne syndrome uncouples early postnatal development from segmental progeria. Aging Cell 12:1144–7 29. Barnhoorn S, Uittenboogaard LM, Jaarsma D et al. (2014) Cell– autonomous progeroid changes in conditional mouse models for repair endonuclease XPG deficiency. PLoS Genet 10:e1004686 30. Jaarsma D, van der Pluijm I, de Waard MC et al. (2011) Age-related neuronal degeneration: complementary roles of nucleotide excision repair and transcription-coupled repair in preventing neuropathology. PLoS Genet 7:e1002405 31. Neilan EG, Delgado MR, Donovan A et al. (2008) Response of motor complications in Cockayne syndrome to carbidopa levodopa. Arch Neurol 65:1117–21 32. Scheibye-Knudsen M, Mitchell SJ, Fang EF et al. (2014) A high-fat diet and NAD(+) activate Sirt1 to rescue premature aging in Cockayne syndrome. Cell Metab 20:840–55

435

36

Vanishing White Matter Disease Marianna Bugiani,1 James M. Powers,2 and Marjo S. van der Knaap3 1 Departments

of Child Neurology and Pathology, VU University Medical Center, Amsterdam, the Netherlands of Pathology and Laboratory Medicine, University Rochester Medical Center, Rochester, NY, USA 3 Department of Child Neurology, VU University Medical Center, Amsterdam, the Netherlands 2 Department

Definition Vanishing white matter (VWM) is a leukoencephalopathy with an autosomal-recessive mode of inheritance, characterized by progressive rarefaction and cystic degeneration of the cerebral white matter. The condition is related to mutations in any of the five genes encoding the subunits of eukaryotic translation initiation factor 2B (eIF2B) [1].

Sex and age distribution Males and females are affected with the same frequency, although the disease may be clinically more severe in males, and females are overrepresented among older patients [8]. VWM affects patients of all ages, from prenatal onset to senescence. Age at onset is most often between two and six years, but early infantile, late childhood, adolescent, and adult onsets are also possible [1]. Cree encephalopathy invariably has its onset within the first year of life [9], but more severe variants with symptoms already evident at birth and early demise have also been described outside the Cree population [10].

Synonyms and historical annotations VWM is also referred to as “childhood ataxia with central nervous system hypomyelination” [2] and “myelinopathia periaxialis diffusa” [3]. “Cree leukoencephalopathy” is a severe variant of VWM, related to one particular mutation [4]. “Ovarioleukodystrophy” is defined as a leukoencephalopathy with primary or secondary ovarian failure [5]. The large majority of affected patients have mutations in eIF2B subunit genes and in fact, have VWM [6].

Epidemiology Incidence and prevalence VWM is one of the most prevalent inherited white matter disorders [7]. In some countries, including the Netherlands, the incidence is similar to that of metachromatic leukodystrophy, with an estimated carrier frequency of 1 in 100 and a disease incidence of 1 in 40 000. The impression given from published case series is that VWM is more common among whites and that the incidence may be lower in people of different ethnic background.

Risk factors VWM is characterized by chronic neurologic deterioration with additional episodes of more rapid and serious deterioration, most often provoked by febrile infections or minor head trauma, which may lead to coma and death [1].

Genetics VWM has an autosomal-recessive mode of inheritance. The disease is related to defects in the translation initiation factor eIF2B. eIF2B consists of five non-identical subunits (eIF2Bα, eIF2Bβ, eIF2Bγ, eIF2Bδ, and eIF2Bϵ), encoded by different genes (EIF2B1, EIF2B2, EIF2B3, EIF2B4, and EIF2B5, respectively) located on different chromosomes (12q24.3, 14q24, 1p34.1, 2p23.3, and 3q27, respectively). Mutations in any of these genes can independently cause the disease [11,12]. Most are missense mutations often affecting nonconserved amino acid residues. Until now, frameshift and nonsense mutations, which prevent the expression of full-length eIF2B subunits, have only been observed in a compound-heterozygous state with a missense mutation as the second mutation [4,8,11–14].

Developmental Neuropathology, Second Edition. Edited by Homa Adle-Biassette, Brian N. Harding, and Jeffrey A. Golden. © 2018 John Wiley & Sons Ltd. Published 2018 by John Wiley & Sons Ltd.

437

Developmental Neuropathology There are a few known founder effects. Most Dutch patients come from the eastern part of the Netherlands (EN haplotype) and share a common ancestor from around 1800. [11,15]. Patients with the EN haplotype have the same mutation in EIF2B5 (271A > G). A second founder effect has been identified in the southern part of the Netherlands (SN haplotype). Patients with the SN haplotype have the same mutation in EIF2B2 (638A > G) [11]. Patients with Cree leukoencephalopathy all have the same mutation in EIF2B5 (584 G > H) [4]. There is a clear genotype– phenotype correlation. Some mutations are consistently related to a mild phenotype and others to a severe phenotype [8]. However, there is also wide phenotypic variability among patients with the same mutations, even among siblings, suggesting that environmental and/or other genetic factors may significantly influence the phenotype [8].

Clinical features Signs and symptoms VWM is a disease with a wide phenotypic variation [1]. As often occurs in genetic disorders, age at onset is predictive of disease severity. The classical VWM phenotype has its onset in early childhood and is characterized by progressive neurological deterioration with cerebellar ataxia, usually less prominent spasticity and relatively mild cognitive decline [1,2,16–18]. Optic atrophy may occur, but not in all patients. Epilepsy is present in most patients, but is rarely a prominent feature. In adult patients, the disease may start with psychiatric symptoms, presenile dementia or complicated migraine. The disease is slowly progressive with superimposed episodes of major and rapid deterioration following stresses such as minor head trauma and, especially, febrile infections [1,17,18]. During these episodes, patients lose motor faculties. Irritability, vomiting, and seizures ensue, usually progressing to somnolence and loss of consciousness. These episodes may end in coma. Death usually follows an episode of coma. If recovery occurs, it is usually incomplete. After infancy, the disease reportedly affects the central nervous system only, except for the ovaries. Ovarian failure is frequent among female patients with VWM and may even precede the neurologic decline [14,17]. Prenatal forms of VWM are characterized by primary microcephaly and extraneurologic signs (renal hypoplasia, hepatosplenomegaly, pancreatitis, and cataract) in addition to ovarian dysgenesis and leukoencephalopathy [10]. Imaging Magnetic resonance imaging (MRI) of the brain (Figure 36.1) shows extensive cerebral white matter signal changes from the presymptomatic stage onward [17,19]. Over time, MRI demonstrates disappearance of the affected white matter, which is eventually replaced by fluid [1,17,18]. On T2-weighted images, abnormal white matter and cystic white matter have similar

438

high signal intensity and cannot be distinguished. Proton density or fluid-attenuated inversion recovery (FLAIR) images are necessary to demonstrate the white matter rarefaction and cystic degeneration. Abnormal white matter has a high signal on proton density and FLAIR images, whereas cystic white matter has low signal intensity, similar to that of cerebrospinal fluid (CSF). Rarefied white matter has intermediate signal intensity, not as low as CSF. Within the rarefied and cystic white matter, radial stripes stretching from the ventricular wall to the subcortical regions are faintly visible, suggesting remaining tissue strands [1,17,18]. Contrast enhancement has not been reported. The cerebellar white matter is often mildly abnormal, but not cystic. The cortical gray matter is always spared. There are often signal abnormalities in the thalamus, midbrain, and pons. Typically, the central tegmental tracts are often involved [1,17,18]. Unlike the cerebral and cerebellar white matter abnormalities, the thalamus and brainstem signal abnormalities when present may improve again and even disappear. MRI-visible spinal cord lesions are rare, but may occur [1,18].

Laboratory findings Laboratory tests are unrevealing in VWM. A consistent abnormality is elevated CSF glycine [20]. This elevation is probably a nonspecific finding related to ongoing excitatory brain damage. CSF asialotransferrin may be decreased [21]. This is thought to be produced exclusively in the brain, possibly by astrocytes and oligodendrocytes, and its reduction may reflect disturbances in these cells.

Pathologic findings The neuropathology of VWM is now well established [2,3, 17,18,22–28]. While cavitating white matter disorders are not synonymous with VWM, there are some archival cases that might also be VWM [29–34]. The neuropathologic features of Cree leukoencephalopathy are comparable, but oligodendroglial hypercellularity and astroglial abnormal morphology have not been reported [4,9].

Macroscopy The brains are generally of normal size and weight, but a few cases are slightly heavier or lighter than controls. At autopsy the gyri are usually normal or atrophic, in contrast to the in vivo imaging studies that may display broad and flattened gyri due to expansion of the white matter. Brain swelling with flattening of the gyri is common in neonates, infants and young children, whereas cortical and subcortical atrophy is common in adults [35]. The lesions are found primarily in cerebral white matter; the cerebellum and brainstem are much less involved. The cerebral white matter appears grayish and varies from gelatinous to cystic to frankly cavitary (Figure 36.2). The frontoparietal white

Vanishing White Matter Disease Chapter 36

Figure 36.1 T2-weighted (upper row) and fluid-attenuated inversion recovery (FLAIR; lower row) magnetic resonance images of a patient with vanishing white matter (VWM; three columns on the right) compared with a healthy child (left column). The T2-weighted images show that in VWM all cerebral white matter has an abnormally high signal from the inital stage. Because cerebrospinal fluid also

has a high signal on T2-weighted images, no distinction can be made between abnormal and vanished white matter. FLAIR images in VWM show the cerebral white matter initially has an abnormally high signal, which gradually decreases, until it is as low as cerebrospinal fluid. In the end stage, all cerebral white matter has vanished in this patient; only ependymal lining and cortex remain.

matter appears to be more severely involved, particularly in the deep and periventricular regions, while the temporal lobe is relatively spared (Figures 36.3, 36.4).Other white matter areas that are characteristically spared include the optic system, corpus callosum, anterior commissure, and internal capsule. The subcortical arcuate fibers also tend to be spared, but not as consistently (Figures 36.2–36.4) [35]. Gray matter structures appear unaffected, and the ventricles are usually of normal size. Rarely, the cerebellar cortex and basal ganglia are atrophic and the ventricles enlarged [26]. The spinal cord has been grossly spared in the few cases reported [17,29], except for one case in which diffuse atrophy was mentioned [33].

Histopathology reveals that white matter oligodendrocytes and astrocytes bear the brunt of the disease. The oligodendrocytes demonstrate a paradoxical increase in some areas and marked loss in others. Increased numbers of oligodendrocytes are found most consistently in the arcuate fibers and other regions bordering the cystic or cavitated lesions; in other patients, morphologically intact or much less affected areas such as the anterior commissure (Figure 36.6), internal capsule or corpus callosum, also demonstrate an increase in oligodendroglial cellularity. One of the first neuropathologic reports of VWM (in a 45-year-old woman) already noted an increased number of oligodendrocytes, confirmed by cell counts, which was so impressive that the authors briefly considered an oligodendroglial neoplasm or designating the lesion as “oligodendrogliomatosis” [34]. Several other groups have confirmed an increase in oligodendrocyte density, often in the absence of obvious mitotic activity [22,25,32,36,37], while others have emphasized their loss [3,28]. A large proportion of these oligodendrocytes have been characterized as oligodendrocyte progenitors that do not mature into myelin-forming cells [22]. The

Histopathology, immunohistochemistry and ultrastructure VWM is a cavitating orthochromatic leukoencephalopathy [35]. Microscopically, the grossly affected white matter of cerebrum and cerebellum has usually shown lack of myelin, thin myelin sheaths, vacuolation, cystic changes, and, rarely, loss of myelin with numerous lipophages (Figure 36.5 ).

439

Developmental Neuropathology

Figure 36.2 Coronal brain slices through the frontal lobe, thalamus, and striatum, and parietal lobe show extensive cavitation of the white matter. Small cysts are present next to the lateral fold of the frontal horn. In the depth of the frontal slice and in the middle frontal gyrus, the white matter is gray and gelatinous. Most of the U-fibers, the internal capsule and the medullary laminae of the striatum are spared. The ependymal wall remains as a thin membrane.

death of oligodendrocytes in VWM has been reported to be apoptotic, but these data were derived from a macrophagerich lesion in an undefined area of pontine white matter, and the authors were not certain about its primary or secondary (Wallerian degeneration) nature [3]. An early study using terminal uridine nucleotide end labeling and activated caspase-3 immunohistochemistry showed that oligodendrocytes in cerebral white matter undergo apoptosis, but also showed immunohistochemical up-regulation in the anti-apoptotic molecule bcl2 [37]. “Foamy” (vacuolated) oligodendrocytes (Figure 36.7) have been identified in both VWM and Cree leukoencephalopathy [4,28] and are proposed to be specific markers for diseases in which this translational defect is operative. These foamy oligodendrocytes are not consistently detected, and are more easily visualized with a periodic acid–Schiff–Alcian blue 8GX stain with which their cytoplasm is light blue and their Golgi is bright red [4]. The vacuolated or foamy oligodendrocytes were initially identified with antibodies to myelin–oligodendrocyte glycoprotein and to myelin basic protein [28]. This same study reported proteolipid protein mRNA in both foamy and “normal” oligodendrocytes, and its expression appeared increased over controls. The “normal” oligodendrocytes expressed proteolipid

440

protein, myelin basic protein, myelin–oligodendrocyte glycoprotein, 2’-3’ cyclic nucleotide 3-phosphodiesterase (CNP-ase) and carbonic anhydrase II [25]. At the ultrastructural level, the vacuoles in foamy oligodendrocytes were reported to be membranous structures associated with mitochondrial membranes, and, in places, contiguous with myelin lamellae [28]. These oligodendrocytes also contained many mitochondria and fingerprint structures (Figure 36.8) [20,28]. The latter two attributes also were noted in nonvacuolated oligodendrocytes. Myelin sheaths are abnormal and vary from pale to thin to vacuolated to absent [2,17,18,24,28]. Vacuolated myelin is noted in areas of preserved myelin, often near the cystic or cavitated lesions. The vacuoles are bordered by Luxol fast blue- or myelin basic protein-positive material, suggesting that the vacuoles contain fluid and reflect intramyelinic edema. Intramyelinic edema was confirmed by ultrastructural studies showing areas of noncompact myelin surrounding the axonal membrane [25,28]. Characteristically, there is a meager response by astrocytes and microglial cells in VWM, even in areas near the cavitation. Lipophages are usually scant, arguing against demyelination. Astrocytes are often reduced in number and typically appear

Vanishing White Matter Disease Chapter 36

(a)

(b)

Figure 36.3 Luxol fast blue (myelin) preparation of coronal section through thalamus and striatum. Severe destruction of corona radiata, pallor of the temporal white matter, partial sparing of the U-fibers, and sparing of the internal capsule and medullary laminae of the striatum can be appreciated (reprinted from van der Knaap et al., 1997 [17], with permission of Wolters Kluwer Health, Inc.).

Figure 36.5 Hematoxylin and eosin-stained sections show spongiform changes and moderate increase of oligodendrocyte numbers in the U-fibers (a) and severe tissue rarefaction in the periventricular white matter (b).

Figure 36.4 Myelin basic protein immunostaining of a coronal section through thalamus and striatum. Pallor, rarefaction of white matter, and paraventricular microcystic change are seen. U-fibers, callosal body, internal capsule, medullary laminae of striatum are spared. The changes are milder than the cases shown in Figures 36.1 and 36.2 (reprinted from van der Knaap et al., 1998 [18], with permission of Wolters Kluwer Health, Inc.).

dysmorphic with blunt broad processes (Figure 36.9) rather than their typical delicate processes, and often have multiple nuclei. These dysmorphic astrocytes are present in cerebral white matter, but are fewer in the relatively spared cerebellum and typically absent in gray matter structures (Figure 36.9) [38]. In primary cell cultures from the brain of a patient with VWM and in normal human glial progenitors manipulated by RNA interference targeting EIF2B5, there were few astrocytes, generation of new astrocytes was compromised and the few astrocytes generated showed an abnormal morphology [39]. A 2013 immunohistochemical study provided evidence that these dysmorphic astrocytes are precursor cells expressing the immature markers nestin and CD44 (Figure 36.10) [38]. Dysmorphic astrocytes characteristically also overexpress the delta isoform of the glial acidic fibrillary protein (GFAP) delta (Figure 36.10) [22]. GFAPδ results from alternative splicing of the GFAP gene

441

Developmental Neuropathology

Figure 36.6 Hematoxylin and eosin section through anterior commissure displaying distinctly increased number of oligodendrocytes.

Figure 36.8 Electron micrograph. Next to the oligodendrocyte nucleus fingerprint-like profiles are seen (reprinted from van der Knaap et al., 1998 [18] with permission of Wolters Kluwer Health, Inc.). (a)

and is normally expressed below detectable levels in white matter astrocytes. In VWM, by contrast, dysmorphic astrocytes overexpress GFAPδ, but not the major isoform GFAPα or total GFAP [22]. This raises the possibility that the intermediate filament network may be disturbed in VWM astrocytes, which could explain their abnormal morphology and the lack of reactive gliosis. While axonal loss is complete in areas of cavitation, less involved areas show a more variable loss of axons. In some, axonal loss has been considered to be as severe as the myelin loss [18], but others have reported axonal sparing [2,25,28] or axonal swellings and spheroids [23,40]. In all but one case [18], blood vessels have been unremarkable, and there is no significant traditional inflammatory response (lymphocytes, plasma cells, neutrophils, eosinophils). As surmised grossly, the gray matter is generally spared or greatly

Figure 36.7 In the center, vacuolated (foamy) oligodendrocytes are visible.

442

(b)

Figure 36.9 Immunostaining for glial fibrillary acidic protein revealing astrocytes with broad, blunted processes in the white matter (a). Astrocytes in the cortical gray matter have a normal morphology (b).

Vanishing White Matter Disease Chapter 36

(a)

(c)

(b)

Figure 36.10 Immunofluorescence microscopy showing dysmorphic white matter astrocytes double-positive for glial fibrillary acidic protein (GFAP) and the immature astrocyte markers nestin (a) and CD44 (b). Immunofluorescence microscopy revealing strong GFAPδ expression in dysmorphic white matter astrocytes (c).

preserved in comparison with white matter. However, some cases have reported astrocytosis and microgliosis of overlying cortex, and there is one report of Alzheimer type II glia in the dentate nucleus [36]. In the brainstem, the central tegmental tract at the level of the pons often shows characteristic bilateral and symmetrical equal loss of axons and myelin (Figure 36.11) [17,18]; other brainstem tracts (e.g., pyramidal and crossing fibers) can show more variable spongy involvement. The spinal cord also has demonstrated some vacuolation or pallor of myelin microscopically [17,33]. Systemic findings have been nonspecific and due to terminal events.

it is highly suggestive, if not diagnostic, of this disease. When the extensive gross cavitation is combined with the presence of dysmorphic astrocytes and a true microscopic increase in the number of oligodendrocytes, particularly if vacuolated or foamy, then a firm neuropathologic diagnosis can be made. Time

Differential diagnosis A cystic to cavitating alteration of cerebral white matter can be seen in Alexander disease and mitochondrial leukoencephalopathies, and, occasionally, in severe forms of adrenoleukodystrophy and Krabbe disease, as well as in other leukoencephalopathies (e.g., vascular lesions), in which both myelin sheaths and their axons are severely depleted or totally lost. The extent of cavitation seen in VWM, appears to be so characteristic that

Figure 36.11 Luxol fast blue staining of a section through the pons. The central tegmental tracts show symmetrical pallor (reprinted from van der Knaap et al., 1997 [17] with permission of Wolters Kluwer Health, Inc.).

443

Developmental Neuropathology and further investigations may prove the dysmorphic astrocytes and vacuolated or foamy oligodendrocytes to be pathognomonic of VWM.

Experimental models Naturally occurring animal models have not been described. A knock-in strategy has been employed to generate a mouse model for VWM [41]. This model carries a homozygous R132H point mutation in Eif2b5, corresponding to the human R136H mutation associated with a childhood-onset form of disease. Eif2b5R132H/R132H mice have no clinical phenotype, but show higher proportions of small-caliber nerve fibers, fail to recover from cuprizone-induced toxic demyelination and show meager astrogliosis after injection of lipopolysaccharide [41,42].

Pathogenesis Translation of mRNA into polypeptides is one of the major energy-consuming processes in the cell and is, therefore, a tightly regulated process [43]. The initiation phase, in which ribosomes are assembled on mRNA, is controlled via several different signaling pathways [44]. Multiple so-called eukaryotic initiation factors (eIFs) are involved in translation initiation in which a crucial step is the delivery by eIF2 of the methionyl initiator transfer RNA (Met-tRNAi) to the small ribosomal subunit. Upon recognition of the start codon, the eIF2bound guanosine triphosphate (GTP) is hydrolyzed and eIF2 is released in its inactive guanosine diphosphate (GDP)-bound form. To bind another Met-tRNAi, active eIF2 must be regenerated by exchange of GDP for GTP. This step is catalyzed by the guanine nucleotide exchange factor eIF2B, a factor that plays a key role in translation initiation [45]. The exchange of GDP for GTP by eIF2B is required for each round of translation initiation, and regulation of this step can control global rates of protein synthesis under diverse conditions [46]. Protein synthesis is markedly inhibited under a variety of stress conditions and in the recovery phase that follows. This response is part of a cell-protective mechanism elicited by various stimuli, including thermal, chemical, oxidative or physical trauma, called the cellular stress response [47]. Stress may lead to misfolding and denaturation of proteins, contributing to cell dysfunction and death. The inhibition of normal RNA translation during stress is thought to enhance cell survival by limiting the accumulation of denatured proteins and saving cellular energy. Inhibition of mRNA translation can be achieved through the modification of several initiation factors. Most stress conditions, including heat stress [48,49], lead to the activation of specific kinases that phosphorylate eIF2 on its α-subunit. In its phosphorylated form, eIF2 is a competitive inhibitor of eIF2B, preventing the recycling of eIF2 [50]. The concentration of eIF2 exceeds that

444

of eIF2B [51]. Therefore, even modest levels of eIF2α phosphorylation can potentially lead to a complete inhibition of translation initiation and protein synthesis [51,52]. Inactivation of eIF2B at 40–41◦ C, in the febrile range for human beings, can be achieved without changes in eIF2α phosphorylation in certain cell types [49]. eIF2B activity can also be regulated through other pathways, such as phosphorylation at different sites, which can enhance or suppress eIF2B activity [43]. Whether these latter pathways are involved in the regulation of eIF2B activity under stress conditions is unclear. The essential role of eIF2B, both in normal protein production and in its regulation under different conditions, including elevated temperature, is reflected by the evolutionary conservation of the complex [50] and the nonviability of yeast null mutants for each of the subunits except eIF2Bα [46]. In patients with VWM, serious deteriorations often follow febrile infections, which could correlate to the regulating role of eIF2B on translation upon heat stress. However, the pathogenesis of the disease is still poorly understood. The functional effects of VWM mutations are very diverse, including defects in eIF2B complex integrity, binding to the regulatory α-subunit, substrate binding, and guanine nucleotide exchange factor activity. Some mutations do not affect these parameters at all, even though they cause severe disease [53]. Decreased eIF2B activity does not affect the rate of global protein synthesis, the regulation of protein synthesis during or after stress, or the ability of patients’ cells to proliferate and survive [54–56]. Finally, it is still unclear why the brain is selectively affected, whereas other organs with high metabolic activity, such as the liver and the bone marrow, are spared.

Future directions and therapy At present, there is no effective treatment for VWM. It is advisable to prevent or avoid the known stress conditions that may provoke an episode of deterioration: fever and head trauma.

References 1. van der Knaap MS, Pronk JC, Scheper GC (2006) Vanishing white matter disease. Lancet Neurol 5:413–23 2. Schiffmann R, Moller JR, Trapp BD et al. (1994) Childhood ataxia with diffuse central nervous system hypomyelination. Ann Neurol 35:331–40 3. Br¨uck W, Herms J, Brockmann K et al. (2001) Myelinopathia centralis diffusa (vanishing white matter disease): evidence of apoptotic oligodendrocyte degeneration in early lesion development. Ann Neurol 50:532–6 4. Fogli A, Wong K, Eymard-Pierre E et al. (2002) Cree leukoencephalopathy and CACH/VWM disease are allelic at the EIF2B5 locus. Ann Neurol 52:506–10 5. Schiffmann R, Tedeschi G, Kinkel P et al. (1997) Leukodystrophy in patients with ovarian dysgenesis. Ann Neurol 41:654–61

Vanishing White Matter Disease Chapter 36 6. Fogli A, Boespflug-Tanguy O. The large spectrum of eIF2B-related diseases (2006) Biochem Soc Trans 34:22–9 7. Van der Knaap MS, Breiter SN, Naidu S et al. (1999) Defining and categorizing leukoencephalopathies of unknown origin: MR imaging approach. Radiology 213:121–33 8. van der Lei HD, van Berkel CG, van Wieringen WN et al. (2010) Genotype–phenotype correlation in vanishing white matter disease. Neurology 75:1555–9 9. Black DN, Booth F, Watters GV et al. (1988) Leukoencephalopathy among native Indian infants in northern Quebec and Manitoba. Ann Neurol 24:490–6 10. van der Knaap MS, van Berkel CG, Herms J et al. (2003) eIF2Brelated disorders: antenatal onset and involvement of multiple organs. Am J Hum Genet 73:1199–207 11. Leegwater PA, Vermeulen G, K¨onst AA et al. (2001) Subunits of the translation initiation factor eIF2B are mutant in leukoencephalopathy with vanishing white matter. Nature Genet 29:383–8 12. Van der Knaap MS, Leegwater PA, K¨onst AA et al. (2002) Mutations in each of the five subunits of translation initiation factor eIF2B can cause leukoencephalopathy with vanishing white matter. Ann Neurol 51:264–70 13. Fogli A, Dionisi-Vici C, Deodato F et al. (2002) A severe variant of childhood ataxia with central hypomyelination/vanishing white matter leukoencephalopathy related to EIF21B5 mutation. Neurology 59:1966–8 14. Fogli A, Rodriguez D, Eymard-Pierre E et al. (2003) Ovarian failure related to eukaryotic initiation factor 2B mutations. Am J Hum Genet 72:1544–50. 15. Leegwater PA, K¨onst AA, Kuyt B et al. (1999) The gene for leukoencephalopathy with vanishing white matter is located on chromosome 3q27. Am J Hum Genet 65:728–34 16. Hanefeld F, Holzbach U, Kruse B et al. (1993) Diffuse white matter disease in three children: an encephalopathy with unique features on magnetic resonance imaging and proton magnetic resonance spectroscopy. Neuropediatrics 24:244–8 17. Van der Knaap MS, Barth PG, Gabreels FJ et al. (1997) A new leukoencephalopathy with vanishing white matter. Neurology 48:845–55 18. Van der Knaap MS, Kamphorst W, Barth PG et al. (1998) Phenotypic variation in leukoencephalopathy with vanishing white matter. Neurology 51:540–7 19. van der Lei HD, Steenweg ME, Barkhof F et al. (2012) Characteristics of early MRI in children and adolescents with vanishing white matter. Neuropediatrics 43:22–6 20. Van der Knaap MS, Wevers RA, Kure S et al. (1999) Increased cerebrospinal fluid glycine: a biochemical marker for a leukoencepohalopathy with vanishing white matter. J Child Neurol 14: 728–31 21. Vanderver A, Schiffmann R, Timmons M et al. (2005) Decreased asialotransferrin in cerebrospinal fluid of patients with childhood-onset ataxia and central nervous system hypomyelination/vanishing white matter disease. Clin Chem 51:2031–42 22. Bugiani M, Boor I, van Kollenburg B et al. (2011) Defective glial maturation in vanishing white matter disease. J Neuropathol Exp Neurol 70:69–82 23. Francalanci P, Eymard-Pierre E, Dionisi-Vici C et al. (2001) Fatal infantile leukodystrophy: a severe variant of CACH/ VWM syndrome, allelic to chromosome 3q27. Neurology 57: 265–70

24. Harding BN, Surtees R (2002) Metabolic and neurodegenerative diseases of childhood. In: DI Graham, PL Lantos, eds, Greenfield’s Neuropathology, 7th ed. London, Arnold, pp. 485–517 25. Rodriguez D, Gelot A, della Gaspera B et al. (1999) Increased density of oligodendrocytes in childhood ataxia with diffuse central hypomyelination (CACH) syndrome: neuropathological and biochemical study of two cases. Acta Neuropathol 97:469–80 26. Sugiura C, Miyata H, Oka A et al. (2001) A Japanese girl with leukoencephalopathy with vanishing white matter. Brain Develop 23:56–61 27. Topc¸u I, Saatci I, Anil Apak R, S¨oylemezoglu F (2000) A case of leukoencephalopathy with vanishing white matter. Neuropadiatrics 31:100–3 28. Wong K, Armstrong RC, Gyure KA et al. (2000) Foamy cells with oligodendroglial phenotype in childhood ataxia with diffuse central nervous system hypomyelination syndrome. Acta Neuropathol 100:635–46 29. Anzil AP, Gessaga E (1972) Late-life cavitating dystrophy of the cerebral and cerebellar white matter. Eur Neurol 7:79–94 30. Deisenhammer E, Jellinger K (1976) H¨ohlen-bindende neurtralfettleukodystrophie mit schubverlauf. Neuropediatrics 7:111–21 31. Eicke WJ (1962) Polycystische umwandlung des marklagers mit progredientem verlauf. Atypische diffuse sklerose? Arch Psychiat Nervenkr 203:599–602 32. Gautier JC, Gray F, Awada A, Escourolle R (1984) Leucodystrophie orthichromatique cavitaire de l’adulte. Prolif´eration et inclusions oligodendrogliales. Rev Neurol (Paris) 140:493–501 33. Girard PF, Tommassi M, Rochet M, Boucher M (1968) Leucoenc´ephalopathie avec cavitations massives, bilat´erales et sym´etriques. Syndrome de d´ecortication post-traumatique. Press Med 76:163–6 34. Watanabe I, Muller J (1967) Cavitating “diffuse sclerosis.” J Neuropathol Exp Neurol 26:437–55 35. Bugiani M, Boor I, Powers JM et al. (2010) Leukoencephalopathy with vanishing white matter: a review. J Neuropathol Exp Neurol 69:987–96 36. Graveleau P, Gray F, Plas J et al. (1985) Leucodystrophie orthochromatique cavitaire avec modifications oligodendrogliales. Un cas sporadique adulte. Rev Neurol (Paris) 141:713–18 37. Van Haren KP, van der Voorn P, van der Knaap MS, Powers JM (2004) Life and death of oligodendrocytes in vanishing white matter disease. J Neuropathol Exp Neurol 63:618–30 38. Bugiani M, Postma N, Polder E et al. (2013) Hyaluronan accumulation and arrested oligodendrocyte progenitor maturation in vanishing white matter disease. Brain 136:209–22 39. Dietrich J, Lacagnina M, Gass D et al. (2005) EIF2B5 mutations compromise GFAP+ astrocyte generation in vanishing white matter leukodystrophy. Nat Med 11:277–83 40. Prass K, Bruck W, Schroder NW et al. (2001) Adult onset leukoencephalopathy with vanishing white matter presenting with dementia. Ann Neurol 50:665–8 41. Geva M, Cabilly Y, Assaf Y et al. (2010) A mouse model for eukaryotic translation initiation factor 2B-leucodystrophy reveals abnormal development of brain white matter. Brain 133:2448–61 42. Cabilly Y, Barbi M, Geva M et al. (2012) Poor cerebral inflammatory response in eIF2B knock-in mice: implications for the aetiology of vanishing white matter disease. PLoS One 7:e46715 43. Proud CG (2002) Regulation of mammalian translation factors by nutrients. Eur J Biochem 269:5338–49

445

Developmental Neuropathology 44. Kleijn M, Scheper GC, Voorma HO, Thomas AAM (1998) Regulation of translation initiation factors by signal transduction. Eur J Biochem 253:531–44 45. Proud CG (2001) Regulation of eukaryotic initiation factor eIF2B. Prog Mol Subcell Biol 26:95–114 46. Hinnebusch AG (2000) Mechanism and regulation of initiator methionyl-tRNA binding to ribosomes. In: N Sonenberg, JWB Hershey, MB Mathews, eds, Translational Control of Gene Expression. Cold Spring Harbor, NY, Cold Spring Harbor Laboratory Press, pp. 185–243 47. Welch WJ (1992) Mammalian stress response: cell physiology, structure/function of stress proteins, and implications for medicine and disease. Physiol Rev 72:1063–81 48. Duncan RF, Hershey JW (1989) Protein synthesis and protein phosphorylation during heat stress, recovery, and adaptation. J Cell Biol 109:1467–81 49. Scheper GC, Mulder J, Kleijn M et al. (1997) Inactivation of eIF2B and phosphorylation of PHAS-I in heat-shocked rat hepatoma cells. J Biol Chem 272:26850–6 50. Hershey JWB, Merrick WC (2000) The pathway and mechanism of initiation of protein synthesis. In: N Sonenberg, JWB Hershey, MB Mathews, eds, Translational Control of Gene Expression.

446

51.

52.

53.

54.

55.

56.

Cold Spring Harbor, NY, Cold Spring Harbor Laboratory Press, pp. 33–88 Oldfield S, Jones BL, Tanton D, Proud CG (1994) Use of monoclonal antibodies to study the structure and function of eukaryotic protein synthesis initiation factor eIF-2B. Eur J Biochem 221:399–410 Rowlands AG, Montine KS, Henshaw EC, Panniers R (1988) Physiological stresses inhibit guanine-nucleotide-exchange factor in Ehrlich cells. Eur J Biochem 175:93–9 Liu R, van der Lei HD, Wang X et al. (2011) Severity of vanishing white matter disease does not correlate with deficits in eIF2B activity or the integrity of eIF2B complexes. Hum Mutat 32:1036–45 Krishnamoorthy T1, Pavitt GD, Zhang F et al. (2001) Tight binding of the phosphorylated alpha subunit of initiation factor 2 (eIF2alpha) to the regulatory subunits of guanine nucleotide exchange factor eIF2B is required for inhibition of translation initiation. Mol Cell Biol 21:5018–30 van Kollenburg B1, Thomas AA, Vermeulen G et al. (2006) Regulation of protein synthesis in lymphoblasts from vanishing white matter patients. Neurobiol Dis 21:496–504 Kantor L, Harding HP, Ron D et al. (2005) Heightened stress response in primary fibroblasts expressing mutant eIF2B genes from CACH/VWM leukodystrophy patients. Hum Genet 118:99–106

37

Alexander Disease James E. Goldman1 and Mel B. Feany2 1 Department 2 Department

of Pathology and Cell Biology, Columbia University, New York, NY, USA of Pathology, Brigham and Women’s Hospital, Harvard Medical School, Boston, MA, USA

Definition Alexander disease is a leukodystrophy that primarily affects children. The typical form presents between birth and two years of age, and produces a progressive psychomotor restriction and megalencephaly, and death usually within the first decade. There are also rarer juvenile and adult onset forms. Most cases are caused by dominant mutations in the gene for glial fibrillary acidic protein (GFAP), which encodes an intermediate filament protein of astrocytes.

become manifest late in the first decade or in the teenage years, and rarer late-onset forms, which present in adult life. There are no known risk factors.

Embryology GFAP is expressed in radial glia throughout the embryonic primate brain. Thus, abnormal accumulations of the GFAP protein may begin during fetal development. This accumulation apparently has no effect on central nervous system (CNS) development itself, since patients do not show gross or microscopic abnormalities of cortical gyration, neuronal migration, or location of specific nuclei.

Synonyms and historical annotations Genetics Alexander disease was first described in 1949 [1], and was called “progressive fibrinoid degeneration of fibrillary astrocytes associated with mental restriction and hydrocephalus.” Other early reports termed the disease, “dysmyelinogenic leukodystrophy with megalobarencephaly” [2] or “megalencephaly associated with hyaline pan-neuropathy” [3]. The eponym was applied first in 1964 by Friede [4].

Epidemiology It is difficult to estimate incidence and prevalence for such a rare disorder. A single study of prevalence has been reported, with an estimate of 1 case per 2.7 million [5]. More than 120 different mutations have been reported in approximately 250 patients to date. Both males and females are affected in approximately equal numbers. The majority of patients present within the first two years of life, although there are juvenile forms, which

Patients with Alexander disease are typically heterozygous for single base pair mutations in the GFAP gene [6–12]. Most mutations lead to amino acid substitutions, although a small number of deletions and insertions have been described [13–17]. GFAP is a member of a class of intermediate filament proteins, which polymerize into filaments of 10 < ns > nm diameter [18]. Within the CNS, GFAP is expressed only in astrocytes. Each member of the intermediate filament family has a relatively conserved central rod region, which interacts side to side to form the core of the filament, and variably sized N-terminal head and C-terminal tail regions. Mutations are found predominantly throughout the rod region, but have also been found in the head and tail domains (Figure 37.1). Since patients are heterozygous, the mutations are dominant. The parents of all patients with infantile or childhood onset Alexander disease are themselves neurologically normal, and those whose GFAP has been sequenced are wild type at the site mutated in their child. Thus, the common forms of Alexander disease represent de novo mutations. There are a few

Developmental Neuropathology, Second Edition. Edited by Homa Adle-Biassette, Brian N. Harding, and Jeffrey A. Golden. © 2018 John Wiley & Sons Ltd. Published 2018 by John Wiley & Sons Ltd.

447

Developmental Neuropathology

N

early onset

K R T T M73 V F F L76 Y S S N77 L S G C C C C C C C C C C C C H H H H H H H H H P H I C C C C C C C C C S P P P P

R79 Y83 V87 R88 L90 Q93 L97

1 A

late onset K63 E69 R70

Q K WWW Q

M74

T T

D78 R79

E H H C

V87 R88 L101

G G L C C P

D128

N

E205 E207 E210

K K Q K K K K

K86_V87 delinsEF

R126_L127dupRL

1 B

P C C C C C C C C C H H H H H H H H H L L P P P C C C C C C C C C D V G

L235 R239 Y242 A244 A253

L231 H L235 P R239 C 2A

P R258

Y349_Q350insHL

E K279 P L331

L352 A358 L359 D360 E362 V A364 Y366 C E371 Q K K K E373 E374 N386 W R416 R417fsX431 P V V V D P H G K G I W

2 B

C Figure 37.1 Locations of glial fibrillary acidic protein (GFAP) mutations in Alexander disease in relation to the protein structure. The four open rectangular boxes represent the helical coiled-coil rod domains of GFAP; these structural motifs are highly conserved among most intermediate filament proteins. The solid lines joining these segments are nonhelical linker regions, and the solid lines at either end are the nonconserved, random coil, N-terminal and C-terminal regions. The gray box just before segment 1A is a nonconserved prehelical sequence important for initiation of rod formation at the start of 1A; the gray box at the end of 2B represents a highly conserved sequence that includes the end of the coiled coil 2B

448

A244 S247 Y257 R258 L264 A267 R276 E332 D351 L357 L359 R376 T383

V P C C P P [R330G; E332K] L K [E332Kdel] H P P P W I

S393 I S398 Y F R416 W W W W W W W D417 A Q426 L

segment. The wild-type amino acid is indicated next to the structure, and amino acid replacements within symbols on either side. Early-onset cases (first symptom before the age of 2 years) are on the left, shown as blue circles, and late-onset cases (first symptom after the age of 2 years) are on the right, shown as red circles. Each symbol represents a single patient, except that familial cases, including identical twins, are represented by a single symbol coded for the onset type of the proband (adapted with permission from Brenner et al. [11] and Messing et al. [12]). Reproduced with permission of Springer.

Alexander Disease Chapter 37

Figure 37.2 Typical magnetic resonance imaging for Alexander disease in a patient juvenile onset disease, illustrating the characteristic abnormalities of frontal white matter including increased signal on T2-weighted images and decreased signal on T1-weighted images (courtesy of Dr. M.S. van der Knaap, VU University Medical Center, Amsterdam, the Netherlands).

families with adult onset Alexander disease in which the mutation is inherited in a dominant Mendelian fashion [19–21].

Clinical features Signs and symptoms The typical infantile form of the disease presents with progressive psychomotor restriction, which is accompanied by megalencephaly, seizures, and eventually spasticity and quadriparesis [1,3,4, 22–29]. The juvenile form is likely to present with bulbar and cerebellar signs, including dysarthria, dysphagia, pernicious vomiting, hiccupping, and ataxia, with eventual cognitive deterioration [23,26,30–34]. Patients with the adult onset form have a slower progression and a variety of signs and symptoms, including ataxia, spastic para- or quadriparesis, abnormal eye movements, and palatal myoclonus, also with eventual cognitive decline [20,21,35–38]. Imaging Magnetic resonance imaging (MRI) of children with Alexander disease shows marked and widespread changes in the white

matter, with the most pronounced pathology in the frontal hemispheres, usually characterized by low T1 and high T2 signals (Figure 37.2) [26,39]. Often, there are cystic changes in the aberrant white matter. Scans also show a high T1 signal and low T2 signal in the immediate periventricular region, abnormalities in the basal ganglia and brain stem, both usually showing increased T2 signal. As the disease progresses, cystic change in the hemispheric white matter becomes more pronounced, the cortical ribbon thins, the ventricles enlarge, and the basal ganglia, thalamus, brainstem, and cerebellum lose volume. There is a variable amount of contrast enhancement, often in the periventricular tissues. Older children and adults with Alexander disease do not necessarily show these hemispheric changes, but display abnormalities in the medulla, spinal cord, and cerebellum [13].

Laboratory findings The only specific laboratory finding is a mutation in GFAP. These have been found in about 95% of cases examined, and are considered a definitive diagnosis [11]. However, absence of a GFAP mutation does not rule out the disorder. GFAP levels are increased in cerebrospinal fluid and blood, but variability in levels from patient to patient makes GFAP more suitable as

449

Developmental Neuropathology a potential biomarker for therapy monitoring than a diagnostic test [17].

Macroscopy Note that descriptions of gross and microscopic pathology prior to 2001 are not accompanied by GFAP mutational analysis. We include these descriptions because of the characteristic pathological changes. However, it is possible that some of these patients would not have a GFAP mutation. The brains of infants with rapid clinical courses display severe hemispheric white matter destruction, often to the extent of cavitation, and form little myelin [25,29]. Patients in the infant years with more prolonged courses or those who develop clinical manifestations in early childhood typically have enlarged brains with widespread white matter loss in the hemispheres, brain stem, cerebellum, and cord, the extent of which is dependent upon the period of survival [1–4,23,28]. In many children, degenerative changes in the frontal lobe are initially more severe than those in the parietal and occipital lobes. Those children who survive a number of years also show greater degrees of cortical thinning and basal ganglia and thalamic atrophy. The myelination of arcuate fibers appears relatively spared. Older children and adults show patchy to severe loss of myelin and in some cases, even gliovascular scars in hemispheric and cerebellar white matter. There appears to be a predilection for brain stem and cerebellar pathology in older children and adults [20,30,35,36].

Histopathology The signature histopathology of Alexander disease is the presence of enormous numbers of Rosenthal fibers in the setting of a lack of or loss of myelin (Figure 37.3a–d). Rosenthal fibers are intracellular inclusions within astrocytes and appear on routine hematoxylin and eosin-stained sections as eosinophilic, refractile, oval to rod-shaped profiles that vary in size from well under a micron to tens of microns in length (Figure 37.3a,b). They accumulate particularly around blood vessels in white matter and at the pial surface, locations reflecting their presence in astrocyte end-feet. Astrocytes in the neocortex do not accumulate Rosenthal fibers to this degree, but often display very small inclusions (Figure 37.3c). Astrocyte processes and end-feet swell with Rosenthal fibers in many other parts of the CNS as well, including striatum, thalamus, cerebellum, brain stem (particularly the tegmentum, Figure 37.3b), and spinal cord. These inclusions were first described by Rosenthal in 1898 [40], but in the gliosis surrounding syrinx cavities, and not in a patient with a leukodystrophy. In infants, myelination does not occur normally; in fact, the white matter can appear partially or entirely cystic, containing

450

enlarged astrocytes bearing small Rosenthal fibers in their cell bodies (Figure 40.3d), macrophages, and scattered oligodendrocytes, but little myelin [23–25,29]. Children who present in early childhood and who survive a number of years show large areas of myelin pallor in the white matter, accompanied by a gliovascular scar with many Rosenthal fibers (Figure 37.3a) [23,28]. It is not clear whether those areas ever began to myelinate initially; in infants, it is likely that little myelination actually took place. In children who develop the disorder after infancy, some degree of myelination undoubtedly has occurred. Arcuate fibers are relatively spared, as is the case in many leukodystrophies. There is a variable degree of axonal loss throughout the affected white matter, ranging from minimal to severe, but axonal loss has not been systematically investigated. Neuronal loss also appears to be variable, with reports of severe loss in cortex, striatum, hippocampus, and brain stem in a few cases. Although case reports have mentioned some degree of oligodendrocyte loss, this has also not been investigated systematically. Some Alexander brain tissues show lymphocytic cuffing around blood vessels, but other cases do not, so this type of inflammatory pathology is variable. Both human and mouse Alexander disease contain large numbers of activated microglia, indicating a widespread inflammatory response [41]. Patients with a juvenile form with brain stem and cerebellar signs can show Rosenthal fiber deposition in the tegmentum of the brain stem and in the molecular layer and deep white matter of the cerebellum, in addition to patchy myelin pallor and gliosis with Rosenthal fibers in the hemispheric white matter [30–33]. Patients with the adult form of Alexander disease show patches of myelin pallor in the deep white matter or small cavities or in some cases, more widespread myelin pallor with gliosis and Rosenthal fiber accumulation [36,37]. There are no descriptions of specific extra-CNS pathology in Alexander disease.

Immunohistochemistry and ultrastructural findings Rosenthal fibers appear ultrastructurally to be osmiophilic deposits without notable internal structure lying in the astrocyte cytoplasm [42]. They are surrounded by, and in intimate contact with, normal-appearing 10-nm intermediate filaments (Figure 37.4). Rosenthal fibers are neither membrane-bound nor associated with any membranous organelles. The molecular components of Rosenthal fibers include GFAP itself and the two small heat shock proteins, αB-crystallin and hsp27, plectin, and the 20S proteasome subunit [43–48]. The filaments at the periphery are composed of both GFAP and vimentin [49], the latter being another intermediate filament type also expressed by astrocytes. Typically, antibodies against GFAP and small heat shock proteins react with the periphery of Rosenthal fibers at the light microscopic level, but this pattern likely reflects difficulty

Alexander Disease Chapter 37

(a)

(b)

(c)

(d)

Figure 37.3 Light microscopic appearance of Alexander disease tissues. (a) Deep hemispheric white matter contains many Rosenthal fibers, many surrounding blood vessels, and a paucity of myelin. Insert: Higher magnification of Rosenthal fibers, astrocyte nuclei (open chromatin pattern) and (presumed) dark oligodendrocyte nuclei. (b) Rosenthal fibers accumulate at subventricular location, here shown in the periacqueductal tissue. (c) The neocortex usually contains sparse and small Rosenthal fibers. (d) In infantile Alexander disease, central nervous system tissues examined early in life show small Rosenthal fibers in astrocyte cell bodies (adapted from Messing et al. [67]).

with antibody penetration into the dense fiber, since electron microscopy immunocytochemistry with the same antibodies reacts over the entire Rosenthal fiber matrix [45].

Steady-state mRNA levels for the small heat shock proteins are elevated, suggesting a chronic stress state.

Differential diagnosis Biochemistry There are no specific biochemical abnormalities in patients with Alexander disease that could be used for diagnostic purposes, except the GFAP mutations. Analyses of CNS tissues show a massive accumulation of GFAP, αB-crystallin, and hsp27 proteins, clearly related to the numbers of Rosenthal fibers (50).

The differential diagnosis includes a number of the leukodystrophies, although few present in early childhood with megalencephaly. Many of the other leukodystrophies can be ruled out by biochemical tests in conjunction with MRI. While the diagnosis of Alexander disease in the past required a brain biopsy, at present a diagnosis suspected on the basis of clinical and MRI findings, can be confirmed by GFAP sequencing.

451

Developmental Neuropathology

Figure 37.4 Ultrastructural appearance of Rosenthal fibers, dense, osmiophilic structures surrounded by skeins of intermediate filaments. These fibers are observed in an astrocyte cell body from a 17-month-old child with Alexander disease.

Experimental models There have been rare reports in the neuropathology literature of nonhuman mammals suffering from progressive neurological deterioration that looked pathologically similar to Alexander disease. Transgenic mice that overexpress human GFAP driven by the human GFAP promoter and knock-in mice that model the human R239H mutation accumulate Rosenthal fibers but do not apparently show myelination abnormalities [51,52]. A mouse generated by the mating of these two lines experiences seizures and dies in about 30 days [51]. A Drosophila model based on overexpression of human wild-type and Alexander disease-linked mutant GFAP in glia has been created and used for genetic analysis [53,54].

Pathogenesis The remarkable accumulation of GFAP in Alexander brain tissue in the context of a heterozygous mutation in the gene suggests that the mutant GFAP transcription is altered, or the synthesis of the protein is increased or the turnover decreased. The R239C mutant protein accumulates in cultured cells to a far greater extent than the wild type, and also polymerizes more efficiently in vitro, suggesting that turnover may be slowed. In

452

fact, accumulation of oligomeric forms of GFAP result in significant inhibition of proteasome activity [46,55–57]. Alexander tissues also show a striking increase in steady-state transcript levels for αB-crystallin and hsp27 [50]. This suggests that the increased levels of these proteins is due in part to transcriptional activation, and not just the accumulation of these proteins in the Rosenthal fiber and a decrease in turnover. The activation of small heat shock protein transcription appears to be a component of a wider stress response characterized by upregulation of MAPK pathways, constitutive activation of JNK and p38 kinases and increased autophagy [46,55,58]. αB-crystallin, a molecular chaperone, normally binds to intermediate filaments in astrocytes [59], and can unbundle skeins of filaments when overexpressed in astrocytes [60,61]. Similar conclusions have been reached from in vitro GFAP polymerization assays [62]. Thus, αB-crystallin may serve in some way to keep filaments in their normally spread architecture. Consistent with a chaperone function for αB-crystallin toward GFAP, the reduced expression of αB-crystallin enhances toxicity, while the overexpression of αBcrystallin protects from toxicity of GFAP in mouse models [63]. However, the development of Rosenthal fibers, which contain both GFAP and αB-crystallin, appears not to fit a model of filament unbundling by the chaperone. Rosenthal fibers represent a massive aggregation of filaments and chaperones. It may be that the chaperone-mediated unbundling of filaments occurs at early stages of GFAP accumulation, but at some point, levels of GFAP and αB-crystallin increase so much that aggregation begins to take place. It is important to note that the genesis of Rosenthal fibers does not depend on the presence of a mutant form of GFAP. These inclusions are found in pilocytic astrocytomas and in glial scars in patients who do not have Alexander disease. The transgenic mouse that overexpresses wild-type human GFAP forms Rosenthal fibers. Thus, the important variable appears to be a large accumulation of GFAP, either normal or normal plus mutant. Alexander disease is unusual among neurological disorders in the sense that the mutant GFAP protein is expressed in a cell type (astrocyte) that appears to survive, but not expressed in other cell types (oligodendrocytes and neurons) that degenerate. Thus, the accumulation of GFAP in astrocytes is somehow toxic to other CNS cells. We do not know full details of the mechanisms yet, but reduction in the levels of the major astrocytic glutamate transporter, GLT-1, is likely to play a role [64]. Seizures are a prominent feature of patients with the infantile form of Alexander disease, which may reflect inadequate glutamate buffering by astrocytes as well as the sequelae of excitotoxic neuronal loss as the disease progresses. Loss of GLT-1 is unlikely to be the only physiological dysfunction in affected astrocytes. Astrocytes in Alexander disease also show a reduced capacity for potassium buffering, a loss of astrocyte-astrocyte coupling, an increased expression and secretion of chemokines and early response proteins, and a significant change in cell shape, in which astrocytes lose their fine processes that normally enwrap synapses [64,65]. Since GFAP accumulation leads to a stress response, the cells

Alexander Disease Chapter 37

may begin to express a variety of molecules, including cytokines and chemokines, that may be toxic to other cells [41,54,66]. Alternatively, astrocytes in Alexander disease may cease producing growth or survival factors for oligodendrocytes or neurons.

12.

Future directions and therapy

14.

There is currently no therapy for Alexander disease. Parents who bear a child with the disorder can be counseled that there is an extremely low probability of having another affected child. In utero testing for a GFAP mutation is possible for subsequent pregnancies if desired. Future directions center on understanding the mechanism of degeneration, with the view of illuminating altered astrocyte physiology. Manipulation of astrocyte glutamate transport may represent one therapeutic opportunity arising from recent advances in understanding cellular defects in Alexander disease astrocytes. Inhibiting the inflammatory response may be another adjunctive strategy. Additionally, since the absence of GFAP does not appear to be deleterious, based on studies with knockout mice, one might imagine a potential therapy to inhibit GFAP transcription or GFAP synthesis.

References 1. Alexander WS (1949) Progressive fibrinoid degeneration of fibrillary astrocytes associated with mental retardation in a hydrocephalic infant. Brain J Neurol 72:373–81 2. Wohlwill FJ, Bernstein J, Yakovlev PI (1959) Dysmyelinogenic leukodystrophy; report of a case of a new, presumably familial type of leukodystrophy with megalobarencephaly. J Neuropathol Exp Neurol 18:359–83 3. Crome L (1953) Megalencephaly associated with hyaline panneuropathy. Brain J Neurol 76:215–28 4. Friede RL (1964) Alexander’s disease. Arch Neurol 11:414–22 5. Yoshida T, Sasaki M, Yoshida M et al. (2011) Nationwide survey of Alexander disease in Japan and proposed new guidelines for diagnosis. J Neurol 258:1998–2008 6. Aoki Y, Haginoya K, Munakata M et al. (2001) A novel mutation in glial fibrillary acidic protein gene in a patient with Alexander disease. Neurosci Lett 312:71–4 7. Brenner M, Johnson AB, Boespflug-Tanguy O et al. (2001) Mutations in GFAP, encoding glial fibrillary acidic protein, are associated with Alexander disease. Nat Genet 27:117–20 8. Gorospe JR, Naidu S, Johnson AB et al. (2002) Molecular findings in symptomatic and pre-symptomatic Alexander disease patients. Neurology 58:1494–500 9. Li R, Messing A, Goldman JE, Brenner M (2002) GFAP mutations in Alexander disease. Int J Dev Neurosci 20:259–68 10. Rodriguez D, Gauthier F, Bertini E et al. (2001) Infantile Alexander disease: spectrum of GFAP mutations and genotype–phenotype correlation. Am J Hum Genet 69:1134–40 11. Brenner M, Goldman JE, Quinlan RA, Messing A (2009) Alexander disease: a genetic disorder of astrocytes. In: V Parpura, Haydon

13.

15.

16. 17.

18. 19.

20.

21.

22.

23. 24. 25. 26. 27.

28.

29. 30. 31.

32. 33.

