The Human Hypothalamus: Middle and Posterior Region [1 ed.] 012820107X, 9780128201077

Citation preview

THE HUMAN HYPOTHALAMUS: MIDDLE AND POSTERIOR REGION

HANDBOOK OF CLINICAL NEUROLOGY Series Editors

MICHAEL J. AMINOFF, FRANÇOIS BOLLER, AND DICK F. SWAAB VOLUME 180

THE HUMAN HYPOTHALAMUS: MIDDLE AND POSTERIOR REGION Series Editors

MICHAEL J. AMINOFF, FRANÇOIS BOLLER, AND DICK F. SWAAB

Volume Editors

DICK F. SWAAB, FELIX KREIER, PAUL J. LUCASSEN, AHMAD SALEHI, AND RUUD M. BUIJS VOLUME 180 3rd Series

ELSEVIER Radarweg 29, PO Box 211, 1000 AE Amsterdam, Netherlands The Boulevard, Langford Lane, Kidlington, Oxford OX5 1GB, United Kingdom 50 Hampshire Street, 5th Floor, Cambridge, MA 02139, United States Copyright © 2021 Elsevier B.V. All rights reserved. No part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system, without permission in writing from the publisher. Details on how to seek permission, further information about the Publisher’s permissions policies and our arrangements with organizations such as the Copyright Clearance Center and the Copyright Licensing Agency can be found at our website: www.elsevier.com/permissions. This book and the individual contributions contained in it are protected under copyright by the Publisher (other than as may be noted herein). Notices Knowledge and best practice in this field are constantly changing. As new research and experience broaden our understanding, changes in research methods, professional practices, or medical treatment may become necessary. Practitioners and researchers must always rely on their own experience and knowledge in evaluating and using any information, methods, compounds, or experiments described herein. In using such information or methods they should be mindful of their own safety and the safety of others, including parties for whom they have a professional responsibility. With respect to any drug or pharmaceutical products identified, readers are advised to check the most current information provided (i) on procedures featured or (ii) by the manufacturer of each product to be administered, to verify the recommended dose or formula, the method and duration of administration, and contraindications. It is the responsibility of practitioners, relying on their own experience and knowledge of their patients, to make diagnoses, to determine dosages and the best treatment for each individual patient, and to take all appropriate safety precautions. To the fullest extent of the law, neither the Publisher nor the authors, contributors, or editors assume any liability for any injury and/or damage to persons or property as a matter of products liability, negligence or otherwise, or from any use or operation of any methods, products, instructions, or ideas contained in the material herein. British Library Cataloguing-in-Publication Data A catalogue record for this book is available from the British Library Library of Congress Cataloging-in-Publication Data A catalog record for this book is available from the Library of Congress ISBN: 978-0-12-820107-7 For information on all Elsevier publications visit our website at https://www.elsevier.com/books-and-journals

Publisher: Nikki Levy Editorial Project Manager: Kristi Anderson Production Project Manager: Punithavathy Govindaradjane Cover Designer: Alan Studholme Typeset by SPi Global, India

Handbook of Clinical Neurology 3rd Series Available titles Vol. 79, The human hypothalamus: Basic and clinical aspects, Part I, D.F. Swaab, ed. ISBN 9780444513571 Vol. 80, The human hypothalamus: Basic and clinical aspects, Part II, D.F. Swaab, ed. ISBN 9780444514905 Vol. 81, Pain, F. Cervero and T.S. Jensen, eds. ISBN 9780444519016 Vol. 82, Motor neurone disorders and related diseases, A.A. Eisen and P.J. Shaw, eds. ISBN 9780444518941 Vol. 83, Parkinson’s disease and related disorders, Part I, W.C. Koller and E. Melamed, eds. ISBN 9780444519009 Vol. 84, Parkinson’s disease and related disorders, Part II, W.C. Koller and E. Melamed, eds. ISBN 9780444528933 Vol. 85, HIV/AIDS and the nervous system, P. Portegies and J. Berger, eds. ISBN 9780444520104 Vol. 86, Myopathies, F.L. Mastaglia and D. Hilton Jones, eds. ISBN 9780444518996 Vol. 87, Malformations of the nervous system, H.B. Sarnat and P. Curatolo, eds. ISBN 9780444518965 Vol. 88, Neuropsychology and behavioural neurology, G. Goldenberg and B.C. Miller, eds. ISBN 9780444518972 Vol. 89, Dementias, C. Duyckaerts and I. Litvan, eds. ISBN 9780444518989 Vol. 90, Disorders of consciousness, G.B. Young and E.F.M. Wijdicks, eds. ISBN 9780444518958 Vol. 91, Neuromuscular junction disorders, A.G. Engel, ed. ISBN 9780444520081 Vol. 92, Stroke – Part I: Basic and epidemiological aspects, M. Fisher, ed. ISBN 9780444520036 Vol. 93, Stroke – Part II: Clinical manifestations and pathogenesis, M. Fisher, ed. ISBN 9780444520043 Vol. 94, Stroke – Part III: Investigations and management, M. Fisher, ed. ISBN 9780444520050 Vol. 95, History of neurology, S. Finger, F. Boller and K.L. Tyler, eds. ISBN 9780444520081 Vol. 96, Bacterial infections of the central nervous system, K.L. Roos and A.R. Tunkel, eds. ISBN 9780444520159 Vol. 97, Headache, G. Nappi and M.A. Moskowitz, eds. ISBN 9780444521392 Vol. 98, Sleep disorders Part I, P. Montagna and S. Chokroverty, eds. ISBN 9780444520067 Vol. 99, Sleep disorders Part II, P. Montagna and S. Chokroverty, eds. ISBN 9780444520074 Vol. 100, Hyperkinetic movement disorders, W.J. Weiner and E. Tolosa, eds. ISBN 9780444520142 Vol. 101, Muscular dystrophies, A. Amato and R.C. Griggs, eds. ISBN 9780080450315 Vol. 102, Neuro-ophthalmology, C. Kennard and R.J. Leigh, eds. ISBN 9780444529039 Vol. 103, Ataxic disorders, S.H. Subramony and A. Durr, eds. ISBN 9780444518927 Vol. 104, Neuro-oncology Part I, W. Grisold and R. Sofietti, eds. ISBN 9780444521385 Vol. 105, Neuro-oncology Part II, W. Grisold and R. Sofietti, eds. ISBN 9780444535023 Vol. 106, Neurobiology of psychiatric disorders, T. Schlaepfer and C.B. Nemeroff, eds. ISBN 9780444520029 Vol. 107, Epilepsy Part I, H. Stefan and W.H. Theodore, eds. ISBN 9780444528988 Vol. 108, Epilepsy Part II, H. Stefan and W.H. Theodore, eds. ISBN 9780444528995 Vol. 109, Spinal cord injury, J. Verhaagen and J.W. McDonald III, eds. ISBN 9780444521378 Vol. 110, Neurological rehabilitation, M. Barnes and D.C. Good, eds. ISBN 9780444529015 Vol. 111, Pediatric neurology Part I, O. Dulac, M. Lassonde and H.B. Sarnat, eds. ISBN 9780444528919 Vol. 112, Pediatric neurology Part II, O. Dulac, M. Lassonde and H.B. Sarnat, eds. ISBN 9780444529107 Vol. 113, Pediatric neurology Part III, O. Dulac, M. Lassonde and H.B. Sarnat, eds. ISBN 9780444595652 Vol. 114, Neuroparasitology and tropical neurology, H.H. Garcia, H.B. Tanowitz and O.H. Del Brutto, eds. ISBN 9780444534903 Vol. 115, Peripheral nerve disorders, G. Said and C. Krarup, eds. ISBN 9780444529022 Vol. 116, Brain stimulation, A.M. Lozano and M. Hallett, eds. ISBN 9780444534972 Vol. 117, Autonomic nervous system, R.M. Buijs and D.F. Swaab, eds. ISBN 9780444534910 Vol. 118, Ethical and legal issues in neurology, J.L. Bernat and H.R. Beresford, eds. ISBN 9780444535016 Vol. 119, Neurologic aspects of systemic disease Part I, J. Biller and J.M. Ferro, eds. ISBN 9780702040863 Vol. 120, Neurologic aspects of systemic disease Part II, J. Biller and J.M. Ferro, eds. ISBN 9780702040870 Vol. 121, Neurologic aspects of systemic disease Part III, J. Biller and J.M. Ferro, eds. ISBN 9780702040887 Vol. 122, Multiple sclerosis and related disorders, D.S. Goodin, ed. ISBN 9780444520012 Vol. 123, Neurovirology, A.C. Tselis and J. Booss, eds. ISBN 9780444534880 Vol. 124, Clinical neuroendocrinology, E. Fliers, M. Korbonits and J.A. Romijn, eds. ISBN 9780444596024 Vol. 125, Alcohol and the nervous system, E.V. Sullivan and A. Pfefferbaum, eds. ISBN 9780444626196 Vol. 126, Diabetes and the nervous system, D.W. Zochodne and R.A. Malik, eds. ISBN 9780444534804 Vol. 127, Traumatic brain injury Part I, J.H. Grafman and A.M. Salazar, eds. ISBN 9780444528926 Vol. 128, Traumatic brain injury Part II, J.H. Grafman and A.M. Salazar, eds. ISBN 9780444635211 Vol. 129, The human auditory system: Fundamental organization and clinical disorders, G.G. Celesia and G. Hickok, eds. ISBN 9780444626301

vi Vol. Vol. Vol. Vol. Vol. Vol. Vol. Vol. Vol. Vol. Vol. Vol. Vol. Vol. Vol. Vol. Vol.

AVAILABLE TITLES (Continued) 130, 131, 132, 133, 134, 135, 136, 137, 138, 139, 140, 141, 142, 143, 144, 145, 146,

Neurology of sexual and bladder disorders, D.B. Vodušek and F. Boller, eds. ISBN 9780444632470 Occupational neurology, M. Lotti and M.L. Bleecker, eds. ISBN 9780444626271 Neurocutaneous syndromes, M.P. Islam and E.S. Roach, eds. ISBN 9780444627025 Autoimmune neurology, S.J. Pittock and A. Vincent, eds. ISBN 9780444634320 Gliomas, M.S. Berger and M. Weller, eds. ISBN 9780128029978 Neuroimaging Part I, J.C. Masdeu and R.G. González, eds. ISBN 9780444534859 Neuroimaging Part II, J.C. Masdeu and R.G. González, eds. ISBN 9780444534866 Neuro-otology, J.M. Furman and T. Lempert, eds. ISBN 9780444634375 Neuroepidemiology, C. Rosano, M.A. Ikram and M. Ganguli, eds. ISBN 9780128029732 Functional neurologic disorders, M. Hallett, J. Stone and A. Carson, eds. ISBN 9780128017722 Critical care neurology Part I, E.F.M. Wijdicks and A.H. Kramer, eds. ISBN 9780444636003 Critical care neurology Part II, E.F.M. Wijdicks and A.H. Kramer, eds. ISBN 9780444635990 Wilson disease, A. Członkowska and M.L. Schilsky, eds. ISBN 9780444636003 Arteriovenous and cavernous malformations, R.F. Spetzler, K. Moon and R.O. Almefty, eds. ISBN 9780444636409 Huntington disease, A.S. Feigin and K.E. Anderson, eds. ISBN 9780128018934 Neuropathology, G.G. Kovacs and I. Alafuzoff, eds. ISBN 9780128023952 Cerebrospinal fluid in neurologic disorders, F. Deisenhammer, C.E. Teunissen and H. Tumani, eds. ISBN 9780128042793 Vol. 147, Neurogenetics Part I, D.H. Geschwind, H.L. Paulson and C. Klein, eds. ISBN 9780444632333 Vol. 148, Neurogenetics Part II, D.H. Geschwind, H.L. Paulson and C. Klein, eds. ISBN 9780444640765 Vol. 149, Metastatic diseases of the nervous system, D. Schiff and M.J. van den Bent, eds. ISBN 9780128111611 Vol. 150, Brain banking in neurologic and psychiatric diseases, I. Huitinga and M.J. Webster, eds. ISBN 9780444636393 Vol. 151, The parietal lobe, G. Vallar and H.B. Coslett, eds. ISBN 9780444636225 Vol. 152, The neurology of HIV infection, B.J. Brew, ed. ISBN 9780444638496 Vol. 153, Human prion diseases, M. Pocchiari and J.C. Manson, eds. ISBN 9780444639455 Vol. 154, The cerebellum: From embryology to diagnostic investigations, M. Manto and T.A.G.M. Huisman, eds. ISBN 9780444639561 Vol. 155, The cerebellum: Disorders and treatment, M. Manto and T.A.G.M. Huisman, eds. ISBN 9780444641892 Vol. 156, Thermoregulation: From basic neuroscience to clinical neurology Part I, A.A. Romanovsky, ed. ISBN 9780444639127 Vol. 157, Thermoregulation: From basic neuroscience to clinical neurology Part II, A.A. Romanovsky, ed. ISBN 9780444640741 Vol. 158, Sports neurology, B. Hainline and R.A. Stern, eds. ISBN 9780444639547 Vol. 159, Balance, gait, and falls, B.L. Day and S.R. Lord, eds. ISBN 9780444639165 Vol. 160, Clinical neurophysiology: Basis and technical aspects, K.H. Levin and P. Chauvel, eds. ISBN 9780444640321 Vol. 161, Clinical neurophysiology: Diseases and disorders, K.H. Levin and P. Chauvel, eds. ISBN 9780444641427 Vol. 162, Neonatal neurology, L.S. De Vries and H.C. Glass, eds. ISBN 9780444640291 Vol. 163, The frontal lobes, M. D’Esposito and J.H. Grafman, eds. ISBN 9780128042816 Vol. 164, Smell and taste, Richard L. Doty, ed. ISBN 9780444638557 Vol. 165, Psychopharmacology of neurologic disease, V.I. Reus and D. Lindqvist, eds. ISBN 9780444640123 Vol. 166, Cingulate cortex, B.A. Vogt, ed. ISBN 9780444641960 Vol. 167, Geriatric neurology, S.T. DeKosky and S. Asthana, eds. ISBN 9780128047668 Vol. 168, Brain-computer interfaces, N.F. Ramsey and J. del R. Millán, eds. ISBN 9780444639349 Vol. 169, Meningiomas, Part I, M.W. McDermott, ed. ISBN 9780128042809 Vol. 170, Meningiomas, Part II, M.W. McDermott, ed. ISBN 9780128221983 Vol. 171, Neurology and pregnancy: Pathophysiology and patient care, E.A.P. Steegers, M.J. Cipolla and E.C. Miller, eds. ISBN 9780444642394 Vol. 172, Neurology and pregnancy: Neuro-obstetric disorders, E.A.P. Steegers, M.J. Cipolla and E.C. Miller, eds. ISBN 9780444642400 Vol. 173, Neurocognitive development: Normative development, A. Gallagher, C. Bulteau, D. Cohen and J.L. Michaud, eds. ISBN 9780444641502 Vol. 174, Neurocognitive development: Disorders and disabilities, A. Gallagher, C. Bulteau, D. Cohen and J.L. Michaud, eds. ISBN 9780444641489 Vol. 175, Sex differences in neurology and psychiatry, R. Lanzenberger, G.S. Kranz, and I. Savic, eds. ISBN 9780444641236 Vol. 176, Interventional neuroradiology, S.W. Hetts and D.L. Cooke, eds. ISBN 9780444640345 Vol. 177, Heart and neurologic disease, J. Biller, ed. ISBN 9780128198148 Vol. 178, Neurology of vision and visual disorders, J.J.S. Barton and A. Leff, eds. ISBN 9780128213773 Vol. 179, The human hypothalamus: Anterior region, D.F. Swaab, F. Kreier, P.J. Lucassen, A. Salehi and R.M. Buijs, eds. ISBN 9780128199756 All volumes in the 3rd Series of the Handbook of Clinical Neurology are published electronically, on Science Direct: http://www.sciencedirect.com/science/handbooks/00729752.

Foreword

Few areas of neuroscience have received so much attention and yielded so many new and important findings in recent years as the hypothalamus and related structures. Earlier series of the Handbook of Clinical Neurology included over 100 entries under the keyword “hypothalamus,” dispersed within all 75 volumes. The current third series started with two volumes dedicated entirely to basic and clinical aspects of the hypothalamus, the first (Volume 79) dealing with the hypothalamic nuclei and the second (Volume 80) with its neuropathology. They were authored by Professor Dick Swaab and were published almost 20 years ago (in 2003 and 2004). As series editors, we felt that the number of new developments since that time required that the entire topic be reviewed once more. These new developments include a better understanding of the anatomy and connections of the human hypothalamus based on novel imaging techniques and the accumulating molecular information on the hypothalamus. Also, it is now apparent that the hypothalamus regulates more hormones than previously recognized and is the key structure in clinical neuroendocrinology. Above all, the hypothalamus is now seen to relate to a large number of neurologic domains—including memory, sleep, epilepsy, Parkinson disease and other neurodegenerative disorders, and headaches, as well as behavioral issues such as eating behavior, depression, and aggression. Last but not least, the hypothalamus plays a crucial role in reproduction and shows sexual dimorphisms in various nuclei. These advances and the associated vast expansion of knowledge that has resulted have required an increase in coverage from two to four volumes of the Handbook. We thank and congratulate Dick Swaab who is the Chief Editor of these four new multiauthored volumes. They were prepared in collaboration with four other highly experienced neuroscientists. Ruud Buijs is in the Institute for Biomedical Investigation, Universidad Nacional Autónoma de Mexico, Mexico City; Felix Kreier is in the Department of Pediatrics, OLVG hospital, Amsterdam; Paul Lucassen is at the Center for Neuroscience, Swammerdam Institute for Life Sciences, University of Amsterdam; and Ahmad Salehi is in the Department of Psychiatry and Behavioral Sciences, Stanford Medical School, California. Together they have gathered a remarkable group of contributing authors, thus assuring the right mix of continuity and highly updated information about the human hypothalamus. As series editors, we reviewed all the chapters in the volumes and made suggestions for improvement, but we are delighted that the volume editors and chapter authors produced such scholarly and comprehensive accounts of different aspects of the topic. We hope that the volumes will appeal to clinicians as a state-of-the-art reference that summarizes the clinical features and management of the many neurologic, neuroendocrine, and psychiatric manifestations of hypothalamic dysfunction. We are also sure that basic researchers will find within them the foundations for new approaches to the study of the complex issues involved. In addition to the print version, the volumes are available electronically on Elsevier’s Science Direct website, which is popular with readers and will improve the books’ accessibility. Indeed, all of the volumes in the present series of the Handbook are available electronically on this website. This should make them more accessible to readers and facilitate searches for specific information. As always, it is a pleasure to thank Elsevier, our publisher, and in particular Michael Parkinson in Scotland, Nikki Levy and Kristi Anderson in San Diego, and Punithavathy Govindaradjane at Elsevier Global Book Production in Chennai, for their assistance in the development and production of the Handbook of Clinical Neurology. Michael J. Aminoff Franc¸ois Boller

This page intentionally left blank

Preface

I know very well that the reader has no great need to know all this; it is I who have a need to tell him J.J. Rousseau

THE HCN VOLUMES ON THE HUMAN HYPOTHALAMUS More than 20 years ago, I (DFS) had written a monograph on the human hypothalamus, meant as a starting point for my PhD students and the students of my former students, my scientifically gifted children and grandchildren. Traditionally the hypothalamus was considered to be a neuroendocrine structure of limited interest to neurologists. In addition, this extremely complex structure, which consists of a large number of very different functional nuclei, was not included in the standard neuropathologic investigation of the human brain. Neuropathologists were trained to cut right through the optic chiasma, thereby destroying the hypothalamus. During the period in which I wrote the monograph, it became clear, however, that the hypothalamus not only regulates hormone levels but also contributes to the memory and attention deficits in the dementias; that a disorder of the orexin/hypocretin system is the cause of narcolepsy; that hypothalamic hamartomas are responsible for gelastic epilepsy; that the subthalamic nucleus is a good target to place depth electrodes in parkinsonian patients; and that the source of cluster headache may be situated in the posterior hypothalamus. Moreover, the hypothalamus appeared to be the basis of many signs and symptoms of disorders situated on the border between neurology and psychiatry, such as depression, eating disorders, aggression, and mental retardation. As a consequence, the hypothalamus became a meeting point for neuroscientists, neurologists and psychiatrists, neuropathologists, endocrinologists, and pediatricians. It was the vision of my friend Professor George Bruyn that my monograph would be a starting point for a new (third) series of the Handbook of Clinical Neurology (HCN). The monography was published in two HCN volumes, 79 and 80. Together with my fellow series editors Michael J. Aminoff and Franc¸ois Boller and the staff of Elsevier, more than 100 additional volumes in this new series have since been published. The other two series editors have asked me repeatedly to consider a follow-up of my two earlier HCN volumes. Since they were published, there has indeed been great progress in the field, e.g., in deep brain stimulation, molecular biology (including gene and cell therapy, the various omics, transgenic animal models, and generation of hypothalamic neurons from human-induced pluripotent stem cells), molecular genetics, advanced scanning techniques (e.g., functional connectivity of hypothalamic nuclei), central effects of neuropeptides in health and disease, human brain donation, and brain banking (e.g., putative confounding factors for hypothalamic research). Other topics were simply not dealt with in Volumes 79 and 80, such as the history of neuroendocrinology/hypothalamic research, or are absolutely necessary to place the other chapters in perspective, such as microscopic neuroanatomy of the hypothalamus, borders, and markers of nuclei. Only after my friends and excellent colleagues Paul Lucassen, Ruud Buijs, Felix Kreier, and Ahmad Salehi agreed to participate as covolume editors did I feel that we could face this challenge. From the start, the Covid-19 pandemic interfered with the composition of the volumes. We are very grateful for the authors that managed to deliver their chapters, in spite of the often extremely difficult circumstances. We are also grateful for the continuous and essential help and support of Michael J. Aminoff and Franc¸ois Boller and Michael Parkinson during the entire process. The new volumes are again subdivided into a basic part (The Nuclei of the Hypothalamus) and a clinical part (Neuropathology, Neuropsychiatric disorders), but now as multiauthored volumes, consisting of in-depth reviews of topics that were novel, had progressed markedly since the earlier volumes, or needed to be reviewed in a critical way. Because of the large number of crucial topics, four volumes emerged. They are in many aspects still complementary to HCN volumes 79 and 80, as is indicated later.

The hypothalamus: Arbitrary borders The exact borders of the hypothalamus are rather arbitrary and the exact terminology has often been controversial (see HCN volume 79 for references and details). As stated by Crosby et al. (1962): Nomenclature is man-made; there is

x

PREFACE

strictly speaking no correct and no incorrect way of designating nuclear groups of a region, except as certain names are sanctioned by usage. The borders are generally considered to be: rostrally, the lamina terminalis, and caudally, the plane through the posterior edge of the mamillary body or mamillothalamic tract or the bundle of Vicq d’Azyr. The hypothalamic sulcus is generally looked upon as the dorsal border. However, the paraventricular nucleus is often found partially dorsally of the hypothalamic sulcus. Cells do not respect hypothalamic boundaries. The anterior commissure has also been mentioned as a dorsal border of the hypothalamus, but this structure might penetrate the third ventricle on different levels and the central nucleus of the bed nucleus of the stria terminalis is partly situated dorsally and partly ventrally of the anterior commissure. The ventral border of the hypothalamus includes the floor of the third ventricle that blends into the infundibulum of the neurohypophysis. The exact location of the lateral boundaries, i.e., the nucleus basalis of Meynert, striatum/nucleus accumbens, amygdala, the posterior limb of the internal capsule and basis pedunculi, and, more caudodorsally, the border of the subthalamic nucleus is not a matter of clear-cut certainty either. Finally, there is great variability: no two hypothalami are alike as Gr€ unthal remarked earlier (1950). Since I (DFS) founded the Netherlands Brain Bank in 1985, the brain has been dissected fresh in more than 100 pieces along anatomical borders. The Netherlands Brain Bank has provided more than 100,000 clinically and neuropathologically well-characterized brain samples from more than 4500 rapid autopsies to research projects in 25 countries. My own main interest was the hypothalamus. Because of this personal focus and the delineation problems mentioned previously, we did not deal with the question of which structure does or does not belong to the hypothalamus sensu stricto or sensu lato based on their embryology or adult hypothalamic borders. The way we dissected the hypothalamus en bloc during an autopsy (Fig. 1)

Fig. 1. A block of tissue (frontal cut) containing the hypothalamus and adjacent structures; OC, optic chiasm; OVLT, organum vasculosum lamina terminalis (note that the third ventricle is shining through the thin lamina terminalis); ac, anterior commissure, on top of which the septum with the fornix at both sides is located. The lateral ventricles containing plexus choroids are present and both sides of the septum and under the CC, corpus callosum.

PREFACE

xi

resulted in a hypothalamus and surrounding structures that are also included for pragmatic reasons in these volumes and provides a basis for neurobiological and neuropathological research of this brain region. This means that we include in these HCN volumes structures that are not traditionally considered to be components of the hypothalamus but are surrounding and often strongly interconnected to the core hypothalamic nuclei. An example is the basal cholinergic nuclei that are included in spite of the fact that the diagonal band of Broca and the nucleus basalis of Meynert are telencephalic. In addition, the bed nucleus of the stria terminalis is included, although it is only partly localized below the anterior commissure. Others introduce the concept of a wider hypothalamic region that includes parts of the former mesencephalic ventral thalamus such as the zona incerta and the subthalamic nucleus, based upon the fact that these structures have few common anatomic and developmental features with the thalamus. Moreover, the preoptic area that originates from the telencephalon is included since it has an intimate relationship with the anterior and other portions of the hypothalamus, with which it forms a functional unit. Dick F. Swaab Felix Kreier Paul J. Lucassen Ahmad Salehi Ruud M. Buijs

REFERENCES Crosby EC, Humphrey T, Lauer EW (1962). Correlative Anatomy of the Nervous System. MacMillan, NY, 310. Gr€ unthal E (1950). In: WR Hess (Ed.), Symposion €uber das Zwischenhirn. Helv Physiol Pharm Acta. Suppl VI: 1–80.

This page intentionally left blank

Contributors

R.M. Ahmed Memory and Cognition Clinic, Department of Clinical Neurosciences, Royal Prince Alfred Hospital; Central Sydney Medical School and Brain & Mind Centre, Faculty of Medicine and Health, University of Sydney, Sydney, NSW, Australia

A. Boelen Laboratory of Endocrinology, Department of Clinical Chemistry, Amsterdam Gastroenterology, Endocrinology, and Metabolism, Amsterdam UMC, University of Amsterdam, Amsterdam, The Netherlands

H. Akram Unit of Functional Neurosurgery, UCL Queen Square Institute of Neurology & The National Hospital for Neurology and Neurosurgery, London, United Kingdom

R.M. Buijs Hypothalamic Integration Mechanisms Laboratory, Department of Cellular Biology and Physiology, Instituto de Investigaciones Biom
edicas, Universidad Nacional Autónoma de M
exico (UNAM), Ciudad de M
exico, Mexico

A. Alkemade Integrative Model-Based Cognitive Neuroscience Research Unit, University of Amsterdam, Amsterdam, The Netherlands

S. Chometton Department of Biological Sciences, University of Southern California, Los Angeles, CA, United States

F. Althammer Neuroscience Department, Center for Neuroinflammation and Cardiometabolic Diseases, Georgia State University, Atlanta, GA, United States N.J.M. Arts Centre of Excellence for Korsakoff and Alcohol-Related Cognitive Disorders, Vincent van Gogh Institute for Psychiatry, Venray; Neuropsychiatry Center Thalamus, Institution for Integrated Mental Health Care Pro Persona, Wolfheze, The Netherlands J. Bakker GIGA Neurosciences, Liège University, Liège, Belgium M. Barbier Seaver Autism Center, Icahn School of Medicine, Mount Sinai, New York, NY, United States F. Benedetti Department of Neuroscience, University of Turin Medical School, Turin, Italy; Medicine and Physiology of Hypoxia, Plateau Rosà, Switzerland

A. Di Cristofori Neurosurgery Unit, Department of Neuroscience, Azienda Socio Sanitaria Territoriale Monza, Ospedale San Gerardo, Monza, Italy L. Dupuis Universit
e de Strasbourg, Inserm, UMR-S 1118, Centre de Recherches en Biom
edecine, Strasbourg, France M. Eliava Department of Neuropeptide Research in Psychiatry, Central Institute of Mental Health, Medical Faculty Mannheim, University of Heidelberg, Mannheim, Germany A.V. Ferguson Department of Biomedical and Molecular Sciences and Centre for Neuroscience Studies, Queen's University, Kingston, ON, Canada E.C. Finger Department of Clinical Neurological Sciences, Lawson Health Research Institute; Department of Clinical Neurological Sciences, Schulich School of Medicine and Dentistry, University of Western Ontario, London, ON, Canada

xiv CONTRIBUTORS S. Fischer S. Hu Clinical Psychology and Psychotherapy, University of Department of Psychiatry, First Affiliated Hospital, Zurich, Zurich, Switzerland Zhejiang University School of Medicine; The Key Laboratory of Mental Disorder Management in Zhejiang Province; Brain Research Institute of Zhejiang E. Fliers University, Hangzhou, China Department of Endocrinology and Metabolism, Amsterdam Gastroenterology, Endocrinology, and A. Kalsbeek Metabolism, Amsterdam UMC, University of Department of Endocrinology and Metabolism, Amsterdam, Amsterdam, The Netherlands Amsterdam University Medical Centers (Amsterdam UMC), University of Amsterdam; Department of B.U. Forstmann Hypothalamic Integration Mechanisms, Netherlands Integrative Model-Based Cognitive Neuroscience Institute for Neuroscience, an Institute of the Royal Research Unit, University of Amsterdam, Amsterdam, Netherlands Academy of Arts and Sciences, Amsterdam, The Netherlands The Netherlands E. Frisaldi Department of Neuroscience, University of Turin Medical School, Turin, Italy R. Fronczek Department of Neurology, Leiden University Medical Centre, Leiden; Sleep Wake Centre SEIN, Heemstede, The Netherlands W.M. Fry Department of Biological Sciences, University of Manitoba, Winnipeg, MB, Canada C. Giussani Department of Medicine and Surgery, Neurosurgery Unit, Università degli Studi Milano Bicocca, Milan; Neurosurgery Unit, Department of Neuroscience, Azienda Socio Sanitaria Territoriale Monza, Ospedale San Gerardo, Monza, Italy V. Grinevich Department of Neuropeptide Research in Psychiatry, Central Institute of Mental Health, Medical Faculty Mannheim, University of Heidelberg, Mannheim, Germany L.A. Henderson Department of Anatomy & Histology, Brain and Mind Centre, University of Sydney, Sydney, NSW, Australia S. Hescham Department of Neurosurgery, Maastricht University Medical Center, Maastricht, The Netherlands E. Hrabovszky Laboratory of Reproductive Neurobiology, Institute of Experimental Medicine, Budapest, Hungary

M.J.T. Kalsbeek Laboratory of Endocrinology, Amsterdam University Medical Center, University of Amsterdam, Amsterdam Gastroenterology Metabolism; Netherlands Institute for Neuroscience, an Institute of the Royal Netherlands Academy of Arts and Sciences, Amsterdam, The Netherlands L. Kerem Neuroendocrine Unit, Department of Medicine, Massachusetts General Hospital and Harvard Medical School; Division of Pediatric Endocrinology, Massachusetts General Hospital for Children, Boston, MA, United States R.P.C. Kessels Centre of Excellence for Korsakoff and Alcohol-Related Cognitive Disorders, Vincent van Gogh Institute for Psychiatry, Venray; Donders Institute for Brain, Cognition and Behaviour, Radboud University, Nijmegen, The Netherlands H.-W. Korf Center for Anatomy and Brain Research, Institute for Anatomy, D€usseldorf, Germany F. Kreier Department Pediatrics, OLVG Hospitals, Amsterdam, The Netherlands G.J. Lammers Department of Neurology, Leiden University Medical Centre, Leiden; Sleep Wake Centre SEIN, Heemstede, The Netherlands E.A. Lawson Neuroendocrine Unit, Department of Medicine, Massachusetts General Hospital and Harvard Medical School, Boston, MA, United States

CONTRIBUTORS xv
Rumpler Q. Lu E. Department of Psychiatry, Hangzhou Seventh People's Laboratory of Reproductive Neurobiology, Institute of Hospital, Hangzhou, China Experimental Medicine, Budapest, Hungary P.J. Lucassen Brain Plasticity Group, Swammerdam Institute for Life Sciences, University of Amsterdam, Amsterdam, The Netherlands V.G. Macefield Baker Heart & Diabetes Institute; Department of Anatomy and Physiology, University of Melbourne, Melbourne, VIC, Australia R. McGregor Neuropsychiatric Institute and Brain Research Institute, University of California; Neurobiology Research, Veterans Administration Greater Los Angeles Healthcare System, Los Angeles, CA, United States M. Møller Department of Neuroscience, Faculty of Health and Medical Sciences, University of Copenhagen, Copenhagen, Denmark P.E. Mosley Systems Neuroscience Laboratory, QIMR Berghofer Medical Research Institute, Brisbane, QLD, Australia R. Nogueiras Department of Physiology, CIMUS, University of Santiago de Compostela-Instituto de Investigación Sanitaria, Santiago de Compostela, Spain P. Panula Department of Anatomy, University of Helsinki, Helsinki, Finland A.-L. Pitel Normandie Univ, UNICAEN, PSL Universit
e, EPHE, INSERM, U1077, CHU de Caen, GIP Cyceron, NIMH, Caen; Institut Universitaire de France, Paris, France V. Prevot Univ. Lille, Inserm, CHU Lille, Laboratory of Development and Plasticity of the Neuroendocrine Brain, Lille Neuroscience & Cognition, UMR-S1172, DISTALZ, EGID, Lille, France P.-Y. Risold EA481, Integrative and Clinical Neurosciences, UFR Sant
e, Universit
e de Bourgogne Franche-Comt
e, Besanc¸on, France

A. Salehi Department of Psychiatry and Behavioral Sciences, Stanford Medical School, Palo Alto, CA, United States M. Schinkelshoek Department of Neurology, Leiden University Medical Centre, Leiden; Sleep Wake Centre SEIN, Heemstede, The Netherlands M. Schwaninger University of L€ubeck, Institute for Experimental and Clinical Pharmacology and Toxicology, L€ubeck, Germany L. Shan Department of Neurology, Leiden University Medical Centre, Leiden; Sleep Wake Centre SEIN, Heemstede; Department Neuropsychiatric Disorders, Netherlands Institute for Neuroscience, an Institute of the Royal Netherlands Academy of Arts and Sciences, Amsterdam, The Netherlands J.M. Siegel Neuropsychiatric Institute and Brain Research Institute, University of California; Neurobiology Research, Veterans Administration Greater Los Angeles Healthcare System, Los Angeles, CA, United States K. Skrapits Laboratory of Reproductive Neurobiology, Institute of Experimental Medicine, Budapest, Hungary C.A. Stewart Department of Clinical Neurological Sciences, Lawson Health Research Institute; Graduate Program in Neuroscience, Schulich School of Medicine and Dentistry, University of Western Ontario, London, ON, Canada F. Steyn School of Biomedical Sciences, University of Queensland; Department of Neurology, Royal Brisbane & Women's Hospital, Brisbane, QLD, Australia D.F. Swaab Department Neuropsychiatric Disorders, Netherlands Institute for Neuroscience, an Institute of the Royal Netherlands Academy of Arts and Sciences, Amsterdam, The Netherlands

xvi

CONTRIBUTORS

M. Szyf Department of Pharmacology and Therapeutics, McGill University, Montreal, QC, Canada S. Takács Laboratory of Reproductive Neurobiology, Institute of Experimental Medicine, Budapest, Hungary Y. Temel Department of Neurosurgery, Maastricht University Medical Center, Maastricht, The Netherlands T.C. Thannickal Neuropsychiatric Institute and Brain Research Institute, University of California; Neurobiology Research, Veterans Administration Greater Los Angeles Healthcare System, Los Angeles, CA, United States

A.S.P. van Trotsenburg Department of Pediatric Endocrinology, Emma Children's Hospital, Amsterdam UMC, University of Amsterdam, Amsterdam, The Netherlands C.-X. Yi Laboratory of Endocrinology, Amsterdam University Medical Center, University of Amsterdam, Amsterdam Gastroenterology Metabolism; Netherlands Institute for Neuroscience, an Institute of the Royal Netherlands Academy of Arts and Sciences; Department of Endocrinology and Metabolism, Amsterdam University Medical Center, University of Amsterdam, Amsterdam, The Netherlands D.A. Zamfira Department of Neuroscience, University of Turin Medical School, Turin, Italy

Contents Foreword vii Preface ix Contributors xiii 1. Introduction: The middle and posterior hypothalamus D.F. Swaab, F. Kreier, P.J. Lucassen, A. Salehi, and R.M. Buijs (Amsterdam, The Netherlands, Palo Alto, United States and Ciudad de México, Mexico) SECTION 6

1

Supraoptic and paraventricular nucleus more than a neuroendocrine system

2. Vasopressin and oxytocin beyond the pituitary in the human brain M. Møller (Copenhagen, Denmark) 3. Central and peripheral release of oxytocin: Relevance of neuroendocrine and neurotransmitter actions for physiology and behavior F. Althammer, M. Eliava, and V. Grinevich (Atlanta, United States and Mannheim, Germany)

7

25

4. Organization of the neuroendocrine and autonomic hypothalamic paraventricular nucleus A. Kalsbeek and R.M. Buijs (Amsterdam, The Netherlands and Ciudad de M
exico, Mexico)

45

5. Sex differences of oxytocin and vasopressin in social behaviors Q. Lu and S. Hu (Hangzhou, China)

65

6. Oxytocin, eating behavior, and metabolism in humans L. Kerem and E.A. Lawson (Boston, United States)

89

7. The supraoptic and paraventricular nuclei in healthy aging and neurodegeneration C.A. Stewart and E.C. Finger (London, Canada)

105

8. Perinatal stress and epigenetics M. Szyf (Montreal, Canada)

125

9. The hypothalamus in anxiety disorders S. Fischer (Zurich, Switzerland)

149

10. Congenital isolated central hypothyroidism: Novel mutations and their functional implications A. Boelen, A.S.P. van Trotsenburg, and E. Fliers (Amsterdam, The Netherlands) SECTION 7

161

Zona incerta

11. The zona incerta system: Involvement in attention and movement S. Chometton, M. Barbier, and P.-Y. Risold (Los Angeles and New York, United States and Besanc¸on, France)

173

xviii SECTION 8

CONTENTS Ventromedial nucleus and dorsomedial nucleus

12. The role of the dorsomedial and ventromedial hypothalamus in regulating behaviorally coupled and resting autonomic drive L.A. Henderson and V.G. Macefield (Sydney and Melbourne, Australia) SECTION 9

187

Circumventricular organs of the hypothalamus

13. The subfornical organ and organum vasculosum of the lamina terminalis: Critical roles in cardiovascular regulation and the control of fluid balance W.M. Fry and A.V. Ferguson (Winnipeg and Kingston, Canada)

203

14. Lamina terminalis fenestration: An important neurosurgical corridor C. Giussani and A. Di Cristofori (Milan and Monza, Italy)

217

15. Arcuate nucleus, median eminence, and hypophysial pars tuberalis H.-W. Korf and M. Møller (D€ usseldorf, Germany and Copenhagen, Denmark)

227

16. Tanycytes in the infundibular nucleus and median eminence and their role in the blood–brain barrier 253 V. Prevot, R. Nogueiras, and M. Schwaninger (Lille, France, Santiago de Compostela, Spain and L€ ubeck, Germany) 17. The human hypothalamic kisspeptin system: Functional neuroanatomy and clinical perspectives
Rumpler, and K. Skrapits (Budapest, Hungary) E. Hrabovszky, S. Takács, E. 18. Kisspeptin and neurokinin B expression in the human hypothalamus: Relation to reproduction and gender identity J. Bakker (Liège, Belgium)

275

297

19. The infundibular peptidergic neurons and glia cells in overeating, obesity, and diabetes M.J.T. Kalsbeek and C.-X. Yi (Amsterdam, The Netherlands)

315

20. Hypothalamus and weight loss in amyotrophic lateral sclerosis R.M. Ahmed, F. Steyn, and L. Dupuis (Sydney and Brisbane, Australia and Strasbourg, France)

327

SECTION 10

Lateral tuberal nucleus

SECTION 11

Lateral hypothalamic area, perifornical area

21. The orexin/hypocretin system in neuropsychiatric disorders: Relation to signs and symptoms R. Fronczek, M. Schinkelshoek, L. Shan, and G.J. Lammers (Leiden, Heemstede and Amsterdam, The Netherlands)

343

22. Pleasure, addiction, and hypocretin (orexin) R. McGregor, T.C. Thannickal, and J.M. Siegel (Los Angeles, United States)

359

SECTION 12

Tuberomamillary complex

23. Histamine receptors, agonists, and antagonists in health and disease P. Panula (Helsinki, Finland)

377

24. The tuberomamillary nucleus in neuropsychiatric disorders L. Shan, R. Fronczek, G.J. Lammers, and D.F. Swaab (Leiden, Heemstede and Amsterdam, The Netherlands)

389

CONTENTS SECTION 13

xix

Subthalamic nucleus

25. Imaging of the human subthalamic nucleus A. Alkemade and B.U. Forstmann (Amsterdam, The Netherlands)

403

26. Neuropsychiatric effects of subthalamic deep brain stimulation P.E. Mosley and H. Akram (Brisbane, Australia and London, United Kingdom)

417

27. The subthalamic nucleus and the placebo effect in Parkinson's disease E. Frisaldi, D.A. Zamfira, and F. Benedetti (Turin, Italy and Plateau Rosà, Switzerland)

433

SECTION 14

Corpora mamillaria, fornix, and mamillothalamic tract

28. Electrical stimulation of the fornix for the treatment of brain diseases S. Hescham and Y. Temel (Maastricht, The Netherlands) 29. The contribution of mamillary body damage to Wernicke's encephalopathy and Korsakoff's syndrome N.J.M. Arts, A.-L. Pitel, and R.P.C. Kessels (Venray, Wolfheze, and Nijmegen, The Netherlands and Caen and Paris, France) Index

447

455

477

This page intentionally left blank

Contents of related volumes Volume 179 (The Human Hypothalamus: Anterior Region) Volume 181 (The Human Hypothalamus: Neuroendocrine Disorders) Volume 182 (The Human Hypothalamus: Neuropsychiatric Disorders)

Contents of Volume 179 Foreword vii Preface ix Contributors xiii SECTION 1

Introduction

1. Introduction: The anterior hypothalamus D.F. Swaab, R.M. Buijs, F. Kreier, P.J. Lucassen, and A. Salehi (Amsterdam, The Netherlands, Ciudad de M
exico, Mexico and Palo Alto, United States)

3

2. History of hypothalamic research: “The spring of primitive existence” F. Kreier and D.F. Swaab (Amsterdam, The Netherlands)

7

3. Anatomy and cytoarchitectonics of the human hypothalamus B. Dudás (Erie, United States and Szeged, Hungary)

45

4. Morphology and distribution of hypothalamic peptidergic systems B. Dudás and I. Merchenthaler (Erie and Baltimore, United States and Szeged, Hungary)

67

5. MRI maps, segregation, and white matter connectivity of the human hypothalamus in health J.-J. Lemaire and A. De Salles (Clermont-Ferrand, France, Los Angeles, United States and São Paulo, Brazil)

87

6. Magnetic resonance imaging of the hypothalamo–pituitary region M. Perosevic, P.S. Jones, and N.A. Tritos (Boston, United States)

95

7. Resting-state functional connectivity of the human hypothalamus S. Kullmann and R. Veit (T€ ubingen and Neuherberg, Germany)

113

8. Neurogenesis in the adult hypothalamus: A distinct form of structural plasticity involved in metabolic and circadian regulation, with potential relevance for human pathophysiology 125 A. Sharif, C.P. Fitzsimons, and P.J. Lucassen (Lille, France and Amsterdam, The Netherlands) 9. Matching of the postmortem hypothalamus from patients and controls D.F. Swaab and A.-M. Bao (Amsterdam, The Netherlands and Hangzhou, China)

141

xxxii SECTION 2

CONTENTS OF RELATED VOLUMES CONTINUED The basal forebrain cholinergic system

10. Spatial topography of the basal forebrain cholinergic projections: Organization and vulnerability to degeneration T.W. Schmitz and L. Zaborszky (London, Canada and Newark, United States)

159

11. The diagonal band of Broca in health and disease A.K.L. Liu and S.M. Gentleman (London, United Kingdom)

175

12. Nucleus basalis of Meynert degeneration predicts cognitive impairment in Parkinson's disease H. Wilson, E.R. de Natale, and M. Politis (London, United Kingdom)

189

13. Enlargement of early endosomes and traffic jam in basal forebrain cholinergic neurons in Alzheimer's disease A. Fahimi, M. Noroozi, and A. Salehi (Los Angeles and Palo Alto, United States) 14. Gene and cell therapy for the nucleus basalis of Meynert with NGF in Alzheimer's disease M. Eriksdotter and S. Mitra (Stockholm and Huddinge, Sweden) SECTION 3

207

219

The circadian system

15. The circadian system: From clocks to physiology R.M. Buijs, E.C. Soto Tinoco, G. Hurtado Alvarado, and C. Escobar (Ciudad de M
exico, Mexico)

233

16. Development of the circadian system and relevance of periodic signals for neonatal development C. Escobar, A. Rojas-Granados, and M. Angeles-Castellanos (Ciudad de M
exico, Mexico)

249

17. Disrupted circadian rhythms and mental health W.H. Walker II, J.C. Walton, and R.J. Nelson (Morgantown, United States)

259

18. Diurnal and seasonal molecular rhythms in the human brain and their relation to Alzheimer disease 271 A.S.P. Lim (Toronto, Canada) 19. Circadian changes in Alzheimer's disease: Neurobiology, clinical problems, and therapeutic opportunities K. Toljan and J. Homolak (Cleveland, United States and Zagreb, Croatia)

285

20. The circadian system in Parkinson's disease, multiple system atrophy, and progressive supranuclear palsy K. Fifel and T. De Boer (Ibaraki, Japan and Leiden, The Netherlands)

301

21. Retina and melanopsin neurons C. La Morgia, V. Carelli, and A.A. Sadun (Bologna, Italy and Los Angeles, United States)

315

22. Melatonin and the circadian system: Keys for health with a focus on sleep P. Pevet, E. Challet, and M.-P. Felder-Schmittbuhl (Strasbourg, France)

331

23. Melatonin receptors, brain functions, and therapies A. Oishi, F. Gbahou, and R. Jockers (Paris, France)

345

24. Chronotherapy D.P. Cardinali, G.M. Brown, and S.R. Pandi-Perumal (Buenos Aires, Argentina and Toronto, Canada)

357

CONTENTS OF RELATED VOLUMES CONTINUED 25. The use of melatonin to mitigate the adverse metabolic side effects of antipsychotics F. Romo-Nava, R.M. Buijs, and S.L. McElroy (Mason and Cincinnati, United States and Ciudad de M
exico, Mexico) SECTION 4

xxxiii 371

Bed nucleus of the stria terminalis and the fear circuit

26. Chemoarchitecture of the bed nucleus of the stria terminalis: Neurophenotypic diversity and function S.E. Hammack, K.M. Braas, and V. May (Burlington, United States)

385

27. Functional anatomy of the bed nucleus of the stria terminalis–hypothalamus neural circuitry: Implications for valence surveillance, addiction, feeding, and social behaviors I. Maita, A. Bazer, J.U. Blackford, and B.A. Samuels (Piscataway and Nashville, United States)

403

28. Roles of the bed nucleus of the stria terminalis and amygdala in fear reactions A.M. Hulsman, D. Terburg, K. Roelofs, and F. Klumpers (Nijmegen and Utrecht, The Netherlands and Cape Town, South Africa) SECTION 5

419

Preoptic area

29. The median preoptic nucleus: A major regulator of fluid, temperature, sleep, and cardiovascular homeostasis M.J. McKinley, G.L. Pennington, and P.J. Ryan (Parkville, Australia) 30. The neuroendocrinology of the preoptic area in menopause: Symptoms and therapeutic strategies M. Modi and W.S. Dhillo (London, United Kingdom)

435

455

31. The intermediate nucleus in humans: Cytoarchitecture, chemoarchitecture, and relation to sleep, sex, and Alzheimer disease C.B. Saper (Boston, United States)

461

Index

471

Contents of Volume 181 Foreword vii Preface ix Contributors xiii 1. Introduction: The human hypothalamus and neuroendocrine disorders D.F. Swaab, R.M. Buijs, P.J. Lucassen, A. Salehi, and F. Kreier (Amsterdam, The Netherlands, Ciudad de M
exico, Mexico and Palo Alto, United States) SECTION 15

1

Structural disorders of the hypothalamo-pituitary region

2. Pituitary stalk interruption syndrome A. Voutetakis (Thrace, Greece) 3. Empty sella syndrome: Multiple endocrine disorders S. Chiloiro, A. Giampietro, A. Bianchi, and L. De Marinis (Rome, Italy)

9

29

xxxiv

CONTENTS OF RELATED VOLUMES CONTINUED

4. Pituitary dysfunction after aneurysmal subarachnoidal hemorrhage S. Bacigaluppi, C. Robba, and N.L. Bragazzi (Genova, Italy and Toronto, Canada)

41

5. Septo-optic dysplasia I. Sataite, S. Cudlip, J. Jayamohan, and M. Ganau (Oxford, United Kingdom)

51

SECTION 16

Tumors of the hypothalamus

6. Hypothalamic hormone-producing tumors S.L. Asa and S. Ezzat (Cleveland, United States and Toronto, Canada)

67

7. Craniopharyngiomas primarily affecting the hypothalamus J.M. Pascual, R. Prieto, and M. Rosdolsky (Madrid, Spain and Jenkintown, United States)

75

SECTION 17

Neuroimmunological disorders

8. The stress-axis in multiple sclerosis: Clinical, cellular, and molecular aspects J. Melief, I. Huitinga, and S.M. Gold (Stockholm, Sweden, Amsterdam, The Netherlands and Berlin and Hamburg, Germany)

119

9. Neuroendocrine manifestations of Langerhans cell histiocytosis M.P. Yavropoulou, M. Tsoli, and G. Kaltsas (Athens, Greece)

127

10. Neuroendocrine manifestations of Erdheim–Chester disease K. Manaka, J. Sato, and N. Makita (Tokyo, Japan)

137

11. Hypothalamitis and pituitary atrophy S. Chiloiro, T. Tartaglione, A. Giampietro, and A. Bianchi (Rome, Italy)

149

12. Narcolepsy Type I as an autoimmune disorder B.R. Kornum (Copenhagen, Denmark)

161

13. Neuromyelitis optica, aquaporin-4 antibodies, and neuroendocrine disorders R. Iorio and C. Papi (Rome, Italy)

173

14. Antibodies against the pituitary and hypothalamus in boxers F. Kelestimur (İstanbul, Turkey)

187

15. Autoimmune diabetes insipidus W.A. Scherbaum (Duesseldorf, Germany)

193

SECTION 18

Drinking disorders

16. Neuroimaging of central diabetes insipidus T.P. Farrell, N.C. Adams, and S. Looby (Philadelphia, United States and Dublin, Ireland)

207

17. Differential diagnosis of familial diabetes insipidus G.L. Robertson (Chicago, United States)

239

18. The vasopressin–aquaporin-2 pathway syndromes G. Valenti and G. Tamma (Bari, Italy)

249

CONTENTS OF RELATED VOLUMES CONTINUED

xxxv

19. Adipsic diabetes insipidus V. Kothari, Z. Cardona, and Y. Eisenberg (Chicago, United States)

261

20. Animal models for diabetes insipidus J. Mahía and A. Bernal (Granada, Spain)

275

21. Nocturnal enuresis in children: The role of arginine–vasopressin K. Kamperis (Aarhus, Denmark)

289

SECTION 19

Eating disorders

22. Monogenic human obesity syndromes I.S. Farooqi (Cambridge, United Kingdom)

301

23. Hypothalamic microinflammation D. Cai and S. Khor (Bronx, United States)

311

24. Glucose and fat sensing in the human hypothalamus A.M. van Opstal (Leiden, The Netherlands)

323

25. Hypothalamus and neuroendocrine diseases: The use of human-induced pluripotent stem cells for disease modeling R. de Souza Santos, A.R Gross, and D. Sareen (Los Angeles and West Hollywood, United States)

337

26. Prader–Willi syndrome: Hormone therapies M. Tauber and G. Diene (Toulouse, France)

351

27. Transcriptomics of the Prader–Willi syndrome hypothalamus E.G. Bochukova (London, United Kingdom)

369

28. Disorders of hypothalamic function: Insights from Prader–Willi syndrome and the effects of craniopharyngioma J.E. Whittington and A.J. Holland (Cambridge, United Kingdom)

381

29. Animal models for Prader–Willi syndrome S. Zahova and A.R. Isles (Cardiff, United Kingdom)

391

30. Is there a hypothalamic basis for anorexia nervosa? V. Tolle, N. Ramoz, and J. Epelbaum (Paris and Brunoy, France)

405

SECTION 20

Reproduction, olfaction and sexual behavior

31. Sexual differentiation of the human hypothalamus: Relationship to gender identity and sexual orientation D.F. Swaab, S.E.C. Wolff, and A.-M. Bao (Amsterdam, The Netherlands and Hangzhou, China)

427

32. Klinefelter syndrome or testicular dysgenesis: Genetics, endocrinology, and neuropsychology A. Skakkebæk, M. Wallentin, and C.H. Gravholt (Aarhus, Denmark)

445

33. Neurobiology of puberty and its disorders S.F. Witchel and T.M. Plant (Pittsburgh, United States)

463

Index

497

xxxvi

CONTENTS OF RELATED VOLUMES CONTINUED

Contents of Volume 182 Foreword vii Preface ix Contributors xiii 1. Introduction: The human hypothalamus and neuropsychiatric disorders D.F. Swaab, R.M. Buijs, F. Kreier, P.J. Lucassen, and A. Salehi (Amsterdam, The Netherlands, Ciudad de M
exico, Mexico and Palo Alto, United States) SECTION 21

Trauma and iatrogenic disorders

2. Chronic traumatic encephalopathy and the nucleus basalis of Meynert E.J. Mufson, C. Kelley, and S.E. Perez (Phoenix, United States) SECTION 22

1

9

Neurobehavioral disorders

3. Hypothalamic stress systems in mood disorders F. Holsboer and M. Ising (Munich, Germany)

33

4. Light therapy for mood disorders B. Bais, W.J.G. Hoogendijk, and M.P. Lambregtse-van den Berg (Rotterdam, The Netherlands)

49

5. Neurobiology of peripartum mental illness J.L. Pawluski, J.E. Swain, and J.S. Lonstein (Rennes, France and Stony Brook and East Lansing, United States)

63

6. The hypothalamo–pituitary–adrenal axis and the autonomic nervous system in burnout A. Sj€ ors Dahlman, I.H. Jonsdottir, and C. Hansson (Gothenburg, Sweden)

83

7. Posterior hypothalamus as a target in the treatment of aggression: From lesioning to deep brain stimulation M. Rizzi, O. Gambini, and C.E. Marras (Milan and Rome, Italy) 8. The implications of hypothalamic abnormalities for schizophrenia H.-G. Bernstein, G. Keilhoff, and J. Steiner (Magdeburg, Germany) 9. The promiscuity of the oxytocin–vasopressin systems and their involvement in autism spectrum disorder A.M. Borie, C. Theofanopoulou, and E. Andari (Atlanta, New York, and Toledo, United States) SECTION 23

95

107

121

Epilepsy

10. Gelastic seizures and the hypothalamic hamartoma syndrome: Epileptogenesis beyond the lesion? J. Scholly and F. Bartolomei (Marseille, France)

143

11. The interactions between reproductive hormones and epilepsy E. Taubøll, J.I.T. Isoj€ arvi, and A.G. Herzog (Oslo, Norway, Oulu, Finland and Boston, United States)

155

SECTION 24

Neurodegenerative disorders

12. Alternative splicing in aging and Alzheimer's disease: Highlighting the role of tau and estrogen receptor a isoforms in the hypothalamus T.A. Ishunina (Kursk, Russia)

177

CONTENTS OF RELATED VOLUMES CONTINUED 13. Cholinergic neurodegeneration in Alzheimer disease mouse models A. Shekari and M. Fahnestock (Hamilton, Canada)

xxxvii 191

14. Autonomic disorders in Parkinson disease: Disrupted hypothalamic connectivity as revealed from resting-state functional magnetic resonance imaging E. Dayan and M. Sklerov (Chapel Hill, United States)

211

15. Hypothalamic a-synuclein and its relation to autonomic symptoms and neuroendocrine abnormalities in Parkinson disease E. De Pablo-Fernández and T.T. Warner (London, United Kingdom)

223

16. Lewy bodies in the olfactory system and the hypothalamus 235 M.G. Cersosimo, E.E. Benarroch, and G.B. Raina (Buenos Aires, Argentina and Rochester, United States) 17. Hypothalamic pathology in Huntington disease D.J. van Wamelen and N.A. Aziz (London, United Kingdom, Nijmegen, The Netherlands and Bonn, Germany)

245

18. Endocrine dysfunction in adrenoleukodystrophy M. Engelen, S. Kemp, and F. Eichler (Amsterdam, The Netherlands and Boston, United States)

257

19. Hypothalamic symptoms of frontotemporal dementia disorders R.M. Ahmed, G. Halliday, and J.R. Hodges (Sydney, Australia)

269

SECTION 25

Olfactory system

20. The vomeronasal organ: History, development, morphology, and functional neuroanatomy G.S. Stoyanov, N.R. Sapundzhiev, and A.B. Tonchev (Varna, Bulgaria)

283

21. Pheromone effects on the human hypothalamus in relation to sexual orientation and gender Y. Ye, Z. Lu, and W. Zhou (Beijing and Shenzhen, China)

293

22. Kallmann syndrome and idiopathic hypogonadotropic hypogonadism: The role of semaphorin signaling on GnRH neurons A. Cariboni and R. Balasubramanian (Milan, Italy and Boston, United States) 23. Olfaction as an early marker of Parkinson's disease and Alzheimer's disease I.M. Walker, M.E. Fullard, J.F. Morley, and J.E. Duda (Philadelphia and Aurora, United States) SECTION 26

307

317

Autonomic and sleep disorders

24. The hypothalamus and its role in hypertension V.D. Goncharuk (Moscow, Russia and Amsterdam, The Netherlands)

333

25. The heart is lost without the hypothalamus S. Pyner (Durham, United Kingdom)

355

26. Sleep disorders and the hypothalamus S. Overeem, R.R.L. van Litsenburg, and P.J. Reading (Heeze, Eindhoven, Utrecht, and Amsterdam, The Netherlands and Middlesbrough, United Kingdom)

369

xxxviii SECTION 27

CONTENTS OF RELATED VOLUMES CONTINUED Addiction and pain

27. Molecular genetics of neurotransmitters and neuropeptides involved in Internet use disorders including first insights on a potential role of hypothalamus’ oxytocin hormone C. Sindermann, R. Sariyska, J.D. Elhai, and C. Montag (Ulm, Germany and Toledo, United States) 28. The neurobiology of cluster headache M. Leone, S. Ferraro, and A.P. Cecchini (Milan, Italy) SECTION 28

389

401

Critical care and brain-death

29. Endocrine interventions in the intensive care unit A. T
eblick, L. Langouche, and G. Van den Berghe (Leuven, Belgium)

417

30. Hypothalamic function in patients diagnosed as brain dead and its practical consequences M. Nair-Collins and A.R. Joffe (Tallahassee, United States and Edmonton, Canada)

433

Index

447

Handbook of Clinical Neurology, Vol. 180 (3rd series) The Human Hypothalamus: Middle and Posterior Region D.F. Swaab, F. Kreier, P.J. Lucassen, A. Salehi, and R.M. Buijs, Editors https://doi.org/10.1016/B978-0-12-820107-7.00001-X Copyright © 2021 Elsevier B.V. All rights reserved

Chapter 1

Introduction: The middle and posterior hypothalamus DICK F. SWAAB1*, FELIX KREIER2, PAUL J. LUCASSEN3, AHMAD SALEHI4, AND RUUD M. BUIJS5 1

Department Neuropsychiatric Disorders, Netherlands Institute for Neuroscience, an Institute of the Royal Netherlands Academy of Arts and Sciences, Amsterdam, The Netherlands 2

Department Pediatrics, OLVG Hospitals, Amsterdam, The Netherlands

3

Brain Plasticity Group, Swammerdam Institute for Life Sciences, University of Amsterdam, Amsterdam, The Netherlands 4

Department of Psychiatry and Behavioral Sciences, Stanford Medical School, Palo Alto, CA, United States

5

Hypothalamic Integration Mechanisms Laboratory, Department of Cellular Biology and Physiology, Instituto de Investigaciones Biomedicas, Universidad Nacional Autónoma de Mexico (UNAM), Ciudad de Mexico, Mexico

SECTION 6#: SUPRAOPTIC AND PARAVENTRICULAR NUCLEUS: MORE THAN A NEUROENDOCRINE SYSTEM Vasopressin and oxytocin are synthesized in the classic hypothalamic magnocellular supraoptic and paraventricular neuroendocrine nuclei (SON and PVN) and transported to the posterior pituitary. For basic information on this system in the human brain, on the development and the involvement of the fetal SON and PVN in the process of birth, on the effect of glucocorticoids on these neuroendocrine nuclei and on the periventricular nucleus (see Swaab, 2003, Chapter 8). An elaborate accessory magnocellular neuroendocrine system is present in the human hypothalamus, with contacts to the posterior pituitary and blood vessels. The neurohormone vasopressin acts as antidiuretic hormone on the kidney, while oxytocin is involved in uterine contractions in labor and in milk ejection. In addition, vasopressin and oxytocin, by their extra-hypothalamic projections, are critically involved in the modulation of socioemotional behavior, partner preference and sexual orientation, memory modulation, emotion regulation, and trust and pain perception and anticipation, often in a sexual dimorphic way. The PVN harbors, moreover, sympathetic and parasympathetic preautonomic neurons. Subpopulations of preautonomic neurons exist that are dedicated to, e.g., the control

of visceral and subcutaneous adipose tissue, liver and adrenal, skeletal muscle, and white and brown adipose tissue. Oxytocin is a multifunctional neuropeptide involved in the regulation of eating behavior, energy homeostasis, and metabolism, and oxytocin-based therapeutics are studied for treating obesity. Neuropeptide alterations in the SON and PVN during aging and in neurodegenerative disorders lead to different functional sequelae. Three PVN systems have emerged as particularly relevant for anxiety disorders. First the oxytocin system is involved in social anxiety disorders. Second, the hypothalamic–pituitary–adrenal (HPA) axis is altered in in patients with panic disorder and generalized anxiety disorder by both, the central corticotropin-releasing hormone system and by cortisol. The third system is the hypothalamic–pituitary–thyroid axis that is likely to be compromised in panic disorder. Animal models and humans exposed to stress in early life are more likely to suffer from long-term behavioral, mental, metabolic, immune, and cardiovascular health consequences. Early life experience is epigenetically programming stress–response genes in the central nodes of the HPA-axis, the hippocampus, and hypothalamus. Through DNA methylation, early life stress affects multiple physiological systems, including the metabolic and immune systems (Lucassen et al., 2013). DNA methylation programming early in life might cause gene

#

Continued from the previous volume.

*Correspondence to: Prof. Dr. Dick F. Swaab, Netherlands Institute for Neuroscience, Meibergdreef 47, 1105 BA Amsterdam, The Netherlands. Tel: +31-20-566-5500, Fax: +31-20-566-6121, E-mail: [email protected]

2 D.F. SWAAB ET AL. expression changes later in life when and if the appropriate organ to stimulate drinking and increase blood pressure. signals are present, e.g., under as stress. The interactions Antibodies against the NaX-channel expressing neurons between the epigenetic matrix triggered by early stress of the subfornical organ cause hypernatremia in patients. and later events that modify the epigenome might result There is also evidence suggesting roles for regions of in reversal of early life adverse events or their aggravation. the lamina terminalis in reproduction, respiration, food Also the tripeptide thyrotropin-releasing hormone intake, and immune function. The lamina terminalis is is produced by neurons in the PVN and has dense nowadays used as a surgical corridor for the microsurgifiber networks in the hypothalamus (see Swaab, 2003, cal treatment of several conditions such as obstructive Chapter 8). As thyroid hormone is essential for the develhydrocephalus and diencephalic tumors. opment of the brain in the first years of life, undiagnosed The infundibular nucleus (in rodents “arcuate congenital hypothyroidism is one of the most common nucleus”) forms a morphologic and functional entity preventable causes of cognitive dysfunction. The identiwith the median eminence and contains neurons that are fication of families with isolated central hypothyroidism, controlling prolactin release, food intake, and metabolism facilitated by T4-based neonatal screening, has helped as well as reproduction and onset of puberty. The median to pinpoint several genes implicated in the pathogenesis eminence lacks a blood–brain barrier and provides an of isolated central hypothyroidism. entry to monitor for peripheral signals, such as nutrients, leptin, and ghrelin. Neurons of the infundibular nucleus contribute to the tubero-infundibular tract terminating in SECTION 7: ZONA INCERTA the median eminence on the hypophysial portal vessels The zona incerta is integrated within a basal ganglia netand regulate the anterior lobe of the pituitary. Infundibular work and plays a role in sensory-motor programming. In nucleus neurons are reciprocally connected with several addition, zona incerta’s connections with the superior colother hypothalamic nuclei, the brainstem, and reward liculus and the cerebral cortex as well as recent behavioral pathways. The hypophysial pars tuberalis is attached to studies point to this region as playing a role in cognitive the stalk of the pituitary and median eminence and conprocesses related to attention toward salient stimuli. The trols seasonal functions. In the median eminence, tanyzona incerta, dorsal to the subthalamic nucleus, has become cytes take over the blood–brain barrier function. These a target for deep brain stimulation in Parkinson’s disease. ependymo-glial cells form the wall of the third ventricle and send long extensions into the parenchyma to contact SECTION 8: VENTROMEDIAL NUCLEUS blood vessels and hypothalamic neurons. Tanycytes conAND DORSOMEDIAL NUCLEUS trol the transport of hormones and key metabolites in and out of the hypothalamus and play a key role in regulating The dorsomedial and ventromedial hypothalamic nuclei glucose balance, food intake, endocrine axes, seasonal are critical for numerous behaviors, including those in changes, reproductive function, and aging. response to psychological stressors. These behaviors are Kisspeptin neurons regulate the onset of puberty, coupled with changes in autonomic function, such as account for the pulsatile secretion of gonadotropinaltered blood pressure, thermogenesis, heart rate, sympareleasing hormone (GnRH) and mediate negative and thetic nerve activity, resetting of the baroreflex, and positive estrogen feedback signals to GnRH neurons. changes in pituitary function. In addition, they could also There are major kisspeptin cell groups in the preoptic be responsible for hypertension associated with obesity. area/rostral hypothalamus and infundibular nucleus that Recent human brain imaging studies have provided the integrate various types of environmental, endocrine, and first evidence that the DMH and VMH are involved in regmetabolic signals that can influence fertility. Kisspeptin ulating resting muscle sympathetic nerve activity levels in neurons that cosynthesize the tachykinin peptide neuroawake, healthy humans in the absence of any ongoing kinin B (NKB) and the opioid peptide dynorphin are stress. The cytoarchitecture and in and output of these called KNDy neurons. The absence of negative estronuclei are described in Swaab (2003, Chapters 9 and 10). gen feedback in menopause cause considerably higher kisspeptin and NKB activities. The increased NKB sigSECTION 9: CIRCUMVENTRICULAR naling toward the preoptic thermoregulatory centers ORGANS OF THE HYPOTHALAMUS plays a crucial role in hot flushes. NK3R is the recepThe subfornical organ, the organum vasculosum of the tor for NKB. NK3R antagonists are emerging as a terminalis, and the median preoptic nucleus comprise new treatment for menopausal hot flushes. In addition, the lamina terminalis. These structures play a role in intekisspeptin-54 has been used in clinical practice as a safe grating the control of multiple autonomic systems and trigger of oocyte maturation in women undergoing so in cardiovascular regulation and the control of fluid in vitro fertilization treatment who are also at high risk balance. Circulating angiotensin II acts at the subfornical of developing ovarian hyperstimulation syndrome.

INTRODUCTION: THE MIDDLE AND POSTERIOR HYPOTHALAMUS Women have higher kisspeptin and NKB expression in the infundibular nucleus than men. Kisspeptin and NKB are often coexpressed but not with dynorphin in postmortem material, thereby challenging the KDNy concept in humans. Female-typical expression of both kisspeptin and NKB were observed in the infundibular nucleus of male to female transsexual people, suggesting an atypical sexual differentiation of the brain. In addition, the infundibular nucleus is the main regulator of energy homeostasis. The peptidergic neurons, astrocytes and microglia in the infundibular nucleus receive from the circulation metabolic cues like insulin, leptin, glucose, and fatty acids, to monitor the energy state of the body. All these metabolic cues are integrated into an output signal regulating energy homeostasis through the release of neuropeptides. Proopiomelanocortin expressing neurons inhibit food intake and stimulate energy expenditure, whereas the agouti-related protein/ neuropeptide Yexpressing neurons stimulate food intake and inhibit energy expenditure. Dysfunctional regulation of energy homeostasis results in increased bodyweight and obesity, eventually leading to type 2 diabetes mellitus. Neurodegenerative disorders such as amyotrophic lateral sclerosis and frontotemporal dementia develop prominent changes in weight and eating behavior. These changes include alterations in metabolism, lipid levels, and insulin resistance. Emerging research suggests that these alterations may be mediated through changes in the hypothalamic function, but the exact mechanisms have still to be revealed.

SECTION 10: LATERAL TUBERAL NUCLEUS The lateral tuberal nucleus (LTN) is a hypothalamic region that has been identified with certainty only in humans and primates. For information on the human LTN in health and disease (see Swaab, 2003, Chapter 12). A parvalbuminpositive (PV1) nucleus in the lateral hypothalamus of rodents was proposed to be possibly homologous to the human LTN. The human LTN is intensely immunoreactive for somatostatin and FF1, but only weakly so or not at all for parvalbumin, calbindin, and calretinin. The rodent PV1-nucleus is intensely immunoreactive for parvalbumin but is not immunoreactive for either somatostatin, FF1, calbindin, or calretinin. Consequently, it was so far not possible to demonstrate a neurochemical homology between the human LTN and the rodent PV1-nucleus (Gerig and Celio, 2007; Meszár et al., 2012).

SECTION 11: LATERAL HYPOTHALAMIC AREA, PERIFORNICAL AREA Hypocretin-1 and -2 (or “orexin A and B”) are neuropeptides exclusively produced by a group of neurons in the

3

lateral and dorsomedial hypothalamus that project throughout the brain and act by their respective receptors. The hypocretin system is involved in sleep–wake regulation, reward mechanisms, food intake and metabolism, autonomic regulation including thermoregulation, and pain. The sleep disorder narcolepsy type 1, which is likely due to an autoimmune process, is caused by a 90% loss of hypocretin neurons. In addition, may the hypocretin system also be affected, but to a lesser extent and less specifically, in neurodegenerative diseases such as Alzheimer’s, Huntington’s, and Parkinson’s disease, immune-mediated disorders such as multiple sclerosis, neuromyelitis optica and anti-Ma2 encephalitis, and genetic disorders such as type 1 diabetus mellitus and Prader–Willi syndrome. Human heroin addicts have an increased number of hypocretin neurons, which are substantially smaller than those staining for hypocretin in control brains. It is proposed that an increased number of hypocetin neurons may underlie and maintain opioid or cocaine use disorders. Human narcoleptics, despite their prescribed use of several commonly addictive drugs, do not show significant evidence of dose escalation or substance use disorder.

SECTION 12: TUBEROMAMILLARY COMPLEX The tuberomamillary neurons in the posterior hypothalamus produces histamine and this nucleus innervates a large number of brain areas. The histaminergic system controls several basal physiological functions, including the sleep–wake cycle, energy and endocrine homeostasis, sensory and motor functions, and cognitive functions such as attention, learning and memory. Four G proteincoupled receptors mediate these effects. The two classic histamine postsynaptic receptors, the H1 and H2 receptors, mediate many of the central effects of histamine on, e.g., alertness and wakefulness. The H3 receptor is a pre- and postsynaptic receptor, which regulates release of histamine and several other neurotransmitters, including serotonin, GABA, and glutamate. H4 receptor is found in cerebral blood vessels and microglia. The H3 receptor antagonist Pitolisant is used to treat narcolepsy and hypersomnia. H1 receptor antagonists have been used to treat insomnia and H2 receptor antagonists have shown efficacy in treatment of schizophrenia. Histaminergic dysfunction may contribute to clinical disorders such as Parkinson’s disease, Alzheimer’s disease, Huntington’s disease, narcolepsy type 1, schizophrenia, Tourette syndrome, and autism spectrum disorder. H4R might indeed be a therapeutic target for Parkinson’s disease acting via microglia inhibition.

4

D.F. SWAAB ET AL.

SECTION 13: SUBTHALAMIC NUCLEUS The subthalamic nucleus increases the inhibitory drive of the basal ganglia. It has gained substantial interest as a target for deep brain stimulation for a variety of movement disorders. The level of overlap between putative limbic, associative, and motor zones within this nucleus is still debated. MRI and postmortem microscopic anatomical are now combined into an integrative research approach of this problem. Deep brain stimulation of the subthalamic nucleus is generally successful but can be complicated by adverse neuropsychiatric side effects, most commonly characterized by impulsivity and mood elevation, although depression, anxiety, apathy, and cognitive changes have also been reported. Deep brain stimulation of the subthalamic nucleus is now also employed to treat obsessive–compulsive disorder. The high rate of placebo responses in Parkinson’s disease clinical trials provided the impetus for investigating the underlying mechanisms. Specifically, placebo effects are associated with dopamine release in the striatum and changes in neuronal activity in the subthalamic nucleus, substantia nigra pars reticulata, and motor thalamus in Parkinson’s disease, as assessed through positron emission tomography and single-neuron recording during deep brain stimulation. Conversely, verbal suggestions of clinical worsening or drug dose reduction induce nocebo responses in Parkinson’s disease, which have been detected at both behavioral and electrophysiologic level. An important implication of this knowledge is that in clinical trials patients’ expectations should always be assessed.

SECTION 14: CORPORA MAMILLARIA, FORNIX, AND MAMILLOTHALAMIC TRACT The corpora mamillaria, their input by the fornix, and the mamillothalamic tract are involved in cognition. For the changes in these structures in Alzheimer’s disease and other neurodegenerative disorders, schizophrenia, and the effects of lesions (see Swaab, 2003,

Chapter 16). The fornix has gained recent interest as potential deep brain stimulation target to decrease rates of cognitive decline, enhance memory, aid visuospatial memorization and improve verbal recollection. Studies reported enhanced hippocampal acetylcholine release, synaptic plasticity, and decreased inflammatory responses in cortex and hippocampus. However, it is still premature to conclude that fornix deep brain stimulation can be used in the treatment of cognitive disorders. Histopathological alterations of the mamillary bodies occur after severe thiamine deficiency in combination with alcohol abuse in Wernicke’s encephalopathy and in the residuals of this disorder Korsakoff’s syndrome and Korsakoff’s psychosis. As a rule, Wernicke’s encephalopathy always occurs alongside neuropathology in other subcortical gray matter structures, notably the medial thalamus. When Wernicke’s encephalopathy is treated early in the process by large quantities of thiamine, full clinical recovery is likely to occur. As a residual syndrome after Wernicke’s encephalopathy, Korsakoff’s syndrome should be considered a form of acquired brain damage and pharmacological therapy is not effective. However, there is accumulating evidence that memory rehabilitation may possible in Korsakoff’s syndrome to a certain degree.

REFERENCES Gerig AT, Celio MR (2007). The human lateral tuberal nucleus: immunohistochemical characterization and analogy to the rodent PV1-nucleus. Brain Res 1139: 110–116. Mesza´r Z, Girard F, Saper CB et al. (2012). The lateral hypothalamic parvalbumin-immunoreactive (PV1) nucleus in rodents. J Comp Neurol 520: 798–815. Lucassen PJ, Naninck EFG, van Goudoever JB et al. (2013). Perinatal programming of hippocampal structure and function; emerging roles of stress, neurogenesis, epigenetics and early nutrition. Trends Neurosci 36: 621–631. Swaab DF (2003). The human hypothalamus. Basic and clinical aspects. In: MJ Aminoff, F Boller, DF Swaab (Eds.), Part I: nuclei of the hypothalamus, handbook of clinical neurology. vol. 79. Elsevier, Amsterdam.

Section 6 Supraoptic and paraventricular nucleus more than a neuroendocrine system

This page intentionally left blank

Handbook of Clinical Neurology, Vol. 180 (3rd series) The Human Hypothalamus: Middle and Posterior Region D.F. Swaab, F. Kreier, P.J. Lucassen, A. Salehi, and R.M. Buijs, Editors https://doi.org/10.1016/B978-0-12-820107-7.00002-1 Copyright © 2021 Elsevier B.V. All rights reserved

Chapter 2

Vasopressin and oxytocin beyond the pituitary in the human brain MORTEN MØLLER* Department of Neuroscience, Faculty of Health and Medical Sciences, University of Copenhagen, Copenhagen, Denmark

Abstract Vasopressin and oxytocin are primarily synthesized in the magnocellular supraoptic and paraventricular nuclei of the hypothalamus and transported to the posterior pituitary. In the human, an extensive accessory magnocellular neuroendocrine system is present with contact to the posterior pituitary and blood vessels in the hypothalamus itself. Vasopressin and oxytocin are involved in social and behavioral functions. However, only few neocortical areas are targeted by vasopressinergic and oxytocinergic nerve fibers, which mostly project to limbic areas in the forebrain, where also their receptors are located. Vasopressinergic/oxytocinergic perikarya in the forebrain project to the brain stem and spinal cord targeting nuclei and areas involved in autonomic functions. Parvocellular neurons containing vasopressin are located in the suprachiasmatic nucleus and synchronize the activity of the pacemaker in this nucleus. From the suprachiasmatic nucleus fibers project to the parvocellular part of the paraventricular nucleus, where preautonomic neurons project to the intermediolateral nucleus in the thoracic spinal cord, from where the superior cervical ganglion is reached whose noradrenergic fibers terminate in the pineal gland to stimulate melatonin secretion at night. The pineal gland is also innervated by vasopressin- and oxytocin-containing fibers reaching the gland via the “central innervation” in the pineal stalk, which might be involve in an annual regulation of melatonin secretion.

More than hundred years have passed since it was shown that aqueous extracts of the infundibular portion of the pituitary gland contain certain constituents, which exert a powerful physiological action on the blood pressure and on smooth muscle contraction (Dale, 1906). Fifty years later, the two posterior pituitary hormones, vasopressin and oxytocin, responsible for these effects were isolated from bovine posterior pituitaries and their molecular structures determined (du Vigneaud et al., 1953). The vasopressin and oxytocin genes are both located on chromosome 20 and are encoding two larger preprohormones, which are cleaved and modified to produce the active vasopressin and oxytocin molecules together with their carrier proteins, neurophysin II, and neurophysin I (Summar et al., 1990).

Vasopressin and oxytocin are predominantly expressed in nerve cell bodies and nerve fibers in the hypothalamus (Figs. 2.1 and 2.2). By use of antibodies against vasopressin and oxytocin or their neurophysins, it was already 40 years ago demonstrated, first in rodents (Swaab et al., 1975; Swaab and Pool, 1975; Vandesande et al., 1977), and later in primates including humans (Sofroniew et al., 1981), that vasopressin and oxytocin are synthesized in the nerve cell bodies of the classical magnocellular hypothalamic nuclei, the supraopticand paraventricular nuclei, both located in the medial area of the hypothalamus (Fig. 2.2A–D). From these nerve cell bodies, the hormones are transported, together with their neurophysins, to the posterior pituitary via anterograde axonal transport (Figs. 2.2A–C and 2.3A).

*Correspondence to: Morten Møller, Professor Emeritus, M.D., DrMedSci, Panum Institute 24.6.04, Blegdamsvej 3, DK-2200 Copenhagen, Denmark, Tel: +45-24-48-02-76, E-mail: [email protected]

8

M. MØLLER

Fig. 2.1. Montage of photomicrographs of coronal 6 mm thick paraffin sections through the rostral part of the human hypothalamus. The sections are stained with luxol fast blue for myelin and counterstained with cresyl violet according to the method of Kl€ uver-Barrera. (A) Coronal section of the human hypothalamus through the optic chiasm (Och) and the third ventricle (3.ventr). The perikarya in the magnocellular supraoptic (SO) and paraventricular nuclei (Pa) are visible. Columna fornicis (cf ) is dividing hypothalamus in a medial and lateral part. Lateral to the hypothalamus, the myelinated nerve fibers of the inferior thalamic peduncle (ithpe) are visible. IC, internal capsule; VA, ventral anterior thalamic nucleus. Bar ¼ 2 mm. (B) Coronal section of the most caudal suprachiasmatic region of human hypothalamus at the level of the supraoptic decussation (arrowheads). The paraventricular nucleus (Pa) is seen on the right side of the third ventricle. Cf, columna fornicis; IC, internal capsule; mtr, mamillothalamic tract; SO, supraoptic nucleus. Bar ¼ 1.5 mm. (C) Coronal section of the tuberal part of the human hypothalamus at the level of the neurohypophysis (Nh). The supraoptic nucleus (SO) is located lateral to the optic tract (ot). IC, internal capsule; If, lenticular fasciculus; mtr, mamillothalamic tract. Bar ¼ 1.5 mm. Modified from Møller M, Busch JR, Jacobsen C et al. (2018). The accessory magnocellular neurosecretory system of the human hypothalamus. Cell Tissue Res 1: 487–498, with permission.

Fig. 2.2. Montage of photomicrographs of rostro-caudal (A–E) coronal sections immunoreacted with an antibody against both neurophysin I + II, and selected areas in higher magnifications (D–F). (A) A rostral section through the optic chiasm (Och) with visualization of the paraventricular (Pa) and supraoptic (SO) nuclei. Among the immunoreactive nerve fibers of the hypothalamo– hypophysial tract in the right side of the photomicrograph, the circular nucleus (Cn) is seen. Several accessory magnocellular immunoreactive neurons and clusters of neurons are seen among the fibers of the hypothalamic–hypophysial tract. cf, columna fornicis. Bar¼ 2 mm. (B) Coronal section through the caudal part of the optic chiasm at the level of the supraoptic decussation (arrowhead). The optic tracts (ot) are now in this more caudal section emerging from the optic chiasm. In the left side of the picture, ovoid immunoreactive accessory magnocellular collections of neurons are present with close relation to the vascular system (arrows). Bar ¼ 2 mm. (C) Coronal section caudal to the optic chiasm at the tuberal level of the hypothalamus. In the right side of the picture an immunoreactive accessory magnocellular cluster of neurons is present (arrowhead). On the left side, the hypothalamo–hypophysial tract from the supraoptic nucleus (SO) is seen. cf, columna fornicis; ot, optic tract. Bar¼ 2 mm. (D) Higher magnification of the paraventricular nucleus seen in (A). The periventricular part (PaP), termed parvocellular subnucleus with low cell density of cell bodies, and the lateral part (PaM), termed magnocellular subnucleus, are seen. Bar ¼ 250 mm. (E) Higher magnification of the circular nucleus seen in (B). Bar ¼ 100 mm. (F) Part of a coronal section slightly caudal to the section seen in (B). A number of large collections of accessory magnocellular neuroendocrine neurons (arrows) are present in the hypothalamo–hypophysial tract. Bar ¼ 50 mm. From Møller M, Busch JR, Jacobsen C et al. (2018). The accessory magnocellular neurosecretory system of the human hypothalamus. Cell Tissue Res 1: 487–498, with permission.

10

M. MØLLER

Fig. 2.3. Montage of photomicrographs of coronal sections of human hypothalamus immunoreacted with antibodies against neurophysin I or neurophysin II. The photomicrographs are illustrating some of the unnamed magnocellular accessory neuroendocrine collections of nerve cell bodies in the human hypothalamus. (A) Hemisection of the hypothalamus at the middle of the optic chiasm (Och) illustrating a basal accessory magnocellular collection of nerve cells (Ba) immunoreactive for neurophysin I. cf, columna fornicis; Pa, paraventricular nucleus; SO, supraoptic nucleus; 3.ventr, third ventricle. Scale bar 1 mm. (B) Higher magnification of the basal accessory magnocellular nucleus (Ba) seen in A. Note the close relationship to the blood vessels. Scale bar 250 mm. (C) Part of a hypothalamic section with a large basal accessory magnocellular nucleus (Bae) immunoreactive for neurophysin II with close vascular connection. Scale bar 1 mm. (D) Single neurophysin II immunoreactive neuron with a process penetrating the endothelium of an arteriole (arrow). Scale bar 100 mm. (E) Higher magnification of the nucleus seen in (C). Note the close connection to the blood vessel. Scale bar 500 mm. From Møller M, Busch JR, Jacobsen C et al. (2018). The accessory magnocellular neurosecretory system of the human hypothalamus. Cell Tissue Res 1: 487–498, with permission.

VASOPRESSIN AND OXYTOCIN BEYOND THE PITUITARY IN THE HUMAN BRAIN

11

Fig. 2.4. Electron micrograph showing large axon terminal (Herring body) in the posterior pituitary containing large secretory granules close to fenestrated capillaries. The granules in the terminal bouton could be either vasopressin or oxytocin. A pituicyte is seen in the lower right corner. The inset shows part of a terminal bouton at a higher magnification. Bar ¼ 1 mm.

From the posterior pituitary the hormones are then released to the blood stream (Swanson and Sawchenko, 1983). On the ultrastructural level, vasopressin and oxytocin are found in nerve cell bodies and in nerve terminals stored in large electron dense granules with a diameter of about 100–150 nm (Fig. 2.4). Vasopressin and oxytocin are also synthesized in neurons located outside the classical magnocellular paraventricular and supraoptic nuclei. The largest collection of magnocellular neurons outside the paraventricular and supraoptic nuclei is confined to the accessory magnocellular neuroendocrine system in the rostral hypothalamus (Møller et al., 2018). Magnocellular vasopressin- and oxytocin-containing neurons in the hypothalamus target—in addition to the posterior pituitary, neurons located in other hypothalamic areas, in the limbic system including the amygdala, hippocampus, brain stem, spinal cord, and some areas of the neocortex, e.g., insular cortex, frontal operculum, subgenual cingulate gyrus, and the primary olfactory cortex. Finally, vasopressin is located in parvocellular neurons in the suprachiasmatic nucleus of rodents (Swaab et al., 1975) and humans (Mai et al., 1993; Dai et al., 1997). This nucleus contains the primary endogenous circadian oscillator of the central nervous system (Klein et al., 1991; Hastings and Herzog, 2004). In addition, in the paraventricular nucleus, vasopressin is colocalized with CRH in parvocellular neurons projecting to the median eminence (Raadsheer et al., 1993).

ACCESSORY MAGNOCELLULAR NEUROENDOCRINE SYSTEM OF THE HYPOTHALAMUS A high number of vasopressin and oxytocin synthesizing magnocellular neurons are located as single cells or in clusters outside the hypothalamic paraventricular and supraoptic nuclei in the preoptic and tuberal area of the hypothalamus (Saper, 2012; Møller et al., 2018) (Figs. 2.2A–F and 2.3A–E). These clusters and single cells are called the accessory magnocellular neurosecretory system of the hypothalamus. In human, this system was already observed by Gagel (1928) and called “Intermedi€arer Kern des Nucleus Supraopticus.” In a later study using Nissl and lipofuchsin staining, these aggregates of magnocellular neurons were described by Braak and Braak (1987). The major part of the accessory magnocellular neurosecretory neurons consists of elongated clusters of neurons located between the nerve fibers of the hypothalamo–hypophysial tract (Fig. 2.2B and F). The cell bodies are mostly immunoreactive for neurophysin II (vasopressin). Additional magnocellular clusters are located in the ventral part of the preoptic and tuberal part of the hypothalamus, lateral to the hypothalamo– hypophysial tract (Figs. 2.2B, C, and 2.3B, C), often in the close relationship to the blood vessels (Fig. 2.3D). Several of the neuronal clusters in the accessory magnocellular neurosecretory system have been named in

12

M. MØLLER

lower vertebrates (Peterson, 1966; Rhodes et al., 1981; Sofroniew and Glasmann, 1981; Castel and Morris, 1988; Rovsing et al., 2013) and primates (Antunes and Zimmerman, 1978), e.g., the anterior commissural nucleus, the anterior and posterior fornical nuclei, the lateral hypothalamic perivascular nucleus, the nucleus of the forebrain bundle, the mouse accessory nucleus, and the circular nucleus (Fig. 2.2A, E). In humans, most of these accessory nuclei have not been described or named, but it is possible in humans to identify the circular nucleus. This nucleus has a characteristic location in the preoptic area close to the hypothalamo–hypophysial tract and a circular appearance as well as a close apposition of the nucleus to a blood vessel (Fig. 2.2E). The termination of the projections of the accessory magnocellular neuroendocrine neurons has been studied in the rat using three-dimensional reconstruction of the hypothalamic pituitary connections after injections of horseradish peroxidase injections into the neurointermediate lobe (Fisher et al., 1979). In addition, the projections of these accessory neurons to the neurohypophysis were studied by postmortem application of a cyano dye in the human posterior lobe (Makarenko et al., 2002) and by their retrograde alterations following hypophysectomy (Swaab, 2003). Although some projections were found to the posterior pituitary, the studies agree that the accessory magnocellular neuroendocrine neurons mainly project to the median eminence with release of vasopressin and oxytocin to the portal capillary system. Moreover, in the rat, several of the accessory neurons, especially from the circular nucleus (Makarenko et al., 2002), project to other areas of the forebrain. Some of the neurophysin–immunoreactive terminals from neurons in the accessory magnocellular neuroendocrine system penetrate the vascular endothelium (Fig. 2.3D) suggesting a release of vasopressin to the blood stream (Møller et al., 2018). Such a vascular release may reflect a role for these neurons in regulating, e.g., blood osmolarity and electrolyte composition.

VASOPRESSIN- AND OXYTOCIN-CONTAINING NEURONS AND FIBERS IN THE FOREBRAIN Early immunohistochemical studies of the rat brain showed that both the paraventricular and supraoptic nucleus contribute with nerve fibers to limbic areas, e.g., fibers entering the stria terminalis (Buijs et al., 1978) to terminate in the central nucleus of the amygdala (Buijs and Swaab, 1979; Hernández et al., 2016) (Fig. 2.5). Vasopressin fibers in rats were also found in the lateral septum and the habenula (Buijs and Swaab, 1979). In rats treated with colchicine, a fairly high number of vasopressin immunoreactive perikarya were in addition seen in the dorsal portion of the medial amygdala, the nucleus

of the diagonal band, the lateral septum, anterior commissural nucleus, and the suprachiasmatic nucleus (Sofroniew, 1985). Most of the major projections in the brain from vasopressinergic, oxytocinergic, and neurophysin neurons, as described in rat, can also be localized in several primates, e.g., tree shrew (Ni et al., 2014), squirrel monkey, rhesus monkey (Sofroniew et al., 1981) including the human (Sofroniew et al., 1981; Fliers et al., 1986; Mai et al., 1993). Further, vasopressinergic but no oxytocinergic neurons are located in parvocellular neurons in the suprachiasmatic nucleus of six mammalian orders including in the human (Sofroniew, 1980). In the rat, parvocellular vasopressinergic neurons are also present in the bed nucleus of stria terminalis and the amygdala (DeVries et al., 1985). From the bed nucleus of stria terminalis, vasopressin fibers project to the lateral septum, the anterior amygdaloid area, the lateral habenular nucleus, the dorsal raphe nucleus, and the locus coeruleus (DeVries and Buijs, 1983). It is also of interest that studies of castrated rat indicate that vasopressin projections from the bed nucleus of the stria terminalis and the medial amygdaloid nucleus require the presence of gonadal hormones for their normal appearance (DeVries et al., 1985). Outside the paraventricular nucleus, some oxytocinergic cell bodies are located in the so-called parafornical group (Mai et al., 1993), which could be considered an extension of the paraventricular nucleus. Fibers from this cluster of neurons project laterally, caudally, and dorsally to the globus pallidus and do not appear to join fibers entering the hypothalamo–hypophysial tract (Mai et al., 1993). In a recent immunohistochemical paper (Rogers et al., 2018) detected vasopressin- and oxytocin-containing nerve fibers in several neocortical regions of the human, chimpanzee, and rhesus macaque. In human, vasopressin immunoreactive nerve fibers were found in all layers of the insular cortex (Fig. 2.6) as well as in the frontal operculum, the subgenual cingulate gyrus, and the primary olfactory cortex. In addition, oxytocin immunoreactive nerve fibers were observed in the same study in the gyrus rectus and the ventral and subgenual anterior cingulate gyrus (Brodmann’s areas 24 and 25). In the chimpanzee, vasopressin immunoreactive nerve fibers were also found in the insular cortex, but in the rhesus monkey, such fibers were not present (Rogers et al., 2018). Oxytocin immunoreactive cortical fibers were found in the gyrus rectus of the ventral anterior cingulate gyrus of chimpanzee as well as in the subgenual cingulate gyrus and superior frontal gyrus (Rogers et al., 2018). Older biochemical studies detecting vasopressin and oxytocin in acetic acid extracts from the forebrain by use of radioimmunoassay have detected high levels of vasopressin and oxytocin in the hypothalamus but very

VASOPRESSIN AND OXYTOCIN BEYOND THE PITUITARY IN THE HUMAN BRAIN

13

Fig. 2.5. Montage of serial Nissl-stained coronal sections through the forebrain, mesencephalon, and rostral medulla oblongata of a primate brain (African green monkey, Chlorocebus aethiops). On the right side of the picture, neurophysin immunoreactive perikarya (red circles) and fibers (red lines) are shown, projecting from the paraventricular nucleus and the accessory magnocellular neuroendocrine system to forebrain as well as mesencephalic and brain stem nuclei. The majority of these neurons contain vasopressin. From the suprachiasmatic nucleus, vasopressin containing perikarya (green circle) and fibers (green line) are seen projecting to nuclei in the forebrain. AcM, accessory magnocellular neurosecretory neuron; LR, lateral reticular nucleus; NT, nuclei of the solitary tract; PV, hypothalamic paraventricular nucleus; SCN, suprachiasmatic nucleus; X, dorsal nucleus of vagus nerve; XII, nucleus of hypoglossal nerve. The Nissl-stained sections are from: BrainMaps: An Interactive Multiresolution Brain Atlas; http://brainmaps.org [retrieved on December 15, 2019, with permssion].

low levels in some neocortical areas, e.g., the prefrontal cortex (Jenkins et al., 1984). Despite the study by Rogers et al. (2018), proof of a direct innervation of the larger areas of the neocortex with vasopressinergic and oxytocinergic fibers is still missing. Therefore, local release from dendrites and subsequent diffusion has been proposed to present an additional route of action (Stoop, 2012). Vasopressin and oxytocin are at the ultrastructural level located in large 100–150 nm large dense core granules (Fig. 2.4) both in the cytoplasm of the nerve cell bodies, as well as in the dendrites, axons, and axonal terminals (Buijs and Swaab, 1979; Ludwig and Leng, 2006). Somatodendritic exocytosis of vasopressin and oxytocin from nerve cell bodies in the hypothalamus allows the peptides to diffuse

to hypothalamic and possibly extrahypothalamic sites. In mammals, the close relationship of the vasopressin- and oxytocin-containing nerve cell bodies to the third ventricle would facilitate peptides to be released release directly into the cerebrospinal fluid and spread via diffusion of the cerebrospinal fluid circulation to areas outside the hypothalamus (Knobloch and Grinevich, 2014). Both vasopressin and oxytocin are present in the human cerebrospinal fluid, where the concentrations are a little higher than their concentrations in plasma (Kagerbauer et al., 2013). A correlation between the concentrations in the plasma and the cerebrospinal fluid has not been found (Kagerbauer et al., 2013). A possible direct secretion of vasopressin and oxytocin into the brain ventricles is further supported by the

14

M. MØLLER

Fig. 2.6. Montage of photomicrographs showing vasopressin immunoreactive nerve fibers in three neocortical areas of the insula in a human brain. Each Nissl-stained neocortical area shown in, (A), (B), and (C), contains an inset, which is shown at higher magnification in the next. (A) Radially oriented fibers in the neocortical layer II. (B) Fibers in the human insula oriented tangentially in the neocortical layer I. (C) A radial fiber that appears to end with tangential branches in cortical layer I. Scale bars ¼ 100 mm. From Rogers CN, Ross AP, Sahu SP et al. (2018). Oxytocin- and arginine vasopressin-containing fibers in the cortex of humans, chimpanzees, and rhesus macaques. Am J Primatol 80: e22875. https://doi.org/10.1002/ajp.22875, with permission.

demonstration of a circadian rhythm of vasopressin and oxytocin in the cerebrospinal fluid in the rhesus monkey (Perlow et al., 1982) and vasopressin in the rat (Schwartz and Reppert, 1985) with a peak during the day time. Lesion in the rat of the suprachiasmatic nucleus deleted the daytime peak of vasopressin (Schwartz and Reppert, 1985). Although the vasopressin neurons in the suprachiasmatic nucleus do not project dendrites or axons into the ventricular lumen (Dai et al., 1998), the results indicate a diffusion of the peptide via the brain parenchyma and the ependyma into the third ventricle. Vasopressin could also be released from the paraventricular nucleus, which is targeted by projections from the suprachiasmatic nucleus. It should also be noted that in study of human patients, where few patients had a clear temperature and cortisol rhythm, no indication of a vasopressin or oxytocin rhythm was observed in ventricular CSF (Swaab et al., 1987).

VASOPRESSIN AND OXYTOCIN ARE INVOLVED IN MOTIVATED SOCIAL BEHAVIOR Vasopressin and oxytocin are associated with several behavioral and physiological functions both in lower vertebrates (Johnson and Young, 2017) as well as in humans (Feldman et al., 2016). Across species, the oxytocin and vasopressin systems have consistently been linked to the modulation of motivated social behaviors (Donaldson

and Young, 2008; Caldwell and Albers, 2016). Further, while in normal aging and Alzheimer’s disease, a pronounced stability has been described of the vasopressin cells (Lucassen et al., 1994, 1997), disruption of the vasopressin and oxytocin systems in humans have been linked with neurological and psychiatric disorders such as autism (Guastella and Hickie, 2016), Williams syndrome (Dai et al., 2012), schizophrenia (MeyerLindenberg et al., 2011), depression (Zhou et al., 2001; Meynen et al., 2006), Huntington (van Wamelen et al., 2012), Parkinson (Purba et al., 1994), and Prader–Willi syndrome (Swaab, 1997). Different techniques have been used in investigations associating vasopressin and oxytocin with behavioral and physiological functions. Several studies have measured peripheral levels of vasopressin and oxytocin and correlating these levels with the behavioral symptoms (Zak et al., 2005). In other studies, the peptides have been applied via the nasal mucosa (Johnson and Young, 2017) and, especially in human studies, this intranasal administration has been the primary source of information about the effects of these peptides (Brambilla et al., 1989). Although effects on behavior using this procedure have varied, several themes have emerged. Oxytocin appears to facilitate social cognition and social approach (Preckel et al., 2014) and vasopressin has primarily been implicated in male-typical social behaviors, including aggression and pair-bond formation, and also mediates anxiogenic effects (Heinrichs and Domes, 2008).

VASOPRESSIN AND OXYTOCIN BEYOND THE PITUITARY IN THE HUMAN BRAIN

LOCALIZATION OF THE NEURONS INVOLVED IN BEHAVIORAL AND PHYSIOLOGICAL FUNCTIONS AND TARGETED BY VASOPRESSINERGIC AND OXYTOCINERGIC NERVE FIBERS From a neurobiological point of view, it would be logical that neocortical brain areas should be involved in the

15

behavioral and physiological brain functions initiated by vasopressin and oxytocin. However, immunohistochemical localization of vasopressin- and oxytocin-containing nerve fibers and cell bodies in the forebrain of higher vertebrates is mostly confined to hypothalamic and limbic areas including the hippocampus. Further, localization of vasopressin and oxytocin receptors by autoradiography or in situ hybridization confirmed the confinement of these receptors to mostly subcortical hypothalamic and limbic areas in the forebrain.

Forebrain

Fig. 2.7. Drawing of a coronal section of a monkey brain through the preoptic hypothalamic area at the level of the optic chiasm. The location of vasopressin receptors is shown on the right side (red circles), and the location of the oxytocin receptors are shown on the left side (green circles). Bl, basolateral amygdaloid nucleus; Bm, basomedial amygdaloid nucleus; C, cortical amygdaloid nucleus; Gp, globus pallidus; LS, lateral septal nucleus; Meynert, nucleus basalis of Meynert; NC, caudate nucleus; Opt ch, optic chiasm; Pic, piriform cortex; Pu, putamen; ST, bed nucleus of stria terminalis.

The actions of vasopressin are mediated by stimulation of three tissue-specific G protein-coupled receptors classified into the V1A, V1B, and V2 receptor subtypes, of which the vasopressin receptors V1A and V1B are expressed in the brain. In humans, receptor autoradiography using [3H]vasopressin on sections from human autopsy brains has shown binding in the lateral septal nucleus, the bed nucleus of stria terminalis, the basal amygdaloid nucleus, the basal nucleus of Meynert (Fig. 2.7), the arcuate nucleus, and the hilus of the dentate gyrus (Fig. 2.8) (Loup et al., 1991). Moreover, strong binding was also observed in the midline- and intralaminar nuclei of the thalamus (Fig. 2.8) (Loup et al., 1991). In a later study, using in situ hybridization for mRNA encoding the V1B receptor, expression was, in addition, found in the CA2 and CA3 subfields of the hippocampus (Young et al., 2006). The human studies are generally in accord

Fig. 2.8. Drawing of a coronal section of a monkey brain through the tuberal hypothalamic area at the level of the arcuate nuclei (Arc). The location of vasopressin receptor is shown on the right side (red circles), and the location of the oxytocin receptors is shown on the left side (green circles). AV, anteroventral thalamic nucleus; DG, dentate gyrus; ENT, entorhinal cortex; Gp, globus pallidus; Pu, putamen; SN, pars compacta of substantia nigra; Tmn, midline thalamic nuclei.

16

M. MØLLER

with studies in rats (Corbani et al., 2018), mice, and monkeys (Young et al., 2006). However, in situ hybridization for the V1A receptor in rats have also found binding in the neocortex and the cerebellar cortex (Szot et al., 1994). Oxytocin target only one G-protein-coupled receptor called the oxytocin receptor (OXTR). The human OXTR gene is located on chromosome 3p25. The gene contains four exons and three introns and has many polymorphic sites (Inoue et al., 1994). In humans, receptor autoradiography using [3H]oxytocin itself or the oxytocin antagonist [125I] D(CH2)51[Tyr(Me)2,Thr4,Tyr-NH29] vasotocin on sections from autopsy brains showed binding in the basal nucleus of Meynert, the amygdala, the nucleus of the vertical limb of the diagonal band of Broca, the ventral part of the lateral septal nucleus (Fig. 2.7), hippocampus, and the entorhinal cortex (Fig. 2.8) (Loup et al., 1989, 1991). Interestingly oxytocin binding was also found in motor areas such as the pars compacta of substantia nigra (Fig. 2.8). In a later immunohistochemical study using a monoclonal antibody on human brain sections, oxytocin receptors were further detected in perikarya and nerve fibers in the anterior cingulate and the piriform cortex (Fig. 2.7) (Boccia et al., 2013). Recent studies of the human dorsolateral prefrontal cortex have also detected mRNA encoding the oxytocin receptors in this part of the neocortex (Lee et al., 2018). However, characterization of neurons containing the receptors has not yet been performed. The localization of oxytocin receptors in the human brain is generally in accord with the localization in lower vertebrates. In rodents the oxytocin receptor is in the olfactory pathways such as olfactory bulb, accessory olfactory nucleus, and amygdala as well as in the lateral septum, anterior olfactory nucleus, anterior cingulate cortex, central and medial nuclei of the amygdala, and ventral subiculum (Shapiro and Insel, 1989). Subsequently, mRNA encoding oxytocin was localized in steroid-hormone sensitive hypothalamic nuclei of the rat such as in the substantia nigra, an area involve in the planning of motor function, and dopaminergic ventral tegmental area from which the mesolimbic and mesocortical dopaminergic pathways originate (Ostrowski, 1998). In the rhesus monkey, oxytocin receptor expression is, as in the human, located in the nucleus basalis of Meynert and the ventromedial hypothalamus, but also in the pedunculopontine tegmental nucleus, the superficial gray layer of the superior colliculus, and the trapezoid body (Freeman et al., 2014). Summarizing, vasopressin and oxytocin receptors in vertebrates including humans have mostly been detected in the basal forebrain, in limbic, and hypothalamic areas, although in a recent immunohistochemical study, oxytocin receptors have also been localized in the anterior cingulate cortex (Boccia et al., 2013). This indicates that vasopressin and oxytocin influence the neocortex via

both an activation of hypothalamic and limbic areas, which then project to neocortical areas. However, a direct innervation of some neocortical areas with vasopressinand oxytocin-containing nerve fibers have been shown (Rogers et al., 2018).

Brainstem and spinal cord Early studies in rodents showed that the majority of neurophysin immunoreactive nerve fibers in the brain stem originate from perikarya in the hypothalamic paraventricular nucleus with a contribution from the accessory magnocellular neurons (Fig. 2.5) (Swanson, 1977; Sofroniew, 1980; Voorn and Buijs, 1983; Buijs, 1978; Buijs et al., 1990). The majority of the brain stem nuclei targeted by vasopressinergic and oxytocinergic projections are involved in autonomic functions such as the sensory nucleus of the solitary tract, the commissural nucleus, and the parasympathetic dorsal nucleus of the vagus nerve (Fig. 2.5). Immunoreactive nerve fibers are also consistently present along the marginal zone of substantia gelatinosa of the trigeminal nucleus, an area continuing caudally into the spinal (Sofroniew, 1980). In addition, the lateral reticular nucleus (Sofroniew, 1980) (Fig. 2.5), a nucleus projecting to the cerebellum and involved in motor reflexes, receives neurophysincontaining fibers. Also in the mouse, the spinal trigeminal nucleus and nucleus of the solitary tract are heavily innervated by oxytocin immunoreactive nerve fibers as well as the lateral reticular nucleus and most of the raphe nuclei (Rood and De Vries, 2011). By specific antibodies against vasopressin and oxytocin it was observed that the density of oxytocin immunoreactive nerve terminals in the target areas is higher than the number of vasopressin immunoreactive nerve terminals (Sofroniew, 1980; Sofroniew et al., 1981). In the spinal cord of rodents, vasopressin immunoreactive fibers are located in lamina I–III of the dorsal horn (Swanson and McKellar, 1979; Sofroniew, 1983). At the thoracic level, vasopressin fibers are also present in lamina X close to the central canal. From lamina X, fibers project laterally toward the sympathetic intermediolateral nucleus (Fig. 2.9). Few oxytocin immunoreactive nerve fibers are also present in the spinal cord in the dorsal horn and lamina X (Fig. 2.9) (Sofroniew, 1983). The immunohistochemical studies of human and monkeys (Swanson and McKellar, 1979; Sofroniew, 1980) are in accord with studies of in rodents. Thus vasopressin and oxytocin immunoreactive nerve fibers are innervating the parabrachial nucleus (van Zwieten et al., 1996), the dorsal motor nucleus of vagus, the nucleus of the solitary tract, the lateral reticular nucleus, lamina X, and Lamina I–III of the spinal cord (Sofroniew, 1980). In the thoracic part of the spinal cord, Neurophysin immunoreactive nerve fibers project in most primates

VASOPRESSIN AND OXYTOCIN BEYOND THE PITUITARY IN THE HUMAN BRAIN

Fig. 2.9. Vasopressin (red lines) and oxytocin (green lines) immunoreactive nerve fibers in the mammalian spinal cord. Immunoreactive fibers are seen lamina I–III (La1–III) of the dorsal horn and in lamina X (LaX), close to the central canal. Form lamina X, immunoreactive nerve fibers project toward the sympathetic intermediolateral nucleus (IML).

laterally to the autonomic intermediolateral nucleus (Swanson and McKellar, 1979; Sofroniew et al., 1981). The immunohistochemical localization of vasopressin and oxytocin fibers is also in accord with the receptor autoradiographical studies visualizing the location of vasopressin and oxytocin receptors in the brain stem and spinal cord (Loup et al., 1989, 1991) obtained, using [3H]vasopressin and the oxytocin antagonist [125I] D (CH2)51[Tyr(Me)2,Thr4,Tyr-NH29] vasotocin on sections from human autopsy brains. Vasopressin receptors were present in the nucleus of the solitary tract and area postrema, the substantia gelatinosa of the caudal spinal trigeminal nucleus (Fig. 2.10). In addition, weak labeling was found in the pars compacta of the substantia nigra and in the dorsal nucleus of vagus nerve. In the thoracic spinal cord the receptors were located in the substantia gelatinosa of the dorsal horn (Loup et al., 1991). Oxytocin receptors were found essentially in the same areas as the vasopressin receptors (Fig. 2.10). However, oxytocin receptors were in addition seen in the hypoglossal nucleus (Fig. 2.10). These brain stem and spinal cord binding sites indicate a neuromodulatory function of oxytocin and vasopressin in spinal and brain reflexes involved in sensory, autonomic, and motor functions in the human central nervous system.

VASOPRESSIN IN THE SUPRACHIASMATIC NUCLEUS The suprachiasmatic nucleus of vertebrates (Fig. 2.5) contains an endogenous circadian oscillator generating daily rhythms of melatonin and cortisol secretion,

17

Fig. 2.10. Drawing of a coronal section of the primate brain stem at the caudal part of the fourth ventricle at the level of area postrema (AP). The location of vasopressin receptor is shown on the right side (red circles), and the location of the oxytocin receptors are shown on the left side (green circles). A high number of oxytocin and a lower number of vasopressin receptors are located in the area of area postrema, the nucleus of the solitary tract (NS), and the vagus nucleus (DX). Oxytocin receptors are also seen in the hypoglossal nucleus (XII. IO, inferior olivary complex; LR, lateral reticular nucleus; Py, pyramis).

temperature, and running activity (Klein et al., 1991; Reppert and Weaver, 2002). Circadian oscillations are generated by a set of clock genes located in the suprachiasmatic neurons forming a transcriptional autoregulatory feedback loop (Lowrey and Takahashi, 2011). The rhythm of the SCN can be phase changed by light activating melanopsin containing intrinsically photosensitive retinal ganglionic cells (Hattar et al., 2002). From these retinal photoreceptors impulses are transmitted via the retinohypothalamic tract, located in the optic nerve to the suprachiasmatic nucleus (Mikkelsen et al., 1995; Dai et al., 1998; Rovsing et al., 2010). Two neurotransmitters, glutamate and pituitary adenylate cyclase activating peptide (PACAP) located in the retinohypothalamic tract are able to phase change the rhythm of the SCN (Hannibal et al., 2000). In lower vertebrates the SCN is relatively large and located in the basal part of the preoptic area of the hypothalamus. In the human, the SCN is smaller and located in the most ventral part of the rostral preoptic hypothalamic area where the hypothalamus is connected to the optic chiasm via a slender extension of the basal part of the brain (Swaab et al., 1985). Parvocellular neurons containing vasopressin, but not oxytocin, are present in the suprachiasmatic nucleus in both humans and lower vertebrates (Dierickx and Vandesande, 1977; Stopa et al., 1984; Swaab, 2003; Rovsing et al., 2010). The vasopressin gene in the suprachiasmatic nucleus contains an E-box in the promoter region, which is targeted by transcription factors encoded by the clock genes (Hastings and Herzog, 2004). Thereby, vasopressin neurons in the suprachiasmatic

18

M. MØLLER

nucleus exhibit a circadian expression and are involved in coupling the activity of the neurons in the suprachiasmatic nucleus (Li et al., 2009) as well as projecting the circadian output to other parts of the brain. A major vasopressinergic projection from the suprachiasmatic nucleus is to the paraventricular nucleus itself (Fig. 2.5), medial preoptic area, the subparaventricular area, the dorsomedial nucleus, and the paraventricular nucleus of the thalamus (DeVries et al., 1985; Vrang et al., 1995; Dai et al., 1997; Kalsbeek et al., 2010). However, lesion studies in the rat have shown that the vasopressin terminals in the lateral septum and lateral habenula do originate from the bed nucleus of stria terminalis (Hoorneman and Buijs, 1982; de Vries and Buijs, 1983). From the paraventricular nucleus, neurons project via the central gray of the mesencephalon and the brain stem to the intermediolateral nucleus of the thoracic spinal cord (Larsen et al., 1998; Teclemariam-Mesbah et al., 1999). The transmitter in the projection from the paraventricular nucleus to the intermediolateral nucleus of the thoracic spinal cord has not been determined in the human. However, in rats, neurons in the paraventricular nucleus projecting to the posterior horn contain either vasopressin or oxytocin (Teclemariam-Mesbah et al., 1999; Gamal-Eltrabily et al., 2018). From the intermediolateral nucleus, fibers enter the sympathetic trunk to make synapses in the superior cervical ganglion. From the superior cervical ganglion postganglionic fibers follow the internal carotid to innervate the pineal gland with sympathetic nerve fibers containing noradrenalin and in most fibers also neuropeptide Y (Møller and Baeres, 2002).

VASOTOCIN, VASOPRESSIN, AND OXYTOCIN IN THE MAMMALIAN PINEAL GLAND Vasotocin is an oligopeptide in nonmammalian lower vertebrates, e.g., fishes, birds, and amphibians and is a homologue to vasopressin and oxytocin in the higher vertebrates. Vasotocin, like vasopressin and oxytocin, serves in lower vertebrates as an important modulator of social behavior in addition to its peripheral functions related to osmoregulation, reproductive physiology (Wilczynski et al., 2017). Vasotocin was in earlier studies proposed to be present in the pineal gland of mammals. This suggestion was based on that the extracts of bovine pineals had antidiuretic, hydroosmotic, and rat uterine activities (Pavel and Petrescu, 1966). However, later immunohistochemical studies of sections of the bovine pineal by use of specific antibodies raised against vasotocin, vasopressin, and oxytocin could not confirm the

Fig. 2.11. Vasopressin immunoreactive nerve fibers (arrows) in the pineal gland of the hedgehog. Both perivascular and intraparenchymal immunoreactive nerve fibers are seen. Bar ¼ 10 mm. Reproduced from N€ urnberger F, Korf HW (1981). Oxytocin- and vasopressin immunoreactive nerve fibers in the pineal gland of the hedgehog, Erinaceus europaeus L. Cell Tissue Res 220: 87–97, with permission.

presence of vasotocin in the bovine pineal gland. Only vasopressin and oxytocin could be visualized (Pevet et al., 1981). By use of immunohistochemistry, vasopressin and oxytocin molecules were detected in nerve fibers in the pineal gland of the rat (Buijs and Pevet, 1980), dog (Matsuura et al., 1983), hedgehog (Figs. 2.11 and 2.12) (N€urnberger and Korf, 1981), and macaque (Rønnekleiv, 1988). The location of the vasopressinergic and oxytocinergic pineal nerve fibers in the investigated animals indicates that the fibers enter the pineal from the brain via the pineal stalk and thus belong to the “central innervation” of the pineal gland. The “central innervation” of the pineal gland has been demonstrated in several morphological (Møller, 1976, 1981; Rønnekleiv and Møller, 1979; Schneider et al., 1981) and electrophysiological studies (Rønnekleiv et al., 1980; Semm et al., 1981). The perikaryal origin of the vasopressin- and oxytocin-containing nerve fibers

VASOPRESSIN AND OXYTOCIN BEYOND THE PITUITARY IN THE HUMAN BRAIN

19

Fig. 2.13. Electron micrograph of a sympathetic nerve terminal in the superficial pineal gland of the rat. The noradrenalin containing transmitter vesicles (arrows), 40–60 nm in diameter, contain a small dense core granule, often located in an eccentric position. The transmitter vesicle with the large granule (arrowhead) contains a neuropeptide, probably neuropeptide Y. Ast, astrocyte; Mi, mitochondria. Bar ¼ 300 nm. Fig. 2.12. Oxytocin immunoreactive nerve fibers (arrowheads) in the pineal gland of the hedgehog. Most of the nerve fibers are located in a perivascular position. Bar ¼ 10 mm. Reproduced from N€urnberger F, Korf HW (1981). Oxytocinand vasopressin immunoreactive nerve fibers in the pineal gland of the hedgehog, Erinaceus europaeus L. Cell Tissue Res 220: 87–97, with permission.

in the mammalian pineal gland probably originate in the hypothalamic paraventricular nucleus because retrograde tracing studies with injection of horseradish peroxidase into the pineal gland has traced nerve fibers back to the hypothalamic paraventricular nucleus of the guinea pig (Korf and Wagner, 1980). The vasopressinergic and oxytocinergic mammalian pineal innervation is, therefore, part of the pineal central innervation. The secretion of melatonin in the mammalian pineal gland is primarily regulated by sympathetic nerve fibers originating from perikarya located in the superior cervical ganglion and the fibers entering the brain together with the internal carotid artery (Møller and Baeres, 2002; Klein, 2007). The majority of sympathetic nerve fibers also contain, in addition to noradrenalin, neuropeptide Y (NPY) (Zhang et al., 1991; Møller et al., 1994). However, in addition to the sympathetic pineal innervation, anatomical (Kenny, 1961; Rønnekleiv and Møller, 1979; Shiotani et al., 1986) and biochemical (Laitinen et al., 1995; Phansuwan-Pujito et al., 1999) data have

shown the presence of a cholinergic pineal innervation. The cholinergic, parasympathetic nerve fibers also contain vasoactive intestinal peptide (VIP) and peptide histidine isoleucine (Møller et al., 1985). The VIPergic nerve fibers have in the rat been traced back to the parasympathetic sphenopalatine ganglion (in humans called the pterygopalatine ganglion) (Shiotani et al., 1986). As further evidence for the presence of a parasympathetic cholinergic innervation of the mammalian pineal gland, an intrapineal ganglion is present in several primate species, e.g., human fetus (Møller, 1976), monkey (Le Gros Clark, 1940; Hartmann, 1957; Ichimura, 1992), and nonprimate species, e.g., rabbit (Romijn, 1973), ferret (David and Herbert, 1973), and ground squirrel (Matsushima and Reiter, 1978). Ultrastructural studies show that nerve terminals are making synaptic contacts with the ganglionic perikarya. In the ferret, the nerve fibers making synaptic contacts with the intrapineal gangionic cells originate in the habenular nuclei and enter the pineal via the pineal stalk (David and Herbert, 1973). It is neuroanatomically interesting that some intrapineal nerve fibers containing the neuropeptide PACAP and calcitonin gene-related peptide (Møller et al., 1999) originate in the trigeminal ganglion, which is a classical sensory ganglion (Møller and Baeres, 2003). At the ultrastructural level intrapineal nerve terminals containing noradrenalin and adrenalin contain transmitter vesicles with a diameter of 40–60 nm, which are

20

M. MØLLER located on the cell membranes. Signals in the pinealocytes can spread to other pinealocytes via gap junctions connecting the cells (Møller, 1976; Eugenin et al., 2019).

CONCLUSIONS

Fig. 2.14. Electron micrograph of a nonsympathetic nerve terminal in the superficial pineal gland of the rat. The terminal contains small electron lucent transmitter vesicles and large granular vesicles (arrowheads), which might contain vasopressin or oxytocin. Bar ¼ 300 nm.

endowed with an excentric located dense core (Rønnekleiv and Møller, 1979) (Fig. 2.13). The nonsympathetic nerve terminals contain 40–60 nm clear vesicles and in addition larger granules, 100–150 nm in diameter, with an electron dense core (Fig. 2.14) containing the neuropeptide. Regarding vasopressin and oxytocin, these peptides are able to modulate the secretion of melatonin from periperfused rat pineal gland (Simonneaux et al., 1990). Further, vasopressin is also able to increase intracellular calcium concentration in 5% of cells in the superficial rat pineal gland (Schomerus et al., 1995). In addition, daily variations of vasopressin not triggered by light have been described in the rat pineal gland (Liu and Burbach, 1987) and seasonal variations in both vasopressin and oxytocin have been observed in the hedgehog pineal (Figs. 2.11 and 2.12) (N€ urnberger and Korf, 1981). It is also of interest that quantitative studies of vasopressin immunoreactive nerve cell bodies in the human suprachiasmatic nucleus have shown a decrease in the number of immunoreactive perikarya during summertime (Hofman et al., 1993). Thus vasopressin and oxytocin might be involved in modulating both daily and seasonal melatonin secretion. It must finally be mentioned that the sympathetic nerve fibers innervating the pineal gland are not making synaptic contacts with the pinealocytes, but the transmitters diffuse from the nerve fiber terminals to receptors

Vasopressin and oxytocin are hypothalamic hormones primarily synthesized in the supraoptic and paraventricular nuclei, but also in additional hypothalamic nuclei and clusters, called the accessory magnocellular neuroendocrine system. The hormones from the supraoptic and paraventricular nuclei are primarily transported to the posterior pituitary lobe to be released to the systematic blood stream. However, the vasopressinergic and oxytocinergic neurons also innervate the median eminence, hypothalamic, limbic, and hippocampal areas, where the neurons contain receptors for the hormones and could be involved in behavioral and physiological functions associated with a stimulation of the brain with these hormones. Further, neuronal projections from the vasopressinergic and oxytocinergic nuclei and clusters, including the vasopressinergic neurons in the suprachiasmatic nucleus, target areas in the brain stem and spinal cord involved in primarily autonomic function. A projection containing vasopressin or oxytocin to the pineal gland via the pineal stalk might be involved in annual changes in melatonin production.

REFERENCES Antunes JL, Zimmerman EA (1978). The hypothalamic magnocellular system of the rhesus monkey: an immunocytochemical study. J Comp Neurol 181: 539–566. Boccia ML, Petrusz P, Suzuki K et al. (2013). Immunohistochemical localization of oxytocin receptors in human brain. Neuroscience253: 155–164. https://doi. org/10.1016/j.neuroscience.2013.08.048. Braak H, Braak E (1987). The hypothalamus of the human adult: chiasmatic region. Anat Embryol 176: 315–330. Brambilla F, Bondiolotti GP, Maggioni M et al. (1989). Vasopressin (DDAVP) therapy in chronic schizophrenia: effects on negative symptoms and memory. Neuropsychobiology 20: 113–119. https://doi.org/10.1159/000118483. Buijs RM (1978). Intra- and extrahypothalamic vasopressin and oxytocin pathways in the rat. Pathways to the limbic system, medulla oblongata and spinal cord. Cell Tissue Res 192: 423–435. Buijs RM, Pevet P (1980). Vasopressin- and oxytoeincontaining fibres in the pineal gland and subcommissural organ of the rat. Cell Tissue Res 205: 11–17. Buijs RM, Swaab DF (1979). Immuno-electron microscopical demonstration of vasopressin and oxytocin synapses in the limbic system of the rat. Cell Tissue Res 204: 355–365. Buijs RM, Swaab DF, Dogterom J et al. (1978). Intra- and extrahypothalamic vasopressin and oxytocin pathways in the rat. Cell Tissue Res 186: 423–433.

VASOPRESSIN AND OXYTOCIN BEYOND THE PITUITARY IN THE HUMAN BRAIN Buijs RM, Van der Beek EM, Renaud LP et al. (1990). Oxytocin localization and function in the A1 noradrenergic cell group: ultrastructural and electrophysiological studies. Neuroscience 39: 717–725. Caldwell HK, Albers HE (2016). Oxytocin, vasopressin, and the motivational forces that drive social behaviors. Curr Top Behav Neurosci 27: 51–103. Castel M, Morris JF (1988). The neurophysin-containing innervation of the forebrain of the mouse. Neuroscience 24: 931–966. Corbani M, Marir R, Trueba M et al. (2018). Neuroanatomical distribution and function of the vasopressin V1B receptor in the rat brain deciphered using specific fluorescent ligands. Gen Comp Endocrinol 258: 15–32. Dai J, Swaab DF, Buijs RM (1997). Distribution of vasopressin and vasoactive intestinal polypeptide (VIP) fibers in the human hypothalamus with special emphasis on suprachiasmatic nucleus efferent projections. J Comp Neurol 383: 397–414. Dai J, Swaab DF, Van der Vliet J et al. (1998). Human retinohypothalamic tract as revealed by in vitro postmortem tracing. J Comp Neurol 397: 357–370. Dai L, Carter CS, Ying J et al. (2012). Oxytocin and vasopressin are dysregulated in Williams Syndrome, a genetic disorder affecting social behavior. PLoS One 7: e38513. https:// doi.org/10.1371/journal.pone.0038513. Epub 2012 Jun 12, PMID: 22719898; PMCID: PMC3373592. Dale HH (1906). On some Physiological actions of ergot. J Physiol 34: 163–206. David GFX, Herbert J (1973). Experimental evidence for a synaptic connection between habenula and pineal ganglion in the ferret. Brain Res 64: 327–343. de Vries GJ, Buijs RM (1983). The origin of the vasopressinergic and oxytocinergic innervation of the rat brain with special reference to the lateral septum. Brain Res 273: 307–317. https://doi.org/10.1016/0006-8993(83)90855-7. PMID: 6311351. DeVries GJ, Buijs RM (1983). The origin of the vasopressinergic and oxytocinergic innervation of the rat brain with special reference to the lateral septum. Brain Res 273: 307–317. DeVries GJ, Buijs RM, Van Leeuwen FW et al. (1985). The vasopressinergic innervation of the brain in normal and castrated rats. J Comp Neurol 233: 236–254. Dierickx K, Vandesande F (1977). Immunocytochemical localization of the vasopressinergic and the oxytocinergic neurons in the human hypothalamus. Cell Tissue Res 184: 15–27. Donaldson ZR, Young LJ (2008). Oxytocin, vasopressin, and the neurogenetics of sociality. Science 322: 900–904. Du Vigneaud V, Ressler C, Swan JM et al. (1953). The synthesis of an octapeptide amide with the hormonal activity of oxytocin. J Am Chem Soc 75: 4879–4880. Eugenin EA, Valdebenito S, Gorska AM et al. (2019). Gap junctions coordinate the propagation of glycogenolysis induced by norepinephrine in the pineal gland. J Neurochem 151: 558–569. https://doi.org:10.1111/jnc. 14846.

21

Feldman R, Monakhov M, Pratt M et al. (2016). Oxytocin pathway genes: evolutionary ancient system impacting on human affiliation, sociality, and psychopathology. Biol Psychiatry 79: 174–184. Fisher AW, Price PG, Burford GD et al. (1979). A 3-dimensional reconstruction of the hypothalamoneurohypophysial system of the rat. The neurons projecting to the neuro/intermediate lobe and those containing vasopressin and somatostatin. Cell Tissue Res 204: 343–354. Fliers E, Guldenaar SE, van de Wal N et al. (1986). Extrahypothalamic vasopressin and oxytocin in the human brain; presence of vasopressin cells in the bed nucleus of the stria terminalis. Brain Res 375: 363–367. Freeman SM, Inoue K, Smith AL et al. (2014). Psychoneuroendocrinology The neuroanatomical distribution of oxytocin receptor binding and mRNA in the male rhesus macaque (Macaca mulatta). Psychoneuroendocrinology 45: 128–141. Gagel O (1928). Zur Topik und feineren Histologie der vegetativen Kerne des Zwischenhirns. Z Anat Entwicklungsgesch 37: 548–584. Gamal-Eltrabily M, Ma´rquez-Morales C, Martı´nez-Lorenzana G et al. (2018). Peptidergic nature of nociception-related projections from the hypothalamic paraventricular nucleus to the dorsal horn of the spinal cord. Neurosci Lett 685: 124–130. Guastella AJ, Hickie IB (2016). Oxytocin treatment, circuitry, and autism: a critical review of the literature placing oxytocin into the autism context. Biol Psychiatry 79: 234–242. Hannibal J, Møller M, Ottersen OP et al. (2000). PACAP and glutamate are co-stored in the retinohypothalamic tract. J Comp Neurol 418: 147–155. € Hartmann F (1957). Uber die Innervation der Epiphysis cerebri einiger S€augetiere. Z Zellforsch 46: 416–429. Hastings MH, Herzog ED (2004). Clock genes, oscillators, and cellular networks in the suprachiasmatic nucleus. J Biol Rhythms 19: 400–413. Hattar S, Liao HW, Takao M et al. (2002). Melanopsincontaining retinal ganglion cells: architecture, projections, and intrinsic photosensitivity. Science (New York, NY) 295: 1065–1070. Heinrichs M, Domes G (2008). Neuropeptides and social behaviour: effects of oxytocin and vasopressin in humans. Prog Brain Res 170: 337–350. Herna´ndez VS, Herna´ndez OR, Perez de la Mora M et al. (2016). Hypothalamic vasopressinergic projections innervate central amygdala GABAergic neurons: implications for anxiety and stress coping. Front Neural Circuits 10: 92. eCollection 2016. Hofman MA, Furba JS, Swaab DF (1993). Annual variations in the vasopressin neuron population of the human suprachiasmatic nucleus. Neuroscience 53: 1103–l112. Hoorneman EM, Buijs RM (1982). Vasopressin fiber pathways in the rat brain following suprachiasmatic nucleus lesioning. Brain Res 243: 235–241. Ichimura T (1992). The ultrastructure of neuronal-pinealocytic interconnections in the monkey pineal. Microsc Res Tech 21: 124–135.

22

M. MØLLER

Inoue T, Kimura T, Azuma C et al. (1994). Structural organization of the human oxytocin receptor gene. J Biol Chem 269: 32451–32456. Jenkins JS, Ang VT, Hawthorn J et al. (1984). Vasopressin, oxytocin and neurophysins in the human brain and spinal cord. Brain Res 291: 111–117. Johnson ZV, Young LJ (2017). Oxytocin and vasopressin neural networks: implications for social behavioral diversity and translational neuroscience. Neurosci Biobehav Rev 76: 87–98. Kagerbauer SM, Martin J, Schuster T et al. (2013). Plasma oxytocin and vasopressin do not predict neuropeptide concentrations in human cerebrospinal fluid. J Neuroendocrinol 25: 668–673. Kalsbeek A, Fliers E, Hofman MA et al. (2010). Vasopressin and the output of the hypothalamic biological clock. J Neuroendocrinol 22: 362–372. Kenny GCT (1961). The “nervus conarii” of the monkey (an experimental study). J Neuropathol Exp Neurol 20: 563–570. Klein DC (2007). Arylalkylamine N-acetyltransferase: “the Timezyme”. J Biol Chem 282: 4233–4237. Klein D, Moore RY, Reppert SM (1991). Suprachiasmatic nucleus: the mind’s clock, Oxford University Press, New York. Knobloch HS, Grinevich V (2014). Evolution of oxytocin pathways in the brain of vertebrates. Front Behav Neurosci 14: 31. https://doi.org/10.3389/fnbeh.2014.00031. Korf HW, Wagner U (1980). Evidence for a nervous connection between the brain and the pineal organ in the guinea pig. Cell Tissue Res 209: 505–510. Laitinen JT, Laitinen KS, Kokkola T (1995). Cholinergic signaling in the rat pineal gland. Cholinergic signaling in the rat pineal gland. Cell Mol Neurobiol 15: 177–192. Larsen PJ, Enquist LW, Card JP (1998). Characterization of the multisynaptic neuronal control of the rat pineal gland using viral transneuronal tracing. Eur J Neurosci 10: 128–145. Le Gros Clark WE (1940). The nervous and vascular relation of the pineal gland. J Anat 74: 470–494. Lee MR, Sheskier MB, Farokhnia M et al. (2018). Oxytocin receptor mRNA expression in dorsolateral prefrontal cortex in major psychiatric disorders: a human post-mortem study. Psychoneuroendocrinology 96: 143–147. Li JD, Burton KJ, Zhang C et al. (2009). Vasopressin receptor V1a regulates circadian rhythms of locomotor activity and expression of clock-controlled genes in the suprachiasmatic nuclei. Am J Physiol Regul Integr Comp Physiol 296: R824–R830. Liu B, Burbach JPH (1987). Detection and HPLC characterization of summer rises of vasopressin- and oxytocinimmunoreactivity in the rat pineal gland. Endocrinology 121: 1716–1720. Loup F, Tribollet E, Dubois-Dauphin M et al. (1989). Localization of oxytocin binding sites in the human brainstem and upper spinal cord: an autoradiographic study. Brain Res 500: 223–230.

Loup F, Tribollet E, Dubois-Dauphin M et al. (1991). Localization of high affinity binding sites for oxytocin and vasopressin in the human brain: an autoradiographic study. Brain Res 555: 220–233. Lowrey PL, Takahashi JS (2011). Genetics of circadian rhythms in Mammalian model organisms. Adv Genet 74: 175–230. Lucassen PJ, Salehi A, Pool CW et al. (1994). Activation of vasopressin neurons in aging and in Alzheimer’s disease. J Neuroendocrinol 6: 673–679. Lucassen PJ, Van Heerikhuize JJ, Guldenaar SEF et al. (1997). No change in the total amounts of vasopressin mRNA in the supraoptic and paraventricular nucleus in aging and Alzheimer’s disease. J Neuroendocrinol 9: 297–305. Ludwig M, Leng G (2006). Dendritic peptide release and peptide-dependent behaviours. Nat Rev Neurosci 7: 126–136. Mai JK, Berger K, Sofroniew MV (1993). Morphometric evaluation of neurophysin-immunoreactivity in the human brain: pronounced inter-individual variability and evidence for altered staining patterns in schizophrenia. J Hirnforsch 34: 133–154. Makarenko IG, Ugryumov MV, Kalas A (2002). Involvement of accessory neurosecretory nuclei of hypothalamus in the formation of hypothalamohypohysial system during prenatal and postnatal development in rats. Russ J Dev Biol 33: 37–42. Matsushima S, Reiter RJ (1978). Electron microscopic observations on neuron-like cells in the ground squirrel pineal gland. J Neural Transm 42: 223–237. Matsuura T, Kawata M, Yamada H et al. (1983). Immunohistochemical studies on the peptidergic nerve fibers in the pineal organ of the dog. Arch Histol Jpn 46: 373–379. Meyer-Lindenberg A, Domes G, Kirsch P et al. (2011). Oxytocin and vasopressin in the human brain: social neuropeptides for translational medicine. Nat Rev Neurosci 12: 524–538. Meynen G, Unmehopa UA, van Heerikhuize JJ et al. (2006). Increased arginine vasopressin mRNA expression in the human hypothalamus in depression: a preliminary report. Biol Psychiatry 60: 892–895. Mikkelsen JD, Larsen PJ, Mick G et al. (1995). Gating of retinal inputs through the suprachiasmatic nucleus: role of excitatory neurotransmission. Neurochem Int 27: 263–272. Møller M (1976). The ultrastructure of the human fetal pineal gland. II. Innervation and cell junctions. Cell Tissue Res 169: 7–21. Møller M (1981). The ultrastructure of the deep pineal gland of the Mongolian gerbil and mouse: granular vesicle localization and innervation. In: CD Matthews, RF Seamark (Eds.), Pineal function. Elsevier, Amsterdam, pp. 257–266. Møller M, Baeres FMM (2002). The anatomy and innervation of the mammalian pineal gland. Cell Tissue Res 309: 139–150. Møller M, Baeres FMM (2003). PACAP-containing intrapineal nerve fibers originate predominantly in the trigeminal

VASOPRESSIN AND OXYTOCIN BEYOND THE PITUITARY IN THE HUMAN BRAIN ganglion: a combined retrograde tracing- and immunohistochemical study of the rat. Brain Res 984: 160–169. Møller M, Mikkelsen JD, Fahrenkrug J et al. (1985). The presence of vasoactive intestinal polypeptide (VIP)-like immunoreactive nerve fibres and VIP-receptors in the pineal gland of the Mongolian gerbil (Meriones unguiculatus). An immunohistochemical and receptor-autoradiographic study. Cell Tissue Res 241: 333–340. Møller M, Phansuwan Pujito P, Pramaulkijja S et al. (1994). Innervation of the cat pineal gland by neuropeptide Y-immunoreactive nerve fibers. An experimental immunohistochemical study. Cell Tissue Res 276: 545–550. Møller M, Fahrenkrug J, Hannibal J (1999). Innervation of the rat pineal gland by pituitary adenylate cyclase-activating polypeptide (PACAP)-immunoreactive nerve fibres. Cell Tissue Res 296: 247–257. Møller M, Busch JR, Jacobsen C et al. (2018). The accessory magnocellular neurosecretory system of the human hypothalamus. Cell Tissue Res 1: 487–498. Ni RJ, Shu YM, Wang J et al. (2014). Distribution of vasopressin, oxytocin and vasoactive intestinal polypeptide in the hypothalamus and extrahypothalamic regions of tree shrews. Neuroscience 265: 124–136. N€ urnberger F, Korf HW (1981). Oxytocin- and vasopressin immunoreactive nerve fibers in the pineal gland of the hedgehog, Erinaceus europaeus L. Cell Tissue Res 220: 87–97. Ostrowski NL (1998). Oxytocin receptor mRNA expression in rat brain: implications for behavioral integration and reproductive success. Psychoneuroendocrinology 23: 989–1004. Pavel S, Petrescu S (1966). Inhibition of gonadotrophin by a highly purified pineal peptide and by synthetic arginine vasotocin. Nature 212: 1054. Perlow MJ, Reppert SM, Artman HA et al. (1982). Oxytocin, vasopressin, and estrogen-stimulated neurophysin: daily patterns of concentration in cerebrospinal fluid. Science 216: 1416–1418. Peterson RP (1966). Magnocellular neurosecretory centers in the rat hypothalamus. J Comp Neurol 128: 181–190. Pevet P, Neacs¸ u C, Holder FC et al. (1981). The vasotocin-like biological activity present in the bovine pineal is due to a compound different from vasotocin. J Neural Transm 51: 295–302. Phansuwan-Pujito P, Møller M, Govitrapong P (1999). Cholinergic innervation and function in the mammalian pineal gland. Microsc Res Tech 46: 281–295. Preckel K, Scheele D, Kendrick KM et al. (2014). Oxytocin facilitates social approach behavior in women. Front Behav Neurosci 8: 191. https://doi.org/10.3389/fnbeh.2014.00191. Purba JS, Hofman MA, Swaab F (1994). Decreased number of oxytocin-immunoreactive neurons in the paraventricular nucleus of the hypothalamus in Parkinson’s disease. Neurology 44: 84–89. https://doi.org/10.1212/wnl.44.1.84. Raadsheer FC, Sluiter AA, Ravid R et al. (1993). Localization of corticotropin-releasing hormone (CRH) neurons in the paraventricular nucleus of the human hypothalamus; agedependent colocalization with vasopressin. Brain Res 615: 50–62.

23

Reppert SM, Weaver DR (2002). Coordination of circadian timing in mammals. Nature 418: 935–941. Rhodes CH, Morrell JI, Pfaff DW (1981). Immunohistochemical analysis of magnocellular elements in rat hypothalamus: distribution and numbers of cells containing neurophysin, oxytocin, and vasopressin. J Comp Neurol 198: 45–64. Rogers CN, Ross AP, Sahu SP et al. (2018). Oxytocin- and arginine vasopressin-containing fibers in the cortex of humans, chimpanzees, and rhesus macaques. Am J Primatol 80: e22875. https://doi.org/10.1002/ajp.22875. Romijn HJ (1973). Parasympathetic innervation of the rabbit pineal gland. Brain Res 55: 431–436. Rønnekleiv OK (1988). Distribution in the macaque pineal of nerve fibers containing immunoreactive substance P, vasopressin, oxytocin, and neurophysins. J Pineal Res 5: 259–271. Rønnekleiv OK, Møller M (1979). Brain-pineal nervous connection in the rat: an ultrastructure study following habenular lesion. Exp Brain Res 37: 551–562. Rønnekleiv OK, Kelly MJ, Wuttke W (1980). Single unit recordings in the rat pineal gland: evidence for habenulopineal neural connections. Exp Brain Res 39: 187–192. Rood BD, De Vries GJ (2011). Vasopressin innervation of the mouse (Mus musculus) brain and spinal cord. J Comp Neurol 519: 2434–2474. Rovsing L, Rath MF, Lund-Andersen C et al. (2010). Neuroanatomical and physiological study of the non-image forming visual system of the cone-rod homeobox gene (Crx) knock out mouse. Brain Res 1343: 54–65. Rovsing L, Rath MF, Møller M (2013). Hypothalamic neurosecretory and circadian vasopressinergic neuronal systems in the blind cone-rod homeobox knock out mouse (Crx /) and the 129sv wild type mouse. J Comp Neurol 521: 4061–4074. Saper CB (2012). Hypothalamus. In: JK Mai, G Paxinos (Eds.), The human nervous system, second edn. Elsevier, Amsterdam, pp. 548–583. Schneider T, Semm P, Vollrath L (1981). Ultrastructural observations on the central innervation of the guinea-pig pineal gland. Cell Tissue Res 220: 41–49. Schomerus C, Laedtke E, Korf HW (1995). Calcium responses of isolated, immunocytochemically identified rat pinealocytes to noradrenergic, cholinergic and vasopressinergic stimulations. Neurochem Int 27: 163–175. Schwartz WJ, Reppert SM (1985). Neural regulation of the circadian vasopressin rhythm in cerebrospinal fluid: a preeminent role for the suprachiasmatic nuclei. J Neurosci 5: 2771–2778. Semm P, Schneider T, Vollrath L (1981). Morphological and electrophysiological evidence for habenular influence on the guinea-pig pineal gland. J Neural Transm 50: 247–266. Shapiro LE, Insel TR (1989). Ontogeny of oxytocin receptors in rat forebrain: a quantitative study. Synapse 4: 259–266. Shiotani Y, Yamano M, Shiosaka S et al. (1986). Distribution and origins of substance P (SP)-, calcitonin gene-related peptide (CGRP)-, vasoactive intestinal polypeptide (VIP)- and neuropeptide Y (NPY)-containing nerve fibers in the pineal gland of gerbils. Neurosci Lett 70: 187–192.

24

M. MØLLER

Simonneaux V, Ouichou A, Burbach JP et al. (1990). Vasopressin and oxytocin modulation of melatonin secretion from rat pineal glands. Peptides 11: 1075–1079. Sofroniew MV (1980). Projections from vasopressin, oxytocin, and neurophysin neurons to neural targets in the rat and human. J Histochem Cytochem 28: 475–478. Sofroniew MV (1983). Vasopressin and oxytocin in the mammalian brain and spinal cord. Trends Neurosci 6: 467–472. Sofroniew MV (1985). Vasopressin- and neurophysinimmunoreactive neurons in the septal region, medial amygdala and locus coeruleus in colchicine-treated rats. Neuroscience 15: 347–358. Sofroniew MV, Glasmann W (1981). Golgi-like immunoperoxidase staining of hypothalamic magnocellular neurons that contain vasopressin, oxytocin or neurophysin in the rat. Neuroscience 6: 619–643. Sofroniew MV, Weindl A, Schrell U et al. (1981). Immunohistochemistry of vasopressin, oxytocin and neurophysin in the hypothalamus and extrahypothalamic regions of the human and primate brain. Acta Histochem Suppl 24: 79–95. Stoop R (2012). Neuromodulation by oxytocin and vasopressin. Neuron 76: 142–159. https://doi.org/10.1016/j. neuron.2012.09.025. Stopa EG, King JC, Lydic R et al. (1984). Human brain contains vasopressin and vasoactive intestinal polypeptide neuronal subpopulations in the suprachiasmatic region. Brain Res 297: 159–163. Summar ML, Phillips JA, Battey J (1990). Linkage relationships of human arginine vasopressin-neurophysin-II and oxytocin-neurophysin-I to prodynorphin and other loci on chromosome 20. Mol Endocrinol 4: 947–950. Swaab DF (1997). Prader-Willi syndrome and the hypothalamus. Acta Paediatr Suppl 423: 50–54. Swaab DF (2003). Handbook of clinical neurology, vol. 79: Elsevier, 163–164. Swaab DF, Pool CW (1975). Specificity of oxytocin and vasopressin immunofluorescence. J Endocrinol 66: 263–272. Swaab DF, Pool CW, Nijveldt F (1975). Immunofluorescence of vasopressin and oxytocin in the rat hypothalamo-neurohypophypopseal system. J Neural Transm 36: 195–215. Swaab DF, Fliers E, Partiman TS (1985). The suprachiasmatic nucleus of the human brain in relation to sex, age and senile dementia. Brain Res 342: 37–44. Swaab DF, Roozendaal B, Ravid R et al. (1987). Suprachiasmatic nucleus in aging, Alzheimer’s disease, transsexuality and Prader–Willi syndrome. Prog Brain Res 72: 301–310. Swanson LW (1977). Immunohistochemical evidence for a neurophysin-containing autonomic pathway arising in the

paraventricular nucleus of the hypothalamus. Brain Res 128: 346–353. Swanson LW, McKellar S (1979). The distribution of oxytocin- and neurophysin-stained fibers in the spinal cord of the rat and monkey. J Comp Neurol 188: 87–106. Swanson LW, Sawchenko PE (1983). Hypothalamic integration: organization of the paraventricular and supraoptic nuclei. Annu Rev Neurosci 6: 269–324. Szot P, Bale TL, Dorsa DM (1994). Distribution of messenger RNA for the vasopressin V1a receptor in the CNS of male and female rats. Brain Res Mol Brain Res 24: 1–10. Teclemariam-Mesbah R, Ter Horst GJ, Postema F et al. (1999). Anatomical demonstration of the suprachiasmatic nucleus-pineal pathway. J Comp Neurol 406: 171–182. van Wamelen DJ, Aziz NA, Anink JJ et al. (2012). Paraventricular nucleus neuropeptide expression in Huntington’s disease patients. Brain Pathol 2: 654–661. van Zwieten EJ, Ravid R, Swaab DF (1996). Differential vasopressin and oxytocin innervation of the human parabrachial nucleus: no changes in Alzheimer’s disease. Brain Res 711: 146–152. Vandesande F, Dierickx K, DeMey I (1977). The origin of the vasopressinergic and oxytocinergic fibres of the external region of the median eminence of the rat hypophysis. Cell Tissue Res 80: 443–452. Voorn P, Buijs RM (1983). An immuno-electronmicroscopical study comparing vasopressin, oxytocin, substance P and enkephalin containing nerve terminals in the nucleus of the solitary tract of the rat. Brain Res 270: 169–173. Vrang N, Larsen PJ, Møller M et al. (1995). Topographical organization of the rat suprachiasmatic-paraventricular projection. J Comp Neurol 353: 585–603. Wilczynski W, Quispe M, Matı´as I et al. (2017). Arginine vasotocin, the social neuropeptide of amphibians and reptiles. Front Endocrinol 8: 1–17. Young WS, Li J, Wersinger SR et al. (2006). The vasopressin 1b receptor is prominent in the hippocampal area CA2 where it is unaffected by restraint stress or adrenalectomy. Neuroscience 143: 1031–1039. Zak PJ, Kurzban R, Matzner WT (2005). Oxytocin is associated with human trustworthiness. Horm Behav 48: 522–527. Zhang E, Mikkelsen JD, Møller M (1991). Tyrosine hydroxylase and neuropeptide Y-immunoreactive nerve fibers in the pineal complex of untreated rats and rats following removal of the superior cervical ganglia. Cell Tissue Res 265: 63–71. Zhou JN, Riemersma RF, Unmehopa UA et al. (2001). Alterations in arginine vasopressin neurons in the suprachiasmatic nucleus in depression. Arch Gen Psychiatry 587: 655–662.

Handbook of Clinical Neurology, Vol. 180 (3rd series) The Human Hypothalamus: Middle and Posterior Region D.F. Swaab, F. Kreier, P.J. Lucassen, A. Salehi, and R.M. Buijs, Editors https://doi.org/10.1016/B978-0-12-820107-7.00003-3 Copyright © 2021 Elsevier B.V. All rights reserved

Chapter 3

Central and peripheral release of oxytocin: Relevance of neuroendocrine and neurotransmitter actions for physiology and behavior FERDINAND ALTHAMMER1, MARINA ELIAVA2, AND VALERY GRINEVICH2* 1

Neuroscience Department, Center for Neuroinflammation and Cardiometabolic Diseases, Georgia State University, Atlanta, GA, United States 2

Department of Neuropeptide Research in Psychiatry, Central Institute of Mental Health, Medical Faculty Mannheim, University of Heidelberg, Mannheim, Germany

Abstract The hypothalamic neuropeptide oxytocin (OT) is critically involved in the modulation of socio-emotional behavior, sexual competence, and pain perception and anticipation. While intracellular signaling of OT and its receptor (OTR), as well as the functional connectivity of hypothalamic and extra-hypothalamic OT projections, have been recently explored, it remains elusive how one single molecule has pleotropic effects from cell proliferation all the way to modulation of complex cognitive processes. Moreover, there are astonishing species-dependent differences in the way OT regulates various sensory modalities such as touch, olfaction, and vision, which can be explained by differences in OTR expression in brain regions processing sensory information. Recent research highlights a small subpopulation of OT-synthesizing cells, namely, parvocellular cells, which merely constitute 1% of the total number of OT cells but act as “master cells’ that regulate the activity of the entire OT system. In this chapter, we summarize the latest advances in the field of OT research with a particular focus on differences between rodents, monkeys and humans and highlight the main differences between OT and its “sister” peptide arginine–vasopressin, which often exerts opposite effects on physiology and behavior.