PG, eds, Astrocytes in (Patho)Physiology of the Nervous System. New York, NY, Springer. pp. 591–648 Messing A, Brenner M, Feany MB et al. (2012) Alexander disease. J Neurosci 32:5017–23 van der Knaap MS, Ramesh V, Schiffmann R et al. (2006) Alexander disease: ventricular garlands and abnormalities of the medulla and spinal cord. Neurology 66:494–8 Jefferson RJ, Absoud M, Jain R et al. (2010) Alexander disease with periventricular calcification: a novel mutation of the GFAP gene. Dev Med Child Neurol 52:1160–3 Flint D, Li R, Webster LS, Naidu S et al. (2012) Splice site, frameshift, and chimeric GFAP mutations in Alexander disease. Hum Mutat 33:1141–8 Schmidt H, Kretzschmar B, Lingor P et al. (2013) Acute onset of adult Alexander disease. J Neurol Sci 331:152–4 Jany PL, Agosta GE, Benko WS et al. (2015) CSF and Blood levels of GFAP in Alexander disease. eNeuro 2(5):ENEURO.0080-15. 2015 Quinlan R, Hutchison C, Lane B (1995) Intermediate filament proteins. Protein Profile 2:795–952 Howard RS, Greenwood R, Gawler J et al. (1993) A familial disorder associated with palatal myoclonus, other brainstem signs, tetraparesis, ataxia and Rosenthal fibre formation. J Neurol Neurosurg Psychiatr 56:977–81 Schwankhaus JD, Parisi JE, Gulledge WR et al. (1995) Hereditary adult-onset Alexander’s disease with palatal myoclonus, spastic paraparesis, and cerebellar ataxia. Neurology 45:2266–71 Okamoto Y, Mitsuyama H, Jonosono M et al. (2002) Autosomal dominant palatal myoclonus and spinal cord atrophy. J Neurol Sci 195:71–6 Arend AO, Leary PM, Rutherfoord GS (1991) Alexander’s disease: a case report with brain biopsy, ultrasound, CT scan and MRI findings. Clin Neuropathol 10:122–6 Borrett D, Becker LE (1985) Alexander’s disease: a disease of astrocytes. Brain 108(Pt 2):367–85 French TA, Bower BD, Cameron AH (1976) Alexander’s disease presenting as astrocytoma. J Neurol Neurosurg Psychiatry 9:803–9 Klein EA, Anzil AP (1994) Prominent white matter cavitation in an infant with Alexander’s disease. Clin Neuropathol 13:31–8 Pridmore CL, Baraitser M, Harding B et al. (1993) Alexander’s disease: clues to diagnosis. J Child Neurol 8:134–44 Schochet SS, Lampert PW, Earle KM (1968) Alexander’s disease: a case report with electron microscopic observations. Neurology 18:543–9 Towfighi J, Young R, Sassani J et al. (1983) Alexander’s disease: further light-, and electron-microscopic observations. Acta Neuropathol 61:36–42 Townsend JJ, Wilson JF, Harris T et al. (1985) Alexander’s disease. Acta Neuropathol 67:163–6 Duckett S, Schwartzman RJ, Osterholm J et al. (1992) Biopsy diagnosis of familial Alexander’s disease. Pediatr Neurosurg 18:134–8 Iwaki A, Iwaki T, Goldman JE et al. (1992) Accumulation of alpha B-crystallin in brains of patients with Alexander’s disease is not due to an abnormality of the 5’-flanking and coding sequence of the genomic DNA. Neurosci Lett 140:89–92 Russo LS, Aron A, Anderson PJ (1976) Alexander’s disease: a report and reappraisal. Neurology 26:607–14 Vogel F, Hallervorden J (1962) Leukodystrophy with diffuse Rosenthal fiber formation. Acta Neuropathol 2:126–43

453

Developmental Neuropathology 34. Wilson J, Manners BT, Robins DG, Erdohazi M (1981) Persistent hiccup as a presenting feature of Alexander’s leucodystrophy. Dev Med Child Neurol 23:660–1 35. Namekawa M, Takiyama Y, Aoki Y et al. (2002) Identification of GFAP gene mutation in hereditary adult-onset Alexander’s disease. Ann Neurol 52:779–85 36. Seil FJ, Schochet SS, Earle KM (1968) Alexander’s disease in an adult. Report of a case. Arch Neurol 19:494–502 37. Spalke G, Mennel HD (1982) Alexander’s disease in an adult: clinicopathologic study of a case and review of the literature. Clin Neuropathol 1:106–12 38. Schmidt S, Wattjes MP, Gerding WM, van der Knaap M (2011) Late onset Alexander’s disease presenting as cerebellar ataxia associated with a novel mutation in the GFAP gene. J Neurol 258:938–40 39. van der Knaap MS, Naidu S, Breiter SN et al. (2001) Alexander disease: diagnosis with MR imaging. AJNR 22:541–52 ¨ 40. Rosenthal W (1898) Uber eine eigentumliche, mit Syringomyelie komplizierte Geschwulst des Ruckenmarks. Beitr Zur Pathol Anat Allgem Pathol 23:11–143 41. Olabarria M, Putilina M, Riemer EC, Goldman JE (2015) Astrocyte pathology in Alexander disease causes a marked inflammatory environment. Acta Neuropathol 130:469–86 42. Herndon RM, Rubinstein LJ, Freeman JM, Mathieson G (1970) Light and electron microscopic observations on Rosenthal fibers in Alexander’s disease and in multiple sclerosis. J Neuropathol Exp Neurol 29:524–51 43. Iwaki T, Iwaki A, Tateishi J et al. (1993) Alpha B-crystallin and 27-kd heat shock protein are regulated by stress conditions in the central nervous system and accumulate in Rosenthal fibers. Am J Pathol 143:487–95 44. Iwaki T, Kume-Iwaki A, Liem RK, Goldman JE (1989) Alpha Bcrystallin is expressed in non-lenticular tissues and accumulates in Alexander’s disease brain. Cell 57:71–8 45. Johnson AB, Bettica A (1989) On-grid immunogold labeling of glial intermediate filaments in epoxy-embedded tissue. Am J Anat 185:335–41 46. Tang G, Perng MD, Wilk S et al. (2010) Oligomers of mutant glial fibrillary acidic protein (GFAP) Inhibit the proteasome system in alexander disease astrocytes, and the small heat shock protein alphaB-crystallin reverses the inhibition. J Biol Chem 285:10527–37 47. Tian R, Gregor M, Wiche G, Goldman JE (2006) Plectin regulates the organization of glial fibrillary acidic protein in Alexander disease. Am J Pathol 168:888–97 48. Heaven MR, Flint D, Randall SM et al. (2016) Composition of Rosenthal fibers, the protein aggregate hallmark of Alexander disease. J Proteome Res 15:2265–82 49. Tomokane N, Iwaki T, Tateishi J et al. (1991) Rosenthal fibers share epitopes with alpha B-crystallin, glial fibrillary acidic protein, and ubiquitin, but not with vimentin. Immunoelectron microscopy with colloidal gold. Am J Pathol 138:875–85 50. Head MW, Corbin E, Goldman JE (1993) Overexpression and abnormal modification of the stress proteins alpha B-crystallin and HSP27 in Alexander disease. Am J Pathol 143:1743–53 51. Messing A, Head MW, Galles K et al. (1998) Fatal encephalopathy with astrocyte inclusions in GFAP transgenic mice. Am J Pathol 152:391–8

454

52. Hagemann TL, Connor JX, Messing A (2006) Alexander disease– associated glial fibrillary acidic protein mutations in mice induce Rosenthal fiber formation and a white matter stress response. J Neurosci 26:11162–73 53. Wang L, Colodner KJ, Feany MB (2011) Protein misfolding and oxidative stress promote glial–mediated neurodegeneration in an Alexander disease model. J Neurosci 31:2868–77 54. Wang L, Hagemann TL, Kalwa H et al. (2015) Nitric oxide mediates glial-induced neurodegeneration in Alexander disease. Nat Commun 6:8966 55. Tang G, Xu Z, Goldman JE (2006) Synergistic effects of the SAPK/JNK and the proteasome pathway on glial fibrillary acidic protein (GFAP) accumulation in Alexander disease. J Biol Chem 281(50):38634–43 56. Cho W, Messing A (2009) Properties of astrocytes cultured from GFAP over-expressing and GFAP mutant mice. Exp Cell Res 315:1260–72 57. Chen Y-S, Lim S-C, Chen M-H et al. (2011) Alexander disease causing mutations in the C-terminal domain of GFAP are deleterious both to assembly and network formation with the potential to both activate caspase 3 and decrease cell viability. Exp Cell Res 317: 2252–66 58. Tang G, Yue Z, Talloczy Z et al. (2008) Autophagy induced by Alexander disease-mutant GFAP accumulation is regulated by p38/MAPK and mTOR signaling pathways. Hum Mol Genet 17:1540–55 59. Wisniewski T, Goldman JE (1998) Alpha B-crystallin is associated with intermediate filaments in astrocytoma cells. Neurochem Res 23:385–92 60. Head MW, Hurwitz L, Kegel K, Goldman JE (2000) AlphaBcrystallin regulates intermediate filament organization in situ. Neuroreport 11:361–5 61. Koyama Y, Goldman JE (1999) Formation of GFAP cytoplasmic inclusions in astrocytes and their disaggregation by alphaBcrystallin. Am J Pathol 154:1563–72 62. Perng MD, Cairns L, van den IJssel P et al. (1999) Intermediate filament interactions can be altered by HSP27 and alphaB–crystallin. J Cell Sci 112:2099–112 63. Hagemann TL, Boelens WC, Wawrousek EF, Messing A (2009) Suppression of GFAP toxicity by alphaB-crystallin in mouse models of Alexander disease. Hum Mol Genet 18:1190–9 64. Tian R, Wu X, Hagemann TL et al. (2010) Alexander disease mutant glial fibrillary acidic protein compromises glutamate transport in astrocytes. J Neuropathol Exp Neurol 69:335–45 65. Sosunov AA, Guilfoyle E, Wu X et al. (2013) Phenotypic conversions of “protoplasmic” to “reactive” astrocytes in Alexander disease. J Neurosci 33:7439–50 66. Olabarria M, Goldman JE (2016) Disorders of astrocytes: Alexander disease as a model. Annu Rev Pathol 12:131–52 67. Messing A, Goldman JE, Johnson AB, Brenner M (2001) Alexander disease: new insights from genetics. J Neuropathol Exp Neurol 60:563–73 68. Eng LF, Lee YL, Kwan H et al. (1998) Astrocytes cultured from transgenic mice carrying the added human GFAP gene contain Rosenthal fibers. J Neurosci Res 53:353–60

38

Neuroaxonal Dystrophy/Neurodegeneration with Brain Iron Accumulation Abi Li,1 Sarah Wiethoff,2,3 Charles Arber,2 Henry Houlden,2 Tamas Revesz,1 and Janice L. Holton1 1

Queen Square Brain Bank and Reta Lila Weston Institute of Neurological Studies, UCL Institute of Neurology, University College London, London, UK 2 Department of Molecular Neuroscience, UCL Institute of Neurology, University College London, London, UK 3 Center for Neurology and Hertie Institute for Clinical Brain Research, Eberhard-Karls-University, T¨ ubingen, Germany

Definition The neuroaxonal dystrophies form a rare and heterogeneous group of inherited neurodegenerative disorders classically considered to include infantile neuroaxonal dystrophy, neuroaxonal dystrophy and pantothenate kinase-associated neurodegeneration (PKAN, previously known as Hallervorden–Spatz syndrome). As these diseases have marked brain iron accumulation, they are now classified within the group known as neurodegeneration with brain iron accumulation (NBIA). Overall, the NBIA group of disorders shows considerable phenotypic variability often with childhood onset and including age-dependent phenotypes. They are clinically characterized by progressive hypo- and/or hyperkinetic movement disorders. Neuropathological hallmarks include the presence of axonal spheroids and abnormal brain iron accumulation, primarily affecting the basal ganglia. Molecular advances have identified eight NBIA-associated genes: pantothenate kinase 2 (PANK2), phospholipase A2 group VI (PLA2G6), chromosome 19 open reading frame 12 (C19orf12), ferritin light chain (FTL), cerulosplasmin (CP), fatty acid hydroxylase (FA2H), WD repeat-containing protein 45 (WDR45) and coenzyme A (CoA) synthase (COASY) (Table 38.1). Kufor–Rakeb syndrome and Woodhouse–Sakati syndrome, caused by autosomal-recessive mutations in ATPase type 13A2 (ATP13A2) and DDB1-and CUL4-associated factor 17 (DCAF17), respectively, are sometimes considered as members of the NBIA group, as neuroimaging studies variably indicate brain iron accumulation. However, studies are few, brain iron

accumulation appears to be inconsistent and no neuropathological descriptions are available, so these are not considered further in this chapter.

Normal embryology No studies describing the role of proteins associated with NBIA disorders in human embryology have been published.

Epidemiology An estimated prevalence for NBIA of 1 in 1 000 000 has been suggested, with no preference for gender or ethnicity, thus classifying NBIAs as ultra-rare diseases. Distribution data from North America indicate that four subtypes of NBIA contribute 87% of the total cases. NBIA1 (PANK2) represents 50% of cases, NBIA2 (PL2G6) contributes 20%, NBIA4 (C19orf12) accounts for 10% and NBIA5 (WDR45) a further 7% [1].

Clinical features The clinical features vary between different diseases in the NBIA group. These are summarized, together with the genetic basis of each disease, in Table 38.1. The most frequently occurring NBIA is pantothenate kinase-associated neurodegeneration (PKAN; NBIA1), which has two main subtypes, an early onset and rapidly progressive “classical” or “typical” disease, and an

Developmental Neuropathology, Second Edition. Edited by Homa Adle-Biassette, Brian N. Harding, and Jeffrey A. Golden. © 2018 John Wiley & Sons Ltd. Published 2018 by John Wiley & Sons Ltd.

455

Gene and Chromosome Pantothenate kinase-2 (PANK2), 20p13

Phospholipase A2 PLA2G6, 22q12

FTL, 19q13.33

C19orf12, 19q12

Synonyms

NBIA1 (formerly called Hallervorden–Spatz syndrome

NBIA2, Karak syndrome, PARK14

NBIA3, FTL

NBIA4

NBIA Type

Pantothenate kinase-associated neurodegeneration (PKAN)

Phospholipase A2-associated neurodegeneration (PLAN)

Neuroferritinopathy

Mitochondrial membrane protein-associated neurodegeneration (MPAN)

AR

AD

C19orf12

Ferritin light chain

Mitochondrial membrane protein

Iron homeostasis

Catalyzes the release of fatty acids from phospholipids

Ca2+ independent phospholipase A2 (iPLA2)

AR

Biosynthesis of coenzyme A

Pantothenate kinase 2

AR

Protein function

Protein

Inheritance

Table 38.1 Genetic and clinical features of neurodegeneration with brain iron accumulation.

Childhood to adulthood

Adult: dystonia parkinsonism Adolescence, early adulthood

Childhood (atypical: neuroaxonal dystrophy (NAD))

Young adult (atypical PKAN) Early infancy (typical: classical infantile neuroaxonal dystrophy (INAD))

Usually early childhood (typical PKAN)

Age of onset

57

Typical: severe dystonia, spasticity parkinsonism, chorea, cognitive decline, behavioral abnormalities Atypical: later-onset, slower progression INAD: hypotonia, developmental regression, progressive psychomotor delay, cerebellar features, bulbar dysfunction, progressive spastic tetraparesis NAD: less aggressive, less progressive phenotype with ataxia, spastic tetraparesis, dystonia, speech delay and behavioral abnormalities PLA2G6-related adult-onset dystonia parkinsonism Dystonia, oromandibular dyskinesias, blepharospasm and chorea Speech and gait difficulties, progressive pyramidal– extrapyramidal movement disorder, cognitive decline, dysarthria, optic atrophy, psychiatric features and motor neuropathy

10, 11

8, 9

7

6, 23

33

References

Main clinical features

AR

AR

FA2H, 16q23.1

CP, 3q24-q25

FAHN, spastic paraplegia 35, SPG35

Hemosiderosis, ceruloplasmin deficiency

Fatty acid hydroxylaseassociated neurodegeneration (FAHN)

Aceruloplasminemia

Ceruloplasmin

Fatty acid 2-hydroxylase

Coenzyme A synthase

WIPI4

AD, autosomal-dominant; AR, autosomal-recessive; NBIA, neurodegeneration with brain iron accumulation.

AR

COASY, 17q21

NBIA6, CoPAN

Coenzyme A synthase protein-associated neurodegeneration (CoPAN)

XLD

WDR45, Xp11.23

NBIA5, SENDA

Beta-propeller protein-associated neurodegeneration (BPAN)

Infancy

Adolescence to adulthood

Aids iron transport across cell membrane

Adult

Early childhood Catalyzes the synthesis of 2hydroxyspingolipids

Coenzyme A synthesis

Role in autophagy

Psychomotor developmental delay stable in early childhood than rapidly progressive adult-onset dystonia, parkinsonism and spasticity Spastic paraplegia, progressive cognitive impairment, oromandibular dystonia, dysarthria Pyramidal–extrapyramidal movement disorder, progressive intellectual impairment, optic atrophy, seizures Cerebellar ataxia, craniofacial dyskinesia, cognitive impairment, dysarthria, retinal degeneration. Microcytic anemia and diabetes mellitus may precede neurological symptoms

18, 19, 20

15, 16, 17

14

12, 13

Developmental Neuropathology “atypical” later-onset form, which is less severe and more slowly progressive. Clinical features of the classical variant of PKAN develop before the first decade of life. Patients present with gait or postural abnormalities progressing to oromandibular and lower limb dystonia, chorea, parkinsonism, pyramidal tract signs, and cognitive decline [2,3]. Late-onset (atypical) PKAN cases have an adult onset with a milder form of motor involvement. Overall, the manifestations are more variable, although there have been few reported adult-onset cases. Symptoms at presentation can include oromandibular dystonia, dysarthria, and neuropsychiatric symptoms. Later in the course of the disease, patients can exhibit dystonia, parkinsonism, and spasticity. The progression of atypical PKAN is much slower than in the classical disease and patients often live up to their fifties [2]. PKAN cases can show characteristic magnetic resonance imaging (MRI) appearances reflecting iron accumulation in the globus pallidus. A T2∗ -weighted MRI can show the “eye of the

tiger” sign, represented by a central hyperintense region surrounded by an area of hypointensity (Figure 38.1a). The substantia nigra can show hypointensity (Figure 38.1b); however, this is not pathognomonic for PKAN [4]. Almost two-thirds of patients with PKAN have abnormal electroretinograms, slow saccadic pursuits and 40% have signs of pigmentary retinopathy [5]. The second most common form of NBIA is PLA2G6associated neurodegeneration (PLAN) or NBIA2. Like NBIA1, there are age-dependent phenotypes providing three subtypes; classical, atypical and adult-onset forms. The classical earlyonset phenotype was historically known as infantile neuroaxonal dystrophy or Seitelberger’s disease. Children present between the ages of six months to three years with truncal hypotonia, progressive motor and mental restriction, cerebellar syndrome, pyramidal signs, and optic atrophy. Hypotonia and areflexia later progress to spastic tetraplegia. Death is often before the age of 10 years. The second subtype of PLAN is

(a)

(c)

(e)

(g)

(b)

(d)

(f)

(h)

Figure 38.1 Imaging characteristics of the four major subtypes of neurodegeneration with brain iron accumulation. (a) T2-weighted magnetic resonance imaging in pantothenate kinase-associated neurodegeneration shows globus pallidus hypointensity indicating iron accumulation with an area of central hyperintensity known as the “eye of the tiger” sign. (b) Substantia nigra in the same patient showing hypointensity in the medial aspect of the nucleus. (c) T2-weighted sequence in a patient with mitochondrial protein-associated neurodegeneration showing pallidal hypointensity with hyperintense streaking in the region of the medial medullary lamina. (d) Iron accumulation in the substantia nigra. (e) T2 sequence of globus pallidus. (f) Substantia nigra in a nine-year-old

458

child with PLA2G6-associated neurodegeneration showing iron accumulation. Imaging performed earlier in the disease course had shown no signal changes. Inset (f) showing cerebellar atrophy in the same child. (g) T2 imaging showing the globus pallidus in a young adult with beta-propeller protein-associated neurodegeneration, after the onset of parkinsonian symptoms. (h) Note the marked hypointensity in the substantia nigra, and, in the inset in (h), the same region on T1-weighted sequence showing the characteristic hyperintense “halo”, thought to represent neuromelanin release from degenerating neurons. All images performed on 3.0T magnet except (g) and (h), which were performed on 1.5T (reproduced with permission from Hogarth [1] and Journal of Movement Disorders).

Neuroaxonal Dystrophy/Neurodegeneration with Brain Iron Accumulation Chapter 38

represented by atypical cases presenting in mid-late childhood and also known as neuroaxonal dystrophy. This disorder progresses more slowly than the classical type, and has a heterogeneous presentation with cerebellar symptoms, hypotonia, dysarthria, and areflexia as typical features. The third subtype of PLAN has an adult-onset and is characterized by dystonia and parkinsonism with spasticity, cognitive decline, and psychiatric symptoms [2,6,7]. MRI in classical and atypical PLAN patients demonstrates cerebellar atrophy in the early stages of the disease. Additional imaging features include diffuse T2 white matter hyperintensities with thinning of the corpus callosum and optic chiasm. Iron deposits may be noted in the globus pallidus and substantia nigra (Figures 38.1e,f) [2,6]. Neuroferritinopathy, also known as NBIA3, is the only autosomal-dominant inherited disorder in the NBIA group. This disorder has a late onset (second decade of life), and patients present with extrapyramidal features including choreoathetosis, dystonia, spasticity, and rigidity, with low levels of serum ferritin [8]. Abnormal mitochondrial respiratory chain enzyme activity has been demonstrated in skeletal muscle [9]. Mitochondrial membrane protein-associated neurodegeneration (MPAN) is a more recent addition to the NBIA disease spectrum, and is also known as NBIA4 [10]. Patients present in childhood to early adulthood with gait abnormalities progressing to spastic paraparesis and extrapyramidal signs (dystonia and parkinsonism). Other manifestations can include optic atrophy, cognitive decline, and neuropsychiatric symptoms. Patients survive well into adulthood [11]. MRI T2-weighted images show hypointense lesions in the globus pallidus and substantia nigra, and in some cases, a hyperintense streaking of the medial medullary lamina between the globus pallidus interna and externa (Figures 38.1c,d) [11]. NBIA5, known as beta-propeller protein-associated neurodegeneration (BPAN) is another recently described form of NBIA, previously known as static encephalopathy of childhood with neurodegeneration in adulthood. BPAN is an X-linked inherited disorder characterized by global developmental delay with slow motor and cognitive deterioration during childhood and dystonia, which often shows some response to levodopa for a few years. The disorder remains fairly stable until early adulthood (20s to 30s) when patients develop progressive dystonia, extrapyramidal signs, dementia and loss of ambulation. MRI reveals iron accumulation in the globus pallidus and substantia nigra (Figures 38.1g,h) [12,13]. NBIA6, also known as CoA synthase protein-associated neurodegeneration (CoPAN) is extremely rare. The two unrelated patients reported presented with early-onset spastic dystonia, paraparesis, and they later developed parkinsonism, oromandibular dystonia, dysarthria, axonal neuropathy, cognitive impairment, and behavioral abnormalities. Patients are so far known to survive at least into the third decade of life. MRI findings include signs reminiscent of the eye of the tiger sign [14]. Fatty acid hydroxylase-associated neurodegeneration (FAHN) is a very rare childhood-onset disorder and is

considered to be a complicated form of spastic paraplegia [15]. Early spasticity, ataxia, dystonia, ocular motor abnormalities, and optic atrophy are presented. The disorder is variable, with some patients having milder less progressive disease while others are more severely affected. Later in the disease course, there are progressive cognitive impairment and seizures. Neuroimaging reveals iron accumulation in the globus pallidus and substantia nigra in some. However, other cases do not have this finding with cerebellar, brain stem, and corpus callosum atrophy, as well as white matter changes also being reported [16,17]. Aceruloplasminemia is caused by mutations in the ceruloplasmin gene [18]. This is an adult-onset disorder (25–60 years of age), with neurological disease, retinal degeneration, and diabetes mellitus. Neurological manifestations include progressive dementia, extrapyramidal signs, cerebellar ataxia, tremor, facial and neck dystonia. Biochemical analysis can distinguish aceruloplasminemia from other forms of NBIA due to low serum concentrations of copper and iron and high serum levels of ferritin. T1- and T2-weighted MRI show iron deposits in the dentate nucleus, thalamus, globus pallidus and liver [19,20].

Pathology Genetically confirmed NBIA cases are rare and care should be taken in interpreting older studies as, without genetic confirmation, the classification of cases is uncertain. Neuropathological reports in genetically defined cases are limited and the findings are summarized below. A unifying pathological feature is the presence of excess iron deposition in the brain, in particular in the basal ganglia structures. In addition, there may be axonal spheroids or swellings. The major neuropathological features are summarized in Table 38.2. In NBIA1, the brain weight is usually within the normal range and macroscopic examination of brain slices shows rusty brown discoloration of the globus pallidus. Histological changes are predominately observed in the globus pallidus, where there is rarefaction of the neuropil, neuronal loss, astrogliosis, and sometimes cavitation. Iron deposition is confirmed by Perls staining, which highlights iron pigment in the neuropil, macrophages and astrocytes. Neocortices, cerebellum and brain stem are relatively spared. Two subtypes of eosinophilic spheroids have been described. Larger structures (20–70 μm diameter), ubiquitinpositive and often containing iron, resembling the ovoid bodies found in other conditions of iron overload such as superficial siderosis, are found in the globus pallidus (Figures 38.2a– c). Smaller homogenous eosinophilic and weakly ubiquitinpositive structures (10–30 μm diameter) immunoreactive for phosphorylated neurofilaments and amyloid precursor protein and, with surrounding myelin sheaths, represent the classical axonal spheroid (Figures 38.2d,e). These are observed in the globus pallidus, putamen and variably in other regions. Historical cases reported as Hallervorden–Spatz syndrome described

459

Cortical and cerebellar atrophy. Rusty discoloration of globus pallidus. Pallor of substantia nigra (inconsistent). Mild reduction brain weight, frontal atrophy and mild ventricular dilatation. Atrophy of frontal lobe, caudate and cerebellum. Pallor of substantia nigra. Cavitation of the globus pallidus, putamen and dentate nucleus. Not described.

NBIA2, PLAN

Globus pallidus, substantia nigra

Neocortex striatum, thalamus, globus pallidus, cerebellar dentate nucleus, cerebellar cortex

Globus pallidus, substantia nigra

Striatum, thalamus

Iron/ferritin bodies in: globus pallidus, caudate, putamen, cerebellar dentate nucleus, red nucleus

Globus pallidus, putamen, caudate, cerebellar dentate nucleus

Globus pallidus, substantia nigra

Globus pallidus, substantia nigra

Widespread

Globus pallidus, substantia nigra

Globus pallidus

Globus pallidus

Iron Deposition

Tau in neurofibrillary tangles.

None described.

None described

α-synuclein in Lewy bodies. Tau in neurons.

None described.

α-synuclein in Lewy bodies. Tau in neurofibrillary tangles in some cases.

Tau in neurofibrillary tangles in some cases.

Protein Aggregation

Globus pallidus, striatum, neocortex, brain stem, spinal cord Globus pallidus, substantia nigra

Globus pallidus, putamen, neocortex, hippocampus, substantia nigra Neocortex, globus pallidus, striatum, cerebellum, brain stem, spinal cord, peripheral nerves Globus pallidus, putamen

Axonal Spheroids

BPAN, beta-propeller protein-associated neurodegeneration; FTL, ferritin light chain; MPAN, membrane protein-associated neurodegeneration; NBIA, neurodegeneration with brain iron accumulation; PKAN, pantothenate kinase-associated neurodegeneration; PLAN, PLA2G6-associated neurodegeneration.

Aceruloplasminemia

NBIA5, BPAN

NBIA4, MPAN

Mild reduction brain weight, frontal atrophy and mild ventricular dilatation. Mild atrophy hippocampus and amygdala. Dark discoloration globus pallidus and substantia nigra. Pale locus ceruleus. Brown discoloration and cavitation putamen and thalamus.

Normal brain weight. Rusty brown discoloration of globus pallidus.

NBIA1, PKAN

NBIA3, FTL, Neuroferritinopathy

Macroscopic Features

Disease

Brain Regions Predominantly Affected

Table 38.2 Neuropathological features in neurodegeneration with brain iron accumulation.

Neuroaxonal Dystrophy/Neurodegeneration with Brain Iron Accumulation Chapter 38

(a)

(b)

(c)

(d)

(e)

(f)

Figure 38.2 Histological features of pantothenate kinase-associated neurodegeneration. Axonal swellings are readily apparent in the globus pallidus (a–e). Large eosinophilic swellings with a granular appearance are present (a; arrows, hematoxylin and eosin) in addition to smaller swellings with a homogeneous appearance (a; arrowhead). There is widespread iron deposition often within large axonal swellings (b; arrow, Perls stain). Axonal swellings are

readily identified using ubiquitin immunohistochemistry (c). Small axonal swellings contain phosphorylated neurofilaments (d; arrows, SMI31 immunohistochemistry) and express amyloid precursor protein (e; arrow) while this is absent from large swellings (e; arrowhead). Tau-positive neurofibrillary tangle pathology varies in severity between cases (f; frontal cortex, AT8 immunohistochemistry).Bar in (a) represents 50 μm in (a), (c) and (f); 25 μm in (b), (d) and (e).

461

Developmental Neuropathology Lewy body and tau pathology. The few genetically characterized PKAN cases have confirmed only the presence of neurofibrillary tangles and threads, which vary in number between cases, possibly relating to age at death (Figure 38.2f). PKAN is no longer considered as a member of the group of α-synucleinopathies [21,22]. Neuropathological findings have been reported in only a few PLAN cases. The macroscopic findings are those of variable cortical and cerebellar atrophy, which is consistent with the MRI appearances. Rusty discoloration of the globus pallidus may be observed. Histology shows widespread neuronal loss and astrogliosis with variable iron deposition in the globus pallidus and substantia nigra. Axonal spheroids containing neurofilaments are observed in most cases in the basal ganglia, brainstem, cerebellum and spinal cord (Figures 38.3a–c) and may also be found in peripheral nerves. α-Synuclein-positive Lewy pathology resembling idiopathic Parkinson’s disease is described in most cases, this may be widespread and severe (Figure 38.3d). Hyperphosphorylated tau-positive neurofibrillary tangles and neuropil threads have also been described. Tau pathology varies between cases but may be extensive (Figure 38.3e) [23,24]. The neuropathological features of neuroferritinopathy/ NBIA3 have been described in a small number of cases. Macroscopic findings include slightly reduced brain weight and mild dilatation of the lateral ventricles. Atrophy of the frontal lobe, caudate and cerebellum has been described and also pallor of the substantia nigra. There may be cavitation of the globus pallidus, putamen, and cerebellar dentate nucleus. There is widespread neuronal loss, most severe in the caudate, putamen, and globus pallidus. The most distinctive finding is that of intranuclear and intracytoplasmic eosinophilic inclusions, largely in glia, but also in some neurons, most frequent in the caudate, putamen, globus pallidus, cerebellar dentate nucleus, and red nucleus. These contain iron (Perls positive) and many are immunoreactive for ferritin. There is gliosis around areas of cavitation and many of the reactive glial cells contain iron pigment. Ubiquitinated axonal swellings are found in the putamen and globus pallidus [8,25,26]. Neuropathological findings of two NBIA4/MPAN cases have been published. Severe neuronal loss in the globus pallidus and substantia nigra is observed and there is myelin loss in the optic nerve and tract and the pyramidal tract. Iron deposits in association with neurons and astrocytes and in macrophages are present in the globus pallidus and substantia nigra. Axonal spheroids are widespread including the globus pallidus, striatum, neocortex, brainstem and spinal cord. Lewy pathology is extensive in the brainstem and in the neocortex where the pathological load is greater than in PLAN cases. Tau pathology is inconsistent with tau-positive neurons described in the hippocampus of one case (aged 23 years at death). These neurons are largely unstained in silver impregnation preparations indicating that the inclusions do not represent mature neurofibrillary tangles [11].

462

Reports of the neuropathological features of NBIA5/BPAN are limited. Macroscopic examination shows some reduction in brain weight and ventricular dilatation together with mild atrophy of the frontal lobe, amygdala and hippocampus. There is striking brown discoloration of the globus pallidus and substantia nigra and also pallor of the locus coeruleus. Neuronal loss is most pronounced in the substantia nigra (Figure 38.4a), but is also evident in the neocortex, globus pallidus, locus coeruleus and Purkinje cell layer. There are axonal swellings prominent in the substantia nigra (Figure 38.4b), but these are also described in the globus pallidus and rarely in other regions. Iron deposition is severe in the substantia nigra and globus pallidus (Figure 38.4c,d). Neurofibrillary tangle pathology is widespread in neocortex and subcortical gray structures (Figure 38.4 e,f) [13,27]. Neuropathological descriptions of aceruloplasminemia are few. Macroscopic findings are of dark discoloration and cavitation of the putamen and thalamus. Histological studies show extensive iron deposition involving the neocortex, striatum, thalamus, globus pallidus, cerebellar dentate nucleus, and cerebellar cortex. Iron deposition involves astrocytes preferentially and these often show ballooned, iron-containing processes. Neuronal iron is less conspicuous. Axonal spheroids, tau or αsynuclein pathology are not described [20,28]. There are no neuropathological descriptions of NBIA6/CoPan or FAHN available in the literature.

Genetics The eight forms of NBIA discussed in this chapter are associated with mutations in genes encoding proteins which are components of several cellular pathways. Ferritin light chain and ceruloplasmin are directly involved in iron metabolism. Proteins encoded by PANK2, PLA2G6, FA2H, COASY and C19orf12 are directly, or indirectly, involved in lipid metabolism and lipid membrane homeostasis. PANK2, PLA2G6 and c19orf12 proteins are localized to the mitochondria and may play a role in mitochondrial bioenergetics. The WDR45 protein is involved in autophagy. It remains unclear how dysfunction of all of these pathways contributes to excess iron accumulation in the brain [29]. Most forms of NBIA have an autosomal-recessive mode of inheritance; however, NBIA3 is inherited in an autosomaldominant manner and NBIA5 is X-linked. Approximately half of NBIA cases have PANK2 mutations. PANK2 is located on chromosome 20p13 and two mutations (1231G > A and 1253C > T) are responsible for one-third of all PKAN cases and the inheritance is autosomal-recessive [2,30]. PANK2, encoded by PANK2, is a mitochondrial protein highly expressed in neurons of the cerebral cortex, globus pallidus, nucleus basalis of Meynert and the pontine nuclei. PANK2 is a key enzyme required for the first conversion step of de novo biosynthesis of CoA, an essential cofactor involved in over 100 metabolic reactions, including the Krebs cycle, fatty acid,

Neuroaxonal Dystrophy/Neurodegeneration with Brain Iron Accumulation Chapter 38

(a)

(b)

(c)

(d)

(e)

Figure 38.3 Histological appearances in phospholipase A2, group VI-associated neurodegeneration. In the globus pallidus (a–c), axonal swellings are numerous and are visible in hematoxylin and eosin stained preparations (a). These may be highlighted by ubiquitin immunohistochemistry (b) and staining for phosphorylated neurofilaments (c; SMI31). Iron deposition is evident on hematoxylin and eosin

staining (a; arrow). α-Synuclein-immunoreactive Lewy bodies and neurites may be widespread (d; nucleus basalis of Meynert) and there may also be tau-positive neurofibrillary tangles and neuropil threads (e; nucleus basalis of Meynert). Bar in (a) represents 50 μm in (a), (b) and (e); 25 μm in (c) and (d).

463

Developmental Neuropathology

(a)

(b)

(c)

(d)

(e)

(f)

Figure 38.4 Histological findings in beta-propeller protein-associated neurodegeneration. There is severe loss of pigmented neurons in the substantia nigra with abundant free pigment in the neuropil and axonal spheroids (a; arrow). Axonal spheroids are highlighted using immunohistochemistry for phosphorylated neurofilaments (b; arrow, SMI31). Perls staining demonstrates abundant iron deposition in both the substantia nigra (c) and globus pallidus (d). There is widespread tau immunoreactive neurofibrillary tangle pathology in cortical and subcortical structures (e, fusiform gyrus; F, caudate; AT8). Bar in (a) represents 45 μ m in (d); 30 μ m in (a–c) and (f); 15 μ m in (e).

and amino acid synthesis. It has been suggested that defective CoA synthesis is damaging in high energy cells such as retinal photoreceptor cells and the neurons of the globus pallidus [31,32]. NBIA2 is caused by autosomal-recessive mutations in PLA2G6 located on chromosome 22q12 and comprising 17

464

exons encoding a calcium independent phospholipase 2 group VI β enzyme, iPLA2β [33]. iPLA2β is part of a group of enzymes that catalyze the hydrolysis of glycerophospholipids creating free fatty acids such as arachidonic acid and lysophopholipids. They also play a major role in phospholipid membrane remodeling, cell signaling, cell proliferation, and apoptosis. The pathogenic

Neuroaxonal Dystrophy/Neurodegeneration with Brain Iron Accumulation Chapter 38

mechanism is likely to be due to abnormal regulation of phospholipid membrane homeostasis of the cell and mitochondria [34]. NBIA3 is caused by mutations in the gene encoding ferritin light chain (FTL) located on chromosome 19q13. The disease has an autosomal-dominant mode of inheritance [8]. The ferritins are major iron storage proteins and are composed of light and heavy chain subunits, the variation in the subunits can affect the rates of iron uptake and release in different tissues [35]. A phenotype–genotype correlation has been observed. Patients with hyperferritinemia cataract syndrome have mutations altering the iron response element of the ferritin light chain. This leads to a decreased affinity of the iron response protein binding, causing over production of FTL protein, which aggregates in the ocular lens [36]. Truncating mutations in exon 4 of FTL result in the accumulation of ferritin inclusions in the brain and other organs [8]. Iron accumulation has been seen in presymptomatic carriers of the mutation, suggesting that iron is part of the disease process rather than an outcome of tissue degeneration [37]. MPAN/NBIA4 is caused by mutations in the orphan gene C19orf12 located on chromosome 19 and has an autosomalrecessive inheritance. Little is known about the function of the protein encoded by this gene. In silico analysis predicts a mitochondrial transmembrane localization and possible involvement in fatty acid synthesis [10,11]. NBIA5 is caused by mutations in WDR45, which is located at Xp11.23 (12). NBIA5 does not follow a typical X-linked inheritance pattern; three males have been reported to exhibit a homogenous phenotype likely caused by somatic mosaicism. Males with germline WDR45 mutations are thought to be nonviable [12]. Lymphoblasts have been shown to have exclusive expression of the mutant transcript, suggesting X inactivation of the wild-type allele and protein instability in females. WDR45 mutations have been linked to impaired autophagy due to abnormal autophagic flux with accumulation of autophagic structures in patient lymphoblasts [38]. NBIA6 is inherited in an autosomal-recessive pattern and is caused by the mutations in the COASY gene, located at 17q.21. Mutations result in a significant decrease of de novo production of CoA. COASY is a bifunctional enzyme, which catalyzes the last two steps of the CoA biosynthesis pathway. COASY is expressed throughout the body; however, one variant, COASYβ, is mainly expressed in the brain [39]. The COASY enzyme is associated with the outer mitochondrial membrane. This is the second disorder directly linked to CoA biosynthesis implicated in NBIA, the other being PKAN/NBIA1 [14]. FAHN is caused by autosomal-recessive mutations in the fatty acid-2 hydroxylase, FA2H located on chromosome 16q23. FA2H plays an important role in the formation of sphingolipids, a critical component for myelin sheath maintenance. In addition, FA2H protein contains conserved iron-binding histidine motifs. The mechanism underlying brain iron accumulation remains unresolved [16].

Aceruloplasminemia is inherited in an autosomal-recessive manner and results from mutations in the ceruloplasmin (CP) gene, located at chromosome 3q24. CP encodes ceruloplasmin, a ferroxidase protein that binds up to 95% of plasma copper and is also involved in the peroxidation of transferrin Fe2+ to Fe3+. Mutations lead to the disruption of ferroxidase function, resulting in the decrease of iron transport toward the plasma and increased storage of intracellular iron [40, 41].

Animal models Several NBIA animal models have been studied and have recapitulated some of aspects of human disease. PANK2 and PLA2G6 were the first NBIA genes to be discovered and the majority of NBIA in vivo research has been carried out in these two disorders. To date, there are no published reports from animal models for COASY. PKAN has been studied in PANK2 knockout mouse and drosophila models. Homozygous null mice develop retinal problems and are infertile, but no abnormalities in gait or changes within the basal ganglia are observed [42]. Mouse pantothenate kinase 2 was shown to localize to the mitochondria of the retina and sperm. Furthermore, the association of pantothenate kinase 2 and mitochondria has been shown in pantothenate kinase 2 defective neurons which have altered mitochondrial membrane potential, and there are swollen mitochondria in the central and peripheral nervous systems [43]. A PKAN drosophila model also demonstrates mitochondrial dysfunction, increased oxidative damage and a reduced lifespan [44]. Interestingly, a knockout mouse model for PLA2G6 also revealed the degeneration of the inner mitochondrial membrane, increasing significantly at distal portions of axons and resulting in axonal swellings. Mice exhibited progressive motor deficits and an increased production of reactive oxygen species [45,46]. Investigation of aceruloplasminemia and the role in iron homeostasis using a knockout mutant mouse model revealed a progressive increase of serum iron and iron content in the liver and spleen, but no neurological phenotype [18]. A second CP mutant mouse model demonstrated increased iron deposits in the cerebellum and brain stem and deficits in motor coordination associated with loss of brainstem dopaminergic neurons. These models indicate that ceruloplasmin is important for iron homeostasis [47]. Transgenic mice have been used to investigate neuroferritinopathy. A mutant mouse model (498-499InsTC), overexpressing human ferritin, developed ferritin inclusion bodies throughout the central nervous system, increased iron in the brain, a decrease in motor performance and had a shorter life span [48]. These results are supported by further transgenic FTL mutant mouse models showing iron ferritin inclusion bodies and signs of oxidative damage in the brain. Hippocampal neurons were found to be particularly vulnerable to cell death [49].

465

Developmental Neuropathology Animal models of other forms of NBIA are limited. FA2H knock out mice have been shown to develop late-onset axon and myelin sheath degeneration in the spinal cord and sciatic nerve. This demonstrates 2-hydoxylated sphingolipids are essential for long term maintenance of axons [50]. ATP13A2 null mice have been shown to develop early neuropathological changes of widespread gliosis, increased lipofuscin granules and large ubiquitin-positive aggregates. The mice also developed an age-dependent motor impairment [51]. Currently, very little is known about the function or activity of the C19orf12 mitochondrial protein which is mutated in MPAN. A double C19orf12 homolog impaired drosophila has displayed locomotor problems due to neuromuscular and mitochondrial deficits and has a shorten life span. Neuropathological examination demonstrated vacuoles in the brain and optic nerve [52]. WDR45 knockout mice were generated to investigate BPAN. These mice showed behavioral abnormalities, including motor and cognitive impairments. Neuropathological studies revealed ubiquitin-positive inclusions in the hippocampus and caudate nucleus and the presence of axonal swellings. These knockout mice also demonstrated autophagy defects, which were less severe in hetrozygous mice [53]. Overall, the current NBIA animal models give greater insight into disease pathogenesis. There are several overlapping observations in the animal models that support the pathogenic mechanisms observed in humans, such as iron toxicity, mitochondrial dysfunction, phospholipid membrane homeostasis.

Treatment, future perspective, conclusions There is no disease modifying treatment for NBIA, although treatments to alleviate the symptoms are available. Dystonia and spasticity are managed with anticholinergic drugs, benzodiazepines, baclofen, and botulinum toxin injections for targeted relief. Stereotactic surgery has resulted in some benefit. The most effective target to date is deep brain stimulation of the posteroventral globus pallidus internus and in a trial of 23 NBIA patients who underwent this surgery dystonia symptoms improved by 20% [54]. A preliminary study of deferiprone, an iron chelator which is able to cross the blood–brain barrier, has shown a reduction of brain iron, but no change in clinical symptoms were observed [55]. Conversely, deferiprone treatment in a small cohort of patients with PKAN stabilized the motor symptoms in the majority of patients and showed a reduction in the hypointensity in the globus pallidus [56]. Treatment of adult PLAN dystonia parkinsonism with levodopa is effective in the early stages of the disease, but patients often develop early dyskinesias with this treatment [24]. Although NBIA is rare, this group of diseases causes significant morbidity and mortality and most types of NBIA affect children. Increased availability of genetic testing in patients with complex movement disorders will clarify the prevalence

466

of these disorders. The factors leading to age-dependent phenotypes in diseases such as PKAN and PLAN are not understood and further research to elucidate disease modifying factors will be important. The diseases comprising the NBIA group are caused by mutations in genes influencing iron and lipid metabolism, membrane homeostasis, mitochondrial function, and autophagy. Clarification of how these diverse pathways contribute to the common feature of brain iron accumulation and neuroaxonal degeneration will contribute to our understanding of the disease mechanisms underlying NBIA. The observation of pathological α-synuclein deposition in the form of Lewy pathology in PLAN and MPAN and of tau pathology in PKAN, PLAN, MPAN and BPAN is intriguing. Further insight into these diseases may be relevant to our understanding of a broad range of other neurodegenerative diseases. Improved understanding of the pathomechanisms involved in the NBIA disease group is a prerequisite for the development of disease-modifying therapies.

References 1. Hogarth P (2015) Neurodegeneration with brain iron accumulation: diagnosis and management. J Mov Disord 8:1–13 2. Gregory A, Polster BJ, Hayflick SJ (2009) Clinical and genetic delineation of neurodegeneration with brain iron accumulation. J Med Genet 46:73–80 3. Hayflick SJ, Westaway SK, Levinson B et al. (2003) Genetic, clinical, and radiographic delineation of Hallervorden–Spatz syndrome. N Engl J Med 348:33–40 4. Kumar N, Boes CJ, Babovic-Vuksanovic D, Boeve BF (2006) The “eye-of-the-tiger” sign is not pathognomonic of the PANK2 mutation. Arch Neurol 63:292–3 5. Egan RA, Weleber RG, Hogarth P et al. (2005) Neuroophthalmologic and electroretinographic findings in pantothenate kinase-associated neurodegeneration (formerly Hallervorden– Spatz syndrome). Am J Ophthalmol 140:267–74 6. Kurian MA, Morgan NV, MacPherson L et al. (2008) Phenotypic spectrum of neurodegeneration associated with mutations in the PLA2G6 gene (PLAN). Neurology 70:1623–9 7. Pais´an-ruiz C, Bhatia KP, Li A et al. (2009) Characterization of PLA2G6 as a locus for dystonia-parkinsonism. Ann Neurol 65:19– 23. 8. Curtis AR, Fey C, Morris CM et al. (2001) Mutation in the gene encoding ferritin light polypeptide causes dominant adult-onset basal ganglia disease. Nat Genet 28:350–4 9. Chinnery PF, Curtis ARJ, Fey C et al. (2003) Neuroferritinopathy in a French family with late onset dominant dystonia. J Med Genet 40:e69 10. Hartig MB, Iuso A, Haack T et al. (2011) Absence of an orphan mitochondrial protein, c19orf12, causes a distinct clinical subtype of neurodegeneration with brain iron accumulation. Am J Hum Genet 89:543–50 11. Hogarth P, Gregory A, Kruer MC et al. (2013) New NBIA subtype: genetic, clinical, pathologic, and radiographic features of MPAN. Neurology 80:268–75

Neuroaxonal Dystrophy/Neurodegeneration with Brain Iron Accumulation Chapter 38 12. Haack TB, Hogarth P, Kruer MC et al. (2012) Exome sequencing reveals de novo WDR45 mutations causing a phenotypically distinct, X-linked dominant form of NBIA. Am J Hum Genet 91: 1144–9 13. Hayflick SJ, Kruer MC, Gregory A et al. (2013) β-Propeller proteinassociated neurodegeneration: a new X-linked dominant disorder with brain iron accumulation. Brain 136:1708–17 14. Dusi S, Valletta L, Haack TB et al. (2014) Exome sequence reveals mutations in CoA synthase as a cause of neurodegeneration with brain iron accumulation. Am J Hum Genet 94:11–22 15. Edvardson S, Hama H, Shaag A et al. (2008) Mutations in the fatty acid 2-hydroxylase gene are associated with leukodystrophy with spastic paraparesis and dystonia. Am J Hum Genet 83: 643–8 16. Kruer MC, Pais´an-Ruiz C, Boddaert N et al. (2010) Defective FA2H leads to a novel form of neurodegeneration with brain iron accumulation (NBIA). Ann Neurol 68:611–8 17. Dick KJ, Eckhardt M, Pais´an-Ruiz C et al. (2010) Mutation of FA2H underlies a complicated form of hereditary spastic paraplegia (SPG35). Hum Mutat 31:E1251–60 18. Harris ZL, Takahashi Y, Miyajima H et al. (1995) Aceruloplasminemia: molecular characterization of this disorder of iron metabolism. Proc Natl Acad Sci U S A 92:2539–43 19. Logan JI, Harveyson KB, Wisdom GB et al. (1994) Hereditary caeruloplasmin deficiency, dementia and diabetes mellitus. QJM Mon J Assoc Physicians 87:663–70 20. Morita H, Ikeda S, Yamamoto K et al. (1995) Hereditary ceruloplasmin deficiency with hemosiderosis: a clinicopathological study of a Japanese family. Ann Neurol 37:646–56 21. Kruer MC, Hiken M, Gregory A et al. (2011) Novel histopathologic findings in molecularly-confirmed pantothenate kinase-associated neurodegeneration. Brain 134:947–58 22. Li A, Paudel R, Johnson R et al. (2013) Pantothenate kinaseassociated neurodegeneration is not a synucleinopathy. Neuropathol Appl Neurobiol 39: 121–31 23. Gregory A, Westaway SK, Holm IE et al. (2008) Neurodegeneration associated with genetic defects in phospholipase A(2). Neurology 71:1402–9 24. Pais´an-Ruiz C, Li A, Schneider SA et al. (2012) Widespread Lewy body and tau accumulation in childhood and adult onset dystonia-parkinsonism cases with PLA2G6 mutations. Neurobiol Aging 33:814–23 25. Vidal R, Ghetti B, Takao M et al. (2004) Intracellular ferritin accumulation in neural and extraneural tissue characterizes a neurodegenerative disease associated with a mutation in the ferritin light polypeptide gene. J Neuropathol Exp Neurol 63:363–80 26. Keogh MJ, Morris CM, Chinnery PF (2013) Neuroferritinopathy. Int Rev Neurobiol 110:91–123 27. Paudel R, Li A, Wiethoff S et al. (2015) Neuropathology of Betapropeller protein associated neurodegeneration (BPAN): a new tauopathy. Acta Neuropathol Commun 3:39 28. Kaneko K, Hineno A, Yoshida K et al. (2012) Extensive brain pathology in a patient with aceruloplasminemia with a prolonged duration of illness. Hum Pathol 43:451–6 29. Rouault TA (2013) Iron metabolism in the CNS: implications for neurodegenerative diseases. Nat Rev Neurosci 14:551–64 30. Hartig MB, H¨ortnagel K, Garavaglia B et al. (2006) Genotypic and phenotypic spectrum of PANK2 mutations in patients with

31.

32.

33.

34.

35.

36.

37. 38.

39.

40.

41. 42.

43.

44.

45.

46.

47.

neurodegeneration with brain iron accumulation. Ann Neurol 59:248–56 Leonardi R, Zhang Y-M, Lykidis A et al. (2007) Localization and Regulation of Mouse Pantothenate Kinase 2. FEBS Lett 581:4639– 44 Kotzbauer PT, Truax AC, Trojanowski JQ, Lee VM-Y (2005) Altered neuronal mitochondrial coenzyme A synthesis in neurodegeneration with brain iron accumulation caused by abnormal processing, stability, and catalytic activity of mutant pantothenate kinase 2. J Neurosci 25:689–98 Morgan NV, Westaway SK, Morton JEV et al. (2006) PLA2G6, encoding a phospholipase A2, is mutated in neurodegenerative disorders with high brain iron. Nat Genet 38:752–4 Farooqui AA, Yang HC, Rosenberger TA, Horrocks LA (1997) Phospholipase A2 and its role in brain tissue. J Neurochem 69:889– 901 Arosio P, Levi S (2010) Cytosolic and mitochondrial ferritins in the regulation of cellular iron homeostasis and oxidative damage. Biochim Biophys Acta 1800:783–92 Girelli D, Bozzini C, Zecchina G et al. (2001) Clinical, biochemical and molecular findings in a series of families with hereditary hyperferritinaemia–cataract syndrome. Br J Haematol 115: 334–40 Keogh MJ, Jonas P, Coulthard A et al. (2012) Neuroferritinopathy: a new inborn error of iron metabolism. Neurogenetics 13:93–6 Saitsu H, Nishimura T, Muramatsu K et al. (2013) De novo mutations in the autophagy gene WDR45 cause static encephalopathy of childhood with neurodegeneration in adulthood. Nat Genet 45:445–9, 449e1 Nemazanyy I, Panasyuk G, Breus O et al. (2006) Identification of a novel CoA synthase isoform, which is primarily expressed in the brain. Biochem Biophys Res Commun 341:995–1000 De Domenico I, Ward DM, di Patti MCB et al. (2007) Ferroxidase activity is required for the stability of cell surface ferroportin in cells expressing GPI-ceruloplasmin. EMBO J 26:2823–31 McNeill A, Pandolfo M, Kuhn J et al. (2008) The neurological presentation of ceruloplasmin gene mutations. Eur Neurol 60:200–5 Kuo Y-M, Duncan JL, Westaway SK et al. (2005) Deficiency of pantothenate kinase 2 (Pank2) in mice leads to retinal degeneration and azoospermia. Hum Mol Genet 14:49–57 Brunetti D, Dusi S, Morbin M et al. (2012) Pantothenate kinaseassociated neurodegeneration: altered mitochondria membrane potential and defective respiration in Pank2 knock-out mouse model. Hum Mol Genet 21:5294–305 Bosveld F, Rana A, van der Wouden PE et al. (2008) De novo CoA biosynthesis is required to maintain DNA integrity during development of the Drosophila nervous system. Hum Mol Genet 17:2058– 69 Shinzawa K, Sumi H, Ikawa M et al. (2008) Neuroaxonal dystrophy caused by group VIA phospholipase A2 deficiency in mice: a model of human neurodegenerative disease. J Neurosci 28:2212–20 Malik I, Turk J, Mancuso DJ et al. (2008) Disrupted membrane homeostasis and accumulation of ubiquitinated proteins in a mouse model of infantile neuroaxonal dystrophy caused by PLA2G6 mutations. Am J Pathol 172:406–16 Patel BN, Dunn RJ, Jeong SY et al. (2002) Ceruloplasmin regulates iron levels in the CNS and prevents free radical injury. J Neurosci 22:6578–86

467

Developmental Neuropathology 48. Vidal R, Miravalle L, Gao X et al. (2008) Expression of a mutant form of the ferritin light chain gene induces neurodegeneration and iron overload in transgenic mice. J Neurosci 28:60–7 49. Maccarinelli F, Pagani A, Cozzi A et al. (2015) A novel neuroferritinopathy mouse model (FTL 498InsTC) shows progressive brain iron dysregulation, morphological signs of early neurodegeneration and motor coordination deficits. Neurobiol Dis 81:119–33 50. Z¨oller I, Meixner M, Hartmann D et al. (2008) Absence of 2hydroxylated sphingolipids is compatible with normal neural development but causes late-onset axon and myelin sheath degeneration. J Neurosci 28:9741–54 51. Kett LR, Stiller B, Bernath MM et al. (2015) α-Synucleinindependent histopathological and motor deficits in mice lacking the endolysosomal Parkinsonism protein Atp13a2. J Neurosci 35:5724–42 52. Iuso A, Sibon OCM, Gorza M et al. (2014) Impairment of Drosophila orthologs of the human orphan protein C19orf12 induces bang sensitivity and neurodegeneration. PloS One 9:e89439

468

53. Zhao YG, Sun L, Miao G et al. (2015) The autophagy gene Wdr45/Wipi4 regulates learning and memory function and axonal homeostasis. Autophagy 11:881–90 54. Timmermann L, Pauls KA, Wieland K et al. (2010) Dystonia in neurodegeneration with brain iron accumulation: outcome of bilateral pallidal stimulation. Brain 133:701–12 55. Zorzi G, Zibordi F, Chiapparini L et al. (2011) Iron-related MRI images in patients with pantothenate kinase-associated neurodegeneration (PKAN) treated with deferiprone: results of a phase II pilot trial. Mov Disord 26:1756–9 56. Cossu G, Abbruzzese G, Matta G et al. (2014) Efficacy and safety of deferiprone for the treatment of pantothenate kinase-associated neurodegeneration (PKAN) and neurodegeneration with brain iron accumulation (NBIA): results from a four years follow-up. Parkinsonism Relat Disord 20:651–4 57. Zhou B, Westaway SK, Levinson B et al. (2001) A novel pantothenate kinase gene (PANK2) is defective in Hallervorden–Spatz syndrome. Nat Genet 28:345–9

39

Spinal Muscular Atrophy Brian N. Harding Department of Pathology and Laboratory Medicine, Children’s Hospital of Philadelphia, Philadelphia, PA, USA

Definition and classification

Genetics

Progressive degeneration of lower motor neurons (in the spinal cord and, in the more severe forms, the bulbar motor nuclei) characterizes this heterogeneous group of disorders that together constitute the second most frequently lethal genetic disorder of childhood. Classification and nomenclature remain problematic, the most useful being based on a combination of type of inheritance, age of onset, distribution of weakness, and clinical progression. This chapter is principally concerned with the most common childhood forms of proximal, recessively inherited spinal muscular atrophy (SMA): the “acute” early onset severe type 1 SMA or Werdnig–Hoffman disease; the chronic intermediate type 2 SMA, and the chronic mild type 3 SMA or Kugelberg–Welander disease [1]. All three types of SMA result from homozygous deletions or mutations in the telomeric copy of the survival motor neuron gene at chromosome 5q11.2-13.3. In this region of the chromosome, which is especially prone to large-scale deletions, at least three SMA-related genes have been identified: the survival motor neuron gene, the neuronal apoptosis inhibitory protein (NAIP) gene and the p44t gene.

SMA is caused by a deficiency of the ubiquitous protein survival of motor neuron (SMN), which is encoded by the SMN gene, comprising two almost identical copies on chromosome 5, designated SMN1 and SMN2. The critical difference between the two genes is a single nucleotide transition in exon 7 that affects a splice site enhancer, such that 90% of transcripts from the SMN2 copy gene lose exon 7 (SMNΔ7) resulting in greatly reduced levels of SMN protein [3–5]. The entire SMN2 gene can be deleted without affecting health, but homozygous absence of SMN1 results in considerably reduced levels of full-size SMN protein and an SMA phenotype. Ninety-six percent of SMA patients show mutations in SMN1, while 4% are unlinked to 5q13. Of the patients with 5q13-linked SMA, the majority show homozygous absence of SMN1 exons 7 and 8 or exon 7 only, whereas a minority present a compound heterozygosity with a subtle mutation on one chromosome and a deletion/gene conversion on the other [6]. A small number of point mutations have also been reported. The presence of SMN1 deletions in all three types of SMA and the rare occurrence of asymptomatic members of SMA families having the same haplotype as affected family members suggest the operation of modifier genes. Two have been implicated. NAIP can reduce neuronal apoptosis under experimental conditions, and two-thirds of patients with SMA type 1 have large deletions, which include the telomeric form of NAIP, compared with less than 5% of types 2 and 3. A second potential modifier gene is SMN2, a gene which is unique to the human species [7]. The observation that there are more copies of SMN2 in chromosomes from the chronic forms of SMA than in chromosomes of type 1 SMA provides a correlation between

Epidemiology SMA is the second most common neuromuscular disease of childhood affecting approximately 1 in 10 000–12 000 live births and with a gene carrier frequency of approximately 1 in 50 [2].

Developmental Neuropathology, Second Edition. Edited by Homa Adle-Biassette, Brian N. Harding, and Jeffrey A. Golden. © 2018 John Wiley & Sons Ltd. Published 2018 by John Wiley & Sons Ltd.

469

Developmental Neuropathology genotype and phenotype and has given rise to the suggestion that gene conversion rather than deletion is the underlying defect in SMA types 2 and 3 [8]. SMN gene dosage analysis can determine the copy number of SMN1 to detect carriers and patients heterozygous for the absence of SMN1. More recently, copy number determination of SMN2 has also become available. Owing to the genetic complexity and the high carrier frequency, comprehensive SMA genetic testing combined with appropriate genetic counseling and risk assessment provides the most complete evaluation of patients and families at this time [2].

Clinical features In 1992, the International SMA Consortium defined three main types according to age of onset and achieved motor milestones, which correlate with disease severity [9]. Further subclassification has been proposed [10,11].

SMA type 1 (Werdnig–Hoffman disease) The classical form of severe SMA has a very consistent clinical phenotype. Presentation is within the first six months of life, and often at birth or within the first few weeks; there may be a period of normal motor development before onset of weakness. Affected infants never achieve independent sitting. There is generalized hypotonia with marked axial and symmetrical limb weakness, affecting legs more than arms and proximal muscles more than distal. Significant contractures are not a feature of SMA type 1 but mild contractures frequently occur, including the internal rotators of the shoulders giving a rather characteristic “jug handle” appearance of the arms with the hands facing outwards. Additionally, there may be mild limitation of hip abduction, knee and elbow extension. Facial muscles are spared so that the infant has a bright, alert expression. However, bulbar weakness is usually present. There is always intercostal muscle involvement, with relative sparing of the diaphragm resulting in a characteristic lateral flattening of the thorax and paradoxical abdominal breathing pattern. Cardiac muscle is not affected and intellect is normal. Death occurs by two years of age from respiratory insufficiency precipitated by intercurrent infection and aspiration from bulbar involvement; median survival is around seven months and about 80% of children die within the first year [1]. The electromyogram (EMG) shows neurogenic changes with spontaneous rhythmical activity or fibrillations in many, as well as polyphasic potentials. Nerve conduction velocity may be normal or reduced. Evidence for sensory involvement includes reports of reduced sensory nerve conduction velocity [12], inexcitability of motor and sensory nerves, and axonal degeneration in sural nerve [13]. Serum creatine kinase levels are usually normal. Muscle biopsy is less often indicated, except where genetic tests are unavailable or equivocal, or the clinical course is atypical.

470

The revised classification takes into account new clinical variants that have been recognized which do not conform to the classical clinical phenotype of type 1 but are nevertheless genetically confirmed. One of these is the in utero severe fatal form with prenatal onset (reduced fetal movements) and intrauterine death, or presentation at birth, with severe asphyxia, more generalized weakness, including facial weakness and external ophthalmoplegia, and arthrogryposis. This subtype, originally denominated type 0 SMA [14], is now referred to as SMA-1A, separating it from SMA-1B with onset by three months of age, and SMA-1C with onset after three months [10].

SMA type 2 Children with SMA type 2 have an onset of symptoms before 18 months of age and achieve independent sitting – the majority in the first year – but do not achieve independent standing or walking. Symptom onset may be insidious or occasionally more acute with a sudden deterioration and subsequent plateau in motor skills. A similar pattern to SMA type 1 is seen, with symmetrical proximal muscle weakness, affecting lower limbs more than upper. Fasciculation and atrophy of the tongue is a common feature, as is hand tremor. Survival into adolescence and adulthood is increasingly common [15], with improved recognition and proactive management of complications, such as recurrent chest infections (more frequent in early childhood), nocturnal hypoventilation (more frequent in the adolescent age group), bulbar dysfunction affecting swallowing and nutrition (variable involvement in type 2), gastroesophageal reflux and spinal deformity (progressive kyphoscoliosis). The EMG is similar to type 1 but fasciculation is common and pseudomyotonic discharges may be observed. SMA type 3 (Kugelberg–Welander disease) This is the mildest form – further subdivided into types 3A and 3B according to symptom onset before and after three years of age, respectively – as onset may span childhood and adolescence into adult life. Motor milestones are usually normal in the first year of life. Independent ambulation is achieved but there is mild to moderate proximal muscle weakness (may be asymmetrical) which causes functional difficulties, such as climbing stairs and getting up from the floor. Waddling gait and “neurogenic” calf pseudohypertrophy may give a resemblance to muscular dystrophy. Hand tremor is a common feature. There is usually no respiratory compromise. Life expectancy is therefore not significantly less than the normal population. However, a significant proportion of children may eventually lose independent walking due to factors such as increased body size during growth spurts, rather than progression of weakness per se [15]. Serum creatine kinase may be mildly to moderately elevated and secondary myopathic features in muscle biopsies may make differentiation from dystrophy difficult, but the EMG and genetic studies are diagnostic. SMA types 2 and 3 are generally considered relatively static conditions in contrast to the inexorable progression and

Spinal Muscular Atrophy Chapter 39

(a)

(b)

(c)

Figure 39.1 Muscle pathology in spinal muscular atrophy (SMA). (a) Deltoid muscle from a case of SMA-1A. (b,c) Established changes in the quadriceps muscle of a child with SMA-1C surviving 10 years. There is type grouping but also small rounded fibers, of both fiber types. Immunohistochemistry for myosin heavy chain isoforms (b) slow myosin (c) fast myosin.

outcome of type 1, although as highlighted there is an overall continuum of severity.

Pathology Muscle pathology The effects of deafferentation on involved muscles vary with the patient’s age. In neonates, the classical changes of group atrophy or angulated fibers may not be present. Scattered small fibers are observed throughout the fascicles (Figure 39.1a); they may be type 1 fibers reminiscent of congenital fiber type disproportion, or alternatively small type 2 fibers. From six weeks of age one finds the typical changes of neurogenic atrophy, including large groups of atrophic fibers, mostly type 2 fibers with scattered type 1 fibers interspersed with larger, often hypertrophic type 1 fibers singly or in groups. With immunocytochemistry, the larger fibers express slow myosin exclusively, while the small fibers co-express fetal, slow and fast myosins suggesting that they may have become denervated before they are fully mature. With longer survival, fiber type grouping becomes more evident (Figure 39.1b,c). Similar changes are present in both SMA types 2 and 3. In patients with chronic disease, endomysial connective tissue may be markedly increased and with fatty infiltration gives an appearance reminiscent of congenital muscular dystrophy, but unlike congenital muscular dystrophy, the small fibers are mostly type 2. In SMA type 3, the pathology is more variable with less atrophy and fiber type grouping. With longstanding disease, myopathic changes increase the resemblance to dystrophy. In one patient with SMA-1B, motor end-plates in diaphragm were examined with acetyl choline receptor immunochemistry and found to be smaller and less complex than controls [3]. Macroscopic neuropathology Autopsy reports are largely confined to SMA type 1. The brain appears normal externally, but dissection of the spinal cord reveals thin gray anterior roots (Figure 39.2).

Figure 39.2 Anterior view of the spinal cord at autopsy: note the thin atrophic anterior roots in comparison with the normal appearing posterior roots.

Microscopic neuropathology Anterior horn cell loss and gliosis is usually profound and extensive throughout most levels of the spinal cord but, in the early stages of the more acute examples, one may observe ballooning and chromatolysis, microglial activation and occasionally neuronophagia (Figure 39.3). Bulbar motor neurons and Clarke’s column can be affected, while chromatolytic changes are regularly seen in dorsal root ganglion sensory cells (Figure 39.4.), and in the thalamus (Figure 39.5), a useful diagnostic pointer when the spinal cord is unavailable for examination.

471

Developmental Neuropathology

Figure 39.3 A ballooned chromatolytic anterior horn cell in the lumbar cord. Hematoxylin and eosin.

Figure 39.5 Degenerating neurons are also present in the thalamus. Hematoxylin and eosin.

In a correlative genotype–phenotype study of SMN1, using gene dosage studies of both SMN1 and SMN2, reduced levels of SMN protein in muscle was demonstrated by Western blot. In contrast to three cases of SMA-1B, each with two copies of SMN2, and exhibiting classical pathologic changes, two severe connatal cases, or type 1A, each had only one copy of SMN2 and a very extended phenotype, with chromatolytic changes additionally present in neurons of cerebral cortex, basal ganglia, brainstem, deep cerebellar nuclei, and pigmented nuclei (Figure 39.6a,b) [3]. Such pathologic findings have appeared in the literature occasionally in pregenetic times, and were problematic to classify [16,17].

Differential diagnosis In neonatal muscle biopsies, type 1 fiber atrophy is also a feature of myotonic dystrophy, nemaline myopathy, and congenital fiber type disproportion. At the other end of the spectrum, older patients with longstanding disease may have markedly myopathic changes resembling congenital muscular dystrophy that require immunocytochemical study with a panel of sarcolemmal markers. Neuropathological differential diagnosis is aimed at discriminating the rarer forms and variants of lower motor neuron disorder (Table 39.1) where genetic diagnosis is absent or equivocal.

Animal models

Figure 39.4 Chromatolytic sensory neurons and nodules of Nageotte in a dorsal root ganglion. Hematoxylin and eosin.

472

Clinical evidence suggests that SMA type 1 may begin in utero when the neuromuscular axis is immature, a situation that can to some extent be reproduced in the neonatal rat, where the neuromuscular system matures later than man. Cutting the unmyelinated rat sciatic nerve at birth causes motor neurons to die, although not four weeks later when motor neurons are mature and at the time when Schwann cells start to produce ciliary neurotrophic factor, which has been shown to delay cell death occurring after axotomy and in a hereditary mouse model of motor neuron disease. Axotomy in the immature rat is not followed by reinnervation, and after a few weeks the muscle has a similar appearance to that of human SMA type 1 [8]. There are no naturally occurring animal models of SMA. Man is unique, even differing from other large primates, in having two not quite identical copies of SMN [7]. The mouse homologue Smn, exists as a single copy [14]: homozygous knockout

Spinal Muscular Atrophy Chapter 39

(a)

(b)

Figure 39.6 Neonate with SMA-1A. The extensive pathology includes chromatolytic neurons in locus ceruleus (a) and occipital cortex (b). Hematoxylin and eosin ×200.

mice are not viable following massive apoptosis in the early blastocyst while heterozygotes display a mild spinal muscular atrophy [18]. In consequence, transgenic mice have been generated with homozygous deletions of Smn rescued by cDNA fragments of human SMN2. In these experiments, reduced proprioceptive reflexes, deafferentation of motor neurons, and subtle defects in cultures of sensory neurons have been demonstrated [19,20], but not the extensive neuropathology of type 1A.

Pathogenesis Much controversy has attended the morphogenetic basis of SMA, the underlying degenerative process in motor neurons and the pathogenetic relationship between neuron and muscle degeneration. Is the ballooning merely a manifestation of

chromatolysis similar to that resulting from loss of synaptic target or a dying back axonopathy, or is it evidence for an intrinsic disturbance? Reduced synaptophysin expression in anterior horns [21,22] may indicate loss of presynaptic terminals, but this does not distinguish between primary motor neuron degeneration and loss of its peripheral target. On the other hand, immunohistochemical studies of ballooned motor neurons have demonstrated accumulations of ubiquitinated degradation products in the center of the perikaryon [23] displacing phosphorylated neurofilaments to the periphery where glycosylation may be abnormally reduced, promoting abnormal neurofilament assembly, neuron–glia adhesion and failure of synapse formation [24]. Is the process of cell death apoptosis or necrosis? The two principal candidate genes for SMA located in the deleted region of chromosome 5q13, SMN and NAIP, both have anti-apoptotic

Table 39.1 Differential diagnosis of lower motor neuron disorders in childhood (inheritance, recessive). Pathology Gene

Type

Presentation

Duration (years)

Usual

Occasional

SMN1 IGHMBP2

SMA SMARD1

2 1

Sc Bu Th DRG Sc Bu Th

Type 1-A: Cx, Bg, Bs, Cbd, SN

VRK1

SMA with cerebellar hypoplasia, Fazio-Londe, BHN 2 Brown–Vialetto–van Laere, BHN 1

Early hypotonia Early diaphragmatic weakness Hypotonia at birth, mental restriction Bulbar weakness Nerve deafness, lower cranial nerve palsies

2

Sc Bu Cbh

Th

1–5 1–30

Bu Sc Coch Bu Sc

Cba Th Bg Cba

SLC52A3 SLC52A3

Bg, basal ganglia; BHN, bulbar hereditary neuropathy; Bs, brain stem; Bu, bulbar; Coch, cochlear; Cba, cerebellar atrophy; Cbd, deep cerebellar nuclei; Cbh, cerebellar hypoplasia, DRG, dorsal root ganglion; Sc, spinal cord; SMA, spinal muscular atrophy; SMARD1, spinal muscular atrophy with respiratory distress type 1; SN, substantia nigra; Th, Thalamus.

473

Developmental Neuropathology properties [25]. Simic et al. [26] have presented convincing morphological and terminal uridine nucleotide end labeling evidence for apoptosis, as well as loss of bcl-2 and upregulation of p53 immunostaining, in spinal motor neurons from children with genetically confirmed SMA type 1, in addition to describing morphological changes of necrotic cell death. The integrity of the neuron–muscle unit may also be compromised by the loss of SMN within muscle cells. Co-culture systems of cloned human muscle satellite cells and fibroblasts with embryonic rat spinal cord form innervated myotubes. Myofibers derived from SMA types 1 and 2 patients degenerate one to three weeks after innervation, but this degeneration can be prevented by adding 50% cloned satellite cells from normal donors, indicating an important role for muscle cells in the establishment and degeneration of the neuromuscular junction and a potential target for therapy [27]. A major question in considering the pathogenesis of SMA concerns the selective vulnerability of some cell populations. SMN is a housekeeping gene, and homozygous mouse knockouts are embryonically lethal [18]. The protein SMN is ubiquitously expressed with high levels in spinal motor neurons, and also throughout the brain and spinal cord including large pyramidal cells of human layer V cortex. If SMA is viewed as a protein deficiency disorder, it can be hypothesized that differential thresholds between different cell types could manifest as differential sensitivity to reduction in SMN expression [3]. Thus, in patients with only one copy of the less efficient SMN2 gene, as in connatal SMN1A, the extremely low protein levels might recruit cell types with lesser protein requirement than spinal motor neurons to the phenotype.

Management and future directions No curative treatment for SMA is presently available. Management of infants with type 1 is supportive (nasogastric tube feeding, suction and postural management) and in the UK, mechanical ventilatory support is generally not recommended or offered. However, there are differences in practice worldwide, with proponents for both noninvasive (mask ventilation) and invasive (tracheostomy) respiratory support for these children [28]. In terms of specific treatments, up until recently, no therapeutic trials had been attempted. However, results from a phase1 trial of riluzole (a glutamate inhibitor shown to slow rate of decline in patients with amyotrophic lateral sclerosis) in infants with SMA type 1, and a pilot trial of phenylbutyrate (shown to significantly increase SMN2 gene expression in vitro [29]) in children with SMA type 2 have been published [30,31]. In the former study, riluzole appeared to be well tolerated but numbers were too small to draw any firm conclusions as to benefit; in the latter, preliminary findings showed an improvement in motor function after nine weeks of treatment without major adverse effects.

474

Atypical forms of SMA This is a group of rare disorders with early onset and rapidly fatal course not linked to deletions of chromosome 5q [13].

Infantile spinal muscular atrophy with respiratory distress type 1 Autosomal recessive spinal muscular atrophy with respiratory distress type 1 is the second anterior horn cell disease in infants in which the genetic defect has been defined, resulting from mutations in the gene encoding the immunoglobulin micro-binding protein 2 on chromosome 11q13. On clinical and genetic review [32], most patients presented at the age of one to six months with respiratory distress due to diaphragmatic paralysis and progressive muscle weakness, with predominantly distal lower limb muscle involvement. Intrauterine growth restriction, weak cry, and foot deformities were the earliest symptoms. Sensory and autonomic nerves may be also affected. SMA with cerebellar hypoplasia SMA with cerebellar hypoplasia [33,34] has a congenital onset, often arthrogryposis, severe mental restriction and evidence of slowed motor conduction velocity, with rapid course and major cerebellar pathology (Chapter 15). Brown–Vialetto–van Laere syndrome, bulbar hereditary neuropathy type 1 Brown–Vialetto–van Laere syndrome is a rare recessively inherited disorder that presents late in childhood or adolescence with bilateral neural deafness and vestibular areflexia, then later involvement of cranial nerves resulting in facial palsy, dysarthria, and dysphagia; although the course is slow, death may result from swallowing problems and respiratory insufficiency. There are degenerative changes in auditory and vestibular pathways, as well as bulbar cranial nerve nuclei and anterior horns [35,36]. Fazio–Londe disease, bulbar hereditary neuropathy type 2 Dominant inheritance, in Fazio–Londe disease, is exceptional. Most cases are recessive, either with early onset before two years with respiratory symptoms and stridor, and rapid fatal progression, or a later onset of dysarthria, dysphagia and facial weakness with protracted clinical course of many years. The few neuropathological reports consistently describe widespread degeneration of motor cranial nerve nuclei through the brainstem, worst caudally but extending to the III nuclei, and usually anterior horn cell degeneration, with variable involvement of cerebellum, thalamus and basal ganglia [37].

Spinal Muscular Atrophy Chapter 39

References 1. Dubowitz V (1995) Disorders of the lower motor neurone: the spinal muscular atrophies 8:325-369. In: V Dubowitz, ed., Muscle Disorders in Childhood, 2nd ed. London, WB Saunders 2. Ogino S, Wilson RB (2004) Spinal muscular atrophy: molecular genetics and diagnostics. Expert Rev Mol Diagn 4:15–29 3. Harding BN, Kariya S, Monani UR et al. (2015) Spectrum of neuropathophysiology in spinal muscular atrophy type I. J Neuropathol Exp Neurol 74:15–24 4. Lorson CL, Strasswimmer J, Yao JM, et al. (1998) SMN oligomerization defect correlates with spinal muscular atrophy severity. Nat Genet 19:63–6 5. Monani UR, Lorson CL, Parsons DW et al. (1999) A single nucleotide difference that alters splicing patterns distinguishes the SMA gene SMN1 from the copy gene SMN2. Hum Mol Genet 8:1177–83 6. Wirth B (2000) An update of the mutation spectrum of the survival motor neuron gene (SMN1) in autosomal recessive spinal muscular atrophy (SMA). Hum Mutat 15:228–37 7. Rochette CF, Gilbert N, Simard LR (2001) SMN gene duplication and the emergence of the SMN2 gene occurred in distinct hominids: SMN2 is unique to Homo sapiens. Hum Gen 108:255–66 8. Schmalbruch H, Haase G (2001) Spinal muscular atrophy: present state. Brain Pathol 11:231–47 9. Munsat T, Davies KE (1992) Report on International SMA Consortium Meeting held in Bonn, Germany, June 1992. Neuromusc Disord 2:423–8 10. Bertini E, Burghes A, Bushby K et al. (2005) 134th ENMC International Workshop: Outcome Measures and Treatment of Spinal Muscular Atrophy, 11–13 February 2005, Naarden, the Netherlands. Neuromusc Disord 15:802–16 11. Dubowitz V (1999) Very severe spinal muscular atrophy (SMA type 0): an expanding clinical phenotype. Europ J Paediatr Neurol 3: 49–51 12. Anagnostou E, Miller SP, Guiot MC, et al. (2005) Type I spinal muscular atrophy can mimic sensory-motor axonal neuropathy. J Child Neurol 20:147–50 13. Rudnik Schoneborn S, Forkert R, Hahnen E et al. (1996) Clinical spectrum and diagnostic criteria of infantile spinal muscular atrophy: further delineation on the basis of SMN gene deletion findings. Neuropediatrics 27:8–15 14. DiDonato CJ, Ingraham SE, Mendell JR et al. (1997) Deletion and conversion in spinal muscular atrophy patients: is there a relationship to severity? Ann Neurol 41:230–7 15. Zerres K, Rudnik-Schoneborn S, Forrest E et al. (1997) A collaborative study on the natural history of childhood and juvenile onset proximal spinal muscular atrophy (type II and III SMA): 569 patients. J Neurol Sci 146:67–72 16. Steimann GS, Rorke LB, Brown MJ (1980) Infantile neuronal degeneration masquerading as Werdnig–Hoffmann disease. Ann Neurol 8:317–24 17. Towfighi J, Young RS, Ward RM (1985) Is Werdnig–Hoffmann disease a pure lower motor neuron disorder? Acta Neuropathol 65: 270–80 18. Schrank B, Gotz R, Gunnersen JM et al. (1997) Inactivation of the survival motor neuron gene, a candidate gene for human

19.

20.

21.

22.

23.

24.

25.

26.

27. 28. 29.

30. 31. 32.

33.

34.

35.

36.

37.

spinal muscular atrophy, leads to massive cell death in early mouse embryos. Proc Natl Acad Sci U S A 94:9920–5 Jablonka S, Karle K, Sandner B et al. (2006) Distinct and overlapping alterations in motor and sensory neurons in a mouse model of spinal muscular atrophy. Hum Mol Genet 15:511–18 Mentis GZ, Blivis D, Liu W et al. (2011) Early functional impairment of sensory–motor connectivity in a mouse model of spinal muscular atrophy. Neuron 69:453–67 Ikemoto A, Hirano A, Matsumoto S et al. (1996) Synaptophysin expression in the anterior horn of Werdnig–Hoffmann disease. J Neurol Sci 136:94–100 Yamanouchi Y, Yamanouchi H, Becker LE (1996) Synaptic alterations of anterior horn cells in Werdnig–Hoffmann disease. Pediatr Neurol 15:32–5 Murayama S, Bouldin TW, Suzuki K (1991) Immunocytochemical and ultrastructural studies of Werdnig–Hoffmann disease. Acta Neuropathol 81:408–17 Chou SM, Wang HS (1997) Aberrant glycosylation/phosphorylation in chromatolytic motoneurons of Werdnig–Hoffmann disease. J Neurol Sci 152:198–209 Iwahashi H, Eguchi Y, Yasuhara N et al. (1997) Synergistic antiapoptotic activity between Bcl-2 and SMN implicated in spinal muscular atrophy. Nature 390:413–17 Simic G, Seso-Simic D, Lucassen PJ et al (2000) Ultrastructural analysis and TUNEL demonstrate motor neuron apop- tosis in Werdnig–Hoffmann disease. J Neuropathol Exp Neurol 59:398– 407 Guettier-Sigrist S, Coupin G, Braun S et al. (1988) Muscle could be the therapeutic target in SMA treatment. J Neurosci Res 53:663–9 Bach JR, Baird JS, Plosky D et al. (2002) Spinal muscular atrophy type 1: management and outcomes. Pediatr Pulmonol 34:16–22 Andreassi C, Angelozzi C, Tiziano FD et al. (2004) Phenylbutyrate increases SMN expression in vitro: relevance for treatment of spinal muscular atrophy. Eur J Hum Genet 12:59–65 Russman BS, Lannaccone ST, Samaha FJ (2003) A phase 1 trial of riluzole in spinal muscular atrophy. Arch Neurol 60:1601–3 Mercuri E, Bertini E, Messina S et al. (2004) Pilot trial of phenylbutyrate in spinal muscular atrophy. Neuromusc Disord 14:130–5 Grohmann K, Varon R, Stolz P et al. (2003) Infantile spinal muscular atrophy with respiratory distress type1 (SMARD1). Ann Neurol 54:719–24 Gouti`eres F, Aicardi J, Farkas E (1977) Anterior horn cell disease associated with pontocerebellar hypoplasia in infants. J Neurol Neurosurg Psych 40:370–8 Kamoshita S, Takei Y, Miyao M et al. (1990) Pontocerebellar hypoplasia associated with infantile motor neuron disease (Norman’s disease). Pediatr Pathol 10:133–42 Gallai V, Hockaday JM, Hughes JT et al. (1981) Ponto-bulbar palsy with deafness (Brown–Vialetto–Van Laere syndrome). J Neurol Sci 50:259–75 Summers BA, Swash M, Schwartz MS, Ingram DA (1987) Juvenileonset bulbospinal muscular atrophy with deafness: Vialetta–van Laere syndrome or Madras-type motor neuron disease? J Neurol 234:440–2 McShane MA, Boyd S, Harding B et al. (1992) Progressive bulbar paralysis of childhood: a reappraisal of Fazio–Londe disease. Brain 115:1889–900

475

40

Autism Spectrum Disorders Matthew P. Anderson Department of Pathology, Division of Neuropathology, Harvard Medical School, Beth Israel Deaconess Medical Center, Boston, MA, USA

Definition, major synonyms and historical perspective Autism, the prototypic pervasive developmental disorder, is usually evident by the age of three to four years. It is defined by the triad of limited or absent verbal communication, lack of reciprocal social interaction/responsiveness, and restricted, stereotypic, often ritualized patterns of behavior and/or interest. Autism spectrum disorder (ASD) covers a broader range of phenotypes, which also encompass the less severe Asperger syndrome and pervasive developmental disorder not otherwise specified. Intellectual disabilities coexist with ASD, but are absent from Asperger syndrome.

Epidemiology Approximately 1 in 68 children has been diagnosed with ASD according to the Centers for Disease Control’s Autism and Developmental Disabilities Monitoring Network. The disorder exists in all socioeconomic, racial, and ethnic groups, and is almost five times more common in boys than girls. Among identical twins, if one child has ASD, the other will be affected 40–95% of the time, whereas for non-identical twins, the concordance rate is 0–30%. On average, children identified with ASD were not diagnosed until after the age of four, although a diagnosis can occur as early as age two years. In the United States, the cost per year for children with ASD is estimated at US$10– 60 billion.

Clinical features ASD represents a diverse group of behaviorally defined conditions that share reduced social interaction and communication and increased repetitive and restrictive behaviors. There are numerous neurological, psychiatric, and medical conditions variably comorbid with behavioral deficits that define ASD. Intellectual and language (cognitive) disabilities, disruptive, and sometimes self-injurious or aggressive behavior, epilepsy, disorders of movement, feeding, sleep or anxiety, depression, bipolar disorder, and schizophrenia are sometime found. Genetic defects, mostly de novo, but sometimes inherited, have been identified in approximately 20% of ASD cases. Each genetic abnormality accounts for less than 1% of cases arguing that ASD is extremely heterogeneous. Before the established genetic etiologies are discussed, some of the potential pathogenetic mechanisms are described.

Epilepsy comorbidity There is a resemblance between children with “autistic regression” and those with Landau–Kleffner syndrome, an epileptic encephalopathy where children (usually after age three) lose language skills in associated with an epileptiform electroencephalogram (EEG) pattern of continuous spike-and-wave in sleep. Multiple lines of evidence support a causal link between seizures and the behavioral deficits that define ASD: 1. There is high incidence of epileptiform activity seen in the cortical EEG in ASD, particularly during sleep, despite the absence of clinical seizures [1].

Developmental Neuropathology, Second Edition. Edited by Homa Adle-Biassette, Brian N. Harding, and Jeffrey A. Golden. © 2018 John Wiley & Sons Ltd. Published 2018 by John Wiley & Sons Ltd.

477

Developmental Neuropathology

Table 40.1 Genes: autism spectrum disorder and epileptic encephalopathy (Mendelian Inheritance in Man, SFARI). EIEE Type

Gene

1 2 4 6 7 8 9 11 12 13 14 24 27 30 33

ARX CDKL5 STXBP1 SCN1A KCNQ2 ARHGEF9 PCDH19 SCN2A PLCB1 SCN8A KCNT1 HCN1 GRIN2B SIK1 EEF1A2

EIEE, early infantile epileptic encephalopathy

2. Dysplastic changes in cerebral cortex or in cerebellum (often seizure foci in the focal epilepsies, e.g., focal cortical dysplasia or cortical tubers of tuberous sclerosis) are found in a higher proportion of ASD than control postmortem brains [2]. 3. Many rare single gene loss of function and missense mutations have been found in ASD that were independently discovered in cohorts with infantile epileptic encephalopathy (Table 40.1). 4. A number of the genomic copy number variations associated with ASD have epilepsy as a major comorbidity. Bilateral sharp or spike waves during sleep are prominent in one of the more frequent copy number variations found in ASD, 15q11-13 duplication (Thibert, personal communication). In a case reported by Kamiya et al. with SCN2A mutation (one of the single-gene defects found in ASD), frequent bilateral sharp or spike waves with maximum amplitude over the centroparietotemporal region were semicontinuous during sleep [3]. 5. Epilepsy was responsible for 7–30% of deaths (sudden unexpected death in epilepsy) in individuals with ASD [4]. In a case series of ASD/intellectual disability, epileptiform activity overall was found in 52% of individuals who died prematurely and 25% of deaths were attributed to epilepsy [5]. 6. Regression during fever, illness, or vaccination as is sometimes reported in ASD, is well described in Dravet syndrome, an early infantile epileptic encephalopathy due to heterozygous SCN1A loss of function mutations [6], another gene harboring mutations in ASD cohorts.

Mitochondrial disease A study published in 2009 found that a subgroup of patients with mitochondrial disorders are also at risk for autistic regression, especially around periods of fever [7].

478

Autoimmune disease and maternal–fetal antibodies Elevated maternal serum C-reactive protein, a biomarker of inflammation, may increase the risk of ASD [8], as may maternal celiac disease or rheumatoid arthritis [9]. Serum antibodies to cerebellar proteins are reported in some individuals with ASD, but are also found in controls [10]. Antibodies recognizing 45 and 62 kDa proteins correlate with more severe ASD behavioral symptoms [11]. Maternal–fetal autoantibodies correlate to brain enlargement in ASD [12]. Furthermore, introducing these human maternal–fetal auto-antibodies into pregnant monkeys produced aberrant social behaviors and enlarged brains in the progeny [13]. In utero infections ASD has been associated with congenital cytomegalovirus infection [14]. Cases of ASD and enterovirus encephalitis have also been reported [15]. As a step towards establishing causality between the viral infections and some of the ASD-associated brain postmortem findings, Fatemi et al. [16] reported that in utero exposure to maternal influenza virus infection leads to pyramidal cell atrophy and macrocephaly in adulthood. To evaluate whether activating the viral innate immune response alters growth of the cerebral cortex, Smith et al. [17] introduced the viral analog polyI:C into pregnant mice and found an excess number of neurons in the cerebral cortex at birth. The changes triggered by in utero viral infection/analog in these experimental animal models resemble those reported in ASD postmortem brain, including the early increased brain growth reported by magnetic resonance imaging (MRI) studies in vivo (reviewed in Amaral et al. [18]), reduced neuronal soma size (confirmed in Wegiel et al. [19]), and increased frontal lobe neuronal densities [20]. Activating the maternal immune system during pregnancy is also sufficient to cause behavioral deficits of reduced social interaction [21,22] and increased fetal and early postnatal brain cytokines [23] resembling changes found in the human ASD postmortem brain [24]. Social deficits and increased repetitive behavior were also produced in monkeys when the mother was exposed to a viral analogue during the first trimester [25]. Microbiome Gastrointestinal bacteria generate short chain fatty acids that can cross the blood–brain barrier as fermentation products; these metabolites can effect gene expression in the cell culture system [26], and are reportedly found at increased concentrations in the stool of individuals with ASD compared with controls [27]. Like the anti-epileptic agent valproate, short chain fatty acids can inhibit histone deacetylase [28]. In an experiment aimed at implicating the microbiota in ASD, mice were exposed in utero to a viral analog (polyI:C), resulting in compromise of the gastrointestinal mucosal barrier so that the microbiota composition (e.g., Bacteroides fragilis), presumably through various fermented metabolites, could promote ASD-related behaviors in mice [29]. On their own, these “non-genetic” causes of ASD could be sufficient to produce the behavioral deficits

Autism Spectrum Disorders Chapter 40

that define ASD, but they might also enable the expression of genetic defects, helping to explain the variable penetrance of some genetic forms of ASD.

Pathology The genetic and clinical heterogeneity of ASD predicts heterogeneity in the neuropathological findings. Future studies of the brain in ASD will be strengthened by examining cohorts with the same genetic defect: a goal of Autism BrainNET (Simons Foundation, Autism Speaks, and Autism Science Foundation) and Neurobiobank (National Institutes of Health), a joint brainbanking program to understand the neuropathology of ASD. A brain study comparing maternal 15q11-13 duplication, extranumerary isodicentric chromosome 15, (idic(15)), idiopathic ASD, and controls is used to illustrate the heterogeneity of ASD and to emphasize the importance of using genetically defined ASD case cohorts. Such results will build a framework for understanding the pathogenic mechanisms that may be shared across distinct clusters of ASD genetic subtypes, and may lay a solid foundation for understanding the presumed more complex nongenetic ASD subtypes. Genetic causes of ASD are identified as genomic copy number variations (CNVs; microdeletions and microduplications; search SFARI CNV) and sequence variants in the coding sequences of genes (search SFARI Gene). The list of genetic etiologies will continue to expand as more cases resulting in somatic mutations are identified [30]. With the reduced cost of whole genome sequencing, many genetic causes within the much larger intronic and intergenic genomic DNA sequence (98% of the human genome is “non-coding sequence”; outside the protein coding exons) are likely to be identified underlying peaks within the genome wide association studies. The efforts of the Simon Foundation and Autism Speaks will facilitate applying the technique to large ASD cohorts. There is also evidence that ASD may result from multiple hits (e.g., somatic mutations causing focal brain malformations) as neuropathologic studies of genetic ASD sometimes reveals concurrent focal brain malformations (e.g., 15q11-13 duplication, [2]). Many neuropathologic studies of ASD have focused on the cerebral cortex, hippocampus, or cerebellum; a few have focused on the amygdala or brainstem. Future studies should focus on the neuronal circuits that underlie the major behavioral deficits of ASD (impaired sociability and increased repetitive behavior) and the frequent ASD comorbidities (movement disorders, disruptive behavior, intellectual and language disabilities, epilepsy). Each behavioral and neurologic problem will likely arise from a distinct and/or specific distributed neuronal circuitry, for which we currently have limited understanding. The neuronal circuits controlling aggression have been explored in depth [31,32]. This example illustrates that behaviorally relevant neuronal circuits are likely to be first mapped using genetically defined subtypes of ASD in experimental animals, allowing precise molecular genetic interventions within particular brain regions and cell

types of a neuronal circuitry. Each subtype of ASD will have a distinctive set of behavioral deficits that demand studies aimed at distinct neuronal circuitries. This approach should be considered when interpreting the significance of the current literature and designing future studies. The significance of the published transcriptional, protein, and morphologic results, should also take into account the clinical conditions preceding death. Agonal events producing episodic hypoxia and/or ischemia can alter morphology, inflammatory cell reactions, mRNA and protein expression levels. Similarly, if a patient died following recent seizures, changes due to excitotoxic cell death, seizure-induced innate immune responses, changes related to neural circuit plasticity, anti-epileptic (homeostatic) and pro-epileptic changes in mRNA and protein expression are likely to be present. Depending on the seizure type, these changes will variably impact different brain regions and cell types. Seizures produce changes that can last for months. As described below, many genetic subtypes and idiopathic forms of ASD have comorbid epilepsy. ASD is one of the most strongly genetic of the behavioral disorders, with monozygotic twin concordance rates of 88% in males and 100% in females in one report [33]. Importantly, many genetic causes of ASD have variable penetrance (e.g., 16p11.2 deletion, interstitial duplication 15q11-13). An initial question was whether ASD might arise from a limited number of genetic defects, making the group relatively uniform, or if, instead, it might be a highly heterogeneous group of behavioral disorders. Clinical characterization already suggested heterogeneity with variable severities (ASD to Asperger’s disorder) and comorbidities (as described earlier). A common feature across published studies in ASD is the varied neuropathologies within each case cohort: 1. brain size: megalencephaly to microcephaly 2. malformations: cortical, hippocampal, cerebellar dysplasia, or altered brainstem nuclei 3. Purkinje neuron loss: severe to none 4. innate immune activity: severe, moderate, mild, or none.

Neuropathology of idic(15) and undefined ASD In 1991, Gillberg et al. [34] reported six boys with a partial trisomy of chromosome 15 who had severe to moderate intellectual disability, autistic behavior, and epilepsy. In 1997, Cook et al. [35] reported a family with a mother who transmitted an extranumerary isodicentric chromosome 15, or idic(15) (with two extra copies of the genomic region) to two children with ASD but not to an unaffected sibling. The unaffected mother inherited the idic(15) chromosome from her father suggesting an imprinted gene. Angelman syndrome results from deletions of the same maternally derived 15q11-13 region and is characterized by severe intellectual and language disability, ataxia and tremor, seizures, and a strong interest in social interactions (despite the severe language impairment). A subset of Angelman syndrome cases have mutations in the imprinted gene Ube3a [36,37]. That same year, it was reported that Ube3a was

479

Developmental Neuropathology expressed exclusively from the maternal allele in neurons of mice and humans [38]. In 2009, Glessner et al. [39] performed a genome-wide analysis of copy number variants (CNVs) in a large cohort of idiopathic ASD cases (2195 cases and 2519 controls) and found 15q11-13 duplications were the most frequent CNV in ASD, occurring in 0.7% of that ASD case cohort while being absent from controls. Hogart et al. [40] reported that 81% of individuals with idic(15) meet strict criteria for ASD using the Autism Diagnostic Interview—Revised. In 2011, Smith et al. [41] added extra copies of the Ube3a gene alone to mice and found this recreated ASD-related behavioral deficits (including reduced sociability) and impaired excitatory synaptic transmission in upper cortical layers. Idic(15) ASD brains (7 of 9 with epilepsy) were microcephalic (1177 g) relative to an idiopathic ASD cohort (1477 g, 4 of 10 with epilepsy) [2]. Hippocampal granule neuron heterotopias (in CA4 alveus or dentate molecular layer) were found 8 of 9 idic(15) cases, but only in 1 of 10 idiopathic ASD cases [2]. Other hippocampal dysplasias were also more frequent in idic(15) than undefined ASD. By contrast, cerebellar white matter heterotopias were found at an equivalent frequency [50% of cases for both idic(15) and idiopathic ASD]. Dysplasia in parts of the nodulus and flocculus, vermis, and focal polymicrogyria of the cerebellum were found in both ASDs. By contrast, 50% of idiopathic ASD cases had dysplasia of the cerebral cortex whereas cortical dysplasia was not found in any of the idic(15) ASD cases. Importantly, the severity of the behavioral deficits correlated with seizures in idic(15) [42] and seizure-associated regression is also reported [43]. Below, we describe the reported neuropathologic and molecular findings in undefined ASD cases. Then we emphasize specific genetic subtypes of ASD and their known neuropathologic features where available. As epilepsy is a common comorbidity in ASD, we also describe recently identified ASD genes independently identified in epilepsy. Additional neuropathological findings in ASD can also be found in other reviews [18,44–46].

(a)

(b)

(c)

Megalencephaly and microcephaly in ASD In some cases, brains from individuals with ASD have been larger (megalencephalic) or smaller (microcephalic) than those of control cohorts (Figure 40.1). Brain size could serve as a biomarker for ASD subtypes and could cluster genetic subtypes into specific cellular and molecular pathways. Seizures in some cases of ASD correlate with a progressive decrease in brain size concurrent with regression.

Figure 40.1 Megalencephaly and micrencephaly in autism spectrum disorder. (a) Lateral view of a brain weighing 1200 g from a 23-year-old male with autism. Note the broad simplified gyral pattern, small frontal lobe and foreshortened occipital lobe. (b) Lateral view of a brain weighing 1300 g from a 35-year-old male control. (c) Lateral view from a brain weighing 1660 g from a 42-year-old male with autism (reference range for brain weights, with 95% inclusion, 1179–1621 g).

Megalencephaly Phosphatase and tensin homologue deleted in chromosome 10 (PTEN) has dual protein and lipid phosphatase activity, and its lipid phosphatase activity, that negatively regulates the phosphatidylinositol 3-kinase/Akt pathway. Mutations in PTEN are associated with a broad spectrum of disorders, including Cowden disease-1, Bannayan-Riley-Ruvalcaba syndrome, and Lhermitte-Duclos disease. Klein et al. [47] found that 22% patients with ASD and macrocephaly harbor heterozygous

mutations in PTEN in association with extreme macrocephaly (>3 SD, 99.7th percentile). O’Roak et al. [48] found that the overall incidence among ASD cases is relatively low at 0.1%. Proteus syndrome associated with gigantism of hands and feet, nevi, hemihypertrophy, and macrocephaly is due to somatic activating mutations in a gene within the PTEN pathway, AKT1. Epilepsy is a frequent comorbidity in cases of ASD with PTEN mutations.

480

Autism Spectrum Disorders Chapter 40

Chromodomain helicase DNA-binding protein 8 (CHD8) mutations were originally discovered in whole exome sequencing studies of ASD cohorts [48,49] and CHD8 was disrupted in a rare balanced chromosome rearrangement [50]. Subsequent sequencing of larger ASD and control cohorts identified 15 truncating mutations in 3730 cases (0.4%), while finding no mutations in 8792 unaffected siblings [51]. ASD was diagnosed in 13 of the 15 mutation-carrying individuals indicating strong penetrance of the core behavioral deficits. Macrocephaly was also strongly penetrant (80%), but seizures were known to have occurred in only 20% of these cases. Individual cases of increased brain weight have been reported in various cohorts of undefined ASD [52–56]. Microcephaly Christianson syndrome, due to loss of function mutations in the X-linked gene NHE6 (a sodium-proton exchange protein) presents as ASD or Angelman syndrome with mutism, intellectual disability, and generalized tonic clonic epilepsy and is characterized by a delayed-onset postnatal microcephaly and cerebellar and brainstem atrophy [57]. Delayed microcephaly and epilepsy are also characteristic of Rett syndrome [58,59]. SCN1A mutation-associated migrating partial epilepsy of infancy, an epileptic encephalopathy, show regression and an acquired progressive microcephaly [60]. Cohen syndrome (ASD with mild facial dysmorphism and joint laxity) [61], idic(15) [2], DYRK1A heterozygous truncating mutations [62,63], and Kleefstra syndrome resulting from haploinsufficiency of EHMT1 on chromosome 9q34.3 [64] also have microcephaly.

Cerebral cortex The cortical thickness is reduced in a number of the genetic disorders with microcephaly/micrencephaly described above. By contrast, increased cortical thickness with heterotopic and maloriented pyramidal neurons were reported in a subset of undefined ASD cases by Bailey et al. [54]. Polymicrogyria was seen in two postmortem cases [55,65]. MRI studies have observed developmental cortical abnormalities in a small proportion of ASD cases (65–67). Bailey et al. [54] reported cortical dysgenesis in a subset of ASD cases. Somatic mutations in PIK3CA, AKT3, and mTOR have been identified in the neurons isolated from cases of focal brain overgrowth disorders, including hemimegalencephaly, which is often associated with severe drug-refractory epilepsy, intellectual disability, and ASD-related behavioral problems [69–71]. More numerous, smaller, and less compact minicolumns were reported in nine patients with ASD (seven with intellectual disability, five with epilepsy, four with megalencephaly) compared with nine controls [72]. The density of neurons in prefrontal cortex was reported to be increased by Courchesne et al. [20]. In a follow-up study using cortical layer-specific in-situ probes, the Courchesne group [73] found patches of morphologically normal cortex where mRNA expression was reduced. The dispersal of layer specific markers anticipated for a cortical dysplasia

was not reported. The foci of reduced mRNA staining did not correspond to decreased cell numbers and were not observed in occipital cortex. The significance of the findings remains unclear, but resemble the patches of NeuN mRNA loss reported in cortical seizure foci resected to treat epilepsy [74]. Hustler and Zhang [75] reported increased spine density in cortical layer II pyramidal neurons.

Cerebellum Reduced Purkinje cell density was seen in one case of ASD with concomitant epilepsy and profound intellectual disability by Williams et al. [52]. Autistic cases showed a decreased number of Purkinje cells in the cerebellar hemisphere and vermis in a case series from Ritvo et al. [65]. Six cases of ASD showed decreased numbers of Purkinje cells in a study by Kemper and Bauman [55]. Bailey et al. [54] reported low Purkinje cell counts in all five of their adult ASD cases. Lee et al. [76] observed a decreased number of Purkinje cells in two ASD cases (both with intellectual disability, one with epilepsy). Fatemi et al. [77] reported a 24% decrease in mean Purkinje cell size in a group of patients with ASD. Neurotransmitter systems Blatt [44] reported on the GABAergic, serotonergic, cholinergic and glutamatergic receptor binding sites in the postmortem ASD brain. Guptill et al. [78] reported a 20% decrease in GABAA receptor binding sites in hippocampus of ASD subjects compared with controls. Oblak et al. [79,80] reported reduced GABAA receptor binding sites in superficial layers of anterior cingulate cortex (47%), posterior cingulate cortex (49%), and fusiform gyrus (31%) in ASD. Oblak et al. [81] reported reduced GABAB receptor binding sites in cingulate cortex, most prominent in superficial layers (35% decrease). Fatemi et al. [82] reported reduced glutamic acid decarboxylase (GAD), the ratelimiting enzyme responsible for the conversion of glutamate to GABA in ASD. GAD65 protein was reduced by 50% in the cerebellum and by 48% in the parietal cortex of ASD. GAD67 protein was reduced by 51% in the cerebellum and by 61% in the parietal cortex of ASD. Yip et al. [83] reported a 40% decrease in GAD67 mRNA in Purkinje neurons of posterolateral cerebellum in ASD. These authors also reported a 28% increase in GAD67 mRNA staining of cerebellar basket cells [84]. Azmitia et al. [85] reported an increased serotonergic axon terminal density in the cortex of individuals with ASD. Oblak et al. [86] reported reduced 5HT1A and 5HT2A receptor binding sites in the posterior cingulate cortex and fusiform gyrus in ASD. A subset of ASD presents with hippocampal sclerosis (two cases of idic(15), [87]). An SCN2A case of Kamiya et al. [3] showed moderate diffuse atrophy by MRI, associated epilepsy, intellectual disability, psychomotor restriction, and ASD features. Cases of SCN1A regression show an acquired progressive microcephaly [60] and delayed mesial temporal sclerosis [88] that then evolves into mesial temporal lobe epilepsy

481

Developmental Neuropathology (focal seizures arising from the hippocampal formation and medial temporal lobe). Unilateral cortical status epilepticus can cause crossed cerebellar diaschisis with Purkinje cell and internal granule neuron depletion and associated cerebellar atrophy [89,90]. The innate immune system and glia are activated in the ASD brain. There is increased expression of immune system-related genes in ASD postmortem brain relative to controls [91,92]. Voineagu et al. [92] identified a large increase of S100a8 mRNA, a native toll-like receptor 4 (TLR4) agonist in ASD. Systemic or direct brain inoculation with the TLR4 agonist lipopolysaccharide is a robust stimulant of the brain’s innate immune system. TLR4 receptors have been shown to be important as an acute and chronic pro-epilepsy pathway [93]. TLR4 is increased in neurons and astroglia (not in microglia) of human mesial temporal lobe resections removed for temporal lobe epilepsy and the levels correlate with the frequency of seizures per month prior to removal [94]. Cytokines, chemokines, and growth factors are increased in ASD brain tissue and cerebrospinal fluid [24]. Li et al. [95] confirmed an increase of tumor necrosis factor alpha, interleukin 6, granulocyte-macrophage colony-stimulating factor, interferon gamma and interleukin 8 in ASD compared with control brains. Microglia-specific transcripts that are involved in microglial cellular responses were elevated in ASD relative to controls [96]: TREM2 (1.75-fold), DAP12 (1.5-fold), and CX3CR1 (1.34-fold). Experimental evidence indicates microglia play a critical role in synaptic pruning during postnatal brain development. In mice lacking CX3CR1, a chemokine receptor expressed by microglia in the brain, microglial cell numbers were transiently decreased and synaptic pruning was concurrently delayed [97]. The transient deficit in microglial numbers and increase in spine density was associated with a persistent deficit of prefrontal cortex–hippocampus connectivity as assessed by local field potential oscillatory coherence and functional MRI measurements, as well as some deficits in social behavior [98]. These studies suggest that defects in microglial function might underlie the increase spine density reported in ASD. The marker of activated astroglia, GFAP mRNA, was elevated 1.7-fold in ASD relative to controls. In the cerebellum, GFAP mRNA was more increased (2.63-fold) in ASD, yet surprisingly, microglial markers were slightly decreased [96]. GFAP protein was also increased 45–75% in frontal, parietal, and cerebellar cortex of ASD relative to control by Laurence and Fatemi [99]. Astroglial cell density was increased whereas the number of branching processes and the branch lengths were reduced in frontal cortex of ASD brains relative to controls, yet these structural properties of astroglia were not altered in cerebellum [100]. Aquaporin 4 protein, a water channel localized to the astroglial perivascular foot processes was reduced in ASD cerebellum relative to controls [101], whereas in hypoxia/ischemia and central nervous system, trauma aquaporin 4 is increased [102]. Rare subsets of ASD have intraneuronal tau inclusions. ASD due to NHE6 mutations (Christianson syndrome) display tau-positive inclusions [103]. Similarly, ASD due to

482

haploinsufficiency of ADNP [104] is associated with tau inclusions in animal models of this condition. Genetics, animal models and pathogenesis Neurodevelopmental disorders have been found in some cases to result from too little or too much of the same genetic locus, suggesting a steep gene dose-dependent effect in disease pathogenesis; for example, maternally derived deleted/duplicated/triplicated 15q11-13 (UBE3A: Angelman syndrome/ASD), deleted/duplicated 7q11.23 (GTF2I-GTF2IRD1: Williams syndrome/ASD), deleted/duplicated DYRK1A (DYRK1A: truncating mutations in ASD/trisomy 21-Downs syndrome) and deleted/duplicated 16p11.2. Such reciprocal gene deletion/mutation/duplication loci are likely to harbor a single specialized gene capable of this steep dosage-dependent effect as observed for UBE3A in 15q11-13 [41]. Imprinted genes like UBE3A that are expressed exclusively from one of the parental alleles in neurons (UBE3A is expressed only from the maternal allele in neurons) implicate a dose-dependent effect on neuronal function and duplications of that single expressed allele will have the greatest impact. Multiple genetic defects can coexist in a single individual and may interact in disease pathogenesis. For example, Autism BrainNET Case AN16641 has multiple simultaneous mutations in ARID1B, CACNA1C, and SLC6A8 that have individually been implicated in ASD [30]. Focal brain malformations (due in most cases to a somatic gene mutation) often coexist with pathogenic germline genetic defects (e.g., SCN1A-Dravet [105]; 15q duplication [2]), implicating a two-hit model of ASD pathogenesis. Such focal regions of dysplastic circuit development are often prone to epileptiform discharge. Somatic defects, in addition to germline genetic defects, have been found in ASD brain tissues, suggesting that a subset of genetically determined ASD cases may go undetected until the affected brain tissue itself is tested for genetic defects [30]. Unlike morphologically visible forms of somatic mutation, such as cortical dysplasia, these may simply impair the function of a subset of neurons within a focal brain region causing epilepsy or other circuit dysfunctions. Genetic defects found in ASD overlap with those found in epileptic encephalopathy (Table 40.1), mental restriction (Table 40.2), intellectual disability (Table 40.3), schizophrenia [106] and a subset of early onset muscular dystrophies. Genetic overlap of ASD with early infantile epileptic encephalopathy and the existence of other genetic ASD–epilepsy comorbid disorders also suggest that epileptiform discharges may play an etiologic role in the behavioral problems present in subsets of ASD cases. Examples of this ASD–epilepsy overlap are described below. Mutations of SCN1A were found in multiple independent exome sequencing studies of individuals with ASD [107]. SCN1A inactivating mutations are also responsible for Dravet syndrome (also called generalized epilepsy with febrile seizures plus), or severe myoclonic epilepsy of infancy. SCN1A encodes the voltage-gated sodium channel Nav1.1. Many GABAergic neurons are fast spiking, firing action potentials with very high

Autism Spectrum Disorders Chapter 40

Table 40.2 Genes: autism spectrum disorder and mental restriction (Mendelian Inheritance in Man, SFARI). Heritance Autosomal-dominant Autosomal-dominant Autosomal-dominant Autosomal-dominant Autosomal-dominant Autosomal-dominant Autosomal-dominant Autosomal-dominant Autosomal-dominant Autosomal-dominant Autosomal-dominant Autosomal-dominant Autosomal-dominant Autosomal-dominant Autosomal-dominant Autosomal-dominant Autosomal-recessive Autosomal-recessive Autosomal-recessive Autosomal-recessive

EIEE Type 1 5 6 7 12 19 21 23 24 26 29 30 32 33 37 39 3 6 37 38

Gene Mbd5 Syngap1 Grin2b Dyrk1a Arid1b Ctnnb1 Ctcf Setd5 Deaf1 Auts2 Setbp1 Zmynd11 Kat6a Dpp6 Pogz Myt1l Cc2d1a Grik2 Ank3 Herc2