INTRODUCTION Oxytocin (OT)—“one of the molecules of the decade” (Grinevich et al., 2016a)—nowadays attracts the attention of both the neuroscience community and the general public. This interest is especially due to its prominent prosocial effects in various vertebrate species, including humans. Well before the synthesis of OT by Du Vigneaud et al. (1953), the German anatomist Ernst Scharrer discovered “glandule-like” giant cells in the hypothalamus of teleost fish. These “magnocellular neurons” were later shown to produce OT and its homologues (as well as arginine–

vasopressin, AVP, and its homologues), and they have been found in representative species of all classes of vertebrates (Knobloch and Grinevich, 2014; Grinevich et al., 2016a). As a neuroendocrine factor, OT is transported to the posterior pituitary lobe from where it is released into the blood as a neurohormone to act on peripheral targets to control sodium reuptake in the kidney, modulate autonomic functions and metabolism, nociception, and immune response (Verbalis et al., 1991; Yang et al., 2013; Carter, 2014). During pregnancy and lactation, OT is important for the delivery of newborns in most mammals, eggs in fish,

*Correspondence to: Prof. Dr. Valery Grinevich, M.D, Ph.D, Chair, Department of Neuropeptide Research in Psychiatry, Central Institute of Mental Health, Medical Faculty Mannheim, University of Heidelberg, J5, Mannheim 68159, Germany. Tel: +49-621-1703-2995, Fax: +49-0621-1703-1325, E-mail: [email protected]

26

F. ALTHAMMER ET AL.

amphibians, reptiles, and birds, and it is critical for milk ejection (Moore, 1992; Knobloch and Grinevich, 2014; Leng et al., 2015). In mammals, magnocellular OT (magnOT) cells are large neuroendocrine cells with a somatic diameter of 25–30 mm (Armstrong, 2004). MagnOT can be found in the supraoptic (SON), paraventricular (PVN), and accessory (AN) nuclei of reptilian, avian, and mammalian hypothalamus (Grinevich and Polenov, 1994; Knobloch and Grinevich, 2014; Grinevich et al., 2016a; Grinevich and Stoop, 2018b). In the rat (Fig. 3.1A), one PVN harbors roughly 1200 OT-positive cells, amounting to 2400 cells for both PVN in total ((Rhodes et al., 1981). The SON (Fig. 3.1B) contains 1600 OT cells per side, resulting in a total of 3200 cells. Thus rat PVN and SON together contain 5600 OT cells. According to the same authors

(Rhodes et al., 1981), 1000 OTcells reside in the accessory groups of the anterior hypothalamic area and
700 OT cells reside in the anterior commissural nucleus. Therefore, the total number of OT neurons in the rat hypothalamus is 7600 cells (indicated here as “about 8000 cells”), which is relatively similar to the number of AVP neurons (Rhodes et al., 1981).

ORGANIZATION OF THE HYPOTHALAMIC– NEUROHYPOPHYSEAL (HHNS) SYSTEM Despite the similarity of the chemical structure of OT and its homologues in evolution, the pathways of OT delivery through the brain were drastically transformed over time: while in basal vertebrates (fish and amphibians)

Fig. 3.1. Fluorescent labeling of OT and AVP neurons via cell type-specific viral vectors. Images show Venus (green) and TdTomato (red) expression in OT and AVP neurons, respectively, after injection of recombinant adeno-associated viruses (rAAVs), equipped with the OT and AVP promoters, into the PVN (A) and SON (B) of adult female rat. The OT and AVP neurons were costained with antibodies against these two peptides (blue), resulting in the overlay in purple. Modified from Grinevich V, Knobloch-Bollmann HS, Roth LC et al. (2016b). Somatic transgenesis (viral vectors). John Wiley & Sons.

CENTRAL AND PERIPHERAL RELEASE OF OXYTOCIN OT homologues are preferentially released into the cerebrospinal fluid, in advanced vertebrates, the neuropeptide is released from elaborated dendrites of OT neurons. Within the hypothalamic SON and PVN, it was calculated that local concentrations of OT are about 100–1000-fold higher than in the blood plasma (Landgraf and Neumann, 2004). This together with the fact that peripherally circulating endogenous OT seems not to cross the blood–brain barrier in behaviorally relevant amounts (Kagerbauer et al., 2013; Leng and Ludwig, 2016) makes the local somatodendritic OT/AVP release from PVN and SON neurons (Fig. 3.2) an important factor for the modulation of neighboring hypothalamic nuclei, which coordinate socioemotional behaviors. Pow and Morris were the first to provide electron microscopic evidence for such dendritic release of OT from magnocellular neurons within the SON (Morris and Pow, 1991). Physiologic, stressful, and pharmacologic stimuli of somatodendritic OT release have been extensively studied and discussed (for review, see Engelmann et al., 2004; Neumann, 2007, 2008; Veenema and Neumann, 2008; Jurek and Neumann, 2018). So far, all reproductive stimuli, such as birth, suckling in the lactating animal, and mating in males and females, which were all shown to activate OT secretion as a neurohormone into the blood stream, were also found to trigger OT release within the brain in a region-dependent manner. Thus in response to the reproductive stimuli mentioned previously, increased OT release has been shown in the PVN, SON, septum, dorsal hippocampus, bed nucleus of the stria terminalis (BNST), olfactory bulb, nucleus

27

accumbens, and medial preoptic area (Moos et al., 1989; Neumann and Landgraf, 1989; Kendrick et al., 1992; Neumann et al., 1993a; Waldherr and Neumann, 2007; Ross et al., 2009; Nyuyki et al., 2011), for review see Landgraf and Neumann (2004)). Also, various physiologic (Landgraf et al., 1988; Ludwig et al., 1994), physical, and emotional (Neumann et al., 1993b; Engelmann et al., 2004) stressors stimulate not only OT secretion into blood but also its intracerebral release. Relevant emotional stimuli include forced swimming (the PVN, SON, amygdala) (Wotjak et al., 1996, 1998; Wigger and Neumann, 2002; Ebner et al., 2005; Torner et al., 2016), and shaker stress (the PVN) (Nishioka et al., 1998). Also, OT release is stimulated in male rats by exposure to aggressive male resident in social defeat paradigm (the SON, septum, but not the PVN) (Engelmann et al., 1999; Ebner et al., 2000) and in virgin female rats in the presence of aggressive lactating female resident (maternal defeat) (in the PVN, but not in amygdala or septum) (Bosch et al., 2004). The latter examples provide striking evidence for region-dependent release of OT, which, notably, occurs independent of peripheral OT secretion. Along with the local somatodendritic release, OT neurons of mammals acquired long-range axonal projections to forebrain, brainstem, and spinal cord regions to release OT distantly (so-called axonal release) (Knobloch and Grinevich, 2014) (Fig. 3.2). This evolutionary determined phenomenon provides a new conceptual insight into the organization of the central OT system. Indeed, addressed axonal OT release allows to modulate various

A somato-dendritic B en passant release

release

C axonal release oxytocin

oxytocin receptor INTERNEURON

PYRAMIDAL CELL

Fig. 3.2. Different types of release from magnocellular OT and AVP neurons. Magnocellular neurons release OT and AVP from somas and dendrites (A), from axons passing by (en passant, B) and from long-range axons (C). OTRs have been found in various neuronal cell types, including GABAergic interneurons (Huber et al., 2005) and pyramidal cells (Lin et al., 2017). The precise preand postsynaptic mechanisms, as well as the location of OTRs remain elusive.

28

F. ALTHAMMER ET AL.

behaviors in a brain region-specific manner. In line with the described long-range axonal projections of OT neurons that innervate various forebrain structures, OT receptors (OTRs) have been described in regions innervated by OT axons (Grinevich et al., 2016a) (Fig. 3.3). For the most part, OTR expression patterns and axonal innervation overlap (Mitre et al., 2016; Marlin and Froemke, 2017), proving the functionally relevant release–effect relationship in those brain regions. In line with these findings, direct and/or neuromodulatory effects of locally released OT have been described for the central amygdala (Hasan et al., 2019), hippocampus (Tirko et al., 2018), lateral septum (Menon et al., 2018), prefrontal cortex (Sabihi et al., 2014), and anterior olfactory nucleus (Oettl et al., 2016). While OT neurons in the mammalian brain are almost exclusively located in the PVN, SON, and accessory nuclei of the hypothalamus, AVP neurons are found in the suprachiasmatic nucleus (SCN) and various extrahypothalamic forebrain nuclei, including the BNST and the medial nucleus of amygdala (De Vries et al., 1984b) (Fig. 3.3). Interestingly, these discrete AVPsynthesizing nuclei respond differently to external stimuli and fulfill a number of important physiological functions including modulation of circadian rhythmicity in retinal ganglion cells (Tsuji et al., 2017; Wacker and Ludwig, 2019), olfactory processing in the olfactory bulb and anterior olfactory nucleus (Tobin et al., 2010), integration of olfactory information in the piriform cortex

(Dumais and Veenema, 2016), and modulation of emotional and physiological stress in the locus coeruleus (Hernandez-Perez et al., 2019). The interaction between the gonadal hormones estrogen and testosterone with AVP neurons of the lateral septum has been described for social recognition (Gabor et al., 2012), where testosterone and its metabolites fine tune various behaviors such as social bonds and hierarchies. In addition, it was demonstrated that different stages of the estrous cycle change the AVP-dependent regulation of cardiac output, blood pressure, and hypertension (Share and Crofton, 1993), but the respective AVP-synthesizing nuclei have not been specifically addressed. It is well established that AVP neuron activity and AVP synthesis is regulated by testosterone (de Vries et al., 1984a; Szot and Dorsa, 1993; Wang et al., 1993), and male rat brains display higher expression of AVP than female brains (Mayes et al., 1988; De Vries et al., 2002). In males, the lateral septum receives projections from brain areas containing testosterone-dependent AVP neurons in rats, such as the BNST and the medial amygdala, which directly influence social recognition (Everts et al., 1997). In addition, castration-reduced AVP expression in several limbic brain areas and this effect could be reversed by application of testosterone (Zhou et al., 1994). It is important to note that while the role of OT in the facilitation of birth is generally accepted, the respective role of AVP is underappreciated. In fact, OTR or OT-deleted mutants are still capable of SSC MC

A

C

OB

LS NAcc AON

AC

Tu

LC SCN

AN PVN SON

SON

CB

HC

BNST vDB

ICj

AN PVN

VC

PV

PFC

CeAMeA Arc BLA

BS SC

PP

RGC

B

OB

oxytocin-expressing neurons and their projections LS PC BNST AON

vDB

SCN AN

LC

PVN

SON

MeA

vasopressin-expressing neurons an their projections overlap of oxytocin and vasopressin projections

Fig. 3.3. Overview of OT- and AVP-synthesizing nuclei as well as their distinct and overlapping axonal projection sites in the rat brain. Brain schemes highlight the location of OT-synthesizing nuclei (A), AVP-synthesizing nuclei (B), as well as their projection sites (C). AN, accessory nuclei; AON, anterior olfactory nucleus; Arc, arcuate hypothalamic nucleus; AC, auditory cortex; BNST, bed nucleus of stria terminalis; BLA, basolateral amygdala; BS, Brainstem; CeA, central amygdala; CB, cerebellum; HC, hippocampus; iCj, Island of Calleja; LH, lateral hypothalamic area; LS, lateral septum; LC, locus coeruleus; MeA, medial amygdala; MC, motor cortex; NAcc, nucleus accumbens; OB, olfactory bulb; Tu, olfactory tubercele; PVN, paraventricular nucleus of the hypothalamus; PV, paraventricular thalamus; PC, piriform cortex; PFC, prefrontal cortex; RGC, retina ganglion cells; SSC, somatosensory cortex; SC, spinal cord; SCN, Suprachiasmatic nucleus; SON, supraoptic nucleus; vDB, Ventral diagonal band of Broca; VS, ventral subiculum; VC, visual cortex.

CENTRAL AND PERIPHERAL RELEASE OF OXYTOCIN giving birth and blocking of OTRs does not prevent a premature birth (Nishimori et al., 1996; Young et al., 1996), suggesting that OT actions at the OTR are not solely responsible for labor induction. In fact, a synergistic activation of both the OTR and V1aR seem to be required for proper induction of labor (Akerlund, 2006; Lee et al., 2008). A comprehensive overview about the effects of AVP on uterine contractions, birth, aggression, social learning, and parental and reproductive behavior can be found in Carter (2017). In a transgenic rat line, expressing AVP-GFP-fused protein under the control of the AVP promoter (Ueta et al., 2005), GFP-expressing neurons were found in the olfactory bulb, where AVP was shown to modulate the processing of olfactory social signals (Tobin et al., 2010). Very recently, the same group led by Ludwig also showed that a small fraction of ganglionic cells in the retina expresses AVP and through its projections to the SCN modulates circadian rhythmicity (Tsuji et al., 2017). The low density and sparse innervation of axonal AVP projections in many brain regions make it technically difficult to dissect the origin of respective axons. A study by W. Scott Young III and colleagues showed that magnocellular AVP neurons of the PVN project to CA2 of the dorsal hippocampus (Smith et al., 2016). Moreover, an elegant study combining extracellular recording of CA2, juxtacellular labeling, post hoc anatomo-immunohistochemical analysis, and camera lucida reconstruction demonstrated various extrahypothalamic AVP projections to the preoptic area, SCN, lateral habenula, amygdala, and other brain regions (Hernandez et al., 2015). Importantly, the projections of OT and AVP neurons largely overlap (Fig. 3.3), suggesting possibility of release of both neuropeptides in the same brain structures. The question whether the release is simultaneous and recruits the same type of target cells requires further investigations (Stoop, 2012; Dumais and Veenema, 2016; Grinevich and Stoop, 2018a).

PARVOCELLULAR OXYTOCIN NEURONS OF THE HYPOTHALAMUS: MASTER CELLS ARRANGING THE HHNS ACTIVITY? ParvOT cells are smaller neurons with a diameter of somas of 10–20 mm and are located in selective subdivisions of the caudal PVN (Swanson and Kuypers, 1980; Swanson and Sawchenko, 1983). It is well established that parvOT neurons send axons to the brainstem and spinal cord (Fig. 3.4), where they modulate a variety of different vital and physiologic functions. For example, some parvOT neurons project to the brainstem, exerting food intake regulation (Blevins et al., 2004), as well as other autonomic functions, such as

29

? ParvOT Midbrain

MagnOT

Brainstem Spinal cord Blood

Fig. 3.4. Projections from ParvOT and MagnOT neurons in the CNS. Brain scheme depicts the currently known magnOT (red) and parvOT (green) neuron interconnectivity within the PVN and SON and their distinct projections to the pituitary, forebrain, midbrain, brainstem, and spinal cord. The green dashed line and the question mark highlight potential, but not yet confirmed parvOT projections to the forebrain. Modified from Althammer F, Grinevich V (2017). Diversity of oxytocin neurons: beyond magno- and parvocellular cell types? J Neuroendocrinol 30.

breathing (Mack et al., 2002), erection and copulation (Melis et al., 1986), cardiovascular reactions (Petersson, 2002), gastric reflexes (Sabatier et al., 2013), and pain perception (Rash et al., 2014). All of these projections arise from a small group of parvOT neurons residing within several subnuclei in the PVN, some of which give rise to long-range axonal projections to the dorsal vagal complex and the nucleus of solitary tract, where they respond to cholecystokinin and glucagon-like peptide-1mediated satiety signals to affect feeding behavior (Valassi et al., 2008; Motojima et al., 2016) and the spinal cord (Eliava et al., 2016b), where parvOT neurons coordinate somatosensory processing. ParvOT neurons have been implicated to orchestrate and control the activity of magnOT cells (Eliava et al., 2016b) via PVN➔SON connections. The OTR is being expressed in many different cells throughout the brain and, in most cases, oxytocin receptor distribution could be matched with OT-ergic projections in rodents and primates in the periaqueductal gray (PAG) (Juif et al., 2013; Juif and Poisbeau, 2013), hypoglossal nucleus of the medulla (Wrobel et al., 2010), brainstem (the nucleus tractus solitarii (NTS) and ventrolateral medulla (VLM)) ((Blevins et al., 2004, Meddle et al., 2007), NTS (Ho et al., 2014), and spinal cord (Tribollet et al., 1997; Eliava et al., 2016b). However, it is not clear whether these parvOT projections are terminating exclusively in the respective brain regions or if they are simultaneously projecting to other brain regions by forming collaterals. The first study demonstrating that parvOT neurons of the PVN can give rise to axonal projections, which target two distinct brain regions, was conducted by Xiao et al. (2017), who showed that two separate subpopulations of parvOT terminate in the ventral tegmental area or in the substantia nigra pars compacta.

30

F. ALTHAMMER ET AL.

In humans, however, the precise number of parvOT neurons and their targets remains unknown, both due to the lack of feasible postmortem tools to label parvOT neurons and the existence of intermixed populations of larger and smaller neurons within the PVN (Wierda et al., 1991), which makes it difficult to discriminate between the two cell types. As for AVP, the intricate interaction between neurosecretory networks and preautonomic neurons has been demonstrated in the PVN. The authors could convincingly show that activity-dependent AVP release from hypothalamic neuroendocrine neurons stimulated neighboring presympathetic neurons (within the range of 100 mm), thereby mediating interpopulation crosstalk (Son et al., 2013). The described mechanism plays a pivotal role in the vasopressin-dependent polymodal neurohumoral response to hyperosmotic challenge. Similar intricate interactions can be observed among OT neurons, namely, parvOT neurons, synaptically innervating magnOT neurons in the SON (Eliava et al., 2016b). In addition, we recently described the putative innervation of magnOT neurons by parvOT neurons in the PVN (Tang et al., 2020). This finding was further corroborated with pseudotyped rabies virus-mediated retrograde labeling, quantification, and three-dimensional reconstruction of parvOT inputs. Our data provide strong evidence that parvOT neurons receive more overall synaptic input, than magnOT neurons, and receive input from distinct brain regions (PVT, Ins, Hb), which do not innervate magnOT neurons (Fig. 3.5). In addition, our study convincingly demonstrated that parvOT neurons act as “first responders” (or “master” cells), conveying somatosensory signal onto the much larger population of magnOT neurons, thereby triggering a global response of the OT system, which ultimately leads to the promotion of social motivation (Tang et al., 2020).

HETEROGENEITY OF OXYTOCIN NEURONS Although numerous studies were focused on connections of OT neurons with the midbrain, brainstem, or spinal cord structures, the types of OT neurons projecting to these regions often have been not precisely classified (Peters et al., 2008; Rosen et al., 2008; Blouet et al., 2009; Yoshida et al., 2009; Shahrokh et al., 2010; Condes-Lara et al., 2012; Freeman et al., 2014). Despite the fact that the origin of the projections described are putative parvOT neurons (Peters et al., 2008, Rosen et al., 2008, Blouet et al., 2009, Yoshida et al., 2009, Shahrokh et al., 2010, Condes-Lara et al., 2012, Freeman et al., 2014), the exact characterization of the cell types in congruency with their functional relevance has been demonstrated in a rather limited number of

studies (Mack et al., 2002; Petersson, 2002; Blevins et al., 2004; Eliava et al., 2016b). In this respect, a most recent, elegant study applying two-color Retrobeads™ (Xiao et al., 2017) demonstrated two distinct populations of Fluorogold-negative parvOT neurons projecting to the ventral tegmental area and substantia nigra, where they differently modulate dopamine neuron activity. Most surprisingly, a novel study (Hung et al., 2017) reported that a subset of magnOT (e.g., Fluorogold-positive) neurons of the caudal PVN project their axons to the hindbrain (the ventral tegmental area of the midbrain), once again challenging the current and likely oversimplified view on parvOT and magnOT neuron systems and opening perspectives for dissection of functional relevance of newly described circuits. In line with the anatomical divergence studies, Romanov et al. (2017) recently reported four (!) distinct types of OT neurons, based on the expression of several genetic markers. Indeed, analyzing a total of 1194 genes via t-student stochastic neighbor embedding algorithmic sorting, Romanov and colleagues identified four clusters of OT neurons displaying distinct genetic patterns. However, the individual genes separating these clusters have not been shown yet. A full description of the various methods available for the discrimination between parvOT and magnOT neurons, as well as a discussion on types of OT cells, can be found in Althammer and Grinevich (2017).

SYNAPTIC VS NONSYNAPTIC OXYTOCIN RELEASE Although controversial discussion still remains (Buijs, 1983), the central release of OT may potentially occur in a synaptic or nonsynaptic fashion, probably, as a combination of presynaptic terminal, axons en passant, dendritic, and somatic release (Grinevich and Neumann, 2020). These different modes of release contribute to the enormous functional complexity of the brain OT system. Although parvOT neurons projecting to the hindbrain synapse onto target cells (Buijs, 1983; Swanson and Sawchenko, 1983; Voorn and Buijs, 1983), the nature of axonal contacts with OT-sensitive neurons in the forebrain remains elusive despite electron microscopic reports showing the presence of large dense-core vesicles containing OTin synaptic terminals within the SON, ventromedial hypothalamus, lateral septum, amygdala, and nucleus of the solitary tract (Buijs and Swaab, 1979; Voorn and Buijs, 1983; Theodosis, 1985; Peters et al., 2008; Griffin et al., 2010; Knobloch et al., 2012). Further, in vivo evidence for extracellular Ca2+-dependent synaptic release of the nonapeptides within the septum, SON, and PVN demonstrates that depolarizing stimuli, or omission of Ca2+, can stimulate, or prevent, such local release (Buijs and Van Heerikhuize, 1982; Neumann and Landgraf,

A

* OT

Reconstruction

Zoom

Merge Sphere

1,5x10 6

*** 1x10 6 5x105 0 1 107

*

*

8 106

*

OT

GFP

Reconstruction

SYN+ voxels

*

SYN+ voxels (adjusted)

*

Merge Spheres

SFO

C

2x10 6

B

0-1 % of inputs to ParvOT 1-2 2-4 4-7 CgC 7-10 PLC OC SFO >10

parvOT inputs

6 106 4 106

*** ***

2 106 0

PVT

5 20 5 20 Distance from soma center (µm)

Hb

ILC SEP PIL

Ins NAC

PAG

OVLT

0-1 % of inputs to MagnOT 1-2 2-4 4-7 CgC 7-10 PLC OC SFO >10

D

DRN

ZI PVN DMH PH ParvOT HDB MMB MPO ARC LH AMY VS

LMN

NST

MRN

PBN PPT RMg

magnOT inputs

1500 1000

E

SEP

500

ZI PVN DMH PH MagnOT HDB MMB MPO ARC LH AMY VS

hypothalamic inputs extra-hypothalamic inputs

PAG

DTT

ParvOT

DRN LMN

NST

OVLT

2000

ParvOT MagnOT

PIL NAC

2500

0

ILC Cl

*

3000

Total number of inputs

Cl DTT

SN

MRN

MagnOT

PBN PPT

673

263

RMg

1291

431

Fig. 3.5. Three-dimensional reconstruction of connectivity between ParvOT and MagnOT neurons in rat PVN and neuronal inputs to these cell types. (A) Panels show the quantification of synaptophysin (SYN) fluorescence at somatic (top) and dendritic (bottom) locations. OT neurons have been visualized using an anti-OT antibody, while parvOT neurons have been labeled using a retrograde, virus-based approach (Eliava et al., 2016a,b). To quantify the amount of SYN fluorescence, a sphere was placed at the respective location and the IMARIS software was used to calculate the number of SYN-positive voxels within the sphere. (B) Bar graphs show the quantification of the obtained data at somatic (top) and dendritic locations (bottom, 5 and 20 mm from soma) of parvOT and magnOT neurons, respectively. (C) Schema representing the proportion of inputs (number of inputs from one brain area/total number of inputs) from each brain area to parvOT and magnOT neurons, assessed by pseudorabies virus retrograde tracing, respectively. Brain areas projecting only to parvOT or magnOT are circled in green or purple, respectively. (D) Quantification of the total number of inputs to parvOT and magnOT neurons (t-test, P < 0.05. (E) Proportion of inputs to parvOT and magnOT neurons located in the hypothalamus or outside the hypothalamus. Numbers indicate average number of neurons. AMY, amygdala; Arc, arcuate hypothalamic nucleus; NST, nucleus of stria terminalis; CgC, cingulate cortex; Cl, claustrum; DRN, dorsal raphe nucleus; DTT, dorsal tenia tecta; DMH, dorsomedial hypothalamic area; Hb, habenular nucleus; HDB, horizontal limb of diagonal band nucleus; ILC, infralimbic cortex; Ins, insular cortex; LH, lateral hypothalamic area; LMN, lateral lemniscus nucleus; SEP, Lateral septum: MMB, mamillary body; MPO, medial preoptic area; MRN, median raphe nucleus; MC, motor cortex; NAc, nucleus accumbens; OC, orbital cortex; PBN, parabrachial nucleus; PVN, paraventricular nucleus of the hypothalamus; PVT, paraventricular thalamus; PPT, pedunculopontine tegmental nucleus; PAG, periaqueductal gray area; PH, posterior hypothalamic nucleus; PIL, posterior intralaminar thalamus; PLC, prelimbic cortex; RMg, raphe magnus nucleus; SFO, subfornical organ; SN, substancia nigra; OVLT, vascular organ of lamina terminalis; VS, ventral subiculum; ZI, zona incerta. Modified with permission from Tang Y, Benusiglio D, Lefevre A et al. (2020). Social touch promotes inter-female communication via oxytocin parvocellular neurons. Nat Neurosci 23: 1125–1137 (in press).

32

F. ALTHAMMER ET AL.

1989; Neumann et al., 1993b). In contrast, so far, large dense-core OT vesicles have not been located in the active zones of presynapses (i.e., near presynaptic membrane) in the few OT-containing synapses found in the SON (Theodosis, 1985; Wotjak et al., 1996) and at the border of the ventromedial hypothalamic nucleus (Griffin et al., 2010). Moreover, the precise localization of OTRs at postsynaptic sites has not been shown yet. Together with electrophysiologic evidence speaking against synaptic neurotransmission of nonapeptides (Son et al., 2013), it is likely that microscopically visualized synapses (Theodosis, 1985; Knobloch et al., 2012; Eliava et al., 2016a) operate by classic neurotransmitters rather than OT. In fact, already in the 1980s it was suggested that OT might not be axonally released from OT fibers that project from the SON and PVN to the neurohypophysis (Buijs and Van Heerikhuize, 1982). Therefore, intracerebral release of OT from magnOT neurons should occur mainly nonsynaptically, either from axonal collaterals en passant or axonal terminals within the forebrain and other limbic regions, as well as from dendrites and somas within the PVN and SON (Fig. 3.2). When released nonsynaptically, OT molecules diffuse into the surrounding space and act as neuromodulators rather than classic neurotransmitters by binding to nearby OTRs in virtually all major forebrain brain regions (Landgraf and Neumann, 2004; Leng and Ludwig, 2008; Neumann and Landgraf, 2012; Mitre et al., 2016) (Fig. 3.3). Recent calculations on the effective OT concentrations around the site of OT release have revealed a radius of about 55–120 mM (Chini et al., 2017); beyond this radius, OT concentrations are not sufficient to activate local OTRs. This excludes the possibility that diffusion of OT to neighboring or even further distant brain regions significantly contributes to its neuronal or behavioral actions. However, given the broad expression of OTRs (at least in the rodent brain) in various brain regions, it seems plausible that OT has a general, neuromodulatory function in these regions and that some of the sparse OT fibers cannot be reliably detected without more sensitive methods. Both parvOT and magnOT neurons coexpress glutamate as a conventional neurotransmitter. However, the balance between expression of OT and glutamate in the same neuron remains unclear. Although parallel elevation of vesicular glutamate transporter vGlutT2 and OT mRNAs levels has been reported in magnOT cell bodies after osmotic challenge (Kawasaki et al., 2006) and the presence of vGluT2 immunosignal was detected in axonal terminals of the posterior lobe of the pituitary (Hrabovszky et al., 2007), to our knowledge only two functional studies tackled the question of axonal OT and glutamate corelease (Knobloch et al., 2012; Hasan et al., 2019). The first work (Knobloch et al., 2012)

reported the presence of putative (asymmetric) glutamatergic synapses formed by axons of hypothalamic OT neurons in the central nucleus of amygdala. To validate this finding functionally, the authors activated OT axonal terminals in acute amygdala slices from rats subjected to fear conditioning and found a small glutamate-mediated response of postsynaptic cells. Intriguingly, Hasan and colleagues showed activated glutamatergic over OT-ergic transmission in rats, which previously experienced fear. This shift was further confirmed in vivo via evoking OT and glutamate release demonstrating that the rapid onset of mobility in fear-conditioned rats remained even after the block of OTRs by its selective antagonist. Thus these two reports support the general notion that coordinated release of slow-acting neuropeptide neuromodulator and fast-acting amino acid neurotransmitters is a mechanism essential for the modulation of cognitive, emotional, and metabolic processes (van den Pol, 2012). However, potentially simultaneous synaptic glutamate and nonsynaptic OT releases from the very same axon request further studies implementing high resolution imaging in vitro and ex vivo techniques in combination with selective labeling of LDCVs (Persoon et al., 2019). It is important to note, however, that despite the general consensus in the field, it is not entirely clear whether OT is exclusively packaged in and released from LDCVs.

INTRACELLULAR OXYTOCIN RECEPTOR SIGNALING OTR are widely distributed in the rodent brains, and OTR expressing cells have been mapped in virtually all major forebrain regions (Newmaster et al., 2020). As it was summarized in recent reviews (Busnelli and Chini, 2018; Jurek and Neumann, 2018; Grinevich and Neumann, 2020), OTR belongs to a G (guanine nucleotide-binding) protein-coupled receptor (GPCR) family. OTR interacts with heterotrimeric G protein complexes (Ga, Gb, and Gg) and can be linked to multiple signaling pathways depending on the specific G protein complex involved (e.g., Gaq, Gao, Gai). For example, activation of the Ga proteins Gaq and Ga11, which are both expressed in the brain, stimulates phospholipase C resulting in the generation of inositol 1,4,5-triphosphate (IP3) and 1,2-diacylglycerol (DAG). Whereas IP3 mobilizes Ca2+ from intracellular stores, DAG activates protein kinase C (PKC) and, thus, phosphorylates a number of other downstream target proteins (Fig. 3.6). Another important intraneuronal signaling pathway activated by OTR both via Gaq/11 and Gai/o is the mitogen-activated protein kinase (MAPK) cascade. This pathway, which requires both, transactivation of the epidermal growth factor receptor (EGFR) (Blume et al., 2008) and the influx of extracellular Ca2+ through

CENTRAL AND PERIPHERAL RELEASE OF OXYTOCIN

33

Fig. 3.6. Scheme of intraneuronal signaling cascades involved in OT-induced anxiolysis in the PVN. The activation of OTR elevates—via the Gb/g protein subunit—extracellular Ca2+ influx via incorporation of TRPV2 calcium channels into the cellular membrane and subsequent activation of calcium-dependent cascades (PKC, CaMKI, II, IV). OTR activation via its G-protein aq subunit also leads to transactivation of EGFR and subsequent activation of the MAPK pathway via MEK1/2. The signaling cascades converge on downstream regulation of transcriptions factors, such as MEF2, CREB, and its co-factor CRTC3 and, consequently, modulate neuronal gene expression. Central infusion OT resulted in 157 up- and 204 downregulated genes identified in rat PVN tissue punches (Martinetz et al., 2019). Moreover, through activation of PKC and MEK1/2, OT regulates eEF2 activity to promote protein de novo synthesis, e.g., synthesis of the NPY5R. Abbreviations: OT, oxytocin; OTR, OT receptor; Ca2 +, calcium; TRPV2, transient receptor potential cation channels of vanilloid type 2; EGFR, epidermal growth factor receptor; CaMK, calcium/ calmodulin-dependent kinase; PKC, protein kinase C; MAPK, mitogen activated protein kinase pathway; MEK, mitogen-activated protein kinase kinase; CRTC, cyclic AMP-regulated transcriptional coactivators; CREB, cyclic AMP responsive element binding protein; MEF-2, myocyte enhancer factor 2; EF2, eukaryotic elongation factor 2; NPY5R, neuropeptide Y 5 receptor. Reproduced from Grinevich V, Neumann I (2020). Brain oxytocin: how puzzle stones from animal studies translate into psychiatry. Mol Psychiatry 26: 265–279. doi:10.1038/s41380-020-0802-9.

transient receptor potential vanilloid type 2 (TRPV2) channels (van den Burg et al., 2015), is essential for the anxiolytic effect of OT (Blume et al., 2008; Jurek et al., 2012; van den Burg et al., 2015). Moreover, the hippocampal MEK1/2–MAPK cascade has also been associated with OT-regulated spatial memory formation in lactating animals, which was shown to be CREB dependent (Tomizawa et al., 2003). The MAPK activation by MEK1/2 finally leads to the stimulation of the transcription factor CREB and subsequent regulation of its cofactor CRTC (TORC) in the nucleus and gene expression (Bakos et al., 2014; Havranek et al., 2015; Jurek et al., 2015; Martinetz et al., 2019). Central infusion of OT leads to upregulation and downregulation of identified 157 and 204 genes within the PVN, respectively (Martinetz et al., 2019). As was demonstrated by Inga Neumann’ group, among the upregulated genes was the neuropeptide Y receptor 5 (Martinetz et al.,

2019), which activity is necessary for the anxiolytic effect of OT in particular. Interestingly, OTR-activated de novo protein synthesis further involves the eukaryotic elongation factor eEF2 (Busnelli and Chini, 2018), which was stimulated by OT in a PKC-dependent manner within the PVN (Martinetz et al., 2019). The previously listed intraneuronal pathways linking acute OTR activation to cytoplasmic or nuclear targets are essential to induce a neuron-specific response ultimately resulting in behavioral or physiologic responses. These pathways identified so far in hypothalamic or hippocampal neurons have mainly been associated with OT-induced anxiolysis, memory formation, and stressrelated behaviors. The detailed intraneuronal responses to OT in any other behavioral context are far from being understood. Neuron-specific OTR-coupled pathways, which are likely to depend on the predominant quality of expressed OTR-coupled G-proteins (e.g., Gaq or Gai),

F. ALTHAMMER ET AL.

34

the neuron-specific density of OTR, and the regiondependent types of neurons expressing the OTR may contribute to the variability and specificity of neuronal responses resulting in multiple, sometimes even opposing behavioral effects of the neuropeptide. These and other factors, such as the duration of OTR activation (acute versus chronic OT action) or differential modulation of local regional circuits via acting on OTR-expressing cells integrated into the local neuronal ensembles, may be responsible for the phenomenon that a single-molecule binding to a single type of GPCR can induce an enormous variability of effects in the brain.

OXYTOCIN SYSTEM IN MONKEYS AND HUMAN In accordance with data obtained from studies in rats, in nonhuman primates, the synthesis of OT is mostly restricted to the SON, PVN, and AN. However, in the rhesus macaque, the crab-eating macaque and the Japanese macaque OT-producing neurons have also been found in the perifornical area and globus pallidus (Ragen and Bales, 2013). Furthermore, in these three primate species, OT cell bodies have also been identified in the BNST (Ragen and Bales, 2013). Intriguingly, a small population of OT-ergic neurons in the BNST has been recently found also in mice (Duque-Wilckens et al., 2017) and rats (DiBenedictis et al., 2017) suggesting anatomic similarity of this extrahypothalamic OT-ergic cell group in rodents and primates. Although the precise number of OT neurons in nonhuman primates has not been established yet, their number can be estimated to amount to approximately 10,000– 20,000 neurons (given the numbers of OT neurons in rat vs human brain, and the respective brain sizes). While the total number of parvOT neurons in rats account for approximately 1% of all OT neurons (Althammer and Grinevich, 2017), this number has not been systematically addressed yet for primates. The human hypothalamus contains less OT cells than AVP cells, and the anatomic organization of the OT system in humans is similar to monkeys (but distinct from rodents): the human SON is composed of three anatomically distinct parts: dorsolateral, dorsomedial, and ventromedial parts. The largest dorsolateral part of the SON is composed of 53,000 magnocellular neurons, but only 10% of them are OT-immunoreactive cells; the remaining 90% represent AVP cells. Thus the OT population in the dorsolateral part amounts to 5300 cells (Dierickx and Vandesande, 1977; Rhodes et al., 1981). The dorsomedial part and ventromedial part of the SON together contain 23,000 magnocellular neurons. However, only 15% of

them are OT-immunoreactive, while 85% belong to AVP-immunoreactive cells. Altogether, both dorsomedial and ventromedial parts of the SON contain 3450 OT neurons (Dierickx and Vandesande, 1977; Rhodes et al., 1981; Althammer and Grinevich, 2017). Hence the total number of OT cells in the human SON is 17,500 OT cells in both hemispheres. These calculations are in line with Morton’s paper (Morton, 1969), where he states that the SON contains 75,000 neurons (per hemisphere), of which 10%– 15% are OT-ergic neurons, while OT and AVP cells in the PVN are equally represented. The human PVN contains approximately 42,000 OT cells bilaterally (Wierda et al., 1991; Van der Woude et al., 1995). Thus the SON and PVN in both hemispheres contain approximately 50,000 OT neurons in total. With respect to parvOT neurons, their total number in humans has not been estimated due to composition of the PVN by intermingled populations of “bigger and smaller” OT neurons (Wierda et al., 1991) and the lack of immunohistochemical markers to exclusively label parvOT neurons postmortem. A recent study performed a postmortem assessment of human brain sections via immunohistochemical staining for the OT/AVP carrier proteins neurophysin I and II and found a prominent system of accessory nuclei, while some of the neurons are in close proximity to blood vessels (Moller et al., 2018). Moreover, Møller and colleagues describe that the accessory nuclei system in humans is larger than in rodents or other primates and displays a considerably different morphology. Already in the late 1970s and early 1980s of the last century, the magnocellular system of monkeys has been extensively studied and first OT and AVP-immunopositive fibers have been reported in extrahypothalamic areas, including the BNST, globus pallidus, suprachiasmatic nucleus, and caudate nucleus (Antunes and Zimmerman, 1978; Antunes et al., 1979; Kawata and Sano, 1982). Studies investigating the respective distribution of OTRs in various monkey species followed much later, due to the technical difficulties in developing reliable techniques to label OTRs in the monkey brain (Freeman et al., 2014; Freeman and Young, 2016). In humans, OT neurons have been reported to project to various brain areas including the BNST, hippocampus, amygdala, or prefrontal cortex (Sofroniew, 1980; Meyer-Lindenberg et al., 2011; Lin et al., 2017). In line with the described projections, OTRs in humans have been found in the amygdala, medial preoptic area, hypothalamus, olfactory nucleus, vertical limb of the diagonal band, ventrolateral septum, anterior cingulate, and hypoglossal and solitary nuclei (Boccia et al., 2013). Moreover, a recent study identified human OTR expression in the hippocampus, putamen, pallidum, and cerebellum (Quintana et al., 2017).

CENTRAL AND PERIPHERAL RELEASE OF OXYTOCIN

35

Fig. 3.7. Overview of different OTR expression patterns in rodents, monkeys, and humans. Brain schemes depict OTR expression in the rat, monkey, and human brain. The selected key socially relevant brain regions are depicted and their colors represent the respective levels of the OTR expression/binding: low ¼ green, moderate ¼ blue, high ¼ red. Amy, Amygdala: ACC, Anterior cingulate cortex; AON, anterior olfactory nucleus: Arc, arcuate hypothalamic nucleus; AC, auditory cortex; BNST, bed nucleus of stria terminalis; Cpu, caudate putamen; CeL; central lateral amygdala; GP, Globus pallidus; HC, hippocampus; HDB, horizontal limb of diagonal band nucleus; HYP, hypothalamus; iCj, Island of Calleja; LS, lateral septum; NBM, nucleus basalis of Meynert; NAcc, nucleus accumbens; OFC, Orbifrontal cortex; PVN, paraventricular nucleus of the hypothalamus; PFC, prefrontal cortex; PLC, prelimbic cortex; PAC, primary auditory cortex; PVC, primary visual cortex; SuC, superior colliculus; SON, supraoptic nucleus; VC, visual cortex; VP, ventral pallidum; VMH, ventromedial nucleus of the hypothalamus.

In contrast to rodents, the most prominent appearance of OTR in monkeys and human was found in brain areas that are involved in visual attention, processing of visual inputs, and control of eye movement (Freeman et al., 2014, Freeman and Young, 2016). (Fig. 3.7). This salient discrepancy in receptor distribution patterns highlights a fundamental difference in the processing of sensory cues by OT in these two species. While OTR activation of olfactory pathways in rats seems to mainly promote the recognition of conspecifics, OT mediates visual attention in primates, potentially to improve in-group communication and flatten hierarchy (Jiang and Platt, 2018). Similar studies in humans have shown that OT mediates out-group attacks and intergroup conflicts (Zhang et al., 2019). Therefore it is tempting to speculate that OT differently affects sensory pathways in rodents and primates, as the interpretation of visual cues became increasingly important during the evolution of higher mammals, which predominantly rely on their eyes for both communication and the detection of potential threats (Grinevich and Stoop, 2018b). It is plausible that intraspecies differences in the contribution of peripheral OT release may account for distinct behavioral traits. However, since the correlative link between peripheral and central release is highly questionable (Landgraf

and Neumann, 2004; Kagerbauer et al., 2013) and the current methods give reasons for serious concerns (McCullough et al., 2013; Leng and Sabatier, 2016), we decided to not discuss this topic in our current review and would like to kindly refer the reader to additional literature (Crockford et al., 2014). Furthermore, comprehensive information about the interplay of oxytocin and sensory systems in various species can be found in our recent review (Grinevich and Stoop, 2018a).

OXYTOCIN IN HUMAN HEALTH In humans, a variety of diseases has been directly or indirectly linked to pathologic alterations in the OT system: in Prader–Willi syndrome, a genetic disease characterized by autistic traits and obesity, impairment of OT signaling seems to play a crucial role (Swaab, 1997; Martin et al., 1998; Grinevich et al., 2014; Johnson et al., 2016). It is speculated that a pathologic alteration in parvOT neurons, which project to the nucleus of the solitary tract, might be the cause of elevated food intake, as mainly the parvocellular divisions of the PVN seem to be affected by the disease (Swaab et al., 1995). In fact, various studies could show that OT signaling is involved in regulation of meal size, control of food intake, satiety, and energy

36

F. ALTHAMMER ET AL.

balance (Blevins et al., 2004; Valassi et al., 2008; Atasoy et al., 2012; Sabatier et al., 2013; Motojima et al., 2016). Furthermore, genetic mutations of the OTR in humans have been associated with higher risk of autism spectrum disorders (Kirsch et al., 2005; Meyer-Lindenberg et al., 2011) and females suffering from depression displayed significantly lower plasma OT levels than healthy subjects (Frasch et al., 1995; Ozsoy et al., 2009). Given the manifold positive effects of OT on mood, social behavior and attention, the neuropeptide has been proposed for a variety of treatments, including autism, schizophrenia, and posttraumatic stress disorder (MeyerLindenberg et al., 2011; Frijling et al., 2014; Neumann and Slattery, 2016; Frijling, 2017; Krystal et al., 2017). Looking back at 700 million years of evolution, it is remarkable that in nematodes (C. elegans), 5% (!) among all neurons express OT-like peptides (Garrison et al., 2012), while in human this number corresponds to about 0.00006% of total neuronal cells (Althammer and Grinevich, 2017). In the light of these numbers, it is unbiasedly impressive that intranasal application of OT in humans modulates a plethora of complex behaviors including fear (Kirsch et al., 2005; Petrovic et al., 2008), empathy (Barraza and Zak, 2009; Hurlemann et al., 2010), trust (Kosfeld et al., 2005), and even xenophobia (De Dreu et al., 2011; Marsh et al., 2017) often accompanied by altered neuronal activity (Eckstein et al., 2015). However, it might seem highly surprising that despite these clear effects in humans, the route by which OT reaches the brain (if at all) is anything but clear at this point in time and this controversial topic is still being heavily discussed among the scientific community (Leng and Ludwig, 2016). The two major theories regarding the intranasally induced effects of OT are: (a) activation of peripheral OTRs and subsequent induction of endogenous release of OT within the brain and (b) small, but still physiologically relevant amounts of OT can cross the blood–brain barrier to exert its functions. In 2018, a study showed that OT reaches the rhesus macaque brain through intravenous and intranasal routes, but was criticized for the inconsistent results, since OT was only detected in the brain of a few monkeys after application (Lee et al., 2018). In a very recent study the same authors demonstrated more convincingly (Lee et al., 2020) that OT reaches the brain in significant amounts through the intranasal route in rhesus macaques. However, there might be species-dependent differences and future studies are needed to evaluate how these findings could be translated to intranasal application of OT in humans. In addition, more research is needed to show how the technical progress in animal experimental research could be extrapolated to humans, namely, the

treatment of various diseases, in which the OT system is compromised (Swaab et al., 1995; Swaab, 1997; Meyer-Lindenberg et al., 2011).

PERSPECTIVES Research over the last decades led to significant advances in the understanding of the mammalian OT and AVP systems. The generation of transgenic mice and rats allowing the expression of fluorescent proteins under the AVP/OT promoter and/or the promoters for their receptors will soon make it possible to create an entire map of the respective neuropeptide projections within several rodent species (Venkatesh et al., 1997; Young et al., 1999; Ueta et al., 2005; Katoh et al., 2010; Hidema et al., 2016). Recent advances in viral vector-based techniques allow for cell-specific expression and manipulation of OT neurons (Grinevich et al., 2016b), and the viral vectors were independently developed by teams of Harold Gainer (Fields et al., 2012) and Valery Grinevich (Knobloch et al., 2012). These viral constructs, equipped with short sequences of OT and AVP promoters, have been used to successfully label and manipulate neuronal activity and behavior in rodents (Knobloch et al., 2012; Eliava et al., 2016b; Menon et al., 2018; Hasan et al., 2019). Notably, these viral vectors can be used in a wide range of subjects, including labeling and manipulation of OT neurons in monkeys (Fig. 3.8). While the neuronal effects of OT (as well as AVP) are extensively described, only recently researchers focused on effects of these neuropeptides on nonneuronal cells in the CNS. Although OTRs have not been found in microglia in adult rodents (Gosselin et al., 2014; Bennett et al., 2016; Matcovitch-Natan et al., 2016; Tay et al., 2017), several studies demonstrated various interactions between the neuropeptide and microglia (Yuan et al., 2016; Inoue et al., 2019; Mairesse et al., 2019), suggesting a potential indirect effect. Further, a few recent studies suggest interactions between OT and oligodendrocyte progenitor cells (Havranek et al., 2017; Palanisamy et al., 2018) or OT-induced oligodendrocyte cell death (Hirayama et al., 2020). Several studies investigating the effect of OT on astrocytes demonstrated far-reaching consequences of this neuropeptide–glia relationship, including glutamate receptor interaction (Kuo et al., 2009), astrocyte plasticity (Wang and Hatton, 2009), and GFAP plasticity (Wang et al., 2017). Moreover, our recent work demonstrated that OTR-expressing astrocytes in the central amygdala are necessary for the modulation of positive emotions and comfort (Wahis et al., 2021). Recent advances in stem cells research have now made it possible to generate

CENTRAL AND PERIPHERAL RELEASE OF OXYTOCIN

37

hypothalamic oxytocin neurons

oxytocin receptor

Gα11 Gαq Gαo Gαi

homeostatic adaptation metabolism energy expediture blood pressure breathing appetite

Fig. 3.8. Cell-type specific tagging of OT neurons by viral vectors in various mammalian species. The expression of fluorescent marker Venus (green) in PVN OT neurons after their infection by OT promoter-containing rAAV in a vole (A), mouse (B), rat (C), and marmoset (D). The sections were processed with the antibody against OT (red), resulting in overlay of both signals in yellow. The images of mouse, rat, and marmoset PVN were generated in Grinevich’s laboratory, while the image of vole PVN is reproduced from Bosch et al. (2016).

various hypothalamic neurons (including OT cells) from human pluripotent stem cells (Merkle et al., 2015). One of the remaining key question is how one single neuropeptide can have so many different neuromodulatory and whole-body physiologic effects, as well as impact a plethora of different homeostatic functions (McCormack et al., 2020) and various behaviors described in this chapter. The broader picture of OT actions remains far from being complete and data supports pleiotropic effects of OT and other neuropeptides. This phenomenon, typical for many of about 100 identified neuropeptides (http://www.neuropeptides.nl), may emerge from the anatomic divergence of OT neurons, their multiple central projections, corelease of the neuropeptide and glutamate, distinct oxytocin-sensitive cell types in different brain regions, and multiple intraneuronal signaling pathways determining the specific neuronal

modulated behaviors stress-coping sociability reproduction feeding

Fig. 3.9. Pleiotropic effects of OT. The hypothetic scheme depicts how OT exerts its pleiotropic effects on the mammalian body. The OT is produced by at least two types of OT neurons—magnOT and parvOT cells and released either centrally or peripherally, acting on OTRs-expressing cells. Notably, the similar forms of OTR–GPCR subunits have been found in both peripheral and nervous tissues (Busnelli and Chini, 2018). Blood OT, originating from magnOT neurons that project to the posterior pituitary, acts on diverse body cells (such as adipocytes, cells of blood vessels, cardiomyocytes, smooth muscle cells), while centrally released OT affects all components of nervous system (i.e., neurons, astrocytes, microglial cells, and oligodendrocytes). Such simultaneous although diverse effects of the neuropeptide might result in counter-balanced homeostatic adaptations and behavioral challenges.

responses as summarized in Fig. 3.9 and explained in details in Box 3.1.

ACKNOWLEDGMENTS The authors thank Dr. Kumi Kuroda for the permission to present the results on viral vector-based labeling of oxytocin neurons in marmosets (Fig. 3.8), obtained during long-lasting cooperation with Valery Grinevich, and Prof. Javier E. Stern for his valuable feedback to this book chapter. The work was supported by German Research Foundation (DFG) postdoctoral fellowship AL 2466/1-1 to FA, DFG Collaborative Research Center (SFB) 1158, DFG grants GR 3619/701, GR 3619/4-1, SFB 1158, SNSF-DFG grant GR 3619/8-1, and Fritz Thyssen Foundation grant 10.16.2.018 MN to VG.

38

F. ALTHAMMER ET AL.

BOX 3.1. UNIQUE FEATURES OF OT AND ITS RECEPTOR AND ITS MECHANISM OF RELEASE THAT CONTRIBUTE TO THE FAR-REACHING CONSEQUENCES OF THE NEUROPEPTIDE EXERTS IN THE MAMMALIAN BODY. ●

● ●

● ● ● ● ●

OT is coexpressed and released with glutamate from the same neurons (Meeker et al., 1991; Knobloch et al., 2012). Although the common release frequencies are different (high-frequency action potential discharge triggers preferentially neuropeptide release: 5 Hz for glutamate and 20–50 Hz for OT (H€ okfelt, 1991), the precisely tuned actions of these transmitters allow for sophisticated control and modulation of postsynaptic neurons (van den Pol, 2012). The corelease of OT with other neurotransmitters, such as glutamate, allows for fine-tuned neuronal responses that could regulate the strength of sustained synaptic inputs to OTR-expressing cells (Hasan et al., 2019). The somatodendritic release is thought to be one of the key features of OT neurons, which allow them to locally control activity of distinct OT populations and coordinated responses during physiologic challenges, such as lactation (FreundMercier and Richard, 1984; Moos et al., 1989). The release or inhibition of release of OT from axonal endings allows rapid effects to elicit brain-region-specific behaviors (Knobloch et al., 2012; Eliava et al., 2016a,b; Menon et al., 2018; Hasan et al., 2019). OTR expression was found in substantial numbers of different brain regions (Gimpl and Fahrenholz, 2001; Marlin and Froemke, 2017) and peripheral organs (Gimpl and Fahrenholz, 2001), which modulate a plethora of different functions. OTR expressing neurons come in a variety of different types such as GABAergic interneurons (Huber et al., 2005), glutamatergic pyramidal cells (Lin et al., 2017), or other neuroendocrine cells (Ludwig and Leng, 2006; Jurek et al., 2015). OT has a very high affinity to its own receptor (EC50 1–100 nM) and can thereby trigger responses as well as exert finetuning of neuronal networks via release of only a few molecules (Chini et al., 2017). Despite its single receptor, OT can differently affect the target cell via activation of different G-protein-induced signaling cascades via Ga11, Gaq, Gai, and Gao subunits (Gimpl and Fahrenholz, 2001).

REFERENCES Akerlund M (2006). Targeting the oxytocin receptor to relax the myometrium. Expert Opin Ther Targets 10: 423–427. Althammer F, Grinevich V (2017). Diversity of oxytocin neurons: beyond magno- and parvocellular cell types? J Neuroendocrinol 30. Antunes JL, Zimmerman EA (1978). The hypothalamic magnocellular system of the rhesus monkey: an immunocytochemical study. J Comp Neurol 181: 539–565. Antunes JL, Carmel PW, Zimmerman EA et al. (1979). Regeneration of the magnocellular system of the rhesus monkey following hypothalamic lesions. Ann Neurol 5: 462–469. Armstrong WE (2004). Hypothalamic supraoptic and paraventricular nuclei. Atasoy D, Betley JN, Su HH et al. (2012). Deconstruction of a neural circuit for hunger. Nature 488: 172–177. Bakos J, Lestanova Z, Strbak V et al. (2014). Neonatal manipulation of oxytocin prevents lipopolysaccharide-induced decrease in gene expression of growth factors in two developmental stages of the female rat. Neuropeptides 48: 281–286. Barraza JA, Zak PJ (2009). Empathy toward strangers triggers oxytocin release and subsequent generosity. Ann N Y Acad Sci 1167: 182–189. Bennett ML, Bennett FC, Liddelow SA et al. (2016). New tools for studying microglia in the mouse and human CNS. Proc Natl Acad Sci U S A 113: E1738–E1746. Blevins JE, Schwartz MW, Baskin DG (2004). Evidence that paraventricular nucleus oxytocin neurons link hypothalamic leptin action to caudal brain stem nuclei controlling

meal size. Am J Physiol Regul Integr Comp Physiol 287: R87–R96. Blouet C, Jo YH, Li X et al. (2009). Mediobasal hypothalamic leucine sensing regulates food intake through activation of a hypothalamus-brainstem circuit. J Neurosci 29: 8302–8311. Blume A, Bosch OJ, Miklos S et al. (2008). Oxytocin reduces anxiety via ERK1/2 activation: local effect within the rat hypothalamic paraventricular nucleus. Eur J Neurosci 27: 1947–1956. Boccia ML, Petrusz P, Suzuki K et al. (2013). Immunohistochemical localization of oxytocin receptors in human brain. Neuroscience 253: 155–164. Bosch OJ, Kromer SA, Brunton PJ et al. (2004). Release of oxytocin in the hypothalamic paraventricular nucleus, but not central amygdala or lateral septum in lactating residents and virgin intruders during maternal defence. Neuroscience 124: 439–448. Bosch OJ, Dabrowska J, Modi ME et al. (2016). Oxytocin in the nucleus accumbens shell reverses CRFR2-evoked passive stress-coping after partner loss in monogamous male prairie voles. Psychoneuroendocrinology 64: 66–78. Buijs RM (1983). Vasopressin and oxytocin—their role in neurotransmission. Pharmacol Ther 22: 127–141. Buijs RM, Swaab DF (1979). Immuno-electron microscopical demonstration of vasopressin and oxytocin synapses in the limbic system of the rat. Cell Tissue Res 204: 355–365. Buijs RM, Van Heerikhuize JJ (1982). Vasopressin and oxytocin release in the brain—a synaptic event. Brain Res 252: 71–76. Busnelli M, Chini B (2018). Molecular basis of oxytocin receptor signalling in the brain: what we know and what we need to know. Curr Top Behav Neurosci 35: 3–29.

CENTRAL AND PERIPHERAL RELEASE OF OXYTOCIN Carter CS (2014). Oxytocin pathways and the evolution of human behavior. Annu Rev Psychol 65: 17–39. Carter CS (2017). The oxytocin-vasopressin pathway in the context of love and fear. Front Endocrinol (Lausanne) 8: 356. Chini B, Verhage M, Grinevich V (2017). The action radius of oxytocin release in the mammalian CNS: from single vesicles to behavior. Trends Pharmacol Sci 38: 982–991. PMID: 28899620. Condes-Lara M, Rojas-Piloni G, Martinez-Lorenzana G et al. (2012). Functional interactions between the paraventricular hypothalamic nucleus and raphe magnus. A comparative study of an integrated homeostatic analgesic mechanism. Neuroscience 209: 196–207. Crockford C, Deschner T, Ziegler TE et al. (2014). Endogenous peripheral oxytocin measures can give insight into the dynamics of social relationships: a review. Front Behav Neurosci 8: 68. De Dreu CK, Greer LL, Van Kleef GA et al. (2011). Oxytocin promotes human ethnocentrism. Proc Natl Acad Sci U S A 108: 1262–1266. De Vries GJ, Buijs RM, Sluiter AA (1984a). Gonadal hormone actions on the morphology of the vasopressinergic innervation of the adult rat brain. Brain Res 298: 141–145. De Vries GJ, Buijs RM, Van Leeuwen FW (1984b). Sex differences in vasopressin and other neurotransmitter systems in the brain. Prog Brain Res 61: 185–203. De Vries GJ, Rissman EF, Simerly RB et al. (2002). A model system for study of sex chromosome effects on sexually dimorphic neural and behavioral traits. J Neurosci 22: 9005–9014. Dibenedictis BT, Nussbaum ER, Cheung HK et al. (2017). Quantitative mapping reveals age and sex differences in vasopressin, but not oxytocin, immunoreactivity in the rat social behavior neural network. J Comp Neurol 525: 2549–2570. Dierickx K, Vandesande F (1977). Immunocytochemical localization of the vasopressinergic and the oxytocinergic neurons in the human hypothalamus. Cell Tissue Res 184: 15–27. Du Vigneaud V, Ressler C, Trippett S (1953). The sequence of amino acids in oxytocin, with a proposal for the structure of oxytocin. J Biol Chem 205: 949–957. Dumais KM, Veenema AH (2016). Vasopressin and oxytocin receptor systems in the brain: sex differences and sex-specific regulation of social behavior. Front Neuroendocrinol 40: 1–23. Duque-Wilckens N, Steinman M, Grinevich V et al. (2017). The role of oxytocin neurons in the bed nucleus of the stria terminalis in mediating social withdrawal. Biol Psychiatry 81: S44–S45. Ebner K, Wotjak CT, Landgraf R et al. (2000). A single social defeat experience selectively stimulates the release of oxytocin, but not vasopressin, within the septal brain area of male rats. Brain Res 872: 87–92. Ebner K, Bosch OJ, Kromer SA et al. (2005). Release of oxytocin in the rat central amygdala modulates stress-coping behavior and the release of excitatory amino acids. Neuropsychopharmacology 30: 223–230. Eckstein M, Becker B, Scheele D et al. (2015). Oxytocin facilitates the extinction of conditioned fear in humans. Biol Psychiatry 78: 194–202.

39

Eliava M, Melchior M, Knobloch-Bollmann HS et al. (2016a). A new population of parvocellular oxytocin neurons controlling magnocellular neuron activity and inflammatory pain processing. Neuron 89: 1291–1304. Eliava M, Melchior M, Knobloch-Bollmann HS et al. (2016b). A new population of parvocellular oxytocin neurons controlling magnocellular neuron activity and inflammatory pain processing. Neuron 89: 1291–1304. Engelmann M, Ebner K, Landgraf R et al. (1999). Emotional stress triggers intrahypothalamic but not peripheral release of oxytocin in male rats. J Neuroendocrinol 11: 867–872. Engelmann M, Landgraf R, Wotjak CT (2004). The hypothalamic-neurohypophysial system regulates the hypothalamic-pituitary-adrenal axis under stress: an old concept revisited. Front Neuroendocrinol 25: 132–149. Everts HG, De Ruiter AJ, Koolhaas JM (1997). Differential lateral septal vasopressin in wild-type rats: correlation with aggression. Horm Behav 31: 136–144. Fields RL, Ponzio TA, Kawasaki M et al. (2012). Cell-type specific oxytocin gene expression from AAV delivered promoter deletion constructs into the rat supraoptic nucleus in vivo. PLoS One 7: e32085. Frasch A, Zetzsche T, Steiger A et al. (1995). Reduction of plasma oxytocin levels in patients suffering from major depression. Adv Exp Med Biol 395: 257–258. Freeman SM, Young LJ (2016). Comparative perspectives on oxytocin and vasopressin receptor research in rodents and primates: translational implications. J Neuroendocrinol 28. Freeman SM, Walum H, Inoue K et al. (2014). Neuroanatomical distribution of oxytocin and vasopressin 1a receptors in the socially monogamous coppery titi monkey (Callicebus cupreus). Neuroscience 273: 12–23. Freund-Mercier MJ, Richard P (1984). Electrophysiological evidence for facilitatory control of oxytocin neurones by oxytocin during suckling in the rat. J Physiol 352: 447–466. PMID: 6747898. Frijling JL (2017). Preventing PTSD with oxytocin: effects of oxytocin administration on fear neurocircuitry and PTSD symptom development in recently trauma-exposed individuals. Eur J Psychotraumatol 8: 1302652. Frijling JL, Van Zuiden M, Koch SB et al. (2014). Efficacy of oxytocin administration early after psychotrauma in preventing the development of PTSD: study protocol of a randomized controlled trial. BMC Psychiatry 14: 92. Gabor CS, Phan A, Clipperton-Allen AE et al. (2012). Interplay of oxytocin, vasopressin, and sex hormones in the regulation of social recognition. Behav Neurosci 126: 97–109. Garrison JL, Macosko EZ, Bernstein S et al. (2012). Oxytocin/ vasopressin-related peptides have an ancient role in reproductive behavior. Science 338: 540–543. Gimpl G, Fahrenholz F (2001). The oxytocin receptor system: structure, function, and regulation. Physiol Rev 81: 629–683. PMID: 11274341. Gosselin D, Link VM, Romanoski CE et al. (2014). Environment drives selection and function of enhancers controlling tissue-specific macrophage identities. Cell 159: 1327–1340.

40

F. ALTHAMMER ET AL.

Griffin GD, Ferri-Kolwicz SL, Reyes BA et al. (2010). Ovarian hormone-induced reorganization of oxytocin-labeled dendrites and synapses lateral to the hypothalamic ventromedial nucleus in female rats. J Comp Neurol 518: 4531–4545. Grinevich V, Neumann I (2020). Brain oxytocin: how puzzle stones from animal studies translate into psychiatry. Mol Psychiatry, 26, 265–279. https://doi.org/10.1038/s41380020-0802-9. Grinevich VV, Polenov AL (1994). The evolution of the nonapeptidergic neurosecretory formations of the hypothalamus in vertebrate animals. Zh Evol Biokhim Fiziol 30: 270–292. Grinevich V, Stoop R (2018a). Interplay between oxytocin and sensory systems in the orchestration of socio-emotional behaviors. Neuron 99: 887–904. Grinevich V, Stoop R (2018b). Interplay between oxytocin and sensory systems in the orchestration of socio-emotional behaviors. Neuron 99: 887–904. Grinevich V, Desarmenien MG, Chini B et al. (2014). Ontogenesis of oxytocin pathways in the mammalian brain: late maturation and psychosocial disorders. Front Neuroanat 8: 164. Grinevich V, Knobloch-Bollmann HS, Eliava M et al. (2016a). Assembling the puzzle: pathways of oxytocin signaling in the brain. Biol Psychiatry 79: 155–164. Grinevich V, Knobloch-Bollmann HS, Roth LC et al. (2016b). Somatic transgenesis (viral vectors), John Wiley & Sons. Hasan MT, Althammer F, Silva Da Gouveia M et al. (2019). A fear memory engram and its plasticity in the hypothalamic oxytocin system. Neuron 103: 133–146 e8. Havranek T, Zatkova M, Lestanova Z et al. (2015). Intracerebroventricular oxytocin administration in rats enhances object recognition and increases expression of neurotrophins, microtubule-associated protein 2, and synapsin I. J Neurosci Res 93: 893–901. Havranek T, Lestanova Z, Mravec B et al. (2017). Oxytocin modulates expression of neuron and glial markers in the rat hippocampus. Folia Biol (Praha) 63: 91–97. Hernandez VS, Vazquez-Juarez E, Marquez MM et al. (2015). Extra-neurohypophyseal axonal projections from individual vasopressin-containing magnocellular neurons in rat hypothalamus. Front Neuroanat 9: 130. Hernandez-Perez OR, Hernandez VS, Nava-Kopp AT et al. (2019). A synaptically connected hypothalamic magnocellular vasopressin-locus coeruleus neuronal circuit and its plasticity in response to emotional and physiological stress. Front Neurosci 13: 196. Hidema S, Fukuda T, Hiraoka Y et al. (2016). Generation of Oxtr cDNA(HA)-ires-cre mice for gene expression in an oxytocin receptor specific manner. J Cell Biochem 117: 1099–1111. Hirayama T, Hiraoka Y, Kitamura E et al. (2020). Oxytocin induced labor causes region and sex-specific transient oligodendrocyte cell death in neonatal mouse brain. J Obstet Gynaecol Res 46: 66–78. Ho JM, Anekonda VT, Thompson BW et al. (2014). Hindbrain oxytocin receptors contribute to the effects of circulating oxytocin on food intake in male rats. Endocrinology 155: 2845–2857.

H€ okfelt T (1991). Neuropeptides in perspective: the last ten years. Neuron 7: 867–879. PMID: 1684901. Hrabovszky E, Deli L, Turi GF et al. (2007). Glutamatergic innervation of the hypothalamic median eminence and posterior pituitary of the rat. Neuroscience 144: 1383–1392. Huber D, Veinante P, Stoop R (2005). Vasopressin and oxytocin excite distinct neuronal populations in the central amygdala. Science 308: 245–248. Hung LW, Neuner S, Polepalli JS et al. (2017). Gating of social reward by oxytocin in the ventral tegmental area. Science 357: 1406–1411. Hurlemann R, Patin A, Onur OA et al. (2010). Oxytocin enhances amygdala-dependent, socially reinforced learning and emotional empathy in humans. J Neurosci 30: 4999–5007. Inoue T, Yamakage H, Tanaka M et al. (2019). Oxytocin suppresses inflammatory responses induced by lipopolysaccharide through inhibition of the eIF-2-ATF4 pathway in mouse microglia. Cell 8: 527. Jiang Y, Platt ML (2018). Oxytocin and vasopressin flatten dominance hierarchy and enhance behavioral synchrony in part via anterior cingulate cortex. Sci Rep 8: 8201. Johnson L, Manzardo AM, Miller JL et al. (2016). Elevated plasma oxytocin levels in children with Prader-Willi syndrome compared with healthy unrelated siblings. Am J Med Genet A 170: 594–601. Juif PE, Poisbeau P (2013). Neurohormonal effects of oxytocin and vasopressin receptor agonists on spinal pain processing in male rats. Pain 154: 1449–1456. Juif PE, Breton JD, Rajalu M et al. (2013). Long-lasting spinal oxytocin analgesia is ensured by the stimulation of allopregnanolone synthesis which potentiates GABA(A) receptormediated synaptic inhibition. J Neurosci 33: 16617–16626. Jurek B, Neumann ID (2018). The oxytocin receptor: from intracellular signaling to behavior. Physiol Rev 98: 1805–1908. Jurek B, Slattery DA, Maloumby R et al. (2012). Differential contribution of hypothalamic MAPK activity to anxietylike behaviour in virgin and lactating rats. PLoS One 7: e37060. Jurek B, Slattery DA, Hiraoka Y et al. (2015). Oxytocin regulates stress-induced Crf gene transcription through CREBregulated transcription coactivator 3. J Neurosci 35: 12248–12260. Kagerbauer SM, Martin J, Schuster T et al. (2013). Plasma oxytocin and vasopressin do not predict neuropeptide concentrations in human cerebrospinal fluid. J Neuroendocrinol 25: 668–673. Katoh A, Fujihara H, Ohbuchi T et al. (2010). Specific expression of an oxytocin-enhanced cyan fluorescent protein fusion transgene in the rat hypothalamus and posterior pituitary. J Endocrinol 204: 275–285. Kawasaki A, Shutoh F, Nogami H et al. (2006). VGLUT2 expression is up-regulated in neurohypophysial vasopressin neurons of the rat after osmotic stimulation. Neurosci Res 56: 124–127. Kawata M, Sano Y (1982). Immunohistochemical identification of the oxytocin and vasopressin neurons in the hypothalamus of the monkey (Macaca fuscata). Anat Embryol 165: 151–167.

CENTRAL AND PERIPHERAL RELEASE OF OXYTOCIN Kendrick KM, Keverne EB, Hinton MR et al. (1992). Oxytocin, amino acid and monoamine release in the region of the medial preoptic area and bed nucleus of the stria terminalis of the sheep during parturition and suckling. Brain Res 569: 199–209. Kirsch P, Esslinger C, Chen Q et al. (2005). Oxytocin modulates neural circuitry for social cognition and fear in humans. J Neurosci 25: 11489–11493. Knobloch HS, Grinevich V (2014). Evolution of oxytocin pathways in the brain of vertebrates. Front Behav Neurosci 8: 31. Knobloch HS, Charlet A, Hoffmann LC et al. (2012). Evoked axonal oxytocin release in the central amygdala attenuates fear response. Neuron 73: 553–566. Kosfeld M, Heinrichs M, Zak PJ et al. (2005). Oxytocin increases trust in humans. Nature 435: 673–676. Krystal JH, Davis LL, Neylan TC et al. (2017). It is time to address the crisis in the pharmacotherapy of posttraumatic stress disorder: a consensus statement of the PTSD psychopharmacology working group. Biol Psychiatry 82: e51–e59. Kuo J, Hariri OR, Micevych P (2009). An interaction of oxytocin receptors with metabotropic glutamate receptors in hypothalamic astrocytes. J Neuroendocrinol 21: 1001–1006. Landgraf R, Neumann ID (2004). Vasopressin and oxytocin release within the brain: a dynamic concept of multiple and variable modes of neuropeptide communication. Front Neuroendocrinol 25: 150–176. Landgraf R, Neumann I, Schwarzberg H (1988). Central and peripheral release of vasopressin and oxytocin in the conscious rat after osmotic stimulation. Brain Res 457: 219–225. Lee HJ, Caldwell HK, Macbeth AH et al. (2008). Behavioural studies using temporal and spatial inactivation of the oxytocin receptor. Prog Brain Res 170: 73–77. Lee MR, Scheidweiler KB, Diao XX et al. (2018). Oxytocin by intranasal and intravenous routes reaches the cerebrospinal fluid in rhesus macaques: determination using a novel oxytocin assay. Mol Psychiatry 23: 115–122. Lee MR, Shnitko TA, Blue SW et al. (2020). Labelled oxytocin administered via the intranasal route reaches the brain in rhesus macaques. Nat Commun 11: 2783. Leng G, Ludwig M (2008). Neurotransmitters and peptides: whispered secrets and public announcements. J Physiol 586: 5625–5632. Leng G, Ludwig M (2016). Intranasal oxytocin: myths and delusions. Biol Psychiatry 79: 243–250. Leng G, Sabatier N (2016). Measuring oxytocin and vasopressin: bioassays, immunoassays and random numbers. J Neuroendocrinol 28. Leng G, Pineda R, Sabatier N et al. (2015). 60 years of neuroendocrinology: the posterior pituitary, from Geoffrey Harris to our present understanding. J Endocrinol 226: T173–T185. Lin YT, Chen CC, Huang CC et al. (2017). Oxytocin stimulates hippocampal neurogenesis via oxytocin receptor expressed in CA3 pyramidal neurons. Nat Commun 8: 537. Ludwig M, Callahan MF, Neumann I et al. (1994). Systemic osmotic stimulation increases vasopressin and oxytocin release within the supraoptic nucleus. J Neuroendocrinol 6: 369–373.

41

Ludwig M, Leng G (2006). Dendritic peptide release and peptidedependent behaviours. Nat Rev Neurosci 7: 126–136. PMID: 16429122. Mack SO, Kc P, Wu M et al. (2002). Paraventricular oxytocin neurons are involved in neural modulation of breathing. J Appl Physiol (1985) 92: 826–834. Mairesse J, Zinni M, Pansiot J et al. (2019). Oxytocin receptor agonist reduces perinatal brain damage by targeting microglia. Glia 67: 345–359. Marlin BJ, Froemke RC (2017). Oxytocin modulation of neural circuits for social behavior. Dev Neurobiol 77: 169–189. Marsh N, Scheele D, Feinstein JSet al. (2017). Oxytocin-enforced norm compliance reduces xenophobic outgroup rejection. Proc Natl Acad Sci U S A 114: 9314–9319. Martin A, State M, Anderson GM et al. (1998). Cerebrospinal fluid levels of oxytocin in Prader-Willi syndrome: a preliminary report. Biol Psychiatry 44: 1349–1352. Martinetz S, Meinung CP, Jurek B et al. (2019). De novo protein synthesis mediated by the eukaryotic elongation factor 2 is required for the anxiolytic effect of oxytocin. Biol Psychiatry 85: 802–811. Matcovitch-Natan O, Winter DR, Giladi A et al. (2016). Microglia development follows a stepwise program to regulate brain homeostasis. Science 353: aad8670. Mayes CR, Watts AG, McQueen JK et al. (1988). Gonadal steroids influence neurophysin II distribution in the forebrain of normal and mutant mice. Neuroscience 25: 1013–1022. McCormack SE, Blevins JE, Lawson EA (2020). Metabolic effects of oxytocin. Endocr Rev 41: 121–145. McCullough ME, Churchland PS, Mendez AJ (2013). Problems with measuring peripheral oxytocin: can the data on oxytocin and human behavior be trusted? Neurosci Biobehav Rev 37: 1485–1492. Meddle SL, Bishop VR, Gkoumassi E et al. (2007). Dynamic changes in oxytocin receptor expression and activation at parturition in the rat brain. Endocrinology 148: 5095–5104. Meeker RB, Swanson DJ, Greenwood RS et al. (1991). Ultrastructural distribution of glutamate immunoreactivity within neurosecretory endings and pituicytes of the rat neurohypophysis. Brain Res 564: 181–193. PMID: 1687373. Melis MR, Argiolas A, Gessa GL (1986). Oxytocin-induced penile erection and yawning: site of action in the brain. Brain Res 398: 259–265. Menon R, Grund T, Zoicas I et al. (2018). Oxytocin signaling in the lateral septum prevents social fear during lactation. Curr Biol 28: 1066–1078 e6. Merkle FT, Maroof A, Wataya T et al. (2015). Generation of neuropeptidergic hypothalamic neurons from human pluripotent stem cells. Development 142: 633–643. Meyer-Lindenberg A, Domes G, Kirsch P et al. (2011). Oxytocin and vasopressin in the human brain: social neuropeptides for translational medicine. Nat Rev Neurosci 12: 524–538. Mitre M, Marlin BJ, Schiavo JK et al. (2016). A distributed network for social cognition enriched for oxytocin receptors. J Neurosci 36: 2517–2535. Moller M, Busch JR, Jacobsen C et al. (2018). The accessory magnocellular neurosecretory system of the rostral human hypothalamus. Cell Tissue Res 373: 487–498.

42

F. ALTHAMMER ET AL.

Moore FL (1992). Evolutionary precedents for behavioral actions of oxytocin and vasopressin. Ann N Y Acad Sci 652: 156–165. Moos F, Poulain DA, Rodriguez F et al. (1989). Release of oxytocin within the supraoptic nucleus during the milk ejection reflex in rats. Exp Brain Res 76: 593–602. Morris JF, Pow DV (1991). Widespread release of peptides in the central nervous system: quantitation of tannic acidcaptured exocytoses. Anat Rec 231: 437–445. Morton A (1969). A quantitative analysis of the normal neuron population of the hypothalamic magnocellular nuclei in man and of their projections to the neurohypophysis. J Comp Neurol 136: 143–157. Motojima Y, Kawasaki M, Matsuura T et al. (2016). Effects of peripherally administered cholecystokinin-8 and secretin on feeding/drinking and oxytocin-mRFP1 fluorescence in transgenic rats. Neurosci Res 109: 63–69. Neumann ID (2007). Stimuli and consequences of dendritic release of oxytocin within the brain. Biochem Soc Trans 35: 1252–1257. Neumann ID (2008). Brain oxytocin: a key regulator of emotional and social behaviours in both females and males. J Neuroendocrinol 20: 858–865. Neumann I, Landgraf R (1989). Septal and hippocampal release of oxytocin, but not vasopressin, in the conscious lactating rat during suckling. J Neuroendocrinol 1: 305–308. Neumann ID, Landgraf R (2012). Balance of brain oxytocin and vasopressin: implications for anxiety, depression, and social behaviors. Trends Neurosci 35: 649–659. Neumann ID, Slattery DA (2016). Oxytocin in general anxiety and social fear: a translational approach. Biol Psychiatry 79: 213–221. Neumann I, Ludwig M, Engelmann M et al. (1993a). Simultaneous microdialysis in blood and brain: oxytocin and vasopressin release in response to central and peripheral osmotic stimulation and suckling in the rat. Neuroendocrinology 58: 637–645. Neumann I, Russell JA, Landgraf R (1993b). Oxytocin and vasopressin release within the supraoptic and paraventricular nuclei of pregnant, parturient and lactating rats: a microdialysis study. Neuroscience 53: 65–75. Newmaster KT, Nolan ZT, Chon U et al. (2020). Quantitative cellular-resolution map of the oxytocin receptor in postnatally developing mouse brains. Nat Commun 11: 1885. Nishimori K, Young LJ, Guo Q et al. (1996). Oxytocin is required for nursing but is not essential for parturition or reproductive behavior. Proc Natl Acad Sci U S A 93: 11699–11704. Nishioka T, Anselmo-Franci JA, Li P et al. (1998). Stress increases oxytocin release within the hypothalamic paraventricular nucleus. Brain Res 781: 57–61. Nyuyki KD, Waldherr M, Baeuml S et al. (2011). Yes, I am ready now: differential effects of paced versus unpaced mating on anxiety and central oxytocin release in female rats. PLoS One 6: e23599. Oettl LL, Ravi N, Schneider M et al. (2016). Oxytocin enhances social recognition by modulating cortical control of early olfactory processing. Neuron 90: 609–621.

Ozsoy S, Esel E, Kula M (2009). Serum oxytocin levels in patients with depression and the effects of gender and antidepressant treatment. Psychiatry Res 169: 249–252. Palanisamy A, Kannappan R, Xu Z et al. (2018). Oxytocin alters cell fate selection of rat neural progenitor cells in vitro. PLoS One 13: e0191160. Persoon CM, Hoogstraaten RI, Nassal JP et al. (2019). The RAB3-RIM pathway is essential for the release of neuromodulators. Neuron 104: 1065–1080 e12. Peters JH, McDougall SJ, Kellett DO et al. (2008). Oxytocin enhances cranial visceral afferent synaptic transmission to the solitary tract nucleus. J Neurosci 28: 11731–11740. Petersson M (2002). Cardiovascular effects of oxytocin. Prog Brain Res 139: 281–288. Petrovic P, Kalisch R, Singer T et al. (2008). Oxytocin attenuates affective evaluations of conditioned faces and amygdala activity. J Neurosci 28: 6607–6615. Quintana DS, Rokicki J, Van Der Meer D et al. (2017). Genetic networks of the oxytocin system in the human brain: a gene expression and large-scale fMRI meta-analysis study. bioRxiv (pre-print). Ragen BJ, Bales KL (2013). Oxytocin and vasopressin in nonhuman primates. In: E Choleris, DW Pfaff, M Kavaliers (Eds.), Oxytocin, vasopressin and related peptides in the regulation of behavior. Cambridge University Press. Rash JA, Aguirre-Camacho A, Campbell TS (2014). Oxytocin and pain: a systematic review and synthesis of findings. Clin J Pain 30: 453–462. Rhodes CH, Morrell JI, Pfaff DW (1981). Immunohistochemical analysis of magnocellular elements in rat hypothalamus: distribution and numbers of cells containing neurophysin, oxytocin, and vasopressin. J Comp Neurol 198: 45–64. Romanov RA, Zeisel A, Bakker J et al. (2017). Molecular interrogation of hypothalamic organization reveals distinct dopamine neuronal subtypes. Nat Neurosci 20: 176–188. Rosen GJ, De Vries GJ, Goldman SL et al. (2008). Distribution of oxytocin in the brain of a eusocial rodent. Neuroscience 155: 809–817. Ross HE, Cole CD, Smith Y et al. (2009). Characterization of the oxytocin system regulating affiliative behavior in female prairie voles. Neuroscience 162: 892–903. Sabatier N, Leng G, Menzies J (2013). Oxytocin, feeding, and satiety. Front Endocrinol (Lausanne) 4: 35. Sabihi S, Durosko NE, Dong SM et al. (2014). Oxytocin in the prelimbic medial prefrontal cortex reduces anxiety-like behavior in female and male rats. Psychoneuroendocrinology 45: 31–42. Shahrokh DK, Zhang TY, Diorio J et al. (2010). Oxytocindopamine interactions mediate variations in maternal behavior in the rat. Endocrinology 151: 2276–2286. Share L, Crofton JT (1993). Interactions between the gonadal steroid hormones and vasopressin and oxytocin. Ann N Y Acad Sci 689: 438–454. Smith AS, Williams Avram SK, Cymerblit-Sabba A et al. (2016). Targeted activation of the hippocampal CA2 area strongly enhances social memory. Mol Psychiatry 21: 1137–1144. Sofroniew MV (1980). Projections from vasopressin, oxytocin, and neurophysin neurons to neural targets in the rat and human. J Histochem Cytochem 28: 475–478.

CENTRAL AND PERIPHERAL RELEASE OF OXYTOCIN Son SJ, Filosa JA, Potapenko ES et al. (2013). Dendritic peptide release mediates interpopulation crosstalk between neurosecretory and preautonomic networks. Neuron 78: 1036–1049. Stoop R (2012). Neuromodulation by oxytocin and vasopressin. Neuron 76: 142–159. Swaab DF (1997). Prader-Willi syndrome and the hypothalamus. Acta Paediatr Suppl 423: 50–54. Swaab DF, Purba JS, Hofman MA (1995). Alterations in the hypothalamic paraventricular nucleus and its oxytocin neurons (putative satiety cells) in Prader-Willi syndrome: a study of five cases. J Clin Endocrinol Metab 80: 573–579. Swanson LW, Kuypers HG (1980). The paraventricular nucleus of the hypothalamus: cytoarchitectonic subdivisions and organization of projections to the pituitary, dorsal vagal complex, and spinal cord as demonstrated by retrograde fluorescence double-labeling methods. J Comp Neurol 194: 555–570. Swanson LW, Sawchenko PE (1983). Hypothalamic integration: organization of the paraventricular and supraoptic nuclei. Annu Rev Neurosci 6: 269–324. Szot P, Dorsa DM (1993). Differential timing and sexual dimorphism in the expression of the vasopressin gene in the developing rat brain. Brain Res Dev Brain Res 73: 177–183. Tang Y, Benusiglio D, Lefevre A et al. (2020). Social touch promotes inter-female communication via oxytocin parvocellular neurons. Nat Neurosci 23: 1125–1137. (in press). Tay TL, Mai D, Dautzenberg J et al. (2017). A new fate mapping system reveals context-dependent random or clonal expansion of microglia. Nat Neurosci 20: 793–803. Theodosis DT (1985). Oxytocin-immunoreactive terminals synapse on oxytocin neurones in the supraoptic nucleus. Nature 313: 682–684. Tirko NN, Eyring KW, Carcea I et al. (2018). Oxytocin transforms firing mode of CA2 hippocampal neurons. Neuron 100: 593–608 e3. Tobin VA, Hashimoto H, Wacker DW et al. (2010). An intrinsic vasopressin system in the olfactory bulb is involved in social recognition. Nature 464: 413–417. Tomizawa K, Iga N, Lu YF et al. (2003). Oxytocin improves long-lasting spatial memory during motherhood through MAP kinase cascade. Nat Neurosci 6: 384–390. Torner L, Plotsky PM, Neumann ID et al. (2016). Forced swimming-induced oxytocin release into blood and brain: effects of adrenalectomy and corticosterone treatment. Psychoneuroendocrinology 77: 165–174. Tribollet E, Barberis C, Arsenijevic Y (1997). Distribution of vasopressin and oxytocin receptors in the rat spinal cord: sex-related differences and effect of castration in pudendal motor nuclei. Neuroscience 78: 499–509. Tsuji T, Allchorne AJ, Zhang M et al. (2017). Vasopressin casts light on the suprachiasmatic nucleus. J Physiol 595: 3497–3514. Ueta Y, Fujihara H, Serino R et al. (2005). Transgenic expression of enhanced green fluorescent protein enables direct visualization for physiological studies of vasopressin neurons and isolated nerve terminals of the rat. Endocrinology 146: 406–413.

43

Valassi E, Scacchi M, Cavagnini F (2008). Neuroendocrine control of food intake. Nutr Metab Cardiovasc Dis 18: 158–168. Van Den Burg EH, Stindl J, Grund T et al. (2015). Oxytocin stimulates extracellular Ca influx through TRPV2 channels in hypothalamic neurons to exert its anxiolytic effects. Neuropsychopharmacology 40: 2938–2947. Van Den Pol AN (2012). Neuropeptide transmission in brain circuits. Neuron 76: 98–115. Van Der Woude PF, Goudsmit E, Wierda M et al. (1995). No vasopressin cell loss in the human hypothalamus in aging and Alzheimer’s disease. Neurobiol Aging 16: 11–18. Veenema AH, Neumann ID (2008). Central vasopressin and oxytocin release: regulation of complex social behaviours. Prog Brain Res 170: 261–276. Venkatesh B, Si-Hoe SL, Murphy D et al. (1997). Transgenic rats reveal functional conservation of regulatory controls between the Fugu isotocin and rat oxytocin genes. Proc Natl Acad Sci U S A 94: 12462–12466. Verbalis JG, Stricker EM, Robinson AG et al. (1991). Cholecystokinin activates C-fos expression in hypothalamic oxytocin and corticotropin-releasing hormone neurons. J Neuroendocrinol 3: 205–213. Voorn P, Buijs RM (1983). An immuno-electronmicroscopical study comparing vasopressin, oxytocin, substance P and enkephalin containing nerve terminals in the nucleus of the solitary tract of the rat. Brain Res 270: 169–173. Wacker D, Ludwig M (2019). The role of vasopressin in olfactory and visual processing. Cell Tissue Res 375: 201–215. Wahis J, Baudon A, Althammer F et al. (2021). Astrocytes mediate the effect of oxytocin in the central amygdala on neuronal activity and affective states in rodents. Nat Neurosci 24: 529–541. Waldherr M, Neumann ID (2007). Centrally released oxytocin mediates mating-induced anxiolysis in male rats. Proc Natl Acad Sci U S A 104: 16681–16684. Wang YF, Hatton GI (2009). Astrocytic plasticity and patterned oxytocin neuronal activity: dynamic interactions. J Neurosci 29: 1743–1754. Wang Z, Bullock NA, De Vries GJ (1993). Sexual differentiation of vasopressin projections of the bed nucleus of the stria terminals and medial amygdaloid nucleus in rats. Endocrinology 132: 2299–2306. Wang P, Qin D, Wang YF (2017). Oxytocin rapidly changes astrocytic GFAP plasticity by differentially modulating the expressions of pERK 1/2 and protein kinase A. Front Mol Neurosci 10: 262. Wierda M, Goudsmit E, Van Der Woude PF et al. (1991). Oxytocin cell number in the human paraventricular nucleus remains constant with aging and in Alzheimer’s disease. Neurobiol Aging 12: 511–516. Wigger A, Neumann ID (2002). Endogenous opioid regulation of stress-induced oxytocin release within the hypothalamic paraventricular nucleus is reversed in late pregnancy: a microdialysis study. Neuroscience 112: 121–129.

44

F. ALTHAMMER ET AL.

Wotjak CT, Kubota M, Liebsch G et al. (1996). Release of vasopressin within the rat paraventricular nucleus in response to emotional stress: a novel mechanism of regulating adrenocorticotropic hormone secretion? J Neurosci 16: 7725–7732. Wotjak CT, Ganster J, Kohl G et al. (1998). Dissociated central and peripheral release of vasopressin, but not oxytocin, in response to repeated swim stress: new insights into the secretory capacities of peptidergic neurons. Neuroscience 85: 1209–1222. Wrobel LJ, Reymond-Marron I, Dupre A et al. (2010). Oxytocin and vasopressin enhance synaptic transmission in the hypoglossal motor nucleus of young rats by acting on distinct receptor types. Neuroscience 165: 723–735. Xiao L, Priest MF, Nasenbeny J et al. (2017). Biased oxytocinergic modulation of midbrain dopamine systems. Neuron 95: 368–384.e5. Yang HP, Wang L, Han L et al. (2013). Nonsocial functions of hypothalamic oxytocin. ISRN Neurosci 2013: 179272.

Yoshida M, Takayanagi Y, Inoue K et al. (2009). Evidence that oxytocin exerts anxiolytic effects via oxytocin receptor expressed in serotonergic neurons in mice. J Neurosci 29: 2259–2271. Young WS 3rd, Shepard E, Amico J et al. (1996). Deficiency in mouse oxytocin prevents milk ejection, but not fertility or parturition. J Neuroendocrinol 8: 847–853. Young WS 3rd, Iacangelo A, Luo XZ et al. (1999). Transgenic expression of green fluorescent protein in mouse oxytocin neurones. J Neuroendocrinol 11: 935–939. Yuan L, Liu S, Bai X et al. (2016). Oxytocin inhibits lipopolysaccharide-induced inflammation in microglial cells and attenuates microglial activation in lipopolysaccharidetreated mice. J Neuroinflammation 13: 77. Zhang H, Gross J, De Dreu C et al. (2019). Oxytocin promotes coordinated out-group attack during intergroup conflict in humans. Elife 8. Zhou L, Blaustein JD, De Vries GJ (1994). Distribution of androgen receptor immunoreactivity in vasopressin- and oxytocin-immunoreactive neurons in the male rat brain. Endocrinology 134: 2622–2627.

Handbook of Clinical Neurology, Vol. 180 (3rd series) The Human Hypothalamus: Middle and Posterior Region D.F. Swaab, F. Kreier, P.J. Lucassen, A. Salehi, and R.M. Buijs, Editors https://doi.org/10.1016/B978-0-12-820107-7.00004-5 Copyright © 2021 Elsevier B.V. All rights reserved

Chapter 4

Organization of the neuroendocrine and autonomic hypothalamic paraventricular nucleus ANDRIES KALSBEEK1,2* AND RUUD M. BUIJS3 1

Department of Endocrinology and Metabolism, Amsterdam University Medical Centers (Amsterdam UMC), University of Amsterdam, Amsterdam, The Netherlands 2

Department of Hypothalamic Integration Mechanisms, Netherlands Institute for Neuroscience, an Institute of the Royal Netherlands Academy of Arts and Sciences, Amsterdam, The Netherlands

3

Hypothalamic Integration Mechanisms Laboratory, Department of Cellular Biology and Physiology, Instituto de Investigaciones Biom
edicas, Universidad Nacional Autónoma de M
exico (UNAM), Ciudad de M
exico, Mexico

Abstract A major function of the nervous system is to maintain a relatively constant internal environment. The distinction between our external environment (i.e., the environment that we live in and that is subject to major changes, such as temperature, humidity, and food availability) and our internal environment (i.e., the environment formed by the fluids surrounding our bodily tissues and that has a very stable composition) was pointed out in 1878 by Claude Bernard (1814–1878). Later on, it was indicated by Walter Cannon (1871–1945) that the internal environment is not really constant, but rather shows limited variability. Cannon named the mechanism maintaining this limited variability homeostasis. Claude Bernard envisioned that, for optimal health, all physiologic processes in the body needed to maintain homeostasis and should be in perfect harmony with each other. This is illustrated by the fact that, for instance, during the sleep–wake cycle important elements of our physiology such as body temperature, circulating glucose, and cortisol levels show important variations but are in perfect synchrony with each other. These variations are driven by the biologic clock in interaction with hypothalamic target areas, among which is the paraventricular nucleus of the hypothalamus (PVN), a core brain structure that controls the neuroendocrine and autonomic nervous systems and thus is key for integrating central and peripheral information and implementing homeostasis. This chapter focuses on the anatomic connections between the biologic clock and the PVN to modulate homeostasis according to the daily sleep–wake rhythm. Experimental studies have revealed a highly specialized organization of the connections between the clock neurons and neuroendocrine system as well as preautonomic neurons in the PVN. These complex connections ensure a logical coordination between behavioral, endocrine, and metabolic functions that helps the organism maintain homeostasis throughout the day.

INTRODUCTION1 The autonomic nervous system (ANS) and the neuroendocrine system together are responsible for organizing homeostasis. Traditionally the ANS is divided into two

main branches, the sympathetic (SNS) and parasympathetic (PNS) nervous systems. Homeostasis is dependent on the dynamic balance between these two branches. Indeed, the sympathovagal or autonomic balance is

1

Abbreviations used in the chapter are listed at the end of the chapter before References section.

*Correspondence to: Andries Kalsbeek, Department of Endocrinology and Metabolism, Amsterdam UMC, University of Amsterdam, Meibergdreef 9, 1105 AZ Amsterdam, The Netherlands. Tel: +31-20-5665500, E-mail: [email protected]

46

A. KALSBEEK AND R.M. BUIJS

considered a major determinant of metabolic health. Within the hypothalamus and in fact within the whole central nervous system (CNS), the paraventricular nucleus of the hypothalamus (PVN) is the most important site for autonomic and endocrine homeostasis and integration. Although the PVN makes up only about 1% of the rodent brain (and much less than 1% of the human brain), it can play such an important role since it has direct connections with the posterior lobe of the pituitary gland, as well as indirect connections with the anterior lobe via the median eminence. In addition, it has connections with brain sites controlling the outputs to the sympathetic and parasympathetic branches of the autonomic nervous system. Ultimately, all these autonomic and endocrine outputs are synchronized with behavior. Based on structure–function criteria, PVN neurons are divided into the large magnocellular neuroendocrine and the smaller parvocellular neurons. Most neurons in the magnocellular division project to the posterior lobe of the pituitary gland and produce either vasopressin or oxytocin as a neurohormone. The neurons in the parvocellular division give rise to projections to either the median eminence (i.e., neuroendocrine neurons) or the brainstem and spinal cord (i.e., the preautonomic neurons). However, whereas the magnocellular and parvocellular neurons can be clearly distinguished anatomically, at least in rats, the neuroendocrine and preautonomic parvocellular neurons cannot be determined so easily. Thus far, these two types of parvocellular neurons can only be discriminated diffusely based on their localization within the PVN, but no morphologic, neurochemical, or molecular characteristics are known that differentiate the neuroendocrine from the preautonomic neurons; this holds true even more for markers to differentiate the sympathetic and parasympathetic preautonomic neurons. Despite the fact that the parvocellular neurons can contain a large variety of neuropeptides, such as vasopressin, oxytocin, thyrotrophin-releasing hormone (TRH), corticotropin-releasing hormone (CRH), somatostatin, neurotensin, substance-P, growth hormone-releasing hormone (GHrH), galanin, dynorphin, enkephalin, or CCK, none of these has been found to be specific for one of the subtypes (Simmons and Swanson, 2009). A clear example of such a “lack of specificity” is the neuropeptide vasopressin. When released from the magnocellular neurons in the systemic circulation, via the posterior pituitary, vasopressin functions as a neurohormone. On the other hand, it serves as a releasing factor when released from the neuroendocrine parvocellular neurons that project to the median eminence and stimulates the release of adrenocorticotrophic hormone (ACTH) from the anterior pituitary, but it serves as a neurotransmitter/neuromodulator when released from preautonomic parvocellular neurons that project to the brainstem or spinal cord. In a recent study, Zhang et al. (2020) further subdivided

vasopressin neurons based on the coexpression of glutamatergic or GABAergic traits. The neuroendocrine parvocellular neurons of the hypophysiotropic system usually are subdivided further, based on their involvement in the control of hormone release from the anterior pituitary, i.e., CRH and AVP neurons control the hypothalamo–pituitary–adrenal (HPA) axis, TRH neurons control the hypothalamo–pituitary–thyroid (HPT) axis and prolactin release, and GHrH-, somatostatin-, and dopamine-containing neurons control the release of growth hormone and prolactin. However, no such molecular markers for a subdivision of the preautonomic SNS and PNS neurons are known. In early studies, it was suggested that single preautonomic neurons influence both branches of the ANS (Swanson and Kuypers, 1980). The finding of similar neuropeptides, such as oxytocin, orexin, MCH, and AVP, in both SNS and PNS preautonomic neurons (Jansen et al., 1997; Buijs et al., 2001) seemed to confirm this idea. However, using simultaneous injections of two different retrograde tracers in the dorsal motor nucleus and spinal cord, Portillo et al. (1998) observed no colocalization and proposed that SNS and PNS preautonomic neurons represent separate populations. Indeed, using a combination of transneuronal tracing and selective denervation of autonomic nerves, it was shown that individual preautonomic neurons project to only one autonomic branch, thus exhibiting a (complete) separation of presympathetic and preparasympathetic neurons (Buijs et al., 2003). So, for now the only possibility to differentiate the neuroendocrine from the preautonomic neurons anatomically, next to electrophysiologic recordings (Tasker and Dudek, 1991; Hermes and Renaud, 1993; Stern, 2001; Price et al., 2009; Feetham et al., 2018), is injection of a retrograde tracer in the spinal cord and/or systemic circulation. Using transsynaptic tracing studies similar to those discussed already, we have shown previously that within the hypothalamus and PVN an astonishing capacity exists with respect to the specialization of the preautonomic neurons. For example, separate neurons are present that project to either the visceral or subcutaneous fat depots (Kreier et al., 2002, 2006). On the other hand, other studies using tracing from two tissues at the same time have shown that single-labeled as well as different combinations of double-labeled neurons can be found (Buijs et al., 2003; Stanley et al., 2010; Nguyen et al., 2017; Wiedmann et al., 2017; Doslikova et al., 2019). These results indicate that preautonomic neurons can be either organ-specific or have a more general, shared function. In the PVN, these double-labeled neurons, sometimes called “central command” neurons (Jansen et al., 1995), were shown to express oxytocin or CRF, but not CART or vasopressin (Stanley et al., 2010). Like most neuropeptidergic neurons, PVN neurons not only colocalize GABA or glutamate as a fast

ORGANIZATION OF THE NEUROENDOCRINE AND AUTONOMIC PVN neurotransmitter, but also one or more other neuropeptides, with examples of up to 10 different neuropeptides (Lee et al., 2013). Further complexity is accomplished by the fact that the proportion and pattern of colocalization may differ depending on the behavioral, endocrine, physiologic, or clinical state (Kiss, 1998). Recent single-cell RNA sequencing studies are now confirming these older data by distinguishing at least 35–60 different neuronal subtypes in the hypothalamus, based on their neuropeptide and/or neurotransmitter content (Romanov et al., 2016; Campbell et al., 2017; Chen et al., 2017; Mickelsen et al., 2017, 2019; Moffitt et al., 2018). These studies, for instance, recognized at least five subtypes of CRH neurons (Romanov et al., 2017) and four subtypes of oxytocin and dopamine neurons (Romanov et al., 2016). Although no scRNA-seq study has yet specifically focused on the PVN, from the preceding it seems likely that a large variety of neuronal subtypes can also be distinguished in the PVN. However, it is still unknown how these subtypes relate to hypothalamic output. For instance, most PVN neurons projecting to the rostral ventrolateral medulla (RVLM) were shown to contain oxytocin, but multiple peptide colocalization patterns were found and that same study showed that the “peptide coding” of PVN-RVLM neurons differed between PVN subdivisions (Lee et al., 2013). A similar observation was made for PVN neurons projecting to the spinal cord (Hallbeck and Blomqvist, 1999; Hallbeck et al., 2001). Recently, Romanov et al. (2016) identified a subclass of PVN dopamine neurons that uniquely expressed the Onecut3 and Nmur2 genes, receiving inputs from the neuromedin S+ neurons in the SCN and sending projections to the median eminence. Most likely, this specific group of dopamine neurons is involved in the control of the daily pattern of prolactin release. In conclusion, anatomic data support a high level of differentiation and specificity in the PVN neurons. As for the neuroendocrine parvocellular neurons, the mere existence of different hypothalamo-pituitary axes provides clear evidence that this anatomic differentiation also has functional consequences. In the next paragraph, we will see functional evidence that also supports a high level of specialization for the preautonomic neurons. It is to be expected that, by combining scRNA-seq with neuroanatomic tracing techniques in future experiments, unique markers will also be discovered for these functionally segregated subclasses of preautonomic neurons.

HYPOTHALAMIC INTEGRATION OF TIME OF DAY According to the original concept of homeostasis, physiologic variables are regulated to remain close to a certain predefined value, in order to be constant over time. However, there is no single ideal set of steady-state conditions

47

in life and different circumstances ask for different set points, a process referred to as allostasis (Goldstein and McEwen, 2002). Circadian rhythms provide a clear example that homeostasis is not about regulation around a fixed set point, but according to a value set for optimal functioning, in this case appropriate for the time of day. To join these two processes Mrosovsky (1990) coined the term rheostasis. The regular 24-h rotation of the earth has led to the evolution of autonomous circadian (i.e., from the Latin words circa and dies, meaning about 1 day) clocks in virtually all life-forms, from prokaryotes to eukaryotes. This circadian clock mechanism serves to coordinate and anticipate behavior and metabolism according to the predictable environmental periodicity induced by the earth’s rotation. In mammals, including humans, the central pacemaker for this circadian clock system is located in the brain, in the suprachiasmatic nuclei (SCN) of the anterior hypothalamus. The SCN consists of several clusters of small and densely packed neurons in which various peptidergic transmitters are expressed (Van Den Pol and Tsujimoto, 1985; Moore, 1996). Synchronization of the
24-h rhythms generated by the SCN to the exact 24-h rhythms in the environment occurs via a direct projection from the retina. Subsequently, the SCN uses its efferent projections to different combinations of intermediate, neuroendocrine, and preautonomic neurons in the hypothalamus to translate its circadian activity into daily rhythms in, among other things, sleep–wake and feeding behavior, hormone release, body temperature, and energy metabolism (Buijs and Kalsbeek, 2001). Information on the distribution of SCN projections was initially obtained from neuroanatomic studies using tracing, immunocytochemistry, SCN lesions, or a combination of these methods (Hoorneman and Buijs, 1982; Watts and Swanson, 1987; Kalsbeek et al., 1993). All these studies showed that the outflow of SCN information was in fact surprisingly limited and pertained to the medial hypothalamus, in particular to target areas that contain mainly interneurons, such as the medial preoptic area (MPOA), the dorsomedial nucleus of the hypothalamus (DMH) and the subPVN. Direct connections to both neuroendocrine (Van Der Beek et al., 1993, 1997; De La Iglesia et al., 1995; Vrang et al., 1995; Hermes et al., 1996; Kalsbeek et al., 2000a) and preautonomic neurons (Teclemariam-Mesbah et al., 1997; Vrang et al., 1997) have been reported, but turned out to be scarce. In later experiments, the preautonomic connections were confirmed using the retrograde transsynaptic virus-tracing technique (Larsen et al., 1998; Buijs et al., 1999, 2001; Teclemariam-Mesbah et al., 1999; Ueyama et al., 1999; La Fleur et al., 2000; Rosario et al., 2016). Tracing studies in postmortem brain tissue have shown amazingly similar projections arising from the SCN in the human brain (Dai et al., 1997, 1998).

48

A. KALSBEEK AND R.M. BUIJS

Our animal studies on how the rhythmic information generated by the brain’s master clock is integrated with the rest of the brain, especially the homeostatic output generated by the PVN, provided clear evidence for a high level of differentiation and specificity of the preautonomic neurons in the PVN.

CONTROL OF THE DAILY RHYTHM IN HPA AXIS ACTIVITY Under baseline conditions, plasma concentrations of the glucocorticoid hormones released from the cortex of the adrenal gland vary predictably across the day/night cycle. Both in nocturnal and diurnal species, plasma corticosterone (cortisol in humans) concentrations are highest around the time of arousal (i.e., early morning for humans and evening for rats, mice, and hamsters). Adrenal glucocorticoid hormones have highly integrated effects on both energy metabolism and behavior. It is thought that the increased levels of corticosterone at awakening act to enable foraging behavior by increasing the amount of available energy. CRH is the principal neural signal controlling the activity of the HPA axis by stimulating the release of ACTH from the anterior pituitary and ACTH, in its turn, controls the release of corticosterone through its stimulatory action on the melanocortin receptor type 2 (MC2R) in the adrenal cortex (Watts, 2005). The proximity of SCN nerve endings near CRH-containing neurons in the PVN gave rise to the hypothesis that, via this projection, circadian information would be enforced onto the HPA axis (Buijs et al., 1993; Vrang et al., 1995). Indeed, we were able to demonstrate that VP released from SCN terminals has a strong inhibitory control over basal plasma corticosterone concentrations (Kalsbeek et al., 1992). Further studies on the relation between the SCN release of VP and the daily rhythm in HPA axis activity revealed that the arrest of SCN VP release at the end of the light period is important for enabling the daily corticosterone surge before awakening (Kalsbeek et al., 1996b,c). Although, at first sight, the CRH-containing neurons in the PVN appeared to be the most likely target neurons for the daily modulation of HPA axis activity, several pieces of evidence were inconsistent with such a role for CRH neurons. First, a direct effect of VP on the CRH neuron would imply a clear daily rhythm in plasma ACTH concentrations, but this was not observed (Buijs et al., 1997). Second, the observed inhibitory effect of VP was not in line with the usual excitatory effect of VP on its target neurons. Third, contrary to the expected abundant contacts between SCN-derived VP fibers and CRH neurons, only a limited number of such connections was found (Vrang et al., 1995; Buijs and Van Eden, 2000). Electrophysiologic in vitro experiments in hypothalamic

slices by Hermes et al. (2000) indicated an intermediate role for gamma-aminobutyric acid (GABA) containing neurons in the subPVN and DMH. A detailed anatomic scheme incorporating all of these results and explaining our current view on the SCN control of the daily rhythm in HPA activity is shown in Fig. 4.1. Considering that the phase of SCN activity (including VP release) is similar in nocturnal and diurnal species (Hofman and Swaab, 1994; Dardente et al., 2004; Cuesta et al., 2009), the involvement of intermediate areas also provides a nice explanation for the 12-h shift in the rhythm in HPA axis activity between nocturnal and diurnal species (Kalsbeek et al., 2008b). However, the anatomic scheme presented in Fig. 4.1 does not explain the mismatch between the daily ACTH and corticosterone rhythms: i.e., despite the clear rhythm in plasma corticosterone, often no clear daily rhythm in plasma ACTH is found. This caused us to propose that the ANS might be involved in regulating the daily changes in the sensitivity of the adrenal cortex for ACTH. Indeed, it has been shown that MC2R expression exhibits a daily rhythm in the adrenal (Kalsbeek et al., 2012; Park et al., 2013; Roa et al., 2017). Transneuronal virus tracing from the adrenal indeed revealed secondorder labeling in PVN neurons and third-order labeling in SCN neurons (Buijs et al., 1999). The functional importance of this multisynaptic neural connection between the SCN and the adrenal cortex for the daily rhythm in adrenal corticosterone release was proven by an elegant series of adrenal microdialysis, adrenal denervation, and adrenal transplantation studies (Jasper and Engeland, 1994; Ishida et al., 2005; Oster et al., 2006). Later, Horacio de la Iglesia and coworkers provided additional evidence for the two-stage control of the circadian corticosterone rhythm using the splitting and forceddesynchrony models (Lilley et al., 2011; Wotus et al., 2013). One of the last remaining questions concerns which mechanisms and transmitters are involved in this neural control of corticosterone release from the adrenal cortex. For instance, we still do not know the SCN neurotransmitter responsible for stimulation of corticosterone release at the beginning of the activity period. Neither do we know whether vasopressin inhibits corticosterone release both via an effect on the HPA axis and the adrenal innervation. Finally, we do not know which preautonomic neurons are involved in the control of adrenal sensitivity for ACTH. However, we do know that those preautonomic neurons that project to the adrenal do not express glucocorticoid receptors (GR) (Leon-Mercado et al., 2017), even though CRH neurons are well-known to contain glucocorticoid receptors (Liposits et al., 1987; Wang et al., 2013). This raises the question: if variation in ACTH is not essential for corticosterone secretion, how is the negative feedback for corticosterone

ORGANIZATION OF THE NEUROENDOCRINE AND AUTONOMIC PVN

49

Fig. 4.1. Detailed anatomic scheme of demonstrated and putative connections* of the suprachiasmatic nucleus (SCN) in the nocturnal rat and the diurnal Arvicanthis ansorgei brain to explain the opposite effects of arginine vasopressin (AVP) on the hypothalamic– pituitary–adrenal axis in these two species. AVP is released during the light period, both in the nocturnal rat and the diurnal A. ansorgei. In rats, AVP release during the light period inhibits the corticotropin-releasing hormone (CRH)-containing neurons in the paraventricular nucleus of the hypothalamus (PVN) by contacting gamma-aminobutyric acid (GABA)ergic interneurons in the subPVN and dorsomedial nucleus of the hypothalamus (DMH). On the other hand, in A. ansorgei, AVP release during the light period stimulates CRH-containing neurons because it acts on the glutamatergic (GLU), instead of GABAergic, interneurons in the subPVN and DMH. *Putative connections mainly concern those indicated in Arvicanthis ansorgei. For instance, the glutamatergic projections from the subPVN and DMH in Arvicanthis have not been demonstrated. The vasopressinergic projections from the SCN to the subPVN and DMH have been demonstrated, but it has not been shown whether they contact CRH neurons. The latter is also true for the rat, although here tracing studies have shown direct projections from the SCN to the CRH neurons. From Kalsbeek A, Verhagen LA, Schalij I et al. (2008b). Opposite actions of hypothalamic vasopressin on circadian corticosterone rhythm in nocturnal vs diurnal species. Eur J Neurosci 27: 818–827.

regulated? This question was resolved recently by microdialysis experiments in the arcuate nucleus, demonstrating that by using the limited presence of the blood–brain barrier at the median eminence, corticosterone can penetrate the arcuate area relatively rapidly and act on the GR receptors present in that area. Via their projections to the autonomic neurons in the PVN, these arcuate nucleus neurons were shown to be essential for the fast-negative feedback of corticosterone to its own release (Leon-Mercado et al., 2017). In conclusion, the hypothalamic control of the daily rhythm in plasma corticosterone provides a clear example of the highly integrated function of the PVN and the twostage mechanism used by the biologic clock to control the activity of the HPA axis. At the first stage, the SCN modulates the release of ACTH via the neuroendocrine CRH neurons in the medial parvocellular region of the PVN. In the second stage, the SCN controls the sensitivity of the MC2R for the hormonal signal that it is controlling

(i.e., ACTH), via preautonomic neurons in the dorsal and ventral aspect of the parvocellular region of the PVN. The circadian control of thyroid hormone secretion provides another example of such a two-stage control.

CONTROL OF THE DAILY RHYTHM IN HPT AXIS ACTIVITY The most active thyroid hormone T3 is a result of the deiodination of the prohormone thyroxin (T4) secreted from the thyroid gland under the influence of thyroidstimulating hormone (TSH) originating from the anterior pituitary. TSH secretion shows a clear circadian rhythm with a peak during the sleep phase (Ikegami et al., 2019). The release of TSH in turn is controlled by the stimulatory action of the neuropeptide TRH released from PVN terminals in the median eminence, and the inhibitory action of dopamine and somatostatin, in addition to the negative hypothalamic and pituitary feedback action of

50

A. KALSBEEK AND R.M. BUIJS

the thyroid hormones T4 and T3. Surprisingly, TSH shows a clear daily rhythm, while T4, which is a direct result of the stimulatory action of TSH on the pituitary, is hardly rhythmic (Greenspan et al., 1986). This lack of synchrony between the TSH rhythmicity and T4 may reflect molecular change in the TSH molecule, such as glycosylation, with reduced bioactivity during the night period (Ikegami and Yoshimura, 2017). On the other hand, altered sensitivity of the thyroid gland for TSH during the day/night cycle might be an alternative explanation, as suggested by tracing studies (see the following). However, neurologically complete cervical spinal injury (between C4 and C7) in humans did not disrupt the daily rhythmicity of TSH (or cortisol) secretion, whereas it did cause a complete loss of the plasma melatonin rhythm (Zeitzer et al., 2000). Contrary to the circadian response of the adrenal to ACTH, whereby the ACTH-corticosterone ratio is always low, the TSH– T4 ratio favors the importance of TSH much more strongly, and the TSH rhythm is much more prominent than the rhythm in T4. This indicates more complex mechanisms behind the circadian control of the daily rhythm in thyroid hormones. In addition, circulating concentrations of T4 are about 10 times those of T3 levels, indicating that just a small modulation of the deiodination of T4 to T3 is sufficient to induce a daily rhythm in T3 levels. Indeed, daily rhythms in deiodinase activity have been demonstrated in different tissues (Murakami et al., 1988; Zandieh Doulabi et al., 2004; Kalsbeek et al., 2005; Koenig, 2005). Further research is needed to fully explain the daily rhythm in HPT-axis activity and the feedback mechanisms of thyroid hormones to limit the secretion of TRH. Interestingly, in the feedback of thyroid hormones the arcuate nucleus plays a major role: T4 may pass the median eminence–arcuate nucleus barrier either by diffusion or by active transport via tanycytes. At the border of the third ventricle, T4 is deiodinated to T3 by ependymal cells, which promotes the negative feedback to TRH secretion (Riskind et al., 1987). Neuroanatomic tracing and lesion studies in rats have provided relatively little extra information on the relationship between the biologic clock and thyroid hormone metabolism. Using immunocytochemistry, SCN fibers were seen to contact TRH neurons in the PVN, a connection that may form the anatomic basis for the daily rhythms in hypothalamic TRH mRNA content and plasma TSH (Collu et al., 1977; Martino et al., 1985; Covarrubias et al., 1988, 1994). Secondly, neuroanatomic studies using the retrograde transneuronal viral tracer PRV revealed multisynaptic neural connections between the hypothalamic SCN and the thyroid gland via sympathetic and parasympathetic outflow. In addition, preautonomic neurons in the PVN, including TRH immunoreactive neurons, were labeled after injection

of the PRV tracer into the thyroid gland (Kalsbeek et al., 2000a). These experiments illustrate that, just as for the adrenal, the PVN-driven autonomic control of the thyroid gland may play an important role for the control of its hormone secretion. Frequent blood sampling via permanent cannulas revealed daily rhythms of TSH and thyroid hormones and thermic ablation of the SCN completely eliminated the diurnal peak in circulating TSH and thyroid hormones, providing functional evidence that the SCN drives these diurnal variations (Kalsbeek et al., 2000a). However, targeted hypothalamic infusions of SCN neurotransmitter agonists or antagonists, which had been so helpful in the previously described studies on the HPA axis, thus far have not disclosed any information on how SCN signals are involved in the control of the daily HPT rhythm.

CONTROL OF THE DAILY RHYTHM IN PROLACTIN RELEASE Prolactin is another hormone of the pituitary that is not only involved in a whole variety of functions, but also shows a strong daily rhythm. In females as well as in males, prolactin in the circulation increases just before the active period. Oxytocin is known to stimulate prolactin release and oxytocin receptors in the anterior pituitary are upregulated and become more sensitive by elevated estradiol levels during the preovulatory phase of the estrous cycle (Breton et al., 1995; Kennett et al., 2008). Dopamine released from neurons in the periventricular nucleus and from tubero-infundibular and tubero-hypophysial neurons in the arcuate nucleus is well-known to inhibit prolactin release from the lactotrophs in the anterior pituitary via binding to dopamine D2 receptors. The proestrous afternoon prolactin surge is likely a result of the rising levels of estradiol, which decrease the inhibitory influence of dopamine and at the same time promote the stimulatory effect of oxytocin (Bertram et al., 2010). The daily timing signal for the prolactin rhythm seems to come mainly from vasoactive intestinal peptide (VIP)-producing neurons in the SCN, as they inhibit dopamine neurons in the arcuate nucleus in a circadian fashion (Gerhold et al., 2001, 2002; Egli et al., 2004; Bertram et al., 2006). Moreover, the oxytocin neurons in the PVN are contacted by the VIP neurons in the SCN (Egli et al., 2004; Kennett et al., 2008). Finally, TRH neurons in the PVN could be involved in the rhythmic control of prolactin release as well, as SCN projections reach TRH neurons (Kalsbeek et al., 2000a) and TRH can function as a prolactin-releasing factor (Fr€ohlich and Wahl, 2019). Recently, Romanov et al. (2016) revealed a rhythmic population of dopaminergic neurons in the periventricular nucleus that projects to the median eminence and receives input from neuromedin S (NMS)-containing

ORGANIZATION OF THE NEUROENDOCRINE AND AUTONOMIC PVN neurons in the SCN. These SCN contacts are in agreement with previous results from tracing studies (Horvath, 1997; Abizaid et al., 2004). Intracerebroventricular administration of neuromedin U (NMU), a sister peptide of NMS that acts on the same receptors, inhibits prolactin release (Nakahara et al., 2019). Remarkably, neuromedin U precursor-related peptide (NURP), a peptide derived from the neuromedin U precursor, has a stimulatory effect on prolactin release (Mori et al., 2017). Finally, vasopressin release from the SCN may also be involved in the control of the daily prolactin rhythm, as the decreased release of vasopressin from SCN terminals at the end of the light period seems to allow for a further rise of the prolactin surge (Palm et al., 2001).

CONTROL OF THE DAILY RHYTHM IN PINEAL MELATONIN RELEASE The rhythmic release of melatonin from the pineal gland during the dark phase is a prime example of circadian control through the ANS. Early experiments involving extirpation of the superior cervical ganglion (SCG) and transection of the spinal cord established the functional importance of the sympathetic innervation (Wurtman et al., 1967; Klein et al., 1971; Axelrod, 1974; Kneisley et al., 1978; Moore, 1978; Reiter et al., 1982; Bowers et al., 1984). The first SCN lesion studies quickly proved the indispensability of the SCN for the daily rhythmicity of melatonin synthesis (Moore and Klein, 1974; Bittman et al., 1989; Tessonneaud et al., 1995), but the brain structures connecting SCN and spinal cord remained unclear for quite some time. Only when the first histochemical studies identified SCN projections to the PVN (Swanson and Cowan, 1975; Berk and Finkelstein, 1981; Stephan et al., 1981) and PVN projections to the spinal cord (Buijs, 1978; Swanson and Kuypers, 1980) was the PVN identified as an important relay for SCN output to the spinal cord (Klein et al., 1983). The key importance of the PVN as a target area for SCN information to control the melatonin rhythm was corroborated by a number of subsequent neurotoxin and knife-cut studies (Pickard and Turek, 1983; Lehman et al., 1984; Nunez et al., 1985; Hastings and Herbert, 1986; Badura et al., 1989; Bittman et al., 1989; Johnson et al., 1989; Smale et al., 1989) and by studies involving electrical stimulation of the PVN (Reuss et al., 1985; Olcese et al., 1987; Yanovski et al., 1987). However, it was only at the closing of the 20th century that the retrograde transsynaptic virus-tracing technique made it possible to map out the entire pathway between the SCN and the pineal (Larsen et al., 1998; Teclemariam-Mesbah et al., 1999). Although the viral tracing studies helped to clearly define the total neuronal pathway, the specific role of the PVN in the control of the melatonin synthesis rhythm remained to be determined.

51

Using microinfusion experiments, we were able to demonstrate that the daily rhythm in plasma melatonin concentrations was generated by a combination of glutamatergic and GABAergic SCN projections to the PVN (Kalsbeek et al., 1996a, 1999, 2000b; Perreau-Lenz et al., 2003, 2004). The preautonomic PVN neurons that are at the origin of the sympathetic innervation of the pineal are the prime targets of these SCN projections. The daytime activity of the GABAergic SCN projections to the PVN ensures low melatonin levels during the light period. The nocturnal arrest of these inhibitory GABAergic inputs, combined with a continuously active glutamatergic input from the SCN, causes the preautonomic PVN neurons controlling the sympathetic input to the pineal gland to become active and initiate the synthesis and release of melatonin during the dark period (Fig. 4.2). Therefore, as proposed for the control of the different hypothalamo-pituitary axes, the SCN also uses multiple outputs for the control of melatonin synthesis. Our ideas on the combined inhibitory and stimulatory outputs of the SCN agree with studies indicating that GABA and glutamate also function as inhibitory and stimulatory SCN outputs, respectively, to regions of the preoptic area involved in the control of the sleep–wake rhythm (Sun et al., 2000, 2001). In addition, evidence of glutamate immunoreactivity within presynaptic boutons in the PVN (Van Den Pol, 1991), as well as the demonstration of a specific glutamate release from the SCN onto (preautonomic) PVN neurons (Hermes et al., 1996; Csaki et al., 2000; Cui et al., 2001), shores up the idea of a glutamatergic SCN input to the PVN. In sum, the daily rhythm of plasma melatonin concentration is generated by a combination of stimulatory and inhibitory SCN outputs. The preautonomic PVN neurons that are in charge of the sympathetic input to the pineal gland are controlled by a combination of glutamatergic and GABAergic inputs from the SCN. The circadian and light-induced activity of the GABAergic SCN projections to the PVN ensures low melatonin levels during the light period, or when the light is switched on at night. The nocturnal arrest of the inhibitory GABAergic inputs, combined with the continuously active glutamatergic inputs, enables the preautonomic PVN to become active again and start a new period of melatonin synthesis and release.

CONTROL OF THE DAILY RHYTHM IN PLASMA GLUCOSE CONCENTRATIONS Based on the preceding findings, especially those for the adrenal and pineal gland, we hypothesized that an important part of the action of the SCN to prepare our bodies for the alternating periods of sleep and wakefulness would be through its connections with the hypothalamic

52

A. KALSBEEK AND R.M. BUIJS

Fig. 4.2. Schematic presentation of the daily activity pattern of suprachiasmatic (SCN) populations of GABAergic and glutamatergic (GLU) neurons implicated in the autonomic control of the daily rhythm in pineal melatonin release. The continuous excitatory input to the sympathetic preautonomic neurons in the PVN from the glutamatergic SCN neurons only results in an actual activation of the preautonomic PVN neuron when the GABAergic inhibition from the SCN is absent. The GABAergic SCN neurons thus function like a traffic light, permitting the stimulatory input to the preautonomic neuron to become “visible” or “noticeable” only when the GABAergic neurons allow. From Kalsbeek A, Fliers E (2013). Daily regulation of hormone profiles. In: A Kramer, M Merrow (Eds.), Circadian clocks. Handbook of experimental pharmacology, vol. 217, with permission from Springer Nature.

preautonomic neurons and control of the sympathetic– parasympathetic autonomic balance. Indeed, a series of retrograde viral tracing studies from tissues involved in energy metabolism, such as brown and white adipose tissue, pancreas, stomach, heart, and intestines, revealed many second-order preautonomic neurons in the PVN (Bartness et al., 2001; Buijs et al., 2001; Scheer et al., 2001; Kreier et al., 2006; Rosario et al., 2016). Moreover, by combining viral tracing with selective denervations, we were able to show a clear separation of the preautonomic neurons controlling the sympathetic and parasympathetic branch of the ANS (La Fleur et al., 2000; Buijs et al., 2001; Kalsbeek et al., 2004). The liver plays a pivotal role in maintaining optimum glucose levels by balancing glucose entry into and removal from the circulation, with a clear involvement of the PVN influencing both the sympathetic and parasympathetic input to the liver (Shimazu, 1987; Nonogaki, 2000; Puschel, 2004), which is also under strong circadian control (Akhtar et al., 2002; Kita et al., 2002; Oishi et al., 2002). Using local PVN administration of GABA and glutamate receptor (ant)agonists, we demonstrated that the daily rhythm in plasma glucose concentrations is controlled very similarly to the mechanism described earlier for the SCN control of the daily rhythm in melatonin release (Fig. 4.3), i.e., by a combination of rhythmic GABAergic inputs and continuous glutamatergic stimulation onto

liver-dedicated sympathetic preautonomic neurons in the PVN (Kalsbeek et al., 2004, 2008a). Further experiments showed that administration of bicuculline (a GABA-A receptor antagonist) in the perifornical area lateral to the DMH caused a pronounced increase in hepatic glucose production, and that orexincontaining (but not melanin-concentrating hormone (MCH)-containing) neurons in this area were strongly activated (Yi et al., 2009). In view of the pronounced day/night rhythm in orexin release (Zeitzer et al., 2003; Zhang et al., 2004; Alam et al., 2005) and the fact that orexin fibers impinge upon sympathetic preganglionic neurons in the spinal cord that project to the liver (Van den Top et al., 2003), we hypothesized that orexin is the main connection between the biologic clock and the daily rhythm in plasma glucose concentrations. The perifornical orexin neurons seem to transduce the rhythmic GABA and glutamatergic signals emanating from the SCN into a daily activation of the sympathetic input to the liver, which results in an increased hepatic glucose production at the end of the sleep period, in anticipation of a new period of wakefulness (Fig. 4.4). Remarkably, a recent study by Shiuchi et al. (2009) demonstrated that orexin is able to stimulate glucose uptake in muscle via the ventromedial nucleus of the hypothalamus (VMH) and the sympathetic nervous system. Thus orexin might be an important link in the SCN controlled concomitant increase of glucose production and glucose uptake at

Fig. 4.3. Schematic presentation of the daily activity pattern of suprachiasmatic (SCN) populations of GABAergic and glutamatergic (GLU) neurons implicated in the autonomic control of the daily rhythms in pineal melatonin release and hepatic glucose production. For the control of the daily rhythms in melatonin release and glucose production, the SCN seems to rely on a uniform mechanism of continuous glutamatergic and rhythmic GABAergic inputs to the sympathetic preautonomic neurons. The difference in the timing of the acrophase for the melatonin and glucose production, however, indicates that separate populations of GABAergic neurons should be in contact with the pineal-dedicated and liver-dedicated preautonomic neurons, i.e., separate traffic lights for the pineal and the liver. From Kalsbeek A, Yi CX, Cailotto C et al. (2011). Mammalian clock output mechanisms. Essays Biochem 49: 137–151.

PF

SCN GABA GLU Orexin MCH

IML

Glucose, 6-meal

Liver

Fig. 4.4. Midsagittal view of the rat brain with a hypothesized presentation of the involvement of orexin neurons in the autonomic control of the daily rhythm of hepatic glucose production. (i) The orexin-containing neurons in the perifornical area (PF) are innervated by both glutamatergic and GABAergic projections from the biologic clock (SCN). During the main part of the light period, activation of the orexin neurons by the excitatory glutamatergic inputs is prevented by releasing the inhibitory neurotransmitter GABA (the daily activity pattern of these inputs is indicated by the lines in the blue boxes beside the projections). The circadian withdrawal of the GABAergic input allows the orexin neurons to become active at the onset of darkness. (ii) Subsequently, the excitatory effect of orexin on the preganglionic neurons in the intermediolateral column (IML) of the spinal cord will (iii) activate the sympathetic input to the liver and result in increased hepatic glucose production. Orexin also stimulates glucose uptake in skeletal muscle via action in the ventromedial nucleus of the hypothalamus (VMH) and mediated through the sympathetic nervous system (Shiuchi et al., 2009); but, as it is not clear yet how the message is propagated from the VMH to the autonomic nervous system, this action has not been incorporated in this scheme. From Kalsbeek A, Yi CX, La Fleur SE et al. (2010). The hypothalamic clock and its control of glucose homeostasis. Trends Endocrinol Metab 21: 402–410.

54

A. KALSBEEK AND R.M. BUIJS

the onset of the activity period (La Fleur, 2003). Together, these results indicate that, due to a disinhibition of the orexin system at the end of the light period, the SCN not only promotes arousal, but also causes an increase of endogenous glucose production and peripheral glucose uptake to ensure adequate glucose availability for brain and muscle when the animal wakes up. The major difference between the liver-dedicated and pineal-dedicated preautonomic neurons seems to be the timing of the GABAergic inputs. In the case of the pineal-dedicated preautonomic neurons, this inhibitory input is present during the whole light period with a peak in the middle of the light period, whereas for the liverdedicated preautonomic orexin neurons, the peak of GABAergic inhibition is closer to the beginning of the light period.

CONTROL OF THE DAILY RHYTHM IN FEEDING-INDUCED INSULIN RESPONSES As has become evident from the daily variation in mealinduced insulin responses, intestinal glucose uptake, respiratory functioning, and markers of cardiac vagal activity (Burgess et al., 1997; Kalsbeek and Strubbe,

1998; Hilton et al., 2000; Scheer et al., 2004; Houghton et al., 2006; Bando et al., 2007), the activity of the parasympathetic branch of the ANS is also modulated by the circadian timing system. Using intrahypothalamic infusions, we were able to show that the daily changes in the activity of the parasympathetic preautonomic neurons also involve a combination of GABAergic and glutamatergic inputs (Kalsbeek et al., 2008a). The inhibition of preautonomic neurons, both sympathetic and parasympathetic, by a daily rhythm in GABA release from SCN efferents to the PVN turned out to be a general principle. However, a major difference between the circadian control of parasympathetic and sympathetic preautonomic neurons appears to be the origin of the excitatory glutamatergic inputs. SCN lesion studies proved not only that the excitatory input to the sympathetic pineal-dedicated preautonomic neurons was derived from the SCN neurons (Perreau-Lenz et al., 2003), but also that the glutamatergic inputs to the parasympathetic pancreas-dedicated preautonomic neurons cannot be derived from the SCN (Strubbe et al., 1987). At present, it is not clear from which extra-SCN source the glutamatergic inputs to the parasympathetic pancreasdedicated preautonomic neurons originate, but the most likely candidate is the VMH (Fig. 4.5).

Fig. 4.5. Schematic presentation of the daily activity pattern of hypothalamic populations of GABAergic and glutamatergic (GLU) neurons implicated in the autonomic control of the daily rhythms in hepatic glucose production (left-hand side) and feeding-induced insulin release (right-hand side). Similar to the previously proposed circadian control of the sympathetic preautonomic neurons (lefthand side), the circadian control of the parasympathetic preautonomic neurons also seems to rely on a combination of glutamatergic and GABAergic inputs (right-hand side). However, whereas for both types of neurons the rhythmic GABAergic input is derived from the SCN, the sources of glutamatergic input seem to be different, i.e., SCN for the sympathetic preautonomic neurons and extra-SCN for the parasympathetic ones. Moreover, whereas the glutamatergic input from the SCN to the sympathetic preautonomic neurons is proposed to be continuous, the glutamatergic input from the VMH to the parasympathetic preautonomic neurons is proposed to be dependent on feeding activity. In the figure, the ventromedial nucleus of the hypothalamus (VMH) is indicated as the most likely origin of the glutamatergic input, but at present experimental evidence is lacking for this proposition. From Kalsbeek A, Foppen E, Schalij I et al. (2008a). Circadian control of the daily plasma glucose rhythm: an interplay of GABA and glutamate. PLoS One 3: e3194.

ORGANIZATION OF THE NEUROENDOCRINE AND AUTONOMIC PVN

MULTIPLE SUBPOPULATIONS OF PREAUTONOMIC NEURONS The preceding findings clearly indicate that, in addition to the different populations of neuroendocrine neurons that control the hypothalamo-pituitary axes, the PVN also harbors different populations of preautonomic neurons. In addition to sympathetic and parasympathetic preautonomic neurons, within these two groups of preautonomic neurons multiple subpopulations of preautonomic neurons exist that are dedicated to a specific organ or tissue. Tracing studies have indicated that separate PVN neurons may be involved in the control of visceral and subcutaneous adipose tissue (Kreier et al., 2002, 2006), but also liver and adrenal (Buijs et al., 2003), skeletal muscle and adipose tissue (Doslikova et al., 2019), white and brown adipose tissue (Nguyen et al., 2017; Wiedmann et al., 2017), and liver and adipose tissue (Stanley et al., 2010). On the other hand, these studies also showed many double-labeled neurons, i.e., preautonomic neurons that are connected to more than one organ. As clearly illustrated in Fig. 4.6, the latter is also in agreement with our findings on the SCN control of the preautonomic neurons: whereas the (timing of the) GABAergic input may differ between tissues, (the timing of ) their glutamatergic inputs

55

may be similar. In Fig. 4.6, the GABAergic input to the liver and adrenal also shows a similar timing; however, this might involve largely separated populations of preautonomic neurons. Forced-desynchrony experiments in humans have shown that the timing of the glucose and cortisol peaks can be separated, with the glucose peak following feeding time and the cortisol peak staying entrained to the L/D cycle (Scheer et al., 2009). Restricted-feeding experiments in animals provide similar results, although, depending on the level of food deprivation, a second corticosterone peak may occur just before the food becomes available (Honma et al., 1992; Ritter et al., 2003). Experiments using electrophysiologic and bioluminescence techniques have clearly shown that the SCN consists of multiple neuronal subpopulations with different timings of their circadian periodicity. These subpopulations are mutually synchronized to form separate regional pacemakers, which further couple to produce a coherent 24-h rhythmic output (Meijer et al., 2012; Honma, 2018). It is not clear how many separate pacemakers the SCN is composed of, but usually at least a morning and evening oscillator as well as a midday and midnight oscillator are recognized. These four oscillators would also be sufficient to generate the different rhythms described earlier. The morning oscillator would

Fig. 4.6. Schematic presentation of the daily activity pattern of suprachiasmatic (SCN) populations of GABAergic and glutamatergic (GLU) neurons implicated in the autonomic control of the daily rhythms in pineal melatonin release, hepatic glucose production, and adrenal corticosterone release. For all these rhythms, the SCN seems to rely on a uniform mechanism of continuous glutamatergic and rhythmic GABAergic inputs to the sympathetic preautonomic neurons. The difference in the timing of the acrophase for glucose production and corticosterone release on the one hand, and melatonin release on the other hand, indicates that at least separate populations of GABAergic neurons should be in contact with the liver- and adrenal-dedicated preautonomic vs the pineal-dedicated neurons. Whether separate SCN and PVN neurons are also responsible for the liver and adrenal is not clear when looking at the timing of their rhythmic activity. On one hand, viral tracing simultaneously from those two organs did not show a complete overlap of labeled neurons (Buijs et al., 2003). On the other hand, restricted-feeding experiments have shown that the timing of the glucose and cortisol peaks can be separated, with the glucose peak following feeding time and the cortisol peak staying entrained to the L/D cycle.

56

A. KALSBEEK AND R.M. BUIJS

be involved in the inhibition of the HPA axis and prolactin rhythm, as well as glucose production. The evening oscillator then would be responsible for the stimulation of HPA axis activity, the prolactin surge, and the increased glucose production. The midday and midnight oscillators would be responsible for the trough and acrophase of the melatonin rhythms, as well as the daytime and nighttime rise in plasma TSH. Together these data clearly indicate that the autonomic balance is not just a matter of preautonomic neurons being more or less active. First, the complete separation of sympathetic and parasympathetic preautonomic neurons shows that the two sides of this balance can be controlled separately and independently. Secondly, the autonomic balance may differ per body compartment and organ. For instance, during metabolic syndrome, the autonomic balance has shifted in favor of the sympathetic branch in the cardiovascular system, but in the abdominal compartment, parasympathetic activity prevails (Kreier et al., 2003). Thirdly, the tracing and SCN control experiments show that within the two ANS branches the autonomic output can be differentiated at the organ and tissue level. It is not clear how “detailed” this differentiation of autonomic output is, but considering the fact that in the cortex so-called face-recognition neurons have been identified (Quiroga et al., 2005; Suthana et al., 2015), a high level of differentiation seems possible and likely. For instance, it is likely that, with regard to the autonomic control of the liver, separate neurons project to hepatocytes and hepatic blood vessels.

FUTURE PERSPECTIVE Although nearly all experimental details in this chapter deal with results from animal studies, the presented findings have, in our opinion, a clear relevance for human physiology and pathology. There is considerable evidence that the organization of the human autonomic nervous system is highly comparable with that of the rat in which we have done all our studies. An imbalance of the sympathetic and parasympathetic branches of the autonomic nervous system also predicts the development of metabolic syndrome in humans (Lieb et al., 2012; Licht et al., 2013; Sabath et al., 2015; Wulsin et al., 2016). Important risk factors for an autonomic imbalance are psychosocial factors such as stressful life events and work stress, including circadian stress from shift work and jet lag, hostility, depression, and anxiety. Thus it is essential to understand how stress affects the autonomic balance. Unfortunately, for now the autonomic balance can only be measured at the peripheral level, both in humans and animals. Moreover, usually the autonomic balance is only determined in the cardiovascular system, although autonomic (dis)balance may differ per body

compartment and organ. Therefore it is essential to develop techniques that are able to determine the autonomic balance in individual organs, preferably in vivo. In animals, it should be possible to analyze the first and primary determinants of the autonomic balance, i.e., the preautonomic neurons in the hypothalamus in the brain. In order to be able to determine the autonomic (dis)balance more precisely, it is essential to identify the different sympathetic and parasympathetic preautonomic neurons. However, currently these neurons can only be identified using a combination of tracing, denervation, and/or electrophysiologic techniques; no neurochemical or molecular markers are known that clearly separate the neuroendocrine and preautonomic neurons in the hypothalamus, let alone markers that can differentiate the sympathetic and parasympathetic preautonomic neurons or organ-specific preautonomic neurons within the sympathetic or parasympathetic branch. Therefore in future experiments separate (molecular) markers for the sympathetic and parasympathetic preautonomic neurons should be identified, for instance by using a combination of retrograde tracing and single-cell sequencing. Such (molecular) markers to identify the different preautonomic neurons are a conditio sine qua non to understand how environmental (stress) factors affect the autonomic (dis)balance. Once these unique (combinations of ) markers have been identified, it might even become possible to modulate specific populations of sympathetic and parasympathetic preautonomic neurons to, for instance, correct an autonomic imbalance. Available data indicate that the K-opioid receptor is expressed in only one out of five CRH-positive neuronal cell types (Romanov et al., 2017). If such a unique marker could be linked to the CRH neurons involved in the control of hypertension (Goncharuk et al., 2002), this might provide a drug target to treat hypertension. In addition, there are a number of studies providing evidence that PVN ion channel changes may underlie disease conditions, such as diabetes (Gao et al., 2017), hypertension (Pachuau et al., 2014; Li et al., 2014), heart failure (Pyner, 2014; Wang et al., 2019) and depression (Gao et al., 2013). Some of these channels might emerge as possible therapeutic targets from our studies. Another immediate advantage of the availability of such markers would be to greatly benefit research involving the use of human brain tissue, as at present there is no possibility of differentiating preautonomic and neuroendocrine neurons in the human hypothalamus. Having markers to differentiate the neuroendocrine and PNS and SNS preautonomic neurons in human postmortem hypothalamic tissue thus would help to translate findings from animal models to the human brain and would provide an enormous advantage in understanding hypothalamic

ORGANIZATION OF THE NEUROENDOCRINE AND AUTONOMIC PVN pathologies such as type 2 diabetes and hypertension (Goncharuk et al., 2002; Hogenboom et al., 2019).

ABBREVIATIONS ACTH, adrenocorticotrophic hormone; ANS, autonomic nervous system; AVP, arginine vasopressin; CNS, central nervous system; CRH, corticotrophin-releasing hormone; DMH, dorsomedial nucleus of the hypothalamus; GABA, gamma-aminobutyric acid; HPA, hypothalamo– pituitary–adrenal; HPT, hypothalamo–pituitary–thyroid; MPOA, medial preoptic area; PF, perifornical area; PRV, pseudorabies virus; PVN, paraventricular nucleus of the hypothalamus; SCG, superior cervical ganglion; SCN, suprachiasmatic nucleus; subPVN, subparaventricular PVN; T3, triiodothyronine; T4, thyroxine; TRH, thyrotrophin-releasing hormone; TSH, thyroidstimulating hormone; VIP, vasoactive intestinal polypeptide; VMH, ventromedial nucleus of the hypothalamus; VP, vasopressin; WAT, white adipose tissue.

REFERENCES Abizaid A, Horvath B, Keefe DL et al. (2004). Direct visual and circadian pathways target neuroendocrine cells in primates. Eur J Neurosci 20 (10): 2767–2776. Akhtar RA, Reddy AB, Maywood ES et al. (2002). Circadian cycling of the mouse liver transcriptome, as revealed by cDNA microarray, is driven by the suprachiasmatic nucleus. Curr Biol 12: 540–550. Alam MN, Kumar S, Bashir T et al. (2005). GABA-mediated control of hypocretin- but not melanin-concentrating hormone-immunoreactive neurones during sleep in rats. J Physiol 563: 569–582. Axelrod J (1974). The pineal gland: a neurochemical transducer. Science 184: 1341–1348. Badura LL, Kelly KK, Nunez AA (1989). Knife cuts lateral but not dorsal to the hypothalamic paraventricular nucleus abolish gonadal responses to photoperiod in female hamsters (Mesocricetus auratus). J Biol Rhythms 4: 79–91. Bando H, Nishio T, van der Horst GT et al. (2007). Vagal regulation of respiratory clocks in mice. J Neurosci 27: 4359–4365. Bartness TJ, Song CK, Demas GE (2001). SCN efferents to peripheral tissues: implications for biological rhythms. J Biol Rhythms 16: 196–204. Berk ML, Finkelstein JA (1981). An autoradiographic determination of the efferent projections of the suprachiasmatic nucleus of the hypothalamus. Brain Res 226: 1–13. Bertram R, Egli M, Toporikova N et al. (2006). A mathematical model for the mating-induced prolactin rhythm of female rats. Am J Physiol Endocrinol Metab 290 (3): E573–E582. Bertram R, Helena CV, Gonzalez-Iglesias AE et al. (2010). A tale of two rhythms: the emerging roles of oxytocin in

57

rhythmic prolactin release. J Neuroendocrinol 22 (7): 778–784. Bittman EL, Crandell RG, Lehman MN (1989). Influences of the paraventricular and suprachiasmatic nuclei and olfactory bulbs on melatonin responses in the golden hamster. Biol Reprod 40: 118–126. Bowers CW, Baldwin C, Zigmond RE (1984). Sympathetic reinnervation of the pineal gland after postganglionic nerve lesion does not restore normal pineal function. J Neurosci 4: 2010–2015. Breton C, Pechoux C, Morel G et al. (1995). Oxytocin receptor messenger ribonucleic acid: characterization, regulation, and cellular localization in the rat pituitary gland. Endocrinology 136 (7): 2928–2936. Buijs RM (1978). Intra- and extrahypothalamic vasopressin and oxytocin pathways in the rat. Pathways to the limbic system, medulla oblongata and spinal cord. Cell Tissue Res 192: 423–435. Buijs RM, Kalsbeek A (2001). Hypothalamic integration of central and peripheral clocks. Nat Neurosci Rev 2: 521–526. Buijs RM, Van Eden CG (2000). The integration of stress by the hypothalamus, amygdale and prefrontal cortex: balance between the autonomic nervous system and the neuroendocrine system. Prog Brain Res 126: 117–132. Buijs RM, Markman M, Nunes-Cardoso B et al. (1993). Projections of the suprachiasmatic nucleus to stressrelated areas in the rat hypothalamus: a light and electron microscopic study. J Comp Neurol 335 (1): 42–54. Buijs RM, Wortel J, Van Heerikhuize JJ et al. (1997). Novel environment induced inhibition of corticosterone secretion: physiological evidence for a suprachiasmatic nucleus mediated neuronal hypothalamo-adrenal cortex pathway. Brain Res 758 (1–2): 229–236. Buijs RM, Wortel J, Van Heerikhuize JJ et al. (1999). Anatomical and functional demonstration of a multisynaptic suprachiasmatic nucleus adrenal (cortex) pathway. Eur J Neurosci 11: 1535–1544. Buijs RM, Chun SJ, Niijima A et al. (2001). Parasympathetic and sympathetic control of the pancreas: a role for the suprachiasmatic nucleus and other hypothalamic centers that are involved in the regulation of food intake. J Comp Neurol 431: 405–423. Buijs RM, la Fleur SE, Wortel J et al. (2003). The suprachiasmatic nucleus balances sympathetic and parasympathetic output to peripheral organs through separate preautonomic neurons. J Comp Neurol 464: 36–48. Burgess HJ, Trinder J, Kim Y et al. (1997). Sleep and circadian influences on cardiac autonomic nervous system activity. Am J Physiol 273: H1761–H1768. Campbell JN, Macosko EZ, Fenselau H et al. (2017). A molecular census of arcuate hypothalamus and median eminence cell types. Nat Neurosci 20: 484–496. Chen R, Wu X, Jiang L et al. (2017). Single-cell RNA-Seq reveals hypothalamic cell diversity. Cell Rep 18: 3227–3241. Collu R, Du Ruisseau P, Tach
e Y et al. (1977). Thyrotropinreleasing hormone in rat brain: nyctohemeral variations. Endocrinology 100: 1391–1393.

58

A. KALSBEEK AND R.M. BUIJS

Covarrubias L, Uribe RM, Mendez M et al. (1988). Neuronal TRH synthesis: developmental and circadian TRH mRNA levels. Biochem Biophys Res Comm 151: 615–622. Covarrubias L, Redondo JL, Vargas MA et al. (1994). In vitro TRH release from hypothalamus slices varies during the diurnal cycle. Neurochem Res 19: 845–850. Csaki A, Kocsis K, Halasz B et al. (2000). Localization of glutamatergic/aspartatergic neurons projecting to the hypothalamic paraventricular nucleus studied by retrograde transport of [3H] D-aspartate autoradiography. Neuroscience 101: 637–655. Cuesta M, Clesse D, P
evet P et al. (2009). From daily behavior to hormonal and neurotransmitters rhythms: comparison between diurnal and nocturnal rat species. Horm Behav 55: 338–347. Cui LN, Coderre E, Renaud LP (2001). Glutamate and GABA mediate suprachiasmatic nucleus inputs to spinalprojecting paraventricular neurons. Am J Physiol 281: R1283–R1289. Dai J, Swaab DF, Buijs RM (1997). Distribution of vasopressin and vasoactive intestinal polypeptide (VIP) fibers in the human hypothalamus with special emphasis on suprachiasmatic nucleus efferent projections. J Comp Neurol 383: 397–414. Dai J, Swaab DF, Van der Vliet J et al. (1998). Postmortem tracing reveals the organization of hypothalamic projections of the suprachiasmatic nucleus in the human brain. J Comp Neurol 400: 87–102. Dardente H, Menet JS, Challet E et al. (2004). Daily and circadian expression of neuropeptides in the suprachiasmatic nuclei of nocturnal and diurnal rodents. Mol Brain Res 124: 143–151. De La Iglesia HO, Blaustein JD, Bittman EL (1995). The suprachiasmatic area in the female hamster projects to neurons containing estrogen receptors and GnRH. Neuroreport 6: 1715–1722. Doslikova B, Tchir D, McKinty A et al. (2019). Convergent neuronal projections from paraventricular nucleus, parabrachial nucleus, and brainstem onto gastrocnemius muscle, white and brown adipose tissue in male rats. J Comp Neurol 527: 2826–2842. Egli M, Bertram R, Sellix MT et al. (2004). Rhythmic secretion of prolactin in rats: action of oxytocin coordinated by vasoactive intestinal polypeptide of suprachiasmatic nucleus origin. Endocrinology 45 (7): 3386–3394. Feetham CH, O’Brien F, Barrett-Jolley R (2018). Ion channels in the paraventricular hypothalamic nucleus (PVN); emerging diversity and functional roles. Front Physiol 9: 760. Fr€ ohlich E, Wahl R (2019). The forgotten effects of thyrotropin-releasing hormone: metabolic functions and medical applications. Front Neuroendocrinol 52: 29–43. Gao SF, Klomp A, Wu JL et al. (2013). Reduced GAD(65/67) immunoreactivity in the hypothalamic paraventricular nucleus in depression: a postmortem study. J Affect Disord 149: 422–425. PMID: 23312397. Gao H, Molinas AJR, Miyata K et al. (2017). Overactivity of liver-related neurons in the paraventricular nucleus of the hypothalamus: electrophysiological findings in db/db mice. J Neurosci 37: 11140–11150. PMID: 29038244.

Gerhold LM, Horvath TL, Freeman ME (2001). Vasoactive intestinal peptide fibers innervate neuroendocrine dopaminergic neurons. Brain Res 919 (1): 48–56. Gerhold LM, Sellix MT, Freeman ME (2002). Antagonism of vasoactive intestinal peptide mRNA in the suprachiasmatic nucleus disrupts the rhythm of FRAs expression in neuroendocrine dopaminergic neurons. J Comp Neurol 450 (2): 135–143. Goldstein DS, McEwen B (2002). Allostasis, homeostats, and the nature of stress. Stress 5: 55–58. Goncharuk VD, Van Heerikhuize J, Swaab DF et al. (2002). Paraventricular nucleus of the human hypothalamus in primary hypertension: activation of corticotropin-releasing hormone neurons. J Comp Neurol 443: 321–331. PMID: 11807841. Greenspan SL, Klibansk A, Schoenfeld D et al. (1986). Pulsatile secretion of thyrotropin in man. J Clin Endocrinol Metab 63: 661–668. Hallbeck M, Blomqvist A (1999). Spinal cord-projecting vasopressinergic neurons in the rat paraventricular hypothalamus. J Comp Neurol 411 (2): 201–211. Hallbeck M, Larhammar D, Blomqvist A (2001). Neuropeptide expression in rat paraventricular hypothalamic neurons that project to the spinal cord. J Comp Neurol 433 (2): 222–238. Hastings MH, Herbert J (1986). Neurotoxic lesions of the paraventriculo-spinal projection block the nocturnal rise in pineal melatonin synthesis in the Syrian hamster. Neurosci Lett 69: 1–6. Hermes ML, Renaud LP (1993). Differential responses of identified rat hypothalamic paraventricular neurons to suprachiasmatic nucleus stimulation. Neuroscience 56: 823–832. Hermes MLHJ, Coderre EM, Buijs RM et al. (1996). GABA and glutamate mediate rapid neurotransmission from suprachiasmatic nucleus to hypothalamic paraventricular nucleus in the rat. J Physiol 496: 749–757. Hermes MLHJ, Ruijter JM, Klop A et al. (2000). Vasopressin increases GABAergic inhibition of rat hypothalamic paraventricular nucleus neurons in vitro. J Neurophysiol 83: 705–711. Hilton MF, Umali MU, Czeisler CA et al. (2000). Endogenous circadian control of the human autonomic nervous system. Comput Cardiol 27: 197–200. Hofman MA, Swaab DF (1994). Alterations in circadian rhythmicity of the vasopressin- producing neurons of the human suprachiasmatic nucleus (SCN) with aging. Brain Res 651: 134–142. Hogenboom R, Kalsbeek MJ, Korpel NL et al. (2019). Loss of arginine-vasopressin neurons and glial cells in the suprachiasmatic nuclei of type 2 diabetes patients. Diabetologia 62: 2088–2093. PMID: 31327049. Honma S (2018). The mammalian circadian system: a hierarchical multi-oscillator structure for generating circadian rhythm. J Physiol Sci 68: 207–219. https://doi.org/10.1007/ s12576-018-0597-5. Honma K-I, Noe Y, Honma S et al. (1992). Roles of paraventricular catecholamines in feeding-associated corticosterone rhythm in rats. Am J Physiol 262: E948–E955.