EIEE, early infantile epileptic encephalopathy

rates. Yu et al. [107] showed that heterozygous deficiency of Nav1.1 causes failure to sustain high action potential firing rates of GABAergic neurons in the hippocampus, which would otherwise help to restrain the excitatory circuitry from entering an epileptiform discharge pattern. Nav1.1 is necessary to sustain high firing rates in GABAergic neurons that express somatostatin (Martinotti cells, controlling horizontal action potential transmission) and parvalbumin (controlling vertical action potential transmission) [108]. The overlap between Nav1.2 (encoded by SCN2A that is also mutated in ASD) and Nav1.1 in somatostatin GABAergic neurons that restrict horizontal transmission of excitability suggests that this GABAergic neuron subtype might be critical to limiting the development and dissemination of seizures which may result in the comorbid behavioral decline and disabilities found in the epilepsy-associated subtype of ASD [109]. Han et al. [110] reported that heterozygous SCN1A deletions targeted selectively to forebrain GABAergic neurons caused deficits in social preference/interaction (modeling social deficits of ASD) and in learning and memory tasks (modeling intellectual disability). Importantly, these deficits in social and learned behavior were rescued within 30 minutes of a single dose of the benzodiazepine clonazepam, a positive allosteric modulator of GABAA receptors that would in principle bolster neurotransmission of the failing GABAergic neurons. Of the genes carrying missense mutations by whole exome sequencing in ASD, SCN2A mutations were among the most frequent [63]. SCN2A encodes the Nav1.2 voltage-gated sodium

channel. Kamiya et al. [3] reported mutations in this gene in an individual with refractory infantile epilepsy and mental decline. The mutation truncated to inactivate channel function, but also caused a dominant negative effect on coexpressed wild-type Nav1.2 channels. Importantly, Nav1.2 accumulates at the axon initial segment of the inhibitory somatostatin, but not parvalbumin, GABAergic neurons. Using specific toxin inhibitors, the channel was found to be necessary to sustain high action potential firing rates in somatostatin neurons. Inhibiting the channel also caused runaway epileptiform discharge of cortical neuronal circuits in acute brain slices [111]. Heterozygous mutations in STXBP1 on chromosome 9q34.11 result in an early onset epileptic encephalopathy with burstsuppression on EEG (i.e., “Ohtahara syndrome”). These patients often have severe global developmental delay, lack of speech, hypotonia, and ataxia and can present with choreoballistic movements, generalized tremors, and dystonia. A distinctive movement disorder, a “figure-of-eight” head stereotypy sometimes with concurrent vocalizations, can be a diagnostic clue. They also have hypsarrhythmia and burst suppression on EEG. STXBP1 is a neuron-specific, syntaxin-binding protein that regulates synaptic vesicle docking and fusion. Heterozygous deficiencies of the protein lead to a depletion of the readily releasable pool of synaptic vesicles and therefore would also impair synaptic transmission at high firing rates [112]. Importantly, as shown above, GABAergic neurons must fire at very high rates to prevent the runaway epileptiform discharge of the excitatory, glutamatergic neuronal circuits. Therefore, like the defective inhibitory neuron firing that arises from insufficient voltage-gated sodium channels (SCN1A and SCN2A deficiency), defective rapid neurotransmitter vesicle recycling (STXBP1 deficiency) could also lead to seizures. Mutations in the X-chromosomal protocadherin gene, PCDH19, first found in families with epilepsy and intellectual disability limited to females, have since been found in female patients with SCN1A-negative Dravet syndrome. Cadherins mediate cell–cell adhesion and are involved in intracellular signaling pathways associated with neuropsychiatric disease. Future studies are anticipated to link the loss of cadherin PCDH19 to defective GABAergic neurotransmission based on pathophysiologic mechanisms in other genetic causes (SCN1A) of Dravet syndrome. Mutations in GRIN2B encoding GluN2B, a subunit of the NMDA receptor, have been reported in both ASD and epileptic encephalopathies [113–116]. Mutations occurred in the ion channel-forming reentrant loop and resulted in increased calcium permeability and a gain of function. As an example, one girl presented at seven weeks with spasms, lack of eye contact, and hypotonia. She also showed episodic hyperextension of axial muscles. EEG showed hypsarrhythmia. At age five years, she could not sit independently or speak. She had severe feeding problems, mild microcephaly, autistic-like behavior, and occasional seizures. In mice, selective deletion of GRIN2B in interneurons resulted in seizures after the second postnatal

483

Developmental Neuropathology

Table 40.3 Chromatin remodeling genes in autism spectrum disorder and intellectual disability (adapted from Ronan et al. [190]). Gene

Syndrome

Type of deletion

Gene

ARID1A (BAF complex) ARID1B (BAF complex)

Non-sense, frameshift indel Translocation, frameshift indel, nonsense, missense, microdeletion Partial deletion, missense, intronic alteration

Chromatin-remodeling complex subunit Chromatin-remodeling complex member

Missense In frame deletion, missense

Chromatin-remodeling complex ATPase Chromatin-remodeling complex subunit

Missense Splice site mutation Missense Deletions and mutations

CHD7

Coffin–Siris syndrome Intellectual disability, Coffin–Siris syndrome, autism, schizophrenia Coffin–Siris syndrome, Nicolaides–Baraitser syndrome, schizophrenia Coffin–Siris syndrome Coffin–Siris syndrome, Kleefstra syndrome phenotypic spectrum Autism Autism Autism Autism, intellectual disability, epilepsy, Lennox–Gastaut syndrome CHARGE syndrome, autism

Missense

CHD8

Autism

Non-sense, frameshift indel missense

ATRX

X-linked α-thalassemia/mental restriction syndrome Rubinstein–Taybi syndrome

Missense

Chromatin-remodeling complex subunit Chromatin-remodeling complex subunit Chromatin-remodeling complex subunit Chromodomain-helicase-DNA-binding protein 2 Chromodomain-helicase-DNA-binding protein 7 Chromodomain-helicase-DNA-binding protein 8 Chromatin remodeler

SMARCA2 (BAF complex)

SMARCA4 (BAF complex) SMARCB1 (BAF complex) SMARCC1 (BAF complex) SMARCC2 (BAF complex) PBRM (BAF complex) CHD2

p300 CBP KAT6B (MYST4 or MORF) HDAC4 EZH2 EHMT1 MLL MLL2 MLL3 KDM5C (JARID1C) PHF8 HUWE1 MECP2 MBD5 MED12

MED23 TBR1 ADNP MEF2C

484

Rubinstein–Taybi syndrome Say–Barber–Biesecker–Young–Simpson syndrome (SBBYSS or Ohdo syndrome) Brachydactyly mental restriction syndrome Weaver syndrome (learning disability) Kleefstra syndrome phenotypic spectrum; autism Wiedemann–Steiner syndrome Kabuki syndrome Autism, Kleefstra syndrome Non syndromic X-linked mental restriction X-linked mental restriction XLMR Turner type Rett syndrome, Angelman syndrome, X-linked mental restriction, autism Autism, Kleefstra syndrome Lujan–Fryns syndrome, FG syndrome (also known as Opitz–Kaveggia syndrome) Non-syndromic intellectual disability Autism Autism, mild to severe intellectual disability Autism, nonverbal, severe intellectual disability, seizures

Chromatin-remodeling complex ATPase

Large deletions and duplications, missense Microdeletions, non-sense Frameshift indel, missense

Histone acetyltransferase

Balanced chromosomal translocation; deletion Missense, frameshift indel Microdeletions, non-sense, frameshift, missense Non-sense Non-sense, frameshift Missense, non-sense Missense, frameshift, non-sense, intronic alteration Missense mutation, non-sense Duplications, missense, copy number gains Missense, non-sense, frameshift indel, duplication Frameshift indel, non-sense Missense

Histone deacetylase

Histone acetyltransferase Histone acetyltransferase

Histone methyltransferase Histone methyltransferase Histone methyltransferase Histone methyltransferase Histone methyltransferase Histone lysine demethylase Histone lysine demethylase Histone ubiquitylation DNA methylation binding protein DNA methylation binding protein REST mechanism for disease, Mediator complex subunit

Missense Deletion, mutation Mutations

Mediator complex subunit Activity-induced transcription factor Activity-induced transcription factor

Deletions, mutations

Activity-induced transcription factor

Autism Spectrum Disorders Chapter 40

Table 40.3 (Continued) Gene

Syndrome

Type of deletion

Gene

CTNNB1

Autism, intellectual disability, microcephaly Angelman syndrome, non-verbal autism, epilepsy X-linked mental restriction, epilepsy Intellectual disability, craniofacial anomalies

Mutations

Transcriptional coactivator

Missense, non-sense, frameshift indel, duplication Mutations Translocation and deletion

Transcriptional coactivator

UBE3A ARX PHF21A (BHC80)

week, and death shortly afterwards [117]. Prior to the onset of seizures, glutamatergic synapses onto hippocampal GABAergic interneurons were shown to be reduced, again implicating defective GABAergic inhibition (failure to excite the GABAergic neurons) in this autism–epilepsy comorbidity. GRIN2A mutations (heterozygous) encoding GluN2A, another subunit of the NMDA receptor, have been found in a syndrome that includes focal epilepsy with speech disorder and with or without mental restriction, also designated Landau– Kleffner syndrome, continuous spike and waves during slowwave sleep syndrome, and Rolandic epilepsy, mental restriction, and speech dyspraxia, autosomal-dominant [113,116,118,119]. Evidence suggests this disorder may be due to loss of function and sometimes to dominant negative mutations in a subunit of the NMDA receptor subunit glutamate-gated cation channel, which mediates excitatory synaptic neurotransmission and long-term potentiation of synapses. NMDA receptor antagonists like ketamine cause an anomalous circuit hyperexcitability, presumably by antagonizing receptors on GABAergic neurons that reduces their activity to disinhibit excitatory circuits. Alternately, the mechanism may involve postnatal development of glutamate synapses onto GABAergic neurons as recently implicated for GRIN2B loss [117]. CDKL5 (Xp22.13) mutations cause an X-linked infantile spasm syndrome, also known as X-linked West syndrome, and Rett syndrome, characterized by severe intellectual disability with lack of speech, generalized tonic–clonic seizures (less than five months of age) that progress to infantile spasms and then refractory myoclonic epilepsy in the majority of individuals, and bruxism [120,121]. EEG with CDKL5 mutations is unique at least in neonates and young infants, as the ictal pattern is an initial bilateral, synchronous electrodecrement followed by repetitive spikes and sharp waves [122]. CDKL5 is a member of Ser/Thr protein kinase family and encodes a phosphorylated protein with protein kinase activity. Mouse models of CDKL5 knockout have been generated and were found to show various behavioral deficits [123,124]. Loss of CDKL5 results in decreased phosphorylation of the protein kinase B/glycogen synthase kinase 3β pathway, causing increased activity of the glycogen synthase kinase 3β, and reduced hippocampal dentate gyrus granule neuron precursor survival and maturation

Transcription factor Chromatin reader, histone deacetylase complex member

associated with impaired hippocampal-dependent cognitive performance [125]. These hippocampal defects were restored by treatment with the glycogen synthase kinase 3β inhibitor SB216763 [126]. DYRK1A mutations (heterozygous) have been found in cases of ASD with epilepsy, microcephaly, and severe intellectual and language disabilities. As summarized in Krumm et al. [126], recurrent mutations in DYRK1A have been identified in ASD cohorts (3/2446), but are absent from controls (0/6503) providing strong evidence that haploid insufficiency of this gene underlies a subset of ASD-epilepsy co-morbidities. Van Bon et al. [62,127,128] linked mutations in DYRK1A to a syndromic form of ASD and intellectual disability that includes microcephaly, intrauterine growth restriction, febrile seizures in infancy, impaired speech, stereotypic behavior, hypertonia, and a specific facial gestalt. Hoischen et al. [63] also highlight the many CNVs disrupting this gene. Courcet et al. [129] reported a patient with growth restriction, primary microcephaly, facial dysmorphism, seizures, ataxic gait, absent speech, intellectual disability, and a microdeletion of 69 kb of the genome that disrupts the promoter of DYRK1A, the dual specificity tyrosine-phosphorylationregulated kinase 1A. They also summarize a series of additional cases with similar clinical features where deletions and balanced chromosomal rearrangements localize to DYRK1A. Studies of Down syndrome where its gene dosage is increased provide further evidence that DYRK1A is a dosage-sensitive gene. Normalizing the DYRK1A gene dosage in a mouse model of Down syndrome improved many behavioral, neuromorphological, and electrophysiological parameters [130]. DYRK1A (minibrain kinase in Drosophila) encodes a member of the dual-specificity tyrosine phosphorylation-regulated kinase (DYRK) family. This member contains a nuclear targeting signal sequence, a protein kinase domain, a leucine zipper motif, and a highly conservative 13-consecutive-histidine repeat. It catalyzes its autophosphorylation on serine/threonine and tyrosine residues. DYRK1A acts to phosphorylate and thereby dynamically calibrate synaptojanin function at the Drosophila neuromuscular junction and in turn enhances endocytic capacity to adapt to conditions of high synaptic activity [131]. One could again speculate that deficiencies of this synaptic adaptation in fast-spiking GABA neurons could result in failure of synaptic vesicle recycling to cause

485

Developmental Neuropathology GABAergic synapse disinhibition. This model would be consistent with that attributed to SCN1A, SCN2A, and STXBP1 deficiencies, as described above. However, DYRK1A also regulates gene expression in part through its effects on the REST/NRSFSWI/SNF chromatin remodeling complex [132], and multiple genes implicated in ASD are also involved in the SWI/SNF chromatin remodeling complex including CHD8, ADNP, ARID1B, and SMARCC2 [127].

Microdeletion/microduplication syndromes 15q11-13 deletion/duplication syndromes Angelman syndrome (15q11-13 deletion or UBE3A mutations) presents with severe developmental delay combined with seizures, ataxia, hypermotoric behaviors, and absent speech [133,134]. Cognitive abilities are stronger than receptive language, which is stronger than expressive language skills. Most individuals with Angelman syndrome do not speak; a minority express a few single words, and only a few conduct short threeto four-word sentences. Sleep dysfunction includes difficulty falling asleep and waking up multiple times at night. Behavioral characteristics include a short attention span, an unusually happy demeanor, easily provoked laughter, mouthing of objects, and fascination with water. In contrast to ASD, most children with Angelman syndrome enjoy social interactions and seek eye contact with adults. Regression in Angelman syndrome is unusual unless seizures are poorly controlled. Abnormal EEG patterns are present in all cases. Global white matter abnormalities with delayed myelination and hypoplastic corpus callosum have been found using MRI. In maternally derived idic(15) chromosomes, hypotonia in infancy is pronounced [40]. Developmental delay is seen, including gross and fine motor skills and cognitive impairment ranging from moderate to severe. Seizures, including infantile spasms, occur in at least 50% of cases. Intractable epilepsy is associated with regression in skills. Late-onset Lennox–Gastaut syndrome in a patient with 15q11.2–q13.1 duplication has been reported [135]. Early developmental milestones, motor development, language, and behavior were normal. The child walked at 11 months and, initially, speech and sociability were appropriate for age. However, by two and a half years of age, the child displayed regression with progressive cognitive and behavioral decline. From that time, the girl began to show learning and expressive communication difficulties that progressed to aphasia, poor motor coordination, reduced social interaction, and repetitive and handwashing stereotypic movements. This provides an example of seizure-associated regression. From eight years, she suffered from atypical absences, astatic, tonic and complex partial seizures starting in the left frontal region with secondary generalization. The 20–30 daily episodes sometimes evolved to status epilepticus. In idic(15), behavioral problems, including impulsivity, self-injurious, and/or aggressive behaviors, hyperactivity and anxiety can appear in early childhood and often increase in adolescence. Sudden, unexplained deaths among seemingly healthy individuals with idic(15) have

486

been identified, as well as deaths in patients with idic(15) who manifest a chronic degenerative phenotype associated with relentless seizures. Eighty-one percent with idic(15) meet strict criteria for autism using the combined Autism Diagnostic Interview Revised. Some regression in socialization appears to be part of the phenotype, as infants and toddlers with idic(15) often can appear more social and make better eye contact and vocalize more than they do as older children. Socialization is impaired, with decreased eye contact and lack of reciprocity. Children with idic(15) also typically display numerous repetitive and stereotyped behaviors (rocking and hand flapping) that are often directed toward sensory stimulation. A few cases of hexasomy for chromosome 15q11.2-q13 have been identified with profound mental restriction and intractable epilepsy. Comorbid schizophrenia and epilepsy have also been diagnosed in some individuals with maternal 15q11-13 duplication [136]. Rett syndrome (MECP2 mutation) and MECP2 duplication Heterozygous MECP2 mutations on chromosome Xq28 in girls result in Rett syndrome, presenting with regression that may stabilize or improve over time. This can manifest by the loss of purposeful hand skills and verbal language by the age of five years after a period of normal development (typically in the first six months of life). Patients subsequently lose their ability to ambulate, with some developing quadriplegia resulting in a state of frozen rigidity. Girls with Rett syndrome also develop stereotypic hand movements, abnormal breathing patterns with episodic hypoventilation and hyperventilation. Children with Rett syndrome, like those with Angelman syndrome, seek eye contact, a feature opposite to classic ASD. Many girls with Rett syndrome develop seizures, impaired sleep patterns, and inappropriate laughter. The EEG patterns in Rett syndrome are distinct from those seen in Angelman syndrome and include background slowing. Epilepsy is found in 60% with Rett syndrome, but seldom before the age of two years. Epilepsy incidence progressively increases with age: 30% at 2–5 years, 62% at 5–10 years and 85% by 15–30 years. Individuals with seizures had greater overall clinical severity, and greater impairment of ambulation, hand use, and communication [137], further suggesting that the seizures themselves may contribute to the behavioral disorder. Duplication of the entire MECP2 in males (i.e., “MECP2 duplication syndrome”), diagnosed by a high-resolution chromosomal microarray, results in severe intellectual disability with minimal speech, and developmental regression associated with seizures that may be intractable. There is an initial infantile hypotonia and motor developmental delay with ataxic gait but a subsequent loss of ambulatory skills due to progressive spasticity of the lower limbs, abnormal involuntary movements, such as choreiform movements, and stereotypic hand movements. EEG pattern typically shows an unusually slow background. Some female carriers of MECP2 duplication with random Xinactivation may have intellectual disability, speech delay, and seizures.

Autism Spectrum Disorders Chapter 40

22q13.3 deletion/duplication syndromes (SHANK3) Phelan–McDermid syndrome (22q13.3 deletion, SHANK3) presents with moderate to severe global developmental delay with absent or minimal speech, neonatal hypotonia that may persist into adulthood, feeding difficulties in infancy, and mouthing behaviors [133]. However, they have normal or even rapid physical growth, large ears, large hands, and dysplastic toenails, features that are not typically associated with Angelman syndrome. Most exhibit impaired social interactions and shun eye contact, unusual for people with Angelman syndrome. Some demonstrate age-appropriate babbling with limited vocabulary until three or four years old, when they lose their expressive language skills. Although SHANK3 has been implicated as the critical gene in the chromosome 22q13.3 deletion syndrome, haploinsufficiency of other genes in the region influence the phenotypic expression and severity. Radiologic studies have revealed a thin corpus callosum, abnormal white matter, cerebellar vermis hypoplasia, and enlarged posterior fossa. Approximately half have a history of seizures, mainly the febrile type. An example case of 22q13.3 duplication syndrome is 11-year-old girl diagnosed with attention-deficit hyperactivity disorder who also had developmental delay and learning problems, kleptomania, destructive behavior, auditory overstimulation, and hyperphagia. She experienced two generalized tonic–clonic seizures at the age of two years without associated fever or other precipitants. Shank3 heterozygous knockout mice targeting multiple different subdomains of the gene have been generated and studied and reconstitute a variety of behavioral deficits and produce synaptic defects (reviewed by Jiang and Ehlers [138]). 7q11.3 deletion/duplication syndromes (GTF2IRD1, GTF2I) Berg et al. [139] reported children with 7q11.3 duplication syndrome (GTF2IRD1, GTF2I) with psychomotor and developmental delay, significant language impairment, and multiple behavioral features usually associated with ASD, including difficulties in communication and eye contact, repetitive actions, and anxiety or withdrawal. Some also had disruptive or aggressive behavior. 7q11.3 duplications were also found in a large ASD cohort [140]. 7q11.3 deletion results in Williams–Beuren syndrome, characterized by an unusually cheerful demeanor and ease with strangers. Most individuals are highly verbal relative to their intelligence quotient, and are overly sociable, having what has been described as a “cocktail party” type personality. Individuals with Williams–Beuren syndrome are reported as hyperfocused on the eyes of others in social engagements. In contrast to Angelman syndrome (15q11-13 deletion/Ube3a mutation), the developmental delay in Williams–Beuren syndrome is coupled with strong language skills. Transcription factor (TFII-I) is a multifunctional protein involved in the transcriptional regulation of critical developmental genes, encoded by the GTF2I gene located on chromosome 7q11.23. Heterozygous deletions of this gene alone produce behavioral and transcriptional abnormalities [141–144]. Heterozygous deletions of the related adjacent

gene, GTF2IRD1, within the 7q11.3 region, in mice also implicate this gene in the behavioral symptoms of Williams–Beuren syndrome [145,146]. Duplications of GTF2I also produce behavioral symptoms in mice, but analogs of ASD social deficit were not examined [147]. 2q23.1 deletion/duplication syndromes (MBD5) Individuals with chromosome 2q23.1 deletion (MBD5) present with severe intellectual disability, motor delay, variable degrees of speech delay ranging from completely nonverbal to speaking in short sentences, autistic and maladaptive behaviors, short attention span, aggression and self-injurious behavior, and a high incidence of seizures [148,149]. Microcephaly is reported in over 80%, but in only one of eight with disruptions or intragenic deletions in MBD5 alone; similarly, ataxia or unusual gait is reported in more than 70% with the chromosomal deletion but in none with disruptions or deletions in MBD5 alone. This finding suggests that other genes explain the additional features in chromosome 2q23.1 haploinsufficiency. Individuals with 2q23.1 duplication are similar in many ways with the exception of an unusual likeable/affable personality [149]. 17q21.31 deletion/duplication (KANSL1) Individuals with microdeletion or mutation of the KANSL1 gene, which encodes a nuclear protein that plays a role in chromatin modification and is a member of a histone acetyltransferase complex, have Koolen–De Vries syndrome. This presents with mild to moderate degrees of developmental delays and intellectual disability, but severe speech and language delays, and neonatal/childhood hypotonia. Individuals are friendly and happy, sometimes laughing easily or frequently. Hypersociability is observed in some adults. Point mutations in the KANSL1 gene are sufficient for full manifestations of chromosome 17q21.31 deletion syndrome, and indicated that it is a monogenic disorder caused by haploinsufficiency of KANSL1 [150]. Duplications of 17q21.31 have also been reported, all with some degree of psychomotor restriction, poor social interaction, and communication difficulties reminiscent of ASD [151]. Of four independent cases reported, two had hypotonia, two had hyperactivity, two had aggressive behavior, and one had obsessive behavior. Two had poor motor skills, and two showed tiptoe walking. 9q34.3 deletions (Kleefstra syndrome)/duplication syndrome Kleefstra syndrome results from haploinsufficiency of EHMT1 on chromosome 9q34.3. The clinical features reported in both Angelman and Kleefstra syndromes include moderate to severe intellectual disability with minimal speech but better receptive language, hypotonia in childhood, and sleep disturbances with multiple awakenings [64,133]. Mildly affected individuals have more than 100 words in their vocabulary and speak in sentences. Only 30–40% develop seizures, and no consistent EEG patterns have been identified. Around 50–75% of individuals are characterized by aggression, self-mutilation, emotional outbursts, and

487

Developmental Neuropathology autistic behaviors. Some adolescents with Kleefstra syndrome develop unusual behaviors such as extreme apathy and catatonia, as well as mood disorders. Developmental regression, (unusual in Angelman syndrome) has been observed in some adolescents with Kleefstra syndrome. A mouse model with heterozygous loss of Ehmt1 displayed behavioral abnormalities including increased anxiety and increased sociability [152]. Individuals with dup 9q34 understand simple directions, have developmental delay, psychomotor restriction, learning disabilities, and a limited vocabulary [153]. A subset will display behavioral problems (temper tantrums, head banging, and/or aggressive behavior). Xq13.3–q21.1 deletion/duplication (ATRX) X-linked alpha-thalassemia/intellectual disability (mental restriction) syndrome, due to deletions, duplications, or mutations within the ATRX on chromosome Xq21.1, presents in males with mild to severe intellectual disability, and some never acquire speech or independent ambulation [133]. Microcephaly, skeletal and genital abnormalities, and severe neonatal hypotonia are also present. Most are described as “affable,” but some are “emotionally labile” with episodes of laughter and crying. Epilepsy occurs in a subset of cases [154]. In a mouse model of ATRX deletion, the cerebral cortex is thinned (consistent with microcephaly), and there is increased apoptosis of neurons in the early cortical plate [155]. The mouse model also helped establish that ATRX partners with cohesin and MeCP2 to developmentally silence imprinted genes in the brain [156]. Mouse studies also helped identify a defect in muscle growth as a potential mechanism for the severe neonatal hypotonia [157]. Duplications of Xq13.3-q21.1 encompassing ATRX were reported in two individuals with severe intellectual disability, absent expressive speech, early hypotonia, behavior problems (hyperactivity, repetitive self-stimulatory behavior), postnatal growth deficiency, and microcephaly [158,159].

Single-gene disorders Pitt–Hopkins syndrome (TCF4) Whereas haploinsufficiency or dominant negative mutants of TCF4 are associated with Pitt–Hopkins syndrome [160], genetic evidence suggests that increased levels of TCF4 gene expression might play a role in some cases of schizophrenia [161]. Pitt– Hopkins syndrome presents with severe intellectual disabilities, a complete lack of or minimal expressive language, hypotonia, and an ataxic gate. Seizures are reported in 40–50% with an older age of onset, and some patients have been classified at severe epileptic encephalopathy with autonomic dysfunction [162]. In this study, MRI revealed a thin corpus callosum, marked whitematter hyperintensity in the temporal poles, small hippocampi, and moderate hypoplasia of the frontal lobes. While individuals with Pitt–Hopkins syndrome generally have a “happy personality,” tendencies towards self-aggression and violent behavioral outbursts are also reported. They have stereotypic hand and head movements such as head rolling or rotation.

488

Hypoplasia of the corpus callosum, ventricular dilatation, cerebellar atrophy, and vermian hypoplasia are reported. TCF4 is a transcription factor that mediates neuronal activity-dependent changes in gene expression and is encode by a large gene spanning 437 kb with approximately 40 exons [160]. Pitt–Hopkinslike syndromes result from autosomal recessive deletions of NRXN1 and CNTNAP2 based on the shared clinical features. Interestingly, TCF4 promotes expression of parts of the CNTNAP2 and NRXN1β genes in vitro [163]. Christianson syndrome (SLC9A6) Christianson syndrome, an X-linked Angelman syndrome-like disorder, is caused by loss of function mutations in SLC9A6 on chromosome Xq26.3 [57,164,165]. Individuals have severe intellectual disability, with lack of speech, seizures, hyperkinesis, and truncal ataxia, and acquire microcephaly postnatally. Like Angelman syndrome, they may have a happy disposition with easily-provoked laughter. EEG findings range from normal to changes consistent with Lennox–Gastaut syndrome. Distinguishing features are external ophthalmoplegia manifested as horizontal or vertical gaze palsies and developmental regression with loss of motor skills. There is progressive atrophy of the cerebellar vermis. Some adults have dystonia and tau depositions are reported in their neurons and glial cells. Carrier females have been reported to have learning disabilities and/or behavioral issues and some had Parkinsonian symptoms. Deletion of the sodium-proton exchanger, NHE6 (SLC9A6) in mice resulted in overacidification of the endosomal compartment attenuating TrkB signaling and an attenuated axonal and dendritic structure of neurons associated with reduced synaptic transmission [166]. Mowat–Wilson syndrome (ZEB2) Haploinsufficiency of ZEB2 (ZFHX1B), which encodes zincfinger and homeodomain-like sequence-containing transcription factor, on chromosome 2q22.3, resembles Angelman syndrome in that all individuals have moderate to severe intellectual disability with minimal speech but have much better receptive than expressive language skills. Like Angelman syndrome, they have a “happy affect” and smile easily. Microcephaly, often of gradual postnatal onset, has been observed in more than 80% of individuals, and seizures or EEG abnormalities in up to 89%. In one study of 22 individuals, seizures typically started as focalonset seizures progressing to atypical absence seizures. A mouse model with mutations in ZEB2 displayed multiple defects relevant to Mowat–Wilson syndrome, including craniofacial abnormalities, defective corpus callosum formation, and a decreased number of parvalbumin interneurons in the cortex [167]. Behaviorally, these mice displayed reduced motor activity, increased anxiety, impaired sociability and reduced Barnes maze learning. FOXG1 syndrome Heterozygous loss-of-function mutations in FOXG1 on chromosome 14q12 result in severe intellectual disability with absent

Autism Spectrum Disorders Chapter 40

or minimal expressive language, early-onset postnatal microcephaly, generalized hypotonia, seizures with a variable age of onset ranging from infancy to teenage years, sleep disturbances, bruxism, dyskinesia of the hands, and impaired social interaction with poor eye contact. They additionally have stereotypies and frank dyskinesias with mixed features of athetosis, chorea and dystonia. Brain MRI studies have revealed frontal lobe simplified gyral pattern and reduced white matter volume, a thinned rostral corpus callosum, atrophy of the anterior vermis of the cerebellum, and variable mild frontal pachygyria. Mice with heterozygous deletions of FOXG1 display reduced volumes of cerebral cortex, striatum and hippocampus, and a thinned cerebral cortex due to reduced thickness of the superficial layers (II/III), the latter explained by reduced proliferation in the subventricular zone and intermediate progenitor cell by Tbr2immunostaining [168]. Deletion of FOXG1 in mice reduces the number of cortical GABAergic neurons [169]. A 2015 study of induced pluripotent stem cell samples derived from genetically undefined macrocephalic case of ASD and differentiated into neuronal embryoid bodies found increased production of GABAergic neurons was due to increased FOXG1 [170]. MEF2C syndrome Haploinsufficiency of MEF2C on chromosome 5q14.3 results in severe global developmental delay and intellectual disability with absent speech, hypotonia, seizures, and (in those who are ambulatory) a wide-based gait, all of which are features also seen in Angelman syndrome [171]. Hyperkinetic movements, at times accompanied by dystonia or chorea, are present. Bruxism, stereotypic movements of the hands such as clapping and handwashing movements, and poor eye contact have been identified in almost 50% of the MEF2C syndrome. Fifty-four percent of these individuals had seizures; 33% with early myoclonus (i.e., multifocal spike and wave on EEG) and 21% with epileptic spasms (hypsarrhythmia). Embryonic homozygous (not heterozygous as in the human condition) deletion of Mef2c produce contextual fear memory deficits and increased spine density along with the number of functional glutamatergic synapse [172]. However, a more recent study deleting this gene specifically in forebrain postnatally produced hyperactivity, defective motor performance, and weak clasping, however no defects in social behavior, increased repetitive self-grooming, or contextual fear memory were observed [173]. CNTNAP2 syndrome Homozygous mutations of CNTNAP2 result in cortical dysplasia, focal epilepsy, and relative macrocephaly. Intractable focal seizures beginning in early childhood is followed by language regression, hyperactivity, impulsive and aggressive behavior, and mental restriction [174]. Mice with homozygous mutations in Cntnap2 exhibit epilepsy, neuronal migration defects including reduced GABAergic interneurons, and deficits in sociability and behavior [175]. Importantly, studies in this mouse model

provide some of the strongest experimental evidence that therapies targeted at bolstering the oxytocin system might improve sociability [176]. NRXN1 syndrome Heterozygous microdeletions within the NRXN1 gene are found in cohorts with ASD but are also seen variably with other disorders and cognitive deficits [39,177]. Compound heterozygous deletions in NRXN1 are associated with severe developmental delay and early onset epilepsy [178].

Treatment, future perspective, conclusions Unlike some central nervous system disorders (e.g., motor neuron disease and Parkinson’s disease), where the specific neuronal circuit defects underlying the behavioral problems are defined, the specific circuits underlying the core (social, communication, and repetitive) and comorbid (aggression, sleep disorder, epilepsy, anxiety) behavioral deficits in ASD remain undefined in most cases. It is anticipated that each behavioral problem could map to a distinct neuronal circuitry. This problem should be solved in the near future by targeting the genetic defects to specific neuronal cell types and brain regions using modern methods. Cre-LoxP technologies combined with cell type specific promoters delivered as transgenics and/or as stereotaxic adeno-associated viral vectors into mice with loxP engineered deletion (or activation with extra gene copies modeling microduplications) combined with measurements of these behavioral and neurologic problems in mice represent such modern methods. Progress in this direction has already begun for Rett syndrome [179–182]. While these conditions are classified as neurodevelopmental disorders presenting at a very early age, evidence suggests that some may be due to a continuing effect of the altered gene dosages and therefore could be reversible in adulthood [189]. For example, many of the symptoms associated with Rett syndrome due to MECP2 gene deficiency are reversed when MECP2 is re-expressed in adult mice [183,184]. Restoration of fully functional MECP2 over a four-week period eradicated tremors and normalized breathing, mobility, and gait in mice that had previously been fully symptomatic and, in some cases, only days away from death. Similarly, mouse models of the MECP2 duplication disorder also show reversibility in adulthood [185]. These findings have motivated clinical trials for these conditions testing the efficacy of insulin-like growth factor 1 based on their therapeutic efficacy in the Rett syndrome mouse model [186,187]. One study investigated the reversibility of Angelman syndrome due to a deficiency of the maternal Ube3a allele. Using antisense oligonucleotides aimed at depleting the long non-coding RNA that normally represses expression of the paternal allele in mature neurons, a lasting re-expression of the paternal Ube3a gene and partial amelioration of some cognitive deficits of the disorder in mice was achieved [188]. The paternal allele Ube3a allele has also been re-expressed in neurons using topoisomerase

489

Developmental Neuropathology inhibitors, which suggests that a pharmacological approach may also be feasible [189], although such drugs have known DNA damaging adverse effects.

References

17.

18. 19.

1. Trauner DA (2015) Behavioral correlates of epileptiform abnormalities in autism. Epilepsy Behav 47:163–6 2. Wegiel J, Schanen NC, Cook EH et al. (2012). Differences between the pattern of developmental abnormalities in autism associated with duplications 15q11.2-q13 and idiopathic autism. J Neuropathol Exp Neurol 71:382–97 3. Kamiya K, Kaneda M, Sugawara T et al. (2004) A nonsense mutation of the sodium channel gene SCN2A in a patient with intractable epilepsy and mental decline. J Neurosci 24:2690–8 4. Woolfenden S, Sarkozy V, Ridley G et al. (2012) A systematic review of two outcomes in autism spectrum disorder: epilepsy and mortality. Dev Med Child Neurol 54:306–12 5. Bilder D, Botts EL, Smith KR et al. (2013) Excess mortality and causes of death in autism spectrum disorders: a follow up of the 1980s Utah/UCLA autism epidemiologic study. J Autism Dev Disord 43:1196–204 6. Scheffer IE (2012) Diagnosis and long-term course of Dravet syndrome. Eur J Paediatr Neurol 16(Suppl 1):S5–8 7. Shoffner J, Hyams L, Langley GN et al. (2010) Fever plus mitochondrial disease could be risk factors for autistic regression. J Child Neurol 25:429–34 8. Brown AS, Sourander A, Hinkka-Yli-Salom¨aki S et al. (2014) Elevated maternal C-reactive protein and autism in a national birth cohort. Mol Psychiatry 19:259–64 9. Atlad´ottir HO, Pedersen MG, Thorsen P et al. (2009) Association of family history of autoimmune diseases and autism spectrum disorders. Pediatrics 124:687–94 10. Goines P, Haapanen L, Boyce R et al. (2011) Autoantibodies to cerebellum in children with autism associate with behavior. Brain Behav Immun 25:514–23 11. Piras IS, Haapanen L, Napolioni V et al. (2014) Anti-brain antibodies are associated with more severe cognitive and behavioral profiles in Italian children with autism spectrum disorder. Brain Behav Immun 38:91–9 12. Nordahl CW, Braunschweig D, Iosif AM et al. (2013) Maternal autoantibodies are associated with abnormal brain enlargement in a subgroup of children with autism spectrum disorder. Brain Behav Immun 30:61–5 13. Bauman MD, Iosif AM, Ashwood P et al. (2013) Maternal antibodies from mothers of children with autism alter brain growth and social behavior development in the rhesus monkey. Transl Psychiatry 3:e278 14. Yamashita Y, Fujimoto C, Nakajima E et al. (2003) Possible association between congenital cytomegalovirus infection and autistic disorder. J Autism Dev Disord 33:455–9 15. Marques F, Brito MJ, Conde M et al. (2014) Autism spectrum disorder secondary to enterovirus encephalitis. J Child Neurol 29:708–14 16. Fatemi SH, Earle J, Kanodia R et al. (2002) Prenatal viral infection leads to pyramidal cell atrophy and macrocephaly in

490

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

adulthood: implications for genesis of autism and schizophrenia. Cell Mol Neurobiol 22:25–33 Smith SE, Elliott RM, Anderson MP (2012) Maternal immune activation increases neonatal mouse cortex thickness and cell density. J Neuroimmune Pharmacol 7:529–32 Amaral DG, Schumann CM, Nordahl CW (2008) Neuroanatomy of autism. Trends Neurosci 31:137–45 Wegiel J, Flory M, Kuchna I et al. (2014) Stereological study of the neuronal number and volume of 38 brain subdivisions of subjects diagnosed with autism reveals significant alterations restricted to the striatum, amygdala and cerebellum. Acta Neuropathol Commun 2:141 Courchesne E, Mouton PR, Calhoun ME et al. (2011) Neuron number and size in prefrontal cortex of children with autism. JAMA 306:2001–10 Smith SE, Li J, Garbett K et al. (2007) Maternal immune activation alters fetal brain development through interleukin-6. J Neurosci 27:10695–702 Malkova NV, Yu CZ, Hsiao EY et al. (2012) Maternal immune activation yields offspring displaying mouse versions of the three core symptoms of autism. Brain Behav Immun 26:607–16 Garay PA, Hsiao EY, Patterson PH, McAllister AK (2013) Maternal immune activation causes age- and region-specific changes in brain cytokines in offspring throughout development. Brain Behav Immun 31:54–68 Vargas DL, Nascimbene C, Krishnan C et al. (2005) Neuroglial activation and neuroinflammation in the brain of patients with autism. Ann Neurol 57:67–81 Bauman MD, Iosif AM, Smith SE et al. (2014) Activation of the maternal immune system during pregnancy alters behavioral development of rhesus monkey offspring. Biol Psychiatry 75:332– 41 Nankova BB, Agarwal R, MacFabe DF, La Gamma EF (2014) Enteric bacterial metabolites propionic and butyric acid modulate gene expression, including CREB-dependent catecholaminergic neurotransmission, in PC12 cells: possible relevance to autism spectrum disorders. PLoS One 9:e103740 Wang L, Christophersen CT, Sorich MJ et al. (2012) Elevated fecal short chain fatty acid and ammonia concentrations in children with autism spectrum disorder. Dig Dis Sci 57:2096–102 Kimura I, Ozawa K, Inoue D et al. (2013) The gut microbiota suppresses insulin-mediated fat accumulation via the short-chain fatty acid receptor GPR43. Nat Commun 4:1829 Hsiao EY, McBride SW, Hsien S et al. (2013) Microbiota modulate behavioral and physiological abnormalities associated with neurodevelopmental disorders. Cell 155:1451–63 D’Gama AM, Pochareddy S, Li M et al. (2015) Targeted DNA sequencing from autism spectrum disorder brains implicates multiple genetic mechanisms. Neuron 88:910–17 Lin D, Boyle MP, Dollar P et al. (2011) Functional identification of an aggression locus in the mouse hypothalamus. Nature 470: 221–6 Lee H, Kim DW, Remedios R et al. (2014) Scalable control of mounting and attack by Esr1+ neurons in the ventromedial hypothalamus. Nature 509:627–32 Rosenberg RE, Law JK, Yenokyan G et al. (2009) Characteristics and concordance of autism spectrum disorders among 277 twin pairs. Arch Pediatr Adolesc Med 163:907–14

Autism Spectrum Disorders Chapter 40 34. Gillberg C, Steffenburg S, Wahlstr¨om J et al. (1991) Autism associated with marker chromosome. J Am Acad Child Adolesc Psychiatry 30:489–94 35. Cook EH Jr, Lindgren V, Leventhal BL et al. (1997) Autism or atypical autism in maternally but not paternally derived proximal 15q duplication. Am J Hum Genet 60:928–34 36. Matsuura T, Sutcliffe JS, Fang P et al. (1997) De novo truncating mutations in E6-AP ubiquitin-protein ligase gene (UBE3A) in Angelman syndrome. Nat Genet 15:74–7 37. Kishino T, Lalande M, Wagstaff J (1997) UBE3A/E6-AP mutations cause Angelman syndrome. Nat Genet 15:70–3 38. Albrecht U, Sutcliffe JS, Cattanach BM et al. (1997) Imprinted expression of the murine Angelman syndrome gene, Ube3a, in hippocampal and Purkinje neurons. Nat Genet 17:75–8 39. Glessner JT, Wang K, Cai G et al. (2009) Autism genome-wide copy number variation reveals ubiquitin and neuronal genes. Nature 459:569–73 40. Hogart A, Wu D, LaSalle JM, Schanen NC. The comorbidity of autism with the genomic disorders of chromosome 15q11.2-q13. Neurobiol Dis 2010;38(2):181–91 41. Smith SE, Zhou YD, Zhang G et al. (2011) Increased gene dosage of Ube3a results in autism traits and decreased glutamate synaptic transmission in mice. Sci Transl Med 3:103ra97 42. Dennis NR, Veltman MW, Thompson R et al. (2006) Clinical findings in 33 subjects with large supernumerary marker(15) chromosomes and 3 subjects with triplication of 15q11-q13. Am J Med Genet A 140:434–41 43. Dup15q Alliance. Available at: www.dup15q.org (accessed November 9, 2017) 44. Blatt GJ (2012) The neuropathology of autism. Scientifica (Cairo) 2012:703675 45. Bauman ML, Kemper TL (2005) Neuroanatomic observations of the brain in autism: a review and future directions. Int J Dev Neurosci 23:183–7 46. Palmen SJ, van Engeland H, Hof PR, Schmitz C (2004) Neuropathological findings in autism. Brain 127:2572–83 47. Klein S, Sharifi-Hannauer P, Martinez-Agosto JA (2013) Macrocephaly as a clinical indicator of genetic subtypes in autism. Autism Res 6:51–6 48. O’Roak BJ, Vives L, Fu W et al. (2012) Multiplex targeted sequencing identifies recurrently mutated genes in autism spectrum disorders. Science 338:1619–22 49. Neale BM, Kou Y, Liu L et al. (2012) Patterns and rates of exonic de novo mutations in autism spectrum disorders. Nature 485:242–5 50. Talkowski ME, Rosenfeld JA, Blumenthal I et al. (2012) Sequencing chromosomal abnormalities reveals neurodevelopmental loci that confer risk across diagnostic boundaries. Cell 149:525–37 51. Bernier R, Golzio C, Xiong B et al. (2014) Disruptive CHD8 mutations define a subtype of autism early in development. Cell 158:263–76 52. Williams RS, Hauser SL, Purpura DP et al. (1980) Autism and mental retardation: neuropathologic studies performed in four retarded persons with autistic behavior. Arch Neurol 37:749–53 53. Courchesne E, M¨uller RA, Saitoh O (1999) Brain weight in autism: normal in the majority of cases, megalencephalic in rare cases. Neurology 52:1057–9 54. Bailey A, Luthert P, Dean A et al. (1998) A clinicopathological study of autism. Brain 121:889–905

55. Kemper TL, Bauman M (1998) Neuropathology of infantile autism. J Neuropathol Exp Neurol 57:645–52 56. Casanova MF, Buxhoeveden DP, Switala AE, Roy E (2002) Neuronal density and architecture (Gray Level Index) in the brains of autistic patients. J Child Neurol 17:515–21 57. Pescosolido MF, Stein DM, Schmidt M et al. (2014) Genetic and phenotypic diversity of NHE6 mutations in Christianson syndrome. Ann Neurol 76:581–93 58. Jellinger K, Armstrong D, Zoghbi HY, Percy AK (1988) Neuropathology of Rett syndrome. Acta Neuropathol 76: 142–58 59. Amir RE, Van den Veyver IB, Wan M et al. (1999) Rett syndrome is caused by mutations in X-linked MECP2, encoding methyl-CpGbinding protein 2. Nat Genet 23:185–8 60. Carranza Rojo D, Hamiwka L, McMahon JM et al. (2011) De novo SCN1A mutations in migrating partial seizures of infancy. Neurology 77:380–3 61. Hennies HC, Rauch A, Seifert W et al. (2004) Allelic heterogeneity in the COH1 gene explains clinical variability in Cohen syndrome. Am J Hum Genet 75:138–45 62. van Bon BW, Coe BP, Bernier R et al. (2016) Disruptive de novo mutations of DYRK1A lead to a syndromic form of autism and ID. Mol Psychiatry 21:126–32 63. Hoischen A, Krumm N, Eichler EE (2014) Prioritization of neurodevelopmental disease genes by discovery of new mutations. Nat Neurosci 17:764–72 64. Willemsen MH, Vulto-van Silfhout AT, Nillesen WM et al. (2012) Update on Kleefstra Syndrome. Mol Syndromol 2:202–12 65. Ritvo ER, Freeman BJ, Scheibel AB et al. (1986) Lower Purkinje cell counts in the cerebella of four autistic subjects: initial findings of the UCLA-NSAC Autopsy Research Report. Am J Psychiatry 143:862–6 66. Gaffney GR, Tsai LY (1987) Magnetic resonance imaging of high level autism. J Autism Dev Disord 17:433–8 67. Piven J, Berthier ML, Starkstein SE et al. (1990) Magnetic resonance imaging evidence for a defect of cerebral cortical development in autism. Am J Psychiatry 147:734–9 68. Schifter T, Hoffman JM, Hatten HP Jr et al. (1994) Neuroimaging in infantile autism. J Child Neurol 9:155–61 69. Poduri A, Evrony GD, Cai X et al. (2012) Somatic activation of AKT3 causes hemispheric developmental brain malformations. Neuron 74:41–8 70. Rivi`ere JB, Mirzaa GM, O’Roak BJ et al. (2012) De novo germline and postzygotic mutations in AKT3, PIK3R2 and PIK3CA cause a spectrum of related megalencephaly syndromes. Nat Genet 44:934–40 71. Lee JH, Huynh M, Silhavy JL et al. (2012) De novo somatic mutations in components of the PI3K-AKT3-mTOR pathway cause hemimegalencephaly. Nat Genet 44:941–5 72. Casanova MF, Buxhoeveden DP, Switala AE, Roy E (2002) Asperger’s syndrome and cortical neuropathology. J Child Neurol 17:142–5 73. Stoner R, Chow ML, Boyle MP et al. (2014) Patches of disorganization in the neocortex of children with autism. N Engl J Med 370:1209–19 74. Dachet F, Bagla S, Keren-Aviram G et al. (2015) Predicting novel histopathological microlesions in human epileptic brain through transcriptional clustering. Brain 138:356–70

491

Developmental Neuropathology 75. Hutsler JJ, Zhang H (2010) Increased dendritic spine densities on cortical projection neurons in autism spectrum disorders. Brain Res 1309:83–94 76. Lee M, Martin-Ruiz C, Graham A et al. (2002) Nicotinic receptor abnormalities in the cerebellar cortex in autism. Brain 125:1483– 95 77. Fatemi SH, Halt AR, Realmuto G et al. (2002) Purkinje cell size is reduced in cerebellum of patients with autism. Cell Mol Neurobiol 22:171–5 78. Guptill JT, Booker AB, Gibbs TT et al. (2007) [3H]flunitrazepam-labeled benzodiazepine binding sites in the hippocampal formation in autism: a multiple concentration autoradiographic study. J Autism Dev Disord 37:911–20 79. Oblak A, Gibbs TT, Blatt GJ (2009) Decreased GABAA receptors and benzodiazepine binding sites in the anterior cingulate cortex in autism. Autism Res 2:205–19 80. Oblak AL, Gibbs TT, Blatt GJ (2010) Decreased GABA(B) receptors in the cingulate cortex and fusiform gyrus in autism. J Neurochem 114:1414–23 81. Oblak AL, Gibbs TT, Blatt GJ (2011) Reduced GABAA receptors and benzodiazepine binding sites in the posterior cingulate cortex and fusiform gyrus in autism. Brain Res 1380:218–28 82. Fatemi SH, Halt AR, Stary JM et al. (2002) Glutamic acid decarboxylase 65 and 67 kDa proteins are reduced in autistic parietal and cerebellar cortices. Biol Psychiatry 52:805–10 83. Yip J, Soghomonian JJ, Blatt GJ (2007) Decreased GAD67 mRNA levels in cerebellar Purkinje cells in autism: pathophysiological implications. Acta Neuropathol 113:559–68 84. Yip J, Soghomonian JJ, Blatt GJ (2008) Increased GAD67 mRNA levels in cerebellar interneurons in autism: implications for Purkinje cell dysfunction. J Neurosci Res 86:525–30 85. Azmitia EC, Singh JS, Whitaker-Azmitia PM (2011) Increased serotonin axons (immunoreactive to 5-HT transporter) in postmortem brains from young autism donors. Neuropharmacology 60:1347–54 86. Oblak A, Gibbs TT, Blatt GJ (2013) Reduced serotonin receptor subtypes in a limbic and a neocortical region in autism. Autism Res 6:571–83 87. Boronat S, Mehan WA, Shaaya EA et al. (2015) Hippocampal Abnormalities in Magnetic Resonance Imaging (MRI) of 15q Duplication Syndromes. J Child Neurol 30:333–8 88. Mantegazza M, Gambardella A, Rusconi R et al. (2005) Identification of an Nav1.1 sodium channel (SCN1A) loss-of-function mutation associated with familial simple febrile seizures. Proc Natl Acad Sci U S A 102:18177–82 89. Tien RD, Ashdown BC (1992) Crossed cerebellar diaschisis and crossed cerebellar atrophy: correlation of MR findings, clinical symptoms, and supratentorial diseases in 26 patients. AJR Am J Roentgenol 158:1155–9 90. Samaniego EA, Stuckert E, Fischbein N, Wijman CA (2012) Crossed cerebellar diaschisis in status epilepticus. Neurocrit Care 12:88–90 91. Garbett K, Ebert PJ, Mitchell A et al. (2008) Immune transcriptome alterations in the temporal cortex of subjects with autism. Neurobiol Dis 30:303–11 92. Voineagu I, Wang X, Johnston P et al. (2011) Transcriptomic analysis of autistic brain reveals convergent molecular pathology. Nature 474:380–4

492

93. Maroso M, Balosso S, Ravizza T et al. (2010) Toll-like receptor 4 and high-mobility group box-1 are involved in ictogenesis and can be targeted to reduce seizures. Nat Med 16:413–19 94. Pernhorst K, Herms S, Hoffmann P et al. (2013) TLR4, ATF3 and IL8 inflammation mediator expression correlates with seizure frequency in human epileptic brain tissue. Seizure 22: 675–8 95. Li X, Chauhan A, Sheikh AM et al. (2009) Elevated immune response in the brain of autistic patients. J Neuroimmunol 207:111–16 96. Edmonson C, Ziats MN, Rennert OM (2014) Altered glial marker expression in autistic post-mortem prefrontal cortex and cerebellum. Mol Autism 5:3 97. Paolicelli RC, Bolasco G, Pagani F et al. (2011) Synaptic pruning by microglia is necessary for normal brain development. Science 333:1456–8 98. Zhan Y, Paolicelli RC, Sforazzini F et al. (2014) Deficient neuronmicroglia signaling results in impaired functional brain connectivity and social behavior. Nat Neurosci 17:400–6 99. Laurence JA, Fatemi SH (2005) Glial fibrillary acidic protein is elevated in superior frontal, parietal and cerebellar cortices of autistic subjects. Cerebellum 4:206–10 100. Cao F, Yin A, Wen G et al. (2012) Alteration of astrocytes and Wnt/β-catenin signaling in the frontal cortex of autistic subjects. J Neuroinflammation 9:223 101. Fatemi SH, Folsom TD, Reutiman TJ, Lee S (2008) Expression of astrocytic markers aquaporin 4 and connexin 43 is altered in brains of subjects with autism. Synapse 62:501–7 102. Benarroch EE (2007) Aquaporin-4, homeostasis, and neurologic disease. Neurology 69:2266–8 103. Garbern JY, Neumann M, Trojanowski JQ et al. (2010) A mutation affecting the sodium/proton exchanger, SLC9A6, causes mental retardation with tau deposition. Brain 133:1391–402 104. Helsmoortel C, Vulto-van Silfhout AT, Coe BP et al. (2014) A SWI/SNF-related autism syndrome caused by de novo mutations in ADNP. Nat Genet 46:380–4 105. Barba C, Parrini E, Coras R et al. (2014) Co-occurring malformations of cortical development and SCN1A gene mutations. Epilepsia 55:1009–19 106. Li J, Cai T, Jiang Y et al. (2015) Genes with de novo mutations are shared by four neuropsychiatric disorders discovered from NPdenovo database. Mol Psychiatry 21(2):298 107. O’Roak BJ, Vives L, Girirajan S et al. (2012) Sporadic autism exomes reveal a highly interconnected protein network of de novo mutations. Nature 485:246–50 108. Yu FH, Mantegazza M, Westenbroek RE et al. (2006) Reduced sodium current in GABAergic interneurons in a mouse model of severe myoclonic epilepsy in infancy. Nat Neurosci 9: 1142–9 109. Tai C, Abe Y, Westenbroek RE et al. (2014) Impaired excitability of somatostatin- and parvalbumin-expressing cortical interneurons in a mouse model of Dravet syndrome. Proc Natl Acad Sci U S A 111:E3139–48 110. Han S, Tai C, Westenbroek RE et al. (2012) Autistic-like behaviour in Scn1a± mice and rescue by enhanced GABA-mediated neurotransmission. Nature 489:385–90 111. Li T, Tian C, Scalmani P et al. (2014) Action potential initiation in neocortical inhibitory interneurons. PLoS Biol 12:e1001944

Autism Spectrum Disorders Chapter 40 112. Toonen RF, Wierda K, Sons MS et al. (2006) Munc18-1 expression levels control synapse recovery by regulating readily releasable pool size. Proc Natl Acad Sci U S A 103:18332–7 113. Endele S, Rosenberger G, Geider K et al. (2010) Mutations in GRIN2A and GRIN2B encoding regulatory subunits of NMDA receptors cause variable neurodevelopmental phenotypes. Nature Genet 42: 1021–6 114. Epi4K Consortium and Epilepsy Phenome/Genome Project (2013) De novo mutations in epileptic encephalopathies. Nature 501: 217–21 115. Lemke J R, Hendrickx R, Geider K et al. (2014) GRIN2B mutations in West syndrome and intellectual disability with focal epilepsy. Ann Neurol 75:147–54 116. Lemke JR, Lal D, Reinthaler EM et al. (2013) Mutations in GRIN2A cause idiopathic focal epilepsy with rolandic spikes. Nature Genet 45:1067–72 117. Kelsch W, Li Z, Wieland S et al. (2014) GluN2B-containing NMDA receptors promote glutamate synapse development in hippocampal interneurons. J Neurosci 34:16022–30 118. Carville GL, Regan BM, Yendle SC et al. (2013) GRIN2A mutations cause epilepsy-aphasia spectrum disorders. Nature Genet 45:1073–6 119. Lesca G, Rudolf G, Bruneau N et al. (2013) GRIN2A mutations in acquired epileptic aphasia and related childhood focal epilepsies and encephalopathies with speech and language dysfunction. Nat Genet 45:1061–6 120. Archer HL, Evans J, Edwards S et al. (2006) CDKL5 mutations cause infantile spasms, early onset seizures, and severe mental retardation in female patients. J Med Genet 43:729–34 121. Moseley BD, Dhamija R, Wirrell EC, Nickels KC (2012) Historic, clinical, and prognostic features of epileptic encephalopathies caused by CDKL5 mutations. Pediatr Neurol 46:101–5 122. Guerrini R, Parrini E (2012) Epilepsy in Rett syndrome, and CDKL5- and FOXG1-gene-related encephalopathies. Epilepsia 53:2067–78 123. Wang IT, Allen M, Goffin D et al. (2012) Loss of CDKL5 disrupts kinome profile and event-related potentials leading to autistic-like phenotypes in mice. Proc Natl Acad Sci U S A 109:21516–21 124. Amendola E, Zhan Y, Mattucci C et al. (2014) Mapping pathological phenotypes in a mouse model of CDKL5 disorder. PLoS One 9:e91613 125. Fuchs C, Trazzi S, Torricella R et al. (2014) Loss of CDKL5 impairs survival and dendritic growth of newborn neurons by altering AKT/GSK-3β signaling. Neurobiol Dis 70:53–68 126. Fuchs C, Rimondini R, Viggiano R et al. (2015) Inhibition of GSK3β rescues hippocampal development and learning in a mouse model of CDKL5 disorder. Neurobiol Dis 82:298–310 127. Krumm N, O’Roak BJ, Shendure J, Eichler EE (2014) A de novo convergence of autism genetics and molecular neuroscience. Trends Neurosci 37:95–105 128. van Bon BW, Hoischen A, Hehir-Kwa J et al. (2011) Intragenic deletion in DYRK1A leads to mental retardation and primary microcephaly. Clin Genet 79:296–9 129. Courcet JB, Faivre L, Malzac P et al. (2012) The DYRK1A gene is a cause of syndromic intellectual disability with severe microcephaly and epilepsy. J Med Genet 49:731–6 130. Garc´ıa-Cerro S, Mart´ınez P, Vidal V et al. (2014) Overexpression of Dyrk1A is implicated in several cognitive, electrophysiological

131.

132.

133.

134. 135.

136.

137. 138. 139.

140.

141.

142.

143.

144.

145.

146.

147.

and neuromorphological alterations found in a mouse model of Down syndrome. PLoS One 9:e106572 Chen CK, Bregere C, Paluch J et al. (2014) Activity-dependent facilitation of Synaptojanin and synaptic vesicle recycling by the Minibrain kinase. Nat Commun 5:4246 Lepagnol-Bestel AM, Zvara A, Maussion G et al. (2009) DYRK1A interacts with the REST/NRSF-SWI/SNF chromatin remodelling complex to deregulate gene clusters involved in the neuronal phenotypic traits of Down syndrome. Hum Mol Genet 18: 1405–14 Tan WH, Bird LM, Thibert RI, Williams CA (2014) If not Angelman, What is it? A review of Angelman-like syndromes. Am J Med Genet A 164a:975–92 Bird LM (2014) Angelman syndrome: review of clinical and molecular aspects. Appl Clin Genet 7:93–104 Orrico A, Zollino M, Galli L et al. (2009) Late-onset Lennox– Gastaut syndrome in a patient with 15q11.2-q13.1 duplication. Am J Med Genet A 149A:1033–5 Stewart LR, Hall AL, Kang SH et al. (2011) High frequency of known copy number abnormalities and maternal duplication 15q11-q13 in patients with combined schizophrenia and epilepsy. BMC Med Genet 12:154 Glaze DG, Percy AK, Skinner S et al. (2010) Epilepsy and the natural history of Rett syndrome. Neurology 74:909–12 Jiang YH, Ehlers MD (2013) Modeling autism by SHANK gene mutations in mice. Neuron 78:8–27 Berg JS, Brunetti-Pierri N, Peters SU et al. (2007) Speech delay and autism spectrum behaviors are frequently associated with duplication of the 7q11.23 Williams-Beuren syndrome region. Genet Med 9:427–41 Sanders SJ, Ercan-Sencicek AG, Hus V et al. (2011) Multiple recurrent de novo CNVs, including duplications of the 7q11.23 Williams syndrome region, are strongly associated with autism. Neuron 70:863–85 Li HH, Roy M, Kuscuoglu U et al. (2009) Induced chromosome deletions cause hypersociability and other features of WilliamsBeuren syndrome in mice. EMBO 1:50–65 Sakurai T, Dorr NP, Takahashi N et al. (2011) Haploinsufficiency of Gtf2i, a gene deleted in Williams Syndrome, leads to increases in social interactions. Autism Res 4:28–39 Schneider T, Skitt Z, Liu Y et al. (2012) Anxious, hypoactive phenotype combined with motor deficits in Gtf2ird1 null mouse model relevant to Williams syndrome. Behav Brain Res 233:458– 73 Borralleras C, Sahun I, P´erez-Jurado LA, Campuzano V (2015) Intracisternal Gtf2i Gene Therapy Ameliorates Deficits in Cognition and Synaptic Plasticity of a Mouse Model of Williams-Beuren Syndrome. Mol Ther 23:1691–9 Young EJ, Lipina T, Tam E et al. (2008) Reduced fear and aggression and altered serotonin metabolism in Gtf2ird1-targeted mice. Genes Brain Behav 7:224–34 Howard ML, Palmer SJ, Taylor KM et al. (2012) Mutation of Gtf2ird1 from the Williams-Beuren syndrome critical region results in facial dysplasia, motor dysfunction, and altered vocalisations. Neurobiol Dis 45:913–22 Mervis CB, Dida J, Lam E et al. (2012) Duplication of GTF2I results in separation anxiety in mice and humans. Am J Hum Genet 90:1064–70

493

Developmental Neuropathology 148. Talkowski ME, Mullegama SV, Rosenfeld JA et al. (2011) Assessment of 2q23.1 microdeletion syndrome implicates MBD5 as a single causal locus of intellectual disability, epilepsy, and autism spectrum disorder. Am J Hum Genet 89:551–63 149. Mullegama SV, Rosenfeld JA, Orellana C et al. (2014) Reciprocal deletion and duplication at 2q23.1 indicates a role for MBD5 in autism spectrum disorder. Eur J Hum Genet 22:57–63 150. Zollino M, Orteschi D, Murdolo M et al. (2012) Mutations in KANSL1 cause the 17q21.31 microdeletion syndrome phenotype. Nature Genet 44:636–8 151. Grisart B, Willatt L, Destree A et al. (2009) 17q21.31 microduplication patients are characterised by behavioural problems and poor social interaction. J Med Genet 46:524–30 152. Balemans MC, Huibers MM, Eikelenboom NW et al. (2010) Reduced exploration, increased anxiety, and altered social behavior: autistic-like features of euchromatin histone methyltransferase 1 heterozygous knockout mice. Behav Brain Res 208: 47–55 153. Allderdice PW, Eales B, Onyett H et al. (1983) Duplication 9q34 syndrome. Am J Hum Genet 35:1005–19 154. Guerrini R, Shanahan JL, Carrozzo R et al. (2000) A nonsense mutation of the ATRX gene causing mild mental retardation and epilepsy. Ann Neurol 47:117–21 155. B´erub´e NG, Mangelsdorf M, Jagla M et al. (2005) The chromatinremodeling protein ATRX is critical for neuronal survival during corticogenesis. J Clin Invest 115:258–67 156. Kernohan KD, Jiang Y, Tremblay DC et al. (2010) ATRX partners with cohesin and MeCP2 and contributes to developmental silencing of imprinted genes in the brain. Dev Cell 18:191– 202 157. Huh MS, Price O’Dea T, Ouazia D et al. (2012) Compromised genomic integrity impedes muscle growth after Atrx inactivation. J Clin Invest 122:4412–23 158. Honda S, Satomura S, Hayashi S et al. (2012) Concomitant microduplications of MECP2 and ATRX in male patients with severe mental retardation. J Hum Genet 57:73–7 159. Mart´ınez F, Rosell´o M, Mayo S et al. (2014) Duplication at Xq13.3q21.1 with syndromic intellectual disability, a probable role for the ATRX gene. Am J Med Genet A 164A:918–23 160. Sepp M, Pruunsild P, Timmusk T (2012) Pitt-Hopkins syndromeassociated mutations in TCF4 lead to variable impairment of the transcription factor function ranging from hypomorphic to dominant-negative effects. Hum Mol Genet 21:2873–88 161. Quednow BB, Brz´ozka MM, Rossner MJ (2014) Transcription factor 4 (TCF4) and schizophrenia: integrating the animal and the human perspective. Cell Mol Life Sci 71:2815–35 162. Amiel J, Rio M, de Pontual L et al. (2007) Mutations in TCF4, encoding a class I basic helix-loop-helix transcription factor, are responsible for Pitt-Hopkins syndrome, a severe epileptic encephalopathy associated with autonomic dysfunction. Am J Hum Genet 80:988–93 163. Forrest M, Chapman RM, Doyle AM et al. (2012) Functional analysis of TCF4 missense mutations that cause Pitt-Hopkins syndrome. Hum Mutat 33:1676–86 164. Gilfillan GD, Selmer KK, Roxrud I et al. (2008) SLC9A6 mutations cause X-linked mental retardation, microcephaly, epilepsy, and ataxia, a phenotype mimicking Angelman syndrome. Am J Hum Genet 82:1003–10

494

165. Morrow EM, Yoo SY, Flavell SW et al. (2008) Identifying autism loci and genes by tracing recent shared ancestry. Science 321:218– 23 166. Ouyang Q, Lizarraga SB, Schmidt M et al. (2013) Christianson syndrome protein NHE6 modulates TrkB endosomal signaling required for neuronal circuit development. Neuron 80:97–112 167. Takagi T, Nishizaki Y, Matsui F et al. (2015) De novo inbred heterozygous Zeb2/Sip1 mutant mice uniquely generated by germline conditional knockout exhibit craniofacial, callosal and behavioral defects associated with Mowat-Wilson syndrome. Hum Mol Genet 24:6390–402 168. Siegenthaler JA, Tremper-Wells BA, Miller MW (2008) Foxg1 haploinsufficiency reduces the population of cortical intermediate progenitor cells: effect of increased p21 expression. Cereb Cortex 18:1865–75 169. Yang Y, Shen W, Ni Y, Su Y, Yang Z, Zhao C (2017) Impaired interneuron development after Foxg1 disruption. Cereb Cortex 27(1):793–808 170. Mariani J, Coppola G, Zhang P et al. (2015) FOXG1-dependent dysregulation of GABA/glutamate neuron differentiation in autism spectrum disorders. Cell 162:375–90 171. Zweier M, Rauch A (2012) The MEF2C-Related and 5q14.3q15 Microdeletion Syndrome. Mol Syndromol 2:164–70 172. Barbosa AC, Kim MS, Ertunc M et al. (2008) MEF2C, a transcription factor that facilitates learning and memory by negative regulation of synapse numbers and function. Proc Natl Acad Sci U S A 105:9391–6 173. Adachi M, Lin PY, Pranav H, Monteggia LM (2016) Postnatal loss of Mef2c results in dissociation of effects on synapse number and learning and memory. Biol Psychiatry 80(2):140–8 174. Strauss KA, Puffenberger EG, Huentelman MJ et al. (2006) Recessive symptomatic focal epilepsy and mutant contactin-associated protein-like 2. N Engl J Med 354:1370–7 175. Pe˜nagarikano O, Abrahams BS, Herman EI et al. (2011) Absence of CNTNAP2 leads to epilepsy, neuronal migration abnormalities, and core autism-related deficits. Cell 147:235–46 176. Pe˜nagarikano O, L´azaro MT, Lu XH et al. (2015) Exogenous and evoked oxytocin restores social behavior in the Cntnap2 mouse model of autism. Sci Transl Med 7:271ra8 177. Curran S, Ahn JW, Grayton H et al. (2013) NRXN1 deletions identified by array comparative genome hybridisation in a clinical case series: further understanding of the relevance of NRXN1 to neurodevelopmental disorders. J Mol Psychiatry 1:4 178. Harrison V, Connell L, Hayesmoore J et al. (2011) Compound heterozygous deletion of NRXN1 causing severe developmental delay with early onset epilepsy in two sisters. Am J Med Genet A 155A:2826–31 179. Fyffe SL, Neul JL, Samaco RC et al. (2008) Deletion of Mecp2 in Sim1-expressing neurons reveals a critical role for MeCP2 in feeding behavior, aggression, and the response to stress. Neuron 59:947–58 180. Samaco RC, Mandel-Brehm C, Chao HT et al. (2009) Loss of MeCP2 in aminergic neurons causes cell-autonomous defects in neurotransmitter synthesis and specific behavioral abnormalities. Proc Natl Acad Sci U S A 106:21966–71 181. Chao HT, Chen H, Samaco RC et al. (2010) Dysfunction in GABA signalling mediates autism-like stereotypies and Rett syndrome phenotypes. Nature 468:263–9

Autism Spectrum Disorders Chapter 40 182. Ito-Ishida A, Ure K, Chen H et al. (2015) Loss of MeCP2 in Parvalbumin-and Somatostatin-Expressing Neurons in Mice Leads to Distinct Rett Syndrome-like Phenotypes. Neuron 88:651–8 183. Guy J, Gan J, Selfridge J et al. (2007) Reversal of neurological defects in a mouse model of Rett syndrome. Science 315: 1143–7 184. Robinson L, Guy J, McKay L et al. (2012) Morphological and functional reversal of phenotypes in a mouse model of Rett syndrome. Brain 135:2699–710 185. Sztainberg Y, Chen HM, Swann JW et al. (2015) Reversal of phenotypes in MECP2 duplication mice using genetic rescue or antisense oligonucleotides. Nature 528:123–6 186. Khwaja OS, Ho E, Barnes KV et al. (2014) Safety, pharmacokinetics, and preliminary assessment of efficacy of mecasermin (recombinant human IGF-1) for the treatment of Rett syndrome. Proc Natl Acad Sci U S A 111:4596–601

187. Tropea D, Giacometti E, Wilson NR et al. (2009) Partial reversal of Rett Syndrome-like symptoms in MeCP2 mutant mice. Proc Natl Acad Sci U S A 106:2029–34 188. Meng L, Ward AJ, Chun S et al. (2015) Towards a therapy for Angelman syndrome by targeting a long non-coding RNA. Nature 518:409–12 189. Huang HS, Allen JA, Mabb AM et al. (2011) Topoisomerase inhibitors unsilence the dormant allele of Ube3a in neurons. Nature 481:185–9 190. Ronan JL, Wu W, Crabtree GR (2013) From neural development to cognition: unexpected roles for chromatin. Nat Rev Genet 14(5):347–59

Online resources http://sfari.org www.omim.org http://www.autismbrainnet.com

495

41

Intrauterine Infections Catherine Keohane1 and Homa Adle-Biassette2 1 Department 2 Department

of Pathology and School of Medicine, University College Cork, Cork, Ireland of Pathology, APHP, Lariboisi`ere Hospital, Universit´e Paris Diderot, Paris, France

Definition and synonyms Human cytomegalovirus (HCMV) is a beta herpesvirus, also known as human herpesvirus 5.

parents [1]. The infection may persist for weeks or years following primary infection, when it becomes latent. Recurrent infection may be due to reactivation of latent infection or reinfection with another strain of the virus. Congenital HCMV infection is a major public health concern inducing serious neurodevelopmental sequelae [2]. Severe congenital disease has an infant mortality rate of 20–30%, with death primarily due to multiorgan dysfunction and liver failure. Congenital HCMV may result from maternal primary or recurrent infection. It is the most common intrauterine infection and occurs in about 1% of newborn infants [3–5]. Reported rates of viral transmission to the fetus during maternal primary infection range from 20% to 75%. There is some evidence that transmission to the fetus occurs more readily when maternal primary infection is late in gestation [6,7]. Seropositivity in women of child bearing age varies from 50% to more than 90% in different populations [5]. It is estimated that about 40 000 children (0.2– 2% of all deliveries) are born with HCMV, resulting in about 400 fatal cases each year [8]. In Italy, where serological tests are routinely performed, the prevalence of anti-HCMV immunoglobulin G (IgG) antibodies was 65.87% in Sicily, and was significantly higher in immigrant women [9]. Infection rates are higher in groups with lower socioeconomic status and increase with age. The fetus is severely damaged when infection occurs early in gestation [10,11] and damage can occur with maternal recurrent infection, which is more common than was previously thought [3,4].

Epidemiology HCMV is highly species specific. The virus is endemic worldwide, and there is no seasonal variation. Infection is often acquired in childhood and is readily transmitted from child to child by contact with oral and respiratory secretions. Children can acquire the infection in day care and then transmit it to their

Clinical features Signs and symptoms In an immune competent woman of childbearing age, primary infection is usually subclinical. About 10–15% of children with congenital HCMV infection are symptomatic at birth. Roughly 50% of these will have systemic disease, with petechiae (76%),

Introduction Infection during pregnancy can involve the mother, the placenta, and the fetus. Whether the fetus becomes infected, and the ensuing effects on the fetus depend on a variety of factors. These include the specific infectious agent, the maternal immune response, the effectiveness of the placental barrier, and the fetal immune response. Owing to changes in maternal, placental, and fetal physiology during the course of pregnancy, the effects of an infectious agent on the fetus often depend on when in gestation the infection occurs. Adverse effects of intrauterine infections vary widely and include stillbirth and miscarriage, intrauterine growth restriction, premature birth, developmental anomalies, and congenital disease. Congenital disease may manifest in utero or at birth, or may not become apparent until weeks, months or years after birth. This chapter focuses on infections transmitted from mother to fetus via the placenta. Ascending (transcervical) infections acquired near the time of birth are discussed in Chapter 42.

Cytomegalovirus

Developmental Neuropathology, Second Edition. Edited by Homa Adle-Biassette, Brian N. Harding, and Jeffrey A. Golden. © 2018 John Wiley & Sons Ltd. Published 2018 by John Wiley & Sons Ltd.

497

Developmental Neuropathology jaundice (67%), hepatosplenomegaly (60%), and neurological findings (68%) as the most common signs [12]. Most children (60–90%) with symptomatic infection, and 10–15% of those with asymptomatic infection develop one or more long-term neurological sequelae, such as microcephaly, mental restriction, psychomotor restriction, language disorders, seizures, chorioretinitis and sensorineural hearing loss [2]. Current estimates indicate that approximately 8000 children are affected each year with some neurological sequelae related to in utero HCMV infection [2]. Imaging Prenatal ultrasound may show generalized signs of growth restriction, oligohydramnios, hydrops fetalis, increased signal in gastrointestinal tract and kidneys, hepatomegaly, enlarged placenta, or placental calcifications. Central nervous system (CNS) lesions include brain calcification, head circumference less than the fifth percentile, ventriculomegaly, subependymal cysts, periventricular halo, lenticulostriate vasculopathy (‘candlestick’ sign), abnormal gyration, enlargement of pericerebral spaces, agenesis or hypoplasia of the corpus callosum, and cerebellar hypoplasia. Occipital horn or temporal horn cavities are strongly suggestive of HCMV infection [13–16]. Abnormal findings on ultrasound are a strong predictor for poor fetal outcome [7]. Magnetic resonance imaging (MRI) is more sensitive and reveals additional abnormalities such as hypersignal in the white matter and abnormal gyration. Laboratory findings Diagnosis relies on the assessment of the mother’s immunological status, but may require IgG avidity, immunoblotting and polymerase chain reaction (PCR). The same applies to evaluation of the neonate who is at risk of congenital infection [2]. Prenatal diagnosis aims to determine the timing of maternal infection, whether the fetus is infected, and whether the fetus is likely to have serious disease. Unless there is documented maternal seroconversion for HCMV IgG during pregnancy, it is difficult to determine when primary infection occurred. HCMV immunoglobulin M (IgM) stays high for months after infection and can be increased in recurrent infections. The HCMV IgG avidity index is helpful in determining if infection occurred during pregnancy: low avidity HCMV IgG is present when primary infection has occurred in the previous four to five months, the presence of high avidity HCMV IgG indicates earlier infection [7,17]. To determine whether fetal infection has occurred, both viral culture and PCR of amniotic fluid are highly sensitive and specific [7,11,15,18]. Quantitative analysis of viral load has proved to be the best predictor for neurologic damage in congenital HCMV infection [19–21], but does not predict whether the fetus will have symptomatic disease. Fetal serology for IgM is also a sensitive indicator of fetal infection, but requires obtaining fetal blood [7]. There is a delay between maternal and fetal infection, and fetal

498

diagnostics are most reliable when performed after 21 weeks of gestation and at least 6 weeks after maternal infection [11].

Macroscopy The gross appearance varies with severity and stage of disease. The brain is usually small, and may show destructive lesions such as porencephaly with adjacent micropolygyria. The lateral ventricles are often dilated. Calcifications may be grossly apparent and are frequently seen in the periventricular regions (Figure 41.1a), but can be anywhere in gray or white matter [22,23]. Histopathology Congenital HCMV infection causes multifocal meningoencephalitis, parenchymal atrophy, and ventricular dilatation. [24, 25]. Cortical abnormalities consist of polymicrogyria (Figure 41.1b) associated with areas of dysplastic cortex and neuronal heterotopia. These lesions are associated with meningitis, rupture of the glia limitans, ventricular erosion and radial glial cell loss, calcifications, and inflammation and may be associated with dysplasia of the hippocampus, temporal cysts and olfactory bulb lesions. In cases with ventricular dilatation, necrosis and cell loss may be found in the ventricular and subventricular areas, associated with calcification and inflammation. Resolution of the inflammatory process may cause extensive subependymal gliosis, but this does not frequently lead to obstructive hydrocephalus. It is a hallmark of HCMV that all cell types within the CNS may become infected, including glia, neurons, ependymal cells, choroid plexus epithelium, meningeal cells, and endothelium, but the virus has a predilection for progenitor cells [25]. Cells of the inner ear are often infected [26] and there may also be eye involvement, although chorioretinitis is less common than in toxoplasmosis. Cytomegalic cells have enlarged nuclei and cytoplasm, with a single large intranuclear inclusion surrounded by a clear halo, so-called “owl’s eye” nuclei (Figure 41.1c), and multiple small cytoplasmic inclusions. The nuclear inclusions may appear eosinophilic or basophilic depending on exact staining procedures. The cytoplasmic inclusions are slightly basophilic and periodic acid–Schiff-positive. Immunohistochemical stains for HCMV are now widely available and may identify early antigen, indicating active rather than latent infection. Differential diagnosis Other major congenital infections to be considered in the clinical differential diagnosis are rubella, toxoplasmosis, syphilis, herpes simplex virus, and zika virus. Of these, toxoplasmosis is the most similar to HCMV in clinical and pathological presentation. Pathogenesis The immune response of the fetus is immature [25]. The reason for the predilection of HCMV for progenitor cells is unknown.

Intrauterine Infections Chapter 41 (a)

(b)

(c)

Figure 41.1 Congenital human cytomegalovirus (HCMV) infection. (a) Coronal section showing enlarged lateral ventricles lined by necrotic calcified tissue (arrows). Calcifications and necrosis are also seen in other areas away from the ventricles. (b) Polymicrogyria following intrauterine HCMV infection, Kluver and Barrera stain. (c) Insert: HCMV immunostaining of intranuclear inclusion with surrounding perinuclear halo, “owl’s eye” appearance (image kindly provided by Dr. Brian N. Harding).

It is not understood why some infected fetuses manifest disease and others do not. Infection with HCMV causes direct and indirect cellular injury [25], and viral replication may continue for years after initial infection. Vasculitis can lead to disseminated intravascular coagulation and ischemic necrosis.

Future directions and therapy There are no effective vaccines or therapies, and the destructive lesions are often irreversible. Two studies treating congenitally infected infants postnatally with intravenous ganciclovir described modest success in preventing progression of hearing loss [27,28].

Herpes simplex virus Definition Herpes simplex virus (HSV) 1 (HSV1) and 2 (HSV2) are alpha herpes viruses, also referred to as human herpesvirus 1 and 2. Epidemiology HSVs are present worldwide. They have no seasonal variation and are species-specific for humans. HSV infection is one of the most common viral sexually transmitted diseases worldwide.

After primary infection through mucosa or abraded skin, the virus travels retrogradely in peripheral nerves and establishes latent infection in ganglion cells. Reactivation causes lesions near the site of original exposure [29]. Primary infection in the mother may lead to severe illness in pregnancy and may be associated with virus transmission from mother to fetus or neonate. Classically 75% of cases were due to HSV 2 [30]. HSV2 is mainly found in genital herpes and is almost always sexually transmitted. HSV1 is usually transmitted during childhood via nonsexual contact, but has emerged as a principal causative agent of genital herpes in some developed countries, particularly evident among college age populations of the Midwest (US), where it reached about 78% in 2001 compared with 31% of isolates in 1993 (31,32). Congenital or neonatal HSV can be due to either HSV1 or HSV2, and may occur with maternal primary or recurrent infection. Congenital/neonatal herpes simplex is primarily a perinatally acquired infection and is only briefly described in this chapter. Only about 5% of neonatal HSV cases are acquired in utero as a transplacental or ascending infection, while 85% to 90% are acquired at delivery via direct contact of the baby with infected vaginal secretions during delivery; the remainder are acquired postnatally [29,33]. First-time infection of the mother is the most important factor for the transmission of genital herpes from mother to fetus or neonate [34]. The risk of neonatal

499

Developmental Neuropathology infection varies from 30% to 50% for HSV infections with onset in the third trimester, whereas early pregnancy infection carries a risk of about 1% [35]. The incidence of neonatal HSV disease was found to be 1 in 3200 deliveries, although incidences range from 1 in 1400 to 1 in 30 000 deliveries [36].

Clinical features The mother may be asymptomatic at the time of transmission. Genital herpes during pregnancy has been associated with spontaneous miscarriage, intrauterine growth restriction, premature labor, congenital and neonatal herpes infections. Intrauterine infection usually results in severe disease characterized by skin vesicles or scarring, meningoencephalitis, microcephaly, ventricular enlargement, hemorrhagic infarcts, diffuse porencephalic changes or hydranencephaly, intracranial calcifications, chorioretinitis and optic atrophy. Intrauterine infection is often widely disseminated with liver, adrenal and lung necrosis, encephalitis and multiorgan involvement [33,37]. Lesions limited to the skin or eye are less frequent and are likely due to ascending infection rather than hematogenous transmission (see Chapter 42). Pathology The brain may be small, hydranencephalic, or grossly normal. Meningoencephalitis and small calcifications (Figure 41.2) have been described [38–41]. Viral inclusions may or may not be seen. Therapy In recurrent infections associated with clinical symptoms, the risk of neonatal disease is reduced dramatically by caesarean section [32]. Neonatal HSV is treated with intravenous acyclovir. Treatment is most effective in infants who do not have central nervous system disease and when treatment is begun early in the disease course [42]. However, this does not apply to most cases of intrauterine HSV, which lead to intrauterine demise, postnatal death, sensorineural hearing loss [43], or developmental delay [37]. Acyclovir has been used in pregnancy, but effectiveness has not been proven [37,44].

Toxoplasmosis Definition The etiologic agent of toxoplasmosis is the protozoan Toxoplasma gondii. Epidemiology Toxoplasmosis is a common infection that occurs in all mammals and some birds. The definitive host is the cat. Infection is transmitted to humans by ingestion of raw or undercooked meat that contains tissue cysts (bradyzoites), or inadvertent ingestion of oocysts passed in cat feces through contact with litter boxes or soil [45]. Seroprevalence rates in women of childbearing age

500

Figure 41.2 Intrauterine herpes simplex virus infection. Acute leptomeningitis. Thickened meninges with mixed cell inflammatory infiltrate overlying immature fetal cortex.

range from 15% to 95%, depending on geography, methods of food preparation, maternal age, and historical time [9]. In the United States, seroprevalence in pregnant women is 10–30%. In immunocompetent mothers, the organisms are transmitted to the fetus primarily when initial infection and maternal parasitemia occurs during pregnancy, but transmission from women infected prior to pregnancy, or previously infected with one serotype developing a new infection with a second serotype during pregnancy, has also been described. Intrauterine transmission has been reported as long as 20 years after maternal infection [46]. Immunosuppression can lead to reactivation of infection and increased risk of transmission to the fetus [47]. The incidence of congenital toxoplasmosis increases with the stage of pregnancy. For untreated women, the transmission rate is approximately 25% in the first trimester, 54% in the second trimester, and 65% in the third trimester [48]. Disease in the fetus is more severe when infection occurs in early gestation [45,49]. The time of highest risk for severe congenital toxoplasmosis is between 10 and 24 weeks of gestation. Estimates for the incidence of congenital toxoplasmosis range from 1 per 1000 to 1 per 10 000 live births [47,50].

Intrauterine Infections Chapter 41

Clinical features Signs and symptoms Primary infection in the mother may manifest as lymphadenopathy and fatigue, but is often asymptomatic. The classical presentation of congenital toxoplasmosis is hydrocephalus, chorioretinitis and intracranial calcifications. Severe cases are evident in the fetus and birth may be premature, in the newborn there may be erythroblastosis and hydrops fetalis. The spectrum of disease includes mild cases with either early (infancy) or late (childhood or adolescence) recognition of disease. Infants who are asymptomatic at birth may develop late sequelae, with ocular and neurological symptoms, including seizures, weakness, and mental restriction [45]. There is no conclusive evidence to indicate that toxoplasmosis causes congenital malformations. Imaging Prenatal and postnatal ultrasound, computed tomography (CT) and MRI may demonstrate hydrocephalus and aqueduct stenosis, with calcifications and destructive lesions scattered throughout the brain. Laboratory findings Infection can be diagnosed in the mother by serology on the basis of increasing IgG titers and the presence of toxoplasma IgM, however, IgM can persist for more than a year. Low avidity IgG testing can be useful in identifying recent infection. Fetal IgM is not sensitive to infection in early gestation. PCR of amniotic fluid is a sensitive and specific way of identifying fetal infection [51]. Macroscopy External examination of the brain shows multiple welldemarcated, soft, yellow lesions randomly scattered on the surface, so called ‘’thumbprint” lesions (Figure 41.3a). Sections (Figure 41.3b) show ventricular enlargement and friable, necrotic yellow tissue lining the ventricles often destroying the basal ganglia. Multiple necrotic and calcified lesions are dispersed throughout the brain, brainstem, spinal cord, and cerebellum [24].

Histopathology There is a chronic meningoencephalitis with microglial nodules and extensive vascular involvement with fibrinoid necrosis of blood vessel walls (Figure 41.3d) and intravascular thrombosis. Necrotic areas (Figure 41.3c) are often surrounded by granulation tissue with abundant macrophages. There is extensive periventricular inflammation and a nodular ependymitis that often occludes the aqueduct of Sylvius. There is both fine mineralization outlining cells and blood vessels and coarse mineralization in zones of necrosis. Organisms may be abundant in and near areas of necrosis, microglial nodules, or perivascular inflammation, but may also be seen in non-inflamed tissue

(Figure 41.3e). Intense inflammation is present around tachyzoites, which are oval, 2–3-μm structures in tissue sections. There may be little or no inflammation around cysts and pseudocysts. Tachyzoites may be difficult to identify on hematoxylin and eosin staining, but are clearly visible on immunohistochemistry, particularly when cysts are not identified.

Differential diagnosis Other congenital infections as listed above. Cytomegalovirus infection may closely resemble congenital toxoplasmosis clinically but can be distinguished pathologically. Pathogenesis In the mother, organisms enter the intestines and either invade cells directly or are phagocytosed. Tachyzoites multiply intracellularly, destroy those cells and invade adjacent cells. They eventually invade into all tissues. Both cell-mediated and humoral immunity are required to control the spread of tachyzoites. Cysts may persist in tissues, providing sites for reactivation although it is rare to find such cysts as incidental findings in autopsies. Human placental studies suggest that the extravillous trophoblasts that attach the placenta to the uterus are more vulnerable to infection than syncytiotrophoblasts that are bathed in maternal blood. It is likely that following primary infection, parasitemia leads to intracellular uterine infection, which in turn leads to extravillous trophoblasts infection, as tachyzoites move from cell to cell, with eventual infection of the fetus. It is also possible that tachyzoites in infected maternal leukocytes cross the placenta contributing to fetal infection. Both mechanisms may be involved, which could explain the apparent time delay between primary maternal infection and fetal infection [48]. The three different genotypes of T. gondii, which vary in geographic distribution, may explain conflicting reports from different regions of the world on the relative public health importance of screening and treatment for congenital disease. There may be a trend towards increased transmission by one genotype [48]. Future directions and therapy Preventive measures are important to avoid infection by appropriate food preparation hygiene and avoidance of risk of exposure to the agent during pregnancy. Prenatal treatment with pyrimethamine sulfadiazine reduces the severity of neurological and ophthalmological sequelae of congenital toxoplasmosis, although not all studies have demonstrated effectiveness of this therapy [49,52–54]. Rates of congenital toxoplasmosis have decreased in countries where routine serology testing and education about risks for toxoplasma infection are performed in pregnant women [45]. Spiramycin is recommended for pregnant women with acute toxoplasmosis when fetal infection has not yet occurred, in an attempt to prevent transmission to the fetus. Alternative medications, such as azithromycin, have been used but have not been as extensively studied as spiramycin [48].

501

Developmental Neuropathology (a)

(c)

(b)

(d)

(e)

Figure 41.3 Central nervous system toxoplasmosis (a) Macroscopy “thumbprint” lesions (arrows) with depressed cortical surface and yellow necrotic areas, scattered on the surface of the brain. Premature infant who died at two months. (b) Coronal slice showing extensive necrosis and calcification especially in the periventricular regions. The lateral ventricles are widely dilated. (c) Histopathology of cortex showing a circumscribed necrotic focus extending onto the meningeal surface. Same case as (a). (d) Intense inflammation and necrosis surrounding a blood vessel with fibrinoid necrosis of the wall. (e) Toxoplasma cyst in non-inflamed tissue.

Rubella Definition and synonyms Rubella is an enveloped single-stranded RNA genome virus, a member of the togavirus family. It causes a flu-like illness with a characteristic maculopapular rash, and is also known as German measles and “three-day measles” or “third” disease. Congenital rubella syndrome can occur when the mother becomes infected during pregnancy. Epidemiology Rubella has a worldwide distribution. Humans are the only known host, and infection peaks in spring. In the prevaccination era, epidemics occurred in six- to nine-year cycles, with

502

outbreaks every three to five years. In the 1962–1964 pandemic there were 12.5 million cases of rubella in the United States, and 20 000 infants were born with congenital rubella syndrome [55]. Following the rubella vaccination program introduced in 1969, the number of cases in the United States decreased from 57 686 in 1969 to 271 in 1999 [56]. During 1997–1999, 24 infants were born in the United States with congenital rubella syndrome. Nearly half the mothers were exposed to rubella outside the United States [57]. About 100 000 cases of congenital rubella syndrome are estimated to occur worldwide every year. Rubellacontaining vaccine is highly effective and safe, but incomplete rubella vaccination programs result in continued disease transmission, such as a large outbreak in Japan [58]. About 15 000 cases of rubella and 43 cases of congenital rubella syndrome were reported to the National Epidemiological Surveillance of

Intrauterine Infections Chapter 41

Infectious Diseases in Japan between October 2012 and March 2014, as a result of the 2012–13 rubella outbreak in Japan. There is variability in humoral and cell-mediated innate and adaptive immune responses to rubella-containing vaccine according to haplotypes and single-nucleotide polymorphisms across the human genome. Severe disease including congenital malformations (congenital rubella syndrome) occurs predominantly when maternal primary infection is in the first trimester of pregnancy. Congenital infection occurs in 90% of fetuses with maternal infection before 11 weeks of gestation, 25% between 23 and 26 weeks, and 67% after 31 weeks [59]. Infections before 11 weeks are frequently (90%) teratogenic, while there is a low risk of congenital defects after 16 weeks [60].

Clinical features Signs and symptoms Rubella infection in children or adults causes a mild febrile illness with rash. Congenital rubella has a wide spectrum of disease [61]. Infection early in gestation results in “congenital rubella syndrome” characterized by congenital heart disease, cataracts, blindness, deafness, microcephaly, growth restriction and hepatosplenomegaly. Only 10–20% of infants are symptomatic at birth and symptoms are often not seen until two years of age or later. Late manifestations of the disease include hearing loss, language and behavior disorders, mental restriction, and motor dysfunction. A progressive panencephalitis can develop years after infection and lead to declining mental function, seizures, ataxia and death [62]. Laboratory findings The presence of rubella IgM or low avidity rubella IgG in maternal serum or saliva can be used to identify fetuses at risk for congenital infection [63]. Maternal IgM may persist for a year or more after infection. Fetal serology for IgM can be useful if positive, but is not sensitive early in gestation. Reverse transcription PCR has been used to identify viral RNA in first trimester fetal blood and chorionic villous samples, but appears less sensitive in amniotic fluid [64,65]. Viral culture takes weeks, making it impractical.

Macroscopy The brain may be small and may show atrophy [66,67]. Major structural alterations are rare. Histopathology The histopathology of 89 cases of congenital rubella was reviewed by Rorke [67]. More than half had vascular abnormalities and two-thirds of these had cerebral ischemic injury. The vascular changes consisted of destruction of vascular walls, defects of the internal elastic lamina, endothelial proliferation and perivascular accumulation of granular material, particularly affecting the small penetrating vessels of the cortex and basal ganglia vessels. Necrosis was seen near affected vessels. Inflammation was seen in only 17 of the 89 cases. Tondury

and Smith [68] examined first-trimester fetuses with rubella infection and found CNS endothelial damage with small hemorrhages. Progressive panencephalitis is a late consequence of congenital rubella. Diffuse white matter gliosis, perivascular inflammatory infiltrates, microglial nodules, diffuse neuron loss, amorphous vascular mineralization and cerebellar cortical atrophy have all been described [62].

Pathogenesis Congenital infection is chronic and virus may persist in the fetus and newborn for months. Structural defects result when tissue damage and scarring occur during organogenesis [60]. The pathogenesis of the vascular changes is not understood. Future directions and therapy Immunization can prevent congenital rubella. Treatment with immune globulin in women who did not wish to terminate the pregnancy has not been very effective in preventing fetal disease. Antiviral agents have not yet been shown to be clinically effective in congenital rubella [61].

Varicella Definition and synonyms Primary infection with varicella zoster virus (VZV) causes the acute, highly contagious childhood disease of chickenpox (varicella). Reactivation of latent virus causes zoster (shingles, herpes zoster). VZV is an alpha herpesvirus also known as human herpesvirus 3. Epidemiology Chickenpox is primarily a disease of young children, and over 90% of adults are seropositive. The varicella vaccine has significantly reduced the incidence of primary VZV infection in the population as a whole. Varicella in pregnancy is unusual because most women of childbearing age are immune. The rate of intrauterine transmission of VZV had been thought to be less than 5% [44,69], but Kusterman [70] reported that 5 (36%) of 14 fetuses of mothers with first- or second-trimester varicella were positive by PCR. The incidence of congenital varicella syndrome after maternal varicella during the first two trimesters is less than 1% across multiple cohort studies [71]. The highest risk for spontaneous miscarriage or congenital varicella syndrome is in the first 20 weeks of gestation, but the overall risk in the first 20 weeks is less than 1%. The fetal outcomes include normal healthy infants in almost all cases, possible zoster in infancy, and, rarely, fetal death or congenital varicella syndrome. Maternal infection just before or after delivery presents a high risk for disseminated varicella in the infant. VZV does not appear to cause intrauterine damage to the lungs or liver in infants with congenital varicella syndrome, as it can in perinatal varicella or in other immunocompromised hosts.

503

Developmental Neuropathology

Clinical features Congenital varicella syndome is characterized by skin scars, intrauterine growth restriction, seizures, mental restriction, ventriculomegaly, microcephaly, polymicrogyria, ocular symptoms including microphthalmia, chorioretinitis, cataracts, Horner syndrome, ptosis and nystagmus, limb abnormalities (hypoplasia of bones and muscle, absent digits, club foot), and dysfunction of the autonomic nervous system. The skin scars are usually on the abnormal limb. Affected fetuses are often growth restricted and born prematurely, and infant mortality is high [69]. Pathology Necrotic lesions and scars may be seen on gross examination of the cerebral hemispheres, and there may be ventricular dilatation. Polymicrogyria has been described. The meninges and brain parenchyma show a diffuse chronic inflammatory infiltrate with granulation tissue, microglial nodules, necrosis and gliosis. There is neuron loss in the spinal cord anterior horns and dorsal root ganglia, and posterior and lateral columns of the spinal cord are shrunken and gliotic [72–74]. Muscles in affected limbs show denervation atrophy. Pathogenesis VZV is a neurotropic virus. Many of the defects in congenital varicella syndrome are thought to be a direct result of infection in neural cells of the spinal cord and ganglia, causing destruction of the plexi during embryogenesis, leading to denervation of the limb bud and subsequent hypoplasia [71]. The cutaneous defects are also likely to reflect VZV infection of sensory nerves. Therapy VZV PCR has a poor positive predictive value for fetal disease or disease severity. The prognosis of infants born with congenital varicella syndrome is generally poor. Death in infancy results from intractable gastroesophageal reflux, recurrent aspiration pneumonia and respiratory failure. Both varicella zoster immune globulin and acyclovir have been given to mothers with varicella during pregnancy, but the effectiveness in preventing congenital varicella syndromeis unknown, while varicella zoster immune globulin improves the prognosis of perinatal varicella.

Human immunodeficiency virus (HIV) Transmission of HIV from mother to fetus occurs in 15–40% of untreated mothers. The proportion of intrauterine infection versus perinatal infection is not known, but current belief is that most occurs in the perinatal period [75,76]. HIV has been identified in fetal brain by PCR in a small number of fetuses [77] and there is one report of neonatal meningo-encephalitis attributed to congenital HIV [78]. In a neuropathological study of 134 fetuses from HIV-infected mothers, no specific pathology was identified [79]. In a randomized, placebo-controlled

504

trial, prenatal plus postnatal treatment with zidovudine reduced maternal-fetal transmission from 25.5% to 8.3% [80]. While an HIV embryopathy was originally reported in babies and children with congenital HIV infection, this concept is not now accepted, and the “dysmorphic” craniofacial features described as characteristic of HIV embryopathy can also be attributed to other factors. Chapter 42 discusses perinatal HIV infection.

Enteroviruses Enteroviruses including polioviruses, coxsackieviruses A and B, and echoviruses, can be transmitted transplacentally [81]. Infection with polioviruses can cause disease similar to that seen in older children and adults, with destruction of anterior horn and motor nuclei neurons. Echoviruses and coxsackieviruses B have been shown to cause fetal meningoencephalitis.

Parvovirus B19 Parvovirus B19 is the only parvovirus known to infect humans. When transmitted to the fetus, it has a predilection to infect red blood cell precursors, and is a well-described cause of hydrops fetalis. A few cases of severe CNS abnormalities have been described with parvovirus B19 infection, mainly due to destructive lesions. These included cortical dysplasia, ventricular dilatation, and necrotic lesions or calcification in the basal ganglia, thalamus, periventricular region and spinal cord [82]. Isumi [83] reported cerebral white matter perivascular calcifications and multinucleated giant cells that were positive for parvovirus B19 by in situ PCR in a fetus at 27 weeks of gestational age. Feline parvovirus is known to cause cerebellar hypoplasia.

Zika virus Zika virus is a flavivirus transmitted by the mosquito Aedes aegypti. It was first identified in Brazil in 1947. An outbreak in 2015 spread within Brazil and to other countries in South and Central America in less than one year. Human infection may be asymptomatic, but can cause mild fever, arthralgia, rash, headache, and myalgia. An association has been established between Zika virus infection in the mother and microcephaly and intracerebral calcification in babies, and fetal death. Neuropathological study in four Brazilian babies born to mothers infected in the first or second trimester, who died within 48 hours of birth [84] showed microcephaly, and arthrogyposis. Microcephaly was not obvious in all cases, due to compensatory expansion of head circumference by ventriculomegaly, but there were shallow sulci or agyria, thin cortex, and white matter and calcifications (Figure 41.4a). Histopathology showed chronic lymphocytic

Intrauterine Infections Chapter 41

(a)

(b)

(c)

Figure 41.4 Zika virus infection (a) Macroscopy: coronal section of the left hemisphere in a baby born at 41 weeks of gestation from a mother who presented with a rash at 12 weeks of gestation. The head circumference was 29.5 cm: microcephaly, agyria and ventricular dilatation. (b) Section of the brainstem: nerve cell degeneration, astrocytosis and filamentous calcifications(images kindly provided by Professor L. Chimelli). (c) Personal case terminated at 21 weeks of gestation because of microcephaly associated with Zika virus infection. Note multiple necrotic cavitating inflammatory foci with calcifications.

meningitis, and scattered perivascular CD3-positive lymphocytes and CD68-positive microglia and gliosis. Erosion of the ependyma was also seen and clusters of germinal matrix cells along the ventricular surface also formed irregular nodules towards the cortex. Polymicrogyria and leptomeningeal glioneuronal heterotopia were observed in the hemispheres and brainstem. The thalamus and basal ganglia were small and calcified, the cerebellum was hypoplastic with focal cortical dysplasia. Nerve cell degeneration, coarse and filamentous calcification, were seen in the brainstem (Figure 41.4b). In a personal case terminated at 21 weeks of gestation, neuropathological study showed microcephaly, ventriculomegaly, dysplasia of the cortex and basal ganglia and multiple necrotic inflammatory foci with calcifications (Figure 41.4c). In another ultrastructural study of a 20-week fetus, an intracellular aggregate of particles consistent with viral particles was identified in macrophages in the subventricular zone and white matter [85].

Syphilis Congenital syphilis results from transplacental infection of the fetus with the spirochete Treponema pallidum. Most infected fetuses have no symptoms at birth, but, if untreated, may develop clinical disease months or years later, typically presenting as acute meningitis [86]. The incidence of congenital syphilis varies widely and correlates with the incidence of primary and secondary syphilis in women of childbearing age in the population. Intrauterine infection may cause stillbirth, hydrops fetalis, hepatosplenomegaly and vesicular or bullous skin lesions. Fetal infection affects most organs. The CNS pathology is similar to that of syphilis acquired later in life. Basilar meningitis and

meningeal fibrosis may lead to hydrocephalus, and vasculitis can cause ischemic necrosis. Tabes dorsalis and general paresis are now rare. Penicillin is effective in treating the fetus even when given late in pregnancy. Sequelae of congenital syphilis are rare in countries where women are tested for syphilis during pregnancy and where penicillin is available.

Listeriosis Congenital infection with Listeria monocytogenes may result in miscarriage, stillbirth, prematurity or neonatal disease. The incidence is approximately 13 per 100 000 births [87]. Miliary abscesses are seen throughout the placenta, and in the fetus, including the skin, liver, lungs, kidneys and CNS, and there is often purulent meningitis. Fetal infection is most frequently the result of ingestion of contaminated food by the mother. Avoidance of such foodstuffs (e.g. soft cheeses), is recommended in pregnancy.

Tuberculosis The incidence of tuberculosis during pregnancy in the United States is unknown, but with increases in immigration, HIV infection, and overall tuberculosis incidence, it has likely increased in recent years [88]. Intrauterine transmission of Mycobacterium tuberculosis can be through the placenta and umbilical vein, or by swallowing or aspiration of infected amniotic fluid. Hepatic, pulmonary and lymph node involvement are most common; meningitis is found in about one-third of cases.

505

Developmental Neuropathology

Other infections Measles virus infection (also called rubeola) in the first trimester of pregnancy has been associated with isolated abnormalities of the CNS, including leukodystrophy, cyclopia, and microcephaly [89]. Human intrauterine infection with lymphocytic choriomeningitis virus can cause congenital hydrocephalus and chorioretinitis [90]. Fetal meningitis has been reported with infection by mycoplasmas, Cryptococcus neoformans, Coccidiodes immitis, and Trypanosoma brucei [91–93]. Abundant spirochetes have been described in the brain in congenital Lyme disease (Borrelia burgdorferi), without significant inflammation [94,95].

Differential diagnosis Some of the most common infections associated with congenital anomalies are designated as TORCH, an acronym for toxoplasmosis, others (syphilis, varicella zoster, parvovirus B19), rubella, cytomegalovirus, and herpes infections. An important differential diagnosis is the pseudo-TORCH syndrome, which clinically manifests with band-like calcification with simplified gyration and polymicrogyria (Mendelian Inheritance in Man 251290) caused by homozygous or compound heterozygous mutation in the gene encoding occludin on chromosome 5q13. Similarly, Aicardi–Gouti`eres syndrome is a disorder clinically resembling in utero viral infection. It is a genetically heterogeneous encephalopathy characterized in its most severe form by cerebral atrophy, intracranial calcifications, leukodystrophy, chronic cerebrospinal fluid lymphocytosis and interferonalpha (IFN-α) increase and negative serological investigations for common prenatal infections. Severe neurologic dysfunction becomes clinically manifest in infancy as a progressive microcephaly, profound psychomotor restriction, spasticity, dystonic posturing, and sometimes death in early childhood. Systemic abnormalities include fever, abnormal liver function with hepatosplenomegaly, chilblain-like cutaneous lesions, raised levels of immunoglobulins and autoantibodies, and thrombocytopenia. The pathological finding of focal infarctions together with patchy myelin loss and calcified deposits in small blood vessels (media, adventitia, and perivascular space) suggests a genetic cerebral angiopathy. IFN-α inhibits angiogenesis, and astrocyte-specific chronic overproduction of IFN-α in transgenic mice recapitulates the neuropathological findings of Aicardi–Gouti`eres syndrome, suggesting a role for this cytokine in its pathogenesis [96]. Studies have established that the disorder may result from a defect in the degradation of RNA– DNA heteroduplexes, causing them to accumulate at high levels [97]. Mutations in the following genes are responsible for the disorder: TREX1 (3′ to 5′ single-stranded DNA exonuclease, AGS1, [98]); RNASEH2A, RNASEH2B, and RNASEH2C (RNase

506

H2AGS2-4, [99]); SAMHD1 (3′ to 5′ exonuclease and dNTP hydrolase, AGS5, [100,101]); ADAR1 (RNA adenosine deaminase, AGS6); and gain of function mutations in the cytosolic double-stranded RNA receptor gene IFIH1 [102], mimicking the intracellular response to a viral infection.