ORGANIZATION OF THE NEUROENDOCRINE AND AUTONOMIC PVN Hoorneman EMD, Buijs RM (1982). Vasopressin fiber pathways in the rat brain following suprachiasmatic nucleus lesioning. Brain Res 243: 235–241. Horvath TL (1997). Suprachiasmatic efferents avoid phenestrated capillaries but innervate neuroendocrine cells, including those producing dopamine. Endocrinology 138 (3): 1312–1320. Houghton SG, Zarroug AE, Duenes JA et al. (2006). The diurnal periodicity of hexose transporter mRNA and protein levels in the rat jejunum: role of vagal innervation. Surgery 139: 542–549. Ikegami K, Yoshimura T (2017). The hypothalamic-pituitarythyroid axis and biological rhythms: the discovery of TSH’s unexpected role using animal models. Best Pract Res Clin Endocrinol Metab 31: 475–485. Ikegami K, Refetoff S, Van Cauter E et al. (2019). Interconnection between circadian clocks and thyroid function. Nat Rev Endocrinol 15: 590–600. Ishida A, Mutoh T, Ueyama T et al. (2005). Light activates the adrenal gland: timing of gene expression and glucocorticoid release. Cell Metab 2: 297–307. Jansen AS, Nguyen XV, Karpitskiy V et al. (1995). Central command neurons of the sympathetic nervous system: basis of the fight-or-flight response. Science 270 (5236): 644–646. Jansen AS, Hoffman JL, Loewy AD (1997). CNS sites involved in sympathetic and parasympathetic control of the pancreas: a viral tracing study. Brain Res 766: 29–38. Jasper MS, Engeland WC (1994). Splanchnic neural activity modulates ultradian and circadian rhythms in adrenocortical secretion in awake rats. Neuroendocrinology 59: 97–109. Johnson RF, Smale L, Moore RY et al. (1989). Paraventricular nucleus efferents mediating photoperiodism in male golden hamsters. Neurosci Lett 98: 85–90. Kalsbeek A, Strubbe JH (1998). Circadian control of insulin secretion is independent of the temporal distribution of feeding. Physiol Behav 63: 553–560. Kalsbeek A, Buijs RM, Van Heerikhuize JJ et al. (1992). Vasopressin-containing neurons of the suprachiasmatic nuclei inhibit corticosterone release. Brain Res 580: 62–67. Kalsbeek A, Teclemariam-Mesbah R, P
evet P (1993). Efferent projections of the suprachiasmatic nucleus in the golden hamster (Mesocricetus auratus). J Comp Neurol 332: 293–314. Kalsbeek A, Drijfhout WJ, Westerink BHC et al. (1996a). GABA receptors in the region of the dorsomedial hypothalamus of rats are implicated in the control of melatonin. Neuroendocrinology 63: 69–78. Kalsbeek A, Van Der Vliet J, Buijs RM (1996b). Decrease of endogenous vasopressin release necessary for expression of the circadian rise in plasma corticosterone: a reverse microdialysis study. J Neuroendocrinol 8: 299–307. Kalsbeek A, Van Heerikhuize JJ, Wortel J et al. (1996c). A diurnal rhythm of stimulatory input to the hypothalamopituitary-adrenal system as revealed by timed intrahypothalamic administration of the vasopressin V1 antagonist. J Neurosci 16: 5555–5565.

59

Kalsbeek A, Cutrera RA, Van Heerikhuize JJ et al. (1999). GABA release from SCN terminals is necessary for the light-induced inhibition of nocturnal melatonin release in the rat. Neuroscience 91: 453–461. Kalsbeek A, Fliers E, Franke AN et al. (2000a). Functional connections between the suprachiasmatic nucleus and the thyroid gland as revealed by lesioning and viral tracing techniques in the rat. Endocrinology 141: 3832–3841. Kalsbeek A, Garidou ML, Palm IF et al. (2000b). Melatonin sees the light: blocking GABA-ergic transmission in the paraventricular nucleus induces daytime secretion of melatonin. Eur J Neurosci 12: 3146–3154. Kalsbeek A, La Fleur SE, Van Heijningen C et al. (2004). Suprachiasmatic GABAergic inputs to the paraventricular nucleus control plasma glucose concentrations in the rat via sympathetic innervation of the liver. J Neurosci 24: 7604–7613. Kalsbeek A, Buijs RM, van Schaik R et al. (2005). Daily variations in type II iodothyronine deiodinase activity in the rat brain as controlled by the biological clock. Endocrinology 146: 1418–1427. Kalsbeek A, Foppen E, Schalij I et al. (2008a). Circadian control of the daily plasma glucose rhythm: an interplay of GABA and glutamate. PLoS One 3: e3194. Kalsbeek A, Verhagen LA, Schalij I et al. (2008b). Opposite actions of hypothalamic vasopressin on circadian corticosterone rhythm in nocturnal vs diurnal species. Eur J Neurosci 27: 818–827. Kalsbeek A, van der Spek R, Lei J et al. (2012). Circadian rhythms in the hypothalamo-pituitary-adrenal (HPA) axis. Mol Cell Endocrinol 349: 20–29. PMID: 21782883. Kennett JE, Poletini MO, Freeman ME (2008). Vasoactive intestinal polypeptide modulates the estradiol-induced prolactin surge by entraining oxytocin neuronal activity. Brain Res 1196: 65–73. Kiss JZ (1998). Dynamism of chemoarchitecture in the hypothalamic paraventricular nucleus. Brain Res Bull 20 (6): 699–708. Kita Y, Shiozawa M, Jin WH et al. (2002). Implications of circadian gene expression in kidney, liver and the effects of fasting on pharmacogenomic studies. Pharmacogenetics 12: 55–65. Klein DC, Weller JL, Moore RY (1971). Melatonin metabolism: neural regulation of pineal serotonin: acetyl coenzyme a N-acetyltransferase activity. Proc Natl Acad Sci U S A 68: 3107–3110. Klein DC, Smoot R, Weller JL et al. (1983). Lesions of the paraventricular nucleus area of the hypothalamus disrupt the suprachiasmatic-spinal cord circuit in the melatonin rhythm generating system. Brain Res Bull 10: 647–652. Kneisley LW, Moskowitz MA, Lynch HG (1978). Cervical spinal cord lesions disrupt the rhythm in human melatonin excretion. J Neural Transm Suppl 13: 311–323. Koenig RJ (2005). Regulation of type 1 iodothyronine deiodinase in health and disease. Thyroid 15: 835–840. Kreier F, Fliers E, Voshol PJ et al. (2002). Selective parasympathetic innervation of subcutaneous and intra-abdominal fat - functional implications. J Clin Invest 110: 1243–1250.

60

A. KALSBEEK AND R.M. BUIJS

Kreier F, Kalsbeek A, Ruiter M et al. (2003). Central nervous determination of food storage – a daily switch from conservation to expenditure. Implications for the metabolic syndrome, Eur J Pharmacol 480: 51–65. PMID: 14623350. Kreier F, Kap YS, Mettenleiter T et al. (2006). Tracing from fat tissue, liver, and pancreas: a neuroanatomical framework for the role of the brain in Type2 diabetes. Endocrinology 147: 1140–1147. La Fleur SE (2003). Daily rhythms in glucose metabolism: suprachiasmatic nucleus output to peripheral tissue. J Neuroendocrinol 15: 315–322. La Fleur SE, Kalsbeek A, Wortel J et al. (2000). Polysynaptic neural pathways between the hypothalamus, including the suprachiasmatic nucleus, and the liver. Brain Res 871: 50–56. Larsen PJ, Enquist LW, Card JP (1998). Characterization of the multisynaptic neuronal control of the rat pineal gland using viral transneuronal tracing. Eur J Neurosci 10: 128–145. Lee SK, Ryu PD, Lee SY (2013). Differential distributions of neuropeptides in hypothalamic paraventricular nucleus neurons projecting to the rostral ventrolateral medulla in the rat. Neurosci Lett 556: 160–165. Lehman MN, Bittman EL, Newman SW (1984). Role of the hypothalamic paraventricular nucleus in neuroendocrine responses to daylength in the golden hamster. Brain Res 308: 25–32. Leon-Mercado L, Herrera Moro Chao D, Basualdo MD et al. (2017). The arcuate nucleus: a site of fast negative feedback for corticosterone secretion in male rats. eNeuro 4 (1). Licht CM, de Geus EJ, Penninx BW (2013). Dysregulation of the autonomic nervous system predicts the development of the metabolic syndrome. J Clin Endocrinol Metab 98: 2484–2493. Li DP, Zhu LH, Pachuau J et al. (2014). mGluR5 Upregulation increases excitability of hypothalamic presympathetic neurons through NMDA receptor trafficking in spontaneously hypertensive rats. J Neurosci 34: 4309–4317. PMID: 24647951. Lieb DC, Parson HK, Mamikunian G et al. (2012). Cardiac autonomic imbalance in newly diagnosed and established diabetes is associated with markers of adipose tissue inflammation. Exp Diabetes Res 2012: 878760. Lilley TR, Wotus C, Taylor D et al. (2011). Circadian regulation of cortisol release in behaviorally Split Golden hamsters. Endocrinology 153: 732–738. Liposits Z, Uht RM, Harrison RW et al. (1987). Ultrastructural localization of glucocorticoid receptor (GR) in hypothalamic paraventricular neurons synthesizing corticotropin releasing factor (CRF). Histochemistry 87: 407–412. PMID: 3323142. Martino E, Bambini G, Vaudaga G et al. (1985). Effects of continuous light and dark exposure on hypothalamic thyrotropinreleasing hormone in rats. J Endocrinol Invest 8: 31–33. Meijer JH, Colwell CS, Rohling JHT et al. (2012). Dynamic neuronal network organization of the circadian clock and possible deterioration in disease. Prog Brain Res 199: 143–162. https://doi.org/10.1016/B978-0-444-59427-3.00009-5.

Mickelsen LE, Kolling FW, Chimileski BR et al. (2017). Neurochemical heterogeneity among lateral hypothalamic Hypocretin/orexin and melanin-concentrating hormone neurons identified through single-cell gene expression analysis. eNeuro 4 (5): 0013–0017. Mickelsen LE, Bolisetty M, Chimileski BR et al. (2019). Single-cell transcriptomic analysis of the lateral hypothalamic area reveals molecularly distinct populations of inhibitory and excitatory neurons. Nat Neurosci 22: 642–656. Moffitt JR, Bambah-Mukku D, Eichhorn SW et al. (2018). Molecular, spatial and functional single-cell profiling of the hypothalamic preoptic region. Science 362 (6416): 5324. Moore RY (1978). Neural control of pineal function in mammals and birds. J Neural Transm Suppl 13: 47–58. Moore RM (1996). Entrainment pathways and the functional organization of the circadian system. Prog Brain Res 111: 103–119. Moore RY, Klein DC (1974). Visual pathways and the central neural control of a circadian rhythm in pineal serotonin N-acetyltransferase activity. Brain Res 71: 17–33. Mori K, Ida T, Fudetani M et al. (2017). Identification of neuromedin U precursor-related peptide and its possible role in the regulation of prolactin release. Sci Rep 7 (1): 10468. Mrosovsky N (1990). Rheostasis: the physiology of change, Oxford University Press. Murakami M, Tanaka K, Greer MA (1988). There is a nyctohemeral rhythm of type II iodothyronine 5’-deiodinase activity in rat anterior pituitary. Endocrinology 123: 1631–1635. Nakahara K, Maruyama K, Ensho T et al. (2019). Neuromedin U suppresses prolactin secretion via dopamine neurons of the arcuate nucleus. Biochem Biophys Res Commun [Epub ahead of print]. Nguyen NL, Barr CL, Ryu V et al. (2017). Separate and shared sympathetic outflow to white and brown fat coordinately regulates thermoregulation and beige adipocyte recruitment. Am J Physiol Regul Integr Comp Physiol 312: R132–r145. Nonogaki K (2000). New insights into sympathetic regulation of glucose and fat metabolism. Diabetologia 43: 533–549. Nunez AA, Brown MH, Youngstrom TG (1985). Hypothalamic circuits involved in the regulation of seasonal and circadian rhytms in male golden hamsters. Brain Res Bull 15: 149–153. Oishi K, Miyazaki K, Kadota K et al. (2002). Genomewide expression analysis of mouse liver reveals CLOCKregulated circadian output genes. J Biol Chem 278: 41519–41527. Olcese J, Reuss S, Steinlechner S (1987). Electrical stimulation of the hypothalamic nucleus paraventricularis mimics the effects of light on pineal melatonin synthesis. Life Sci 40: 455–459. Oster H, Damerow S, Kiessling S et al. (2006). The circadian rhythm of glucocorticoids is regulated by a gating mechanism residing in the adrenal cortical clock. Cell Metab 4: 163–173.

ORGANIZATION OF THE NEUROENDOCRINE AND AUTONOMIC PVN Pachuau J, Li DP, Chen SR et al. (2014). Protein kinase CK2 contributes to diminished small conductance Ca2+activated K+ channel activity of hypothalamic presympathetic neurons in hypertension. J Neurochem 130: 657–667. PMID: 24806793. Palm IF, van der Beek EM, Swarts HJ et al. (2001). Control of the estradiol-induced prolactin surge by the suprachiasmatic nucleus. Endocrinology 142 (6): 2296–2302. Park SY, Walker JJ, Johnson NW et al. (2013). Constant light disrupts the circadian rhythm of steroidogenic proteins in the rat adrenal gland. Mol Cell Endocrinol 371: 114–123. Perreau-Lenz S, Kalsbeek A, Garidou ML et al. (2003). Suprachiasmatic control of melatonin synthesis in rats: inhibitory and stimulatory mechanisms. Eur J Neurosci 17: 221–228. Perreau-Lenz S, Kalsbeek A, P
evet P et al. (2004). Glutamatergic clock output stimulates melatonin synthesis at night. Eur J Neurosci 19: 318–324. Pickard GE, Turek FW (1983). The hypothalamic paraventricular nucleus mediates the photoperiodic control of reproduction but not the effects of light on the circadian rhythm of activity. Neurosci Lett 43: 67–72. Portillo F, Carrasco M, Vallo JJ (1998). Separate populations of neurons within the paraventricular hypothalamic nucleus of the rat project to vagal and thoracic autonomic preganglionic levels and express c-Fos protein induced by lithium chloride. J Chem Neuroanat 14 (2): 95–102. Price CJ, Samson WK, Ferguson AV (2009). Neuropeptide W has cell phenotype-specific effects on the excitability of different subpopulations of paraventricular nucleus neurones. J Neuroendocrinol 21 (10): 850–857. Puschel GP (2004). Control of hepatocyte metabolism by sympathetic and parasympathetic hepatic nerves. Anat Rec 280A: 854. Pyner S (2014). The paraventricular nucleus and heart failure. Exp Physiol 99: 332–339. PMID: 24317407. Quiroga RQ, Reddy L, Kreiman G et al. (2005). Invariant visual representation by single neurons in the human brain. Nature 435: 1102–1107. PMID: 15973409. Reiter RJ, King TS, Richardson BA et al. (1982). Studies on pineal melatonin levels in a diurnal species, the eastern chipmunk (Tamias striatus): effects of light at night, propranolol administration or superior cervical ganglionectomy. J Neural Transm 54: 275–284. Reuss S, Olcese J, Vollrath L (1985). Electrical stimulation of the hypothalamic paraventricular nuclei inhibits pineal melatonin synthesis in male rats. Neuroendocrinology 41: 192–196. Riskind PN, Kolodny JM, Larsen PR (1987). The regional hypothalamic distribution of type II 50 -monodeiodinase in euthyroid and hypothyroid rats. Brain Res 420: 194–198. Ritter S, Watts AG, Dinh TT et al. (2003). Immunotoxin lesion of hypothalamically projecting norepinephrine and epinephrine neurons differentially affects circadian and stressorstimulated corticosterone secretion. Endocrinology 144: 1357–1367.

61

Roa SLR, Martinez EZ, Martins CS et al. (2017). Postnatal ontogeny of the circadian expression of the adrenal clock genes and corticosterone rhythm in male rats. Endocrinology 158: 1339–1346. Romanov RA, Zeisel A, Bakker J et al. (2016). Molecular interrogation of hypothalamic organization reveals distinct dopamine neuronal subtypes. Nat Neurosci 20: 176–188. Romanov RA, Alpar A, Hokfelt T et al. (2017). Molecular diversity of corticotropin-releasing hormone mRNA-containing neurons in the hypothalamus. J Endocrinol 232: R161–R172. Rosario W, Singh I, Wautlet A et al. (2016). The brain-topancreatic islet neuronal map reveals differential glucose regulation from distinct hypothalamic regions. Diabetes 65: 2711–2723. Sabath E, Baez-Ruiz A, Buijs RM (2015). Non-alcoholic fatty liver disease as a consequence of autonomic imbalance and circadian desynchronization. Obes Rev 16: 871–882. Scheer FA, Ter Horst GJ, van Der Vliet J et al. (2001). Physiological and anatomic evidence for regulation of the heart by suprachiasmatic nucleus in rats. Am J Physiol 280: H1391–H1399. Scheer FA, Van Doornen LJ, Buijs RM (2004). Light and diurnal cycle affect autonomic cardiac balance in human; possible role for the biological clock. Auton Neurosci 110: 44–48. Scheer FA, Hilton MF, Mantzoros CS et al. (2009). Adverse metabolic and cardiovascular consequences of circadian misalignment. Proc Natl Acad Sci U S A 106: 4453–4458. Shimazu T (1987). Neuronal regulation of hepatic glucose metabolism in mammals. Diabetes Metab Rev 3: 185–206. Shiuchi T, Haque MS, Okamoto S et al. (2009). Hypothalamic orexin stimulates feeding-associated glucose utilization in skeletal muscle via sympathetic nervous system. Cell Metab 10: 466–480. Simmons DM, Swanson LW (2009). Comparison of the spatial distribution of seven types of neuroendocrine neurons in the rat paraventricular nucleus: toward a global 3D model. J Comp Neurol 516: 423–441. Smale L, Cassone VM, Moore RY et al. (1989). Paraventricular nucleus projections mediating pineal melatonin and gonadal responses to photoperiod in the hamster. Brain Res Bull 22: 263–269. Stanley S, Pinto S, Segal J et al. (2010). Identification of neuronal subpopulations that project from hypothalamus to both liver and adipose tissue polysynaptically. Proc Natl Acad Sci U S A 107: 7024–7029. Stephan FK, Berkley KJ, Moss RL (1981). Efferent connections of the rat suprachiasmatic nucleus. Neuroscience 6: 2625–2641. Stern JE (2001). Electrophysiological and morphological properties of pre-autonomic neurones in the rat hypothalamic paraventricular nucleus. J Physiol 537 (Pt 1): 161–177. Strubbe JH, Alingh Prins AJ, Bruggink J et al. (1987). Daily variation of food-induced changes in blood glucose and insulin in the rat and the control by the suprachiasmatic nucleus and the vagus nerve. J Auton Nerv Syst 20: 113–119.

62

A. KALSBEEK AND R.M. BUIJS

Sun X, Rusak B, Semba K (2000). Electrophysiology and pharmacology of projections from the suprachiasmatic nucleus to the ventromedial preoptic area in rat. Neuroscience 98: 715–728. Sun X, Whitefield S, Rusak B et al. (2001). Electrophysiological analysis of suprachiasmatic nucleus projections to the ventrolateral preoptic area in the rat. Eur J Neurosci 14: 1257–1274. Suthana NA, Parikshak NN, Ekstrom AD et al. (2015). Specific responses of human hippocampal neurons are associated with better memory. Proc Natl Acad Sci U S A 112: 10503–10508. PMID: 26240357. Swanson LW, Cowan WM (1975). The efferent connections of the suprachiasmatic nucleus of the hypothalamus. J Comp Neurol 160: 1–12. Swanson LW, Kuypers HGJM (1980). The paraventricular nucleus of the hypothalamus: cytoarchitectonic subdivisions and organization of projections to the pituitary, dorsal vagal complex, and spinal cord as demonstrated by retrograde fluorescence double-labeling methods. J Comp Neurol 194: 555–570. Tasker JG, Dudek FE (1991). Electrophysiological properties of neurones in the region of the paraventricular nucleus in slices of rat hypothalamus. J Physiol 434: 271–293. Teclemariam-Mesbah R, Kalsbeek A, P
evet P et al. (1997). Direct vasoactive intestinal polypeptide-containing projection from the suprachiasmatic nucleus to spinal projecting hypothalamic paraventricular neurons. Brain Res 748: 71–76. Teclemariam-Mesbah R, Ter Horst GJ, Postema F et al. (1999). Anatomical demonstration of the suprachiasmatic nucleus - pineal pathway. J Comp Neurol 406: 171–182. Tessonneaud A, Locatelli A, Caldani M et al. (1995). Bilateral lesions of the suprachiasmatic nuclei alter the nocturnal melatonin secretion in sheep. J Neuroendocrinol 7: 145–152. Ueyama T, Krout KE, Nguyen XV et al. (1999). Suprachiasmatic nucleus: a central autonomic clock. Nat Neurosci 2 (12): 1051–1053. Van Den Pol AN (1991). Glutamate and aspartate immunoreactivity in hypothalamic presynaptic axons. J Neurosci 11: 2087–2101. Van Den Pol AN, Tsujimoto KL (1985). Neurotransmitters of the hypothalamic suprachiasmatic nucleus: immunocytochemical analysis of 25 neuronal antigens. Neuroscience 15 (4): 1049–1086. Van den Top M, Nolan MF, Lee K et al. (2003). Orexins induce increased excitability and synchronisation of rat sympathetic preganglionic neurones. J Physiol 549: 809–821. Van Der Beek EM, Wiegant VM, Van Der Donk HA et al. (1993). Lesions of the suprachiasmatic nucleus indicate the presence of a direct vasoactive intestinal polypeptide- containing projection to gonadotrophin-releasing hormone neurons in the female rat. J Neuroendocrinol 5: 137–144. Van Der Beek EM, Horvath TL, Wiegant VM et al. (1997). Evidence for a direct neuronal pathway from the

suprachiasmatic nucleus to the gonadotropin-releasing hormone system: combined tracing and light and electron microscopic immunocytochemical studies. J Comp Neurol 384: 569–579. Vrang N, Larsen PJ, Mikkelsen JD (1995). Direct projection from the suprachiasmatic nucleus to hypophysiotrophic corticotropin-releasing factor immunoreactive cells in the paraventricular nucleus of the hypothalamus demonstrated by means of Phaseolus vulgaris-leucoagglutinin tract tracing. Brain Res 684: 61–69. Vrang N, Mikkelsen JD, Larsen PJ (1997). Direct link from the suprachiasmatic nucleus to hypothalamic neurons projecting to the spinal cord: a combined tracing study using cholera toxin subunit B and Phaseolus vulgaris-leucoagglutinin. Brain Res Bull 44: 671–680. Wang LA, Nguyen DH, Mifflin SW (2019). Corticotropinreleasing hormone projections from the paraventricular nucleus of the hypothalamus to the nucleus of the solitary tract increase blood pressure. J Neurophysiol 121: 602–608. PMID: 30565964. Wang Q, Van Heerikhuize J, Aronica E et al. (2013). Glucocorticoid receptor protein expression in human hippocampus; stability with age. Neurobiol Aging 34: 1662–1673. PMID: 23290588. Watts AG (2005). Glucocorticoid regulation of peptide genes in neuroendocrine CRH neurons: a complexity beyond negative feedback. Front Neuroendocrinol 26: 109–130. Watts AG, Swanson LW (1987). Efferent projections of the suprachiasmatic nucleus: II.Studies using retrograde transport of fluorescent dyes and simultaneous peptide immunohistochemistry in the rat. J Comp Neurol 258: 230–252. Wiedmann NM, Stefanidis A, Oldfield BJ (2017). Characterization of the central neural projections to brown, white, and beige adipose tissue. FASEB J 31: 4879–4890. Wotus C, Lilley TR, Neal AS et al. (2013). Forced desynchrony reveals independent contributions of suprachiasmatic oscillators to the daily plasma corticosterone rhythm in male rats. PLoS One 8 (7): e68793. Wulsin LR, Horn PS, Perry JL et al. (2016). The contribution of autonomic imbalance to the development of metabolic syndrome. Psychosom Med 78: 474–480. Wurtman RJ, Axelrod J, Sedvall G et al. (1967). Photic and neural control of the 24-hour norepinephrine rhythm in the rat pineal gland. J Pharmacol Exp Ther 157: 487–492. Yanovski J, Witcher J, Adler NT et al. (1987). Stimulation of the paraventricular nucleus area of the hypothalamus elevates urinary 6-hydroxymelatonin during daytime. Brain Res Bull 19: 129–133. Yi CX, Serlie MJ, Ackermans MT et al. (2009). A major role for perifornical orexin neurons in the control of glucose metabolism in rats. Diabetes 58: 1998–2005. Zandieh Doulabi B, Platvoet-Ter Schiphorst M, Kalsbeek A et al. (2004). Diurnal variation in rat liver thyroid hormone receptor (TR)-alpha messenger ribonucleic acid

ORGANIZATION OF THE NEUROENDOCRINE AND AUTONOMIC PVN (mRNA) is dependent on the biological clock in the suprachiasmatic nucleus, whereas diurnal variation of TR beta 1 mRNA is modified by food intake. Endocrinology 145: 1284–1289. Zeitzer JM, Ayas NT, Shea SA et al. (2000). Absence of detectable melatonin and preservation of cortisol and thyrotropin rhythms in tetraplegia. J Clin Endocrinol Metab 85: 2189–2196. Zeitzer JM, Buckmaster CL, Parker KJ et al. (2003). Circadian and homeostatic regulation of hypocretin in a primate

63

model: implications for the consolidation of wakefulness. J Neurosci 23: 3555–3560. Zhang S, Zeitzer JM, Yoshida Y et al. (2004). Lesions of the suprachiasmatic nucleus eliminate the daily rhythm of hypocretin-1 release. Sleep 27: 619–627. Zhang L, Herna´ndez VS, Zetter MA et al. (2020). VGLUTVGAT expression delineates functionally specialized populations of vasopressin-containing neurons including a glutamatergic perforant path-projecting cell group to the hippocampus in rat and mouse brain. J Neuroendocrinol.

This page intentionally left blank

Handbook of Clinical Neurology, Vol. 180 (3rd series) The Human Hypothalamus: Middle and Posterior Region D.F. Swaab, F. Kreier, P.J. Lucassen, A. Salehi, and R.M. Buijs, Editors https://doi.org/10.1016/B978-0-12-820107-7.00005-7 Copyright © 2021 Elsevier B.V. All rights reserved

Chapter 5

Sex differences of oxytocin and vasopressin in social behaviors QIAOQIAO LU1 AND SHAOHUA HU2,3,4* 1

Department of Psychiatry, Hangzhou Seventh People’s Hospital, Hangzhou, China

2

Department of Psychiatry, First Affiliated Hospital, Zhejiang University School of Medicine, Hangzhou, China 3

The Key Laboratory of Mental Disorder Management in Zhejiang Province, Hangzhou, China 4

Brain Research Institute of Zhejiang University, Hangzhou, China

Abstract The neuropeptides oxytocin (OT) and vasopressin (VP) are known to mediate social cognition and behaviors in a sex-dependent manner. This chapter reviews the sex-dependent influence of OT and VP on social behaviors, focusing on (1) partner preference and sexual orientation, (2) memory modulation, (3) emotion regulation, and (4) trust-related behaviors. Most studies suggest that OT promotes familiar (opposite-sex) partner preference, strengthens memory, relieves anxiety, and increases trust. However, VP-regulated social cognition has been studied less than OT. VP facilitates familiar (opposite-sex) partner preference, enhances memory, induces anxiety, and influences happiness/anger perception. Detailed sex differences of these effects are reviewed. There is a male preponderance in the use of animal models and many study results are too complex to draw firm conclusions. Clarifying the complex interplay between the OT/VP system and sex hormones in the regulation of social behaviors is needed.

INTRODUCTION Oxytocin (OT) and vasopressin (VP) are well-known social neuropeptides. They have shared similar and conservative molecular structures during their long evolutionary history across placental mammals (Donaldson and Young, 2008). Their receptors are expressed both in the central nervous system (CNS) and peripheral tissues. This indicates that they are involved as neurotransmitters/neuromodulators in advanced functions and also modulate peripheral physiologic activity as hormones. Here we mainly discuss their effects on the CNS. Although the structures of OT and VP differ by only two amino acids, they act differently on the brain. They have disparate receptor expressions in different brain

regions. Interestingly, many of their actions are sexdependent. They interact with sex-related factors and shape social behaviors. Social behavior is how an organism acts in the presence of members of the same species and environmental stimuli. It is a broad concept in human society, encompassing how we live our lives, interact with others, follow our careers, avoid difficulties, and make use of nature. In animal communities, social behavior mainly involves courtship behavior, foraging behavior, reproduction activity, and territorial competition. All these behaviors are shaped by environmental factors, sociopsychologic factors, and biologic factors. We herein mainly analyze how biologic factors regulate social behavior in an objective way. Cognition, an advanced

*Correspondence to: Shaohua Hu, Department of Psychiatry, the First Affiliated Hospital, Zhejiang University School of Medicine, The Key Laboratory of Mental Disorder’s Management in Zhejiang Province, #79 Qingchun Road, Hangzhou 310003, China. Tel: +86-571-87235989, +86-139-57162903, Fax: +86-571-87235989, E-mail: [email protected]

66

Q. LU AND S. HU

function of the brain, includes all mental abilities and processes in response to outside information and is closely related to the physiologic states of the CNS. An organism performs its behavior based on its cognitive features. This chapter focuses on how the peptide molecules OT and VP influence social cognition and lead to various behaviors. Several aspects of sex-based social cognition and behavior are discussed, including partner preference, memory modulation, emotion regulation, and trust-related behaviors. OT is mainly synthesized by the magnocellular neurons of the supraoptic nucleus (SON) and paraventricular nucleus (PVN) of the mammalian hypothalamus (Eisenberg et al., 2019). After production, OT is stored in the posterior pituitary and released into the blood as a hormone. Parvocellular neurons also produce OT and project fibers to the spinal cord and brainstem (Althammer and Grinevich, 2017). Rogers et al. detected vasopressin- and oxytocin-containing nerve fibers in several neocortical regions. In human, oxytocin immunoreactive nerve fibers were observed in the gyrus rectus and the ventral and subgenual anterior cingulate gyrus (Brodmann’s areas 24 and 25) (Rogers et al., 2018). OT influences various cognitive performances through the central oxytocin receptor (OTR) system. OTR expression is uneven across the brain and is influenced by life stage, species, and sex (Smith et al., 2017; Vaidyanathan and Hammock, 2017). For example, in humans, OTR is most commonly detected in the basal nucleus of Meynert, lateral septal nucleus, substantia nigra pars compacta, nucleus of the solitary tract, and substantia gelatinosa of the trigeminal nucleus (Gimpl and Fahrenholz, 2001). In rats, OTR is mostly detected in the cingulate cortex, retrosplenial cortex, caudoputamen, globus pallidus, dorsal subiculum, lateral mamillary nucleus, medial mamillary nucleus, and substantia gelatinosa of trigeminal nucleus in pups, and in islands of Calleja, peduncular cortex, ventral pallidum cell groups, bed nucleus of stria terminalis (BNST), central amygdaloid nucleus, and ventral subiculum in adults (Gimpl and Fahrenholz, 2001). Male mice had fewer OT-immunoreactive fibers in the limbic system than females (Haussler et al., 1990). In particular, OTimmunostained neurons in the ventral ansa lenticularis, the perifornical region, and the lateral hypothalamus in male mice were almost absent (Haussler et al., 1990). OTR expression is modulated by sex hormones and shows obvious sex differences in brain regions across different species. Brain-region, species-specific, and sex-dependent expression of OTR is closely associated with some sexdimorphic social behaviors (Dumais and Veenema, 2016). These are discussed in the following sections in detail. VP is produced by the magnocellular neurosecretory cells in PVN, SON, and the parvocellular neurosecretory cells in PVN. VP produced by magnocellular

neurosecretory cells is also stored in the posterior pituitary and released into the blood as an antidiuretic hormone. VP produced by parvocellular neurons is transported by the portal system to the anterior pituitary and plays a role in regulation of the hypothalamic–pituitary– adrenal (HPA) axis. In addition, extrahypothalamic VP-producing regions include BNST, the medial amygdala (MeA) (Frank and Landgraf, 2008), the hippocampus, the choroid plexus, the diagonal band of Broca (de Vries and Miller, 1998), and the septum (Buijs, 1978). In human, VP, and not OT, is also produced in the suprachiasmatic nucleus and the biological clock (Swaab et al., 1985), and VP immunoreactive nerve fibers were also found in the insular cortex, the frontal operculum, the subgenual cingulate gyrus, and the primary olfactory cortex (Rogers et al., 2018). Three subtypes of VP receptors (VPRs) have been identified: the V1 receptors (V1Rs), which include subtypes V1aR and V1bR, and the V2 receptor (V2R) (Frank and Landgraf, 2008). V1aR is the subtype most expressed in the brain, and the second-most expressed is V1bR. V2R is scarcely expressed in the CNS and is mainly expressed in the renal collecting ducts (Frank and Landgraf, 2008). Hence, V1aR and V1bR primarily mediate social cognition. VPR expression varies across different brain regions and differs according to species, sex, and life stage. In general, males have more VP synthesis and innervation in the BNST and MeA than females (de Vries and Miller, 1998). Sex differences of V1aR and V1bR expression in brain regions are far from conclusive, because the current studies included few animals, used more males than females, and analyzed only a few brain regions (Dumais and Veenema, 2016). These research studies have been performed in animals and humans. It is interesting that our emotions and behaviors are modulated not only by subjective wishes, but also by several molecules. In this review, we first introduce the experimental models and intervention methods. Then we discuss the sexually dimorphic influences of OT on social cognition, followed by a discussion of VP in the same context. Studies on peripartum females related to these topics are covered in a separate section. Finally, we summarize current studies, discuss the interplay between OT, VP, and sex hormones, and analyze future directions.

MODELS AND INTERVENTIONS Experimental models in this field include rodents, primates, healthy humans, and patients. The prairie vole (Microtus ochrogaster) is an excellent animal model for pair bonding due to its being socially monogamous (Carter and Getz, 1993). The mandarin vole is also monogamous (Jia et al., 2008) and appropriate for studies on pair bonding. These two animal models are often used

SEX DIFFERENCES OF OXYTOCIN AND VASOPRESSIN IN SOCIAL BEHAVIORS in opposite-sex pair-bonding studies. In same-sex partner preference studies, the meadow vole is the primary choice, for its social promiscuousness. Meadow voles secrete reduced levels of gonadal hormones in winter, which is suitable for researching nonreproductive social behaviors. For example, uteri of female meadow voles are smaller in winter than in summer. So in a winter-like day length environment, these females can form partner preference with either same- or opposite-sex roommates (Parker and Lee, 2003; Beery et al., 2008; Beery et al., 2009). Laboratory rats, such as Wistar rats and Sprague Dawley rats, are used in memory-relevant tests. Other less commonly used animal models and human models are listed in Tables 5.1 and 5.2. Methods of intervention in this field (see Table 5.1 and 5.2) include intranasal administration, intracranial injection (intracerebroventricular (ICV) injection), intraperitoneal injection, and subcutaneous injection for rodents, and intranasal spray for marmosets and humans. Intranasal administration proved to elevate central VP concentration in humans (Pietrowsky et al., 1996) as well as central OT level in macaques (Dal Monte et al., 2014).

SEXUALLY DIMORPHIC INFLUENCES OF OT ON SOCIAL COGNITION Partner preference and sexual orientation Partner preference is used for selecting a pair mate. OT mediates partner preference in monogamous animals. This underlies the biologic basis of fidelity and reflects a potential mechanism of sexual orientation in humans. The effects of OT on females and males are discussed separately (Fig. 5.1; Lu et al., 2019). The simplified study designs and results in this field are listed in Table 5.1. Transient administration of OT induced a preference for familiar mates or objects in both sexes of rodents and humans. This effect was more significant in females regardless of the means of intervention. Staying with opposite-sex cage mates, both male and female prairie voles showed familiar partner preference with intracranial OT injections (Williams et al., 1992, 1994; Bales and Carter, 2003; Duclot et al., 2016; Johnson et al., 2016). A left lateral ventricle injection of a selective OTR antagonist (OTA), (d(CH,),-[Tyr(Me),Thr4,TryNHi], blocked this effect (Williams et al., 1994; Insel and Hulihan, 1995). Similar effects can be seen in both sexes of mandarin voles (Jia et al., 2008), with a more prominent effect in females (Cushing and Carter, 2000; Young and Wang, 2004). However, when OT was given subcutaneously, only female prairie voles exhibited familiar partner preference (Cushing and Carter, 2000). In marmoset experiments, intranasal OT-treated females spent more time in intimate contact with the male partner, while males spent less time with both the female stranger and the female partner (Cavanaugh et al., 2014).

67

Intranasal OT-treated marmosets start to contact their familiar opposite-sex partner sooner than the OT antagonisttreated ones or controls in both sexes (Smith et al., 2010). In addition, they huddled with their opposite-sex familiar partner more frequently without sex difference, while OT antagonist treatment blocked this effect (Smith et al., 2010). These results were consistent with those using indirect methods to substantially increase the CNS OT level. For example, viral vector gene transfer of OTR to prepubertal female prairie voles resulted in increased expression of OTR in the nucleus accumbens (NAcc). This overexpression of OTR continued into adulthood and accelerated familiar partner (heterosexual) preference formation (Keebaugh and Young, 2011). Central injection of trichostatin A (TSA) also upregulated OTR expression in NAcc and promoted familiar partner (heterosexual) preference in both male and female prairie voles (Wang et al., 2013; Duclot et al., 2016). A difference between males and females was that, besides OTR expression, TSA also upregulated V1aR gene expression in females. The overexpressed V1aR may also help to facilitate familiar partner (heterosexual) preference in females (Duclot et al., 2016). On receiving a 21-day continuous administration of a low (0.08 IU/kg) or medium (0.8 IU/kg) dosage of OT, only male prairie voles suffered deficits in the formation of partner (opposite-sex) preference with a decrease of contact time with a familiar partner. However, a high dosage of OT (8 IU/kg) did not affect partner preference either in males or females (Bales et al., 2013). This indicates that OT regulates familiar (opposite-sex) partner preference of males in a dose-dependent way. Facing same-sex cage mates, both male and female voles preferred the familiar one (Beery and Zucker, 2010; Triana-Del Rio et al., 2015). Interestingly, this familiar preference for same-sex or opposite-sex partner is also applicable when the “partner” is replaced by an “object” in centrally OT-treated female rats (Madularu et al., 2014b). The familiar object preference is not so obvious in male prairie voles. OT-treated male voles did not show an object preference, while the placebo-treated control showed a novel object preference. However, after intervening with the OT antagonist, male voles also showed a novel object preference (Madularu et al., 2014a). In human studies, OT also promotes familiar partner preference regardless of the sex of the preferred partner (Liu et al., 2013). Heterosexual men in a monogamous relationship will keep further distance from an attractive heterosexual female stranger after receiving intranasally sprayed OT (Scheele et al., 2012), while keeping a closer contact with their heterosexual male friend (Cohen et al., 2017). OT intervention reduced interpersonal distance between heterosexual women in the luteal phase of the menstrual cycle (Riem et al., 2019). Only one study was performed on homosexual men, reporting that following

Table 5.1 Experimental models and species, doses, routes, brain regions, and effects of OT administration Models and species Partner preference Prairie voles ♂

Prairie voles ♀

Doses

Routes

Artificial CSF: 2 mL OTR antagonist: 5 ng; (Gly-NH2d(CH2)5[D-Tyr2,Thr4]OVT) Overexpress OTR in NAcc

Brain regions

Effects

References

ICV injection

Facilitate opposite-sex partner preference

Johnson et al. (2016)

Viral vector gene transfer to NAcc nucleus accumbens (NAcc) injection Intranasally; a 21-day continuous intervention

Facilitated opposite-sex partner preference in females Low, medium dosage decreased same-sex partner preference in males; high dosage treated males and all females unaffected Facilitate opposite-sex partner preference in both sexes Upregulated OTR gene expression in the NAcc and facilitated opposite-sex partner preference Upregulated OTR and V1aR gene expression in the NAcc and facilitated opposite-sex partner preference Facilitated opposite-sex partner preference in females OT group did not show object preference, OTR antagonist group showed a novel object preference, like the placebo group

Keebaugh and Young (2011)

Prairie voles ♂+♀

OT dosage: Low: 0.08 IU/kg Medium: 0.8 IU/kg High: 8 IU/kg

Prairie voles ♂+♀

OT: 100 ng VP: 100 ng

ICV injection

Prairie voles ♂

Trichostatin A (TSA): histone deacetylase inhibitor, 0.08, 0.4, and 4 ng; CSF: 500 nL TSA: 0.08, 0.4, or 4 ng CSF: 500 nL

Lateral ventricle injection

NAcc

ICV injection or NAcc injection

NAcc

Mandarin voles ♂+♀

OT: 3 mg Placebo

Subcutaneous injection

BNST, PVN, MD, mPOA, LSD, CeA

Prairie voles ♂

OT: 0.4 mg/kg OTR antagonist: 20 mg/kg (L-368,899) Placebo

Intraperitoneal injection

Prairie voles ♀

Bales et al. (2013)

Cho et al. (1999)

Duclot et al. (2016)

Wang et al. (2013)

Jia et al. (2008)

Madularu et al. (2014a)

OT: 1.0 mg Placebo Pro8-OT: 25 IU Leu8-OT: 25 IU OT antagonist: 20 mg/kg Placebo Men adults (homosexual and OT: 24 IU heterosexual) Placebo (Four groups)

Left lateral ventricle OT injection Intranasally (OT) Orally (OT antagonist)

Familiar object preference

Madularu et al. (2014b)

Help to keep fidelity in well-established opposite-sex pairs

Cavanaugh et al. (2014)

Intranasally

Thienel et al. (2014)

Men adults

OT: 24 IU Placebo

Intranasally

Men adults

OT: 24 IU Placebo

Intranasally

Women adults

OT: 24 IU Placebo

Intranasally

OT-treated homosexual men were more easily attracted by the male faces despite the facial expressions and the happy female faces Men in a monogamous relationship kept a further distance from a female stranger Prefer to approach a friend but not a stranger with an increased dmPFC activity Reduced same-sex interpersonal distance

OT: 8–28 IU Placebo OT: 0.2 mg per day for 2–3 days OTR antagonist: 5 mg/kg (L-368,899) OT: 24 IU OT: 48 IU Placebo

Intranasally

Long-Evans rats ♀ Marmosets ♂+♀

Memory modulation Macaques ♂+♀ Sprague Dawley rats ♂

Men adults

OT was given intranasally L-368,899 was given intraperitoneally

Hippocampus

OT: 24 IU Placebo

Intranasally

Men adults

OT: 24 IU Placebo

Intranasally

Men and women adults

OT: 24 IU Placebo

Intranasally

OT: low dose (1 ng/h  19 days); high dose (10 ng/h  15 days)

Chronic ICV infusion

Septum, the basolateral and medial amygdala, the median raphe nucleus

Cohen et al. (2017)

Riem et al. (2019)

Increased working memory Simpson et al. (2017) in males, not in females Reduced stress-induced Lee et al. (2015) hippocampal memory and plasticity impairment Increased the recall of specific personal memories and positive social affiliation memories Facilitated memory encoding and retrieval of negative social stimuli Impaired memory of both social and nonsocial visual objects in men Impaired recollection judgments in men only

Intranasally

Men adults

Anxiety regulation Mice ♂

dmPFC

Scheele et al. (2012)

Low dose prevented hyper anxiety, high dose promoted anxiety

Cardoso et al. (2014)

Weigand et al. (2013)

Herzmann et al. (2012)

Herzmann et al. (2013)

Peters et al. (2014)

Continued

Table 5.1 Continued Models and species

Doses

Routes

Brain regions

Effects

References

Rats ♂ Sprague Dawley rats ♂+♀ Wistar rats ♀

OT: 0.01 nmol/0.5 mL Saline: 0.5 mL OT: 1.0 mg/1 mL  1 mL Saline: 1 mL OT: 0.1 mg/5 mL Placebo: Ringer’s solution 5 mL OTR antagonist: 5 ng CSF Oxytocin: 10 ng or 100 ng OTR antagonist: 10 ng or 100 ng (des-Gly-NH2, d(CH2)5[Tyr(Me)2,Thr4] OVT) OT: 24 IU Placebo

PVN injection

Hypothalamus

Reduced anxiety

Blume et al. (2008)

The prelimbic region of the mPFC injection ICV injection

mPFC

Decreased anxiety regardless Sabihi et al. (2014) of sex Reduced trait anxiety and de Jong et al. (2014) aggression

OT: 24 IU Placebo

Intranasally

Prairie voles ♂ Prairie voles ♀

Men and women adults

Men and women adults

Happiness and anger regulation Men adults OT: 24 IU Placebo Men adults OT: 24 IU Placebo

PVN

ICV injection Intra-PVN injection

Paraventricular nucleus

Intranasally

Intranasally Intranasally

Men and women adults

OT: 24 IU Placebo

Intranasally

Men and women patients with chronic depression

OT: 24 IU Placebo

Intranasally

Men adults

OT: 24 IU Placebo

Intranasally

Amygdala, anterior insula

OT reduced aggression in males Reduce stress

Winslow et al. (1993) Smith and Wang (2014)

Grillon et al. (2013) Increased anxiety to unpredictable threat in both sexes Reduced anxiety only in men Chen et al. (2016) and was ineffective in women OT increased happiness perception Increased happiness perception, did not decrease anger perception Reduced negative affect in men; increased anger in women Reduced attention to angry faces, increased attention to happy faces in both sexes OT decreased aversion to angry faces

Domes et al. (2013a) Domes et al. (2013b)

Kubzansky et al. (2012)

Domes et al. (2016)

Evans et al. (2010)

Abbreviations: BNST, bed nucleus of the stria terminalis; CeA, central amygdaloid nucleus; CSF, cerebrospinal fluid; dmPFC, dorsomedial prefrontal cortex; ICV, intracerebroventricular; IU, international unit; Leu, leucine; LSD, the dorsal part of the lateral septal nucleus; MD, mediodorsal thalamic nucleus; mPFC, medial prefrontal cortex; mPOA, medial preoptic area; NAcc, nucleus accumbens; OT, oxytocin; OTR, oxytocin receptor; Pro, proline; TSA, Trichostatin A; PVN, hypothalamic paraventricular nucleus; VP, vasopressin.

Table 5.2 Experimental models and species, doses, routes, brain regions, and effects of VP administration Models and species

Doses

Routes

Partner preference Meadow voles ♂

/V1aR overexpression

Prairie voles ♂

/

Prairie voles ♂+♀

One of three doses of VP: (1, 5, or 10 mg) Saline: 50 mL V1aR antagonist: 5 pg  500 ng (d(CH2)5[Tyr(Me)]) CSF /

Prairie voles ♂

Prairie voles ♂

Prairie voles ♂

V1aR antagonist: 5 ng (d(CH2)5[Tyr(Me)]AVP) Placebo

Callicebus cupreus ♂

Low VP: 40 IU High VP: 80 IU Placebo

Effects

References

Viral vector V1aR gene transfer to the ventral forebrain Viral vector V1aR gene transfer into ventral pallidal area Subcutaneous injection

Facilitated opposite-sex partner preference

Lim and Young (2004)

Facilitated opposite-sex partner preference

Pitkow et al. (2001)

Facilitated opposite-sex preference

Cushing et al. (2001)

ICV injection

A wide range of V1aR antagonists failed to exhibit aggression

Winslow et al. (1993)

Ventral pallidum Viral vectors expressing shRNA sequences targeting Avpr1a mRNA injection into ventral pallidum in juvenile male voles ICV injection at different time points

Downregulated pallidal Barrett et al. (2013) V1aR density and impaired the opposite-sex partner preference, reduced anxiety-like behavior in adulthood Donaldson et al. (2010) Injection prior to cohabitation with mating or immediately prior to partner preference testing failed to display an opposite-sex partner preference, while injection immediately after cohabitation with mating and control animals receiving placebo displayed opposite-sex partner preferences Placebo: opposite-sex Jarcho et al. (2011) stranger preference High VP: opposite-sex partner preference

Intranasally

Brain regions

Continued

Table 5.2 Continued Models and species Memory modulation Spinocerebellar ataxia type 3 mice / Vascular dementia model rats ♂ Han Chinese first-episode schizophrenic patients /

Sprague Dawley rats ♂ Wistar rats ♂

Wistar rats ♂

Brattleboro rats ♂+♀ Wistar rats ♂ Anxiety regulation CD rats ♂+♀ Wistar rats ♀

Doses

Routes

VP: / VPR antagonist: /

Intraventricular injection

VP: 600 ng Placebo

PVN injection

VP: 12-week treatment

Intranasally

Orally given V1bR antagonist: 20 or 60 mg/kg (A-988315) Placebo VP: 0.001 and 0.005 mg/kg Intraperitoneally V1aR antagonist: 1 mg/kg (SR49059) OTR antagonist: 10, 30, and 100 mg/kg (WAY 267464) OTR antagonist: 5 mg/kg (Compound 25) VP: 10 fg ICV injection Pirenzepine or KN-62

/ VP against Abeta: 25 nmol (Ab25–35) VP: 0.1, 1, 10 nmol V1bR antagonist: 1–30 mg/kg, weighing 25–30 g (V1B-30N) V1aR antagonist: [d(CH2)5Tyr(Me)2AVP] 100 ng/0.5 mL (SS149415): 100 ng/0.5 mL

Brain regions

Effects

References

Jiang et al. (2017)

PVN, hippocampus

VP strengthened but VPR antagonist weakened the spatial memory Improved memory

Improved long-term memory, short-term memory, immediate memory, and memory quotient Improved stress-induced retrieval of memory impairment OTR antagonist (WAY 267464): blocked the social recognition memory-enhancing effects of VP via V1aR

Geng et al. (2017)

HPA axis

Ventral hippocampus

/ VP deficiency ICV injection

Intraperitoneal injection

HPA axis

Medial-posterior part of BNST infusion

PVN

Li et al. (2017)

Barsegyan et al. (2015)

Hicks et al. (2015)

VP improved scopolamine-induced impairments of spatial memory Deficits of spatial working memory performance Protected spatial learning and memory

Mishima et al. (2001)

Blockage of V1bR diminished anxiety-like behavior in male rodents Decreased anxiety-like behavior in lactating rats

Hodgson et al. (2014)

Aarde and Jentsch (2006) Pan et al. (2010)

Bayerl et al. (2016)

Men adults

VP: 20 IU Saline

Intranasally

mPFC, amygdala

Syrian hamsters ♂

Anabolic-androgenic steroids pretreated or not pretreated VP: 0.09 uM  0.5 mL V1aR antagonist: 9.0 uM  0.5 mL ([OHPhaa-D-Tyr(Me)-Phe-GlnAsn-Arg-Pro-Arg-NH2]) Ethanol or water pretreatment: 4 g/kg  11 postnatal days OTR agonist: 0, 2.5, 5 mg/kg (WAY-267464 dihydrochloride) V1aR antagonist: 0, 1, 3, 9 mg/kg (SR-49059) V1bR antagonist: 0, 5, 10, 20 mg/kg (SSR-149415) /

Subcutaneous injection for pretreatment; LAH injection for VP/V1aR antagonist treatment

LAH

Pretreatment: intragastrically The others: intraperitoneal injection

Hypothalamus

Sprague Dawley rats ♂+♀

Mice ♂ Men and women

VP: 20 IU Placebo Other emotion perception and trust-related behaviors Men adults VP: 20 IU Placebo

Men adults Men adults

Wild type V1aR and V1bR double knockout (dKO) Intranasally

Decreased suppression effect Brunnlieb et al. (2013) of mPFC toward amygdala, increased V1R activity in amygdala VP regulated aggression and Morrison et al. (2016) anxiety shift after adolescent exposure to anabolic-androgenic steroids

V1bR suppression diminished anxiety-like behavior

dKO mice were less anxious Shimizu et al. (2018) than WT mice Amygdala, anterior insula

Marginal effects on amygdala only in men

Improved memory both for happy and angry information in men VP: 20 IU Intranasally Reduced recognition of Placebo negative emotion in men First step: orally given twice Amygdala, right SRX246 attenuated brain First step: a day temporoparietal junction, activity toward angry V1aR antagonist: precuneus, anterior faces 120 mg  2/day  7 days Second step: intranasally cingulate, and putamen (SRX246) Placebo Second step: VP: 40 IU Placebo Intranasally

Dannenhoffer et al. (2018)

Chen et al. (2016)

Guastella et al. (2010)

Uzefovsky et al. (2012) Lee et al. (2013)

Continued

Table 5.2 Continued Models and species

Doses

Routes

Men adults

VP: / Placebo

Intranasally

Men and women adults

/

Examine the repeat lengths in the intron of V1aR gene

Men and women adults

/

/ DNA extraction and genotyping for VPR1a. RS3 microsatellite

Brain regions

Effects

References

VP enhanced the corrugator Thompson et al. (2004) electromyogram responses evoked by emotionally neutral expressions, which was similar to that of the control group when facing angry expressions Men with a short form of Nishina et al. (2019) V1aR tended to split more money with the opponent. Both men and women with a short form of V1aR preferred to return money to their opponent that trusted them V1aR gene was related to Uzefovsky et al. (2015) cognitive empathy

Abbreviations: AVPA, V1aR antagonist; BNST, bed nucleus of the stria terminalis; CSF, cerebrospinal fluid; dKO, double knockout; DNA, deoxyribonucleic acid; HPA axis, hypothalamic–pituitary–adrenal axis; ICV, intracerebroventricular; IU, international unit; LAH, latero-anterior hypothalamic; mPFC, medial prefrontal cortex; mRNA, messenger-ribonucleic acid; OTR, oxytocin receptor; PVN, hypothalamic paraventricular nucleus; shRNA, short hairpin RNA; V1aR, vasopressin 1a receptor; V1bR, vasopressin 1a receptor; VP, vasopressin; VPR, vasopressin receptor.

SEX DIFFERENCES OF OXYTOCIN AND VASOPRESSIN IN SOCIAL BEHAVIORS

75

Fig. 5.1. Oxytocin (OT) and vasopressin (VP) regulate partner preference in both sexes. Notes: A mouse/rat was treated with OT/VP and then cohabitated with a partner. After cohabitation, it went through a partner preference test in which the familiar partner and the stranger were the same sex. (A) The mouse/rat was female and treated with OT. When it was cohabitated with an opposite-sex (male) partner, it preferred the familiar male rather than the stranger. When it was cohabitated with a same-sex (female) partner, it preferred the familiar female rather than the stranger. When it was cohabitated with an object, it preferred the familiar one rather than the novel one. (B) The mouse/rat was male and treated with OT. When it was cohabitated with an opposite-sex (female) partner, it preferred the familiar female rather than the stranger. When it was cohabitated with a same-sex (male) partner, it preferred the familiar male rather than the stranger. When it was cohabitated with an object, it preferred the familiar one rather the novel one. (C) The mouse/rat was female and treated with VP. It was cohabitated with an opposite-sex (male) partner, and it preferred the familiar male rather than the stranger. (D) The mouse/rat was male and treated with VP. It was cohabitated with an opposite-sex (female) partner, and it preferred the familiar female rather than the stranger. Reproduced from Lu Q, Lai J, Du Y et al. (2019). Sexual dimorphism of oxytocin and vasopressin in social cognition and behavior. Psychol Res Behav Manag 12: 337–349.

OT treatment they were more easily attracted by male faces regardless of whether it was a happy or angry facial expression. However, when they were shown female faces, only the happy expression was attractive to these

homosexual men (Thienel et al., 2014). No study has been performed on homosexual women. In conclusion, OT promoted familiar partner preference independent of the partner sex, an effect that is more

76

Q. LU AND S. HU

obvious in females. The relevant brain regions identified in current studies are different in females and males and are listed in Table 5.1.

Memory modulation Memory is an important capacity of the brain to encode, store, and retrieve information and is closely associated with hippocampal function. Without memory, language never develops, and learning is impossible. OT is one of the factors that regulate memory. Most studies of how OT regulates memory were performed in males, while only a few studies used nonpregnant females. This is because females experience larger fluctuations in sex hormones, along with the menstrual/estrous cycle and pregnancy. Changes in these hormones are also closely associated with memory performance (Koebele and Bimonte-Nelson, 2017). As a result, it is in males that most current studies now report positive effects of OT on social memory. Acute OT administration in macaques increased working memory in males but not in females (Simpson et al., 2017), while in Wistar rats, it improved stressimpaired memory in both males and females (Dayi et al., 2015). Uncontrolled stress is harmful to hippocampal plasticity and memory. Male Sprague Dawley rats showed an improvement in stress-induced memory impairment after receiving an intranasal OT intervention (Lee et al., 2015). However, female Sprague Dawley rats were not used in this study. Human studies only included males and showed that OT facilitated recall of positive social affiliation memories (Cardoso et al., 2014) and promoted memory encoding and retrieval of negative social stimuli (Weigand et al., 2013). However, studies in this area have not come to a conclusion. Some studies

reported contrary results. For example, intranasal OT-treated men suffered impairment of visual object recall (Herzmann et al., 2012). This research group continued to compare recollection judgments between men and women after OT intervention. They found that OT treatment impaired memory recollection in men but not in women (Herzmann et al., 2013). Study designs in this field were variable. Memory has four phases, including memorizing, retention, recall, and recognition. Memory is clinically divided into three subtypes: immediate, recent, and remote memory. Other divisions include long-term memory, short-term memory, working memory, and spatial memory. To obtain more persuasive and reproducible results, multiple series of well-designed studies are needed to systematically clarify how OT influences the different types of memory, from the phenomena to the molecular mechanisms in both sexes.

Emotion regulation Animals, and especially humans, have emotions facing social stimuli, with anxiety, happiness, and anger being the most common. These emotions switch from one to another in daily life and are influenced by OT (Fig. 5.2). In animal experiments, OT relieves anxious states in both sexes. Most animal studies in this field involved single-sex animals with males more frequently used, while only a few studies included both sexes. The brain regions receiving OT injection also differed between studies. Although these studies had different designs and mechanisms of action, they came to the same phenomenologic conclusions. Injecting OT into the PVN and periamygdala regions led to a decrease of amygdala activity and anxiety in male rats and mice (Blume et al., 2008;

Fig. 5.2. Summary of current opinions on how oxytocin (OT)/vasopressin (VP) influence emotions in males and females. Notes: This figure summarizes the mainstream opinions according to current studies; a few had conflicting results, which were unable to be shown here. (A) For females, central OT supplement can relieve anxious and angry emotions. (B) For males, central OT supplement relieves anxiety and anger and increases happiness perception. (C) Higher central VP level has an anxiogenic effect in females. (D) Higher central VP level has an anxiogenic effect in males. Reproduced from Lu Q, Lai J, Du Y et al. (2019). Sexual dimorphism of oxytocin and vasopressin in social cognition and behavior. Psychol Res Behav Manag 12: 337–349.

SEX DIFFERENCES OF OXYTOCIN AND VASOPRESSIN IN SOCIAL BEHAVIORS

77

Peters et al., 2014). ICV infusion of OT also remitted anxiety in female rats (de Jong et al., 2014). OT infusion into the prelimbic region of the medial prefrontal cortex (mPFC) reduced anxiety in both male and female rats (Sabihi et al., 2014). The dosage and treatment duration of OT are also important in determining the study outcome. Peters et al. administered a chronic low-dose (1 ng/h for 19 days) or high-dose (10 ng/h for 15 days) of OT to male mice. Chronic low-dose administration prevented the male mice from developing hyperanxiety, while the high-dose administration promoted anxiogenic phenotype development (Peters et al., 2014). However, these studies did not record the posttreatment concentration of central OT. More detailed study designs are required to clarify how CNS OT concentrations modulate anxiety-related behaviors and how the intervention course influences the maintenance of behavioral changes. Human studies reported different results compared to animal experiments. These are listed in Table 5.1. Studies on how OT modulates happiness and anger are few and are listed in Table 5.1 (also see Fig. 5.2). To sum up, OT reduces anxiety in both sexes, reduces anger perception in males, increases anger perception in females, and increases happiness perception in males. But there are exceptions in peripartum women, as is elaborated on in section “Effects of OT and VP on Social Cognition of Women During Pregnancy, Birth, and Lactation.”

dictator game (Fig. 5.3), prisoner’s dilemma game, the envelope task (Fig. 5.4), and the monetary game (Fig. 5.5). The ultimatum game and dictator game were performed using male participants. In the ultimatum game, OT treatment made the decision maker more generous, as exhibited by transferring more money to the other player in the group. In comparison, the OT-intervened decision maker gave relatively less money to the other member in the dictator game. In the dictator game, there was no significant difference between the OT group and control group (Zak et al., 2007). From this study, we infer that OT promotes decision-making in a context-dependent manner in loss and benefit assessment. The decision maker showed greater generosity in the ultimatum game. The relationships between the two players in these two paradigms were different. In the ultimatum game, the two players maintained a gambling relationship. In the dictator game, the rights were unbalanced between the two players, because the decision maker had full discretion. Several other studies used the prisoner’s dilemma game, the envelope task (Mikolajczak et al., 2010; Feng et al., 2015), and the monetary game (Kosfeld et al., 2005) to test trust-related behaviors. Designs and OT detection methods in these studies were similar, but the criteria were controversial. Altogether, OT increased feelings of trustworthiness in others in men. Few studies in this field included women.

Trust-related behaviors

SEXUALLY DIMORPHIC INFLUENCES OF VP ON SOCIAL COGNITION

People with higher central OT levels tend to show more trust in others. Several paradigms of trust have been used in the experiments, including the ultimatum game, the

Studies of VP effects on social cognition are fewer than those of OT. However, the anxiogenic effect of VP is

Fig. 5.3. Illustration of the ultimatum game and dictator game. Notes: A and B are anonymous in these two games. In the ultimatum game, A gets $10 before the game starts. A proposes how to split the money. He can split the $10 evenly or offer B $2 and keep $8. B can either accept or reject the proposal. If B accepts, the money is divided according to the proposal. If B rejects, neither A nor B achieves anything. The dictator game is a variation of the ultimatum game. A is the money keeper and the only decision maker of the money distribution. Reproduced from Lu Q, Lai J, Du Y et al. (2019). Sexual dimorphism of oxytocin and vasopressin in social cognition and behavior. Psychol Res Behav Manag 12: 337–349.

78

Q. LU AND S. HU

Fig. 5.4. Illustration of the prisoner’s dilemma and envelope tasks. Notes: In the prisoner’s dilemma game, A and B choose to cooperate with each other or defect sequentially. A chooses first, based on which B gives their choice. There are four potential outcomes, which are associated with different payoffs. In the envelope task, the participants put private information into an envelope anonymously and give it to a stranger. They can choose to seal the envelope or not. The openness degree of the envelopes is the assessment criteria, with total open indicating utmost trust. Reproduced from Lu Q, Lai J, Du Y et al. (2019). Sexual dimorphism of oxytocin and vasopressin in social cognition and behavior. Psychol Res Behav Manag 12: 337–349.

Fig. 5.5. Illustration of the monetary game. Notes: There are two steps in this game. A is the investor and B is the trustee in this game. Each of them receives 12 monetary units (MUs) before this game starts. In the first step, A is free to transfer 0, 4, 8, or 12 MUs to B. Then B gets tripled MUs that A sent. In the second step, B has the option of splitting any amount of MUs back to A, between 0 and his total MUs. Reproduced from Lu Q, Lai J, Du Y et al. (2019). Sexual dimorphism of oxytocin and vasopressin in social cognition and behavior. Psychol Res Behav Manag 12: 337–349.

widely studied in depth. Males are also more frequently used in studies than females. The current studies are discussed in the following paragraphs. Experimental models, intervention methods, VP dosages, relevant brain regions, and experimental results are listed in Table 5.2. VP, like OT, also promotes a preference for familiar opposite-sex partners in both males and females (Fig. 5.1) mainly by V1a receptor (V1aR) signaling in the ventral pallidum (Winslow et al., 1993; Lim and Young, 2004). V1aR overexpression in the ventral pallidum promoted males’ preference for a familiar partner. This effect was shown in both monogamous (Pitkow et al., 2001; Lim et al., 2004) and socially promiscuous voles (Lim et al., 2004). This effect was also proven by downregulating

the V1aR activity, resulting in a weakened partner preference of both sexes (Cho et al., 1999; Donaldson et al., 2010; Barrett et al., 2013). Adult male Callicebus cupreus receiving a high dose of VP (80 IU) intranasally increased the frequency of partner contact (Jarcho et al., 2011). While the preceding cited studies all examined intervention in adult animals, Simmons et al. (2017) explored how juvenile intranasal administration of VP influences adult partner preference in both sexes. Their results showed that a 1-week intranasal VP treatment at the doses of 0.5 and 5.0 IU/kg on juveniles blocked partner preference formation when the male voles were sexually mature (Simmons et al., 2017). This was not seen in sexually mature females and was contrary to

SEX DIFFERENCES OF OXYTOCIN AND VASOPRESSIN IN SOCIAL BEHAVIORS those study results using an instant VP intervention (Simmons et al., 2017). This indicates that VP adjusts immature brain development, leading to long-term effects on behavior. How long-term intervention influences mature cerebral function is also a question that remains to be solved. Current studies have shown that VP has a memoryenhancing effect (Mishima et al., 2001; Barsegyan et al., 2015; Hicks et al., 2015). The most relevant brain region for this effect is the hippocampus (Alescio-Lautier et al., 2000). The primary active metabolite of VP-(1–9), VP-(4–9) (Dietrich and Allen, 1997), and its derivative, VP-(4–8) (Vawter et al., 1997), play a role in memory modulation. Most studies only included males, and few reported any sex differences of this memory-enhancing effect. However, in general, men have better spatial memory than women: for instance, a better direction sense or geometric mathematics scores. This goes together with a VP system that is more abundant in males than in females (Dumais and Veenema, 2016). The spatial memoryenhancing effect of VP is stronger in males (Dumais and Veenema, 2016). There are also studies of some disease models, without distinguishing any sex difference, concerning memory impairment, such as spinocerebellar ataxia type 3 (SCA3), vascular dementia (VD), and schizophrenia. VP strengthened, while the VPR antagonist weakened, spatial memory in SCA3 mice dose dependently (Jiang et al., 2017). There was a decrease of VP in the PVN of VD rats and this was hypothesized to lead to an impairment of cognitive function. Indeed, injecting VP into the PVN ameliorated the impairment (Li et al., 2017). There was also a decrease of VP in cerebrospinal fluid (CSF) in first-episode schizophrenic patients. A 4–12-day consecutive treatment of VP kept an elevated level of VP in the CNS and improved these patients’ memory performance (Geng et al., 2017). VP also influences emotions like anxiety, happiness, and anger (see Fig. 5.2 and Table 5.2). Extensive studies have confirmed the anxiogenic effect of VP mediated by V1R. V1bR antagonist injected i.p. diminished anxiety in males but not females (Brunnlieb et al., 2013). Central infusion of V1aR antagonist into the PVN of lactating rats decreased anxiety (Bayerl et al., 2016). V1bR suppression and OTR activation in the hypothalamus reversed anxiety-like behavior caused by adolescent ethanol exposure in males but not females (Dannenhoffer et al., 2018). Knockout of V1aR and V1bR genes in male mice also reduced anxiety-like behavior (Shimizu et al., 2018). A human study with intranasal VP treatment increased the activity of V1R in the amygdala of young men, which indicated an increase of anxiety (Brunnlieb et al., 2013). However, blocking V1aR signaling in the anterior hypothalamus enhanced anxiety and suppressed offensive aggression in male hamsters exposed to anabolic/androgenic steroids

79

during adolescence (Morrison et al., 2016). In addition, activating V1aR enhanced offensive aggression of these hamsters (Morrison et al., 2016). This indicates additional androgenic steroids influenced V1aR expression in anterior hypothalamus and anxiety-like behavior during the period of neural development. Interestingly, prenatal stress led to a higher plasma VP level, while an additional exposure to morphine prenatally resulted in a lower VP concentration in female offspring compared to male offspring. This indicates that morphine also influences VP anxiety-related neural development in a sex-dependent way (Nakhjiri et al., 2017). Socially isolated rats showed an increase of plasma VP level and a decrease of OT level independent of sex while anxiety was induced. Agomelatine, an antidepressant, can reverse these phenomena. That is, agomelatine reversed the elevated plasma VP concentration and suppressed anxiety in both sexes and partially elevated the previously decreased OT concentration in female rats (Harvey et al., 2019). This suggests that regulating VP and OT balance may be one of the mechanisms whereby agomelatine relieves anxiety. VP also regulates cognitive empathy formation, happiness and anger perception, and trust-related behaviors. Studies in these fields are few and evidence is not sufficient for reliable conclusions. Representative studies are listed in Table 5.2.

EFFECTS OF OT AND VP ON SOCIAL COGNITION OF WOMEN DURING PREGNANCY, BIRTH, AND LACTATION OT in peripartum period OT plays an important role in labor. Sometimes women need synthetic OT to enhance uterine contractions during parturition. In addition, OT synthesis is upregulated during late pregnancy and released with impulses during lactation (Russell et al., 2001; Prevost et al., 2014). The mother’s cognitive and behavioral changes due to the change in OT level are a preparation for better care for the offspring. Abnormal release of OT is associated with a series of peripartum symptoms, among which changes in maternal mood, memory, and cognition are the most studied. The estimated rate of postpartum depression in mothers is between 9% and 17% in Western countries (Robertson et al., 2004; Yelland et al., 2010). The mainstream idea holds that higher plasma or central OT level helps to relieve depressive and anxious symptoms. However, study results on how OT influences peripartum depression are inconsistent (Gu et al., 2016). One of the reasons may be that the OT levels in the CSF and plasma in primates are anatomically and functionally separate (Amico et al., 1990). Many studies in this field

80

Q. LU AND S. HU

were performed on peripartum women and used blood samples to test OT level. Only a few studies dealt with pregnant or lactating mothers receiving intranasal OT. While healthy women had a remarkable increase in plasma OT level from the 35th week of pregnancy to the 6th month postpartum, postnatally depressed mothers had a decreased plasma OT level between the 38th week of pregnancy and the 2nd day postpartum (Jobst et al., 2016). Anxiety in female rodents was increased during pregnancy and decreased in the postpartum period (Lonstein et al., 2014). In psychosocially stressed mothers, higher plasma OT level was correlated with fewer depressive symptoms at 12–14 weeks of gestation and 7–9 weeks postpartum (Zelkowitz et al., 2014). The mechanism of this phenomenon is unclear and one of the hypotheses is that OT acts as a buffer against maternal peripartum stress (Cox et al., 2015). The observation that OTR DNA methylation plays a role in human postpartum depression needs further studies (Kimmel et al., 2016). Salivary OT level was negatively associated with postpartum fatigue but not depression (Shishido et al., 2019). This review summarized that OT has an anxiolytic effect on females, which is consistent with the anxiolytic effect of elevated central OT in postpartum animals. For example, in an fMRI study, lactating rats that received ICV OT were exposed to predator scent, as an unconditioned fear stimulus. They experienced an MR scanning 45–60 min later that showed that, concomitant with an anxiolytic effect of OT, many brain regions involved had changes in activity. The BNST, anterior cingulate, and perirhinal area showed increased activity, while the secondary motor cortex, mamillary bodies, prelimbic prefrontal cortex, gustatory cortex, orbital cortex, and anterior olfactory nucleus showed decreased activity (Febo et al., 2009). However, there are also some conflicting studies. OTR knockout mice had the same level of anxiety-like behavior as the wildtype after parturition, which indicates anxiety may not be directly related to OT (Rich et al., 2014). On receiving an intranasal OT spray, postnatally depressed mothers reported sadder mood but better relationship with their baby (Mah et al., 2013). Exposure to synthetic OT in peripartum increased the risk of suffering from depressive or anxiety disorder within the first year postpartum (Gu et al., 2016; KrollDesrosiers et al., 2017). There are also data that do not support the depressive and anxious suppressive effects of OT. For example, synthetic OT is given intravenously to facilitate parturition and given intramuscularly to prevent postpartum hemorrhage in some women in labor. Two months later, these women had greater depressive and anxious symptoms (Gu et al., 2016). Mother rabbits had significantly reduced OTR density in the preoptic area (POA) and PFC during lactation

relative to late pregnancy period (Jimenez et al., 2015). This indicates that only specific brain regions increase OT sensitivity during the peripartum period, while the other brain areas do not. Besides OT, the changes in reproductive hormones and VP are drastic and complex in the female peripartum period. Future studies should take the other hormonal changes as variates into consideration. OT facilitates the onset of maternal behavior (Yoshihara et al., 2018), which includes maternal-fetal bonding, maternal protective behavior, and maternal memory. Elevated plasma OT levels in late pregnancy are related to increased maternal-fetal bonding (Levine et al., 2007) and theory of mind and decreased depressive behaviors (MacKinnon et al., 2014). Administration of intranasal OT, which increased central OT levels, strengthened the maternal protective behavior of postnatally depressed mothers toward their infant (Mah et al., 2015). Pregnancy, parturition, and lactation affect females’ memories. In particular, the phenomenon that maternally experienced females display a faster onset of bonding with their newborn babies is called “maternal memory.” In pregnant mice with maternal experience, an ICV injection of OT improves long-lasting spatial memory through the mitogen-activated protein kinases cascade (Tomizawa et al., 2003). OT release in the olfactory bulb (OB) at delivery may help ewes to recognize lamb odors via acetylcholine (Ach), noradrenaline (NA), and g-aminobutyric acid (GABA) release, which are pivotal in the formation of olfactory memory. Intriguingly, this kind of OT release was reduced in primiparous ewes relative to multiparous ewes, which may be a basis for the decreased modulation of Ach and NA release, and the longer period for maternally inexperienced ewes to selectively bond with their lambs (Levy et al., 1995). Parturitionexperienced mothers had indeed higher central OT levels than primipara, associated with a better maternal retrieval behavior (Lopatina et al., 2011). This difference was regulated by the transmembrane glycoprotein CD38 in the hypothalamus and pituitary (Lopatina et al., 2012). Besides the OB, also the hypothalamus, pituitary, and the NAcc shell contribute to maternal memory enhancement by increasing the OTR activity (D’Cunha et al., 2011). OT increased the mother’s perception of infant cuteness, which involved brain regions that manage face perception, emotion, attention, memory, empathy, theory of mind, reward, attachment, and control of motor responses (Luo et al., 2015). The neural activity of these brain areas was decreased in women with postpartum depression or anxiety (Luo et al., 2015). OT signaling in the mPFC was beneficial for postpartum mothers to improve their cognitive flexibility facing their pups (Albin-Brooks et al., 2017). Neonatal maternal separation influenced their future maternal behavior in adulthood.

SEX DIFFERENCES OF OXYTOCIN AND VASOPRESSIN IN SOCIAL BEHAVIORS For example, lactating female rats who suffered from maternal separation showed more maternal aggression and a decreased central OT neural activity (Veenema et al., 2007). In contrast, maternal separation-experienced males showed less aggression toward intruders (Veenema et al., 2007). All these studies suggest that the increase of OT release and its neural activity modulates mothers’ cognition and behavior so that the mothers’ care for their offspring is improved.

81

V1aR density was correlated with more postpartum licking/grooming behaviors (Curley et al., 2012). In general, like OT, VP and its receptors also contribute to an improved state for peripartum mothers to care for their offspring and is sensitive to early life stress, as reviewed recently (Kompier et al., 2019). This relates to, for example, an earlier onset of maternal care, continuation of maternal behavior, maternal aggression, and improved memory for guarding the pups.

DISCUSSION VP in peripartum period Though VP was studied more in males than in females, there are some studies on VP in maternal behavior and around the peripartum period. The most-studied behaviors are maternal care, maternal aggression, and maternal memory. V1aR plays a major role in these behaviors. VP promotes the onset and ongoing of maternal care via V1aR in the medial POA (mPOA) (Pedersen et al., 1994; Bosch and Neumann, 2012) and the PVN (Bayerl et al., 2016). Chronic VP infusion into the right lateral ventricle of lactating rats also promotes the onset of nursing and increases total maternal care (Coverdill et al., 2012). In addition, this intervention also decreased maternal aggression on day 3 of lactation of mothers experiencing chronic social stress (Coverdill et al., 2012). However, there are opposite results concerning the effect of central VP on maternal aggression, which may be related to the brain region involved and the presence of social stress. Increased expression of VP in the PVN (Bosch, 2011) and increased activity of V1aR in the BNST (Bosch and Neumann, 2012) were positively associated with maternal aggression. VP release in the CeA also facilitated maternal aggression in lactating rats (Bosch and Neumann, 2010). Moreover, V1aR in the PVN induced anxiety-like behaviors in lactating rats (Bayerl et al., 2016). OTR and V1aR genes are critical for the Bruce effect in pregnant mice (Wersinger et al., 2008). This effect relies on olfactory memory. Recently mated females terminate pregnancy when exposed to chemosensory cues of unfamiliar males (Wersinger et al., 2008). V1aR antagonist infused into the MeA impaired maternal memory in pregnant rats (Nephew and Bridges, 2008). Central VP deficiency led to maternal neglect and less depressive behavior in primiparous female rats, an effect in which the mPOA, PVN, amygdala, and shell of the Nacc are involved (Fodor et al., 2012). V1bR antagonist infusion into the lateral ventricle impaired maternal care but not maternal aggression in lactating rats (Bayerl et al., 2014). Specifically, in the dorsal lateral septum, both OTR and V1aR were positively associated with maternal care. Higher OTR density was associated with more frequent nursing and higher

This chapter discusses the socially cognitive features of OT and VP in relation to sex. The effects of OT and VP administration depend on the dosages, intervention durations, and life stages. The majority of the studies used a dose of 24 IU for OT and 40 IU for VP in human studies and the dose in animals was given according to weight. Most studies chose an acute intervention, while some studies involved long-term treatment. An acute intervention was usually applied in adults, whereas the long-term treatment was performed on juveniles that were tested in adulthood. Specifically, the behavioral paradigm plays an important role in guiding the results of the trust-related behaviors. To summarize, OT promotes familiar partner preference, strengthens memory, relieves anxiety, and increases trust. VP facilitates familiar partner preference, enhances memory, induces anxiety, and influences happiness/anger perception. Both OT and VP induce opposite-sex familiar partner preference in both sexes after acute administration. This effect of OT is more significant in females and is also applicable to same-sex partners and object preference. Long-term intervention of VP has contradictory results in males. OT and VP have a cooperative and balanced relationship in regulating the partner preference. Sexual orientation is the preference for sexual partner, with people identifying as homosexual preferring a same-sex partner as their mate. Homosexuality is similar to the phenomenon of OT- or VP-treated animals preferring a familiar same-sex partner rather than a new one. Whether this possible mechanism of same-sex partner preference is implicated in homosexuality needs further research. In addition, the VPR in the PVN and substantia nigra and the OTR in the PFC are related to the frequency of sexual interactions (Acevedo et al., 2019). Sexual satisfaction in pair-bonds was closely associated with the activation of many subcortical structures involved in, e.g., social bonding (ventral pallidum), reward and motivation (substantia nigra, ventral tegmental area and caudate), hormone control (hypothalamus), memory (hippocampus), and emotion (amygdala) (Acevedo et al., 2019). Further studies of OT, OTR, VP, and V1R in the postmortem homosexual brain would be helpful.

82

Q. LU AND S. HU

Estrogens and androgens take part in the regulation of OT and VP systems in either sex. Estrogens promote OTR (Sharma et al., 2019) and OT gene expression (Richard and Zingg, 1990). Testosterone directly inhibits OT expression via the androgen receptor (AR) (Dai et al., 2017). In addition, testosterone is converted into estradiol by the catalysis of aromatase, and estradiol has a positive effect on OT expression (Jirikowski et al., 2018). Moreover, androgen metabolites, such as 3b,17b-diol (3b-diol), and 5a-androstane, are able to upregulate OT expression via the estrogen receptor b (ERb) (Sharma et al., 2012; Hiroi et al., 2013). Though in males several metabolites of testosterone can upregulate OT expression, females have higher estrogen levels than males in physiologic conditions and estrogens have a stronger effect on increasing OT expression. Therefore we may infer that central OT expression in females is more abundant than in males. From this conclusion, we can explain why the familiar partner preference effect of OT is more prominent in females. However, this explanation is not sufficient to explain all phenomena, as some studies reported opposite results in some specific brain regions. In posterior BNST, OTR expression and binding rate were found to be higher in males than in females, which was positively correlated with ER and AR activation (Worley et al., 2019). As previously mentioned, BNST is also involved in the formation of familiar partner preference, but only in females (Jia et al., 2008). The results of these two studies seem to be in conflict with each other. Further studies are needed to solve this problem. The OT expression level runs parallel with the plasma estrogen fluctuation across the female estrous cycle (Sarkar et al., 1992). Estrogen is also an important factor in regulating memory. Most findings support the memory-enhancing effect of OT and VP. As mentioned earlier, OT is significantly higher in female brain regions, such as MeA, due to the higher estrogen concentration. This is consistent with the finding that OTR gene expression in the MeA is essential for the maintenance of normal social recognition in females (Choleris et al., 2007). Other reproduction-related hormones also influence memory. For example, gonadotropin-releasing hormone (GnRH) receptors are also expressed in cerebral areas related to memory, such as in the hippocampus and amygdala (Hough et al., 2017). Applying GnRH agonist to male sheep chronically is beneficial for the retention of long-term spatial memory (Hough et al., 2017). However, in older men with low testosterone, who have a high GnRH level and age-associated poor memory, Resnick and his colleagues observed no memory-improvement effect following a 1-year testosterone treatment (Resnick et al., 2017). This contradicts our conclusions that VP has a memory-enhancing effect on males with respect to

the promotion effect of testosterone on VP expression in the male brain. A reason for these discrepancies may be that not all memory types are significantly influenced by sex hormones, OT, or VP. Different memory types are tested using different paradigms. Also, the memoryrelated brain regions in man and animals may be different. Estrogen deficiency may result in depression- and anxiety-like behaviors and hippocampal inflammation in mice via the nucleotide binding and oligomerization domain-like receptor family pyrin domain-containing 3 (NLRP3) signaling (Xu et al., 2016). This pathologic change translates into a series of symptoms in menopausal women, which is called climacteric syndrome. It includes vasomotor symptoms, somatic symptoms, and psychologic symptoms. The psychologic symptoms include anxiety, depression, and mood swings. In fact, estrogen regulates anxiety in different ways depending on ER subtypes: ERa is anxiogenic, ERb is anxiolytic, and the G-protein-coupled ER (GPR30) can either decrease or increase anxious behaviors (Borrow and Handa, 2017). Despite the promoting effect of estrogen on OTR expression in females (Sharma et al., 2019) and the inhibiting effect of testosterone on OT expression in males (Dai et al., 2017), we may conclude that males experience a more significant anxiolytic effect. Previous studies reported discrepant results on how testosterone influences anxiety, with some studies promoting anxiety (McHenry et al., 2014) while in others anxiety was not affected by testosterone (Filova et al., 2015). Current studies support the inference that VP induces anxiety in both sexes, although the role of testosterone in both sexes is not clear. Drugs like morphine, alcohol, and agomelatine, as well as social isolation, also influence sex hormone-related anxiety regulation by OT and VP. So, besides using antidepressant treatment, we should also encourage patients with anxiety disorders to take an active part in collective activities to avoid isolation. Maternal stress increases plasma VP level in female offspring while morphine decreases it, which means that morphine can reduce anxiety caused by VP. However, morphine has negative effects on the activity of offspring and VP-associated blood pressure control. Hence it is important for women to avoid stress during their pregnancy (Nakhjiri et al., 2017). Women during the peripartum period experience drastic hormonal fluctuations, including reproductive hormones, OT, and VP. Although these changes cause some annoying symptoms like postpartum depression, anxiety, and more aggression toward others, they are all beneficial for caring for the offspring. In humans, severe peripartum symptoms should be addressed with interventions, to avoid poor outcomes of mothers, protecting them from aggressive behaviors toward others, deep depression, or even suicide. Antidepressants can

SEX DIFFERENCES OF OXYTOCIN AND VASOPRESSIN IN SOCIAL BEHAVIORS be a choice at present (Becker et al., 2016). Future studies should focus on how antidepressants interact with reproductive hormones, OT, and VP in remitting these negative symptoms. In conclusion, OT, VP, and sex hormones interact in a complex network, with influences from many social behaviors. Precise research on the interaction of OT/VP and sex hormones is a meaningful and promising avenue for the fields of sexual orientation, affinity regulation, memory manipulation, antidepressive treatment, anxiety therapy, and other social behaviors.

REFERENCES Aarde SM, Jentsch JD (2006). Haploinsufficiency of the arginine-vasopressin gene is associated with poor spatial working memory performance in rats. Horm Behav 49: 501–508. Acevedo BP, Poulin MJ, Geher G et al. (2019). The neural and genetic correlates of satisfying sexual activity in heterosexual pair-bonds. Brain Behav 9: e01289. Albin-Brooks C, Nealer C, Sabihi S et al. (2017). The influence of offspring, parity, and oxytocin on cognitive flexibility during the postpartum period. Horm Behav 89: 130–136. Alescio-Lautier B, Paban V, Soumireu-Mourat B (2000). Neuromodulation of memory in the hippocampus by vasopressin. Eur J Pharmacol 405: 63–72. Althammer F, Grinevich V (2017). Diversity of oxytocin neurons: beyond magno- and parvocellular cell types? J Neuroendocrinol 30: e12549. Amico JA, Challinor SM, Cameron JL (1990). Pattern of oxytocin concentrations in the plasma and cerebrospinal fluid of lactating rhesus monkeys (Macaca mulatta): evidence for functionally independent oxytocinergic pathways in primates. J Clin Endocrinol Metab 71: 1531–1535. Bales KL, Carter CS (2003). Developmental exposure to oxytocin facilitates partner preferences in male prairie voles (Microtus ochrogaster). Behav Neurosci 117: 854–859. Bales KL, Perkeybile AM, Conley OG et al. (2013). Chronic intranasal oxytocin causes long-term impairments in partner preference formation in male prairie voles. Biol Psychiatry 74: 180–188. Barrett CE, Keebaugh AC, Ahern TH et al. (2013). Variation in vasopressin receptor (Avpr1a) expression creates diversity in behaviors related to monogamy in prairie voles. Horm Behav 63: 518–526. Barsegyan A, Atsak P, Hornberger WB et al. (2015). The vasopressin 1b receptor antagonist A-988315 blocks stress effects on the retrieval of object-recognition memory. Neuropsychopharmacology 40: 1979–1989. Bayerl DS, Klampfl SM, Bosch OJ (2014). Central V1b receptor antagonism in lactating rats: impairment of maternal care but not of maternal aggression. J Neuroendocrinol 26: 918–926. Bayerl DS, Honig JN, Bosch OJ (2016). Vasopressin V1a, but not V1b, receptors within the PVN of lactating rats mediate maternal care and anxiety-related behaviour. Behav Brain Res 305: 18–22.

83

Becker M, Weinberger T, Chandy A et al. (2016). Depression during pregnancy and postpartum. Curr Psychiatry Rep 18: 32. Beery AK, Zucker I (2010). Oxytocin and same-sex social behavior in female meadow voles. Neuroscience 169: 665–673. Beery AK, Loo TJ, Zucker I (2008). Day length and estradiol affect same-sex affiliative behavior in the female meadow vole. Horm Behav 54: 153–159. Beery AK, Routman DM, Zucker I (2009). Same-sex social behavior in meadow voles: multiple and rapid formation of attachments. Physiol Behav 97: 52–57. Blume A, Bosch OJ, Miklos S et al. (2008). Oxytocin reduces anxiety via ERK1/2 activation: local effect within the rat hypothalamic paraventricular nucleus. Eur J Neurosci 27: 1947–1956. Borrow AP, Handa RJ (2017). Estrogen receptors modulation of anxiety-like behavior. Vitam Horm 103: 27–52. Bosch OJ (2011). Maternal nurturing is dependent on her innate anxiety: the behavioral roles of brain oxytocin and vasopressin. Horm Behav 59: 202–212. Bosch OJ, Neumann ID (2010). Vasopressin released within the central amygdala promotes maternal aggression. Eur J Neurosci 31: 883–891. Bosch OJ, Neumann ID (2012). Both oxytocin and vasopressin are mediators of maternal care and aggression in rodents: from central release to sites of action. Horm Behav 61: 293–303. Brunnlieb C, Munte TF, Tempelmann C et al. (2013). Vasopressin modulates neural responses related to emotional stimuli in the right amygdala. Brain Res 1499: 29–42. Buijs RM (1978). Intra- and extrahypothalamic vasopressin and oxytocin pathways in the rat. Pathways to the limbic system, medulla oblongata and spinal cord. Cell Tissue Res 192: 423–435. Cardoso C, Orlando MA, Brown CA et al. (2014). Oxytocin and enhancement of the positive valence of social affiliation memories: an autobiographical memory study. Soc Neurosci 9: 186–195. Carter CS, Getz LL (1993). Monogamy and the prairie vole. Sci Am 268: 100–106. Cavanaugh J, Mustoe AC, Taylor JH et al. (2014). Oxytocin facilitates fidelity in well-established marmoset pairs by reducing sociosexual behavior toward opposite-sex strangers. Psychoneuroendocrinology 49: 1–10. Chen X, Hackett PD, DeMarco AC et al. (2016). Effects of oxytocin and vasopressin on the neural response to unreciprocated cooperation within brain regions involved in stress and anxiety in men and women. Brain Imaging Behav 10: 581–593. Cho MM, DeVries AC, Williams JR et al. (1999). The effects of oxytocin and vasopressin on partner preferences in male and female prairie voles (Microtus ochrogaster). Behav Neurosci 113: 1071–1079. Choleris E, Little SR, Mong JA et al. (2007). Microparticlebased delivery of oxytocin receptor antisense DNA in the medial amygdala blocks social recognition in female mice. Proc Natl Acad Sci U S A 104: 4670–4675.

84

Q. LU AND S. HU

Cohen D, Perry A, Gilam G et al. (2017). The role of oxytocin in modulating interpersonal space: a pharmacological fMRI study. Psychoneuroendocrinology 76: 77–83. Coverdill AJ, McCarthy M, Bridges RS et al. (2012). Effects of chronic central arginine vasopressin (AVP) on maternal behavior in chronically stressed rat dams. Brain Sci 2: 589–604. Cox EQ, Stuebe A, Pearson B et al. (2015). Oxytocin and HPA stress axis reactivity in postpartum women. Psychoneuroendocrinology 55: 164–172. Curley JP, Jensen CL, Franks B et al. (2012). Variation in maternal and anxiety-like behavior associated with discrete patterns of oxytocin and vasopressin 1a receptor density in the lateral septum. Horm Behav 61: 454–461. Cushing BS, Carter CS (2000). Peripheral pulses of oxytocin increase partner preferences in female, but not male, prairie voles. Horm Behav 37: 49–56. Cushing BS, Martin JO, Young LJ et al. (2001). The effects of peptides on partner preference formation are predicted by habitat in prairie voles. Horm Behav 39: 48–58. Dai D, Li QC, Zhu QB et al. (2017). Direct involvement of androgen receptor in oxytocin gene expression: possible relevance for mood disorders. Neuropsychopharmacology 42: 2064–2071. Dal Monte O, Noble PL, Turchi J et al. (2014). CSF and blood oxytocin concentration changes following intranasal delivery in macaque. PLoS One 9: e103677. Dannenhoffer CA, Kim EU, Saalfield J et al. (2018). Oxytocin and vasopressin modulation of social anxiety following adolescent intermittent ethanol exposure. Psychopharmacology (Berl) 235: 3065–3077. Dayi A, Cetin F, Sisman AR et al. (2015). The effects of oxytocin on cognitive defect caused by chronic restraint stress applied to adolescent rats and on hippocampal VEGF and BDNF levels. Med Sci Monit 21: 69–75. D’Cunha TM, King SJ, Fleming AS et al. (2011). Oxytocin receptors in the nucleus accumbens shell are involved in the consolidation of maternal memory in postpartum rats. Horm Behav 59: 14–21. de Jong TR, Beiderbeck DI, Neumann ID (2014). Measuring virgin female aggression in the female intruder test (FIT): effects of oxytocin, estrous cycle, and anxiety. PLoS One 9: e91701. de Vries GJ, Miller MA (1998). Anatomy and function of extrahypothalamic vasopressin systems in the brain. Prog Brain Res 119: 3–20. Dietrich A, Allen JD (1997). Vasopressin and memory. I. The vasopressin analogue AVP4-9 enhances working memory as well as reference memory in the radial arm maze. Behav Brain Res 87: 195–200. Domes G, Sibold M, Schulze L et al. (2013a). Intranasal oxytocin increases covert attention to positive social cues. Psychol Med 43: 1747–1753. Domes G, Steiner A, Porges SW et al. (2013b). Oxytocin differentially modulates eye gaze to naturalistic social signals of happiness and anger. Psychoneuroendocrinology 38: 1198–1202.

Domes G, Normann C, Heinrichs M (2016). The effect of oxytocin on attention to angry and happy faces in chronic depression. BMC Psychiatry 16: 92. Donaldson ZR, Young LJ (2008). Oxytocin, vasopressin, and the neurogenetics of sociality. Science 322: 900–904. Donaldson ZR, Spiegel L, Young LJ (2010). Central vasopressin V1a receptor activation is independently necessary for both partner preference formation and expression in socially monogamous male prairie voles. Behav Neurosci 124: 159–163. Duclot F, Wang H, Youssef C et al. (2016). Trichostatin A (TSA) facilitates formation of partner preference in male prairie voles (Microtus ochrogaster). Horm Behav 81: 68–73. Dumais KM, Veenema AH (2016). Vasopressin and oxytocin receptor systems in the brain: sex differences and sex-specific regulation of social behavior. Front Neuroendocrinol 40: 1–23. Eisenberg Y, Murad S, Casagrande A et al. (2019). Oxytocin alterations and neurocognitive domains in patients with hypopituitarism. Pituitary 22: 105–112. Evans S, Shergill SS, Averbeck BB (2010). Oxytocin decreases aversion to angry faces in an associative learning task. Neuropsychopharmacology 35: 2502–2509. Febo M, Shields J, Ferris CF et al. (2009). Oxytocin modulates unconditioned fear response in lactating dams: an fMRI study. Brain Res 1302: 183–193. Feng C, Hackett PD, DeMarco AC et al. (2015). Oxytocin and vasopressin effects on the neural response to social cooperation are modulated by sex in humans. Brain Imaging Behav 9: 754–764. Filova B, Malinova M, Babickova J et al. (2015). Effects of testosterone and estradiol on anxiety and depressive-like behavior via a non-genomic pathway. Neurosci Bull 31: 288–296. Fodor A, Klausz B, Pinter O et al. (2012). Maternal neglect with reduced depressive-like behavior and blunted c-fos activation in Brattleboro mothers, the role of central vasopressin. Horm Behav 62: 539–551. Frank E, Landgraf R (2008). The vasopressin system—from antidiuresis to psychopathology. Eur J Pharmacol 583: 226–242. Geng CH, Wang C, Yang J et al. (2017). Arginine vasopressin improves the memory deficits in Han Chinese patients with first-episode schizophrenia. Peptides 97: 8–15. Gimpl G, Fahrenholz F (2001). The oxytocin receptor system: structure, function, and regulation. Physiol Rev 81: 629–683. Grillon C, Krimsky M, Charney DR et al. (2013). Oxytocin increases anxiety to unpredictable threat. Mol Psychiatry 18: 958–960. Gu V, Feeley N, Gold I et al. (2016). Intrapartum synthetic oxytocin and its effects on maternal well-being at 2 months postpartum. Birth 43: 28–35. Guastella AJ, Kenyon AR, Alvares GA et al. (2010). Intranasal arginine vasopressin enhances the encoding of happy and angry faces in humans. Biol Psychiatry 67: 1220–1222.

SEX DIFFERENCES OF OXYTOCIN AND VASOPRESSIN IN SOCIAL BEHAVIORS Harvey BH, Regenass W, Dreyer W et al. (2019). Social isolation rearing-induced anxiety and response to agomelatine in male and female rats: role of corticosterone, oxytocin, and vasopressin. J Psychopharmacol 33: 640–646. Haussler HU, Jirikowski GF, Caldwell JD (1990). Sex differences among oxytocin-immunoreactive neuronal systems in the mouse hypothalamus. J Chem Neuroanat 3: 271–276. Herzmann G, Young B, Bird CW et al. (2012). Oxytocin can impair memory for social and non-social visual objects: a within-subject investigation of oxytocin’s effects on human memory. Brain Res 1451: 65–73. Herzmann G, Bird CW, Freeman M et al. (2013). Effects of oxytocin on behavioral and ERP measures of recognition memory for own-race and other-race faces in women and men. Psychoneuroendocrinology 38: 2140–2151. Hicks C, Ramos L, Reekie TA et al. (2015). WAY 267,464, a non-peptide oxytocin receptor agonist, impairs social recognition memory in rats through a vasopressin 1A receptor antagonist action. Psychopharmacology (Berl) 232: 2659–2667. Hiroi R, Lacagnina AF, Hinds LR et al. (2013). The androgen metabolite, 5a-androstane-3b,17b-diol (3b-diol), activates the oxytocin promoter through an estrogen receptor-b pathway. Endocrinology 154: 1802–1812. Hodgson RA, Mullins D, Lu SX et al. (2014). Characterization of a novel vasopressin V1b receptor antagonist, V1B-30N, in animal models of anxiety-like and depression-like behavior. Eur J Pharmacol 730: 157–163. Hough D, Bellingham M, Haraldsen IRH et al. (2017). Spatial memory is impaired by peripubertal GnRH agonist treatment and testosterone replacement in sheep. Psychoneuroendocrinology 75: 173–182. Insel TR, Hulihan TJ (1995). A gender-specific mechanism for pair bonding: oxytocin and partner preference formation in monogamous voles. Behav Neurosci 109: 782–789. Jarcho MR, Mendoza SP, Mason WA et al. (2011). Intranasal vasopressin affects pair bonding and peripheral gene expression in male Callicebus cupreus. Genes Brain Behav 10: 375–383. Jia R, Tai F, An S et al. (2008). Neonatal manipulation of oxytocin influences the partner preference in mandarin voles (Microtus mandarinus). Neuropeptides 42: 525–533. Jiang HB, Du AL, Luo HY et al. (2017). Arginine vasopressin relates with spatial learning and memory in a mouse model of spinocerebellar ataxia type 3. Neuropeptides 65: 83–89. Jimenez A, Young LJ, Triana-Del Rı´o R et al. (2015). Neuroanatomical distribution of oxytocin receptor binding in the female rabbit forebrain: variations across the reproductive cycle. Brain Res 1629: 329–339. Jirikowski GF, Ochs SD, Caldwell JD (2018). Oxytocin and steroid actions. Curr Top Behav Neurosci 35: 77–95. Jobst A, Krause D, Maiwald C et al. (2016). Oxytocin course over pregnancy and postpartum period and the association with postpartum depressive symptoms. Arch Womens Ment Health 19: 571–579. Johnson ZV, Walum H, Jamal YA et al. (2016). Central oxytocin receptors mediate mating-induced partner preferences

85

and enhance correlated activation across forebrain nuclei in male prairie voles. Horm Behav 79: 8–17. Keebaugh AC, Young LJ (2011). Increasing oxytocin receptor expression in the nucleus accumbens of pre-pubertal female prairie voles enhances alloparental responsiveness and partner preference formation as adults. Horm Behav 60: 498–504. Kimmel M, Clive M, Gispen F et al. (2016). Oxytocin receptor DNA methylation in postpartum depression. Psychoneuroendocrinology 69: 150–160. Koebele SV, Bimonte-Nelson HA (2017). The endocrinebrain-aging triad where many paths meet: female reproductive hormone changes at midlife and their influence on circuits important for learning and memory. Exp Gerontol 94: 14–23. Kompier NF, Keysers C, Gazzola V et al. (2019). Early life adversity and adult social behavior: focus on arginine vasopressin and oxytocin as potential mediators. Front Behav Neurosci 13: 143. https://doi.org/10.3389/fnbeh.2019.00143. Kosfeld M, Heinrichs M, Zak PJ et al. (2005). Oxytocin increases trust in humans. Nature 435: 673–676. Kroll-Desrosiers AR, Nephew BC, Babb JA et al. (2017). Association of peripartum synthetic oxytocin administration and depressive and anxiety disorders within the first postpartum year. Depress Anxiety 34: 137–146. Kubzansky LD, Mendes WB, Appleton AA et al. (2012). A heartfelt response: oxytocin effects on response to social stress in men and women. Biol Psychol 90: 1–9. Lee RJ, Coccaro EF, Cremers H et al. (2013). A novel V1a receptor antagonist blocks vasopressin-induced changes in the CNS response to emotional stimuli: an fMRI study. Front Syst Neurosci 7: 100. Lee SY, Park SH, Chung C et al. (2015). Oxytocin protects hippocampal memory and plasticity from uncontrollable stress. Sci Rep 5: 18540. Levine A, Zagoory-Sharon O, Feldman R et al. (2007). Oxytocin during pregnancy and early postpartum: individual patterns and maternal-fetal attachment. Peptides 28: 1162–1169. Levy F, Kendrick KM, Goode JA et al. (1995). Oxytocin and vasopressin release in the olfactory bulb of parturient ewes: changes with maternal experience and effects on acetylcholine, gamma-aminobutyric acid, glutamate and noradrenaline release. Brain Res 669: 197–206. Li CY, Zhang L, Li J et al. (2017). Effect of endogenous arginine-vasopressin arising from the paraventricular nucleus on learning and memory functions in vascular dementia model rats. Biomed Res Int 2017: 3214918. Lim MM, Young LJ (2004). Vasopressin-dependent neural circuits underlying pair bond formation in the monogamous prairie vole. Neuroscience 125: 35–45. Lim MM, Wang Z, Olazabal DE et al. (2004). Enhanced partner preference in a promiscuous species by manipulating the expression of a single gene. Nature 429: 754–757. Liu JC, Guastella AJ, Dadds MR (2013). Exploring the role of intra-nasal oxytocin on the partner preference effect in humans. Psychoneuroendocrinology 38: 587–591.

86

Q. LU AND S. HU

Lonstein JS, Maguire J, Meinlschmidt G et al. (2014). Emotion and mood adaptations in the peripartum female:complementary contributions of GABA and oxytocin. J Neuroendocrinol 26: 649–664. Lopatina O, Inzhutova A, Pichugina YA et al. (2011). Reproductive experience affects parental retrieval behaviour associated with increased plasma oxytocin levels in wild-type and CD38-knockout mice. J Neuroendocrinol 23: 1125–1133. Lopatina O, Inzhutova A, Salmina AB et al. (2012). The roles of oxytocin and CD38 in social or parental behaviors. Front Neurosci 6: 182. Lu Q, Lai J, Du Y et al. (2019). Sexual dimorphism of oxytocin and vasopressin in social cognition and behaviour. Psychol Res Behav Manag 12: 337–349. Luo L, Ma X, Zheng X et al. (2015). Neural systems and hormones mediating attraction to infant and child faces. Front Psychol 6: 970. MacKinnon AL, Gold I, Feeley N et al. (2014). The role of oxytocin in mothers’ theory of mind and interactive behavior during the perinatal period. Psychoneuroendocrinology 48: 52–63. Madularu D, Athanassiou M, Yee JR et al. (2014a). Oxytocin and object preferences in the male prairie vole. Peptides 61: 88–92. Madularu D, Athanassiou M, Yee JR et al. (2014b). Centrallyadministered oxytocin promotes preference for familiar objects at a short delay in ovariectomized female rats. Behav Brain Res 274: 164–167. Mah BL, Van Ijzendoorn MH, Smith R et al. (2013). Oxytocin in postnatally depressed mothers: its influence on mood and expressed emotion. Prog Neuro-Psychopharmacol Biol Psychiatry 40: 267–272. Mah BL, Bakermans-Kranenburg MJ, Van Ijzendoorn MH et al. (2015). Oxytocin promotes protective behavior in depressed mothers: a pilot study with the enthusiastic stranger paradigm. Depress Anxiety 32: 76–81. McHenry J, Carrier N, Hull E et al. (2014). Sex differences in anxiety and depression: role of testosterone. Front Neuroendocrinol 35: 42–57. Mikolajczak M, Pinon N, Lane A et al. (2010). Oxytocin not only increases trust when money is at stake, but also when confidential information is in the balance. Biol Psychol 85: 182–184. Mishima K, Tsukikawa H, Inada K et al. (2001). Ameliorative effect of vasopressin-(4-9) through vasopressin V(1A) receptor on scopolamine-induced impairments of rat spatial memory in the eight-arm radial maze. Eur J Pharmacol 427: 43–52. Morrison TR, Ricci LA, Melloni Jr RH (2016). Vasopressin differentially modulates aggression and anxiety in adolescent hamsters administered anabolic steroids. Horm Behav 86: 55–63. Nakhjiri E, Saboory E, Roshan-Milani S et al. (2017). Effect of prenatal restraint stress and morphine co-administration on plasma vasopressin concentration and anxiety behaviors in adult rat offspring. Stress 20: 205–211.

Nephew BC, Bridges RS (2008). Arginine vasopressin V1a receptor antagonist impairs maternal memory in rats. Physiol Behav 95: 182–186. Nishina K, Takagishi H, Takahashi H et al. (2019). Association of polymorphism of arginine-vasopressin receptor 1A (AVPR1a) gene with trust and reciprocity. Front Hum Neurosci 13: 230. Pan YF, Chen XR, Wu MN et al. (2010). Arginine vasopressin prevents against Abeta(25-35)-induced impairment of spatial learning and memory in rats. Horm Behav 57: 448–454. Parker KJ, Lee TM (2003). Female meadow voles (Microtus pennsylvanicus) demonstrate same-sex partner preferences. J Comp Psychol 117: 283–289. Pedersen CA, Caldwell JD, Walker C et al. (1994). Oxytocin activates the postpartum onset of rat maternal behavior in the ventral tegmental and medial preoptic areas. Behav Neurosci 108: 1163–1171. Peters S, Slattery DA, Uschold-Schmidt N et al. (2014). Dose-dependent effects of chronic central infusion of oxytocin on anxiety, oxytocin receptor binding and stress-related parameters in mice. Psychoneuroendocrinology 42: 225–236. Pietrowsky R, Struben C, Molle M et al. (1996). Brain potential changes after intranasal vs. intravenous administration of vasopressin: evidence for a direct nose-brain pathway for peptide effects in humans. Biol Psychiatry 39: 332–340. Pitkow LJ, Sharer CA, Ren X et al. (2001). Facilitation of affiliation and pair-bond formation by vasopressin receptor gene transfer into the ventral forebrain of a monogamous vole. J Neurosci 21: 7392–7396. Prevost M, Zelkowitz P, Tulandi T et al. (2014). Oxytocin in pregnancy and the postpartum: relations to labor and its management. Front Public Health 2: 1. Resnick SM, Matsumoto AM, Stephens-Shields AJ et al. (2017). Testosterone treatment and cognitive function in older men with low testosterone and age-associated memory impairment. JAMA 317: 717–727. Rich ME, deCa´rdenas EJ, Lee HJ et al. (2014). Impairments in the initiation of maternal behavior in oxytocin receptor knockout mice. PLoS One 9: e98839. Richard S, Zingg HH (1990). The human oxytocin gene promoter is regulated by estrogens. J Biol Chem 265: 6098–6103. Riem MME, Kunst LE, Steenbakkers FDF et al. (2019). Oxytocin reduces interpersonal distance: examining moderating effects of childrearing experiences and interpersonal context in virtual reality. Psychoneuroendocrinology 108: 102–109. Robertson E, Grace S, Wallington T et al. (2004). Antenatal risk factors for postpartum depression: a synthesis of recent literature. Gen Hosp Psychiatry 26: 289–295. Rogers CN, Ross AP, Sahu SP et al. (2018). Oxytocin- and arginine vasopressin-containing fibers in the cortex of humans, chimpanzees, and rhesus macaques. Am J Primatol 80: e22875.

SEX DIFFERENCES OF OXYTOCIN AND VASOPRESSIN IN SOCIAL BEHAVIORS Russell JA, Douglas AJ, Ingram CD (2001). Brain preparations for maternity—adaptive changes in behavioral and neuroendocrine systems during pregnancy and lactation. An overview. Prog Brain Res 133: 1–38. Sabihi S, Durosko NE, Dong SM et al. (2014). Oxytocin in the prelimbic medial prefrontal cortex reduces anxiety-like behavior in female and male rats. Psychoneuroendocrinology 45: 31–42. Sarkar DK, Frautschy SA, Mitsugi N (1992). Pituitary portal plasma levels of oxytocin during the estrous cycle, lactation, and hyperprolactinemia. Ann N Y Acad Sci 652: 397–410. Scheele D, Striepens N, Gunturkun O et al. (2012). Oxytocin modulates social distance between males and females. J Neurosci 32: 16074–16079. Sharma D, Handa RJ, Uht RM (2012). The ERb ligand 5a-androstane, 3b,17b-diol (3b-diol) regulates hypothalamic oxytocin (Oxt) gene expression. Endocrinology 153: 2353–2361. Sharma K, LeBlanc R, Haque M et al. (2019). Sexually dimorphic oxytocin receptor-expressing neurons in the preoptic area of the mouse brain. PLoS One 14: e0219784. Shimizu K, Nakamura K, Yokosuka M et al. (2018). Modulation of male mouse sociosexual and anxiety-like behaviors by vasopressin receptors. Physiol Behav 197: 37–41. Shishido E, Shuo T, Takahata K et al. (2019). Changes in salivary oxytocin levels and bonding disorder in women from late pregnancy to early postpartum: a pilot study. PLoS One 14: e0221821. Simmons TC, Balland JF, Dhauna J et al. (2017). Early intranasal vasopressin administration impairs partner preference in adult male prairie voles (Microtus ochrogaster). Front Endocrinol (Lausanne) 8: 145. Simpson EA, Paukner A, Sclafani V et al. (2017). Acute oxytocin improves memory and gaze following in male but not female nursery-reared infant macaques. Psychopharmacology (Berl) 234: 497–506. Smith AS, Wang Z (2014). Hypothalamic oxytocin mediates social buffering of the stress response. Biol Psychiatry 76: 281–288. Smith AS, Agmo A, Birnie AK et al. (2010). Manipulation of the oxytocin system alters social behavior and attraction in pair-bonding primates, Callithrix penicillata. Horm Behav 57: 255–262. Smith CJ, Poehlmann ML, Li S et al. (2017). Age and sex differences in oxytocin and vasopressin V1a receptor binding densities in the rat brain: focus on the social decision-making network. Brain Struct Funct 222: 981–1006. Swaab DF, Fliers E, Partiman TS (1985). The suprachiasmatic nucleus of the human brain in relation to sex, age and senile dementia. Brain Res 342: 37–44. Thienel M, Heinrichs M, Fischer S et al. (2014). Oxytocin’s impact on social face processing is stronger in homosexual than heterosexual men. Psychoneuroendocrinology 39: 194–203.

87

Thompson R, Gupta S, Miller K et al. (2004). The effects of vasopressin on human facial responses related to social communication. Psychoneuroendocrinology 29: 35–48. Tomizawa K, Iga N, Lu YF et al. (2003). Oxytocin improves long-lasting spatial memory during motherhood through MAP kinase cascade. Nat Neurosci 6: 384–390. Triana-Del Rio R, Tecamachaltzi-Silvaran MB, DiazEstrada VX et al. (2015). Conditioned same-sex partner preference in male rats is facilitated by oxytocin and dopamine: effect on sexually dimorphic brain nuclei. Behav Brain Res 283: 69–77. Uzefovsky F, Shalev I, Israel S et al. (2012). Vasopressin selectively impairs emotion recognition in men. Psychoneuroendocrinology 37: 576–580. Uzefovsky F, Shalev I, Israel S et al. (2015). Oxytocin receptor and vasopressin receptor 1a genes are respectively associated with emotional and cognitive empathy. Horm Behav 67: 60–65. Vaidyanathan R, Hammock EA (2017). Oxytocin receptor dynamics in the brain across development and species. Dev Neurobiol 77: 143–157. Vawter MP, De Wied D, Van Ree JM (1997). Vasopressin fragment, AVP-(4-8), improves long-term and short-term memory in the hole board search task. Neuropeptides 31: 489–494. Veenema AH, Bredewold R, Neumann ID (2007). Opposite effects of maternal separation on intermale and maternal aggression in C57BL/6 mice: link to hypothalamic vasopressin and oxytocin immunoreactivity. Psychoneuroendocrinology 32: 437–450. Wang H, Duclot F, Liu Y et al. (2013). Histone deacetylase inhibitors facilitate partner preference formation in female prairie voles. Nat Neurosci 16: 919–924. Weigand A, Feeser M, Gartner M et al. (2013). Effects of intranasal oxytocin prior to encoding and retrieval on recognition memory. Psychopharmacology (Berl) 227: 321–329. Wersinger SR, Temple JL, Caldwell HK et al. (2008). Inactivation of the oxytocin and the vasopressin (Avp) 1b receptor genes, but not the Avp 1a receptor gene, differentially impairs the Bruce effect in laboratory mice (Mus musculus). Endocrinology 149: 116–121. Williams JR, Carter CS, Insel T (1992). Partner preference development in female prairie voles is facilitated by mating or the central infusion of oxytocin. Ann N Y Acad Sci 652: 487–489. Williams JR, Insel TR, Harbaugh CR et al. (1994). Oxytocin administered centrally facilitates formation of a partner preference in female prairie voles (Microtus ochrogaster). J Neuroendocrinol 6: 247–250. Winslow JT, Hastings N, Carter CS et al. (1993). A role for central vasopressin in pair bonding in monogamous prairie voles. Nature 365: 545–548. Worley NB, Dumais KM, Yuan JC et al. (2019). Oestrogen and androgen receptor activation contribute to the masculinisation of oxytocin receptors in the bed nucleus of the stria terminalis of rats. J Neuroendocrinol 31: e12760.

88

Q. LU AND S. HU

Xu Y, Sheng H, Bao Q et al. (2016). NLRP3 inflammasome activation mediates estrogen deficiency-induced depression- and anxiety-like behavior and hippocampal inflammation in mice. Brain Behav Immun 56: 175–186. Yelland J, Sutherland G, Brown SJ (2010). Postpartum anxiety, depression and social health: findings from a population-based survey of Australian women. BMC Public Health 10: 771.

Yoshihara C, Numan M, Kuroda KO (2018). Oxytocin and parental behaviors. Curr Top Behav Neurosci 35: 119–153. Young LJ, Wang Z (2004). The neurobiology of pair bonding. Nat Neurosci 7: 1048–1054. Zak PJ, Stanton AA, Ahmadi S (2007). Oxytocin increases generosity in humans. PLoS One 2: e1128. Zelkowitz P, Gold I, Feeley N et al. (2014). Psychosocial stress moderates the relationships between oxytocin, perinatal depression, and maternal behavior. Horm Behav 66: 351–360.

Handbook of Clinical Neurology, Vol. 180 (3rd series) The Human Hypothalamus: Middle and Posterior Region D.F. Swaab, F. Kreier, P.J. Lucassen, A. Salehi, and R.M. Buijs, Editors https://doi.org/10.1016/B978-0-12-820107-7.00006-9 Copyright © 2021 Elsevier B.V. All rights reserved

Chapter 6

Oxytocin, eating behavior, and metabolism in humans LIYA KEREM1,2 AND ELIZABETH A. LAWSON1* 1

Neuroendocrine Unit, Department of Medicine, Massachusetts General Hospital and Harvard Medical School, Boston, MA, United States

2

Division of Pediatric Endocrinology, Massachusetts General Hospital for Children, Boston, MA, United States

Abstract The hypothalamic peptide oxytocin has been increasingly recognized as a hormone and neurotransmitter with important effects on energy intake, metabolism, and body weight and is under investigation as a potential novel therapeutic agent for obesity. The main neurons producing oxytocin and expressing the oxytocin receptor are strategically located in brain areas known to be critically involved in homeostatic energy balance as well as hedonic and motivational aspects of eating behavior. In this chapter, we will review the central and peripheral physiology of oxytocin and the interaction of oxytocin with key hormones and neural circuitries that affect food intake and metabolism. Next, we will synthesize the available data on endogenous oxytocin levels related to caloric intake, body weight, and metabolic status. We will then review the effects of exogenous oxytocin administration on eating behavior, body weight, and metabolism in humans, including in healthy individuals as well as specific populations with suspected perturbations involving oxytocin pathways. Finally, we will address the promise and fundamental challenges of translating this line of research to clinical care.

OXYTOCIN NEUROPHYSIOLOGY— RELATION TO ENERGY HOMEOSTASIS The hypothalamic neurohormone oxytocin is a highly conserved nine-amino-acid neuropeptide that is predominantly produced in neurons originating in the paraventricular (PVN) and supraoptic (SON) nuclei of the hypothalamus. Current knowledge regarding the neuroanatomy and physiology of oxytocin is mostly derived from animal data (e.g., rodents and nonhuman primates). Oxytocin is dispersed across the brain via distal axonal projections from parvocellular PVN oxytocin neurons as well as local dendritic release of oxytocin into the extracellular space (Ludwig and Leng, 2006; MeyerLindenberg et al., 2011). In addition to having central effects within the brain, oxytocin is also released into the peripheral circulation following activation of magnocellular SON oxytocin neurons that project to the

posterior pituitary (Brown et al., 2013). The seventransmembrane G-protein-coupled oxytocin receptor can be found in a wide range of brain areas that are important for energy regulation and gustatory perception (Jurek and Neumann, 2018), including the nucleus accumbens (NAc; Olazabal and Young, 2006) and ventral tegmental area (VTA; Peris et al., 2017) which are highly involved in reward processing, the hypothalamus (Boccia et al., 2013; Maejima et al., 2014) which is critical for energy homeostasis, the nucleus tractus solitarius (NTS; Ong et al., 2015) which is also involved in energy balance control, and the amygdala (Boccia et al., 2013), a brain area important for decision making and emotional responses. The role of oxytocin in parturition and lactation has been studied extensively. Oxytocin receptor activation in the myometrium stimulates uterine contractions, primarily by inducing an increase in intracellular calcium

*Correspondence to: Elizabeth A. Lawson, Neuroendocrine Unit, Massachusetts General Hospital 55 Fruit Street, BUL457, Boston, MA, 02114, United States. Tel: +1-617-726-3870, Fax: +1-617-726-5072, E-mail: [email protected]

90

L. KEREM AND E.A. LAWSON

levels (Arrowsmith and Wray, 2014). During labor, noradrenaline release in the SON and PVN nuclei activates the oxytocin neurons to fire in synchrony, resulting in a surge of oxytocin release from the posterior pituitary into the peripheral circulation (Douglas et al., 2001; Brunton, 2019). Oxytocin-mediated uterine contractions prior to parturition are regulated by the increase in oxytocin secretion and myometrial oxytocin receptor expression (Blanks and Thornton, 2003). Additionally, a positive feedback loop exists between myometrial contraction and oxytocin release, mediated by spinal and vagal afferents that relay the uterine contraction signaling to A2 noradrenergic NTS neurons, which further project to the magnocellular oxytocin neurons (Brunton, 2019). Postpartum, oxytocin receptor expression decreases rapidly in the myometrium, while it remains upregulated in the mammary glands in preparation for lactation (Larcher et al., 1995). Milk ejection, induced by the contraction of the myoepithelial cells surrounding the mammary alveoli, is regulated by a pulsatile release of oxytocin into the peripheral circulation (Augustine et al., 2018). This pattern of oxytocin secretion results from synchronized electrical activity of magnocellular oxytocin neurons, which occurs in response to suckling (Higuchi et al., 1983; Crowley, 2015). In addition to having an important role in lactation and parturition, oxytocin is also involved in the onset and maintenance of maternal behavior following parturition. Oxytocin is involved in the plasticity of the somatosensory and auditory cortex that occurs in the mother following birth to enable the perception and recognition of infant cues, and it has an important role in the establishment of parent–infant attachment and bonding (Galbally et al., 2011; Marlin et al., 2015; Kim and Strathearn, 2016; MacKinnon et al., 2018; Valtcheva and Froemke, 2019). While oxytocin has been shown to be present in human maternal milk (Takeda et al., 1986; Mishra et al., 2014), very little is known about the effects of ingested oxytocin in the newborn (Ivell and Anand-Ivell, 2017). Studies of the central neurocircuitry in which oxytocin acts reveal the intricate involvement of this hormone in appetite regulation, its interaction with other appetiteregulating neurons and its potential role as an anorexigenic neurohormone. The PVN and SON oxytocinergic neurons have been shown to have direct projections to proopiomelanocortin (POMC) neurons in the arcuate nucleus (ARC) of the hypothalamus (Maejima et al., 2014), with rapidly induced satiety in mice following chemo- or optogenetic manipulation of ARC neurons expressing the oxytocin receptor (Fenselau et al., 2017). These neurons were shown to have projections to PVN neurons expressing the melanocortin 4 receptor and optogenetic stimulation of their terminals in the PVN rapidly decreased feeding (Fenselau et al., 2017).

Accumulating evidence shows that oxytocin contributes to the effects of leptin on food intake and body weight (Blevins and Ho, 2013; Altirriba et al., 2015). Central leptin administration directedly activates a subpopulation of oxytocin neurons in the PVN that innervate the NTS (Perello and Raingo, 2013). Moreover, oxytocin receptor antagonist administration attenuates the anorexigenic effects of leptin (Blevins et al., 2004). A recent study showed that when diet-induced, leptin resistant obese mice were pretreated with oxytocin compared with vehicle, leptin sensitivity was restored, as observed by reduced food intake following acute leptin administration (Labyb et al., 2018). Another mechanism explaining the anorexigenic effects of oxytocin could be related to interaction with central receptors for glucagon-like peptide-1 (GLP-1), a peptide hormone with potent inhibitory effects on food intake in animal models and human subjects (Dailey and Moran, 2013). Central GLP-1 administration activates PVN oxytocin release (Larsen et al., 1997) and hindbrain GLP-1 neurons are activated to express c-Fos after central administration of oxytocin (Rinaman and Rothe, 2002). Additionally, immunolabeled oxytocin-positive fibers and terminals have been demonstrated in close proximity to GLP-1 cell bodies in the NTS (Rinaman and Rothe, 2002). Interestingly, an observed anorexigenic effect of centrally infused oxytocin in rats (to be discussed in detail later in the chapter) was significantly attenuated when animals were pretreated with a central GLP-1 receptor antagonist, while the anorexigenic effects of central GLP-1 administration were not affected by administration of an oxytocin receptor antagonist, suggesting that GLP-1 receptor signaling is an important downstream mediator of oxytocin effects (Rinaman and Rothe, 2002). Suppression of the central effects of oxytocin by agouti-related protein (AgRP) neurons in the ARC may be required to initiate feeding behavior. AgRP neurons respond to circulating signals of energy deficit, such as ghrelin, and their activation stimulates feeding (Aponte et al., 2011). Using optogenetic and pharmacogenetic techniques, Atasoy et al. showed that AgRP neurons target and inhibit oxytocin neurons and that suppression of oxytocin neurons by AgRP neural activation is critically required to initiate feeding behavior in mice (Atasoy et al., 2012). Oxytocin may also inhibit food intake by interacting with cholecystokinin (CCK), a peptide released from the intestinal endocrine cells during a meal to promote satiety. Blevins et al. have shown in rats that oxytocin pathways originating from the PVN and descending to the NTS anatomically interact with NTS neurons that respond to peripheral CCK signals following a meal (Blevins et al., 2003). In the same study, central administration of an oxytocin receptor antagonist

OXYTOCIN, EATING BEHAVIOR, AND METABOLISM IN HUMANS attenuated the inhibitory effects of CCK on food intake, suggesting that intact oxytocin signaling is required for CCK-induced suppression of feeding. Further supporting this notion, targeted insult of cells expressing oxytocin in the NTS in rats attenuated the anorexigenic effects of CCK and eliminated the effects of oxytocin receptor antagonist to induce food intake (Baskin et al., 2010). Oxytocin may specifically affect hedonic food intake, a concept which will be discussed in further detail later in this chapter. From a neurophysiological standpoint, the interplay between oxytocin pathways and the dopaminergic system, which affects reward and motivational behavior, has been well established (Love, 2014; Maejima et al., 2018). Using virus-based and cell type-specific monosynaptic tracing techniques (Beier et al., 2015) as well as optogenetic and electrophysiological approaches (Xiao et al., 2017), studies have shown direct projections of the PVN oxytocin-synthesizing neurons to the VTA, the origin of the mesolimbic dopaminergic system, and an area critically involved in motivational food processing (Meye and Adan, 2014). Glutamate and dopamine oxytocin receptor expressing VTA neurons further project to additional mesocorticolimbic structures such as the NAc, prefrontal cortex, and the amygdala demonstrating the complexity of oxytocin effects on reward pathways (Peris et al., 2017). Administration of oxytocin directly into the VTA in rats significantly suppresses intake of a sucrose solution, while administration of oxytocin receptor antagonists into the VTA results in a significant increase of sucrose intake, suggesting that endogenous oxytocin action within the VTA suppresses palatable food intake (Mullis et al., 2013). Similarly, direct oxytocin infusion into the NAc in rats decreases consumption of palatable sucrose in nondeprived animals, an effect that was abolished in the presence of an oxytocin antagonist (Herisson et al., 2016). The effects of oxytocin on the mesolimbic dopaminergic system have been extensively studied in the context of drug abuse and addiction (McGregor and Bowen, 2012; Bowen and Neumann, 2017; Leong et al., 2018). Oxytocin interferes with reward-related drug actions (e.g., cocaine, opiates, and methamphetamine) within the dopaminergic system, and it can modulate drug-induced perturbations that occur in the dopaminergic circuitry in the setting of addiction. There is evidence from animal studies to suggest that palatable foods rich in sugar induce reward and craving similar to addictive drugs by acting on common neural circuits (Ahmed et al., 2013). Therefore, the effects of oxytocin on hedonic food consumption and drug addiction could represent a common neurobiological mechanism. In addition to having an effect on hedonic energy consumption, there is evidence to show that oxytocin suppresses homeostatic food intake by acting directly

91

on the ventromedial hypothalamic nucleus (VMH), an area critically important for energy balance (Choi et al., 2013) and rich in oxytocin receptors (Boccia et al., 2013). Oxytocin injected directly into the VMH in rats reduced feeding in both fed and fasted states, and this effect was eliminated when animals were pretreated with an oxytocin antagonist (Noble et al., 2014; Klockars et al., 2017). Interestingly, while direct administration of oxytocin into the VMH in rats decreases chow intake, an effect that is reversed by an oxytocin antagonist, it does not modify intake of palatable sweet sucrose solutions in hunger or satiety (Klockars et al., 2017), suggesting that oxytocin acting in the VMH decreases food intake driven by energy deficit and not by palatability. The suppression of homeostatic food intake by oxytocin is likely achieved by activating a neural network that involves both the PVN and the ARC. This is suggested by an increase in Fos immunoreactivity in the VMH itself as well as in the ARC and PVN following direct oxytocin injection into the VMH (Klockars et al., 2017). While oxytocin is mainly produced in the central nervous system, gastrointestinal (GI) oxytocin-producing neurons have been demonstrated in the myenteric and submucous ganglia along the human gastrointestinal tract (Ohlsson et al., 2006b), supporting the involvement of oxytocin in the brain–gut neurohormonal axis. Oxytocin receptors are also present along the GI tract and can mediate GI motility (Ohlsson et al., 2006a; Qin et al., 2009; Xi et al., 2019). Multiple studies have shown that oxytocin injected peripherally (intraperitoneally and subcutaneously) can have similar anorexigenic effects to oxytocin administrated directly into the brain (Maejima et al., 2011; Leslie et al., 2018); however, oxytocin given systemically has very poor blood–brain barrier penetrance, with approximately 0.001% CSF recovery of the dose delivered systemically (Mens et al., 1983). A recent study showed that intraperitoneal administration of oxytocin induces c-Fos expression in PVN oxytocin neurons and inhibits food intake. These effects were blunted by subdiaphragmatic vagotomy, suggesting that peripheral oxytocin exerts its effects on energy intake regulation by activating central oxytocin neurons via vagal afferent nerves (Iwasaki et al., 2019). Most current investigational human trials examining the effects of oxytocin on energy intake and metabolism use oxytocin delivered via an intranasal spray which has been shown to increase oxytocin CSF and peripheral levels in nonhuman primates (Dal Monte et al., 2014; Freeman et al., 2016; Lee et al., 2018). The peripheral effects of oxytocin on energy expenditure, adipose tissue metabolism, glucose, and lipid homeostasis will be discussed later in the chapter.

92

L. KEREM AND E.A. LAWSON

ENDOGENOUS OXYTOCIN LEVELS, EATING BEHAVIOR, AND METABOLIC STATUS IN HUMANS Measurement of oxytocin During the last three decades, numerous studies have been conducted to investigate the role of oxytocin in relation to appetite, metabolism, social bonding, psychological disorders, bone health, and immunity (Olff et al., 2013; Amri and Pisani, 2016; Li et al., 2016; Lawson, 2017). Techniques of measuring endogenous oxytocin levels are still being developed to date and therefore studies examining an association of oxytocin levels with other parameters should be carefully interpreted (MacLean et al., 2019). The lack of validation, reproducibility, and correlation between different methods of oxytocin measurement has been attributed to different methods of plasma/serum preparation (extraction, filtration or none) and detection (commercial or laboratory-based enzyme linked-immunosorbent assay, radioimmunoassay, and mass spectrometry) (Szeto et al., 2011; Leng and Sabatier, 2016). While extraction can remove interfering substances, oxytocin strongly binds to common components in blood samples and therefore a considerable quantity of bound oxytocin may be discarded in the process of sample purification (Lefevre et al., 2017; MacLean et al., 2019). Methods of immunoassays may be less specific due to cross-reactivity of oxytocin antibodies with nonoxytocin analytes/epitopes in the sample (Lefevre et al., 2017); however, they have the potential advantage of identifying bound or partially degraded oxytocin as well as oxytocin fragments, which may have a biological role (Uvnas Moberg et al., 2019). While liquidchromatography–mass-spectrometry (LC–MS) is considered a gold standard test to detect intact molecules with high specificity, bound or partially degraded oxytocin may not be detected using this method. Also, the lower limit of detection using LC–MS may not be sufficient to measure baseline oxytocin levels in nonpregnant humans (Franke et al., 2019). Overall, further research is required to understand the unique physiological roles of free, bound, and partially degraded oxytocin and to develop validated and reliable measurement methods. The synthesis of data regarding levels of oxytocin from different studies should take into account the methodological limitations and interpret the data with caution; however, studies using a single measurement method to compare the levels of oxytocin between groups or alteration of oxytocin levels in response to a stimulus (e.g., meal, exercise) are still valuable. The relation between peripheral oxytocin levels and the levels or actions of oxytocin in the central nervous system is not well understood. In lactating monkeys, the release of oxytocin into the CSF was not associated

with release into the peripheral circulation (Amico et al., 1990), suggesting that central oxytocin effects by dendritic and axonal release from PVN oxytocin nuclei may not be coordinated with the release of oxytocin into the peripheral circulation by SON oxytocin neurons. In another study, a subpopulation of parvocellular PVN oxytocin neurons was found to project simultaneously to spinal cord neurons and magnocellular SON oxytocin neurons that mediate peripheral oxytocin secretion, thus supporting coordinated central and peripheral oxytocin release in rats (Eliava et al., 2016). Studies in humans are also inconsistent in regard to the relationship between central and peripheral oxytocin levels. While two studies found a correlation between peripheral and central oxytocin levels in adults with headaches (Wang et al., 2013) and children undergoing indicated lumbar puncture (Carson et al., 2015), others could not demonstrate such a relationship in adults (Kagerbauer et al., 2013), nonpregnant, or pregnant females in labor or elective c-section (Takagi et al., 1985; Altemus et al., 2004). Another key point when investigating oxytocin levels is that there may be sex differences in oxytocin secretion and females demonstrate variations in oxytocin levels related to menstrual cycle, functional hypothalamic amenorrhea, hormonal contraceptive agents, and menopause (Maestrini et al., 2018; Engel et al., 2019; Aulinas et al., 2019b), a fact that should be taken into consideration when integrating data collected from both genders or when interpreting data from female participants. Finally, the pulsatile nature of oxytocin secretion (Baskaran et al., 2017) and the short half-life of 5 min in the circulation (Ryden and Sjoholm, 1969) necessitate specific methods of analysis (e.g., frequent sampling and deconvolution) in order to provide a more comprehensive and accurate understanding of oxytocin dynamics.

Endogenous oxytocin levels in healthy individuals and patients with disrupted oxytocin pathways Several studies examined the effects of food consumption on endogenous oxytocin secretory patterns in humans. In a recent study, young healthy females demonstrated a significant decrease in endogenous peripheral oxytocin levels following a mixed meal with a balanced macronutrient content (Aulinas et al., 2019b). Further analysis showed that the change in oxytocin from premeal to postprandial levels was significantly associated with subjective postprandial appetite, and more specifically, an attenuated oxytocin excursion was associated with lower rating of hunger. Of note, this association was independent of age and phase of menstrual cycle, which is of importance since oxytocin area under the curve was significantly lower in the early- to mid-follicular

OXYTOCIN, EATING BEHAVIOR, AND METABOLISM IN HUMANS phase compared with other menstrual cycle phases (Aulinas et al., 2019b). In another study, ingestion of corn oil was associated with a significant increase in oxytocin levels in a small cohort of healthy females (Ohlsson et al., 2002), while two other studies did not find a significant change in oxytocin levels following a meal in male and female adults (Miaskiewicz et al., 1989; Stock et al., 1989). Further studies are required to delineate the dynamics of oxytocin secretion following energy intake, the response to different macronutrients, and the relation to appetite regulation. Oxytocin has also been studied extensively in females with anorexia nervosa (AN) (Giel et al., 2018), not only due to its appetite-regulating properties but also because of anxiolytic and antidepressant effects (Bale et al., 2001; Blume et al., 2008; MacDonald and Feifel, 2014; Massey et al., 2016; Sabihi et al., 2017), and involvement in reward neural circuitry which is thought to be aberrant in AN (O’hara et al., 2015). Genetic studies may provide insight into altered oxytocin physiology in patients with AN. In women with active AN, but not in women who recovered from AN, the presence of an A carrier of two small nucleotide polymorphisms (SNPs; rs53576, rs2254298) of the oxytocin receptor has been associated with the severity of eating disorder pathology (Acevedo et al., 2015). Additionally, in another study, lifetime restrictive eating behavior was associated with oxytocin receptor AG/AA rs2253298 genotype (Micali et al., 2017). AN has also been associated with specific epigenetic changes in the oxytocin receptor. More specifically, increased methylation of CpG sites in the oxytocin receptor promoter region has been found in women with AN compared with healthy controls, with a negative association between the level of methylation and BMI across all participants (Kim et al., 2014). In the same study, a multiple regression analysis showed that BMI and eating disorder psychopathology were the main determinants of the methylation level at specific CpG sites. In concert, studies suggest genetic and epigenetic differences in AN; further research is needed to better characterize these differences and their association with the clinical aspects of the disease. Overall, review of the literature shows alterations in oxytocin pathways in females with AN compared with normal-weight healthy individuals. In an early study, CSF oxytocin levels were lower in a small group of females with AN compared with healthy controls (Demitrack et al., 1990). In more recent investigations, morning fasting and nocturnal serum oxytocin levels were also shown to be lower in patients with AN compared with healthy-weight females (Monteleone et al., 2016; Schorr et al., 2017), and positively associated with BMI across the weight spectrum (Schorr et al., 2017), suggesting that oxytocin could be a marker of energy availability. In other studies, fasting oxytocin

93

levels were significantly lower in normal-weight females who had partially or fully recovered from AN compared with healthy controls (Lawson et al., 2012; Afinogenova et al., 2016). Interestingly, in normal-weight partially recovered (but not underweight) females with AN, lower oxytocin levels were associated with a greater degree of anxiety and depressive symptoms as well as disordered eating psychopathology (Afinogenova et al., 2016), suggesting that persistence of psychological abnormalities in these individuals could be partly related to perturbations in oxytocin secretion. Several studies examined the pattern of oxytocin secretion in response to a meal in the setting of AN. Following a meal, patients with AN demonstrated higher oxytocin levels and weight-recovered patients had lower oxytocin levels compared with healthy controls. In this study, mean oxytocin area under the curve was the highest in AN and significantly associated with the severity of disordered eating psychopathology in patients with low-weight and weight-recovered AN (Lawson et al., 2012). A larger study of patients with AN showed that oxytocin levels significantly decreased postprandially with a greater reduction observed in higher weight (85% of expected body weight) atypical AN patients (Aulinas et al., 2019a). In females with atypical AN, an attenuated postprandial reduction in oxytocin was associated with a greater reduction in hunger ratings, independent of estrogen levels; however, this relationship was absent in lower weight individuals with AN, supporting a dissociation between physiological response of oxytocin and subjective appetite in underweight AN (Aulinas et al., 2019a). In sum, studies suggest that patients with AN have lower basal CSF and peripheral oxytocin levels and according to one small study (Lawson et al., 2012), higher postmeal peripheral oxytocin levels compared to controls. It has been hypothesized that lower basal oxytocin levels in AN reflect an adaptive response to chronic starvation (Plessow et al., 2018a). Additional research is required to better understand central and peripheral oxytocin pathways in AN. Abnormalities in oxytocin pathways have also been identified in Prader–Willi syndrome (PWS), a multisystemic genetic disorder characterized by insatiable hunger, hyperphagia, and obesity, as well as emotional and behavioral disturbances (Rice et al., 2018). Of note, hyperghrelinemia has also been observed in PWS; however, the relationship between elevated ghrelin levels, alterations of the oxytocin system, and hyperphagia is still being researched and is not clearly understood (Kabasakalian et al., 2018). PWS is caused by lack of expression of paternally inherited genes on the chromosome 15q11.2-q13 region (Angulo et al., 2015). Patients with PWS have a remarkable reduction of oxytocinsynthesizing neurons in the PVN and a smaller volume

94

L. KEREM AND E.A. LAWSON

of the PVN itself (Swaab et al., 1995) as well as reduced expression of the oxytocin receptor gene as reported by whole genome microarray expression analysis (Bittel et al., 2007). Patients with PWS demonstrate higher fasting plasma oxytocin levels than their healthy siblings (Johnson et al., 2016), as well as elevated CSF oxytocin levels compared with healthy controls (Martin et al., 1998). The finding of elevated basal oxytocin levels in patients with PWS is counterintuitive considering the anorexigenic properties of this hormone and the reduced number of PVN oxytocin neurons. It has been suggested that depletion of central oxytocin receptors may lead to an increase in oxytocin secretion due to loss of regulatory negative feedback (Johnson et al., 2016). Further studies using well-validated oxytocin assays are needed to establish central and peripheral oxytocin levels in PWS and to determine if the reduction in PVN oxytocin neurons contributes to the lack of satiety, hyperphagia, and obesity that patients with PWS manifest. Clinical trials of oxytocin administration in PWS will be discussed later in the chapter. Oxytocin has also been investigated in patients with craniopharyngioma, a rare intracranial tumor that develops from remnant epithelial cells of Rathke’s pouch and frequently affects the hypothalamic and/or pituitary regions (Muller, 2016). Due to the neuroanatomical location of the tumor and the potential treatment-related lesions to hypothalamic structures, alteration in oxytocin pathways is thought to be involved in the pathogenesis of hypothalamic obesity, a complication driven by hyperphagia, and low energy expenditure that is observed in up to 75% of survivors (Lustig, 2011). In a crosssectional study, fasting oxytocin saliva concentrations did not differ in patients with craniopharyngioma and healthy controls; however, they were lower in patients who had surgical damage involving only the anterior hypothalamic area, compared with patients who had surgical damage involving both the anterior and posterior hypothalamic regions or no hypothalamic damage (Daubenbuchel et al., 2016). Additionally, in this study, patients with craniopharyngioma demonstrated a negative correlation between BMI and change in oxytocin levels pre- and postmeal, with the patients with the highest BMI showing smaller oxytocin changes, thus providing a possible link between oxytocin secretion and obesity in these patients (Daubenbuchel et al., 2016). In another study, the change in salivary oxytocin concentration before vs. after a meal was associated with adverse eating behavior in patients with craniopharyngioma (evaluated by validated questionnaires assessing “concerns about eating behavior” and “feelings of external eating pressures”) (Daubenbuchel et al., 2019). Finally, patients with craniopharyngioma exhibited

blunted exercise induced release of salivary oxytocin compared to controls, which was associated with greater anxiety (Gebert et al., 2018). In this study, participants with higher BMI showed smaller changes in oxytocin in response to an exercise stimulation test, supporting the notion that obesity is associated with alterations in oxytocin pathways.

Oxytocin levels in obesity and metabolic disorders Due to its important role in appetite, metabolism, and energy homeostasis, oxytocin has been studied in relation to obesity and associated metabolic comorbidities (McCormack et al., 2019). Several genetic studies have explored the association between oxytocin/oxytocin receptor genotypes and obesity as well as metabolic disorders. Genetic copy-number variants in the oxytocin receptor have been shown to be associated with severe, early-onset obesity (Wheeler et al., 2013). In an exploratory study examining the relationship between oxytocin receptor polymorphisms and overeating, rs2268493 SNP was associated with greater overeating behavior in individuals carrying homozygous T alleles, while the A allele was associated with stronger preference for sweet and fatty foods (Davis et al., 2017). In healthy weight, nondiabetic adults, analysis of the oxytocin receptor SNP rs53576 showed that in participants carrying the A allele, oxytocin levels were significantly correlated with fasting glucose and insulin levels while in subjects carrying the GG genotype, oxytocin levels were significantly and negatively correlated with leptin levels. This study suggests that variations in oxytocin receptor genotype are associated with specific relations between levels of circulating oxytocin and other metabolic hormones (Chang et al., 2019). Additional research is needed to investigate the causal association between these genotypes and metabolic phenotypes. The association between circulating oxytocin levels, obesity, and metabolic health has been investigated in multiple studies. In a recent well-designed study that included 721 adults, serum oxytocin levels were higher in individuals with overweight and obesity compared with lean participants, and positively correlated with multiple parameters of metabolic syndrome (e.g., waist-to-hip ratio, waist circumference, plasma glucose and insulin levels, impaired glucose tolerance and triglycerides) (Weingarten et al., 2019). Similar relationships were found in cohorts of men (Szulc et al., 2016) and women (Schorr et al., 2017; Skinner et al., 2019) with a positive correlation between oxytocin levels and

OXYTOCIN, EATING BEHAVIOR, AND METABOLISM IN HUMANS total, visceral, and subcutaneous fat in women (Schorr et al., 2017). Finally, another small study showed that oxytocin levels were significantly higher in individuals with severe obesity (BMI > 40 kg/m2) compared with lean individuals (Pataky et al., 2019). While the positive correlation between serum oxytocin levels and obesity severity has been established in the aforementioned studies, others have found an opposite relationship. Qian et al. found lower fasting oxytocin levels in adults with obesity compared with healthy-weight individuals, with a negative correlation between serum oxytocin levels and BMI as well as multiple indices of metabolic syndrome (waist circumference, HbA1c, fasting plasma glucose, 2-h plasma glucose, total cholesterol, triglycerides, low-density lipoprotein cholesterol, and HOMA of insulin resistance (HOMA-IR; Qian et al., 2014). Similar findings were shown in a cohort of childbearing age women (Fu-Man et al., 2019), postmenopausal women (Maestrini et al., 2018), and children with obesity (Binay et al., 2017). The discrepancy between studies showing a positive correlation between oxytocin levels and BMI and those that support a negative correlation could be related to several potential factors. The studies reported previously differed in the ethnicity and characteristics of the subjects as well as the storage conditions and analytic techniques of the blood samples. Glucose homeostasis may be another important modifying factor influencing the relation between oxytocin levels and obesity status. In the study by Schorr et al. that supported a positive association between oxytocin and obesity, subjects with T2DM were excluded from the study (Schorr et al., 2017), while in the study by Qian et al., half of the participants had a new diagnosis of T2DM (Qian et al., 2014). Additionally, in a cohort of African American men with overweight and obesity, urine oxytocin levels were found to be significantly lower in patients with T2DM compared with participants who had normal oral glucose tolerance test (Eisenberg et al., 2018). In the same manner, in a cohort of patients with metabolic syndrome, serum oxytocin levels were lower in participants who had prediabetes/T2DM compared with nondiabetic patients (Al-Rawashdeh et al., 2017). Type I diabetes mellitus is also associated with lower oxytocin levels compared with nondiabetic healthy controls (Kujath et al., 2015). Oxytocin mRNA and the oxytocin receptor have been demonstrated in the pancreas in rodents (Suzuki et al., 2013) and cultured human beta-cells (Mohan et al., 2018), suggesting that oxytocin could affect insulin secretion and glucose homeostasis. However, more research is needed to clarify the biological mechanism underlying the interaction between glucose homeostasis, obesity, and oxytocin.

95

THE EFFECTS OF OXYTOCIN ADMINISTRATION ON CALORIC INTAKE AND METABOLISM Animal studies support the role of oxytocin as a potent regulator of caloric intake, body weight, and energy metabolism (Blevins and Baskin, 2015). Genetic knockout studies show that mice deficient in either oxytocin (Camerino, 2009) or oxytocin receptors (Takayanagi et al., 2008) exhibit late-onset obesity and impaired glucose homeostasis (Camerino, 2009). Additionally, mice with single-minded 1 gene haploinsufficiency demonstrate hyperphagia, obesity, and reduction in oxytocin expression in the PVN (Kublaoui et al., 2008). These mice demonstrate a significant reduction in food intake and weight gain following oxytocin treatment (Kublaoui et al., 2008). Furthermore, administration of oxytocin to diet-induced obese rodents results in a decrease in body weight gain with a preferential reduction in fat mass as well as increased adipose tissue lipolysis and reduced insulin resistance (Deblon et al., 2011; Maejima et al., 2011, 2017; Zhang and Cai, 2011; Blevins et al., 2016; Roberts et al., 2017). Similarly, when ob /ob mice were treated with oxytocin, they displayed a dose-dependent reduction in food consumption and body weight gain (Altirriba et al., 2014), and obese diabetic db/db mice showed significant reduction in body fat accumulation and improved glucose and fat metabolism under oxytocin treatment (Plante et al., 2015). Of note, in a recent rigorous meta-analysis of 57 articles, a single dose of oxytocin was found to have a robust inhibitory effect on feeding in animals, regardless of the route of administration (peripheral vs. central) (Leslie et al., 2018). With regard to chronic administration of oxytocin, in another recent systematic review, it was found that animals receiving prolonged central or peripheral oxytocin treatment consistently exhibited reduction in energy intake, weight, adipocyte size, and fat mass together with enhanced fat oxidation and lipolysis, with the exception of several studies failing to demonstrate reduction of body weight or fat mass in nonobese animals (Horta et al., 2019). Furthermore, improvement in glucose intolerance and insulin resistance under chronic oxytocin administration was specifically seen in animal models of diet-induced obesity and diabetes. Chronic oxytocin administration evokes weight loss in diet-induced obese rodents, in part, by reducing energy intake and increasing energy expenditure and lipolysis (Blevins et al., 2016). Whether its effects on energy expenditure require activation of brown adipose tissue will require further investigation. Oxytocin administration has been researched in humans, revealing an anorexigenic effect of oxytocin

96

L. KEREM AND E.A. LAWSON

and suggesting modulation of hedonic and homeostatic food intake. In most human clinical trials, oxytocin has been given with an intranasal spray at a dose of 24 IU, with strong evidence from animal and human studies for a subsequent increase in plasma and CSF oxytocin levels (Striepens et al., 2013; Smith et al., 2019). Two randomized, placebo-controlled crossover studies of a single intranasal (IN) dose of oxytocin in healthy men showed that oxytocin significantly reduced hungerdriven caloric intake. The effect was seen across the weight spectrum (normal weight to overweight) in one study (Lawson et al., 2015), while another study concluded that oxytocin reduces food intake in those with obesity but not in normal-weight participants (Thienel et al., 2016). In addition, oxytocin has been shown to significantly attenuate postprandial palatable snack consumption, reflecting hedonic eating in men across the weight spectrum (Ott et al., 2013; Thienel et al., 2016; Burmester et al., 2018). Notably, the observed hungerand hedonic-driven anorexigenic effects of oxytocin were not associated with appetite (Ott et al., 2013; Lawson et al., 2015; Thienel et al., 2016; Spetter et al., 2018) and a single dose of oxytocin treatment had no effect on levels of appetite-regulating hormones leptin, ghrelin, PYY (Lawson et al., 2015), or GLP-1 (Ott et al., 2013). Insights into the neurobiological mechanism underlying the effects of oxytocin on energy intake in humans are provided by fMRI studies. A single dose of oxytocin compared with placebo was found to suppress the hypothalamic fMRI blood oxygen level-dependent (BOLD) signal in response to images of high-calorie foods in healthy-weight adults (van der Klaauw et al., 2017) and men with overweight and obesity (Plessow et al., 2018b), indicating that oxytocin may reduce caloric intake in part by modulating homeostatic neural circuitries (van der Klaauw et al., 2017). Additionally, in men with overweight and obesity, a single dose of oxytocin vs placebo decreased the fMRI BOLD signal in the VTA and other reward brain regions and increased the signal in cognitive-control areas (dorsal anterior cingulate cortex and frontopolar prefrontal cortex) in response to images of high-calorie foods vs nonfood objects (Plessow et al., 2018b). In another study, when healthy-weight men received a single IN oxytocin dose, an increase in fMRI BOLD signal in cognitive-control brain areas in response to high-calorie vs low-calorie food images was seen, together with an observed decrease in caloric intake during a buffet (Spetter et al., 2018). Taken together, fMRI studies support the role of oxytocin in homeostasis, reward, and cognitive neural processing in response to visual food images, thus providing valuable information regarding the potential neurobiological mechanisms underlying the effects of oxytocin on eating behavior

in humans. A randomized-controlled study examining the neurobiological effects of IN oxytocin administration for 8 weeks in adults with obesity is currently ongoing (NCT03043053). While animal studies show that oxytocin administration modulates lipid metabolism (e.g., increased lipolysis and fat oxidation, reduction in deleterious visceral fat) (Lawson, 2017), little is known in regard to the effects of oxytocin administration on lipid metabolism in humans, although the few studies investigating this question found favorable effects. In an early study conducted in the 1960s, intravenous (IV) administration of oxytocin resulted in an increase in free fatty acids and a decrease in triglyceride levels, suggesting increased lipolysis (Burt et al., 1963). A single IN oxytocin dose given to men with and without obesity increased fat utilization as measured by indirect calorimetry and was also associated with a trend toward reduction of triglyceride levels (Lawson et al., 2015). Finally, in a small pilot study, IN oxytocin given for 8 weeks to adults with obesity significantly reduced serum LDL and total cholesterol levels compared to placebo (Zhang et al., 2013). In this study, oxytocin treatment also resulted in a significant decrease in waist and hip circumferences, suggestive of reduction in visceral adiposity (Zhang et al., 2013). While advantageous effects of oxytocin treatment were seen in these studies, further research is needed to address the knowledge gap regarding the mechanisms underlying the effects of oxytocin treatment on lipid metabolism in humans. Human studies investigating the effects of a single oxytocin dose have been useful as a proof-of-concept to demonstrate the anorexigenic effects of this hormone. To date, only a single pilot study (with nine participants randomized to drug) examined the potential weightreduction effects of prolonged oxytocin treatment in adults with obesity and without diabetes (Zhang et al., 2013). In this study, IN oxytocin taken daily for 8 weeks was found to be safe and well tolerated and resulted in a significant BMI reduction (Zhang et al., 2013). Animal studies suggest that oxytocin-induced weight loss may be partially related to increased energy expenditure (Blevins et al., 2016) and brown adipose tissue thermogenesis (Roberts et al., 2017) leading to net negative energy balance (Lawson et al., 2019). While a single dose of oxytocin given to men across the weight spectrum did not change the resting energy expenditure as measured by indirect calorimetry (Lawson et al., 2015; Thienel et al., 2016), it will be critical to assess the effects of chronic oxytocin administration on energy expenditure in humans. Consistent with data in animal models, human studies show that oxytocin can modulate glucose homeostasis. Early studies examining the effects of IV oxytocin

OXYTOCIN, EATING BEHAVIOR, AND METABOLISM IN HUMANS administration showed discordant results, potentially related to different oxytocin doses, single IV dose vs continuous infusion, and differences in participant characteristics (e.g., gender and weight status). Following IV oxytocin administration, studies have shown a robust decrease in glucose levels (Burt et al., 1963), no effect on glucose or insulin levels (Spellacy et al., 1966), an increase in glucose and insulin levels (Paolisso et al., 1988), and augmented insulin levels in response to IV glucose administration (Chiodera et al., 1984). More recent investigations examined the effects of 24 IU of IN oxytocin on glucose homeostasis. A single dose of IN oxytocin was found to reduce fasting insulin levels and HOMA-IR in men across the weight spectrum (Lawson et al., 2015). Additionally, several studies have shown that a single dose of oxytocin can blunt postprandial glucose elevation in adults with and without obesity (Ott et al., 2013; Thienel et al., 2016; Klement et al., 2017) and augment insulin release following an oral glucose tolerance test in lean individuals (Klement et al., 2017). However, one dose of IN oxytocin failed to induce improvement in beta-cell responsivity in participants with obesity (Brede et al., 2019), suggesting that the effects of oxytocin on insulin sensitivity may be diminished with increasing adiposity. Furthermore, chronic oxytocin treatment for 8 weeks vs placebo had no effect on fasting blood glucose or insulin levels in individuals with obesity despite significant weight loss (Zhang et al., 2013). Additional studies are required to further define the effects of oxytocin on glucose homeostasis and impact of adiposity in humans.