References 1. Adler S (1986) Molecular epidemiology of cytomegalovirus: evidence for viral transmission to parents from children infected at a day care center. Pediatr Infect Dis 5:315–18 2. Cheeran MC, Lokensgard JR, Schleiss MR (2009) Neuropathogenesis of congenital cytomegalovirus infection: disease mechanisms and prospects for intervention. Clin Microbiol Rev 22:99–126 3. Ahlfors K, Ivarsson SA, Harris S (1999) Report on a long-term study of maternal and congenital cytomegalovirus infection in Sweden. Review of prospective studies available in the literature. Scand J Infect Dis 31:443–57 4. Boppana S, Fowler K, Britt W et al. (1999) Symptomatic congenital cytomegalovirus infection in infants born to mothers with preexisting immunity to cytomegalovirus. Pediatrics 104:55–60 5. Stagno S (2001) Cytomegalovirus. In: J Remington and J Klein, eds, Infectious Diseases of the Fetus and Newborn Infant, 5th ed, Philadelphia, PA, WB Saunders. pp. 389–424 6. Bod´eus M, Hubinont C, Goubau P (1999) Increased risk of cytomegalovirus transmission in utero during late gestation. Obstet Gynecol 93:658–60 7. Enders G, B¨ader U, Lindemann L et al. (2001) Prenatal diagnosis of congenital cytomegalovirus infection in 189 pregnancies with known outcome. Prenat Diagn 21:362–77 8. Cannon M J, Davis KF (2005) Washing our hands of the congenital cytomegalovirus disease epidemic. BMC Public Health 5:70 9. Puccio G, Cajozzo C, Canduscio LA et al. (2014) Epidemiology of toxoplasma and CMV serology and of GBS colonizationin pregnancy and neonatal outcome in a Sicilian population. Ital J Pediatr 40:23 10. Ahlfors K, Forsgren M, Ivarsson SA et al. (1983) Congenital cytomegalovirus infection: on the relation between type and time of maternal infection and infant’s symptoms. Scand J Infect Dis 15:129–38 11. Liesnard C, Donner C, Brancart F et al. (2000) Prenatal diagnosis of congenital cytomegalovirus infection: prospective study of 237 pregnancies at risk. Obstet Gynecol 95:881–8 12. Boppana S, Pass R, Britt W et al. (1992) Symptomatic congenital cytomegalovirus infection: neonatal morbidity and mortality. Pediatr Infec Dis J 11:93–9 13. Picone O, Teissier N, Cordier AG et al. (2014) Detailed in utero ultrasound description of 30 cases of congenital cytomegalovirus infection. Prenat Diagn 34:518–24 14. Henrich W, Meckies J, Dudenhausen J et al. (2002) Recurrent cytomegalovirus infection during pregnancy: ultrasonographic diagnosis and fetal outcome. Ultrasound Obstet Gynecol 19:608– 11 15. Lipitz S, Achiron R, Zalel Y et al. (2002) Outcome of pregnancies with vertical transmission of primary cytomegalovirus transmission. Obstet Gynecol 100:428–33

Intrauterine Infections Chapter 41 16. Malinger G, Lev D, Zahalka N et al. (2003) Fetal cytomegalovirus infection of the brain: the spectrum of sonographic findings. AJNR 24:28–32 17. Lazzarotto T, Galli C, Pulvirenti R et al. (2001) Evaluation of the Abbott AxSYM cytomegalovirus (CMV) immunoglobulin M (IgM) assay in conjunction with other CMV IgM tests and a CMV IgG avidity assay. Clin Diagn Lab Immunol 8:196–8 18. Nigro G, Mazzocco M, Anceschi M et al. (1999) Prenatal diagnosis of fetal cytomegalovirus infection after primary or recurrent maternal infection. Obstet Gynecol 94:909–14 19. Gouarin S, Gault E, Vabret A et al. (2002) Real-time PCR quantification of human cytomegalovirus DNA in amniotic fluid samples from mothers with primary infection. J Clin Microbiol 40: 1767–72 20. Guerra B, Lazzarotto T, Quarta S et al. (2000) Prenatal diagnosis of symptomatic congenital cytomegalovirus infection. Am J Obstet Gynecol 183:476–82 21. Lazzarotto T, Varani S, Guerra B et al. (2000) Prenatal indicators of congenital cytomegalovirus infection. J Pediatr 137:90–5 22. Haymaker W, Girdany B, Stephens J et al. (1954) Cerebral involvement with advanced periventricular calcification in generalized cytomegalic inclusion disease in the newborn. J Neuropath Exp Neurol 13:562–86 23. Perlman J, Argyle C (1992) Lethal cytomegalovirus infection in preterm infants: clinical, radiological and neuropathological findings. Ann Neurol 31:64–8 24. Friede RL (1989) In: Developmental Neuropathology. 2nd ed. Berlin, Springer, pp. 158–160 25. Teissier N, Fallet-Bianco C, Delezoide AL et al. (2014) Cytomegalovirus-induced brain malformations in fetuses. J Neuropathol Exp Neurol 73:143–58 26. Teissier N, Delezoide AL, Mas AE et al. (2011) Inner ear lesions in congenital cytomegalovirus infection of human fetuses. Acta Neuropathol 122:763–74 27. Michaels M, Greenberg D, Sabo D, Wald E (2003) Treatment of children with congenital cytomegalovirus infection with ganciclovir. Pediatr Infec Dis J 22:504–8 28. Whitley R, Cloud G, Gruber W et al. (1997) Ganciclovir treatment of symptomatic congenital cytomegalovirus infection: results of a phase II study. J Infect Dis 175:080–6 29. Whitley R (2001) Herpes simplex virus infections. Lancet 357:1513–18 30. Kimberlin D, Lin CY, Jacobs R et al. (2001) Natural history of neonatal herpes simplex virus infections in the acyclovir era. Pediatrics 108:223–9 31. Roberts CM, Pfister JR, Spear SJ (2003) Increasing proportion of herpes simplex virus type 1 as a cause of genital herpes infection incollege students. Sex Transm Dis 30:797–800 32. Anzivino E, Fioriti D, Mischitelli M et al. (2009) Herpes simplex virus infection in pregnancy and in neonate: status of art of epidemiology, diagnosis, therapy and prevention. Virol J 6:40 33. Arvin A, Whitley R (2001) Herpes simplex virus infections. In: J Remington, J Klein (eds) Infectious Diseases of the Fetus and Newborn Infant, 5th ed.,: Philadelphia, PA, WB Saunders, pp. 425–46 34. Avgil M, Ornoy A (2006) Herpes simplex virus and Epstein– Barr virus infections in pregnancy: consequences of neonatal or intrauterine infection. Reprod Toxicol 21:436–45

35. Workowski KA, Bolan GA; Centers for Disease Control and Prevention (2015) Sexually transmitted diseases treatment guidelines, 2015. MMWR Recomm Rep 64(RR-03):1–137 36. Corey L, Wald A (2009) Maternal and Neonatal Herpes Simplex Virus Infections. N Engl J Med 361:1376–85 37. Marquez L, Levy ML, Munoz FM, Palazzi DL (2011) A report of three cases and review of intrauterine herpes simplex virus infection. Pediatr Infect Dis J 30:153–7 38. Chalub E, Baenziger J, Feigen R et al. (1977) Congenital herpes simplex type II infection with extensive hepatic calcification, bone lesions and cataracts: complete postmortem examination. Dev Med Child Neurol 19:527–34 39. Hutto C, Arvin A, Jacobs R et al. (1987) Intrauterine herpes virus infections. J Pediatr 110:97–101 40. Monif G, Kellner K, Donnelly W (1985) Congenital herpes simplex type II infection. Am J Obstet Gynecol 152:1000–2 41. Wolf A, Cowen D (1959) Perinatal infections of the central nervous system. J Neuropathol Exp Neurol 18:191–243 42. Kimberlin D, Lin CY, Jacobs R et al. (2001) Safety and efficacy of high–dose intravenous acyclovir in the management of neonatal herpes simplex virus infections. Pediatrics 108: 230–8 43. Westerberg BD, Atashband S, Kozak FK (2008) A systematic review of the incidence of sensorineural hearing loss in neonates exposed to herpes simplex virus (HSV). Int J Pediatr Otorhinolaryngol 72:931–7 44. Whitley R, Kimberlin D (1997) Treatment of viral infections during pregnancy and the perinatal period. Clin Perinatol 24: 267–83 45. Remington J, McLeod R, Thulliez P, Desmonts G (2001) Toxoplasmosis. In: J Remington, J Klein J, eds, Infectious Diseases of the Fetus and Newborn Infant, 5th ed., Philadelphia, WB Saunders. pp. 205–346 46. Silveira C, Ferreira R, Muccioli C et al. (2003) Toxoplasmosis transmitted to a newborn from the mother infected 20 years earlier. Am J Ophthalmol 136:370–1 47. Mitchell C, Erlich S, Mastrucii M et al. (1990) Congenital toxoplasmosis occurring in infants perinatally infected with human immunodeficiency virus 1. Pediatr Infec Dis J 9:512–18 48. McAuley JB (2014) Congenital toxoplasmosis. J Pediatric Infect Dis Soc 3(Suppl 1):S30–5 49. Foulon W, Villena I, Stray-Petersen B et al. (1999) Treatment of toxoplasmosis during pregnancy: a multicenter study of impact on fetal transmission and children’s sequalae at age 1 year. Am J Obstet Gynecol 180:410–15 50. Guerina N, Hsu H, Meissner H et al. (1994) Neonatal serologic screening and early treatment for congenital Toxoplasma gondii infection. N Engl J Med 330:1858–63 51. Grover C, Thulliez P, Remington J, Boothroyd J (1990) Rapid prenatal diagnosis of congenital Toxoplasma infection by using polymerase chain reaction and amniotic fluid. J Clin Microbiol 28:2297–301 52. Br´ezin A, Thulliez P, Couvreur J, Nobr´e R (2003) Ophthalmic outcomes after prenatal and postnatal treatment of congenital toxoplasmosis. Am J Ophthalmol 135:779–84 53. Daffos F, Forestier F, Capella-Pavlovsky M et al. (1988) Prenatal management of 746 pregnancies at risk for congenital toxoplasmosis. N Engl J Med 318:271–5

507

Developmental Neuropathology 54. Gras L, Gilbert R, Ades A, Dunn D (2001) Effect of prenatal treatment on the risk of intracranial and ocular lesions in children with congenital toxoplasmosis. Int J Epidemiol 30:1309–13 55. Orenstein W, Bart K, Hinman A et al. (1984) The opportunity and obligation to eliminate rubella from the United States. JAMA 251:1988–94 56. Centers for Disease Control and Prevention (2001) Control and prevention of rubella: evaluation and management of suspected outbreaks, rubella in pregnant women, and surveillance for congenital rubella syndrome. MMWR Recomm Rep 50:1–22 57. Centers for Disease Control and Prevention (2000) Measles, rubella and congenital rubella syndrome – United States and Mexico, 1997–1999. MMWR 49:1048–59 58. Lambert N, Strebel P, Orenstein W et al. (2015) Rubella in Japan. Lancet 385:2297–307 59. Miller E, Cradock-Watson J, Pollock T (1982) Consequences of confirmed maternal rubella at successive stages of pregnancy. Lancet 2:781–4 60. Webster W (1998) Teratogen update: congenital rubella. Teratology 58:13–23 61. Cooper L, Alford C (2001) Rubella. In: J Remington, J Klein, eds, Infectious Diseases of the Fetus and Newborn Infant, 5th ed., Philadelphia, PA, WB Saunders. pp. 347–88 62. Townsend J, Wolinsky J, Baringer J (1976) The neuropathology of progressive rubella pancencephalitis of late onset. Brain 99:81–90 63. Akingbade D, Cohen B, Brown D (2003) Detection of lowavidity immunoglobulin G in oral fluid samples: new approach for rubella diagnosis and surveillance. Clin Diagn Lab Immunol 10: 189–90 64. Bosma T, Corbett K, O’Shea S et al. (1995) PCR for detection of rubella virus RNA in clinical samples. J Clin Microbiol 33:1075–9 65. Tanemura M, Suzumori K, Yagami Y, Katow S (1996) Diagnosis of fetal rubella infection with reverse transcription and nested polymerase chain reaction: a study of 34 cases diagnosed in fetuses. Am J Obstet Gynecol 174:578–82 66. Love S, Wiley C (2002) Viral diseases. In: DI Graham, P Lantos, P. eds, Greenfield’s Neuropathology, 7th ed, London, Arnold. pp. 1–105 67. Rorke L (1973) Nervous system lesions in the congenital rubella syndrome. Arch Otolaryngol 98:249–51 68. Tondury G, Smith D (1966) Fetal rubella pathology. J Pediatr 68:867–79 69. Gershon A (2001) Chickenpox, measles and mumps. In: J Remington, J Klein, eds, Infectious Diseases of the Fetus and Newborn Infant, 5th ed., Philadelphia, WB Saunders. pp. 683–732 70. Kustermann A, Zoppini C, Tassis B et al. (1996) Prenatal diagnosis of congenital varicella syndrome. Prenat Diagn 16:71–4 71. Smith CK, Arvin AM (2009) Varicella in the fetus and newborn. Semin Fetal Neonatal Med 14:209–17 72. Harding B, Baumer J (1976) Congenital varicella-zoster: a serologically proven case with necrotizing encephalitis and malformation. Acta Neuropathol 76:311–15 73. Magliocco A, Demetrick D, Sarnat H, Hwang W (1992) Varicella embryopathy. Arch Pathol Lab Med 116:181–6 74. Srabstein J, Morris N, Larke R et al. (1974) Is there a congenital varicella syndrome? J Pediatr 84:239–43 75. CDC (1997) Update: perinatally acquired HIV/AIDS – United States, 1997. MMWR 46:1086–92

508

76. Mueller B, Pizzo P (2001) Acquired Immunodeficiency Syndrome in the Infant. In: J Remington, J Klein, eds, Infectious Diseases of the Fetus and Newborn Infant, 5th ed., Philadelphia, WB Saunders. pp. 447–75 77. Soeiro R, Rubinstein A, Rashbaum W, Lyman W (1992) Maternofetal transmission of AIDS: frequency of human immunodeficiency virus type 1 nucleic acid sequences in human fetal DNA. J Infect Dis 166:699–703 78. Srugo J, Witteck A, Israele V, Brunell P (1992) Meningoencephalitis in a neonate congenitally infected with human immunodeficiency virus type 1. J Pediatr 120:93–5 79. Encha-Ravazi F, Larroche J, Roume J et al. (1997) [Neuropathological study of 134 fetuses in HIV infected mother]. Arch Anat Cytol Path 45:118–20 80. Connor E, Sperling R, Gelber R et al. (1994) Reduction of maternal–infant transmission of human immunodeficiency virus type I with zidovudine treatment. Pediatric AIDS Clinical Trials Group Protocol 076 Group. N Engl J Med 331:1173–80 81. Cherry J (2001) Enteroviruses. In: J Remington, J Klein, eds) Infectious Diseases of the Fetus and Newborn Infant, 5th ed. Philadelphia, WB Saunders. pp. 477–518 82. T¨or¨ok T (2001) Human parvovirus B19. In: J Remington, J Klein, eds, Infectious Diseases of the Fetus and Newborn Infant, 5th ed., Philadelphia, PA, WB Saunders. pp. 779–811 83. Isumi H, Nunoue T, Nishida A, Takashima S (1999) Fetal brain infection with human parvovirus B19. Pediatr Neurol 21: 661–3 84. Chimelli L, Melo ASO, Tovar-Moll F et al. (2016) Neuropathological aspects of congenital Zika virus infection in Brazil. (Abstract) J Neuropathol Exp Neurol 75:567 85. Ames H, Rodriguez F, Hill D, Ho C-Y (2016) Early neuropathological findings in congenital Zika virus infection. J Neuropathol Exp Neurol 75:567 86. Ingall D, S´anchez P (2001) Syphilis. In: J Remington, J Klein, eds, Infectious Diseases of the Fetus and Newborn Infant, 5th ed., Philadelphia, PA, WB Saunders. pp. 643–81 87. Bortolussi R, Schlech W (2001) Listeriosis. In: J Remington, J Klein, eds, Infectious Diseases of the Fetus and Newborn Infant, 5th ed., Philadelphia, PA, WB Saunders. pp. 1157–77 88. Starke J, Smith M (2001) Tuberculosis. In: J Remington, J Klein, eds, Infectious Diseases of the Fetus and Newborn Infant, 5th ed, Philadelphia, PA, WB Saunders. pp. 1179–97 89. Jespersen C, Littauer J, Sagild U (1977) Measles as a cause of fetal defects. Acta Paediatr Scand 66:367–72 90. Arvin A, Maldonado Y (2001) Other viral infections of the fetus and newborn. In: J Remington, J Klein, eds, Infectious Diseases of the Fetus and Newborn Infant, 5th ed. Philadelphia, PA, WB Saunders. pp. 855–66 91. Cassell G, Waites K, Crouse D (2001) Mycoplasmal infections. In: J Remington, J Klein, eds, Infectious Diseases of the Fetus and Newborn Infant, 5th ed., Philadelphia, PA, WB Saunders. pp. 733–67 92. Maldonado Y (2001) Protozoan and helminth infections (including Pneumocystis carinii). In: J Remington, J Klein, eds, Infectious Diseases of the Fetus and Newborn Infant, 5th ed., Philadelphia, PA, WB Saunders. pp. 867–911 93. Miller M (2001) Fungal infections. In: J Remington, J Klein, eds, Infectious Diseases of the Fetus and Newborn Infant, 5th ed., Philadelphia, PA, WB Saunders. pp. 813–53

Intrauterine Infections Chapter 41 94. MacDonald A, Benach J, Burgdorfer W (1987) Stillbirth following maternal Lyme disease. NY State J Med 87:615–16 95. Schlesinger P, Duray P, Burke B et al. (1985) Maternal–fetal transmission of the Lyme disease spirochete, Borrelia bugdorferia. Ann Intern Med 103:67–8 96. Campbell IL, Krucker T, Steffensen S et al. (1999) Structural and functional neuropathology in transgenic mice with CNS expression of IFN-alpha. Brain Res 835:46–61 97. Lim YW, Sanz LA, Xu X et al. (2015) Genome-wide DNA hypomethylation and RNA:DNA hybrid accumulation in Aicardi– Gouti`eres syndrome. Elife. 2015 Jul 16;4. doi: 10.7554/eLife.08007 98. Crow YJ, Hayward BE, Parmar R et al. (2006) Mutations in the gene encoding the 3′ -5′ DNA exonuclease TREX1 cause Aicardi– Gouti`eres syndrome at the AGS1 locus. Nat Genet 38:917–20

99. Crow YJ, Leitch A, Hayward BE et al. (2006) Mutations in genes encoding ribonuclease H2 subunits cause Aicardi–Gouti`eres syndrome and mimic congenital viral brain infection. Nat Genet 38:910–16 100. Rice GI, Bond J, Asipu A et al. (2009) Mutations involved in Aicardi–Gouti`eres syndrome implicate SAMHD1 as regulator of the innate immune response. Nat Genet 41:829–32 101. Goldstone DC, Ennis-Adeniran V, Hedden JJ et al. (2011) HIV-1 restriction factor SAMHD1 is a deoxynucleoside triphosphate triphosphohydrolase. Nature 480:379–82 102. Rice GI, del Toro Duany Y, Jenkinson EM et al. (2014) Gain-offunction mutations in IFIH1 cause a spectrum of human disease phenotypes associated with upregulated type I interferon signaling. Nat Genet 46:503–9

509

42

Perinatal and Postnatal Infections Catherine Keohane Department of Pathology and School of Medicine, University College Cork, Cork, Ireland

Introduction Definition Central nervous system (CNS) infections refer to the involvement of CNS tissues and their surroundings by infectious agents including viruses, bacteria, and parasites. Of these, bacterial meningitis poses the most serious threat to neonates and children, both in terms of numbers of cases affected and mortality rates. This chapter deals with the common pathological changes in CNS related to perinatal and postnatal infections. Parainfectious CNS complications are also summarized. Historical perspective Infections of the CNS have always afflicted humans. Hippocrates described meningitis 25 centuries ago [1]. Robert Whytt wrote about tuberculous meningitis in an article “On the dropsy in the Brain” in 1768. Virus infections and parasites are only more recently described. Widespread immunization programs have led to an overall reduction in the incidence in the United States of specific meningitis serotypes except in babies less than two months of age [2]. With increasing numbers of children undergoing organ transplantation and surviving previously fatal immune deficiency diseases, opportunistic infections due to uncommon bacteria, viruses, and fungi are increasingly described. Most recently, CNS effects of HIN1 virus (swine flu), West Nile virus, Ebola virus and Zika virus in children are recognized, but pathological descriptions are few. Background CNS infections especially meningitis are serious life-threatening diseases in neonates and children. Their incidence and type differ according to maturity and also differ in various geographic regions. In neonatal bacterial meningitis, Group B streptococci

(GBS; Streptococcus agalactiae), Eschericia coli, Gram-negative bacilli other than E. coli, Haemophilus influenzae, and Listeria monocytogenes are the most important causative organisms in most of the world [2,3]; in Africa and Asia, other bacteria, including Streptococcus pneumoniae [4] and salmonella species [5] are also important. As globalization increases, infectious agents are less restricted geographically. Poverty and malnutrition increase susceptibility to infection. Morbidity and mortality vary, indicating differences in health care provision. Multiresistant organisms also contribute to morbidity. In infants and older children, Neisseria meningitidis, H. influenzae (type b) and Streptococcus pneumoniae are the more common responsible organisms. Immunodeficiency is a major factor in childhood CNS infections. HIV infection, chemotherapy and bone marrow transplant in childhood malignancies and immune suppression in organ transplantation bring the risk of fulminant bacterial sepsis, opportunistic viral, parasitic, and fungal infections. New outbreaks of highly virulent organisms such as Ebola virus, HIN1 (swine flu), Zika virus and West Nile virus have affected many children in the twenty-first century and bring the challenge of developing effective treatments and preventions. Babies and young children cannot give a history or document symptoms, so clinical suspicion must be high. Additionally, clinical symptoms reflect the host response, which may be lacking in neonates whose immune response is immature. CNS infections require urgent treatment to avoid the complications caused by brain swelling or diffuse spread of sepsis. Even with effective and rapid treatment, tissue destruction in infections such as herpes encephalitis often cause permanent neurological deficits. Clinically the diagnostic approach is based on syndrome recognition including acute and chronic meningitis, acute and chronic encephalitis, space-occupying lesion, toxin-mediated and postinfectious syndromes [6].

Developmental Neuropathology, Second Edition. Edited by Homa Adle-Biassette, Brian N. Harding, and Jeffrey A. Golden. © 2018 John Wiley & Sons Ltd. Published 2018 by John Wiley & Sons Ltd.

511

Developmental Neuropathology

Routes of infection The skull and spine, meninges, blood–brain and blood– cerebrospinal fluid barriers protect the CNS from entry of organisms. Once these barriers are breached, infection can spread rapidly through the cerebrospinal fluid because the CNS is largely remote from immune protection. The anatomical planes also localize infection (e.g., epidural abscess, subdural empyema, meningitis, encephalitis, brain abscess). Overlap occurs; for example, many viruses cause a meningoencephalomyelitis (inflammation of meninges, brain, and spinal cord), and listeriosis causes meningitis, brainstem encephalitis, and abscesses. The most common routes of infections are: r local spread from contact with an infected birth canal or through an entry site such as an open congenital CNS malformation r hematogenous spread from systemic infection r trauma and surgery especially penetrating injuries or compound fractures (uncommon in children) r venous spread from adjacent septic sites (e.g. sinusitis or mastoiditis) r intravascular catheters or intraventricular shunts r spread along nerves, which especially occurs in some viral infections and also in listeriosis.

Host factors influencing an infection The variation in clinical response to identical infecting pathogens is a result of the combined effects of pathogen and host genetic factors. While host genetic factors are important, the role of nutrition, poverty, and access to medical care must also be emphasized. In neonates and children, overwhelming CNS infection may be the first evidence of an underlying congenital immune disorder (e.g., agammaglobulinemia). Low birthweight and premature newborns, in contrast to mature neonates protected by maternal antibodies, have immature immunity and increased risk of infection, especially meningitis [7]. They have low maternal immunoglobulin and antibodies to bacteria, their neutrophils are functionally immature, and the immune response to bacteremia is immature. They frequently have intravenous catheters, nasogastric or endotracheal tubes, which facilitate bacterial invasion, and metabolic acidosis which compromises the host response [8]. In immune competent hosts, primary invading organisms are virulent, act on defined host cell or humoral receptors and cause specific immune responses, which aid in their diagnosis and allow for specific treatment. Immune deficient hosts develop atypical infections from low virulence or commensal organisms that do not interact with defined host cell or humoral receptors. The congenital immune deficiencies of T- and B-lymphocytes, of complement and of phagocytes are severe and fortunately

512

rare. They represent the tip of the iceberg in predisposition to infection [9]. Milder alterations probably remain unrecognized, but are more common. In the human CNS, pathogen recognition receptors include toll-like receptors (TLRs) expressed in meningeal macrophages, in the choroid plexuses and perivascular spaces [10]. Human TLR2 and TLR4 are activated by bacterial lipopolysaccharide. A retrospective study in 535 preterm infants investigating the role of polymorphic variants in genes that modulate the host response to infection including inflammatory cytokines interleukin 6 (IL6), IL10, IL1B, and tumor necrosis factor (TNF), cytokine receptors Il1 RN, TLR2, TLR4, and TLR5, and cell surface receptors (CD14), found that polymorphisms in TLR2 (rs3804099), TLR5 (rs5744105), IL10(rs1800896), and phospholipase A2 PLA2G2A (rs1891320) genes were associated with neonatal sepsis [11]. Allelic variants in PLA2G2A and TLR2 were associated with Gram-positive infections, whereas those in IL10 were associated with Gram-negative infections. A genome-wide expression study using Mycobacterium tuberculosis (Mtb)-stimulated macrophages, identified that human polymorphism in epiregulin (EREG) a TLR-dependent gene, and a member of the epidermal growth factor family involved in the macrophage response to Mtb infection, was associated with susceptibility to tuberculosis. In addition, the risk of tuberculosis was further increased in individuals who had EREG polymorphism rs7675690 and were infected with the Beijing strain of Mtb [12]. Impaired interferon gamma-mediated immunity in children is found in atypical mycobacterial infection [13] and may also be a risk factor for herpesvirus infection [14]. Mutations in the genes controlling TLR3 pathways may be a risk factor for some cases of childhood herpes simplex encephalitis [15]. Children under five years are particularly susceptible to malaria infection. Seventy percent of all malaria deaths occur in this age group, mainly in sub-Saharan Africa. Between 2000 and 2015, the malaria death rate in the under fives fell by 65% globally, translating into an estimated 5.9 million children’s lives saved. Host genetic mutations and polymorphisms are linked to susceptibility to malaria in humans. These include the protective effect of sickle cell trait, alpha-thalassemia’s and glucose-6 phosphate dehydrogenase deficiency, and increased parasite density in children with different expression profiles in genes encoding for proinflammatory molecules, as well as host gene responses HmOX1HSPCB and TNFRSF6 [16]. Molecular methods have contributed enormously to the diagnosis of CNS infections. Identification of specific organism nucleic acid in cerebrospinal fluid or tissue enables rapid diagnosis and treatment [17]. Analysis of bacterial and cell-derived products in the response to infection enables the development of therapeutic agents to counteract harmful effects. Recognition of aberrant genes in chronic persistent infections explains the lack of clearance from the CNS and increases understanding of the pathogenesis. Elucidation of subtle differences in host and organism factors at a molecular level contributes to the understanding of infectious pathogenesis.

Perinatal and Postnatal Infections Chapter 42

Bacterial infections Neonatal bacterial meningitis The risk of meningitis is greatest in the first 30 days of life [5]. Premature babies with low birth weight who need resuscitation, those born to mothers with premature ruptured membranes or with pyrexia are all at increased risk. The most common organisms are Streptococcus agalactiae (group B), E. coli K1, other E. coli such as Citrobacter diversus koseri, Klebsiella enterobacter species, Listeria monocytogenes, Staphylococcus aureus and Staphylococcus epidermidis. The mother’s birth canal is the source for group B streptococci. Neonates are infected via upward spread through the cervix during delivery, or later from the hospital environment or contaminated hands of personnel. In developing countries Klebsiella, E. coli and S. aureus are more common than group B streptococci [18]. In a series from Africa and South Asia, 75% of cases of late onset meningitis (more than 48 hrs) were due to Gram-negative bacilli [4]. E. coli K1 has an acidic capsular polysaccharide, and is acquired from the mother’s birth canal/perineum or from hospital personnel. L. monocytogenes is a Gram-positive facultative anaerobe widely distributed in soil and stagnant water. Human infection in adults and older children is from foodstuffs, particularly soft cheese. Outbreaks are common. There are two patterns of neonatal listeriosis [19]. The early type causes miscarriage or stillbirth, and affects premature infants within five days of birth, due to transplacental spread or ascending infection from the birth canal. Infants have diffuse sepsis, skin pustules and liver granulomas. The CNS may or may not be involved. Mortality is 20% of live births. The late type affects mature infants from 3 to 35 days of birth. Mothers are asymptomatic, and infection is via the birth canal. Infants develop meningitis and septicemia. The mortality is 10%.

Postnatal bacterial meningitis Seventy percent of bacterial meningitis occurs in children under five years of age. Children 4–12 weeks are affected by S. agalactiae, E. coli, L. monocytogenes, H. influenzae B (HiB), S. pneumoniae and Neisseria meningitidis (meningococcus). Children three months to eight years are mainly infected by H. Influenzae, N. meningitidis and S. pneumoniae. This is similar to adults. Meningococcal epidemics and sporadic cases have a seasonal incidence in winter and spring. Epidemics of serogroup A occur in the meningitis belt of Africa and overall account for 75% of cases worldwide, although vaccination has led to a fall in numbers of cases [2]. Elsewhere, serogroups B and C cause the most disease. The incidence of HiB and meningococcal disease is inversely proportional to the presence of serum bactericidal antibodies. Deficiencies of properidin and complement C5 to C8 also increase risk and result in recurrent meningitis. N. meningitidis is a Gram-negative diplococcus, carried in the nasopharynx in 1% of young children, 5% in adolescents and

20% or higher in adults. It causes meningitis and septicemia. H. influenzae is a small Gram-negative coccobacillus. Most invasive H. influenzae disease is due to group B. It causes meningitis, epiglottitis, septicemia, and pneumonia. Symptoms and signs of bacterial meningitis In neonates there may be no localizing signs of meningeal irritation. Signs include pyrexia, listlessness, high-pitched cry, weak suck, irritability, vomiting and diarrhea, respiratory distress, bulging fontanelle and convulsions. Older children may suffer from headache, lethargy, fever, photophobia, neck stiffness, vomiting, and altered consciousness; in septicemia, a characteristic nonblanching purpuric rash occurs in skin, conjunctivae, and mucous membranes. Septicemia with collapse, shock, and multiorgan failure may develop rapidly, and ischemia of the limbs may require amputation. Diagnosis of bacterial meningitis Examination of the cerebrospinal fluid is the most valuable. Newborns with a white blood cell count of greater than 20/υl, greater than five polymorphs, and protein above100 mg/dl should be regarded as at risk for meningitis [7]. Older children have a high polymorphonuclear leukocytosis, reduced or absent glucose and protein greater than 100–500 mg/dl. Gram stain shows organisms in up to 90% of patients when neutrophils are present. Latex agglutination detects the most common causative bacteria; however, a negative result does not exclude bacterial meningitis. Real-time polymerase chain reaction (RT-PCR) is highly sensitive, specific, and rapidly identifies bacterial DNA in cerebrospinal fluid or blood. RT-PCR and multiplex PCR (mPCR)/reverse line blot assays are more sensitive than bacterial culture and can detect pathogens in cerebrospinal fluid samples with negative culture results [17]. Neuroimaging is of limited value but should be performed prior to lumbar puncture if focal neurological signs are present. Pathogenesis of acute bacterial meningitis and septic shock (meningococcal septicemia) Infection is by nasal droplets from a carrier or infected individual. While nasopharyngeal colonization and carrier status is common, meningitis is not. Variation in bacterial genes is important in occasionally producing a genotype capable of causing meningitis following colonization. S. pneumoniae and N. meningitidis possess a polysaccharide-rich capsule that protects both against mucus entrapment and phagocytosis. Bacterial cell adhesion molecule receptors (adhesins) aid bacteria to enter the bloodstream via the respiratory epithelium, as do the pili in N. meningitidis membrane. Organisms multiply and disseminate to the choroid plexuses and leptomeninges where proinflammatory cytokines IL1 and IL8, and TNF-α disrupt the blood brain barrier allowing bacteria and polymorphs to cross. In severe bacteremia, septic shock with a hemorrhagic rash and disseminated intravascular coagulation develops before clinical evidence of meningitis. The speed of onset is alarming

513

Developmental Neuropathology

Figure 42.1 Streptococcus pneumoniae meningitis. Thickened surface leptomeninges. The pyogenic exudate is confined to the leptomeningeal surface. The inflammatory response is predominantly polymorphonuclear neutrophils.

and may cause organ failure. Bacterial toxins, TNF-α and IL1 all contribute to septic shock. High cytokine levels injure endothelium; activated neutrophils generate adhesion molecules localizing leukocytes to the area. With further damage, IL8 and platelet activating factor attract more cells magnifying the response, causing ongoing endothelial injury. Bacterial eicosanoids also activate neutrophils. Simultaneously the compensatory antiinflammatory response syndrome comes into play and soluble cytokine receptor and antagonists reduce cytokine interaction with membrane-bound receptors. Intravascular coagulation in sepsis results from toxin-induced upregulation of tissue factor. Indirect coagulation activation also occurs through the proinflammatory cascade. Lipid A from Gram-negative and lipoteichoic acid from Gram-positive organisms cause release of the coagulation inducers TNF-α and interleukins [20].

Neuropathology of bacterial meningitis In fatal fulminant meningitis with septicemia there are few gross abnormalities in the brain apart from vascular congestion and cerebral swelling. Gram staining of blood from the leptomeninges or the purpuric skin lesions will reveal organisms, reflecting septicemia. Outside the CNS, adrenal hemorrhages may be found (Waterhouse Friedrichsen syndrome) and the viscera are pale and swollen. If death occurs within a few days, there is a purulent exudate covering the brain and spinal cord surface. Intense engorgement or thrombosis of cortical vessels and leptomeningeal sinuses may be seen, especially with pneumococcal meningitis. Microscopically, in the acute phase, the leptomeninges are infiltrated by an inflammatory exudate consisting of neutrophils, fibrin and macrophages (Figure 42.1). Intra- and extracellular organisms can be found on Gram stain. Choroid plexus inflammation may be associated with ependymitis. Subsequent organization leads to numerous ependymal granulations. Thrombosis of inflamed vessels may lead to foci of cortical necrosis. If death occurs later, complications of meningitis are seen. Glial scarring across the aqueduct of Sylvius with forking or obliteration of the lumen may lead to aqueduct stenosis (Figure 42.2) and hydrocephalus. Exudate becomes organized around the cranial nerves (Figure 42.3) and can cause demyelination and subsequent cranial nerve palsies. In listeria infection, in addition to meningitis, rhomboencephalitis with hemorrhagic necroses or small abscesses may be seen in the brain stem. The inflammatory infiltrate often includes plasma cells. Complications of bacterial meningitis There is a high mortality and incidence of neurological complications in bacterial meningitis. In children, about 50% of the survivors have neurological and other sequelae. Seizures, sensoneural deafness, mental restriction and behavioral difficulties are important complications [21–23]. In Africa, mortality rates for

Figure 42.2 Post-meningitic aqueduct stenosis. Section of midbrain. Aqueduct obliteration on right, normal aged matched control on left.

514

Perinatal and Postnatal Infections Chapter 42

source of infection is vertebral osteomyelitis. In 50% of patients a source is not found. Clinical symptoms are fever and back pain, with difficulty walking or other neurological signs. MRI is the investigation of choice but may be negative. Technetium99 scans of bone may be required. The lumbar and thoracic regions are most affected. Blood culture yields identical organisms to cultures of operative specimens. Pediatric cases have a better outcome than adult [31]. S. aureus is the main pathogen, but group B streptococcus, E. coli, Salmonella enteriditis and Aspergillus flavus or Candida in immune-suppressed patients can also be responsible. Rarely, spinal epidural abscess occurs in neonates [32]. Figure 42.3 Post-meningitic scarring. Fibrous thickening of basal meninges and distortion of brainstem. Child aged two years with past history of bacterial meningitis.

pneumococcal, HiB and meningococcal meningitis were 45%, 29%, and 8%, respectively [2]. In Nigeria, 21% of cases of profound deafness of known cause were due to meningitis, often HiB [24]. Neuronal damage may result from neuronal apoptosis and necrosis due to combined effects of edema, vasculitis and inflammatory cytokines and the effects of bacterial toxins. Hippocampal neuronal apoptosis was seen in humans dying from bacterial meningitis [25]. Experimental studies on mouse brain infected with S. pneumoniae showed that caspases mRNA, which are activators of apoptosis were elevated as early as six hours following infection [26]. Prevention In the United States, vaccination for HiB has succeeded in reducing invasive disease from 41 cases per 100 000 in 1987 to 1.3 per 100 000 in 1997 [27,28]. Vaccination programs for N. meningitidis types A and C in many countries have been successful in reducing meningococcal diseases.

Intracranial epidural abscess Intracranial epidural abscess is rare in young children [29]. It may result from spread of infection in the frontal and nasal sinuses, from mastoiditis, following trauma or where a skull defect exists as a result of neurosurgery, or exceptionally following osteomyelitis complicating fetal scalp monitoring [30]. The responsible organisms are usually streptococci and staphylococci but mixed infections with Gram-negative organisms are common [7]. The abscess expands to separate the dura from the skull and often spreads through the dura to cause subdural empyema. Clinically fever, lethargy, and skull tenderness may be present. The cerebrospinal fluid is usually normal but MRI outlines the abscess even in the early stages. Spinal epidural abscesses Spinal epidural abscesses are rare in children, with an incidence of 0.6 per 10 000 hospital admissions [31]. The most common

Brain abscess About 25% of all brain abscesses occur in children under 15 years of age [33] and are serious infections. The peak incidence is four to seven years [34]. Neonates are rarely affected, but the prognosis in neonates is poor. There is usually an associated Gramnegative bacterial meningitis or septicemia. In infants, brain abscess is associated with CNS malformations such as a congenital dermal sinus, encephalocele or myelomeningocele, which facilitate entry of skin organisms, especially S. aureus. Other predisposing factors in children are cyanotic congenital heart diseases, immune deficiencies such as chronic granulomatous disease and myeloperoxidase deficiency, trauma (especially fracture across a sinus), sinusitis, otitis, dental sepsis, HIV infection, and chemotherapy-induced neutropenia. Contiguous spread of organisms from a local infection in the sinuses or middle ear usually causes a single abscess, the site of which relates to the source e.g. frontal lobe abscesses are consequent to dental sepsis, and temporal lobe (Figure 42.4a) or cerebellar (Figure 42.4b) abscess may complicate mastoiditis. The most common causative agents in contiguous spread are aerobic and anaerobic streptococci, bacteroides, Enterobacteriaceae, haemophilus and staphylococci. Abscesses arising from blood borne spread in endocarditis, respiratory infection, or congenital heart disease tend to be multiple, and often distributed in the course of the middle cerebral artery. They are due to septic emboli, which may be paradoxical in cyanotic congenital heart diseases with right to left shunts. Anaerobic streptococci, S. viridans, haemophilus, nocardia, and actinomyces are most often responsible and in immunocompromised children, toxoplasma, fungi, nocardia, and Enterobacteriaceae are most common. Development of the bacterial abscess cavity goes through a number of stages, according to experimental models [35] starting with cerebritis following local vascular injury after inoculation of organisms. Neutrophil infiltration, perivascular fibrinous exudates and surrounding edema cause raised intracranial pressure. By day 4, the central part of the abscess has become necrotic and purulent, and at the margins, lymphocytes and macrophages are present and edema is marked. A capsule made up of granulation tissue is formed by day 10 and collagen fibers are laid down so that the capsule thickens and by day 14 may be separated from

515

Developmental Neuropathology (a)

(b)

Figure 42.4 Brain abscess. (a) Temporal lobe abscess. Granulation tissue capsule around central cavity containing pus. (b) Cerebellar abscess on the right, which spread from an ear infection.

the brain. The adjacent white matter shows astrocyte proliferation, edema and perivascular dense lymphocytic inflammation. For a detailed description of these stage-specific characteristic see Deckert [10]. In postmortem human cases, the site of origin of infection may be adherent to the brain surface overlying the abscess [36]. Signs and symptoms depend on the virulence of the organism, the strength of host response, whether there is a single or multiple abscesses, the site of the lesion and the primary site of infection (e.g., frontal sinusitis, bacterial endocarditis). In a large series of children, symptom duration ranged from 3 to 120 days, with a median of 13 days [37]. Headache, fever, and focal neurological signs are the classical triad of symptoms, and fever and blood leukocytosis, white blood cell count greater than 10 000/mm3 are more often present in children than adults. Nausea and vomiting, seizures, nuchal rigidity, and papilledema may also be found. In brainstem abscesses, facial palsy and dysphagia may additionally be present. Cerebellar abscess characteristically presents with ataxia, vomiting and nystagmus.

516

Blood culture is often negative. Neuroimaging is the most useful diagnostic aid and MRI with gadolinium enhancement outlines the central necrotic abscess, the surrounding rim and the perilesional edema. MRI is useful in monitoring response to therapy. Lumbar puncture is contraindicated due to raised intracranial pressure. Cerebrospinal fluid data from older series showed reduced glucose, elevated protein, and mononuclear pleocytosis but the fluid may be normal. Biopsy is rarely required, except in atypical cases in the immune suppressed individual.

Tuberculous meningitis Tuberculous meningitis is the most common tuberculosis in children. Tuberculomas and epidural paraspinal tuberculosis also occur. CNS infection is virtually always secondary to hematogenous spread to the meninges (Rich focus) from the lung. Infection is acquired by inhalation of droplets containing M. tuberculosis from an infected human. M. bovis from cattle is now uncommon, and is acquired by drinking unpasteurized cow’s milk, with primary lesions in the tonsils. Atypical mycobacteria (M. kansasii or M. avium intracellulare) are most commonly seen in the setting of HIV infection. In the United States, cases of tuberculosis have fallen by 39% since 1992, but the case rate among foreign-born persons per 100 000 population rose [38]. With increasing numbers of HIV-infected children worldwide, the population at risk for tuberculosis is expanding. Symptoms in children are subacute, develop over days, and include headache (rare in children under three years), altered behavior, nausea, vomiting, and seizures. Meningismus and fever may not be present. The most common focal sign is a sixth cranial nerve palsy. Choroidal tubercles near the optic disc are present in 50% of patients with and 10% without miliary disease [39,40]. Early diagnosis is important, as examination of the cerebrospinal fluid and radiology are often inadequate. The cerebrospinal fluid is clear or cloudy, and forms a cobweb-like surface pellicle in which organisms can be identified by smear or culture. If spinal block is present, the opening pressure may be artificially low, even with hydrocephalus [41]. Glucose is reduced but not as low as in bacterial meningitis. The cell count and content varies with disease stage; in the first few days, polymorphs and lymphocytes can be equal in numbers; later, lymphocytes predominate; the total cell count is 50–200/mm3 but may be over 1000. Protein is elevated, usually 150–200 mg/dl; as the disease progresses, protein often rises in association with spinal block. Smears are often negative for acid-fast bacilli, even using auramine rhodamine fluorescence. Positive culture is diagnostic, but several samples may be required and the delay limits its clinical usefulness. Rapid confirmation by PCR on cerebrospinal fluid is useful, but may be negative in tuberculoma [42]. PCR can also be used on biopsy tissue. Treatment is usually instigated prior to organism identification. In tuberculosis meningitis, the brain shows a whitish exudate on the undersurface of the brain (Figure 42.5) surrounding the

Perinatal and Postnatal Infections Chapter 42

(a)

Figure 42.5 Tuberculous meningitis. A thick fibrous exudate in the basal meninges surrounding the optic nerves and Circle of Willis. Child aged three years with Mycobacterium bovis meningitis. (b)

circle of Willis, and cranial nerves. Small tubercles may sometimes be seen around the Sylvian fissures. A similar exudate is seen around the spinal cord. In chronic or treated cases, the exudate is more fibrous. The brain often shows hydrocephalus and necrotic infarcts in the basal ganglia due to accompanying vasculitis or thrombosis in the middle cerebral arteries. On microscopy, the inflammation is “exudative” with little granulomatous reaction (Figure 42.6a), or may show caseating or noncaseating epithelioid granulomas with Langhans multinucleated giant cells (Figure 42.6b). Inflammation extends into the blood vessel walls (Figure 42.6a) and induces fibrinoid necrosis or endarteritis obliterans. In untreated cases, organisms can be seen on Ziehl–Nielsen stain.

Tuberculoma Tuberculomas accounted for 10% of pediatric intracranial space occupying lesions in one series but are declining in incidence except in endemic regions [43,44]. The pathogenesis is believed to be due to the host granulomatous reaction around organisms that reach the brain from the bloodstream during primary infection. The inflammation expands to form a mass rather than rupturing into the subarachnoid space. Symptoms depend on the site and are those of a space-occupying lesion. Systemic signs of tuberculosis may be absent. Neuroimaging is not diagnostic but can show resolution of lesions. Tuberculomas may be single or multiple and can occur in association with tuberculosis meningitis. The most common site is the cerebellum but cerebral hemispheres, brain stem, and spinal cord [45] may all be affected. The gross lesions vary from several millimeters to centimeters and are well-circumscribed and round. Grossly, the center is yellow and firm or may be calcified. Microscopically, the central region shows caseous necrosis in which organisms are identifiable with surrounding epithelioid histiocytes, Langhans giant cells, lymphocytes and fibroblasts.

Figure 42.6 Tuberculous meningitis. (a) Dense exudative inflammation in the meninges surrounding and involving blood vessels. (b) Granulomatous inflammation with epithelioid cells and Langhans multinucleated giant cell.

Epidural spinal tuberculous abscesses The abscesses spread from infection in the adjacent vertebral body and are most common in the thoracic region. Although more frequently seen in the past, they still occur and may give rise to compression of the spinal cord or a paravertebral “cold abscess”. Borrelia burgdorfi infection Borrelia burgdorfi is a spirochete and causes Lyme disease, transmitted by ticks and common in the eastern United States and Europe. It may cause meningitis, cranial neuritis and radiculoneuritis. There are few neuropathological descriptions in children but perivascular inflammation and multifocal demyelination can occur. Rickettsia ricketsii infection Rickettsia ricketsii is a spirochete and causes “Rocky Mountain spotted fever” in the United States and Canada. Fifty percent of

517

Developmental Neuropathology cases occur in patients under 19 years. Infection is acquired in humans from tick bites. About one day after the bite, a high fever, headache, nausea, and myalgia occur. Conjunctivitis and periorbital edema may be present. The hemorrhagic rash that gives the disease its name appears between three and five days. Paired antibody sera in the acute and convalescent phase, PCR and culture confirm the diagnosis. A meningoencephalitis may occur.

Viral infections Viral meningitis and meningoencephalitis in the perinatal period are the result of prenatal, intranatal or postnatal infection. Most are due to herpes simplex (HSV types 1 and 2) or enteroviruses. The CNS may be involved in isolation but more often as part of a disseminated infection. Prenatal infections are dealt with in Chapter 41. In addition to the usual environmental exposure, babies are exposed to viruses during vaginal delivery and in breast milk. The presence of passive maternal antibodies to viruses is protective but not preventive.

Herpes simplex infections HSV types 1 and 2 are alpha herpes-enveloped double-stranded DNA viruses, with a worldwide distribution. They have a short reproductive cycle that rapidly causes cell lysis and can also remain latent in sensory ganglia. In humans, HSV1 causes cold sores of the orofacial skin, follicular conjunctivitis, corneal ulcers, and necrotizing encephalitis in neonates and adults, and genital infections. HSV2 causes genital infections, aseptic meningitis, and necrotizing encephalitis in neonates. Neonatal herpes virus infection Neonatal HSV infection is defined as infection in a newborn within 28 days after birth. Some of the epidemiological data are reviewed in Chapter 41 and additional information is well presented by Corey and Wald [46]. The incidence of neonatal herpes varies between 8 and 60/100 000 in the USA [46] and 1.6/100 000 in the UK [47,48]. Some 50–80% of cases of neonatal HSV infection result from women who acquire genital HSV1 or 2 infection near term. Neonatal herpes develops in less than 1% of infants delivered vaginally to women with HSV2 shedding at term [49], suggesting a role for transplacental antibodies in abrogating the risk of infection. About 75% of cases are due to HSV2, which causes encephalitis either in isolation or more often as part of systemic disease. Prematurity, primary maternal genital herpes, absence of maternal antibodies, and prolonged ruptured membranes are all risk factors [50]. Placement of fetal scalp monitors may also play a role [51]. Neonates usually present within four weeks of birth. Nearly half of the reported cases present with localized skin, eye, and/or mouth vesicles. Systemic therapy is required to prevent further progression. Encephalitis, which manifests by irritability, lethargy, poor feeding, bulging fontanelle, and seizures, with or without orofacial lesions, accounts for one-third of cases. The

518

morbidity is higher among infants with HSV2 infection and may include developmental delay, epilepsy, blindness, and cognitive disabilities. Acyclovir therapy has substantially improved survival; however, neonates with CNS HSV2 infection still have high rates of developmental problems with moderate to severe neurologic abnormalities [52]. The corollary is that infants with HSV1 infection usually have a better prognosis [53,54]. Disseminated HSV with skin vesicles, keratoconjunctivitis, CNS and visceral involvement of liver and lungs indicates hematogenous dissemination. In contrast, restricted encephalitis is believed to be the result of spread along nerves. The risk of death from disseminated neonatal HSV infection is high (30%), even with antiviral therapy. In those who recover, even with high-dose intravenous acyclovir therapy, considerable disabilities are present, 50% have neurologic sequelae [52]. Diagnosis is confirmed by PCR of the cerebrospinal fluid or blood, and culture. Fatal cases show a swollen congested brain and microscopic diffuse involvement of gray and white matter in the cerebral hemispheres, brain stem and cerebellum. Necrotic or hemorrhagic lesions with macrophages, lymphocytes and viral intranuclear inclusions and antigen are seen. Cystic encephalomalacia may be found in cases with prolonged survival. Postnatal herpes virus infection This infection is usually due to HSV1. There is no seasonal incidence and the virus occurs worldwide. The primary infection is often asymptomatic in infants or children under five years; some children develop extensive orolabial or pharyngeal ulcers. Several studies show that over 90% of children developed HSV antibodies by 15 years. HSV encephalitis is uncommon in young children. The estimated incidence of all cases is 1 in 250 000– 500 000 per year [53] and the peak incidence in children is between six months and three years. Following absorption and replication, virus spreads along axons to sensory ganglia where latent infection occurs. Reactivation of viral replication results in virus spreading anterogradely along the nerve to the skin and mucosa and vesicles erupt. HSV1 gains entry into the brain via the olfactory nerves to the frontal and temporal lobes or along the trigeminal nerves to the brainstem. Reactivation of latent intracerebral virus appears less likely [54]. Clinical symptoms of HSV1 encephalitis are variable and may be acute or subacute. Fever, headache, altered mental status, and convulsions are the most common presenting symptoms. Electroencephalography shows focal spike and slow wave abnormalities; however, they are non-diagnostic; cerebrospinal fluid usually shows an increase in red cells, moderate pleocytosis, predominantly lymphocytes and elevated protein with normal glucose. PCR of cerebrospinal fluid is the most rapid and reliable assay. MRI shows edema and hemorrhagic necrosis, often located in the temporal lobe. Brain biopsy is now very rarely undertaken for diagnosis, but may still be performed in atypical subacute cases.

Perinatal and Postnatal Infections Chapter 42

relatives had not developed HSE. These deficiencies represent underlying genetic etiologies for HSE.

Figure 42.7 Herpes simplex virus encephalitis. Section of cerebral cortex showing several cells with eosinophilic intranuclear viral inclusions. Hematoxylin and eosin.

In acute fatal cases, the brain is swollen and may show medial temporal herniation. Hemorrhagic necrosis of one or both temporal lobes is characteristic, with some cases involving the posterior frontal lobes. In chronic cases, there may be shrunken temporal lobes and cystic cavitation surrounded by brown discoloration. Microscopically in acute cases there is little inflammation in the brain or leptomeninges but numerous eosinophilic viral inclusions are seen in neurons (Figure 42.7), and immunocytochemistry reveals viral antigen in neurons, endothelial, and glial cells. In established infection, cellular infiltration is dense and hemorrhagic necrosis of the affected tissue is evident with numerous macrophages. There are often microglial nodules, as well as neuronophagia. The inflammation at the edges of necrotic lesions has a perivascular distribution with lymphocytic cuffs. Immunopositivity for viral antigen remains for several weeks but may be diminished by treatment with acyclovir [54]. Electron microscopy of infected cells shows a characteristic icosahedral structure. The treatment of choice is acyclovir. Genetic factors in herpes encephalitis of childhood(HSE) show that autosomal-recessive UNC-93B and TLR3 deficiencies and autosomal-dominant TLR3 and TNF receptor-associated factor (TRAF)3 deficiencies underlie HSE in some children [15, 55–58]. The autosomal-recessive form of the disease was found to be due to a homozygous nonsense mutation that resulted in a complete absence of the Toll/IL-IR (TIR) domain-containing adaptor inducing IFN-B(beta) (TRIF) protein. Both the TLR3and the TRIF-dependent TLR4 signaling pathways were abolished. The autosomal-dominant form of disease was found to be due to a heterozygous missense mutation, resulting in a dysfunctional protein. In this form of the disease, the TLR3 signaling pathway was impaired, whereas the TRIF-dependent TLR4 pathway was unaffected. Patients who are TRIF-deficient with HSE have suffered from no other infections and clinical penetrance was incomplete, as some HSV1-infected TRIF-deficient

Enteroviral infections During seasonal outbreaks, 3% of pregnant women excrete enteroviruses at term. Neonates are infected from the mother, infection by personnel in nurseries also occurs. Echoviruses and group B Coxsackie viruses are the most common neonatal viruses [50,59,60]. Clinical signs may be nonspecific and include lethargy, irritability, poor feeding, apnea, and convulsions. Neonates are at greatest risk for morbidity and mortality; in fatal enterovirus infection death is usually associated with hepatic disease (Echoviruses) or myocarditis (Coxsackie) [60]. Microscopic examination of the cerebral hemispheres, brain stem and cerebellum shows a lymphocytic leptomeningitis and inflammatory lymphocytic foci in the cerebrum, brainstem, cerebellum and spinal cord. Viral antigen is less in the brain than in systemic organs [54] and the diagnosis is often established by positive culture from liver or heart. Aseptic meningitis Aseptic meningitis is a benign inflammatory meningeal illness with cerebrospinal fluid pleocytosis and negative microscopy and routine cultures at 48 hours. Enteroviruses, arbovirus, and herpesviruses account for the overwhelming majority. Mumps, lymphocytic choriomeningitis virus and HIV are less common causes [59]. Other viruses, bacteria and noninfective agents can also be responsible. Aseptic meningitis is very common in young children. In about 85–95% of cases in which a virus is identified, enteroviruses are the cause, followed by HSV (HSV2). In tropical countries, infection occurs throughout the year but in the Western world there is a seasonal variation, with the majority of cases occurring in summer or autumn. Infection is by the fecal/oral route and overcrowding is a risk for outbreaks. Enteroviruses, poliovirus, Coxsackie A and B, and echovirus have many different serotypes. All are small icosahedral nonenveloped picornaviruses with a single strand of positive sense RNA. Clinical symptoms include headache, photophobia, neck stiffness and fever. Cerebrospinal fluid shows an initial mixed polymorphonuclear and lymphocytic infiltrate, later becoming lymphocytic. Virus may be isolated from cerebrospinal fluid, blood, throat, and feces. Neutralizing and complement fixation antibodies appear about day 10 after infection. RT-PCR of cerebrospinal fluid is the most useful and highly sensitive test [59]. As the disease is usually mild, there are few pathological studies; these show slight lymphocytic inflammation in the leptomeninges and around superficial cortical blood vessels and in the choroid plexus. Enterovirus polioencephalomyelitis in older children “Polio” indicates gray matter involvement. All of the enteroviruses, poliovirus, echoviruses, Coxsackie A and B,

519

Developmental Neuropathology and the numbered enteroviruses, in addition to causing aseptic meningitis, can cause an encephalitis predominantly involving the gray matter of the brain (polioencephalitis) and spinal cord (poliomyelitis). Infection is by fecal/oral route; poor hygiene and overcrowding predispose to outbreaks. Wild-type virus causes infection in unvaccinated regions. Polio incidence has dropped more than 99% since the launch of global polio eradication efforts in 1988. According to global polio surveillance data from 2016, five wild poliovirus cases were reported in Pakistan and one wild poliovirus case was reported in Afghanistan in 2016. In 2015, 74 cases of wild poliovirus were reported: 54 from Pakistan and 20 from Afghanistan [61]. Poliovirus is a member of the enterovirus subgroup, family Picornaviridae. Enteroviruses are transient inhabitants of the gastrointestinal tract, and are stable at acid pH. Picornaviruses are small, ether-insensitive viruses with an RNA genome. Immunity to one of the three polio serotypes (P1, P2, P3) does not, however, confer significant immunity to the other serotypes. Virus reaches the CNS by neural pathways and the bloodstream, following replication in the Peyer patches of the gut. Following attachment, the virion is coated by cell membrane and viral protein and RNA is released, binds to ribosomes and begins viral protein production within six hours of infection. In addition to systemic general prodromal symptoms of fever and malaise, muscle pain and stiffness is present, followed by flaccid paralysis or bulbar signs, which may persist for weeks but then recover. Most children with poliovirus infection recover muscle strength. Paralytic polio may be predominantly spinal, bulbar or a combination of both. Paralysis present 12 months after onset tends to be permanent. The clinical diagnosis is usually confirmed by virus isolation in stools, or less commonly in blood or cerebrospinal fluid. RTPCR and viral genome sequencing determine if the virus is wild type or vaccine related. Rising antibody titers in serum may also be helpful, but depend on the stage of infection. In acute disease, there is dense inflammation of the leptomeninges by neutrophils, lymphocytes and macrophages. The spinal cord gray matter, brainstem nuclei and precentral gyrus are most often affected. Microglial “stars” and lymphocytes surround anterior horn cells and brain stem neurons and there are perivascular lymphocyte cuffs. Blood vessels are congested and there may be small hemorrhages. In chronic cases, there is loss of motor neurons, together with gliosis. Clusters of microglial cells remain. There is secondary atrophy of affected motor pathways and muscles. Chronic enterovirus encephalitis may occur in children with X-linked or combined immunodeficiency.

HIV infection HIV types 1 and 2, the causative agents of acquired immunodeficiency syndrome (AIDS), are lentiviruses of the Retroviridae family. They are similar to the simian immunodeficiency virus (SIV) in primates. Human infection may have occurred via cross-species transmission of SIVs. HIV1 and 2 cause

520

similar disease and both affect children. Transmissibility is less and disease progression is slower in HIV2. Childhood AIDS has been recognized in children for over 30 years [62]. An estimated 3.2 million children were living with HIV at the end of 2013, mostly in sub-Saharan Africa. There are four routes of infection: i) sexual; ii) via blood or blood products; iii) parenteral exposure to contaminated instruments; and iv) vertical infection from mother to child. Children are infected by all four routes but perinatal infection is the most common. Prior to antiretroviral treatment, transmission rates were between 13% and 30% [63], but with efficient treatment rates fall to 2%. Over 75% of women with AIDS are in the reproductive age group. The majority of childhood HIV infections occur through vertical transmission across the placenta, via direct transmission intrapartum, or postpartum through breast milk. T-lymphocytes, monocytes and macrophages express CD4, the major receptor for HIV1 and 2. CD4 T-lymphocytes are the main source of viral production and spread. The CNS is probably infected at an early stage, at the time of initial infection coinciding with aseptic meningitis. Following a variable asymptomatic interval, viral replication and viral loads increase and CD4 T-lymphocytes are destroyed. When counts fall lower than 200, opportunistic infections and symptoms of HIV-specific disorders develop. A feature of HIV neuropathology is the coexistence of simultaneous multiple pathologies, sometimes in the same site. This reflects both the synergistic interaction of HIV with other infectious agents, and the migration of HIV-infected macrophages to sites of disease. Pathologists should consider more than one pathological process when examining biopsies or brains from children with AIDS. Neurological manifestations are found in most children with AIDS. The most frequent HIV-specific clinical disease is HIV encephalopathy, which occurs in the later stages of immune suppression. Without treatment, it has a progressive downhill course. The main features are delayed developmental milestones, apathy, spastic quadriparesis, and convulsions. Radiological investigations show calcification of the basal ganglia and brain atrophy [64]. In pediatric AIDS, the brain may show lesions due to direct CNS HIV1 infection, to opportunistic infections, or nonspecific changes related to severe systemic disease, complications of wasting or ischemia [65]. Macroscopic examination may be normal, may show atrophy, or signs of opportunistic infections. In HIV encephalitis and myelitis, there are perivascular inflammatory cell infiltrates, microglial nodules in both gray and white matter and characteristic multinucleated giant cells (Figure 42.8), which express HIV antigens and contain the virus. HIV leukoencephalopathy is also often present, the white matter is poorly myelinated and contains HIV-containing macrophages or multinucleated giant cells. Basal ganglia mineralization with calcium and iron deposits is prominent in the majority of children with AIDS but is not specific. Corticospinal tract degeneration without local evidence of HIV is found in children with

Perinatal and Postnatal Infections Chapter 42

Figure 42.8 HIV encephalitis. Characteristic uninucleate (thin arrow) and multinucleated giant cells (thick arrow) adjacent to vessel in cerebral white matter.