THE EFFECTS OF OXYTOCIN ADMINISTRATION IN INDIVIDUALS WITH OBESITY AND SUSPECTED INSULT TO OXYTOCIN PATHWAYS The effects of IN oxytocin administration have been explored in humans with suspected disruptions in oxytocin pathways. In adult patients with Prader Willi syndrome, a single dose of IN oxytocin compared with placebo had no effect on eating behavior as assessed by questionnaires (Tauber et al., 2011). Three additional randomized, double-blind, placebo-controlled clinical trials examined the effects of oxytocin or oxytocin analogue administration in patients with Prader Willi syndrome. While one of these studies did not demonstrate an improvement in hyperphagia or eating behaviors in children and adults following 8 weeks of oxytocin compared with placebo (Einfeld et al., 2014), a second study found an improvement in eating behavior in children younger than 11 years of age treated for 4 weeks, but not in older patients (Kuppens et al., 2016). In a third

97

study, IN carbetocin, a longer-acting oxytocin analogue (Rath, 2009), improved hyperphagia in youth treated for 2 weeks (Dykens et al., 2018). None of these studies examined the effects of drug on body weight. Larger trials assessing efficacy and safety are required to assess the therapeutic potential of oxytocin-based therapeutics in this population. A phase 3 study examining the effects of IN carbetocin is ongoing (NCT03649477). In a case report of a 13-year-old male with hyperphagia and hypothalamic obesity after craniopharyngioma resection, prolonged treatment with IN oxytocin and naltrexone resulted in improvement in hyperphagia as well as significant reduction in BMI (Hsu et al., 2018). A randomized, placebo-controlled study investigating the effects of prolonged IN oxytocin administration as a weight-loss therapy for patients with hypothalamic obesity is under way (NCT02849743).

THERAPEUTIC OPPORTUNITIES AND CHALLENGES Oxytocin is a multifunctional neuropeptide involved in the regulation of eating behavior, energy homeostasis, and metabolism. Strong preclinical evidence along with early human studies supporting tolerability and safety of IN oxytocin with central and peripheral effects have catalyzed further investigation of oxytocin-based therapeutics for obesity management. However, significant challenges remain before translating this to patient care. Specifically, further research is required to better understand the physiological properties of endogenous oxytocin pathways and secretory characteristics in humans, including diurnal rhythm and pulsatility, as well as the effects of adiposity and glucose homeostasis on endogenous oxytocin levels, secretion patterns, and action. Moreover, the peripheral vs central effects of both endogenous and exogenous oxytocin need to be better characterized. The use of oxytocin as a therapeutic agent may be limited by its short half-life (3–6 min in the periphery (Vankrieken et al., 1983) and 19 min in the CNS (Mens et al., 1983)), breakdown by proteolytic enzymes in the gastrointestinal tract if given orally (Viero et al., 2010), the potential need for multiple daily dosing with an IN spray, and by downregulation of the oxytocin receptor in the setting of chronic oxytocin administration (Freeman et al., 2018). Another concern for exogenous oxytocin treatment is the cross-reactivity in binding of oxytocin to the vasopressin receptor (Baribeau and Anagnostou, 2015). Pharmacologic agents targeting the oxytocin system are under development. Long-term, large-scale clinical trials will be important to examine the safety and efficacy of oxytocin-based therapeutics in treating obesity in both sexes, to determine the most

98

L. KEREM AND E.A. LAWSON

effective route and regimen of administration, and to delineate the impact of BMI and metabolic status on therapeutic response.

COMPETING INTERESTS STATEMENT E.A.L. has a financial interest in OXT Therapeutics, a company developing an intranasal oxytocin and longacting analogs of oxytocin to treat obesity and metabolic disease. E.A.L.’s interests were reviewed and are managed by Massachusetts General Hospital and Partners HealthCare in accordance with their conflict of interest policies. This company was not involved in any way in the writing of this chapter. The other author declares no conflict of interest.

REFERENCES Acevedo SF, Valencia C, Lutter M et al. (2015). Severity of eating disorder symptoms related to oxytocin receptor polymorphisms in anorexia nervosa. Psychiatry Res 228: 641–648. Afinogenova Y, Schmelkin C, Plessow F et al. (2016). Low fasting oxytocin levels are associated with psychopathology in anorexia nervosa in partial recovery. J Clin Psychiatry 77: e1483–e1490. Ahmed SH, Guillem K, Vandaele Y (2013). Sugar addiction: pushing the drug-sugar analogy to the limit. Curr Opin Clin Nutr Metab Care 16: 434–439. Al-Rawashdeh A, Kasabri V, Bulatova N et al. (2017). The correlation between plasma levels of oxytocin and betatrophin in non-diabetic and diabetic metabolic syndrome patients: a cross sectional study from Jordan. Diabetes Metab Syndr 11: 59–67. Altemus M, Fong J, Yang R et al. (2004). Changes in cerebrospinal fluid neurochemistry during pregnancy. Biol Psychiatry 56: 386–392. Altirriba J, Poher AL, Caillon A et al. (2014). Divergent effects of oxytocin treatment of obese diabetic mice on adiposity and diabetes. Endocrinology 155: 4189–4201. Altirriba J, Poher AL, Rohner-Jeanrenaud F (2015). Chronic oxytocin administration as a treatment against impaired leptin signaling or leptin resistance in obesity. Front Endocrinol (Lausanne) 6: 119. Amico JA, Challinor SM, Cameron JL (1990). Pattern of oxytocin concentrations in the plasma and cerebrospinal fluid of lactating rhesus monkeys (Macaca mulatta): evidence for functionally independent oxytocinergic pathways in primates. J Clin Endocrinol Metab 71: 1531–1535. Amri EZ, Pisani DF (2016). Control of bone and fat mass by oxytocin. Horm Mol Biol Clin Invest 28: 95–104. Angulo MA, Butler MG, Cataletto ME (2015). Prader-Willi syndrome: a review of clinical, genetic, and endocrine findings. J Endocrinol Invest 38: 1249–1263. Aponte Y, Atasoy D, Sternson SM (2011). AGRP neurons are sufficient to orchestrate feeding behavior rapidly and without training. Nat Neurosci 14: 351–355.

Arrowsmith S, Wray S (2014). Oxytocin: its mechanism of action and receptor signalling in the myometrium. J Neuroendocrinol 26: 356–369. Atasoy D, Betley JN, Su HH et al. (2012). Deconstruction of a neural circuit for hunger. Nature 488: 172–177. Augustine RA, Seymour AJ, Campbell RE et al. (2018). Integrative neurohumoural regulation of oxytocin neurone activity in pregnancy and lactation. J Neuroendocrinol 30: e12569. Aulinas A, Plessow F, Pulumo RL et al. (2019a). Disrupted oxytocin-appetite signaling in females with anorexia nervosa. J Clin Endocrinol Metab 104: 4931–4940. Aulinas A, Pulumo RL, Asanza E et al. (2019b). Endogenous oxytocin levels in relation to food intake, menstrual phase, and age in females. J Clin Endocrinol Metab 104: 1348–1356. Bale TL, Davis AM, Auger AP et al. (2001). CNS regionspecific oxytocin receptor expression: importance in regulation of anxiety and sex behavior. J Neurosci 21: 2546–2552. Baribeau DA, Anagnostou E (2015). Oxytocin and vasopressin: linking pituitary neuropeptides and their receptors to social neurocircuits. Front Neurosci 9: 335. Baskaran C, Plessow F, Silva L et al. (2017). Oxytocin secretion is pulsatile in men and is related to social-emotional functioning. Psychoneuroendocrinology 85: 28–34. Baskin DG, Kim F, Gelling RW et al. (2010). A new oxytocinsaporin cytotoxin for lesioning oxytocin-receptive neurons in the rat hindbrain. Endocrinology 151: 4207–4213. Beier KT, Steinberg EE, Deloach KE et al. (2015). Circuit architecture of VTA dopamine neurons revealed by systematic input-output mapping. Cell 162: 622–634. Binay C, Paketci C, Guzel S et al. (2017). Serum irisin and oxytocin levels as predictors of metabolic parameters in obese children. J Clin Res Pediatr Endocrinol 9: 124–131. Bittel DC, Kibiryeva N, Sell SM et al. (2007). Whole genome microarray analysis of gene expression in Prader-Willi syndrome. Am J Med Genet A 143A: 430–442. Blanks AM, Thornton S (2003). The role of oxytocin in parturition. BJOG 110 (Suppl 20): 46–51. Blevins JE, Baskin DG (2015). Translational and therapeutic potential of oxytocin as an anti-obesity strategy: insights from rodents, nonhuman primates and humans. Physiol Behav 152: 438–449. Blevins JE, Ho JM (2013). Role of oxytocin signaling in the regulation of body weight. Rev Endocr Metab Disord 14: 311–329. Blevins JE, Eakin TJ, Murphy JA et al. (2003). Oxytocin innervation of caudal brainstem nuclei activated by cholecystokinin. Brain Res 993: 30–41. Blevins JE, Schwartz MW, Baskin DG (2004). Evidence that paraventricular nucleus oxytocin neurons link hypothalamic leptin action to caudal brain stem nuclei controlling meal size. Am J Physiol Regul Integr Comp Physiol 287: R87–R96. Blevins JE, Thompson BW, Anekonda VT et al. (2016). Chronic CNS oxytocin signaling preferentially induces fat loss in high-fat diet-fed rats by enhancing satiety

OXYTOCIN, EATING BEHAVIOR, AND METABOLISM IN HUMANS responses and increasing lipid utilization. Am J Physiol Regul Integr Comp Physiol 310: R640–R658. Blume A, Bosch OJ, Miklos S et al. (2008). Oxytocin reduces anxiety via ERK1/2 activation: local effect within the rat hypothalamic paraventricular nucleus. Eur J Neurosci 27: 1947–1956. Boccia ML, Petrusz P, Suzuki K et al. (2013). Immunohistochemical localization of oxytocin receptors in human brain. Neuroscience 253: 155–164. Bowen MT, Neumann ID (2017). Rebalancing the addicted brain: oxytocin interference with the neural substrates of addiction. Trends Neurosci 40: 691–708. Brede S, Fehr S, Dalla-Man C et al. (2019). Intranasal oxytocin fails to acutely improve glucose metabolism in obese men. Diabetes Obes Metab 21: 424–428. Brown CH, Bains JS, Ludwig M et al. (2013). Physiological regulation of magnocellular neurosecretory cell activity: integration of intrinsic, local and afferent mechanisms. J Neuroendocrinol 25: 678–710. Brunton PJ (2019). Endogenous opioid signalling in the brain during pregnancy and lactation. Cell Tissue Res 375: 69–83. Burmester V, Higgs S, Terry P (2018). Rapid-onset anorectic effects of intranasal oxytocin in young men. Appetite 130: 104–109. Burt RL, Leake NH, Dannenburg WN (1963). Effect of synthetic oxytocin on plasma nonesterified fatty acids, triglycerides, and blood glucose. Obstet Gynecol 21: 708–712. Camerino C (2009). Low sympathetic tone and obese phenotype in oxytocin-deficient mice. Obesity (Silver Spring) 17: 980–984. Carson DS, Berquist SW, Trujillo TH et al. (2015). Cerebrospinal fluid and plasma oxytocin concentrations are positively correlated and negatively predict anxiety in children. Mol Psychiatry 20: 1085–1090. Chang HH, Chang WH, Chi MH et al. (2019). The OXTR polymorphism stratified the correlation of oxytocin and glucose homeostasis in non-diabetic subjects. Diabetes Metab Syndr Obes 12: 2707–2713. Chiodera P, Coiro V, Camellini L et al. (1984). Effect of pharmacological doses of oxytocin on insulin response to glucose in normal man. Horm Res 20: 150–154. Choi YH, Fujikawa T, Lee J et al. (2013). Revisiting the ventral medial nucleus of the hypothalamus: the roles of SF-1 neurons in energy homeostasis. Front Neurosci 7: 71. Crowley WR (2015). Neuroendocrine regulation of lactation and milk production. Compr Physiol 5: 255–291. Dailey MJ, Moran TH (2013). Glucagon-like peptide 1 and appetite. Trends Endocrinol Metab 24: 85–91. Dal Monte O, Noble PL, Turchi J et al. (2014). CSF and blood oxytocin concentration changes following intranasal delivery in macaque. PLoS One 9: e103677. Daubenbuchel AM, Hoffmann A, Eveslage M et al. (2016). Oxytocin in survivors of childhood-onset craniopharyngioma. Endocrine 54: 524–531. Daubenbuchel AM, Ozyurt J, Boekhoff S et al. (2019). Eating behaviour and oxytocin in patients with childhood-onset craniopharyngioma and different grades of hypothalamic involvement. Pediatr Obes 14: e12527.

99

Davis C, Patte K, Zai C et al. (2017). Polymorphisms of the oxytocin receptor gene and overeating: the intermediary role of endophenotypic risk factors. Nutr Diabetes 7: e279. Deblon N, Veyrat-Durebex C, Bourgoin L et al. (2011). Mechanisms of the anti-obesity effects of oxytocin in diet-induced obese rats. PLoS One 6: e25565. Demitrack MA, Lesem MD, Listwak SJ et al. (1990). CSF oxytocin in anorexia nervosa and bulimia nervosa: clinical and pathophysiologic considerations. Am J Psychiatry 147: 882–886. Douglas A, Scullion S, Antonijevic I et al. (2001). Uterine contractile activity stimulates supraoptic neurons in term pregnant rats via a noradrenergic pathway. Endocrinology 142: 633–644. Dykens EM, Miller J, Angulo M et al. (2018). Intranasal carbetocin reduces hyperphagia in individuals with PraderWilli syndrome. JCI Insight 3: e98333. Einfeld SL, Smith E, McGregor IS et al. (2014). A doubleblind randomized controlled trial of oxytocin nasal spray in Prader Willi syndrome. Am J Med Genet A 164A: 2232–2239. Eisenberg Y, Dugas LR, Akbar A et al. (2018). Oxytocin is lower in African American men with diabetes and associates with psycho-social and metabolic health factors. PLoS One 13: e0190301. Eliava M, Melchior M, Knobloch-Bollmann HS et al. (2016). A new population of parvocellular oxytocin neurons controlling magnocellular neuron activity and inflammatory pain processing. Neuron 89: 1291–1304. Engel S, Klusmann H, Ditzen B et al. (2019). Menstrual cyclerelated fluctuations in oxytocin concentrations: a systematic review and meta-analysis. Front Neuroendo-crinol 52: 144–155. Fenselau H, Campbell JN, Verstegen AM et al. (2017). A rapidly acting glutamatergic ARC–>PVH satiety circuit postsynaptically regulated by alpha-MSH. Nat Neurosci 20: 42–51. Franke AA, Li X, Menden A et al. (2019). Oxytocin analysis from human serum, urine, and saliva by orbitrap liquid chromatography-mass spectrometry. Drug Test Anal 11: 119–128. Freeman SM, Samineni S, Allen PC et al. (2016). Plasma and CSF oxytocin levels after intranasal and intravenous oxytocin in awake macaques. Psychoneuroendocrinology 66: 185–194. Freeman SM, Ngo J, Singh B et al. (2018). Effects of chronic oxytocin administration and diet composition on oxytocin and vasopressin 1a receptor binding in the rat brain. Neuroscience 392: 241–251. Fu-Man D, Hong-Yu K, Bin-Hong D et al. (2019). Associations of oxytocin with metabolic parameters in obese women of childbearing age. Endokrynol Pol 70: 417–422. Galbally M, Lewis AJ, Ijzendoorn M et al. (2011). The role of oxytocin in mother-infant relations: a systematic review of human studies. Harv Rev Psychiatry 19: 1–14. Gebert D, Auer MK, Stieg MR et al. (2018). De-masking oxytocin-deficiency in craniopharyngioma and assessing

100

L. KEREM AND E.A. LAWSON

its link with affective function. Psychoneuroendo-crinology 88: 61–69. Giel K, Zipfel S, Hallschmid M (2018). Oxytocin and eating disorders: a narrative review on emerging findings and perspectives. Curr Neuropharmacol 16: 1111–1121. Herisson FM, Waas JR, Fredriksson R et al. (2016). Oxytocin acting in the nucleus accumbens core decreases food intake. J Neuroendocrinol 28: e12381. Higuchi T, Honda K, Fukuoka T et al. (1983). Pulsatile secretion of prolactin and oxytocin during nursing in the lactating rat. Endocrinol Jpn 30: 353–359. Horta M, Kaylor K, Feifel D et al. (2019). Chronic oxytocin administration as a tool for investigation and treatment: a cross-disciplinary systematic review. Neurosci Biobehav Rev 108: 1–23. Hsu EA, Miller JL, Perez FA et al. (2018). Oxytocin and naltrexone successfully treat Hypothalamic obesity in a boy post-craniopharyngioma resection. J Clin Endocrinol Metab 103: 370–375. Ivell R, Anand-Ivell R (2017). Neohormones in milk. Best Pract Res Clin Endocrinol Metab 31: 419–425. Iwasaki Y, Kumari P, Wang L et al. (2019). Relay of peripheral oxytocin to central oxytocin neurons via vagal afferents for regulating feeding. Biochem Biophys Res Commun 519: 553–558. Johnson L, Manzardo AM, Miller JL et al. (2016). Elevated plasma oxytocin levels in children with Prader-Willi syndrome compared with healthy unrelated siblings. Am J Med Genet A 170: 594–601. Jurek B, Neumann ID (2018). The oxytocin receptor: from intracellular signaling to behavior. Physiol Rev 98: 1805–1908. Kabasakalian A, Ferretti CJ, Hollander E (2018). Oxytocin and Prader-Willi Syndrome. Curr Top Behav Neurosci 35: 529–557. Kagerbauer SM, Martin J, Schuster T et al. (2013). Plasma oxytocin and vasopressin do not predict neuropeptide concentrations in human cerebrospinal fluid. J Neuroendocrinol 25: 668–673. Kim S, Strathearn L (2016). Oxytocin and maternal brain plasticity. New Dir Child Adolesc Dev 2016: 59–72. Kim YR, Kim JH, Kim MJ et al. (2014). Differential methylation of the oxytocin receptor gene in patients with anorexia nervosa: a pilot study. PLoS One 9: e88673. Klement J, Ott V, Rapp K et al. (2017). Oxytocin improves beta-cell responsivity and glucose tolerance in healthy men. Diabetes 66: 264–271. Klockars OA, Waas JR, Klockars A et al. (2017). Neural basis of ventromedial hypothalamic oxytocin-driven decrease in appetite. Neuroscience 366: 54–61. Kublaoui BM, Gemelli T, Tolson KP et al. (2008). Oxytocin deficiency mediates hyperphagic obesity of Sim1 haploinsufficient mice. Mol Endocrinol 22: 1723–1734. Kujath AS, Quinn L, Elliott ME et al. (2015). Oxytocin levels are lower in premenopausal women with type 1 diabetes mellitus compared with matched controls. Diabetes Metab Res Rev 31: 102–112. Kuppens RJ, Donze SH, Hokken-Koelega AC (2016). Promising effects of oxytocin on social and food-related

behaviour in young children with Prader-Willi syndrome: a randomized, double-blind, controlled crossover trial. Clin Endocrinol (Oxf ) 85: 979–987. Labyb M, Chretien C, Caillon A et al. (2018). Oxytocin administration alleviates acute but not chronic leptin resistance of diet-induced obese mice. Int J Mol Sci 20: 88. Larcher A, Neculcea J, Breton C et al. (1995). Oxytocin receptor gene expression in the rat uterus during pregnancy and the estrous cycle and in response to gonadal steroid treatment. Endocrinology 136: 5350–5356. Larsen PJ, Tang-Christensen M, Jessop DS (1997). Central administration of glucagon-like peptide-1 activates hypothalamic neuroendocrine neurons in the rat. Endocrinology 138: 4445–4455. Lawson EA (2017). The effects of oxytocin on eating behaviour and metabolism in humans. Nat Rev Endocrinol 13: 700–709. Lawson EA, Holsen LM, Santin M et al. (2012). Oxytocin secretion is associated with severity of disordered eating psychopathology and insular cortex hypoactivation in anorexia nervosa. J Clin Endocrinol Metab 97: E1898–E1908. Lawson EA, Marengi DA, Desanti RL et al. (2015). Oxytocin reduces caloric intake in men. Obesity (Silver Spring) 23: 950–956. Lawson EA, Olszewski PK, Weller A et al. (2019). The role of oxytocin in regulation of appetitive behaviour, body weight and glucose homeostasis. J Neuroendocrinol 32: e12805. Lee MR, Scheidweiler KB, Diao XX et al. (2018). Oxytocin by intranasal and intravenous routes reaches the cerebrospinal fluid in rhesus macaques: determination using a novel oxytocin assay. Mol Psychiatry 23: 115–122. Lefevre A, Mottolese R, Dirheimer M et al. (2017). A comparison of methods to measure central and peripheral oxytocin concentrations in human and non-human primates. Sci Rep 7: 17222. Leng G, Sabatier N (2016). Measuring oxytocin and vasopressin: bioassays, immunoassays and random numbers. J Neuroendocrinol 28: 12413. Leong KC, Cox S, King C et al. (2018). Oxytocin and rodent models of addiction. Int Rev Neurobiol 140: 201–247. Leslie M, Silva P, Paloyelis Y et al. (2018). A systematic review and quantitative meta-analysis of the effects of oxytocin on feeding. J Neuroendocrinol 30: e12584. Li T, Wang P, Wang SC et al. (2016). Approaches mediating oxytocin regulation of the immune system. Front Immunol 7: 693. Love TM (2014). Oxytocin, motivation and the role of dopamine. Pharmacol Biochem Behav 119: 49–60. Ludwig M, Leng G (2006). Dendritic peptide release and peptide-dependent behaviours. Nat Rev Neurosci 7: 126–136. Lustig RH (2011). Hypothalamic obesity after craniopharyngioma: mechanisms, diagnosis, and treatment. Front Endocrinol (Lausanne) 2: 60. Macdonald K, Feifel D (2014). Oxytocin’s role in anxiety: a critical appraisal. Brain Res 1580: 22–56.

OXYTOCIN, EATING BEHAVIOR, AND METABOLISM IN HUMANS Mackinnon AL, Carter CS, Feeley N et al. (2018). Theory of mind as a link between oxytocin and maternal behavior. Psychoneuroendocrinology 92: 87–94. Maclean EL, Wilson SR, Martin WL et al. (2019). Challenges for measuring oxytocin: the blind men and the elephant? Psychoneuroendocrinology 107: 225–231. Maejima Y, Iwasaki Y, Yamahara Y et al. (2011). Peripheral oxytocin treatment ameliorates obesity by reducing food intake and visceral fat mass. Aging (Albany NY) 3: 1169–1177. Maejima Y, Sakuma K, Santoso P et al. (2014). Oxytocinergic circuit from paraventricular and supraoptic nuclei to arcuate POMC neurons in hypothalamus. FEBS Lett 588: 4404–4412. Maejima Y, Aoyama M, Sakamoto K et al. (2017). Impact of sex, fat distribution and initial body weight on oxytocin’s body weight regulation. Sci Rep 7: 8599. Maejima Y, Yokota S, Nishimori K et al. (2018). The anorexigenic neural pathways of oxytocin and their clinical implication. Neuroendocrinology 107: 91–104. Maestrini S, Mele C, Mai S et al. (2018). Plasma oxytocin concentration in pre- and postmenopausal women: its relationship with obesity, body composition and metabolic variables. Obes Facts 11: 429–439. Marlin BJ, Mitre M, D’amour JA et al. (2015). Oxytocin enables maternal behaviour by balancing cortical inhibition. Nature 520: 499–504. Martin A, State M, Anderson GM et al. (1998). Cerebrospinal fluid levels of oxytocin in Prader-Willi syndrome: a preliminary report. Biol Psychiatry 44: 1349–1352. Massey SH, Backes KA, Schuette SA (2016). Plasma oxytocin concentration and depressive symptoms: a review of current evidence and directions for future research. Depress Anxiety 33: 316–322. McCormack SE, Blevins JE, Lawson EA (2019). Metabolic effects of oxytocin. Endocr Rev 41: 121–145. McGregor IS, Bowen MT (2012). Breaking the loop: oxytocin as a potential treatment for drug addiction. Horm Behav 61: 331–339. Mens WB, Witter A, Van Wimersma Greidanus TB (1983). Penetration of neurohypophyseal hormones from plasma into cerebrospinal fluid (CSF): half-times of disappearance of these neuropeptides from Csf. Brain Res 262: 143–149. Meye FJ, Adan RA (2014). Feelings about food: the ventral tegmental area in food reward and emotional eating. Trends Pharmacol Sci 35: 31–40. Meyer-Lindenberg A, Domes G, Kirsch P et al. (2011). Oxytocin and vasopressin in the human brain: social neuropeptides for translational medicine. Nat Rev Neurosci 12: 524–538. Miaskiewicz SL, Stricker EM, Verbalis JG (1989). Neurohypophyseal secretion in response to cholecystokinin but not meal-induced gastric distention in humans. J Clin Endocrinol Metab 68: 837–843. Micali N, Crous-Bou M, Treasure J et al. (2017). Association between oxytocin receptor genotype, maternal care, and eating disorder behaviours in a community sample of women. Eur Eat Disord Rev 25: 19–25.

101

Mishra M, Ali S, Das M (2014). Analysis of oxytocin in milk samples and intake pattern in different age groups of Indian population. Toxicol Mech Methods 24: 342–346. Mohan S, Khan D, Moffett RC et al. (2018). Oxytocin is present in islets and plays a role in beta-cell function and survival. Peptides 100: 260–268. Monteleone AM, Scognamiglio P, Volpe U et al. (2016). Investigation of oxytocin secretion in anorexia nervosa and bulimia nervosa: relationships to temperament personality dimensions. Eur Eat Disord Rev 24: 52–56. Muller HL (2016). Craniopharyngioma and hypothalamic injury: latest insights into consequent eating disorders and obesity. Curr Opin Endocrinol Diabetes Obes 23: 81–89. Mullis K, Kay K, Williams DL (2013). Oxytocin action in the ventral tegmental area affects sucrose intake. Brain Res 1513: 85–91. Noble EE, Billington CJ, Kotz CM et al. (2014). Oxytocin in the ventromedial hypothalamic nucleus reduces feeding and acutely increases energy expenditure. Am J Physiol Regul Integr Comp Physiol 307: R737–R745. O’hara CB, Campbell IC, Schmidt U (2015). A reward-centred model of anorexia nervosa: a focussed narrative review of the neurological and psychophysiological literature. Neurosci Biobehav Rev 52: 131–152. Ohlsson B, Forsling ML, Rehfeld JF et al. (2002). Cholecystokinin stimulation leads to increased oxytocin secretion in women. Eur J Surg 168: 114–118. Ohlsson B, Bjorgell O, Ekberg O et al. (2006a). The oxytocin/ vasopressin receptor antagonist atosiban delays the gastric emptying of a semisolid meal compared to saline in human. BMC Gastroenterol 6: 11. Ohlsson B, Truedsson M, Djerf P et al. (2006b). Oxytocin is expressed throughout the human gastrointestinal tract. Regul Pept 135: 7–11. Olazabal DE, Young LJ (2006). Oxytocin receptors in the nucleus accumbens facilitate “spontaneous” maternal behavior in adult female prairie voles. Neuroscience 141: 559–568. Olff M, Frijling JL, Kubzansky LD et al. (2013). The role of oxytocin in social bonding, stress regulation and mental health: an update on the moderating effects of context and interindividual differences. Psychoneuroendocrinology 38: 1883–1894. Ong ZY, Alhadeff AL, Grill HJ (2015). Medial nucleus tractus solitarius oxytocin receptor signaling and food intake control: the role of gastrointestinal satiation signal processing. Am J Physiol Regul Integr Comp Physiol 308: R800–R806. Ott V, Finlayson G, Lehnert H et al. (2013). Oxytocin reduces reward-driven food intake in humans. Diabetes 62: 3418–3425. Paolisso G, Sgambato S, Passariello N et al. (1988). Pharmacological doses of oxytocin affect plasma hormone levels modulating glucose homeostasis in normal man. Horm Res 30: 10–16. Pataky Z, Guessous I, Caillon A et al. (2019). Variable oxytocin levels in humans with different degrees of obesity and impact of gastric bypass surgery. Int J Obes (Lond) 43: 1120–1124.

102

L. KEREM AND E.A. LAWSON

Perello M, Raingo J (2013). Leptin activates oxytocin neurons of the hypothalamic paraventricular nucleus in both control and diet-induced obese rodents. PLoS One 8: e59625. Peris J, Macfadyen K, Smith JA et al. (2017). Oxytocin receptors are expressed on dopamine and glutamate neurons in the mouse ventral tegmental area that project to nucleus accumbens and other mesolimbic targets. J Comp Neurol 525: 1094–1108. Plante E, Menaouar A, Danalache BA et al. (2015). Oxytocin treatment prevents the cardiomyopathy observed in obese diabetic male db/db mice. Endocrinology 156: 1416–1428. Plessow F, Eddy KT, Lawson EA (2018a). The neuropeptide hormone oxytocin in eating disorders. Curr Psychiatry Rep 20: 91. Plessow F, Marengi DA, Perry SK et al. (2018b). Effects of intranasal oxytocin on the blood oxygenation leveldependent signal in food motivation and cognitive control pathways in overweight and obese men. Neuropsychopharmacology 43: 638–645. Qian W, Zhu T, Tang B et al. (2014). Decreased circulating levels of oxytocin in obesity and newly diagnosed type 2 diabetic patients. J Clin Endocrinol Metab 99: 4683–4689. Qin J, Feng M, Wang C et al. (2009). Oxytocin receptor expressed on the smooth muscle mediates the excitatory effect of oxytocin on gastric motility in rats. Neurogastroenterol Motil 21: 430–438. Rath W (2009). Prevention of postpartum haemorrhage with the oxytocin analogue carbetocin. Eur J Obstet Gynecol Reprod Biol 147: 15–20. Rice LJ, Einfeld SL, Hu N et al. (2018). A review of clinical trials of oxytocin in Prader-Willi syndrome. Curr Opin Psychiatry 31: 123–127. Rinaman L, Rothe EE (2002). GLP-1 receptor signaling contributes to anorexigenic effect of centrally administered oxytocin in rats. Am J Physiol Regul Integr Comp Physiol 283: R99–106. Roberts ZS, Wolden-Hanson T, Matsen ME et al. (2017). Chronic hindbrain administration of oxytocin is sufficient to elicit weight loss in diet-induced obese rats. Am J Physiol Regul Integr Comp Physiol 313: R357–R371. Ryden G, Sjoholm I (1969). Half-life of oxytocin in blood of pregnant and non-pregnant women. Acta Endocrinol 61: 425–431. Sabihi S, Dong SM, Maurer SD et al. (2017). Oxytocin in the medial prefrontal cortex attenuates anxiety: anatomical and receptor specificity and mechanism of action. Neuropharmacology 125: 1–12. Schorr M, Marengi DA, Pulumo RL et al. (2017). Oxytocin and its relationship to body composition, bone mineral density, and hip geometry across the weight spectrum. J Clin Endocrinol Metab 102: 2814–2824. Skinner JA, Garg ML, Dayas CV et al. (2019). Is weight status associated with peripheral levels of oxytocin? A pilot study in healthy women. Physiol Behav 212: 112684. Smith AS, Korgan AC, Young WS (2019). Oxytocin delivered nasally or intraperitoneally reaches the brain and plasma of normal and oxytocin knockout mice. Pharmacol Res 146: 104324.

Spellacy WN, Carlson KL, Birk SA (1966). Effect of posterior pituitary hormones on blood glucose and plasma insulin levels in postpartum patients. Obstet Gynecol 28: 355–359. Spetter MS, Feld GB, Thienel M et al. (2018). Oxytocin curbs calorie intake via food-specific increases in the activity of brain areas that process reward and establish cognitive control. Sci Rep 8: 2736. Stock S, Granstrom L, Backman L et al. (1989). Elevated plasma levels of oxytocin in obese subjects before and after gastric banding. Int J Obes (Lond) 13: 213–222. Striepens N, Kendrick KM, Hanking V et al. (2013). Elevated cerebrospinal fluid and blood concentrations of oxytocin following its intranasal administration in humans. Sci Rep 3: 3440. Suzuki M, Honda Y, Li MZ et al. (2013). The localization of oxytocin receptors in the islets of Langerhans in the rat pancreas. Regul Pept 183: 42–45. Swaab DF, Purba JS, Hofman MA (1995). Alterations in the hypothalamic paraventricular nucleus and its oxytocin neurons (putative satiety cells) in Prader-Willi syndrome: a study of five cases. J Clin Endocrinol Metab 80: 573–579. Szeto A, McCabe PM, Nation DA et al. (2011). Evaluation of enzyme immunoassay and radioimmunoassay methods for the measurement of plasma oxytocin. Psychosom Med 73: 393–400. Szulc P, Amri EZ, Varennes A et al. (2016). High serum oxytocin is associated with metabolic syndrome in older men—the Minos study. Diabetes Res Clin Pract 122: 17–27. Takagi T, Tanizawa O, Otsuki Y et al. (1985). Oxytocin in the cerebrospinal fluid and plasma of pregnant and nonpregnant subjects. Horm Metab Res 17: 308–310. Takayanagi Y, Kasahara Y, Onaka T et al. (2008). Oxytocin receptor-deficient mice developed late-onset obesity. Neuroreport 19: 951–955. Takeda S, Kuwabara Y, Mizuno M (1986). Concentrations and origin of oxytocin in breast milk. Endocrinol Jpn 33: 821–826. Tauber M, Mantoulan C, Copet P et al. (2011). Oxytocin may be useful to increase trust in others and decrease disruptive behaviours in patients with Prader-Willi syndrome: a randomised placebo-controlled trial in 24 patients. Orphanet J Rare Dis 6: 47. Thienel M, Fritsche A, Heinrichs M et al. (2016). Oxytocin’s inhibitory effect on food intake is stronger in obese than normal-weight men. Int J Obes (Lond) 40: 1707–1714. Uvnas Moberg K, Handlin L, Kendall-Tackett K et al. (2019). Oxytocin is a principal hormone that exerts part of its effects by active fragments. Med Hypotheses 133: 109394. Valtcheva S, Froemke RC (2019). Neuromodulation of maternal circuits by oxytocin. Cell Tissue Res 375: 57–68. Van Der Klaauw AA, Ziauddeen H, Keogh JM et al. (2017). Oxytocin administration suppresses hypothalamic activation in response to visual food cues. Sci Rep 7: 4266. Vankrieken L, Godart A, Thomas K (1983). Oxytocin determination by radioimmunoassay. Gynecol Obstet Invest 16: 180–185.

OXYTOCIN, EATING BEHAVIOR, AND METABOLISM IN HUMANS Viero C, Shibuya I, Kitamura N et al. (2010). Review: oxytocin: crossing the bridge between basic science and pharmacotherapy. CNS Neurosci Ther 16: e138–e156. Wang YL, Yuan Y, Yang J et al. (2013). The interaction between the oxytocin and pain modulation in headache patients. Neuropeptides 47: 93–97. Weingarten MFJ, Scholz M, Wohland T et al. (2019). Circulating oxytocin is genetically determined and associated with obesity and impaired glucose tolerance. J Clin Endocrinol Metab 104: 5621–5632. Wheeler E, Huang N, Bochukova EG et al. (2013). Genomewide Snp and CNV analysis identifies common and lowfrequency variants associated with severe early-onset obesity. Nat Genet 45: 513–517.

Xi

103

TF, Li DN, Li YY et al. (2019). Central 5-hydroxytryptamine (5-HT) mediates colonic motility by hypothalamus oxytocin-colonic oxytocin receptor pathway. Biochem Biophys Res Commun 508: 959–964. Xiao L, Priest MF, Nasenbeny J et al. (2017). Biased oxytocinergic modulation of midbrain dopamine systems. Neuron 95: 368–384.e5. Zhang G, Cai D (2011). Circadian intervention of obesity development via resting-stage feeding manipulation or oxytocin treatment. Am J Physiol Endocrinol Metab 301: E1004–E1012. Zhang H, Wu C, Chen Q et al. (2013). Treatment of obesity and diabetes using oxytocin or analogs in patients and mouse models. PLoS One 8: e61477.

This page intentionally left blank

Handbook of Clinical Neurology, Vol. 180 (3rd series) The Human Hypothalamus: Middle and Posterior Region D.F. Swaab, F. Kreier, P.J. Lucassen, A. Salehi, and R.M. Buijs, Editors https://doi.org/10.1016/B978-0-12-820107-7.00007-0 Copyright © 2021 Elsevier B.V. All rights reserved

Chapter 7

The supraoptic and paraventricular nuclei in healthy aging and neurodegeneration CHLOE A. STEWART1,2 AND ELIZABETH C. FINGER1,3* 1

Department of Clinical Neurological Sciences, Lawson Health Research Institute, London, ON, Canada

2

Graduate Program in Neuroscience, Schulich School of Medicine and Dentistry, University of Western Ontario, London, ON, Canada 3

Department of Clinical Neurological Sciences, Schulich School of Medicine and Dentistry, University of Western Ontario, London, ON, Canada

Abstract The supraoptic (SON) and paraventricular (PVN) nuclei of the hypothalamus undergo structural and functional changes over the course of healthy aging. These nuclei and their connections are also heterogeneously affected by several different neurodegenerative diseases. This chapter reviews the involvement of the SON and PVN, the hypothalamic–pituitary axes, and the peptide hormones produced in both nuclei in healthy aging and in neurodegeneration, with a focus on Alzheimer’s disease (AD), frontotemporal dementia (FTD), amyotrophic lateral sclerosis, progressive supranuclear palsy, Parkinson’s disease (PD), dementia with Lewy bodies (DLB), multiple system atrophy, and Huntington’s disease. Although age-related changes occur in several regions of the hypothalamus, the SON and PVN are relatively preserved during aging and in many neurodegenerative disorders. With aging, these nuclei do undergo some sexually dimorphic changes including changes in size and levels of vasopressin and corticotropin-releasing hormone, likely due to age-related changes in sex hormones. In contrast, oxytocinergic cells and circulating levels of thyrotropin-releasing hormone remain stable. A relative resistance to many forms of neurodegenerative pathology is also observed, in comparison to other hypothalamic and brain regions. Mirroring the pattern observed in aging, pathologic hallmarks of AD, and some subtypes of FTD are observed in the PVN, though to a milder degree than are observed in other brain regions, while the SON is relatively spared. In contrast, the SON appears more vulnerable to alpha-synuclein pathology of DLB and PD. The consequences of these alterations may help to inform several of the physiologic changes observed in aging and neurodegenerative disease.

THE SUPRAOPTIC AND PARAVENTRICULAR NUCLEI IN HEALTHY AGING Morphological changes in aging The hypothalamus has been observed to undergo a number of morphological changes during healthy aging, notably volume loss and inflammation, which occur across a number of its nuclei. These changes correspond to alterations in

homeostatic regulation and various behavioral changes that have been observed in aging (Tang et al., 2015; Kim and Choe, 2019). The paraventricular (PVN) and supraoptic nuclei (SON) of the hypothalamus, by contrast, are noted for being largely untouched in the course of healthy aging. Numerous studies of the aging brain have found preservation of the cytoskeletal architecture, neuronal activity, and connectivity of both nuclei (Goudsmit

*Correspondence to: Elizabeth C. Finger, M.D., Parkwood Institute, St. Joseph’s Health Care London, P.O. Box 5777, Stn B, London, ON N6A 4 V2, Canada. Tel: +1-519-685-8500x66032, Fax: +1-519-646-6226, E-mail: [email protected]

106

C.A. STEWART AND E.C. FINGER

Fig. 7.1. Frontal tuberoinfundibular region of the hypothalamus in an apparently unaffected 77-year-old male, stained with AT8 to display abnormally phosphorylated tau and AT8-immunoreactive perikarya in and between the PVN and SON (A). Sections of the tuberal hypothalamus of an unaffected 73-year-old male showing NFT pathology in PVN and SON labeled with AT8 immunostaining (B) and the Gallyas silver technique (C). NFT pathology in the PVN, labeled with AT8 (D), PHF-1 (E), and Gallyas silver technique (F). PVN pathology at higher magnification: AT8-immunostained neuronal perikaryal and dystrophic neurites and axons (G). PHF-1-immunostained neuronal perikaryal and dystrophic neurites and axons (H). NFTs and dystrophic neurites labeled with the Gallyas silver technique (I). NFT, neurofibrillary tangles; PVN, paraventricular nucleus; SON, supraoptic nucleus. Reprinted with permission from Schultz, C., Ghebremedhin, E., Braak, H., et al., 1997a. Neurofibrillary pathology in the human paraventricular and supraoptic nuclei. Acta Neuropathol 94, 99–102. doi: 10.1007/s004010050679.

et al., 1990; Wierda et al., 1991, Lucassen et al., 1994, 1997; Van der Woude et al., 1995; Cai, 2018). There is some evidence that a subpopulation of apparently healthily aging individuals may develop aggregations of neurofibrillary tangles (NFTs) in the magnocellular nuclei of both the PVN and SON. These individuals appear to have no other Alzheimer-like pathological changes in other regions of the brain, and no history of reported cognitive or behavioral changes associated with these pathological alterations. This appears to be a sex-linked finding, with men significantly more likely than women to exhibit the most severe of these seemingly harmless changes (Fig. 7.1) (Schultz et al., 1997a, b, c). The exact causes or implications of these accumulations are as yet unclear (Swaab, 1999).

Sex differences in aging VASOPRESSIN While there is no notable degradation of the PVN or SON in healthy aging, there are sexually dimorphic changes that occur over the course of the lifespan. In young people, vasopressinergic (VP) cells in the PVN and SON are present in similar numbers across genders (Ishunina et al., 2000a). However, these neurons have been observed to be significantly larger in men than women (Swaab et al., 2001). Over the course of aging, and particularly postmenopause, the VP neurons of the PVN and SON in women grow significantly, approaching the size of those in elderly men, indicating a gradual activation of these neurons (Ishunina and Swaab, 1999). This age-

THE SUPRAOPTIC AND PARAVENTRICULAR NUCLEI IN NEURODEGENERATION related size change does not occur in men, who maintain a relatively steady VP cell size throughout the life course (Ishunina et al., 1999). This size change in women, and stability in men, is believed to be due to the action of estrogen. Several pathologic studies have found estrogen suppresses the growth and activation of VP neurons in young women via estrogen receptor ß inhibition of p75 neurotrophin receptors in both the PVN and SON (Ishunina et al., 2000a,b, 2001).

OXYTOCIN Though the VP system is highly sexually dimorphic across the lifespan, the same is not true of the morphology of the oxytocinergic (OT) system (Van der Woude et al., 1995). OT neurons in the PVN and SON occur in similar numbers across genders, and their number remains constant during healthy aging (Wierda et al., 1991). Similarly, they do not exhibit any growth or atrophy with aging for either gender (Fliers et al., 1985).

CORTICOTROPIN-RELEASING HORMONE Like VP neurons, the corticotropin-releasing hormone (CRH)-producing neurons in the PVN show a sexually dimorphic pattern; however, the difference begins in young adulthood and persists into senescence (Bao and Swaab, 2007). Men have significantly more CRH neurons than women beginning in early adulthood, and the total number of CRH neurons continues to increase during aging in men only, indicating an increased activity; women show no age-related changes in CRH neuron number (Bao and Swaab, 2007). This difference appears to be driven by the interplay between sex hormones, particularly the inhibitory action of androgens and enhancing effect of estrogens (Lund et al., 2004; Bao et al., 2006; Bao and Swaab, 2007). In both sexes, there has been a strong indication that the percentage of CRH neurons that coexpress VP increases progressively with age. This is interpreted as a mark of the increasing activation of CRH with age, as CRH neurons begin to coexpress VP when chronically activated (Raadsheer et al., 1993; Zhou and Swaab, 1999; Swaab et al., 2005).

THYROTROPIN-RELEASING HORMONE No published study to date has investigated morphological changes in the thyrotropin-releasing hormone (TRH)producing neurons in the PVN in relation to age or sex.

Hormone changes in aging VASOPRESSIN Vasopressin is predominantly expressed in the SON, though some VP neurons are also present in the PVN

107

(Swaab and Bao, 2011). While there is a general increase in circulating levels of VP in the blood with age, the results supporting this finding are mixed (Sannino et al., 2017; Plasencia et al., 2019). It is possible that this lack of clarity on VP secretion in aging is due to the initial sex difference in VP: men typically produce more VP than women, attributable to the previously noted size differences of VP neurons. With aging and the increasing size of VP neurons, women show a substantial increase in circulating VP levels that men do not; however, men continue to have higher levels of VP overall (Asplund and Åberg, 1991; Ishunina and Swaab, 1999). Vasopressin is secreted in a diurnal rhythm with biphasic peaks, one at night and one during the day; in young people, the nocturnal peak of VP in plasma is greater than the diurnal peak (Nadal et al., 1994). Over the course of healthy aging, the nocturnal peak of VP is attenuated and may vanish altogether (Asplund, 2002; Duffy et al., 2016).

OXYTOCIN Oxytocin is predominantly produced in the PVN, with some OT neurons also present in the SON (Ishunina and Swaab, 1999). While studies in rats and monkeys have indicated a decline or increase in OT production with age (Ebner et al., 2013), numerous studies in humans have found stable OT production and circulating levels in the blood in healthy aging (Chiodera et al., 1994; Plasencia et al., 2019). This is further supported by the preservation of cell architecture in OT neurons during aging (Dumais and Veenema, 2016). There is a proposed sex difference in early life that may continue into senescence: women have sometimes been found to have more circulating oxytocin than men (Miller et al., 2013; Marazziti et al., 2019; Plasencia et al., 2019). This finding remains controversial, as many other studies have reported the opposite finding or no difference between the sexes (Taylor et al., 2010; Zhong et al., 2012; Weisman et al., 2013). In contrast to other neuropeptides synthesized in the PVN and SON, there does not appear to be any circadian rhythmicity to the release of OT (Graugaard-Jensen et al., 2014, 2017).

CORTICOTROPIN-RELEASING HORMONE CRH is secreted in a circadian and ultradian rhythm by the PVN, as well as in a pulsatile fashion in response to external stimuli (Watabe et al., 1987; Swaab et al., 2005). Regardless of age, women have been observed to have greater CRH activity and secretion than men (Antonijevic et al., 1999; Laughlin and Barrett-Connor, 2000). Studies using immunohistochemical staining of postmortem human and rat brains have established that CRH levels increase with aging in both genders (Swaab et al., 1994, 2005; Ceccatelli et al., 1996;

108

C.A. STEWART AND E.C. FINGER

Aguilera, 2011). This age-related increase has been measured and confirmed directly through CRH levels in blood and cerebrospinal fluid (CSF) and indirectly through measurements of cortisol, a glucocorticoid that is stimulated by CRH (Laughlin and Barrett-Connor, 2000; Tsigos and Chrousos, 2002; Erkut et al., 2004; Swaab et al., 2005). The diurnal rhythmicity observed in CRH production, in which there is a morning peak and nighttime dip, is still be present with aging, although its amplitude is attenuated (Van Cauter et al., 1996; Ferrari et al., 2001). The timing of the morning peak and evening dip are also observed to change, occurring earlier in the day as age increases (Van Cauter et al., 1996).

accompanied by a concomitant increase in circulating levels of TSH in the blood (Bremner et al., 2012). There is a notable decrease in the response of TSH to the action of TRH in both sexes; elderly individuals show a blunted TSH response to TRH stimulation (Olsson et al., 1989; Monzani et al., 1996; Mazzoccoli et al., 2010). These alterations appear to be directed in large part by the PVN, given its central role in the control and regulation of HPT activity (Chiamolera and Wondisford, 2009).

THYROTROPIN-RELEASING HORMONE

Though early research suggested that the PVN and SON were largely uninvolved in the course of pathological aging, subsequent studies have suggested that they are affected by a number of different neurodegenerative conditions.

TRH is secreted by the PVN in ultradian and circadian rhythm, as observed directly via TRH measurement, and by its effects on thyroid-stimulating hormone (TSH), a pituitary hormone that is stimulated by TRH (Greenspan et al., 1986; Brabant et al., 1990; Mazzoccoli et al., 2010; Kalsbeek and Fliers, 2013). Levels of TRH do not appear to be altered in aging, as recent studies have found no actual change in either TRH secretion or circulation itself, nor free in thyroxine serum levels in either sex (Mazzoccoli et al., 2010; Bremner et al., 2012). Despite this, circulating levels of TSH are generally higher or present in a wider range of values with a higher maximum, in elderly individuals (Surks and Hollowell, 2007; Atzmon et al., 2009). The daily rhythmicity of TRH secretion, with a nocturnal peak and daytime nadir, does not seem to be altered in aging (Mazzoccoli et al., 2010; Kalsbeek and Fliers, 2013).

Circuit changes in aging

THE SUPRAOPTIC AND PARAVENTRICULAR NUCLEI IN NEURODEGENERATION

Alzheimer’s disease ABNORMAL INCLUSIONS AND AGGREGATES The most common pathological findings in Alzheimer’s disease (AD) are amyloid ß (Aß)-positive plaques and tau-positive NFTs. The PVN appears to be particularly susceptible to these pathological inclusions, as it is often observed to develop plaques, NFTs, and dystrophic neurites (Fig. 7.2); the SON remains relatively untouched until very late in the disease, if it is involved at all (Swaab et al., 1992; van de Nes et al., 1993; Iwatsubo et al., 1996; Diodati et al., 2012; Baloyannis et al., 2015). The amyloid plaques and NFTs in these hypothalamic nuclei often come quite late in the disease progression and are typically

HYPOTHALAMIC–PITUITARY–ADRENAL AXIS Healthy aging is associated with increasing basal hyperactivity of the hypothalamic–pituitary–adrenal (HPA) axis that is nonetheless within clinically normal boundaries (Dodt et al., 1994; Deuschle et al., 1997). This increase in activity is higher in men than in women (Traustadóttir et al., 2003; Bao and Swaab, 2007). The exact mechanism of this age-related increase is as yet unclear; it may be related to degraded inhibitory hippocampal feedback to the PVN, increasing adrenal sensitivity to HPA activity, or increasing insensitivity of the HPA circuit to glucocorticoid feedback (Swaab and Bao, 2011). It is clear, however, that there is increased activation of CRH neurons in the PVN with age, which indicates the PVN’s central role in HPA hyperactivity (Bao et al., 2008).

HYPOTHALAMIC–PITUITARY–THYROID AXIS Over the course of healthy aging the hypothalamic–pituitary–thyroid (HPT) appears to undergo a gradual decrease in activity (van den Beld et al., 2018). This is

Fig. 7.2. Histological section of the PVN of a patient with Alzheimer’s disease demonstrating tau-paired helical filament immunostaining (x400 magnification). PVN, paraventricular nucleus. Reprinted with permission from Diodati, D., CynAng, L., Kertesz, A., et al., 2012. Pathologic evaluation of the supraoptic and paraventricular nuclei in dementia. Can J Neurol Sci 39, 213–219. doi: 10.1017/s0317167100013251.

THE SUPRAOPTIC AND PARAVENTRICULAR NUCLEI IN NEURODEGENERATION observed to be less severe than the pathology observed in more classically involved regions of the cortex or other regions of the hypothalamus such as the suprachiasmatic nucleus (SCN), the lateral hypothalamic area, and the tuberomamillary nucleus (Braak and Braak, 1991; Baloyannis et al., 2015, 2018; Ishii and Iadecola, 2015).

NEURONAL LOSS AND OTHER PATHOLOGICAL CHANGES As with the pathological aggregates and inclusions observed in AD, the alterations of neurons and their connections are often mild and slow to develop in the PVN and SON (Baloyannis et al., 2018). Some evidence suggests that there may be a pattern of gradual cell loss and decline in nucleolar volume in the PVN and SON (Mann et al., 1985; Baloyannis et al., 2015). However, these results are controversial, as numerous studies have found that there is no notable change in cell counts or volumes in AD relative to controls (Goudsmit et al., 1990; Wierda et al., 1991; Van der Woude et al., 1995; Diodati et al., 2012). It is possible that the progression of atrophy in these nuclei is simply so slow and mild that any changes in number or volume are too small to be reliably captured (Baloyannis et al., 2018). Studies have also suggested that there are marked morphological alterations of the neurons in the PVN and SON. Both nuclei experience significant loss of dendritic spines and branches, though as before these changes are milder than in other nuclei (Baloyannis et al., 2015, 2018). Again, the PVN appears to be particularly susceptible to morphological alterations, as in AD it is observed to have notable Golgi apparatus fragmentation and atrophy, as well as mitochondrial alterations consisting primarily of disrupted cristae and fibrillary accumulations (Baloyannis et al., 2015, 2018). The SON, by contrast, does not seem to experience these pathological changes and may indeed become more active, with cells expressing VP having been found to have increased nucleolar and Golgi apparatus size (Fliers et al., 1985; Lucassen et al., 1994; Baloyannis et al., 2015).

HORMONAL CHANGES Studies have consistently found that over the course of AD there is a significant decline in circulating levels of VP in the plasma and CSF (Raskind et al., 1986; North et al., 1992; Edvinsson et al., 1993). However, evidence for changes in the levels of VP in the SON, PVN, and hypothalamus is equivocal, with some studies finding no changes in these nuclei, while others suggest agerelated changes in VP levels that are due to impairment in the PVN or SON (Mann et al., 1985; Goudsmit et al., 1990; North et al., 1992). Variations in OT levels in AD are less clearly established, but generally are the same as in healthy aging, or marginally decreased in AD (Raskind et al., 1986;

109

North et al., 1992). In AD, OT levels do not appear to increase or decrease in the PVN or SON (Mazurek et al., 1987; Wierda et al., 1991). Research suggests CRH levels in plasma and CSF tend to decline or remain stable in AD relative to healthy aging, and the same is true throughout much of the brain (Bissette et al., 1985; May et al., 1987; Nemeroff et al., 1989). However, the PVN shows much higher CRH concentrations in patients with AD, which may drive the HPA hyperactivity discussed later (Raadsheer et al., 1995). Relatively little research has examined the alterations in TRH levels in AD. Two studies that investigated TRH levels in the brain and in CSF did not find differences in nondepressed AD patients relative to controls (Nemeroff et al., 1989; Banki et al., 1992). One of these studies often struggled to detect TRH in the brain, thus further research is needed to identify the role of TRH in AD (Nemeroff et al., 1989).

CIRCUIT ALTERATIONS AND DISRUPTION Perhaps one of the most consistent findings in AD research is an early hyperactivity of the HPA axis that is tightly tied to progression of the disease (Gil-Bea et al., 2010; Popp et al., 2015; Newhouse and Chemali, 2019). No research has clearly indicated which specific element of the HPA axis is overactive, but it appears to be a combination of high cortisol levels and insensitivity to glucocorticoid feedback (Chowen and Garcia-Segura, 2019). Given the early onset of HPA perturbance and late onset of pathological changes in the PVN, it may be that the PVN is responding to dysfunction at some other point in the axis, and not the source of dysfunction itself (Swaab et al., 2005; Popp et al., 2015). However, some research has indicated that the PVN’s CRH neurons may be partially responsible for HPA axis hyperactivity (Hatzinger et al., 1995; Raadsheer et al., 1995). Function of the HPT axis is less clear than for the HPA, but overall studies suggest that there is hypoactivity of the HPT axis in AD (Yong-Hong et al., 2013; Wang et al., 2016; Newhouse and Chemali, 2019). As this has been predominantly assessed through circulating levels of TSH and thyroid hormones in the blood and CSF, the level of involvement of the PVN is unclear. However, the low levels of both TRH and TSH suggest a hypothalamic dysfunction, either in secretion of TRH or insensitivity to thyroid hormone feedback (Ishii and Iadecola, 2015).

Frontotemporal dementia ABNORMAL INCLUSIONS AND AGGREGATES The PVN and SON appear to be relatively resistant to TDP-43 pathology in frontotemporal dementia (FTD), though there is some evidence that TDP-43 may aggregate in the parvocellular neurons of the PVN in

110

C.A. STEWART AND E.C. FINGER

Fig. 7.3. Magnocellular cells of the PVN and SON in the hypothalamus of FTLD patients with C9orf72 (A, B, D, and E) and nonC9orf72 (C and F) mutations. The magnocellular cells of the SON (A–C) and PVN (D–F) were spared from poly(GA) and pTDP-43 pathology in both subtypes. Insets in D and F show poly(GA) (D) and pTDP-43 (F) pathology in the smaller neurons between the magnocellular neurons. Scale bar represents 100 mm in (A–F) and 10 mm in insets. FTLD, frontotemporal lobar degeneration; PVN, paraventricular nucleus; SON, supraoptic nucleus. Reprinted with permission from Dedeene, L., Van Schoor, E., Vandenberghe, R., et al., 2019. Circadian sleep/wake-associated cells show dipeptide repeat protein aggregates in C9orf72-related ALS and FTLD cases. Acta Neuropathol Commun 7, 189. doi: 10.1186/s40478-019-0845-9.

frontotemporal lobar degeneration (FTLD) of the TDP43 type associated with an FTD-ALS spectrum disorder (TDP-43 type B) (Fig. 7.3) (Diodati et al., 2012; Dedeene et al., 2019). Dipeptide repeat protein polyGA aggregations have also been found in the parvocelluar PVN in individuals with FTLD caused by mutations in C9orf72 (Dedeene et al., 2019). The parvocellular PVN may also accumulate dipeptide repeat protein aggregations, though these results have yet to be replicated (Dedeene et al., 2019). Though ubiquitin staining has been observed in both the PVN and SON in FTLD, in neither nucleus is the level significantly different from controls (Diodati et al., 2012). Significant tau staining of the whole hypothalamus has been found in FTLD of the tau subtype, with occasional Pick body inclusions (Piguet et al., 2011). No published research to date has investigated FUS-positive inclusion bodies in the PVN or SON.

NEURONAL LOSS AND OTHER PATHOLOGICAL CHANGES Hypothalamic atrophy is frequently observed in FTD, particularly in the behavioral variant of FTD, and in FTLD of the TDP-43 subtype (Piguet et al., 2011; Ahmed et al., 2015). In a study that segmented the hypothalamus into smaller subsections, the PVN was found to be among the atrophied nuclei, while the SON’s volume remained stable (Bocchetta et al., 2015). Although there is atrophy in the volume of the PVN, it does not appear to be due to cell loss, as studies have found no decrease in

total number of neuropeptide-immunoreactive cells in the PVN (Piguet et al., 2011; Diodati et al., 2012).

HORMONAL CHANGES A single study identified a significant decrease in VP levels in the CSF in FTD relative to healthy controls (Edvinsson et al., 1993). OT levels in the blood appear to remain fairly consistent in FTD, with patients showing no OT change relative to healthy controls or to individuals with AD (Ahmed et al., 2015). No studies have yet examined CSF OT levels in FTD. CRH levels appear to decline significantly in FTD relative to both healthy controls and individuals with AD (Edvinsson et al., 1993). No research published to date has examined alterations to the level of TRH peripherally or centrally in FTD. Given the scant evidence on hormonal alterations in FTD, at present it is difficult to establish the etiology of the observed changes, and whether they might be related specifically to the PVN or SON.

CIRCUIT ALTERATIONS AND DISRUPTION No published research to date has examined the HPA or HPT axes in any subtype of FTD. There are numerous indications of autonomic and endocrine disturbances in this disorder, but the degree of involvement of

THE SUPRAOPTIC AND PARAVENTRICULAR NUCLEI IN NEURODEGENERATION PVN connections to the PA or PT axis are unknown (Newhouse and Chemali, 2019).

Amyotrophic lateral sclerosis ABNORMAL INCLUSIONS AND AGGREGATES As in FTD, the PVN and the SON appear to be relatively unaffected by the TDP-43 pathology that is otherwise prevalent throughout the hypothalamus and other regions of the brain in amyotrophic lateral sclerosis (ALS) (Cykowski et al., 2014). However, as noted previously, Dedeene et al. (2019) found possible TDP-43 and dipeptide repeat protein inclusions in the parvocellular neurons of the PVN associated with an FTD-ALS spectrum disorder.

NEURONAL LOSS AND PATHOLOGICAL CHANGES ALS is associated with an early atrophy of the hypothalamus that is observable even in presymptomatic gene carriers (Gorges et al., 2017). Whether the PVN or SON is atrophied is unknown, as no studies have yet examined cell counts or neuronal size alterations in either nucleus in ALS.

HORMONAL CHANGES Currently the only evidence for VP levels in ALS come from a SOD1-G86R mouse model, which indicates a decline in extrahypothalamic VP relative to healthy controls (González de Aguilar et al., 1999). There appears to be no change in the level of VP in the hypothalamus; this dichotomy is suggestive of a degradation of transportation mechanisms rather than of the nuclei themselves (González de Aguilar et al., 1999, 2003). No published study to date has investigated the role of oxytocin in ALS. There is some evidence of alteration of CRH in ALS. In particular, it appears that CSF CRH is significantly lower in patients with ALS than in healthy controls (Klimek et al., 1986). There is also notable evidence of a CRH dysrhythmia, discussed in relation to HPA perturbation later. In a study investigating TRH in the brain, TRH was found to be decreased in ALS relative to healthy controls (Mitsuma et al., 1986). However, another study found no difference in levels of TRH in the spinal cord or CSF (Jackson et al., 1986). These results present a mixed picture of TRH function in ALS but suggest that TRH neurons may be affected incidentally as the disease progresses (Jackson et al., 1986).

111

CIRCUIT ALTERATIONS AND DISRUPTION There is some evidence to suggest that there is a perturbance of the HPA axis in ALS, though research is sparse. Alterations in circulating cortisol levels have been found, though whether this is marked by hyperactivity in the morning or evening, or hypoactivity in the morning, is unclear (Patacchioli et al., 2003; Roozendaal et al., 2012; Spataro et al., 2015). These findings do suggest a circadian rhythm disruption at some point on the HPA axis, though the exact location of disturbance is as yet unknown (Spataro et al., 2015). There is very little research on the HPT axis in ALS, though TRH was at one point considered as a treatment target for ALS (Engel et al., 1983; Brooke et al., 1986). Studies appear to suggest hypoactivity of the HPT axis in ALS, which is in part supported by the subjective improvements felt by patients receiving TRH (Brooke et al., 1986; Mitsuma et al., 1986).

Progressive supranuclear palsy ABNORMAL INCLUSIONS AND AGGREGATES There is a great deal of evidence that the hypothalamus is susceptible to tau-positive NFTs in progressive supranuclear palsy (PSP), often more so than individuals with AD (Bugiani et al., 1979; Kida et al., 1988, 1992; Daniel et al., 1995; Li et al., 1996, 1998; Dickson et al., 2007). There is also some evidence to suggest that Pick bodies occasionally aggregate in the hypothalamus in PSP (Mori et al., 1986; Arima et al., 1992). Some small autopsy studies have also suggested that Lewy bodies aggregate in the hypothalamus in PSP, though these appear to be a rare finding and may be associated with comorbid dementia with Lewy bodies (DLB) (Mori et al., 1986, 2002; De Bruin and Lees, 1994). However, no published study to date has examined which nuclei of the hypothalamus are affected in PSP, or the extent to which they might be affected.

NEURONAL LOSS AND PATHOLOGICAL CHANGES Hypothalamic gliosis and neuronal loss secondary to the accumulation of NFTs has been observed microscopically in PSP (Homma et al., 1987). It is clear from MRI studies that the hypothalamus is atrophied in PSP, but the degree to which it involves the PVN and SON are not clear (Golbe, 1997; Lower et al., 1997; Price et al., 2004).

HORMONAL CHANGES No studies have yet investigated the alterations undergone by VP or OT in PSP.

112

C.A. STEWART AND E.C. FINGER

In a single study investigating cortical CRH concentrations, individuals with PSP were found to have a significant decline in CRH relative to controls. This may be due to dysfunction in the PVN, but this is not yet known (Whitehouse et al., 1987). No published research studies to date have investigated TRH levels in PSP, although several case studies have suggested that thyroid function is largely unaffected in PSP (Chiu et al., 2016; Li et al., 2016).

CIRCUIT ALTERATIONS AND DISRUPTION Very little research has explored the integrity of the HPA or HPT axis in PSP. Of the research that exists, there is the suggestion of HPA axis hyperactivity that may be due to dysfunction of CRH neurons in the PVN; more research is needed to clarify the involvement of the HPA in PSP (Whitehouse et al., 1987; Florio et al., 2007).

Parkinson’s disease ABNORMAL INCLUSIONS AND AGGREGATES In rats and mice, the progression of alpha-synuclein (a-syn) pathology inevitably includes the hypothalamus; however, these studies have not investigated aggregations in specific nuclei (Ulusoy et al., 2013). In humans with Parkinson’s disease (PD), the picture is mixed. Some studies indicate that there are no or very few a-syn-positive Lewy bodies in the PVN (Purba et al., 1994; Braak et al., 2011). However, others have identified Lewy body pathology in both nuclei which increases in severity with disease progression in PD (Langston and Forno, 1978; de Pablo Fernandez, 2019). It is possible that these different results arise from investigations of patients at different disease stages, as Lewy body pathology in the hypothalamus tends to begin later and be milder than elsewhere in the brain, and thus may not be present or noticeable at earlier stages (Masuda-Suzukake et al., 2013; de Pablo Fernandez, 2019). It is also likely that different histological methods play a role in these contradictory findings (de Pablo Fernandez, 2019).

NEURONAL LOSS AND PATHOLOGICAL CHANGES Hypothalamic volume is reduced in PD (Breen et al., 2016; Chowen and Garcia-Segura, 2019). This is due in part to neuronal population declines in both the PVN and the SON (Langston and Forno, 1978; Mann and Yates, 1983; Ansorge et al., 1997). Cells immunoreactive for OT in particular seem to be susceptible to decline in PD, while CRH-expressing neurons appear to be spared (Purba et al., 1994; Hoogendijk et al., 1998). Concomitant with the loss of neurons in the SON is a neuronal change that has been noted in one study of PD patients. Following death of SON neurons, the surviving

neurons have been observed to increase in somatic and nuclear size, suggesting compensatory hyperactivation in the SON (Ansorge et al., 1997). Whether this compensation also exists in the PVN is not yet known.

HORMONAL CHANGES A study investigating extrahypothalamic VP in the brain found no difference between PD and healthy controls (Rossor et al., 1982). Studies investigating CSF VP, on the other hand, indicate a decrease in PD relative to controls (Sundquist et al., 1983; Olsson et al., 1987). A subsequent study comparing patients treated with a dopamine agonist with treatment naïve patients suggested that CSF VP levels in treated patients were significantly higher than untreated patients. The study further found that treated men had higher VP concentrations than treated women, mirroring the sex difference seen in VP levels in healthy aging (Arai, 2011). This perhaps suggests that VP production is endogenously low in the hypothalamus in PD, and dopamine treatment can return VP levels in PD to normal; this is supported by the finding that dopamine is important for VP production in the SON (Yang et al., 1991). Though there are known alterations of the OT system in PD, it is as yet unclear whether circulating levels of OT change in the brain, blood, or CSF. The evidence for CRH alterations in PD is mixed. Measurements of extrahypothalamic CRH in the brain suggest either no change in CRH levels in PD or a decrease (Whitehouse et al., 1987; Leake et al., 1991). In contrast, indirect measurements of CRH activity via cortisol and adrenocorticotropin (ACTH) in plasma indicate an increase in CRH activity (Stypuła et al., 1996). Given the observed integrity of CRH neurons in the PVN, it seems possible that the hyperactivity observed peripherally may be due to an external stimulus (Hoogendijk et al., 1998). Research on TRH alterations in PD is equivocal, although no published research to date has specifically investigated TRH levels in the hypothalamus. There appears to be perturbance of TRH function in PD, with higher levels of free thyroxine and TSH, and absent or attenuated TSH responses to TRH stimulation (Otake et al., 1994; Aziz et al., 2011; Daimon et al., 2013). The exact extent or origin of these changes is not yet known.

CIRCUIT ALTERATIONS AND DISRUPTION There is relatively little research on integrity of the HPA axis in PD, and what exists presents a mixed picture, with some suggesting hypoactivation while others indicate hyperactivation (Du and Pang, 2015). The circadian dysrhythmia of cortisol in PD may be due in part to HPA

THE SUPRAOPTIC AND PARAVENTRICULAR NUCLEI IN NEURODEGENERATION dysregulation, but whether this is driven by the PVN is unknown (De Pablo-Fernández et al., 2017; Newhouse and Chemali, 2019). There is some evidence that the HPT axis is affected in PD, though the picture is again mixed, with no clearly established change in TRH levels in the blood or CSF at any stage of the disease (Daimon et al., 2013).

Dementia with Lewy bodies ABNORMAL INCLUSIONS AND AGGREGATES Patients diagnosed with DLB typically have a-syn- positive Lewy bodies and neurites distributed throughout the brain (Weil et al., 2017). These aggregations have also been found in both the PVN and the SON, but the SON seems to be particularly susceptible to a-syn pathology (Fig. 7.4) (Diodati et al., 2012; de Pablo Fernandez, 2019). Both the PVN and the SON have also stained positive for tau pathology; the PVN was more affected than the SON, with DLB patients showing similar amounts of tau inclusions in the PVN to patients with AD (Diodati et al., 2012).

NEURONAL LOSS AND PATHOLOGICAL CHANGES Though the SON and to a lesser extent the PVN experience a moderate burden of pathological aggregation over the course of DLB, they do not seem to exhibit significant concomitant atrophy (Diodati et al., 2012).

113

HORMONAL CHANGES Very little research has investigated VP in the blood or CSF in DLB. A study comparing DLB to AD, PD, and healthy controls found no difference in cortical concentrations of VP (Leake et al., 1991). One study has found that individuals with DLB have an intact VP response to hypotension, suggesting intact connections to and from the hypothalamus (Palma et al., 2015). In a single study investigating CRH in DLB, concentrations of CRH in the cortex were found to be decreased relative to controls (Leake et al., 1991). Whether this is driven by the hypothalamus is unknown. No published studies to date have investigated OT or TRH alterations in DLB.

CIRCUIT ALTERATIONS AND DISRUPTION No published research to date has specifically examined either the HPA or HPT in DLB. Given its closeness to PD, it is possible that the pathological alterations in these circuits seen in PD are similar in DLB (Newhouse and Chemali, 2019).

Multiple system atrophy ABNORMAL INCLUSIONS AND AGGREGATES a-Syn glial cytoplasmic inclusions are typically found throughout the brain in multiple system atrophy (MSA) (Trojanowski et al., 2007). Neuronal inclusions that are positive for a-syn have been observed in both the PVN and the SON in MSA, though they are involved less frequently and less severely than other brain regions (Cykowski et al., 2015, 2016). There is some evidence that tau-positive Pick body-like inclusions may accumulate in the hypothalamus in MSA, but this may be due to an atypical or FTLD subtype (Aoki et al., 2015).

NEURONAL LOSS AND PATHOLOGICAL CHANGES There is some evidence that the VP-immunoreactive neurons in the posterior part of the PVN are susceptible to atrophy in MSA, while VP- and non-VP-immunoreactive neurons in the anterior PVN and the SON remain largely intact (Fig. 7.5) (Benarroch et al., 2006).

HORMONAL CHANGES Fig. 7.4. Histological section of the SON of a patient with Lewy body disease demonstrating alpha-synuclein immunostaining (x400 magnification). SON, supraoptic nucleus. Reprinted with permission from Diodati, D., Cyn-Ang, L., Kertesz, A., et al., 2012. Pathologic evaluation of the supraoptic and paraventricular nuclei in dementia. Can J Neurol Sci 39, 213–219. doi: 10.1017/s0317167100013251.

Basal levels of VP in the brain and blood appear to be unchanged in MSA relative to healthy controls (Rossor et al., 1982; Kimber et al., 1999). There is a clear perturbation of VP’s circadian rhythm, as the nighttime peak observed in healthy adults is attenuated in MSA, which is possibly due to disturbed SCN function (Ozawa et al., 1999). VP also has unusual responses to challenge.

114

C.A. STEWART AND E.C. FINGER

Fig. 7.5. Upper panel: Sections of the posterior hypothalamus of a 64-year-old woman with no neurological disease (A and C) and a 54-year-old woman with MSA (B and D). The PVN, stained for VP alone (A and B) or costained with thionin (C and D), shows a marked loss of VP-immunoreactive neurons. Bar represents 20 mm. Lower panel: Count of VP-immunoreactive cells per PVNp section in controls and individuals with MSA, demonstrating a consistent loss of VP-staining neurons in the PVNp in MSA. ∗∗ P < 0.01. MSA, multiple system atrophy; PVNp, posterior paraventricular nucleus; VP, vasopressin. Reprinted with permission from Benarroch, E.E., Schmeichel, A.M., Sandroni, P., et al., 2006. Differential involvement of hypothalamic vasopressin neurons in multiple system atrophy. Brain 129, 2688–2696. doi: 10.1093/brain/awl109.

Individuals with MSA show a blunted VP response to head up tilt, a test of cardiovascular function (Kaufmann et al., 1992; Deguchi et al., 2004). They have also been shown to have an abnormal reaction to clonidine, a drug which suppresses VP in healthy individuals; individuals with MSA showed a blunted response to clonidine, experiencing no decrease in plasma VP (Kimber et al., 1999). Taken together, these results suggest that there is a failure of inputs to the SON that dysregulate its VP output (Kimber et al., 1999; Deguchi et al., 2004).

In individuals with MSA, there is a significant reduction in the morning peak of cortisol observed in healthy individuals, which may be related to a disturbed SCN’s circadian function (Ozawa et al., 2001). Individuals with MSA have been shown to have a blunted ACTH response to hypoglycemia (Polinsky et al., 1987). This ACTH response has been previously shown to be stimulated by CRH neurons, and to a lesser extent by VP neurons, in the PVN (Caraty et al., 1990). This suggests that in MSA there is a failure of the PVN to secrete CRH or VP.

THE SUPRAOPTIC AND PARAVENTRICULAR NUCLEI IN NEURODEGENERATION No published research to date has investigated OT or TRH in MSA.

CIRCUIT ALTERATIONS AND DISRUPTION Little research has explored the HPA axis in MSA. What exists suggests hypoactivity of the HPA axis, observed through significantly attenuated corticotrophin response to insulin challenge relative to healthy controls (Polinsky et al., 1987). This may be driven specifically by PVN dysfunction, but this is as yet unconfirmed (Caraty et al., 1990; Leone et al., 1991). No research has directly investigated the HPT axis in MSA. Research on TRH infusion in MSA has suggested that there may be a lack of TRH in MSA, but the origin or extent of this dysfunction is unknown (Kimura et al., 2011).

Huntington’s disease ABNORMAL INCLUSIONS AND AGGREGATES In Huntington’s disease (HD) aggregates of Huntingtin protein (Htt) form nuclear inclusions that are observed throughout the brain (Rubinsztein and Carmichael, 2003). Intranuclear and cytoplasmic Htt inclusions have been observed in both the PVN and SON, though they, like the rest of the hypothalamus, are generally more sparsely affected than other regions of the brain (Fig. 7.6) (Aziz et al., 2008; Gabery et al., 2010).

NEURONAL LOSS AND PATHOLOGICAL CHANGES The hypothalamus is susceptible to atrophy and selective neuronal loss in HD, though whether this involves the PVN or SON is unclear (Peters
en and Gabery, 2012). Some evidence suggests a significant reduction in the number of neurons in the PVN based on cell numbers and immunohistochemistry. However, these studies do

115

not suggest any size changes nor gliosis. Using immunostaining, OT neurons were found to be particularly affected by this loss, while the remaining neurons were atrophied; VP neurons also declined in number, but did not show the atrophy of OT neurons (Fig. 7.7) (Gabery et al., 2010). Another study found that there was no change in the actual number of immunoreactive PVN neurons in HD. However, patients with HD did not show the positive correlation between OT and VPimmunoreactive neurons in the PVN that healthy adults did (van Wamelen et al., 2012).

HORMONAL CHANGES Some studies on changes in VP in HD indicate a decline in VP in the hypothalamus relative to controls in both human and animal models; the evidence for an alteration of the VP-immunoreactive neurons in the PVN is equivocal (Peters
en and Gabery, 2012; van Wamelen et al., 2012; Gabery et al., 2015). Levels of OT similarly are observed either to remain the same or to decline (Gabery et al., 2015). Alterations in both VP and OT appear to be directly related to PVN and SON activity, though evidence for this alteration is unclear (van Wamelen et al., 2012, 2014). While levels of CRH in the PVN itself appear unchanged relative to controls, CRH mRNA expression in the PVN increases significantly in HD (van Wamelen et al., 2012). There is also an increase of CRH in CSF that is believed to be mediated by the PVN (Aziz et al., 2007). TRH levels in the brain have been observed to be increased in HD, although TRH mRNA appears unchanged (Nemeroff et al., 1983; Aziz et al., 2010; van Wamelen et al., 2012). The source of these changes is unknown, particularly as evidence for TSH alterations, or altered responses to TRH stimuli are unclear (Lavin et al., 1981; Saleh et al., 2009).

Fig. 7.6. EM48-positive Huntingtin inclusions in the PVN (A) and SON (B) and IC2-positive aggregates in the PVN (C) of a patient with HD, counterstained with CV. Arrows indicate Huntingtin inclusions. Scale bar represents 20 mm (A and B) and 10 mm (C). CV, cresyl violet; HD, Huntington’s disease; PVN, paraventricular nucleus; SON, supraoptic nucleus. Reprinted with permission from Gabery, S., Murphy, K., Schultz, K., et al., 2010. Changes in key hypothalamic neuropeptide populations in Huntington disease revealed by neuropathological analyses. Acta Neuropathol 120, 777–788. doi: 10.1007/s00401-010-0742-6.

116

C.A. STEWART AND E.C. FINGER

Fig. 7.7. Analysis of OT- (A and B) and VP- (D and E) containing populations of the hypothalamus in patients with HD (B and E) and controls (A and D), assessed using stereology in immunohistochemically processed sections of the whole hypothalamus. Number of neurons expressing OT (C) and VP (F) are expressed as mean  SEM. ∗P < 0.05. Scale bar represents 500 mm. HD, Huntington’s disease; OT, oxytoxin; VP, vasopressin. Reprinted with permission from Gabery, S., Murphy, K., Schultz, K., et al., 2010. Changes in key hypothalamic neuropeptide populations in Huntington disease revealed by neuropathological analyses. Acta Neuropathol 120, 777–788. doi: 10.1007/s00401-010-0742-6.

CIRCUIT ALTERATIONS AND DISRUPTION The HPA axis has been consistently shown to be hyperactive in HD, and this is believed to manifest early in the disease course (Heuser et al., 1991; Aziz et al., 2009). This change has been hypothesized to be directly attributable to hypothalamic dysfunction, particularly to PVN insensitivity to glucocorticoid feedback due to impaired low-affinity glucocorticoid receptors (Aziz et al., 2007, 2009). The HPT axis has similarly been found to be hyperactive in HD, though this hyperactivity has been found to be mild, and in some studies it has not been found at all (Saleh et al., 2009; Aziz et al., 2010; van Wamelen et al., 2012). The mild hyperactivity of the HPT axis has been proposed to be due to loss of dopamine receptors in the PVN, which thus fail to inhibit TRH secretion (Aziz et al., 2010).

CONCLUSION The SON and PVN of the hypothalamus undergo changes in normal aging and are involved in a number of pathological processes related to numerous neurodegenerative diseases. In particular, with healthy aging, sexually dimorphic changes in these nuclei and probable

interactions with sex hormones result in alterations in circulating levels of VP and CRH that may underlie agerelated changes in physiology. Evidence to date indicates that, in comparison to the PVN, the SON is relatively resistant to neurodegenerative pathology in AD and FTD, yet more susceptible to alpha-synuclein pathology hallmark of Lewy Body and PD. Much yet remains to be known about these nuclei and their roles in rarer disorders. To date, very little research has investigated the involvement of the PVN or SON in chronic traumatic encephalopathy, corticobasal degeneration, limbicpredominant age-related TDP-43 encephalopathy, primary age-related tauopathy, or the spinocerebellar ataxias. Further exploration of the presence and patterns of neurodegenerative changes in the SON and PVN may elucidate the basis for the selective vulnerability of certain cell populations and address a significant gap in knowledge in neurodegenerative diseases. Further study of the consequences of pathology in the SON and PVN for CNS and systemic levels and patterns of the related hormones in these disorders may also delineate the basis for many of the metabolic and physiologic changes that are currently poorly understood during the course of neurodegenerative disorders.

THE SUPRAOPTIC AND PARAVENTRICULAR NUCLEI IN NEURODEGENERATION

REFERENCES Aguilera G (2011). HPA axis responsiveness to stress: implications for healthy aging. Exp Gerontol 46: 90–95. https:// doi.org/10.1016/j.exger.2010.08.023. Ahmed RM, Latheef S, Bartley L et al. (2015). Eating behavior in frontotemporal dementia: peripheral hormones vs hypothalamic pathology. Neurology 85: 1310–1317. https://doi. org/10.1212/WNL.0000000000002018. Ansorge O, Daniel SE, Pearce RK (1997). Neuronal loss and plasticity in the supraoptic nucleus in Parkinson’s disease. Neurology 49: 610–613. https://doi.org/10.1212/wnl.49. 2.610. Antonijevic IA, Murck H, Frieboes R et al. (1999). On the gender differences in sleep-endocrine regulation in young normal humans. Neuroendocrinology 70: 280–287. https://doi.org/10.1159/000054487. Aoki N, Boyer PJ, Lund C et al. (2015). Atypical multiple system atrophy is a new subtype of frontotemporal lobar degeneration: frontotemporal lobar degeneration associated with a-synuclein. Acta Neuropathol 130: 93–105. https://doi.org/10.1007/s00401-015-1442-z. Arai M (2011). Increased plasma arginine vasopressin levels in dopamine agonist-treated Parkinson’s disease patients. Neuro Endocrinol Lett 32: 39–43. Arima K, Murayama S, Oyanagi S et al. (1992). Presenile dementia with progressive supranuclear palsy tangles and pick bodies: an unusual degenerative disorder involving the cerebral cortex, cerebral nuclei, and brain stem nuclei. Acta Neuropathol 84: 128–134. https://doi.org/10.1007/ bf00311384. Asplund R (2002). Diuresis pattern, plasma vasopressin and blood pressure in healthy elderly persons with nocturia and nocturnal polyuria. Neth J Med 60: 276–280. ˚ berg H (1991). Diurnal variation in the levels of Asplund R, A antidiuretic hormone in the elderly. J Intern Med 229: 131–134. https://doi.org/10.1111/j.1365-2796.1991.tb0 0320.x. Atzmon G, Barzilai N, Hollowell JG et al. (2009). Extreme longevity is associated with increased serum thyrotropin. J Clin Endocrinol Metab 94: 1251–1254. https://doi.org/ 10.1210/jc.2008-2325. Aziz NA, Swaab DF, Pijl H et al. (2007). Hypothalamic dysfunction and neuroendocrine and metabolic alterations in Huntington’s disease: clinical consequences and therapeutic implications. Rev Neurosci 18: 223–251. https://doi.org/ 10.1515/revneuro.2007.18.3-4.223. Aziz A, Fronczek R, Maat-Schieman M et al. (2008). Hypocretin and melanin-concentrating hormone in patients with Huntington disease. Brain Pathol 18: 474–483. https:// doi.org/10.1111/j.1750-3639.2008.00135.x. Aziz NA, Pijl H, Fr€olich M et al. (2009). Increased hypothalamicpituitary-adrenal axis activity in Huntington’s disease. J Clin Endocrinol Metab 94: 1223–1228. https://doi.org/10.1210/ jc.2008-2543. Aziz NA, Pijl H, Fr€olich M et al. (2010). Altered thyrotropic and lactotropic axes regulation in Huntington’s disease. Clin Endocrinol (Oxf ) 73: 540–545. https://doi.org/10.1111/ j.1365-2265.2010.03836.x.

117

olich M et al. (2011). Diurnal secretion Aziz NA, Pijl H, Fr€ profiles of growth hormone, thyrotrophin and prolactin in Parkinson’s disease. J Neuroendocrinol 23: 519–524. https://doi.org/10.1111/j.1365-2826.2011.02134.x. Baloyannis SJ, Mavroudis I, Mitilineos D et al. (2015). The hypothalamus in Alzheimer’s disease: a Golgi and electron microscope study. Am J Alzheimers Dis Other Demen 30: 478–487. https://doi.org/10.1177/1533317514556876. Baloyannis SJ, Mavroudis I, Mitilineos D et al. (2018). The hypothalamus in Alzheimer’s disease. In: SJ Baloyannis, J Oxholm Gordeladze (Eds.), Hypothalamus in health and diseases. IntechOpen. https://doi.org/10.5772/intechopen. 81475. Banki CM, Karmacsi L, Bissette G et al. (1992). Cerebrospinal fluid neuropeptides in dementia. Biol Psychiatry 32: 452–456. https://doi.org/10.1016/0006-3223(92)90132-j. Bao AM, Swaab DF (2007). Gender difference in age-related number of corticotropin-releasing hormone-expressing neurons in the human hypothalamic paraventricular nucleus and the role of sex hormones. Neuroendocrinology 85: 27–36. https://doi.org/10.1159/000099832. Bao AM, Fischer DF, Wu YH et al. (2006). A direct androgenic involvement in the expression of human corticotropinreleasing hormone. Mol Psychiatry 11: 567–576. https:// doi.org/10.1038/sj.mp.4001800. Bao AM, Meynen G, Swaab DF (2008). The stress system in depression and neurodegeneration: focus on the human hypothalamus. Brain Res Rev 57: 531–553. https://doi. org/10.1016/j.brainresrev.2007.04.005. Benarroch EE, Schmeichel AM, Sandroni P et al. (2006). Differential involvement of hypothalamic vasopressin neurons in multiple system atrophy. Brain 129: 2688–2696. https://doi.org/10.1093/brain/awl109. Bissette G, Reynolds GP, Kilts CD et al. (1985). Corticotropinreleasing factor-like immunoreactivity in senile dementia of the Alzheimer type. JAMA 254: 3067. https://doi.org/ 10.1001/jama.1985.03360210083036. Bocchetta M, Gordon E, Manning E et al. (2015). Detailed volumetric analysis of the hypothalamus in behavioral variant frontotemporal dementia. J Neurol 262: 2635–2642. https://doi.org/10.1007/s00415-015-7885-2. Braak H, Braak E (1991). Neuropathological stageing of Alzheimer-related changes. Acta Neuropathol 82: 239–259. https://doi.org/10.1007/bf00308809. Braak H, Thal DR, Del Tredici K (2011). Nerve cells immunoreactive for p62 in select hypothalamic and brainstem nuclei of controls and Parkinson’s disease cases. J Neural Transm 118: 809–819. https://doi.org/10.1007/s00702010-0508-2. Brabant G, Prank K, Ranft U et al. (1990). Physiological regulation of circadian and pulsatile thyrotropin secretion in normal man and woman. J Clin Endocrinol Metab 70: 403–409. https://doi.org/10.1210/jcem-70-2-403. Breen DP, Nombela C, Vuono R et al. (2016). Hypothalamic volume loss is associated with reduced melatonin output in Parkinson’s disease. Mov Disord 31: 1062–1066. https:// doi.org/10.1002/mds.26592. Bremner AP, Feddema P, Leedman PJ et al. (2012). Agerelated changes in thyroid function: a longitudinal study

118

C.A. STEWART AND E.C. FINGER

of a community-based cohort. J Clin Endocrinol Metab 97: 1554–1562. https://doi.org/10.1210/jc.2011-3020. Brooke MH, Florence JM, Heller SL et al. (1986). Controlled trial of thyrotropin releasing hormone in amyotrophic lateral sclerosis. Neurology 36: 146–151. https://doi.org/ 10.1212/wnl.36.2.146. Bugiani O, Mancardi GL, Brusa A et al. (1979). The fine structure of subcortical neurofibrillary tangles in progressive supranuclear palsy. Acta Neuropathol 45: 147–152. https://doi.org/10.1007/bf00691893. Cai ZJ (2018). The peripheral hypotheses of hypothalamic aging. Open Access Libr J 05: 1–15. https://doi.org/ 10.4236/oalib.1104445. Caraty A, Grino M, Locatelli A et al. (1990). Insulin-induced hypoglycemia stimulates corticotropin-releasing factor and arginine vasopressin secretion into hypophysial portal blood of conscious, unrestrained rams. J Clin Invest 85: 1716–1721. https://doi.org/10.1172/JCI114626. Ceccatelli S, Calza´ L, Giardino L (1996). Age-related changes in the expression of corticotropin-releasing hormone receptor mRNA in the rat pituitary. Brain Res Mol Brain Res 37: 175–180. https://doi.org/10.1016/0169328x(95)00304-b. Chiamolera MI, Wondisford FE (2009). Minireview: thyrotropin-releasing hormone and the thyroid hormone feedback mechanism. Endocrinology 150: 1091–1096. https://doi.org/10.1210/en.2008-1795. Chiodera P, Volpi R, Capretti L et al. (1994). Oxytocin response to challenging stimuli in elderly men. Regul Pept 51: 169–176. https://doi.org/10.1016/0167-0115(94)90206-2. Chiu YW, Lee SH, Yeh TH (2016). Diversified psychiatric presentation in a case of progressive supranuclear palsy. J Clin Gerontol Geriatr 7: 164–167. https://doi.org/10.1016/j.jcgg. 2016.05.001. Chowen JA, Garcia-Segura LM (2019). Microglia, neurodegeneration and loss of neuroendocrine control. Prog Neurobiol 184: 101720. https://doi.org/10.1016/j.pneurobio. 2019.101720. Cykowski MD, Coon EA, Powell SZ et al. (2015). Expanding the spectrum of neuronal pathology in multiple system atrophy. Brain 138: 2293–2309. https://doi.org/10.1093/ brain/awv114. Cykowski MD, Takei H, Schulz PE et al. (2014). TDP-43 pathology in the basal forebrain and hypothalamus of patients with amyotrophic lateral sclerosis. Acta Neuropathol Commun 2: 171. https://doi.org/10.1186/s40478-014-0171-1. Cykowski MD, Takei H, Van Eldik LJ et al. (2016). Hippocampal sclerosis but not normal aging or Alzheimer disease is associated with TDP-43 pathology in the basal forebrain of aged persons. J Neuropathol Exp Neurol 75: 397–407. https://doi.org/10.1093/jnen/nlw014. Daimon CM, Chirdon P, Maudsley S et al. (2013). The role of thyrotropin releasing hormone in aging and neurodegenerative diseases. Am J Alzheimers Dis 1. https://doi.org/ 10.7726/ajad.2013.1003. Daniel SE, de Bruin VM, Lees AJ (1995). The clinical and pathological spectrum of Steele-Richardson-Olszewski syndrome (progressive supranuclear palsy): a reappraisal. Brain 118: 759–770. https://doi.org/10.1093/brain/118.3.759.

De Bruin VM, Lees AJ (1994). Subcortical neurofibrillary degeneration presenting as Steele-Richardson-Olszewski and other related syndromes: a review of 90 pathologically verified cases. Mov Disord 9: 381–389. https://doi.org/ 10.1002/mds.870090402. de Pablo Fernandez E (2019). Pathophysiological mechanisms of non-motor features and their role on the pathogenic process in Parkinson’s disease. Undergraduate thesis. De Pablo-Ferna´ndez E, Breen DP, Bouloux PM et al. (2017). Neuroendocrine abnormalities in Parkinson’s disease. J Neurol Neurosurg Psychiatry 88: 176–185. https://doi. org/10.1136/jnnp-2016-314601. Dedeene L, Van Schoor E, Vandenberghe R et al. (2019). Circadian sleep/wake-associated cells show dipeptide repeat protein aggregates in C9orf72-related ALS and FTLD cases. Acta Neuropathol Commun 7: 189. https:// doi.org/10.1186/s40478-019-0845-9. Deguchi K, Sasaki I, Touge T et al. (2004). Abnormal baroreceptor-mediated vasopressin release as possible marker in early diagnosis of multiple system atrophy. J Neurol Neurosurg Psychiatry 75: 110–115. Deuschle M, Gotthardt U, Schweiger U et al. (1997). With aging in humans the activity of the hypothalamus-pituitaryadrenal system increases and its diurnal amplitude flattens. Life Sci 61: 2239–2246. https://doi.org/10.1016/S00243205(97)00926-0. Dickson DW, Rademakers R, Hutton ML (2007). Progressive supranuclear palsy: pathology and genetics. Brain Pathol 17: 74–82. https://doi.org/10.1111/j.1750-3639.2007.00054.x. Diodati D, Cyn-Ang L, Kertesz A et al. (2012). Pathologic evaluation of the supraoptic and paraventricular nuclei in dementia. Can J Neurol Sci 39: 213–219. https://doi.org/ 10.1017/s0317167100013251. Dodt C, Theine KJ, Uthgenannt D et al. (1994). Basal secretory activity of the hypothalamo-pituitary-adrenocortical axis is enhanced in healthy elderly. An assessment during undisturbed night-time sleep. Eur J Endocrinol 131: 443–450. https://doi.org/10.1530/eje.0.1310443. Du X, Pang TY (2015). Is dysregulation of the HPA-axis a core pathophysiology mediating co-morbid depression in neurodegenerative diseases? Front Psych 6: 32. https://doi. org/10.3389/fpsyt.2015.00032. Duffy JF, Scheuermaier K, Loughlin KR (2016). Age-related sleep disruption and reduction in the circadian rhythm of urine output: contribution to nocturia? Curr Aging Sci 9: 34–43. https://doi.org/10.2174/1874609809666151130220343. Dumais KM, Veenema AH (2016). Vasopressin and oxytocin receptor systems in the brain: sex differences and sex-specific regulation of social behavior. Front Neuroendocrinol 40: 1–23. https://doi.org/10.1016/j.yfrne.2015.04.003. Ebner NC, Maura GM, Macdonald K et al. (2013). Oxytocin and socioemotional aging: current knowledge and future trends. Front Hum Neurosci 7: 487. https://doi.org/ 10.3389/fnhum.2013.00487. Edvinsson L, Minthon L, Ekman R et al. (1993). Neuropeptides in cerebrospinal fluid of patients with Alzheimer’s disease and dementia with frontotemporal lobe degeneration. Dementia 4: 167–171. https://doi.org/ 10.1159/000107318.

THE SUPRAOPTIC AND PARAVENTRICULAR NUCLEI IN NEURODEGENERATION Engel WK, Siddique T, Nicoloff JT (1983). Effect on weakness and spasticity in amyotrophic lateral sclerosis of thyrotropin-releasing hormone. Lancet 2: 73–75. https:// doi.org/10.1016/s0140-6736(83)90060-0. Erkut ZA, Klooker T, Endert E et al. (2004). Stress of dying is not suppressed by high-dose morphine or by dementia. Neuropsychopharmacology 29: 152–157. https://doi.org/ 10.1038/sj.npp.1300299. Ferrari E, Cravello L, Muzzoni B et al. (2001). Age-related changes of the hypothalamic-pituitary-adrenal axis: pathophysiological correlates. Eur J Endocrinol 144: 319–329. https://doi.org/10.1530/eje.0.1440319. Fliers E, Swaab DF, Pool CW et al. (1985). The vasopressin and oxytocin neurons in the human supraoptic and paraventricular nucleus; changes with aging and in senile dementia. Brain Res 342: 45–53. https://doi.org/10.1016/0006-8993 (85)91351-4. Florio P, Zatelli MC, Reis FM et al. (2007). Corticotropin releasing hormone: a diagnostic marker for behavioral and reproductive disorders? Front Biosci 12: 551–560. https://doi.org/10.2741/2081. Gabery S, Murphy K, Schultz K et al. (2010). Changes in key hypothalamic neuropeptide populations in Huntington disease revealed by neuropathological analyses. Acta Neuropathol 120: 777–788. https://doi.org/10.1007/s00401010-0742-6. Gabery S, Halliday G, Kirik D et al. (2015). Selective loss of oxytocin and vasopressin in the hypothalamus in early Huntington disease: a case study. Neuropathol Appl Neurobiol 41: 843–848. https://doi.org/10.1111/nan. 12236. Gil-Bea FJ, Aisa B, Solomon A et al. (2010). HPA axis dysregulation associated to apolipoprotein E4 genotype in Alzheimer’s disease. J Alzheimers Dis 22: 829–838. https://doi.org/10.3233/JAD-2010-100663. Golbe LI (1997). Progressive supranuclear palsy. In: RL Watts, WC Koller (Eds.), Movement disorders. McGraw-Hill, New York, pp. 279–295. Gonza´lez de Aguilar JL, Gordon JW, Ren
e F et al. (1999). A mouse model of familial amyotrophic lateral sclerosis expressing a mutant superoxide dismutase 1 shows evidence of disordered transport in the vasopressin hypothalamo-neurohypophysial axis. Eur J Neurosci 11: 4179–4187. https://doi.org/10.1046/j.1460-9568.1999. 00840.x. Gonza´lez De Aguilar JL, Ren
e F, Dupuis L et al. (2003). Neuroendocrinology of neurodegenerative diseases. Insights from transgenic mouse models. Neuroendocrinology 78: 244–252. https://doi.org/10.1159/000074445. Gorges M, Vercruysse P, M€uller HP et al. (2017). Hypothalamic atrophy is related to body mass index and age at onset in amyotrophic lateral sclerosis. J Neurol Neurosurg Psychiatry 88: 1033–1041. https://doi.org/ 10.1136/jnnp-2017-315795. Goudsmit E, Hofman MA, Fliers E et al. (1990). The supraoptic and paraventricular nuclei of the human hypothalamus in relation to sex, age and Alzheimer’s disease. Neurobiol Aging 11: 529–536. https://doi.org/10.1016/0197-4580 (90)90114-f.

119

Graugaard-Jensen C, Hvistendahl GM, Frøkiaer J et al. (2014). Urinary concentration does not exclusively rely on plasma vasopressin. A study between genders. Gender and diurnal urine regulation. Acta Physiol (Oxf ) 212: 97–105. https:// doi.org/10.1111/apha.12337. Graugaard-Jensen C, Hvistendahl GM, Frøkiær J et al. (2017). Oral contraceptives and renal water handling: a diurnal study in young women. Physiol Rep 5: e13547. https:// doi.org/10.14814/phy2.13547. Greenspan SL, Klibanski A, Schoenfeld D et al. (1986). Pulsatile secretion of thyrotropin in man. J Clin Endocrinol Metab 63: 661–668. https://doi.org/10.1210/ jcem-63-3-661. Hatzinger M, Z’Brun A, Hemmeter U et al. (1995). Hypothalamic-pituitary-adrenal system function in patients with Alzheimer’s disease. Neurobiol Aging 16: 205–209. https://doi.org/10.1016/0197-4580(94)00159-6. Heuser IJ, Chase TN, Mouradian MM (1991). The limbichypothalamic-pituitary-adrenal axis in Huntington’s disease. Biol Psychiatry 30: 943–952. https://doi.org/10.1016/00063223(91)90007-9. Homma Y, Takahashi H, Takeda S et al. (1987). An autopsy case of progressive supranuclear palsy showing “pure akinesia without rigidity and tremor and with no effect by L-dopa therapy (Imai)”. No To Shinkei 39: 183–187. Hoogendijk WJ, Purba JS, Hofman MA et al. (1998). Depression in Parkinson’s disease is not accompanied by more corticotropin-releasing hormone expressing neurons in the hypothalamic paraventricular nucleus. Biol Psychiatry 43: 913–917. Ishii M, Iadecola C (2015). Metabolic and non-cognitive manifestations of Alzheimer’s disease: the hypothalamus as both culprit and target of pathology. Cell Metab 22: 761–776. https://doi.org/10.1016/j.cmet.2015.08.016. Ishunina TA, Swaab DF (1999). Vasopressin and oxytocin neurons of the human supraoptic and paraventricular nucleus: size changes in relation to age and sex. J Clin Endocrinol Metab 84: 4637–4644. https://doi.org/10.1210/jcem.84. 12.6187. Ishunina TA, Salehi A, Hofman MA et al. (1999). Activity of vasopressinergic neurones of the human supraoptic nucleus is age- and sex-dependent. J Neuroendocrinol 11: 251–258. https://doi.org/10.1046/j.1365-2826.1999.00318.x. Ishunina TA, Kruijver FP, Balesar R et al. (2000a). Differential expression of estrogen receptor alpha and beta immunoreactivity in the human supraoptic nucleus in relation to sex and aging. J Clin Endocrinol Metab 85: 3283–3291. https:// doi.org/10.1210/jcem.85.9.6826. Ishunina TA, Salehi A, Swaab DF (2000b). Sex- and agerelated P75 neurotrophin receptor expression in the human supraoptic nucleus. Neuroendocrinology71: 243–251. https://doi.org/10.1159/000054542. Ishunina TA, Dolzhikov AA, Grinevich VV (2001). Activity of vasopressin neurons in the human supraoptic nucleus: estrogen inhibitory effect. Usp Fiziol Nauk 32: 48–59. Iwatsubo T, Saido TC, Mann DM et al. (1996). Full-length amyloid-beta (1-42(43)) and amino-terminally modified and truncated amyloid-beta 42(43) deposit in diffuse plaques. Am J Pathol 149: 1823–1830.

120

C.A. STEWART AND E.C. FINGER

Jackson IM, Adelman LS, Munsat TL et al. (1986). Amyotrophic lateral sclerosis: thyrotropin-releasing hormone and histidyl proline diketopiperazine in the spinal cord and cerebrospinal fluid. Neurology 36: 1218–1223. https://doi.org/10.1212/wnl.36.9.1218. Kalsbeek A, Fliers E (2013). Daily regulation of hormone profiles. Handb Exp Pharmacol 185–226, 217. https://doi. org/10.1007/978-3-642-25950-0_8. Kaufmann H, Oribe E, Miller M et al. (1992). Hypotensioninduced vasopressin release distinguishes between pure autonomic failure and multiple system atrophy with autonomic failure. Neurology 42: 590–593. https://doi.org/ 10.1212/wnl.42.3.590. Kida M, Koo H, Grossniklaus HE et al. (1988). Neuropathologic findings in progressive supranuclear palsy. A brief review with two additional case reports. J Clin Neuroophthalmol 8: 161–170. Kida E, Barcikowska M, Niemczewska M (1992). Immunohistochemical study of a case with progressive supranuclear palsy without ophthalmoplegia. Acta Neuropathol 83: 328–332. https://doi.org/10.1007/bf00296797. Kim K, Choe HK (2019). Role of hypothalamus in aging and its underlying cellular mechanisms. Mech Ageing Dev 177: 74–79. https://doi.org/10.1016/j.mad.2018.04.008. Kimber J, Watson L, Mathias CJ (1999). Abnormal suppression of arginine-vasopressin by clonidine in multiple system atrophy. Clin Auton Res 9: 271–274. https://doi.org/ 10.1007/bf02319457. Kimura N, Kumamoto T, Masuda T et al. (2011). Evaluation of the effects of thyrotropin releasing hormone (TRH) therapy on regional cerebral blood flow in the cerebellar variant of multiple system atrophy using 3DSRT. J Neuroimaging 21: 132–137. https://doi.org/10.1111/j.1552-6569.2009. 00411.x. Klimek A, Cie
slak D, Szulc-Kuberska J et al. (1986). Reduced lumbar cerebrospinal fluid corticotropin releasing factor (CRF) levels in amyotrophic lateral sclerosis. Acta Neurol Scand 74: 72–74. https://doi.org/10.1111/j.16000404.1986.tb04629.x. Langston JW, Forno LS (1978). The hypothalamus in Parkinson disease. Ann Neurol 3: 129–133. https://doi. org/10.1002/ana.410030207. Laughlin GA, Barrett-Connor E (2000). Sexual dimorphism in the influence of advanced aging on adrenal hormone levels: the Rancho Bernardo study. J Clin Endocrinol Metab 85: 3561–3568. https://doi.org/10.1210/jcem.85.10.6861. Lavin PJ, Bone I, Sheridan P (1981). Studies of hypothalamic function in Huntington’s chorea. J Neurol Neurosurg Psychiatry 44: 414–418. https://doi.org/10.1136/jnnp.44. 5.414. Leake A, Perry EK, Perry RH et al. (1991). Neocortical concentrations of neuropeptides in senile dementia of the Alzheimer and Lewy body type: comparison with Parkinson’s disease and severity correlations. Biol Psychiatry 29: 357–364. https://doi.org/10.1016/00063223(91)90221-7. Leone M, Zappacosta BM, Valentini S et al. (1991). The insulin tolerance test and the ovine corticotrophin-releasing

hormone test in episodic cluster headache. Cephalalgia 11: 269–274. https://doi.org/10.1046/j.1468-2982.1991.110 6269.x. Li F, Iseki E, Kosaka K et al. (1996). Progressive supranuclear palsy with fronto-temporal atrophy and various taupositive abnormal structures. Clin Neuropathol 15: 319–323. Li F, Iseki E, Odawara T et al. (1998). Regional quantitative analysis of tau-positive neurons in progressive supranuclear palsy: comparison with Alzheimer’s disease. J Neurol Sci 159: 73–81. https://doi.org/10.1016/s0022510x(98)00136-1. Li Y, Dai CB, Wang LJ et al. (2016). Idiopathic basal ganglia calcification presented with progressive supranuclear palsy-like features. Chin Med J (Engl) 129: 478–479. https://doi.org/10.4103/0366-6999.176067. Lower J, Lennox G, Leigh PN (1997). Disorders of movement and system degeneration. In: DI Graham, PL Lantos (Eds.), Greenfield’s neuropathology. vol. II. Arnold, London, pp. 281–366. Lucassen PJ, Salehi A, Pool CW et al. (1994). Activation of vasopressin neurons in aging and Alzheimer’s disease. J Neuroendocrinol 6: 673–679. https://doi.org/10.1111/ j.1365-2826.1994.tb00634.x. Lucassen PJ, Van Heerikhuize JJ, Guldenaar SEF et al. (1997). No change in the total amounts of vasopressin mRNA in the supraoptic and paraventricular nucleus in aging and Alzheimer’s disease. J Neuroendocrinol 9: 297–305. Lund TD, Munson DJ, Haldy ME et al. (2004). Androgen inhibits, while oestrogen enhances, restraint-induced activation of neuropeptide neurones in the paraventricular nucleus of the hypothalamus. J Neuroendocrinol 16: 272–278. https://doi.org/10.1111/j.0953-8194.2004.01167.x. Mann DM, Yates PO (1983). Pathological basis for neurotransmitter changes in Parkinson’s disease. Neuropathol Appl Neurobiol 9: 3–19. https://doi.org/10.1111/j.13652990.1983.tb00320.x. Mann DM, Yates PO, Marcyniuk B (1985). Changes in Alzheimer’s disease in the magnocellular neurones of the supraoptic and paraventricular nuclei of the hypothalamus and their relationship to the noradrenergic deficit. Clin Neuropathol 4: 127–134. Marazziti D, Baroni S, Mucci F et al. (2019). Sex-related differences in plasma oxytocin levels in humans. Clin Pract Epidemiol Ment Health 15: 58–63. https://doi.org/10.2174/ 1745017901915010058. Masuda-Suzukake M, Nonaka T, Hosokawa M et al. (2013). Prion-like spreading of pathological a-synuclein in brain. Brain 136: 1128–1138. https://doi.org/10.1093/brain/ awt037. May C, Rapoport SI, Tomai TP et al. (1987). Cerebrospinal fluid concentrations of corticotropin-releasing hormone (CRH) and corticotropin (ACTH) are reduced in patients with Alzheimer’s disease. Neurology 37: 535–538. https://doi.org/10.1212/wnl.37.3.535. Mazurek MF, Beal MF, Bird ED et al. (1987). Oxytocin in Alzheimer’s disease: postmortem brain levels. Neurology 37: 1001–1003. https://doi.org/10.1212/wnl.37.6.1001.

THE SUPRAOPTIC AND PARAVENTRICULAR NUCLEI IN NEURODEGENERATION Mazzoccoli G, Pazienza V, Piepoli A et al. (2010). Hypothalamus-hypophysis-thyroid axis function in healthy aging. J Biol Regul Homeost Agents 24: 433–439. Miller M, Bales KL, Taylor SL et al. (2013). Oxytocin and vasopressin in children and adolescents with autism spectrum disorders: sex differences and associations with symptoms. Autism Res 6: 91–102. https://doi.org/10.1002/aur.1270. Mitsuma T, Adachi K, Mukoyama M et al. (1986). Concentrations of thyrotropin-releasing hormone in the brain of patients with amyotrophic lateral sclerosis. J Neurol Sci 76: 277–281. https://doi.org/10.1016/0022510x(86)90175-9. Monzani F, Del Guerra P, Caraccio N et al. (1996). Age-related modifications in the regulation of the hypothalamicpituitary-thyroid axis. Horm Res 46: 107–112. https://doi. org/10.1159/000185005. Mori H, Yoshimura M, Tomonaga M et al. (1986). Progressive supranuclear palsy with Lewy bodies. Acta Neuropathol 71: 344–346. https://doi.org/10.1007/bf00688061. Mori H, Oda M, Komori T et al. (2002). Lewy bodies in progressive supranuclear palsy. Acta Neuropathol 104: 273–278. https://doi.org/10.1007/s00401-002-0555-3. Nadal M, Wikstr€om L, Ruthstr€om L (1994). Secretory pattern of vasopressin in plasma and cerebrospinal fluid of patients with dementia and of two control groups. Eur J Endocrinol 130: 346–349. https://doi.org/10.1530/eje.0. 1300346. Nemeroff CB, Youngblood WW, Manberg PJ et al. (1983). Regional brain concentrations of neuropeptides in Huntington’s chorea and schizophrenia. Science 221: 972–975. https://doi.org/10.1126/science.6136092. Nemeroff CB, Kizer JS, Reynolds GP et al. (1989). Neuropeptides in Alzheimer’s disease: a postmortem study. Regul Pept 25: 123–130. https://doi.org/10.1016/0167-0115(89)90254-1. Newhouse A, Chemali Z (2019). Neuroendocrine disturbances in neurodegenerative disorders: a scoping review. Psychosomatics 61: 105–115. North WG, Harbaugh R, Reeder T (1992). An evaluation of human neurophysin production in Alzheimer’s disease: preliminary observations. Neurobiol Aging 13: 261–265. https://doi.org/10.1016/0197-4580(92)90038-y. Olsson JE, Forsling ML, Lindvall B et al. (1987). Cerebrospinal fluid arginine vasopressin in Parkinson’s disease, dementia, and other degenerative disorders. Adv Neurol 45: 239–242. Olsson T, Viitanen M, H€agg E et al. (1989). Hormones in “young” and “old” elderly: pituitary-thyroid and pituitaryadrenal axes. Gerontology 35: 144–152. https://doi.org/ 10.1159/000213013. Otake K, Oiso Y, Mitsuma T et al. (1994). Hypothalamic dysfunction in Parkinson’s disease patients. Acta Med Hung 50: 3–13. Ozawa T, Tanaka H, Nakano R et al. (1999). Nocturnal decrease in vasopressin secretion into plasma in patients with multiple system atrophy. J Neurol Neurosurg Psychiatry 67: 542–545. https://doi.org/10.1136/jnnp.67.4.542. Ozawa T, Soma Y, Yoshimura N et al. (2001). Reduced morning cortisol secretion in patients with multiple system

121

atrophy. Clin Auton Res 11: 271–272. https://doi.org/ 10.1007/BF02298961. Palma JA, Martinez J, Percival L et al. (2015). Hypotensioninduced vasopressin release distinguishes Lewy body disorders from multiple system atrophy. Auton Neurosci 192: 126. https://doi.org/10.1016/j.autneu.2015.07.230. Patacchioli FR, Monnazzi P, Scontrini A et al. (2003). Adrenal dysregulation in amyotrophic lateral sclerosis. J Endocrinol Invest 26: RC23–5. https://doi.org/10.1007/BF03349149. Peters
en A, Gabery S (2012). Hypothalamic and limbic system changes in Huntington’s disease. J Huntingt Dis 1: 5–16. https://doi.org/10.3233/JHD-2012-120006. Piguet O, Peters
en A, Yin Ka Lam B et al. (2011). Eating and hypothalamus changes in behavioral-variant frontotemporal dementia. Ann Neurol 69: 312–319. https://doi.org/ 10.1002/ana.22244. Plasencia G, Luedicke JM, Nazarloo HP et al. (2019). Plasma oxytocin and vasopressin levels in young and older men and women: functional relationships with attachment and cognition. Psychoneuroendocrinology 110: 104419. https:// doi.org/10.1016/j.psyneuen.2019.104419. Polinsky RJ, Brown RT, Lee GK et al. (1987). Beta-endorphin, ACTH, and catecholamine responses in chronic autonomic failure. Ann Neurol 21: 573–577. https://doi.org/10.1002/ ana.410210608. Popp J, Wolfsgruber S, Heuser I et al. (2015). Cerebrospinal fluid cortisol and clinical disease progression in MCI and dementia of Alzheimer’s type. Neurobiol Aging 36: 601–607. https://doi.org/10.1016/j.neurobiolaging.2014. 10.031. Price S, Paviour D, Scahill R et al. (2004). Voxel-based morphometry detects patterns of atrophy that help differentiate progressive supranuclear palsy and Parkinson’s disease. Neuroimage 23: 663–669. https://doi.org/10.1016/ j.neuroimage.2004.06.013. Purba JS, Hofman MA, Swaab DF (1994). Decreased number of oxytocin-immunoreactive neurons in the paraventricular nucleus of the hypothalamus in Parkinson’s disease. Neurology 44: 84–89. https://doi.org/10.1212/wnl.44.1.84. Raadsheer FC, Sluiter AA, Ravid R et al. (1993). Localization of corticotropin-releasing hormone (CRH) neurons in the paraventricular nucleus of the human hypothalamus; age-dependent colocalization with vasopressin. Brain Res 615: 50–62. https://doi.org/10.1016/0006-8993(93) 91113-7. Raadsheer FC, van Heerikhuize JJ, Lucassen PJ et al. (1995). Corticotropin-releasing hormone mRNA levels in the paraventricular nucleus of patients with Alzheimer’s disease and depression. Am J Psychiatry 152: 1372–1376. https:// doi.org/10.1176/ajp.152.9.1372. Raskind MA, Peskind ER, Lampe TH et al. (1986). Cerebrospinal fluid vasopressin, oxytocin, somatostatin, and beta-endorphin in Alzheimer’s disease. Arch Gen Psychiatry 43: 382–388. https://doi.org/10.1001/archpsyc. 1986.01800040092013. Roozendaal B, Kim S, Wolf OT et al. (2012). The cortisol awakening response in amyotrophic lateral sclerosis is blunted and correlates with clinical status and depressive

122

C.A. STEWART AND E.C. FINGER

mood. Psychoneuroendocrinology 37: 20–26. https://doi. org/10.1016/j.psyneuen.2011.04.013. Rossor MN, Hunt SP, Iversen LL et al. (1982). Extrahypothalamic vasopressin is unchanged in Parkinson’s disease and Huntington’s disease. Brain Res 253: 341–343. https://doi. org/10.1016/0006-8993(82)90706-5. Rubinsztein DC, Carmichael J (2003). Huntington’s disease: molecular basis of neurodegeneration. Expert Rev Mol Med 5: 1–21. https://doi.org/10.1017/S1462399403006549. Saleh N, Moutereau S, Durr A et al. (2009). Neuroendocrine disturbances in Huntington’s disease. PLoS ONE 4: e4962. https://doi.org/10.1371/journal.pone.0004962. Sannino S, Chini B, Grinevich V (2017). Lifespan oxytocin signaling: maturation, flexibility, and stability in newborn, adolescent, and aged brain. Dev Neurobiol 77: 158–168. https://doi.org/10.1002/dneu.22450. Schultz C, Ghebremedhin E, Braak H et al. (1997a). Neurofibrillary pathology in the human paraventricular and supraoptic nuclei. Acta Neuropathol 94: 99–102. https://doi.org/10.1007/s004010050679. Schultz C, Koppers D, Braak E et al. (1997b). Neurofibrillary degeneration in hypophysiotrophic nuclei of the aging human hypothalamus. In: H-W Korf, K-H Usadel (Eds.), Neuroendocrinology. Springer Berlin Heidelberg, Berlin, Heidelberg, pp. 115–126. https://doi.org/10.1007/978-3642-60915-2_10. Schultz C, Koppers D, Braak H et al. (1997c). Cytoskeletal alterations in the aged human neurohypophysis. Neurosci Lett 237: 93–96. https://doi.org/10.1016/s0304-3940(97) 00817-3. Spataro R, Volanti P, Vitale F et al. (2015). Plasma cortisol level in amyotrophic lateral sclerosis. J Neurol Sci 358: 282–286. https://doi.org/10.1016/j.jns.2015.09.011. Stypuła G, Kunert-Radek J, Stepie
n H et al. (1996). Evaluation of interleukins, ACTH, cortisol and prolactin concentrations in the blood of patients with parkinson’s disease. Neuroimmunomodulation 3: 131–134. https://doi. org/10.1159/000097237. Sundquist J, Forsling ML, Olsson JE et al. (1983). Cerebrospinal fluid arginine vasopressin in degenerative disorders and other neurological diseases. J Neurol Neurosurg Psychiatry 46: 14–17. https://doi.org/10.1136/ jnnp.46.1.14. Surks MI, Hollowell JG (2007). Age-specific distribution of serum thyrotropin and antithyroid antibodies in the US population: implications for the prevalence of subclinical hypothyroidism. J Clin Endocrinol Metab 92: 4575–4582. https:// doi.org/10.1210/jc.2007-1499. Swaab DF (1999). The human hypothalamo-neurohypophysial system in health and disease. In: Advances in brain vasopressin, progress in brain research, Elsevier, pp. 577–618. https://doi.org/10.1016/S0079-6123(08)61594-0. Swaab DF, Bao AM (2011). (Re-)activation of neurons in aging and dementia: lessons from the hypothalamus. Exp Gerontol 46: 178–184. https://doi.org/10.1016/j.exger.2010. 08.028. Swaab DF, Grundke-Iqbal I, Iqbal K et al. (1992). Tau and ubiquitin in the human hypothalamus in aging and Alzheimer’s disease. Brain Res 590: 239–249. https://doi.org/10.1016/ 0006-8993(92)91101-j.

Swaab DF, Raadsheer FC, Endert E et al. (1994). Increased cortisol levels in aging and Alzheimer’s disease in postmortem cerebrospinal fluid. J Neuroendocrinol 6: 681–687. https:// doi.org/10.1111/j.1365-2826.1994.tb00635.x. Swaab DF, Chung WC, Kruijver FP et al. (2001). Structural and functional sex differences in the human hypothalamus. Horm Behav 40: 93–98. https://doi.org/10.1006/hbeh.2001. 1682. Swaab DF, Bao AM, Lucassen PJ (2005). The stress system in the human brain in depression and neurodegeneration. Ageing Res Rev 4: 141–194. https://doi.org/10.1016/ j.arr.2005.03.003. Tang Y, Purkayastha S, Cai D (2015). Hypothalamic microinflammation: a common basis of metabolic syndrome and aging. Trends Neurosci 38: 36–44. https://doi.org/10.1016/ j.tins.2014.10.002. Taylor SE, Saphire-Bernstein S, Seeman TE (2010). Are plasma oxytocin in women and plasma vasopressin in men biomarkers of distressed pair-bond relationships? Psychol Sci 21: 3–7. https://doi.org/10.1177/0956797609356507. Traustado´ttir T, Bosch PR, Matt KS (2003). Gender differences in cardiovascular and hypothalamic-pituitaryadrenal axis responses to psychological stress in healthy older adult men and women. Stress 6: 133–140. https:// doi.org/10.1080/1025389031000111302. Trojanowski JQ, Revesz T, Neuropathology Working Group on MSA (2007). Proposed neuropathological criteria for the post mortem diagnosis of multiple system atrophy. Neuropathol Appl Neurobiol 33: 615–620. https://doi. org/10.1111/j.1365-2990.2007.00907.x. Tsigos C, Chrousos GP (2002). Hypothalamic-pituitaryadrenal axis, neuroendocrine factors and stress. J Psychosom Res 53: 865–871. https://doi.org/10.1016/ s0022-3999(02)00429-4. Ulusoy A, Rusconi R, P
erez-Revuelta BI et al. (2013). Caudorostral brain spreading of a-synuclein through vagal connections. EMBO Mol Med 5: 1119–1127. https://doi.org/ 10.1002/emmm.201302475. Van Cauter E, Leproult R, Kupfer DJ (1996). Effects of gender and age on the levels and circadian rhythmicity of plasma cortisol. J Clin Endocrinol Metab 81: 2468–2473. https:// doi.org/10.1210/jcem.81.7.8675562. van de Nes JA, Kamphorst W, Ravid R et al. (1993). The distribution of Alz-50 immunoreactivity in the hypothalamus and adjoining areas of Alzheimer’s disease patients. Brain 116: 103–115. https://doi.org/10.1093/brain/116.1.103. van den Beld AW, Kaufman J-M, Zillikens MC et al. (2018). The physiology of endocrine systems with ageing. Lancet Diabetes Endocrinol 6: 647–658. https://doi.org/10.1016/ S2213-8587(18)30026-3. Van der Woude PF, Goudsmit E, Wierda M et al. (1995). No vasopressin cell loss in the human hypothalamus in aging and Alzheimer’s disease. Neurobiol Aging 16: 11–18. https://doi.org/10.1016/0197-4580(95)80003-a. van Wamelen DJ, Aziz NA, Anink JJ et al. (2012). Paraventricular nucleus neuropeptide expression in Huntington’s disease patients. Brain Pathol 22: 654–661. https://doi.org/10.1111/j.1750-3639.2012.00565.x. van Wamelen DJ, Aziz NA, Roos RAC et al. (2014). Hypothalamic alterations in Huntington’s disease patients: comparison with

THE SUPRAOPTIC AND PARAVENTRICULAR NUCLEI IN NEURODEGENERATION genetic rodent models. J Neuroendocrinol 26: 761–775. https:// doi.org/10.1111/jne.12190. Wang Y, Sheng Q, Hou X et al. (2016). Thyrotropin and Alzheimer’s disease risk in the elderly: a systematic review and meta-analysis. Mol Neurobiol 53: 1229–1236. https:// doi.org/10.1007/s12035-014-9078-x. Watabe T, Tanaka K, Kumagae M et al. (1987). Diurnal rhythm of plasma immunoreactive corticotropin-releasing factor in normal subjects. Life Sci 40: 1651–1655. https:// doi.org/10.1016/0024-3205(87)90013-0. Weil RS, Lashley TL, Bras J et al. (2017). Current concepts and controversies in the pathogenesis of Parkinson’s disease dementia and dementia with Lewy bodies. [version 1; peer review: 2 approved]. F1000Res 6: 1604. https:// doi.org/10.12688/f1000research.11725.1. Weisman O, Zagoory-Sharon O, Schneiderman I et al. (2013). Plasma oxytocin distributions in a large cohort of women and men and their gender-specific associations with anxiety. Psychoneuroendocrinology 38: 694–701. https:// doi.org/10.1016/j.psyneuen.2012.08.011. Whitehouse PJ, Vale WW, Zweig RM et al. (1987). Reductions in corticotropin releasing factor-like immunoreactivity in cerebral cortex in Alzheimer’s disease, Parkinson’s disease,

123

and progressive supranuclear palsy. Neurology 37: 905–909. https://doi.org/10.1212/wnl.37.6.905. Wierda M, Goudsmit E, Van der Woude PF et al. (1991). Oxytocin cell number in the human paraventricular nucleus remains constant with aging and in Alzheimer’s disease. Neurobiol Aging 12: 511–516. https://doi.org/10.1016/ 0197-4580(91)90081-t. Yang CR, Bourque CW, Renaud LP (1991). Dopamine D2 receptor activation depolarizes rat supraoptic neurones in hypothalamic explants. J Physiol (Lond) 443: 405–419. https://doi.org/10.1113/jphysiol.1991.sp018840. Yong-Hong L, Xiao-Dong P, Chang-Quan H et al. (2013). Hypothalamic-pituitary-thyroid axis in patients with Alzheimer disease (AD). J Invest Med 61: 578–581. https://doi.org/10.2310/JIM.0b013e318280aafb. Zhong S, Monakhov M, Mok HP et al. (2012). U-shaped relation between plasma oxytocin levels and behavior in the trust game. PLoS ONE 7: e51095. https://doi.org/ 10.1371/journal.pone.0051095. Zhou JN, Swaab DF (1999). Activation and degeneration during aging: a morphometric study of the human hypothalamus. Microsc Res Tech 44: 36–48. doi:10.1002/(SICI)10970029(19990101)44:1 3.0.CO;2-F.

This page intentionally left blank

Handbook of Clinical Neurology, Vol. 180 (3rd series) The Human Hypothalamus: Middle and Posterior Region D.F. Swaab, F. Kreier, P.J. Lucassen, A. Salehi, and R.M. Buijs, Editors https://doi.org/10.1016/B978-0-12-820107-7.00008-2 Copyright © 2021 Elsevier B.V. All rights reserved

Chapter 8

Perinatal stress and epigenetics MOSHE SZYF* Department of Pharmacology and Therapeutics, McGill University, Montreal, QC, Canada

Abstract Animal and humans exposed to stress early in life are more likely to suffer from long-term behavioral, mental health, metabolic, immune, and cardiovascular health consequences. The hypothalamus plays a nodal role in programming, controlling, and regulating stress responses throughout the life course. Epigenetic reprogramming in the hippocampus and the hypothalamus play an important role in adapting genome function to experiences and exposures during the perinatal and early life periods and setting up stable phenotypic outcomes. Epigenetic programming during development enables one genome to express multiple cell type identities. The most proximal epigenetic mark to DNA is a covalent modification of the DNA itself by enzymatic addition of methyl moieties. Cell-type-specific DNA methylation profiles are generated during gestational development and define cell and tissue specific phenotypes. Programming of neuronal phenotypes and sex differences in the hypothalamus is achieved by developmentally timed rearrangement of DNA methylation profiles. Similarly, other stations in the life trajectory such as puberty and aging involve predictable and scheduled reorganization of DNA methylation profiles. DNA methylation and other epigenetic marks are critical for maintaining cell-type identity in the brain, across the body, and throughout life. Data that have emerged in the last 15 years suggest that like its role in defining cellspecific phenotype during development, DNA methylation might be involved in defining experiential identities, programming similar genes to perform differently in response to diverse experiential histories. Early life stress impact on lifelong phenotypes is proposed to be mediated by DNA methylation and other epigenetic marks. Epigenetic marks, as opposed to genetic mutations, are reversible by either pharmacological or behavioral strategies and therefore offer the potential for reversing or preventing disease including behavioral and mental health disorders. This chapter discusses data testing the hypothesis that DNA methylation modulations of the HPA axis mediate the impact of early life stress on lifelong behavioral and physical phenotypes.

INTRODUCTION: DNA METHYLATION AND CELLULAR IDENTITY While there has been a large body of evidence that early exposure to stress could result in long-term phenotypic changes in a wide range of animals as well as humans (Power and Hertzman, 1997), there was no mechanism that could explain how transient stressors would be embedded biologically to establish long-term and stable phenotypes. While genetics provided a compelling explanation for germ line inherited phenotypic variations, how

experience such as stress could lead to phenotypic differences that emerge at a relatively rapid time scale remained unresolved. Epigenetics emerged as a concept explaining how one genome can encode the numerous phenotypes expressed by multicellular organisms. The term was first proposed by Waddington (Waddington, 1959; Jablonka and Lamb, 2002). The first cue that DNA contains information beyond the four letters that encode the genetic code came from the discovery of a minor base in DNA, 5-methylcytosine, by Hotchkiss (1948) (Fig. 8.1). The possible role of this

*Correspondence to: Moshe Szyf, Department of Pharmacology and Therapeutics, McGill University, 3655 Sir William Osler Promenade, Montreal, QC H3G1Y6, Canada. Tel: +1-514-398-7107, Fax: +1-514-398-3622, E-mail: [email protected]

Fig. 8.1. Regulation of transcription by DNA methylation. Top panel: DNA methyltransferase (DNMT) catalyzes the transfer of a methyl moiety from the methyl donor SAM to cytosines in DNA. 5-methylcytosine could be further modified by an oxygenase enzyme TET to 5-hydroxymethylcytosine, 5-formylcytosine, and 5-carboxycytosine. TDG glycosylase removes and base excision repair (BER) can replace 5-carboxycytosine with unmethylated cytosine causing demethylation back to cytosine. Bottom panel: Expressed genes (horizontal green arrow) have unmethylated promoters and enhancers. Transcription factors bind enhancers (TF) and RNApolII machinery and transcription factors and factors such as the TATA binding protein (TBP) bind promoters; interaction between enhancers and promoters activates transcription (blue arrow). The body of expressed genes is methylated (red balloons). It is believed that this methylation prevents spurious transcription from cryptic promoters. Methylation of cytosines (red balloons) at promoters or enhancers leads to suppression of transcription (green arrow indicates transcription and the red X indicates suppression). Two mechanisms were proposed. Methylation can interfere with binding of transcription factors to promoters and enhancers, alternatively methylated DNA binding proteins (MBD) bind methylated regions and recruit other chromatin modifying enzymes (Sin3A HDAC and others) that silence transcription. Further modification of 5-methylcytosine by TET enzymes (light green balloons; COOH in the balloons indicates carboxylation of cytosine by sequence of oxidations catalyzed by TET enzymes) might lead to activation of enhancers or promoters, However, this is not yet proven.

PERINATAL STRESS AND EPIGENETICS additional chemical information in DNA has been intriguing and inspired Holliday and Pugh to theorize in 1974 that DNA methylation plays a role in regulating gene activity through selective modification of promoters and that it can therefore function as a developmental clock as well as a clock for aging (Holliday and Pugh, 1975). Both ideas were confirmed decades later. Remarkably, this model was proposed well before there was little evidence even for the presence of DNA methylating enzymes in eukaryotes nor was there yet evidence for cell-type-specific DNA methylation patterns. The discovery of methyl-sensitive restriction enzymes HpaII and its isoschizomer MspI combined with Southern blotting method for visualizing restriction patterns of specific genomic regions provided the first potent tool to probe DNA methylation in vertebrates (Waalwijk and Flavell, 1978). The first evidence that cell-type-specific methylation correlated with cell-typespecific gene expression came from studies of human gamma delta beta-Globin genes by van der Ploeg and Flavell in the late 1970s (van der Ploeg and Flavell, 1980); the genes were fully modified in germ line (sperm) DNAwhile in DNA of tissues expressing the globin genes, the region surrounding and including the expressed genes was hypomethylated (van der Ploeg and Flavell, 1980). Using the HpaII/MspI assay other tissue-specific genes were examined, further confirming cell-type-specific profiles of DNA methylation. These data also revealed that 50 regulatory regions of genes are less methylated in expressing tissues (Razin and Szyf, 1984). Taken together, these data formed the basis for the hypothesis proposed by Razin and Riggs (1980) that DNA methylation in promoters silences gene expression and that cell-type-specific demethylation during development activates cell-typespecific gene expression and plays a role in cellular differentiation (Razin and Riggs, 1980). This paradigm of control of cell-type-specificity by DNA methylation has been supported by numerous studies (Lister et al., 2009, 2013; Mo et al., 2015; Roadmap Epigenomics consortium et al., 2015), albeit the relationship between gene expression, DNA methylation and cell type specificity has proven to be far more complex (Massart et al., 2017a). The idea that epigenetic processes confer numerous cell-type identities to otherwise indistinguishable sequences provides a plausible explanation for the development of a multicellular organism from a single genome which is a hybrid of maternal and paternal genetic contributions. However, since epigenetic processes are responsible for maintaining the integrity of the organism, they should be highly predictable and immune to sporadic change that can disrupt normal development and maintenance of body physiological functions. There must be precise mechanisms for generation of DNA methylation profiles as well as maintenance of these patterns within a dividing cell lineage. There must be as well

127

mechanisms for translating the DNA methylation profiles to a gene expression readout. The two layers of information contained in the covalent structure of a DNA, the genetic sequence encoded in the four letters of DNA, and the DNA methylation profiles are generated and copied by different enzymatic machineries. While DNA sequence is directly inherited from paternal and maternal ancestors, the DNA methylation pattern is generated during development by enzyme-catalyzed transfer of methyl moieties from the ubiquitous methyl donor S-adenosyl methionine (SAM) to 50 position in cytosine in DNA (Wu and Santi, 1985) (Fig. 8.1). Two classes of DNA methyltransferases (DNMT) catalyze DNA methylation in vertebrates, maintenance DNA methyltransferase 1 is instructed by the methylation state of a CpG dinucleotide from the parental strand template to position a methyl moiety on its palindromic partner on the daughter strand (a CG sequence is a palindrome since across a 50 CG30 in the parental strand there is a 30 GC50 on the daughter strand; CG is the most commonly methylated CX dinucleotide in vertebrate DNA) (Gruenbaum et al., 1981). Early studies have suggested that maintenance methylation is possible because of the biochemical preference of DNMT1 to hemimethylated substrates; such substrates are generated when a DNA strand containing a methylated CG is copied during cell division. Later studies have suggested that another protein (ubiquitin-like, containing PHD and RING finger domains 1) URHF1 is involved as well in tethering DNMT1 to hemimethylated sites in the replication fork (Bostick et al., 2007). Generation of new DNA methylation patterns requires two de novo DNMTs 3A and 3B, which can methylate CG sequences that are unmethylated on both strands as well Cs at other dinucleotide sequence contexts CC, CA, and CT, which unlike CG are not palindromic and are not methylated at the parental template strand (Ramsahoye et al., 2000). De novo methylation plays an important role early in development (Razin and Shemer, 1995; Okano et al., 1999) and in postmitotic tissues such as neurons, where de novo methylation can expand the plasticity of DNA methylation profiles as will discussed later, since this process adds new methylation sites beyond the ancestral template. CA, CT, and CC dinucleotide methylations tend to accumulate in the brain, participating perhaps in acquiring new gene expression modalities and functions by learned experience (Lister et al., 2013). DNMT3A has indeed been implicated in generating DNA methylation patterns in the hypothalamus during sexual differentiation (Kolodkin and Auger, 2011; Cisternas et al., 2019) as well as other behavioral adaptations of the brain such as in drug addiction (Hopf and Bonci, 2010; LaPlant et al., 2010; Urb et al., 2019).

128

M. SZYF

The 5-methyl moiety could be further modified by oxygenation by members of the ten-eleven translocation (Tet) family of enzymes (1–3) to 5-hydroxymethylcytosine, formyl cytosine, and carboxycytosine derivatives (Kriaucionis and Heintz, 2009; Tahiliani et al., 2009; Maiti and Drohat, 2011) (Fig. 8.1). The oxidized derivatives could be replaced by unmethylated cytosines during cell division if there is no mechanism to maintain such sites by de novo methylation following DNA synthesis, or by base replacement repair activity (BER) that involves BER and thymidine DNA glycosylase (TDG) enzymes that remove the modified base; repair-mediated replacement of oxidized methyl derivative with a cytosine can lead to loss of a methylated position even in nondividing cells such as neurons (He et al., 2011; Lister et al., 2013) (Fig. 8.1). Although the focus in the field has been on the role of oxidized derivatives of methylcytosine in demethylation, it is becoming nevertheless clear that oxidized derivatives of methylcytosine are stably maintained and that they probably serve as epigenetic signals on their own such as activation of enhancer regions during cellular differentiation (Lu et al., 2014; Iurlaro et al., 2016; Wu et al., 2018). The DNA methylation landscape is therefore intertwined with its oxidized derivatives, both playing roles in defining gene function and acquisition of celltype-specific identities and perhaps enhancing the resolution of the epigenetic signal (Iurlaro et al., 2016). It should be noted however that the generation and maintenance of oxidized methylcytosine derivatives is dependent on DNMT as well as Tet. In addition to modification of DNA, a complex system of epigenetic regulation is encoded in the profile of histone modifications which include methylation, acetylation, phosphorylation, and ubiquitination. Histone modifications can increase or decrease the accessibility of genes to transcriptional machineries based on the presence of activating or silencing chromatin marks (Tsukiyama and Wu, 1997). Histone modifications are reversible. The enzymes that modify histones or remove histone marks are important epigenetic regulators. Histone acetylases catalyze the acetylation of histones (Brownell and Allis, 1996, 2021) while Histone deacetylases (HDAC) catalyze deacetylation of histones (Wolffe, 1996). Histone methyltransferases methylate histones (Strahl et al., 1999; Rea et al., 2000), while histone demethylases demethylate histones (Kooistra and Helin, 2012). Histone marks are involved in setting up gene expression programs during cellular differentiation as well as in response to different physiological signals (Jenuwein and Allis, 2001), such as the binding of the glucocorticoid receptor (GR) to its recognition elements in response to stress (Wallberg et al., 2000). The commonly studied histone modifications that are associated with gene activation (Strahl and Allis, 2000) are acetylation of the lysine at position 9 of histone H3 (H3K9ac)

(Brownell and Allis, 1996; Kuo et al., 1996), monomethylation at lysine 4 of H3 (H3K4me1) (Barski et al., 2007; Zhu et al., 2013), which is associated with enhancers, and acetylation at lysine 27 of H3 H3K27ac, which is associated with active enhancers (Barski et al., 2007; Heintzman et al., 2007; Creyghton et al., 2010). Trimethylation at lysine 4 of H3 histone marks active promoters. Methylation at lysine 9 (Shinkai, 2007) or lysine 27 (Rougeulle et al., 2004) in H3 histones is associated with gene silencing and heterochromatin. Interestingly, bivalent chromatin with both activating and silencing histone modifications is abundant on genes in embryonal stem cells (Bernstein et al., 2006), a stage which precedes differentiation at crossroads for multiple cellular trajectories. In contrast to enhancers and promoters, inactivating chromatin marks such as H3K9me3 H3K36me3 methylation are present in bodies of genes, possibly preventing spurious firing of cryptic promoters (Hahn et al., 2011). Similarly, bodies of active genes are methylated (Hahn et al., 2011). There are strong interrelationships between the different chromatin modifications as well as between histone modifications and DNA methylation (D’alessio and Szyf, 2006; Karemaker and Baubec, 2020). A comprehensive discussion of the role of chromatin modification in setting up stable gene expression programs is beyond the scope of this chapter, which will mainly focus on cytosine methylation. A broader definition of epigenetic regulation includes chromatin modeling, microRNAs, and other long noncoding RNAs as well as transcription factors. This review will focus on DNA methylation, which is the only epigenetic modification (with its oxidized derivatives) that is part of the covalent structure of DNA. In summary, the DNA molecule bears two identities, the ancestral identity in the sequence and the cellular identity in the DNA methylation and its oxidized derivatives. The cellular identity is highly predictable and similar across individuals and is progressively generated during prenatal and postnatal developmental and life cycle stations. The prenatal and perinatal windows involve dramatic epigenetic reconfigurations when cell types and organs are formed (Benvenisty et al., 1985a,b; Cedar and Razin, 1990; Reik et al., 2001). However, changes in DNA methylation continue during postnatal development at important stages such as puberty (Lomniczi et al., 2015; Yang et al., 2016, 2018; Luo et al., 2017; Yuan et al., 2019), menopause (Levine et al., 2016; Bacon et al., 2019) and aging (Levine et al., 2016).

DNA METHYLATION AND GENE FUNCTION Early associations of DNA methylation and cell-typespecific gene activity suggested that DNA methylation at 50 regulatory regions and enhancers is inhibitory to

PERINATAL STRESS AND EPIGENETICS gene activity (Razin and Szyf, 1984). Like the role of DNA methylation in restriction/modification systems DNA methylation could block the interaction of transcription factors with their recognition elements in promoters and enhancers leading to silencing of gene expression (Fig. 8.1, bottom panel). The first described example of inhibition of binding of a transcription factor to a methylated recognition element in a promoter is activating enhancer-binding protein 2 (AP2); methylation of the AP2 recognition sequence in the promoter of the Proenkephalin gene inhibits its expression by blocking binding of AP2 (Comb and Goodman, 1990). Cytosine methylation could also target binding of proteins that comprise a methylated DNA recognition domain from the methylated DNA binding domain family (MBD) (Hendrich and Bird, 1998). The first example of gene suppression by MBD binding is the RETT syndrome protein methyl CpG binding protein 2 (MeCP2) (Meehan et al., 1992). Binding of MeCP2 targets a repressive complex that includes histone deacetylases HDAC leading to loss of H3K9ac and silenced chromatin configuration (Nan et al., 1998; Ng et al., 1999). This mechanism involves crosstalk between chromatin modification and DNA methylation. There are many known avenues of crosstalk between these two levels of epigenetic regulation and therefore affecting one level of regulation will impact the other (D’alessio and Szyf, 2006). For example, inhibiting HDAC with sodium butyrate (Szyf et al., 1985), TSA (Cervoni et al., 2001), or valproate (Detich et al., 2003) results in loss of DNA methylation and hypomethylation of promoters. On the other hand, active chromatin marks like H3K4me3 prevent methylation by DNMT3a and DNMT3b, while H3K36me recruits DNMT3a and DNMT3 (Morselli et al., 2015; Rondelet et al., 2016; Weinberg et al., 2019) to methylated gene bodies of actively transcribed genes; methylation of gene bodies in contrast to promoters and enhancers is associated with gene activity and is believed to play a role in silencing of spurious promoters in the gene body (Neri et al., 2017). The histone methyltransferase enhancer of zeste 2 polycomb repressive complex 2 subunit (EZH2), which is a component of polycomb repressive complex 2 that methylates lysine 27 on histone H3 (Etchegaray et al., 2006), a repressive chromatin mark, associates with DNMT1 to methylate EZH2 target promoters (Vire et al., 2006). The intricate bilateral relationships between chromatin modifications and DNA methylation might have evolved to produce highdensity gradients of epigenetic silencing/activation but perhaps also to provide multiple routes for generation of new epigenetic profiles through a rich variety of crosstalk with other signaling pathways in response to different signals, as well as protecting the epigenetic matrix from spurious alterations in one kind of modification. Interestingly, MBDs including MeCP2 and MBD2 might

129

be involved in gene activation as well. For example, phosphorylation of MeCP2 at serine 421 by a CaMKII-dependent mechanism is involved in activitydependent transcriptional induction of BDNF (Chen et al., 2003; Zhou et al., 2006). Mecp2 dysfunction results in changes in gene expression 85% of which involve genes that are activated by Mecp2; possibly through interaction with the transcription factor Creb1 (Chahrour et al., 2008). Similarly, MeCP2 in collaboration with CAMP responsive element-binding protein 1 (Creb1) is required for transcriptional activity of forkhead box P3 (Foxp3), a gene important for regulatory cells function (Lal et al., 2009). Genome-wide mapping of MBD2 binding in embryonal stem cells reveals both repressive and activating functions (Baubec et al., 2013). MBD2 activates the glucocorticoid receptor NR3C117/ promoter in response to maternal care in rat hippocampal cells (Weaver et al., 2014), and MBD2 is required for activation and demethylation of several autism risk genes in the hippocampus (Lax et al., 2018). Methylated DNA binding proteins coordinately regulate both expression and suppression of methylated/demethylated genes. It is hypothesized that different interactions of MBD proteins in either transcriptional activating or repressor complexes define the direction of the impact that they have on transcriptional activity (Baubec et al., 2013). In summary, epigenetic networks comprised of transcription factors, “readers” of DNA methylation and chromatin modifications as well as covalent modification states of histones and the DNA molecule itself, enable a diversity of gene expression outputs and cellular phenotypes to emerge during embryonal and organismal development and life trajectories.

EPIGENETIC PROGRAMMING OF THE HPA AXIS BY MATERNAL BEHAVIOR AND ITS REVERSAL BY EPIGENETIC REGULATORS The study of DNA methylation during development shows a stable mechanism that enables different states of activity to identical sequences of DNA, thus allowing two identical genes to express different phenotypes. Could such a mechanism provide differential identity to identical genes as a function of experience and specifically early life experience? DNA methylation profiles are predictable in time and space during the life course. The developmental programming of epigenetic modification is believed to be critical for maintaining cell type identity. Thus it is expected that epigenetic programs would be faithfully maintained to preserve developmental and cellular integrity (Cedar and Razin, 1990). There are basic biochemical mechanisms that protect against a drift in DNA methylation profiles; the preference of

130

M. SZYF

DNMT1 to hemimethylated sites generated only when a methylated template strand is replicated (Gruenbaum et al., 1982; Stein et al., 1982), low expression of de novo DNMTs in differentiated cells, and the tight interrelationship between chromatin modification and DNA methylation could protect against the consequences of spurious drift in DNA methylation. In cancer cellular identity is disrupted and so is the DNA methylation profile (Baylin, 1992; Baylin et al., 1998), and it was hypothesized that the alterations in DNA methylation are involved in the broad transcriptional and phenotypic changes in cancer cells. Is it possible that DNA methylation which is so critical for maintaining the integrity of cellular identity would be susceptible to the vagaries of dynamic social and physical environments? The hypothalamus controls stress response through release of corticotropin-releasing hormone (CRH) by the paraventricular nucleus. The release of this hormone is regulated by a negative feedback by glucocorticoids acting on the glucocorticoid receptor encoded by the Nr3c1 gene in the hippocampus and hypothalamus. Disruption of the negative feedback loop can result in a heightened stress response (McEwen et al., 2015). First evidence that DNA methylation could be mediating between early life stressful social exposure and lifelong phenotypes came from studies of maternal care in rats, which varies across individual animals. Maternal care in rats consists of licking and grooming (LG), arch back nursing, and nipple switching, and there is an interindividual variation in maternal care intensity. Adult offspring of high maternal care rats show a reduced stress responsivity as compared with low maternal care offspring (Liu et al., 1997). Thus the distribution of maternal behaviors is translated into a distribution in stress responsivity in the offspring. This effect is not purely genetic since cross fostering studies reveal that the offspring phenotypes are triggered by the fostering mothers (Francis et al., 1999). In summary, stress response phenotypes are triggered by maternal behavior. Maternal care variations result in differences in DNA methylation and histone acetylation and expression of the exon17 promoter of Nr3c1 gene encoding the glucocorticoid receptor in offspring hippocampus (Weaver et al., 2004). Reversal of this epigenetic programming in the adult offspring by injecting the HDAC inhibitor trichostatin A converts the stress response of the offspring of low maternal care mothers into a stress behavior characteristic of the high maternal care, while injection of methionine changes the behavior of the high maternal care animals to become indistinguishable from the low maternal care animals (Weaver et al., 2004, 2005). These data imply first that epigenetic modification is mediating the effect of maternal behavior on the offspring phenotype since the stress phenotype can be reversed by epigenetic

modulation. Second, although epigenetic programming results in stable lifelong phenotypes, they are nevertheless reversible by epigenetic modulation. This has implications on future directions for preventing and treating HPArelated disorders. The changes in DNA methylation in response to differences in maternal care are directed and programmed and not just a consequence of a random drift triggered by stress. Epigenetic programming by maternal care elicits a cascade of specific molecular events in the offspring hippocampus that link the maternal behavior and epigenetic changes at specific positions in the offspring DNA. Maternal behavior triggers a serotoninergic signaling pathway, leading to activation of a transcription factor (nerve growth factor induced A) NGFIA that binds the nr3c1 exon 7 region promoter (Weaver et al., 2007). NGFIA recruits the histone acetyl transferase CREB binding protein and the methylated DNA binding protein MBD2 (Weaver et al., 2014), which results in increased acetylation and reduced methylation. Could the lessons derived from epigenetic programming by maternal behavior guide early life intervention that will improve lifelong stress responsivity, and would this also be mediated via epigenetic programming of HPA axis? A recent study has shown that augmented maternal care in the first postnatal week improves stress-resilience and memory as compared with standard maternal care. This was associated with broad changes in methylation in the hypothalamus; differentially methylated genes were enriched for neuronal pathways as well as stress response genes. These data suggest that early life behavioral intervention might be utilized to reverse effects of low maternal care and improve behavioral traits in adulthood beyond the normal standard (Vogel Ciernia et al., 2018). This touches on the important issue of medical science biased preference for disorders and disease versus attempts to use physiological mechanisms to enhance and increase “well-being.” The long-term phenotypes triggered by maternal behavior could be transferred across multiple generations. The cross-generational transmission of the effects of maternal care is associated with epigenetic changes in methylation of estrogen receptor-alpha1b promoter gene in the sexually dimorphic medial preoptic area of the hypothalamus of female offspring (Champagne et al., 2006). High maternal licking and grooming is associated with demethylation of a signal transducer and activator of transcription 5B (Stat5b) transcription factor binding site in the estrogen receptor alpha (ERalpha) promoter (Champagne et al., 2006). Methylation of this site in offspring of low LG mothers results in reduced binding of the transcription factor and reduced expression (Champagne et al., 2006). Since ERalpha is important for maternal behavior, this provides a molecular

PERINATAL STRESS AND EPIGENETICS mechanism by which a mother transmits to her offspring maternal behavior traits. These in turn will determine stress responsivity in the next generation. Such a mechanism could plausibly transmit behavioral traits across multiple generations without germ line transmission of genetic or epigenetic signals.

EARLY LIFE STRESS-MEDIATED EPIGENETIC REPROGRAMMING OF THE HPA AXIS: EPIGENETIC CHANGES IN THE HYPOTHALAMUS Epigenetic programming associated with the natural distribution of maternal behaviors provides evidence for adjustment of lifelong phenotypes to experiential and environmental cues through epigenetic mechanisms. Early life stress (ELS) in mice results in increased and endured corticosterone response, coping and memory deficits, which models the impacts of early life stress in humans (Nugent et al., 2011; Targum and Nemeroff, 2019). Murgatroyd et al. examined whether epigenetic mechanisms in the hypothalamus are mediating the establishment of these “traits” (Murgatroyd et al., 2009). Early life stress was associated with demethylation of a DNA binding recognition site for MeCP2 at a downstream enhancer of the arginine vasopressin (AVP) gene and its persistent induced expression in neurons of the paraventricular nucleus in the hypothalamus (Murgatroyd et al., 2009). During normal neuronal development, this region is targeted for methylation and silencing by binding of polycomb complexes, while ELS triggers loss of this developmentally regulated methylation (Murgatroyd and Spengler, 2014). In contrast, early life stress activates the expression of the Nr3c1 in hypothalamic neurons that produce CRH by methylating a CpG island shore which regulates the gene. Methylation of this region inhibits the binding of a transcriptional repressor YY1 to an “insulator” sequence and activates Nr3c1 resulting in inhibition of CRH release under chronic stress situations (Bockm€ uhl et al., 2015). Thus ELS epigenetic programming activates Avp and Pomc, which causes sustained activity of HPA axis while also inducing GR in the hypothalamus to reduce stimulation of CRH release under chronic stress. However, GR-mediated negative feedback is less efficient in ELS (Bockm€ uhl et al., 2015). It remains to be explained how the enhanced expression of Nr3c1 in the hypothalamus, which is expected to repress CRH release is consistent with the observation that ELS in mice results in increased and endured corticosterone response. There is evidence in rodents that the next station in the HPA axis, the pituitary, is epigenetically programmed as well by early life stress. POMC is a prohormone that

131

gives rise to the ACTH peptide, which triggers stress hormone release by the adrenal gland (McEwen et al., 2015). Early life stress administered by daily separation of pups from their dam 1 to 10 days after birth causes demethylation of a regulatory region at the Pomc gene, which persists into adulthood and up to 1 year and is associated with induced expression and increased basal levels of ACTH (Wu et al., 2014). Interestingly there is an age-dependent reduction in methylation from 3 months to 1 year in the control mice that were not exposed to early life stress, but this demethylation did not contribute to elevation of basal ACTH (Wu et al., 2014) (see Fig. 8.2 for summary). Early life adversity results in lasting differences in DNA methylation in other brain regions and other genes involved in learning and brain function. Early life maltreatment in rats by exposure to an abusive mother triggered persistent reduction in Bdnf expression and increased methylation in exon IX and IVof the prefrontal cortex that remains into adulthood as well as changes in DNA methylation in offspring of females that were exposed to early life maltreatment (Roth et al., 2009).

EARLY LIFE STRESS/ADVERSITY CAUSES BROAD EPIGENETIC CHANGES ACROSS THE GENOME WHICH ARE EVOLUTIONARY CONSERVED Changes in DNA methylation in the HPA axis are evolutionary conserved. Differences in DNA methylation were reported at the Nr3c1 gene exon 1f, a homologous promoter to the rat exon 17, in postmortem samples from people who committed suicide, who were abused as children, and controls and suicide victims were not abused (McGowan et al., 2009). Changes in DNA methylation in response to early life stress are broad and are not limited to classical candidate genes known to regulate the HPA axis. Analysis of transcriptional changes in the hippocampus between low and high LG maternal care adult offspring revealed hundreds of differentially expressed genes (Weaver et al., 2006). rRNA genes are hypermethylated in hippocampi of adults who were abused as children (McGowan et al., 2008), examination of genomic regions that flank the Nr3c1 locus in both humans and rats hippocampi revealed broad changes in methylation and histone acetylation that span several million base pairs, and included in this region is the entire family of more than 70 protocadherin genes, which are involved in neuronal development and synaptic transmission (Frank and Kemler, 2002) (McGowan et al., 2011). These broad changes in DNA methylation are evolutionary conserved in both humans and rats (Suderman et al., 2012).

132

M. SZYF

Fig. 8.2. Changes in DNA methylation in the HPA axis in response to early life stress (ELS) reprogram the stress response (results from animal studies). HPA axis is regulated by feedback inhibition by cortisol. In response to ELS the gene encoding GR Nr3c1 is methylated in the hippocampus, resulting in reduced expression of GR and reduced inhibitory downstream signaling to the hypothalamus. AVP and CRH are expressed in the hypothalamus and both get demethylated and activated by ELS increasing POMC release from the pituitary. The POMC gene is activated in the pituitary by demethylation in response to ELS. FKBP5 is induced by glucocorticoids and acts as a proximal negative regulator of GR responsivity. Demethylation of an intronic enhancer in the hippocampus and hypothalamus activates FKBP5 and thus the negative feedback of FKBP5 on GR and reduces GR-mediated inhibition of the HPA. Combined, these epigenetic changes increase HPA activation and reduce negative feedback regulation.

THE EFFECT OF EARLY LIFE STRESS/ADVERSITY-RELATED BROAD EPIGENETIC CHANGES IN PERIPHERAL TISSUES OUTSIDE THE HPA AXIS One of the challenges of studying epigenetics of HPA axis in humans is that analysis of DNA methylation requires biological samples from hippocampus, hypothalamus, and pituitary, which are inaccessible in living humans. Since DNA methylation and other epigenetic marks are responsible for cell-type specific gene expression programming (Lister et al., 2009, 2013; Roadmap Epigenomics Consortium et al., 2015; Mo et al., 2015), they are predicted to be cell-type specific as discussed previously. It is expected therefore that epigenetic changes in HPA axis will also be specific to the brain regions regulating the HPA. This hypothesis was examined by comparing DNA methylation changes in adult rhesus macaque monkeys who were reared either by their natural mother or in a nursery (Ruppenthal et al., 1976; Suomi et al., 1976). The differentially methylated sites were mostly tissue specific; however, common differentially methylated regions between control and maternally deprived monkeys in the brain and T cells were reported as well (Provencal et al., 2012). Several

studies that followed examined differences in DNA methylation associated with early life or parental stress and neuropsychiatry disease that were altered in blood in humans, for example (Mehta et al., 2011, 2013). The idea that epigenetic programming relevant to HPA or associated with HPA disorders could be detected in the immune system has encountered resistance. The DNA methylation alterations seen peripherally might be just surrogates of DNA methylation alterations in the hypothalamus or hippocampus or part of the systemic response to stress that is perceived by the HPA axis but has implications across many systems including metabolic, cardiovascular, and immune functions. Stress hormones that are also epigenetic modulators through their nuclear receptors (Manteuffel-Cymborowska, 1999; Hebbar and Archer, 2003) are natural candidates that translate stress signals perceived by the hippocampus and hypothalamus to epigenetic programming across multiple systems as will be discussed later. Thus the epigenetic response to early life stress is proposed to play a role in adapting physiological systems across the organism to anticipated lifelong social and physical environments. The immune system might be an important component of this organismal adaptation.

PERINATAL STRESS AND EPIGENETICS A longitudinal examination of DNA methylation profiles in rhesus Macaque T cells during development of maternally reared and nursery reared male and female monkeys revealed that DNA methylation profiles in the peripheral immune systems evolve during postnatal development into adolescence, that they are sex specific, and that the normal developmental trajectory of evolution of DNA methylation profiles in T cells is altered by maternal deprivation early in life (Massart et al., 2016). Differences in DNA methylation in response to social stress cues are detected as early as birth, not only in the immune system but in the placentae as well. Differences in DNA methylation are seen in placenta between newborns of mothers of high and low social rank (Massart et al., 2017b). Since the brain is inaccessible to epigenetic research in living people, many studies examined associations of the DNA methylation of candidate HPA genes in blood and early life adversity. For example, NR3C1 (Perroud et al., 2011; Cicchetti and Handley, 2017) exon 1f, the proximal regulator of glucocorticoid receptor FK506 binding protein 5 (FKBP5) (Klengel et al., 2013; Yehuda et al., 2016; Harms et al., 2017; Parade et al., 2017; Tozzi et al., 2018), brain-derived neurotrophic factor (BDNF) (Unternaehrer et al., 2015), OXTR (Unternaehrer et al., 2015; Gouin et al., 2017; Kraaijenvanger et al., 2019), serotonin transporter SLC6A4 (Provenzi et al., 2016), and ionotropic glutamate receptor NMDA type subunit 2B (GRIN2B) (Engdahl et al., 2020). POMC DNA hypermethylation in blood was associated with alcohol and tobacco dependence possibly linking early life adversity and smoking and alcohol dependence (Muschler et al., 2018; Gangisetty et al., 2019). The main criticisms of published candidate gene methylation studies in the periphery are: the relatively limited number of cases, possible genetic and other exposures confounding the analysis, and the small differences in methylation. Methylation is a binary measure, a site can either be methylated or unmethylated; the percentage of methylation which is measured in a population of cells provides a measure of the fraction of cells that are methylated. So, for example, if a candidate gene promotor in a blood sample from a case group is 5% methylated and a control group is 0% methylated, it indicates that 5% of the cells have 100% methylation at this site. That is, the differences in methylation occurs only in a small fraction of cells. One possible explanation is that the methylation differences emerged in response to an early stress exposure in a specific subset of cells and are thereafter faithfully copied across cell divisions of their lineage, thus bearing the memory of the exposure. To date such subset of cells has not been identified. Alternatively, 5% differentially methylated cells are randomly

133

distributed across different cell types, and this stochastic variation is maintained across cell divisions. The memory of early stress in this case is not preserved in a specific subset of cells but as a 5% overall difference in the capacity for preservation of methylation during cell division that will randomly affect different cells at each cell division. There is no evidence however for such a mechanism. Single-cell bisulfite sequencing might be needed to test these two alternative possibilities and to determine whether a subset cell-type exists that bears DNA methylation marks of early stress exposure. It is also interesting to note that the candidate genes examined in these studies have specific roles in the brain. It is unclear whether the change in methylation in blood has any functional role in blood or whether it is just a surrogate of similar changes occurring in the brain. Mechanisms of such surrogacy of brain epigenetics in blood are unclear yet but will be discussed later. In addition to candidate gene approaches, unbiased “agnostic” genome-wide studies examined associations of early life stress and blood or saliva DNA methylation. Whole-genome methylation analysis in whole blood of 14 children raised by biological parents and institutionalized children revealed a differential methylation profile that includes genes involved in the immune response as well as brain function. A larger study comparing 29 children raised in orphanages and 29 raised in biological families using Illumina Epic Arrays found significant differences in child adaptive behavior domains and DNA methylation (Naumova et al., 2012, 2019). To address the challenge of randomization of case and control groups for known and unknown confounders, King and colleagues examined the impact of a natural disaster, the Quebec ice storm of 1998 as a random objective stressor of mothers during the perinatal period (King et al., 2000). Genome-wide DNA methylation profiles in T cells in adolescent offspring of mothers exposed to the ice storm during the perinatal period were correlated with the objective stress of the mothers (Cao-Lei et al., 2014). The study identified a large number of CG sites whose methylation level correlated with objective stress (Cao-Lei et al., 2014). Functional analysis of the pathways affected revealed gene pathways representing metabolism and obesity, immunity, which was as expected predominant, and behavior (Cao-Lei et al., 2014). Mediation analyses revealed that DNA methylation mediates the effect of maternal cognitive appraisal of the disaster on children insulin secretion (Cao-Lei et al., 2018), children BMI and central adiposity (CaoLei et al., 2016a), maternal stress on childhood asthma (Turcotte-Tremblay et al., 2014), and cytokine production (Cao-Lei et al., 2016b). Unbiased analyses, which are not restricted by hypotheses-driven candidate genes which focus on the brain, unravel the system wide nature

134

M. SZYF

of methylation differences. These data are consistent with the idea that DNA methylation profiles in the periphery that are associated with early life stress are not just surrogates of central events but that they represent a multisystem response to early life stress that is potentially mediating body wide adaptations of physiology to early life stress. Animal studies suggested that early life stress affects stress reactivity in adults. A genome-wide association study examined associations between DNA methylation in blood and cortisol stress reactivity in adults. A locus which showed the highest association in the Kit ligand gene (KITLG) was discovered (He et al., 2018). Methylation of KITLG in blood correlates with methylation in the frontal cortex. This association was replicated in a separate set of independent samples and in DNA from buccal cells. Importantly the differentially methylated region is positioned within an enhancer consistent with a regulatory role for this differential methylation. KITLG methylation partly mediates the association between early life traumatic events and cortisol stress reactivity (He et al., 2018). These data support the hypothesis that in humans like in rats, DNA methylation mediates the effect of early life stress on adult stress reactivity. Such associations could be discovered in adult blood DNA as well as buccal cells. Interestingly the KIT ligand is expressed in hematopoietic progenitors that proliferate in response to chronic stress in mice, increasing the relative level of neutrophils and inflammatory monocytes (Heidt et al., 2014). These data link systemic DNA methylation reprogramming by early life stress with a systemic response and specifically immune/inflammation readjustment.

SOCIOECONOMIC POSITIONING AND EPIGENETIC PROGRAMMING Low social rank is an important stressor. Low socioeconomic positioning at early childhood is a risk factor for mental and physical challenges during adulthood (Power and Hertzman, 1997; Power et al., 2006). A study in nonhuman primates rhesus macaques showed broad differences in DNA methylation at birth in placentae between monkey born to high- and low-ranking mothers suggesting that maternal social rank affects DNA methylation at birth even before the newborn has formed any social interactions (Massart et al., 2017b). A study that examined genomewide DNA methylation in a small number of people from the British Birth cohort of 1948 delineated associations with early life socioeconomic status (SES) (Borghol et al., 2012). A follow-up study in the same cohort revealed associations of adult DNA methylation profiles with early child abuse (Suderman et al., 2014). Interestingly, the DNA methylation profiles of poverty and child abuse were

different suggesting that different forms of social stressors might have different signatures in DNA methylation (Suderman et al., 2014). Several studies have confirmed differential DNA methylation associations with socioeconomic status (Stringhini et al., 2015; Fiorito et al., 2017; Smith et al., 2017). Examining genome-wide promoter DNA methylation profiles from peripheral blood monocytes in a community cohort stratified by socioeconomic positioning revealed associations with perceived stress, cortisol output, and early-life socioeconomic status. DNA methylation was strongly correlated with the inflammatory response of peripheral blood mononuclear cells to microbial products. Interestingly no correlation with gene expression was seen in this study (Lam et al., 2012). DNA methylation across hundreds of sites in the genome are correlated with age, and DNA methylation age clocks were previously described (Horvath, 2013). Accelerated DNA methylation clock was associated with increased risk for age-related disease (Perna et al., 2016). A very recent study examined shifts in the “DNA methylation aging” clock in response to low socioeconomic positioning and demonstrated accelerated biological aging in people with low socioeconomic positioning (Fiorito et al., 2017). Acceleration of DNA methylation aging with low socioeconomic positioning might be mediating effects of social position on longevity (Stringhini et al., 2017). These studies again support the hypothesis that perinatal stress affects DNA systemically outside the immediate HPA axis and DNA methylation might be mediating the long-term physiological responses to stress signals across different physiological systems and tissues.

EARLY LIFE STRESS AND PTSD Posttraumatic stress disorder is a paradigmatic example of long-term phenotypic changes triggered by a temporary stressor, a traumatic event. It is therefore a candidate for mediation by DNA methylation changes. It has been well established that early life adversity increases susceptibility for developing PTSD. It was hypothesized that early life stress causes DNA methylation changes that increase the likelihood of developing PTSD in adulthood. In support of this hypothesis, genome-wide DNA methylation analysis in peripheral leukocytes of adults with posttraumatic stress disorder revealed a distinct profile of DNA methylation in those that were exposed to maltreatment in childhood (Mehta et al., 2013). The FK506 binding protein 5 is a proximal negative regulator of glucocorticoid responsivity. FKBP5 is upregulated in response to elevation in glucocorticoid levels in blood and binds and sequesters the glucocorticoid receptor leading to reduced expression of GR responsive genes and “glucocorticoid resistance” (Tatro et al., 2009).

PERINATAL STRESS AND EPIGENETICS A polymorphism of FKBP5 was shown to associate with PTSD only in people who suffered from early life adversity (Klengel et al., 2013). Klengel et al. demonstrated that early life trauma causes demethylation of a GR response element in the gene when the risk polymorphism is present, which increases its response to stress and causes “glucocorticoid resistance” in PTSD patients (Klengel et al., 2013). A plausible mechanism is that demethylation in FKBP5 is caused by exposure to glucocorticoids in response to early life stress. Indeed, FKBP5 is demethylated and induced in mice hypothalamus and hippocampus in response to chronic exposure to corticosterone, which leads to anxiety-like behavior in these mice (Lee et al., 2010). Exposure of hippocampal neuronal stem cells but not differentiated neurons to glucocorticoids resulted in demethylation at the same element that is found demethylated in the blood of patients with PTSD, suggesting a developmental window for demethylation of FKBP5 in the brain which might explain why early life stress is important for developing PTSD in people carrying the FKBP5 risk allele (Klengel et al., 2013). The parallel changes in DNA methylation in peripheral blood cells and undifferentiated neuronal precursors are consistent with the idea that HPA-related DNA methylation alterations would be found in the periphery, supporting the examination of peripheral DNA methylation in human behavioral and neuropsychiatry studies (Klengel et al., 2013). DNA methylation of FKBP5 was recently found to be associated with obesity and insulin resistance consistent with the fact that disruption of HPA contributes to obesity and metabolic disorders later in life. CG sites in intron 7 of FKBP5 were hypermethylated in DNA from abdominal subcutaneous adipose tissue and gluteal subcutaneous adipose tissue biopsies from 27 obese compared to 27 normal weight South African women and DNA methylation was inversely correlated with expression of the FKBP5 mRNA and positively correlated with adiposity, metabolic and inflammatory parameters (Willmer et al., 2020). This study is consistent with the idea that early life stress affects DNA methylation across peripheral tissues and physiological systems as well as the brain and that these changes in DNA methylation affect several physiological systems, in this case metabolic response and regulation. However, a recent large study of DNA methylation epigenome-wide association in blood after early childhood trauma found no significant association after correction for confounding factors such as cigarette smoking (Marzi et al., 2018). It is still unclear how to reconcile this with a large body of data published before, and following this article, showing associations with candidate genes as well as genome-wide associations

135

cited previously. It is also difficult to reconcile with a large body of animal model data showing epigenetic rearrangements in response to early life adversity in the hippocampus and hypothalamus in response to early life stress cited previously. One possibility that is discussed in the paper is that the genome-wide association was performed on blood DNA and that blood DNA does not reflect the HPA-related brain region-specific DNA methylation alterations. In addition, it should be noted that although significant associations were observed in blood and saliva DNA in response to early adversity in the cited studies, the degree of DNA methylation change though significant is small as discussed previously. It is also possible that smaller genome-wide studies were confounded by smoking and other environmental exposures that were not corrected as they were in the larger study (Marzi et al., 2018).

CROSS-GENERATIONAL EFFECTS OF TRAUMA Interestingly, this link of FKBP5 and early life adversity was recently extended to cross-generational effects of trauma. Yehuda et al. examined methylation at FKP5 intron 7 in blood DNA from holocaust survivors and their offspring and compared it with a matched control group (Yehuda et al., 2016). Interestingly, holocaust exposure caused hypermethylation in the parents and demethylation in the offspring suggesting more complex responses in this group that in people carrying the FKBP5 risk allele, which might reflect a “resilience” response in the parent group, while in the offspring that were not themselves exposed to the holocaust the direction of the methylation change was similar to PTSD with childhood adversity (Yehuda et al., 2016). The differential methylation of FKBP5 intron 7 was replicated in a recent study on a larger group of subjects and FKBP5 methylation was found to be significantly lower in the offspring of holocaust survivors, methylation correlated with glucocorticoid sensitivity rather than basal expression of FKBP5 (Bierer et al., 2020). It is important to note that methylation, particularly of an enhancer, might play a role only in induced responsive expression such as response to hormone and not affect steady-state mRNA; DNA methylation might be regulating the response to elevated stress-induced GC rather than baseline expression (Fig. 8.3).

EARLY LIFE STRESS METHYLATION AND METABOLIC CONTROL The HPA axis coordinates metabolic and stress responses. Early life stress is associated with metabolic disruption later in life, while food restriction is a potent stressor. The crh gene in the hypothalamus is

136

M. SZYF

Fig. 8.3. Changes in DNA methylation program future context-dependent gene expression. Early life stress triggers changes in DNA methylation but these might not lead to changes in gene expression. Demethylation enables binding of transcription factors to enhancers, but if the transcription factor is absent in the cell or the signal that activates the transcription factor such as activation of the glucocorticoid receptor by stress-induced cortisol, then the gene will remain poorly expressed (narrow green arrow). However, later in development or in a specific tissue context or in response to an external trigger, the transcription factor is activated, and the gene is induced. This could be a transient response to a transient future signal, such as acute stress. Thus DNA methylation could program future rather than current responses to anticipated and unanticipated future encounters.

demethylated and upregulated when rats are food restricted and exposed to frustration stress (Pucci et al., 2016). Prenatal stress predisposes to binge eating (Keski-Rahkonen and Mustelin, 2016). Prenatal overexpression of CRH in late-pregnant female mice choroid plexus triggers high-circulating corticosterone, which mimics prenatal maternal stress (Schroeder et al., 2017). CRH-induced prenatal stress leads to downregulation of the three DNMTs in the female offspring hypothalamus, hypomethylation of a microRNA mir-1a which in turn causes deregulation of the melanocortin system through downregulation of Pax/Pa3 (Schroeder et al., 2017). Interestingly, the demethylation of mir-1a is normalized by a methyl-balanced diet during adolescence, which also protects from development of binge eating behavior. This provides an example for “lifestyle” developmentally timed interventions to reverse epigenetically programmed phenotypes. The crosstalk between stress and metabolic regulation is not limited to the brain, as discussed previously; the proximal GR negative regulator FKBP5 is hypermethylated in DNA from fat cells in obese women (Willmer et al., 2020).

EPIGENETIC PROCESSES MEDIATING EFFECTS OF STRESS DURING ADULTHOOD Although large dynamic changes in DNA methylation are known to occur during embryonal development and therefore exposures and experiences are anticipated to have larger effects during early life, epigenetic

modulations of the HPA axis, which might mediate behavioral changes, occur later in life as well. Corticotrophin-releasing hormone, which is secreted from the paraventricular nucleus in the hypothalamus integrates brain stress response pathways. Chronic release of CRH has been linked to stress-related disorders such as depression (De Kloet et al., 2005). Expression of CRH is regulated by a CRE (cAMP response element) in the 50 upstream region of the Crh gene, which binds the cAMP response element-binding protein (CREB) transcription factor only when it is unmethylated (Elliott et al., 2010). This region is demethylated and induced in mice that are exposed to 10 days of social defeat (Elliott et al., 2010), a model known to induce depressive-like symptoms in mice such as anhedonia and social avoidance. Interestingly, a subset of “resilient” mice that did not exhibit the depressive-like behaviors were not demethylated and did not show induced Crh expression illustrating that resilience could result from a heterogeneous epigenetic response to stress. Knockdown of Crh in hypothalamus PVN attenuates the depressive-like behaviors suggesting that CRH plays a causal role in mediating the impact of exposure to chronic stress on depressive-like behaviors at least in this mouse model (Elliott et al., 2010). The study points to the possibility that the HPA axis is responsive to epigenetic reprogramming even later in life. This has implications to our understanding of the long-term consequences of behavioral exposures later in life and the potential for behavioral intervention to reverse adverse epigenetic programming in adults. Chronic stress in rats exposed to 2 weeks of chronic variable mild stress

PERINATAL STRESS AND EPIGENETICS paradigm causes sex and brain-region-specific changes in methylation of Crh as well as sex specific changes in Crh mRNA and peptide; however, in this specific study, there was no obvious correlation between changes in methylation and expression (Sterrenburg et al., 2011).

DNA METHYLATION PROGRAMMING OF THE HPA AXIS IN RESPONSE TO ELS: A GENOMIC MECHANISM OF ADAPTATION DNA methylation, a covalent modification of DNA that is hard wired into its chemical structure, is a mechanism of genome plasticity that does not require sequence changes. DNA methylation emerged in the prokaryotic world as a mechanism for recognizing invading bacteriophages in the “restriction-modification” system (Arber and Linn, 1969), a tool-set of bacterial enzymes that cleave certain sequences and enzymes that methylate the same sequences and protect them from cleavage. The restriction modification system illustrates two principles. First, DNA methylation provides different identities to otherwise identical sequences. Second, it does so by altering the interaction of the sequences with proteins that recognize them, in this case restriction enzymes. Classic genetic studies focused on correlating changes in genetic sequences to trait or phenotypic changes. Animal studies discussed previously illustrated how stress early in life causes DNA methylation alterations in stress control mechanisms in the hippocampus, hypothalamus, and pituitary gland and that these define stable behavioral phenotypes such as stress responsivity, depression, and anxiety. The focus in neuroscience was obviously on brainrelated mechanisms, which guided the candidate genes and brain regions that were interrogated. The hypothalamus is obviously at the core of the HPA axis, and changes in DNA methylation of critical genes are obviously of high interest. However, with the obvious emerging interest in unraveling the role of HPA axis epigenetics in mediating the well-known effects of early life stress on human behavioral and neuropsychiatric disorders, and the fact that the brain is unavailable for obvious reason for DNA research in living humans, noninvasive samples from peripheral tissues were being increasingly used as discussed previously. Whereas peripheral sources such as saliva are a perfect choice for genetic studies since the genetic information is practically identical in brain and peripheral tissues, this is not the case for epigenetics. As discussed previously, epigenetics is the primary mechanism for generating cell-type identities (Razin and Szyf, 1984; Roadmap Epigenomics Consortium et al., 2015); it is therefore almost “sacrilegious” so to

137

speak to even assume that DNA methylation profiles in blood or saliva will be informative on the state of methylation of a gene in the hypothalamus or hippocampus. Indeed, the rationale for investigating peripheral blood DNA methylation as a “surrogate” of brain regionspecific methylation profiles has been almost uniformly a priori rejected. Nevertheless, several studies reviewed in this chapter and discussed previously investigated associations between DNA methylation in saliva and blood and early life adversity/stress at the level of candidate genes as well as genome-wide methylation association studies. If the large body of data examining candidate genes in blood has some validity, the question remains what is the relevance of a DNA methylation change in a brain-specific gene, in blood cells? An additional challenge is that changes in DNA methylation associated with early life stress are small and since DNA methylation is a binary signal, the small changes in methylation imply that a small fraction of blood cells had 100% of change in these specific sites. Suppose, indeed, the changes in DNA methylation in HPA candidate genes are so pervasive that they target not only the specialized HPA brain regions but also indiscriminately other systems. Why are so few cells affected? Perhaps there is a very small subset of cells that is responsive to early life stress in blood. Even though changes in DNA methylation are pervasive across tissues, they only target a subset of cells within a given tissue. Is it alternatively possible that the changes in methylation occur stochastically across the population of cells as a result of a small change in the DNA methylation/demethylation equilibrium? White blood cells are derived from dividing stem cells. It is reasonable to assume that if an early life stress affected DNA methylation, it affected the stem cell population. Maintenance methylation is believed, however, to faithfully replicate DNA methylation profiles across cell divisions. We have no known mechanism that explains a predictable but stochastic mechanism for altering the accuracy of copying of DNA methylation at specific sites during DNA replication. What is the biological role of changes in DNA methylation of brain-specific genes in peripheral tissues? The system-wide DNA methylation changes in brain-specific genes in peripheral tissues might not necessarily predispose system-wide changes in expression. Changes in DNA methylation are insufficient to turn on gene expression. Gene expression requires the presence of transcription factors that might be present only in specific brain regions. It is possible that a system-wide change in DNA methylation triggered by stress will only be effective in a specific brain region and under a specific context (Fig. 8.3). Changes in DNA methylation determine the set point for expression but not expression. If this is the case, then DNA methylation changes of HPA-

138

M. SZYF

specific genes in blood might be informative about their stable reprogramming in the brain. Agnostic genome-wide DNA methylation analyses have expanded our understanding of the effect of early life stress beyond the usual suspects. What emerges in several studies is a clear immune/inflammatory signature as well as other signatures as, for example, in the Quebec 1998 ice storm study (Cao-Lei et al., 2014). These data point out to a broader system-wide understanding of the impact of early life stress. Psychological stress has been linked with immune function in a large body of data (Segerstrom and Miller, 2004). The impact of early life stress on long-term programming of the immune system has been noted in previous studies that examined immune function and gene expression profiles and were defined as an “ELA immune phenotype” (Elwenspoek et al., 2017). Early life stress in mouse affects immune reactivity (natural killer), NK-cell activity, and T-cell mitogenesis in adulthood (Neveu et al., 1994). In a rat model of early life stress induced by maternal separation maternal separation led to increase in activation of renal neutrophils, CD44 + cells, toll-like receptor 4 (TLR4) IL1 beta as well as induction of 17 inflammatory genes in the kidney in response to LPS stimulation, whereby no changes were observed in control rats (De Miguel et al., 2018). In the nonhuman primate rhesus macaque, maternal deprivation by rearing the newborns in a nursery affected cellular immunity functions such as lower proportions of CD8 cells and lower NK-cell activity. The effects could not be reversed by behavioral rehabilitation, suggesting a stable lasting change in cellular immunity in response to early life adversity (Lubach et al., 1995). In humans, in the European Prospective Investigation into Cancer and Nutrition in Norfolk, adverse experiences in childhood were positively associated with lymphocyte counts in adults aged 40–80 (Surtees et al., 2003). Childhood adversity leads to increased plasma IL-6 in women with breast cancer (Witek Janusek et al., 2013). Adolescents (n ¼ 30) with a history of childhood adversity had elevated immune activation and proinflammatory profile compared to controls that included elevated T-cell activation markers (CD3 + CD4 + CD25 + and CD3 + CD69 +) and senescent T cells (CD8 + CD28  and CD4 + CD28 ), reduction in NK (CD3  CD56 +) and NK T cells (CD3 + CD56 +), higher levels of IL-2, IL-4, IFN-gamma, and IL-17 in response to stimulation and elevated MAPK ERK and NF-kappaB signaling (Do Prado et al., 2017). Using a pilot sample of 114 people from the Health and Retirement Study showed an association between a composite score of three proinflammatory genes (PTSG2, IL1B, and IL8) and childhood trauma (Levine et al., 2015). Although in this study there was no association with childhood socioeconomic status, childhood trauma exacerbated the effect of SES on inflammatory gene

expression (Levine et al., 2015). Transcriptional profiling of cord blood of newborns to low socioeconomic position mothers revealed a profile of higher immune activation and downregulation of transcriptional pathways involved in myeloid differentiation (Miller et al., 2017). Early life socioeconomic stress was associated with upregulation of CREB/ATF and NF-kappaB responsive genes and greater stimulated secretion of proinflammatory cytokine IL6, suggesting that ELS programs a heighted inflammatory responsivity that continues into late adulthood (Miller et al., 2009). Examination of changes in the transcriptome in response to acute stress in monocytes in healthy adults (n ¼ 30) with a history of early childhood adversity, and a control group (n ¼ 30) revealed differential gene expression of cytokine and chemokine activity suggesting that early life stress programs a different set point for the response of immune and inflammatory genes to acute stress at later times in life (Schwaiger et al., 2016). A recent study showed that early life exposure to stress hormone results in reduced CD8 T cell response in adults, which was associated with decreased CD8 T cell function and reduced antitumor and antibacterial responses (Hong et al., 2020). DNA methylation alterations in peripheral white blood cells are therefore proposed here to play a role in setting up an immune/inflammatory phenotype in response to early life stress signals in the HPA axis. The HPA axis response to early life stress sets systemic lifelong phenotypic alterations that involve critical homeostatic physiological systems including the immune/inflammatory systems and as discussed previously the metabolic system (Schroeder et al., 2017) (Fig. 8.4). For example, differentially methylated genes associated with early life stress in the ice storm study included genes involved in insulin response and a mediation analysis suggested that DNA methylation is mediating the effects of early life stress on BMI (CaoLei et al., 2016a). What is the functional role of this reprogramming of the immune system by perinatal stress? The epigenetic changes in the immune system might be an adverse consequence of the HPA axis modulation and one might propose a similar role for the DNA methylation changes in the hippocampus and hypothalamus. Activation of the HPA axis releases stress hormones and adrenergic signaling that could potentially disrupt epigenetic programming at a critical stage of development. Alternatively, perinatal stress is a cue that is interpreted by HPA epigenetic programming to organize a long-term adaptation of the genome to the environment at early life, modifying the phenotype to fit with the anticipated lifelong environment (Fig. 8.4). The response of DNA methylation to environment is thus suggested to be an organized organismal response that is mediated by well-defined signaling cascades that coordinate a multisystem response to

PERINATAL STRESS AND EPIGENETICS

139

Stress Early life environment HPA GC

GC/Nr3c1

Hippocampus/hypothalamus/ WBC pituitary gland M

M

MM

M

M

Fat cells/muscle/cardio

MM

M

M

MM

Developmen t Behavior

Immune/inflammation

Metabolism

Fig. 8.4. Putative mechanism for system-wide epigenetic reprogramming by ELS. Early life stress activates HPA and glucocorticoid release systemically. Glucocorticoid (GC) bound glucocorticoid receptors (GR) epigenetically reprogram target genes in different tissues modulating DNA methylation, gene expression, and phenotypic physiological function including the immune system and the cardiovascular system adapting the organism to the environment anticipated by exposure to early life stress. Crosstalk between the immune system and the brain could in turn reprogram behavior.

environmental cues. The HPA axis which concurrently impacts behavioral, immune/inflammatory, and metabolic physiology is well positioned to be epigenetically programmed to adapt its activity to environmental variations that are predictive of a lifelong environment. If this is the case, then peripheral DNA methylation changes are not just a surrogate of central DNA methylation changes but are of physiological relevance on their own for the long-term functioning of the entire system. The main contribution of DNA methylation is by translating transient environmental exposures during the perinatal period to lifelong new physiological set points. DNA methylation changes in the immune system might in turn also program brain function. Thus changes in programming of cytokine expression by activation of the HPA might mediate changes in brain function and behavior (Raber et al., 1998; Bilbo and Schwarz, 2012). Experimental studies in laboratory rodents suggest that sympathetic nervous system activation following early-life stress exposure causes a shift in the profile of innate immune cells, with an increase in proinflammatory monocytes (Mondelli and Vernon, 2019). In turn, these cells traffic to the brain and influence neural circuitry, which manifests as increased anxiety and other relevant-behavioral phenotypes. Cytokines secreted by dedicated immune cells were long being postulated to play role in neural development (Garay and McAllister, 2010; Bilbo and Schwarz, 2012). Several lines of evidence are emerging that link innate

immune states to neurodegenerative and psychiatric diseases (Novellino et al., 2020). Chronic inflammation has been associated with incident dementia, and a recent study showed that the methylation profile of the inflammatory C-reactive protein (CRP) is a reliable marker of chronic inflammation and a negative correlation was revealed between CRP methylation score and cognitive ability (Stevenson et al., 2020). Although the focus of research in the field of early life experience has been on adversity and disease, if indeed the response of the HPA to stress is adaptive, enrichment of early life experience should provide benefits later in life. As discussed previously, augmented maternal care in the first postnatal week improves stress resilience and memory (Vogel Ciernia et al., 2018). It is worthwhile to further this hypothesis since it could have important implications on shifting the focus of intervention and policy from prevention of adversity and adverse effects to enrichment and improvement of well-being.

DYNAMIC EPIGENETIC PROGRAMMING IN RESPONSE TO EARLY LIFE EXPERIENCE: SETTING UP FUTURE GENE EXPRESSION TRAJECTORIES Epigenetic programming in response to early life experience is a long-term memory in the genome of a shortlived experience. Initial studies suggested that epigenetic

140

M. SZYF

memory is maintained by methyl moieties that are introduced at the perinatal age to specific positions and remain through life (Weaver et al., 2004). This can explain how a gene expression program that was triggered by a transient signal is preserved. However, later data suggest that epigenetic memory of early life experience could also be dynamic, whereby early life experience alters the developmental trajectory of the normal evolution of the DNA methylation pattern. Thus a DNA methylation pattern introduced early in life does not remain stable but changes the course of the developmental evolution of the DNA methylation pattern so that the profile that emerges in adulthood is different from the early profile of the response to adversity but is also different from controls who were not exposed to adversity (Massart et al., 2016). Dynamic DNA methylation memories can explain how early life experiences have phenotypic consequences only later during adulthood. It should also be noted that DNA methylation differences are insufficient in many cases to trigger differences in gene expression (Fig. 8.3). Thus although DNA methylation might be necessary for transcription or enhancer function, it is insufficient. Transcription or activation of enhancers requires presence of appropriate signals acting through transcription factors that might not be present at this time but only in response to certain external triggers or later in development (Fig. 8.3). Thus changes in DNA methylation define future gene expression programs by carving a different set point for future triggers and inducers. For example, demethylation of a glucocorticoid response element would prime the gene for induction only when there is stress and a surge of glucocorticoids (Thomassin et al., 2001), and glucocorticoid-induced demethylation in progenitor hippocampal neurons change the range of response to a second wave of glucocorticoid exposure (Provencal et al., 2019).

WHAT ARE THE MECHANISMS THAT MEDIATE BETWEEN EXPERIENCE AND DNA METHYLATION ALTERATIONS IN THE HPA AXIS AND SOMATIC TISSUES AND WHAT ARE THEIR FUNCTIONAL ROLES? Stress is perceived by the brain; therefore, it is expected that DNA methylation differences in newborns exposed to an early stressful environment will be detected only in the brain. How can early life stress produce a systemwide epigenetic response? One possible mechanism could involve the glucocorticoid stress hormone which has receptors (GR) in almost all tissues and cell types including sperm (Kaufmann et al., 1992). An infant exposed to a

stressful experience will respond by release of glucocorticoid hormone which will act on receptors in different tissues; binding of GR to targets in different tissues will result in epigenetic reprogramming. A fetus is exposed to maternal GC; however, this is regulated by the glucocorticoid metabolizing enzyme 11b-hydroxysteroid dehydrogenase type 2 (11b-HSD2), which protects the fetus from high maternal glucocorticoids but is potentially epigenetically regulated (Seckl, 2004; Holmes et al., 2006; Drake et al., 2012; Marsit et al., 2012; Togher et al., 2014; Paquette et al., 2015). Since glucocorticoids are well established to act on metabolic (Vegiopoulos and Herzig, 2007), immune (Oppong and Cato, 2015), and brain pathways (Fietta and Fietta, 2007), this could provide an explanation for the methylome and phenotypic effects seen in children exposed to early life stress. In support of this hypothesis, perinatal exposure of guinea pig fetuses to maternal dexamethasone resulted in genome-wide changes in promoter methylation in the hippocampus as well as multigenerational changes in DNA methylation in different organs (Crudo et al., 2012, 2013a,b). More recently as discussed previously, Provencal et al. (2019) demonstrated that exposure of progenitor hippocampal neurons to glucocorticoids alters the expression of a large number of CG positions in the genome. If this hypothesis is true, the glucocorticoid receptor nr3c1 should be important for the evolution of DNA methylation during development. In accordance with this hypothesis, a mouse knockout of one copy of nr3c1 in the fetuses but not the mothers resulted in sexdependent changes in DNA methylation across many loci in the fetal placenta (Schmidt et al., 2019). DNA methylation changed in both males and females, however, in the opposite direction. Differences in methylation between male and females who had one copy of nr3c1 were dramatically larger than in wild type, a sex by genotype effect on DNA methylation. This study supports the idea that nr3c1 plays a role in shaping the DNA methylation profile in peripheral tissues and positions glucocorticoid hormone and the receptor as a possible mediator of the system-wide effect of maternal stress on offspring DNA methylation (Schmidt et al., 2019). Glucocorticoid hormone is not the only system-wide signal that can mediate system-wide epigenetic programming, there are obviously other neuroendocrine pathways that can be activated by exposures such an insulin and insulin receptor, oxytocin and oxytocin receptor, norepinephrine, and adrenergic receptors. microRNAs are known to circulate peripherally and might be delivering epigenetic signals across the body and to the germline (Rodgers et al., 2013). Future studies are needed to test these hypotheses.

PERINATAL STRESS AND EPIGENETICS

SUMMARY AND PERSPECTIVE Evidence emerging in the last 15 years supports the idea that early life experience is epigenetically programming stress-response genes in the central nodes of the HPA axis, the hippocampus, and hypothalamus, thus defining lifelong trajectories of stress responsivity. It is also becoming clear that changes in DNA methylation in response to early life stress are not limited to just usual suspects, but that many genes and genome regions are involved, implying a more comprehensive impact than just on HPA regulatory genes. This is consistent with the idea that early life stress affects multiple physiological systems including the metabolic and immune system. Although animal data provide some evidence for causation, many unknowns remain that need to be addressed in animal models to solidify the role of DNA methylation in embedding in the genome the long-term impact of early life stress experience. First, we need to have a developmental time course of the methylome response to perinatal stress in specific brain regions associated with HPA at high resolution using the same stress paradigm. Compiling a molecular map of stress response requires that all components are mapped using the same experimental paradigm. These studies should also provide a map of the functional gene pathways involved and should examine whether these pathways provide a basis for deciphering the potential impact of stress on other physiological systems. The data that we currently have is spotty and derived from different experimental perinatal stress paradigms. Second, the brain is a complex tissue with multiple cell types and numerous neuronal circuitries. Even in animal studies, differences in DNA methylation are small, suggesting that they are derived from a small number of neurons or glial cells. Both cell type- and single cell-methylation and transcriptional data, as well as anatomical positioning of the cells within circuitries, are required to comprehend how DNA methylation changes are associated with structural, functional features of the developing hypothalamus and hippocampus. Third, the emerging picture of an impact of stress on peripheral physiological systems such as metabolism and immunity requires performing in parallel developmental mapping of the evolution of DNA methylation profiles in response to early life stress in peripheral white blood cells and other metabolically important tissues and cell types such as liver, fat cells, heart, muscle, and others. Fourth, although animal studies both past and those proposed here test a causal link between perinatal stress and differential DNA methylation, they do not address the causal link between these changes and phenotypes. Previous studies used epigenetic drugs to demonstrate that phenotypic

141

changes were driven by epigenetic changes (Weaver et al., 2004, 2005, 2006); however, these studies did not test the role of specific DNA methylation alterations since the drugs target multiple DNA methylation marks. Epigenetic editing using CRISPR-Cas targeting of DNMT3A, a de novo methyltransferase might provide a possible line of evidence for the causal role of specific methylation marks altered by perinatal stress gene expression and adult behavioral phenotype. This approach has been recently used to test a causal relationship between addiction-specific DNA methylation difference within the IRX2 gene body and CTCF protein binding (Vaillancourt et al., 2020). It should be noted that currently epigenetic editing has its own confounding issues such as off target methylation (Galonska et al., 2018) and low methylation efficiency in vivo. Fifth, several of the DNA methylation differences might fall outside commonly recognized genes or their 5’ regions, but nevertheless they might be critical for long-distance cis and trans regulatory elements, which could be deciphered using 3C chromatin conformation mapping (Galonska et al., 2018). Lastly, a standard attempt to discover the regulatory function of a DNA methylation change is to determine its effect on steady-state mRNA levels by quantifying mRNA levels. However, many of the important regulatory genes are expressed only in response to physiological or experiential, environmental time- and context-limited signals as discussed previously (Fig. 8.3). DNA methylation might regulate this induced response rather than steady-state mRNA levels. It is especially relevant for stress regulation and the transitory elevation in stress hormone levels as well as other regulated processes in immunity and metabolism as discussed previously (Provencal et al., 2019). The idea that DNA methylation programming early in life might cause gene expression changes later in life when and if the appropriate signals are present is critical for understanding the impact of early life experience on adult phenotype but is often sidestepped in comparative analyses of expression and DNA methylation. We need to examine the impact of DNA methylation on gene expression under conditions of a challenge such as stress, immune challenge, or metabolic challenge as well as steady-state conditions. Human studies are more challenging than animal research. The main problem is the limited access to the main brain regions controlling the HPA axis such as the hippocampus and hypothalamus. Many of the studies are limited by sample number and naturally confounded by the complexity of human genetic variation and environmental exposures. Postmortem studies provide access to brain material but cannot recapitulate the dynamic trajectories of living people. Human studies reluctantly

142

M. SZYF

used peripheral tissues as proxies of the brain. These were criticized for small numbers and small differences in DNA methylation associated with early life stress and behavioral disorders. Nevertheless, several studies reviewed here showed a consistent signal in candidate HPA-regulated genes, although these associations were questioned in a recent analysis of a large cohort as discussed previously. We need to better understand the impact of the small differences in DNA methylation observed in these studies. Are the methylation differences limited to a specific subset of cells that bear the epigenetic memory or are these maintained stochastically in the heterogenous white blood cell population? An important question that needs to be addressed is how DNA methylation changes induced by early life stress and its phenotypic consequences interact with later life events and exposures that will also in turn modify the epigenome. Animal studies have shown as discussed previously that adult stress can also induce long-term epigenetic changes in HPA (Elliott et al., 2010) and that the DNA methylation evolves during childhood and adolescence (Massart et al., 2016) and is changing with aging (Horvath, 2013). These interactions between the epigenetic matrix triggered by early stress and later events that modify the epigenome might result in reversal of early life adverse events or their aggravation. Understanding the dynamic evolution of the epigenetic matrix laid down by HPA responses early in life through a life course of dynamic environments is required to be able to design possible intervention and prevention strategies. Collecting such data will necessitate longitudinal studies with more comprehensive documentation of exposures and experiences, which might be accomplished using “live on-line data collection” combined with current machine learning methods.

ACKNOWLEDGMENT Studies from MS lab were funded by a grant from the Canadian Institute of Health Research PJT-159583.

REFERENCES Arber W, Linn S (1969). DNA modification and restriction. Annu Rev Biochem 38: 467–500. Bacon ER, Mishra A, Wang Y et al. (2019). Neuroendocrine aging precedes perimenopause and is regulated by DNA methylation. Neurobiol Aging 74: 213–224. Barski A, Cuddapah S, Cui K et al. (2007). High-resolution profiling of histone methylations in the human genome. Cell 129: 823–837. Baubec T, Ivanek R, Lienert F et al. (2013). Methylationdependent and -independent genomic targeting principles of the MBD protein family. Cell 153: 480–492.

Baylin SB (1992). Abnormal regional hypermethylation in cancer cells. AIDS Res Hum Retrovir 8: 811–820. Baylin SB, Herman JG, Graff JR et al. (1998). Alterations in DNA methylation: a fundamental aspect of neoplasia. Adv Cancer Res 72: 141–196. Benvenisty N, Mencher D, Meyuhas O et al. (1985a). Sequential changes in DNA methylation patterns of the rat phosphoenolpyruvate carboxykinase gene during development. Proc Natl Acad Sci U S A 82: 267–271. Benvenisty N, Razin A, Reshef L (1985b). Developmental changes in the methylation pattern, chromatin conformation and expression of the rat phosphoenolpyruvate carboxykinase gene. Prog Clin Biol Res 198: 185–199. Bernstein BE, Mikkelsen TS, Xie X et al. (2006). A bivalent chromatin structure marks key developmental genes in embryonic stem cells. Cell 125: 315–326. Bierer LM, Bader HN, Daskalakis NP et al. (2020). Intergenerational effects of maternal holocaust exposure on FKBP5 methylation. Am J Psychiatry 177: 744–753. https://doi.org/10.1176/appi.ajp.2019.19060618. Bilbo SD, Schwarz JM (2012). The immune system and developmental programming of brain and behavior. Front Neuroendocrinol 33: 267–286. Bockm€ uhl Y, Patchev AV, Madejska A et al. (2015). Methylation at the CpG island shore region upregulates Nr3c1 promoter activity after early-life stress. Epigenetics 10: 247–257. Borghol N, Suderman M, Mcardle W et al. (2012). Associations with early-life socio-economic position in adult DNA methylation. Int J Epidemiol 41: 62–74. Bostick M, Kim JK, Esteve PO et al. (2007). UHRF1 plays a role in maintaining DNA methylation in mammalian cells. Science 317: 1760–1764. Brownell JE, Allis CD (1996). Special HATs for special occasions: linking histone acetylation to chromatin assembly and gene activation. Curr Opin Genet Dev 6: 176–184. Brownell JE, Allis CD (2021). HAT discovery: heading toward an elusive goal with a key biological assist. Biochim Biophys Acta, Gene Regul Mech 1864: 194605. https://doi.org/10.1016/j.bbagrm.2020.194605. Cao-Lei L, Massart R, Suderman MJ et al. (2014). DNA methylation signatures triggered by prenatal maternal stress exposure to a natural disaster: project ice storm. PLoS One 9: e107653. Cao-Lei L, Dancause KN, Elgbeili G et al. (2016a). Pregnant women’s cognitive appraisal of a natural disaster affects their children’s BMI and central adiposity via DNA methylation: project ice storm. Early Hum Dev 103: 189–192. Cao-Lei L, Veru F, Elgbeili G et al. (2016b). DNA methylation mediates the effect of exposure to prenatal maternal stress on cytokine production in children at age 13(1/2) years: project ice storm. Clin Epigenetics 8: 54. Cao-Lei L, Dancause KN, Elgbeili G et al. (2018). DNA methylation mediates the effect of maternal cognitive appraisal of a disaster in pregnancy on the child’s C-peptide secretion in adolescence: project ice storm. PLoS One 13: e0192199.

PERINATAL STRESS AND EPIGENETICS Cedar H, Razin A (1990). DNA methylation and development. Biochim Biophys Acta 1049: 1–8. Cervoni N, Sang-Beom S, Chakravarti D et al. (2001). A novel regulatory role for Set/TAF-Iß oncoprotein integrating histone hypoacetylation and DNA hypermethylation in transcriptional silencing. submitted for publication. Chahrour M, Jung SY, Shaw C et al. (2008). MeCP2, a key contributor to neurological disease, activates and represses transcription. Science 320: 1224–1229. Champagne FA, Weaver IC, Diorio J et al. (2006). Maternal care associated with methylation of the estrogen receptor-alpha1b promoter and estrogen receptor-alpha expression in the medial preoptic area of female offspring. Endocrinology 147: 2909–2915. Chen WG, Chang Q, Lin Y et al. (2003). Derepression of BDNF transcription involves calcium-dependent phosphorylation of MeCP2. Science 302: 885–889. Cicchetti D, Handley ED (2017). Methylation of the glucocorticoid receptor gene, nuclear receptor subfamily 3, group C, member 1 (NR3C1), in maltreated and nonmaltreated children: associations with behavioral undercontrol, emotional lability/negativity, and externalizing and internalizing symptoms. Dev Psychopathol 29: 1795–1806. Cisternas CD, Cortes LR, Bruggeman EC et al. (2019). Developmental changes and sex differences in DNA methylation and demethylation in hypothalamic regions of the mouse brain. Epigenetics 1–13. Comb M, Goodman HM (1990). CpG methylation inhibits proenkephalin gene expression and binding of the transcription factor AP-2. Nucleic Acids Res 18: 3975–3982. Creyghton MP, Cheng AW, Welstead GG et al. (2010). Histone H3K27ac separates active from poised enhancers and predicts developmental state. Proc Natl Acad Sci U S A 107: 21931–21936. Crudo A, Petropoulos S, Moisiadis VG et al. (2012). Prenatal synthetic glucocorticoid treatment changes DNA methylation States in male organ systems: multigenerational effects. Endocrinology 153: 3269–3283. Crudo A, Petropoulos S, Suderman M et al. (2013a). Effects of antenatal synthetic glucocorticoid on glucocorticoid receptor binding, DNA methylation, and genome-wide mRNA levels in the fetal male hippocampus. Endocrinology 154: 4170–4181. Crudo A, Suderman M, Moisiadis VG et al. (2013b). Glucocorticoid programming of the fetal male hippocampal epigenome. Endocrinology 154: 1168–1180. D’alessio AC, Szyf M (2006). Epigenetic tete-a-tete: the bilateral relationship between chromatin modifications and DNA methylation. Biochem Cell Biol 84: 463–476. De Kloet ER, Joels M, Holsboer F (2005). Stress and the brain: from adaptation to disease. Nat Rev Neurosci 6: 463–475. De Miguel C, Obi IE, Ho DH et al. (2018). Early life stress induces immune priming in kidneys of adult male rats. Am J Physiol Ren Physiol 314: F343–F355. Detich N, Bovenzi V, Szyf M (2003). Valproate induces replication-independent active DNA demethylation. J Biol Chem 278: 27586–27592.

143

Do Prado CH, Grassi-Oliveira R, Daruy-Filho L et al. (2017). Evidence for immune activation and resistance to glucocorticoids following childhood maltreatment in adolescents without psychopathology. Neuropsychopharmacology 42: 2272–2282. Drake AJ, McPherson RC, Godfrey KM et al. (2012). An unbalanced maternal diet in pregnancy associates with offspring epigenetic changes in genes controlling glucocorticoid action and foetal growth. Clin Endocrinol 77: 808–815. Elliott E, Ezra-Nevo G, Regev L et al. (2010). Resilience to social stress coincides with functional DNA methylation of the Crf gene in adult mice. Nat Neurosci 13: 1351–1353. Elwenspoek MMC, Kuehn A, Muller CP et al. (2017). The effects of early life adversity on the immune system. Psychoneuroendocrinology 82: 140–154. Engdahl E, Alavian-Ghavanini A, Forsell Y et al. (2020). Childhood adversity increases methylation in the GRIN2B gene. J Psychiatr Res 132: 38–43. Etchegaray JP, Yang X, Debruyne JP et al. (2006). The polycomb group protein EZH2 is required for mammalian circadian clock function. J Biol Chem 281: 21209–21215. Fietta P, Fietta P (2007). Glucocorticoids and brain functions. Riv Biol 100: 403–418. Fiorito G, Polidoro S, Dugue PA et al. (2017). Social adversity and epigenetic aging: a multi-cohort study on socioeconomic differences in peripheral blood DNA methylation. Sci Rep 7: 16266. Francis D, Diorio J, Liu D et al. (1999). Nongenomic transmission across generations of maternal behavior and stress responses in the rat. Science 286: 1155–1158. Frank M, Kemler R (2002). Protocadherins. Curr Opin Cell Biol 14: 557–562. Galonska C, Charlton J, Mattei AL et al. (2018). Genome-wide tracking of dCas9-methyltransferase footprints. Nat Commun 9: 597. Gangisetty O, Sinha R, Sarkar DK (2019). Hypermethylation of proopiomelanocortin and period 2 genes in blood are associated with greater subjective and behavioral motivation for alcohol in humans. Alcohol Clin Exp Res 43: 212–220. Garay PA, McAllister AK (2010). Novel roles for immune molecules in neural development: implications for neurodevelopmental disorders. Front Synaptic Neurosci 2: 136. Gouin JP, Zhou QQ, Booij L et al. (2017). Associations among oxytocin receptor gene (OXTR) DNA methylation in adulthood, exposure to early life adversity, and childhood trajectories of anxiousness. Sci Rep 7: 7446. Gruenbaum Y, Stein R, Cedar H et al. (1981). Methylation of CpG sequences in eukaryotic DNA. FEBS Lett 124: 67–71. Gruenbaum Y, Cedar H, Razin A (1982). Substrate and sequence specificity of a eukaryotic DNA methylase. Nature 295: 620–622. Hahn MA, Wu X, Li AX et al. (2011). Relationship between gene body DNA methylation and intragenic H3K9me3 and H3K36me3 chromatin marks. PLoS One 6: e18844. Harms MB, Birn R, Provencal N et al. (2017). Early life stress, FK506 binding protein 5 gene (FKBP5) methylation, and

144

M. SZYF

inhibition-related prefrontal function: a prospective longitudinal study. Dev Psychopathol 29: 1895–1903. He YF, Li BZ, Li Z et al. (2011). Tet-mediated formation of 5-carboxylcytosine and its excision by TDG in mammalian DNA. Science 333: 1303–1307. He Y, Vinkers CH, Houtepen LC et al. (2018). Childhood adversity is associated with increased KITLG methylation in healthy individuals but not in bipolar disorder patients. Front Psych 9: 743. Hebbar PB, Archer TK (2003). Chromatin remodeling by nuclear receptors. Chromosoma 111: 495–504. Heidt T, Sager HB, Courties G et al. (2014). Chronic variable stress activates hematopoietic stem cells. Nat Med 20: 754–758. Heintzman ND, Stuart RK, Hon G et al. (2007). Distinct and predictive chromatin signatures of transcriptional promoters and enhancers in the human genome. Nat Genet 39: 311–318. Hendrich B, Bird A (1998). Identification and characterization of a family of mammalian methyl-CpG binding proteins. Mol Cell Biol 18: 6538–6547. Holliday R, Pugh JE (1975). DNA modification mechanisms and gene activity during development. Science 187: 226–232. Holmes MC, Abrahamsen CT, French KL et al. (2006). The mother or the fetus? 11beta-hydroxysteroid dehydrogenase type 2 null mice provide evidence for direct fetal programming of behavior by endogenous glucocorticoids. J Neurosci 26: 3840–3844. Hong JY, Lim J, Carvalho F et al. (2020). Long-Term Programming of CD8 T Cell Immunity by Perinatal Exposure to Glucocorticoids. Cell 180: 847–861 e15. Hopf FW, Bonci A (2010). Dnmt3a: addiction’s molecular forget-me-not? Nat Neurosci 13: 1041–1043. Horvath S (2013). DNA methylation age of human tissues and cell types. Genome Biol 14: R115. Hotchkiss RD (1948). The quantitative separation of purines, pyrimidines, and nucleosides by paper chromatography. J Biol Chem 175: 315–332. Iurlaro M, Mcinroy GR, Burgess HE et al. (2016). In vivo genome-wide profiling reveals a tissue-specific role for 5-formylcytosine. Genome Biol 17: 141. Jablonka E, Lamb MJ (2002). The changing concept of epigenetics. Ann N Y Acad Sci 981: 82–96. Jenuwein T, Allis CD (2001). Translating the histone code. Science 293: 1074–1080. Karemaker ID, Baubec T (2020). DNA methyltransferases hitchhiking on chromatin. Swiss Med Wkly 150: w20329. Kaufmann SH, Wright WW, Okret S et al. (1992). Evidence that rodent epididymal sperm contain the Mr approximately 94,000 glucocorticoid receptor but lack the Mr approximately 90,000 heat shock protein. Endocrinology 130: 3074–3084. Keski-Rahkonen A, Mustelin L (2016). Epidemiology of eating disorders in Europe: prevalence, incidence, comorbidity, course, consequences, and risk factors. Curr Opin Psychiatry 29: 340–345.

King S, Barr RG, Brunet A et al. (2000). The ice storm: an opportunity to study the effects of prenatal stress on the baby and the mother. Sante Ment Que 25: 163–185. Klengel T, Mehta D, Anacker C et al. (2013). Allele-specific FKBP5 DNA demethylation mediates gene-childhood trauma interactions. Nat Neurosci 16: 33–41. Kolodkin MH, Auger AP (2011). Sex difference in the expression of DNA methyltransferase 3a in the rat amygdala during development. J Neuroendocrinol 23: 577–583. Kooistra SM, Helin K (2012). Molecular mechanisms and potential functions of histone demethylases. Nat Rev Mol Cell Biol 13: 297–311. Kraaijenvanger EJ, He Y, Spencer H et al. (2019). Epigenetic variability in the human oxytocin receptor (OXTR) gene: a possible pathway from early life experiences to psychopathologies. Neurosci Biobehav Rev 96: 127–142. Kriaucionis S, Heintz N (2009). The nuclear DNA base 5-hydroxymethylcytosine is present in Purkinje neurons and the brain. Science 324: 929–930. Roadmap Epigenomics Consortium, Kundaje A, Meuleman W et al. (2015). Integrative analysis of 111 reference human epigenomes. Nature 518: 317–330. Kuo MH, Brownell JE, Sobel RE et al. (1996). Transcriptionlinked acetylation by Gcn5p of histones H3 and H4 at specific lysines. Nature 383: 269–272. Lal G, Zhang N, Van Der Touw W et al. (2009). Epigenetic regulation of Foxp3 expression in regulatory T cells by DNA methylation. J Immunol 182: 259–273. Lam LL, Emberly E, Fraser HB et al. (2012). Factors underlying variable DNA methylation in a human community cohort. Proc Natl Acad Sci U S A 109 (Suppl. 2): 17253–17260. Laplant Q, Vialou V, Covington 3rd HE et al. (2010). Dnmt3a regulates emotional behavior and spine plasticity in the nucleus accumbens. Nat Neurosci 13: 1137–1143. Lax E, Do-Carmo S, Enuka Y et al. (2018). Methyl-CpG binding domain 2 (Mbd2) deficiency causes cognitive, social and emotional functional deficits. bioRxiv 247197. https://doi.org/10.1101/247197. Lee RS, Tamashiro KL, Yang X et al. (2010). Chronic corticosterone exposure increases expression and decreases deoxyribonucleic acid methylation of FKBP5 in mice. Endocrinology 151: 4332–4343. Levine ME, Cole SW, Weir DR et al. (2015). Childhood and later life stressors and increased inflammatory gene expression at older ages. Soc Sci Med 130: 16–22. Levine ME, Lu AT, Chen BH et al. (2016). Menopause accelerates biological aging. Proc Natl Acad Sci U S A 113: 9327–9332. Lister R, Pelizzola M, Dowen RH et al. (2009). Human DNA methylomes at base resolution show widespread epigenomic differences. Nature 462: 315–322. Lister R, Mukamel EA, Nery JR et al. (2013). Global epigenomic reconfiguration during mammalian brain development. Science 341: 1237905. Liu D, Diorio J, Tannenbaum B et al. (1997). Maternal care, hippocampal glucocorticoid receptors, and hypothalamic-

PERINATAL STRESS AND EPIGENETICS pituitary-adrenal responses to stress. Science 277: 1659–1662. Lomniczi A, Wright H, Ojeda SR (2015). Epigenetic regulation of female puberty. Front Neuroendocrinol 36: 90–107. Lu F, Liu Y, Jiang L et al. (2014). Role of Tet proteins in enhancer activity and telomere elongation. Genes Dev 28: 2103–2119. Lubach GR, Coe CL, Ershler WB (1995). Effects of early rearing environment on immune responses of infant rhesus monkeys. Brain Behav Immun 9: 31–46. Luo L, Yao Z, Ye J et al. (2017). Identification of differential genomic DNA Methylation in the hypothalamus of pubertal rat using reduced representation Bisulfite sequencing. Reprod Biol Endocrinol 15: 81. Maiti A, Drohat AC (2011). Thymine DNA glycosylase can rapidly excise 5-formylcytosine and 5-carboxylcytosine: potential implications for active demethylation of CpG sites. J Biol Chem 286: 35334–35338. Manteuffel-Cymborowska M (1999). Nuclear receptors, their coactivators and modulation of transcription. Acta Biochim Pol 46: 77–89. Marsit CJ, Maccani MA, Padbury JF et al. (2012). Placental 11-beta hydroxysteroid dehydrogenase methylation is associated with newborn growth and a measure of neurobehavioral outcome. PLoS One 7: e33794. Marzi SJ, Sugden K, Arseneault L et al. (2018). Analysis of DNA methylation in young people: limited evidence for an association between victimization stress and epigenetic variation in blood. Am J Psychiatry 175: 517–529. Massart R, Nemoda Z, Suderman MJ et al. (2016). Early life adversity alters normal sex-dependent developmental dynamics of DNA methylation. Dev Psychopathol 28: 1259–1272. Massart R, Suderman M, Mongrain V et al. (2017a). DNA methylation and transcription onset in the brain. Epigenomics 9: 797–809. Massart R, Suderman MJ, Nemoda Z et al. (2017b). The signature of maternal social rank in placenta deoxyribonucleic acid methylation profiles in rhesus monkeys. Child Dev 88: 900–918. McEwen BS, Bowles NP, Gray JD et al. (2015). Mechanisms of stress in the brain. Nat Neurosci 18: 1353–1363. McGowan PO, Sasaki A, Huang TC et al. (2008). Promoterwide hypermethylation of the ribosomal RNA gene promoter in the suicide brain. PLoS One 3: e2085. McGowan PO, Sasaki A, D’alessio AC et al. (2009). Epigenetic regulation of the glucocorticoid receptor in human brain associates with childhood abuse. Nat Neurosci 12: 342–348. McGowan PO, Suderman M, Sasaki A et al. (2011). Broad epigenetic signature of maternal care in the brain of adult rats. PLoS One 6: e14739. Meehan RR, Lewis JD, Bird AP (1992). Characterization of MeCP2, a vertebrate DNA binding protein with affinity for methylated DNA. Nucleic Acids Res 20: 5085–5092. Mehta D, Gonik M, Klengel T et al. (2011). Using polymorphisms in FKBP5 to define biologically distinct subtypes

145

of posttraumatic stress disorder: evidence from endocrine and gene expression studies. Arch Gen Psychiatry 68: 901–910. Mehta D, Klengel T, Conneely KN et al. (2013). Childhood maltreatment is associated with distinct genomic and epigenetic profiles in posttraumatic stress disorder. Proc Natl Acad Sci U S A 110: 8302–8307. Miller GE, Chen E, Fok AK et al. (2009). Low earlylife social class leaves a biological residue manifested by decreased glucocorticoid and increased proinflammatory signaling. Proc Natl Acad Sci U S A 106: 14716–14721. Miller GE, Borders AE, Crockett AH et al. (2017). Maternal socioeconomic disadvantage is associated with transcriptional indications of greater immune activation and slower tissue maturation in placental biopsies and newborn cord blood. Brain Behav Immun 64: 276–284. https://doi.org/ 10.1016/j.bbi.2017.04.014. Mo A, Mukamel EA, Davis FP et al. (2015). Epigenomic signatures of neuronal diversity in the mammalian brain. Neuron 86: 1369–1384. Mondelli V, Vernon AC (2019). From early adversities to immune activation in psychiatric disorders: the role of the sympathetic nervous system. Clin Exp Immunol 197: 319–328. Morselli M, Pastor WA, Montanini B et al. (2015). Targeting of DNA methylation by histone modifications in yeast and mouse. elife 4: e06205. https://doi.org/ 10.7554/eLife.06205. Murgatroyd C, Spengler D (2014). Polycomb binding precedes early-life stress responsive DNA methylation at the Avp enhancer. PLoS One 9: e90277. Murgatroyd C, Patchev AV, Wu Y et al. (2009). Dynamic DNA methylation programs persistent adverse effects of early-life stress. Nat Neurosci 12: 1559–1566. Muschler M, Rhein M, Ritter A et al. (2018). Epigenetic alterations of the POMC promoter in tobacco dependence. Eur Neuropsychopharmacol 28: 875–879. Nan X, Ng HH, Johnson CA et al. (1998). Transcriptional repression by the methyl-CpG-binding protein MeCP2 involves a histone deacetylase complex [see comments]. Nature 393: 386–389. Naumova OY, Lee M, Koposov R et al. (2012). Differential patterns of whole-genome DNA methylation in institutionalized children and children raised by their biological parents. Dev Psychopathol 24: 143–155. Naumova OY, Rychkov SY, Kornilov SA et al. (2019). Effects of early social deprivation on epigenetic statuses and adaptive behavior of young children: a study based on a cohort of institutionalized infants and toddlers. PLoS One 14: e0214285. Neri F, Rapelli S, Krepelova A et al. (2017). Intragenic DNA methylation prevents spurious transcription initiation. Nature 543: 72–77. Neveu PJ, Deleplanque B, Puglisi-Allegra S et al. (1994). Influence of early life events on immune reactivity in adult mice. Dev Psychobiol 27: 205–213.

146

M. SZYF

Ng HH, Zhang Y, Hendrich B et al. (1999). MBD2 is a transcriptional repressor belonging to the MeCP1 histone deacetylase complex. Nat Genet 23: 58–61. Novellino F, Sacca` V, Donato A et al. (2020). Innate immunity: a common denominator between neurodegenerative and neuropsychiatric diseases. Int J Mol Sci 21: 1115. Nugent NR, Tyrka AR, Carpenter LL et al. (2011). Geneenvironment interactions: early life stress and risk for depressive and anxiety disorders. Psychopharmacology 214: 175–196. Okano M, Bell DW, Haber DA et al. (1999). DNA methyltransferases Dnmt3a and Dnmt3b are essential for de novo methylation and mammalian development. Cell 99: 247–257. Oppong E, Cato AC (2015). Effects of glucocorticoids in the immune system. Adv Exp Med Biol 872: 217–233. https:// doi.org/10.1007/978-1-4939-2895-8_9. Paquette AG, Lester BM, Lesseur C et al. (2015). Placental epigenetic patterning of glucocorticoid response genes is associated with infant neurodevelopment. Epigenomics 7: 767–779. Parade SH, Parent J, Rabemananjara K et al. (2017). Change in FK506 binding protein 5 (FKBP5) methylation over time among preschoolers with adversity. Dev Psychopathol 29: 1627–1634. Perna L, Zhang Y, Mons U et al. (2016). Epigenetic age acceleration predicts cancer, cardiovascular, and all-cause mortality in a German case cohort. Clin Epigenetics 8: 64. Perroud N, Paoloni-Giacobino A, Prada P et al. (2011). Increased methylation of glucocorticoid receptor gene (NR3C1) in adults with a history of childhood maltreatment: a link with the severity and type of trauma. Transl Psychiatry 1: e59. Power C, Hertzman C (1997). Social and biological pathways linking early life and adult disease. Br Med Bull 53: 210–221. Power C, Jefferis BJ, Manor O et al. (2006). The influence of birth weight and socioeconomic position on cognitive development: does the early home and learning environment modify their effects? J Pediatr 148: 54–61. Provencal N, Suderman MJ, Guillemin C et al. (2012). The signature of maternal rearing in the methylome in rhesus macaque prefrontal cortex and T Cells. J Neurosci 32: 15626–15642. Provencal N, Arloth J, Cattaneo A et al. (2019). Glucocorticoid exposure during hippocampal neurogenesis primes future stress response by inducing changes in DNA methylation. Proc Natl Acad Sci U S A 117: 23280–23285. Provenzi L, Giorda R, Beri S et al. (2016). SLC6A4 methylation as an epigenetic marker of life adversity exposures in humans: a systematic review of literature. Neurosci Biobehav Rev 71: 7–20. Pucci M, Micioni Di Bonaventura MV, Giusepponi ME et al. (2016). Epigenetic regulation of nociceptin/orphanin FQ and corticotropin-releasing factor system genes in frustration stress-induced binge-like palatable food consumption. Addict Biol 21: 1168–1185.

Raber J, Sorg O, Horn TF et al. (1998). Inflammatory cytokines: putative regulators of neuronal and neuro-endocrine function. Brain Res Brain Res Rev 26: 320–326. Ramsahoye BH, Biniszkiewicz D, Lyko F et al. (2000). NonCpG methylation is prevalent in embryonic stem cells and may be mediated by DNA methyltransferase 3a. Proc Natl Acad Sci U S A 97: 5237–5242. Razin A, Riggs AD (1980). DNA methylation and gene function. Science 210: 604–610. Razin A, Shemer R (1995). DNA methylation in early development. Hum Mol Genet 4: 1751–1755. Razin A, Szyf M (1984). DNA methylation patterns. Formation and function. Biochim Biophys Acta 782: 331–342. Rea S, Eisenhaber F, O’carroll D et al. (2000). Regulation of chromatin structure by site-specific histone H3 methyltransferases. Nature 406: 593–599. Reik W, Dean W, Walter J (2001). Epigenetic reprogramming in mammalian development. Science 293: 1089–1093. Rodgers AB, Morgan CP, Bronson SL et al. (2013). Paternal stress exposure alters sperm microRNA content and reprograms offspring HPA stress axis regulation. J Neurosci 33: 9003–9012. Rondelet G, Dal Maso T, Willems L et al. (2016). Structural basis for recognition of histone H3K36me3 nucleosome by human de novo DNA methyltransferases 3A and 3B. J Struct Biol 194: 357–367. Roth TL, Lubin FD, Funk AJ et al. (2009). Lasting epigenetic influence of early-life adversity on the BDNF gene. Biol Psychiatry 65: 760–769. Rougeulle C, Chaumeil J, Sarma K et al. (2004). Differential histone H3 Lys-9 and Lys-27 methylation profiles on the X chromosome. Mol Cell Biol 24: 5475–5484. Ruppenthal GC, Arling GL, Harlow HF et al. (1976). A 10-year perspective of motherless-mother monkey behavior. J Abnorm Psychol 85: 341–349. Schmidt M, Lax E, Zhou R et al. (2019). Fetal glucocorticoid receptor (Nr3c1) deficiency alters the landscape of DNA methylation of murine placenta in a sex-dependent manner and is associated to anxiety-like behavior in adulthood. Transl Psychiatry 9: 23. Schroeder M, Jakovcevski M, Polacheck T et al. (2017). A methyl-balanced diet prevents CRF-induced prenatal stress-triggered predisposition to binge eating-like phenotype. Cell Metab 25: 1269–1281 e6. Schwaiger M, Grinberg M, Moser D et al. (2016). Altered stressinduced regulation of genes in monocytes in adults with a history of childhood adversity. Neuropsychopharmacology 41: 2530–2540. Seckl JR (2004). Prenatal glucocorticoids and long-term programming. Eur J Endocrinol 151 (Suppl. 3): U49–U62. Segerstrom SC, Miller GE (2004). Psychological stress and the human immune system: a meta-analytic study of 30 years of inquiry. Psychol Bull 130: 601–630. Shinkai Y (2007). Regulation and function of H3K9 methylation. Subcell Biochem 41: 337–350. Smith JA, Zhao W, Wang X et al. (2017). Neighborhood characteristics influence DNA methylation of genes involved in

PERINATAL STRESS AND EPIGENETICS stress response and inflammation: the multi-ethnic study of atherosclerosis. Epigenetics 12: 662–673. Stein R, Gruenbaum Y, Pollack Y et al. (1982). Clonal inheritance of the pattern of DNA methylation in mouse cells. Proc Natl Acad Sci U S A 79: 61–65. Sterrenburg L, Gaszner B, Boerrigter J et al. (2011). Chronic stress induces sex-specific alterations in methylation and expression of corticotropin-releasing factor gene in the rat. PLoS One 6: e28128. Stevenson AJ, McCartney DL, Hillary RF et al. (2020). Characterisation of an inflammation-related epigenetic score and its association with cognitive ability. Clin Epigenetics 12: 113. Strahl BD, Allis CD (2000). The language of covalent histone modifications. Nature 403: 41–45. Strahl BD, Ohba R, Cook RG et al. (1999). Methylation of histone H3 at lysine 4 is highly conserved and correlates with transcriptionally active nuclei in Tetrahymena. Proc Natl Acad Sci U S A 96: 14967–14972. Stringhini S, Polidoro S, Sacerdote C et al. (2015). Life-course socioeconomic status and DNA methylation of genes regulating inflammation. Int J Epidemiol 44: 1320–1330. Stringhini S, Carmeli C, Jokela M et al. (2017). Socioeconomic status and the 25 x 25 risk factors as determinants of premature mortality: a multicohort study and meta-analysis of 1.7 million men and women. Lancet 389: 1229–1237. Suderman M, McGowan PO, Sasaki A et al. (2012). Conserved epigenetic sensitivity to early life experience in the rat and human hippocampus. Proc Natl Acad Sci U S A 109: 17266–17272. Suderman M, Borghol N, Pappas JJ et al. (2014). Childhood abuse is associated with methylation of multiple loci in adult DNA. BMC Med Genet 7: 13. Suomi SJ, Collins ML, Harlow HF et al. (1976). Effects of maternal and peer separations on young monkeys. J Child Psychol Psychiatry 17: 101–112. Surtees P, Wainwright N, Day N et al. (2003). Adverse experience in childhood as a developmental risk factor for altered immune status in adulthood. Int J Behav Med 10: 251–268. Szyf M, Eliasson L, Mann V et al. (1985). Cellular and viral DNA hypomethylation associated with induction of Epstein-Barr virus lytic cycle. Proc Natl Acad Sci U S A 82: 8090–8094. Tahiliani M, Koh KP, Shen Y et al. (2009). Conversion of 5-methylcytosine to 5-hydroxymethylcytosine in mammalian DNA by MLL partner TET1. Science 324: 930–935. Targum SD, Nemeroff CB (2019). The effect of early life stress on adult psychiatric disorders. Innov Clin Neurosci 16: 35–37. Tatro ET, Everall IP, Kaul M et al. (2009). Modulation of glucocorticoid receptor nuclear translocation in neurons by immunophilins FKBP51 and FKBP52: implications for major depressive disorder. Brain Res 1286: 1–12. Thomassin H, Flavin M, Espinas ML et al. (2001). Glucocorticoid-induced DNA demethylation and gene memory during development. EMBO J 20: 1974–1983.

147

Togher KL, Togher KL, O’keeffe MM et al. (2014). Epigenetic regulation of the placental HSD11B2 barrier and its role as a critical regulator of fetal development. Epigenetics 9: 816–822. Tozzi L, Farrell C, Booij L et al. (2018). Epigenetic changes of FKBP5 as a link connecting genetic and environmental risk factors with structural and functional brain changes in major depression. Neuropsychopharmacology 43: 1138–1145. Tsukiyama T, Wu C (1997). Chromatin remodeling and transcription. Curr Opin Genet Dev 7: 182–191. Turcotte-Tremblay AM, Lim R, Laplante DP et al. (2014). Prenatal maternal stress predicts childhood asthma in girls: project ice storm. Biomed Res Int 2014: 201717. Unternaehrer E, Meyer AH, Burkhardt SC et al. (2015). Childhood maternal care is associated with DNA methylation of the genes for brain-derived neurotrophic factor (BDNF) and oxytocin receptor (OXTR) in peripheral blood cells in adult men and women. Stress 18: 451–461. Urb M, Niinep K, Matsalu T et al. (2019). The role of DNA methyltransferase activity in cocaine treatment and withdrawal in the nucleus accumbens of mice. Addict Biol 25: e12720. Vaillancourt K, Yang J, Chen GG et al. (2020). Cocainerelated DNA methylation in caudate neurons alters 3D chromatin structure of the IRXA gene cluster. Mol Psychiatry. https://doi.org/10.1038/s41380-020-00909-x. Van Der Ploeg LH, Flavell RA (1980). DNA methylation in the human gamma delta beta-globin locus in erythroid and nonerythroid tissues. Cell 19: 947–958. Vegiopoulos A, Herzig S (2007). Glucocorticoids, metabolism and metabolic diseases. Mol Cell Endocrinol 275: 43–61. https://doi.org/10.1016/j.mce.2007.05.015. Vire E, Brenner C, Deplus R et al. (2006). The Polycomb group protein EZH2 directly controls DNA methylation. Nature 439: 871–874. Vogel Ciernia A, Laufer BI, Dunaway KW et al. (2018). Experience-dependent neuroplasticity of the developing hypothalamus: integrative epigenomic approaches. Epigenetics 13: 318–330. Waalwijk C, Flavell RA (1978). MspI, an isoschizomer of hpaII which cleaves both unmethylated and methylated hpaII sites. Nucleic Acids Res 5: 3231–3236. Waddington CH (1959). Canalization of development and genetic assimilation of acquired characters. Nature 183: 1654–1655. Wallberg AE, Flinn EM, Gustafsson JA et al. (2000). Recruitment of chromatin remodelling factors during gene activation via the glucocorticoid receptor N-terminal domain. Biochem Soc Trans 28: 410–414. Weaver IC, Cervoni N, Champagne FA et al. (2004). Epigenetic programming by maternal behavior. Nat Neurosci 7: 847–854. Weaver IC, Champagne FA, Brown SE et al. (2005). Reversal of maternal programming of stress responses in adult offspring through methyl supplementation: altering epigenetic marking later in life. J Neurosci 25: 11045–11054.

148

M. SZYF

Weaver IC, Meaney MJ, Szyf M (2006). Maternal care effects on the hippocampal transcriptome and anxiety-mediated behaviors in the offspring that are reversible in adulthood. Proc Natl Acad Sci U S A 103: 3480–3485. Weaver IC, D’alessio AC, Brown SE et al. (2007). The transcription factor nerve growth factor-inducible protein a mediates epigenetic programming: altering epigenetic marks by immediate-early genes. J Neurosci 27: 1756–1768. Weaver IC, Hellstrom IC, Brown SE et al. (2014). The methylated-DNA binding protein MBD2 enhances NGFI-A (egr-1)-mediated transcriptional activation of the glucocorticoid receptor. Philos Trans R Soc Lond Ser B Biol Sci 369: 20130513. Weinberg DN, Papillon-Cavanagh S, Chen H et al. (2019). The histone mark H3K36me2 recruits DNMT3A and shapes the intergenic DNA methylation landscape. Nature 573: 281–286. Willmer T, Goedecke JH, Dias S et al. (2020). DNA methylation of FKBP5 in South African women: associations with obesity and insulin resistance. Clin Epigenetics 12: 141. Witek Janusek L, Tell D, Albuquerque K et al. (2013). Childhood adversity increases vulnerability for behavioral symptoms and immune dysregulation in women with breast cancer. Brain Behav Immun 30 (Suppl): S149–S162. Wolffe AP (1996). Histone deacetylase: a regulator of transcription. Science 272: 371–372.

Wu JC, Santi D (1985). On the mechanism and inhibition of DNA cytosine methyltransferases. Prog Clin Biol Res 198: 119–129. Wu Y, Patchev AV, Daniel G et al. (2014). Early-life stress reduces DNA methylation of the Pomc gene in male mice. Endocrinology 155: 1751–1762. Wu X, Li G, Xie R (2018). Decoding the role of TET family dioxygenases in lineage specification. Epigenetics Chromatin 11: 58. Yang C, Ye J, Li X et al. (2016). DNA methylation patterns in the hypothalamus of female pubertal goats. PLoS One 11: e0165327. Yang C, Ye J, Liu Y et al. (2018). Methylation pattern variation between goats and rats during the onset of puberty. Reprod Domest Anim 53: 793–800. Yehuda R, Daskalakis NP, Bierer LM et al. (2016). Holocaust exposure induced intergenerational effects on FKBP5 methylation. Biol Psychiatry 80: 372–380. Yuan X, Li Z, Ye S et al. (2019). Genome-wide DNA methylation analysis of pituitaries during the initiation of puberty in gilts. PLoS One 14: e0212630. Zhou Z, Hong EJ, Cohen S et al. (2006). Brain-specific phosphorylation of MeCP2 regulates activity-dependent Bdnf transcription, dendritic growth, and spine maturation. Neuron 52: 255–269. Zhu Y, Sun L, Chen Z et al. (2013). Predicting enhancer transcription and activity from chromatin modifications. Nucleic Acids Res 41: 10032–10043.

Handbook of Clinical Neurology, Vol. 180 (3rd series) The Human Hypothalamus: Middle and Posterior Region D.F. Swaab, F. Kreier, P.J. Lucassen, A. Salehi, and R.M. Buijs, Editors https://doi.org/10.1016/B978-0-12-820107-7.00009-4 Copyright © 2021 Elsevier B.V. All rights reserved

Chapter 9

The hypothalamus in anxiety disorders SUSANNE FISCHER* Clinical Psychology and Psychotherapy, University of Zurich, Zurich, Switzerland

Abstract Of all mental disorders, anxiety disorders are currently the strongest contributors to the global burden of disease, with 7.3% of the general population affected worldwide. The hypothalamus is crucial hub of a network of neural structures modulating fear conditioning and extinction and, as such, highly relevant to the pathophysiology of these conditions. Three hypothalamic systems have emerged as particularly relevant in this context. First, the oxytocin system is highly likely to be involved anxiety disorders and in particular in the cognitive and behavioral deficits pertaining to social anxiety disorder. Second, peripheral markers of the hypothalamic–pituitary–adrenal axis appear altered in patients with panic disorder and generalized anxiety disorder, which may denote aberrant functioning of their central corticotropin-releasing hormone system. Furthermore, cortisol seems to augment the effects of exposure therapy in patients with specific phobia. Third, the integrity of the hypothalamic–pituitary–thyroid axis is likely compromised in panic disorder. Further, cross-disciplinary research efforts are required to shed more light on how, exactly, these hypothalamic systems interact with the neural structures involved in fear conditioning and extinction, which should ultimately open up new avenues for the prevention and treatment of anxiety disorders.

INTRODUCTION Of all mental disorders, anxiety disorders are currently the strongest contributors to the global burden of disease (Whiteford et al., 2013), with 7.3% of the general population affected worldwide (Baxter et al., 2013). Indeed, the most prevalent adult anxiety disorders present with numerous debilitating symptoms. In social anxiety disorder, a fear of embarrassing or humiliating social interactions predominates (APA, 2013). Panic disorder is defined by recurrent, unexpected panic attacks, which are characterized by symptoms such as palpitations, shortness of breath, sweating, nausea, dizziness, and fear of losing control or dying. Typically, these symptoms arise within 10 min, and panic attacks are followed by at least 1 month of anxiety about additional attacks. Individuals with generalized anxiety disorder present with persistent worries related to a wide range of subjects (e.g., health, finances), which are accompanied by

symptoms such as fatigue, impaired concentration, and irritability. Finally, in specific phobia, individuals experience intense, persistent and excessive fear of a particular object or situation. The annual direct and indirect cost of anxiety disorders is currently estimated at up to $30,041 per patient (Konnopka et al., 2009). In addition to causing tremendous suffering in affected individuals, anxiety disorders are paralleled by frequent healthcare use and reduced productivity, and as such inflict a substantial financial burden on society. Therefore, a more in-depth understanding of these conditions appears imperative in order to alleviate their individual and societal consequences. The fact that all anxiety disorders are typified by the presence of fear/anxiety and ancillary behavioral, cognitive, and/or social symptoms suggests a common pathophysiological mechanism (Craske and Stein, 2016). A number of neural structures have received particular attention in this context: the amygdala, which coordinates

*Correspondence to: Dr Susanne Fischer, University of Zurich, Institute of Psychology, Clinical Psychology and Psychotherapy, Binzmuehlestrasse 14/Box 26, 8050 Zurich, Switzerland. Tel: +41-44-635-74-60, Fax: +41-44-637-73-59, E-mail: s.fischer@ psychologie.uzh.ch

150

S. FISCHER

fear and anxiety responses, the hippocampus, which governs learning and memory processes, and the prefrontal cortex, which mediates cognitive (including executive) function. Together, they constitute a network of areas, which is presumably involved in the development and perpetuation of clinical anxiety by facilitating fear generalization and hampering fear extinction. A crucial hub of this network is the hypothalamus, with several neuropeptides expressed within its nuclei projecting to and receiving input from these very same structures (Bao and Swaab, 2019). The present chapter will outline three major hypothalamic systems that have emerged as particularly relevant to this context: oxytocin, corticotropin-releasing hormone (CRH) and thyrotropin-releasing hormone (TRH). Each subchapter will begin with a brief overview of the physiology and function of the system, before continuing with a comprehensive summary of studies in anxiety disorders. The three sections will be integrated in the final part of the chapter and suggestions for future research agendas will be made.

OXYTOCIN Oxytocin is produced by magnocellular and parvocellular neurons in the supraoptic nucleus and in the paraventricular nucleus (PVN) of the hypothalamus (Jurek and Neumann, 2018). Oxytocin is transferred to the posterior pituitary via axons, where it is stored and released into the bloodstream. The physiological functions of oxytocin are manifold. In the periphery, its main effects lie in promoting uterine contractions during labor and milk ejection during lactation by binding to G protein-coupled oxytocin receptors. In the central nervous system, the same receptors are expressed in several extrahypothalamic areas that are relevant to anxiety disorders, including the amygdala, the hippocampus, the nucleus accumbens, and the cerebral cortex (e.g., Knobloch et al., 2012; Eckstein et al., 2015; Hasan et al., 2019). The main effects of oxytocin in humans appear to be in modulating social cognition and behavior. For instance, in healthy adults, intranasal oxytocin has been shown to enhance emotion recognition (Leppanen et al., 2017), to promote in-group trust (Van IJzendoorn and Bakermans-Kranenburg, 2012), and to facilitate the expression of positive emotions (Leppanen et al., 2017). Given these effects, the last 2 decades have witnessed a considerable interest in the oxytocin system within research on social anxiety disorder, at the core of which are deficits in social functioning (see also MacDonald and Feifel, 2014; Neumann and Slattery, 2016; Gottschalk and Domschke, 2018). One important line of research has focused on endogenous oxytocin. A seminal study in this regard measured plasma oxytocin in patients with generalized social anxiety disorder (GSAD) and healthy controls. While no

significant differences in oxytocin levels were found between the two groups, oxytocin was positively related to anxiety severity in the patients with GSAD (Hoge et al., 2008). A larger subsequent study by the same group did find lower oxytocin levels in patients when compared to healthy controls, but only after a trust game (Hoge et al., 2012). Notably, neither of these studies extracted the plasma samples before biochemical analysis, which could have falsely inflated oxytocin values and masked potential abnormalities in patients (Engel et al., 2019). Moreover, plasma measures of oxytocin are not necessarily representative of centrally available levels, as oxytocin does not readily cross the blood–brain barrier, thus impeding the ability to draw any inferences about the central availability of oxytocin (Valstad et al., 2017). Notwithstanding these limitations, the findings are in line with an epigenetic study in a Caucasian sample (Ziegler et al., 2015). The study showed hypomethylation within exon 3 of the oxytocin receptor gene OXTR in patients with social anxiety disorder relative to healthy controls, along with a negative association between methylation levels and anxiety severity. Together, these findings suggest an upregulation of oxytocin receptors in social anxiety disorder and a dose–response relationship between circulating oxytocin and anxiety severity in the resting state in concert with diminished levels of oxytocin in response to signals of social trust. Notably, as gene expression was not measured in the Ziegler et al. (2015) study, it can only be speculated whether the observed hypomethylation in patients does indeed reflect increased expression of OXTR. Furthermore, it remains questionable whether DNA methylation measured in peripheral blood cells is representative of DNA methylation in brain tissue. Interestingly, in the same study, genotype distribution of a single nucleotide polymorphism (SNP) within OXTR previously implicated in social cognition and behavior (rs53576) did not differ between patients and controls (Ziegler et al., 2015). This null finding was replicated in a Japanese sample; however, individuals with the GG genotype of the rs53576 and rs2254298 SNPs had an increased risk of panic disorder (Onodera et al., 2015). A caveat concerning the latter study is that the distribution of genotypes did not fit the Hardy–Weinberg equilibrium, indicating sample selection bias. These preliminary findings should therefore not be interpreted as definitive evidence for an absence of genetic influences on oxytocin abnormalities in social anxiety. Another, experimental line of research has employed exogenous oxytocin in patients with social anxiety disorder (see De Cagna et al., 2019 for a systematic review). Although the question of whether intranasally applied oxytocin reaches effective concentrations in the brain is still a subject of ongoing scientific debate, a burgeoning literature of placebo-controlled studies supports this

THE HYPOTHALAMUS IN ANXIETY DISORDERS notion. In the first two functional magnetic resonance imaging (fMRI) studies, administering oxytocin to patients with generalized social anxiety was found to normalize a previously observed hyperactivity of the amygdala while viewing fearful faces (Labuschagne et al., 2010) as well as a previously observed hyperactivity of the anterior cingulate cortex (ACC) and medial prefrontal cortex (mPFC) while viewing sad faces (Labuschagne et al., 2012). Subsequent resting-state fMRI analyses added to these observations by showing that oxytocin also appears to strengthen the connectivity between the amygdala and the rostral ACC and mPFC after intranasal oxytocin administration, reversing a previously recorded hypoconnectivity (Dodhia et al., 2014). Importantly, correlation analyses revealed that this improvement was strongest in patients with high levels of anxiety, suggesting that the most severely afflicted individuals may benefit the most from intranasal oxytocin. These findings are complemented by another report which demonstrated that oxytocin reinstated an initial hypoconnectivity between the amygdala, insula, and the mid/dorsal ACC (Gorka et al., 2015). In sum, these results may be interpreted such that oxytocin ameliorates the top-down control of exaggerated fear responses to social signals in individuals with social anxiety disorder. In addition, intranasal oxytocin has been demonstrated to foster social cooperation in such individuals (Fang et al., 2014). However, this was only true when attachment avoidance was relatively low, whereas oxytocin facilitated the detection of neutral and disgusted faces when attachment avoidance was high. Similarly, oxytocin led to prosocial behavior only in patients with lower levels of anxiety (Fang et al., 2017). When taken together, the previously mentioned studies suggest that any beneficial effects of exogenous oxytocin are highly dependent on the severity of social anxiety disorder; while highly anxious individuals are more likely to benefit from oxytocin in terms of improved responses to potentially threatening social cues, only individuals with lower levels of anxiety appear to benefit from its prosocial effects. Importantly, all of the previous studies recruited exclusively male samples, thus limiting the generalizability of the findings. A third line of research, consisting of clinical studies, has attempted to integrate exogenous oxytocin into cognitive behavioral therapy (CBT) for anxiety disorders (De Cagna et al., 2019). Guastella et al. (2009) were the first to test intranasal oxytocin as an adjunct to exposure therapy compared to placebo, and—again—in patients with social anxiety disorder. Although the oxytocin group showed more favorable evaluations of their own appearance and speech performance following the four exposure sessions, the overall social anxiety at the end of treatment and at 1-month follow-up was

151

comparable to the placebo group. Similarly, Acheson et al. (2015) were unable to demonstrate any effects of intranasal oxytocin on either self-reported or clinically evaluated arachnophobia after a single-session exposure therapy; in fact, the placebo group showed better outcomes than the verum group at 1-month follow-up. Together, these studies suggest that oxytocin may exert beneficial effects on dysfunctional cognitions in patients with social anxiety disorder, while apparently failing to enhance fear extinction in specific phobia. Notably, both of the aforementioned studies employed a limited number of weekly administrations of oxytocin. The jury is thus still out on whether oxytocin as an augmentation to CBT may prove useful if daily dosing regimens are employed and/or in multisession trials of longer duration. In sum, the converging fields of oxytocin research point to a complex role of the system in social anxiety disorder, with a lower central oxytocin level potentially paralleling some of the cognitive and behavioral deficits these patients exhibit in social situations (e.g., negative appraisal of the self and others). Further research is warranted to delineate oxytocin’s regional and cell-type specific actions in the brain, and in different contexts relevant to social anxiety disorders, before findings may potentially prove useful to clinical practitioners.

CORTICOTROPIN-RELEASING HORMONE The preeminent function of CRH lies in coordinating the physiological responses to external and internal demands (Deussing and Chen, 2018). As such, CRH is a key orchestrator of both sympatho-adrenomedullary (SAM) and hypothalamic–pituitary–adrenal (HPA) axis activity. In terms of the SAM system, CRH directly increases the firing rate of noradrenergic neurons in the locus coeruleus, which ultimately results in increased sympathetic innervation of organs, including the heart and lungs, as well as in the secretion of adrenaline from the adrenal medulla. In terms of the HPA axis, adrenocorticotropic hormone (ACTH) is released from the anterior pituitary upon transportation of CRH from parvocellular cells of the PVN via the portal vasculature. Adrenocorticotropic hormone, in turn, enters the peripheral bloodstream, and initiates the synthesis and secretion of the glucocorticoid cortisol from the adrenal cortex, which, as the end product of this HPA axis, not only mediates metabolic and immune function in the periphery, but feeds back to inhibit the release of CRH and ACTH at the central level. However, above and beyond its function as a secretagogue, CRH has direct neuromodulatory properties. Indeed, its G protein-coupled receptor 1 (CRHR1) has been located in extrahypothalamic areas such as various brainstem nuclei, the amygdala, the hippocampus, and

152

S. FISCHER

cortical structures. The cognate receptor 2 (CRHR2) is less abundantly expressed, but has been found in the raphe nucleus, the mamillary nucleus, and the lateral septum. Corresponding to the distribution of the CRHR1, research in animals has shown that injected CRH causes heightened arousal, enhanced fear conditioning, and increased and decreased behavioral activity in familiar and unfamiliar environments, respectively. On account of these anxiogenic-like properties, and given the presence of anxiety-related symptoms in patients with clinical conditions presenting with HPA axis malfunctioning (e.g., Cushing’s disease), it is unsurprising that the last 4 decades have produced a wealth of data on the role of CRH and its successor hormones in anxiety disorders, and especially in panic and generalized anxiety disorder (Risbrough and Stein, 2006; Binder and Nemeroff, 2010). Due to the limited access to brain tissue, the bulk of research has focused on the pituitary and adrenal levels (see later). However, two studies used lumbar punctures to gain an approximate measure of the apex of the HPA axis, CRH. The first study, which was conducted in patients with panic disorder, failed to demonstrate any significant differences in cerebrospinal fluid (CSF) levels of CRH between patients and controls, and found no evidence for a relationship between CRH and anxiety severity (Jolkkonen et al., 1993); the second study extended these null-findings to a mixed group of patients with different anxiety disorders (Fossey et al., 1996). Notably, lumbar punctures are highly stressful procedures, which may have concealed any resting-state differences in CRH between anxious and non-anxious individuals. Furthermore, nothing is known so far about the dynamics of CRH in situations that are highly pertinent to anxiety disorders, such as panic attacks. However, initial inquiries into potential changes in CRH as invoked by treatment with benzodiazepines or antidepressants have likewise been disappointing (Jolkkonen et al., 1993). More recently, (epi-)genetic research into CRH has gained momentum. A pioneering investigation in this field reported relative hypomethylation of the CRHR1 promoter region in panic disorder, and this pattern was confirmed to confer increased in vitro expression of the CRHR1 (Schartner et al., 2017). Again, further research is warranted to clarify whether peripheral blood constitutes a valid proxy for brain tissue. In terms of genetic studies, Keck et al. (2008) investigated genetic variation within CRH, CRHR1, and CRHR2, and found the G allele of a particular SNP within CRHR1 (rs878886) to be more frequent in patients with panic disorder than in controls. Weber et al. (2016) went one step further, by conducting a multilevel study on the role of the CRHR1 in panic disorder. They found four SNPs

to be significantly linked to panic disorder. Of these, the A allele of rs17689918, which is in linkage disequilibrium with rs878886, not only had the strongest association with panic disorder, but carriers were characterized by reduced CRHR1 expression in the amygdala and forebrain of postmortem samples, reduced frontal cortex activity during differential conditioning, and a dissociation between psychological and physiological measures during a behavioral avoidance task. By contrast, three repeat polymorphisms within CRHR2 examined in an earlier study by Tharmalingam et al. (2006) did not differentiate patients with panic disorder from healthy controls. Overall, the current state of research is conflicted as to whether the CRHR1 is up- or downregulated in panic disorder, with epigenetic studies suggesting the former and genetic studies suggesting the latter. Correspondingly, initial attempts at testing a selective CRHR1 antagonist in patients with generalized anxiety disorders have failed (Coric et al., 2010). Thus, further scientific efforts are necessary to unravel the regional distribution and functional role of the CRHR1 in panic disorder, and more studies encompassing multiple levels of research will undoubtedly be pivotal to achieve this. As regards ACTH, a considerable body of research investigating this parameter has accumulated over the last 4 decades, although, as with CRH, research efforts have almost exclusively been directed at panic disorder. As evident from Table 9.1, basal circulating levels do not seem to differentiate patients with panic disorder from healthy controls, although the vast majority of these studies employed single time point measures and often immediately before experimental testing, which could have impacted the values due to anticipatory anxiety. Indeed, the most elaborate study measuring ACTH over the course of 24 h did observe subtle abnormalities in circadian rhythms (i.e., elevated circadian amplitudes), which not only distinguished patients from healthy controls, but also distinguished between patients with a low vs high number of panic attacks (Abelson and Curtis, 1996). Stimulation of ACTH secretion by means of psychosocial stress yielded ambiguous findings, with some evidence for diminished responses in patients, whereas stimulation via CRH was not effective in revealing any potential abnormalities, even when individuals were pretreated with dexamethasone, a synthetic analog of cortisol. The same was true when the system was probed with metyrapone, a cortisol inhibitor, or with panicogenic substances, such as pentagastrin. Cognitive interventions seemed to be effective in reducing ACTH responses to panicogenic substances. Finally, challenges with serotonin receptor agonists were equivocal, while there were some indications that longer-term treatment with benzodiazepines alleviates previously enhanced ACTH

THE HYPOTHALAMUS IN ANXIETY DISORDERS

153

Table 9.1 Findings of studies comparing adrenocorticotropic hormone (ACTH) between patients with anxiety disorders and healthy controls. Study Baseline Abelson et al. (1991) Gurguis et al. (1991)

Kahn et al. (1991) Roy-Byrne et al. (1991)

Brambilla et al. (1992) Lesch et al. (1992) Abelson and Curtis (1996)

Abelson et al. (1996b) Curtis et al. (1997) Condren et al. (2002) Abelson et al. (2005) van Duinen et al. (2007) Siegmund et al. (2011) Petrowski et al. (2013) Wichmann et al. (2017b) Psychological stress Condren et al. (2002) Young et al. (2004)

Siegmund et al. (2011) Petrowski et al. (2013) Wichmann et al. (2017a)

Sample

Results

N ¼ 5 panic disorder N ¼ 5 control N ¼ 32 panic disorder with agoraphobia N ¼ 12 control N ¼ 20 panic disorder N ¼ 20 control N ¼ 13 panic disorder N ¼ 8 generalized anxiety disorder N ¼ 13 control N ¼ 17 panic disorder N ¼ 14 control N ¼ 14 panic disorder N ¼ 14 control N ¼ 20 panic disorder N ¼ 12 control

Elevated baseline ACTH in patients when compared to controls

N ¼ 16 panic disorder N ¼ 16 control N ¼ 20 panic disorder N ¼ 12 control N ¼ 15 social phobia N ¼ 15 control N ¼ 14 panic disorder N ¼ 14 control N ¼ 16 panic disorder N ¼ 16 control N ¼ 10 panic disorder N ¼ 10 control N ¼ 32 panic disorder N ¼ 32 control N ¼ 28 panic disorder N ¼ 32 control N ¼ 15 social phobia N ¼ 15 control N ¼ 15 anxiety disorder N ¼ 48 control N ¼ 10 panic disorder N ¼ 10 control N ¼ 32 panic disorder N ¼ 32 control N ¼ 29 panic disorder N ¼ 46 control

No group differences in baseline ACTH

No group differences in baseline ACTH No group differences in baseline ACTH

Elevated baseline ACTH in patients when compared to controls No group differences in baseline ACTH Elevated circadian ACTH amplitudes in patients when compared to controls Patients with a lower number of panic attacks exhibited elevated daytime mean ACTH when compared to controls Patients with a higher number of panic attacks exhibited an earlier circadian ACTH acrophase when compared to controls Elevated baseline ACTH in patients when compared to controls No group differences in baseline ACTH No group differences in baseline ACTH No correlation between baseline ACTH and anxiety severity No group differences in baseline ACTH No group differences in baseline ACTH No group differences in baseline ACTH No group differences in baseline ACTH No correlation between baseline ACTH and anxiety severity No group differences in baseline ACTH

No group differences in ACTH responses to psychosocial stress No correlation between ACTH responses and anxiety severity No group differences in ACTH responses to psychosocial stress

Attenuated ACTH levels after an exposure session in patients when compared to controls Attenuated ACTH responses to psychosocial stress in patients when compared to controls No group differences in ACTH responses to psychosocial stress

Continued

154

S. FISCHER

Table 9.1 Continued Study

Sample

Results

Wichmann et al. (2017b)

N ¼ 28 panic disorder N ¼ 32 control

No group differences in ACTH responses to psychosocial stress

N ¼ 17 panic disorder N ¼ 14 control N ¼ 20 panic disorder N ¼ 12 control

No group differences in ACTH responses to CRH infusion

CRH Brambilla et al. (1992) Curtis et al. (1997)

Dexamethasone Petrowski et al. (2012) Wichmann et al. (2018) Dexamethasone/CRH Schreiber et al. (1996) Erhardt et al. (2006) Petrowski et al. (2012)

Wichmann et al. (2018) Metyrapone Kellner et al. (2004)

N ¼ 32 panic disorder N ¼ 32 control N ¼ 34 panic disorder N ¼ 34 control

No group differences in ACTH responses to oral intake of dexamethasone

N ¼ 13 panic disorder N ¼ 10 control N ¼ 30 panic disorder N ¼ 30 control N ¼ 32 panic disorder N ¼ 32 control

No differences in ACTH responses to CRH infusion between dexamethasone pretreated patients and controls No differences in ACTH responses to CRH infusion between dexamethasone pretreated patients and controls No differences in ACTH responses to CRH infusion between dexamethasone pretreated patients and controls Negative correlation between ACTH response and anxiety severity No differences in ACTH responses to CRH infusion between dexamethasone pretreated patients and controls

N ¼ 34 panic disorder N ¼ 34 control N ¼ 14 panic disorder N ¼ 14 control

Metyrapone/dexamethasone Kellner et al. (2004) N ¼ 14 panic disorder N ¼ 14 control Panicogens Abelson et al. (1994) N ¼ 10 panic disorder N ¼ 10 control Kellner et al. (1998) N ¼ 10 panic disorder N ¼ 10 control Wiedemann et al. (2001) N ¼ 9 panic disorder N ¼ 9 control Abelson et al. (2005) N ¼ 14 panic disorder N ¼ 14 control van Duinen et al. (2007)

Earlier and higher ACTH peak in response to CRH infusion in patients when compared to controls No correlation between ACTH response and anxiety severity

N ¼ 16 panic disorder N ¼ 16 control

Psychotherapy Abelson et al. (1996b)

N ¼ 16 panic disorder

Abelson et al. (2005)

N ¼ 14 panic disorder

Siegmund et al. (2011)

N ¼ 10 panic disorder

Wichmann et al. (2017b)

N ¼ 28 panic disorder N ¼ 32 control

No group differences in ACTH responses to oral intake of dexamethasone

No group differences in ACTH responses to oral intake of metyrapone

No differences in ACTH responses to oral intake of dexamethasone between metyrapone pretreated patients and controls No group differences in ACTH responses to pentagastrin infusion No correlation between ACTH and symptom response No group differences in ACTH responses to sodium lactate infusion Elevated ACTH responses to cholecystokinin tetrapeptide infusion in patients when compared to controls No group differences in ACTH responses to pentagastrin infusion Positive correlation between ACTH responses and anxious distress and panic symptoms No group differences in ACTH responses to CO2 inhalation

Reduction of ACTH responses to doxapram infusion after a cognitive intervention Reduction of ACTH responses to pentagastrin infusion after a cognitive intervention Negative correlation between ACTH levels during the first exposure session and symptom improvement at posttreatment and follow-up No correlation between pretreatment ACTH responses to psychosocial stress and symptom improvement after 5 weeks of cognitive behavioral therapy

THE HYPOTHALAMUS IN ANXIETY DISORDERS

155

Table 9.1 Continued Study Serotonin agonists Kahn et al. (1991) Lesch et al. (1992) Benzodiazepines Roy-Byrne et al. (1991)

Brambilla et al. (1992) Klein et al. (1995)

Abelson et al. (1996a) Curtis et al. (1997) Strohle et al. (1999)

Sample

Results

N ¼ 20 panic disorder N ¼ 20 control N ¼ 14 panic disorder N ¼ 14 control

Elevated ACTH response to the oral intake of a serotonin receptor agonist only in female patients when compared to controls Attenuated ACTH response to the oral intake of a serotonin receptor agonist in patients when compared to controls

N ¼ 13 panic disorder N ¼ 8 generalized anxiety disorder N ¼ 13 control N ¼ 17 panic disorder N ¼ 14 control N ¼ 36 panic disorder N ¼ 35 generalized anxiety disorder N ¼ 20 panic disorder N ¼ 20 panic disorder N ¼ 12 control N ¼ 8 panic disorder N ¼ 8 control

No group differences in ACTH responses to diazepam infusion

Normalization (reduction) of baseline ACTH after treatment with alprazolam No effect of 9 weeks of treatment with alprazolam on baseline ACTH

Higher evening ACTH levels after 12 weeks of treatment with alprazolam Normalization (reduction) of ACTH responses to CRH infusion after 12 weeks of treatment with alprazolam No group differences in ACTH responses to a benzodiazepine receptor antagonist

concentrations. To summarize, the current state of literature in panic disorder may be interpreted as suggesting subtle elevations of ACTH when observed over the course of the day and decreases in ACTH when applying psychological or pharmacological anxiolytics. The largest proportion of research on the CRH system has been devoted to the study of cortisol in anxiety disorders (see Elnazer and Baldwin, 2014 for a review). In panic disorder, there is accumulating evidence for elevated resting levels of cortisol, a diminished cortisol secretion upon exogenous stimulation with CRH or during acute psychosocial stress, and for higher cortisol to predict panic attacks after infusion with sodium lactate (mimicking rising levels of serum lactate during panic attacks). In generalized anxiety disorder, findings point to a lowering of cortisol with successful pharmacological or psychological treatment. In addition, another recent investigation reported increased methylation within the glucocorticoid receptor (GR) gene (NR3C1) 1F promoter in patients with generalized anxiety disorder when compared to healthy controls, and the more pronounced this pattern, the lower the amount and sensitivity of patients’ GRs (Wang et al., 2017). By contrast, in a pharmacogenetic study of generalized anxiety disorder, none of the three tested SNPs within NR3C1 were found to predict antidepressant response (Perlis et al., 2013). In social anxiety disorder, enhanced levels during psychosocial stress were observed upon comparison with healthy controls. Finally, in

specific phobia, there is consistent evidence for attenuated stimulus-induced fear when cortisone is administered immediately before psychotherapeutic exposure sessions. The latter finding mirrors clinical studies, which revealed that the higher a patient’s endogenous cortisol levels during exposure sessions, the better their general treatment outcome (see Fischer and Cleare, 2017 for a systematic review and meta-analysis). These results make sense in light of the fact that cortisol, via its mineralocorticoid and GR-mediated actions in the hippocampus, is a facilitator of fear extinction, a key ingredient in the successful psychological treatment of any anxiety disorder (de Quervain et al., 2017). In conclusion, the diverse literature suggests elevated ACTH and cortisol in panic disorder, which could be indicative of central dysregulation of CRHR1 expression. There is tentative evidence that psychological and pharmacological treatments alleviate ACTH and cortisol in panic disorder and generalized anxiety disorder. Lastly, cortisol appears to aid the extinction of fear memories in the context of psychological interventions targeting specific phobia.

THYROTROPIN-RELEASING HORMONE TRH is mainly expressed in the hypothalamus, where it is produced by parvocellular neurons in the PVN and stimulates the release of thyroid-stimulating hormone (TSH)

156 S. FISCHER in the anterior pituitary (Fekete and Lechan, 2014; In sum, the present state of research can be summarized as suggesting lowered stimulated TSH in panic disOrtiga-Carvalho et al., 2016; Frohlich and Wahl, order. However, both central and peripheral endpoints of 2019). TSH, in turn, acts as a secretagogue of the thyroid the HPT axis remain sorely understudied, and important hormones triiodothyronine (T3) and thyroxine (T4) in the thyroid gland. Triiodothyronine is released at a questions for future research concern the distribution of fraction of 20%, whereas T4 is released at a fraction of the TRHR as well as the actions of TRH in the central 80%. However, the less biologically active T4 can be nervous system and how they may relate to other types converted into T3, which occurs by means of three of anxiety disorders. tissue-specific types of deiodinase. In the periphery, the multifaceted actions of thyroid hormones include CONCLUSION the regulation of cardiovascular, bone, and liver function; in the brain, they are inhibitors of TRH and TSH release The present chapter outlined three major neuroendocrine in the hypothalamus and pituitary, respectively. The role systems originating in the hypothalamus, which have of nonhypophysiotropic TRH in the brain is less clear, been a major focus of research into the neurobiological although the TRH receptor (TRHR) has been located underpinnings of anxiety disorders. While since the 1980s, it was mainly the close phenomenological ties in extrahypothalamic regions, such as parts of the SAM between manifest endocrine and anxiety disorders that system and the medulla, and TRH injections exert demonstrable anxiogenic effects. Nevertheless, these findings as gave rise to extensive investigations into the CRH and well as the fact that anxiety symptoms, such as palpitations, TRH systems, from the turn of the millennium, exciting shortness of breath, fatigue, and impaired concentration are findings originating from basic research have spurred myriad in association with overt hyperthyroidism (e.g., clinical researchers’ interest in the oxytocin system. Graves’ disease), have led to a steadily growing awareness The following picture emerges concerning the role of of the potential importance of the TRH system in these these systems in anxiety disorders: First, oxytocin is highly likely to be involved in the cognitive and behavconditions (Simon et al., 2002; Fischer and Ehlert, 2018). ioral deficits pertaining to social anxiety disorder, but At the central level, only one study to date has attempted to compare CSF TRH between patients with there is little evidence to date with regard to a potential anxiety disorders and healthy controls: No apparent role in other anxiety disorders. Second, an almost overabnormalities were detected in a mixed patient group, whelming body of research documents alterations in and TRH concentrations were not correlated with anxiety peripheral HPA axis markers in patients with panic disseverity (Fossey et al., 1993). Again, this null-finding is order and generalized anxiety disorder, which may tentative and not to be interpreted as unequivocal evidenote aberrant functioning of the central CRH system. dence of the irrelevance of TRH in anxiety, given the In addition, while not generally found to be abnormal heterogeneity of the sample and the incomplete underin specific phobia, the glucocorticoid system of these standing of how TRH is distributed and utilized in the patients can apparently be exploited to augment the central nervous system. A general consensus that resting effects of CBT during exposure sessions. Third, the levels of TSH are normal in social anxiety disorder and integrity of the HPT axis is likely compromised in panic panic disorder can be derived from the current state of the disorder, while virtually no research has been devoted to literature (see Fischer and Ehlert, 2018 for a systematic other types of anxiety disorders. However, these findings review). However, TSH reactivity appears to be deficient should be interpreted in light of a host of limitations. in the same patients, such that attenuated levels prevail From a clinical perspective, there is a lack of subtypafter probing with exogenous TRH. Interestingly, in ing within subtypes of different anxiety disorders. For one of these studies, this pattern was only present in instance, in social anxiety disorder, the bulk of research patients with panic disorder but not in those with agorahas been dedicated to GSAD, rendering it unclear phobia (Hofmann et al., 2001). Other studies reported a whether findings are applicable to more specific types negative relation between the TSH response and anxiety of this disorder. In panic disorder, the number of panic severity in the same patient population (Tukel et al., attacks as well as the concomitant presence of agorapho1999) and an enhancement of TSH responses during bia has insufficiently been accounted for, despite their antidepressant treatment (Stein and Uhde, 1991). Finally, importance as shown by a handful of carefully designed two studies evaluating thyroid hormones in social anxistudies. In generalized anxiety disorder, comorbidity ety disorder (Tancer et al., 1990) and panic disorder with major depressive disorder, which presents with dis(Stein et al., 1991), respectively, produced nonsignificant tinct abnormalities in all three hypothalamic systems, is results, although another study observed declines in T4 generally high, and not all studies have been mindful of after treatment with benzodiazepines and tricyclic this aspect. Related to this, states of (chronic) anxiety and antidepressants (Balon et al., 1991). (acute) fear have rarely been disentangled in the same

THE HYPOTHALAMUS IN ANXIETY DISORDERS studies, which might explain some of the inconsistencies between resting and postchallenge measures in the literature. From a methodological point of view, the most glaring impediment to gaining further insight into the role of oxytocin, CRH, and TRH in anxiety disorders is the incomplete account of their regional distribution in the brain. This problem is amplified by the fact that, although CSF and plasma markers are often claimed to represent a window into the brain, the possible inferences about central processes are restricted by the failure of this methodology to account for tissue specificity. Another limitation is that single endocrine measures have dominated all three lines of research, and this does not enable the interactions to be examined across multiple levels and layers of these systems. Luckily, investigations complementing multiple endocrine with (epi-)genetic and functional imaging measures have resuscitated the field more recently and should help to resolve some of the still existing conundrums in this literature. Finally, significant caution is encouraged in drawing causal inferences from the previously summarized findings, since longitudinal studies are sparse and it is thus largely unknown whether the observed abnormalities constitute antecedents or consequences of anxiety disorders. To conclude, there is substantial evidence that the oxytocin, CRH, and TRH systems are instrumental in the pathophysiology of anxiety disorders, although their exact role in the development and maintenance of symptoms is only beginning to be understood. Further investigations into the hypothalamus are required, not only in order to do justice to the complexity of patients’ experience and hence to enhance their sense of plausibility regarding their symptoms but also because biological markers in general hold the promise of improving the recognition and assessment of anxiety disorders. Furthermore, although CBT is the gold standard treatment for anxiety disorders (NICE, 2011, 2013), around 50% of patients do not respond sufficiently (Carpenter et al., 2018). There is thus much hope that a more nuanced view of the neurobiology of anxiety disorders will inform the choice of currently available treatments and aid the development of novel treatments. Indeed, pretreatment cortisol levels have shown some promise in predicting CBT outcomes in these patients. However, whether the oxytocin, CRH, and TRH systems are viable therapeutic targets remains to be established; while the monotherapeutic administration of a CRHR1 antagonist has not proven to be efficacious so far, there seems to be potential in the adjunctive use of glucocorticoids during CBT. Further, cross-disciplinary research efforts will undoubtedly shed more light on how, exactly, hypothalamic systems interact with the neural structures involved in fear conditioning and extinction, which should ultimately open up new avenues for the prevention and treatment of anxiety disorders.

157

ACKNOWLEDGMENT The author would like to acknowledge the assistance of Raphaela Sch€opfer in compiling Table 9.1.

REFERENCES Abelson JL, Curtis GC (1996). Hypothalamic-pituitaryadrenal axis activity in panic disorder. 24-hour secretion of corticotropin and cortisol. Arch Gen Psychiatry 53: 323–331. Abelson JL, Curtis GC, Cameron OG (1996a). Hypothalamicpituitary-adrenal axis activity in panic disorder: effects of alprazolam on 24 h secretion of adrenocorticotropin and cortisol. J Psychiatr Res 30: 79–93. Abelson JL, Liberzon I, Young EA et al. (2005). Cognitive modulation of the endocrine stress response to a pharmacological challenge in normal and panic disorder subjects. Arch Gen Psychiatry 62: 668–675. Abelson JL, Nesse RM, Vinik A (1991). Stimulation of corticotropin release by pentagastrin in normal subjects and patients with panic disorder. Biol Psychiatry 29: 1220–1223. Abelson JL, Nesse RM, Vinik AI (1994). Pentagastrin infusions in patients with panic disorder. II. Neuroendocrinology. Biol Psychiatry 36: 84–96. Abelson JL, Weg JG, Nesse RM et al. (1996b). Neuroendocrine responses to laboratory panic: cognitive intervention in the doxapram model. Psychoneuroendocrinology 21: 375–390. Acheson DT, Feifel D, Kamenski M et al. (2015). Intranasal oxytocin administration prior to exposure therapy for arachnophobia impedes treatment response. Depress Anxiety 32: 400–407. APA (2013). Diagnostic and statistical manual of mental disorders, 5th ed. APA, Washington, DC. Balon R, Pohl R, Yeragani VK et al. (1991). The changes of thyroid hormone during pharmacological treatment of panic disorder patients. Prog Neuro-Psychopharmacol Biol Psychiatry 15: 595–600. Bao AM, Swaab DF (2019). The human hypothalamus in mood disorders: the HPA axis in the center. IBRO Rep 6: 45–53. Baxter AJ, Scott KM, Vos T et al. (2013). Global prevalence of anxiety disorders: a systematic review and meta-regression. Psychol Med 43: 897–910. Binder EB, Nemeroff CB (2010). The CRF system, stress, depression and anxiety-insights from human genetic studies. Mol Psychiatry 15: 574–588. Brambilla F, Bellodi L, Perna G et al. (1992). Psychoimmunoendocrine aspects of panic disorder. Neuropsychobiology 26: 12–22. Carpenter JK, Andrews LA, Witcraft SM et al. (2018). Cognitive behavioral therapy for anxiety and related disorders: a meta-analysis of randomized placebo-controlled trials. Depress Anxiety 35: 502–514. Condren RM, O’Neill A, Ryan MC et al. (2002). HPA axis response to a psychological stressor in generalised social phobia. Psychoneuroendocrinology 27: 693–703. Coric V, Feldman HH, Oren DA et al. (2010). Multicenter, randomized, double-blind, active comparator and

158

S. FISCHER

placebo-controlled trial of a corticotropin-releasing factor receptor-1 antagonist in generalized anxiety disorder. Depress Anxiety 27: 417–425. Craske MG, Stein MB (2016). Anxiety. Lancet 388: 3048–3059. Curtis GC, Abelson JL, Gold PW (1997). Adrenocorticotropic hormone and cortisol responses to corticotropin-releasing hormone: changes in panic disorder and effects of alprazolam treatment. Biol Psychiatry 41: 76–85. De Cagna F, Fusar-Poli L, Damiani S et al. (2019). The role of intranasal oxytocin in anxiety and depressive disorders: a systematic review of randomized controlled trials. Clin Psychopharmacol Neurosci 17: 1–11. de Quervain D, Schwabe L, Roozendaal B (2017). Stress, glucocorticoids and memory: implications for treating fearrelated disorders. Nat Rev Neurosci 18: 7–19. Deussing JM, Chen A (2018). The corticotropin-releasing factor family: physiology of the stress response. Physiol Rev 98: 2225–2286. Dodhia S, Hosanagar A, Fitzgerald DA et al. (2014). Modulation of resting-state amygdala-frontal functional connectivity by oxytocin in generalized social anxiety disorder. Neuropsychopharmacology 39: 2061–2069. Eckstein M, Becker B, Scheele D et al. (2015). Oxytocin facilitates the extinction of conditioned fear in humans. Biol Psychiatry 78: 194–202. Elnazer HY, Baldwin DS (2014). Investigation of cortisol levels in patients with anxiety disorders: a structured review. Curr Top Behav Neurosci 18: 191–216. Engel S, Laufer S, Miller R et al. (2019). Demographic, sampling- and assay-related confounders of endogenous oxytocin concentrations: a systematic review and metaanalysis. Front Neuroendocrinol 54: 100775. Erhardt A, Ising M, Unschuld PG et al. (2006). Regulation of the hypothalamic-pituitary-adrenocortical system in patients with panic disorder. Neuropsychopharmacology 31: 2515–2522. Fang A, Hoge EA, Heinrichs M et al. (2014). Attachment style moderates the effects of oxytocin on social behaviors and cognitions during social rejection: applying an RDoC framework to social anxiety. Clin Psychol Sci 2: 740–747. Fang A, Treadway MT, Hofmann SG (2017). Working hard for oneself or others: effects of oxytocin on reward motivation in social anxiety disorder. Biol Psychol 127: 157–162. Fekete C, Lechan RM (2014). Central regulation of hypothalamic-pituitary-thyroid axis under physiological and pathophysiological conditions. Endocr Rev 35: 159–194. Fischer S, Cleare AJ (2017). Cortisol as a predictor of psychological therapy response in anxiety disorders—systematic review and meta-analysis. J Anxiety Disord 47: 60–68. Fischer S, Ehlert U (2018). Hypothalamic-pituitary-thyroid (HPT) axis functioning in anxiety disorders. A systematic review. Depress Anxiety 35: 98–110. Fossey MD, Lydiard RB, Ballenger JC et al. (1993). Cerebrospinal fluid thyrotropin-releasing hormone concentrations in patients with anxiety disorders. J Neuropsychiatr Clin Neurosci 5: 335–337.

Fossey MD, Lydiard RB, Ballenger JC et al. (1996). Cerebrospinal fluid corticotropin-releasing factor concentrations in patients with anxiety disorders and normal comparison subjects. Biol Psychiatry 39: 703–707. Frohlich E, Wahl R (2019). The forgotten effects of thyrotropinreleasing hormone: metabolic functions and medical applications. Front Neuroendocrinol 52: 29–43. Gorka SM, Fitzgerald DA, Labuschagne I et al. (2015). Oxytocin modulation of amygdala functional connectivity to fearful faces in generalized social anxiety disorder. Neuropsychopharmacology 40: 278–286. Gottschalk MG, Domschke K (2018). Oxytocin and anxiety disorders. Curr Top Behav Neurosci 35: 467–498. Guastella AJ, Howard AL, Dadds MR et al. (2009). A randomized controlled trial of intranasal oxytocin as an adjunct to exposure therapy for social anxiety disorder. Psychoneuroendocrinology 34: 917–923. Gurguis GN, Mefford IN, Uhde TW (1991). Hypothalamicpituitary-adrenocortical activity in panic disorder: relationship to plasma catecholamine metabolites. Biol Psychiatry 30: 502–506. Hasan MT, Althammer F, Silva da Gouveia M et al. (2019). A fear memory engram and its plasticity in the hypothalamic oxytocin system. Neuron 103: 133–146. Hofmann PJ, Nutzinger DO, Kotter MR et al. (2001). The hypothalamic-pituitary-thyroid axis in agoraphobia, panic disorder, major depression and normal controls. J Affect Disord 66: 75–77. Hoge EA, Lawson EA, Metcalf CA et al. (2012). Plasma oxytocin immunoreactive products and response to trust in patients with social anxiety disorder. Depress Anxiety 29: 924–930. Hoge EA, Pollack MH, Kaufman RE et al. (2008). Oxytocin levels in social anxiety disorder. CNS Neurosci Ther 14: 165–170. Jolkkonen J, Lepola U, Bissette G et al. (1993). CSF corticotropin-releasing factor is not affected in panic disorder. Biol Psychiatry 33: 136–138. Jurek B, Neumann ID (2018). The oxytocin receptor: from intracellular signaling to behavior. Physiol Rev 98: 1805–1908. Kahn RS, Wetzler S, Asnis GM et al. (1991). Pituitary hormone responses to meta-chlorophenylpiperazine in panic disorder and healthy control subjects. Psychiatry Res 37: 25–34. Keck ME, Kern N, Erhardt A et al. (2008). Combined effects of exonic polymorphisms in CRHR1 and AVPR1B genes in a case/control study for panic disorder. Am J Med Genet B Neuropsychiatr Genet 147B: 1196–1204. Kellner M, Knaudt K, Jahn H et al. (1998). Atrial natriuretic hormone in lactate-induced panic attacks: mode of release and endocrine and pathophysiological consequences. J Psychiatr Res 32: 37–48. Kellner M, Schick M, Yassouridis A et al. (2004). Metyrapone tests in patients with panic disorder. Biol Psychiatry 56: 898–900. Klein E, Zinder O, Colin V et al. (1995). Clinical similarity and biological diversity in the response to alprazolam in

THE HYPOTHALAMUS IN ANXIETY DISORDERS patients with panic disorder and generalized anxiety disorder. Acta Psychiatr Scand 92: 399–408. Knobloch HS, Charlet A, Hoffmann LC et al. (2012). Evoked axonal oxytocin release in the central amygdala attenuates fear response. Neuron 73: 553–566. Konnopka A, Leichsenring F, Leibing E et al. (2009). Cost-ofillness studies and cost-effectiveness analyses in anxiety disorders: a systematic review. J Affect Disord 114: 14–31. Labuschagne I, Phan KL, Wood A et al. (2010). Oxytocin attenuates amygdala reactivity to fear in generalized social anxiety disorder. Neuropsychopharmacology 35: 2403–2413. Labuschagne I, Phan KL, Wood A et al. (2012). Medial frontal hyperactivity to sad faces in generalized social anxiety disorder and modulation by oxytocin. Int J Neuropsychopharmacol 15: 883–896. Leppanen J, Ng KW, Tchanturia K et al. (2017). Meta-analysis of the effects of intranasal oxytocin on interpretation and expression of emotions. Neurosci Biobehav Rev 78: 125–144. Lesch KP, Wiesmann M, Hoh A et al. (1992). 5-HT1A receptor-effector system responsivity in panic disorder. Psychopharmacology 106: 111–117. MacDonald K, Feifel D (2014). Oxytocin’s role in anxiety: a critical appraisal. Brain Res 1580: 22–56. Neumann ID, Slattery DA (2016). Oxytocin in general anxiety and social fear: a translational approach. Biol Psychiatry 79: 213–221. NICE (2011). Generalised anxiety disorder and panic disorder in adults: management [Online]. Available: https://www. nice.org.uk/guidance/cg113. NICE (2013). Social anxiety disorder: recognition, assessment and treatment [Online]. Available: https://www.nice.org. uk/guidance/cg159. Onodera M, Ishitobi Y, Tanaka Y et al. (2015). Genetic association of the oxytocin receptor genes with panic, major depressive disorder, and social anxiety disorder. Psychiatr Genet 25: 212. Ortiga-Carvalho TM, Chiamolera MI, Pazos-Moura CC et al. (2016). Hypothalamus-pituitary-thyroid axis. Compr Physiol 6: 1387–1428. Perlis RH, Fijal B, Dharia S et al. (2013). Pharmacogenetic investigation of response to duloxetine treatment in generalized anxiety disorder. Pharm J 13: 280–285. Petrowski K, Wintermann GB, Kirschbaum C et al. (2012). Dissociation between ACTH and cortisol response in DEX-CRH test in patients with panic disorder. Psychoneuroendocrinology 37: 1199–1208. Petrowski K, Wintermann GB, Schaarschmidt M et al. (2013). Blunted salivary and plasma cortisol response in patients with panic disorder under psychosocial stress. Int J Psychophysiol 88: 35–39. Risbrough VB, Stein MB (2006). Role of corticotropin releasing factor in anxiety disorders: a translational research perspective. Horm Behav 50: 550–561. Roy-Byrne PP, Cowley DS, Hommer D et al. (1991). Neuroendocrine effects of diazepam in panic and generalized anxiety disorders. Biol Psychiatry 30: 73–80. Schartner C, Ziegler C, Schiele MA et al. (2017). CRHR1 promoter hypomethylation: an epigenetic readout of panic disorder? Eur Neuropsychopharmacol 27: 360–371.

159

Schreiber W, Lauer CJ, Krumrey K et al. (1996). Dysregulation of the hypothalamic-pituitary-adrenocortical system in panic disorder. Neuropsychopharmacology 15: 7–15. Siegmund A, Koster L, Meves AM et al. (2011). Stress hormones during flooding therapy and their relationship to therapy outcome in patients with panic disorder and agoraphobia. J Psychiatr Res 45: 339–346. Simon NM, Blacker D, Korbly NB et al. (2002). Hypothyroidism and hyperthyroidism in anxiety disorders revisited: new data and literature review. J Affect Disord 69: 209–217. Stein MB, Muir-Nash J, Uhde TW (1991). The QKd interval in panic disorder: an assessment of end-organ thyroid hormone responsivity. Biol Psychiatry 29: 1209–1214. Stein MB, Uhde TW (1991). Endocrine, cardiovascular, and behavioral-effects of intravenous protirelin in patients with panic disorder. Arch Gen Psychiatry 48: 148–156. Strohle A, Kellner M, Holsboer F et al. (1999). Behavioral, neuroendocrine, and cardiovascular response to flumazenil: no evidence for an altered benzodiazepine receptor sensitivity in panic disorder. Biol Psychiatry 45: 321–326. Tancer ME, Stein MB, Gelernter CS et al. (1990). The hypothalamic-pituitary-thyroid axis in social phobia. Am J Psychiatry 147: 929–933. Tharmalingam S, King N, De Luca V et al. (2006). Lack of association between the corticotrophin-releasing hormone receptor 2 gene and panic disorder. Psychiatr Genet 16: 93–97. Tukel R, Kora K, Hekim N et al. (1999). Thyrotropin stimulating hormone response to thyrotropin releasing hormone in patients with panic disorder. Psychoneuroendocrinology 24: 155–160. Valstad M, Alvares GA, Egknud M et al. (2017). The correlation between central and peripheral oxytocin concentrations: a systematic review and meta-analysis. Neurosci Biobehav Rev 78: 117–124. van Duinen MA, Schruers KR, Maes M et al. (2007). CO2 challenge induced HPA axis activation in panic. Int J Neuropsychopharmacol 10: 797–804. Van IJzendoorn MH, Bakermans-Kranenburg MJ (2012). A sniff of trust: meta-analysis of the effects of intranasal oxytocin administration on face recognition, trust to in-group, and trust to out-group. Psychoneuroendocrinology 37: 438–443. Wang W, Feng J, Ji C et al. (2017). Increased methylation of glucocorticoid receptor gene promoter 1F in peripheral blood of patients with generalized anxiety disorder. J Psychiatr Res 91: 18–25. Weber H, Richter J, Straube B et al. (2016). Allelic variation in CRHR1 predisposes to panic disorder: evidence for biased fear processing. Mol Psychiatry 21: 813–822. Whiteford HA, Degenhardt L, Rehm J et al. (2013). Global burden of disease attributable to mental and substance use disorders: findings from the Global Burden of Disease Study 2010. Lancet 382: 1575–1586. Wichmann S, Bornstein SR, Lorenz T et al. (2018). Stress hormone response to the DEX-CRH test and its relation to psychotherapy outcome in panic disorder patients with and without agoraphobia. Transl Psychiatry 8: 37.

160

S. FISCHER

Wichmann S, Kirschbaum C, Bohme C et al. (2017a). Cortisol stress response in post-traumatic stress disorder, panic disorder, and major depressive disorder patients. Psychoneuroendocrinology 83: 135–141. Wichmann S, Kirschbaum C, Lorenz T et al. (2017b). Effects of the cortisol stress response on the psychotherapy outcome of panic disorder patients. Psychoneuroendocrinology 77: 9–17. Wiedemann K, Jahn H, Yassouridis A et al. (2001). Anxiolyticlike effects of atrial natriuretic peptide on

cholecystokinin tetrapeptide-induced panic attacks: preliminary findings. Arch Gen Psychiatry 58: 371–377. Young EA, Abelson JL, Cameron OG (2004). Effect of comorbid anxiety disorders on the hypothalamic-pituitaryadrenal axis response to a social stressor in major depression. Biol Psychiatry 56: 113–120. Ziegler C, Dannlowski U, Brauer D et al. (2015). Oxytocin receptor gene methylation: converging multilevel evidence for a role in social anxiety. Neuropsychopharmacology 40: 1528–1538.

Handbook of Clinical Neurology, Vol. 180 (3rd series) The Human Hypothalamus: Middle and Posterior Region D.F. Swaab, F. Kreier, P.J. Lucassen, A. Salehi, and R.M. Buijs, Editors https://doi.org/10.1016/B978-0-12-820107-7.00010-0 Copyright © 2021 Elsevier B.V. All rights reserved

Chapter 10

Congenital isolated central hypothyroidism: Novel mutations and their functional implications ANITA BOELEN1, A.S. PAUL VAN TROTSENBURG2, AND ERIC FLIERS3* 1

Laboratory of Endocrinology, Department of Clinical Chemistry, Amsterdam Gastroenterology, Endocrinology, and Metabolism, Amsterdam UMC, University of Amsterdam, Amsterdam, The Netherlands 2

Department of Pediatric Endocrinology, Emma Children’s Hospital, Amsterdam UMC, University of Amsterdam, Amsterdam, The Netherlands

3

Department of Endocrinology and Metabolism, Amsterdam Gastroenterology, Endocrinology, and Metabolism, Amsterdam UMC, University of Amsterdam, Amsterdam, The Netherlands

Abstract Congenital hypothyroidism is the most frequent endocrine disorder in newborns, occurring in 1 per 3000–4000 newborns. In the Netherlands, the neonatal screening program is based primarily on heel prick thyroxine (T4). In contrast to thyroid-stimulating hormone-based programs, this approach allows for the detection of both primary and central congenital hypothyroidism. Over the past decade, the identification of families with isolated congenital central hypothyroidism enabled the identification of novel genetic causes of this condition, in addition to mutations in the TSHb-subunit gene and thyrotropin-releasing hormone receptor gene reported earlier. In 2012, loss-of-function mutations in the immunoglobulin superfamily, member 1 (IGSF1) gene, were reported as a genetic cause of a syndrome including X-linked congenital central hypothyroidism and adult macroorchidism. IGSF1 encodes a hypothalamic plasma membrane glycoprotein. Mutations in IGSF1 represent the most prevalent genetic cause of isolated central hypothyroidism to date. In 2016, mutations in the transducin b-like 1X (TBL1X) gene were identified in patients with a combination of mild central hypothyroidism and sensorineural hearing loss. TBL1X is an essential subunit of the NCoR/SMRT corepressor complex and expressed in many tissues including the human hypothalamus and pituitary. In 2018, mutations in the insulin receptor substrate 4 (IRS4) gene were reported in cases of familial isolated central hypothyroidism. IRS4 encodes a hypothalamic protein that is part of the insulin and leptin signaling cascade. These recent developments will broaden our understanding of the role of the hypothalamus in hypothalamus–pituitary–thyroid axis regulation and will help to improve diagnosis and treatment of isolated central hypothyroidism.

INTRODUCTION The hypothalamus–pituitary–thyroid axis Serum free thyroxine (FT4) levels are kept within a narrow range by the hypothalamic–pituitary–thyroid (HPT) axis via a classical multiple-loop negative feedback pathway. The tripeptide thyrotropin-releasing hormone (TRH) is

produced by neurons in the hypothalamic paraventricular nucleus. A subset of these neurons, the hypophysiotropic TRH neurons, project to the median eminence where TRH is released and transported by the hypophyseal portal vessel to the anterior pituitary gland. Upon delivery to the anterior pituitary, TRH stimulates the production and release of thyroid-stimulating hormone (TSH). TSH is a

*Correspondence to: Eric Fliers, Department of Endocrinology and Metabolism, Amsterdam Gastroenterology, Endocrinology, and Metabolism, Amsterdam UMC, University of Amsterdam, Amsterdam, The Netherlands. Tel: 0031-20-566-6071, Fax: 0031-20-691-7682, E-mail: [email protected]

162

A. BOELEN ET AL.

glycoprotein consisting of a unique beta-subunit and an alpha-subunit that is shared with luteinizing hormone (LH) and follicle-stimulating hormone. TSH increases the uptake of iodine in the thyroid, ultimately leading to increased production of T4 and triiodothyronine (T3) (Zoeller et al., 2007). The production and release of both TRH and TSH is stimulated by decreasing levels of circulating thyroid hormones and inhibited by increasing levels of thyroid hormones, leading to tightly controlled circulating thyroid hormone levels. The thyroid hormone concentration in a healthy individual barely fluctuates within the population reference interval throughout life, indicating that the HPT axis strictly maintains an individual optimal concentration, the so-called HPT axis set-point (Hansen et al., 2004; Medici et al., 2015; Zwaveling-Soonawala et al., 2015a). Optimal thyroid hormone concentrations may change, however, during exceptional circumstances including food deprivation and disease. This may explain why, in addition to thyroid hormone levels, TRH is regulated by physiological factors, including the availability of food, time of day, environmental temperature, and inflammation; with that, the HPT axis set-point is flexible ensuring an optimal adaptation to the environment (Zoeller et al., 2007; Fliers et al., 2014).

Cellular thyroid hormone metabolism and action While serum FT4 represents the overall activity of the HPT axis, thyroid hormone availability and action can vary at the level of individual tissues and organs, resulting from differential expression of the iodothyronine deiodinase enzymes types 1, 2, and 3 (D1, D2, and D3, respectively) and the thyroid hormone receptor (TR). Deiodinases are selenoproteins that convert thyroid hormones by removing iodine molecules from either the inner or outer ring of the substrate. D2 is known as the activating deiodinase, converting the prohormone T4 into biologically active T3 by outer ring deiodination. Conversely, D3 acts as an inactivating deiodinase, converting T4 into biologically inactive reverse T3 (rT3) and T3 into inactive T2 by inner ring deiodination. Finally, D1 is able to deiodinate both the inner and outer ring. The preferred substrate of D1 is rT3 and—to a lesser extent—T4. For many years, D1 was considered the major T3 producing enzyme, but recent studies suggest that D1 is more important for thyroid hormone clearance during hyperthyroidism (Bianco and Kim, 2006). Intracellular activity of T3 may vary widely depending on the deiodinase type and level of expression (Bianco et al., 2005). While some tissues, such as the liver, mainly depend on circulating thyroid hormone concentrations, other tissues—including muscle and brain—depend on the deiodinase enzymes to provide tissue-specific

regulation of intracellular T3 concentrations. To some extent, this occurs independent of systemic thyroid hormone levels. The ability to generate locally varying concentrations of intracellular T3 allows for acute or chronic adaptation to circumstances with extreme metabolic requirements, such as starvation, cold, or illness (Bianco et al., 2005; Boelen et al., 2011). The effects of T3 in the cell nucleus are mediated by the thyroid hormone receptors alpha and beta. TRs and their isoforms have a highly tissue-specific distribution. The TR-alpha isoform TRa1 is widely expressed, most abundantly in the central nervous system, gastro-intestinal tract, bone and skeletal muscle. While TRa2 is expressed in many tissues including brain, it is unable to bind thyroid hormone, and its biological function is unclear. TRb1 is expressed predominantly in the liver and kidney, while TRb2 is expressed exclusively in hypothalamus, pituitary, inner ear, and retina (Moran and Chatterjee, 2015). Following binding of T3 to the TRs with high affinity and specificity, the TRs in turn bind to regulatory DNA sequences known as thyroid response elements (TREs) in promotor regions of thyroid hormone target genes. Depending on whether the ligand T3 is present or not, TRs recruit coactivators or corepressors, serving as coregulatory factors for either transcriptional gene activation or silencing. TREs may be either positively or negatively regulated, depending on whether binding of T3 induces or inhibits gene transcription. When unbound, positive TREs associate with corepressors, which inhibit gene transcription by inducing histone deacetylation. In the presence of T3, the corepressors dissociate and instead coactivators are recruited, which activate gene transcription by inducing histone acetylation (Hollenberg et al., 1995; Tagami et al., 1997). In genes with negative TREs, including TSHB and TRH, corepressors are critical in activation of gene transcription in absence of T3, although the precise mechanism is still unknown (Tagami et al., 1997).

HYPOTHYROIDISM Hypothyroidism is characterized by a decreased production or secretion of thyroid hormones. During hypothyroidism, a wide range of metabolic processes is slowed down, and the clinical manifestations reflect this decreased metabolism (Braverman and Cooper, 2013). Hypothyroid patients may complain of fatigue, decreased concentration and mood, cold intolerance, constipation, weight gain, hair loss, and facial swelling. Symptoms commonly progress slowly and can go unnoticed for a long time (Vaidya and Pearce, 2008). Hypothyroidism is most often due to abnormalities of the thyroid gland, which is referred to as primary hypothyroidism. Primary hypothyroidism is mostly an acquired condition, developing in the context of autoimmunity or treatments that destroy thyroid tissue. In adults,

CONGENITAL ISOLATED CENTRAL HYPOTHYROIDISM it is a common endocrine disorder, with a female-to-male ratio of 4 to 1. In a small minority of patients, it may occur as a congenital disorder, caused either by developmental abnormalities of the thyroid gland (dys- or agenesis) or by hormone synthesis disorders (dyshormonogenesis) (Rastogi and Lafranchi, 2010; Gruters and Krude, 2011). Even more rarely, hypothyroidism stems from an acquired or congenital dysfunction of the pituitary or hypothalamus, which is referred to as central hypothyroidism (Beck-Peccoz et al., 2017). Central hypothyroidism is characterized by insufficient stimulation by TSH of an otherwise normal thyroid gland. TSH deficiency may be qualitative or quantitative. As damage or malformation of the pituitary usually affects the entire gland, central hypothyroidism is most often accompanied by multiple pituitary hormone deficiency. However, in rare cases, the hypothyroidism occurs isolated (Persani, 2012).

Congenital hypothyroidism Congenital hypothyroidism is the most frequent endocrine disorder in newborns (Weiss and Refetoff, 2015), occurring in 1 per 3000–4000 newborns (Fisher, 1983). As thyroid hormone is essential for the development of the brain in the first years of life, undiagnosed congenital hypothyroidism is one of the most common preventable causes of mental retardation (Lafranchi, 1999). Since maternal thyroid hormones cross the placenta to some extent, and most neonates have residual functioning of their own thyroid gland, the development of the fetus is somewhat protected during pregnancy. After birth, however, the deficiency will result in progressive damage. Neonates with congenital hypothyroidism may present with signs and symptoms like prolonged jaundice, feeding difficulties, lethargy, or umbilical hernia (Alm et al., 1984). However, signs and symptoms are often subtle or even absent, leaving the congenital hypothyroidism unnoticed (Lafranchi, 1979). The most common cause of primary congenital Table 10.1

Candidate genes and related phenotypes associated with isolated congenital central hypothyroidism. Gene

Protein

Inheritance

Endocrine phenotype

Associated conditions

TSHb TRHR IGSF1

TSH TRH receptor IGSF1

AD AD XLa

TBL1X IRS4

TBL1X IRS4

XL XL

Isolated CH Isolated CH Isolated CH +/ PRL, transient GHD Mild isolated CH Mild isolated CH

— — Macroorchidism Ovarian cysts Hearing defect

a

163

hypothyroidism is thyroid dysgenesis, a group of entities including absence (thyroid agenesis), hypoplastic development (thyroid hypoplasia), and misplacement (thyroid ectopy) of the thyroid gland. Approximately 80%–85% of cases of congenital primary hypothyroidism is due to dysgenesis, with a third due to thyroid ectopy (Fisher, 1983). Mutations in several genes are associated with dysgenesis, including genes coding for transcription factors involved in thyroid gland morphogenesis (PAX8, NKX2–1, FOXE1, NKX2–5, and HHEX), and thyroid differentiation (TSHR). A less common cause of congenital primary hypothyroidism are inborn errors of T4 synthesis or dyshormonogenesis, accounting for approximately 10% of cases. Such a defect may occur during any step of thyroid hormone synthesis or secretion, including iodine trapping, oxidation and organification, and coupling (Lafranchi, 1999). The final 5% of cases include transient cases caused by maternal TSH receptor blocking antibodies or iodine excess (Brown et al., 1996). In contrast to congenital primary hypothyroidism, congenital central hypothyroidism is due to developmental disorders of the hypothalamus and/or the pituitary. The development of the pituitary gland requires expression of a number of transcription factors and signaling molecules, prompting a complex signaling cascade. Several mutations in genes coding for proteins involved in this signaling cascade are known to disrupt development of the normal pituitary and lead to multiple pituitary hormone deficiency. Depending on the involved gene, additional syndromic features may occur. However, only a minority of patients with multiple pituitary hormone deficiency have known mutations, including POU1F1, PROP1, HESX1, LHX3, LHX4, SOX3, and OTX2 (Schoenmakers et al., 2015). Until recently, the genetic basis of isolated congenital central hypothyroidism was largely unknown, with only a few reported families with mutations in the TSH b-subunit gene (TSHB) or in the TRH receptor gene (TRHR). However, in the past decade, several additional genetic causes have been reported (Table 10.1).

One-third of females affected to some degree. AD, autosomal dominant; CH, central hypothyroidism; GHD, growth hormone deficiency; PRL, prolactin deficiency; XL, X-linked.

164

A. BOELEN ET AL.

GENETICS OF CONGENITAL ISOLATED CENTRAL HYPOTHYROIDISM TSH b-subunit gene Mutations in the TSH b-subunit gene cause severe congenital central hypothyroidism. Patients have elevated a-subunit concentrations, an impaired TSH response to TRH, and a hyperplastic pituitary gland on MRI. Although serum TSH concentrations may be within reference range, the profound hypothyroid state of affected patients indicates a total inability to stimulate the TSH receptor (Bonomi et al., 2001). These mutations were first described in 1989 in patients with congenital central hypothyroidism (Hayashizaki et al., 1989). Since then, several additional naturally occurring mutations in TSHB have been reported worldwide, and all seem to disrupt the specific “seatbelt” configuration of the b-subunit around the a-subunit (Bonomi et al., 2001).

TRH receptor gene A far less common genetic cause of congenital central hypothyroidism are mutations in the TRH receptor gene. Since the first description by Collu et al. (1997), only a handful of cases have been described. While absence of TRH signaling in the pituitary should significantly hamper the production of TSH, these patients have TSH concentrations within reference range. Additionally, cases that were detected at a later age had normal neurological development, suggesting that sufficient thyroid hormones were produced to guarantee a normal brain development in childhood. Clinical manifestations predominantly concern growth abnormalities. Currently, there is no evidence for an extrapituitary phenotype in patients with TRHR mutations (Schoenmakers et al., 2015).

IGSF1 gene More recently, loss-of-function mutations in the immunoglobulin superfamily, member 1 (IGSF1), gene were described as a genetic cause of congenital central hypothyroidism by Sun et al. (2012). In males, the phenotype consists of congenital central hypothyroidism, delayed rise of serum testosterone in puberty, adult macroorchidism, partial GH deficiency, and obesity. Although it is an X-linked disease, a minority of female heterozygous carriers are affected. IGSF1 encodes a hypothalamic plasma membrane glycoprotein, and all described mutations impair proper glycosylation and trafficking of the protein to the cell surface (Sun et al., 2012). Within 3 years after the discovery of this syndrome, 30 mutation carrying families were described (Joustra et al., 2013), making it the most common genetic cause of isolated central hypothyroidism of date. The specific function of IGSF1

is still unknown, and the pathophysiology of mutations is unclear. A recent study suggests that loss of IGSF1 may lead to an impairment in TRH1 expression and TRH action (Bernard et al., 2018); however, an altered thyroid hormone feedback and/or signaling have also been suggested (Joustra et al., 2015).