AIDS, regarded as due to secondary Wallerian degeneration [66]. Vacuolar myelopathy of the dorsal and lateral columns similar to subacute combined degeneration of the cord is rare in pediatric cases [66,67]. The mechanisms of neuronal dysfunction in HIV encephalitis/leukoencephalopathy are not fully understood, but direct injury from HIV neurotoxic receptor-blocking proteins and indirect toxicity from macrophage/microglial cell-derived inflammatory mediators affecting glutamate-related excitotoxicity and oxidative stress are involved [68]. Since the onset of highly active antiretroviral treatment (HAART), the incidence of HIV- related neuropathology has changed [69]. Opportunistic infections in childhood AIDS are less common than in adults and have reduced since HAART treatment. In many countries affected by AIDS, HAART is not available and the spectrum of neuropathology is closer to that described in the early series of AIDS [70]. Vascular lesions in children with AIDS include ischemic infarcts, Leigh-like changes, and endarteritis obliterans. Primary CNS lymphomas in AIDS are most often high-grade diffuse non-Hodgkin’s B-cell lymphomas, which occur late in the disease when immunosuppression is profound. They are regarded as due to uncontrolled Epstein–Barr virus-driven proliferation of B lymphocytes, in the absence of T cells. The Epstein–Barr virus genome is consistently expressed in these neoplasms, and while the overall reported incidence of primary CNS lymphomas is 1000 times greater in HIV-infected individuals than in the general population, the incidence has decreased since the era of HAART.

Cytomegalovirus infection Cytomegalovirus is a double-stranded DNA herpes virus. Human infection is usually acquired via the oral or respiratory tracts, by blood transfusion or organ transplant. Infection

in immune competent hosts causes hepatitis, pneumonia and lymphadenopathy. Neurological symptoms are rare. The virus remains latent in bone marrow. Cytomegalovirus is the most common CNS opportunistic infection in children with AIDS [64,67,71], and is acquired in the perinatal period or postnatally [72]. It is also a serious complication of other immune-suppressive states, including organ transplants. Clinically, cytomegalovirus may cause encephalopathy, but most pathologically-studied cases also had HIV encephalopathy. Infection is associated with more rapid progression of HIV disease. Serologic methods are not reliable indicators of cytomegalovirus infection in immunocompromised hosts. PCR detection of cytomegalovirus DNA in plasma correlates with clinical severity of infection in AIDS. Cytomegalovirus encephalitis in immune suppression is usually associated with systemic cytomegalovirus infection in other organs. Macroscopically, in the brain, there may be periventricular necrotic lesions or ependymitis, or it may appear normal. Microscopically, there are diffuse nodular lesions in the cerebral hemispheres and periventricular regions, suggesting reactivation of latent infection [73] or single scattered cytomegalic cells sometimes surrounded by microglial nodules. In general, there is a lack of inflammatory response. Cytomegalovirus myelitis and radiculitis also occurs. Infected cells have typical “owl’s eye” intranuclear inclusions (Figure 42.9a). Immunocytochemistry using antibody directed against early- expressed antigen helps to identify infected cells (Figure 42.9b) and only cells in active phase of viral replication will be immunopositive.

Progressive multifocal leukoencephalopathy Progressive multifocal leukoencephalopathy is a fatal demyelinating disease of the CNS caused by polyomaviruses JC and BK that infect oligodendrocytes. Children with inherited immune deficiency disease and AIDS can develop the disease. There is multifocal demyelination and, microscopically, bizarre astrocytes and inclusion-bearing oligodendrocytes are seen in white matter. The virus can be identified by PCR in cerebrospinal fluid, in situ hybridization in CNS tissue, or by electron microscopy. Measles infections Measles virus is highly contagious. Humans and primates are the hosts, but only human-to-human infection occurs. Measles virus causes acute and subacute infections. Infants and young children are most at risk, and in outbreaks 40% of cases occur in children under 16 months [74]. Measles is a single-stranded RNA antisense nonsegmented enveloped Morbilli virus of the Paramyxoviridae family. It contains six major structural proteins. Neutralizing antibodies are produced to the Hemagglutinin (H) and fusion (F) proteins and the cell surface glycoproteins and give lifelong immunity. The matrix protein (M) on the inner aspect of the envelope is important for viral assembly [75].

521

Developmental Neuropathology (a)

(b)

Figure 42.9 Cytomegalovirus (CMV) ependymitis. (a) Numerous cytomegalic ependymal and periventricular cells, some with perinuclear clearing “owl’s eye” appearance (thin arrows). One enlarged cell has two inclusions (thick arrow). Hematoxylin and eosin. (b) Immunocytochemistry for CMV. Many, but not all, cytomegalic cells in the ependymal are immunopositive for CMV early antigen. Alkaline phospatase.

Infection is acquired when viral-laden fomites infect conjunctival or respiratory cells followed by viral replication and viremia. Fever and upper respiratory tract symptoms appear 10–14 days following initial infection. A morbilliform skin rash appears a few days later. Recovery is the normal outcome. Acute measles may involve the CNS in two ways; aseptic meningitis that has a benign outcome, or acute disseminated encephalomyelitis, which occurs in 1 in 1000 cases of measles infection and 1 in 2 000 000 following measles vaccination [76]. Neurological sequelae may be severe in 25% of cases and death rate was 1 per 10 000 reported cases of measles infection [77]. Successful vaccination programs using live attenuated virus have been instituted in many countries, but measles in some countries has re-emerged following unproven concerns about complications of the triple vaccine [78]. Subacute sclerosing panencephalitis Subacute sclerosing panencephalitis is a rare subacute or chronic, slowly progressive, measles infection of the CNS in children with an onset 5–10 years after initial infection. Some cases occur following vaccination. Incidence is one case per million population per year [79]. There is a 16-fold increased risk in children whose primary infection occurs at less than two years of age, compared with over five years. Persistence of maternal antibody may promote the development of mutant clones. Vaccination reduces the risk 10–20-fold [80]. Despite occasional reports [81], proof of direct CNS infection in acute measles is lacking, although EEG abnormalities and cerebrospinal fluid lymphocytosis suggest that virus accesses the CNS [82]. In subacute sclerosing panencephalitis, sequence analysis of measles virus genomes shows mutations throughout the genome; infected neurons and oligodendroglia contain mutated viral genome and truncated genome with deletions. These result in absence of the matrix protein

522

M on the inner aspect of the envelope; the hemagglutinin protein H, responsible for binding to the cell surface receptor and the fusion protein F, are also altered. In subacute sclerosing panencephalitis, measles is unable to assemble infectious virus or bud normally. Cell to cell transfer is possible, so virus spreads and invokes high antibody levels. Demyelination may be due to viral or cytokine-induced oligodendroglial or myelin damage, or other immune response against oligodendrocytes or myelin [83]. There is behavioral disturbance, progressive intellectual impairment, myoclonus and chorioretinitis. Drop attacks may occur. Eventually mutism, ataxia spasticity or choreoathetosis develop. Median survival is 1.8 years but short fulminant cases and prolonged survival for many years are recorded. Characteristic slow waves are found on EEG. Measles virus antibody titer is increased in cerebrospinal fluid as compared with serum (normal ratio 1 : 200). Measles RNA can be identified in fresh frozen biopsy or postmortem brain tissue by immunofluorescence using monoclonal antibody to the antinucleocapsid protein (N) and can be detected by sequencing the RT-PCR amplicon from fresh frozen brain. Autopsied brain may be macroscopically normal, or in prolonged survival may show atrophy. The white matter is demyelinated and gliotic. Microscopic examination shows infiltration of the leptomeninges, gray and white matter by lymphocytes, often arranged in a perivascular distribution. Microglial cells are proliferated and intranuclear or cytoplasmic eosinophilic viral inclusions are seen, mainly in neurons and oligodendroglia (Figure 42.10a). Immunocytochemistry for measles antigen enables infected cells to be easily seen (Figure 42.10b). Astrocytic gliosis is severe in longstanding cases and neuronal loss with taupositive neurofibrillary tangles similar to Alzheimer disease may be found. A fulminant case from 2003, with both neuronal and oligodendrocyte immunostaining of measles antigen, showed

Perinatal and Postnatal Infections Chapter 42 (a)

(b)

Figure 42.10 Subacute sclerosing panencephalitis. (a) Eosinophilic inclusions in multiple cells (arrows) surrounded by a few microglial cells. Hematoxylin and eosin. Girl aged 16 years. (b) Immunocytochemistry for measles antigen is concentrated in inclusions (arrows) and nuclei.

that the measles virus genotype was D7, possibly one of the wild types of measles virus in the UK in the 1980s [84]. Measles inclusion body encephalitis Subacute measles encephalitis develops in immunocompromised hosts especially those with congenital or acquired Tlymphocyte deficiency including HIV-infected, cancer, and chemotherapy patients. Most are children with leukemia or lymphoma. Following initial recovery from the acute illness, seizures, confusion, altered consciousness, and visual defects develop one to seven months post-infection. The EEG changes are non-specific. Cerebrospinal fluid is generally normal. Most cases are fatal within weeks of onset. Gross examination of the brain shows softened white matter, but microscopic examination shows numerous eosinophilic viral nuclear inclusion bodies in neurons and glial cells (Figure 42.11). Cytoplasmic inclusions may also be found. There is mild microglial or inflammatory response and reactive astrocytosis. Molecular analysis of virus isolated from CNS shows mutations of M, H and F proteins similar to subacute sclerosing panencephalitis [82]. Measles vaccination is contraindicated in immunosuppressed children, except those with HIV infection, since their morbidity and mortality from natural measles is very high [75].

Progressive rubella panencephalitis Progressive rubella panencephalitis is a very rare entity that has been described following congenital [85] and acquired rubella infection [86] in childhood. Symptoms occur years after infection and dementia develops similar to subacute sclerosing panencephalitis. Age at onset is 8–20 years. There is no evidence of immune disorder. High levels of antibody to rubella

Figure 42.11 Measles inclusion body encephalitis. Almost every cell shows an intranuclear eosinophilic inclusion.

are present in serum and cerebrospinal fluid. Neuropathological examination shows white matter loss of myelin and perivascular lymphocytic inflammation, marked cerebellar atrophy and mineralization of basal ganglia vessels. Viral inclusion bodies are not seen and definite evidence of persistent rubella infection in the CNS is lacking.

Arbovirus infections Arthropod-borne RNA viruses of the families Flaviviridiae, Togaviride, Bunyaviridae and Reoviridae cause acute hemorrhagic encephalitis and have a specific geographic distribution. In North America, there is a seasonal incidence in late summer and autumn. Eastern equine encephalitis in United States,

523

Developmental Neuropathology Central and South America is a special risk for children, who develop an acute encephalitic syndrome with convulsions one to two weeks after infection. Mortality is 75%. The brain shows panencephalitis, with neuronophagia, microglial nodules and perivascular inflammation sometimes with thrombosis. Virus is identified by fluorescence in situ hybridization [87]. Similar pathology occurs in western and Venezuelan equine encephalitis, and Japanese B encephalitis. West Nile virus is most commonly transmitted to humans by mosquitoes that have bitten infected birds or animals. Infection is usually asymptomatic but approximately 1% of infected people develop encephalitis. Older people are more at risk than children.

Rabies infection Rabies is an enveloped single stranded RNA rhabdovirus. Most cases occur in Asia. Humans are infected by a bite or contact with saliva from an infected animal, most often dogs, foxes, skunks, and bats. The virus replicates locally and reaches the CNS via nerve pathways. The incubation period varies from a week to months. Children have a shorter incubation period than adults. Eighty percent of cases have “furious” rabies; patients are agitated, hydrophobic and develop autonomic signs including hypersalivation. In “dumb” rabies there is ascending paralysis. The mortality rate without treatment is 100%. In the brain, there is neuronophagia and characteristic intraneuronal cytoplasmic eosinophilic Negri bodies are found in the Purkinje cells, hippocampus and brain stem. Varicella zoster virus Varicella zoster virus is a herpes virus. A small number of patients with chickenpox develop transient cerebellar ataxia. It is a rare cause of transverse myelitis, Guillain Barr´e syndrome and acute disseminated encephalomyelitis (ADEM). Severe neonatal chickenpox is generally attributed to late intrauterine infection and to lack of maternal antibodies. Devastating generalized chickenpox can occur with cerebral and pulmonary injury. Perinatal mortality is high in cases of neonatal varicella. Ebola virus Neonates born to women with Ebola are often premature, and typically do not survive for more than a few weeks. It is not clear whether these deaths are due to transmission of Ebola virus from mother to the neonate or to other factors that contribute to high infant mortality rates in Ebola-affected countries, but it is suggested that in utero transmission of Ebola virus to the fetus occurs [88]. Because of the highly infectious nature of the agent and risk to healthcare workers, postmortem data are lacking. Post-infectious syndromes Acute disseminated encephalomyelitis ADEM is regarded as a para-infectious immune mediated disorder that often follows measles, rubella, EBV, mumps and VZV infection or vaccination. In Mycoplasma pneumoniae

524

and B. burgdorferi, ADEM may overlap with acute infection. ADEM is uncommon in children less than one year of age and rates have declined since the introduction of measles, mumps, and rubella vaccination in the United States. The neuropathology is of diffuse perivenular demyelination accompanied by T-cell inflammatory infiltration in the cerebral white matter. Many of the older literature accounts of CNS disease, regarded as being due to direct infection, were examples of ADEM [89]. Transverse myelitis Transverse myelitis is an immune-mediated inflammatory disorder similar to ADEM but affecting a few segments of the spinal cord that often follows a viral or mycoplasma infection.

CNS mycosis Cryptococcus neoformans infection Cryptococcus neoformans is an encapsulated yeast found worldwide and present in pigeon feces that contaminate soil. Infection is by inhalation, with hematogenous spread to the CNS. It is a common cause of meningitis in children with advanced immunosuppression, including AIDS [90]. The course is indolent, with fever, intermittent headache abnormal mental status and vomiting. Signs of meningeal irritation may be lacking. Occasionally, a cryptococcoma mass with an associated granulomatous inflammatory response can develop; this is more common in immunecompetent individuals. Cerebrospinal fluid cell count, protein and glucose may be normal but lymphocytic pleocytosis and low glucose are common. Cerebrospinal fluid cytology with India ink staining outlines the thick capsules. Cryptoccocal antigen testing and culture are all useful. In fatal cases, there is little inflammatory response to the yeast and the meninges are covered with a gelatinous layer that contains the organisms. They may also accumulate in the Virchow– Robin space around deep vessels, particularly those of the basal ganglia and thalamus. Periodic acid–Schiff and mucicarmine stain the capsule and are useful for identifying the fungi in tissues. Aspergillus infection The Aspergillus fungus is found worldwide in soil. CNS disease mostly arises from hematogenous spread from the lung and causes hemorrhagic necrotic cerebral abscesses [91,92]. Neutropenic children with bone marrow or solid organ transplants are most at risk. The outcome of CNS disease is usually fatal. There is little inflammatory reaction and the fungi have numerous branches and have septa (Figures 42.12a,b). Candida albicans infection Candida albicans can cause meningitis or brain microabscesses (Figure 42.13a) in the neonate, either transmitted maternally, or

Perinatal and Postnatal Infections Chapter 42

(a)

(b)

from intravascular catheters, or endocarditis. In immunocompromised patients with cancer, organisms are endogenous [93]. Suppurative microabscesses can be found disseminated in the brain and there may be a surrounding granulomatous reaction [91]. Characteristic pseudohyphae with periodic constrictions and 2–3 μm oval budding spores are seen (Figure 42.13b). In treated cases, residual microabscesses with degenerating fungi may occur.

Coccidioidomycosis, sporotrichosis, blastomycosis, paracoccidioidomycosis infections For reviews, see Chimelli and Mahler-Araujo [91] and Lucas [92].

Parasitic infections

Figure 42.12 Aspergillus. (a) Numerous fungi within a meningeal blood vessel and eroding through the wall (arrow). Hematoxylin and eosin. (b) Grocott’s stain outlines the fungi and demonstrates branching hyphae and septae.

(a)

(b)

Figure 42.13 Candida abscess. (a) Microabscesses in the cortex with slight granulomatous reaction in a premature neonate with candida septicemia (treated). Hematoxylin and eosin. (b) Periodic acid–Schiff stain showing candida within a blood vessel. Note constrictions.

Neurocysticercosis Neurocysticercosis is the most common CNS parasitic disease. It is endemic in Central and South America and Asia but occurs worldwide. The prevalence in some countries is 10% [94]. Man is infected as an intermediate host by ingesting food or water contaminated with human feces containing ova of the porcine tapeworm Taenia solium. Following intestinal penetration, ova are disseminated by blood and form cystic masses in organs. There are three CNS stages: i) cystic, in which cysticercus is viable and there is a single scolex; ii) necrotic, in which the parasite degenerates and provokes inflammation; and iii) fibrocalcified, in which there is a calcified nodule. Small subcutaneous nodules and myositis are common. Children with CNS disease usually have a solitary parenchymal cyst and present with focal or generalized seizures. Encephalitis is seen in children with numerous inflamed cysts [94]. Diagnosis is by radiology, detection of anticysticercal antibodies or histological demonstration of a parasite from the brain or cord. Cysts are round or oval, often multiple. The cyst wall is 2– 3 mm and has three layers, an outer eosinophilic cuticular layer with bundles of muscle fibers beneath, a middle cellular layer with small dark nuclei, an inner reticular layer with loose fibrils, excretory canaliculi, and small round calcareous corpuscles [95]. Humans can also ingest larva from undercooked pork, which results in an adult worm colonizing the human small intestine. It is very rare for an individual to be self-infected by these eggs. Hydatid disease, echinococcosis The larval form of Taenia echinococcus causes echinococcosis. Humans are infected by ingesting eggs from feces of the definitive host, usually dogs. The disease is most common in regions breeding cattle and sheep. Echinococcus granulosus species causes cystic echinococcosis, which occurs worldwide [96]. Echinococcus multilocularis causes alveolar echinococcosis in the northern hemisphere. Liver cysts are found in 50% of

525

Developmental Neuropathology cases. CNS disease involves the brain, the cord and the orbit. Infection is mostly acquired in childhood but as cysts enlarge slowly, clinical presentation is usually in adults. Some CNS and eye cysts are symptomatic even when small so most CNS cases are diagnosed in children [97]. Histologically, the cyst shows an outer periodic acid–Schiff-positive cuticle layer and an inner thin granular layer that generates the scolices with a double row of hooks. There is usually little inflammatory response, but a granulomatous reaction can occur [96].

Schistosomiasis, toxocariasis, trichinosis Schistosomiasis, toxocariasis and trichinosis may all involve the CNS in children. For review see [92]. Malaria Of the four malaria species, Plasmodium falciparum is responsible for cerebral malaria, with estimated deaths of one to two million annually, 80% in children less than five years of age. It is largely a tropical or equatorial disease, endemic in areas of subSaharan Africa, and is transmitted by mosquitoes. It occurs in nonimmune individuals from or visitors to endemic regions. Clinical manifestations are of altered mental status, coma, seizures, respiratory distress, hypoglycemia, anemia, and shocklike syndrome. Diagnosis is by parasite identification in erythrocytes in blood films. Several samples may be required. In fatal cases of cerebral malaria, the brain is swollen, the leptomeninges are deeply congested and may show petechial hemorrhages. Histologically the capillaries and venules are engorged (Figure 42.14) and parasites are visible “lined up” and adherent to the endothelial surface. There may be malaria pigment or Durck’s granulomas, necrotic white matter lesions with lymphocytes and macrophages, some containing parasites, surrounded

Figure 42.14 Plasmodium falciparum malaria. Tiny brown organisms within cerebral microvessels. Some are aligned directly beneath the endothelium. Hematoxylin and eosin.

526

by astrogliosis [98]. The generation of coma in cerebral malaria is thought to be related to specific endothelial receptors causing parasite adhesion and vascular adhesion molecule-induced blood–brain barrier damage. Genome-wide linkage studies in malaria infection [16] show that absolute neutrophil count correlated with expression of certain genes, including those for cytokine receptors (ILIR2, ILI8R1, IL6R) TLRs, heat shock proteins, HSPA1A and HSPA1L), acute phase proteins ferritin (FTL) and alkaline phosphatase (ALPL) and intracellular signaling factors (NFKB1A, JUNB and FOSL2). Based on changes in expression patterns of these and other genes, febrile and convalescent children could be assigned to distinct groups. Another gene cluster was associated with parasite density in children. Genome-wide linkage studies in malaria infection severity also showed significant linkages to chromosome 10p15.3-14 and chromosome 13q. Polymorphisms in genes relating to host immunity including human leukocyte antigen, are associated with susceptibility or resistance to malaria (for review, see Driss et al. [16]).

Amoebiasis Acanthamoeba histolytica can cause a hemorrhagic meningoencephalitis in immune competent or deficient adults, and rare cases occur in children. Acanthamoeba keratitis is a sightthreatening infection that occurs in contact lens wearers from contaminated lens solution or swimming. Acanthamoeba can be detected histologically in corneal scrapings. Toxoplasmosis Toxoplasma gondii infection of the CNS is due to reactivated postnatal primary latent infection. It occurs almost exclusively in immune suppression but is less frequent in pediatric than in adult AIDS [70,71]. Focal signs are common, including hemiparesis and convulsions with fever and headache. CT or MRI shows an abscess-like necrotic mass, which can be single, or multiple or more diffuse encephalitic multifocal lesions. The cerebrospinal fluid shows mild mononuclear pleocytosis, reduced glucose and elevated protein and increased local antibodies to T. gondii. The gross and microscopic pathology depends on the disease stage, duration and treatment. Active lesions show necrotizing encephalitis with fibrinoid necrosis of blood vessels and dense inflammatory infiltrate of polymorphs, mononuclear cells and macrophages. Encysted organisms containing bradyzoites (Figure 42.15) or free tachyzoites are also found, which can be confirmed by immunocytochemistry. Older lesions are cavitated, surrounded by gliotic scarring, and may not contain organisms. Trypanosomiasis African trypanosomiasis (sleeping sickness) transmitted by the tsetse fly is relatively uncommon in children, but fulminant CNS disease may occur. American trypanosomiasis (Chagas disease)

Perinatal and Postnatal Infections Chapter 42

9.

10.

11.

12.

13.

Figure 42.15 Toxoplasmosis. Necrotizing encephalitis with encysted organisms containing bradyzoites (arrows) in parenchyma. Hematoxylin and eosin.

transmitted by reduviid bugs affects children and can cause a fatal myocarditis and meningoencephalitis.

14. 15.

16. 17.

Acknowledgments We wish to acknowledge the children who lost their lives due to CNS infections, their parents, and also the clinicians and pathologists who referred cases.

18.

19.

References 1. Pappas G, Kiriaze IJ, Falagas ME (2008) Insights into infectious disease in the era of Hippocrates. Int J Infect Dis 12:347–50 2. Thigpen MC, Whitney CG, Messonnier NE et al. (2011) Emerging Infections Programs Network. Bacterial meningitis in the United States, 1998-2007. N Engl J Med 364:2016–25 3. Gaschignard J, Levy C, Romain O et al. (2011) Neonatal bacterial meningitis: 444 cases in 7 years. Pediatr Infect Dis J 30:212–17 4. Tiskumara R, Fakharee SH, Liu C et al. (2009) Neonatal infections in Asia. Arch Dis Child Fetal Neonatal Ed 94:F144–8 5. Nhanatmbo AA, Cantarelli VV, Caireao J et al. (2015) Frequency of pathogenic paediatric bacterial meningitis in Mozambique: The critical role of multiplex real-time polymerase chain reaction to estimate the burden of disease. PLoS One. 10:e0138249 6. Durack DT, Whitley RJ, Scheld WM (1991) Approach to the patient with central nervous system infection. In: WM Scheld, RJ Whitley, DT Durack, eds, Infections of the Central Nervous System. New York, NY, Raven, pp. 2–4 7. Smith AJ, Haas J (1991) Neonatal Bacterial Meningitis. In: WM Scheld, RJ Whitley, DT Durack, eds, Infections of the Central Nervous System. New York, NY, Raven Press. pp. 313–3 8. Edwards MS, Baker CJ (2003) Bacterial Infections in the neonate. In: SS Long, LK Pickering, CG Prober, eds, Principles and Practice

20. 21.

22.

23.

24. 25.

26.

27.

of Pediatric Infectious Diseases, 2nd ed., Philadelphia, PA, Churchill Livingstone, pp. 536–42 Somech R, Amariglio N, Spirer Z, Rechvi (2003) Genetic predisposition to infectious pathogens: a review of less familiar variants. Pediatr Infect Dis J 22:457–61 Deckert M (2015) Bacterial infections. In: S. Love, H. Budka, J. W. Ironside. A. Perry, eds, Greenfield’s Neuropathology, 9th ed., New York, NY, CRC Press, pp. 1197 Abu-Maziad A, Schaa K, Bell EF et al. (2010) Role of polymorphic variants as genetic modulators of infection in neonatal sepsis. Pediatr Res 68:323–9 Thuong NTT, Hawn TR, Chau TTH et al. (2012) Epiregulin (EREG) variation is associated with susceptibility to GE tuberculosis. Genes Immun 13:275–81 Levin M, Newport MJ, D’Souza S et al. (1999) Familial disseminated atypical mycobacterial infection in childhood; a human mycobacterial susceptibility gene? Lancet 345:79–83 Dorman SE, Uzel G, Roesler J et al. (1999) Viral infections in interferon-gamma receptor deficiency. J Pediatr 135:640–3 Sancho-Shimizu V, P´erez de Diego R, Lorenzo L et al. (2011) Herpes simplex encephalitis in children with autosomal recessive and dominant TRIF deficiency. J Clin Invest 121:4889–902 Driss A, Hibbert JM, Wilson NO et al. (2011) Genetic polymorphisms linked to susceptibility to malaria. Malar J 10:271 Wang Y Guo G, Wang H et al. (2014) Comparative study of bacteriological culture and real-time fluorescence quantitative PCR (RT-PCR) and multiplex PCR-based reverse line blot (mPCR/RLB) hybridization assay in the diagnosis of bacterial neonatal meningitis. BMC Pediatr 14:224 Zaida AK, Thayer D, Ali SA, Khan TA (2009) Pathogens associated with sepsis in newborns and young infants in developing countries. Pediatr Infect Dis J 1(Suppl):S10–18 Schlech WF 3rd (2000) Foodborne listeriosis. Clin Infect Dis 31:770–5 Dellinger RP (2003) Inflammation and coagulation: Implications for the septic patient. Clin Inf Dis 36:1259–65 Edwards MS, Rench MA, Haffar AAM et al. (1985) Longterm sequelae of group B streptococcal meningitis in infants. J Pediatr 106:717–22 Grimwood K, Anderson P, Anderson V et al. (2000) Twelve year outcomes following bacterial meningitis: further evidence for persisting effects. Arch Dis Child 83:111–16 Taylor HG, Michaels RH, Mazur PM et al. (1984) Intellectual, neuropsychological and achievement outcomes in children six to eight years after recovery from haemophilus influenzae meningitis. Pediatrics 74:198–205 Obiako MN (1987) Profound childhood deafness in Nigeria: a three year survey. Ear Hearing 8:74–7 Nau R, Soto A, Bruck W (1999) Apoptosis of neurons in the dentate gyrus in humans dying from bacterial meningitis. J Neuropathol Exp Neurol 58:265–74 Von Mering M, Wellmer A, Michel U et al. (2001) Transcriptional regulation of caspases in experimental pneumococcal meningitis. Brain Pathol 11:282–95 Centers for Disease Control and Prevention (1995) Progress towards elimination of Haemophilus influenzae type B disease among infants and children-United States, 1983–1994. MMWR 44:545

527

Developmental Neuropathology 28. Centers for Disease Control and Prevention (1998) Progress towards elimination of Haemophilus influenzae type B disease among infants and children-United States, 1987–1997 MMWR 47:993 29. Smith HP, Hendrick EB (1983) Subdural empyema and epidural abscess in children. J Neurosurg 58:392–7 30. Listinsky JL, Wood BP, Ekholm SC (1986). Parietal osteomyelitis and epidural abscess; a delayed complication of fetal monitoring. Pediatr Radiol 16:150–1 31. Auletta JJ, John CC (2001) Spinal epidural abscesses in children; a 15 year experience and review of the literature. Clin Inf Dis 32:9–17 32. Nejat F, Ardkani SJ, Khotaei GT, Roodsari NN (2002) Spinal epidural abscess in a neonate. Pediatr Inf Dis J 21:797–8 33. Nielsen H, Gyldensted C, Harmsen A (1982) Cerebral abscess: etiology and pathogenesis, symptoms, diagnosis and treatment. Acta Neurolog Scand 65:609–22 34. Kaplan K (1985) Brain abscess. Med Clin North Am 69:345–60 35. Britt RH, Enzmann DR, Yeager AS (1981) Neuropathological and computerized tomographic findings in experimental brain abscess. J Neurosurg 55:590–603 36. Gray F, Alonso J-M (2002) Bacterial infections of the central nervous system. In: DI Graham, PL Lantos, eds, Greenfield’s Neuropathology 7th ed., vol 2, London, Arnold, pp. 151–93 37. Saez-Llorens XJ, Umann Ma, Odio CM et al. (1989) Brain abscess in infants and children. Pediatr Infect Dis J 8:449–58 38. Centers for Disease Control and Prevention (2002) Tuberculosis morbidity among US-born and foreign-born populations: United States, 2000 MMWR 51:101–4 39. Lincoln EM, Sordillo SVR, Davies PA (1960). Tuberculous meningitis in children. A review of 167 untreated and 74 treated patients with special reference to early diagnosis. J Pediatr 57:807–23 40. Smith AL (1975) Tuberculous meningitis in childhood. Med J Aust 1:57–60 41. Zuger A, Lowy FD (1991) Tuberculosis of the central nervous system In: WM Scheld, RJ Whitley, DT Durack, eds, Infections of the Central Nervous System. New York, NY, Raven Press, pp. 425–56 42. Baker CA, Cartwright CP, Williams DN et al. (2002) Early detection of central nervous system tuberculosis with the geneprobe nucleic acid amplification assay; utility in an Inner City Hospital. Clin Inf Dis 35:339–42 43. Dastur HM, Desai AD (1965) A comparative study of brain tuberculomas and gliomas based upon 107 case records of each. Brain 88:375–86 44. Maheswari V, Srivastava VK, Prasad S, Alam K (2002) Tuberculoma: a significant diagnostic entity in brain biopsies of intracranial space occupying lesions in children. J Trop Pediatr 48:242–4 45. Den BI, Chandra S, Mangia S et al. (2002) Spinal intramedullary tuberculoma and abscess; a rare cause of paraparesis. Neurol India 50:494–6 46. Corey L, Wald A (2009) Maternal and neonatal herpes simplex virus infections. N Engl J Med 361:1376–85 47. Tookey P, Peckham CS (1996) Neonatal herpes simplex virus infection in the British Isles. Paediatr Perinat Epidemiol 10:432–42 48. Kimberlin DW (2004) Neonatal herpes simplex infection. Clin Microbiol Rev 17:1–13 49. Brown ZA, Wald A, Morrow RA et al. (2003) Effect of serologic status and cesarean delivery on transmission rates of herpes simplex virus from mother to infant. J Am Med Assoc 289:203–9

528

50. Whitley RJ, Stagno S (1991). Perinatal Viral Infections. In: WM Scheld, RJ Whitley, DT Durack, eds, Infections of the Central Nervous System. New York, NY, Raven Press. pp. 167–200 51. Kaye EM, Dooling EC (1981) Neonatal herpes simplex meningoencephalitis associated with fetal monitor scalp electrodes. Neurology 31:1045–55 52. Kimberlin DW, Lin CY, Jacobs RF, et al. (2001) Natural history of neonatal herpes simplex virus infections in the acyclovir era. Pediatrics 108:223–9 53. Whitley RJ, Lakeman F (1995) Herpes simplex virus infections of the central nervous system: therapeutic and diagnostic considerations. Clin Infect Dis 20:414–20 54. Love S, Wiley CA (2002). Viral diseases. In: DI Graham, PL Lantos, eds, Greenfield’s Neuropathology, 7th ed., vol 2, London, Arnold. pp. 1–105 55. Casrouge A, Zhang SY, Eidenschenk C et al. (2006) Herpes simplex virus encephalitis in human UNC-93B deficiency. Science 314:308– 12 56. Zhang SY, Jouanguy E, Ugolini S et al. (2007) TLR3 deficiency in patients with herpes simplex encephalitis. Science 317:1522–7 57. Perez de Diego R, Sancho-Shimizu V, Lorenzo L et al. (2010) Human TRAF3 adaptor molecule deificiency leads to impaired toll like receptor 3 response and susceptibility to herpes simplex encephalitis. Immunity 33:400–11 58. Guo Y, Audry M, Ciancanelli M, et al. (2011) Herpes simplex encephalitis in a patient wth complete TLR 3 deficiency: TLR is otherwise redundant in protective immunity. J Exp Med 208: 2083–98 59. Rotbart HA (2003) Aseptic and Viral meningitis. In: SS Long, LK Pickering, CG Prober, eds, Principles and Practice of Pediatric Infectious Diseases, 2nd ed., Philadelphia, PA, Churchill Livingstone. pp. 285–90 60. Kaplan MH, Klein SW, McPhee J, Harper RG (1983) Group B coxsackie virus infections in infants younger than three months of age: A serious childhood illness. Rev Infect Dis 5:1019–32 61. Centers for Disease Control and Prevention (2016) Updates on CDC’s Polio Eradication Efforts. https://www.cdc.gov/polio/ updates (accessed December 1, 2017) 62. Centers for Disease Control and Prevention (1982) Unexplained immunodeficiency and opportunistic infections in infants. New York, New Jersey, California MMWR 31:665–7 63. European Collaborative Study (1994) Natural history of vertically acquired human immunodeficiency virus-1 infection. Pediatrics 94:815–19 64. Epstein LG, Sharer LR, Oleske JM et al. (1986) Neurological manifestations of human immunodeficiency virus infection in children. Pediatrics 78:678–87 65. Keohane C, Lacroix-Jousselin C, Dickson DW (1993). Central nervous system in pediatric AIDS. In: F Gray, ed., Atlas of the Neuropathology of HIV Infection. Oxford, Oxford University Press. pp. 177–208 66. Dickson DW, Belman Al, Kim TS et al. (1989) Spinal cord pathology in pediatric acquired immunodeficiency syndrome. Neurology 9:227–325 67. Sharer LR (1992) Pathology of HIV-1 infection of the central nervous system: a review. J Neuropathol Exp Neurol 51:3–11 68. Kaul M, Garden GA, Lipton SA (2001) Pathways to neuronal injury and apoptosis in HIV-associated dementia. Nature 410:988–94

Perinatal and Postnatal Infections Chapter 42 69. Gray F, Keohane C (2003) The neuropathology of HIV infection in the era of highly active antiretroviral therapy (HAART). Brain Pathol 13:79–83 70. Bell JE, Lowrie S, Koffi K et al. (1997) The neuropathology of HIVinfected African children in Abidjan, Cote d’Ivoire. J Neuropathol Exp Neurol 56:686–92 71. Dickson DW, Llena JF, Nelson SJ, Weidenheim KM (1995) Central nervous system pathology in pediatric AIDS. Ann NY Acad Sci 693:93–106 72. Kovacs A, Schluchter M, Easley K et al. (1999) Cytomegalovirus infection and HIV-1 disease progression in infants born to HIV-1 infected women. N Engl J Med 341:77–84 73. Belec L, Tayot J, Tron P et al. (1990) Cytomegalovirus encephalopathy in an infant with congenital acquired immunodeficiency syndrome. Neuropediatrics 21:124–9 74. Mason WH, Ross LA, Lanson J, Wright HT Jr (1993) Epidemic measles in the postvaccine era: Evaluation of epidemiology, clinical presentation and complications of suspected measles in hospitalized children. Pediatr Inf Dis J 12:42–8 75. Maldonado YA (2003) Rubeola virus (measles and subacute sclerosing panencephalitis). In: SS Long, LK Pickering, CG Prober, eds, Principles and Practice of Pediatric Infectious Diseases, 2nd ed., Philadelphia, PA, Churchill Livingstone. pp. 1148–55 76. Centers for Disease Control (1981) Measles encephalitis: United States, 1962–79. MMWR 30:362–4 77. Centers for Disease Control (1998) Measles mumps and rubellavaccine use and strategies for elimination of measles rubella and congenital rubella syndrome and control of mumps; recommendations of the Advisory Committee on Immunisation Practices (ACIP) MMWR 47:1–57 78. McBrien J, Murphy J, Gill D et al. (2003) Measles outbreak in Dublin, 2000. Pediatr Infect Dis J 22:580–4 79. Dimova P, Bojinova V (2000) Subacute sclerosing panencephalitis with atypical onset. Clinical computed tomographic and magnetic imaging correlations. Child Neurol 15:258–60 80. Farrington CP (1991) Subacute sclerosing panencephalitis in England and Wales transient effects and risk estimates. Statist Med 10:1733–44 81. Ter Meulen V, Muller D, Kackell Y, Katz M (1972) Isolation of infectious measles virus in measles encephalitis. Lancet 2:1172–5 82. Suryanaryana K, Baczko K, ter Meulen V, Wagner RR (1994) Transcription inhibition and other properties of matrix proteins

83. 84.

85.

86. 87.

88.

89.

90.

91. 92.

93.

94. 95. 96. 97. 98.

expressed by M genes cloned from measles virus and diseased human brain tissue. J Virol 68:1532–43 Stohlman SA, Hinton DR (2001) Viral induced demyelination. Brain Pathol 11:92–6 Wight C, Jin L, Nelson SC et al. (2003) An autopsy-proven case of fulminant subacute sclerosing panencephalitis. Neuropathol Appl Neurobiol 29:312–20 Townsend JJ, Wolinsky JS, Baringer JR (1976) The neuropathology of progressive rubella panencephalitis of late onset. Brain 99: 81–90 Wolinsky JS, Berg BO, Maitland CJ (1976) Progressive rubella panencephalitis. Arch Neurol 33:722–3 Garen PD, Tsai F, Powers JM (1999). Human eastern equine encephalitis; immunohistochemstry and ultrastructure. Mod Pathol 12:646–52 Baggi FM, Taybi A, Kurth A et al. (2014) Management of pregnant women infected with Ebola virus in a treatment centre in Guinea, June 2014. Euro Surveill 19(49): pii: 20983 Prineas JW, McDonald WI, Franklin RJM (2002) Demyelinating diseases. In: DI Graham, PL Lantos, eds, Greenfield’s Neuropathology, 7th ed., vol 2, London, Arnold. pp. 471–550 Hajjeh RA, Warnock DW (2003) Cryptococcus neoformans. In: SS Long, LK Pickering, CG Prober, eds, Principles and Practice of Pediatric Infectious Diseases, 2nd ed., Philadelphia, PA, Churchill Livingstone. pp. 1229–33 Chimelli L, Mahler-Araujo MB (1997). Fungal infections. Brain Pathol 7:613–27 Lucas S (2015) Fungal Infections. In: S Love, H Budka, JW Ironside, A Perry, eds, Greenfield’s Neuropathology, 9th ed., Boca Raton, Fl, CRC, pp. 1281–96 Hostetter MK (2003) Candida species. In: SS Long, LK Pickering, CG Prober, eds, Principles and Practice of Pediatric Infectious Diseases, 2nd ed., Philadelphia, PA, Churchill Livingstone. pp. 1203–13 Rosenfeld E (2003) Neurocysticercosis update. Pediatr Infect Dis J 22:181–2 Pitella JEH (1997) Neurocysticercosis. Brain Pathol 7:681–93 Taratuto AL, Venturiello SM (1997). Echinococcosis. Brain Pathol 7:673–9 Slim MS, Khayat G, Nasr AT, Jidejian YD (1971) Hydatid disease in childhood. J Paediatr Surg 6:440–8 Turner G (1997) Cerebral malaria. Brain Pathol 7:569–82

529

43

Rasmussen Encephalitis Harry V. Vinters,1,2 Shino D. Magaki,1 and Geoffrey C. Owens2 1 Section

of Neuropathology, Department of Pathology and Laboratory Medicine, David Geffen School of Medicine, Los Angeles, California, USA 2 Department of Neurosurgery, David Geffen School of Medicine at UCLA and Ronald Reagan UCLA Medical Center, Los Angeles, California, USA

Definition

Genetics

Rasmussen encephalitis is a rare, medically intractable seizure disorder characterized clinically by progressive unilateral neurologic deficit (usually hemiparesis), and epilepsia partialis continua, and neuropathologically by chronic inflammatory and destructive changes almost always confined to one cerebral hemisphere. Corticectomy or even hemispherectomy is often the “treatment of choice” for affected patients, yielding abundant material for detailed neuropathologic study. A definitive etiology for the condition has not yet been established.

There is no clear evidence that Rasmussen encephalitis is an inherited disease; identical twins discordant for Rasmussen encephalitis have been reported (4, 5) although genetic factors that may predispose to the condition cannot be completely ruled out (5). There is one report of brothers with alternating bilateral epilepsia partialis continua, in whom a brain biopsy showed chronic inflammation (4). However, the limited morphologic information and uncharacteristic semiology in these cases renders a diagnosis of Rasmussen encephalitis uncertain.

Synonyms and historical annotations Synonyms for Rasmussen encephalitis include Rasmussen syndrome, smoldering encephalitis, and chronic (pathogen-free) encephalitis. Rasmussen et al. [1,2] described the first cases 60 years ago; he and his colleagues observed the characteristic enigmatic chronic localized encephalitis after documenting the inflammatory features in cortical tissue from some patients with intractable epilepsy treated by surgical resection.

Clinical features

Epidemiology

Clinical presentation The rate of disease progression is variable, ranging from months to many years [6]. Typically, children present between the ages of 5–10 years with seizures, usually focal (unilateral) motor seizures, which prove to be resistant to anticonvulsant medication. In association with focal motor seizures, there is evidence of a progressive hemiparesis, and in many cases an associated cognitive decline. Epilepsia partialis continua ultimately occurs in 60% of the diagnosed cases.

The age range of presentation with first seizure is usually five to ten years, although presentation outside this range is not inconsistent with the diagnosis, and there are a small number of adult onset cases [3]. There is no apparent gender bias. The underlying etiology is unknown and considerably debated. Some reports suggest a possible link to an initial viral infection, including reports of associated iritis at presentation. No particular risk factors have been reported.

Imaging It appears that early involvement of the head of the caudate nucleus may be pathognomonic. Thereafter magnetic resonance imaging (MRI) signal changes and progressive atrophy of one cerebral hemisphere are considered to be consistent with a clinical diagnosis. However, consistent progressive changes are not seen in all cases; thus, a direct correlative MRI signal change may not be linked to clinical presentation.

Developmental Neuropathology, Second Edition. Edited by Homa Adle-Biassette, Brian N. Harding, and Jeffrey A. Golden. © 2018 John Wiley & Sons Ltd. Published 2018 by John Wiley & Sons Ltd.

531

Developmental Neuropathology

(a)

(c)

(b)

(d)

Figure 43.1 Surgically resected specimens from patients with Rasmussen encephalitis; all panels (except b) are from hematoxylin and eosin-stained sections. (a) Severely gliotic region of neocortex, with extensive astrocytic gliosis. Arrows indicate a region of cystic cavitation. Underlying white matter is largely unaffected, but showed evidence of Wallerian degeneration. (b) Glial fibrillary acidic protein

Neuropathology Macroscopy The overwhelming majority of neuropathological observations have been gained from surgical material from diagnostic biopsies and functional hemispherectomies [7,8]. Gyral damage is patchy and varies from indiscernible, through slight granularity or discoloration and thinning in the early stages, to extensive hemi-atrophy, striatal atrophy and ventricular dilatation in longstanding cases. The neuropathological substrate of cerebral atrophy is multifocal neuron loss with astrocytic gliosis, sometimes with laminar necrosis (Figure 43.1).

Histopathology Microscopic changes (Figure 43.2) are much more extensive than is evident macroscopically, although patchy and often sharply demarcated from normal areas; lesions may be abundant and extensive in large cortical resections. Active lesions show perivascular lymphomonocytic cuffs in the leptomeninges,

532

immunostained section of microcystic change, highlighting reactive gliosis. (c) Microglial nodule (arrows) in relatively intact region of cortex. (d) Perivenous lymphocytic cuffing. Note pronounced astrocytic gliosis adjacent to the vessel, with prominent gemistocytic astrocytes within neuropil (×200).

randomly scattered through the cortical layers, and less often in subjacent white matter. In the cortex, there are activated microglia, microglial nodules, and prominent neuronophagia, with a variable degree of neuronal loss and gliosis. Longstanding “burnt out” lesions show little inflammation but show spongy destruction of the cortical ribbon, marked capillary proliferation and gliosis. Active and burnt out areas can occur in close proximity to each other in a biopsy. Rare postmortem studies have confirmed that the inflammatory process is, with rare exceptions [9], strictly unilateral, involving the cerebral hemisphere very widely from frontal pole to occipital pole. This “unilaterality” of the neuropathological features distinguishes Rasmussen encephalitis from other immune-mediated central nervous system disorders [10]. The basal ganglia and occasionally the brain stem are involved. There is also cerebellar atrophy without inflammation, which may be secondary to the neocortical destruction (i.e. a trans-synaptic phenomenon), or due to hypoxia resulting from seizures. It has been proposed that the neuropathological features of Rasmussen encephalitis comprise four merging stages [11– 13]. The earliest of these is characterized by inflammation,

Rasmussen Encephalitis Chapter 43

(a)

(c)

(b)

(d)

Figure 43.2 Surgically resected specimens; all micrographs are from hematoxylin and eosin-stained sections (a × 20; b-d × 40). (a,b) Prominent cystic cavitation shown at low (a) and high (b) magnification. Notice that an immediately adjacent gyrus (arrows) appears to be entirely unaffected. (c) A large collection of macrophages in an area of cortical encephalomalacia. (d) Prominent collections of lymphocytes (arrowheads) adjacent to a region of tissue destruction. Arrow indicates perivenous lymphocytic cuffing.

especially the presence of perivascular lymphocytes, and accumulation of microglial nodules within the brain substance, but little evidence of neuronal destruction. In the second stage, lymphocytic inflammation becomes more prominent and both astrogliosis and microgliosis become widespread, involving all cortical layers; patchy neuronal loss may be present. In the third stage, severe neocortical degeneration becomes apparent with a patchy panlaminar pattern and astrocytic gliosis. In the fourth stage, there is profound cortical atrophy with gliosis and vacuolization of the neuropil, and (in many cases) cystic cavitation, especially within the neocortex. However, it bears emphasis that areas of relatively normal cortex can separate areas with severe abnormalities; the occipital cortex may be relatively spared. Figure 43.3 shows a hypothetical pathogenetic sequence of changes in the brain of a patient with Rasmussen encephalitis.

Immunohistochemical and ultrastructural findings CD68+, Iba1+ microglia, and CD3+ T cells are readily apparent in the Rasmussen encephalitis brain, although variable in different segments of the neocortex; fewer B cells are present [13–15]. A quantitative analysis of Iba1+ immunoreactivity in resected Rasmussen encephalitis, dysplastic cortex and tuberous sclerosis tubers provided clearcut evidence for an accentuated microglial activation in Rasmussen encephalitis compared with other surgically treated seizure disorders [16]. On ultrastructural examination of Rasmussen encephalitis brain tissue there is generally no evidence of definite viral particles, though structures resembling measles virus have been described [17]. Deposition of immune complexes, damage to the blood–brain barrier, and some curious electron microscopic findings, including rare endothelial tubuloreticular inclusions, have been described [14,17].

533

Developmental Neuropathology

Figure 43.3 Hypothetical events in the pathogenesis and progression of Rasmussen encephalitis, for which the etiology is not known. Intervention might be possible at times indicated by any of the “arrows” showing progression.

Differential diagnosis The major clinical and morphologic differential diagnoses are any neocortical lesion that may be “epileptogenic” (e.g. focal cortical dysplasia), although the neuroimaging features of Rasmussen encephalitis are quite distinct from those of focal cortical dysplasia. Cases of dual Rasmussen encephalitis and focal cortical dysplasia pathology have, however, been reported [18–21]. A European consensus conference on Rasmussen encephalitis has formulated helpful diagnostic criteria that incorporate clinical, electroencephalogram, MRI and histopathologic findings [11,12].

Pathogenesis The pathological features of the Rasmussen encephalitis brain suggest a chronic viral infection or autoimmunity [10,14,22–24]. Very low levels of Herpesviridae genomic sequences have been detected in some patients, which may or may not be of pathological significance [25,26]. A recent search of RNASeq data from six Rasmussen encephalitis brain specimens failed to find any evidence for actively transcribed viral genes (results unpublished at the time of writing). However, after clearance of a virus from the brain, some of the memory T cells that persist may cross react with a brain-specific self-antigen due to molecular mimicry. Autoreactive bystander T cells that have escaped tolerance could also traffic to the brain during the response to an infection. It has recently been shown that many of the T cells in resected brain tissue are resident memory T cells, consistent with the notion that an acute immune response occurs at a very early stage of the disease [27]. Evidence that Rasmussen encephalitis may be an autoimmune disease came from the detection of circulating autoantibodies against the glutamate ionotropic receptor AMPA type subunit

534

3 in patients with Rasmussen encephalitis [28]. In subsequent studies, these antibodies were found to occur in other seizure disorders, leading to the conclusion that their formation was a function of the receptors being “shed” into the circulation during seizures [29–31]. Autoantibodies to the glutamate ionotropic receptor NMDA type subunit 2 and the synaptic protein Munc18 have also been described in some patients with Rasmussen encephalitis, often accompanied by a B lymphocyte and plasma cell infiltrate [32–34]. Aggressive plasmapheresis treatment and anti-CD20 monoclonal antibody therapy has been tried in some affected individuals [35,36]. Evidence exists for the role of cellular immunity in disease pathogenesis. Brain lymphocytic infiltrates in Rasmussen encephalitis consist predominantly of CD8+ αβ T cells and CD4-/CD8- γδ T cells [14,15,37,38]. Polarized T cells containing granzyme B have been described in close apposition to neurons and astrocytes [37,39]. With rare exceptions, B lymphocytes, immunoglobulins, and complement are usually not found in Rasmussen encephalitis brain tissue [15]. The T cell infiltrate is of comparatively restricted clonality [38, 40–42]. The observation of CD8+ clones within brain infiltrates in Rasmussen encephalitis tissue, combined with the finding of their persistence in the periphery, has suggested the hypothesis that Rasmussen encephalitis may be caused by an antigen-driven major histocompatibility complex (MHC) class-I restricted CD8+ T cell-mediated attack upon CNS astrocytes and neurons [42]. Gamma delta T cells also appear to be involved in the immune response in Rasmussen encephalitis [38]. These innate-like immune cells are early responders to an inflammatory reaction, and are not MHC-restricted. They recognize MHC-like molecules that are upregulated on stressed cells, and produce proinflammatory cytokines such as interferon-gamma (IFN-γ). Increased levels of IFN-γ mRNA in resected Rasmussen encephalitis brain tissue compared with focal cortical dysplasia have been reported [43], and elevated levels of IFN-γ and tumor necrosis factor alpha have been

Rasmussen Encephalitis Chapter 43

measured in cerebrospinal fluid from patients with Rasmussen encephalitis [44]. Transcripts for CCL5, CXCL9 and CXCL10, chemokines that attract cytotoxic CD8+ T cells, are increased in Rasmussen encephalitis brain tissue compared with focal cortical dysplasia [43], which is consonant with the expression of their cognate receptors on brain-infiltrating T cells [45,46]. It is likely that activated microglia also produce proinflammatory cytokines and chemokines, thus resulting in an inflammatory milieu that may promote seizures [47]. Activated microglia may also influence neuron excitability by forming hemichannels with pyramidal neurons [48]. The interesting observation of arterial intimal thickening in patients with Rasmussen encephalitis who have undergone serial corticectomies over months or years may be related to an autoimmune mechanism directed against blood vessel wall components [49].

Future directions and therapy In view of a possible autoimmune etiology for Rasmussen encephalitis, various immunosuppressive therapies have been proposed, including intravenous immunoglobulin and highdose steroids [50]. It appears that long-term treatment protocols involving immunoglobulin and steroids may put off other treatments and gain time, but problems with adverse effects raise concerns about long-term therapy with these modalities. Use of antiviral [3,51,52] and alternative immunosuppressive therapies [36,53–55] have been reported. Given the comparative rarity of Rasmussen encephalitis, many such reports are anecdotal, highlighting the importance of case registries and multi-center research efforts. However, surgery in the form of hemispherectomy still appears to be the only long-term cure for such seizures, and early referral of children to epilepsy surgery centers is recommended for decisions on the optimal timing of such treatment. Further evaluation of possible autoimmune mechanisms and alternative immunosuppressive therapy is required.

References 1. Rasmussen T (1978) Further observations on the syndrome of chronic encephalitis and epilepsy. Appl Neurophysiol 41:1–12 2. Rasmussen T, Olszewski J, Lloydsmith D (1958) Focal seizures due to chronic localized encephalitis. Neurology 8:435–45 3. McLachlan RS, Girvin JP, Blume WT, Reichman H (1993) Rasmussen’s chronic encephalitis in adults. Arch Neurol 50:269–74 4. Silver K, Andermann F, Meagher-Villemure K (1998) Familial alternating epilepsia partialis continua with chronic encephalitis: another variant of Rasmussen syndrome? Arch Neurol 55:733–6 5. Takahashi Y, Mogami Y, Mine J et al. (2013) Genetic variations of immunoregulatory genes associated with Rasmussen syndrome. Epilepsy Res 107:238–43 6. Oguni H, Andermann F, Rasmussen TB (1992) The syndrome of chronic encephalitis and epilepsy. A study based on the MNI series of 48 cases. Adv Neurol 57:419–33

7. Honavar M, Janota I, Polkey CE (1992) Rasmussen’s encephalitis in surgery for epilepsy. Dev Med Child Neurol 34:3–14 8. Robitaille Y (1991) Neuropathologic aspects of chronic encephalitis. In: F Andermann (ed) Chronic Encephalitis and Epilepsy: Rasmussen’s Syndrome. Butterworth-Heinemann: London. pp. 79– 110 9. Tobias SM, Robitaille Y, Hickey WF et al. (2003) Bilateral Rasmussen encephalitis: postmortem documentation in a five–year– old. Epilepsia 44:127–30 10. Junker A, Bruck W (2012) Autoinflammatory grey matter lesions in humans: cortical encephalitis, clinical disorders, experimental models. Curr Opin Neurol 25:349–57 11. Bien CG, Granata T, Antozzi C et al. (2005) Pathogenesis, diagnosis and treatment of Rasmussen encephalitis: a European consensus statement. Brain 128:454–71 12. Bien CG, Urbach H, Deckert M et al. (2002) Diagnosis and staging of Rasmussen’s encephalitis by serial MRI and histopathology. Neurology 58:250–7 13. Pardo CA, Vining EP, Guo L et al. (2004) The pathology of Rasmussen syndrome: stages of cortical involvement and neuropathological studies in 45 hemispherectomies. Epilepsia 45:516–26 14. Farrell MA, Droogan O, Secor DL et al. (1995) Chronic encephalitis associated with epilepsy: immunohistochemical and ultrastructural studies. Acta Neuropathol 89:313–21 15. Prayson RA, Frater JL (2002) Rasmussen encephalitis: a clinicopathologic and immunohistochemical study of seven patients. Am J Clin Pathol 117:776–82 16. Wirenfeldt M, Clare R, Tung S et al. (2009) Increased activation of Iba1+ microglia in pediatric epilepsy patients with Rasmussen’s encephalitis compared with cortical dysplasia and tuberous sclerosis complex. Neurobiol Dis 34:432–40 17. Park SH, Vinters HV (2002) Ultrastructural study of Rasmussen encephalitis. Ultrastruct Pathol 26:287–92 18. Hart YM, Andermann F, Robitaille Y et al. (1998) Double pathology in Rasmussen’s syndrome: a window on the etiology? Neurology 50:731–5 19. Prayson RA (2012) Dual pathology in Rasmussen’s encephalitis: a report of coexistent focal cortical dysplasia and review of the literature. Case Rep Pathol 2012:569170 20. Takei H, Wilfong A, Malphrus A et al. (2010) Dual pathology in Rasmussen’s encephalitis: a study of seven cases and review of the literature. Neuropathology 30:381–91 21. Yacubian EM, Rosemberg S, Marie SK et al. (1996) Double pathology in Rasmussen’s encephalitis: etiologic considerations. Epilepsia 37:495–500 22. Amrom D, Kinay D, Hart Y et al. (2014) Rasmussen encephalitis and comorbid autoimmune diseases: A window into disease mechanism? Neurology 83:1049–55 23. Ganor Y, Goldberg-Stern H, Blank M et al. (2005) Antibodies to glutamate receptor subtype 3 (GluR3) are found in some patients suffering from epilepsy as the main disease, but not in patients whose epilepsy accompanies antiphospholipid syndrome or Sneddon’s syndrome. Autoimmunity 38:417–24 24. Pardo CA, Nabbout R, Galanopoulou AS (2014) Mechanisms of epileptogenesis in pediatric epileptic syndromes: Rasmussen encephalitis, infantile spasms, and febrile infection–related epilepsy syndrome (FIRES). Neurotherapeutics 11:297–310 25. Jay V, Becker LE, Otsubo H et al. (1995) Chronic encephalitis and epilepsy (Rasmussen’s encephalitis): detection of cytomegalovirus

535

Developmental Neuropathology

26.