TBL1X gene Recently, mutations in the transducin b-like 1X (TBL1X) gene were identified in patients with mild congenital central hypothyroidism (Heinen et al., 2016). TBL1X is a WD40 repeat-containing protein and an essential subunit of the NCoR/SMRT corepressor complex, which is the most important thyroid hormone receptor corepressor complex. This complex is responsible for regulating gene expression of T3 target genes, depending on whether T3 is absent or present (Astapova et al., 2011). Mutations in TBL1X result in relatively mild disease, consisting of congenital isolated central hypothyroidism and sensorineural hearing loss. Mice with mutations in NCoR, resulting in a defective NCoR/ SMRT corepressor complex, display low FT4 concentrations in combination with normal TSH concentrations, suggesting the presence of central hypothyroidism (Astapova et al., 2011). These findings indicate that normal functioning of the NCoR/SMRT corepressor complex is vital for HPT axis regulation. So far, all mutations in TBL1X were located in the protein’s WD40 domain (Fig. 10.1). This domain is thought to be responsible for tethering the NCoR/SMRT corepressor complex to the histones in the DNA, thereby allowing for acetylation or deacetylation (Yoon et al., 2003). Most likely, these mutations cause an inability of the entire corepressor complex to properly adhere to the DNA, and the complex might thereby be unable to initiate its effects on the histones. While this remains unconfirmed at this stage, the function of TBL1X makes it quite likely that the isolated central hypothyroidism results from the described mutations. Several patients with mutations in TBL1X were only diagnosed with central hypothyroidism as teenagers or adults. Although their biochemical hypothyroidism was untreated during the first 3 years of life and during puberty, mental and physical development was normal, although not formally tested. All individuals reached normal adult heights, and several affected female mutation carriers conceived and carried their pregnancies fullterm without difficulties. This suggests an absence of severe thyroid hormone deficiency at the cellular level, and the necessity of thyroid hormone replacement therapy remains uncertain at this stage. However, one of these women showed visible improvement of myxedema, concentration, and energy level after starting

CONGENITAL ISOLATED CENTRAL HYPOTHYROIDISM

165

Fig. 10.1. Schematic representations of the mutated amino acids. The mutated amino acids are shown on the crystal structure of the TBL1XR1 WD40 domain (PDB ID 4LG9). In each case, the mutated residues are shown in orange, TBL1XR1 in gray, and water molecules in the crystal structure in cyan. The numbering of the amino acids is as for TBL1X. (A) N365, (B) H453, (C) W369, and (D) c.1312-1G > A splice mutation with the missing amino acids in cyan (starting at asterisk). (E) Surface representation of the WD40 domain to show the mutations that are on the surface. (F) Transparent representation of the WD40 domain to show the buried and surface mutations. Source: Heinen, C.A., Losekoot, M., Sun, Y., et al., 2016. Mutations in TBL1X are associated with central hypothyroidism. J Clin Endocrinol Metab 101, 4564–4573.

levothyroxine replacement therapy, suggesting she had been experiencing thyroid hormone deficiency (Heinen et al., 2016).

IRS4 gene The insulin receptor substrate 4 (IRS4) gene is another intriguing candidate to be involved in the pathogenesis of central hypothyroidism; four nonsense mutations (one nonsense, three frame-shift) in IRS4 have been identified in a number of male patients with isolated

central hypothyroidism from five families. All male carriers had central hypothyroidism with plasma free thyroxine (FT4) concentrations below the reference interval. Magnetic resonance imaging of the hypothalamus and pituitary showed no structural abnormalities (Heinen et al., 2018). All carriers had relatively low thyroid volumes, in keeping with relative TSH deficiency. IRS4 is part of the insulin receptor substrate (IRS) family, which encompasses six members (IRS1–6). Proteins of the IRS family contain phosphotyrosine-binding domains and multiple glycosylation sites necessary in

166

A. BOELEN ET AL.

hormone signaling pathways. Indeed, IRS1 and IRS2 are involved in insulin signaling while IRS3 and IRS4 were found to influence actions of IRS1 and IRS2 (Araki et al., 1994; Withers et al., 1998). IRS4 is located on the X-chromosome and encodes a 1257 amino-acid protein that is activated by the insulin, IGF-1, and leptin receptor. IRS4 mRNA is expressed in a variety of tissues, including the pituitary gland and hypothalamus (Heinen et al., 2018). At present, the mechanism underlying the central hypothyroidism in patients carrying IRS4 mutations is unclear, but it is tempting to speculate about a role for hypothalamic leptin signaling, which is closely associated with central regulation of thyroid function. Leptin is known to exert its role on the HPT axis via binding to the leptin receptor on hypophysiotropic TRH neurons. It also has indirect actions on TRH neurons via the leptin receptor on proopiomelanocortin and neuropeptide Y/agouti-related peptide expressing neurons in the hypothalamic arcuate nucleus that project to hypothalamic TRH neurons (Legradi et al., 1998). Leptin is able to restore the fasting-induced suppression of TSH pulsatility in healthy persons (Chan et al., 2003). Of note, suppressed 24-h TSH secretion profiles were also observed in two adult male CH patients carrying an IRS4 mutation. This suggests that the central downregulation of the HPT axis in patients with IRS4 mutations may be mediated by a subtle disturbance of hypothalamic leptin signaling. Patients with mutations in the leptin receptor display congenital central hypothyroidism but also morbid obesity, growth hormone deficiency, and hypogonadotropic hypogonadism (Farooqi et al., 2007). These differences may be due to the restricted expression pattern of IRS4 compared to the broader expression of the leptin receptor.

CENTRAL HYPOTHYROIDISM AND NEONATAL SCREENING The discovery that the IQ of a child with congenital hypothyroidism depends on the timing of its detection and the timely start of levothyroxine replacement therapy prompted the development of neonatal screening programs. These programs have now been implemented in many countries worldwide. The most commonly used screening strategy is measurement of TSH. This method is excellent for the detection of primary hypothyroidism but does not detect central hypothyroidism. Other strategies either consecutively or simultaneously measure T4 and TSH and detect both primary and central hypothyroidism (Gruters and Krude, 2011). A primary T4-based screening is less sensitive and therefore less accepted; low T4 levels in neonates can be caused by several other factors than central hypothyroidism, such as prematurity, severe illness, or thyroxine-binding globulin (TBG) deficiency. Another argument against primary

T4 screening is based on the assumption that central hypothyroidism is only a mild disease, which is not likely associated with an increased risk of damage to brain development and does not warrant immediate detection (Price and Weetman, 2001; Lafranchi, 2011). Finally, the prevalence of congenital central hypothyroidism is considered so low that adaptation of the screening would not be cost effective (Asakura et al., 2002; Lanting et al., 2005). In contrast, in 2015, Zwaveling et al. demonstrated that the severity of central hypothyroidism as assessed by the level of initial serum FT4 can be classified as moderate in over half of patients with central hypothyroidism, very similar to those with primary hypothyroidism (Zwaveling-Soonawala et al., 2015b), suggesting that central hypothyroidism can be no less damaging than primary hypothyroidism. More importantly, central hypothyroidism is associated with other pituitary hormone deficiencies in 80% of cases, including growth hormone and adrenocorticotropic hormone, resulting in growth hormone deficiency and central adrenal insufficiency. These deficiencies are not easily identifiable based solely on clinical characteristics, and they are associated with acute life-threatening illness (Persani, 2012). Van Tijn et al. found that delayed detection of these neonates resulted in significant morbidity, including hypoglycemia and neonatal hepatitis. None of these patients had symptoms indicating pituitary dysfunction either in the neonatal period, or during follow-up (Van Tijn et al., 2005). One effective method of finding these neonates is through neonatal screening for central hypothyroidism. To combat the high numbers of false-positives in primary T4 screening programs, the Netherlands developed a unique three-step screening. In 1995, after a 1-year trial period, the Dutch neonatal screening was extended from an initial T4 and subsequent TSH measurement by adding determination of TBG. Dried blood spots are obtained via heel puncture 3–7 days after birth, and T4 is determined in all. TSH is measured in the lowest 20% T4 concentrations, and TBG in the lowest 5%. The ratio between T4 and TBG is then used as indicator of the FT4 concentration. Clearly abnormal results are immediately referred to a pediatrician, while neonates with dubious results undergo a second heel puncture. A second dubious result is also reason for referral to a pediatrician (Van Tijn et al., 2005). This screening method was found to be both clinically effective and cost effective (Lanting et al., 2005). Further studies proved that this method was highly sensitive for detecting both primary and central hypothyroidism, and the incidence of the latter was increased from 1 in 106,394 (Hanna et al., 1986) to 1 in 16,404 since implementation (Van Tijn et al., 2005), making it a very effective screening program.

CONGENITAL ISOLATED CENTRAL HYPOTHYROIDISM The incidence of congenital central hypothyroidism depends on the method of screening. In countries with a TSH-based screening, such as the United States and Italy, the incidence is between 1 in 50,000 and 70,000 (Zamboni et al., 2004; Ford and Lafranchi, 2014). In countries that use programs including both (F)T4 and TSH, such as the Netherlands and Japan, have a far higher incidence between 1 in 14,000 and 22,000 (Lanting et al., 2005; Van Tijn et al., 2005; Fujiwara et al., 2008), suggesting that a (F)T4-based screening method is effective in detecting milder disease (BeckPeccoz et al., 2017).

CONGENITAL CENTRAL HYPOTHYROIDISM—CLINICAL ASPECTS

167

preferred by direct sequencing or by next generation sequencing using a panel of candidate genes. Whole exome or genome sequencing can be considered in sporadic or familial cases with negative candidate gene analysis. If a causative mutation in any candidate gene is found, the analysis should be extended to first-degree relatives (Persani et al., 2018). Genetic testing can also support the diagnosis of very mild forms of central hypothyroidism, when the clinical picture is unclear and the distinction with an FT4 in the low normal or even subnormal range without an endocrine abnormality is difficult to make. Additional findings useful to support the diagnosis of central hypothyroidism in these cases include a positive family history, delayed growth, macroorchidism, impaired hearing, blunted or delayed TSH response to TRH, and a blunted nocturnal TSH surge.

Diagnosis The diagnosis central hypothyroidism should be suspected in all individuals with a subnormal serum FT4 concentration in combination with inappropriately low serum TSH. Although the clinical context in patients with multiple pituitary hormone deficiencies may make the diagnosis of central hypothyroidism rather easy, the situation in isolated central hypothyroidism is more challenging. Of note, reference values for adults cannot be simply used for neonates and children, as serum thyroid hormone serum concentrations change in the first weeks to months of life. In a recent study performed to establish neonatal FT4 and TSH reference intervals, the lower limit of the FT4 reference interval was found to be higher than previously thought (Naafs et al., 2020). Although most patients with central hypothyroidism have low FT4 levels in combination with low or normal TSH, some patients may have high serum immunoreactive TSH concentrations devoid of full biological activity (Faglia et al., 1979). This situation may lead to misdiagnosis. Other sources of erroneous diagnosis of central hypothyroidism may occur in patients in whom FT4 concentrations are low due to a variety of other reasons. These include nonthyroidal illness (Boelen et al., 2011), isolated maternal hypothyroxinemia, recovery from recent thyrotoxicosis, assay interference, a wide range of drugs including glucocorticoids and dopamine, and premature birth (Persani et al., 2018). Once the diagnosis central hypothyroidism is established, an important question is in which patients with congenital central hypothyroidism genetic analysis should be performed. The 2018 European Thyroid Association (ETA) Guidelines on the diagnosis and management of central hypothyroidism advises to perform genetic analysis in congenital and familial cases as well as in cases with onset during childhood, whenever the condition remains unexplained. In index cases, genetic analysis based on the phenotype is

Management and treatment When the diagnosis of isolated central hypothyroidism is confirmed, replacement therapy with levothyroxine (LT4) should be started. Of note, in all patients with central hypothyroidism the additional presence of central adrenal insufficiency should be excluded before LT4 therapy is started. Treatment of central hypothyroidism should restore appropriate serum concentrations of thyroid hormones. As combination therapy with LT4 and LT3 was not proven superior to LT4 alone in patients with central hypothyroidism (Slawik et al., 2007), LT4 monotherapy remains the standard treatment. In exceptional cases, combination therapy may be considered as an experimental approach in patients with persistent complaints. However, with TSH as an unreliable monitor in patients with central hypothyroidism, the risk of overtreatment with this approach is higher than in patients with primary hypothyroidism. The determination of serum FT4 is of major significance in monitoring LT4 treatment in patients with central hypothyroidism. Blood should be withdrawn before or at least 4 h after LT4 administration. Several studies reported that FT4 concentrations in the upper part of the normal range represent an appropriate target in most patients, and this was included in the treatment recommendations by the ETA. Pediatric patients should undergo FT4 monitoring according to age-related reference ranges. Annual monitoring is mostly sufficient in adult patients. During pregnancy, a 25%–50% increase of the LT4 dose is advised to minimize the risk of under replacement for the fetus. Likewise, uptitration should be considered after the introduction of growth hormone treatment, estrogen replacement therapy or oral contraceptives, and after weight gain. Conversely, downtitration should be considered in elderly patients, especially in patients with

168

A. BOELEN ET AL.

cardiovascular morbidities, and after parturition or menopause, or after weight loss (Persani et al., 2018).

CONCLUDING REMARKS The identification of families with isolated central hypothyroidism, facilitated by T4-based neonatal screening, has helped to uncover several genes implicated in the pathogenesis of isolated central hypothyroidism. These recent developments will broaden our understanding of the role of the hypothalamus in HPT axis regulation and will help to improve diagnosis and treatment of isolated central hypothyroidism. In the coming years, the prevalence of these genetic variants as well as their clinical implications will have to be delineated.

REFERENCES Alm J, Hagenfeldt L, Larsson A et al. (1984). Incidence of congenital hypothyroidism: retrospective study of neonatal laboratory screening versus clinical symptoms as indicators leading to diagnosis. Br Med J (Clin Res Ed) 289: 1171–1175. Araki E, Lipes MA, Patti ME et al. (1994). Alternative pathway of insulin signalling in mice with targeted disruption of the IRS-1 gene. Nature 372: 186–190. Asakura Y, Tachibana K, Adachi M et al. (2002). Hypothalamo-pituitary hypothyroidism detected by neonatal screening for congenital hypothyroidism using measurement of thyroid-stimulating hormone and thyroxine. Acta Paediatr 91: 172–177. Astapova I, Vella KR, Ramadoss P et al. (2011). The nuclear receptor corepressor (NCoR) controls thyroid hormone sensitivity and the set point of the hypothalamic-pituitarythyroid axis. Mol Endocrinol 25: 212–224. Beck-Peccoz P, Rodari G, Giavoli C et al. (2017). Central hypothyroidism—a neglected thyroid disorder. Nat Rev Endocrinol 13: 588–598. Bernard DJ, Brule E, Smith CL et al. (2018). From consternation to revelation: discovery of a role for igsf1 in pituitary control of thyroid function. J Endocr Soc 2: 220–231. Bianco AC, Kim BW (2006). Deiodinases: implications of the local control of thyroid hormone action. J Clin Invest 116: 2571–2579. Bianco AC, Maia AL, Da Silva WS et al. (2005). Adaptive activation of thyroid hormone and energy expenditure. Biosci Rep 25: 191–208. Boelen A, Kwakkel J, Fliers E (2011). Beyond low plasma T3: local thyroid hormone metabolism during inflammation and infection. Endocr Rev 32: 670–693. Bonomi M, Proverbio MC, Weber G et al. (2001). Hyperplastic pituitary gland, high serum glycoprotein hormone alpha-subunit, and variable circulating thyrotropin (TSH) levels as hallmark of central hypothyroidism due to mutations of the TSH beta gene. J Clin Endocrinol Metab 86: 1600–1604. Braverman LE, Cooper DS (2013). Introduction to hypothyroidism. In: LE Braverman, DS Cooper (Eds.), Werner

& Ingbar’s the thyroid a fundamental and clinical text, 10 edn. Lippincott Williams & Wilkins, Philedelphia, pp. 523–535. Brown RS, Bellisario RL, Botero D et al. (1996). Incidence of transient congenital hypothyroidism due to maternal thyrotropin receptor-blocking antibodies in over one million babies. J Clin Endocrinol Metab 81: 1147–1151. Chan JL, Heist K, Depaoli AM et al. (2003). The role of falling leptin levels in the neuroendocrine and metabolic adaptation to short-term starvation in healthy men. J Clin Invest 111: 1409–1421. Collu R, Tang J, Castagne J et al. (1997). A novel mechanism for isolated central hypothyroidism: inactivating mutations in the thyrotropin-releasing hormone receptor gene. J Clin Endocrinol Metab 82: 1561–1565. Faglia G, Bitensky L, Pinchera A et al. (1979). Thyrotropin secretion in patients with central hypothyroidism: evidence for reduced biological activity of immunoreactive thyrotropin. J Clin Endocrinol Metab 48: 989–998. Farooqi IS, Wangensteen T, Collins S et al. (2007). Clinical and molecular genetic spectrum of congenital deficiency of the leptin receptor. N Engl J Med 356: 237–247. Fisher DA (1983). Second international conference on neonatal thyroid screening: progress report. J Pediatr 102: 653–654. Fliers E, Kalsbeek A, Boelen A (2014). Beyond the fixed setpoint of the hypothalamus-pituitary-thyroid axis. Eur J Endocrinol 171: R197–R208. Ford G, Lafranchi SH (2014). Screening for congenital hypothyroidism: a worldwide view of strategies. Best Pract Res Clin Endocrinol Metab 28: 175–187. Fujiwara F, Fujikura K, Okuhara K et al. (2008). Central congenital hypothyroidism detected by neonatal screening in Sapporo, Japan (2000–2004): it’s prevalence and clinical characteristics. Clin Pediatr Endocrinol 17: 65–69. Gruters A, Krude H (2011). Detection and treatment of congenital hypothyroidism. Nat Rev Endocrinol 8: 104–113. Hanna CE, Krainz PL, Skeels MR et al. (1986). Detection of congenital hypopituitary hypothyroidism: ten-year experience in the Northwest Regional Screening Program. J Pediatr 109: 959–964. Hansen PS, Brix TH, Bennedbaek FN et al. (2004). Genetic and environmental causes of individual differences in thyroid size: a study of healthy Danish twins. J Clin Endocrinol Metab 89: 2071–2077. Hayashizaki Y, Hiraoka Y, Endo Y et al. (1989). Thyroidstimulating hormone (TSH) deficiency caused by a single base substitution in the CAGYC region of the beta-subunit. EMBO J 8: 2291–2296. Heinen CA, Losekoot M, Sun Y et al. (2016). Mutations in TBL1X are associated with central hypothyroidism. J Clin Endocrinol Metab 101: 4564–4573. Heinen CA, De Vries EM, Alders M et al. (2018). Mutations in IRS4 are associated with central hypothyroidism. J Med Genet 55: 693–700. Hollenberg AN, Monden T, Wondisford FE (1995). Ligandindependent and -dependent functions of thyroid hormone receptor isoforms depend upon their distinct amino termini. J Biol Chem 270: 14274–14280.

CONGENITAL ISOLATED CENTRAL HYPOTHYROIDISM Joustra SD, Schoenmakers N, Persani L et al. (2013). The IGSF1 deficiency syndrome: characteristics of male and female patients. J Clin Endocrinol Metab 98: 4942–4952. Joustra SD, Meijer OC, Heinen CA et al. (2015). Spatial and temporal expression of immunoglobulin superfamily member 1 in the rat. J Endocrinol 226: 181–191. Lafranchi SH (1979). Hypothyroidism. Pediatr Clin North Am 26: 33–51. Lafranchi S (1999). Congenital hypothyroidism: etiologies, diagnosis, and management. Thyroid 9: 735–740. Lafranchi SH (2011). Approach to the diagnosis and treatment of neonatal hypothyroidism. J Clin Endocrinol Metab 96: 2959–2967. Lanting CI, Van Tijn DA, Loeber JG et al. (2005). Clinical effectiveness and cost-effectiveness of the use of the thyroxine/ thyroxine-binding globulin ratio to detect congenital hypothyroidism of thyroidal and central origin in a neonatal screening program. Pediatrics 116: 168–173. Legradi G, Emerson CH, Ahima RS et al. (1998). Arcuate nucleus ablation prevents fasting-induced suppression of ProTRH mRNA in the hypothalamic paraventricular nucleus. Neuroendocrinology 68: 89–97. Medici M, Visser WE, Visser TJ et al. (2015). Genetic determination of the hypothalamic-pituitary-thyroid axis: where do we stand? Endocr Rev 36: 214–244. Moran C, Chatterjee K (2015). Resistance to thyroid hormone due to defective thyroid receptor alpha. Best Pract Res Clin Endocrinol Metab 29: 647–657. Naafs JC, Heinen CA, Zwaveling-Soonawala N et al. (2020). Age-specific reference intervals for plasma free thyroxine and thyroid stimulating hormone in term neonates during the first two weeks of life. Thyroid 30: 1106–1111. https://doi.org/10.1089/thy.2019.0779. Persani L (2012). Clinical review: central hypothyroidism: pathogenic, diagnostic, and therapeutic challenges. J Clin Endocrinol Metab 97: 3068–3078. Persani L, Brabant G, Dattani M et al. (2018). 2018 European thyroid association (ETA) guidelines on the diagnosis and management of central hypothyroidism. Eur Thyroid J 7: 225–237. Price A, Weetman AP (2001). Screening for central hypothyroidism is unjustified. BMJ 322: 798. Rastogi MV, Lafranchi SH (2010). Congenital hypothyroidism. Orphanet J Rare Dis 5: 17. Schoenmakers N, Alatzoglou KS, Chatterjee VK et al. (2015). Recent advances in central congenital hypothyroidism. J Endocrinol 227: R51–R71.

169

Slawik M, Klawitter B, Meiser E et al. (2007). Thyroid hormone replacement for central hypothyroidism: a randomized controlled trial comparing two doses of thyroxine (T4) with a combination of T4 and triiodothyronine. J Clin Endocrinol Metab 92: 4115–4122. Sun Y, Bak B, Schoenmakers N et al. (2012). Loss-of-function mutations in IGSF1 cause an X-linked syndrome of central hypothyroidism and testicular enlargement. Nat Genet 44: 1375–1381. Tagami T, Madison LD, Nagaya T et al. (1997). Nuclear receptor corepressors activate rather than suppress basal transcription of genes that are negatively regulated by thyroid hormone. Mol Cell Biol 17: 2642–2648. Vaidya B, Pearce SH (2008). Management of hypothyroidism in adults. BMJ 337: a801. Van Tijn DA, De Vijlder JJ, Verbeeten Jr B et al. (2005). Neonatal detection of congenital hypothyroidism of central origin. J Clin Endocrinol Metab 90: 3350–3359. Weiss RE, Refetoff S (2015). Thyroid function testing. In: JL Jameson, LJ De Groot (Eds.), Endocrinology adult and pediatric, seventh edn. Elsevier, Philadelphia, pp. 1444–1492. Withers DJ, Gutierrez JS, Towery H et al. (1998). Disruption of IRS-2 causes type 2 diabetes in mice. Nature 391: 900–904. Yoon HG, Chan DW, Huang ZQ et al. (2003). Purification and functional characterization of the human N-CoR complex: the roles of HDAC3, TBL1 and TBLR1. EMBO J 22: 1336–1346. Zamboni G, Zaffanello M, Rigon F et al. (2004). Diagnostic effectiveness of simultaneous thyroxine and thyroidstimulating hormone screening measurements. Thirteen years’ experience in the Northeast Italian Screening Programme. J Med Screen 11: 8–10. Zoeller RT, Tan SW, Tyl RW (2007). General background on the hypothalamic-pituitary-thyroid (HPT) axis. Crit Rev Toxicol 37: 11–53. Zwaveling-Soonawala N, Van Beijsterveldt CE, Mesfum ET et al. (2015a). Fetal environment is a major determinant of the neonatal blood thyroxine level: results of a large Dutch twin study. J Clin Endocrinol Metab 100: 2388–2395. Zwaveling-Soonawala N, Van Trotsenburg AS, Verkerk PH (2015b). The severity of congenital hypothyroidism of central origin should not be underestimated. J Clin Endocrinol Metab 100: E297–E300.

This page intentionally left blank

Section 7 Zona incerta

This page intentionally left blank

Handbook of Clinical Neurology, Vol. 180 (3rd series) The Human Hypothalamus: Middle and Posterior Region D.F. Swaab, F. Kreier, P.J. Lucassen, A. Salehi, and R.M. Buijs, Editors https://doi.org/10.1016/B978-0-12-820107-7.00011-2 Copyright © 2021 Elsevier B.V. All rights reserved

Chapter 11

The zona incerta system: Involvement in attention and movement SANDRINE CHOMETTON1, MARIE BARBIER2, AND PIERRE-YVES RISOLD3* 1

Department of Biological Sciences, University of Southern California, Los Angeles, CA, United States 2

Seaver Autism Center, Icahn School of Medicine, Mount Sinai, New York, NY, United States

EA481, Integrative and Clinical Neurosciences, UFR Sant
e, Universit
e de Bourgogne Franche-Comt
e, Besanc¸on, France

3

Abstract The zona incerta (ZI) is a large structure made of four neurochemically defined regions (at least, in rodents). It is globally involved in complex connections with telencephalic and brainstem centers. In this work, we focus on some of the anatomical links this structure develops with the cerebral cortex and the tectum. We also point to its integration within a larger basal ganglia network. The functions of this region are still mysterious, even if recent works suggest its participation in behavioral expression. Studies about the functional organization of the vibrissal system have provided the first integrated model, illustrating the ZI’s role in sensory-motor programing. In addition, ZI connections with the superior colliculus and the cerebral cortex as well as recent behavioral studies point to this region playing a role in cognitive processes related to attention toward salient stimuli.

INTRODUCTION The zona incerta (ZI) is a poorly differentiated structure that has long been considered part of the ventral (or pre-) thalamic region. Indeed, 20th-century concepts divided the diencephalon into the dorsal thalamus, ventral thalamus, and hypothalamus (Swanson, 1987). However, theories are evolving, and the concept of a wider hypothalamic region that includes parts of the former ventral thalamus, such as the ZI and the subthalamic nucleus, has emerged (Allen Institute, 2004; Swanson, 2004). Apart from the nomenclature or semantic issues, this change in our understanding of the anatomical position of the ZI reflects the fact that this structure has few anatomical and developmental features in common with the thalamus. Therefore, the ZI is no longer perceived as a part of the thalamus, although the thalamus has an important role in mediating ZI functions. Development will not be dealt with here. This work will focus on anatomy and functions, describing some of the main connections of

the ZI with regards to its integration in a complex network devoted to attentional processes.

ORGANIZATION OF THE ZONA INCERTA Most of what is known about the cyto- and chemoarchitecture of the ZI comes from studies on rats. Thus the short summary of the ZI organization provided later is largely based on observations of rodents made by Mitrofanis (2005), but readers may refer to other works (e.g., Chivileva and Gorbachevskaya, 2008; Watson et al., 2014) for additional information on other species. The ZI is a thick slice of tissue horizontally oriented between the thalamus dorsally and the caudal hypothalamus and rostral midbrain ventrally. Its proximity to motor tracts and centers has been described by previous authors and includes the superior cerebellar peduncle (the H fields of Forel), the cerebral peduncle, the nigrostriatal tract, the medial lemniscus, the thalamic fasciculus (the h1 field), and the lenticular fasciculus (the h2 field)

*Correspondence to: Pierre-Yves Risold, EA481, Neurosciences Int
egratives et Cliniques, UFR Sant
e, 19 rue Ambroise Par
e, Universit
e de Bourgogne Franche-Comt
e, 25030 Besanc¸on, cedex, France. Tel: +33-363-08-22-23, E-mail: [email protected]

174

S. CHOMETTON ET AL.

(Watson et al., 2014). It is a heterogeneous region in which several parts can be recognized: rostromedial (ZIrm), ventral (ZIv), dorsal (ZId), and caudal (ZIc). The divisions of the ZI are clearly delineated using neurochemistry (Mitrofanis, 2005). Indeed, the distribution of calbindin, parvalbumin, and calretinin varies within the boundaries of the ZI (Fig. 11.1). The ZIv is characterized by an intense parvalbumin expression (Fig. 11.1B0 –I0 ). Parvalbumin antibodies label cell bodies as well as axons within this region. By contrast, the rostral ZId contains almost no parvalbumin-labeled cells (Fig. 11.1B0 –F0 ). Caudally, the neuropil of the ZId is inconsistently stained (Fig. 11.1G0 –I0 ), with a region just ventral to the medial lemniscus being darker. The ZId, as a whole, is better characterized by NO expression (Mitrofanis, 2005). The ZIv and ZId contain little calbindin signals, and dispersed cell bodies can be observed with anti-calretinin antibodies. Both calbindin and calretinin appear more intensely expressed in the ZIrm (Fig. 11.1A and A00 ), and calbindin neurons are abundant only in the ZIc (Fig. 11.1J). Most of the ZI cells are medium-sized and often fusiform, oval, or multipolar in shape. Large cells are observed medially, many of which correspond to neurons containing melaninconcentrating hormone peptide (MCH). The MCH population is distributed over the ZIrm/rostral ZIv and the tuberal lateral hypothalamic area (LHA) (Swanson et al., 2005). Notably, the ZIrm also contains a tight group of dopaminergic neurons (A13). Other peptides are reported in the ZI of other species, such as galanin present in mice (Allen Institute, 2004) as well as in humans (Gai et al., 1990) but not in rats (Ryan and Gundlach, 1996). The cytoarchitecture of the ZI also shows interspecies differences. Indeed, the work of Goncharuk et al. (2004) describes the human ZI as organized as a chain of cell clusters linked by rows of single cells, many of these cells expressing the hFF1 receptor (Goncharuk et al., 2004).

CONNECTIONS OF THE ZONA INCERTA Each division in the ZI has a specific connection pattern. However, a widespread interconnection also exists, which indicates that divisions of the ZI are not independent entities. Therefore, in this chapter, instead of describing the connections successively for each ZI part, we highlight some of the most relevant mono- and bidirectional anatomical links.

Connections with the cerebral cortex The cerebral cortex is one of the main sources of afferents to the ZI (Shammah-Lagnado et al., 1985; Mitrofanis and Mikuletic, 1999). Most cortical areas send a few axons into this structure, but main cortical inputs arise from cingulate and frontal somatomotor areas. These projections come from layer 5, while cortical layer 6 is solely

involved in corticothalamic projections. Therefore, the cortical axons innervating the ZI are collateral branches of those descending toward the caudal brainstem and spinal cord. Projections from the cingulate cortex enter the ZI rostrodorsally and innervate both the rostral and dorsal parts thoroughly. The secondary motor area (MOs) innervates the ZI with a pattern similar to that of the cingulate cortex. Fig. 11.2 illustrates the distribution of axons in the ZI after injection of the anterograde tracer Phaseolus vulgaris leucoagglutinin (PHAL) into the part of the MOs named the “Frontal Eye Field” in rats (Chometton et al., 2017). From the cerebral peduncle, axons enter the dorsolateral ZI at hypothalamic tuberal levels and can then be traced to the ZIrm and ZId (Fig. 11.2A and B). Some axons traveling at the edge of the cerebral peduncle and ventral ZI participate in the innervation of the STN (Fig. 11.2C). Caudal to the STN, a large trunk of axons leaves the cerebral peduncle and divides into two branches, one arching laterally, participating in the innervation of the ZI, and the other one running in medial and then dorsal directions toward the tectum (Fig. 11.2D). The innervation from the frontal motor (MO) and somatosensory (SS) areas enters the ZI through its ventral part (Aronoff et al., 2010). A single axon study of primates showed that many axons from the MO innervated the STN before entering the ZIv (Coud
e et al., 2018). After PHAL injections into the MO of a rat, we also observed that many axons reach the ZIv from the STN, but others take routes similar to those previously described for the projections from the MOs. The inputs from the MO and SS are somatotopically organized, with an overrepresentation of the face, including the whiskers (Shaw and Mitrofanis, 2002). Some of these projections extend beyond the border of the ZIv into the ZId, but convergence on the same cells as those innervated by the cingulate or MOs has not yet been illustrated. Therefore, cortical axons reach the ZI through different routes, and some of those are common to pathways innervating the STN. The ZI may reciprocate some of these cortical projections (Chen and Kriegstein, 2015), but the extent and organization of the incertocortical projection are not entirely clear yet. Part of these connections may arise from MCH cell bodies in the ZIrm. These cells were confused with a-MSH cell bodies in the past because many polyclonal serums against a-MSH cross-reacted with NEI, a peptide cleaved from the ppMCH (Risold et al., 1992). This information was unknown prior to 1989 (as in Saper et al., 1986), but it is now clear that there are no MSH cell bodies in the ZI. Other neurons in the ZId may also send axons through the cerebral cortex.

THE ZONA INCERTA SYSTEM

175

Fig. 11.1. Photomicrographs of adjacent coronal rat brain sections labeled with antibodies against calbindin, parvalbumin, and calretinin and organized from rostral (A–A00 ) to caudal (J–J00 ). These photographs illustrate the anatomical position of the ZI as well as some of its neurochemical characteristics that help define the boundaries of its subdivisions. See text for additional information. Abbreviations: AHN, anterior hypothalamic nucleus; cpd, cerebral peduncle; DMH, dorsomedial hypothalamic nucleus; FF, field of Forel; fx, fornix; LHA, lateral hypothalamic area; ml, medial lemniscus; mtt, mamillothalamic tract; NRT, thalamic reticular nucleus; PSTN, parasubthalamic nucleus; PVH, paraventricular hypothalamic nucleus; SN, substantia nigra; STN, subthalamic nucleus; vlt, ventrolateral hypothalamic tract; ZI, zona incerta; ZIc, zona incerta caudal part; ZId, zona incerta dorsal part; ZIrm, zona incerta rostromedial part; ZIv, zona incerta ventral part.

176

S. CHOMETTON ET AL.

Fig. 11.1—Cont’d

Connections with the superior colliculus (SC) The ZI is bidirectionally connected to the SC (Kim et al., 1992; Kolmac et al., 1998). The SC outputs arise from the intermediate and deep layers, are mostly glutamatergic, and primarily target the ZIv (although they often spread into the ZId). The SC projections to the ZI are topographically organized. The rostral SC projects into the rostral ZI as well

as into the ZIrm from its most rostral pole (Chometton et al., 2017). The lateral SC innervates the medial ZI abundantly, and the medial SC projects into the lateral ZI. After injection of a retrograde tracer into the ZIv, Watson et al. (2015) reported labeled cells in both somesthetic and visually responsive regions of the SC, suggesting that both sensory modalities are relayed to the ZI from the SC.

THE ZONA INCERTA SYSTEM

177

Fig. 11.2. Photomicrographs illustrating the distribution of PHAL-labeled axons in the rat ZI after injection of the anterograde tracer in the secondary motor cortex (MOs). The PHAL was revealed by immunohistochemistry and observed under darkfield illumination. (A) Axons leaving the cpd to enter the rostral ZI. (B) Axons innervate a large part of the ZI, but the densest labeling is in the ZId. (C) Some axons from the MOs take a dorsal route (arrowhead) in the cpd and enter the STN and, in some cases, the ZI. Other axons traveling ventrally in the ZI (double arrowhead) also enter the STN. (D) Caudal to the STN level, a large trunk of axons arches dorsally from the cpd and provides an innervation of the ZIc, while others take a medial route. See text for additional information. Abbreviations: cpd, cerebral peduncle; fx, fornix; LHA, lateral hypothalamic area; ml, medial lemniscus; mtt, mamillothalamic tract; STN, subthalamic nucleus; TH, thalamus; vlt, ventrolateral hypothalamic tract; ZIc, zona incerta caudal part; ZId, zona incerta dorsal part; ZIv, zona incerta ventral part.

The incertotectal projections arise from the ZIv. They are GABAergic and target, in return, the intermediate and deep layers of the SC (Ricardo, 1981; Romanowski et al., 1985; Kim et al., 1992; May et al., 1997; Mitrofanis, 2005). The ZI has also been identified as being involved in the generation of saccades in the SC (May et al., 1997; Bangash et al., 2019).

Connections with the thalamus Since first reported by Ricardo (1981), the thalamus has been identified as an important output station mediating ZI functions. Subsequent anterograde studies of the ZI have also reported a dense, dorsally directed contingent of axons innervating thalamic nuclei (Barthó et al., 2002). However, injection sites in different ZI subdivisions innervate distinct

sets of thalamic nuclei. Generally, the thalamus is divided into first- and second-order nuclei, depending on the nature of the information transmitted to the telencephalon. The former (also named relay nuclei) are involved in transmitting sensory information to the isocortex. The latter send projections to the cerebral cortex and the striatum. They are also involved in complex information processing related to learning, memory, and attention. Ricardo (1981) described projections from the ZI into the intralaminar and midline nuclei (second-order nuclei), but the isotope method was not precise enough to identify terminals. Subsequent studies confirmed the large input from the ZI to the thalamus. A retrograde study by Power et al. (1999) clearly illustrated that the ZId and ZIv have distinct patterns of outputs into the dorsal thalamus. According to this study, the ZId has a larger

178

S. CHOMETTON ET AL.

contingent of thalamic projection than the ZIv into the intralaminar nuclei of the thalamus, including the parafascicular nucleus (PF). The ZId also has stronger projections into the posterior association nuclei, including the lateral dorsal, lateral posterior, and posterior (PO) nuclei. Patterns of outputs from the ZIrm and ZIc remain less clear, but it has been shown that the ZIr projects into the reuniens nucleus of the midline thalamus, like the ZId (Sita et al., 2007). To conclude, the projections from the ZI into relay or first-order nuclei are weak, but the ZI might be involved in complex sensory coding of higherorder thalamic nuclei.

Other connections of the zona incerta Besides the aforementioned connections, the ZI is also linked to many other brain structures that can be grouped into four main subparagraphs: basal ganglia, cerebellum, sensory afferents, and hypothalamus/reticular formation.

RELATED TO THE BASAL GANGLIA The relationship between the ZI and the cerebral cortex, the parafascicular nucleus of the thalamus, and the SC indicates that the ZI network is connected to the basal ganglia network. However, tract-tracing studies of rodents suggest only moderate connections between the ZI and the classic structures of the basal ganglia (Takada et al., 1994). The ZI has weak reciprocal connections with the globus pallidus (GP), some ZI axons reported in this nucleus, and some retrograde-labeled cells in the entopeduncular nucleus from ZI injections. Shammah-Lagnado et al. (1996) found that PHAL injections in the caudal GP of rats revealed axons in the ventral ZI. Some connections with the reticular part of the substantia nigra have also been signaled (Romanowski et al., 1985). Pallidal projections into the ZI in primates have been discussed (Haber, 2016).

RELATED TO MOTOR AND CEREBELLUM The connection between the ZI and the cerebellar network is well documented. The ZI receives projections from deep cerebellar nuclei and, in turn, innervates the red nucleus (Pong et al., 2008). We do not focus on this part of the ZI network in the present study, but these projections may be involved in the control of face and neck musculature. Pong et al. (2008) suggested that the ZI may serve as an intermediary between the basal ganglia and the cerebellar networks of cats.

RELATED TO SOMATOSENSORY AFFERENTS Mitrofanis (2005) reported that visual and auditory modalities can reach specific, mostly lateral, sectors of the ZI. However, the bulk of ZI sensory afferents are

Fig. 11.3. Diagram schematizing the paralemniscal and lemniscal pathways, and the disinhibition of the PO by glutamatergic-descending MO axons targeting ZI GABAergic interneurons. Abbreviations: MO, motor area of the cerebral cortex; PO, posterior medial thalamic nucleus; PrV, principal trigeminal nucleus; SpVi, interpolar region of the spinal trigeminal division; SS, somatosensory area of the cerebral cortex; VPM, ventral posteromedial nucleus of the thalamus; ZI, zona incerta.

somatosensory. The ZI receives a contingent of inputs from the dorsal horn of the spinal cord. These projections carry somatic and visceral sensory information and terminate mostly in the ZIv. Besides, trigeminal afferents are well described in the ZI. These projections arise from the interpolar region of the spinal trigeminal division (SpVi) of the trigeminal complex and form the paralemniscal pathway (Desch^enes et al., 2005; Moore et al., 2015) (Fig. 11.3). They target the ZI, mostly the ventral part, as well as the PO in the thalamus, which itself projects into the dysgranular barrel field (SS) and MO cortices. From the SpVi, axons also innervate brainstem nuclei, including the SC.

RELATED TO THE HYPOTHALAMUS AND RETICULAR FORMATION

The ZI has bidirectional links with many brainstem reticular nuclei. Among those, the neighboring mesencephalic reticular nucleus appears to be a favored partner as the bidirectional connections between the two structures are very intense, but many other nuclei of the pons and medulla are also concerned (Kolmac et al., 1998).

THE ZONA INCERTA SYSTEM This includes the pedunculopontine tegmental nucleus (PPN), a target of the basal ganglia network (Mori et al., 2016). Finally, we conclude this short, nonexhaustive review of ZI connectivity with the hypothalamus. As part of the reticular formation, some links exist between the LHA and the ZI. We have already mentioned the fact that MCH cells are distributed over the two structures. Projections from the LHA into the ZI are also mentioned in the literature (Swanson et al., 2005). Many other nuclei of the hypothalamus distribute at least a few axons into the ZI (Risold et al., 1994; Canteras et al., 1995). These projections transit by the lateral regions of the ZI to reach the ventral geniculate. It is worth mentioning that a large component of the hypothalamic output to the periaqueductal gray transit through the dorsal and posterior hypothalamus; this pathway largely spills into the rostromedial region of the ZI (Thompson et al., 1996). Therefore, the ZIrm can be included within the medial hypothalamic network (Sita et al., 2007).

OVERALL FUNCTIONAL NETWORKS OF THE ZONA INCERTA Functionally, the ZI has mostly been investigated in regard to its predominant connections with the somatosensory/whiskers and SC/motor networks.

179

individuals, the position and movement of whiskers reflect the behavioral state of the animal (Dominiak et al., 2019). It is likely that the ZI and PO are influenced by a large range of inputs, playing an important role in determining how ZI regulates PO activity. Several sources of input may inform the ZI about the current behavioral context and modulate the activity of the paralemniscal pathway (Diamond and Ahissar, 2007). Among those sources, the cerebral cortex plays a significant role and can disinhibit or inhibit the PO through its projections into the ZI (Urbain et al., 2015; Escudero and Nuñez, 2019). The glutamatergic neurons in the motor cortex stimulate the GABAergic interneurons in the ZI, resulting in the inhibition of the GABAergic ZIv neurons that project to the PO, disinhibiting the PO (Urbain and Desch^enes, 2007). Projection from the medial prefrontal areas (prelimbic) might have an opposite action by stimulating GABAergic ZIv and PO neurons and, therefore, inhibiting the PO (Escudero and Nuñez, 2019). The convergence of primary sensory cortex and deep cerebellar projections into the ZI is also thought to play a role in the fine-tuning of sensory-motor programs and to help in spatial navigation through the environment (Sch€afer and Hoebeek, 2018). The SC is another structure that receives trigeminal inputs and may control the activity of the paralemniscal pathway through its projections into the ZI. However, the SC is also known to react to unexpected stimuli. This function will be discussed further in the next section.

The zona incerta within the vibrissal system Whisking is a typical rodent behavior involved in the exploration of the environment during locomotion. The vibrissal system is characterized by highly segregated projections arising from the principal trigeminal nucleus to the ventral posterior medial nucleus of the thalamus (first-order nucleus) and then to the barrel cortex in the somatosensory area (Desch^enes et al., 2005; Bosman et al., 2011; Moore et al., 2015) (lemniscal pathway, Fig. 11.3). This system ensures the whisking cycle, encodes self-motion and whisker deflection (Moore et al., 2015). The role of the paralemniscal pathway—which involves the ZI and was previously briefly described — is not entirely clear (Masri et al., 2008; Moore et al., 2015) (Fig. 11.3). The spinal division of the trigeminal complex conveys signals corresponding to whisker deflection to the PO and ZIv. However, GABAergic neurons of the ZIv innervate and inhibit the PO to such an extent that its neurons only weakly respond to the trigeminal signal. Therefore, in anesthetized animals, the paralemniscal pathway requires the disinhibition of PO neurons to transmit information to the telencephalon (Lavall
ee et al., 2005). In conscious and behaving

The zona incerta and the switching of behaviors The SC is a multisensory structure that guides gaze shifts to bring prominent events into central vision (Stein et al., 1975). It is also involved in approach reactions, either associated with appetitive tasks (reach for food, caring for pups) or nocifensive responses toward a source of pain by orienting the head toward the painful stimulus; it is also involved in defense/avoidance reactions to a threat, such as a predator (Comoli et al., 2012; Savage et al., 2017). Lateral parts of the SC, in relation to the lower visual field, are involved in approach reactions. Whiskerrelated somatosensory stimuli are detected in the lateral approach region of the SC. This region projects into the medial ZIv; this pathway has been associated with prey catching (Favaro et al., 2011; Shang et al., 2019). The lateral part of the SC is also involved in orienting the head toward a painful stimulus in order to bite or lick the affected body part (Telford et al., 1996; Wang and Redgrave, 1997). In comparison, the stimulation of the medial SC triggers active avoidance through a complex pathway that

180

S. CHOMETTON ET AL.

converges onto the PPN, the locomotor center (Hormigo et al., 2019). Retinal input to the medial SC relates to the upper visual field, supposedly, from where danger may loom for rodents. The medial SC is connected to the lateral ZI. Experimental evidence linking the lateral ZI to avoidance is lacking. However, retrograde tract-tracing experiments from the lateral and medial SC have confirmed that both regions receive afferents compatible with their opposite responses (Comoli et al., 2012; Savage et al., 2017). The reaction to an unexpected stimulus supposes that the animal switches behavior to adapt to the new environmental condition. This switch of attention between the perceptual dimensions of complex stimuli requires the intervention of the basal ganglia network. In fact, one of the cognitive impairments associated with Parkinson’s disease is difficulty in shifting attention. Tait et al. (2017) illustrated that lesions of the dorsomedial striatum and STN impaired rats’ response to attentional set-shifting paradigms. In this study, STN lesions also extended into the ZI. Conversely, the substantia nigra pars reticulata (SNr) reduces threat recognition through direct nigrotectal projections and causes a shift from avoidance to approach behavior (Almada et al., 2018). The basal ganglia network includes a direct pathway from the striatum to the SNr, and an indirect pathway with a relay in the GPe that innervates the STN, which, in turn, innervates the SNr (Kita and Kitai, 1987; Parent and Hazrati, 1995) (Fig. 11.4). Authors also often refer to these pathways as “go” or “no-go,” as the former promotes motor behavioral responses while the latter inhibits the same responses. These two pathways are completed by a hyperdirect path from the isocortex to the STN; as previously mentioned, collaterals of cortical axons in the STN extend into the ZI. Besides the cerebral cortex and the mesencephalic dopaminergic neurons, the striatum is also under strong influence from high-order thalamic nuclei. Among those, the PF and the PO have opposite influences on the striatum (Fig. 11.4). Intralaminar projections to the striatum are thought to play a role in salient stimuli, redirecting attention through shifts in cortically driven action selection (blue pathway on Fig. 11.4). Stimulation of these nuclei causes a pause in striatal activity and then an increase in the responsiveness of the indirect pathways to cortical input (Ding et al., 2010). The action of the PF in shifting behavior is reported in the dorsomedial striatum, although this nucleus is involved in complex thalamo-corticostriatal loops (Mandelbaum et al., 2019). In addition, the PF has a strong direct projection into the STN (Kita et al., 2016) and, therefore, stimulates the no-go pathway directly through this nucleus. The behavioral function of the PO is still not fully understood, but it projects only in dorsolateral sectors of the striatum (red

pathway on Fig. 11.4). It is essential for sensorimotor habit learning, perhaps through the direct pathway. Watson et al. (2015) and Watson and Alloway (2018) showed that the SC and ZI cooperate in order to influence the striatal complex through the thalamus. First, the SC sends glutamatergic projections into the PF, stimulating the striatal no-go pathway (Kita et al., 2016) (Fig. 11.4). It also stimulates GABAergic neurons in the ZIv, which inhibit the PO (Fig. 11.4). This forward inhibition can be involved in modifying motor behavior from the dorsolateral striatum. The network also involves the ZId that innervates the intralaminar nuclei of the thalamus, including the PF (Fig. 11.4). The ZId receives fewer projections from the SC (although in the squirrel monkey, the ZId is as heavily innervated by collicular projections as the ZIv (May and Basso, 2018)). However, as previously mentioned, the SC directly innervates the PF, which, in turn, innervates the dorsomedial striatum. Interestingly, this last structure receives afferents from the cingulate cortex, the region involved in avoidance, and that projects into the ZId and the medial SC (Comoli et al., 2012; Savage et al., 2017). To conclude, the ZI is involved in a complex network largely intricated with the basal ganglia network, as illustrated in Fig. 11.4. The functional roles of the ZI are slowly emerging. Recent data indicate that other parts of the ZI, and the ZIrm in particular, might be involved in other aspects of attentional processes.

The case of the rostromedial zona incerta The SC guides gaze shifts to bring specific stimuli into central vision, but it is also involved in gaze fixation. Previous studies on monkeys and cats identified fixation cells in the rostral pole of the SC. These cells project onto omnipause neurons in the raphe interpositus (Wang et al., 2013; Shinoda et al., 2019). Fixation cells discharge continuously during visual fixation. Their stimulation suppresses the generation of saccadic eye movements generated in the SC. The rostral pole of the SC projects onto the rostral ZI, including the ZIrm. Chometton et al. (2017) showed that the rostral ZI (mostly ZIrm) expressed higher levels of the c-Fos protein, a marker of neuronal activation, when rats explored an unknown environment and when hungry rats were teased by palatable food enclosed in a mesh box. The expression of c-Fos was lower when animals actually consumed the food. Similar c-Fos responses were obtained in the rostral SC. It was concluded that the ZIrm is involved in attentional processes related to goal-oriented behavior. This conclusion was in accordance with previous data showing that the ZIrm is imbedded within the medial hypothalamic networks that initiate goal-oriented behaviors (Sita et al., 2007).

THE ZONA INCERTA SYSTEM

181

Fig. 11.4. Summary diagrams illustrating the complex networks interconnecting the ZI, basal ganglia, SC, and cerebral cortex. See text for details. Abbreviations: Cing, cingulate cortex; dl, dorsolateral sector of the striatum; dm, dorsomedial sector of the striatum; GABA, gamma-aminobutyric acid; Glu, glutamate; MD, mediodorsal thalamic nucleus; MO, motor area of the cerebral cortex; PF, parafascicular nucleus thalamus; PO, posterior medial thalamic nucleus; SC, superior colliculus; SNr, substantia nigra pars reticulata; VM, ventromedial nucleus of the thalamus; STN, subthalamic nucleus; ZId, zona incerta dorsal part; ZIv, zona incerta ventral part.

Recently, the SC-ZIrm pathway was identified as being involved in hunting, with the ZIrm integrating SC and auditive/visual information to induce an appetitive response and promoting hunting (Zhao et al., 2019). The strong motivational drive described by these authors is coherent with the observation of Chometton et al. (2017), in particular concerning the response of teased rats. Interestingly, Chometton et al. (2017) illustrated that

the rostral pole of the SC not only projects into the rostral ZI but also has widespread projections through the ventral and dorsal ZI regions. However, the ZIrm receives projections only from the rostral pole of the SC. Considering that (a) the lateral ZI is also involved in processing sensory afferents (trigeminal, auditory, or visual (see Mitrofanis, 2005)), (b) the ZIrm appears to promote action toward a specific goal present in an arena

182

S. CHOMETTON ET AL.

(hunting a cockroach or trying to reach food enclosed in a mesh box), and finally (c) the ZI is characterized by widespread interconnections, it is plausible to hypothesize that the lateral ZI processes those sensory stimuli and relays the information to the ZIrm for attention and motivation in a way that is rhetorically similar to the SC, which guides gaze shifts to bring prominent events into central vision represented in the rostral SC. Furthermore, both structures are hugely and topographically interconnected, and the ZI has a permissive role in the generation of saccades (May et al., 1997; Bangash et al., 2019). However, the exact mechanisms and the respective roles of each ZI compartment and the ZI network as a whole still need to be investigated.

CONCLUSIONS We still have an incomplete understanding of ZI functions. What is known of its connections demonstrates that this structure is integrated into a network related to the isocortex and basal ganglia. Most studies in rodents implicate this region in complex (cognitive) processes probably related to attention. Studies on the vibrissal system in rodents are at the origin of significant progress. In primates, the vibrissal system is irrelevant. However, the ZI has become an important target for deep brain stimulation in Parkinson’s disease, along with the STN (Ossowska, 2019). In human patients, the stimulation of ZI regions dorsal to the STN benefits many of the symptoms of this debilitating disease, including tremors, akinesia, and speech intelligibility (Sandstr€ om et al., 2015; Blomstedt et al., 2018). The mechanisms by which the ZI is involved in these responses are still not clear (see Ossowska, 2019). One possibility is an indirect involvement via the cerebellothalamic pathway that is collateral in the ZI. Indeed, data in rodents previously reviewed suggest that the ZI may indirectly control striatonigral neurons and, in return, might enhance the imbalance between the direct and indirect pathways in Parkinson’s disease.

REFERENCES Allen Institute (2004). Allen mouse brain atlas. http://mouse. brain-map.org/. Almada RC, Genewsky AJ, Heinz DE et al. (2018). Stimulation of the nigrotectal pathway at the level of the superior colliculus reduces threat recognition and causes a shift from avoidance to approach behavior. Front Neural Circuits 12: 36. https://doi.org/10.3389/fncir.2018.00036. Aronoff R, Matyas F, Mateo C et al. (2010). Long-range connectivity of mouse primary somatosensory barrel cortex. Eur J Neurosci 31: 2221–2233. https://doi.org/10.1111/ j.1460-9568.2010.07264.x.

Bangash OK, Dissanayake AS, Knight S et al. (2019). Modulation of saccades in humans by electrical stimulation of the posterior subthalamic area. J Neurosurg 132: 1–9. https://doi.org/10.3171/2018.12.JNS18502. Bartho´ P, Freund TF, Acsa´dy L (2002). Selective GABAergic innervation of thalamic nuclei from zona incerta. Eur J Neurosci 16: 999–1014. https://doi.org/10.1046/j.14609568.2002.02157.x. Blomstedt P, Stenmark Persson R, Hariz G-M et al. (2018). Deep brain stimulation in the caudal zona incerta versus best medical treatment in patients with Parkinson’s disease: a randomised blinded evaluation. J Neurol Neurosurg Psychiatry 89: 710–716. https://doi.org/10.1136/jnnp2017-317219. Bosman LWJ, Houweling AR, Owens CB et al. (2011). Anatomical pathways involved in generating and sensing rhythmic whisker movements. Front Integr Neurosci 5: 53. https://doi.org/10.3389/fnint.2011.00053. Canteras NS, Simerly RB, Swanson LW (1995). Organization of projections from the medial nucleus of the amygdala: a PHAL study in the rat. J Comp Neurol 360: 213–245. https://doi.org/10.1002/cne.903600203. Chen J, Kriegstein AR (2015). A GABAergic projection from the zona incerta to cortex promotes cortical neuron development. Science 350: 554–558. https://doi.org/10.1126/ science.aac6472. Chivileva OG, Gorbachevskaya AI (2008). Projections of the basal ganglia to the zona incerta of the dog diencephalon. Neurosci Behav Physiol 38: 743–746. https://doi.org/ 10.1007/s11055-008-9040-3. Chometton S, Charrie`re K, Bayer L et al. (2017). The rostromedial zona incerta is involved in attentional processes while adjacent LHA responds to arousal: c-Fos and anatomical evidence. Brain Struct Funct 222: 2507–2525. https://doi.org/10.1007/s00429-016-1353-3. Comoli E, Das Neves Favaro P, Vautrelle N et al. (2012). Segregated anatomical input to sub-regions of the rodent superior colliculus associated with approach and defense. Front Neuroanat 6: 9. https://doi.org/10.3389/fnana.2012.00009. Coud
e D, Parent A, Parent M (2018). Single-axon tracing of the corticosubthalamic hyperdirect pathway in primates. Brain Struct Funct 223: 3959–3973. https://doi.org/ 10.1007/s00429-018-1726-x. Desch^enes M, Timofeeva E, Lavall
ee P et al. (2005). The vibrissal system as a model of thalamic operations. Prog Brain Res 149: 31–40. https://doi.org/10.1016/S0079-6123 (05)49003-2. Diamond ME, Ahissar E (2007). When outgoing and incoming signals meet: new insights from the zona incerta. Neuron 56: 578–579. https://doi.org/10.1016/j.neuron.2007.11.006. Ding JB, Guzman JN, Peterson JD et al. (2010). Thalamic gating of corticostriatal signaling by cholinergic interneurons. Neuron 67: 294–307. https://doi.org/10.1016/j.neuron. 2010.06.017. Dominiak SE, Nashaat MA, Sehara K et al. (2019). Whisking asymmetry signals motor preparation and the behavioral state of mice. J Neurosci 39: 1809–1819. https://doi.org/ 10.1523/JNEUROSCI.1809-19.2019.

THE ZONA INCERTA SYSTEM Escudero G, Nun˜ez A (2019). Medial prefrontal cortical modulation of whisker thalamic responses in anesthetized rats. Neuroscience 406: 626–636. https://doi.org/10.1016/j. neuroscience.2019.01.059. Favaro PDN, Gouv^ea TS, de Oliveira SR et al. (2011). The influence of vibrissal somatosensory processing in rat superior colliculus on prey capture. Neuroscience 176: 318–327. https://doi.org/10.1016/j.neuroscience.2010.12.009. Gai WP, Geffen LB, Blessing WW (1990). Galanin immunoreactive neurons in the human hypothalamus: colocalization with vasopressin-containing neurons. J Comp Neurol 298: 265–280. https://doi.org/10.1002/cne.902980302. Goncharuk V, Zeng Z, Wang R et al. (2004). Distribution of the neuropeptide FF1 receptor (hFF1) in the human hypothalamus and surrounding basal forebrain structures: immunohistochemical study. J Comp Neurol 474: 487–503. https://doi.org/10.1002/cne.20132. Haber S (2016). Perspective on basal ganglia connections as described by Nauta and Mehler in 1966: where we were and how this paper effected where we are now. Brain Res 1645: 4–7. https://doi.org/10.1016/j.brainres.2016.04.016. Hormigo S, Vega-Flores G, Rovira V et al. (2019). Circuits that mediate expression of signaled active avoidance converge in the pedunculopontine tegmentum. J Neurosci 39: 4576–4594. https://doi.org/10.1523/JNEUROSCI.004919.2019. Kim U, Gregory E, Hall WC (1992). Pathway from the zona incerta to the superior colliculus in the rat. J Comp Neurol 321: 555–575. https://doi.org/10.1002/cne.90321 0405. Kita H, Kitai ST (1987). Efferent projections of the subthalamic nucleus in the rat: light and electron microscopic analysis with the PHA-L method. J Comp Neurol 260: 435–452. https://doi.org/10.1002/cne.902600309. Kita T, Shigematsu N, Kita H (2016). Intralaminar and tectal projections to the subthalamus in the rat. Eur J Neurosci 44: 2899–2908. https://doi.org/10.1111/ejn.13413. Kolmac CI, Power BD, Mitrofanis J (1998). Patterns of connections between zona incerta and brainstem in rats. J Comp Neurol 396: 544–555. https://doi.org/10.1002/ (sici)1096-9861(19980713)396:43.0. co;2-g. Lavall
ee P, Urbain N, Dufresne C et al. (2005). Feedforward inhibitory control of sensory information in higher-order thalamic nuclei. J Neurosci 25: 7489–7498. https://doi. org/10.1523/JNEUROSCI.2301-05.2005. Mandelbaum G, Taranda J, Haynes TM et al. (2019). Distinct cortical-thalamic-striatal circuits through the parafascicular nucleus. Neuron 102: 636–652.e7. https://doi.org/ 10.1016/j.neuron.2019.02.035. Masri R, Bezdudnaya T, Trageser JC et al. (2008). Encoding of stimulus frequency and sensor motion in the posterior medial thalamic nucleus. J Neurophysiol 100: 681–689. https://doi.org/10.1152/jn.01322.2007. May PJ, Basso MA (2018). Connections between the zona incerta and superior colliculus in the monkey and squirrel. Brain Struct Funct 223: 371–390. https://doi.org/10.1007/ s00429-017-1503-2.

183

May PJ, Sun W, Hall WC (1997). Reciprocal connections between the zona incerta and the pretectum and superior colliculus of the cat. Neuroscience 77: 1091–1114. https:// doi.org/10.1016/s0306-4522(96)00535-0. Mitrofanis J (2005). Some certainty for the “zone of uncertainty”? Exploring the function of the zona incerta. Neuroscience 130: 1–15. https://doi.org/10.1016/j.neuroscience.2004.08.017. Mitrofanis J, Mikuletic L (1999). Organisation of the cortical projection to the zona incerta of the thalamus. J Comp Neurol 412: 173–185. Moore JD, Mercer Lindsay N, Desch^enes M et al. (2015). VIBRISSA self-motion and touch are reliably encoded along the same somatosensory pathway from brainstem through thalamus. PLoS Biol 13: e1002253. https://doi.org/ 10.1371/journal.pbio.1002253. Mori F, Okada K-I, Nomura T et al. (2016). The pedunculopontine tegmental nucleus as a motor and cognitive interface between the cerebellum and basal ganglia. Front Neuroanat 10: 109. https://doi.org/10.3389/fnana.2016.00109. Ossowska K (2019). Zona incerta as a therapeutic target in Parkinson’s disease. J Neurol 267: 591–606. https://doi. org/10.1007/s00415-019-09486-8. Parent A, Hazrati LN (1995). Functional anatomy of the basal ganglia. II. The place of subthalamic nucleus and external pallidum in basal ganglia circuitry. Brain Res Brain Res Rev 20: 128–154. https://doi.org/10.1016/0165-0173(94) 00008-d. Pong M, Horn KM, Gibson AR (2008). Pathways for control of face and neck musculature by the basal ganglia and cerebellum. Brain Res Rev 58: 249–264. https://doi.org/10.1016/ j.brainresrev.2007.11.006. Power BD, Kolmac CI, Mitrofanis J (1999). Evidence for a large projection from the zona incerta to the dorsal thalamus. J Comp Neurol 404: 554–565. Ricardo JA (1981). Efferent connections of the subthalamic region in the rat. II. The zona incerta. Brain Res 214: 43–60. https://doi.org/10.1016/0006-8993(81)90437-6. Risold PY, Fellmann D, Rivier J et al. (1992). Immunoreactivities for antisera to three putative neuropeptides of the rat melanin-concentrating hormone precursor are coexpressed in neurons of the rat lateral dorsal hypothalamus. Neurosci Lett 136: 145–149. Risold PY, Canteras NS, Swanson LW (1994). Organization of projections from the anterior hypothalamic nucleus: a Phaseolus vulgaris-leucoagglutinin study in the rat. J Comp Neurol 348: 1–40. https://doi.org/10.1002/cne.903480102. Romanowski CA, Mitchell IJ, Crossman AR (1985). The organisation of the efferent projections of the zona incerta. J Anat 143: 75–95. Ryan MC, Gundlach AL (1996). Localization of preprogalanin messenger RNA in rat brain: identification of transcripts in a subpopulation of cerebellar Purkinje cells. Neuroscience 70: 709–728. https://doi.org/10.1016/s03064522(96)83009-0. Sandstr€ om L, H€agglund P, Johansson L et al. (2015). Speech intelligibility in Parkinson’s disease patients with zona incerta deep brain stimulation. Brain Behav 5: e00394. https://doi.org/10.1002/brb3.394.

184

S. CHOMETTON ET AL.

Saper CB, Akil H, Watson SJ (1986). Lateral hypothalamic innervation of the cerebral cortex: immunoreactive staining for a peptide resembling but immunochemically distinct from pituitary/arcuate alpha-melanocyte stimulating hormone. Brain Res Bull 16: 107–120. Savage MA, McQuade R, Thiele A (2017). Segregated frontocortical and midbrain connections in the mouse and their relation to approach and avoidance orienting behaviors. J Comp Neurol 525: 1980–1999. https://doi.org/10.1002/ cne.24186. Sch€afer CB, Hoebeek FE (2018). Convergence of primary sensory cortex and cerebellar nuclei pathways in the whisker system. Neuroscience 368: 229–239. https://doi.org/ 10.1016/j.neuroscience.2017.07.036. Shammah-Lagnado SJ, Negra˜o N, Ricardo JA (1985). Afferent connections of the zona incerta: a horseradish peroxidase study in the rat. Neuroscience 15: 109–134. https:// doi.org/10.1016/0306-4522(85)90127-7. Shammah-Lagnado SJ, Alheid GF, Heimer L (1996). Efferent connections of the caudal part of the globus pallidus in the rat. J Comp Neurol 376: 489–507. https://doi.org/10.1002/ (SICI)1096-9861(19961216)376:3 < 489::AID-CNE10 > 3.0.CO;2-H. Shang C, Liu A, Li D et al. (2019). A subcortical excitatory circuit for sensory-triggered predatory hunting in mice. Nat Neurosci 22: 909–920. https://doi.org/10.1038/s41593019-0405-4. Shaw V, Mitrofanis J (2002). Anatomical evidence for somatotopic maps in the zona incerta of rats. Anat Embryol 206: 119–130. https://doi.org/10.1007/s00429-002-0280-7. Shinoda Y, Takahashi M, Sugiuchi Y (2019). Brainstem neural circuits for fixation and generation of saccadic eye movements. Prog Brain Res 249: 95–104. https://doi.org/10.1016/ bs.pbr.2019.04.007. Sita LV, Elias CF, Bittencourt JC (2007). Connectivity pattern suggests that incerto-hypothalamic area belongs to the medial hypothalamic system. Neuroscience 148: 949–969. https://doi.org/10.1016/j.neuroscience.2007.07.010. Stein BE, Magalhaes-Castro B, Kruger L (1975). Superior colliculus: visuotopic-somatotopic overlap. Science 189: 224–226. https://doi.org/10.1126/science.1094540. Swanson LW (1987). The hypothalamus. In: A Bj€orklund, T H€okfelt, LW Swanson (Eds.), Handbook of chemical neuroanatomy, integrated systems of the CNS, Part I. Elsevier, Amsterdam. Swanson LW (2004). Brain maps: structure of the rat brain, third edn. Elsevier, San Diego. Swanson LW, Sanchez-Watts G, Watts AG (2005). Comparison of melanin-concentrating hormone and hypocretin/orexin mRNA expression patterns in a new parceling scheme of the lateral hypothalamic zone. Neurosci Lett 387: 80–84. https://doi.org/10.1016/j.neulet.2005.06.066. Tait DS, Phillips JM, Blackwell AD et al. (2017). Effects of lesions of the subthalamic nucleus/zona incerta area and dorsomedial striatum on attentional set-shifting in the rat.

Neuroscience 345: 287–296. https://doi.org/10.1016/j. neuroscience.2016.08.008. Takada M, Tokuno H, Ikai Y et al. (1994). Direct projections from the entopeduncular nucleus to the lower brainstem in the rat. J Comp Neurol 342: 409–429. https://doi.org/ 10.1002/cne.903420308. Telford S, Wang S, Redgrave P (1996). Analysis of nociceptive neurones in the rat superior colliculus using c-fos immunohistochemistry. J Comp Neurol 375: 601–617. https://doi.org/10.1002/(SICI)1096-9861(19961125) 375:4< 601::AID-CNE4 > 3.0.CO;2–5. Thompson RH, Canteras NS, Swanson LW (1996). Organization of projections from the dorsomedial nucleus of the hypothalamus: a PHA-L study in the rat. J Comp Neurol 376: 143–173. https://doi.org/10.1002/(SICI)10969861(19961202)376:1 3.0.CO;2–3. Urbain N, Desch^enes M (2007). Motor cortex gates vibrissal responses in a thalamocortical projection pathway. Neuron 56: 714–725. https://doi.org/10.1016/j.neuron. 2007.10.023. Urbain N, Salin PA, Libourel P-A et al. (2015). Whiskingrelated changes in neuronal firing and membrane potential dynamics in the somatosensory thalamus of awake mice. Cell Rep 13: 647–656. https://doi.org/10.1016/j.celrep. 2015.09.029. Wang S, Redgrave P (1997). Microinjections of muscimol into lateral superior colliculus disrupt orienting and oral movements in the formalin model of pain. Neuroscience 81: 967–988. https://doi.org/10.1016/s0306-4522(97) 00191-7. Wang N, Perkins E, Zhou L et al. (2013). Anatomical evidence that the superior colliculus controls saccades through central mesencephalic reticular formation gating of omnipause neuron activity. J Neurosci 33: 16285–16296. https://doi. org/10.1523/JNEUROSCI.2726-11.2013. Watson GDR, Alloway KD (2018). Opposing collicular influences on the parafascicular (Pf ) and posteromedial (POm) thalamic nuclei: relationship to POm-induced inhibition in the substantia nigra pars reticulata (SNR). Brain Struct Funct 223: 535–543. https://doi.org/10.1007/s00429-0171534-8. Watson C, Lind CRP, Thomas MG (2014). The anatomy of the caudal zona incerta in rodents and primates. J Anat 224: 95–107. https://doi.org/10.1111/joa.12132. Watson GDR, Smith JB, Alloway KD (2015). The zona incerta regulates communication between the superior colliculus and the posteromedial thalamus: implications for thalamic interactions with the dorsolateral striatum. J Neurosci 35: 9463–9476. https://doi.org/10.1523/JNEUROSCI.160615.2015. Zhao Z-D, Chen Z, Xiang X et al. (2019). Zona incerta GABAergic neurons integrate prey-related sensory signals and induce an appetitive drive to promote hunting. Nat Neurosci 22: 921–932. https://doi.org/10.1038/s41593019-0404-5.

Section 8 Ventromedial nucleus and dorsomedial nucleus

This page intentionally left blank

Handbook of Clinical Neurology, Vol. 180 (3rd series) The Human Hypothalamus: Middle and Posterior Region D.F. Swaab, F. Kreier, P.J. Lucassen, A. Salehi, and R.M. Buijs, Editors https://doi.org/10.1016/B978-0-12-820107-7.00012-4 Copyright © 2021 Elsevier B.V. All rights reserved

Chapter 12

The role of the dorsomedial and ventromedial hypothalamus in regulating behaviorally coupled and resting autonomic drive LUKE A. HENDERSON1* AND VAUGHAN G. MACEFIELD2,3 1

Department of Anatomy & Histology, Brain and Mind Centre, University of Sydney, Sydney, NSW, Australia 2 3

Baker Heart & Diabetes Institute, Melbourne, VIC, Australia

Department of Anatomy and Physiology, University of Melbourne, Melbourne, VIC, Australia

Abstract Nearly a century ago it was reported that stimulation of the hypothalamus could evoke profound behavioral state changes coupled with altered autonomic function. Since these initial observations, further studies in animals have revealed that two hypothalamic regions—the dorsomedial and ventromedial hypothalamic nuclei—are critical for numerous behaviors, including those in response to psychological stressors. These behaviors are coupled with changes in autonomic functions, such as altered blood pressure, heart rate, sympathetic nerve activity, resetting of the baroreflex and changes in pituitary function. There is also growing evidence that these two hypothalamic regions play a critical role in thermogenesis, and suggestions they could also be responsible for the hypertension associated with obesity. The aim of this chapter is to review the anatomy, projection patterns, and function of the dorsomedial and ventromedial hypothalamus with a particular focus on their role in autonomic regulation. While most of what is known about these two hypothalamic regions is derived from laboratory animal experiments, recent human studies will also be explored. Finally, we will describe recent human brain imaging studies that provide evidence of a role for these hypothalamic regions in setting resting sympathetic drive and their potential role in conditions such as hypertension.

Almost a century ago it was demonstrated that lesions or electrical stimulation of the hypothalamus evoked a range of behaviors, most of which were coupled with significant changes in autonomic function (for review, see Hess and Brugger, 1981). More modern neural exploratory techniques such as chemical stimulation, optogenetics, and human brain imaging techniques have confirmed that neurons within discrete regions of the hypothalamus regulate a range of autonomically coupled behavioral responses, even contributing to basal autonomic tone. The aim of this chapter is to review two discrete parts of the hypothalamus, the dorsomedial and ventromedial regions, and their roles in regulating autonomic function coupled to behavioral state changes and

during quiet rest. While the vast majority of what we currently know about the roles of these regions in autonomic function stems from studies in rodents, more recently, human brain imaging studies are also confirming and in some areas extending our understanding of the autonomic function of these hypothalamic regions.

DMH AND VMH LOCATIONS AND ANATOMICAL PROJECTION PATTERNS The dorsomedial hypothalamus (DMH) and ventromedial hypothalamus (VMH) nuclei lie rostrocaudally in the tuberal part of the hypothalamus and mediolaterally in the intermediate region between the periventricular

*Correspondence to: Luke A. Henderson, Department of Anatomy and Histology, Brain and Mind Centre, M02, University of Sydney, NSW, Australia. Tel: +612-9351-7063, Fax: +612-9351-6556, E-mail: [email protected]

188

L.A. HENDERSON AND V.G. MACEFIELD

Fig. 12.1. Locations of the dorsomedial hypothalamus (DMH) and ventromedial hypothalamus (VMH) shown on a sagittal and a coronal human brain wet specimen, T1-weighted anatomical magnetic resonance image (MRI), T2-weighted anatomical MRI, and a blood oxygen level-dependent (BOLD) functional MRI image. The lower row shows an expanded view of the hypothalamus with the location of the DMH indicated by the green outline and the VMH by the red outline. Note that the DMH and VMH lie medial to the fornix and optic tract and lateral to the median eminence and the third ventricle. The MRI images were collected on a 7 T MRI, and the spatial resolution of the BOLD fMRI is of the highest currently available.

and lateral zones. The hypothalamus is its widest at the tuberal part, and the VMH is the largest nucleus in this region, lying between the optic tract and the median eminence. It is oval in shape and is relatively easily identifiable using ultra-high-field human anatomical magnetic resonance imaging (MRI) techniques (Fig. 12.1). The DMH lies immediately dorsal to the VMH between the border of the third ventricle and the lateral preoptic area and fornix laterally. While in rodents the DMH can be divided into numerous subregions and is easily distinguishable (Swanson, 2004), in humans it is less differentiated and more difficult to identify using anatomical MRI techniques (Baroncini et al., 2012; Saper, 2012). Tract tracing studies have revealed that both the DMH and VMH have multiple ascending and descending

projection targets. In general, connections between both the DMH and VMH and higher brain regions are small, with the DMH sending projections to the thalamic paraventricular nucleus (PVN), septal region, hippocampus, and amygdala and receiving inputs primarily from the septal area, prefrontal cortex, and subiculum (ter Horst and Luiten, 1986; Thompson et al., 1996; Thompson and Swanson, 1998). Similarly, the VMH sends projections to and receives inputs from the periventricular thalamic nucleus, septal region, amygdala, and bed nucleus of the stria terminalis (Saper et al., 1976; Shimogawa et al., 2015). While it is challenging to use current human brain imaging techniques to explore regional hypothalamic connectivity, a relatively recent investigation used diffusion tensor imaging to show similar DMH and VMH

CONTRIBUTIONS OF THE HYPOTHALAMUS TO AUTONOMIC CONTROL ascending projection patterns to those derived from laboratory animal investigations (Lemaire et al., 2011). Similar to the ascending connections of the DMH and VMH, the descending projections are relatively sparse, although they have been studied more extensively. Indeed, for the main part, the descending projections of these two hypothalamic regions target brainstem sites that are well known to play significant roles in autonomic regulation. The DMH projects directly to the midbrain periaqueductal gray matter (PAG), in particular to the region of the lateral and dorsomedial PAG columns (ter Horst et al., 1984; ter Horst and Luiten, 1986; Bandler et al., 2000; de Git et al., 2018). In addition, the DMH projects directly to the dorsolateral pons, nucleus tractus solitarius (NTS), and to the regions of the rostral ventrolateral medulla (RVLM), caudal ventrolateral medulla (CVLM), and nucleus raphe pallidus (Hosoya and Matsushita, 1981; Hosoya, 1985; ter Horst and Luiten, 1986). Brainstem inputs to the DMH arise from the PAG, dorsolateral pons, and the regions of the RVLM and CVLM (Thompson and Swanson, 1998). While the VMH also projects to the PAG, it appears that this projection targets primarily the region of the dorsolateral column. Furthermore, whereas the DMH projects strongly to more caudal brainstem regions, the VMH projects lightly at best to regions such as the locus coeruleus and NTS and does not appear to project to regions such as the RVLM, CVLM, or nucleus raphe pallidus (Saper et al., 1976; Luiten et al., 1987; Canteras et al., 1994; Shimogawa et al., 2015). In addition to numerous descending and ascending projections, there are extensive interconnections between hypothalamic nuclei. For example, the DMH has major projections to all nuclei in the periventricular zone apart from the arcuate nucleus (Thompson et al., 1996) and receives inputs from almost all the major nuclei in the anterior, middle, and posterior parts of the hypothalamus (Thompson and Swanson, 1998). Similarly, the VMH is interconnected extensively with the periventricular region, including the arcuate nucleus, as well as with almost all regions covering the rostrocaudal extent of the hypothalamus (Saper et al., 1976; Shimogawa et al., 2015). Interestingly, while some tract tracing studies have shown reciprocal connections between the DMH and VMH, others have suggested such projections appears weak at best, suggesting that there is little direct interplay between these two hypothalamic regions (ter Horst and Luiten, 1986; Thompson and Swanson, 1998; Shimogawa et al., 2015). Overall, both the DMH and VMH have extensive projections to the PAG, albeit to different locations, and the DMH also has robust direction connections with multiple brainstem sites involved in cardiovascular regulation – including the dorsolateral pons, NTS and the region containing almost all sympathetic premotor

189

neurons in the brain, the RVLM. These projections provide the DMH and VMH with the ability to directly regulate behaviorally coupled or resting cardiovascular function.

DMH AND CARDIOVASCULAR CONTROL Acute stressors Early studies reported that electrical stimulation of the hypothalamus evoked increases in blood pressure (BP), heart rate (HR), and sympathetic and respiratory activity. This configuration of cardiorespiratory changes is the characteristic of those expressed when an animal is confronted with a threatening stimulus (Yardley and Hilton, 1986) or during acute psychological stressors such as air jet stress in conscious rats (Morin et al., 2001) or mental stress in humans (Callister et al., 1992; Carter et al., 2005). The hypothalamic region stimulated to evoke such a response comprises the DMH and the immediately lateral perifornical region, and was named the “hypothalamic defense area.” More recent studies have confirmed that direct chemical activation or disinhibition of the DMH evokes increases in BP, HR, respiratory activity as well as adrenocorticotropic hormone (ACTH) secretion from the anterior pituitary (Soltis and DiMicco, 1991; Bailey and Dimicco, 2001). Critically, DMH inhibition diminishes stress-induced increases in BP, HR, respiratory activity, and ACTH (Stotz-Potter et al., 1996a,b) and resets the baroreflex to a higher set-point, similar to that which occurs during acute stressors (McDowall et al., 2006; Kanbar et al., 2007). Furthermore, acute stressors evoke neural activation (as demonstrated by an increase in c-fos expression) in the DMH as well as in multiple sites that receive direct DMH connections, including the amygdala, hypothalamic paraventricular nucleus, PAG, parabrachial nucleus in the dorsolateral pons, NTS, RVLM, and the nucleus raphe pallidus (Dayas et al., 2001; Albutaihi et al., 2004; McDougall et al., 2004; Ulrich-Lai and Herman, 2009; Furlong et al., 2014; Dampney, 2019). While all of these brainstem regions with DMN connections can alter at least one autonomic parameter upon activation (Dampney, 1994), whether all or some of these connections mediate the autonomic changes during emotional stressors is not known. It is clear, however, that blocking the projection from the DMH to the PVN of the hypothalamus eliminates the neuroendocrine part of the stress response but does not change the autonomic responses (Stotz-Potter et al., 1996a). Since it appears from tract tracing experiments that the DMH does not send direct projections to the spinal sympathetic preganglionic neurons in the intermediolateral cell column of the spinal cord, the DMH must mediate its sympathetic effects via premotor neurons. The major

190

L.A. HENDERSON AND V.G. MACEFIELD

source of sympathetic premotor neurons in the brain lies in the region of the RVLM and, for the main part, most behaviorally coupled changes in sympathetic drive occur via this region (Dampney, 1994). However, while acute stressors activate neurons in the ventrolateral medulla, very few lie in the RVLM; the majority are located caudally in the CVLM (Carrive and Gorissen, 2008; Furlong et al., 2014). Also, while it has been shown that inhibition of the RVLM reduces the increase in sympathetic activity evoked by DMH activation (Fontes et al., 2001; Cao et al., 2004), this appears to result from the role of the RVLM setting basal sympathetic drive and resetting of the baroreflex set-point rather than a direct DMHRVLM effect on sympathetic drive (Dampney, 2019). Furthermore, though it is possible that the DMH regulates the RVLM via the CVLM or the NTS, there is some evidence that the response of these regions to psychological stressors results from descending inputs from the amygdala and paraventricular hypothalamus, respectively (Dayas and Day, 2002; Dayas et al., 2004). Furthermore, given that the NTS-CVLM-RVLM pathway is critical for the expression of the baroreflex, it is possible that activation of these regions during emotional stressors underlies resetting of the baroreflex that occurs during both DMH stimulation and acute stressors (McDowall et al., 2006; Kanbar et al., 2007). Although the pathways responsible for DMH-evoked sympathetic responses remain unknown, there is growing evidence that the HR responses to DMH activation and acute psychological stressors are mediated by the nucleus raphe pallidus. Stimulation of the raphe pallidus and the immediately dorsal raphe obscurus can evoke increases or decreases in BP and HR depending on the anesthesia (Coleman and Dampney, 1995; Henderson et al., 1998; Heslop et al., 2002), and tracing studies provide evidence for a direct projection from the raphe pallidus to the cardiac sympathetic preganglionic neurons in the lateral horn of the thoracic spinal cord (Loewy, 1981; Bacon et al., 1990; Heslop et al., 2004). Critically, inhibition of the raphe pallidus suppresses HR increases evoked by DMH stimulation (Samuels et al., 2002) and abolishes the HR increase during an acute psychological stressor (Zaretsky et al., 2003), consistent with the hypothesis that acute stressors increase HR via a DMH-raphe palliduslateral horn pathway. Interestingly, there is some evidence that DMH control of HR is lateralized since right DMH inhibition abolishes stress-evoked increases in HR, whereas inhibition of the left DMH does not (Xavier et al., 2013; Fontes et al., 2017). Furthermore, while the DMH projects directly to the raphe pallidus, it also projects to the PAG; this nucleus itself produces increase in HR upon activation and projects directly to the raphe pallidus, providing an alternative route via which the DMH can alter HR (Carrive et al., 1987; Henderson et al., 1998).

Defining the role of the DMH in stress responses is important from a clinical perspective since there is strong and consistent human evidence that acute stressors can evoke sudden cardiac dysfunction and death (Mittleman et al., 1995; Leor et al., 1996). Moreover, chronic emotional stress is correlated with the development of cardiovascular diseases such as hypertension and cardiac arrhythmias (Bunker et al., 2003; Esler et al., 2008; Lampert, 2009). Furthermore, it has been reported that DMH activation can trigger cardiac arrhythmias, particularly ventricular ectopic beats evoked from the right side of the DMH (Xavier et al., 2013). A lack of DMH-evoked arrhythmias during blockade of cardiac beta-adrenergic receptors suggests that increased sympathetic output from the hypothalamus to the ventricles of the heart can precipitate cardiac arrhythmias. Although there have been many human brain imaging studies exploring brain activation during numerous psychological stressors, few have reported hypothalamic signal changes (Dedovic et al., 2009). However, Pruessner and colleagues reported, using fMRI, that blood oxygen level-dependent (BOLD) signal intensity decreases throughout the limbic system, including the left hypothalamus, in humans during the Montreal Imaging stress task, a social evaluative threat paradigm (Pruessner et al., 2008). Interestingly, this decrease in left hypothalamic BOLD signal intensity occurred only in those subjects that also displayed an increase in cortisol release during the acute stressor. The lack of consistent reports of hypothalamic changes during emotional stressors in human imaging studies may appear at odds with the extensive data obtained from laboratory animals, it is likely that the limited spatial acuity of current human brain imaging techniques has meant that hypothalamic changes are difficult to resolve using standard analysis procedures. Furthermore, the finding by Pruessner and colleagues that hypothalamic changes occur only when cortisol release also increases suggests that it is critical that an experimental measure of sympathetic output is desirable (Pruessner et al., 2008). Indeed, we have explored human brain activation patterns during the emotional stress of viewing images depicting mutilation, while at the same time recording skin sympathetic nerve activity (SSNA), which supplies cutaneous blood vessels, sweat glands and hairs, directly using microneurography (Henderson et al., 2012). Viewing emotionally charged images evoked significant increases in sweat release and skin vasoconstriction, and increased BOLD signal intensity in the amygdala and hypothalamus (Fig. 12.2A and B). Furthermore, SSNA significantly increased, and this was coupled to increases in amygdala and hypothalamic activity (Fig. 12.2C). While at the time of our study, the available MRI technology meant we were not able to determine the precise hypothalamic region that was

CONTRIBUTIONS OF THE HYPOTHALAMUS TO AUTONOMIC CONTROL

191

Fig. 12.2. (A) Changes in autonomic measures during viewing of emotional images (mutilation) in healthy control subjects. Note that viewing stressful images evoked a small increase in heart rate and larger increases in skin vasoconstriction, skin potential as well as respiratory rate and depth. Furthermore, direct recording of skin sympathetic activity (SSNA) revealed significant increases in total bursts frequency and amplitude during viewing images of mutilation compared with neutral images. (B) Increases in BOLD signal intensity (SI) during viewing images of mutilation compared with controls. Note the increases in signal intensity in the right hypothalamus and amygdala. The coronal slice location in Montreal Neurological Institute space is indicated at the top right of the image. (C) Increases in signal intensity coupled to bursts in SSNA during mutilation image viewing also occurred in the hypothalamus and amygdala. Modified with permission from Henderson LA, Stathis A, James C et al. (2012). Real-time imaging of cortical areas involved in the generation of increases in skin sympathetic nerve activity when viewing emotionally charged images. Neuroimage 62: 30–40.

activated, relatively recent increases in MRI field strengths means that submillimeter spatial resolutions are now possible (Fig. 12.1). Interestingly, we found BOLD signal intensity increases on the right side, whereas Pruessner and colleagues reported signal decreases on the left side. This apparent lateralization may underlie differences in processing different forms of acute stressors and may relate to lateralized hypothalamic functions. Indeed, we also found that signal intensity increased in the right amygdala when viewing images of mutilation but increased on the left side when viewing images of erotica, which have a positive emotional valence (Henderson et al., 2012). Furthermore, while an increase in fMRI signal intensity is usually ascribed to an increase in neural activity, since fMRI signal likely reflects changes in synaptic activity, it may also reflect an increase in inhibitory neurotransmitter release

(Logothetis and Pfeuffer, 2004). Therefore, our results may reflect an overall increase in excitatory drive whereas the findings of Pruessner and colleagues may reflect a disinhibition of the DMH during different forms of acute psychological stressors.

Thermoregulation In addition to its role in mediating the autonomic responses to acute stressors, it has been proposed that the DMH integrates thermoregulatory responses to cold and fever. It is essential that humans maintain core body temperature within a narrow range so that cellular metabolism is not adversely affected. Changes in the temperature of an individual’s environment activate homeostatic mechanisms designed to maintain core body temperature by motivating an individual to alter the environment, to

192

L.A. HENDERSON AND V.G. MACEFIELD

regulate physical activity and energy expenditure, and bring about sympathetically mediated changes in skin blood flow (and sweat release in humans) and piloerection. Previous investigations have shown that the preoptic area of the hypothalamus is critical for maintaining core body temperature as it contains temperaturesensitive neurons and receives ascending inputs from cutaneous thermoreceptors (Boulant, 2000). Two mechanisms via which the brain can regulate body temperature are via cutaneous vasoconstriction or vasodilatation, and via shivering or nonshivering thermogenesis, all of which are controlled by the sympathetic nervous system. Nonshivering heat production occurs with brown adipose tissue (BAT), which has the specific metabolic function to dissipate energy in the form of heat (Dimicco and Zaretsky, 2007; RezaiZadeh and Munzberg, 2013; Morrison, 2016). Although many details of the preoptic area microcircuitry remain unknown, preoptic neurons regulate the output of shivering and nonshivering thermogenesispromoting neurons by modulating activity on the DMH. For example, acute skin cooling, which does not reduce core body temperature, evokes increases in BAT sympathetic nerve activity, BAT temperature, and HR, which are inhibited by bilateral inhibition of the preoptic area, DMH, or the nucleus raphe pallidus (Nakamura and Morrison, 2007). Since the DMH projects directly to the nucleus raphe pallidus, it is thought that the DMH exerts its sympathetic control of BAT thermogenesis via sympathetic premotor neurons in the raphe pallidus (Cao et al., 2004; Dimicco and Zaretsky, 2007). This DMH-raphe pallidus pathway may be direct, but there is evidence that the PAG also plays a critical role in modulating BAT sympathetic drive; PAG inhibition increases BAT sympathetic activity and the increase in BAT sympathetic activity evoked by the DMH is abolished by altering PAG activity (Rathner and Morrison, 2006). More specifically, it has been reported that the increase in BAT sympathetic activity and core body temperature that results from DMH disinhibition is significantly reduced by activation of ventral PAG sites (Rathner and Morrison, 2006) and inhibition of the caudal lateral and dorsolateral PAG sites (da Silva et al., 2003). This suggests that the DMH increases BAT sympathetic drive through a combination of inhibitory and excitatory drive onto different parts of the PAG. Since most of the basic circuitry responsible for thermoregulation has derived from experiments in rodents, it remains unknown if these circuits are conserved in humans, although a number of human brain imaging studies have explored brain activation patterns associated with thermoregulation. For example, it has been shown that prolonged cooling of the body evokes fMRI

signal intensity increases in the region of the nucleus raphe pallidus (McAllen et al., 2006). The presence of functional BAT has only recently been demonstrated in humans (Nedergaard et al., 2007) and a recent positron emission tomography study found no significant difference in hypothalamic or brainstem activity associated with activated BAT (Huang et al., 2011). Given the limited spatial resolution of positron emission tomography, further investigations using high-resolution fMRI techniques in addition to sympathetic recordings are needed to determine if the same circuitry shown to be involved in BAT sympathetic activity in rodent studies, namely the DMH, PAG, and nucleus raphe pallidus, are also critical in humans.

VMH AND CARDIOVASCULAR CONTROL Similarly to the DMH, the VMH has been implicated in behaviorally coupled autonomic changes. An early study in anesthetized rabbits revealed that electrical stimulation of the VMH could evoke an increase in BP and HR (Gellman et al., 1981) and recently a study in anesthetized rats revealed a positive relationship between slow activity oscillations in the VMH and sympathetic activity (Iigaya et al., 2017). Although the VMH is not considered to be a part of the classic “hypothalamic defense area,” there is emerging evidence that it may also be involved in mediating defensive behavior. For example, it has been reported that predator exposure and exposure to an aggressive conspecific significantly activates the VMH as determined by c-fos expression (Silva et al., 2013). Critically, unlike the DMH, the VMH is not activated by foot shock, an acute psychological stressor, suggesting that this nucleus is specifically recruited during predator and social fear. Furthermore, inactivation of the VMH reduces freezing responses in rats when faced with a natural predator (Silva et al., 2013), and VMH activation can elicit immobility which is mediated through projections to the midbrain PAG (Lin et al., 2011; Wang et al., 2015). Consistent with these findings from laboratory animals, electrical stimulation of the VMH in awake humans elicits panic attacks (Wilent et al., 2010) and so it appears that VMH function is conserved across species. As with the DMH, the behavioral responses elicited from the VMH are associated with autonomic changes. Low-level VMH stimulation evokes an increase in respiratory rate as well as a significant increase in HR (Wang et al., 2015). However, more intense VMH stimulation evoked primarily decreases in HR and so it appears that the autonomic function of the VMH may be complex. Interestingly, it was shown that VMH-evoked immobility could be associated with either an increase or decrease in HR,

CONTRIBUTIONS OF THE HYPOTHALAMUS TO AUTONOMIC CONTROL suggesting that the context of the immobility may result in divergent HR responses. Similar to the DMH, the VMH plays a critical role in the regulation of glucose and energy balance and has also been implicated in control of sympathetic activity, including the control of BAT (Choi et al., 2013). Over 30 years ago, it was shown in rodents that VMH lesions resulted in decreased sympathetic activity in a number of tissues including BAT (Vander Tuig et al., 1982), whereas electrical stimulation of VMH induces liver gluconeogenesis and increases in BP and HR (Gellman et al., 1981; Shimazu and Ishikawa, 1981). Leptin, a circulating hormone produced by adipose tissue, is known to reduce appetite and increase energy expenditure, which are coupled to sympathetic activation of BAT (Pelleymounter et al., 1995; Haynes et al., 1997). While evidence in laboratory animals suggests that the arcuate nucleus of the hypothalamus mediates the satiety effect of leptin (Satoh et al., 1997), microinjection of leptin in the VMH and DMH evoked increases in BP, with leptin injections into the VMH also evoking increases in renal sympathetic nerve activity (Montanaro et al., 2005). Importantly, leptin does not alter BP or sympathetic activity when injected into the hypothalamic arcuate nucleus (Marsh et al., 2003). Consistent with this finding, it was previously reported that leptin injections into the VMH evoked increases in plasma epinephrine and norepinephrine (Satoh et al., 1999). Furthermore, the increases in catecholamine levels that occur following intravenous leptin administration were eliminated by VMH lesions, suggesting that leptininduced increase in catecholamine secretion is mediated primarily via the VMH. While the action of leptin on the VMH is not likely involved in mediating behaviorally coupled changes in BP or sympathetic drive—such as those that occur during defensive behaviors—it has been proposed that it is involved in autonomic changes in disease states such as obesity. Indeed, obesity is associated with elevated levels of muscle sympathetic nerve activity (MSNA) at rest (Jones et al., 1996; Alvarez et al., 2002); leptin is associated with body fat mass and circulating leptin concentrations are correlated with adiposity-associated elevations in MSNA in humans (Solin et al., 1997; Monroe et al., 2000). Given that leptin injections into the VMH or DMH increase BP and sympathetic vasoconstrictor drive (Montanaro et al., 2005), it has been proposed that continuously altered functioning of the DMH and/or VMH is associated with increased resting MSNA in clinical conditions such as obesity (Mark et al., 1999; Marsh et al., 2003). Although there is no human evidence of a relationship between activity levels in the VMH and leptin-induced changes in sympathetic drive, there is evidence that the hypothalamus has an altered structure in obesity. For example, human MRI

193

investigations have reported that obesity is associated with markers of gliosis and inflammation in the region of the VMH (Thaler et al., 2012; Kreutzer et al., 2017) and these changes are associated with microglial infiltration and astrocyte activation in a mouse model of obesity (Valdearcos et al., 2014). While it remains unknown if such changes are associated with altered autonomic function, it is now well known that astrocyte activation is associated with altered calcium wave synchronicity and gliotransmitter release, which can significantly modulate neural activity and alter brain dynamics (Halassa et al., 2007; Crunelli et al., 2012). It is possible that chronic VMH astrocyte activation results in altered synaptic activity, which in turn results in increased output to areas such as the PAG and ultimately increase sympathetic drive and hypertension.

Resting sympathetic drive While it is clear that both the DMH and VMH alter autonomic function associated with behavioral state changes, until recently there was no evidence that either the DMH or VMH played a significant role in resting autonomic drive. In a series of investigations, we used concurrent recordings of brain activity using fMRI and sympathetic nerve activity using microneurography to determine brain regions in which activity was associated with spontaneous bursts of either muscle or skin sympathetic activity. In an initial study, we focused our investigation specifically on the lower brainstem and found, as expected, positive covariations between signal intensity and spontaneous bursts of muscle sympathetic nerve activity in the region of the RVLM and negative covariations in the region of the CVLM and NTS (Macefield and Henderson, 2010). More recently we focused our investigations on regions above the midbrain and determined brain regions in which signal covaried with either MSNA or SSNA (James et al., 2013a,b). While we found that hypothalamic activity did not covary with spontaneous bursts in SSNA, we found robust covariations between hypothalamic signal intensity, restricted to the left and right VMH and the left DMH, and spontaneous MSNA bursts. That is, when MSNAwas high, activity in the VMH and DMH was high and vice versa (Fig. 12.3). This data strongly suggests that although the DMH and VMH do not appear to set resting AP and HR, it is likely involved in setting and/or modulating resting muscle sympathetic drive. In addition to playing a role in setting resting muscle sympathetic (vasoconstrictor) drive in healthy control subjects, DMH and/or VMH activity may be critical in setting altered resting sympathetic levels in some disease states. For example, as mentioned previously, obesity is associated with elevated levels of resting MSNA (Jones et al., 1996; Alvarez et al., 2002) and the DMH and VMH

194

L.A. HENDERSON AND V.G. MACEFIELD

Fig. 12.3. (A) Recording of spontaneous muscle sympathetic nerve activity (MSNA) in a subject while performing fMRI of the brain. The filtered neurogram is shown in the top trace, the root-mean-square processed signal in the bottom trace. The dark areas represent scanning artifacts during functional magnetic resonance image (fMRI) collection. Bursts of spontaneous MSNA were measured in-between fMRI collection periods. The hemodynamic delay and time between brain activity and nerve traffic meant that the MSNA collection period corresponded to the following 4 s fMRI collection period (see Macefield and Henderson, 2010 for details). (B) Signal intensity (SI) increases in the dorsomedial hypothalamus (DMH) and ventromedial hypothalamus (VMH) correlated with spontaneous bursts of MSNA in 14 healthy control subjects. The hot color scale indicates regions in which SI was high during periods of high MSNA and low during low MSNA. Slice locations in Montreal Neurological Space are shown on the top right of each section. (C) Percentage SI change in individual subjects (black circles) and mean of all subjects (gray circle  SEM) in the dorsomedial and ventromedial hypothalamus during the periods of spontaneous MSNA bursts compared with the periods of no MSNA bursts. Note the consistent effect, i.e., signal intensity increased during periods of high MSNA in at least 11 of the 14 subjects for both hypothalamic regions. Modified with permission from James C, Macefield VG, Henderson LA (2013b). Real-time imaging of cortical and subcortical control of muscle sympathetic nerve activity in awake human subjects. Neuroimage 70: 59–65.

CONTRIBUTIONS OF THE HYPOTHALAMUS TO AUTONOMIC CONTROL

195

Fig. 12.4. (A) Brain regions in which gray matter volume was significantly lower in healthy normotensive individuals with higher resting muscle sympathetic nerve activity (MSNA) compared with lower MSNA. Plots of mean  SEM gray matter volumes in higher (pink) and lower (light blue) MSNA groups for the dorsomedial hypothalamus (DMH), anterior cingulate cortex (ACC), and midbrain periaqueductal gray (PAG) are also shown. (B) Brain regions in which gray matter volume was significantly lower or higher in healthy normotensive individuals with higher resting blood pressure (BP) compared with lower BP. Plots of mean  SEM gray matter volumes in higher (red) and lower (dark blue) BP groups for the rostral ventrolateral medulla (RVLM), raphe, and nucleus tractus solitaries (NTS) are also shown. Slice locations in Montreal Neurological Institute space are shown at the top right of each image. *P < 0.05. Modified from Kobuch S, Fatouleh RH, Macefield JM et al. (2020). Differences in regional grey matter volume of the brain are related to mean blood pressure and muscle sympathetic nerve activity in normotensive humans. J Hypertens 38: 303–313, with permission.

have been implicated in producing this obesity-related sympathetic activity increases. We recently provided evidence that even in healthy normotensive individuals, the structure of the DMH is related to variations in overall resting MSNA. More specifically, we found that subtle differences in overall resting MSNA levels are associated with regional gray matter volume differences (Kobuch et al., 2020). That is, those normotensive subjects with higher resting MSNA had lower gray matter volumes

in the DMH as well as in the cingulate cortex and PAG (Fig. 12.4A). Interestingly, while resting MSNA levels were not associated with differences in gray matter volume in medullary autonomic regions, resting mean BP levels were. More specifically, in the region of the medullary raphe, higher resting BP was associated with reduced gray matter volume and vice versa for the regions of the NTS and RVLM (Fig. 12.4B). As described earlier, the PAG, RVLM, NTS, and raphe nuclei all receive direct

196

L.A. HENDERSON AND V.G. MACEFIELD

Fig. 12.5. (A) The simplest form of resting-state functional connectivity is an analysis in which in each subject a Pearson’s correlation of functional magnetic resonance imaging (fMRI) signals between a seed region and each other voxel in the brain is determined. The resulting correlation coefficient maps are then placed into second-level group analysis to determine regions that display consistent positive or negative signal covariations with the seed region. (B) Brain regions in which signal intensity covaries significantly with signal intensity within the dorsomedial hypothalamus (DMH) and ventromedial hypothalamus (VMH). Note that signal intensity within the DMH and VMH covaries with signal intensity in the anterior insula, dorsolateral prefrontal cortex (dlPFC), and also in the region encompassing the rostroventrolateral medulla (RVLM). In addition, signal within the VMH covaries with signal in the orbitofrontal cortex (OFC) and precuneus. Slice locations in Montreal Neurological Space are shown on the top right of each section. The location of the DMH and VMH seed regions are indicated by the green shading on the coronal section to the left. (C) Slices through the rostral medulla showing the location of Angiotensin 2 binding in the RVLM as well as signal changes in the same regions during MSNA bursts at rest as well as resting functional connectivity with the VMH and DMH. Panel (C) Modified from James C, Henderson L, Macefield VG (2013a). Real-time imaging of brain areas involved in the generation of spontaneous skin sympathetic nerve activity at rest. Neuroimage 74: 188–194. Figure of Angiotensin 2 imaging reproduced, with permission, from Allen AM, Moeller I, Jenkins TA, et al. (1998). Angiotensin receptors in the nervous system. Brain Res Bull 47: 17–28.

CONTRIBUTIONS OF THE HYPOTHALAMUS TO AUTONOMIC CONTROL inputs from the DMH. It has been shown that early increases in sympathetic drive without concurrent increases in BP often occur in conditions such as essential hypertension (Julius and Nesbitt, 1996), which itself is characterized by increased sympathetic drive in the absence of any obvious peripheral trigger. It is possible that subtle structural and functional changes in areas including the DMH and its lower brainstem targets play a causal role in the development of essential hypertension, although further investigations are need to determine if such a link exists. As well as determining regional hypothalamic activity related to resting MSNA, we have also explored resting functional connectivity of the DMH and VMH. Restingstate functional imaging measures signal covariations between different brain regions and maps the strength of these “functional” connections (Fig. 12.5A; Allen et al., 1998). While two brain regions may not have a direct anatomical connection, a strong signal coupling between them indicates that at some level they are likely to be connected and involved in mediating the same function. In our study, we used the DMH and VMH regions, which covaried significantly with resting MSNA, as “seeding” regions. We found that resting signal intensity fluctuations in the DMH covaried with other regions known to regulate MSNA such as the anterior insula and dorsolateral prefrontal cortex as well as the region of the RVLM (Fig. 12.5B). Similarly, resting signal in the VMH also covaried with the anterior insula, dorsolateral prefrontal cortex, and the region of the RVLM. The coupling between both DMH and VMH and the RVLM was in the same location as the coupling between resting MSNA bursts and fMRI signal intensity, as well as the location shown to contain robust angiotensin II binding in humans (Fig. 12.5C). Interestingly, as noted above, while there is evidence of a direct projection from the DMH to the RVLM, no such connection exists for the VMH and thus it is likely that the VMH is involved in resting MSNA regulation via another brain region such as the DMH or the PAG. Since we did not find any significant resting signal covariations between the VMH and other brainstem regions besides the RVLM, our data suggest that VMH may influence resting MSNA via the DMH or higher brain regions, such as the insula or dorsolateral prefrontal cortex.

CONCLUSIONS Studies in laboratory animals have provided substantive evidence that the DMH and VMH are critical for the expression of various behaviorally coupled changes in autonomic function. Evidence to date shows that these regions are involved in mediating autonomic changes during acute stressors via projections to those lower

197

brainstem sites that contain premotor sympathetic neurons, such as the RVLM. In addition, both the DMH and VMH play a role in thermogenesis via their control over BAT sympathetic activity. There is evidence that this control is mediated via projections to the region of the nucleus raphe pallidus in the medulla, either directly or via projections to the PAG. While it remains unknown if similar functions and pathways also exits in humans, current human evidence suggests that DMH and VMH functions and pathways are indeed conserved across species. In addition, recent human brain imaging studies have provided the first evidence that the DMH and VMH are involved in regulating resting MSNA levels in awake, healthy humans in the absence of any ongoing stress. More studies are needed if we are to elucidate the precise roles of the DMH and VMH in both behaviorally coupled and resting autonomic control, particularly in humans. While human brain imaging studies also indicate that at least the output pathways of the DMH are lateralized with respect to autonomic output, further investigations are needed to determine what such lateralized function means for hypothalamic function and whether unbalanced autonomic output to the heart could produce inhomogeneity of repolarization, cardiac electrical instability, and arrhythmias. Furthermore, the role of the DMH and VMH in regulating autonomic function in disease states such as chronic stress and anxiety as well as obesity and thermoregulatory conditions needs to be explored in laboratory animals as well as in humans. The development of ultra-high-field strength MRI scanners that can measure brain activity at sub-millimeter and sub-second scales, and the ability to concurrently record sympathetic nerve activity in awake humans, will provide an invaluable method to investigate the role of these hypothalamic regions during numerous challenges and in a variety of clinical conditions.

REFERENCES Albutaihi IA, DeJongste MJ, Ter Horst GJ (2004). An integrated study of heart pain and behavior in freely moving rats (using fos as a marker for neuronal activation). Neurosignals 13: 207–226. Allen AM, Moeller I, Jenkins TA et al. (1998). Angiotensin receptors in the nervous system. Brain Res Bull 47: 17–28. Alvarez GE, Beske SD, Ballard TP et al. (2002). Sympathetic neural activation in visceral obesity. Circulation 106: 2533–2536. Bacon SJ, Zagon A, Smith AD (1990). Electron microscopic evidence of a monosynaptic pathway between cells in the caudal raphe nuclei and sympathetic preganglionic neurons in the rat spinal cord. Exp Brain Res 79: 589–602. Bailey TW, Dimicco JA (2001). Chemical stimulation of the dorsomedial hypothalamus elevates plasma ACTH in

198

L.A. HENDERSON AND V.G. MACEFIELD

conscious rats. Am J Physiol Regul Integr Comp Physiol 280: R8–15. Bandler R, Keay KA, Floyd N et al. (2000). Central circuits mediating patterned autonomic activity during active vs. passive emotional coping. Brain Res Bull 53: 95–104. Baroncini M, Jissendi P, Balland E et al. (2012). MRI atlas of the human hypothalamus. Neuroimage 59: 168–180. Boulant JA (2000). Role of the preoptic-anterior hypothalamus in thermoregulation and fever. Clin Infect Dis 31: S157–S161. Bunker SJ, Colquhoun DM, Esler MD et al. (2003). “Stress” and coronary heart disease: psychosocial risk factors. Med J Aust 178: 272–276. Callister R, Suwarno NO, Seals DR (1992). Sympathetic activity is influenced by task difficulty and stress perception during mental challenge in humans. J Physiol 454: 373–387. Canteras NS, Simerly RB, Swanson LW (1994). Organization of projections from the ventromedial nucleus of the hypothalamus: a Phaseolus vulgaris-leucoagglutinin study in the rat. J Comp Neurol 348: 41–79. Cao WH, Fan W, Morrison SF (2004). Medullary pathways mediating specific sympathetic responses to activation of dorsomedial hypothalamus. Neuroscience 126: 229–240. Carrive P, Gorissen M (2008). Premotor sympathetic neurons of conditioned fear in the rat. Eur J Neurosci 28: 428–446. Carrive P, Dampney RA, Bandler R (1987). Excitation of neurones in a restricted portion of the midbrain periaqueductal grey elicits both behavioural and cardiovascular components of the defence reaction in the unanaesthetised decerebrate cat. Neurosci Lett 81: 273–278. Carter JR, Kupiers NT, Ray CA (2005). Neurovascular responses to mental stress. J Physiol 564: 321–327. Choi YH, Fujikawa T, Lee J et al. (2013). Revisiting the ventral medial nucleus of the hypothalamus: the roles of SF-1 neurons in energy homeostasis. Front Neurosci 7: 71. Coleman MJ, Dampney RA (1995). Powerful depressor and sympathoinhibitory effects evoked from neurons in the caudal raphe pallidus and obscurus. Am J Physiol 268: R1295–R1302. Crunelli V, Lorincz ML, Errington AC et al. (2012). Activity of cortical and thalamic neurons during the slow (55 years) groups showed a conspicuous female dominance in the regional abundance of KP and NKB cell bodies, the density of KP-IR fibers, and the incidence of contacts these fibers established with the somatodendritic compartment of GnRH neurons (Fig. 17.3B) (Hrabovszky et al., 2011). A somewhat milder but significant sex difference was also noticed in the labeling of NKB, with more IR perikarya detected in postmenopausal women compared with men of similar age (Hrabovszky et al., 2011). In addition, KP and NKB perikarya were also larger in females than in males. Further, immunofluorescent studies established that the overlap between KP- and NKB-IR perikarya is only partial and differs significantly between sexes; 84% of NKB neurons in postmenopausal women and 69% in age-matched men exhibited KP immunoreactivity (Fig. 17.3A). KP labeling can also be observed in higher percentages of NKB-IR afferents to GnRH neurons in women (31%) compared with men (9%) and the percentage of KP afferents containing NKB is also higher in women (26%) than in men (10%) (Hrabovszky et al., 2011). These sex differences are thought to be partly activational and attributable to the lack of estrogen’s suppressive effects on the KISS1 and TAC3 genes in postmenopausal women. In contrast, men of similar age likely have enough testosterone action to suppress these genes. The sexual dimorphism of the human hypothalamic NKB system has also been observed by other

288

E. HRABOVSZKY ET AL.

investigators (Taziaux et al., 2012). These authors found higher NKB-IR input to the Inf and lower innervation to the pars tuberalis in adult human females compared with males (Taziaux et al., 2012). Furthermore, the Inf volume occupied by NKB immunoreactivity was lower in adult men than in adult women and in adult male-to-female transsexuals (Taziaux et al., 2012). These anatomic differences were detected in samples from young adults with intact negative sex steroid feedback, making it likely that they partly reflect organization sex steroid effects earlier in development.

Aging-dependent changes in men The aging-related decline in reproductive functions is less dramatic in men than in postmenopausal women. Testosterone production is often well sustained in elderly ages (Araujo and Wittert, 2011). Nevertheless, there is a declining trend in the negative feedback response of the reproductive axis (Veldhuis et al., 2010). When hypogonadism is present, common clinical symptoms include decreased morning erections, erectile dysfunction, and decreased frequency of sexual thoughts (Wu et al., 2010). Altered serum parameters often include decreased levels of free testosterone and dihydrotestosterone and increased levels of LH, FSH, and sex hormone binding globulin (Feldman et al., 2002; Wu et al., 2010). Reproductive aging in men is associated with reduced pulsatile and enhanced basal LH secretion, a decline in the LH secretory burst mode (Veldhuis et al., 2010) and a more irregular and asynchronous LH and testosterone secretion (Pincus et al., 1996; Liu et al., 2006). In the light of aging-related alterations in negative feedback and LH pulsatility, the plastic changes of the KP and NKB neuronal systems were addressed with quantitative immunohistochemical studies of hypothalamic samples from men of different age groups (Molnar et al., 2012). In samples categorized arbitrarily as “young” ( male; Ruiz-Pino et al., 2012). Interestingly sex differences in NKB expression and projections emerge progressively

300

J. BAKKER

during puberty and appear to be dependent on pubertal exposure to the rising levels of estradiol in females and testosterone in males (Ciofi et al., 2007). In this chapter, an overview will be provided on KP and NKB expression in the human hypothalamus and their potential role in human reproductive function.

KISSPEPTIN EXPRESSION IN THE HUMAN HYPOTHALAMUS In humans, immunohistochemistry on postmortem tissues have revealed a dense network of KP immunoreactive ( ir) fibers in the medial hypothalamus (Hrabovszky et al., 2010; Taziaux et al., 2016). The highest fiber densities are observed around the third ventricle, including the organum vasculosum of the lamina terminalis (OVLT), ventral periventricular area, anteromedial and anterolateral POA, paraventricular nucleus (PVN), at both magnocellular and parvocellular subdivisions, INF and stalk, dorsomedial hypothalamic nucleus, and the dorsal hypothalamic area. Overall, the density of KP-ir fibers diminishes with increasing distance from the third ventricle and labeled fibers become scarce in the ventromedial hypothalamic nucleus (VMN) and the lateral hypothalamic area. The majority of KP-ir cell bodies are observed in the INF. Some intensely labeled cell bodies are also found scattered periventricularly throughout the rostrocaudal extent of the hypothalamus. Furthermore, some less intensely immunostained KP cell bodies can be found in the PVN (Hrabovszky et al., 2010). Strong sex differences have been observed in the number of KP fibers and cell bodies (Hrabovszky et al., 2010; Taziaux et al., 2016; Fig. 18.1). Overall, consistently fewer ir fibers are observed in the hypothalamic sections of male individuals, compared with the patterns found in female brains. Furthermore, whereas KP-ir cell bodies are consistently observed in the rostral periventricular zone of female brains, no such labeled cells are found in males in this region (Hrabovszky et al., 2010). A further obvious sex difference is that the INF contain very few, if any, rather faintly labeled cells in males, compared with many heavily labeled cells in females (Hrabovszky et al., 2010; Taziaux et al., 2016). These sex differences are present in young adults (Hrabovszky et al., 2010) as well as in more aged subjects (Hrabovszky et al., 2011). Furthermore, KP expression appears to change over the lifetime, with a greater number of KP cell bodies in the infant/prepubertal and elderly period compared to the adult period (Taziaux et al., 2016; Figs. 18.1 and 18.2A–C). Thus the developmental pattern of INF KP expression throughout life seems to be quite similar between men and women, i.e., a moderate number of KP neurons in the infant/prepuberal period

followed by a lower number in the adult period and an increasing number in the elderly period. It is difficult to discern at which stage of life these sex differences emerge, but based on mean numbers, females have quantitatively more KP cell bodies compared to age-matched males at all stages of life analyzed (Taziaux et al., 2016), raising the possibility that this sex difference is already present before birth and might reflect organizational effects of sex steroids during early development. KP expression (as well as that of its receptor) has already been observed in fetuses at 15 weeks of gestation (Guimiot et al., 2012), but no sex comparison was made in this particular study. The changes in KP expression in the INF likely reflect fluctuations in the circulating concentration of sex steroid hormones throughout life. This is most obvious in females, with KP expression following a U-shaped curve, i.e., greater numbers of KP-expressing neurons in the infant/prepubertal and elderly periods compared to the adult period, thereby showing an inverted relationship with circulating estrogen levels. Accordingly, low KP mRNA levels are observed in estradiol-treated ovariectomized monkeys and in premenopausal women, whereas they are elevated in untreated ovariectomized monkeys and postmenopausal women (Rometo et al., 2007). Remarkably, in postmenopausal women, KP neurons show a hypertrophied morphology. This is observed by using in situ hybridization to measure KP mRNA levels (Rometo et al., 2007) as well as by using immunohistochemistry to measure KP peptide levels (Taziaux et al., 2016). This cellular hypertrophy suggests increased neuronal activity as well as that they are part of the neural circuitry regulating estrogen negative feedback. Since there does not seem to be a homologue of the rodent AVPv/PeN KP population in humans (Hrabovszky et al., 2010; Taziaux et al., 2016), it is very likely that the INF represents the hypothalamic site that mediates both negative and positive feedback actions of estradiol in humans, as was already suggested for primates (Knobil, 1980). This is further supported by the fact that GnRH neurons can predominantly be found in this same brain region, whereas in rodents, the majority of the GnRH neurons are located in the OVLT/POA. In both monkeys (Shahab et al., 2005) and humans (Guimiot et al., 2012), the KP receptor GPR54 has also been identified in the mediobasal hypothalamus.

NEUROKININ B EXPRESSION IN THE HUMAN HYPOTHALAMUS NKB-ir fibers can be found widely distributed throughout the human hypothalamus, whereas cell bodies are confined to a few specific nuclei (Taziaux et al., 2012; Fig. 18.3). Cell bodies expressing NKB vary in size

KISSPEPTIN AND NEUROKININ B EXPRESSION IN THE HUMAN HYPOTHALAMUS Male

Female

A

Infant/prepubertal

301

B

IIIV

IIIV

13 year-old girl

6 year-old boy

C

D IIIV

Adult

IIIV

27 year-old man

E

25 year-old woman

F

IIIV

Elderly

IIIV

76 year-old man

90 year-old woman

Fig. 18.1. Representative photomicrographs of kisspeptin expression in infundibular nucleus (INF) during the infant/prepubertal (A, B), adult (C, D), and elderly (E, F) periods in both sexes. The boxes in panels illustrate the area photographed at higher magnification. Note the numerous and intensely labeled kisspeptin neurons in the infant/prepubertal (B) and elderly periods (F) compared to the adult period (D) in the female INF. IIIV, third ventricle. Reproduced from Taziaux M, Staphorsius AS, Ghatei MA et al. (2016). Kisspeptin expression in the human infundibular nucleus in relation to sex, gender identity, and sexual orientation. J Clin Endocrinol Metab 101: 2380–2389, with permission of Endocrine Society, Oxford University Press.

depending on their localization. Small, oval-to-round NKB neurons are numerous in the central and medial portions of the bed nucleus of the stria terminalis (BST) and in the INF/ME complex, with the majority of the NKB cells being found in the middle and caudal part. Less intensely labeled cell bodies are also found

scattered periventricularly throughout the caudal extent of the hypothalamus. A small number of medium-sized round NKB neurons are observed in the nucleus basalis of Meynert (NBM), the anteroventral hypothalamus, the medial POA, and the posterior BST. Immunoreactive fibers for NKB exhibit varicosities and are abundantly

J. BAKKER Estimation of the total number of kisspeptin-ir neurons (mean ± SEM)

A

45000

Male Female

40000 35000 30000 25000 20000 15000 10000 5000

D

20000

Estimation of the total number of kisspeptin-ir neurons (mean ± SEM)

302

18000

0

Adult period

E 45000 Birth 40000 Infant/ prepubertal 35000 period 30000

Adult period

Estimation of the total number of kisspeptin-ir neurons (mean ± SEM)

Estimation of the total number of kisspeptin-ir neurons (mean ± SEM)

12000 10000

Elderly period

25000 20000 15000 10000 5000 0 0

10

20

30

40

50

60

70

80

90

T7

2000

S7 T3

Birth Adult period

T4

T6

18000 T1 T2

16000 14000 12000 10000 8000 6000

T5

T8 T7

4000 2000 T4

0 2

4

6

T3 T6 8 10 12 14 16 18 20

Duration of hormonal treatment (yrs)

F Infant/ prepubertal period

Trans man

0

Estimation of the total number of kisspeptin-ir neurons (mean ± SEM)

Estimation of the total number of kisspeptin-ir neurons (mean ± SEM)

T5

4000

100

C

30000

T8

6000

Age (yrs)

35000

*

8000

Elderly period

B

40000

T2

14000

0

Infant/ prepubertal period

45000

T1

16000

Elderly period

25000 20000 15000 10000 5000 0

9000 8000 7000 6000 5000 4000 3000 2000 1000 0

0

10

20

30

40

50

60

Age (yrs)

70

80

90

100

Men Men Men Women hetero hetero homo hetero HIV- HIV+ HIV+ HIV-

Fig. 18.2. (A) Estimation of the total number of kisspeptin-ir neurons in the INF of males and females during the infant/pubertal period (between 5 months and 14 years), the adult period (between 22 and 44 years), and the elderly period (between 58 and 90 years). Relationship between the total number of kisspeptin-ir neurons in the INF and age in females (B) and in males (C). The lines represent the quadratic regression curve. (D) Estimation of the total number of kisspeptin-ir neurons in the INF of adult men, women, and trans women. The open circle represents the number of KP neurons in the trans man sample. (E) Relationship between the total number of kisspeptin-ir neurons in the INF and duration of the hormonal treatment in transwomen. (F) Estimation of the total number of kisspeptin-ir neurons in the INF of HIV heterosexual men, HIV heterosexual women, HIV+ heterosexual men, and HIV+ homosexual men. Black dots represent individual data. * P < 0.05 vs adult men. S7, nontreated trans woman; T1–T8, transgender individuals. Panels (B) and (C) reproduced from Taziaux M, Staphorsius AS, Ghatei MA et al. (2016). Kisspeptin expression in the human infundibular nucleus in relation to sex, gender identity, and sexual orientation. J Clin Endocrinol Metab 101: 2380–2389, with permission of Endocrine Society, Oxford University Press.

Fig. 18.3. Schematic drawing from anterior to posterior (A–H) in a representative female hypothalamus to illustrate the distribution of neurokinin-B immunoreactive (NKB-ir) cells and fibers. NKB-ir cells are represented by close circles, whereas NKB-ir fibers are represented by dotted lines (single fibers or low density), continuous lines (moderate density), and crossed lines (high density). Abbreviations: AC, anterior commissure; AVH, anteroventral hypothalamic area; BSTc, bed nucleus of the stria terminalis, central part; BSTm, bed nucleus of the stria terminalis, medial part; BSTp, bed nucleus of the stria terminalis, posterior part; BSTv, bed nucleus of the stria terminalis, ventral part; DMH, dorsomedial nucleus of the hypothalamus; FO, fornix; INF, infundibular nucleus; LS, lateral septum; MPO, medial preoptic nucleus; NBM, nucleus basalis of Meynert; NTL, lateral tuberal nucleus; OC, optic chiasm; OT, optic tract; PEN, periventricular nucleus; PH, posterior hypothalamic nucleus; PVN, paraventricular nucleus; SCN, suprachiasmatic nucleus; SON, supraoptic nucleus; TMN, tubero mamillary nucleus; VMH, ventromedial nucleus of the hypothalamus. Reproduced from Taziaux M, Swaab DF, Bakker J (2012). Sex differences in the neurokinin B system in the human infundibular nucleus. J Clin Endocrinol Metab 97: E2210–2220, with permission of Endocrine Society, Oxford University Press.

304

J. BAKKER

Fig. 18.4. Representative photomicrographs illustrating the main localization of NKB-ir cells and fibers in the human hypothalamus: BSTm and BSTc (A), BSTv (B), INF (D), and ME (E). (C) and the insets in (D) and (E) show higher magnifications. Scale bars: 1 mm in (A), (D), and E; 0.5 mm in (B); 0.05 in (C) and in the insets. Abbreviations: AC, anterior commissure; BSTc, bed nucleus of the stria terminalis, central part; BSTm, bed nucleus of the stria terminalis, medial part; BSTv, bed nucleus of the stria terminalis, ventral part; INF; infundibular nucleus; LV, lateral ventricle; ME, median eminence; IIIV, third ventricle. Reproduced from Taziaux M, Swaab DF, Bakker J (2012). Sex differences in the neurokinin B system in the human infundibular nucleus. J Clin Endocrinol Metab 97: E2210–2220, with permission of Endocrine Society, Oxford University Press.

present throughout the hypothalamus, notably in the medial POA, lateral septum, anteroventral hypothalamus, NBM, VMN, the dorsomedial hypothalamus, and in the periventricular area but are particularly prominent in the BST (Fig. 18.4A–C) and the INF/ME complex (Fig. 18.4D and E). The distribution of NKB peptide is in agreement with the distribution of NKB mRNA (Chawla et al., 1997) and that of preprotachikinin B protein (Hrabovszky et al., 2010). Prominent sex differences can be observed at two sites, i.e., the INF and the pars tubularis (PT). Although the PT is totally devoid of NKB-ir during the infant/ pubertal period in both sexes, adult men display a dense

NKB innervation in the PT, which is totally absent in women (Taziaux et al., 2012; Fig. 18.5). By contrast, in elderly subjects, both men and women show high levels of NKB-immunoreactivity in this area. This sex difference is thus only visible in adulthood. Similar observations have been reported in the rat, in which a male-specific NKB innervation is observed around blood vessels of the external zone of the ME, compared with a more diffuse axonal wiring in the female (Ciofi et al., 2006). Moreover, in rats, the masculine phenotype emerges only at puberty and is activated by androgens (Ciofi et al., 2007). Based upon the pattern observed in adults, it is likely that androgens stimulate, whereas

KISSPEPTIN AND NEUROKININ B EXPRESSION IN THE HUMAN HYPOTHALAMUS

305

Fig. 18.5. Sexually dimorphic NKB innervation in the PT. (A) Schematic drawing depicting the location of the human PT. The box in (A) illustrates the area photographed from a 39-year-old man (B) and a 34-year-old woman (C). The boxes in (B) and (C) illustrate the area rephotographed at higher magnification in panels (b) and (c). The boundaries of the PT are drawn in dotted line (C). Scale bars: 0.5 mm in (A) and (B); 0.05 in (a) and (b). FO, fornix; INF, infundibular nucleus; NTL, lateral tuberal nucleus; OT, optic tract; PC, portal capillaries; PT, pars tuberalis. Reproduced from Taziaux M, Swaab DF, Bakker J (2012). Sex differences in the neurokinin B system in the human infundibular nucleus. J Clin Endocrinol Metab 97: E2210–2220, with permission of Endocrine Society, Oxford University Press.

estrogens inhibit this NKB innervation in the PT. Although the PT can be considered as a gateway uniquely placed to influence communication between the hypothalamus and the pituitary, there is currently no insight into the function of the male-specific NKB innervation of the PT. In the INF, there are strong sex differences in NKB-ir fibers with females showing a much higher density of NKB-ir fibers than men (measured as total volume of NKB-ir), but no significant sex difference in the number of NKB-expressing cell bodies (Taziaux et al., 2012; Figs. 18.6 and 18.7A–D). However, one study reported a somewhat higher regional density of NKB-ir somata in women compared to men (Hrabovszky et al., 2011). In contrast to KP expression, sex differences in NKB expression seem to emerge later in life, i.e., progressively from puberty to adulthood, where the female-dominant sex difference appears for the first time and continues over the years, well into the elderly period (Taziaux et al., 2012). Although it is generally thought that the sexual differentiation of the neuroendocrine hypothalamus does not proceed beyond the early postnatal period, puberty has been recognized as another period of development during which sex steroid hormones organize the nervous system (Sisk and Zehr, 2005). Several hypothalamic structures seem to differentiate later in life, such as the sexually dimorphic nucleus of the POA

(between 4 years of age and puberty; Swaab and Hofman, 1988), the darkly stained posteromedial components of the BST (around puberty; Allen and Gorski, 1990), and the central part of the BST (into adulthood; Chung et al., 2002). Furthermore, the sexually dimorphic NKB innervation in the rat does not become visible before puberty (Ciofi et al., 2007). Therefore, it is likely that the sex difference in human NKB expression also reflects organizational actions of sex steroid hormones during the prenatal period. However, this sex difference is only revealed postpubertally, suggesting that it needs to be activated by sex steroid hormones as well. As was also shown for KP-ir cell bodies (Taziaux et al., 2016), a hypertrophy of NKB-ir neurons was observed in postmenopausal women compared to premenopausal women, most likely reflecting the loss of ovarian estradiol (Taziaux et al., 2012; Fig. 18.7F). Taken together, sex differences in NKB expression can be observed (female > male) but are clearly less prominent in comparison to KP expression.

KDNy NEURONS IN HUMANS? In many species, KP and NKB can be found to be co-expressed with dynorphin A in a population of neurons in the ARC (or INF), also known as the KNDy neurons (Krajewski et al., 2005; Goodman et al., 2007;

306

J. BAKKER

Fig. 18.6. Representative photomicrographs of the sexually dimorphic NKB immunoreactivity in the INF between adult men (A) and women (B). The boxes in panels (A) and (B) illustrate the area rephotographed at higher magnification in (a) and (b). Scale bars: 0.5 mm in (A) and (B); 0.25 mm in (a) and (c); 0.05 mm in (b) and (d). IIIV, third ventricle. Reproduced from Taziaux M, Swaab DF, Bakker J (2012). Sex differences in the neurokinin B system in the human infundibular nucleus. J Clin Endocrinol Metab 97: E2210–2220, with permission of Endocrine Society, Oxford University Press.

Navarro et al., 2009; Wakabayashi et al., 2010). Indeed, overlap between NKB-ir and KP-ir perikarya have been observed when using dual fluorescence immunohistochemistry (Molnar et al., 2012). In this particular study, in which brain tissues from young men (50 years) and aged, postmenopausal women (>55 years) were compared, the majority of KP-ir perikarya (72.7%  6.0% in young men, 77.9%  5.9% in aged men, and 83.7%  3.7% in postmenopausal women) also contain NKB-ir. Similarly, the majority of NKB-ir neurons in aged human subjects (68.1%  6.8% in aged men and 71.3%  5.9% in postmenopausal women) contain KP-ir. However, in young men, most of the NKB-ir perikarya are single-labeled and only 35.8%  5.1% contain KP-ir. This suggests that KP expression in NKB neurons is highly sex and age dependent. Remarkably, additional colocalization studies have revealed unexpectedly low levels (if any) of dynorphin (A or B) signal in neuronal cell bodies in the INF from

young men (Hrabovszky et al., 2012) or in postmenopausal women (Skrapits et al., 2015). Dynorphin signal is absent from most KP neurons and fibers, in contrast with the extensive coexpression previously reported in rodents (Navarro et al., 2009), sheep (Goodman et al., 2007), and goats (Wakabayashi et al., 2010). By contrast, using in situ hybridization, Dynorphin expressing cells show a similar postmenopausal hypertrophy as KP and NKB neurons in the INF (Rometo and Rance, 2008), suggesting a possible role for Dynorphin in the negative feedback regulation of GnRH and LH release. Whether the absence of dynorphin in KP/NKB neurons indeed represents important species differences or whether it is caused by an increased postmortem degradation of dynorphin in KP/NKB neurons still needs to be clarified. By contrast, the tachykinin peptide substance P (SP) shows important sex differences in the INF, with greater numbers of SP-ir neurons in postmenopausal women compared to age-matched men (Hrabovszky et al., 2013). Furthermore, SP-ir can be detected in large

KISSPEPTIN AND NEUROKININ B EXPRESSION IN THE HUMAN HYPOTHALAMUS

NKB-ir volume (mm3)

0.14

Male Female

*#φ

0.12

*φ

0.10 0.08 0.06 0.04 0.02 0.00

5

5

7

Infant/ pubertal period

C

0.18

NKB-ir volume (mm3)

0.16 0.14

7

B

20000

Total number of NKB-ir neurons

0.16

15000 12500 10000 7500 5000 2500

6

6

0

Birth Infant/ pubertal period

Elderly period

Adult period

D

0.12 0.10 0.08 0.06 0.04

40000 35000 30000

5

Birth Infant/ pubertal period

7

7

6

6

Adult period Elderly period

Adult period

Elderly period

25000 20000 15000 10000 5000 0

0.00 0

10

20 30 40 50 60 70

0.10

*

0

80 90 100

Age (years)

NKB-ir volume (mm3)

5

Infant/ pubertal period

Adult period Elderly period

0.02

E

Male Female

17500

Total number of NKB-ir cells

A

307

Male Female

10

20 30 40 50 60 70

80 90 100

Age (years)

F

*

0.08 0.06 0.04 0.02 0.00

7

5

7

Men Trans Women Women

Fig. 18.7. Estimation of the volume occupied by NKB-ir (A) and of the total number of NKB-ir cells (B) in the INF of males and females during the infant/pubertal period (between 5 months and 14 years), the adult period (between 22 and 44 years), and the elderly period (between 58 and 90 years). * P < 0.05 vs male from the adult period; # P < 0.05 vs female from the adult period; f P < 0.05 vs male from the elderly period. Estimation of the total NKB-ir volume (C) and of the total number of NKB-ir cells (D) in the INF of both sexes from the postnatal period (5 months) to the elderly period (90 years). The lines represent the regression line for each sex. (E) Estimation of the volume occupied by NKB-ir in the INF of men, women, and trans women in adulthood. * P < 0.05 vs male. (F) Representative microphotographs illustrating the increase and hypertrophy of neurons expressing NKB in postmenopausal women (panel 2; 90-year-old) compared to premenopausal women (panel 1; 34-year-old) in the INF (40x objective). The boxes in panels (1) and (2) illustrate the area rephotographed at higher magnification (100 objective) in the insets. Scale bars: 0.05 mm in panels 1 and 2; 0.01 in the insets. Reproduced from Taziaux M, Swaab DF, Bakker J (2012). Sex differences in the neurokinin B system in the human infundibular nucleus. J Clin Endocrinol Metab 97: E2210–2220, with permission of Endocrine Society, Oxford University Press.

subsets of KP-ir and NKB-ir neurons in the INF: 31% of KP-ir and 25% of NKB-ir perikarya contain SP, whereas 16.5% of all labeled cell bodies exhibit all three neuropeptides (Skrapits et al., 2015). Dual- and triple-labeled

fibers are also detected in the infundibular stalk, raising the possibility that these peptides are coreleased into the portal circulation. The absence of dynorphin, but the presence of SP in KP/NKB neurons suggests that

308

J. BAKKER

compared to rodents and sheep, humans might use different neuropeptide signaling mechanisms to regulate sex steroid feedback on GnRH secretion.

CONNECTIONS OF KP AND NKB NEURONS WITH GnRH NEURONS Double immunohistochemistry for KP and GnRH has shown that KP-ir neurons in the INF are in contact with other KP-ir neurons (cell bodies and dendrites) as well as that they innervate somatic and dendritic compartments of GnRH neurons (Hrabovszky et al., 2010). There is a robust sex difference in the number of KP-ir contacts with higher innervation in postmenopausal women compared to age-matched men. Furthermore, 26% of these contacts of KP-ir neurons on GnRH bodies also contain NKB-ir in postmenopausal women, whereas it is lower (10%) in age-matched men (Hrabovszky et al., 2013). At present there is no immunohistochemical evidence for the presence of GPR54 on GnRH axons, albeit that NK3 receptors have been localized on GnRH axons in the rat (Krajewski et al., 2005). Furthermore, KP-ir axons have been found to form sporadic appositions onto GnRH-ir fibers in the INF stalk and around the portal capillary vessels of the postinfundibular ME (Hrabovszky et al., 2010). In contrast to rodents, in which most GnRH axons in the ME terminate in the external zone, many GnRH neurons in humans and monkeys travel long distances in the INF stalk and descend all the way down to the posterior pituitary (King and Anthony, 1984). GnRH fibers in this descending tract are also accompanied and occasionally contacted by KP-ir axons (Hrabovszky et al., 2010). The observation that KP-ir axons can be found around the portal vasculature of the human postinfundibular ME suggests that KP might be secreted into the hypophysial portal circulation and could act directly on gonadotrophs to affect LH secretion from the pituitary.

CLINICAL IMPLICATIONS OF KP AND NKB IN HUMAN REPRODUCTION The discovery of KP and NKB being potent regulators of GnRH secretion and thus critically involved in pubertal maturation and the maintenance of adult reproductive function has led to a huge scientific interest in the role of KP and NKB in controlling the reproductive axis in numerous animal models as well as in humans. (Hunjan and Abbara, 2019). This has opened up the possibility of manipulating KP signaling in disorders related to decreased GnRH signaling, such as hypogonadotropic hypogonadism and subfertility but also in disorders in which the reproductive axis needs to be suppressed, such as hormone-sensitive cancers.

A variety of clinical studies have shown that exogenously administered KP, either intravenously or through a bolus injection subcutaneously, reliably increases LH secretion in both men and women (reviewed in Clarke et al., 2015). This effect seems to be more pronounced in the preovulatory phase of the menstrual cycle (Dhillo et al., 2007), suggesting that it is dependent on circulating estradiol levels. Furthermore, it was found that KP increases LH pulsatility in women with hypothalamic amenorrhea (Jayasena et al., 2009), although chronic administration induces tachyphylaxis, i.e., desensitization due to a downregulation of the KP receptor. Furthermore, KP administration induces egg maturation in a dose-dependent manner in women undergoing in vitro fertilization treatment providing hope that KP may be successfully used to develop new or to improve existing fertility treatments. By contrast, NKB infusion has been shown to induce hot flushes in a group of healthy women (Jayasena et al., 2015) suggesting that NKB plays a role in thermoregulation which has also been shown in mice (Padilla et al., 2018). This opens the possibility of new treatments in which NKB signaling can be blocked pharmaceutically to reduce the appearance of hot flushes during the menopause or during treatment of sex-steroid sensitive cancers.

KP AND NKB EXPRESSION IN RELATION TO GENDER IDENTITY AND SEXUAL ORIENTATION In many animal species, including humans, strong sex differences are evident in KP and to a lesser extent in NKB expression. Most likely these sex differences reflect organizational actions of sex steroids during early development. KP expression can already be detected in human fetuses, but there is currently no information available on the presence of any sex differences before birth. However, in many species, sex differences in KP expression do not become visible before puberty, such as the sexually dimorphic KP population in the AVPv/ PeN in the mouse. Interestingly, it has been shown that perinatal exposure to testosterone or estradiol derived from neural aromatization of testosterone completely masculinizes this population in female rodents (Kaufmann et al., 2007). Conversely, estradiol is required during a specific prepubertal period to induce female-typical numbers of kisspeptin neurons (Clarkson et al., 2009; Bakker et al., 2010). Estradiol has thus both masculinizing and feminizing effects on this particular KP population depending on when it is present during development: when present perinatally in males, it induces masculinization but when present postnatally (after P15) in females, it induces feminization. This is highlighted by

KISSPEPTIN AND NEUROKININ B EXPRESSION IN THE HUMAN HYPOTHALAMUS results obtained in ArKO mice, i.e., both male and female ArKO mice show low numbers of KP neurons (Bakker et al., 2010), which suggests that this population has not been sexually differentiated in this mouse model due to the absence of estrogens in both sexes. The observed sex difference in KP expression reflecting organizational actions of sex steroid hormones in animal models makes KP an interesting candidate neuropeptide to study also in relationship to the sexual differentiation of the human brain and subsequently for potential effects of sex steroid hormones on the human brain. This question has in particularly been raised with regard to fundamental features of human existence, such as gender identity and sexual orientation. Gender dysphoria or gender incongruence has been defined as a marked and persistent incongruence between an individual’s experienced gender and the at birth assigned sex (DSM-5; ICD-11). A prominent hypothesis on its etiology proposes that the condition is related to the sexual differentiation of the brain and, specifically, to the fact that different critical periods exist for the development of the reproductive organs vs the brain, thereby positing that these processes have been affected differentially in individuals with gender incongruence (Swaab, 2007). Postmortem studies investigating the brains of individuals with gender incongruence have generally confirmed the sexual differentiation hypothesis. For example, a female-typical volume and number of neurons in the central subdivision of the bed nucleus of the stria terminalis and the third interstitial nucleus of the anterior hypothalamus (INAH-3) have been observed in transgender females (male sex assigned at birth and female gender identity) (Zhou et al., 1995; Kruijver et al., 2000; Garcia-Falgueras and Swaab, 2008). By contrast, functional and structural neuroimaging studies have shown more mixed results. Some studies reported a sex reversal, i.e., hypothalamic responses to the malechemosignal androstadienone were in line with their experienced gender in both transgender boys and girls (Burke et al., 2014), as well as masculinized neural activity patterns while performing a mental rotation task in transgender boys (female sex assigned at birth, male gender identity, treated with GnRH agonists to inhibit puberty at the time of the study; Burke et al., 2016). By contrast, at the structural level, gray matter volumes were largely concordant with their sex assigned at birth (e.g., Hoekzema et al., 2015). In two postmortem studies (Taziaux et al., 2012, 2016), KP and NKB expression were analyzed in transgender individuals. The MTF transsexual group consisted of seven sex-reassigned and estrogen-treated individuals and one individual who was not orchidectomized but hormonally treated. A nontreated individual with strong cross-gender identity feelings, which were

309

already present since early childhood, and one trans man (female sex assigned at birth, male gender identity) were also analyzed. Interestingly, female-typical numbers of KP-ir and NKB-ir neurons were observed in the INF of trans women who had undergone estrogen treatment and sex-reassignment in adulthood (Taziaux et al., 2012, 2016; Fig. 18.2D and E, Fig. 18.7D). Linear regression analyses indicated that the number of KP-ir neurons was not correlated with age in transgender individuals but is negatively, albeit not significantly, correlated with the duration of hormonal treatment (Fig. 18.2E), such that long-term estrogen treatment was associated with lower number of KP-ir neurons. Finally, it is interesting to note that the one trans man subject had a number of KP-ir neurons (144 neurons) in the male range, while the untreated trans gender subject (S7) showed an intermediate number of KP-ir neurons (1465 neurons). The sex reversal of the KP and NKB neuronal populations in the INF might be explained either by the presence of higher estrogen concentrations in the blood due to prolonged estrogen treatment or the lack of androgens due to orchidectomy (or antiandrogen treatment). Due to the low number of individuals as well as the strong variability in the duration of hormone treatment and KP/NKB expression, it is difficult to draw any strong conclusions on this particular observation of a female-typical KP and NKB population in trans women. It could indeed indicate a sex-atypical differentiation of the hypothalamus. However, it has previously been reported that the LH surge in trans women was male-typical, i.e., no LH surge, before sex reassignment and almost female-typical afterwards, i.e., significant rise in LH, suggesting that long-term estrogen treatment could feminize the gonadotropin response (Gooren, 1986). This could suggest that LH responses to estrogens in humans might not be as perinatally fixed as is the case in rodents but might depend on the nature of circulating sex steroids with androgens generally being inhibitory.

KP EXPRESSION IN RELATION TO SEXUAL ORIENTATION One theory of homosexual orientation is that it results from low fetal exposure to testosterone and that the absence of organizational effects of testosterone in homosexual men is responsible for a feminization of certain brain regions. Indeed, hypothalamic differences in relation to sexual orientation have been observed. For example, in accordance with this theory, a smaller INAH-3 (LeVay, 1991) and a larger anterior commissure (Allen and Gorski, 1992), both thus “female-like,” were observed in homosexual compared to heterosexual men.

310

J. BAKKER

By contrast, a larger suprachiasmatic nucleus (Swaab and Hofman, 1990) was observed in homosexual men compared to heterosexual men, whereas there is no sex difference observed in this nucleus. In addition, another hypothalamic nucleus that shows sex differences, i.e., the sexually dimorphic nucleus (SDN: male > female) did not differ between homosexual and heterosexual men. These findings do not support the hypothesis of a female-typical hypothalamus in homosexual men. In the study of Taziaux et al. (2016), the number of KP-ir neurons was compared between homosexual and heterosexual men. The number of KP-ir neurons was higher in homosexual men compared to heterosexual men and appeared to be “female-like” (Fig. 18.2F). However, since all homosexual subjects died of acquired immunodeficiency syndrome (AIDS), which has been associated with subnormal testosterone levels and hypogonadism (Sellmeyer and Grunfeld, 1996) and thus most likely a reduced negative feedback, brain tissues from heterosexual men who died from AIDS were included as additional controls. Increased KP expression was observed in HIV + heterosexual men and when compared to HIV + homosexual men, there were no significant differences, suggesting that the number of KP neurons does not vary with sexual orientation. Furthermore, this particular finding is not in support of the hypothesis of a female-typical hypothalamus in homosexual men.

CONCLUDING REMARKS As has been shown in many animal species, important sex differences can be observed in KP and NKB expression in the human hypothalamus. These sex differences might be related to sex differences in GnRH functioning, i.e., cyclical in women and tonic in men. It also most likely suggests that any positive feedback actions of estrogens on GnRH secretion are mediated by KP neurons in the INF and that there is no such role for more rostral KP neurons. Interestingly, although these sex differences probably reflect organizational actions of sex steroid hormones, the human INF KP system appears to remain sensitive to gonadal hormones throughout life since KP expression is higher in the infant/prepubertal and elderly periods, which are both characterized by low levels of circulating hormones. The sex reversal observed in KP and NKB expression in trans women might reflect, at least partially, an atypical sexual differentiation of the brain. It is important to note that these observations are based on postmortem brain material derived from a rather heterogeneous patient population. The high variability in the number of KP- and NKB-ir neurons especially in females could be partially explained by limitations related to the use of postmortem brain tissue, for which

conditions at death cannot be tightly controlled. Other confounding factors are the known sex differences in brain weights, but no such sex differences were observed for the INF between adult men and women. Finally, postmortem studies using immunohistochemistry and/or in situ hybridization remain important tools to increase our understanding of the human reproductive axis and associated neuroendocrine disorders. Future studies should focus on a further characterization of INF KP neurons in humans, such as whether they express steroid hormone receptors and to which brain areas they project, i.e., whether they project to other areas in the brain in addition to the median eminence. Interestingly, GPR54, the kisspeptin receptor, has been shown to be expressed outside the hypothalamus as well (Muir et al., 2001). This is particularly interesting in light of the recent finding of a specific, stimulatory role for kisspeptin in female sexual behavior in the mouse (Hellier et al., 2018). This latest discovery leads to the question whether kisspeptin might play a very similar role in women.

REFERENCES Allen LS, Gorski RA (1990). Sex difference in the bed nucleus of the stria terminalis of the human brain. J Comp Neurol 302: 697–706. Allen LS, Gorski RA (1992). Sexual orientation and the size of the anterior commissure in the human brain. Proc Natl Acad Sci U S A 89: 7199–7202. Bakker J, Pierman S, Gonzalez-Martinez D (2010). Effects of aromatase mutation (ArKO) on the sexual differentiation of kisspeptin neuronal numbers and their activation by same versus opposite sex urinary pheromones. Horm Behav 57: 390–395. Bao AM, Swaab DF (2011). Sexual differentiation of the human brain: relation to gender identity, sexual orientation and neuropsychiatric disorders. Front Neuroendocrinol 32: 214–226. Brock O, Bakker J (2013). The two kisspeptin neuronal populations are differentially organized and activated by estradiol in mice. Endocrinology 154: 2739–2749. Burke SM, Cohen-Kettenis PT, Veltman DJ et al. (2014). Hypothalamic response to the chemo-signal androstadienone in gender dysphoric children and adolescents. Front Endocrinol 5: 60. Burke SM, Kreukels BP, Cohen-Kettenis PT et al. (2016). Male-typical visuospatial functioning in gynephilic girls with gender dysphoria—organizational and activational effects of testosterone. J Psychiatry Neurosci 41: 395–404. Chawla MK, Gutierrez GM, Young 3rd WS et al. (1997). Localization of neurons expressing substance P and neurokinin B gene transcripts in the human hypothalamus and basal forebrain. J Comp Neurol 384: 429–442. Cheng G, Coolen LM, Padmanabhan V et al. (2010). The kisspeptin/neurokinin B/dynorphin (KNDy) cell population of the arcuate nucleus: sex differences and effects of prenatal testosterone in sheep. Endocrinology 151: 301–311.

KISSPEPTIN AND NEUROKININ B EXPRESSION IN THE HUMAN HYPOTHALAMUS Chung WC, De Vries GJ, Swaab DF (2002). Sexual differentiation of the bed nucleus of the stria terminalis in humans may extend into adulthood. J Neurosci 22: 1027–1033. Ciofi P, Leroy D, Tramu G (2006). Sexual dimorphism in the organization of the rat hypothalamic infundibular area. Neuroscience 141: 1731–1745. Ciofi P, Lapirot OC, Tramu G (2007). An androgen-dependent sexual dimorphism visible at puberty in the rat hypothalamus. Neuroscience 146: 630–642. Clarke H, Dhillo WS, Jayasena CN (2015). Comprehensive review on kisspeptin and its role in reproductive disorders. Endocrinol Metab 30: 124–141. Clarkson J, Herbison AE (2006). Postnatal development of kisspeptin neurons in mouse hypothalamus; sexual dimorphism and projections to gonadotropin-releasing hormone neurons. Endocrinology 147: 5817–5825. Clarkson J, d’Anglemont de Tassigny X, Moreno AS et al. (2008). Kisspeptin-GPR54 signaling is essential for preovulatory gonadotropin-releasing hormone neuron activation and the luteinizing hormone surge. J Neurosci 28: 8691–8697. Clarkson J, Boon WC, Simpson ER et al. (2009). Postnatal development of an estradiol-kisspeptin positive feedback mechanism implicated in puberty onset. Endocrinology 150: 3214–3220. D’Anglemont de Tassigny X, Fagg LA, Carlton MB et al. (2008). Kisspeptin can stimulate gonadotropin-releasing hormone (GnRH) by a direct action at GnRH nerve terminals. Endocrinology 149: 3926–3932. de Roux N, Genin E, Carel JC et al. (2003). Hypogonadotropic hypogonadism due to loss of function of the KiSS1-derived peptide receptor GPR54. Proc Natl Acad Sci U S A 100: 10972–10976. Dellovade TL, Merchenthaler I (2004). Estrogen regulation of neurokinin B gene expression in the mouse arcuate nucleus is mediated by estrogen receptor alpha. Endocrinology 145: 736–742. Dhillo WS, Chaudhri OB, Thompson EL et al. (2007). Kisspeptin-54 stimulates gonadotropin release most potently during the preovulatory phase of the menstrual cycle in women. J Clin Endocrinol Metab 92: 3958–3966. Dungan HM, Clifton DK, Steiner RA (2006). Minireview: kisspeptin neurons as central processors in the regulation of gonadot-tropin releasing hormone secretion. Endocrinology 147: 1154–1158. Garcia-Falgueras A, Swaab DF (2008). A sex difference in the hypothalamic uncinate nucleus: relationship to gender identity. Brain 131: 3132–3146. Goh HH, Ratnam SS (1988). The LH surge in humans: its mechanism and sex difference. Gynecol Endocrinol 2: 165–182. Goldsmith PC, Thind KK, Song T et al. (1990). Location of the neuroendocrine gonadotropin-releasing hormone neurons in the monkey hypothalamus by retrograde tracing and immunostaining. J Neuroendocrinol 2: 157–168. Goodman RL, Lehman MN, Smith JT et al. (2007). Kisspeptin neurons in the arcuate nucleus of the ewe express both dynorphin A and neurokinin B. Endocrinology 148: 5752–5760.

311

Gooren L (1986). The neuroendocrine response of luteinizing hormone to estrogen administration in the human is not sex specific but dependent on the hormonal environment. J Clin Endocrinol Metab 63: 589–593. Goubillon ML, Forsdike RA, Robinson JE et al. (2000). Identification of neurokinin B-expressing neurons as an highly estrogen-receptive, sexually dimorphic cell group in the ovine arcuate nucleus. Endocrinology 141: 4218–4225. Guimiot F, Chevrier L, Dreux S et al. (2012). Negative fetal FSH/LH regulation in late pregnancy is associated with declined kisspeptin/KISS1R expression in the tuberal hypothalamus. J Clin Endocrinol Metab 97: E2221–E2229. Hellier V, Brock O, Candlish M et al. (2018). Female sexual behavior in mice is controlled by kisspeptin neurons. Nat Commun 9: 400. Herbison AE (2006). Physiology of the gonadotropinreleasing hormone neuronal network, Academic Press, San Diego. Herbison AE, Theodosis DT (1992). Localization of oestrogen receptors in preoptic neurons containing neurotensin but not tyrosine hydroxylase, cholecystokinin or luteinizing hormone-releasing hormone in the male and female rat. Neuroscience 50: 283–298. Hoekzema E, Schagen SE, Kreukels BP et al. (2015). Regional volumes and spatial volumetric distribution of gray matter in the gender dysphoric brain. Psychoneuroendocrinology 55: 59–71. Hoffman GE, Le WW, Franchescini I et al. (2011). Expression of fos and in vivo median eminence release of LHRH identifies an active role for preoptic area kisspeptin neurons in synchronized surges of LH and LHRH in the ewe. Endocrinology 152: 214–222. Hrabovszky E, Ciofi P, Vida B et al. (2010). The kisspeptin system of the human hypothalamus: sexual dimorphism and relationship with gonadotropin-releasing hormone and neurokinin B. Eur J Neurosci 31: 1984–1998. Hrabovszky E, Molna´r CS, Sipos MT et al. (2011). Sexual dimorphism of kisspeptin and neurokinin B immunoreactive neurons in the infundibular nucleus of aged men and women. Front Endocrinol 2: 80. Hrabovszky E, Sipos MT, Molnar CS et al. (2012). Low degree of overlap between kisspeptin, neurokinin B, and dynorphin immunoreactivities in the infundibular nucleus of young male human subjects challenges the KNDy neuron concept. Endocrinology 153: 4978–4989. Hrabovszky E, Borsay BA, Racz K et al. (2013). Substance P immunoreactivity exhibits frequent colocalization with kisspeptin and neurokinin B in the human infundibular region. PloS One 8: e72369. Hunjan T, Abbara A (2019). Clinical translational studies of kisspeptin and Neurokinin B. Semin Reprod Med 37: 119–124. Jayasena CN, Nijher GM, Chaudhri OB et al. (2009). Subcutaneous injection of kisspeptin-54 acutely stimulates gonadotropin secretion in women with hypothalamic amenorrhea, but chronic administration causes tachyphylaxis. J Clin Endocrinol Metab 94: 4315–4323.

312

J. BAKKER

Jayasena CN, Comninos AN, Stefanopoulou E et al. (2015). Neurokinin-B administration induces hot flushes in women. Sci Rep 5: 8466. Kauffman AS (2010). Gonadal and non-gonadal regulation of sex differences in hypothalamic Kiss1 neurons. J Neuroendocrinol 22: 682–691. Kaufmann AS, Gottsch ML, Roa J et al. (2007). Sexual differentiation of Kiss1 gene expression in the brain of the rat. Endocrinology 148: 1774–1783. King JC, Anthony EL (1984). LHRH neurons and their projections in humans and other mammals: species comparisons. Peptides 5: 195–207. Knobil E (1980). The neuroendocrine control of the menstrual cycle: recent progress in hormone research. Recent Prog Horm Res 36: 53–88. Krajewski SJ, Anderson MJ, Iles-Shih L et al. (2005). Morphologic evidence that neurokinin B modulates gonadotropin-releasing hormone secretion via neurokinin 3 receptors in the rat median eminence. J Comp Neurol 489: 372–386. Kruijver FP, Zhou JN, Pool CW et al. (2000). Male-to-female transsexuals have female neuron numbers in a limbic nucleus. J Clin Endocrinol Metab 85: 2034–2041. Lamprecht SA, Kohen F, Ausher J et al. (1976). Hormonal stimulation of oestradiol-17 beta release from the rat ovary during early postnatal development. J Endocrinol 68: 343–344. Lehman MN, Karsch FJ (1993). Do gonadotropin-releasing hormone, tyrosine hydroxylase-, and beta-endorphinimmunoreactive neurons contain estrogen receptors? A double-label immunocytochemical study in the Suffolk ewe. Endocrinology 133: 887–895. LeVay S (1991). A difference in hypothalamic structure between heterosexual and homosexual men. Science 253: 1034–1037. Molnar CS, Vida B, Sipos MT et al. (2012). Morphological evidence for enhanced kisspeptin and neurokinin B signaling in the infundibular nucleus of the aging man. Endocrinology 153: 5428–5439. Moore AM, Coolen LM, Porter DT et al. (2018). KNDy cells revisited. Endocrinology 159: 3219–3234. Muir AI, Chamberlain L, Elshourbagy NA et al. (2001). AXOR12, a novel human G protein-coupled receptor, activated by the peptide KiSS-1. J Biol Chem 276: 28969–28975. Navarro VM, Gottsch ML, Chavkin C et al. (2009). Regulation of gonadotropin-releasing hormone secretion by kisspeptin/dynorphin/neurokinin B neurons in the arcuate nucleus of the mouse. J Neurosci 29: 11859–11866. Navarro VM, Gottsch ML, Wu M et al. (2011). Reguation of NKB pathways and their roles in the control of Kiss1 neurons in the arcuate nucleus of the male mouse. Endocrinology 152: 4265–4275. Padilla SL, Johnson SW, Barker FD et al. (2018). A neural circuit underlying the generation of hot flushes. Cell Rep 24: 271–277. Ramaswamy S, Guerriero KA, Gibbs RB et al. (2008). Structural interactions between kisspeptin and GnRH neurons in the mediobasal hypothalamus of the male rhesus

monkey (Macaca mulatta) as revealed by double immunofluorescence and confocal microscopy. Endocrinology 149: 4387–4395. Rance NE, McMullen NT, Smialek JE et al. (1990). Postmenopausal hypertrophy of neurons expressing the estrogen receptor gene in the human hypothalamus. J Clin Endocrinol Metab 71: 79–85. Robertson JL, Clifton DK, de la Iglesia HO et al. (2009). Circadian regulation of Kiss1 neurons: implications for timing the preovulatory GnRH/LH surge. Endocrinology 150: 3664–3671. Rometo AM, Rance NE (2008). Changes in prodynorphin gene expression and neuronal morphology in the hypothalamus of postmenopausal women. J Neuroendocrinol 20: 1376–1381. Rometo AM, Krajewski SJ, Voytko ML et al. (2007). Hypertrophy and increased kisspeptin gene expression in the hypothalamic infundibular nucleus of postmenopausal women and ovariectomized monkeys. J Clin Endocrinol Metab 92: 2744–2750. Roseweir AK, Kauffman AS, Smith JT et al. (2009). Discovery of potent kisspeptin antagonists delineate physiological mechanisms of gonadotropin regulation. J Neurosci 29: 3920–3929. Ruiz-Pino F, Navarro VM, Bentsen AH et al. (2012). Neurokinin B and the control of the gonadotropic axis in the rat: developmental changes, sexual dimorphism, and regulation by gonadal steroids. Endocrinology 153: 4818–4829. Sellmeyer DE, Grunfeld C (1996). Endocrine and metabolic disturbances in human immunodeficiency virus infection and the acquired immune deficiency syndrome. Endocr Rev 17: 518–532. Seminara SB, Messager S, Chatzidaki EE et al. (2003). The GPR54 gene as a regulator of puberty. N Engl J Med 349: 1614–1627. Shahab M, Mastronardi C, Seminara SB et al. (2005). Increased hypothalamic GPR54 signaling: a potential mechanism for initiation of puberty in primates. Proc Natl Acad Sci U S A 102: 2129–2134. Simonneaux V, Ansel L, Revel FG et al. (2009). Kisspeptin and the seasonal control of reproduction in hamsters. Peptides 30: 146–153. Sisk CL, Zehr JL (2005). Pubertal hormones organize the adolescent brain and behavior. Front Neuroendocrinol 26: 163–174. Skrapits K, Borsay BA, Herczeg L et al. (2015). Neuropeptide co-expression in hypothalamic kisspeptin neurons of laboratory animals and the human. Front Neurosci 9: 29. Smith JT, Cunningham MJ, Rissman EF et al. (2005a). Regulation of Kiss1 gene expression in the brain of the female mouse. Endocrinology 146: 3686–3692. Smith JT, Dungan HM, Stoll EA et al. (2005b). Differential regulation of KiSS-1 mRNA expression by sex steroids in the brain of the male mouse. Endocrinology 146: 2976–2984. Smith JT, Popa SM, Clifton DK et al. (2006). Kiss1 neurons in the forebrain as central processors for generating the preovulatory luteinizing hormone surge. J Neurosci 26: 6687–6694.

KISSPEPTIN AND NEUROKININ B EXPRESSION IN THE HUMAN HYPOTHALAMUS Smith JT, Shahab M, Pereira A et al. (2010). Hypothalamic expression of KISS1 and gonadotropin inhibitory hormone genes during the menstrual cycle of a non-human primate. Biol Reprod 83: 568–577. Stamou MI, Georgopoulos NA (2018). Kallmann syndrome: phenotype and genotype of hypogonadotropic hypogonadism. Metabolism 86: 124–134. Sullivan KA, Witkin JW, Ferin M et al. (1995). Gonadotropinreleasing hormone neurons in the rhesus macaque are not immunoreactive for the estrogen receptor. Brain Res 685: 198–200. Swaab DF (2007). Sexual differentiation of the brain and behavior. Best Pract Res Clin Endocrinol Metab 21: 431–444. Swaab DF, Hofman MA (1988). Sexual differentiation of the human hypothalamus: ontogeny of the sexually dimorphic nucleus of the preoptic area. Brain Res Dev Brain Res 44: 314–318. Swaab DF, Hofman MA (1990). An enlarged suprachiasmatic nucleus in homosexual men. Brain Res 537: 141–148. Taziaux M, Swaab DF, Bakker J (2012). Sex differences in the neurokinin B system in the human infundibular nucleus. J Clin Endocrinol Metab 97: E2210–E2220. Taziaux M, Staphorsius AS, Ghatei MA et al. (2016). Kisspeptin expression in the human infundibular nucleus in relation to sex, gender identity, and sexual orientation. J Clin Endocrinol Metab 101: 2380–2389.

313

Tena-Sempere M (2005). Hypothalamic KiSS-1: the missing link in gonadotropin feedback control? Endocrinology 146: 3683–3685. Tolson KP, Garcia C, Yen S et al. (2014). Impaired kisspeptin signaling decreases metabolism and promotes glucose intolerance and obesity. J Clin Invest 124: 3075–3079. Topaloglu AK, Reimann F, Guclu M et al. (2009). TAC3 and TACR3 mutations in familial hypogonadotropic hypogonadism reveal a key role for neurokinin B in the central control of reproduction. Nat Genet 41: 354–358. Vreeburg JT, van der Vaart PD, van der Schoot P (1977). Prevention of central defeminization but not masculinization in male rats by inhibition neonatally of oestrogen biosynthesis. J Endocrinol 74: 375–382. Wakabayashi Y, Nakada T, Murata K et al. (2010). Neurokinin B and dynorphin A in kisspeptin neurons of the arcuate nucleus participate in generation of periodic oscillation of neural activity driving pulsatile gonadotropin-releasing hormone secretion in the goat. J Neurosci 30: 3124–3132. Wintermantel TM, Campbell RE, Porteous R et al. (2006). Definition of estrogen receptor pathway critical for estrogen positive feedback to gonadotropin-releasing hormone neurons and fertility. Neuron 52: 271–280. Zhou JN, Hofman MA, Gooren LJ et al. (1995). A sex difference in the human brain and its relation to transsexuality. Nature 378: 68–70.

This page intentionally left blank

Handbook of Clinical Neurology, Vol. 180 (3rd series) The Human Hypothalamus: Middle and Posterior Region D.F. Swaab, F. Kreier, P.J. Lucassen, A. Salehi, and R.M. Buijs, Editors https://doi.org/10.1016/B978-0-12-820107-7.00019-7 Copyright © 2021 Elsevier B.V. All rights reserved

Chapter 19

The infundibular peptidergic neurons and glia cells in overeating, obesity, and diabetes MARTIN J.T. KALSBEEK1,2* AND CHUN-XIA YI1,2,3 1

Laboratory of Endocrinology, Amsterdam University Medical Center, University of Amsterdam, Amsterdam Gastroenterology Metabolism, Amsterdam, The Netherlands

2

Netherlands Institute for Neuroscience, an Institute of the Royal Netherlands Academy of Arts and Sciences, Amsterdam, The Netherlands 3

Department of Endocrinology and Metabolism, Amsterdam University Medical Center, University of Amsterdam, Amsterdam, The Netherlands

Abstract Dysfunctional regulation of energy homeostasis results in increased bodyweight and obesity, eventually leading to type 2 diabetes mellitus. The infundibular nucleus (IFN) of the hypothalamus is the main regulator of energy homeostasis. The peptidergic neurons and glia cells of the IFN receive metabolic cues concerning energy state of the body from the circulation. The IFN can monitor hormones like insulin and leptin and nutrients like glucose and fatty acids. All these metabolic cues are integrated into an output signal regulating energy homeostasis through the release of neuropeptides. These neuropeptides are released in several inter- and extrahypothalamic brain regions involved in regulation of energy homeostasis. This review will give an overview of the peripheral signals involved in the regulation of energy homeostasis, the peptidergic neurons and glial cells of the IFN, and will highlight the main intra-hypothalamic projection sites of the IFN.

INTRODUCTION1 Obese individuals are at high risk of developing a variety of comorbidities including type 2 diabetes mellitus (T2DM). Losing the excess weight by intensive weight management showed a remission to a nondiabetic state in almost half of the individuals (Lean et al., 2018), indicating obesity causes T2DM. The excess weight in obesity results from an unbalance between energy intake and energy expenditure. While complex biologic systems regulate both energy intake and energy expenditure, the hypothalamus sits in the center of these systems

controlling them. The crucial role of the hypothalamus in feeding behavior was first shown in the 1940s, when hypothalamic lesions resulted in abnormal feeding behavior in rats (Brooks et al., 1946; Bertozzi, 1950). Since then, our understanding of the hypothalamic control of food intake and energy expenditure has expanded greatly, bulk of pharmacological and genetic approaches have revealed a variety of neuropeptides involved. Pharmaceutical companies spend tremendous efforts in developing therapeutics regulating neuropeptide function in order to combat obesity and T2DM; however, so far with limited results.

1

Abbreviations used in the chapter are listed at the end of the chapter before References section.

*Correspondence to: Martin Jan Kalsbeek, M.Sc., Laboratory of Endocrinology, Nederlands Herseninstituut, Meibergdreef 47, Amsterdam 1105BA, The Netherlands. Tel: +061-724-2141, E-mail: [email protected]

316

M.J.T. KALSBEEK AND C.-X. YI

The infundibular nucleus (IFN) is a brain region in the mediobasal part of the hypothalamus, situated adjacent to the third ventricle and above the median eminence (Krisch and Leonhardt, 1978; Broadwell et al., 1983; Shaver et al., 1992; Swaab, 2003). In this region metabolic cues from the periphery, like insulin and leptin, but also nutritional signals like glucose and fatty acids can penetrate the brain (Cottrell and Ferguson, 2004). The IFN integrates these signals into an output signal to second-order neurons controlling energy homeostasis. In the IFN, two distinct populations of peptidergic neurons reside which have opposing effects on energy homeostasis. The proopiomelanocortin (POMC) expressing neurons inhibit food intake and stimulate energy expenditure, whereas the agouti-related protein (AgRP)/ neuropeptide Y (NPY) expressing neurons stimulate food intake and inhibit energy expenditure. Peptidergic neurons exert their effect mainly through the release of neuropeptides to second-order neurons in innerhypothalamic and extrahypothalamic brain regions. Furthermore, both neuronal populations in the IFN possess a variety of receptors for peripheral metabolic signals but respond differently to the same stimuli. Next to the neuronal regulation of energy homeostasis, nonneuronal cells like astrocytes and microglia also play an active role in the control of metabolism (Freire-Regatillo et al., 2017). With their wide range of cell surface receptors, both cell types play an important role in the functional interactions of specific neuronal subpopulations (GarcíaCáceres et al., 2019). This review will focus on the role of the main neuropeptides in the IFN in the human setting of feeding behavior, obesity, and diabetes. First, we will give an overview of the peripheral signals that can affect the activity of infundibular peptidergic neurons and glial cells. Next, we will discuss the role of the different neuropeptides involved in food intake and energy expenditure. Last, we will highlight some of the main hypothalamic downstream projecting brain areas of the IFN.

PERIPHERAL SIGNALS THAT INFLUENCE THE IFN The IFN receives information about the energy status of the body in two ways, either directly through the sensing of available nutrients or indirectly through the action of hormones, which are produced in response to certain cues concerning energy intake or expenditure (Blouet and Schwartz, 2010). Some of these molecules are quite large, and how they pass the blood–brain barrier (BBB) is still not completely clear (Banks, 2001; Murphy, 2015). However, the IFN locates right on top of the median eminence, which is devoid of a BBB (Rodriguez et al., 2010), and this lack of a BBB is thought to enable large molecules to passively diffuse and reach the IFN.

Nevertheless, the IFN cells express receptors for glucose and fatty acids and they have receptors for the hormones insulin, leptin, ghrelin, peptide YY (PYY), and glucagonlike peptide 1 (Perry and Wang, 2012; Timper and Br€uning, 2017). All of these hormones and nutrients modify the IFN regulation of energy homeostasis; we will discuss the most important ones later.

Insulin Insulin is probably best understood for its use as an antidiabetic agent in the treatment of type 1 and type 2 diabetes mellitus (T1DM and T2DM), which is vital to maintain a “healthy” energy balance in these patients. In T1DM, the insulin producing b-cells of the pancreas are destroyed and very little to no insulin is produced. As insulin is vital for survival, these patients need to inject endogenous insulin on a daily basis to survive. In T2DM, the b-cells still function and can produce insulin, but long-lasting episodes of hyperglycemic events have rendered the target cells insensitive to insulin, e.g., insulin resistant. These patients use high amounts of endogenous insulin or drugs that can increase insulin sensitivity to still provoke an insulin response and activate the downstream signaling pathways. Insulin is the only glucose-lowering hormone known in the human body, and the pancreatic b-cells release insulin in response to the postprandial rise of plasma glucose levels. Insulin then decreases plasma glucose through a whole pallet of actions in peripheral cells, like increased uptake of glucose into muscle and fat tissue, stimulation of glycogenesis, and inhibition of gluconeogenesis (Petersen and Shulman, 2018). Next to these direct glucose-lowering properties in the periphery, central administration of insulin can indirectly lower glucose by inhibition of food intake (Woods et al., 1979). Furthermore, intranasal insulin was shown to decrease body fat in men, unfortunately not in women (Hallschmid et al., 2004). Even though glucose uptake in the brain is thought to be independent of insulin signaling, insulin receptors (IRs) are widely expressed in the brain, especially in the hypothalamus (Havrankova et al., 1978). Both POMC neurons and AgRP/NPY neurons have been shown to express IRs on their cell surface (Benoit et al., 2002; Lam et al., 2017). The anorexigenic function of insulin is mediated by a stimulation of the POMC neurons and an inhibition of the AgRP/NPY neurons (Schwartz et al., 1992; Wang and Leibowitz, 1997; Havel et al., 2000; Benoit et al., 2002). Interestingly, while the ablation of all neuronal IRs resulted in obese mice (Bruning et al., 2000), the phenotype of mice lacking IRs specifically in POMC or AgRP neurons was unaltered (Konner et al., 2007). This indicated that insulin signaling in NPY neurons is crucial in the regulation

THE INFUNDIBULAR PEPTIDERGIC NEURONS AND GLIA CELLS of food intake and energy expenditure. Indeed, specific knock out of IRs on NPY neurons resulted in obesity in mice (Loh et al., 2017). Astrocytes also express IRs, which play a supporting role in the glucose sensing mechanism of the IFN. Insulin signaling in astrocytes is important in the regulation of glucose uptake into the brain; ablation of IR in astrocytes reduces glucose-induced activation of POMC neurons and impairs physiologic responses to changes in glucose availability (García-Cáceres et al., 2016). The IR is also expressed on microglia, and the effect of insulin seems to be antiinflammatory (Brabazon et al., 2018). This is interesting in relation to the control of energy homeostasis, as hypothalamic microglia activity has been shown to play a role in the dysfunctional regulation of energy homeostasis in diet-induced obesity, not only in rodents but also in humans (Thaler et al., 2012; Gao et al., 2014, 2017a, b; Reis et al., 2015; Valdearcos et al., 2015; Baufeld et al., 2016; Yi et al., 2017).

Leptin Adipocytes are the main site of leptin production, and circulating leptin levels are highly correlated with body mass index (BMI) (Maffei et al., 1995). The function of leptin was discovered in the early 1990s, when the leptin gene was identified (Zhang et al., 1994). The ob/ob and db/db mice do not produce leptin or leptin receptors, respectively, and both mutations result in a severely obese phenotype indicating that in the absence of leptin signaling, food intake is not inhibited. Obese humans seem to be resistant to leptin as they have high levels of the circulating hormone indicating its failure to decrease food intake (Myers et al., 2010). In humans, congenital leptin deficiency is very rare, but it results in early onset obesity and can be treated with leptin substitution therapy (Funcke et al., 2014), interestingly in such condition no leptin resistance is detected. The leptin molecule is quite large, and how it crosses the BBB and reaches the neurons in the IFN remains unclear. Specialized ependymal cells in the median eminence called tanycytes have been proposed to play an important role in BBB transport of leptin (Langlet et al., 2013; Balland et al., 2014). It was also shown that this transport is still intact in obese mice, indicating leptin resistance is not due to impaired transport (Harrison et al., 2019). Even though the mechanism of BBB transport of leptin is not completely known, it is clear that both POMC neurons and NPY/AgRP neurons express leptin receptors and they respond to leptin (Balthasar et al., 2004; Takahashi and Cone, 2005; Caron et al., 2018). Leptin stimulates POMC neuron activity and inhibits NPY/AgRP neurons, resulting in an anorexigenic effect. Leptin can influence the neuronal control of energy homeostasis indirectly by its action on both astrocytes

317

and microglia. Specific knockout of leptin receptor in astrocytes leads to altered feeding behavior; diminished leptin-induced anorexia; and enhanced fasting- or ghrelin-induced hyperphagia (Kim et al., 2014). In microglia, specific leptin receptor knockout results in mice with higher body weight and hyperphagia with an associated decrease of POMC neurons (Gao et al., 2018).

Gut hormones The brain also receives hormonal signals from the gastrointestinal (GI) tract; the GI tract produces these hormones mainly in relation to food ingestion. Ghrelin, also known as the hunger hormone, is produced by the stomach and stimulates feeding behavior (Tschop et al., 2000). Ghrelin levels are highest right before an anticipated meal, and the ingested meal suppresses ghrelin levels proportionately to the caloric content of the meal (Kirchner et al., 2012). Peripheral ghrelin administration induces feelings of hunger and increases food intake in lean and obese individuals. Ghrelin exerts its orexigenic effect through the IFN neurons, where it stimulates NPY/ AgRP neurons and inhibits POMC neurons (Cowley et al., 2003). While ghrelin receptors are expressed by almost all NPY/AgRP neurons, only a small proportion of the POMC neurons express ghrelin receptors (Willesen et al., 1999). Additionally, ghrelin can mediate its metabolic action partly through astrocytes (Frago and Chowen, 2017). By changing astrocytic glucose and glutamate metabolism, ghrelin alters the levels of signals/ nutrients reaching the neighboring neurons (FuenteMartin et al., 2016). Furthermore, ghrelin has antiinflammatory properties, which has also been shown in microglia (Theil et al., 2009). Specialized enteroendocrine cells in the distal GI tract, called L-cells, produce peptide YY in response to food ingestion. Intravenous infusion showed PYY inhibits food intake in both lean and obese individuals (Batterham et al., 2003). In the brain, PYY signals through the NPY receptors, and in the IFN, the Y2-receptor subtype is highly expressed (Batterham et al., 2002). The NPY/ AgRP neurons express the Y2-subtype receptor and binding with PYY inhibits these neurons. Even though PYY was shown to stimulate POMC neurons, this does not seem to be essential for its anorexigenic effect, as PYY was shown to still inhibit food intake in MC4R and POMC knockout mice (Ghamari-Langroudi et al., 2005). When PYY is injected ICV it can bind to the Y1 and Y5 subtype receptor of second-order neurons, which results in a stimulatory effect on food intake, similar to NPY (Karra et al., 2009). Opposed to leptin, PYY levels are low in obese individuals and PYY infusion inhibits food intake (Batterham et al., 2003), which makes it interesting as a potential antiobesity drug. However, the endogenous form of PYY has

318

M.J.T. KALSBEEK AND C.-X. YI

a short half-life, rendering it unpractical as a therapeutic compound. Several efforts to prolong the half-life of PYY have been performed and are still under investigation (Hope et al., 2018). The L-cells from the distal GI tract also produce glucagon-like peptide-1 (GLP-1), again in response to food ingestion (Holst, 2007). The release of GLP-1 has many peripheral effects including an increase of glucosestimulated insulin production (Jones et al., 2018). In the brain, GLP-1 directly activates POMC neurons and indirectly inhibits NPY/AgRP neurons, collectively inhibiting food intake (Baggio and Drucker, 2014). In overweight and obese humans, treatment with GLP-1 receptor agonists showed to induce significant weight loss, irrespective of whether the patients had T2DM or not (Vilsboll et al., 2012), making it an interesting target as a potential antiobesity drug.

Nutrients The (unhealthy) western diet contains high amounts of sugar and fats, which are reflected in increased levels of these components in the general circulation, and the IFN can sense them directly. Glucokinase is a key enzyme in the neuronal glucose sensing machinery and it is expressed by both POMC and NPY/AgRP neurons (De Backer et al., 2016). POMC neurons are glucose responsive and upon stimulation; they release a-melanocyte-stimulating hormone (MSH) in a dose-dependent manner (Parton et al., 2007). Furthermore, glucose sensing by POMC neurons became defective in diet-induced obese animals, which indicates that neuronal glucose sensing plays a role in the development of T2DM (Parton et al., 2007). Glucose sensing neurons in the IFN can directly stimulate the pancreatic B-cells to secrete insulin (Rosario et al., 2016). NPY/AgRP neurons seem to be inhibited by glucose (Cheng et al., 2008). The glucose transporter 2 (GLUT2) is important for glucose sensing, and inhibition of GLUT2-mediated glucose sensing increases food intake (Bady et al., 2006; Stolarczyk et al., 2010). Interestingly, GLUT2 signaling seems to be required specifically in astrocytes (Marty et al., 2005). Several studies have shown that astrocytes have an important role in glucose sensing in the IFN (Allard et al., 2014; Camandola, 2018). Fluctuating glucose levels have also been shown to activate microglia (Hsieh et al., 2019), fatty acid signaling is mediated mainly by free fatty acid receptors (FFARs) (Falomir-Lockhart et al., 2019), and both neuronal populations in the IFN express FFARs (Dragano et al., 2017). The short-chain fatty acid acetate can increase POMC and reduce AgRP expression in the hypothalamus (Frost et al., 2014), and the long-chain fatty acid increases POMC neuron excitability (Jo et al., 2009). Next to direct effects on IFN neurons, fatty acids

influence the immune response in the hypothalamus, indirectly affecting central control of energy homeostasis. Several studies have shown that fatty acids can either cause or attenuate hypothalamic inflammation in diet-induced obesity (Mendes et al., 2018). Amino acid levels are also a cue of nutrient availability and in the periphery cells use mTOR signaling to sense amino acids (Goberdhan et al., 2016). In the IFN, mTOR signaling has been shown to play an active role in the regulation of food intake and can be activated by the amino acid L-leucine (Cota et al., 2006). While it seems clear that the amino acid leucine is sensed by the brain, its role in food intake remains controversial (Zampieri et al., 2013; Laeger et al., 2014; Maurin et al., 2014).

NEUROPEPTIDES AND THEIR RECEPTORS Melanocortins are peptide hormones derived from POMC and induce their effect through the melanocortin receptors (MCRs). Over 2 centuries ago, it was already discovered that mutations in the POMC gene are associated with early onset obesity and severe overeating (Krude et al., 1998). POMC is a precursor polypeptide, which undergoes posttranslational modifications and can gives rise to the signal peptides a, b, and Y-MSH, adrenocorticotropic hormone and b-endorphin. The main melanocortin related to food intake is a-MSH, which inhibits food intake. Injection of an a-MSH analog into the third ventricle inhibits food intake in food deprived rats (Brown et al., 1998). In humans, the expression of a-MSH in postmortem brain tissue did not correlate with BMI but was significantly decreased in diabetic patients (Alkemade et al., 2012). Next to the effect on neuronal control of food intake, a-MSH also has an antiinflammatory effect (Delgado et al., 1998). Both astrocytes and microglia express the MCRs, and activation of this receptor results in decreased inflammation in several different ways (Carniglia et al., 2013; Caruso et al., 2013). There are five MCRs and the brain only harbors two of them, MC3R and MC4R. Of these two MCRs, the role of MC4R in energy homeostasis has historically received the most scientific attention. Actually, MC4R deficiency is the most common monogenic form of obesity (Farooqi et al., 2003) and several studies have shown MC4R mutations in obesity (Hinney et al., 1999; Farooqi et al., 2000; Vaisse et al., 2000). Additionally, a huge genome wide association study revealed over 50 new candidate genes associated with BMI, most of them are enriched in the CNS and many of them belong to the melanocortin system, specifically the MC4Rs (Locke et al., 2015). The MC4Rs are widely expressed throughout the human

THE INFUNDIBULAR PEPTIDERGIC NEURONS AND GLIA CELLS hypothalamus, with highest expression in the paraventricular nucleus (PVN), supraoptic nucleus (SON) and the nucleus basalis of Meynert (NBM) (Siljee et al., 2013). Researchers studied the role of MC4R in the PVN intensively and showed it to be important for feeding behavior (Cowley et al., 1999; Atasoy et al., 2012). MC4R knockout mice have an obese phenotype characterized with hyperphagia and increased adiposity (Butler et al., 2001), which underscores the importance of MC4R in healthy energy homeostasis. Surprisingly, since their discovery, nobody investigated the role of MC4Rs in the SON and NBM and their function therefore remains unclear. Opposed to MC4R, mutations in MC3R gene did not show up in an association to obesity (Calton et al., 2009) and MC3R knockout studies did not show a clear phenotype (they display moderate adiposity without being hyperphagic) (Butler et al., 2000; Chen et al., 2000). This probably explains the historic lack of attention for the role of MC3R in energy homeostasis. However, in more recent years, research into the role of MC3R has increased and a systematic review showed that loss of function mutations in the MC3R gene confers a three times increased risk of obesity in humans (Ehtesham et al., 2019). Another neuropeptide cleaved from the precursor POMC is b-endorphin, which is not considered to be melanocortin because it uses the m-opioid receptor. It does, however, affect the central control of food intake (Appleyard et al., 2003). The acute effect of b-endorphin seems to be the stimulation of food intake by antagonizing the effect of a-MSH (Dutia et al., 2012). However, under chronic conditions this increase in food intake was not sustained (Dutia et al., 2012). Transgenic mice lacking b-endorphin, but with normal melanocortin signaling, became hyperphagic and obese, indicating b-endorphin has an inhibitory effect on food intake (Appleyard et al., 2003). It has also been suggested that b-endorphin plays a motivational role in feeding behavior (Veening and Barendregt, 2015).

Agouti-related protein AgRP and its related protein agouti are the only known endogenous reverse agonists of the MCRs. By decreasing activity of MCR expressing cells, AgRP can decrease the inhibition of food intake, effectively increasing food intake. In animal studies, AgRP is a strong appetite inducer and results in hyperphagia and the development of obesity when transgenically overexpressed or when administered intracerebroventricular (ICV). In humans, however, little is known about mutations in the AgRP gene that correlate with human obesity or diabetes. One

319

study investigated a mutation which induced enhanced promotor activity of the AgRP gene and found that this mutation was increased in Siera Leonians with a “high BMI” (a BMI above the mean BMI), compared to individuals with a “normal BMI” (a BMI below the mean) (Mayfield et al., 2001). In anorectic patients, they found an increase of mutations in the AgRP gene possibly affecting receptor binding, indicating that a decreased affinity of AgRP protein for the MCR results in decreased food intake (Vink et al., 2001).

Neuropeptide Y Even though 90% of the NPY neurons coexpress AgRP, the neuropeptide NPY itself does not bind to MCRs. There are six NPY receptors known and three of them are highly expressed in the brain (Y1, Y2, and Y5). Animal studies revealed NPY as a potent stimulator of food intake; chronic infusion of NPY led to hyperphagia and obesity (Beck et al., 1992) and ICV injections of NPY stimulated food intake (Levine and Morley, 1984). NPY in the brain mainly produces by the IFN neurons (Chronwall et al., 1985), and it elicits consumptive behavior through several second-order neuronal nuclei, like the PVN, ventromedial hypothalamus (VMH), and LH (Stanley et al., 1985). NPY neurons express insulin and leptin receptors, and both hormones inhibit NPY expression (Jang et al., 2000; Loh et al., 2017), ghrelin has the opposite effect and stimulates NPY expression (Hashiguchi et al., 2017). In nondiabetic humans the expression of NPY shows a negative correlation with BMI (Alkemade et al., 2012) and obese individuals have significantly less NPY expression compared to lean controls (Goldstone et al., 2002). Conversely, T2DM subjects have significantly increased NPY expression compared to nondiabetic controls (Saderi et al., 2012). Together, these studies indicate that there is a negative feedback of adiposity on NPY expression, which is lost in T2DM. As leptin is the main signal of adiposity, it seems likely that NPY neurons in T2DM have become leptin resistant. Five different Y-receptors have been identified, of which Y1, Y2, and Y5 have been implicated in the central regulation of energy homeostasis. Both Y1 and Y5 receptors are mainly expressed on second-order neurons and have been shown to mediate NPY-induced hyperphagia (Nguyen et al., 2012). Conversely, Y2 receptors are mainly found on NPY neurons itself and activation by NPY or PYY results in decreased neurotransmitter release (Yulyaningsih et al., 2011). Thus stimulation of the Y2 receptor on NPY neurons leads to inhibition of food intake, which has fueled the development of Y2 agonists (Feletou et al., 2006).

320

M.J.T. KALSBEEK AND C.-X. YI

SECOND-ORDER NEURONS Upon activation by peripheral signals, the IFN neurons release their neuropeptides to several second-order neurons in different brain areas to exert their control on energy homeostasis. The main hypothalamic projection sites are the PVN, the VMH, and the lateral hypothalamus (LH).

play a role in the regulation of glucose homeostasis. Activation of a specific subset of VMH neurons by designer receptor exclusively activated by designer drugs (DREADD), increased insulin sensitivity, and depressed glucose production (Coutinho et al., 2017). On the contrary, optogenetic stimulation of a (different) subset of VMH neurons caused diabetes-range hyperglycemia (Meek et al., 2016).

Paraventricular nucleus

Lateral hypothalamus

The PVN combines the metabolic input from the IFN, with input from many other hypothalamic regions, like information concerning circadian rhythms from the suprachiasmatic nucleus (Buijs et al., 1998). The rodent PVN consists of two major sections, the parvocellular and magnocellular division (Swanson and Kuypers, 1980). The parvocellular neurons project to the autonomic nervous system, thereby linking the PVN to all the major peripheral organs involved in energy homeostasis (Hill, 2012). The magnocellular neurons project to the pituitary, where they can release oxytocin and vasopressin into the systemic circulation. Vasopressin in the circulation increases water reabsorption in the kidneys and constrict arterioles to increase blood pressure. In the brain, activation of vasopressin neurons can inhibit food intake (Yoshimura et al., 2017). Oxytocin, also known as the love-hormone, has many functions, including a role in social bonding, childbirth, milk ejection, and sexual reproduction. In the brain, oxytocin signaling has a role in feeding behavior, and intranasal application of oxytocin was shown to decrease food intake in humans (Ott et al., 2013; Lawson et al., 2015). In Prader–Willi patients, there is a deficiency of oxytocin in the PVN (Swaab et al., 1995). In general, activation of the PVN neurons decreases food intake, which was first shown by lesions in the PVN, which resulted in hyperphagic and obese rats (Leibowitz et al., 1981). The neurons in the PVN express MCRs (Siljee et al., 2013) and Y-receptors (Yoshimura et al., 2017). Both neuropeptide receptors in the PVN have been shown to be involved in the control of energy homeostasis (Reichenbach et al., 2012; Shi et al., 2013).

The LH is generally known as the hunger center, and two of its main functions are the stimulation of feeding behavior and arousal. Electrical stimulation of the LH results in ravenous eating behavior, and animals are extremely motivated to work for a food reward (Stuber and Wise, 2016). The neurons of the LH are mainly orexin expressing neurons and they respond to both melanocortins and NPY (Campbell et al., 2003; Morgan et al., 2015). Orexin neurons stimulate wakefulness, and a loss of orexin neurons causes narcolepsy (Shan et al., 2015), which in turn is associated with increased risk on the development of T2DM (Mohammadi et al., 2019). Together, these properties of orexin promote alertness in a fasted state, which is crucial for food-seeking behavior (Tsujino and Sakurai, 2013).

Ventromedial hypothalamus The role of the VMH in feeding behavior was the topic of discussion in the early 1970s (Gold, 1973), but we now know that the VMH can be classified as the satiety center (King, 2006). The VMH neurons express NPY receptors and NPY inhibits the activity of VMH neurons, which increases food intake (Chee et al., 2010). The neurons in the VMH are also leptin responsive, and leptin infusion in the fourth ventricle results in VMH-mediated weight loss (Seamon et al., 2019). The VMH neurons

CONCLUDING REMARKS The regulation of energy homeostasis is a complex interplay between numerous peripheral and central biological systems. Several endogenous and exogenous factors can deregulate energy homeostasis, which, in combination with a western diet, usually results in a positive energy balance. Long-term positive energy balance eventually leads to obesity, and prolonged obesity often results in T2DM. This review focused mainly on the control of energy homeostasis itself, as flawed control is the key underlying factor in the development of obesity and T2DM. A large part of the research listed in this review was obtained through animal studies, which is not always directly translatable to the human situation. The nomenclature is a little bit different; the animal equivalent of the IFN is called the arcuate nucleus. Furthermore, some successful antiobesity therapies in animal models proofed to have no effect in humans. Additionally, animal models do not completely mimic the human situation of T2DM development, unless a genetic or pharmacologic intervention is performed. Rodents only on an obesogenic diet do become obese and hyperglycemic but do not develop T2DM with pancreatic b-cell failure like humans do. On the other hand, human in vivo data lack satisfactory spatiotemporal resolution to study specific cell types or molecules. Studies using human postmortem brain tissue do provide the possibility to study cell

THE INFUNDIBULAR PEPTIDERGIC NEURONS AND GLIA CELLS types and molecules but only allow for association studies. Collectively, this review gives an overview of the latest human and animal data concerning the IFN control of energy homeostasis, which possess several targets in the combat against obesity and T2DM.

ABBREVIATIONS ACTH, adrenocorticotropic hormone; AgRP, agoutirelated protein; BBB, blood–brain barrier; BMI, body mass index; DREAD, designer receptor exclusively activated by designer drugs; FFAR, free fatty acid receptor; GI, gastrointestinal; GLP-1, glucagon-like peptide 1; GLUT2, glucose transporter 2; ICV, intracerebroventricular; IFN, infundibular nucleus; LH, lateral hypothalamus; MCRs, melanocortin receptors; MSH, melanocyte-stimulating hormone; NBM, nucleus basalis of Meynert; NPY, neuropeptide Y; POMC, proopiomelanocortin; PVN, paraventricular nucleus; PYY, peptide YY; SON, supraoptic nucleus; T1DM, type 1 diabetes mellitus; T2DM, type 2 diabetes mellitus; VMH, ventromedial hypothalamus.

REFERENCES Alkemade A, Yi C-X, Pei L et al. (2012). AgRP and NPY expression in the human hypothalamic infundibular nucleus correlate with body mass index, whereas changes in aMSH are related to type 2 diabetes. J Clin Endocrinol Metab 97: E925–E933. Allard C, Carneiro L, Grall S et al. (2014). Hypothalamic astroglial connexins are required for brain glucose sensinginduced insulin secretion. J Cereb Blood Flow Metab 34: 339–346. Appleyard SM, Hayward M, Young JI et al. (2003). A role for the endogenous opioid b-endorphin in energy homeostasis. Endocrinology 144: 1753–1760. Atasoy D, Betley JN, Su HH et al. (2012). Deconstruction of a neural circuit for hunger. Nature 488: 172–177. Bady I, Marty N, Dallaporta M et al. (2006). Evidence from glut2-null mice that glucose is a critical physiological regulator of feeding. Diabetes 55: 988–995. Baggio LL, Drucker DJ (2014). Glucagon-like peptide-1 receptors in the brain: controlling food intake and body weight. J Clin Invest 124: 4223–4226. Balland E, Dam J, Langlet F et al. (2014). Hypothalamic tanycytes are an ERK-gated conduit for leptin into the brain. Cell Metab 19: 293–301. Balthasar N, Coppari R, McMinn J et al. (2004). Leptin receptor signaling in POMC neurons is required for normal body weight homeostasis. Neuron 42: 983–991. Banks WA (2001). Leptin transport across the blood-brain barrier: implications for the cause and treatment of obesity. Curr Pharm Des 7: 125–133. Batterham RL, Cohen MA, Ellis SM et al. (2003). Inhibition of food intake in obese subjects by peptide YY3-36. N Engl J Med 349: 941–948.

321

Batterham RL, Cowley MA, Small CJ et al. (2002). Gut hormone PYY(3-36) physiologically inhibits food intake. Nature 418: 650–654. Baufeld C, Osterloh A, Prokop S et al. (2016). High-fat dietinduced brain region-specific phenotypic spectrum of CNS resident microglia. Acta Neuropathol 132: 361–375. Beck B, Stricker-Krongrad A, Nicolas JP et al. (1992). Chronic and continuous intracerebroventricular infusion of neuropeptide Y in Long-Evans rats mimics the feeding behaviour of obese Zucker rats. Int J Obes Relat Metab Disord 16: 295–302. Benoit SC, Air EL, Coolen LM et al. (2002). The catabolic action of insulin in the brain is mediated by melanocortins. J Neurosci 22: 9048–9052. Bertozzi S (1950). Abnormal hunger due to brain lesion: pituitary tumor and paroxystic bulimia. Riv Patol Nerv Ment 71: 343–34366. Blouet C, Schwartz GJ (2010). Hypothalamic nutrient sensing in the control of energy homeostasis. Behav Brain Res 209: 1–12. Brabazon F, Bermudez S, Shaughness M et al. (2018). The effects of insulin on the inflammatory activity of BV2 microglia. PLoS One 13: e0201878. Broadwell RD, Balin BJ, Salcman M et al. (1983). Brain-blood barrier? Yes and no. Proc Natl Acad Sci U S A 80: 7352–7356. Brooks CM, Lockwood RA, Wiggins ML (1946). A study of the effect of hypothalamic lesions on the eating habits of the albino rat. Am J Physiol 147: 735–741. Brown KS, Gentry RM, Rowland NE (1998). Central injection in rats of alpha-melanocyte-stimulating hormone analog: effects on food intake and brain Fos. Regul Pept 78: 89–94. Bruning JC, Gautam D, Burks DJ et al. (2000). Role of brain insulin receptor in control of body weight and reproduction. Science 289: 2122–2125. Buijs RM, Hermes MH, Kalsbeek A (1998). The suprachiasmatic nucleus-paraventricular nucleus interactions: a bridge to the neuroendocrine and autonomic nervous system. Prog Brain Res 119: 365–382. Butler AA, Kesterson RA, Khong K et al. (2000). A unique metabolic syndrome causes obesity in the melanocortin-3 receptor-deficient mouse. Endocrinology 141: 3518–3521. Butler AA, Marks DL, Fan W et al. (2001). Melanocortin-4 receptor is required for acute homeostatic responses to increased dietary fat. Nat Neurosci 4: 605–611. Calton MA, Ersoy BA, Zhang S et al. (2009). Association of functionally significant melanocortin-4 but not melanocortin-3 receptor mutations with severe adult obesity in a large North American case-control study. Hum Mol Genet 18: 1140–1147. Camandola S (2018). Astrocytes, emerging stars of energy homeostasis. Cell Stress 2: 246–252. Campbell RE, Smith MS, Allen SE et al. (2003). Orexin neurons express a functional pancreatic polypeptide Y4 receptor. J Neurosci 23: 1487–1497. Carniglia L, Durand D, Caruso C et al. (2013). Effect of NDP-a-MSH on PPAR-g and -b expression and antiinflammatory cytokine release in rat astrocytes and microglia. PLoS One 8: e57313.

322

M.J.T. KALSBEEK AND C.-X. YI

Caron A, Dungan Lemko HM, Castorena CM et al. (2018). POMC neurons expressing leptin receptors coordinate metabolic responses to fasting via suppression of leptin levels. eLife 7: e33710. Caruso C, Carniglia L, Durand D et al. (2013). Astrocytes: new targets of melanocortin 4 receptor actions. J Mol Endocrinol 51: R33–R50. Chee MJS, Myers Jr MG, Price CJ et al. (2010). Neuropeptide Y suppresses anorexigenic output from the ventromedial nucleus of the hypothalamus. J Neurosci 30: 3380–3390. Chen AS, Marsh DJ, Trumbauer ME et al. (2000). Inactivation of the mouse melanocortin-3 receptor results in increased fat mass and reduced lean body mass. Nat Genet 26: 97–102. Cheng H, Isoda F, Belsham DD et al. (2008). Inhibition of agouti-related peptide expression by glucose in a clonal hypothalamic neuronal cell line is mediated by glycolysis, not oxidative phosphorylation. Endocrinology 149: 703–710. Chronwall BM, DiMaggio DA, Massari VJ et al. (1985). The anatomy of neuropeptide-Y-containing neurons in rat brain. Neuroscience 15: 1159–1181. Cota D, Proulx K, Smith KAB et al. (2006). Hypothalamic mTOR signaling regulates food intake. Science 312: 927–930. Cottrell GT, Ferguson AV (2004). Sensory circumventricular organs: central roles in integrated autonomic regulation. Regul Pept 117: 11–23. Coutinho EA, Okamoto S, Ishikawa AW et al. (2017). Activation of SF1 neurons in the ventromedial hypothalamus by DREADD technology increases insulin sensitivity in peripheral tissues. Diabetes 66: 2372–2386. Cowley MA, Pronchuk N, Fan W et al. (1999). Integration of NPY, AGRP, and melanocortin signals in the hypothalamic paraventricular nucleus: evidence of a cellular basis for the adipostat. Neuron 24: 155–163. Cowley MA, Smith RG, Diano S et al. (2003). The distribution and mechanism of action of ghrelin in the CNS demonstrates a novel hypothalamic circuit regulating energy homeostasis. Neuron 37: 649–661. De Backer I, Hussain SS, Bloom SR et al. (2016). Insights into the role of neuronal glucokinase. Am J Physiol Endocrinol Metab 311: E42–E55. Delgado R, Carlin A, Airaghi L et al. (1998). Melanocortin peptides inhibit production of proinflammatory cytokines and nitric oxide by activated microglia. J Leukoc Biol 63: 740–745. Dragano N, Solon C, Ramalho AF et al. (2017). Polyunsaturated fatty acid receptors, GPR40 and GPR120, are expressed in the hypothalamus and control energy homeostasis and inflammation. J Neuroinflammation 14: 91. Dutia R, Meece K, Dighe S et al. (2012). b-Endorphin antagonizes the effects of a-MSH on food intake and body weight. Endocrinology 153: 4246–4255. Ehtesham S, Qasim A, Meyre D (2019). Loss-of-function mutations in the melanocortin-3 receptor gene confer risk for human obesity: a systematic review and meta-analysis. Obes Rev 20: 1085–1092.

Falomir-Lockhart LJ, Cavazzutti GF, Gimenez E et al. (2019). Fatty acid signaling mechanisms in neural cells: fatty acid receptors. Front Cell Neurosci 13: 162. Farooqi IS, Keogh JM, Yeo GSH et al. (2003). Clinical spectrum of obesity and mutations in the melanocortin 4 receptor gene. N Engl J Med 348: 1085–1095. Farooqi IS, Yeo GS, Keogh JM et al. (2000). Dominant and recessive inheritance of morbid obesity associated with melanocortin 4 receptor deficiency. J Clin Invest 106: 271–279. Feletou M, Galizzi J-P, Levens NR (2006). NPY receptors as drug targets for the central regulation of body weight. CNS Neurol Disord Drug Targets 5: 263–274. Frago LM, Chowen JA (2017). Involvement of astrocytes in mediating the central effects of ghrelin. Int J Mol Sci 18: 536. Freire-Regatillo A, Argente-Arizo´n P, Argente J et al. (2017). Non-neuronal cells in the hypothalamic adaptation to metabolic signals. Front Endocrinol 8: 51. Frost G, Sleeth ML, Sahuri-Arisoylu M et al. (2014). The short-chain fatty acid acetate reduces appetite via a central homeostatic mechanism. Nat Commun 5: 3611. Fuente-Martin E, Garcia-Caceres C, Argente-Arizon P et al. (2016). Ghrelin regulates glucose and glutamate transporters in hypothalamic astrocytes. Sci Rep 6: 23673. Funcke J-B, von Schnurbein J, Lennerz B et al. (2014). Monogenic forms of childhood obesity due to mutations in the leptin gene. Mol Cell Pediatr 1: 3. Gao Y, Bielohuby M, Fleming T et al. (2017a). Dietary sugars, not lipids, drive hypothalamic inflammation. Mol Metab 6: 897–908. Gao Y, Ottaway N, Schriever SC et al. (2014). Hormones and diet, but not body weight, control hypothalamic microglial activity. Glia 62: 17–25. Gao Y, Vidal-Itriago A, Kalsbeek MJ et al. (2017b). Lipoprotein lipase maintains microglial innate immunity in obesity. Cell Rep 20: 3034–3042. Gao Y, Vidal-Itriago A, Milanova I et al. (2018). Deficiency of leptin receptor in myeloid cells disrupts hypothalamic metabolic circuits and causes body weight increase. Mol Metab 7: 155–160. Garcı´a-Ca´ceres C, Balland E, Prevot V et al. (2019). Role of astrocytes, microglia, and tanycytes in brain control of systemic metabolism. Nat Neurosci 22: 7–14. Garcı´a-Ca´ceres C, Quarta C, Varela L et al. (2016). Astrocytic insulin signaling couples brain glucose uptake with nutrient availability. Cell 166: 867–880. Ghamari-Langroudi M, Colmers WF, Cone RD (2005). PYY3–36 inhibits the action potential firing activity of POMC neurons of arcuate nucleus through postsynaptic Y2 receptors. Cell Metab 2: 191–199. Goberdhan DCI, Wilson C, Harris AL (2016). Amino acid sensing by mTORC1: intracellular transporters mark the spot. Cell Metab 23: 580–589. Goldstone AP, Unmehopa UA, Bloom SR et al. (2002). Hypothalamic NPY and agouti-related protein are increased in human illness but not in Prader-Willi syndrome and other obese subjects. J Clin Endocrinol Metab 87 (2): 927–937. https://doi.org/10.1210/jcem.87.2.8230. PMID: 11836343.

THE INFUNDIBULAR PEPTIDERGIC NEURONS AND GLIA CELLS Gold RM (1973). Hypothalamic obesity: the myth of the ventromedial nucleus. Science 182: 488–490. Hallschmid M, Benedict C, Schultes B et al. (2004). Intranasal insulin reduces body fat in men but not in women. Diabetes 53: 3024–3029. Harrison L, Schriever SC, Feuchtinger A et al. (2019). Fluorescent blood–brain barrier tracing shows intact leptin transport in obese mice. Int J Obes (Lond) 43: 1305–1318. Hashiguchi H, Sheng Z, Routh V et al. (2017). Direct versus indirect actions of ghrelin on hypothalamic NPY neurons. PLoS One 12: e0184261. Havel PJ, Hahn TM, Sindelar DK et al. (2000). Effects of streptozotocin-induced diabetes and insulin treatment on the hypothalamic melanocortin system and muscle uncoupling protein 3 expression in rats. Diabetes 49: 244–252. Havrankova J, Roth J, Brownstein M (1978). Insulin receptors are widely distributed in the central nervous system of the rat. Nature 272: 827–829. Hill JW (2012). PVN pathways controlling energy homeostasis. Indian J Endocrinol Metab 16: S627–S636. Hinney A, Schmidt A, Nottebom K et al. (1999). Several mutations in the melanocortin-4 receptor gene including a nonsense and a frameshift mutation associated with dominantly inherited obesity in humans. J Clin Endocrinol Metab 84: 1483–1486. Holst JJ (2007). The physiology of glucagon-like peptide 1. Physiol Rev 87: 1409–1439. Hope DCD, Tan TMM, Bloom SR (2018). No guts, no loss: toward the ideal treatment for obesity in the twenty-first century. Front Endocrinol 9: 442. Hsieh C-F, Liu C-K, Lee C-T et al. (2019). Acute glucose fluctuation impacts microglial activity, leading to inflammatory activation or self-degradation. Sci Rep 9: 840. Jang M, Mistry A, Swick AG et al. (2000). Leptin rapidly inhibits hypothalamic neuropeptide Y secretion and stimulates corticotropin-releasing hormone secretion in adrenalectomized mice. J Nutr 130: 2813–2820. Jo Y-H, Su Y, Gutierrez-Juarez R et al. (2009). Oleic acid directly regulates POMC neuron excitability in the hypothalamus. J Neurophysiol 101: 2305–2316. Jones B, Bloom SR, Buenaventura T et al. (2018). Control of insulin secretion by GLP-1. Peptides 100: 75–84. Karra E, Chandarana K, Batterham RL (2009). The role of peptide YY in appetite regulation and obesity. J Physiol 587: 19–25. Kim JG, Suyama S, Koch M et al. (2014). Leptin signaling in astrocytes regulates hypothalamic neuronal circuits and feeding. Nat Neurosci 17: 908–910. King BM (2006). The rise, fall, and resurrection of the ventromedial hypothalamus in the regulation of feeding behavior and body weight. Physiol Behav 87: 221–244. Kirchner H, Heppner KM, Tschop MH (2012). The role of ghrelin in the control of energy balance. Handb Exp Pharmacol 209: 161–184. Konner AC, Janoschek R, Plum L et al. (2007). Insulin action in AgRP-expressing neurons is required for suppression of hepatic glucose production. Cell Metab 5: 438–449. Krisch B, Leonhardt H (1978). The functional and structural border of the neurohemal region of the median eminence. Cell Tissue Res 192: 327–339.

323

Krude H, Biebermann H, Luck W et al. (1998). Severe earlyonset obesity, adrenal insufficiency and red hair pigmentation caused by POMC mutations in humans. Nat Genet 19: 155–157. Laeger T, Reed SD, Henagan TM et al. (2014). Leucine acts in the brain to suppress food intake but does not function as a physiological signal of low dietary protein. Am J Physiol Regul Integr Comp Physiol 307: R310–R320. Lam BYH, Cimino I, Polex-Wolf J et al. (2017). Heterogeneity of hypothalamic pro-opiomelanocortin-expressing neurons revealed by single-cell RNA sequencing. Mol Metab 6: 383–392. Langlet F, Levin BE, Luquet S et al. (2013). Tanycytic VEGFA boosts blood-hypothalamus barrier plasticity and access of metabolic signals to the arcuate nucleus in response to fasting. Cell Metab 17: 607–617. Lawson EA, Marengi DA, DeSanti RL et al. (2015). Oxytocin reduces caloric intake in men. Obesity (Silver Spring) 23: 950–956. Lean ME, Leslie WS, Barnes AC et al. (2018). Primary careled weight management for remission of type 2 diabetes (DiRECT): an open-label, cluster-randomised trial. Lancet (London, England) 391: 541–551. Leibowitz SF, Hammer NJ, Chang K (1981). Hypothalamic paraventricular nucleus lesions produce overeating and obesity in the rat. Physiol Behav 27: 1031–1040. Levine AS, Morley JE (1984). Neuropeptide Y: a potent inducer of consummatory behavior in rats. Peptides 5: 1025–1029. Locke AE, Kahali B, Berndt SI et al. (2015). Genetic studies of body mass index yield new insights for obesity biology. Nature 518: 197–206. Loh K, Zhang L, Brandon A et al. (2017). Insulin controls food intake and energy balance via NPY neurons. Mol Metab 6: 574–584. Maffei M, Halaas J, Ravussin E et al. (1995). Leptin levels in human and rodent: measurement of plasma leptin and ob RNA in obese and weight-reduced subjects. Nat Med 1: 1155–1161. Marty N, Dallaporta M, Foretz M et al. (2005). Regulation of glucagon secretion by glucose transporter type 2 (glut2) and astrocyte-dependent glucose sensors. J Clin Invest 115: 3545–3553. Maurin A-C, Benani A, Lorsignol A et al. (2014). Hypothalamic eIF2alpha signaling regulates food intake. Cell Rep 6: 438–444. Mayfield DK, Brown AM, Page GP et al. (2001). A role for the agouti-related protein promoter in obesity and type 2 diabetes. Biochem Biophys Res Commun 287: 568–573. Meek TH, Nelson JT, Matsen ME et al. (2016). Functional identification of a neurocircuit regulating blood glucose. Proc Natl Acad Sci U S A 113: E2073–E2082. Mendes NF, Kim Y-B, Velloso LA et al. (2018). Hypothalamic microglial activation in obesity: a mini-review. Front Neurosci 12: 846. Mohammadi S, Dolatshahi M, Zare-Shahabadi A et al. (2019). Untangling narcolepsy and diabetes: pathomechanisms with eyes on therapeutic options. Brain Res 1718: 212–222.

324

M.J.T. KALSBEEK AND C.-X. YI

Morgan DA, McDaniel LN, Yin T et al. (2015). Regulation of glucose tolerance and sympathetic activity by MC4R signaling in the lateral hypothalamus. Diabetes 64: 1976–1987. Murphy EJ (2015). Blood-brain barrier and brain fatty acid uptake: role of arachidonic acid and PGE2. J Neurochem 135: 845–848. Myers MGJ, Leibel RL, Seeley RJ et al. (2010). Obesity and leptin resistance: distinguishing cause from effect. Trends Endocrinol Metab 21: 643–651. Nguyen AD, Mitchell NF, Lin S et al. (2012). Y1 and Y5 receptors are both required for the regulation of food intake and energy homeostasis in mice. PLoS One 7: e40191. Ott V, Finlayson G, Lehnert H et al. (2013). Oxytocin reduces reward-driven food intake in humans. Diabetes 62: 3418–3425. Parton LE, Ye CP, Coppari R et al. (2007). Glucose sensing by POMC neurons regulates glucose homeostasis and is impaired in obesity. Nature 449: 228–232. Perry B, Wang Y (2012). Appetite regulation and weight control: the role of gut hormones. Nutr Diabetes 2: e26. Petersen MC, Shulman GI (2018). Mechanisms of insulin action and insulin resistance. Physiol Rev 98: 2133–2223. Reichenbach A, Stark R, Andrews Z (2012). Hypothalamic control of food intake and energy metabolism. In: The Human Hypothalamus: Anatomy, Functions and Disorders. Reis WL, Yi C-X, Gao Y et al. (2015). Brain innate immunity regulates hypothalamic arcuate neuronal activity and feeding behavior. Endocrinology 156: 1303–1315. Rodriguez EM, Blazquez JL, Guerra M (2010). The design of barriers in the hypothalamus allows the median eminence and the arcuate nucleus to enjoy private milieus: the former opens to the portal blood and the latter to the cerebrospinal fluid. Peptides 31: 757–776. Rosario W, Singh I, Wautlet A et al. (2016). The brain-topancreatic islet neuronal map reveals differential glucose regulation from distinct hypothalamic regions. Diabetes 65: 2711–2723. Saderi N, Salgado-Delgado R, Avendan˜o-Pradel R et al. (2012). NPY and VGF immunoreactivity increased in the arcuate nucleus, but decreased in the nucleus of the Tractus Solitarius, of type-II diabetic patients. PLoS One 7: e40070. Schwartz MW, Sipols AJ, Marks JL et al. (1992). Inhibition of hypothalamic neuropeptide Y gene expression by insulin. Endocrinology 130: 3608–3616. Seamon M, Ahn W, Li A-J et al. (2019). Leptin receptorexpressing neurons in ventromedial nucleus of the hypothalamus contribute to weight loss caused by fourth ventricle leptin infusions. Am J Physiol Metab 317: E586–E596. Shan L, Dauvilliers Y, Siegel J (2015). Interactions of the histamine and hypocretin systems in CNS disorders. Nat Rev Neurol 11: 401–413. Shaver SW, Pang JJ, Wainman DS et al. (1992). Morphology and function of capillary networks in subregions of the rat tuber cinereum. Cell Tissue Res 267: 437–448.

Shi Y-C, Lau J, Lin Z et al. (2013). Arcuate NPY controls sympathetic output and BAT function via a relay of tyrosine hydroxylase neurons in the PVN. Cell Metab 17: 236–248. Siljee JE, Unmehopa UA, Kalsbeek A et al. (2013). Melanocortin 4 receptor distribution in the human hypothalamus. Eur J Endocrinol 168: 361–369. Stanley BG, Chin AS, Leibowitz SF (1985). Feeding and drinking elicited by central injection of neuropeptide Y: evidence for a hypothalamic site(s) of action. Brain Res Bull 14: 521–524. Stolarczyk E, Guissard C, Michau A et al. (2010). Detection of extracellular glucose by GLUT2 contributes to hypothalamic control of food intake. Am J Physiol Endocrinol Metab 298: E1078–E1087. Stuber GD, Wise RA (2016). Lateral hypothalamic circuits for feeding and reward. Nat Neurosci 19: 198–205. Swaab D (2003). The human hypothalamus: basic and clinical aspects. Part I. Nuclei of the human hypothalamus and adjacent structures. Handb Clin Neurol 79: 502c. Elsevier. Swaab DF, Purba JS, Hofman MA (1995). Alterations in the hypothalamic paraventricular nucleus and its oxytocin neurons (putative satiety cells) in Prader-Willi syndrome: a study of five cases. J Clin Endocrinol Metab 80: 573–579. Swanson LW, Kuypers HG (1980). The paraventricular nucleus of the hypothalamus: cytoarchitectonic subdivisions and organization of projections to the pituitary, dorsal vagal complex, and spinal cord as demonstrated by retrograde fluorescence double-labeling methods. J Comp Neurol 194: 555–570. Takahashi KA, Cone RD (2005). Fasting induces a large, leptin-dependent increase in the intrinsic action potential frequency of orexigenic arcuate nucleus neuropeptide Y/Agouti-related protein neurons. Endocrinology 146: 1043–1047. Thaler JP, Yi C-X, Schur EA et al. (2012). Obesity is associated with hypothalamic injury in rodents and humans. J Clin Invest 122: 153–162. Theil M-M, Miyake S, Mizuno M et al. (2009). Suppression of experimental autoimmune encephalomyelitis by ghrelin. J Immunol 183: 2859–2866. Timper K, Br€ uning JC (2017). Hypothalamic circuits regulating appetite and energy homeostasis: pathways to obesity. Dis Model Mech 10: 679–689. Tschop M, Smiley DL, Heiman ML (2000). Ghrelin induces adiposity in rodents. Nature 407: 908–913. Tsujino N, Sakurai T (2013). Role of orexin in modulating arousal, feeding, and motivation. Front Behav Neurosci 7: 28. Vaisse C, Clement K, Durand E et al. (2000). Melanocortin-4 receptor mutations are a frequent and heterogeneous cause of morbid obesity. J Clin Invest 106: 253–262. Valdearcos M, Xu AW, Koliwad SK (2015). Hypothalamic inflammation in the control of metabolic function. Annu Rev Physiol 77: 131–160. Veening JG, Barendregt HP (2015). The effects of betaendorphin: state change modification. Fluids Barriers CNS 12: 3.

THE INFUNDIBULAR PEPTIDERGIC NEURONS AND GLIA CELLS Vilsboll T, Christensen M, Junker AE et al. (2012). Effects of glucagon-like peptide-1 receptor agonists on weight loss: systematic review and meta-analyses of randomised controlled trials. BMJ 344: d7771. Vink T, Hinney A, van Elburg AA et al. (2001). Association between an agouti-related protein gene polymorphism and anorexia nervosa. Mol Psychiatry 6: 325–328. Wang J, Leibowitz KL (1997). Central insulin inhibits hypothalamic galanin and neuropeptide Y gene expression and peptide release in intact rats. Brain Res 777: 231–236. Willesen MG, Kristensen P, Romer J (1999). Co-localization of growth hormone secretagogue receptor and NPY mRNA in the arcuate nucleus of the rat. Neuroendocrinology 70: 306–316. Woods SC, Lotter EC, McKay LD et al. (1979). Chronic intracerebroventricular infusion of insulin reduces food intake and body weight of baboons. Nature 282: 503–505.

325

Yi C-X, Walter M, Gao Y et al. (2017). TNFa drives mitochondrial stress in POMC neurons in obesity. Nat Commun 8: 15143. Yoshimura M, Nishimura K, Nishimura H et al. (2017). Activation of endogenous arginine vasopressin neurons inhibit food intake: by using a novel transgenic rat line with DREADDs system. Sci Rep 7: 15728. Yulyaningsih E, Zhang L, Herzog H et al. (2011). NPY receptors as potential targets for anti-obesity drug development. Br J Pharmacol 163: 1170–1202. Zampieri TT, Pedroso JAB, Furigo IC et al. (2013). Oral leucine supplementation is sensed by the brain but neither reduces food intake nor induces an anorectic pattern of gene expression in the hypothalamus. PLoS One 8: e84094. Zhang Y, Proenca R, Maffei M et al. (1994). Positional cloning of the mouse obese gene and its human homologue. Nature 372: 425–432.

This page intentionally left blank

Handbook of Clinical Neurology, Vol. 180 (3rd series) The Human Hypothalamus: Middle and Posterior Region D.F. Swaab, F. Kreier, P.J. Lucassen, A. Salehi, and R.M. Buijs, Editors https://doi.org/10.1016/B978-0-12-820107-7.00020-3 Copyright © 2021 Elsevier B.V. All rights reserved

Chapter 20

Hypothalamus and weight loss in amyotrophic lateral sclerosis REBEKAH M. AHMED1,2, FREDERIK STEYN3,4, AND LUC DUPUIS5* 1

Memory and Cognition Clinic, Department of Clinical Neurosciences, Royal Prince Alfred Hospital, Sydney, NSW, Australia 2

Central Sydney Medical School and Brain & Mind Centre, Faculty of Medicine and Health, University of Sydney, Sydney, NSW, Australia 3

School of Biomedical Sciences, University of Queensland, Brisbane, QLD, Australia

4

Department of Neurology, Royal Brisbane & Women’s Hospital, Brisbane, QLD, Australia

5

Universite de Strasbourg, Inserm, UMR-S 1118, Centre de Recherches en Biomedecine, Strasbourg, France

Abstract Amyotrophic lateral sclerosis (ALS) is a devastating progressive neurodegenerative disorder. While initially pathophysiology was thought to be restricted to motor deficits, it is increasingly recognized that patients develop prominent changes in weight and eating behavior that result from and mediate the underlying neurodegenerative process. These changes include alterations in metabolism, lipid levels, and insulin resistance. Emerging research suggests that these alterations may be mediated through changes in the hypothalamic function, with atrophy of the hypothalamus shown in both ALS patients and also presymptomatic genetic at-risk patients. This chapter reviews the evidence for hypothalamic involvement in ALS, including melanocortin pathways and potential treatment targets.

OVERVIEW OF ALS Amyotrophic lateral sclerosis (ALS) is a devastating progressive neurodegenerative disorder, with involvement of the corticospinal tract, brainstem, and anterior horn cells of the spinal cord. The peak onset of the condition is between 50 and 75 years with involvement of both the upper and lower motor neurons; 65% of patients present with limb involvement, while 30% have involvement of bulbar muscles and 5% involvement of respiratory muscles (Hardiman et al., 2011; Kiernan et al., 2011). Onset is typically asymmetric, with the dominant limb more likely to be affected first and the pattern of spreading affected by dominance of the first affected limb (Turner et al., 2011; Devine et al., 2014). Patients diagnosed with ALS typically exhibit limb or bulbar symptoms at initial presentation (Kiernan et al., 2011; Turner et al., 2013b; Vucic et al., 2014). There are varying reports on the incidence of cognitive changes ∗

in ALS (behavioral, cognitive, language) with estimates upwards of 5% (Strong, 2008; Montuschi et al., 2015), while up to 15% of patients may satisfy the criteria for a diagnosis of concomitant frontotemporal dementia (FTD) (Ringholz et al., 2005). Conversely, 10%–15% of FTD patients have ALS, with varying estimates, between 25% and 30%, of motor neuron dysfunction in FTD insufficient to reach criteria for ALS (Lomen-Hoerth et al., 2002; Burrell et al., 2011). FTD and ALS often share a common pathology, TDP-43 protein deposition, which is present in the majority of ALS patients and in up to 50% of cases of behavioral variant of FTD (bvFTD), the most common form of FTD (Mackenzie et al., 2010). Recent research has suggested that FTD and ALS may potentially result from a contiguous spread (Braak et al., 2013; Ludolph and Brettschneider, 2015; Tan et al., 2015) in a recognized centrifugal pattern, with four stages of spread in ALS, beginning in the motor neocortex, progressing to the spinal

Correspondence to: Dr Luc Dupuis, Universite de Strasbourg, Inserm, UMR-S1118, Faculte de medecine, 11 rue Humann, 67085 Strasbourg, France. Tel: +33-3-68-85-34-57, E-mail: [email protected]

328

R.M. AHMED ET AL.

cord and brainstem, with involvement of frontal–parietal regions, and finally the temporal lobes (Brettschneider et al., 2013). Such a pattern of spread may further explain the development of cognitive symptoms in ALS. This overlap has been further reinforced with the discovery of the C9ORF72 gene abnormality in individuals with familial FTD and ALS (Hodges, 2012). Currently there are at least 15 genes associated with familial ALS, with the five major genes being C9ORF72, SOD1, FUS, TARDBP, and TBK1 (Hardiman et al., 2011; Devenney et al., 2015). Longitudinal imaging studies in ALS have shown a decrease in the cortical thickness in motor-temporal and frontoparietal cortices over time (Filippi et al., 2015). Severe precentral gyrus volume loss is an early sign of ALS, but atrophy spreads to nonmotor regions over time. Structural and functional imaging studies have indicated that networks involved in the primary motor deficits of ALS functionally and anatomically involve regions associated with the motor cortices (Schmidt et al., 2014). On diffusion tensor imaging the corpus callosum has been found to be consistently affected in ALS, with the hypothesis that it is integral for the pathologic spread in the brain (Filippini et al., 2010). It is becoming increasingly recognized that the changes in ALS are not restricted to only motor and cognitive involvement (Ahmed et al., 2018b), with increasing research suggesting patients with the disease can develop changes in energy homeostasis; this may manifest as altered body mass index and unexplained weight loss, altered lipid metabolism, and impairments in insulin secretion and glucose control. These changes are likely the result of interactions between the neurodegenerative process, with involvement of key structures within the brain, including the hypothalamus, and peripheral mechanisms that would normally modulate energy metabolism in ALS, resulting in increased energy expenditure (hypermetabolism) associated with qualitative and/or quantitative alterations in eating behavior. The following sections review metabolic and appetite changes in ALS and discuss the role of the hypothalamus as a potential key structure that interacts with other central and peripheral mechanisms of energy balance.

METABOLIC CHANGES IN ALS Energy balance is maintained through the regulation of energy intake (via food intake, nutrient absorption, and shuttling and cellular uptake of nutrients) and energy expenditure (primarily as resting, but also activity and postprandial energy expenditure). Energy homeostasis is also intrinsically linked to glucose and lipid metabolism, with insulin being integral to cellular uptake of nutrients, and insulin resistance resulting in decreased

sensitivity of peripheral cells (e.g., muscle) to nutrient uptake, leading to decreased energy stores. ALS patients are generally lean and lose body mass through loss of muscle and fat mass as the disease progresses (Dupuis et al., 2011; Ioannides et al., 2016). Despite this, patients have been found to have increased lipid levels, glucose intolerance, and insulin resistance. Paradoxically, this is normally associated with obesity. Therefore, patients with ALS may present with a complex metabolic phenotype with varying degrees of disruption in body weight balance and metabolic and hormonal pathways.

Body mass index Patients with ALS typically have a normal or low BMI (Desport et al., 2000) and lose weight and body fat as the disease progresses (Desport et al., 1999; Gallo et al., 2013), which in turn negatively affects prognosis (Reich-Slotky et al., 2013). Such weight loss often precedes onset of motor symptoms by several years (Peter et al., 2017). Low BMI in ALS has been attributed to a number of causes, including loss of muscle mass (Dupuis et al., 2011), swallowing difficulties, decreased nutritional intake (Kuhnlein et al., 2008), a state of hypermetabolism (Dupuis et al., 2011), and reduced food intake due to loss of appetite (see the following text for a discussion of these different mechanisms) (Ngo et al., 2019). It has also been suggested that the effect of BMI on survival in ALS may form a U-shaped relationship, with both low BMI and BMI > 35 associated with increased mortality, perhaps secondary to an increased incidence of cardiovascular disease or complications of type 2 diabetes (Paganoni et al., 2011). Two large population-based studies found that risk of developing ALS was increased by lower premorbid body weight (Gallo et al., 2013; O’Reilly et al., 2013). Conversely, high prediagnostic BMI and weight gain have been associated with lower ALS risk several decades later (Nakken et al., 2019). Indirect epidemiologic evidence also suggests that body weight and adiposity are critical in ALS risk. Indeed, levels of leptin, an adipose-derived hormone positively correlated with body fat mass, are inversely associated with risk of ALS in a German case control register study (Nagel et al., 2017). Conversely, levels of adiponectin, an adipose-derived hormone whose levels are inversely correlated with adiposity, are inversely associated with risk of ALS in the same study. How hormone levels could be related to ALS risk requires further investigation, but they could relate to factors such as change in lipid levels and increased energy stores in the setting of a hypermetabolic state. BMI is, however, a poor predictor of body composition in ALS (Ioannides et al., 2017), and thus BMI may not serve as a reliable indicator of overall energy balance

HYPOTHALAMUS AND WEIGHT LOSS IN AMYOTROPHIC LATERAL SCLEROSIS

329

Fig. 20.1. Changes in fat distribution in ALS. Patient showing loss of subcutaneous fat in arm and classical wasting of first dorsal interossei in ALS (black arrows), with redistribution of fat to abdomen and visceral region.

in ALS. This may be due to changes in the distribution of body fat in ALS, where a disproportionate loss of subcutaneous fat and increased abdominal fat correlates with progression (Lindauer et al., 2013) (as illustrated in Fig. 20.1). In ALS patients, the amount of subcutaneous fat generally correlates with functional status and survival (Lindauer et al., 2013). Given variations in predictive outcomes from BMI, it is likely that changes in body weight may be more informative; loss of body weight throughout the disease course is consistently associated with shorter survival (Stambler et al., 1998; Marin et al., 2011; Clavelou et al., 2013; Roubeau et al., 2015; Fasano et al., 2017; Peter et al., 2017; Moglia et al., 2019). A progressive decline in body weight is usually observed in patients with ALS, and likely reflects a state of undernutrition (Kasarskis et al., 1996). This is consistently associated with worsening disease outcome (Kirk et al., 2019), and thus factors that contribute to weight loss are of critical clinical concern.

Lipids Whether changes in steady-state blood lipids contribute to ALS pathophysiology remains an ongoing source of debate. In a French cohort of 369 patients with ALS, two-thirds of patients had increased LDL cholesterol, decreased HDL concentration, or a combination of the two (Dupuis et al., 2008). In the same cohort, 38% of patients had an elevated LDL-to-HDL ratio and increased concentrations of apolipoprotein E (Dupuis et al., 2008). In a German cohort, elevated triglyceride and total cholesterol levels were associated with a positive effect on survival (Dorst et al., 2011). Other studies have suggested that increased cholesterol levels may be associated with slower functional decline and increased survival (Ikeda et al., 2012), but these elevations in cholesterol levels may be gender specific, e.g., present only in females in a Japanese cohort (Ikeda et al., 2012). Increased blood

lipids have not been consistently observed in all studies. In some studies, although patients were not dyslipidemic, having a higher HDL/LDL ratio was correlated with improved survival (Sutedja et al., 2011), while in Italian (Chio et al., 2009) and Korean populations, lower blood lipids were associated with worsened outcomes, and in particular in respiratory function. Thus, while steadystate blood lipids might vary according to population ethnicity, BMI, dietary habits, or other environmental factors, a positive correlation between blood lipids and survival or function occurs in most studies. Indeed, adding support to the hypothesis that circulating lipids may play a modulating role in ALS is the finding in a number of epidemiologic studies that treatment with statins is associated with increased incidence of ALS (Colman et al., 2008; Golomb et al., 2009). Given the variation of nutritional guidelines across countries, it remains difficult to ascertain an effect of diet and BMI on lipid levels. Recently, it has been found across the ALS- FTD spectrum that higher cholesterol levels correlate to improved survival and are mediated by fat intake (Ahmed et al., 2018a). Further research is required to examine longitudinal changes in lipid levels with disease progression as a function of nutritional intervention.

Insulin and insulin resistance There is controversy as to whether there is an increased incidence of diabetes and insulin resistance in ALS or whether diabetes may be protective for the onset of ALS and affect disease progression. Several studies have shown that insulin resistance or diabetes may be protective (Reyes et al., 1984; Jawaid et al., 2010), with later onset of ALS observed in those with diabetes (Jawaid et al., 2010). A large Danish case control study found the estimated odds ratio (OR) for ALS in association with diabetes was 0.61 (95% confidence interval (CI)

R.M. AHMED ET AL.

330

0.46–0.80) (Kioumourtzoglou et al., 2015). In a large Swedish case control study, type 2 diabetes was associated with a decreased risk of ALS (OR 0.79, 95% CI 0.68–0.91), whereas type 1 insulin-dependent diabetes was associated with an increased risk (OR 5.38, 95% CI 1.87–15.51), suggesting protective effects may be restricted to type 2 diabetes, which is associated with insulin resistance, rather than type 1, which has an autoimmune pathophysiology and may drive ALS (Turner et al., 2013a; Mariosa et al., 2015). Of interest, an association between ALS risk and obesity was observed (OR 0.81, 95% CI 0.57–1.16) within the Danish case control study. Given that obesity is a risk factor for type 2 diabetes, it is possible that the underlying modifying factor for ALS risk relative to diabetes could be underscored by associations with obesity (Kioumourtzoglou et al., 2015). A phase II clinical trial found that pioglitazone, an oral antidiabetic drug as an add-on therapy to riluzole, does not improve survival in ALS. Rather, this resulted in a 21% increased hazard risk for mortality (Dupuis et al., 2012). It was hypothesized that this drug would be effective due to its antioxidant and antiinflammatory properties. One reason why this effect was not seen could be secondary to its effect on insulin resistance and glucose homeostasis (Jawaid et al., 2014; Vercruysse et al., 2016). Despite these findings, a number of other studies have suggested, that while there may be an increased incidence of diabetes, it is not protective nor a prognostic factor associated with ALS (Lekoubou et al., 2014; Paganoni et al., 2015). To determine the association of diabetes and insulin resistance with ALS and whether this metabolic change drives neurodegeneration or is protective, large prospective longitudinal multiple center studies are required across multiple countries and ethnic groups.

Causes of weight loss Weight balance is the result of energy intake and expenditure, both being affected in ALS in a differential manner according to disease progression.

HYPERMETABOLISM Patients with ALS generally develop hypermetabolism, an unexplained and supraphysiologic rise in resting energy expenditure observed in up to 50% of patients (Bouteloup et al., 2009; Vaisman et al., 2009). Hypermetabolism in ALS has been shown to increase risk of earlier death (Steyn et al., 2018). The finding of increased prevalence of hypermetabolism in ALS seems somewhat paradoxic, given that as disease progresses, denervation and inactivity contributes to muscle atrophy and an

overall loss of free fat mass. Muscle contributes significantly to energy use, and thus loss of muscle mass would be expected to contribute to decreased resting energy expenditure (Bouteloup et al., 2009). Several variables have been hypothesized to contribute to hypermetabolism, including uncontrolled fasciculations (Dupuis et al., 2011), increased respiratory muscle work (Kasarskis et al., 1996), denervation of muscles leading to a switch in muscle-energy use (Steyn et al., 2018), and mitochondrial dysfunction (Menzies et al., 2002). It is also possible that a hypermetabolic state is intrinsically linked to the process of neurodegeneration, with several genetic animal models exhibiting hypermetabolism and weight loss (Dupuis et al., 2004; Chiang et al., 2010; Shan et al., 2010; Xu et al., 2010). Energy expenditure has also been found to change with disease progression, body composition, and physical activity (Kasarskis et al., 2014). Further studies are required to ascertain the factors affecting energy metabolism, as these may change with disease progression and both peripheral and central pathologic spread.

EATING BEHAVIOR AND NUTRITIONAL INTAKE IN ALS AS A FUNCTION OF PROGRESSIVE DISABILITY

ALS is regarded as a disease associated with malnutrition. It was found that nutritional intake decreases as the disease progresses, and decreased intake is observed in those with lower functional levels (Park et al., 2015). Empirical evidence regarding optimal energy intake in ALS remains to be defined, and dietary advice is prioritized for patients presenting with or that develop bulbar symptoms (Burgos et al., 2018). However, it has been accepted that patients with ALS may develop reduced dietary intake secondary to dysphagia (Kuhnlein et al., 2008), loss of appetite (Holm et al., 2013), and/or difficulty consuming food due to weakness of their hands (Ngo et al., 2017). Decreased food intake may also result from the ongoing disease progression. Indeed, progressive loss of cranial nerves results in a significant impairment in the process of swallowing (Robbins, 1987). The resulting loss of food bolus control could occur very early in the disease and may impact a patient’s capacity to consume sufficient calories during a meal. When considered alongside impaired respiratory function, bulbar dysfunction could compound fear of choking, which in turn may further reduce calorie and fluid intake, leading to dehydration and malnutrition (Neudert et al., 2001). Moreover, increased or prolonged meal times as a consequence of dysphagia may further impair appetite, as meal size is reduced in response to fatigue and endogenous processes that regulate and signal satiety (Ngo et al., 2017).

HYPOTHALAMUS AND WEIGHT LOSS IN AMYOTROPHIC LATERAL SCLEROSIS In ALS, 25%–35% of patients experience bulbar onset of weakness with dysphagia, and 70%–80% of patients develop dysphagia throughout the course of the disease (Muscaritoli et al., 2012; Korner et al., 2013). Initial management of dysphagia involves modification of food and fluid consistency (blending food, adding thickeners to liquids), and education of patients and carers in swallowing techniques (Heffernan et al., 2004). Swallowing difficulties can be managed through enteral feeding, but this too may present unique challenges and complications that would result in malnutrition and energy deficit (Blumenstein et al., 2014). Disability can also result in social isolation and embarrassment, further compromising a patient’s nutritional status. Social aspects of eating can be affected by dysphagia or time-consuming and exhausting practices associated with eating. Introduction of adapted eating utensils can help a patient to maintain autonomy (Greenwood, 2013). Sialorrhea (drooling or excessive salivation) can further burden patients during mealtimes, especially within social settings (Heffernan et al., 2004; Andersen et al., 2007), lessening the desire to eat and the enjoyment of meals. Treatment options range from anticholinergic drugs and botulinum toxin injections to reduce saliva production, to radiologic and surgical interventions (Andersen et al., 2007). Gastric emptying and colonic transit times can also be delayed in patients with ALS (Toepfer et al., 2009). While contributing to discomfort associated with eating, impairments in gut function or delayed gastric emptying can also impact the release of gut hormones (Ngo et al., 2015) associated with appetite control (Chaudhri et al., 2006). This may alter the perception of hunger and satiety. It is thus undisputed that disease progression affects eating behavior, thus exacerbating weight loss.

EXISTENCE OF ALTERATIONS IN APPETITE AND FOOD INTAKE IN ALS, INDEPENDENT OF DISEASE PROGRESSION

Besides difficulties in eating associated with disease progression, ALS patients independently develop alterations in appetite. Two recent studies demonstrated that loss of appetite was present in many ALS patients independently of dysphagia, and was correlated with greater weight loss, lower energy intake, and worsened ALS functional rating scale (Mezoian et al., 2020; Ngo et al., 2019). Contrastingly, early symptomatic ALS patients may have increased total daily energy intake compared to control subjects (Huisman et al., 2015), similar to what has been observed in mouse models of ALS (Dupuis et al., 2004; Vercruysse et al., 2016). That these alterations in appetite and food intake occur independently of dysphagia (which is caused by the progression of symptoms) is

331

highly suggestive of defects in hypothalamic networks that control appetite.

INVOLVEMENT OF THE HYPOTHALAMUS IN ALS The hypothalamus is the primary brain region that regulates energy balance through homeostatic control of food intake and energy expenditure. The hypothalamus constitutes an integration node of an expansive network of neurocircuitry that controls our overall behavior toward food. Given that both energy expenditure and eating behavior appears to be profoundly affected in ALS, recent studies began to investigate the potential occurrence of defects in the hypothalamus in relation to body weight.

Structural and pathologic defects in ALS hypothalamus There is increasing evidence for involvement of the hypothalamus in several neurodegenerative disorders including Alzheimer’s disease, Huntington’s disease, and FTD (Ahmed et al., 2015; Vercruysse et al., 2018). Recent studies have shown that the hypothalamus undergoes region-specific degeneration, with imaging studies showing that at least a 22% reduction is found in hypothalamic volume in patients with ALS and presymptomatic gene carriers compared to controls (Gorges et al., 2017). This atrophy involves both the anterior and posterior regions and is associated with changes in BMI, especially in familial forms of the disease (Gorges et al., 2017). Importantly, hypothalamic atrophy is also present in presymptomatic gene carriers, thus preceding motor symptoms, and is associated with earlier disease onset (Gorges et al., 2017). Few pathology studies have been performed in the hypothalamus. Cykowski and collaborators have observed TDP-43 inclusions in the hypothalamus, with inclusions in the lateral hypothalamus associated with lower BMI (Cykowski et al., 2014). Further studies are required to ascertain whether this finding is simply related to changes in BMI or whether hypothalamic atrophy affects pathologic spread and disease pathogenesis. Furthermore, it is necessary to characterize the presence of neurodegeneration, and whether specific neuronal populations might be affected prior to and during the course of disease.

Altered melanocortin pathways in ALS The arcuate nucleus (ARC, or infundibular nucleus in human) of the hypothalamus is critically involved in the homeostatic regulation of energy intake and expenditure, mostly through the antagonistic action of POMCand AgRP-expressing neurons, two major neuronal types

332

R.M. AHMED ET AL.

constituting together the melanocortin system. POMC neurons produce a number of anorexigenic peptides, including a-MSH (melanocyte stimulating hormone) (Diano, 2011; Lu et al., 2011; Ryan et al., 2011; Long et al., 2014), and AgRP neurons produce the antagonistic and endogenous a-MSH antagonist. Together they integrate central and peripheral signals that stimulate (for AgRP neurons) or inhibit (for POMC neurons) food intake. In particular, the melanocortin system is modulated by peripheral hormones such as leptin, ghrelin, and adiponectin, and centrally derived neurotransmitters such as serotonin (Fig. 20.2) (Chaptini and Peikin, 2008). Neuronal projections from the ARC connect with other hypothalamic and extrahypothalamic nuclei to convey signals that balance energy needs with energy intake. The action of the melanocortinergic system on food intake and energy expenditure is mediated by melanocortin-3 (MC3R) and -4 (MC4R) receptors,

which are activated by POMC-derived peptides and antagonized by AgRP and are expressed broadly in the hypothalamus and brainstem. Direct evidence of melanocortin system impairment is currently lacking in ALS patients. The antidiabetic drug pioglitazone is known to increase food intake through direct inhibition of POMC neurons (Diano, 2011; Long et al., 2014). In humans, this action is directly translated through the observation of 3–5 kg of weight gain in multiple clinical trials. Importantly, while all the peripheral responses to pioglitazone could be observed in ALS patients, pioglitazone did not lead to weight gain in ALS patients and was unable to increase food intake in mouse models of ALS (Vercruysse et al., 2016). This correlated with loss of POMC expression and of POMC immunoreactive neurons, and increased AgRP immunoreactivity and expression in the same mouse models. Such alterations in melanocortin

Fig. 20.2. Schematic representation of hypothalamic pathways that control appetite and metabolism. In ALS TDP-43 deposition has been shown in the lateral hypothalamus. Increased AgRP levels (green) and decreased POMC (green) levels have been shown in ALS. Currently, research is focusing on possible connections between the hypothalamus and motor cortex. 5HTR, serotonin receptor; ARC, arcuate nucleus (in human: infundibular nucleus); CCK, cholecystokinin; GHSR, growth hormone secretagogue receptor; GIT, gastrointestinal tract; IGF-1, insulin-like growth factor-1; LEPR, leptin receptor; LHA, lateral hypothalamic area; MCH, melanin-concentrating hormone; MCR, melanocortin receptor; ORX, orexin; OXT, oxytocin; PVN, paraventricular nucleus; TDP43, TAR DNA binding protein 43.

HYPOTHALAMUS AND WEIGHT LOSS IN AMYOTROPHIC LATERAL SCLEROSIS pathways could underlie increased eating behavior in some early symptomatic ALS patients (Nelson et al., 2000; Huisman et al., 2015), and could orientate lipid metabolic pathways through effects on hepatic metabolism (Perez-Tilve et al., 2010; Leckstrom et al., 2011). Causes of melanocortin impairment are still incompletely understood. Pharmacologic evidence suggested that increasing serotonergic tone could rescue POMC expression in mouse models of ALS (Vercruysse et al., 2016), while decreased leptin, as a consequence of decreased fat mass, could also constitute a primary cause of loss of POMC expression. Interestingly, genetically decreasing leptin mitigates weight loss and energy expenditure in ALS mice (Lim et al., 2014). Conversely, deletion of the MC4R, the major melanocortin receptor with effects on energy balance, does not (Doshi et al., 2017). Further work is needed to clarify the role of the melanocortin pathway in ALS and its possible therapeutic potential. It would be especially important to perform careful pathologic examination of the hypothalamus in ALS, as has been performed, for instance, in type 2 diabetes (Alkemade et al., 2012), Prader–Willi syndrome (Goldstone et al., 2002), and FTD (Piguet et al., 2011).

Peripheral signaling to the hypothalamus could mediate weight changes in ALS Numerous peripheral signals interact to control appetite; many of these converge at the level of the hypothalamus to modulate energy intake and expenditure. Factors released proportionate to fat mass (including leptin and insulin) signal peripheral calorie storage and regulate central mechanisms that control appetite. Gut peptides (such as peptide YY [PYY], glucagon-like peptide 1 [GLP1], cholecystokinin [CCK], and ghrelin) guide our perception of hunger and satiety and regulate the size and frequency of our meals. Endogenous mediators of food intake enhance signaling of these peripheral peptides. These include proinflammatory cytokines (e.g., interleukin-6 (IL6) (Shirazi et al., 2013) and tumor necrosis factor alpha (TNF-a) (Romanatto et al., 2007)) and the nutrients that we consume (including sugars, fats, and proteins). Oxytocin, which is produced in the hypothalamus, is an important component of the pathways activated by leptin and is believed to decrease food intake (Sabatier et al., 2013). Emerging data suggest that the production and/or release of many of these peripheral regulators may change in patients with ALS. For example, a decrease in gastric inhibitory polypeptide (GIP) and ghrelin and an increase in IL-6 and TNF-a are observed in patients with ALS (Ngo et al., 2015). Moreover, levels of fasting AgRP increase in patients with FTD (Ahmed et al., 2015).

333

It was recently found that circulating levels of leptin and adiponectin are strongly and independently correlated with ALS risk, and that leptin is associated with ALS survival (Nagel et al., 2017). It was also found that circulating levels of leptin and NPY may inform differential diagnosis across the ALS-FTD spectrum, with levels of NPY increased in ALS and decreased in bvFTD, potentially in response to increased food intake in FTD and reduced energy stores in ALS (Ahmed et al., 2019). While reinforcing the notion that circulating factors could alter energy balance in ALS, these studies provide additional insights to suggest that circulating regulators of appetite may also contribute to pathophysiologic mechanisms that contribute to the progression of disease. Additional studies are, however, needed to clarify the roles of these factors, and specifically to differentiate the impact of these factors from expected physiologic changes in the release and actions of these factors that may occur secondary to a change in body composition or pathophysiology that arises secondary to disease.

Central pathways of food reward converging to the hypothalamus in ALS Brain regions outside of the hypothalamus, including limbic (i.e., nucleus accumbens, amygdala, and hippocampus) and cortical (i.e., orbitofrontal cortex, cingulate gyrus, and insula) regions, and neurotransmitter systems (including dopaminergic and serotonergic pathways) direct rewards associated with food (Volkow et al., 2011). These brain regions and pathways modify eating behaviors, and may override hypothalamic control of energy homeostasis, resulting in altered calorie intake (Volkow et al., 2008). In this instance, palatability is a key motivator for food consumption, and loss of pleasure associated with eating may have a dramatic effect on a patient’s desire to consume sufficient calories to maintain a stable body weight. This is especially relevant to those patients that experience early and progressively worsening bulbar symptoms, or when PEG feeding is the only option. While the extent to which altered reward systems contribute to impairments in energy consumption in ALS remains mostly unexplored, it is likely that this is relevant to some patients. For example, altered taste is observed in some individuals with ALS (Tarlarini et al., 2019), and taste is a critical factor in determining our motivation to maintain a healthy intake (Brug, 2008). Changes in cognition, anxiety, and depression (Kurt et al., 2007) may also impact motivation and behaviors around food. This is particularly relevant, as patients with frontotemporal involvement develop cognitive changes associated with increased caloric intake and sugar preference (Ahmed et al., 2016).

334

R.M. AHMED ET AL.

Genetic variants could influence weight loss through hypothalamic signaling Variability within patients and between controls may mask differences in the operation of mechanisms that control appetite and energy homeostasis. These variances may reflect, in part, genetic heterogeneity. Numerous gene variants are associated with altered control of metabolic processes including BMI (Locke et al., 2015), glucose homeostasis (International Schizophrenia Consortium, 2009; Saxena et al., 2010; Scott et al., 2012), and type-2 diabetes (Morris et al., 2012). The capacity of ALS patients to direct energy supply may depend on similar genetic interactions. Gene sets within BMI-associated loci are enriched for expression in the brain and central nervous system. Of significant interest to ALS, many of these gene variants also encode central control of locomotor activity, the regulation of postsynaptic membrane potentials, and neurotransmitter release (Locke et al., 2015). For now, these genetic associations, relative to appetite control and body weight regulation, remain mostly unexplored. However, given that premorbid BMI is consistently shown to be associated with ALS risk and that genetics could define BMI, this cannot be excluded.

TREATMENT TARGETS FOR PREVENTING WEIGHT LOSS The convergence of epidemiologic, clinical, and experimental research evidence suggests that preventing or reversing weight loss could slow down disease progression. Despite the perception of relative ease of intervention, very few interventional studies have investigated the value of a nutritional intervention. A high-caloric diet is generally promoted and is supported by two different randomized clinical trials. Wills and coworkers have studied the safety and efficacy of high-calorie diets in gastrostomized ALS patients. These authors found that a high-caloric, high-carbohydrate diet was safe and well tolerated by patients and that this was associated with increased survival. However, this pilot study included by design very advanced patients and few patients per arm (Wills et al., 2014). Similarly, non-double blind, nonrandomized studies have reported that a high-carbohydrate diet and high-fat diet could result in stabilization of BMI, with no effect on functional decline (as assessed by a change in ALSFRS, the ALS Functional Rating Scale) and muscle mass (Dorst et al., 2013). A very recent randomized, double blind clinical trial study focused on safety and efficacy of a high-calorie, fatty dietary supplementation in 200 ALS patients. While the study did not meet its primary endpoint (improved survival at 18 months), a post hoc analysis revealed that dietary

supplementation improved survival in normal to normal-to-fast progressing ALS patients (loss of >0.62 ALSFRS point/month at baseline, i.e., half of the patients included) but not in slow-progressing ALS patients. Importantly, dietary supplementation increased BMI in fast progressors but not in slow progressors (Ludolph et al., 2020). Thus increased calorie intake could improve survival of a large subset of ALS patients. Further clinical studies are needed to provide definitive evidence. A recent study utilizing protein supplementation resulted in increased BMI and stabilization of the ALSFRS, suggesting a possible role for protein supplementation that requires further investigation (Silva et al., 2010). Currently, patients are unable to be advised on the ideal diet to slow progression in ALS, or on the effects that diet may have on metabolic changes. Nevertheless, recommendations are to maintain sufficient energy intake to prevent weight loss (Burgos et al., 2018). Further longitudinal studies are required to ascertain the relationships between eating behavior, metabolic changes, and weight changes in ALS and the role that the hypothalamus plays in this. Understanding these interactions will aid in our understanding the pathogenesis of ALS and providing druggable treatment targets that can modify disease progression and survival. Given that metabolic and hypothalamic structural changes potentially appear decades before disease onset, an understanding of the interaction between metabolism and ALS pathogenesis is crucial as we attempt to prevent and cure this devastating condition.

REFERENCES Ahmed RM et al. (2016). Cognition and eating behavior in amyotrophic lateral sclerosis: effect on survival. J Neurol 263: 1593–1603. Ahmed RM et al. (2018a). Lipid metabolism and survival across the frontotemporal dementia-amyotrophic lateral sclerosis spectrum: relationships to eating behavior and cognition. J Alzheimers Dis 61: 773–783. Ahmed RM et al. (2018b). Physiological changes in neurodegeneration—mechanistic insights and clinical utility. Nat Rev Neurol 14: 259–271. Ahmed RM et al. (2015). Eating behavior in frontotemporal dementia: peripheral hormones vs hypothalamic pathology. Neurology 85: 1310–1317. Ahmed RM et al. (2019). Eating peptides: biomarkers of neurodegeneration in amyotrophic lateral sclerosis and frontotemporal dementia. Ann Clin Transl Neurol 6: 486–495. Alkemade A et al. (2012). AgRP and NPY expression in the human hypothalamic infundibular nucleus correlate with body mass index, whereas changes in alphaMSH are related to type 2 diabetes. J Clin Endocrinol Metab 97: E925–E933.

HYPOTHALAMUS AND WEIGHT LOSS IN AMYOTROPHIC LATERAL SCLEROSIS Andersen PM et al. (2007). Good practice in the management of amyotrophic lateral sclerosis: clinical guidelines. An evidence-based review with good practice points. EALSC working group. Amyotroph Lateral Scler 8: 195–213. Blumenstein I, Shastri YM, Stein J (2014). Gastroenteric tube feeding: techniques, problems and solutions. World J Gastroenterol 20: 8505–8524. Bouteloup C et al. (2009). Hypermetabolism in ALS patients: an early and persistent phenomenon. J Neurol 256: 1236–1242. Braak H et al. (2013). Amyotrophic lateral sclerosis—a model of corticofugal axonal spread. Nat Rev Neurol 9: 708–714. Brettschneider J et al. (2013). Stages of pTDP-43 pathology in amyotrophic lateral sclerosis. Ann Neurol 74: 20–38. Brug J (2008). Determinants of healthy eating: motivation, abilities and environmental opportunities. Fam Pract 25: i50–i55. Burgos R et al. (2018). ESPEN guideline clinical nutrition in neurology. Clin Nutr 37: 354–396. Burrell JR, Kiernan MC, Vucic S et al. (2011). Motor neuron dysfunction in frontotemporal dementia. Brain 134: 2582–2594. Chaptini L, Peikin S (2008). Neuroendocrine regulation of food intake. Curr Opin Gastroenterol 24: 223–229. Chaudhri O, Small C, Bloom S (2006). Gastrointestinal hormones regulating appetite. Philos Trans R Soc Lond Ser B Biol Sci 361: 1187–1209. Chiang PM et al. (2010). Deletion of TDP-43 down-regulates Tbc1d1, a gene linked to obesity, and alters body fat metabolism. Proc Natl Acad Sci U S A 107: 16320–16324. Chio A et al. (2009). Lower serum lipid levels are related to respiratory impairment in patients with ALS. Neurology 73: 1681–1685. Clavelou P et al. (2013). Rates of progression of weight and forced vital capacity as relevant measurement to adapt amyotrophic lateral sclerosis management for patient result of a French multicentre cohort survey. J Neurol Sci 331: 126–131. Colman E et al. (2008). An evaluation of a data mining signal for amyotrophic lateral sclerosis and statins detected in FDA’s spontaneous adverse event reporting system. Pharmacoepidemiol Drug Saf 17: 1068–1076. Cykowski MD et al. (2014). TDP-43 pathology in the basal forebrain and hypothalamus of patients with amyotrophic lateral sclerosis. Acta Neuropathol Commun 2: 171. Desport JC et al. (2000). Nutritional assessment and survival in ALS patients. Amyotroph Lateral Scler Other Motor Neuron Disord 1: 91–96. Desport JC et al. (1999). Nutritional status is a prognostic factor for survival in ALS patients. Neurology 53: 1059–1063. Devenney E, Vucic S, Hodges JR et al. (2015). Motor neuron disease-frontotemporal dementia: a clinical continuum. Expert Rev Neurother 15: 509–522. Devine MS et al. (2014). Study of motor asymmetry in ALS indicates an effect of limb dominance on onset and spread of weakness, and an important role for upper motor

335

neurons. Amyotroph Lateral Scler Frontotemporal Degener 15: 481–487. Diano S (2011). New aspects of melanocortin signaling: a role for PRCP in alpha-MSH degradation. Front Neuroendocrinol 32: 70–83. Dorst J, Cypionka J, Ludolph AC (2013). High-caloric food supplements in the treatment of amyotrophic lateral sclerosis: a prospective interventional study. Amyotroph Lateral Scler Frontotemporal Degener 14: 533–536. Dorst J et al. (2011). Patients with elevated triglyceride and cholesterol serum levels have a prolonged survival in amyotrophic lateral sclerosis. J Neurol 258: 613–617. Doshi S, Gupta P, Kalb RG (2017). Genetic induction of hypometabolism by ablation of MC4R does not suppress ALSlike phenotypes in the G93A mutant SOD1 mouse model. Sci Rep 7: 13150. Dupuis L et al. (2008). Dyslipidemia is a protective factor in amyotrophic lateral sclerosis. Neurology 70: 1004–1009. Dupuis L et al. (2012). A randomized, double blind, placebocontrolled trial of pioglitazone in combination with riluzole in amyotrophic lateral sclerosis. PLoS One 7: e37885. Dupuis L et al. (2004). Evidence for defective energy homeostasis in amyotrophic lateral sclerosis: benefit of a high-energy diet in a transgenic mouse model. Proc Natl Acad Sci U S A 101: 11159–11164. Dupuis L, Pradat PF, Ludolph AC et al. (2011). Energy metabolism in amyotrophic lateral sclerosis. Lancet Neurol 10: 75–82. Fasano A et al. (2017). Percutaneous endoscopic gastrostomy, body weight loss and survival in amyotrophic lateral sclerosis: a population-based registry study. Amyotrophic Lateral Scler Frontotemporal Degener 18: 233–242. Filippi M et al. (2015). Progress towards a neuroimaging biomarker for amyotrophic lateral sclerosis. Lancet Neurol 14: 786–788. Filippini N et al. (2010). Corpus callosum involvement is a consistent feature of amyotrophic lateral sclerosis. Neurology 75: 1645–1652. Gallo V et al. (2013). Prediagnostic body fat and risk of death from amyotrophic lateral sclerosis: the EPIC cohort. Neurology 80: 829–838. Goldstone AP, Unmehopa UA, Bloom SR et al. (2002). Hypothalamic NPY and agouti-related protein are increased in human illness but not in Prader-Willi syndrome and other obese subjects. J Clin Endocrinol Metab 87: 927–937. Golomb BA, Kwon EK, Koperski S et al. (2009). Amyotrophic lateral sclerosis-like conditions in possible association with cholesterol-lowering drugs: an analysis of patient reports to the University of California, San Diego (UCSD) statin effects study. Drug Saf 32: 649–661. Gorges M et al. (2017). Hypothalamic atrophy is related to body mass index and age at onset in amyotrophic lateral sclerosis. J Neurol Neurosurg Psychiatry 88: 1033–1041. Greenwood DI (2013). Nutrition management of amyotrophic lateral sclerosis. Nutr Clin Pract 28: 392–399.

336

R.M. AHMED ET AL.

Hardiman O, Van Den Berg LH, Kiernan MC (2011). Clinical diagnosis and management of amyotrophic lateral sclerosis. Nat Rev Neurol 7: 639–649. Heffernan C et al. (2004). Nutritional management in MND/ ALS patients: an evidence based review. Amyotroph Lateral Scler Other Motor Neuron Disord 5: 72–83. Hodges J (2012). Familial frontotemporal dementia and amyotrophic lateral sclerosis associated with the C9ORF72 hexanucleotide repeat. Brain 135: 652–655. Holm T et al. (2013). Severe loss of appetite in amyotrophic lateral sclerosis patients: online self-assessment study. Interact J Med Res 2: e8. Huisman MH et al. (2015). Effect of presymptomatic body mass index and consumption of fat and alcohol on amyotrophic lateral sclerosis. JAMA Neurol 72: 1155–1162. Ikeda K et al. (2012). Relationships between disease progression and serum levels of lipid, urate, creatinine and ferritin in Japanese patients with amyotrophic lateral sclerosis: a cross-sectional study. Intern Med 51: 1501–1508. International Schizophrenia Consortium et al. (2009). Common polygenic variation contributes to risk of schizophrenia and bipolar disorder. Nature 460: 748–752. Ioannides ZA et al. (2016). Altered metabolic homeostasis in amyotrophic lateral sclerosis: mechanisms of energy imbalance and contribution to disease progression. Neurodegener Dis 16: 382–397. Ioannides ZA et al. (2017). Anthropometric measures are not accurate predictors of fat mass in ALS. Amyotroph Lateral Scler Frontotemporal Degener 18: 486–491. Jawaid A, Paganoni S, Hauser C et al. (2014). Trials of antidiabetic drugs in amyotrophic lateral sclerosis: proceed with caution? Neurodegener Dis 13: 205–208. Jawaid A et al. (2010). ALS disease onset may occur later in patients with pre-morbid diabetes mellitus. Eur J Neurol 17: 733–739. Kasarskis EJ et al. (1996). Nutritional status of patients with amyotrophic lateral sclerosis: relation to the proximity of death. Am J Clin Nutr 63: 130–137. Kasarskis EJ et al. (2014). Estimating daily energy expenditure in individuals with amyotrophic lateral sclerosis. Am J Clin Nutr 99: 792–803. Kiernan MC et al. (2011). Amyotrophic lateral sclerosis. Lancet 377: 942–955. Kioumourtzoglou MA et al. (2015). Diabetes mellitus, obesity, and diagnosis of amyotrophic lateral sclerosis: a population-based study. JAMA Neurol 72: 905–911. Kirk SE, Tracey TJ, Steyn FJ et al. (2019). Biomarkers of metabolism in amyotrophic lateral sclerosis. Front Neurol 10: 191. Korner S et al. (2013). Weight loss, dysphagia and supplement intake in patients with amyotrophic lateral sclerosis (ALS): impact on quality of life and therapeutic options. BMC Neurol 13: 84. Kuhnlein P et al. (2008). Diagnosis and treatment of bulbar symptoms in amyotrophic lateral sclerosis. Nat Clin Pract Neurol 4: 366–374.

Kurt A, Nijboer F, Matuz T et al. (2007). Depression and anxiety in individuals with amyotrophic lateral sclerosis: epidemiology and management. CNS Drugs 21: 279–291. Leckstrom A, Lew PS, Poritsanos NJ et al. (2011). Central melanocortin receptor agonist reduces hepatic lipogenic gene expression in streptozotocin-induced diabetic mice. Life Sci 88: 664–669. Lekoubou A, Matsha TE, Sobngwi E et al. (2014). Effects of diabetes mellitus on amyotrophic lateral sclerosis: a systematic review. BMC Res Notes 7: 171. Lim MA et al. (2014). Genetically altering organismal metabolism by leptin-deficiency benefits a mouse model of amyotrophic lateral sclerosis. Hum Mol Genet 23: 4995–5008. Lindauer E et al. (2013). Adipose tissue distribution predicts survival in amyotrophic lateral sclerosis. PLoS One 8: e67783. Locke AE et al. (2015). Genetic studies of body mass index yield new insights for obesity biology. Nature 518: 197–206. Lomen-Hoerth C, Anderson T, Miller B (2002). The overlap of amyotrophic lateral sclerosis and frontotemporal dementia. Neurology 59: 1077–1079. Long L et al. (2014). PPARgamma ablation sensitizes proopiomelanocortin neurons to leptin during high-fat feeding. J Clin Invest 124: 4017–4027. Lu M et al. (2011). Brain PPAR-gamma promotes obesity and is required for the insulin-sensitizing effect of thiazolidinediones. Nat Med 17: 618–622. Ludolph AC, Brettschneider J (2015). TDP-43 in amyotrophic lateral sclerosis—is it a prion disease? Eur J Neurol 22: 753–761. Ludolph AC et al. (2020). Effect of high-caloric nutrition on survival in amyotrophic lateral sclerosis. Ann Neurol 87: 206–216. https://doi.org/10.1002/ana.25661. Epub 2020 Jan 6. Mackenzie IR, Rademakers R, Neumann M (2010). TDP-43 and FUS in amyotrophic lateral sclerosis and frontotemporal dementia. Lancet Neurol 9: 995–1007. Marin B et al. (2011). Alteration of nutritional status at diagnosis is a prognostic factor for survival of amyotrophic lateral sclerosis patients. J Neurol Neurosurg Psychiatry 82: 628–634. Mariosa D et al. (2015). Association between diabetes and amyotrophic lateral sclerosis in Sweden. Eur J Neurol 22: 1436–1442. Menzies FM, Ince PG, Shaw PJ (2002). Mitochondrial involvement in amyotrophic lateral sclerosis. Neurochem Int 40: 543–551. Mezoian T et al. (2020). Loss of appetite in amyotrophic lateral sclerosis is associated with weight loss and decreased calorie consumption independent of dysphagia. Muscle Nerve 61: 230–234. https://doi.org/10.1002/mus.26749. Moglia C et al. (2019). Early weight loss in amyotrophic lateral sclerosis: outcome relevance and clinical correlates in a population-based cohort. J Neurol Neurosurg Psychiatry 90: 666–673. https://doi.org/10.1136/jnnp2018-319611. Montuschi A et al. (2015). Cognitive correlates in amyotrophic lateral sclerosis: a population-based study in Italy. J Neurol Neurosurg Psychiatry 86: 168–173.

HYPOTHALAMUS AND WEIGHT LOSS IN AMYOTROPHIC LATERAL SCLEROSIS Morris AP et al. (2012). Large-scale association analysis provides insights into the genetic architecture and pathophysiology of type 2 diabetes. Nat Genet 44: 981–990. Muscaritoli M et al. (2012). Nutritional and metabolic support in patients with amyotrophic lateral sclerosis. Nutrition 28: 959–966. Nagel G et al. (2017). Adipokines, C-reactive protein and amyotrophic lateral sclerosis—results from a populationbased ALS registry in Germany. Sci Rep 7: 4374. Nakken O, Meyer HE, Stigum H et al. (2019). High BMI is associated with low ALS risk: a population-based study. Neurology 93: e424–e432. Nelson LM, Matkin C, Longstreth Jr WT et al. (2000). Population-based case-control study of amyotrophic lateral sclerosis in western Washington State. II. Diet. Am J Epidemiol 151: 164–173. Neudert C, Oliver D, Wasner M et al. (2001). The course of the terminal phase in patients with amyotrophic lateral sclerosis. J Neurol 248: 612–616. Ngo ST et al. (2017). Exploring targets and therapies for amyotrophic lateral sclerosis: current insights into dietary interventions. Degener Neurol Neuromuscul Dis 7: 95–108. Ngo ST et al. (2015). Altered expression of metabolic proteins and adipokines in patients with amyotrophic lateral sclerosis. J Neurol Sci 357: 22–27. Ngo ST et al. (2019). Loss of appetite is associated with a loss of weight and fat mass in patients with amyotrophic lateral sclerosis. Amyotroph Lateral Scler Frontotemporal Degener 20: 497–505. O’Reilly EJ et al. (2013). Premorbid body mass index and risk of amyotrophic lateral sclerosis. Amyotroph Lateral Scler Frontotemporal Degener 14: 205–211. Paganoni S et al. (2011). Body mass index, not dyslipidemia, is an independent predictor of survival in amyotrophic lateral sclerosis. Muscle Nerve 44: 20–24. Paganoni S et al. (2015). Pre-morbid type 2 diabetes mellitus is not a prognostic factor in amyotrophic lateral sclerosis. Muscle Nerve 52: 339–343. Park Y et al. (2015). Association between nutritional status and disease severity using the amyotrophic lateral sclerosis (ALS) functional rating scale in ALS patients. Nutrition 31: 1362–1367. Perez-Tilve D et al. (2010). Melanocortin signaling in the CNS directly regulates circulating cholesterol. Nat Neurosci 13: 877–882. Peter RS et al. (2017). Life course body mass index and risk and prognosis of amyotrophic lateral sclerosis: results from the ALS registry Swabia. Eur J Epidemiol 32: 901–908. Piguet O et al. (2011). Eating and hypothalamus changes in behavioral-variant frontotemporal dementia. Ann Neurol 69: 312–319. Reich-Slotky R et al. (2013). Body mass index (BMI) as predictor of ALSFRS-R score decline in ALS patients. Amyotroph Lateral Scler Frontotemporal Degener 14: 212–216. Reyes ET et al. (1984). Insulin resistance in amyotrophic lateral sclerosis. J Neurol Sci 63: 317–324. Ringholz GM et al. (2005). Prevalence and patterns of cognitive impairment in sporadic ALS. Neurology 65: 586–590.

337

Robbins J (1987). Swallowing in ALS and motor neuron disorders. Neurol Clin 5: 213–229. Romanatto T et al. (2007). TNF-alpha acts in the hypothalamus inhibiting food intake and increasing the respiratory quotient—effects on leptin and insulin signaling pathways. Peptides 28: 1050–1058. Roubeau V et al. (2015). Nutritional assessment of amyotrophic lateral sclerosis in routine practice: value of weighing and bioelectrical impedance analysis. Muscle Nerve 51: 479–484. Ryan KK et al. (2011). A role for central nervous system PPAR-gamma in the regulation of energy balance. Nat Med 17: 623–626. Sabatier N, Leng G, Menzies J (2013). Oxytocin, feeding, and satiety. Front Endocrinol (Lausanne) 4: 35. Saxena R et al. (2010). Genetic variation in GIPR influences the glucose and insulin responses to an oral glucose challenge. Nat Genet 42: 142–148. Schmidt R et al. (2014). Correlation between structural and functional connectivity impairment in amyotrophic lateral sclerosis. Hum Brain Mapp 35: 4386–4395. Scott RA et al. (2012). Large-scale association analyses identify new loci influencing glycemic traits and provide insight into the underlying biological pathways. Nat Genet 44: 991–1005. Shan X, Chiang PM, Price DL et al. (2010). Altered distributions of Gemini of coiled bodies and mitochondria in motor neurons of TDP-43 transgenic mice. Proc Natl Acad Sci U S A 107: 16325–16330. Shirazi R et al. (2013). Glucagon-like peptide 1 receptor induced suppression of food intake, and body weight is mediated by central IL-1 and IL-6. Proc Natl Acad Sci U S A 110: 16199–16204. Silva LB et al. (2010). Effect of nutritional supplementation with milk whey proteins in amyotrophic lateral sclerosis patients. Arq Neuropsiquiatr 68: 263–268. Stambler N, Charatan M, Cedarbaum JM (1998). Prognostic indicators of survival in ALS. ALS CNTF Treatment Study Group. Neurology 50: 66–72. Steyn FJ et al. (2018). Hypermetabolism in ALS is associated with greater functional decline and shorter survival. J Neurol Neurosurg Psychiatry 89: 1016–1023. Strong MJ (2008). The syndromes of frontotemporal dysfunction in amyotrophic lateral sclerosis. Amyotroph Lateral Scler 9: 323–338. Sutedja NA et al. (2011). Beneficial vascular risk profile is associated with amyotrophic lateral sclerosis. J Neurol Neurosurg Psychiatry 82: 638–642. Tan RH et al. (2015). TDP-43 proteinopathies: pathological identification of brain regions differentiating clinical phenotypes. Brain 138: 3110–3122. Tarlarini C et al. (2019). Taste changes in amyotrophic lateral sclerosis and effects on quality of life. Neurol Sci 40: 399–404. Toepfer M, Folwaczny C, Klauser A et al. (2009). Gastrointestinal dysfunction in amyotrophic lateral sclerosis. Amyotroph Lateral Scler Other Motor Neuron Disord 1: 15–19.

338

R.M. AHMED ET AL.

Turner MR et al. (2013a). Autoimmune disease preceding amyotrophic lateral sclerosis: an epidemiologic study. Neurology 81: 1222–1225. Turner MR et al. (2013b). Controversies and priorities in amyotrophic lateral sclerosis. Lancet Neurol 12: 310–322. Turner MR et al. (2011). Concordance between site of onset and limb dominance in amyotrophic lateral sclerosis. J Neurol Neurosurg Psychiatry 82: 853–854. Vaisman N et al. (2009). Do patients with amyotrophic lateral sclerosis (ALS) have increased energy needs? J Neurol Sci 279: 26–29. Vercruysse P et al. (2016). Alterations in the hypothalamic melanocortin pathway in amyotrophic lateral sclerosis. Brain 139: 1106–1122. Vercruysse P et al. (2018). Hypothalamic alterations in neurodegenerative diseases and their relation to abnormal energy metabolism. Front Mol Neurosci 11: 2.

Volkow ND, Wang GJ, Baler RD (2011). Reward, dopamine and the control of food intake: implications for obesity. Trends Cogn Sci 15: 37–46. Volkow ND, Wang GJ, Fowler JS et al. (2008). Overlapping neuronal circuits in addiction and obesity: evidence of systems pathology. Philos Trans R Soc Lond Ser B Biol Sci 363: 3191–3200. Vucic S, Rothstein JD, Kiernan MC (2014). Advances in treating amyotrophic lateral sclerosis: insights from pathophysiological studies. Trends Neurosci 37: 433–442. Wills AM et al. (2014). Hypercaloric enteral nutrition in patients with amyotrophic lateral sclerosis: a randomised, double-blind, placebo-controlled phase 2 trial. Lancet 383: 2065–2072. Xu YF et al. (2010). Wild-type human TDP-43 expression causes TDP-43 phosphorylation, mitochondrial aggregation, motor deficits, and early mortality in transgenic mice. J Neurosci 30: 10851–10859.

Section 10 Lateral tuberal nucleus See Introduction to the volume

This page intentionally left blank

Section 11 Lateral hypothalamic area, perifornical area

This page intentionally left blank

Handbook of Clinical Neurology, Vol. 180 (3rd series) The Human Hypothalamus: Middle and Posterior Region D.F. Swaab, F. Kreier, P.J. Lucassen, A. Salehi, and R.M. Buijs, Editors https://doi.org/10.1016/B978-0-12-820107-7.00021-5 Copyright © 2021 Elsevier B.V. All rights reserved

Chapter 21

The orexin/hypocretin system in neuropsychiatric disorders: Relation to signs and symptoms ROLF FRONCZEK1,2*, MINK SCHINKELSHOEK1,2, LING SHAN1,2,3, AND GERT JAN LAMMERS1,2 1

Department of Neurology, Leiden University Medical Centre, Leiden, The Netherlands 2

Sleep Wake Centre SEIN, Heemstede, The Netherlands

3

Department Neuropsychiatric Disorders, Netherlands Institute for Neuroscience, an Institute of the Royal Netherlands Academy of Arts and Sciences, Amsterdam, The Netherlands

Abstract Hypocretin-1 and 2 (or orexin A and B) are neuropeptides exclusively produced by a group of neurons in the lateral and dorsomedial hypothalamus that project throughout the brain. In accordance with this, the two different hypocretin receptors are also found throughout the brain. The hypocretin system is mainly involved in sleep–wake regulation, but also in reward mechanisms, food intake and metabolism, autonomic regulation including thermoregulation, and pain. The disorder most strongly linked to the hypocretin system is the primary sleep disorder narcolepsy type 1 caused by a lack of hypocretin signaling, which is most likely due to an autoimmune process targeting the hypocretin-producing neurons. However, the hypocretin system may also be affected, but to a lesser extent and less specifically, in various other neurological disorders. Examples are neurodegenerative diseases such as Alzheimer’s, Huntington’s and Parkinson’s disease, immune-mediated disorders such as multiple sclerosis, neuromyelitis optica, and anti-Ma2 encephalitis, and genetic disorders such as type 1 diabetus mellitus and Prader–Willi Syndrome. A partial hypocretin deficiency may contribute to the sleep features of these disorders.

INTRODUCTION The hypocretin system was discovered in 1998 simultaneously by two research groups. One group named these newfound peptides hypocretins, because they are exclusively produced in the hypothalamus and have a weak sequence homology to the incretin hormone family (de Lecea et al., 1998). Another group named the same peptides orexins (οrEjZ ¼ appetite), because intracerebroventricular injection of these peptides induced food intake in rats (Sakurai et al., 1998). Hypocretin-1 and -2 (or orexin A and B) are derived from the precursor peptide preprohypocretin. Hypocretin-1 is 33 amino acids in

length and has an N-terminal pyroglutamyl residue and an amidated C-terminal. Four cysteine residues in the peptide form two sets of intrachain disulfide bonds. Hypocretin-2 is 28 amino acids in length having an amidated C-terminal (Fig. 21.1). There are two hypocretin 7-transmembrane G-protein-coupled receptors encoded by seven exons. Receptor 1 has a preferential affinity for hypocretin-1, whereas receptor 2 binds both hypocretins with equal affinity (Sakurai et al., 1998). Hypocretin-1 can be measured in the cerebrospinal fluid (CSF), which is mostly done with a commercially available radioimmunoassay (RIA). However, with the currently available RIAs, it is technically not possible to reliably measure

*Correspondence to: Rolf Fronczek, Leiden University Medical Centre, Albinusdreef 2, 2333 RC Leden, The Netherlands. Tel: +31-71-526-2197, Fax: +31-(0)71-524-8253, E-mail: [email protected]

344

R. FRONCZEK ET AL. s

s

qplpdccrqktcscrlyellhgagnhaagiltl-NH2 s

1

rsgppglqgrlqrllqasgnhaagiltm-NH

s

signal

34

hypocre n-1

66 70 GKR

hypocre n-2

97 GRR

131

Fig. 21.1. Hypocretin-1 and -2 (or orexin A and B) are derived from the precursor peptide preprohypocretin. Hypocretin-1 is 33 amino acids in length and has an N-terminal pyroglutamyl residue and an amidated C-terminal. Four cysteine residues in the peptide form two sets of intrachain disulfide bonds. Hypocretin-2 is 28 amino acids in length having an amidated C-terminal. Both hypocretin-1 and -2 are produced by the same hypocretin neurons, and both receptor 1 and 2 are found throughout the brain. Drawn by R. Fronczek.

hypocretin-2, because lacking disulfide bonds it is probably less stable than hypocretin-1 (Nishino, 2005).

NEUROANATOMY OF HYPOCRETIN NEURONS Hypocretin-1 and -2 are both exclusively produced by a group of neurons located in the lateral and dorsomedial hypothalamus (Hunt et al., 2014), centered around the fornix and more caudally near the mamillary bodies (Peyron et al., 2000). Hypocretin-producing neurons among other peptides also contain neuronal activityregulated pentraxin and dynorphin (Blouin et al., 2005; Crocker et al., 2005). In humans, the total number of hypocretin-producing neurons at one side of the hypothalamus is estimated to range from 15,000–20,000 (using in situ hybridization for preprohypocretin) (Peyron et al., 2000), to 50,000–80,000 (using immunocytochemistry with an antibody against hypocretin-1) (Fronczek et al., 2005). Although these neurons reside in a small area, they are involved in various functions (sleep–wake regulation and metabolism, among others) and thus project throughout the central nervous system to hypocretin receptors 1 and 2 (Peyron et al., 1998; Adamantidis and de Lecea, 2008; Luppi et al., 2013). In the lateral hypothalamic area, the melanin-concentrating hormone (MCH)producing neurons are intermingled with hypocretinproducing neurons (Aziz et al., 2008).

THE FUNCTIONS OF HYPOCRETIN Sleep–wake regulation Sleep is hypothesized to be mainly regulated by a “flip-flop” mechanism in which the hypocretin system plays a key role (Saper et al., 2005). In this hypothesis, brain areas involved in sleep and wake regulation form a network with the characteristics of a bistable “flip-flop” switch, a term derived from electronics. Wake–active areas that produce acetylcholine and monoaminergic neurotransmitters (such as the locus coeruleus, dorsal raphe nucleus, and tuberomamillary nucleus) inhibit, but vice versa are also inhibited by the GABAergic

sleep–active ventrolateral preoptic area. This system leads to the evolutionary desirable situation in which transitions occur fast and intermediate situations are avoided. However, such a “flip-flop”switch is inherently instable and small disturbances will result in too frequent transitions. Hypocretin neurons have strong projections to the wake–active brain areas, such as the locus coeruleus, dorsal raphe nucleus, and tuberomamillary nucleus and others and form the external stabilizer for this switch. Without this stabilizing hypocretin signal, as is the case in narcolepsy type 1, transitions between sleep and wake occur too often (Fig. 21.2). A decline in hypocretin neurotransmission might play a role in reduced sleep quality in the elderly. Although normal hypocretin-1 levels in the CSF have been found in normal aging (Kanbayashi et al., 2002a,b), a lower number of hypocretin neurons has been described in adults (not elderly) compared with infants and children (Hunt et al., 2014). Alterations in hypocretin neurotransmission might play a role in sleep problems in the elderly. However, there is no direct proof for this, yet.

Other functions Because local injection of hypocretin-1 in several hypothalamic areas induced feeding in rodents (Dube et al., 1999), the hypocretin system was first thought to be mainly involved in food intake. However, prolonged hypocretin-1 administration does not alter 24-h food consumption or body weight in rats (Yamanaka et al., 1999). The hypocretin system does seem to play a major role in arousal and motivational behavior (Harris et al., 2005), which includes feeding behavior. Furthermore, the hypocretin system is involved in autonomic control (Grimaldi et al., 2013) and thermoregulation (Kuwaki, 2015).

PRIMARY SLEEP DISORDERS Soon after its discovery, the important role of the hypocretin system in the pathogenesis of primary sleep disorders was demonstrated. The hallmark disease developing as a result of the disruption of the hypocretin system is

THE OREXIN/HYPOCRETIN SYSTEM IN NEUROPSYCHIATRIC DISORDERS Locus Coeruleus Dorsal Raphe Nucleus Tuberomamillary Nucleus

345

Ventrolateral Preop c Nucleus

Wake

Sleep

hypocre n

disturbance

Wake

Sleep

Wake

Sleep

Fig. 21.2. Sleep is hypothesized to be mainly regulated by a “flip-flop” mechanism in which the hypocretin system plays a key role (Saper et al., 2005). Brain areas involved in sleep and wake regulation form a network with the characteristics of a bistable “flip-flop” switch. Wake–active areas that produce acetylcholine and monoaminergic neurotransmitters (such as the locus coeruleus, dorsal raphe nucleus, and tuberomamillary nucleus) inhibit but vice versa are also inhibited by the GABAergic sleep–active ventrolateral preoptic area. This system leads to the evolutionary desirable situation in which transitions occur fast and intermediate situations are avoided. However, such a “flip-flop” switch is inherently instable and small disturbances will result in too frequent transitions. Hypocretin neurons have strong projections to the wake–active brain areas and form the external stabilizer for this switch. Without this stabilizing hypocretin signal (as is the case in narcolepsy type 1 and possibly in the other disorders discussed in this chapter), transitions between sleep and wake occur too often. Redrawn by R. Fronczek based upon Saper CB, Scammell TE, Lu J (2005). Hypothalamic regulation of sleep and circadian rhythms. Nature 437: 1257–1263.

narcolepsy type 1. Idiopathic hypersomnia and Kleine– Levin syndrome, the other two primary hypersomnolence disorders, are considered to be nonhypocretinergic disorders (Fronczek et al., 2009) (Fig. 21.3).

Narcolepsy type 1 and 2 Narcolepsy is a sleep–wake disorder that can be subdivided in two types (American Academy of Sleep Medicine, 2014). Narcolepsy type 1 is characterized by five core symptoms: excessive daytime sleepiness, cataplexy, sleep paralysis, hypnagogic hallucinations, and disturbed night sleep. Cataplexy is a sudden episode of voluntary muscle weakness typically triggered by strong, mostly positive emotions, such as laughter or surprise. In narcolepsy type 2, cataplexy is typically absent. Shortly after the hypocretins were discovered, a mutation in the gene coding for the hypocretin receptor 2 was demonstrated to be the cause of canine narcolepsy (Lin et al., 1999). This finding was soon followed by the first reports in human narcolepsy, where hypocretinproducing neurons were found to be almost absent in the hypothalamus of deceased narcolepsy type 1 patients (Peyron et al., 2000; Thannickal et al., 2000).

The absence of hypocretin-1 in the cerebrospinal fluid of narcolepsy type 1 patients was subsequently shown (Nishino et al., 2000; Ripley et al., 2001). Hypocretin-1 measurement in the cerebrospinal fluid using a dedicated radioimmunoassay became a diagnostic tool that allows for confirming the clinical diagnosis of narcolepsy type 1 and is incorporated in the diagnostic criteria for the disease (American Academy of Sleep Medicine, 2014). A hypocretin-1 concentration below 110 pg/mL is regarded as hypocretin deficiency. In narcolepsy type 2, hypocretin-1 is not absent in the cerebrospinal fluid (Bassetti et al., 2003; Dauvilliers et al., 2003; Heier et al., 2007). However, a reduced number of hypocretin-producing neurons was reported in only the posterior hypothalamus in two hypothalami from narcolepsy type 2 patients (Thannickal et al., 2009). The hypothesis that the hypocretin system is partly affected in narcolepsy type 2 and thus not reflected by lowered hypocretin-1 concentrations in in vivo CSF compared with healthy controls (Mignot et al., 2002; Dauvilliers et al., 2003). As mentioned earlier, evidence on the role of hypocretin deficiency in the instability of wakefulness points to it as the loss of stabilization of the “flip-flop” switch

346

R. FRONCZEK ET AL.

Fig. 21.3. Overview of hypocretin-1 values of different patient populations in six publications mentioned in the main text (Nishino et al., 2000; Ripley et al., 2001; Mignot et al., 2002; Bassetti et al., 2003; Dauvilliers et al., 2003; Heier et al., 2007). (A) Narcolepsy type 1. (B) Narcolepsy type 2. (C) Idiopathic hypersomnia. (D) Healthy controls. (E) Patient with other neurological disorders. Percentages of values below the detection limit of the assay are depicted in red; the percentage below 110 pg/mL (but above the detection limit) is shown in green. The percentages of values that are below 200 pg/mL (but above 110 pg/mL) are depicted in blue. Drawn by M.S. Schinkelshoek.

between wake and sleep. However, it remains largely unknown how hypocretin deficiency in narcolepsy type 1 leads to the pathognomonic symptom of the disorder, cataplexy. Most hypotheses are derived from mouse or dog models and include a suggested role for the loss of

hypocretinergic activation of dopaminergic (Burgess et al., 2010), serotonergic (Hasegawa et al., 2014), noradrenergic (Wu et al., 1999), GABAergic (Black et al., 2014) pathways in addition to a disinhibition of prefrontal cortical and medullar processes (Siegel et al.,

THE OREXIN/HYPOCRETIN SYSTEM IN NEUROPSYCHIATRIC DISORDERS 1991; Vassalli et al., 2013) in the development of cataplexy in narcolepsy type 1. Developing causal treatment for narcolepsy has proven challenging. In canine narcolepsy, intravenous administration of hypocretins was shown to be ineffective in improving cataplexy, most probably due to poor blood–brain barrier penetration of hypocretins (Fujiki et al., 2003). The development of nonpeptide hypocretin receptor agonists is currently underway. One newly synthesized example of this group, YNT-185, was recently shown to ameliorate narcolepsy symptoms in a narcolepsy mouse model (Irukayama-Tomobe et al., 2017).

Idiopathic hypersomnia Idiopathic hypersomnia is a primary hypersomnolence disorder that is distinguished from narcolepsy type 2 by the absence of sleep-onset rapid-eye-movement (REM) episodes during the short naps of the multiple sleep latency test (MSLT). This is a diagnostic criterion for both narcolepsy type 1 and 2. Features of idiopathic hypersomnia include excessive daytime sleepiness despite normal, undisturbed sleep and long sleep time during the 24-h period with sleep inertia. Classification of idiopathic hypersomnia has changed considerably over the past decades, leaving a highly heterogeneous group of patients and scientific evidence that is hard to compare. The pathophysiology of idiopathic hypersomnia remains elusive. Given the fact that hypocretin-1 concentrations in the cerebrospinal fluid were not decreased compared with healthy controls (Mignot et al., 2002; Kanbayashi et al., 2002a,b, 2009a; Dauvilliers et al., 2003), the role of the hypocretin system in idiopathic hypersomnia seems limited. However, studies focusing on hypocretin in idiopathic hypersomnia are scarce.

Kleine–Levin syndrome Episodes of hypersomnia combined with behavioral or cognitive disturbances are typical for the Kleine–Levin syndrome, a rare disorder of adolescence and young adulthood that frequently resolves spontaneously over time (Arnulf et al., 2012). Hypocretin-1 measurements in the cerebrospinal fluid are not decreased in between episodes. During episodes, a decrease in hypocretin-1 is reported in several studies, even though concentrations below 110 pg/mL, the threshold for the diagnosis of narcolepsy type 1, are only seldomly reached (Dauvilliers et al., 2003; Podestá et al., 2006; Lopez et al., 2015; Wang et al., 2016; Usuda et al., 2018). It remains unclear whether the episodes can be regarded as the clinical correlate of a temporary decrease in hypocretin concentrations.

347

PRIMARY HEADACHE DISORDERS Hypocretin neurons project to areas important in pain processing, such as the periaqueductal gray and spinal dorsal horns (Peyron et al., 1998). Hypocretin knockout mice have a lowered pain threshold after peripheral local inflammation (Watanabe et al., 2005). Stimulation of the hypocretin receptors with hypocretin-1 has antinociceptive effects on trigeminal pain processing, which are negated by a hypocretin receptor 1 antagonist (Holland et al., 2005). The relation between hypocretin and pain has raised interest in the possible role of the hypocretin system in nociception in primary headache disorders (Holland, 2017; Razavi and Hosseinzadeh, 2017).

Cluster headache The extremely severe headache attacks that characterize cluster headache occur mainly during the night, linked to sleep (de Coo et al., 2019). Because of this intriguing relation with the clock and sleep, the hypothalamus is often considered to be the culprit in cluster headache since it contains the biological clock and hypocretin neurons (Naber et al., 2019). A link between a polymorphism in the hypocretin receptor 2 gene and cluster headache is debated, because this genetic association was only shown in one South European study, becoming weak in a meta-analysis of four different studies in which the association was not seen at first (Rainero et al., 2004; Sch€urks et al., 2007a,b; Weller et al., 2015). In vivo lumbar hypocretin-1 CSF levels were normal with a tendency toward higher levels in chronic cluster headache (Cevoli et al., 2011). Another study reports lowered hypocretin levels in both episodic and chronic cluster headache, independent of attack frequency (Barloese et al., 2015). If there is a role of hypocretin or its receptors in the development of cluster headache remains to be established. Interestingly, a case series describes beneficial effects of the narcolepsy medication sodium oxybate on nocturnal attacks in four cluster headache patients (Khatami et al., 2011).

Migraine Hypocretin-1 in vivo lumbar CSF concentrations have been studied in medication-overuse-headache and in chronic migraine, a disorder often accompanied by overuse of pain medication. In both patient groups, hypocretin-1 values were higher compared to controls. There was a correlation between hypocretin-1 levels and scores on a substance dependence questionnaire (Leeds Dependence Questionnaire), and with monthly drug intake (Sarchielli et al., 2007). These findings may point toward a relation between hypocretin and chronic pain (or headache) but may also be a reflection

R. FRONCZEK ET AL.

348

of reward mechanisms, drug use, and possibly addiction in these groups. Interestingly, migraine was reported to occur more frequently in patients with proven narcolepsy: 44% of female and 28% of narcolepsy patients fulfilled the criteria of migraine, with an onset of narcolepsy symptoms 12.3  11.4 SD years before migraine symptoms (Dahmen et al., 2003). This finding indirectly points to a possible contribution of the hypocretin system to the development of migraine in people with narcolepsy at later time in life than in the normal population. Of note, the human leukocyte antigen (HLA)-type associated with narcolepsy type 1 (HLA DQB1*0602) is not more prevalent in people with migraine (Coelho et al., 2007).

NEURODEGENERATIVE DISORDERS Sleep disturbances are frequently seen in a variety of neurodegenerative disorders. In various of those, the hypocretin system has been studied. In general, hypocretin neurons seem more vulnerable compared to other hypothalamic neurons, especially compared to the MCH neurons in the same region. In the various disorders studied, the hypocretin system does seem affected to different extents (Fronczek et al., 2009). Recurring questions are whether this loss of hypocretin neurosignaling is (partly) responsible for the sleep symptoms and whether narcolepsy medication would of beneficial.

Parkinson’s disease Sleep problems occur frequently in Parkinson’s disease (PD) and are reported to sometimes precede the characteristic motor symptoms of the disease (Arnulf et al., 2008). Nocturnal sleep is often fragmented with a reduced sleep efficiency (Wienecke et al., 2012) and hypnagogic hallucinations (Leu-Semenescu et al., 2011), while during the day, many patients experience excessive daytime sleepiness with sleep attacks (Arnulf, 2005). Shortened sleep latencies and sleep-onset rapid-eyemovement periods occur during the multiple sleep latency test (Arnulf, 2005; Wienecke et al., 2012). This has led to the hypothesis that the hypocretin system may be involved in the sleep disturbances in PD (Arnulf et al., 2008). In vivo studies on lumbar cerebrospinal fluid did not show altered hypocretin-1 levels (Ripley et al., 2001; Yasui et al., 2006; Compta et al., 2009; Poceta et al., 2009), even in patients with clearly described sleep disturbances (Overeem et al., 2002). In contrast, low or even absent levels were described in intraoperatively acquired ventricular CSF in patients with late-stage PD (Drouot et al., 2011). Two postmortem studies have shown the hypocretin system to be affected in PD. The total number of hypocretin-1 immunoreactive neurons was almost 50% decreased in one study

(Fig. 21.4), with a 40% reduction in the hypocretin-1 concentration in the prefrontal cortex and a 25% reduction in ventricular CSF levels. There was a correlation between ventricular hypocretin-1 CSF levels and the number of hypocretin-1 immunoreactive neurons. Lewy bodies, the pathophysiological hallmark of PD, were abundantly present in the perifornical region, sometimes localized in hypocretin neurons (Fronczek et al., 2007). Another postmortem study found a similar decrease in both hypocretin-1 and MCH immunoreactive neurons, related to disease severity (Thannickal et al., 2007). In rodent studies, a reduction of 15% of hypocretin-producing neurons does not lead to alterations in CSF hypocretin levels or sleep patterns, while a reduction of 50%–60% does lead to REM-sleep alterations (Gerashchenko et al., 2003). The question remains whether in human PD, the observed reduction in hypocretin-1 at least partly explains the sleep problems. The dopamine deficiency and resulting motor symptoms, certain medication (such as dopamine agonists), and psychiatric comorbidity may also be responsible (Baumann et al., 2008). Clinically, the hypnagogic hallucinations in PD are often of another nature than the ones in narcolepsy type 1 (LeuSemenescu et al., 2011). However, several studies point toward a role for hypocretin in the sleep problems seen in PD. Polymorphisms in the preprohypocretin gene are described in PD patients reporting sleep attacks (Rissling et al., 2005). In a prospective human study, there was a reduction in CSF hypocretin-1 level comparing late with early stage PD patients, and a correlation between objective sleepiness (quantified by an MSLT) and decrease of CSF hypocretin-1 level. In two PD patients, a repeated lumbar puncture showed a further reduction in CSF hypocretin-1 level. The authors conclude that despite an only partial loss of hypocretin-1 in PD, disturbed hypocretin signaling is likely to contribute to excessive daytime sleepiness in PD (Wienecke et al., 2012). However, other authors report no correlation between objective sleepiness (daytime sleep latency during an MSLT) and CSF hypocretin-1 levels (Drouot et al., 2011). Another study does report an association between lower CSF hypocretin-1 levels and disease duration in PD patients but no direct relation to the occurrence of sleep attacks (Asai et al., 2009). Of note, the functional meaning of hypocretin-1 in the CSF is not clear and may not have any role at all, leaving the brain by the CSF after having performed its function, showing only a reduction when the number of hypocretinproducing neurons is severely reduced (Fronczek et al., 2008). The question remains whether narcolepsy medication is beneficial in PD. A few trials showed a beneficial effect of a stimulant often used in the treatment of narcolepsy, modafinil, on subjective daytime sleepiness in PD (Rodrigues et al., 2016). Sodium oxybate, a compound

THE OREXIN/HYPOCRETIN SYSTEM IN NEUROPSYCHIATRIC DISORDERS

349

Fig. 21.4. Examples of hypocretin-1 cell bodies in the lateral hypothalamus of a control (A) and a PD patient (B). (C) Sample distribution of hypocretin neurons in the lateral hypothalamus for a control (left) and a PD patient (right). For each slide, the total number of hypocretin neurons is shown. The total number of cells is determined by calculating the total area under the curve. (D) A schematic sagittal view of the hypothalamus; the paraventricular nucleus and the mamillary bodies are indicated in dark. The lines depict the first and the last slide in which hypocretin cell bodies were present. From: Fronczek R, Baumann CR, Lammers GJ, Bassetti CL, Overeem S (2009). Hypocretin/orexin disturbances in neurological disorders. Sleep Med Rev 13: 9–22. Elsevier Journal.

that improves sleep fragmentation and cataplexy in narcolepsy type 1, improved nocturnal sleep and subjective daytime sleepiness in PD, both in an open study (Ondo et al., 2008) and in a randomized trial (B€ uchele et al., 2017).

Alzheimer’s disease Sleep problems and circadian disturbances are common in AD. Often these problems are the primary reason for institutionalization (der Lek et al., 2008). Spinal CSF hypocretin-1 levels in AD were conflictingly reported to be normal (Ripley et al., 2001; Slats et al., 2012; Schmidt et al., 2013; Deuschle et al., 2014), lowered (Wennstr€ om et al., 2012), or increased (Liguori et al., 2016, 2018; Gabelle et al., 2017). A relation between hypocretin-1 spinal CSF level has been found with

amyloid-b42 (Slats et al., 2012) and (p)Tau (Deuschle et al., 2014; Osorio et al., 2016). One study reports a correlation between CSF hypocretin-1 values and wake fragmentation (Friedman et al., 2007). Increased CSF hypocretin-1 levels in mild cognitive impairment due to AD were associated with rapid eye movement-sleep disruption and fragmentation (Liguori et al., 2016). In AD patients with neuropsychiatric symptoms, CSF hypocretin-1 levels were increased and had relation to sleep fragmentation (Liguori et al., 2018). One postmortem study has quantified the number of hypocretin-1 immunoreactive neurons in the lateral hypothalamus and measured the hypocretin-1 concentration in ventricular CSF. A 40% reduction in the number of hypocretinexpressing neurons was found with 15% lowered ventricular CSF levels. Two patients with clearly documented

350

R. FRONCZEK ET AL.

sleep–wake problems had the lowest CSF concentrations (Fronczek et al., 2011). As in PD, the question is whether this reduction in hypocretin signaling is responsible for the sleep problems in AD. Surprisingly, in a study of 48 AD patients, higher in vivo lumbar CSF hypocretin-1 levels were reported in moderate to severe AD, accompanied by parallel sleep deterioration and cognitive decline (Liguori et al., 2014). Despite the clear relationship between sleep problems and cognitive functioning, a direct and causal role of the hypocretin-1 system was not proven. In an amyloid precursor protein/Presenilin 1 (PS1) transgenic hypocretin knockout mice model, hypocretin signaling was reinstated by a lentiviral vector causing focal overexpression of hypocretin driven by an ubiquitin promoter. This led to increased wakefulness and an increase in the amount of Ab pathology in the brain (Roh et al., 2014). The main consistent finding seems to be a relationship between AD pathology and sleep disturbances. This has led to the hypothesis that sleep plays an important role in the clearance of amyloid-b from the brain (ShokriKojori et al., 2018), possibly through the glymphatic pathway (Xie et al., 2013). There is a paucity of evidence-based data regarding the treatment of sleep problems in AD. Treatment is often circadian rhythm related to melatonin or bright light therapy (der Lek et al., 2008; Urrestarazu and Iriarte, 2016). In relation to the emerging role of sleep in AD pathology, hypocretin-based therapies might be of value. In a case series of 4 AD patients with nocturnal delirium, administration of a hypocretin-antagonist immediately led to improvement of nocturnal sleep (Hanazawa and Kamijo, 2019).

Dementia with lewy bodies Lumbar in vivo hypocretin-1 CSF levels were normal in two studies in dementia with lewy bodies (DLB) (Baumann et al., 2004; Yasui et al., 2006) and reduced comom pared to AD and controls in another study (Wennstr€ et al., 2012). In a postmortem study, hypocretin-1 levels were reduced in the neocortex of 21 postmortem DLB brains compared to control brains (n ¼ 3), and AD (n ¼ 19) (Lessig et al., 2010). In line with this, a reduction in the total number of hypocretin-1-producing neurons was found in 15 DLB brains compared to AD (n ¼ 14) and controls (n ¼ 7), with a negative correlation between the number of neurons and the neurofibrillary stage (Kasanuki et al., 2014). These findings suggest that in DLB the hypocretin system is affected to an even greater extent than in AD. The sleep problems in DLB often resemble the ones in PD, such as daytime sleepiness with sleep attacks and REM-sleep behavior disorder. As in PD, it is an intriguing question whether damage to the hypocretin system is (partly) responsible for these symptoms, and whether narcolepsy medication might be beneficial.

Progressive supranuclear palsy A case report described a 74-year-old woman with progressive supranuclear palsy (PSP) who also suffered from excessive daytime sleepiness, with abnormally short latencies during a multiple sleep latency test and an absence of hypocretin-1 in the CSF (Hattori et al., 2003). However, in a larger series of 16 patients with PSP, in vivo lumbar CSF hypocretin-1 levels were normal (Yasui et al., 2006). There have been no postmortem studies yet.

Huntington’s disease Patients with Huntington’s disease (HD) often suffer from sleep disturbances (Aziz et al., 2007). The hypocretin system was hypothesized to be involved when it was found that the most commonly used HD rodent model, the R6/2 mice model, shows a progressive loss to up to 70% of hypocretin-expressing neurons. In the same study, atrophy and a reduction density of hypocretin-1 immunoreactive neurons were found in several postmortem hypothalamic sections from 4 HD patients (Petersen et al., 2004). Another postmortem study quantified the total number of hypocretin or MCH-expressing neurons in HD patients and controls and found a 30% decrease in the number of hypocretin neurons with a normal number of MCH neurons. Furthermore, the hypocretin-1 concentration in the prefrontal cortex was 33% lower than in controls, with normal ventricular postmortem CSF levels. Huntington aggregates were found in several hypothalamic areas, but not specifically in the hypocretin region (Aziz et al., 2008). Several studies have reported normal in vivo spinal lumbar CSF hypocretin-1 levels in HD (Ripley et al., 2001; Gaus et al., 2005; Meier et al., 2005; Baumann et al., 2006). CSF hypocretin-1 does thus not seem to be a suitable in vivo biomarker for HD (Bj€orkqvist et al., 2006). Whether the hypocretin system is involved in the sleep problems in HD also remains to be determined.

Multiple system atrophy Although multiple system atrophy (MSA) is mainly characterized by autonomic dysfunction, cerebellar features, and Parkinsonism, sleep problems can also be present in the form of sleep-related breathing disorders, REM-sleep behavior disorder, and excessive daytime sleepiness (Ghorayeb et al., 2005). Especially, the daytime sleep attacks could be related to the hypocretin system. In a postmortem study, a 70% reduction in the number of hypocretin-producing neurons was found with abundant alpha-synuclein-containing cytoplasmatic inclusions (Benarroch et al., 2007). In line with this, another postmortem MSA study found reduced hypocretin-1 immunoreactivity in the nucleus basalis of Meynert region

THE OREXIN/HYPOCRETIN SYSTEM IN NEUROPSYCHIATRIC DISORDERS

351

(Mishima et al., 2017). Lumbar in vivo CSF hypocretin-1 levels were normal in two studies (Martinez-Rodriguez et al., 2007; Abdo et al., 2008).

exclusive inflammation and tissue damage in the hypothalamus, with CD8 + cells probably being responsible for the tissue damage (Dauvilliers et al., 2013).

IMMUNE-MEDIATED DISORDERS

Guillain–Barre syndrome

Although narcolepsy type 1 probably is an autoimmune disorder in which hypocretin cells are selectively destroyed, there are multiple other immune-mediated disorders in which hypocretin transmission may be impaired. They are listed below. In most of these disorders, there is evidence for immune-mediated destruction of hypocretin cells, but in contrast to narcolepsy type 1, the destruction is never limited to the hypocretin cells nor is the vast majority of cells destroyed. In some disorders, such as Guillain–Barre syndrome it is more likely that there is a functional impairment.

There are several case series showing hypocretin-1 deficiency in the CSF in severe Asian, but not in Caucasian Guillain–Barre cases (Ripley et al., 2001; Nishino et al., 2003). All subjects with low hypocretin were Japanese, had tetraplegia, bulbar symptoms, and/or respiratory failure. Follow-up data have never been published. There is no clue why and how hypocretin deficiency may develop in these Guillain–Barre syndrome patients.

Multiple sclerosis/neuromyelitis optica There are no postmortem brain studies focusing on hypocretin cells in multiple sclerosis (MS). However, there are several case studies and series that show that narcolepsy type 1 and MS may co-occur. However, interestingly, reports in the literature are scarcer than expected based on the presumed prevalence of both disorders (Kallweit et al., 2018). Symptomatic cases have also been described, showing MS lesions in the hypothalamus either uni- or bilateral. All these patients developed excessive daytime sleepiness, in some cases with established hypocretin-1 deficiency in the CSF (Kanbayashi et al., 2009b). Surprisingly, cataplexy has never been described in such cases (Kallweit et al., 2018). Based on the number of published cases, symptomatic bilateral hypothalamic lesions seem to occur more often in neuromyelitis optica (NMO) than in MS. Recovery of symptoms and hypocretin-1 deficiency has been reported in symptomatic MS and NMO cases that were treated with immunosuppressants (Kanbayashi et al., 2011).