27.

28.

29.

30.

31.

32.

33.

34. 35. 36.

37.

38.

39.

40.

536

and herpes simplex virus 1 by the polymerase chain reaction and in situ hybridization. Neurology 45:108–17 Vinters HV, Wang R, Wiley CA (1993) Herpesviruses in chronic encephalitis associated with intractable childhood epilepsy. Hum Pathol 24:871–9 Owens GC, Chang JW, Huynh MN et al. (2016) Evidence for resident memory T cells in Rasmussen encephalitis. Front Immunol 7:64 Rogers SW, Andrews PI, Gahring LC et al. (1994) Autoantibodies to glutamate receptor GluR3 in Rasmussen’s encephalitis. Science 265:648–51 Mantegazza R, Bernasconi P, Baggi F et al. (2002) Antibodies against GluR3 peptides are not specific for Rasmussen’s encephalitis but are also present in epilepsy patients with severe, early onset disease and intractable seizures. J Neuroimmunol 131: 179–85 Watson R, Jiang Y, Bermudez I et al. (2004) Absence of antibodies to glutamate receptor type 3 (GluR3) in Rasmussen encephalitis. Neurology 63:43–50 Wiendl H, Bien CG, Bernasconi P et al. (2001) GluR3 antibodies: prevalence in focal epilepsy but no specificity for Rasmussen’s encephalitis. Neurology 57:1511–14 Alvarez-Baron E, Bien CG, Schramm J et al. (2008) Autoantibodies to Munc18, cerebral plasma cells and B-lymphocytes in Rasmussen encephalitis. Epilepsy Res 80:93–7 Takahashi Y, Mori H, Mishina M et al. (2003) Autoantibodies to NMDA receptor in patients with chronic forms of epilepsia partialis continua. Neurology 61:891–6 Yang R, Puranam RS, Butler LS et al. (2000) Autoimmunity to munc-18 in Rasmussen’s encephalitis. Neuron 28:375–83 Andrews PI, Dichter MA, Berkovic SF et al. (1996) Plasmapheresis in Rasmussen’s encephalitis. Neurology 46:242–6 Thilo B, Stingele R, Knudsen K et al. (2009) A case of Rasmussen encephalitis treated with rituximab. Nat Rev Neurol 5:458– 62 Bien CG, Bauer J, Deckwerth TL et al. (2002) Destruction of neurons by cytotoxic T cells: a new pathogenic mechanism in Rasmussen’s encephalitis. Ann Neurol 51:311–8 Owens GC, Erickson KL, Malone CC et al. (2015) Evidence for the involvement of gamma delta T cells in the immune response in Rasmussen encephalitis. J Neuroinflammation 12:134 Bauer J, Elger CE, Hans VH et al. (2007) Astrocytes are a specific immunological target in Rasmussen’s encephalitis. Ann Neurol 62:67–80 Li Y, Uccelli A, Laxer KD et al. (1997) Local–clonal expansion of infiltrating T lymphocytes in chronic encephalitis of Rasmussen. J Immunol 158:1428–37

41. Schneider–Hohendorf T, Mohan H, Bien CG et al. (2016) CD8(+) T-cell pathogenicity in Rasmussen encephalitis elucidated by largescale T-cell receptor sequencing. Nat Commun 7:11153 42. Schwab N, Bien CG, Waschbisch A et al. (2009) CD8+ T-cell clones dominate brain infiltrates in Rasmussen encephalitis and persist in the periphery. Brain 132:1236–46 43. Owens GC, Huynh MN, Chang JW et al. (2013) Differential expression of interferon-gamma and chemokine genes distinguishes Rasmussen encephalitis from cortical dysplasia and provides evidence for an early Th1 immune response. J Neuroinflamm 10:56 44. Takahashi Y, Mine J, Kubota Y et al. (2009) A substantial number of Rasmussen syndrome patients have increased IgG, CD4+ T cells, TNFalpha, and Granzyme B in CSF. Epilepsia 50:1419–31 45. Martin-Blondel G, Brassat D, Bauer J et al. (2016) CCR5 blockade for neuroinflammatory diseases: beyond control of HIV. Nat Rev Neurol 12:95–105 46. Trebst C, Staugaitis SM, Tucky B et al. (2003) Chemokine receptors on infiltrating leucocytes in inflammatory pathologies of the central nervous system (CNS). Neuropathol Appl Neurobiol 29:584–95 47. Vezzani A, French J, Bartfai T, Baram TZ (2011) The role of inflammation in epilepsy. Nat Rev Neurol 7:31–40 48. Cepeda C, Chang JW, Owens GC et al. (2015) In Rasmussen encephalitis, hemichannels associated with microglial activation are linked to cortical pyramidal neuron coupling: a possible mechanism for cellular hyperexcitability. CNS Neurosci Ther 21:152–63 49. Wagner AS, Yin NS, Tung S et al. (2012) Intimal thickening of meningeal arteries after serial corticectomies for Rasmussen encephalitis. Hum Pathol 43:1308–13 50. Hart YM, Cortez M, Andermann F et al. (1994) Medical treatment of Rasmussen’s syndrome (chronic encephalitis and epilepsy): effect of high–dose steroids or immunoglobulins in 19 patients. Neurology 44:1030–6 51. DeToledo JC, Smith DB (1994) Partially successful treatment of Rasmussen’s encephalitis with zidovudine: symptomatic improvement followed by involvement of the contralateral hemisphere. Epilepsia 35:352–5 52. McLachlan RS, Diosy D, Levin S (2011) Early treatment of a progressive Rasmussen’s like syndrome with ganciclovir. Can J Neurol Sci 38:296–8 53. Bien CG, Gleissner U, Sassen R et al. (2004) An open study of tacrolimus therapy in Rasmussen encephalitis. Neurology 62: 2106–9 54. Bittner S, Simon OJ, Gobel K et al. (2013) Rasmussen encephalitis treated with natalizumab. Neurology 81:395–7 55. Lagarde S, Villeneuve N, Trebuchon A et al. (2016) Anti– tumor necrosis factor alpha therapy (adalimumab) in Rasmussen’s encephalitis: An open pilot study. Epilepsia 57:956–66

Index

Note: Page numbers followed by f, t, and b indicate figures, tables, and boxes, respectively. abusive head trauma, 241. See also pediatric head injury aceruloplasminemia, 457t, 459 acquired immunodeficiency syndrome (AIDS), 520–521 acrodynia, 285 acute disseminated encephalomyelitis, 524 adrenoleukodystrophy, 381, 382f. See also peroxisomal disorders agenesis of corpus callosum (ACC), 36–37 clinical presentation, 36 definition, 36 differential diagnosis, 37 embryology and, 37 epidemiology, 36 genetics, 36 isolated, 36 neuroimaging, 36 pathogenesis and experimental models, 37 pathology, 36–37, 36f syndromic, 36–37 Aicardi–Gouti`eres leukoencephalopathy, 421 Alder–Reilly granules, 337 Alexander disease, 447–453 clinical features, 449–450, 449f differential diagnosis, 451 genetics, 447–449 pathogenesis, 452–453 pathology, 450–451, 451f allogenic bone marrow transplantation, 350 Alpers–Huttenlocher disease, 395t, 397–398, 398f–399f. See also mitochondrial disorders Alpers syndrome, 393 ambroxol, 332

amino acid disorders, 403–412 detection of, 404 homocystinuria, 406–408, 407f maple syrup urine disease, 409–410 neurological manifestations, 403 nonketotic hyperglycinemia, 405–406, 406f phenylketonuria, 404–405 prenatal diagnosis, 404 propionic and methylmalonic acidemias, 410–412 spongy myelinopathy, 403–404, 406f urea cycle disorders, 408–409, 408f–409f, 408t Amish infantile epilepsy syndrome, 357 amoebiasis, 526 amphetamines, 285, 287, 290 Anderson–Fabry disease, see Fabry disease anencephaly, 1, 17–18, 18f Angelman syndrome (15q11-13 deletion syndrome), 479, 481, 486 antenatal disruptive lesions, 199–201 anterior encephalocele, 20, 20f antiepileptic drugs adverse effects of, 180–181 during pregnancy, 46 aqueductal stenosis, 190–191, 191f, 514f aqueduct gliosis, 190 aqueduct of Sylvius, 190–191 arachnoid cysts, 146 arbovirus infections, 523–524 Arnold–Chiari malformations, 133 arrhinencephaly, 38 arteriovenous aneurysms, 251 arteriovenous malformations, 251, 257 arylsulfatase A, 332, 334 aseptic meningitis, 519

aspartyl glucosaminuria, 296 aspergillus infection, 524, 525f astrocytes, 231–232 atelencephaly/aprosencephaly, 35–36 definition and synonyms, 35 differential diagnosis, 35–36 genetics, 35 pathogenesis, 36 pathology, 35 general, 35 macroscopic neuropathology, 35 autism spectrum disorder (ASD), 477–490 clinical features, 477 epilepsy comorbidity, 477–478 in-utero infections, 478 maternal–fetal autoantibodies, 478 microbiome, 478–479 mitochondrial disease, 478 epidemiology, 477 and epileptic encephalopathy, 478, 478t genetic defects, 482–486, 484t–485t and intellectual disability, 484t–485t megalencephaly and microcephaly in, 480–481, 480f and mental restriction, 482, 483t pathology, 479–489 treatment and future perspective, 489–490 autosomal aneuploidies, 1–2, 3t tetraploidy, 3t, 6 triploidy, 3t, 5–6, 5f–6f trisomy 8, 2 trisomy 9, 2 trisomy 13, 2–3 trisomy 18, 3 trisomy 21, 3, 5, 5f autosomal recessive hydrocephalus, 192–193

Developmental Neuropathology, Second Edition. Edited by Homa Adle-Biassette, Brian N. Harding, and Jeffrey A. Golden. © 2018 John Wiley & Sons Ltd. Published 2018 by John Wiley & Sons Ltd.

537

Index autosomal recessive primary microcephaly (MCPH), 41, 43, 44t, 48f. See also microcephaly axial anomalies, 1 axonal sprouting, in hippocampal sclerosis, 102 Back-to-Sleep campaign, 271 bacterial infections, 513–518 bacterial meningitis, 513–515, 514f–515f Borrelia burgdorfi infection, 517 brain abscess, 515–516, 516f epidural spinal tuberculous abscesses, 517 intracranial epidural abscess, 515 Rickettsia ricketsii infection, 517–518 spinal epidural abscesses, 515 tuberculomas, 517 tuberculous meningitis, 516–517, 517f banana bodies, 349, 350f Baraitser–Winter syndrome, 66 benign external hydrocephalus, 244 berry aneurysms, 253, 255, 257 beta-propeller protein-associated neurodegeneration (BPAN), 457t, 459, 464 D-bifunctional protein deficiency (BPD), 381. See also peroxisomal disorders bilateral hemimegalencephaly, 55–60. See also dysplastic megalencephaly (DMEG) bilirubin encephalopathy, see kernicterus Blake’s pouch cyst (BPC), 141–149. See also Dandy–Walker malformation (DWM) bone marrow transplantation, 299, 345 bone morphogenetic protein (BMP), 29 Borrelia burgdorfi infection, 517 Botox injections, 434 brain abscess, 515–516, 516f brain development, in gestation period, 229–230 Cajal–Retzius neurons, 65, 67 Canavan disease, 412–413 advanced, 412 clinical features, 412 definition, 412 diagnosis, 412 epidemiology and genetics, 412 imaging, 412 neonatal/infantile, 412 pathogenesis, 413 pathology, 412–413, 413f treatment, 413 Candida albicans infection, 524–525, 525f Cantrell syndrome, 181 capillary telangiectasias, 253 carbidopa levodopa, 434 carbohydrate metabolism, disorders of, 293–308

538

congenital disorders of glycosylation, 304–308 lysosomal diseases, 293–299, 294t–295t polyglucosan disorders, 299–304, 300t catalase-positive neurons, 382 caudal regression syndrome, 182–183 cavernous hemangiomas, 252–253, 257–258, 259f. See also vascular malformations, CNS cebocephaly, 32 cell-based therapy, 315 cell death, in mitochondrial disorders, 400 ceramides, 313 ceramide synthase 1 deficiency, 356–357 ceramide synthase 2 deficiency, 357 cerebellar dysplasias, 159, 161 neurodevelopmental syndromes with, 161–162, 162t occipital encephalocele and, 162 septo-optic dysplasia and, 162 cerebellar heterotopia, 159 histologic changes, 159, 160f of infancy, 159–161 nodular, 161f cerebral heterotopia, 91–98. See also heterotopia cerebral palsy, 213–214 cerebrospinal fluid (CSF), 187. See also hydrocephalus production and circulation, 187–188 shunting, 194 Chiari malformation, 133–138 clinical features, 134–135 definition, 133 differential diagnosis, 137 embryology and, 133–134 epidemiology, 134 experimental models, 137–138 future directions and therapy, 138 genetics, 137 historical perspective, 133 imaging, 135 incidence and prevalence, 134 macroscopy and histopathology, 135–137 pathogenesis, 138 risk factors, 134 surgical management, 138 type 0, 133 type I, 133, 135, 135f type II, 133, 135–136, 136f type III, 133, 136–137, 137f type IV, 133 chickenpox, 503–504 cholesterol modulator, 348 choroid plexus, 187, 203 Christianson syndrome, 481, 488 chromosomal defects, in CNS anomalies, 2t

chromosomal structural microcephalies, 43, 46t, 48 cigarette smoking, 285 prenatal exposure to, 287, 290 ciliopathy, 19–20 circle of Willis, 252 CNTNAP2 syndrome, 489 cobblestone lissencephaly, see lissencephaly, type II cocaine, 285, 287, 290 Cockayne syndrome, 421, 427–434 animal models, 433–434 biochemistry, 429 cellular and molecular biology, 431–433 clinical presentation, 428–429, 428f definition, 427 differential diagnosis, 429–430 future perspectives, 434 genetics, 430–431 incidence and prevalence, 427 major criteria, 427 minor criteria, 427 neuroimaging, 429, 429f neuropathology, 430 pathogenesis, 433 pathology, 430 synonyms and historical perspective, 427 treatment, 434 types, 427 coenzyme A synthase protein-associated neurodegeneration (CoPAN), 457t, 459 COL4A1 mutation-associated small vessel disease, 255 coloboma, 5f congenital disorders of glycosylation (CDG), 293, 304–308 biochemistry, 307 clinical features, 305 definitions, 304 differential diagnosis, 306–307 epidemiology, 305 experimental models, 307 genetics, 305 histopathology, 306 immunohistochemistry and ultrastructural findings, 306, 306f macroscopy, 305–306 pathogenesis, 307–308 synonyms and historical annotations, 304–305 treatment, 308 Cornelia de Lange syndrome, 9 cortical dysplasias, 106–113 crack babies, 285 craniofacial complex, 1 craniorachischisis, 16–17, 17f, 22–23, 23f–24f Cree encephalopathy, 437 cri du chat syndrome, 7, 8t

Index Crigler–Najjar syndrome type I, 281 Cryptococcus neoformans infection, 524 Currarino syndrome, 181–182 cyclodextrins, 315 cyclopia, 31 cytokines, 223 cytomegalovirus, 521, 522f Dandy–Walker malformation (DWM), 2, 6f, 9, 141–149, 142f clinical features, 142–143 differential diagnosis, 146 embryology and, 146–147 epidemiology, 144 experimental models, 148 genetic models, 148–149 teratogen exposure, 148 features, 141 future investigations, 149 genetics, 144 histopathology cerebellum and brainstem, 146 cyst walls, 146 macroscopy, 144 cerebellar hypoplasia, 145 choroid plexus, 145 enlarged/cystic retrocerebellar space, 144–145, 145f enlarged posterior fossa, 145 hydrocephalus and foraminal occlusion, 145 morphometry, 145 other CNS anomalies, 145–146 somatic malformations, 146 neuroimaging, 143 children and adults, 143–144, 143f fetal, 144 pathogenesis, 147, 148f of cerebellar hypoplasia, 148 of cystic fourth ventricle, 147 of enlarged posterior fossa, 148 synonyms and historical annotations, 141 treatment, 149 deferiprone, 466 deletion syndromes, 6, 8t 3p- deletions, 6 4p- deletions, 6–7 5p- deletions, 7 9p- deletions, 7 18p- deletions, 9 11q- deletions, 7, 9 13q- deletions, 9 18q- deletions, 9 21q- deletions, 9 DeMorsier syndrome, 37. See also septo-optic dysplasia dental amalgams, 286 dentate bilamination, 273, 273f

dentate–olivary dysplasia, 167–168 clinical features and genetics, 168 definition, 167–168 embryology and pathogenesis, 168 histopathology, 168 macroscopy, 168 dentato–olivary dysplasia with intractable seizures in infancy, 168–170 clinical features, 169 definition, 168 differential diagnosis, 170 epidemiology, 168 genetics, 169 histopathology, 169, 169f historical annotation, 168 imaging, 169 laboratory findings, 169 macroscopy, 169 therapy and future directions, 170 diastematomyelia, 179, 181f. See also spinal cord lesions diffuse cerebral white matter gliosis (DWMG), 213–214 diffuse traumatic axonal injury, 244–245 diffuse white matter injury (DWMI), 213–224 clinical features, 215–216 cystic PVL, 213, 215, 216f differential diagnosis, 218–219 experimental models, 220 genetics, 215 histopathology, 216–217, 217f historical annotations, 213–214 imaging, 215–216 immunohistochemistry, 217–218, 219f incidence and prevalence, 214 macroscopy, 216, 216f–217f pathogenesis, 220–223, 220t–221t, 222f risk factors, 214–215 sex and age distribution, 214 treatment and future perspective, 223–224 dimyelia, 179, 180f. See also spinal cord lesions diplomyelia, 20, 21f, 179, 182f. See also spinal cord lesions Down syndrome, see trisomy 21 duplication of central canal of spinal cord, 180 duplications, 9, 10t dysplasias of dentate and olivary nuclei, see dentate–olivary dysplasia dysplastic megalencephaly (DMEG), 55 clinical features, 55–56 epidemiology, 55 experimental models, 59 genetics, 55 histopathology, 57–59, 58f imaging studies, 56–57, 57f immunohistochemistry, 59 macroscopy, 57, 58f

pathogenesis, 59 therapies and future investigations, 59–60 Ebola virus, 524 Edwards syndrome, see trisomy 18 eliglustat tartrate, 332 embryology of forebrain patterning, 29–30, 30f encephalomyopathies, mitochondrial, 393 encephalopathy of prematurity, 213 encephalotrigeminal angiomatosis, see Sturge–Weber–Dimitri syndrome enteroviral infections, 519–520 enteroviruses, 504 enterovirus polioencephalomyelitis, in older children, 519–520 enzyme replacement therapy, 299, 316, 328, 345 Gaucher disease, 332 epidural spinal tuberculous abscesses, 517 epilepsy-related hippocampal sclerosis, 102. See also hippocampal sclerosis erethism, 285 erythropoietin, 223 ethmocephaly, 32 eukaryotic initiation factors (eIFs), 444 excitotoxicity, 222 exencephaly, 17, 23, 23f, 25f extra-axial fluid collection, 244 extradural hematomas, 243 Fabry disease, 325–328 biochemistry, 326 definition, 325 differential diagnosis, 326 epidemiology, 325 genetics, 327–328 neuroimaging, 325–326 pathogenesis and animal models, 328 pathology, 326–327, 327f signs and symptoms, 325 synonyms and historical perspective, 325 treatment and future perspective, 328 Farber disease, 348–350 biochemistry, 349 definition, 348 differential diagnosis, 349 epidemiology, 348 genetics, 349 neuroimaging, 349 pathogenesis and animal models, 349–350 pathology, 349, 350f signs and symptoms, 348–349 treatment and future perspective, 350 Farber’s lipogranulomatosis, see Farber disease fatty acid hydroxylase-associated neurodegeneration (FAHN), 457t, 459, 465

539

Index fatty acid 2-hydroxylase deficiency (FA2H deficiency), 356 fetal alcohol syndrome, 46, 285–291 fetal tobacco exposure, 287 focal cortical dysplasias (FCD), 106–113 animal models, 112–113 clinical features, 107 definition, 106 differential diagnosis, 108 epidemiology, 107 future perspectives, 113 genetics and pathogenesis, 111–112 ILAE classification of, 106, 107t imaging and electrophysiology, 107–108 isolated, 107 normal embryology and, 106–107 pathology, 108–111, 109f seizures with, 107 surgical treatments and outcome, 113 synonyms, 106 type I, 107t, 108, 109f type II, 107t, 108–110, 109f type III, 107t, 110–111, 111f–112f folic acid, neural tube defects and, 26 forebrain patterning, 29–30, 30f Fowler syndrome, 254 FOXG1 syndrome, 488–489 fucosidosis, 296 Fukuyama congenital muscular dystrophy (FCMD), 75. See also lissencephaly, type II fungal infections, 524–525 GABAergic neurons, 229 galactosialidosis, 296, 313 galactosylceramidase deficiency, 339 Gaucher disease, 328–332 animal models and pathogenesis, 331–332 biochemistry, 329 definition, 328 epidemiology, 328 genetics, 331 neuroimaging, 329 pathology, 329–331, 330f signs and symptoms, 328–329 synonyms and historical perspective, 328 treatment and future perspective enzyme replacement therapy, 332 pharmacologic chaperone therapy, 332 substrate reduction therapy, 332 gene therapy, 315, 336, 350, 424 genetic counseling, 1 gene transfer, 299 gene trap technology, 22 Gilbert syndrome, 281 Glasgow Coma Scale (GCS), 242 glial fibrillary acidic protein (GFAP), 275, 447, 448f

540

gliding contusions, 244, 244f globoid-cell leukodystrophy, 338–342 biochemistry, 339 definition, 338 differential diagnosis, 339 epidemiology, 338 genetics, 341 infantile form, 338–339 later-onset, 339 neuroimaging, 339 pathogenesis and animal models, 341 pathology, 339–341, 341f signs and symptoms, 338–339 synonyms and historical perspective, 338 treatment and future perspective, 342 globoid cells, 340–341 glucose-6-phosphate-dehydrogenase (G6PD) deficiency, 281 glycine, 406 glycine encephalopathy, see nonketotic hyperglycinemia glycolipid storage, 326 GM1 gangliosidosis, 318–321 biochemistry, 319 definition, 318 differential diagnosis, 319 epidemiology, 318 genetics, 319–320 histopathology, 319, 320f infantile type I, 318 macroscopy, 319 neuroimaging, 318–319 magnetic resonance imaging, 318 magnetic resonance spectroscopy (MRS), 318–319 pathogenesis and animal models, 320 signs and symptoms, 318 synonyms and historical perspective, 318 treatment and future perspective, 320–321 type II, 318 type III, 318 GM2 gangliosidosis, 321–325 biochemistry, 323 definition, 321–322 differential diagnosis, 323 epidemiology, 322 genetics, 324 histopathology, 323–324 macroscopy, 323 neuroimaging, 322–323 pathogenesis and animal models, 324 signs and symptoms, 322 synonyms and historical perspective, 322 treatment and future perspective, 324–325 GM2 synthase deficiency, 357–358 GM3 synthase deficiency, 357 granular ependymitis, 189

granule cell aplasia, 174–175 definition and synonyms, 174 embryology, 174 epidemiology and clinical features, 174 genetics, 175 pathogenesis and animal models, 175 pathology, 174–175 granule cell dispersion, 101–106, 273 animal models, 106 clinical features, 102 epidemiology, 101 genetics and pathogenesis, 106 normal embryology and, 101 pathology, 102, 104, 104f treatment and future perspectives, 106 granule cell dysplasia, see granule cell dispersion gray matter lesions, 229–236. See also hypoxic–ischemic encephalopathy (HIE) hemangiomas, 253 hematopoietic stem cell transplantation, 315, 336 hemimegalencephaly (HMEG), 55–60 clinical features, 55–56 epidemiology, 55 experimental models, 59 genetics, 55 histopathology, 57–59, 58f imaging studies, 56–57, 57f immunohistochemistry, 59 macroscopy, 57 occipital sign, 57 pathogenesis, 59 and seizures, 56 therapies and future investigations, 59–60 hemimegamyelia, 180, 181f. See also spinal cord lesions hemorrhagic lesions, 203–209 animal models, 208 clinical features, 204–205 differential diagnosis, 205 genetics, 208 histopathology, 206–207, 207f–208f imaging, 205, 205f immunohistochemistry and ultrastructure, 207 incidence and prevalence, 204 intraventricular hemorrhage, 203 laboratory findings, 205 macroscopy, 205–206, 206f normal development and, 203, 204f pathogenesis, 207–208 periventricular hemorrhage, 203 risk factors, 204 sex and age distribution, 204 treatment and future perspective, 208–209

Index herpes simplex virus infection, 499–500, 500f, 518–519, 519f heterotopia, 91–98 animal models, 96–97 clinical features, 93–94 differential diagnosis, 95 genetics, 95–96 histopathology, 94–95 imaging, 94 incidence and prevalence, 92 leptomeningeal, 91, 92f macroscopy, 94 non-genetic mechanisms in, 98 normal embryology and, 91 pathogenesis cellular mechanisms, 97–98 genetic mechanisms, 97 molecular mechanisms, 98 periventricular, 91, 93f risk factors, 93 sex and age distribution, 92 subcortical-band, 91, 93f synonyms, 91 treatment and future perspectives, 98 hippocampal abnormalities, 273–275, 274f hippocampal formation maldevelopment in SUDP (HFM-SUDP), 274 hippocampal sclerosis, 101–106, 233 animal models, 106 clinical features, 102 epidemiology, 101 genetics and pathogenesis, 104, 106 International LeagueAgainst Epilepsy (ILAE) scheme for, 102, 105t normal embryology and, 101 pathology, 102, 103f with temporal-lobe epilepsy, 102 treatment and future perspectives, 106 HIV infection, 520–521 holomyelia, 180. See also spinal cord lesions holoprosencephaly (HPE), 1, 2f, 30–35 alobar, 32, 33f classic, 30 clinical presentation, 31 definition, 30 differential diagnosis, 34 epidemiology, 30–31 future directions and therapy, 35 genetics, 31 historical annotations, 30 HPE complex, 30 incidence, 30 lobar, 32, 33f microform, 30, 32 middle interhemispheric variant, 32, 34f neuroimaging, 31 pathogenesis and experimental models, 34–35

pathology, 31 general, 31–32 histopathology, 32, 34 macroscopic neuropathology, 32 risk factors, 30–31 semilobar, 32, 33f syndromes associated with, 31 syntelencephaly, 30 homocystinuria, 406–408, 407f human cytomegalovirus (HCMV) infection, 497–499, 499f human immunodeficiency virus (HIV), 504 hydatid disease, echinococcosis, 525–526 hydranencephaly, 199–201, 199f hydrocephalus, 187–194 animal models, 193–194 autosomal recessive, 192–193 causes, 189–190 clinical features, 188 communicating, 187 congenital, 188, 190f CSF overproduction and, 187 definition, 187 detection of, 188, 189f epidemiology, 188 genetics, 192 non-communicating, 187 nonsyndromic, 192 pathology, 190 aqueductal stenosis, 190–191, 191f atresia and/or forking, 191, 191f gliosis, 190 pathophysiology, 188–189 risk factors, 188 syndromic, 192 treatment, 194 X-linked, 192, 193f hydrocephalus ex vacuo, 187, 319 hydromyelia, 20, 21f, 179, 180f. See also spinal cord lesions hyperammonemia, 409 hyperlactatemia, 394 hyperphenylalaninemias (HPAs), 404 hypomyelinating leukodystrophies, 423, 423t hypoxic–ischemic encephalopathy (HIE), 229 animal models, 235–236 clinical features, 230 epidemiology, 230 MRI studies, 230 pathogenesis and genetics, 234–235 pathology, 230–232, 231f patterns of damage in basal nuclei and thalamic damage, 233–234 cerebellum, brainstem, and spinal cord damage, 234

hippocampal damage, 233 neocortical damage, 232–233 risk factors, 230 treatment and future perspective, 236 hypoxic–ischemic lesions, in brain, 275–276 imiglucerase, 332 infantile onset spinocerebellar ataxia (IOSCA), 395t infantile Refsum disease, 381, 381t. See also peroxisomal disorders infantile sialic storage disease, 296 intracranial epidural abscess, 515 intrauterine infections, 497–506 cytomegalovirus, 497–499, 499f differential diagnosis, 506 enteroviruses, 504 fetal meningitis, 506 herpes simplex virus, 499–500, 500f human immunodeficiency virus, 504 listeriosis, 505 parvovirus B19, 504 rubella, 502–503 rubeola, 506 syphilis, 505 toxoplasmosis, 500–501, 502f tuberculosis, 505 varicella, 503–504 Zika virus, 504–505, 505f intraventricular hemorrhage (IVH), 203. See also hemorrhagic lesions isofagomine, 332 Jacobsen syndrome, 7, 8t, 9 Joubert syndrome, 146, 151–156 brain malformations in, 151 definition, 151 differential diagnosis, 155 epidemiology, 151 genetics, 155 allelic Joubert syndrome-related disorders, 156 genotype–phenotype correlations, 155–156 inheritance and genes, 155 neuroimaging, 152 fetal neuroimaging, 153 molar tooth sign, 152–153, 152f other brain malformations, 153 neurological manifestations, 151 neuropathology, 153 brainstem, 154–155 cerebellum, 153–154, 153f–154f cerebral hemispheres and diencephalon, 155 dorsal medullary protuberance, 155 spinal cord, 155 non-neurologic features, 151–152

541

Index Joubert syndrome (Continued) pathogenesis animal models, 156 primary cilia, 156 treatment, 156 Kallmann syndrome., 38 Kearns–Sayre syndrome, 393–394, 396, 396t, 397f. See also mitochondrial disorders kernicterus, 281–283 biochemistry, 283 clinical features, 281–282 differential diagnosis, 283 epidemiology, 281 future directions and therapy, 283 genetics, 281 histopathology, 283, 283f macroscopy, 282, 282f predisposing conditions, 281 ketogenic diet, 59–60 Kleefstra syndrome, 487–488 Klinefelter syndrome, 7t Klippel–Feil syndrome (KFS), 182 Koolen–De Vries syndrome, 487 Krabbe disease, 315. See also globoid-cell leukodystrophy Kufor–Rakeb syndrome, 455 Lafora bodies, 300, 300f, 303–304 Lafora disease, 300–303 lead toxicity, 285–291 Leigh syndrome, 393–394, 395t, 396f–397f. See also mitochondrial disorders Lennox–Gastaut syndrome, 488 leptomeningeal heterotopia, 91, 92f. See also heterotopia Lhermitte–Duclos disease, 182 lipomyelomeningocele, 20 lissencephaly, type I, 63–71 clinical presentation, 66 differential diagnosis, 68 embryology, 64–65, 64f experimental models and pathogenesis, 68–71 future directions and therapy, 71 genetics, 65–66 histopathology, 67–68, 68f imaging, 66 immunohistochemistry, 68 incidence, 63 macroscopy, 66–67, 67f risk factors, 63 sex and age distribution, 63 synonyms and historical annotations, 63 lissencephaly, type II, 75–82 clinical features, 76–77 differential diagnosis, 80–81 embryology, 77, 78f

542

epidemiology, 75 future directions and therapy, 82 genes associated with, 75–76, 76t histopathology, 79–80, 79f–80f imaging, 77 macroscopy, 77–79 pathogenesis, 81 synonyms, 75 listeriosis, 505 Little disease, 213 L1 syndrome, 192 lymphangioleiomyomatosis, 119. See also tuberous sclerosis complex lysosomal diseases, 293–299, 294t–295t biochemistry, 298–299 clinical features, 296–297 differential diagnosis, 299 embryology, 296 epidemiology, 296 future directions and therapy, 299 genetics, 296 histopathology, 297–298 immunohistochemistry and ultrastructural findings, 298 macroscopy, 297 morphological features, 293 pathogenesis, 299 synonyms and historical annotations, 293, 296 mad hatter’s disease, 285 magnesium sulfate, 236 malaria, 526, 526f malformative atresia–forking of central canal of spinal cord, 180, 181f Management of Myelomeningocele Study (MOMS), 25 maple syrup urine disease (MSUD), 409–410 brain damage in, 410 clinical types, 410 marijuana, 285, 287, 290 maternal diabetes mellitus, and spinal lesions, 181 measles inclusion body encephalitis, 523, 523f measles infections, 521–523, 523f Meckel–Gruber syndrome, 19–20, 22, 151 MECP2 duplication syndrome, 486 MEF2C syndrome, 489 mega-cisterna magna (MCM), 141–149, 143f. See also Dandy–Walker malformation (DWM) megalencephaly capillary malformation (MCAP), 55 megalencephaly polymicrogyria polydactyly hydrocephalus (MPPH), 55 meningioangiomatosis, 254, 261, 262f. See also vascular malformations, CNS meningocele, 20

mercury intoxication, 285–291 mesenchymal stem cell transfusion, 299 metachromatic leukodystrophy (MLD), 332–337 adult form, 333 biochemistry, 334 differential diagnosis, 336 epidemiology, 332 genetics, 332–333 juvenile, 333 late infantile, 333 neuroimaging, 333 pathogenesis and animal models, 335–336 pathology and histochemistry, 333–334, 334f–335f signs and symptoms, 333 treatment and future perspective, 336 metalloporphyrins, 283 methylmalonic acidemia, 410–412, 411f–412f methylmercury, 288 microcephaly, 41–52, 504, 505f animal models and pathogenesis, 51–52 autosomal recessive primary, 41, 43, 44t, 48f brain size, decrease in, 41 chromosomal structural, 43, 46t clinical features, 43 definition and synonyms, 41 differential diagnosis, 46, 48, 51 dwarfism and DNA repair deficiency syndromes, 43, 45t epidemiology, 42–43 genetics, 43–44 imaging studies, 43 and microlissencephaly, 46 monogenic syndromic, 43, 47t neuropathology, 44, 46, 49f–51f normal embryology and, 41–42, 42f primary, 41 secondary, 43–44 treatment and future perspectives, 52 microglia, 223 micro-peroxisomes, 382 midline patterning defects, 29–38 agenesis of corpus callosum, 36–37 atelencephaly/aprosencephaly, 35–36 holoprosencephaly, 30–35 septo-optic dysplasia, 37–38 migalastat HCl, 328 miglustat, 332 Miller–Dieker syndrome, 66, 167 Minamata disease, 285 mitochondrial disorders, 393–401 animal models, 400–401 biochemistry, 393–394 clinical features, 394, 395t–396t epidemiology, 393 genetics, 400 neuropathology, 394

Index Alpers–Huttenlocher disease, 395t, 397–398, 398f–399f central nervous system, 394 Kearns–Sayre syndrome, 394, 396, 396t, 397f Leigh syndrome, 394, 395t, 396f–397f MELAS syndrome, 395t, 396–397 MERRF syndrome, 395t, 397–398 peripheral nervous system, 399 skeletal muscle, 399, 399f pathogenesis, 400 prenatal manifestations of, 393 in preschool children, 393 synonyms and historical perspective, 393 treatment and future perspective, 401 mitochondrial encephalomyopathy, lactic acidosis and stroke-like episodes (MELAS), 394, 395t, 396–397. See also mitochondrial disorders mitochondrial membrane protein-associated neurodegeneration (MPAN), 456t, 459 mitochondrial neurogastrointestinal encephalopathy (MNGIE), 396t mitochondrial spinocerebellar ataxia and epilepsy (MSCAE), 396t M¨obius syndrome, 170–171, 287 clinical features, 170 definition, 170 differential diagnosis, 171 experimental models, 171 genetics, 170 histopathology, 170–171, 170f treatment, 171 molar tooth sign, 146, 152–153, 152f monogenic syndromic microcephalies, 43, 47t Morquio B disease, 318 Mowat–Wilson syndrome, 488 mucolipidoses, see lysosomal diseases mucopolysaccharidoses, see lysosomal diseases multicystic encephalopathy, 199–201, 200f multiple sulfatase deficiency (MSD), 334, 336–338 biochemistry, 337 definition, 336–337 differential diagnosis, 337 epidemiology, 337 genetics, 338 neuroimaging, 337 pathogenesis and animal models, 338 pathology and histochemistry, 337 signs and symptoms, 337 treatment gene therapy, 338 microRNA (miR-95), 338 muscle–eye–brain disease, 75. See also lissencephaly, type II myelocystocele, 179. See also spinal cord lesions

myelomeningocele, 18, 19f, 24–25, 26f myocerebrohepatopathy, 398 myoclonus epilepsy and ragged red fibers (MERRF), 394, 395t, 397–398. See also mitochondrial disorders National Institute of Child Health and Human Development (NICHD), 269 neonatal adrenoleukodystrophy, 381, 381t neonatal bacterial meningitis, 513 neonatal encephalopathy, 229. See also hypoxic–ischemic encephalopathy (HIE) neonatal glycine encephalopathy, 405 neonatal herpes virus infection, 518 neonatal jaundice, 281 neural tube defects (NTDs), 13 animal models and pathogenesis, 22 craniorachischisis, 23, 23f–24f exencephaly, 23, 23f, 25f myelomeningocele, 24–25, 26f axial mesodermal defects with neural tube herniation, 18 anterior encephalocele, 20, 20f Meckel–Gruber syndrome, 19–20 meningocele, 20 occipital encephalocele, 18–19, 19f parietal encephalocele/meningocele, 20 clinical presentation, 16 differential diagnosis, 16 epidemiology, 15–16 genetics, 21 human genetic studies, 21, 22f mouse genetic studies, 21–22 of occipital encephalocele, 22 neural tube closure disorders anencephaly, 17–18, 18f craniorachischisis, 16–17, 17f exencephaly, 17 myelomeningocele, 18, 19f non-genetic risk factors, 15–16 normal embryology and, 13–15 prenatal diagnosis, 16 prevalence, 15 primary prevention, 26 sex distribution, 15 surgical treatment, 25–26 tail bud development, defects of, 20, 21f tethered cord, 20–21 neuroaxonal dystrophy, 347 neurocysticercosis, 525 neurodegeneration with brain iron accumulation (NBIA), 455–466 animal models, 465–466 clinical features, 455–459, 456t–457t genetics, 462, 464–465 pathology, 459–462, 460t, 461f neuroferritinopathy, 456t, 459

neurofibrillary tangles, 344f, 347, 523 neuronal ceroid lipofuscinosis (NCL), 369–379, 370f–371f, 372t CLN1, 372–374 CLN2, 374–376 CLN3, 376–377 CLN4, 377 CLN5, 377–378 CLN6, 378 CLN7, 378 CLN8, 378–379 CLN10, 379 forms, 369 neuropathologists, 1 Niemann–Pick disease types A and B, 342–345 biochemistry, 343 definition, 342 differential diagnosis, 343 epidemiology, 342 genetics, 343, 345 neuroimaging, 343 pathogenesis and animal models, 345 pathology, 343 signs and symptoms, 342–343 treatment and future perspective bone marrow transplantation, 345 enzyme replacement therapy, 345 Niemann–Pick type C disease (NPC), 313, 345–348 biochemistry, 346 definition, 345 differential diagnosis, 346–347 epidemiology, 345–346 genetics, 347 neuroimaging, 346 pathogenesis and animal models, 347–348 pathology, 347 signs and symptoms, 346 synonyms and historical perspective, 345 treatment cholesterol modulator, 348 substrate reduction therapy, 348 types of, 345 Nodal signaling, 29 nonketotic hyperglycinemia, 405–406, 406f NRXN1 syndrome, 489 occipital encephalocele, 18–19, 19f, 137 genetics of, 22 Ohtahara syndrome, 56, 170 oligodendrogliomatosis, 439 oligosaccharidosis, see lysosomal diseases olivary heterotopia, 167 clinical features and genetics, 167 embryology and pathogenesis, 167 histopathology, 167, 168f opiates toxicity, 285–286, 288f opiate withdrawal syndrome, 287

543

Index orexin-related abnormalities, in hypothalamus, 275 ornithine transcarbamylase (OTC) deficiency, 408 ovarioleukodystrophy, 437 oxidative phosphorylation (OXPHOS) system, 393 pachygyria, 63, 67–68 pantothenate kinase-associated neurodegeneration (PKAN), 455, 456t, 458, 458f, 461f parietal encephalocele/meningocele, 20 parvovirus B19, 504 Patau syndrome, see trisomy 13 pediatric head injury, 241–247 abusive head trauma, 245–247 accidental head injury, 242–243 diffuse traumatic axonal injury, 244–245 extradural hematoma, 243 parasagittal white matter lesion, 244, 244f subdural hematoma, 243–244, 243f animal models, 247 clinical features, 242 epidemiology, 241–242 pathologies associated with, 241 perinatal non-abusive head injury, 242 Pelizaeus–Merzbacher disease (PMD), 417–424 biochemistry, 421 clinical features, 419 differential diagnosis, 421 epidemiology, 417 experimental models, 421 in females, 423 future direction and therapy, 424 genetics, 417, 418f duplication analysis, 418–419, 419f PLP1 gene, 417–418 histopathology, 420, 420f–421f imaging, 419 immunohistochemical and ultrastructural findings, 420–421 laboratory findings, 419–420 macroscopy, 420, 420f pathogenesis, 421–423 pentasomy X, 7t perinatal arterial ischemic stroke, 230 perinatal extradural hemorrhage, 242 perinatal head injury, 242 perinatal and postnatal infections, 511–527 bacterial infections, 513–518 CNS infections, 511 fungal infections, 524–525 host factors, 512 parasitic infections, 525–527 routes of infections, 512 viral infections, 518–524 perinatal spinal cord injury, 242

544

periventricular hemorrhage (PVH), 203. See also hemorrhagic lesions periventricular hemorrhagic infarction, 213 periventricular heterotopia, 91, 93f. See also heterotopia periventricular leukomalacia (PVL), 213, 275. See also diffuse white matter injury (DWMI) peroxisomal biogenesis disorders (PBD), see peroxisomal disorders peroxisomal disorders, 381–389 animal models, 388–389 biochemistry and laboratory findings, 383–384 differential diagnosis, 384 epidemiology, 382 genetics and pathogenesis, 387–388 histopathology and immunohistochemistry, 384–387, 385f–387f macroscopy, 384, 384f–385f neuroimaging, 382–383, 383f normal embryology and, 382 organ systems affected, 381 peroxisomal biogenesis disorders (PBD), 381, 381t signs and symptoms, 382 single protein deficiencies, 381, 381t treatment and future perspective, 389 peroxisomes, in brain, 382 pharmacological chaperone therapy, 316, 320–321, 332 Phelan–McDermid syndrome (22q13.3 deletion), 487 phenylalanine hydroxylase (PAH), 404 phenylketonuria (PKU), 404–405 phospholipase A2-associated neurodegeneration (PLAN), 456t, 458–459 phototherapy, 283 PI3K-AKT-MTOR pathway, 55, 59 PI3K inhibitors, 60 Pitt–Hopkins syndrome, 488 plumbism, 285 Poland anomaly, 170 polyglucosan disorders, 299–304, 300f, 300t biochemistry, 303 clinical features, 301–302 differential diagnosis, 303 epidemiology, 301 experimental models, 303 future directions and therapy, 304 genetics, 301 histopathology, 302–303 immunohistochemistry and ultrastructural findings, 303 macroscopy, 302 pathogenesis, 303–304 synonyms and historical annotations, 301

polymicrogyria, 68, 85–89 clinical features, 85–86 differential diagnosis, 88 experimental models, 88 future directions and therapy, 89 genetics, 85, 86t histopathology, 87–88, 87f, 499f imaging, 86 incidence and prevalence, 85 laboratory findings, 86–87 macroscopy, 87, 87f pathogenesis, 88–89, 88f perisylvian, 85 and polygyria, 88 risk factors, 85 sex and age distribution, 85 Pompe disease, 296 pontine tegmental cap dysplasia, 171–172 clinical features and investigation, 171 definition, 171 embryology, 171 genetics and pathogenesis, 172 pathology, 171–172, 172f pontocerebellar hypoplasia (PCH), 172–174 animal models, 174 clinical features, 172 definition and synonyms, 172 differential diagnosis, 172 embryology, 172 epidemiology, 172 genetics, 173, 174t historical perspective, 172 pathogenesis, 173–174 pathology, 172–173, 173f treatment, 174 type 1 (PCH1), 172 type 2 (PCH2), 172, 173f pontosubicular necrosis, 233 porencephaly, 199–201, 200f, 233 posterior fossa extradural hematomas, 243 preeclampsia, 2 Pregnavite Forte F, 26 progressive multifocal leukoencephalopathy, 521 progressive rubella panencephalitis, 523 proliferative vasculopathy and hydranencephaly, 254 propionic and methylmalonic acidemias, 410–412, 411f–412f prosaposin (PSAP) deficiency, 352–355, 354f pseudo-Hurler polydystrophy, 296 pseudoperoxisomal biogenesis disorders, 381, 381t. See also peroxisomal disorders Purkinje cell abnormalities, 175 rabies, 524 radial glia, 64 rapamycin analogs, 60

Index Rasmussen encephalitis, 531–535 clinical features, 531 differential diagnosis, 534 neuropathology, 532–533, 532f–534f pathogenesis, 534–535 therapy, 535 reactive nitrogen species (RNS), 218, 220–222 reactive oxygen species (ROS), 218, 220–223 reelin, 70 Reissner’s fiber, 188 Rethor´e syndrome, 9, 10t retinal hemorrhages, 247 Rett syndrome, 486 rhizomelic chondrodysplasia punctata, type 1, classical, 381, 381t. See also peroxisomal disorders rhombencephalosynapsis (RES), 162–164, 163f animal models, 164 clinical features, 164 definition, 162 embryology and, 162 epidemiology, 162 genetics, 164 neuroimaging, 164 pathology, 164 treatment, 164 Rickettsia ricketsii infection, 517–518 rubella, 502–503 Salla disease, 296 Sandhoff disease, 322. See also GM2 gangliosidosis Schindler disease, 299 schizencephaly, 199, 232 Scholz disease, see metachromatic leukodystrophy (MLD) sensory ataxic neuropathy with ophthalmoparesis (SANDO), 396t septo-optic dysplasia, 37–38, 38f, 233 clinical presentation, 37 definition and synonyms, 37 differential diagnosis, 37–38 embryology, pathogenesis and experimental models, 38 genetics, 37 pathology, 37 serine palmytoyl transferase (SPT), 355 sex chromosome aneuploidy, 6, 7t shaken baby syndrome, 241. See also pediatric head injury shaking-impact syndrome, 241. See also pediatric head injury sialidosis, 297, 313 skull fractures, 242 Smith–Lemli–Opitz syndrome, 31 sodium benzoate, 406 somatic cell therapy, 424 sonic hedgehog (SHH), 29 Spemann organizer, 29

sphingolipid activator protein deficiency, 350–351 Sap-A deficiency, 351 Sap-B deficiency, 351–352 Sap-C deficiency, 352 Sap-D deficiency, 352 sphingolipid activator proteins (SAPs), 350–351, 351f sphingolipid biosynthesis deficiencies, 355–356 sphingolipidoses, 313 diagnosis of, 314 disease-causing variants, identification of, 314 functional study, 314 metabolites, measurement of, 314 treatments of, 314, 317f enzyme replacement therapy, 316 gene therapy and cell-based therapy, 315 hematopoietic stem cell transplantation, 315 pharmacological chaperone therapy, 316 storage material, removal of, 315 substrate-reduction therapy, 315 sphingolipids, 313 catabolism, 313, 314t metabolism, 314t, 315f–316f synthesis, 313, 314t spinal cord lesions, 179–185 clinical features, 183 definitions and epidemiology, 179–180 differential diagnosis, 185 genetics, 181–183 histopathology, 183–185, 184f imaging, 183 macroscopy, 183 pathogenesis, 185 risk factors, 180–181 treatment, 185 spinal dysraphism, 20–21, 179 spinal epidural abscesses, 515 spinal muscular atrophy, 469–474 Brown–Vialetto–van Laere syndrome, 474 with cerebellar hypoplasia, 474 Fazio–Londe disease, 474 genetics, 469–470 muscle pathology, 471, 471f neuropathology, 471–472, 471f–472f pathogenesis, 473–474 with respiratory distress type 1, 474 SMA type 1 (Werdnig–Hoffman disease), 470 SMA type 2, 470 SMA type 3 (Kugelberg–Welander disease), 470–471 split notochord syndrome, 179, 183. See also spinal cord lesions status marmoratus, 233

stem cell-based therapy, 223 stillbirth, 230 Sturge–Weber–Dimitri syndrome, 251–252, 254–255, 256f, 262, 263f. See also vascular malformations, CNS subacute sclerosing panencephalitis, 522–523, 523f subarachnoid space, benign enlargement of, 244 subcommissural organ, 187–188 subcortical-band heterotopia, 91, 93f. See also heterotopia subdural hematoma, 243–244, 243f acute, 243 chronic, 243–244, 243f subependymal giant cell astrocytomas (SEGAs), 117, 119f, 122, 124f. See also tuberous sclerosis complex substrate-reduction therapy, 315, 332, 348 sudden infant death syndrome (SIDS), 269–277 definition, 269–270 future directions, 277 genetics, 272 hypoxic–ischemic lesions in brain, 275–276 incidence and prevalence, 270 pathogenesis, 276–277 pathology, 272 brainstem abnormalities, 272–273 cerebral white matter pathology, 275 heavy brain weight, 275 hippocampal abnormalities, 273–275, 274f hypothalamic abnormalities, 275 risk factors, 271–272 triple risk model, 271, 271f sex and age distribution, 270–271 sudden unexplained death in pediatrics (SUDP), 274 syphilis, 505 syringobulbia, 134 syringomyelia, 179. See also spinal cord lesions taliglucerase alfa, 332 Tay–Sachs disease, 322. See also GM2 gangliosidosis tectocerebellar dysraphia, 162 terminal uridine nucleotide end labeling (TUNEL), 231 tethered cord syndrome, 20–21 tethered spinal cord, 179, 183 tetraploidy, 3t, 6 tetrasomy X, 7t thanatophoric dysplasia, 168 therapeutic hypothermia, 223 thimerosal, 286 toll-like receptors, 223 total hemimegalencephaly, 55. See also hemimegalencephaly (HMEG)

545

Index toxins, lesions induced by, 285–291 alcohol, 285 clinical features, 286–287 differential diagnosis, 289–290 experimental models, 290 future directions and therapy, 291 genetics, 286 histopathology, 288–289, 288f–289f illegal substances, 285–286 imaging findings, 287 incidence and prevalence, 285–286 laboratory findings, 287–288 lead, 286–287, 289f macroscopy, 288 mercury, 286–287, 289f nicotine, 285 pathogenesis, 290 toxoplasmosis, 500–501, 502f, 526, 527f transcription-coupled nucleotide excision repair (TC-NER), 432 transverse myelitis, 524 traumatic axonal injury, 244 triploidy, 3t, 5–6, 5f–6f trisomic cell, 1 trisomy 8, 2, 3t trisomy 9, 2, 3t trisomy 13, 2–3, 3t, 4f–5f trisomy 18, 3, 3t trisomy 21, 3, 3t, 5, 5f trisomy X, 7t trypanosomiasis, 526–527 tuberculomas, 517 tuberculosis, 505 tuberculous meningitis, 516–517, 517f tuberous sclerosis complex, 117–129 brain abnormalities in, 117 clinical features, 118–119 cortical tuber, 121, 122f–124f and subependymal nodules, 121f criteria for clinical diagnosis of, 119b definition, 117 differential diagnosis, 126 embryology and, 118 epidemiology, 117–118 experimental models, 126–127, 127f future directions and therapy, 128–129 genetics, 118 histopathology, 121–125, 122f

546

immunohistochemistry findings, 125–126 incidence and prevalence, 117 laboratory findings, 120 macroscopy, 120–121 neuroimaging, 119–120, 119f–120f pathogenesis, 127–128 risk factors, 118 sex and age distribution, 118 synonyms and historical annotations, 117 ultrastructural features, 126 Turner syndrome, 7t type I lissencephaly, see lissencephaly, type I type II lissencephaly, see lissencephaly, type II Ullrich-Turner syndrome, 7t unilateral megalencephaly, see hemimegalencephaly (HMEG) urea cycle disorders, 408–409, 408f–409f, 408t vanishing white matter (VWM), 437–444 clinical features, 438, 439f differential diagnosis, 443–444 epidemiology, 437 genetics, 437–438 pathogenesis, 444 pathologic findings, 438–443, 440f–443f prenatal forms, 438 varicella zoster virus, 503–504, 524 vascular malformations, CNS, 251–265. See also specific malformations animal models, 263 classification of anomalies of cerebral vasculature, 252t clinical features, 253–255 definition, 251 differential diagnosis, 263 embryology and, 252 epidemiology, 251–252 future directions and therapy, 263, 265 genetics, 253 immunohistochemistry, 262–263 neuroimaging, 255 pathology, 255–262 risk factors, 252 vein of Galen aneurysms, 252, 254, 258–260, 259f–261f. See also vascular malformations, CNS

velaglucerase alfa, 332 venous angiomas, 252–253 ventriculomegaly, 187. See also hydrocephalus prenatal diagnosis of, 188 very long chain fatty acids (VLCFA), 383 viral infections, 518–524 arbovirus infections, 523–524 cytomegalovirus, 497–499, 499f, 521, 522f Ebola virus, 524 enteroviral infections, 519–520 herpes simplex infection, 499–500, 500f, 518–519, 519f HIV infection, 520–521 measles infections, 521–523, 523f progressive multifocal leukoencephalopathy, 521 progressive rubella panencephalitis, 523 rabies, 524 varicella zoster virus, 524 Zika virus, 504–505, 505f Walker–Warburg syndrome, 75. See also lissencephaly, type II warfarin, 336 white matter lesion in perinatal period, 213–224. See also diffuse white matter injury (DWMI) Williams–Beuren syndrome, 487 Wolf–Hirschhorn syndrome, 6–7, 8t Woodhouse–Sakati syndrome, 455 Worster–Drought syndrome, 85 Wyburn–Mason syndrome, 253 xeroderma pigmentosum–Cockayne syndrome complex, 430 X-linked adrenoleukodystrophy (X-ALD), 381–382. See also peroxisomal disorders X-linked alpha-thalassemia/intellectual disability syndrome, 488 X-linked hydrocephalus, 192, 193f Zebra bodies, 319, 349 Zellweger spectrum disorders, 381. See also peroxisomal disorders Zellweger syndrome, 85, 162, 168, 381, 381t. See also peroxisomal disorders Zika virus, 504–505, 505f