Anti-Ma2-encephalitis Anti-Ma2 encephalitis is a very rare disorder. Nevertheless, a small case series has been published focusing on hypocretin-1 measurements (Overeem et al., 2004). Four out of 6 patients had a low hypocretin-1 concentration. They all had a complaint of daytime sleepiness, but detailed clinical information was not provided. There is one additional case report with detailed clinical and postmortem histological/immunological information (Dauvilliers et al., 2013). The subject described showed all symptoms of narcolepsy type 1 including cataplexy among other diencephalic symptoms. The hypocretin-1 concentration was undetectable. He died 4 months after disease onset. Postmortem brain study revealed almost

NEUROMUSCULAR DISORDERS: MYOTONIC DYSTROPHY A relatively high percentage of myotonic dystrophy type 1 patients complain about excessive daytime sleepiness, usually without a convincing explanation. In a minority of these subjects, hypocretin-1 deficiency has been reported (Omori et al., 2018). Cataplexy has not been reported, and there are no data on postmortem brain findings.

TRAUMATIC BRAIN INJURY Sleep problems are a common occurrence after traumatic brain injury (TBI). These can vary from insomnia and fatigue to hypersomnia with excessive daytime sleepiness (Imbach et al., 2015). Interestingly, hypothalamic damage after TBI was described already in 1971 (Crompton, 1971). In a series of 44 consecutive patients with TBI, hypocretin-1 was measured in the CSF within 4 h of the head trauma. Surprisingly, levels were low or even undetectable in 20% of cases, returning to normal levels in most cases 6 months later (Baumann et al., 2005). However, in 19% of cases, hypocretin-1 levels remained low, especially in subjects that still suffered from daytime sleepiness (Imbach et al., 2015). In a postmortem study in 4 TBI patients, a 30% reduction in the total number of hypocretin-1 immunoreactive neurons was found (Baumann et al., 2009). Hypocretinproducing neurons seem vulnerable to TBI especially the first days after the trauma. Impaired hypocretin signaling could partly explain the daytime sleepiness complaints patients can have after TBI. It remains an interesting question whether the reduction in hypocretin function after TBI can lead to a full narcolepsy phenotype. In a series of 186 TBI patients, narcolepsy-like multiple sleep latency test findings were described in 5 patients (Guilleminault et al., 2000), without any information about their hypocretin status. In another series of 65 patients with TBI, there were 2 patients with short MSLT sleep latencies, with a low hypocretin-1 CSF

352

R. FRONCZEK ET AL.

concentration in one of them (Imbach et al., 2015). In one rodent TBI model, two different kinds of head injury did not alter the number of hypocretin-1 immunoreactive neurons (Vu et al., 2018). In contrast, two rodent TBI studies did show a reduction in the number of hypocretin neurons after head injury, with altered sleep patterns in wild-type mice, but no effects on sleep in a hypocretin knockout model in one study (Thomasy et al., 2017; Thomasy and Opp, 2019). To conclude, TBI may functionally impair hypocretin neurons, which might very well play a role in daytime sleepiness symptoms.

PSYCHIATRIC DISORDERS Schizophrenia Comorbidity of narcolepsy and schizophrenia has been reported (Canellas et al., 2014; Huang et al., 2014; Plazzi et al., 2015). However, misdiagnosis may occur: the presence of frequent and severe hypnagogic hallucinations in narcolepsy may lead to misdiagnoses of schizophrenia in narcoleptic subjects (Howland, 1997; Talih, 2011). In both disorders, patients experience hallucinations, although in contrast to the “real” hallucinations in schizophrenia, the hypnagogic hallucinations that are typical for narcolepsy can be distinguished from reality by patients (Fortuyn et al., 2009; Leu-Semenescu et al., 2011). Moreover, narcoleptic patients report their hypnagogic hallucinations to be multisensory “holistic” rather than the predominantly verbal–auditory sensory mode in schizophrenia (Fortuyn et al., 2009). Certain variants in the hypocretin receptor 1 gene were found to contribute to schizophrenia symptoms in a Japanese population (Meerabux et al., 2005; Fukunaka et al., 2007). Lower in vivo CSF hypocretin-1 levels were reported in patients with schizophrenia treated with haloperidol compared to unmedicated patients and control subjects, although there were no differences in CSF hypocretin-1 level between unmedicated schizophrenic patients and controls (Dalal et al., 2003). It is hypothesized that neuroinflammation might contribute to the development of schizophrenia (Weickert and Weickert, 2016). Genome-wide association studies have robustly identified markers in the major histocompatibility complex (Stefansson et al., 2009; Yue et al., 2011) linked to schizophrenia, especially showing an association with specific human leukocyte antigen alleles and complement component 4 (Schizophrenia Working Group of the Psychiatric Genomics Consortium et al., 2016). Positron emission tomography (PET) studies show activated microglia in the brain of patients with schizophrenia and of people with an ultrahigh risk of psychosis (Doorduin et al., 2009; Bloomfield et al., 2015). In addition, inflammatory markers such as interleukin, transforming growth factor, or tumor necrosis factor-a are

present in the peripheral blood of people at risk to develop psychosis and during the first episode of psychosis (de Witte et al., 2014; Perkins et al., 2014). In postmortem brain tissue of schizophrenia patients, higher levels of cytokine and microglia markers are found (Fillman et al., 2013; Trepanier et al., 2016). This process of neuroinflammation might also target hypocretin neurons, which seem to be especially vulnerable (Peyron et al., 2000; Thannickal et al., 2000). Whether impaired hypocretin transmission plays a relevant role in the pathophysiology and symptomatology of schizophrenia deserves further study.

Major depression and bipolar disorder In a postmortem study, the total amount of hypothalamic hypocretin-1 staining was increased in female but not in male depressive patients (Fig. 21.5) (Lu et al., 2017). It should be noted that these sex-dependent changes in the hypocretin system were not systematically investigated in the following studies. A genome-wide functional pathway analysis pointed out that the hypocretin receptor 2 gene may be involved in hypersomnia symptoms during a major depressive episode of bipolar disorder (Cho et al., 2015). This is in line with a recent clinical 1b phase trial in which a selective hypocretin receptor 2 antagonist (seltorexant) had positive effects on core depressive symptoms with a trend toward improved self-reported sleep quality (Recourt et al., 2019). Studies investigating in vivo lumbar CSF hypocretin-1 levels in depressive disorders report contradictory findings. Reduced diurnal variation in CSF hypocretin-1 levels was found in a group of major depression and bipolar patients compared to controls, whereas the basal 24 h levels did not differ significantly between the groups (Salomon et al., 2003). Reduced mean in vivo lumbar CSF hypocretin-1 levels were found in suicidal patients with major depressive disorder (Brundin et al., 2007). Interestingly, the same group reported that CSF hypocretin-1 increases the first year after a suicide attempt. This increase was accompanied by an improvement in suicide assessment scale scores (Brundin et al., 2009). It is therefore intriguing to consider a possible link between the hypocretin system and its sex-dependent changes in major depression, bipolar disorder, and suicide, but the evidence for this is still scarce.

PRADER–WILLI SYNDROME Sleep problems are a common feature of Prader–Willi Syndrome (PWS). Most often, this entails sleep disorder breathing. However, there are frequent reports of excessive daytime sleepiness and even the occurrence of cataplexy (Tobias et al., 2002). Lowered (but not absent) in vivo CSF hypocretin-1 levels have been described

THE OREXIN/HYPOCRETIN SYSTEM IN NEUROPSYCHIATRIC DISORDERS

353

Fig. 21.5. Immunocytochemistry of hypocretin-1. (A, B): representative distribution patterns of the integrated optical density (IOD) of hypocretin-1 immunoreactivity from rostral to caudal along the hypothalamus. (A) A female depression patient, (B) a male depression patient. (C–F): representative hypocretin-1 staining in the hypothalamus: (C) a female control subject, (D) a female depression patient, (E) a male control subject, and (F) a male depression patient. Bar ¼ 0.025 mm. Lu J, Zhao J, Balesar R, Fronczek R, Zhu Q-B, Wu X-Y, Hu S-H, Bao A-M, Swaab DF (2017). Sexually dimorphic changes of hypocretin (orexin) in depression. Ebiomedicine 18: 311–319 with permission. (Open access journal).

in cases with documented daytime sleepiness in three studies (Mignot et al., 2002; Nevsimalova et al., 2005; Omokawa et al., 2016). Although hypocretin-1 measurements in serum are not convincingly reliable yet (Nishino, 2005; Sakai et al., 2019), a study has reported increased hypocretin-1 levels in the unextracted serum of 23 children diagnosed with PWS (Manzardo et al., 2016). Only one postmortem study has been performed in which the total number of hypocretin-1 immunoreactive neurons was determined in 8 PWS adults and 3 infants, compared to 11 controls. It was not clear whether these cases had suffered from cataplexy, but the authors conclude that on a group level, the hypocretin system does not seem to be affected in PWS (Fronczek et al., 2005). However, this might be different for individual cases with clear-cut cataplexy.

CONCLUSIONS It is clear that the disorder most strongly linked to the hypocretin system is the primary sleep disorder narcolepsy type 1 caused by a lack of hypocretin signaling, which is most likely due to an autoimmune process targeting the hypocretin-producing neurons. However, the hypocretin system may also be affected, but to a lesser extent and less specifically, in various other neurological disorders. Examples are neurodegenerative diseases such as Alzheimer’s, Huntington’s and Parkinson’s disease, immune-mediated disorders such as MS/NMO, anti-Ma2 encephalitis, and genetic disorders such as DM1 and PWS. A partial hypocretin deficiency may at least partly explain the sleep features of these disorders.

354

R. FRONCZEK ET AL.

REFERENCES Abdo WF, Bloem BR, Kremer HPH et al. (2008). CSF hypocretin-1 levels are normal in multiple-system atrophy. Parkinsonism Relat Disord 14: 342–344. Adamantidis A, de Lecea L (2008). Sleep and metabolism: shared circuits, new connections. Trends Endocrinol Metab 19: 362–370. American Academy of Sleep Medicine (2014). International classification of sleep disorders—3e edition, American Academy of Sleep Medicine, Darien, IL. Arnulf I (2005). Excessive daytime sleepiness in parkinsonism. Sleep Med Rev 9: 185–200. Arnulf I, Leu S, Oudiette D (2008). Abnormal sleep and sleepiness in Parkinson’s disease. Curr Opin Neurol 24: 472–477. Arnulf I, Rico TJ, Mignot E (2012). Diagnosis, disease course, and management of patients with Kleine-Levin syndrome. Lancet Neurol 11: 918–928. Asai H, Hirano M, Furiya Y et al. (2009). Cerebrospinal fluidorexin levels and sleep attacks in four patients with Parkinson’s disease. Clin Neurol Neurosurg 111: 341–344. Aziz NA, Swaab DF, Pijl H et al. (2007). Hypothalamic dysfunction and neuroendocrine and metabolic alterations in Huntington disease: clinical consequences and therapeutic implications. Rev Neurosci 18: 223–252. Aziz A, Fronczek R, Maat-Schieman M et al. (2008). Hypocretin and melanin-concentrating hormone in patients with Huntington disease. Brain Pathol 18: 474–483. Barloese M, Jennum P, Lund N et al. (2015). Reduced CSF hypocretin-1 levels are associated with cluster headache. Cephalalgia 35: 869–876. Bassetti C, Gugger M, Bischof M et al. (2003). The narcoleptic borderland: a multimodal diagnostic approach including cerebrospinal fluid levels of hypocretin-1 (orexin A). Sleep Med 4: 7–12. Baumann CR, Dauvilliers Y, Mignot E et al. (2004). normal csf hypocretin-1 (orexin A) levels in dementia with lewy bodies associated with excessive daytime sleepiness. Eur Neurol 52: 73–76. Baumann CR, Stocker R, Imhof H-G et al. (2005). Hypocretin-1 (orexin A) deficiency in acute traumatic brain injury. Neurology 65: 147–149. Baumann CR, Hersberger M, Bassetti CL (2006). Hypocretin-1 (orexin A) levels are normal in Huntington’s disease. J Neurol 253: 1232–1233. Baumann CR, Scammell TE, Bassetti CL (2008). Parkinson’s disease, sleepiness and hypocretin/orexin. Brain 131: e91. Baumann CR, Bassetti CL, Valko PO et al. (2009). Loss of hypocretin (orexin) neurons with traumatic brain injury. Ann Neurol 66: 555–559. Benarroch EE, Schmeichel AM, Sandroni P et al. (2007). Involvement of hypocretin neurons in multiple system atrophy. Acta Neuropathol 113: 75–80. ˚ , Nielsen J et al. (2006). Cerebrospinal Bj€ orkqvist M, Petersen A fluid levels of orexin-A are not a clinically useful biomarker for Huntington disease. Clin Genet 70: 78–79. Black SW, Morairty SR, Chen T-M et al. (2014). GABAB agonism promotes sleep and reduces cataplexy in murine narcolepsy. J Neurosci 34: 6485–6494.

Bloomfield PS, Selvaraj S, Veronese M et al. (2015). Microglial activity in people at ultra high risk of psychosis and in schizophrenia: an [(11)C]PBR28 PET brain imaging study. Am J Psychiatry 173: 44–52. Blouin AM, Thannickal TC, Worley PF et al. (2005). Narp immunostaining of human hypocretin (orexin) neurons: loss in narcolepsy. Neurology 65: 1189–1192. ˚ , Bj€ Brundin L, Petersen A orkqvist M et al. (2007). Orexin and psychiatric symptoms in suicide attempters. J Affect Disord 100: 259–263. Brundin L, Bj€ orkqvist M, Tr€askman-Bendz L et al. (2009). Increased orexin levels in the cerebrospinal fluid the first year after a suicide attempt. J Affect Disord 113: 179–182. B€ uchele F, Hackius M, Schreglmann SR et al. (2017). Sodium oxybate for excessive daytime sleepiness and sleep disturbance in parkinson disease: a randomized clinical trial. JAMA Neurol 75 (114): 118. Burgess CR, Tse G, Gillis L et al. (2010). Dopaminergic regulation of sleep and cataplexy in a murine model of narcolepsy. Sleep 33: 1295–1304. Canellas F, Lin L, Julia` MR et al. (2014). Dual cases of type 1 narcolepsy with schizophrenia and other psychotic disorders. J Clin Sleep Med 10: 1011–1018. Cevoli S, Pizza F, Grimaldi D et al. (2011). Cerebrospinal fluid hypocretin-1 levels during the active period of cluster headache. Cephalalgia 31: 973–976. Cho C-H, Lee H-J, Woo HG et al. (2015). CDH13 and HCRTR2 may be associated with hypersomnia symptom of bipolar depression: a genome-wide functional enrichment pathway analysis. Psychiatry Investig 12: 402–407. Coelho FMS, Pradella-Hallinan M, Abud PC et al. (2007). Prevalence of HLA DQB1*0602 allele in patients with migraine. Arq Neuropsiquiatr 65: 1123–1125. Compta Y, Santamaria J, Ratti L et al. (2009). Cerebrospinal hypocretin, daytime sleepiness and sleep architecture in Parkinson’s disease dementia. Brain 132: 3308–3317. Crocker A, Espana RA, Papadopoulou M et al. (2005). Concomitant loss of dynorphin, NARP, and orexin in narcolepsy. Neurology 65: 1184–1188. Crompton MR (1971). Hypothalamic lesions following closed head injury. Brain 94: 165–172. Dahmen N, Kasten M, Wieczorek S et al. (2003). Increased frequency of migraine in narcoleptic patients: a confirmatory study. Cephalalgia 23: 14–19. Dalal MA, Schuld A, Pollm€acher T (2003). Lower CSF orexin A (hypocretin-1) levels in patients with schizophrenia treated with haloperidol compared to unmedicated subjects. Mol Psychiatry 8: 836–837. https://doi.org/10.1038/sj.mp. 4001363. PMID:14515133. Dauvilliers Y, Baumann C, Carlander B et al. (2003). CSF hypocretin-1 levels in narcolepsy, Kleine-Levin syndrome, and other hypersomnias and neurological conditions. J Neurol Neurosurg Psychiatry 74: 1667. Dauvilliers Y, Bauer J, Rigau V et al. (2013). Hypothalamic immunopathology in anti-ma–associated diencephalitis with narcolepsy-cataplexy. JAMA Neurol 70: 1305–1310. de Coo IF, van Oosterhout WPJ, Wilbrink LA et al. (2019). Chronobiology and sleep in cluster headache. Headache 59: 1032–1041.

THE OREXIN/HYPOCRETIN SYSTEM IN NEUROPSYCHIATRIC DISORDERS de Lecea L, Kilduff TS, Peyron C et al. (1998). The hypocretins: hypothalamus-specific peptides with neuroexcitatory activity. Proc Natl Acad Sci U S A 95: 322–327. de Witte L, Tomasik J, Schwarz E et al. (2014). Cytokine alterations in first-episode schizophrenia patients before and after antipsychotic treatment. Schizophr Res 154: 23–29. der Lek RFR, Swaab DF, Twisk J et al. (2008). Effect of bright light and melatonin on cognitive and noncognitive function in elderly residents of group care facilities. JAMA 299: 2642. Deuschle M, Schilling C, Leweke FM et al. (2014). Hypocretin in cerebrospinal fluid is positively correlated with Tau and pTau. Neurosci Lett 561: 41–45. Doorduin J, de Vries EFJ, Willemsen ATM et al. (2009). Neuroinflammation in schizophrenia-related psychosis: a PET study. J Nucl Med 50: 1801–1807. Drouot X, Moutereau S, Lefaucheur JP et al. (2011). Low level of ventricular CSF orexin-A is not associated with objective sleepiness in PD. Sleep Med 12: 936–937. Dube MG, Kalra SP, Kalra PS (1999). Food intake elicited by central administration of orexins/hypocretins: identification of hypothalamic sites of action1This study was presented in part at the 28th annual meeting of the Society for Neuroscience, Los Angeles, CA, in November 1998.1. Brain Res 842: 473–477. Fillman SG, Cloonan N, Catts VS et al. (2013). Increased inflammatory markers identified in the dorsolateral prefrontal cortex of individuals with schizophrenia. Mol Psychiatry 18: 206–214. Fortuyn HAD, Lappenschaar GA, Nienhuis FJ et al. (2009). Psychotic symptoms in narcolepsy: phenomenology and a comparison with schizophrenia. Gen Hosp Psychiatry 31: 146–154. Friedman LF, Zeitzer JM, Lin L et al. (2007). In Alzheimer disease, increased wake fragmentation found in those with lower hypocretin-1. Neurology 68: 793–794. Fronczek R, Lammers GJ, Balesar R et al. (2005). The number of hypothalamic hypocretin (orexin) neurons is not affected in Prader-Willi syndrome. J Clin Endocrinol Metab 90: 5466–5470. Fronczek R, Overeem S, Lee SYY et al. (2007). Hypocretin (orexin) loss in Parkinson’s disease. Brain 130: 1577–1585. Fronczek R, Overeem S, Lee SYY et al. (2008). Hypocretin (orexin) loss and sleep disturbances in Parkinson’s Disease. Brain 131: e88. Fronczek R, Baumann CR, Lammers GJ et al. (2009). Hypocretin/orexin disturbances in neurological disorders. Sleep Med Rev 13: 9–22. Fronczek R, van Geest S, Fr€olich M et al. (2011). Hypocretin (orexin) loss in Alzheimer’s disease. Neurobiol Aging 33: 1642–1650. Fujiki N, Yoshida Y, Ripley B et al. (2003). Effects of IV and ICV hypocretin-1 (orexin A) in hypocretin receptor-2 gene mutated narcoleptic dogs and IV hypocretin-1 replacement therapy in a hypocretin-ligand-deficient narcoleptic dog. Sleep 26: 953–959. Fukunaka Y, Shinkai T, Hwang R et al. (2007). The orexin 1 receptor (HCRTR1) gene as a susceptibility gene

355

contributing to polydipsia-hyponatremia in schizophrenia. NeuroMolecular Med 9: 292–297. Gabelle A, Jaussent I, Hirtz C et al. (2017). Cerebrospinal fluid levels of orexin-A and histamine, and sleep profile within the Alzheimer process. Neurobiol Aging 53: 59–66. Gaus SE, Lin L, Mignot E (2005). CSF hypocretin levels are normal in Huntington’s disease patients. Sleep 28: 1607–1608. Gerashchenko D, Murillo-Rodriguez E, Lin L et al. (2003). Relationship between CSF hypocretin levels and hypocretin neuronal loss. Exp Neurol 184: 1010–1016. Ghorayeb I, Bioulac B, Tison F (2005). Sleep disorders in multiple system atrophy. J Neural Transm 112: 1669–1675. Grimaldi D, Silvani A, Benarroch EE et al. (2013). Orexin/ hypocretin system and autonomic control: new insights and clinical correlations. Neurology 82: 271–278. Guilleminault C, Yuen KM, Gulevich MG et al. (2000). Hypersomnia after head-neck trauma: a medicolegal dilemma. Neurology 54: 653. Hanazawa T, Kamijo Y (2019). Effect of suvorexant on nocturnal delirium in elderly patients with Alzheimer’s disease: a case-series study. Clin Psychopharmacol Neurosci 17: 547–550. Harris GC, Wimmer M, Aston-Jones G (2005). A role for lateral hypothalamic orexin neurons in reward seeking. Nature 437: 556–559. Hasegawa E, Yanagisawa M, Sakurai T et al. (2014). Orexin neurons suppress narcolepsy via 2 distinct efferent pathways. J Clin Invest 124: 604–616. Hattori Y, Hattori T, Mukai E et al. (2003). Excessive daytime sleepiness and low CSF orexin-A/hypocretin-I levels in a patient with probable progressive supranuclear palsy. No To Shinkei Brain Nerve 55: 1053–1056 (in Japanese), PMID: 14870576. Heier MS, Evsiukova T, Vilming S et al. (2007). CSF hypocretin-1 levels and clinical profiles in narcolepsy and idiopathic CNS hypersomnia in Norway. Sleep 30: 969–973. Holland PR (2017). Biology of neuropeptides: orexinergic involvement in primary headache disorders. Headache 57 (Suppl. 2): 76–88. Holland PR, Akerman S, Goadsby PJ (2005). Orexin 1 receptor activation attenuates neurogenic dural vasodilation in an animal model of trigeminovascular nociception. J Pharmacol Exp Ther 315: 1380–1385. Howland RH (1997). Sleep-onset rapid eye movement periods in neuropsychiatric disorders: implications for the pathophysiology of psychosis. J Nerv Ment Dis 185: 730–738. Huang Y-S, Guilleminault C, Chen C-H et al. (2014). Narcolepsy–cataplexy and schizophrenia in adolescents. Sleep Med 15: 15–22. Hunt NJ, Rodriguez ML, Waters KA et al. (2014). Changes in orexin (hypocretin) neuronal expression with normal aging in the human hypothalamus. Neurobiol Aging 36: 292–300. Imbach LL, Valko PO, Li T et al. (2015). Increased sleep need and daytime sleepiness 6 months after traumatic brain injury: a prospective controlled clinical trial. Brain 138: 726–735.

356

R. FRONCZEK ET AL.

Irukayama-Tomobe Y, Ogawa Y, Tominaga H et al. (2017). Nonpeptide orexin type-2 receptor agonist ameliorates narcolepsy-cataplexy symptoms in mouse models. Proc Natl Acad Sci U S A 114: 5731–5736. Kallweit U, Bassetti CLA, Oberholzer M et al. (2018). Coexisting narcolepsy (with and without cataplexy) and multiple sclerosis. J Neurol 265: 2071–2078. Kanbayashi T, Inoue Y, Chiba S et al. (2002a). CSF hypocretin-1 (orexin-A) concentrations in narcolepsy with and without cataplexy and idiopathic hypersomnia. J Sleep Res 11: 91–93. Kanbayashi T, Yano T, Ishiguro H et al. (2002b). Hypocretin-1 (Orexin-A) levels in human lumbar CSF in different age groups: infants to elderly persons. Sleep 25: 337–339. Kanbayashi T, Kodama T, Kondo H et al. (2009a). CSF histamine contents in narcolepsy, idiopathic hypersomnia and obstructive sleep apnea syndrome. Sleep 32: 181–187. Kanbayashi T, Shimohata T, Nakashima I et al. (2009b). Symptomatic narcolepsy in patients with neuromyelitis optica and multiple sclerosis: new neurochemical and immunological implications. Arch Neurol 66: 1563–1566. Kanbayashi T, Sagawa Y, Takemura F et al. (2011). The pathophysiologic basis of secondary narcolepsy and hypersomnia. Curr Neurol Neurosci 11: 235–241. Kasanuki K, Iseki E, Kondo D et al. (2014). Neuropathological investigation of hypocretin expression in brains of dementia with Lewy bodies. Neurosci Lett 569: 68–73. Khatami R, Tartarotti S, Siccoli MM et al. (2011). Long-term efficacy of sodium oxybate in 4 patients with chronic cluster headache. Neurology 77: 67–70. Kuwaki T (2015). Thermoregulation under pressure: a role for orexin neurons. Temperature 2: 379–391. Lessig S, Ubhi K, Galasko D et al. (2010). Reduced hypocretin (orexin) levels in dementia with Lewy bodies. Neuroreport 21: 756–760. Leu-Semenescu S, Cock VCD, Masson VDL et al. (2011). Hallucinations in narcolepsy with and without cataplexy: contrasts with Parkinson’s disease. Sleep Med 12: 497–504. Liguori C, Romigi A, Nuccetelli M et al. (2014). Orexinergic system dysregulation, sleep impairment, and cognitive decline in Alzheimer disease. JAMA Neurol 71: 1498–1505. Liguori C, Nuccetelli M, Izzi F et al. (2016). Rapid eye movement sleep disruption and sleep fragmentation are associated with increased orexin-A cerebrospinal-fluid levels in mild cognitive impairment due to Alzheimer’s disease. Neurobiol Aging 40: 120–126. Liguori C, Mercuri NB, Nuccetelli M et al. (2018). Cerebrospinal fluid orexin levels and nocturnal sleep disruption in Alzheimer’s disease patients showing neuropsychiatric symptoms. J Alzheimer’s Dis 66: 993–999. https:// doi.org/10.3233/JAD-180769. PMID:30372684. Lin L, Faraco J, Li R et al. (1999). The sleep disorder canine narcolepsy is caused by a mutation in the hypocretin (orexin) receptor 2 gene. Cell 98: 365–376. Lopez R, Barateau L, Chenini S et al. (2015). Preliminary results on CSF biomarkers for hypothalamic dysfunction in Kleine–Levin syndrome. Sleep Med 16: 194–196.

Lu J, Zhao J, Balesar R et al. (2017). Sexually dimorphic changes of Hypocretin (orexin) in depression. Ebiomedicine 18: 311–319. Luppi P-H, Peyron C, Fort P (2013). Role of MCH neurons in paradoxical (REM) sleep control. Sleep 36: 1775–1776. Manzardo AM, Johnson L, Miller JL et al. (2016). Higher plasma orexin a levels in children with Prader-Willi syndrome compared with healthy unrelated sibling controls. Am J Med Genet A 170: 2328–2333. Martinez-Rodriguez JE, Seppi K, Cardozo A et al. (2007). Cerebrospinal fluid hypocretin-1 levels in multiple system atrophy. Mov Disord 22: 1822–1824. Meerabux J, Iwayama Y, Sakurai T et al. (2005). Association of an orexin 1 receptor 408Val variant with polydipsia– hyponatremia in schizophrenic subjects. Biol Psychiatry 58: 401–407. Meier A, Mollenhauer B, Cohrs S et al. (2005). Normal hypocretin-1 (orexin-A) levels in the cerebrospinal fluid of patients with Huntington’s disease. Brain Res 1063: 201–203. Mignot E, Lammers GJ, Ripley B et al. (2002). The role of cerebrospinal fluid hypocretin measurement in the diagnosis of narcolepsy and other hypersomnias. Arch Neurol 59: 1553. Mishima T, Kasanuki K, Koga S et al. (2017). Reduced orexin immunoreactivity in Perry syndrome and multiple system atrophy. Parkinsonism Relat Disord 42: 85–89. Naber WC, Fronczek R, Haan J et al. (2019). The biological clock in cluster headache: a review and hypothesis. Cephalalgia 39: 033310241985181. Nevsimalova S, Vankova J, Stepanova I et al. (2005). Hypocretin deficiency in Prader-Willi syndrome. Eur J Neurol 12: 70–72. Nishino S (2005). The orexin/hypocretin system, Springer 73–82. Nishino S, Ripley B, Overeem S et al. (2000). Hypocretin (orexin) deficiency in human narcolepsy. Lancet 355: 39–40. Nishino S, Kanbayashi T, Fujiki N et al. (2003). CSF hypocretin levels in Guillain-Barre syndrome and other inflammatory neuropathies. Neurology 61: 823–825. Omokawa M, Ayabe T, Nagai T et al. (2016). Decline of CSF orexin (hypocretin) levels in Prader-Willi syndrome. Am J Med Genet A 170: 1181–1186. Omori Y, Kanbayashi T, Imanishi A et al. (2018). Orexin/hypocretin levels in the cerebrospinal fluid and characteristics of patients with myotonic dystrophy type 1 with excessive daytime sleepiness. Neuropsychiatr Dis Treat 14: 451–457. Ondo WG, Perkins T, Swick T et al. (2008). Sodium oxybate for excessive daytime sleepiness in Parkinson disease: an open-label polysomnographic study. Arch Neurol 65: 1337–1340. Osorio RS, Ducca EL, Wohlleber ME et al. (2016). Orexin-A is associated with increases in cerebrospinal fluid phosphorylated-tau in cognitively normal elderly subjects. Sleep 39: 1253–1260.

THE OREXIN/HYPOCRETIN SYSTEM IN NEUROPSYCHIATRIC DISORDERS Overeem S, van Hilten JJ, Ripley B et al. (2002). Normal hypocretin-1 levels in Parkinson’s disease patients with excessive daytime sleepiness. Neurology 58: 498–499. Overeem S, Dalmau J, Bataller L et al. (2004). Hypocretin-1 CSF levels in anti-Ma2 associated encephalitis. Neurology 62: 138–140. Perkins DO, Jeffries CD, Addington J et al. (2014). Toward a psychosis risk blood diagnostic for persons experiencing high-risk symptoms: preliminary results from the NAPLS project. Schizophr Bull 41: 419–428. ˚ , Gil J, Maat-Schieman MLC et al. (2004). Orexin Petersen A loss in Huntington’s disease. Hum Mol Genet 14: 39–47. Peyron C, Tighe DK, van den Pol AN et al. (1998). Neurons containing hypocretin (orexin) project to multiple neuronal systems. J Neurosci 18: 9996–10015. Peyron C, Faraco J, Rogers W et al. (2000). A mutation in a case of early onset narcolepsy and a generalized absence of hypocretin peptides in human narcoleptic brains. Nat Med 6: 991–997. Plazzi G, Fabbri C, Pizza F et al. (2015). Schizophrenia-like symptoms in narcolepsy type 1: shared and distinctive clinical characteristics. Neuropsychobiology 71: 218–224. Poceta JS, Parsons L, Engelland S et al. (2009). Circadian rhythm of CSF monoamines and hypocretin-1 in restless legs syndrome and Parkinson’s disease. Sleep Med 10: 129–133. Podesta´ C, Ferreras M, Mozzi M et al. (2006). Kleine–Levin syndrome in a 14-year-old girl: CSF hypocretin-1 measurements. Sleep Med 7: 649–651. Rainero I, Gallone S, Valfre W et al. (2004). A polymorphism of the hypocretin receptor 2 gene is associated with cluster headache. Neurology 63: 1286–1288. Razavi BM, Hosseinzadeh H (2017). A review of the role of orexin system in pain modulation. Biomed Pharmacother 90: 187–193. Recourt K, de Boer P, Zuiker R et al. (2019). The selective orexin-2 antagonist seltorexant (JNJ-42847922/MIN-202) shows antidepressant and sleep-promoting effects in patients with major depressive disorder. Transl Psychiatry 9: 216. Ripley B, Overeem S, Fujiki N et al. (2001). CSF hypocretin/ orexin levels in narcolepsy and other neurological conditions. Neurology 57: 2253–2258. Rissling I, K€orner Y, Geller F et al. (2005). Preprohypocretin polymorphisms in Parkinson disease patients reporting “sleep attacks.”. Sleep 28: 871–875. Rodrigues TM, Caldas AC, Ferreira JJ (2016). Pharmacological interventions for daytime sleepiness and sleep disorders in Parkinson’s disease: systematic review and meta-analysis. Parkinsonism Relat Disord 27: 25–34. Roh JH, Jiang H, Finn MB et al. (2014). Potential role of orexin and sleep modulation in the pathogenesis of Alzheimer’s disease. J Exp Med 211: 2487–2496. Sakai N, Matsumura M, Lin L et al. (2019). HPLC analysis of CSF hypocretin-1 in type 1 and 2 narcolepsy. Sci Rep 9: 477. Sakurai T, Amemiya A, Ishii M et al. (1998). Orexins and orexin receptors: a family of hypothalamic neuropeptides

357

and g protein-coupled receptors that regulate feeding behavior. Cell 92: 573–585. Salomon RM, Ripley B, Kennedy JS et al. (2003). Diurnal variation of cerebrospinal fluid hypocretin-1 (Orexin-A) levels in control and depressed subjects. Biol Psychiatry 54: 96–104. Saper CB, Scammell TE, Lu J (2005). Hypothalamic regulation of sleep and circadian rhythms. Nature 437: 1257–1263. Sarchielli P, Rainero I, Coppola F et al. (2007). Involvement of corticotrophin-releasing factor and orexin-A in chronic migraine and medication-overuse headache: findings from cerebrospinal fluid. Cephalalgia 28: 714–722. Schizophrenia Working Group of the Psychiatric Genomics Consortium, Sekar A, Bialas AR et al. (2016). Schizophrenia risk from complex variation of complement component 4. Nature 530: 177–183. Schmidt FM, Kratzsch J, Gertz H-J et al. (2013). Cerebrospinal fluid melanin-concentrating hormone (MCH) and hypocretin-1 (HCRT-1, Orexin-A) in Alzheimer’s disease. PLoS One 8: e63136. Sch€ urks M, Kurth T, Geissler I et al. (2007a). The G1246A polymorphism in the hypocretin receptor 2 gene is not associated with treatment response in cluster headache. Cephalalgia 27: 363–367. Sch€ urks M, Limmroth V, Geissler I et al. (2007b). Association between migraine and the G1246A polymorphism in the hypocretin receptor 2 gene. Headache 47: 1195–1199. Shokri-Kojori E, Wang G-J, Wiers CE et al. (2018). b-Amyloid accumulation in the human brain after one night of sleep deprivation. Proc Nat Acad Sci U S A 115: 201721694. Siegel J, Nienhuis R, Fahringer H et al. (1991). Neuronal activity in narcolepsy: identification of cataplexy-related cells in the medial medulla. Science 252: 1315–1318. Slats D, Claassen JAHR, Lammers GJ et al. (2012). Association between hypocretin-1 and amyloid-b42 cerebrospinal fluid levels in Alzheimer’s disease and healthy controls. Curr Alzheimer Res 9: 1119–1125. Stefansson H, Ophoff RA, Steinberg S et al. (2009). Common variants conferring risk of schizophrenia. Nature 460: 744–747. Talih FR (2011). Narcolepsy presenting as schizophrenia: a literature review and two case reports. Innov Clin Neurosci 8: 30–34. Thannickal TC, Moore RY, Nienhuis R et al. (2000). Reduced number of hypocretin neurons in human narcolepsy. Neuron 27: 469–474. Thannickal TC, Lai Y-Y, Siegel JM (2007). Hypocretin (orexin) and melanin concentrating hormone loss and the symptoms of Parkinson’s disease. Brain 131: e87. Thannickal TC, Nienhuis R, Siegel JM (2009). Localized loss of hypocretin (orexin) cells in narcolepsy without cataplexy. Sleep 32: 993–998. Thomasy HE, Opp MR (2019). Hypocretin mediates sleep and wake disturbances in a mouse model of traumatic brain injury. J Neurotrauma 36: 802–814.

358

R. FRONCZEK ET AL.

Thomasy HE, Febinger HY, Ringgold KM et al. (2017). Hypocretinergic and cholinergic contributions to sleepwake disturbances in a mouse model of traumatic brain injury. Neurobiol Sleep Circadian Rhythm 2: 71–84. Tobias ES, Tolmie JL, Stephenson JBP (2002). Cataplexy in the Prader-Willi syndrome. Arch Dis Child 87: 170-a-170. Trepanier MO, Hopperton KE, Mizrahi R et al. (2016). Postmortem evidence of cerebral inflammation in schizophrenia: a systematic review. Mol Psychiatry 21: 1009–1026. Urrestarazu E, Iriarte J (2016). Clinical management of sleep disturbances in Alzheimer’s disease: current and emerging strategies. Nat Sci Sleep 8: 21–33. Usuda M, Kodaira M, Ogawa Y et al. (2018). Fluctuating CSF hypocretin-1 levels in mild brain trauma-induced KleineLevin syndrome. J Neurol Sci 391: 10–11. Vassalli A, Dellepiane JM, Emmenegger Y et al. (2013). Electroencephalogram paroxysmal theta characterizes cataplexy in mice and children. Brain 136: 1592–1608. Vu PA, Tucker LB, Liu J et al. (2018). Transient disruption of mouse home cage activities and assessment of orexin immunoreactivity following concussive- or blast-induced brain injury. Brain Res 1700: 138–151. Wang JY, Han F, Dong SX et al. (2016). Cerebrospinal fluid orexin A levels and autonomic function in Kleine-Levin syndrome. Sleep 39: 855–860. Watanabe S, Kuwaki T, Yanagisawa M et al. (2005). Persistent pain and stress activate pain-inhibitory orexin pathways. Neuroreport 16: 5–8.

Weickert CS, Weickert TW (2016). What’s hot in schizophrenia research? Psychiatr Clin N Am 39: 343–351. Weller CM, Wilbrink LA, Houwing-Duistermaat JJ et al. (2015). Cluster headache and the hypocretin receptor 2 reconsidered: a genetic association study and metaanalysis. Cephalalgia 35: 741–747. Wennstr€ om M, Londos E, Minthon L et al. (2012). Altered CSF orexin and a-synuclein levels in dementia patients. J Alzheimers Dis 29: 125–132. Wienecke M, Werth E, Poryazova R et al. (2012). Progressive dopamine and hypocretin deficiencies in Parkinson’s disease: is there an impact on sleep and wakefulness? J Sleep Res 21: 710–717. Wu M-F, Gulyani SA, Yau E et al. (1999). Locus coeruleus neurons: cessation of activity during cataplexy. Neuroscience 91: 1389–1399. Xie L, Kang H, Xu Q et al. (2013). Sleep drives metabolite clearance from the adult brain. Science 342: 373–377. Yamanaka A, Sakurai T, Katsumoto T et al. (1999). Chronic intracerebroventricular administration of orexin-A to rats increases food intake in daytime, but has no effect on body weight. Brain Res 849: 248–252. Yasui K, Inoue Y, Kanbayashi T et al. (2006). CSF orexin levels of Parkinson’s disease, dementia with Lewy bodies, progressive supranuclear palsy and corticobasal degeneration. J Neurol Sci 250: 120–123. Yue W-H, Wang H-F, Sun L-D et al. (2011). Genome-wide association study identifies a susceptibility locus for schizophrenia in Han Chinese at 11p11.2. Nat Genet 43: 1228–1231.

Handbook of Clinical Neurology, Vol. 180 (3rd series) The Human Hypothalamus: Middle and Posterior Region D.F. Swaab, F. Kreier, P.J. Lucassen, A. Salehi, and R.M. Buijs, Editors https://doi.org/10.1016/B978-0-12-820107-7.00022-7 Copyright © 2021 Elsevier B.V. All rights reserved

Chapter 22

Pleasure, addiction, and hypocretin (orexin) RONALD MCGREGOR1,2, THOMAS C. THANNICKAL1,2, AND JEROME M. SIEGEL1,2* 1

Neuropsychiatric Institute and Brain Research Institute, University of California, Los Angeles, CA, United States

2

Neurobiology Research, Veterans Administration Greater Los Angeles Healthcare System, Los Angeles, CA, United States

Abstract The hypocretins/orexins were discovered in 1998. Within 2 years, this led to the discovery of the cause of human narcolepsy, a 90% loss of hypothalamic neurons containing these peptides. Further work demonstrated that these neurons were not simply linked to waking. Rather these neurons were active during pleasurable behaviors in waking and were silenced by aversive stimulation. This was seen in wild-type mice, rats, cats, and dogs. It was also evident in humans, with increased Hcrt release during pleasurable activities and decreased release, to the levels seen in sleep, during pain. We found that human heroin addicts have, on average, an increase of 54% in the number of detectable Hcrt neurons compared to “control” human brains and that these Hcrt neurons are substantially smaller than those in control brains. We found that in mice, chronic morphine administration induced the same changes in Hcrt neuron number and size. Our studies in the mouse allowed us to determine the specificity, dose response relations, time course of the change in the number of Hcrt neurons, and that the increased number of Hcrt neurons after opiates was not due to neurogenesis. Furthermore, we found that it took a month or longer for these anatomical changes in the mouse brain to return to baseline. Human narcoleptics, despite their prescribed use of several commonly addictive drugs, do not show significant evidence of dose escalation or substance use disorder. Similarly, mice in which the peptide has been eliminated are resistant to addiction. These findings are consistent with the concept that an increased number of Hcrt neurons may underlie and maintain opioid or cocaine use disorders.

ANATOMY The hypocretin (Hcrt)/orexin peptides were discovered by two independent groups in 1998 (De Lecea et al., 1998; Sakurai et al., 1998; Siegel et al., 2001). The name hypocretin was created because of the hypothalamic localization of all somas containing the peptides and the resemblance of the peptides to secretin (De Lecea et al., 1998). The name orexin was selected because of the hypothesis that these peptides might drive appetite (Sakurai et al., 1998), since early work had shown that damage to the lateral hypothalamus produces anorexia, whereas damage to the medial hypothalamus produces hyperphagia and obesity (Anand and Brobeck, 1951; Teitelbaum and Epstein, 1962). Although the Hcrt

peptides are often erroneously described as being in the “lateral” hypothalamus, these neurons are in fact present through the medial-lateral extent of the hypothalamus. An equal number of Hcrt neurons are present medial and lateral to the fornix, a structure used to define the boundary between the medial and lateral hypothalamus. Mice, rats, and humans all have Hcrt neurons throughout the medial-lateral extent of the hypothalamus (Peyron et al., 1998; Thannickal et al., 2000a,b; McGregor et al., 2011). Hcrt neurons are also present in the zona incerta in primates and other species (Bhagwandin et al., 2011; Dell et al., 2012, 2013, 2016a,b,c; Olateju et al., 2017; Pillay et al., 2017). In the rostro-caudal dimension, Hcrt neurons are present

*Correspondence to: Jerome M. Siegel, Neuropsychiatric Institute and Brain Research Institute, University of California, Los Angeles, Los Angeles, CA 90095, United States. Tel: +1-818-366-8900, E-mail: [email protected]

360

R. MCGREGOR ET AL.

in the tuberal and mamillary regions of the hypothalamus, and though the majority of these neurons are located dorsally to the fornix there are some neuronal somas ventral to this structure. Throughout their distribution, Hcrt neurons are intermingled with many other cell types, not forming a dense homogeneous nucleus. From their hypothalamic location they send extensive projections within the hypothalamus and to the rest of the neuraxis, from the spinal cord to the cerebral cortex (Peyron et al., 1998; Chen et al., 1999; Date et al., 1999; Horvath et al., 1999; van den Pol, 1999). Hcrt signaling is conveyed through two G-protein-coupled receptors (HcrtR1 and, HcrtR2) with a range of distribution that overlaps that of Hcrt fibers (Marcus et al., 2001; Kukkonen and Leonard, 2014). Phylogenetic studies have shown a high degree of receptor homology between different species indicating that this system is evolutionary conserved (Ammoun et al., 2003). Activation of these receptors by Hcrts has short-term effects like depolarization and increase in neuronal firing rate and long-term effects including modulation of cell plasticity (Sakurai et al., 1998; Smart et al., 1999; Eriksson et al., 2001). Hcrt neurons respond to Hcrt peptides directly via the HcrtR2 or indirectly (via HcrtR1) through the release of glutamate (Li et al., 2002; Yamanaka et al., 2010).

Hcrt LINK TO NARCOLEPSY The development of a Hcrt peptide knockout mouse, in which the neurons normally containing hypocretin are present (identified by the cotransmitters dynorphin and neuronal activity regulated pentraxin (Narp) (Chou et al., 2001; Blouin et al., 2005; Crocker et al., 2005), but the Hcrt peptide itself is not (Siegel, 2004; Blouin et al., 2005; Crocker et al., 2005), produced the disappointing observation that these animals were not anorexic, leading Chemelli et al. (1999) to use video observation to determine if there were any other abnormalities in their behavior. They made the striking observation that these mice showed sudden movement arrests. Their further work demonstrated that these were not seizures or losses of consciousness, but rather had electroencephalographic and electromyographic signs of waking, resembling those of cataplexy in human narcoleptics. This led to the discovery that there was a 90% loss of Hcrt neurons in human narcoleptics, amid signs of prior hypothalamic inflammation (Peyron et al., 2000; Thannickal et al., 2000a,b, 2003). This was the first indication of a neuroanatomical abnormality in human narcoleptics, although we had previously identified an abnormality in genetically narcoleptic dogs (Siegel et al., 1999). These dogs have a mutation that disrupts the function of the HcrtR2 (Lin et al., 1999). We found that they had elevated levels of axonal degeneration

and reactive neuronal somata, an indicator of neuronal pathology, in a number of subcortical structures. These degenerative changes precede or coincide with symptom onset. In very rare cases, human narcolepsy can be caused by an Hcrt mutation, impairing peptide trafficking and processing (Peyron et al., 2000). Nearly all human narcolepsy appears to be linked to an autoimmune process that causes destruction of Hcrt neurons (Scammell, 2006). This autoimmune hypothesis stems from the discovery that nearly all (
95%) of all human narcoleptics have an HLA immune subtype (DQB1*0602) present in only about 25% of the general population (Honda et al., 1984; Mignot et al., 2001). This hypothesis received further support from the finding that cases of narcolepsy increased during the H1N1 influenza epidemic in individuals immunized for the virus (Dauvilliers et al., 2010) and in those who contracted H1N1 without immunization (Han et al., 2011).

HYPOCRETIN, REWARD, AND OPIOIDS We (Kiyashchenko et al., 2002; Mileykovskiy et al., 2005; McGregor et al., 2011; Wu et al., 2011a,b) and others (Nestler et al., 2002; Georgescu et al., 2003; Harris et al., 2005; Boutrel and De Lecea, 2008; Borgland et al., 2009; Aston-Jones et al., 2010; Nestler, 2013; Baimel et al., 2015; Hassani et al., 2016; James et al., 2017) have demonstrated that increased neuronal discharge in Hcrt neurons is linked to the performance of rewarded tasks in wild-type (WT) mice, rats, cats, and dogs. Mice in which the Hcrt peptide is genetically knocked out (Hcrt-KO) learn to bar press for food or water as quickly as their WT littermates in the light phase and will respond as well as WT on fixed ratio tasks requiring relatively low effort. This indicates that they experience the rewarding properties of these natural reinforcers. However, when the effort to obtain these rewards is increased in a progressive ratio, the mice invariably stop bar pressing before the end of the 2 h test period, whereas their WT littermates continue until the end of the session (Fig. 22.1). The Hcrt-KO mice never showed cataplexy during the positive reinforcement-tests, but often fell asleep as the amount of work required to receive the reward (progressive ratio) increased. However, surprisingly, the KO mice were unimpaired relative to WT mice when working for a positive reward during the dark phase (Fig. 22.1E). This indicates that Hcrt peptides play a critical role in mediating motivated behaviors during the natural “sleep time” in these animals (McGregor et al., 2011). These behavioral results find striking parallels with the activity of Hcrt neurons. Mirroring the behavioral deficits seen in Hcrt-KO animals, we found that in WT mice, expression of the immediate early gene cFos in

PLEASURE, ADDICTION, AND HYPOCRETIN (OREXIN)

c

WT Food PR5 - Light Phase

10

Lever presses (x102)

Lever presses (x102)

a 8 6 4 2 8

24

40

56

72

88

b

104

120

Time (min)

8 6 4 2

WT Water PR3 – Light phase

10 8 6 4 2

8

d Lever presses (x102)

Lever presses (x102)

KO Food PR5 - Light Phase

10

0

0

24

40

56

72

88

104

120

104

120

Time (min)

KO Water PR3 – Light phase

10 8 6 4 2 0

0 8

24

e Lever presses (x102)

361

40

56

72

88

104

120

Time (min)

8

24

40

56

72

88

Time (min)

KO Food PR5 - Light/Dark Phase

10 8 6 4 2 0 8

24

40

56

72

88

104

120

Time (min)

Fig. 22.1. Operant performance of WT and KO mice on progressive ratio responding for food or water reinforcement paradigm. Hcrt-KO mice are unable to sustain bar pressing for food or water in the light phase, in contrast to littermate WT mice. Representative cumulative records of the performance of a WT animal (A, food, 2817 total presses; B, water, 2494 total presses) and an Hcrt-KO animal (C, food, 456 total presses; D, water, 164 total presses) responding for positive reinforcers. The downward pips on the cumulative record denote food or water deliveries. The Hcrt-KO mouse sessions were terminated when they ceased pressing the lever for 15 min. Hcrt-KOs are unimpaired on the same task in the dark phase (E). Redrawn from McGregor R, Wu M-F, Barber G, Ramanathan L, Siegel JM (2011). Highly specific role of hypocretin (orexin) neurons: differential activation as a function of diurnal phase, operant reinforcement vs. operant avoidance and light level. J Neurosci 31: 15455–15467.

Hcrt neurons, an indirect indicator of neuronal activation, occurs only in the light phase when working for positive reinforcement in a progressive ratio task. In a second set of experiments, we observed that Hcrt-KO mice were unimpaired relative to WT when working to avoid a foot shock in a progressive ratio schedule during the light or dark phase. Analysis of Hcrt activation (cFos) under these conditions revealed that these neurons were not activated during the performance of this task. Furthermore, cFos was not expressed in Hcrt neurons above baseline when expected or unexpected rewards were presented, or when given or expecting an unavoidable foot

shock, even though these conditions elicit maximal electroencephalogram (EEG) arousal (Figs. 22.2–22.3). Together these results from behavioral and anatomical studies point toward an emotional specificity in the recruitment of Hcrt neurons (McGregor et al., 2011; Blouin et al., 2013). Interestingly when light was turned off, cFos was not expressed in Hcrt neurons beyond control levels in the light phase during positive reinforcement, indicating a very specific role of light in Hcrt’s involvement in reinforcement (McGregor et al., 2011; Blouin et al., 2013). This finding is consistent with the lack of

362

R. MCGREGOR ET AL.

Fig. 22.2. Distribution of Hcrt and cFos/Hcrt neurons in the hypothalamus of WT mice under different behavioral conditions. Hcrt neurons express cFos during a food motivated task in the light phase. Neither food nor shock avoidance tasks increase cFos expression in the dark phase. (A) Diagrams of coronal sections of the hypothalamus stained for Hcrt and cFos of six animals each under one of six different experimental conditions during the light and the dark phase: L1, PR food, light phase; L2, shock avoidance, light phase; L3, chamber control, light phase; D1, PR food, dark phase; D2, shock avoidance, dark phase; D3, chamber control, dark phase. Red dots indicate double-labeled cFos/Hcrt neurons; blue triangles correspond to Hcrt neurons. Fx, Fornix; 3V, third ventricle. Scale bar, 150 mm. (B) Photomicrographs of the same hypothalamic region in a section processed for Hcrt and cFos. LH, Lateral hypothalamus; MH, medial hypothalamus. Scale bar, 150 mm. The rectangular region in the LH is magnified in the insert at the lower left. Scale bar, 20 mm. The double-labeled neurons (red arrows) show the characteristic black nucleus due to the presence of cFos protein and a brown precipitate in the cytoplasm, indicating their hypocretinergic nature. These cells are easily distinguishable from single-labeled hypocretin neurons (blue arrows) and single-labeled cFos cells (black arrowheads). Redrawn from McGregor R, Wu M-F, Barber G, Ramanathan L, Siegel JM (2011). Highly specific role of hypocretin (orexin) neurons: differential activation as a function of diurnal phase, operant reinforcement vs. operant avoidance and light level. J Neurosci 31: 15455–15467.

PLEASURE, ADDICTION, AND HYPOCRETIN (OREXIN)

%Hcrt cells expressing Fos

70

**

363

Light Phase

60

Dark Phase

50 40 30 20 10 0 PR Food

Shock Avoidance

Chamber Control

Fig. 22.3. Percentage of Hcrt neurons expressing cFos in the hypothalamus of WT mice under different behavioral conditions. Activation of Hcrt neurons was maximal when bar pressing for food in the light phase, but not during shock avoidance. Comparison of the percentage of hypocretin neurons expressing cFos in the PR food, shock avoidance, and chamber control conditions during the light and dark phases (**P < 0.01, Newman–Keuls post hoc test comparing food task during the light phase with all other conditions; n ¼ 4 in each condition). There is no significant difference between the light and dark phases in shock avoidance and chamber control conditions. Redrawn from McGregor R, Wu M-F, Barber G, Ramanathan L, Siegel JM (2011). Highly specific role of hypocretin (orexin) neurons: differential activation as a function of diurnal phase, operant reinforcement vs. operant avoidance and light level. J Neurosci 31: 15455–15467.

HCRT, DOPAMINE, AND ADDICTION Somewhat similar to Hcrt neurons, dopamine neurons, particularly those located in the ventral tegmental area (VTA), have long been implicated in reinforcement in general and addiction in particular (Beitner-Johnson et al., 1992, 1993; Mignot et al., 1995; Schilstrom et al., 1998; Sarti et al., 2002; Meye et al., 2012; Farahimanesh et al., 2017). Hcrt and dopamine are evolutionarily linked from both a neurochemical and anatomical perspective (Stefano and Kream, 2007). VTA plasticity associated

10

** ** ** **

** **

6

REM phasic

Sleep

Quiet wake

Ea ng

Grooming

0

REM tonic

*

2

Explora on

Firing rate spikes/s

light-induced arousal in human narcoleptics (Hajek et al., 1989), reported prior to the discovery of Hcrt, in contrast to the arousing effects of light in nonnarcoleptics. It has been previously reported that there is a dichotomy in the functions of the Hcrt neuronal population, with the medial group related to arousal and the lateral group to reward (Harris and Aston-Jones, 2006). In our studies we did not observe a restricted distribution in the double-labeled Hcrt/cFos neurons. Rather they were seen homogeneously throughout the medial-lateral extent of the Hcrt field. Recording of Hcrt neurons in freely moving rats showed that they discharged maximally during exploration, grooming, and eating, but ceased discharge during aversive stimulation in waking (Mileykovskiy et al., 2005); all changes consistent with our work on reinforcement in mice (McGregor et al., 2011). They reduced discharge in non-REM sleep with a low level of activity in REM sleep (Fig. 22.4).

Fig. 22.4. Firing rate of Hcrt neurons in waking and sleep behaviors in freely moving rats. Maximal discharge is seen during exploration-approach behavior. Group average of the discharge pattern of Hcrt neurons (n ¼ 9) in different behavioral conditions (*P < 0.05, **P < 0.01 Bonferroni t-test). Error bars indicate SEM. Redrawn from Mileykovskiy BY, Kiyashchenko LI, Siegel JM (2005). Behavioral correlates of activity in identified hypocretin/orexin neurons. Neuron 46: 787–798.

with drug rewards requires functional Hcrt receptors (Baimel et al., 2015). The levels of dopamine and its major metabolites in the nucleus accumbens are markedly increased by the microinjection of Hcrts into the VTA. Hcrt neurons project strongly to the VTA, where the peptides appear to act via volume conduction

364

R. MCGREGOR ET AL.

b ** 8

Hcrt cell discharge (Hz)

12

#

15 5 0

## EEG

Postmorphine

REM sleep

Non REM

##

Quiet wake

0

Morphine

25

EMG

4

Ac ve wake

Hcrt discharge rate (Hz)

a

9:30

10:00

10:30

11:00

11:30

12:00

12:30

Time of day (h)

25 15 0

EEG

EMG

Hcrt cell discharge (Hz)

2.1 ms

Morphine

9:30

9:40

9:50 Time of day (h)

12:30

12:40

12:50 Time of day (h)

Fig. 22.5. Effect of morphine administration on hypocretin cell activity in freely moving rats. A species-appropriate dose of morphine (15 mg/kg) injected into three freely moving rats resulted in an elevated neuronal discharge rate lasting for 3 h accompanied by an increase in EMG activity. (A) Sleep rates are averages of mean rate determined by five 10-s samples in each of five hypocretin neurons from three rats in each sleep state: active waking, quiet waking, non-REM sleep, and REM sleep. Postmorphine injection rate was based on five 10-s samples in each neuron taken 15 min after morphine injection. One-way ANOVA of hypocretin neurons (n ¼ 5): F4,16 ¼ 18.2, **P < 0.0001. Posthoc comparisons with Tukey/Kramer procedure: active waking/quiet waking, #P < 0.05; active waking/non-REM sleep, ##P < 0.01; active waking/REM, ##P < 0.01. (B) Discharge rate of rat hypocretin neurons after morphine administration. Bottom: Traces show EEG activation immediately after morphine injection (left) and 3 h after injection (right). Expanded trace shows the characteristic long average waveform of hypocretin neurons. Redrawn from Thannickal TC, John J, Shan L, Swaab DF, Wu M-F, Ramanathan L, McGregor R, Chew K-T, Cornford M, Yamanaka A, Inutsuka A, Fronczek R, Lammers GJ, Worley PF, Siegel JM (2018). Opiates increase the number of hypocretin-producing cells in mouse and human brain, and reverse cataplexy in a mouse model of narcolepsy. Sci Transl Med 10: p. pii: eaao4953. doi: 10.1126/ scitranslmed.aao4953.

(Del Cid-Pellitero and Garzon, 2014), and to the nucleus accumbens and paraventricular nucleus of the thalamus (Peyron et al., 1998). The paraventricular nucleus also projects directly to the nucleus accumbens (Zhu et al., 2016). Thus via its direct and indirect projections, Hcrt can strongly modulate circuits implicated in addiction (Peyron et al., 1998; Sim-Selley et al., 2011; Ho and Berridge, 2013; Zhu et al., 2016, Chen et al., 2006; Anderson et al., 2017).

Hcrt and opioids An in vitro slice study found that opioids decrease the activity of Hcrt neurons and that blockade of m-opioid receptors enhances the activity of Hcrt neurons.

Morphine pretreatment inhibits subsequent excitatory responses to Hcrt in Hcrt neurons (Li and van den Pol, 2008). However, our current in vivo data (Fig. 22.5) (Thannickal et al., 2018) shows that systemic administration of morphine greatly increases Hcrt unit activity in intact rats. The effects of opioid agonists can be exerted not only in plasma membrane receptors and endosomes but also in the Golgi apparatus (Stoeber et al., 2018), suggesting a possible pathway for the alteration of Hcrt neuronal size after chronic opioid exposure that we have reported (Thannickal et al., 2000a,b, 2018) and for receptor expression (Cai et al., 2019). A large percentage of Hcrt cells also release glutamate (Torrealba et al., 2003), trigger glutamate release from adjacent cells. They also contain corelease dynorphin (Li and Van Den Pol,

PLEASURE, ADDICTION, AND HYPOCRETIN (OREXIN) 2006;Muschamp et al., 2014), a member of the opioid peptide family that preferentially binds to the kappa opioid receptor (KOR) (Schwarzer, 2009). These two neuropeptides have opposing roles in reward related behaviors such as cocaine and alcohol self-administration, cocaine seeking, impulsivity, and brain stimulation reward (Matzeu and Martin-Fardon, 2018; Anderson et al., 2018). The VTA firing rate is increased by Hcrt and decreased by dynorphin, but bath coapplication of both peptides resulted in no net changes in neuronal firing (Muschamp et al., 2014). HcrtR1 and KOR can form receptor heterodimers, altering signal transduction and second messenger activation including increased

Human studies In another study, we found that Hcrt is released in the brain of nonaddict humans when they are engaged in enjoyable tasks, but not when they are aroused by pain or feeling sad (Fig. 22.6) (Blouin et al., 2013). Elevating Hcrt production by self-administration of opioids

a

Visit by family members

Fmol (x10-1) 20 ul sample

10

Interacts with physician

6

Interacts with physician Sleep

2 0

22:00

365

protein kinase A activity and intracellular cAMP levels (Chen et al., 2015). Hcrt neurons also contain neuronal activity regulated pentraxin, involved in aggregating AMPA receptors and thought to have a role in addiction (Blouin et al., 2005; Crocker et al., 2005).

Sleep

Sleep 02:00

06:00

Interacts with staff

10:00

14:00

17:30

09:30

14:00

18:30

55

35

Sleepimg just before wake

Sleeping

Sleep onset

During

Before

Wake while in pain

Wake just before sleep

Awake

0

Wake Onset

15

Social

Fmol (x10-2), 15 ul sample

b 75

Fig. 22.6. (A) Time course of Hcrt release over a 20-h period in patient d378. Maximal release occurred during interactions that the subject rated as pleasurable on a periodically administered questionnaire. Hcrt release was minimal during sleep and during pain. (B) Maximal Hcrt levels in waking are seen during positive emotions, social interactions, and awakening; minimal levels are seen before sleep and when reporting pain. Changes during and after eating are smaller than those during monitored non-eating-related activities. Waking values in shades of green, and sleep values in shades of blue. Awake indicates samples in which subjects were awake but were not exhibiting social interaction or reporting emotion. Redrawn from Blouin AM, Fried I, Wilson CL, Staba RJ, Behnke EJ, Lam HA, Maidment NT, Karlsson KAE, Lapierre JL, Siegel JM (2013). Human hypocretin and melanin-concentrating hormone levels are linked to emotion and social interaction. Nat Commun 4: 1547.

366

R. MCGREGOR ET AL.

(Thannickal et al., 2018) creates a positive mental state. A negative affect is correlated with reduced administration of opioids and a diminishing rate of Hcrt production (C.D.C, 2017). Humans with narcolepsy have greatly elevated levels of depression (Ponz et al., 2010b; Lee et al., 2016; Nordstrand et al., 2019), with similar changes in animal models of narcolepsy (Lutter et al., 2008; James et al., 2018), i.e., both a low rate of Hcrt production (in narcoleptics) and a diminishing rate of Hcrt production (in addicts attempting withdrawal) (Thannickal et al., 2018) are correlated with depression. Similarly it has been shown that humans who have attempted suicide have lower levels of cerebrospinal Hcrt (Brundin et al., 2007, 2009). Circadian, sex-related differences, and brain region-specific changes in Hcrt system functioning have been reported in relation to human depression (Lu et al., 2017). We have shown that Parkinson’s disease patients have a considerable loss of Hcrt neurons (Fronczek et al., 2007; Thannickal et al., 2007, 2008), although not to the extent seen in narcoleptics. This loss may help explain the symptoms that Parkinson’s patients have in common with narcoleptics including daytime sleep attacks, nocturnal insomnia, hallucinations and depression, keeping in mind the much more extensive neuronal loss and symptoms in Parkinson’s. From a medical standpoint, the most critical issue in opiate addicts is the inability of many addicts to successfully withdraw from opioid use (Li and van den Pol, 2008; Editors, 2016; C.D.C, 2017; Chang et al., 2017; Ostling et al., 2018). The difficulty of withdrawal for addicts is not principally caused by the seeking of a pleasurable “high.” Rather it is seeking relief from the symptoms induced by withdrawal. These include insomnia (Valentino and Volkow, 2020), anxiety, irritability, hot flashes/chills, sweating, restlessness, and hyperalgesia. Acute symptoms typically peak 24–48 h after withdrawing from short-acting opioids (e.g., heroin or oxycodone). These acute symptoms may be followed by anhedonia, fatigue, anorexia, depression, and insomnia (Christie, 2008; Shi et al., 2009; Del Bello et al., 2013; Lutz et al., 2014; Zhu et al., 2016), effects that persist for weeks to months or years in humans (Sigmon et al., 2012). These short- and long-term effects drive most subjects who have attempted withdrawal to relapse within 1 year (McLellan et al., 2000; C.D.C, 2017; Volkow et al., 2018), even after medically supervised detoxification and pharmacological intervention.

HUMAN NARCOLEPTICS RARELY GET ADDICTED It has long been noted that narcoleptics, who have an average 90% loss of Hcrt neurons (Thannickal et al., 2000a,b), show little, if any, evidence of drug abuse, addiction or

overdose (Borgland et al., 2009; Guilleminault and Cao, 2011; Brown and Guilleminault, 2011; James et al., 2017), despite their daily prescribed use of gamma hydroxybutyrate, methylphenidate, and amphetamine. These drugs reverse the sleepiness and cataplexy of narcolepsy and are frequently abused in the general population with considerable loss of life (Harris et al., 2007; Borgland et al., 2009; Nishino and Mignot, 2011; Dauvilliers et al., 2013, 2014; Barateau et al., 2016; Darke et al., 2019; Jalal et al., 2018; Turner et al., 2018). Yet dose escalation and overdose are virtually nonexistent in narcoleptics (Galloway et al., 1997; Aston-Jones et al., 2010; Bayard and Dauvilliers, 2013; Baimel et al., 2015). Human narcoleptics have been shown to have a greatly reduced reward activation of the VTA, amygdala, and accumbens (Ponz et al., 2010a,b) and altered processing of humor in the hypothalamus and amygdala (Schwartz et al., 2007). The lack of abuse in human narcoleptics is consistent with the greatly reduced addiction potential in mice and rats with reduced Hcrt function (Sharf et al., 2010; Tabaeizadeh et al., 2013; Zarepour et al., 2014; Bentzley and Aston-Jones, 2015; Bali et al., 2015; Sadeghi et al., 2016; Sadeghzadeh et al., 2016; Guo et al., 2016; Farahimanesh et al., 2017; Alizamini et al., 2017; Assar et al., 2019; Azizbeigi and Haghparast, 2019; Azizbeigi et al., 2019; Pourhamzeh et al., 2019; Farzinpour et al., 2019; Shirazy et al., 2020; Zarrabian et al., 2020). It is also consistent with our recent finding of the converse phenomenon, greatly increased Hcrt cell number in human heroin addicts (Fig. 22.7) (Thannickal et al., 2018). Whereas a reduced number of Hcrt cells in narcoleptics is correlated with a greatly reduced addiction susceptibility in human and mouse narcoleptics, a greatly increased number of detected Hcrt-producing cells is elicited by opioid administration in humans and mice (Thannickal et al., 2018).

Changes in the Hcrt system produced by opioids Morphine had to be given for at least 2 weeks to produce a significant change in the number of Hcrt cells in mice, whereas cell size reduction was seen as soon as 72 h after subcutaneous implant of a morphine tablet (Thannickal et al., 2018). These changes in Hcrt neuron number and size after morphine were accompanied by an increased expression of preprohypocretin mRNA (Fig. 22.8). The opioid antagonist naltrexone (Narayanan et al., 2004; Skoubis et al., 2005; Shoblock and Maidment, 2006, 2007) given alone on the same dose schedule as morphine did not change the number of Hcrt neurons (data not shown) indicating that the maintenance of baseline number of Hcrt neurons does not require m-opioid receptor activation. The increased number of Hcrt neurons persisted for at least 4 weeks after discontinuation of 14 days

PLEASURE, ADDICTION, AND HYPOCRETIN (OREXIN)

Fx

OT

OT

502

25

723 OT

Fx

Fx

OT

OT

dic

l ro

t

MMOT

MM

Addict

189

124

5 0

Control

Ad

Co

192 Fx

nt

20 0

126

Co

100

d

*

nt r A d ol dic t

Hcrt cell size x 101 (μm²)

45

nt r A d ol dic t

Hcrt number % control

180

c

b

** * *

Co

a

367

Fig. 22.7. Postmortem brain tissue from heroin addicts shows an increased number of hypocretin-producing neurons. (A) Immunohistochemistry showed that there was a 54% increase in the number of detectable hypocretin neurons in hypothalamic brain tissue from human heroin addicts (n ¼ 5) relative to hypothalamic tissue from human control subjects (n ¼ 7; ****P ¼ 0.0009, t ¼ 8.89, df ¼ 10, t-test). (B) Immunohistochemical staining of postmortem brain tissue showed that hypocretin cells were 22% smaller in cross-sectional area in brain tissue from heroin addicts compared to control subjects [*P < 0.01, t ¼ 2.78, df ¼ 10 (t-test)]. (C) Neurolucida mapping illustrates the distribution and increased number of hypocretin cells in brain tissue from heroin addicts relative to control subjects. Representative counts are given at three anterior–posterior positions: OT, optic tract; MM, mamillary bodies; Fx, fornix. (D) A representative example of immunohistochemical labeling of hypocretin cells in brain tissue from control individuals and heroin addicts is shown. Hcrt neurons are smaller and more numerous in the addicts. Scale bar, 50 mm. Redrawn from Thannickal TC, John J, Shan L, Swaab DF, Wu M-F, Ramanathan L, McGregor R, Chew K-T, Cornford M, Yamanaka A, Inutsuka A, Fronczek R, Lammers GJ, Worley PF, Siegel JM (2018). Opiates increase the number of hypocretin-producing cells in mouse and human brain, and reverse cataplexy in a mouse model of narcolepsy. Sci Transl Med 10: p. pii: eaao4953. doi: 10.1126/scitranslmed.aao4953.

*

Rela ve mRNA expression

3

morphine treatment in mice, whereas the decrease in Hcrt cell size lasted for 2 weeks. Our data suggests that the increase may last much longer in human addicts than in mice. One of our addicts had 154% of the number of Hcrt neurons in control brains, even though he had not abused opioids for at least 3 years before his death (Thannickal et al., 2018). Self-administration has been shown to produce longer-lasting behavioral changes compared to passive, involuntary administration (Chen et al., 2006; McNamara et al., 2010; Picetti et al., 2012; Smith and Aston-Jones, 2012; James et al., 2013), suggesting that both species and administration differences may underlie these anatomical changes.

1.5

ine rph Mo

Sa

lin e

0

Fig. 22.8. Effect of morphine administration on preprohypocretin mRNA expression in mouse brain. An escalating dose of morphine, starting at 100 mg/kg, was given for 14 days to wild-type mice who were compared to saline injected littermates (*P < 0.05, t ¼ 2.99, df ¼ 5, t-test). Redrawn from Thannickal TC, John J, Shan L, Swaab DF, Wu M-F, Ramanathan L, McGregor R, Chew K-T, Cornford M, Yamanaka A, Inutsuka A, Fronczek R, Lammers GJ, Worley PF, Siegel JM (2018). Opiates increase the number of hypocretin-producing cells in mouse and human brain, and reverse cataplexy in a mouse model of narcolepsy. Sci Transl Med 10: eaao4953. doi: 10.1126/scitranslmed.aao4953.

MORPHINE DOES NOT PRODUCE “NEW” HCRT NEURONS We determined that the increase in the number of detected Hcrt cells was not due to neurogenesis. Both BrdU and doublecortin labeling indicated that no new neurons were produced by morphine (see fig. 4 in (Thannickal et al., 2018)). In a further study, we explored the issue of where the “newly visible” Hcrt neurons are coming from, by giving colchicine to drug naïve mice. Colchicine blocks axonal transport, thereby causing peptide to accumulate in the cell body. We found that this manipulation increased the number of “detectable” Hcrt cells in mice by about 44% (Fig. 22.9A) (McGregor et al., 2017), similar to

R. MCGREGOR ET AL.

Number of Hcrt+ neurons (x103)

a

5

**

4

3

2 1 0

b Number of MCH+ neurons(x103)

368

c

6

Saline 5 4 3

Colchicine 2 1 0

Saline Colchicine

Saline Colchi cine

Fig. 22.9. Colchicine given to mice increases the number of Hcrt neurons by 44%. Melanin-concentrating hormone (MCH) neuronal numbers remain unchanged. (A) Total cell counts for saline and colchicine treated subjects showing a 44% increase in the number of detected Hcrt neurons after colchicine (**P < 0.02, t-test). (B) Average number of MCH neurons in animals with ICV saline vs ICV colchicine injections. Number of MCH neurons remains unchanged. (C) Photomicrographs of the same hypothalamic area immunostained for Hcrt of an animal treated with saline (top) or colchicine (bottom). Calibration bar 250 mm. Inset corresponds to a higher (60) magnification of the selected area (black square) of the animal that received colchicine. Black arrow indicates an Hcrt neuron. Calibration bar 10 mm, Fx, fornix; LH, lateral hypothalamus; MH, medial hypothalamus; PFA, perifornical area. Redrawn from McGregor R, Wu M-F, Barber G, Ramanathan L, Siegel JM (2011). Highly specific role of hypocretin (orexin) neurons: differential activation as a function of diurnal phase, operant reinforcement vs. operant avoidance and light level. J Neurosci 31: 15455–15467.

the amount of increase seen in mice after morphine, i.e., as many as 44% of the neurons capable of producing Hcrt in mice do not produce it at detectable levels under “baseline” conditions. Fig. 22.9B shows that colchicine does not have any effect on the number of melaninconcentrating hormone neurons, a peptide of similar size, whose neurons are intermixed with Hcrt cells. Fig. 22.9C shows a representative hypothalamic section immunostained for Hcrt in a saline (top) and colchicine (bottom)-treated animal.

INSOMNIA IS A MAJOR CAUSE OF OPIOID WITHDRAWAL SYMPTOMS, LEADING TO RELAPSE Increased nocturnal wakefulness is a well-documented effect of opioid withdrawal. Despite progress in treating opioid dependence, sleep disturbance remains an almost universal complaint among withdrawing opioid addicts, persisting for more than 6 weeks and playing a major role in relapse. Longer sleep time is a predictor of increased treatment compliance and better treatment outcome (Gossop and Bradley, 1984; Beswick et al., 2003; Lofwall et al., 2013; Lin et al., 2014). Postaddiction insomnia may be mediated, to some extent, by the increased number of Hcrt-producing neurons, just as the

inability to maintain waking in human narcoleptics is linked to decreased Hcrt receptor activation (Peyron et al., 2000; Thannickal et al., 2000a,b; Sharf et al., 2010; Tabaeizadeh et al., 2013; Zarepour et al., 2014; Bali et al., 2015; Bentzley and Aston-Jones, 2015; Guo et al., 2016; Sadeghi et al., 2016; Sadeghzadeh et al., 2016; Alizamini et al., 2017; Farahimanesh et al., 2017; Assar et al., 2019; Azizbeigi and Haghparast, 2019; Azizbeigi et al., 2019; Farzinpour et al., 2019; Pourhamzeh et al., 2019; Shirazy et al., 2020; Zarrabian et al., 2020).

CONCLUSION The loss of Hcrt neurons causes human narcolepsy. In animal models a clear linkage between the Hcrt system and working for positive reinforcement has been shown. In contrast, Hcrt activity is not strongly altered by working to avoid aversive conditions. A strong circadian modulation of Hcrt function has been shown in both animals and humans. In a human microdialysis study, release was shown to be correlated with pleasurable activities (Blouin et al., 2013). Changes in Hcrt function have been linked to depression (Lu et al., 2017; Thannickal et al., 2018). We found a large increase in the number of Hcrt-producing neurons in human heroin addicts and

PLEASURE, ADDICTION, AND HYPOCRETIN (OREXIN) in mice chronically administered morphine (Thannickal et al., 2018). James et al. reported a nearly identical increase in the number of Hcrt-labeled neurons after chronic cocaine administration in rats (James et al., 2019), suggesting that the increase in Hcrt number may be a correlate of other chemical use disorders. Examining changes in Hcrt anatomy and physiology may shed light on a wide range of behavioral disorders. Researchers have typically characterized Hcrt neurons as a key part of a waking system. The work reviewed previously suggests that this is an oversimplification. Rather Hcrt activity is linked to particular types of waking behavior. In prior work it has been shown that neurons in the classic brainstem “waking arousal” systems are in fact related to very specific movements that occur in waking rather than relating simply to the waking state (Siegel and McGinty, 1976, 1977; Siegel, 1979; Siegel et al., 1979, 1980, 1983; Siegel and Tomaszewski, 1983). The work on Hcrt neurons suggests that other waking or sleep-related neurons may similarly have positive or negative emotional or behavioral roles. Understanding the behavioral roles of these neuronal groups is critical to understanding the waking state itself.

ACKNOWLEDGMENTS Supported by RO1 HL148574 and DA034748. Dr. Siegel is the recipient of a Senior Research Career Scientist Award 1IK6BX005245 from the Department of Veterans Affairs.

REFERENCES Alizamini MM, Farzinpour Z, Ezzatpanah S et al. (2017). Role of intra-accumbal orexin receptors in the acquisition of morphine-induced conditioned place preference in the rats. Neurosci Lett 660: 1–5. Ammoun S, Holmqvist T, Shariatmadari R et al. (2003). Distinct recognition of OXl and Ox2; receptors by orexin peptides. J Pharmacol Exp Ther 305: 507. Anand BK, Brobeck J (1951). Hypothalamic control of food intake in rats and cats. Yale J Biol Med 24: 123–140. Anderson EM, Wissman AM, Chemplanikal J et al. (2017). BDNF-Trk B controls cocaine-induced dendritic spines in rodent nucleus accumbens dissociated from increases in addictive behaviors. Proc Natl Acad Sci U S A 114: 9469–9474. Anderson RI, Moorman DE, Becker HC (2018). Contribution of dynorphin and orexin neuropeptide systems to the motivational effects of alcohol. Handb Exp Pharmacol 248: 473–503. Assar N, Mahmoudi D, Mousavi Z et al. (2019). Role of orexin-1 and -2 receptors within the nucleus accumbens in the acquisition of sensitization to morphine in rats. Behav Brain Res 373: 112090. Aston-Jones G, Smith RJ, Sartor GC et al. (2010). Lateral hypothalamic orexin/hypocretin neurons: a role in reward-seeking and addiction. Brain Res 1314: 74–90.

369

Azizbeigi R, Haghparast A (2019). Involvement of orexin-2 receptor in the ventral tegmental area in stress- and drug priming-induced reinstatement of conditioned place preference in rats. Neurosci Lett 696: 121–126. https://doi.org/ 10.1016/j.neulet.2018.12.029. Epub; %2018 Dec %20. Azizbeigi R, Farzinpour Z, Haghparast A (2019). Role of Orexin-1 receptor within the ventral tegmental area in mediating stress- and morphine priming-induced reinstatement of conditioned place preference in rats. Basic Clin Neurosci 10: 373–382. Baimel C, Bartlett SE, Chiou LC et al. (2015). Orexin/hypocretin role in reward: Implications for opioid and other addictions. Br J Pharmacol 172: 334–348. Bali A, Randhawa PK, Jaggi AS (2015). Stress and opioids: role of opioids in modulating stress-related behavior and effect of stress on morphine conditioned place preference. Neurosci Biobehav Rev 51: 138–150. https://doi.org/10.1016/ j.neubiorev.2014.12.018. Epub;%2015 Jan 27. Barateau L, Jaussent I, Lopez R et al. (2016). Smoking, alcohol, drug use, abuse and dependence in narcolepsy and idiopathic hypersomnia: a case-control study. Sleep 39: 573–580. Bayard S, Dauvilliers YA (2013). Reward-based behaviors and emotional processing in human with narcolepsycataplexy. Front Behav Neurosci 7: 50. Beitner-Johnson D, Guitart X, Nestler EJ (1992). Neurofilament proteins and the mesolimbic dopamine system: common regulation by chronic morphine and chronic cocaine in the rat ventral tegmental area. J Neurosci 12: 2165–2176. Beitner-Johnson D, Guitart X, Nestler EJ (1993). Glial fibrillary acidic protein and the mesolimbic dopamine system: regulation by chronic morphine and Lewis-Fischer strain differences in the rat ventral tegmental area. J Neurochem 61: 1766–1773. Bentzley BS, Aston-Jones G (2015). Orexin-1 receptor signaling increases motivation for cocaine-associated cues. Eur J Neurosci 41: 1149–1156. Beswick T, Best D, Bearn J et al. (2003). The effectiveness of combined naloxone/lofexidine in opiate detoxification: results from a double-blind randomized and placebocontrolled Trial. Am J Addict 12: 295–305. Bhagwandin A, Gravett N, Hemingway J et al. (2011). Orexinergic neuron numbers in three species of African mole rats with rhythmic and arrhythmic chronotypes. Neuroscience 199: 153–165. Blouin AM, Thannickal TC, Worley PF et al. (2005). Narp immunostaining of human hypocretin (orexin) neurons: loss in narcolepsy. Neurology 65: 1189–1192. Blouin AM, Fried I, Wilson CL et al. (2013). Human hypocretin and melanin-concentrating hormone levels are linked to emotion and social interaction. Nat Commun 4: 1547. Borgland SL, Chang SJ, Bowers MS et al. (2009). Orexin A/hypocretin-1 selectively promotes motivation for positive reinforcers. J Neurosci 29: 11215–11225. Boutrel B, De Lecea L (2008). Addiction and arousal: the hypocretin connection. Physiol Behav 93: 947–951. Brown MA, Guilleminault C (2011). A review of sodium oxybate and baclofen in the treatment of sleep disorders. Curr Pharm Des 17: 1430–1435.

370

R. MCGREGOR ET AL.

Brundin L, Bjorkqvist M, Petersen A et al. (2007). Reduced orexin levels in the cerebrospinal fluid of suicidal patients with major depressive disorder. Eur Neuropsychopharmacol 17: 573–579. Brundin L, Bjorkqvist M, Traskman-Bendz L et al. (2009). Increased orexin levels in the cerebrospinal fluid the first year after a suicide attempt. J Affect Disord 113: 179–182. C.D.C (2017). “search “C.D.C. drug overdose epidemic””, Centers for Disease Control and Prevention. Cai NS, Quiroz C, Bonaventura J et al. (2019). Opioid-galanin receptor heteromers mediate the dopaminergic effects of opioids. J Clin Invest 129: 2730–2744. Chang KC, Wang JD, Saxon A et al. (2017). Causes of death and expected years of life lost among treated opioiddependent individuals in the United States and Taiwan. Int J Drug Policy 43: 1–6. https://doi.org/10.1016/j.drugpo. 2016.12.003. Epub;%2017 Feb 1. Chemelli RM, Willie JT, Sinton CM et al. (1999). Narcolepsy in orexin knockout mice: molecular genetics of sleep regulation. Cell 98: 437–451. Chen CT, Dun SL, Kwok EH et al. (1999). Orexin A-like immunoreactivity in the rat brain. Neurosci Lett 260: 161–164. Chen SA, O’Dell LE, Hoefer ME et al. (2006). Unlimited access to heroin self-administration: independent motivational markers of opiate dependence. Neuropsychopharmacology 31: 2692–2707. Chen J, Zhang R, Chen X et al. (2015). Heterodimerization of human orexin receptor 1 and kappa opioid receptor promotes protein kinase A/cAMP-response element binding protein signaling via a Galphas-mediated mechanism. Cell Signal 27: 1426–1438. Chou TC, Lee CE, Lu J et al. (2001). Orexin (hypocretin) neurons contain dynorphin. J Neurosci 21: RC168. Christie MJ (2008). Cellular neuroadaptations to chronic opioids: tolerance, withdrawal and addiction. Br J Pharmacol 2008/04/14: 384–396. Crocker A, Espana RA, Papadopoulou M et al. (2005). Concomitant loss of dynorphin, NARP, and orexin in narcolepsy. Neurology 65: 1184–1188. Darke S, Kaye S, Duflou J et al. (2019). Completed suicide among methamphetamine users: a national study. Suicide Life Threat Behav 49: 328–337. Date Y, Ueta Y, Yamashita H et al. (1999). Orexins, orexigenic hypothalamic peptides, interact with autonomic, neuroendocrine and neuroregulatory systems. Proc Natl Acad Sci U S A 96: 748–753. Dauvilliers Y, Montplaisir J, Cochen V et al. (2010). PostH1N1 narcolepsy-cataplexy. Sleep 33: 1428–1430. Dauvilliers Y, Lopez R, Ohayon M et al. (2013). Hypersomnia and depressive symptoms: methodological and clinical aspects. BMC Med 11: 78. Dauvilliers Y, Siegel JM, Lopez R et al. (2014). Cataplexy: clinical aspects, pathophysiology and management strategy. Nat Rev Neurol 10: 386–395. De Lecea L, Kilduff T, Peyron C et al. (1998). The hypocretins: hypothalamus-specific peptides with neuroexcitatory activity. Proc Natl Acad Sci U S A 95: 322–327.

Del Bello F, Diamanti E, Giannella M et al. (2013). Exploring multitarget interactions to reduce opiate withdrawal syndrome and psychiatric comorbidity. ACS Med Chem Lett 4: 875–879. Del Cid-Pellitero E, Garzon M (2014). Hypocretin1/orexinAimmunoreactive axons form few synaptic contacts on rat ventral tegmental area neurons that project to the medial prefrontal cortex. BMC Neurosci 15: 105. Dell LA, Patzke N, Bhagwandin A et al. (2012). Organization and number of orexinergic neurons in the hypothalamus of two species of Cetartiodactyla: a comparison of giraffe (Giraffa camelopardalis) and harbour porpoise (Phocoena phocoena). J Chem Neuroanat 2012/06/08: 98–109. Dell LA, Kruger JL, Pettigrew JD et al. (2013). Cellular location and major terminal networks of the orexinergic system in the brain of two megachiropterans. J Chem Neuroanat 53: 64–71. Dell LA, Karlsson KA, Patzke N et al. (2016a). Organization of the sleep-related neural systems in the brain of the minke whale (Balaenoptera acutorostrata). J Comp Neurol 524: 2018–2035. Dell LA, Patzke N, Spocter MA et al. (2016b). Organization of the sleep-related neural systems in the brain of the river hippopotamus (Hippopotamus amphibius): a most unusual cetartiodactyl species. J Comp Neurol 524: 2036–2058. Dell LA, Patzke N, Spocter MA et al. (2016c). Organization of the sleep-related neural systems in the brain of the harbour porpoise (Phocoena phocoena). J Comp Neurol 524: 1999–2017. Editors (2016). US drug overdose deaths: a global challenge. Lancet 387: 404–6736. Eriksson KS, Sergeeva O, Brown RE et al. (2001). Orexin/ hypocretin excites the histaminergic neurons of the tuberomammillary nucleus. J Neurosci 21: 9273–9279. Farahimanesh S, Zarrabian S, Haghparast A (2017). Role of orexin receptors in the ventral tegmental area on acquisition and expression of morphine-induced conditioned place preference in the rats. Neuropeptides 17: 10. Farzinpour Z, Taslimi Z, Azizbeigi R et al. (2019). Involvement of orexinergic receptors in the nucleus accumbens, in the effect of forced swim stress on the reinstatement of morphine seeking behaviors. Behav Brain Res 356: 279–287. https://doi.org/10.1016/j.bbr.2018.08.021. Epub;%2018 Sep 5. Fronczek R, Overeem S, Lee SY et al. (2007). Hypocretin (orexin) loss in Parkinson’s disease. Brain 130: 1577–1585. Galloway GP, Frederick SL, Staggers FEJ et al. (1997). Gamma-hydroxybutyrate: an emerging drug of abuse that causes physical dependence [see comments]. Addiction 92: 89–96. Georgescu D, Zachariou V, Barrot M et al. (2003). Involvement of the lateral hypothalamic peptide orexin in morphine dependence and withdrawal. J Neurosci 23: 3106–3111. Gossop M, Bradley B (1984). Insomnia among addicts during supervised withdrawal from opiates: a comparison of oral methadone and electrostimulation. Drug Alcohol Depend 13: 191–198.

PLEASURE, ADDICTION, AND HYPOCRETIN (OREXIN) Guilleminault C, Cao MT (2011). Narcolepsy: diagnosis and management. In: MH Kryger, T Roth, WC Dement (Eds.), Principles and Practice of Sleep Medicine, Fifth edn. Elsevier Saunders, Missouri, pp. 957–968. Guo SJ, Cui Y, Huang ZZ et al. (2016). Orexin A-mediated AKT signaling in the dentate gyrus contributes to the acquisition, expression and reinstatement of morphine-induced conditioned place preference. Addic Biol 21: 547–559. Hajek M, Meier-Ewert K, Wirz-Justice A et al. (1989). Bright white light does not improve narcoleptic symptoms. Eur Arch Psychiatry Neurol Sci 238: 203–207. Han F, Lin L, Warby SC et al. (2011). Narcolepsy onset is seasonal and increased following the 2009 H1N1 pandemic in China. Ann Neurol 70: 410–417. Harris GC, Aston-Jones G (2006). Arousal and reward: a dichotomy in orexin function. Trends Neurosci 29: 571–577. Harris GC, Wimmer M, Aston-Jones G (2005). A role for lateral hypothalamic orexin neurons in reward seeking. Nature 437: 556–559. Harris GC, Wimmer M, Randall-Thompson JF et al. (2007). Lateral hypothalamic orexin neurons are critically involved in learning to associate an environment with morphine reward. Behav Brain Res 183: 43–51. Hassani OK, Krause MR, Mainville L et al. (2016). Orexin neurons respond differentially to auditory cues associated with appetitive versus aversive outcomes. J Neurosci 36: 1747–1757. Ho CY, Berridge KC (2013). An orexin hotspot in ventral pallidum amplifies hedonic ‘liking’ for sweetness. Neuropsychopharmacology 38: 1655–1664. Honda Y, Doi Y, Juji T et al. (1984). Narcolepsy and HLA: Positive DR2 as a prerequisite for the development of narcolepsy. Folia Psychiatr Neurol Jpn 38: 360. Horvath TL, Peyron C, Diano S et al. (1999). Hypocretin (orexin) activation and synaptic innervation of the locus coeruleus noradrenergic system. J Comp Neurol 415: 145–159. Jalal H, Buchanich JM, Roberts MS et al. (2018). Changing dynamics of the drug overdose epidemic in the United States from 1979 through 2016. Science 361: eaau1184. James AS, Chen JY, Cepeda C et al. (2013). Opioid selfadministration results in cell-type specific adaptations of striatal medium spiny neurons. Behav Brain Res 256: 279–283. James MH, Mahler SV, Moorman DE et al. (2017). A decade of orexin/hypocretin and addiction: where are we now? Curr Top Behav Neurosci 33: 247–281. https://doi.org/ 10.1007/7854_2016_57. James MH, Stopper CM, Zimmer BA et al. (2018). Increased number and activity of a lateral subpopulation of hypothalamic orexin/hypocretin neurons underlies the expression of an addicted state in rats. Biol Psychiatry. https://doi. org/10.1016/j.biopsych.2018.07.022. [Epub ahead of print]. James MH, Stopper CM, Zimmer BA et al. (2019). Increased number and activity of a lateral subpopulation of hypothalamic orexin/hypocretin neurons underlies the expression of an addicted state in rats. Biol Psychiatry. https://doi.org/ 10.1016/j.biopsych.2018.07.022. [Epub ahead of print].

371

Kiyashchenko LI, Mileykovskiy BY, Maidment N et al. (2002). Release of hypocretin (orexin) during waking and sleep states. J Neurosci 22: 5282–5286. Kukkonen JP, Leonard CS (2014). Orexin/hypocretin receptor signalling cascades. Br J Pharmacol 171: 314–331. Lee MJ, Lee SY, Yuan SS et al. (2016). Comorbidity of narcolepsy and depressive disorders: a nationwide populationbased study in Taiwan. Sleep Med 39: 95–100. Li, Y. & Van Den Pol A. 2006, Dynorphin inhibits hypocretin/ orexin neurons in hypothalamic brain slice", Abstract viewer/Itinerary Planner Society for Neuroscience p. 157. 19/V5. Li Y, van den Pol AN (2008). Mu-opioid receptor-mediated depression of the hypothalamic hypocretin/orexin arousal system. J Neurosci 28: 2814–2819. Li Y, Gao XB, Sakurai T et al. (2002). Hypocretin/orexin excites hypocretin neurons via a local glutamate neuron potential mechanism for orchestrating the hypothalamic arousal system. Neuron 36: 1169–1181. Lin L, Faraco J, Kadotani H et al. (1999). The sleep disorder canine narcolepsy is caused by a mutation in the hypocretin (orexin) receptor gene. Cell 98: 365–376. Lin SK, Pan CH, Chen CH (2014). A double-blind, placebocontrolled trial of dextromethorphan combined with clonidine in the treatment of heroin withdrawal. J Clin Psychopharmacol 34: 508–512. Lofwall MR, Babalonis S, Nuzzo PA et al. (2013). Efficacy of extended-release tramadol for treatment of prescription opioid withdrawal: a two-phase randomized controlled trial. Drug Alcohol Depend 133: 188–197. Lu J, Zhao J, Balesar R et al. (2017). Sexually Dimorphic Changes of Hypocretin (Orexin) in Depression. EBioMedicine 2017/03/31: 311–319. Lutter M, Krishnan V, Russo SJ et al. (2008). Orexin signaling mediates the antidepressant-like effect of calorie restriction. J Neurosci 28: 3071–3075. Lutz PE, Ayranci G, Chu-Sin-Chung P et al. (2014). Distinct mu, delta, and kappa opioid receptor mechanisms underlie low sociability and depressive-like behaviors during heroin abstinence. Neuropsychopharmacology 39: 2694–2705. Marcus JN, Aschkenasi CJ, Lee CE et al. (2001). Differential expression of orexin receptors 1 and 2 in the rat brain. J Comp Neurol 435: 6–25. Matzeu A, Martin-Fardon R (2018). Drug seeking and relapse: new evidence of a role for orexin and dynorphin co-transmission in the paraventricular nucleus of the thalamus. Front Neurol 9: 720. eCollection;%2018. https://doi. org/10.3389/fneur.2018.00720. McGregor R, Wu M-F, Barber G et al. (2011). Highly specific role of hypocretin (orexin) neurons: differential activation as a function of diurnal phase, operant reinforcement vs. operant avoidance and light level. J Neurosci 31: 15455–15467. McGregor R, Shan L, Wu MF et al. (2017). Diurnal fluctuation in the number of hypocretin/orexin and histamine producing: Implication for understanding and treating neuronal loss. PLoS One 12.

372

R. MCGREGOR ET AL.

McLellan A, Lewis DC, O’Brien CP et al. (2000). Drug dependence, a chronic medical illness: Implications for treatment, insurance, and outcomes evaluation. JAMA 284: 1689–1695. McNamara R, Dalley JW, Robbins TW et al. (2010). Trait-like impulsivity does not predict escalation of heroin selfadministration in the rat. Psychopharmacology (Berl) 212: 453–464. Meye FJ, van Zessen R, Smidt MP et al. (2012). Morphine withdrawal enhances constitutive opioid receptor activity in the ventral tegmental area. J Neurosci 32: 16120–16128. Mignot E, Reid M, Tafti M et al. (1995). Local administration of dopaminergic drugs into the ventral tegmental area modulates cataplexy in narcoleptic canines. Sleep Res 24: 298. Mignot E, Lin L, Rogers W et al. (2001). Complex HLA-DR and -DQ interactions confer risk of narcolepsy-cataplexy in three ethnic groups. Am J Hum Genet 68: 686–699. Mileykovskiy BY, Kiyashchenko LI, Siegel JM (2005). Behavioral correlates of activity in identified hypocretin/ orexin neurons. Neuron 46: 787–798. Muschamp JW, Hollander JA, Thompson JL et al. (2014). Hypocretin (orexin) facilitates reward by attenuating the antireward effects of its cotransmitter dynorphin in ventral tegmental area. Proc Natl Acad Sci U S A 111: E1648–E1655. Narayanan S, Lam H, Christian L et al. (2004). Endogenous opioids mediate basal hedonic tone independent of dopamine D-1 or D-2 receptor activation. Neuroscience 124: 241–246. Nestler EJ (2013). Cellular basis of memory for addiction. Dialogues Clin Neurosci 15: 431–443. Nestler EJ, Barrot M, DiLeone RJ et al. (2002). Neurobiology of depression. Neuron 34: 13–25. Nishino S, Mignot E (2011). Narcolepsy and cataplexy. Handb Clin Neurol 99: 783–814. Nordstrand SEH, Hansen BH, Rootwelt T et al. (2019). Psychiatric symptoms in patients with post-H1N1 narcolepsy type 1 in Norway. Sleep zsz008. Olateju OI, Bhagwandin A, Ihunwo AO et al. (2017). Changes in the cholinergic, catecholaminergic, orexinergic and serotonergic structures forming part of the sleep systems of adult mice exposed to intrauterine alcohol. Front Neuroanat 11: 110. eCollection;%2017. https://doi.org/ 10.3389/fnana.2017.00110. Ostling PS, Davidson KS, Anyama BO et al. (2018). America’s opioid epidemic: a comprehensive review and look into the rising crisis. Curr Pain Headache Rep 22: 32–0685. Peyron C, Tighe DK, van den Pol AN et al. (1998). Neurons containing hypocretin (orexin) project to multiple neuronal systems. J Neurosci 18: 9996–10015. Peyron C, Faraco J, Rogers W et al. (2000). A mutation in a case of early onset narcolepsy and a generalized absence of hypocretin peptides in human narcoleptic brains. Nat Med 6: 991–997. Picetti R, Caccavo JA, Ho A et al. (2012). Dose escalation and dose preference in extended-access heroin self-administration in Lewis and Fischer rats. Psychopharmacology (Berl) 220: 163–172.

Pillay S, Bhagwandin A, Bertelsen MF et al. (2017). Regional distribution of cholinergic, catecholaminergic, serotonergic and orexinergic neurons in the brain of two carnivore species: the feliform banded mongoose (Mungos mungo) and the caniform domestic ferret (Mustela putorius furo). J Chem Neuroanat 82: 12–28. Ponz A, Khatami R, Poryazova R et al. (2010a). Reduced amygdala activity during aversive conditioning in human narcolepsy. Ann Neurol 67: 394–398. Ponz A, Khatami R, Poryazova R et al. (2010b). Abnormal activity in reward brain circuits in human narcolepsy with cataplexy. Ann Neurol 67: 190–200. Pourhamzeh M, Mozafari R, Jamali S et al. (2019). Involvement of orexin receptors within the hippocampal dentate gyrus in morphine-induced reinstatement in food-deprived rats. Behav Brain Res 375: 112155. https://doi.org/10.1016/j.bbr.2019.112155. Epub;%2019 Aug 15. Sadeghi B, Ezzatpanah S, Haghparast A (2016). Effects of dorsal hippocampal orexin-2 receptor antagonism on the acquisition, expression, and extinction of morphine-induced place preference in rats. Psychopharmacology (Berl) 233: 2329–2341. Sadeghzadeh F, Namvar P, Naghavi FS et al. (2016). Differential effects of intra-accumbal orexin-1 and -2 receptor antagonists on the expression and extinction of morphine-induced conditioned place preference in rats. Pharmacol Biochem Behav 142: 8–14. https://doi.org/ 10.1016/j.pbb.2015.12.005. Epub;%2015 Dec 17. Sakurai T, Amemiya A, Ishii M et al. (1998). Orexins and orexin receptors: a family of hypothalamic neuropeptides and G protein-coupled receptors that regulate feeding behavior. Cell 92: 573–585. Sarti F, Borgland SL, Kharazia VN et al. (2002). Acute cocaine exposure alters spine density and long-term potentiation in the ventral tegmental area. Eur J Neurosci 26: 749–756. Scammell TE (2006). The frustrating and mostly fruitless search for an autoimmune cause of narcolepsy. Sleep 29: 601–660. Schilstrom B, Svensson HM, Svensson TH et al. (1998). Nicotine and food induced dopamine release in the nucleus accumbens of the rat: putative role of alpha7 nicotinic receptors in the ventral tegmental area. Neuroscience 85: 1005–1009. Schwartz S, Ponz A, Poryazova R et al. (2007). Abnormal activity in hypothalamus and amygdala during humour processing in human narcolepsy with cataplexy. Brain 131: 514–522. Schwarzer C (2009). 30 years of dynorphins–new insights on their functions in neuropsychiatric diseases. Pharmacol Ther 2009/05/28: 353–370. Sharf R, Guarnieri DJ, Taylor JR et al. (2010). Orexin mediates morphine place preference, but not morphine-induced hyperactivity or sensitization. Brain Res 1317: 24–32. Shi J, Li Sx, Zhang Xl et al. (2009). Time-dependent neuroendocrine alterations and drug craving during the first month of abstinence in heroin addicts. Am J Drug Alcohol Abuse 35: 267–272.

PLEASURE, ADDICTION, AND HYPOCRETIN (OREXIN) Shirazy M, RayatSanati K, Jamali S et al. (2020). Role of orexinergic receptors in the dentate gyrus of the hippocampus in the acquisition and expression of morphine-induced conditioned place preference in rats. Behav Brain Res 379: 112349. https://doi.org/10.1016/j.bbr.2019.112349. Epub; %2019 Nov 9. Shoblock JR, Maidment NT (2006). Constitutively active micro opioid receptors mediate the enhanced conditioned aversive effect of naloxone in morphine-dependent mice. Neuropsychopharmacology 31: 171–177. Shoblock JR, Maidment NT (2007). Enkephalin release promotes homeostatic increases in constitutively active mu opioid receptors during morphine withdrawal. Neuroscience 149: 642–649. Siegel JM (1979). Behavioral functions of the reticular formation. Brain Res Rev 1: 69–105. Siegel JM (2004). Hypocretin (orexin): role in normal behavior and neuropathology. Annu Rev Psychol 55: 125–148. Siegel JM, McGinty DJ (1976). Brainstem neurons without spontaneous unit discharge. Science 193: 240–242. Siegel JM, McGinty DJ (1977). Pontine reticular formation neurons: relationship of discharge to motor activity. Science 196: 678–680. Siegel JM, Tomaszewski KS (1983). Behavioral organization of reticular formation: studies in the unrestrained cat. I. Cells related to axial, limb, eye, and other movements. J Neurophysiol 50: 696–716. Siegel JM, Wheeler RL, McGinty DJ (1979). Activity of medullary reticular formation neurons in the unrestrained cat during waking and sleep. Brain Res 179: 49–60. Siegel JM, Wheeler RL, Breedlove SM et al. (1980). Brainstem units related to movements of the pinna. Brain Res 202: 183–188. Siegel JM, Tomaszewski KS, Wheeler RL (1983). Behavioral organization of reticular formation: Studies in the unrestrained cat: II. Cells related to facial movements. J Neurophysiol 50: 717–723. Siegel JM, Nienhuis R, Gulyani S et al. (1999). Neuronal degeneration in canine narcoleps. J Neurosci 19: 248–257. Siegel JM, Moore R, Thannickal T et al. (2001). A brief history of hypocretin/orexin and narcolepsy. Neuropsychopharmacology 25: S14–S20. Sigmon SC, Bisaga A, Nunes EV et al. (2012). Opioid detoxification and naltrexone induction strategies: recommendations for clinical practice. Am J Drug Alcohol Abuse 38: 187–199. Sim-Selley LJ, Cassidy MP, Sparta A et al. (2011). Effect of DeltaFosB overexpression on opioid and cannabinoid receptor-mediated signaling in the nucleus accumbens. Neuropharmacology 61: 1470–1476. Skoubis PD, Lam HA, Shoblock J et al. (2005). Endogenous enkephalins, not endorphins, modulate basal hedonic state in mice. Eur J Neurosci 21: 1379–1384. Smart D, Jerman JC, Brough SJ et al. (1999). Characterization of recombinant human orexin receptor pharmacology in a Chinese hamster ovary cell-line using FLIPR. Br J Pharmacol 128: 1–3. Smith RJ, Aston-Jones G (2012). Orexin-hypocretin 1 receptor antagonist reduces heroin self-administration and cueinduced heroin seeking. Eur J Neurosci 35: 798–804.

373

Stefano GB, Kream RM (2007). Endogenous morphine synthetic pathway preceded and gave rise to catecholamine synthesis in evolution (Review). Int J Mol Med 20: 837–841. Stoeber M, Jullie D, Lobingier BT et al. (2018). A genetically encoded biosensor reveals location bias of opioid drug action. Neuron 98: 963–976. Tabaeizadeh M, Motiei-Langroudi R, Mirbaha H et al. (2013). The differential effects of OX1R and OX2R selective antagonists on morphine conditioned place preference in naive versus morphine-dependent mice. Behav Brain Res 237: 41–48. https://doi.org/10.1016/j.bbr.2012.09.010. Epub;%2012 Sep 17. Teitelbaum P, Epstein AN (1962). The lateral hypothalamic syndrome: recovery of feeding and drinking after lateral hypothalamic lesions. Psychol Rev 69: 74–90. https://doi. org/10.1037/h0039285. Thannickal TC, Moore RY, Aldrich M et al. (2000a). Human narcolepsy is linked to reduced number, size and synaptic bouton density in hypocretin-2 labeled neurons. Abstr Soc Neurosci 26: 2061. Thannickal TC, Moore RY, Nienhuis R et al. (2000b). Reduced number of hypocretin neurons in human narcolepsy. Neuron 27: 469–474. Thannickal TC, Siegel JM, Moore RY (2003). Pattern of hypocretin (orexin) soma and axon loss, and gliosis, in human narcolepsy. Brain Pathol 13: 340–351. Thannickal TC, Lai YY, Siegel JM (2007). Hypocretin (orexin) cell loss in Parkinson’s disease. Brain 130: 1586–1595. Thannickal TC, Lai YY, Siegel JM (2008). Hypocretin (orexin) and melanin concentrating hormone loss and the symptoms of Parkinson’s disease. Brain 131: e87. Thannickal TC, John J, Shan L et al. (2018). Opiates increase the number of hypocretin-producing cells in mouse and human brain, and reverse cataplexy in a mouse model of narcolepsy. Sci Transl Med 10: eaao4953. https://doi.org/ 10.1126/scitranslmed.aao4953. Torrealba F, Yanagisawa M, Saper CB (2003). Colocalization of orexin a and glutamate immunoreactivity in axon terminals in the tuberomammillary nucleus in rats. Neuroscience 119: 1033–1044. Turner C, Chandrakumar D, Rowe C et al. (2018). Crosssectional cause of death comparisons for stimulant and opioid mortality in San Francisco, 2005. Drug Alcohol Depend 185: 305–312. Valentino RJ, Volkow ND (2020). Drugs, sleep, and the addicted brain. Neuropsychopharmacology 45: 3–5. van den Pol AN (1999). Hypothalamic hypocretin (orexin): robust innervation of the spinal cord. J Neurosci 19: 3171–3182. Volkow ND, Woodcock J, Compton WM et al. (2018). Medication development in opioid addiction: meaningful clinical end points. Sci Transl Med 10: eaan2595. Wu MF, Nienhuis R, Maidment N et al. (2011a). Cerebrospinal fluid hypocretin (orexin) levels are elevated by play but are not raised by exercise and its associated heart rate, blood pressure, respiration or body temperature changes. Arch Ital Biol 149: 492–498.

374

R. MCGREGOR ET AL.

Wu MF, Nienhuis R, Maidment N et al. (2011b). Role of the hypocretin (orexin) receptor 2 (Hcrt-r2) in the regulation of hypocretin level and cataplexy. J Neurosci 31: 6305–6310. Yamanaka A, Tabuchi S, Tsunematsu T et al. (2010). Orexin directly excites orexin neurons through orexin 2 receptor. J Neurosci 30: 12642–12652. Zarepour L, Fatahi Z, Sarihi A et al. (2014). Blockade of orexin-1 receptors in the ventral tegmental area could attenuate the lateral hypothalamic stimulation-induced

potentiation of rewarding properties of morphine. Neuropeptides 48: 179–185. Zarrabian S, Riahi E, Karimi S et al. (2020). The potential role of the orexin reward system in future treatments for opioid drug abuse. Brain Res 1731: 146028. https://doi.org/10.1016/ j.brainres.2018.11.023. Epub;%2018 Nov 23., p. 146028. Zhu Y, Wienecke CF, Nachtrab G et al. (2016). A thalamic input to the nucleus accumbens mediates opiate dependence. Nature 530: 219–222.

Section 12 Tuberomamillary complex

This page intentionally left blank

Handbook of Clinical Neurology, Vol. 180 (3rd series) The Human Hypothalamus: Middle and Posterior Region D.F. Swaab, F. Kreier, P.J. Lucassen, A. Salehi, and R.M. Buijs, Editors https://doi.org/10.1016/B978-0-12-820107-7.00023-9 Copyright © 2021 Elsevier B.V. All rights reserved

Chapter 23

Histamine receptors, agonists, and antagonists in health and disease PERTTI PANULA* Department of Anatomy, University of Helsinki, Helsinki, Finland

Abstract Histamine in the brain is produced by a group of tuberomamillary neurons in the posterior hypothalamus and a limited number of mast cells in different parts of the brain. Four G-protein-coupled receptors mediate the effects of histamine. Two of these receptors, H3 and H4 receptors, are high-affinity receptors in the brain and immune system, respectively. The two classic histamine receptors, H1 receptor and H2 receptor, are well known as drug targets for allergy and gastric ulcer, respectively. These receptors have lower affinity for histamine than the more recently discovered H3 and H4 receptors. The H1 and H2 receptors are important postsynaptic receptors in the brain, and they mediate many of the central effects of histamine on, e.g., alertness and wakefulness. H3 receptor is a pre- and postsynaptic receptor, which regulates release of histamine and several other neurotransmitters, including serotonin, GABA, and glutamate. H4 receptor is found in cerebral blood vessels and microglia, but its expression in neurons is not yet well established. Pitolisant, a H3 receptor antagonist, is used to treat narcolepsy and hypersomnia. H1 receptor antagonists have been used to treat insomnia, but its use requires precautions due to potential side effects. H2 receptor antagonists have shown efficacy in treatment of schizophrenia, but they are not in widespread clinical use. H4 receptor ligands may in the future be tested for neuroimmunological disorders and potentially neurodegenerative disorders in which inflammation plays a role, but clinical tests have not yet been initiated.

HISTAMINE AND HISTAMINE RECEPTORS Histamine is a biogenic amine present in almost all tissues in mammals, including humans. It is produced by a single-step decarboxylation of L-histidine, which is transported in the cells by L-amino acid transporter, by L-histidine decarboxylase (HDC, EC 4.1.1.22), and metabolized by oxidation by diamine-oxidase (DAO, EC 1.4.3.22) mostly in peripheral tissues and by methylation by histamine-N-methyltransferase (HNMT, EC 2.1.1.8) in the brain and some other tissues (Haas and Panula, 2003). Histamine in packed in different types of secretory vesicles in mast cells, basophils, enterochromaffinlike cells (Panula et al., 1985) of the stomach and brain neurons (Panula et al., 1984) by vesicular monoamine

transporter 2 (Anlauf et al., 2006) and secreted by these so-called professional cells. Nascent histamine is also produced in developing tissues by so-called nonprofessional cells, which do not load histamine in secretory granules (Panula et al., 2014). Vesicular histamine is stored in neuronal granules both in the cell bodies and varicose processes (Fig. 23.1) (Kukko-Lukjanov and Panula, 2003). Whereas histamine released from secretory granules into extracellular space mediates its actions through four G protein-coupled receptors (Panula et al., 2015), the actions and roles of nonvesicularly stored histamine are not fully known. The steps of production, uptake in granules, release, receptor activation, and degradation of histamine are subject to regulation by drugs specific for several proteins and genetic variation in

*Correspondence to: Pertti Panula, Department of Anatomy, University of Helsinki, Haartmaninkatu 8, 00290 Helsinki, Finland. Tel: +358-40-592-2323, E-mail: [email protected]

378

P. PANULA

Fig. 23.1. Histaminergic neurons in vitro and in vivo. (A–C) Brightly fluorescent granular histamine immunofluorescence is seen in both the cell body and processes of a cultured rat tuberomamillary neurons (A). The processes are identified as dendrites by microtubule-associated protein 2 (MAP2) immunoreactivity (B). When both are visualized simultaneously, the presence of granular histamine immunofluorescence is seen clearly in MAP2-immunoreactive dendritic varicosities (C). Original data published in Kukko-Lukjanov and Panula (2003). (D) Some of the histamine-immunoreactive neurons in the basal part of the tuberomamillary nucleus of the rat lie attached to the basal surface of the brain. Original data published in Panula et al. (1984). (E) An abnormally dense network of histamine-immunoreactive nerve fibers is seen among neuromelanin-containing dopaminergic neurons in the substantia nigra of a brain of a PD patients. Original data published in Anichtchik et al. (2000).

corresponding genes. Among these are histidine decarboxylase (HDC, necessary for histamine synthesis) (Schayer, 1957), vesicular monoamine transporter 2 (VMAT2, necessary for vesicular storage of histamine) (Erickson et al., 1992), histamine N-methyltransferase (HNMT, particularly important for histamine inactivation in the brain) (Snyder and Axelrod, 1965), and diamino oxidase (DAO, important for histamine oxidation in peripheral organs) (Kapeller-Adler, 1949). In the brain, histamine is produced by tuberomamillary neurons in the posterior hypothalamus (Fig. 23.2), mast cells in limited areas in the brain (dural perivascular space and subpial areas of the human brain, median eminence and dorsal thalamus in the rat brain), ependymal cells, and at least in vitro by some microglia. Binding affinity of histamine to H1 and H2 receptors is much lower than to H3 and H4 receptors likely due to the different transmembrane domain structures of these G protein-coupled receptors (Panula et al., 2015). Distribution of histamine receptor mRNAs generally correlates poorly with corresponding receptor radioligand binding distribution, because receptor mRNAs are concentrated in the neuronal cell bodies, whereas receptor activity and ligand binding occur mostly in dendrites and axon terminals, where receptor proteins are transported (Haas and Panula, 2003). Although all histaminergic neurons are

concentrated in the tuberomamillary neurons of posterior hypothalamus, uptake of L-histidine, and synthesis of histamine occurs in fibers and terminals in processes of these neurons in widespread areas of the brain (Haas and Panula, 2003). Gene-modified animals lacking HDC, HNMT, and any of the four G protein-coupled receptors (GPCR:s) have been produced and characterized (Panula et al., 2015; Naganuma et al., 2017). All these mice display abnormalities in behavior related to the functions of histamine as a neurotransmitter in the brain. Many of the abnormalities are related to sleep–wake cycle, novel object recognition, alertness, or aggression. A family with a dominantly inherited mutation in HDC gene, consisting of a father and eight children, has been reported to have Tourette syndrome (Ercan-Sencicek et al., 2010).

HISTAMINERGIC TUBEROMAMILLARY NEURONS AND BRAIN HISTAMINE The detailed anatomy of the tuberomamillary histaminergic neurons was first described in the rat brain based on immunocytochemical demonstration of the synthesizing enzyme histidine decarboxylase (Watanabe et al., 1984) and independently with antibodies against histamine itself (Panula et al., 1984). Subsequently, the presence

HISTAMINE RECEPTORS, AGONISTS, AND ANTAGONISTS IN HEALTH AND DISEASE

379

Fig. 23.2. Histaminergic neurons in the human brain. Original data published in (Panula et al., 1990). (A) Histamineimmunoreactive fibers pass all layers (I–V) of neocortex in adult human brain. (B) The density of histamine-containing fibers is the highest in the superficial layer I of the neocortex, and some fibers extend to the pia mater. (C) Large histamineimmunoreactive neurons in normal adult human brain, and a dense network of thick and thin histamine-immunoreactive fibers in the TMN. (D) Histamine-containing neurons form a large population of neurons in the basal tuberomamillary nucleus of the hypothalamus.

of histamine in neurons at the same posterior hypothalamic domain was demonstrated in a whole range of vertebrates such as fish, amphibians, lizards, birds, and several other mammals including man (Fig. 23.2). As a generally applicable tool, antibodies against histamine are useful in demonstrating the presence of the amine in neurons (and all other HDC-expressing cells) of all species including humans (Figs. 23.1 and 23.2), whereas antibodies against HDC are species-specific and may detect other enzymes such as aromatic amino acid decarboxylase in species not related to the one used as source of immunogen. Thus the differences in the specificity of these markers should be taken into account when working on different species. The human tuberomamillary

nucleus covers a large area of posterior hypothalamus between the level of the arcuate nucleus rostrally to the level of the caudal part of the medial mamillary nucleus and rostral part of substantia nigra (Airaksinen et al., 1991) (Fig. 23.2). Morphologically the histaminergic neurons in human brain resemble the large noradrenergic, dopaminergic and serotonergic neurons (Airaksinen et al., 1991). The configuration of the tuberomamillary nucleus containing the histaminergic neurons is complicated due to many other small nuclei devoid of histaminergic neurons within the area, and thick nonhistaminergic fiber tracts that traverse the region. The tuberomamillary neurons were identified as GABAergic already before identification of their nature

380

P. PANULA

as histaminergic, when these neurons were found to supply a prominent hypothalamo-cortical pathway (Vincent et al., 1983). In several vertebrates, from fish to humans, the tuberomamillary neurons express several other neuroactive substances, such as neuropeptides galanin, thyrotropin-releasing hormone, substance P and proenkephalin A-derived peptides (Airaksinen et al., 1992; Sundvik and Panula, 2012) to variable degree of coexistence. GABA has been suggested to fine-tune the excitatory actions of histamine at target sites, possibly through extrasynaptic GABA-A receptors, but conclusive evidence of exact mechanisms is lacking. Current evidence suggests that histamine and GABA in TMN neurons are stored in different vesicles (Kukko-Lukjanov and Panula, 2003), suggesting that their release may be differently regulated. It is a demanding task to evaluate the status of the brain histaminergic system in living humans. Levels of histamine and its first metabolite, N-tele-methylhistamine in the spinal fluid can give some information. However, normal levels of histamine in spinal fluid on lumbar level is low, and some immunological methods are not specific for histamine but detect other confounding substances as well. High-pressure liquid chromatography, when the sensitivity is sufficient, and mass spectrometry are among the methods of choice. Receptor occupancy of H1 and H3 receptor can be studied with positron emission tomography using [11C]pyrilamine/[11C]doxepin (Yanai et al., 1992) or [11C]GSK189254 (Jucaite et al., 2013), respectively. Abnormalities found in the histaminergic system in human diseases include increased histamine metabolite levels in the CSF (Prell et al., 1995) and altered cortical H3 receptor radioligand binding (Jin et al., 2009) in schizophrenics, decreased histamine levels in the hypothalamus, hippocampus, and cortex (MazurkiewiczKwilecki and Nsonwah, 1989; Panula et al., 1998), and neurofibrillary tangles in histamine neurons (Airaksinen et al., 1991) in Alzheimer’s disease, increased histamine levels (Rinne et al., 2002) in basal ganglia and increased H3 receptor binding (Anichtchik et al., 2001) and

histaminergic fiber density (Anichtchik et al., 2000) in substantia nigra in Parkinson’s disease. Expression of histidine decarboxylase and histamine receptors (Shan et al., 2015) and drug trials in these and other diseases (Panula and Nuutinen, 2013) are reviewed in detail in recent reviews and described further in Chapter 24 of this volume.

HISTAMINE RECEPTORS Histamine H1 receptor Although H1 receptor is often regarded as primarily peripheral receptor due to its involvement in allergy and immunology, it is very prominently expressed in the brain. The receptor gene was cloned in 1991 (Yamashita et al., 1991), and it encodes for a protein of 487 amino acids. The crystal structure of this receptor complex with doxepin has been determined (Shimamura et al., 2011), so that the binding pockets are well known. H1 receptor is coupled mainly to Gq/11 proteins, but also to Gi/0 proteins, and its activation thus elicits production of 1,2-diacylglycerol and inositol-1,4,5-triphosphate through activation of phospholipase C (Panula et al., 2015). Binding sites for H1 receptor ligand 3[H]mepyramine are particularly abundant in cerebral cortex and infralimbic structures, and in general areas important for behavioral, nutritional, and endocrine control such as the hypothalamic ventromedial, periventricular, and suprachiasmatic nuclei (Panula et al., 2015). In human brain, particularly high densities of H1 receptor binding sites have been found in most internal layers of the neocortex, claustrum, hippocampus, and thalamus (Martinez-Mir et al., 1990). H1 receptor has also been localized in the human brain using PET and [11C]-doxepin or [11C]pyrilamine as a ligand (Yanai et al., 1992). High binding has been detected in the frontal, parietal, and temporal cortices, where clear age-dependent decrease was observed in the PET study (Yanai et al., 1992). H1 receptor mRNA is expressed particularly strongly in the deep layers IV–VI in human neocortex (Fig. 23.3A), whereas the [3H] mepyramine binding is evenly distributed through all

Fig. 23.3. Expression of histamine receptors in the gray matter of normal adult human neocortex. Exposures on film to receptorspecific radioactively labeled oligonucleotide probes. Original data published in (Jin and Panula, 2005). (A) H1 receptor mRNA expression. (B) H2 receptor mRNA expression. (C) H3 receptor mRNA expression.

HISTAMINE RECEPTORS, AGONISTS, AND ANTAGONISTS IN HEALTH AND DISEASE cortical laminae (Jin and Panula, 2005). The strong receptor radioligand binding suggests that histamine through H1 receptor may play a significant role in cognition. Significant expression of H1 receptor is also evident in human thalamus, where the strongest expression is found in the mediodorsal area, and receptor radioligand binding is the strongest in the mediodorsal and posterior parts of the thalamus (Jin et al., 2002). Histaminergic innervation of the thalamus is most prominent in areas close to the ventricular surfaces (Jin et al., 2002). The strong H1 receptor expression and ligand binding in areas controlling areas of sensorimotor integration and cognition suggest that histamine through H1 receptor controls essential functions during the wake state. The effects of first-generation antihistaminic drugs or histamine H1 antagonists caused significant drowsiness, which was in most cases an undesired side effect. However, in cases where allergic symptoms included severe itch, this effect could contribute to improved sleep quality. Since the discovery of the high affinity and selectivity at low doses of doxepin at H1 receptor, it has been tested clinically and found to be a good option as sleep aid with little side effects and no rebound insomnia, when proper doses are used, since at achievable levels it is highly selective for H1 receptor (Roth et al., 2007; Scharf et al., 2008; Krystal et al., 2011). Nevertheless, care should be taken in administration to elderly patients, first-generation antihistaminic drugs such as diphenhydramine, which may increase the risk of delirium (Rothberg et al., 2013). Due to the excitatory actions of H1 receptor, a brain-specific H1 receptor agonist could theoretically have some use in conditions requiring maintained active state. Actions of antipsychotic drugs, which have significant H1 receptor affinity, at this receptor in the hypothalamus, particularly in the ventromedial nucleus, are responsible for weight gain in psychiatric patients (Panula et al., 2015).

Histamine H2 receptor Antagonists of H2 receptor are among the best known medications for gastric and esophageal disorders, including reflux disease and gastric ulcer (Panula et al., 2015). These drugs, the best known of which are cimetidine, ranitidine, and famotidine, are polar and as such do not readily enter the brain through the blood–brain barrier. Despite the significant roles H2 receptor plays in, e.g., long-term potentiation in the hippocampus and cortex (Haas and Panula, 2003) and potential beneficial effects in brain disorders, H2 receptor has not been a target of extensive studies in brain disorders. This is followed by a very interesting case report on a schizophrenic patient and his improvement following treatment of

381

gastric ulcer with famotidine (Kaminsky et al., 1990). This patient had been isolated and with few social contacts for several years, and following hospitalization treated with haloperidol until severe extrapyramidal symptoms occurred and medication had to be discontinued. After a 3-week-long drug-free period, famotidine treatment for ulcer was initiated. This was followed by significant improvement of the mental condition, sociability, and activity. When the famotidine treatment was discontinued 6 months later due to improvement of the ulcer, the patient’s mental condition rapidly deteriorated and he discontinued work training. Reintroduction of famotidine was followed again by significant mental improvement and return to work. Availability of H2 antagonists in the brain has not been a goal for drug companies, because mental confusion was a side effect of these drugs in particularly elderly patients with impaired renal and/or hepatic functions. Nevertheless, it has been established that at the clinically relevant dose (40 mg/day) famotidine can be detected in the CSF and is thus available in the CNS (Kagevi et al., 1987). Later studies have established significant therapeutic effects of famotidine in cohorts of schizophrenics (Deutsch et al., 1993; Rosse et al., 1996; Meskanen et al., 2013). These studies suggest that H2 receptor antagonists may have significant potential in treatment of at least a subgroup of schizophrenics who suffer from negative symptoms. The H2 receptor was cloned in 1991 (Gantz et al., 1991), and the intron-less gene encodes for a protein of 359 amino acids (Panula et al., 2015). H2 receptor activation stimulates cAMP formation through coupling to Gs proteins, and in some cells Gq/11 proteins, which leads to inositol phosphate formation and increase in intracellular Ca2+. In mammals it is expressed highly in the stomach and brain, where the expression level is about 10% of that of the stomach (Karlstedt et al., 2001). In the brain both neurons and astrocytes express H2 receptor. Through H2 receptor, histamine can cause excitation of potentiation of excitation of neurons. It blocks a Ca2+-dependent K+ conductance, which causes a longlasting after-hyperpolarization (Haas and Panula, 2003). As a result of this effect a sensory input into, e.g., cortical neurons can cause a much enhanced response. In the hippocampus, synaptic transmission is enhanced for several hours after exposure to histamine (Selbach et al., 1997; Haas and Panula, 2003). H2 receptor is also expressed in several cell types in the immune system, including T and B cells, monocytes, and dendritic cells, indicating possible involvement in brain disorders linked with the immune system such as MS. Expression in the endothelial cells (Karlstedt et al., 1999) is in agreement with the observed role of histamine in regulation of blood–brain barrier permeability (Dux and Joo, 1982). In addition, astrocytes express H2 receptor

P. PANULA

382

which may be involved in energy regulation of the brain (Inagaki and Wada, 1994). In human brain, H2 receptor radioligand binding is abundant in the hippocampus, cerebral cortex, basal ganglia, and amygdala (Traiffort et al., 1992). In the prefrontal cortex, H2 receptor mRNA expression is the highest in layer II (Fig. 23.3B), whereas receptor radioligand binding is the strongest in layers II and III (Jin and Panula, 2005; Panula, 2020). H2 receptor is expressed early in developing rodent brain, from embryonal day 15 particularly in the developing cortical plate and pons (Karlstedt et al., 2001), sites of early transient histaminergic system originating in the raphe neurons which project to developing forebrain (Auvinen and Panula, 1988; Kinnunen and Panula, 1991). This is in agreement with the established role of histamine in proliferation of neural stem cells (Molina-Hernandez and Velasco, 2008).

Histamine H3 receptor Histamine H3 receptor is the only GPCR initially identified and characterized as a new CNS autoreceptor for histamine and an essential regulator of neurotransmitter release in the brain (Arrang et al., 1983). Subsequent cloning of the receptor in 1999 (Lovenberg et al., 1999) facilitated the development of ligands, both agonists and particularly antagonists, of which most are inverse agonists at this receptor, which displays significant constitutive activity (Panula et al., 2015). Only one of the several tested H3 inverse agonists is currently in clinical use. Pitolisant is used to treat idiopathic hypersomnia and narcolepsy refractory to other stimulants. Despite its relatively long half-life, administration once a day is sufficient to maintain wakefulness and attention during the day without significant sleep disturbance (Lin et al., 2008). A significant advantage of H3 antagonists and difference to many other stimulants is the lack of any reported tendency to addiction. Several clinical trials with other H3 receptor inverse agonists to treat cognitive impairment in Alzheimer’s disease and schizophrenia, attention deficit syndrome of adults, and Tourette syndrome have been completed, but the results have not been encouraging enough to extend trials to larger groups (Panula et al., 2015). Two clinical trials to treat alcohol use disorder have been registered, but both were withdrawn before initiation due to company policy reasons (Panula, 2020) rather than a lack of effect. H3 receptor is coupled to Gi/o proteins and activates mitogen-activated protein kinase pathways (Drutel et al., 2001; Panula et al., 2015), and through G16a activation, it can increase cytosolic Ca2+ concentration (Krueger et al., 2005; Panula et al., 2015). H3 receptor

shows significant agonist-independent constitutive signaling, which inhibits forskolin-induced cAMP accumulation, which is further decreased by agonists and enhanced by inverse agonists (Panula et al., 2015). H3 receptor has an unusual structure for this 445 aa GPCR with several exons and introns, which leads to a large number of isoforms (Coge et al., 2001; Drutel et al., 2001). These isoforms exist in all studied mammals, and they display different ligand binding profiles and signaling properties (Panula et al., 2015). In addition, some of them are not expressed on cell surface but act as dominant negative isoforms at least in vitro (Bakker et al., 2006). These complex properties have rendered it difficult to explore the full significance of all expressed forms of H3 receptor. There are approximately 20 identified shorter forms of H3 receptor in human brain (Leurs et al., 2005; Panula et al., 2015), suggesting that much remains to be studied before the roles of this receptor are known. In rodent brain H3 receptor is abundantly expressed in several aminergic neuron groups including the histaminergic tuberomamillary neurons and noradrenergic locus coeruleus neurons, cerebral cortex, thalamus, striatum, hippocampus, amygdala, and several other areas (Lovenberg et al., 1999; Drutel et al., 2001; Pillot et al., 2002). Interestingly, the expression patterns of the H3 receptor isoform and apparently processing of the transcript differ in rat brain areas (Drutel et al., 2001). As can be seen in fig. 5 of a recent review (Panula et al., 2015), the mRNA expression pattern and receptor radioligand binding patterns in rodent brain fit together very well and show that the thalamocortical system, major part of the limbic system (amygdala and hippocampus), constitutes major elements of H3 receptor regulated systems apart from the autoreceptor function of the H3 receptor. The same is true in the human brain (Jin et al., 2002; Jin and Panula, 2005; Panula and Nuutinen, 2013). In human brain, H3 receptor is highly expressed in dorsal thalamus (Jin et al., 2002), which projects to putamen and cerebral cortex. Many small nuclei and, e.g., the neocortex (Fig. 23.3C) in the brain also express moderate to high levels of H3 receptor mRNA. In the striatonigral system, expression of H3 mRNA is high in putamen and nucleus caudatus and low in substantia nigra. H3 receptor radioligand binding is high in substantia nigra, suggesting that striatonigral GABAergic neurons express the receptor (Anichtchik et al., 2001). H3 receptor binding is higher in substantia nigra of PD patients than in normal control brains (Anichtchik et al., 2001). This abnormal binding pattern in PD is associated with abnormally high histamine levels in basal ganglia on PD patients (Rinne et al., 2002) and abnormally dense histamine fibers (Anichtchik et al., 2000).

HISTAMINE RECEPTORS, AGONISTS, AND ANTAGONISTS IN HEALTH AND DISEASE The abnormalities in the brain histamine system in schizophrenia and PD may be in part related to the close interactions of H3 receptor with dopamine receptors. There is evidence for dopamine D1 and H3 receptor heteromers (Sanchez-Lemus and Arias-Montano, 2004; Moreno et al., 2011). Dopamine D2 receptors also interact closely with H3 receptors (Ferrada et al., 2008). D1 and D2 receptor-mediated activation of extracellular signalregulated kinase 1/2 is absent in the striatum of H3 receptor knockout mice, suggesting that normal dopamine receptor-mediated signaling requires also histamine H3 receptor (Kononoff Vanhanen et al., 2016). Interactions of histamine and dopamine receptors are reviewed in detail in Panula and Nuutinen (2013).

Histamine H4 receptor The most recently found histamine receptor, H4 receptor, was discovered by several independent research groups soon following the identification of the H3 receptor gene (Nakamura et al., 2000; Oda et al., 2000; Liu et al., 2001; Panula et al., 2015). It resembles structurally H3 receptor and has three exons and two introns. Like H3 receptor, it is coupled to Gi/O proteins and uses Ca2+ as a second messenger. The most prominent expression of H4 receptor is found in bone marrow, peripheral blood, spleen, and thymus, but expression has also been reported in the lung, small intestine, and colon. Expression of H4 receptor in the mammalian brain has been reported in neurons, endothelial cells, choroid plexus, but particularly expression in brain neurons is still controversial: while most groups have reported no or trace amounts of H4 receptor mRNA in the brain, no reports have convincingly shown the presence of both the H4 receptor protein and mRNA in control but not in H4 receptor deficient mammals (Panula et al., 2015; Schneider and Seifert, 2016). Expression in endothelial cells has been verified by sequencing of the PCR products in immortalized endothelial cells. H4 receptor has also been reported in microglial cell lines and primary microglia (Ferreira et al., 2012) in agreement with the presence of this receptor in macrophages and cells of the immune system (Panula et al., 2015). The detection of this receptor in tissues and cells is hampered by large changes in its expression depending on the inflammation status, and the fact that many commercial antibodies against it are nonspecific, i.e. immunoreactivity is detected in tissues of gene knockout animals or in absence of corresponding mRNA. Histamine H4 receptor antagonists may have clinical potential in treatment of neuroimmune disorders and potentially also neurodegenerative disorders where inflammation and immune mechanisms play crucial roles. So far, however, these drugs have been tested

383

mostly for immune and dermatological conditions. There is some evidence that H4 receptor mechanisms are important in nociception (Panula et al., 2015).

HISTAMINE RECEPTORS IN AUTOIMMUNE DISEASES Histamine and all four histamine receptors play complex roles in multiple sclerosis and its animal models, like experimental allergic encephalomyelitis (EAE). Increased histamine levels have been reported in the CSF of MS patients (Tuomisto et al., 1983), and brain-penetrating histamine H1R antagonists ameliorate the findings and symptoms of EAE (Dimitriadou et al., 2000), and in a small open-labeled study of MS patients (Logothetis et al., 2005). The use of firstgeneration antihistaminic drugs is also associated with the lower incidence of MS (Alonso et al., 2006). The exact roles of histamine and the four GPCRs in MS have been difficult to reveal due to the involvement of these receptors in almost all the stages of MS. Histamine receptors are found in Th1 and Th2 lymphocytes, dendritic cells, mast cells, brain capillary endothelial cells, neurons, and glial cells (Panula et al., 2014). Histamine H1 receptor has been identified as a factor responsible for the vasoactive amine sensitization elicited by histamine and associated increased permeability of brain vasculature (Ma et al., 2002). Mice lacking the histamine-synthesizing enzyme histidine decarboxylase are more susceptible for EAE than wild-type animals, whereas mice lacking either functional histamine H1 receptor (hrh1) or H2 receptor (hrh2) gene are less susceptible. Mice lacking hrh3 are more susceptible than control animals, whereas mice lacking all four histamine GPCRs are less susceptible to EAE than control animals. These examples and the fact that lack of histamine-synthesizing enzyme (¼all endogenously generated histamine) and lack of all hrh:s result in completely different phenotypes suggest that the involvement of histamine in the mechanisms of EAE and MS is important but complex. One possibility is that so far only partly known mechanisms of genetic compensation (El-Brolosy et al., 2019) may be activated in gene-modified mice. In this mechanism, inactivation of one gene leads to compensatory changes in expression of other genes to compensate for the lack of an essential protein.

HISTAMINE RECEPTORS AND SUBSTANCE ABUSE One of the special features of histamine H3 receptor biology is that despite the role of H3 receptor antagonists in transmitter release, including release of histamine itself and dopamine (Schlicker et al., 1993; Panula and

384

P. PANULA

Nuutinen, 2013), these drugs do not seem to cause dependence. Histamine originating from the hypothalamic tuberomamillary nucleus appears to inhibit reward in rodents (Wagner et al., 1993). A surprising finding of high histamine levels and metabolism accompanied by decreased alcohol intake by H3 receptor antagonists in alcohol-preferring AA rats (Lintunen et al., 2001) initiated a series of studies, which revealed that several histamine H3 receptor antagonists inhibit alcohol self-administration and conditioned place preference in several mouse models using three different drinking paradigms (Nuutinen et al., 2010, 2011a,b; Galici et al., 2011; Panula, 2020). Two clinical trials have been registered for testing whether H3 receptor inverse agonists could be used to treat alcohol use disorder in human, but neither one was initiated due to company policy reasons. The brain histamine levels are also abnormally high in human type 2 alcoholics (Alakarppa et al., 2002). This may be due to liver failure, since histamine levels are also high in patients suffering from liver cirrhosis (Lozeva et al., 2003). In experimental animals, liver failure or portocaval anastomosis can increase the brain histamine levels 4–11-fold higher than in normal conditions, suggesting that the failure to control blood neutral amino acid levels leads to increased availability of L-histidine in the brain (Fogel et al., 2002). The significance of this increase in histamine in the brain has not been evaluated, but it remains possible that excessive histamine contributes to confusion and cognitive decline and sleep disorders of alcoholics.

FUTURE POTENTIAL OF HISTAMINE RECEPTORS IN BRAIN DISORDERS The full potential of H3 receptor antagonists in brain disorders remains to be established. In addition to already initiated use in narcolepsy and sleep disorders, some applications in cognitive disorders are possible despite somewhat negative clinical trials completed so far. Alcohol use disorder is another promising and so far untested area, where all rodent studies and lack of habit-forming properties provide good support. New drugs with affinity to both H3 and dopamine receptors may also become useful in cognitive disorders. Currently H4 receptor antagonists have undergone clinical tests in asthma and allergic rhinitis, atopic dermatitis and pruritus (Panula et al., 2015). In experimental models, antagonists have also been effective in inflammatory pain and neuropathic pain. Expression of histamine H4 receptor in cells of the immune system supports the concept that antagonists may prove beneficial in disorders in which inflammation plays a major role, such as multiple sclerosis, and several neurodegenerative disorders such as Parkinson’s and Alzheimer’s disease. Drugs that target two different receptors, like H4 and H3 receptors, may

prove promising in neurodegenerative disorders in which inflammation and loss of neurons are associated with problems in wakefulness and attention. The full potential of H2 receptor antagonists or drugs in schizophrenia may become evident when the patients are selected based on disease pathology, which is still seldom used in clinical trials. H1 receptor antagonists continue to be useful in limited cases of sleep disorders. Brain-specific drugs that combine the properties of H3 antagonists and H1 agonists might prove useful in cognitive disorders.

REFERENCES Airaksinen MS, Paetau A, Paljarvi L et al. (1991). Histamine neurons in human hypothalamus: anatomy in normal and Alzheimer diseased brains. Neuroscience 44: 465–481. Airaksinen MS, Alanen S, Szabat E et al. (1992). Multiple neurotransmitters in the tuberomamillary nucleus: comparison of rat, mouse, and Guinea pig. J Comp Neurol 323: 103–116. Alakarppa K, Tupala E, Mantere T et al. (2002). Effect of alcohol abuse on human brain histamine and telemethylhistamine. Inflamm Res 51 (Suppl 1): S40–S41. Alonso A, Jick SS, Hernan MA (2006). Allergy, histamine 1 receptor blockers, and the risk of multiple sclerosis. Neurology 66: 572–575. Anichtchik OV, Rinne JO, Kalimo H et al. (2000). An altered histaminergic innervation of the substantia nigra in Parkinson’s disease. Exp Neurol 163: 20–30. Anichtchik OV, Peitsaro N, Rinne JO et al. (2001). Distribution and modulation of histamine H(3) receptors in basal ganglia and frontal cortex of healthy controls and patients with Parkinson’s disease. NeurobiolDis 8: 707–716. Anlauf M, Schafer MK, Schwark T et al. (2006). Vesicular monoamine transporter 2 (VMAT2) expression in hematopoietic cells and in patients with systemic mastocytosis. J Histochem Cytochem 54: 201–213. Arrang JM, Garbarg M, Schwartz JC (1983). Auto-inhibition of brain histamine release mediated by a novel class (H3) of histamine receptor. Nature 302: 832–837. Auvinen S, Panula P (1988). Development of histamineimmunoreactive neurons in the rat brain. J Comp Neurol 276: 289–303. Bakker RA, Lozada AF, van Marle A et al. (2006). Discovery of naturally occurring splice variants of the rat histamine H3 receptor that act as dominant-negative isoforms. Mol Pharmacol 69: 1194–1206. Coge F, Guenin SP, Audinot V et al. (2001). Genomic organization and characterization of splice variants of the human histamine H3 receptor. Biochem J 355: 279–288. Deutsch SI, Rosse RB, Kendrick KA et al. (1993). Famotidine adjunctive pharmacotherapy for schizophrenia: preliminary data. Clin Neuropharmacol 16: 518–524. Dimitriadou V, Pang X, Theoharides TC (2000). Hydroxyzine inhibits experimental allergic encephalomyelitis (EAE) and associated brain mast cell activation. Int J Immunopharmacol 22: 673–684.

HISTAMINE RECEPTORS, AGONISTS, AND ANTAGONISTS IN HEALTH AND DISEASE Drutel G, Peitsaro N, Karlstedt K et al. (2001). Identification of rat H3 receptor isoforms with different brain expression and signaling properties. Mol Pharmacol 59: 1–8. Dux E, Joo F (1982). Effects of histamine on brain capillaries. Fine structural and immunohistochemical studies after intracarotid infusion. Exp Brain Res 47: 252–258. El-Brolosy MA, Kontarakis Z, Rossi A et al. (2019). Genetic compensation triggered by mutant mRNA degradation. Nature 568: 193–197. Ercan-Sencicek AG, Stillman AA, Ghosh AK et al. (2010). L-histidine decarboxylase and Tourette’s syndrome. N Engl J Med 362: 1901–1908. Erickson JD, Eiden LE, Hoffman BJ (1992). Expression cloning of a reserpine-sensitive vesicular monoamine transporter. Proc Natl Acad Sci U S A 89: 10993–10997. Ferrada C, Ferre S, Casado V et al. (2008). Interactions between histamine H3 and dopamine D2 receptors and the implications for striatal function. Neuropharmacology 55: 190–197. Ferreira R, Santos T, Goncalves J et al. (2012). Histamine modulates microglia function. J Neuroinflammation 9: 90. Fogel WA, Michelsen KA, Granerus G et al. (2002). Neuronal storage of histamine in the brain and telemethylimidazoleacetic acid excretion in portocaval shunted rats. J Neurochem 80: 375–382. Galici R, Rezvani AH, Aluisio L et al. (2011). JNJ-39220675, a novel selective histamine H3 receptor antagonist, reduces the abuse-related effects of alcohol in rats. Psychopharmacology (Berl) 214: 829–841. Gantz I, Schaffer M, Delvalle J et al. (1991). Molecularcloning of a gene encoding the histamine-H2-receptor. Proc Natl Acad Sci USA 88: 429–433. Haas H, Panula P (2003). The role of histamine and the tuberomamillary nucleus in the nervous system. Nat Rev Neurosci 4: 121–130. Inagaki N, Wada H (1994). Histamine and prostanoid receptors on glial cells. Glia 11: 102–109. Jin CY, Panula P (2005). The laminar histamine receptor system in human prefrontal cortex suggests multiple levels of histaminergic regulation. Neuroscience 132: 137–149. Jin CY, Kalimo H, Panula P (2002). The histaminergic system in human thalamus: correlation of innervation to receptor expression. Eur J Neurosci 15: 1125–1138. Jin CY, Anichtchik O, Panula P (2009). Altered histamine H3 receptor radioligand binding in post-mortem brain samples from subjects with psychiatric diseases. Br J Pharmacol 157: 118–129. Jucaite A, Takano A, Bostrom E et al. (2013). AZD5213: a novel histamine H3 receptor antagonist permitting high daytime and low nocturnal H3 receptor occupancy, a PET study in human subjects. Int J Neuropsychopharmacol 16: 1231–1239. Kagevi I, Thorhallsson E, Wahlby L (1987). CSF concentrations of famotidine. Br J Clin Pharmacol 24: 849–850. Kaminsky R, Moriarty TM, Bodine J et al. (1990). Effect of famotidine on deficit symptoms of schizophrenia. Lancet 335: 1351–1352. Kapeller-Adler R (1949). Studies on histaminase. Biochem J 44: 70–77.

385

Karlstedt K, Sallmen T, Eriksson KS et al. (1999). Lack of histamine synthesis and down-regulation of H1 and H2 receptor mRNA levels by dexamethasone in cerebral endothelial cells. J Cereb Blood Flow Metab 19: 321–330. Karlstedt K, Senkas A, Ahman M et al. (2001). Regional expression of the histamine H-2 receptor in adult and developing rat brain. Neuroscience 102: 201–208. Kinnunen A, Panula P (1991). Histamine and tyrosine hydroxylase in developing rat brain. Agents Actions 33: 108–111. Kononoff Vanhanen J, Nuutinen S, Tuominen M et al. (2016). Histamine H3 receptor regulates sensorimotor gating and dopaminergic signaling in the striatum. J Pharmacol Exp Ther 357: 264–272. Krueger KM, Witte DG, Ireland-Denny L et al. (2005). G protein-dependent pharmacology of histamine H3 receptor ligands: evidence for heterogeneous active state receptor conformations. J Pharmacol Exp Ther 314: 271–281. Krystal AD, Lankford A, Durrence HH et al. (2011). Efficacy and safety of doxepin 3 and 6 mg in a 35-day sleep laboratory trial in adults with chronic primary insomnia. Sleep 34: 1433–1442. Kukko-Lukjanov TK, Panula P (2003). Subcellular distribution of histamine, GABA and galanin in tuberomamillary neurons in vitro. J Chem Neuroanat 25: 279–292. Leurs R, Bakker RA, Timmerman H et al. (2005). The histamine H3 receptor: from gene cloning to H3 receptor drugs. Nat Rev Drug Discov 4: 107–120. Lin JS, Dauvilliers Y, Arnulf I et al. (2008). An inverse agonist of the histamine H(3) receptor improves wakefulness in narcolepsy: studies in orexin / mice and patients. Neurobiol Dis 30: 74–83. Lintunen M, Hyytia P, Sallmen T et al. (2001). Increased brain histamine in an alcohol-preferring rat line and modulation of ethanol consumption by H(3) receptor mechanisms. FASEB J 15: 1074–1076. Liu C, Ma X, Jiang X et al. (2001). Cloning and pharmacological characterization of a fourth histamine receptor (H(4)) expressed in bone marrow. Mol Pharmacol 59: 420–426. Logothetis L, Mylonas IA, Baloyannis S et al. (2005). A pilot, open label, clinical trial using hydroxyzine in multiple sclerosis. Int J Immunopathol Pharmacol 18: 771–778. Lovenberg TW, Roland BL, Wilson SJ et al. (1999). Cloning and functional expression of the human histamine H3 receptor. Mol Pharmacol 55: 1101–1107. Lozeva V, Tuomisto L, Tarhanen J et al. (2003). Increased concentrations of histamine and its metabolite, tele-methylhistamine and down-regulation of histamine H3 receptor sites in autopsied brain tissue from cirrhotic patients who died in hepatic coma. J Hepatol 39: 522–527. Ma RZ, Gao J, Meeker ND et al. (2002). Identification of Bphs, an autoimmune disease locus, as histamine receptor H1. Science 297: 620–623. Martinez-Mir MI, Pollard H, Moreau J et al. (1990). Three histamine receptors (H1, H2 and H3) visualized in the brain of human and non-human primates. Brain Res 526: 322–327. Mazurkiewicz-Kwilecki IM, Nsonwah S (1989). Changes in the regional brain histamine and histidine levels in postmortem brains of Alzheimer patients. Can J Physiol Pharmacol 67: 75–78.

386

P. PANULA

Meskanen K, Ekelund H, Laitinen J et al. (2013). Antagonism of histamine H2 receptors as a novel approach to treat schizophrenia: a double-blind, randomized clinical trial. J Clin Psychopharmacol 33: 472–478. Molina-Hernandez A, Velasco I (2008). Histamine induces neural stem cell proliferation and neuronal differentiation by activation of distinct histamine receptors. J Neurochem 106: 706–717. Moreno E, Hoffmann H, Gonzalez-Sepulveda M et al. (2011). Dopamine D1-histamine H3 receptor heteromers provide a selective link to MAPK signaling in GABAergic neurons of the direct striatal pathway. J Biol Chem 286: 5846–5854. Naganuma F, Nakamura T, Yoshikawa T et al. (2017). Histamine N-methyltransferase regulates aggression and the sleep-wake cycle. Sci Rep 7: 15899. Nakamura T, Itadani H, Hidaka Y et al. (2000). Molecular cloning and characterization of a new human histamine receptor, Hh4R. Biochem Biophys ResCommun 279: 615–620. Nuutinen S, Karlstedt K, Aitta-Aho T et al. (2010). Histamine and H3 receptor-dependent mechanisms regulate ethanol stimulation and conditioned place preference in mice. Psychopharmacology (Berl) 208: 75–86. Nuutinen S, Lintunen M, Vanhanen J et al. (2011a). Evidence for the role of histamine h3 receptor in alcohol consumption and alcohol reward in mice. Neuropsychopharmacology 36: 2030–2040. Nuutinen S, Vanhanen J, Pigni MC et al. (2011b). Effects of histamine H3 receptor ligands on the rewarding, stimulant and motor-impairing effects of ethanol in DBA/2J mice. Neuropharmacology 60: 1193–1199. Oda T, Morikawa N, Saito Y et al. (2000). Molecular cloning and characterization of a novel type of histamine receptor preferentially expressed in leukocytes. J Biol Chem 275: 36781–36786. Panula P (2020). Histamine, histamine H3 receptor, and alcohol use disorder. Br J Pharmacol 177: 634–641. Panula P, Nuutinen S (2013). The histaminergic network in the brain: basic organization and role in disease. Nat Rev Neurosci 14: 472–487. Panula P, Yang HY, Costa E (1984). Histamine-containing neurons in the rat hypothalamus. Proc. Natl. Acad. Sci. U.S.A 81: 2572–2576. Panula P, Kaartinen M, Macklin M et al. (1985). Histaminecontaining peripheral neuronal and endocrine systems. J Histochem Cytochem 33: 933–941. Panula P, Airaksinen MS, Pirvola U et al. (1990). A histaminecontaining neuronal system in human brain. Neuroscience 34: 127–132. Panula P, Rinne J, Kuokkanen K et al. (1998). Neuronal histamine deficit in Alzheimer’s disease. Neuroscience 82: 993–997. Panula P, Sundvik M, Karlstedt K (2014). Developmental roles of brain histamine. Trends Neurosci 37: 159–168. Panula P, Chazot PL, Cowart M et al. (2015). International union of basic and clinical pharmacology. XCVIII. Histamine receptors. Pharmacol Rev 67: 601–655.

Pillot C, Heron A, Cochois V et al. (2002). A detailed mapping of the histamine H(3) receptor and its gene transcripts in rat brain. Neuroscience 114: 173–193. Prell GD, Green JP, Kaufmann CA et al. (1995). Histamine metabolites in cerebrospinal fluid of patients with chronic schizophrenia: their relationships to levels of other aminergic transmitters and ratings of symptoms. Schizophr Res 14: 93–104. Rinne JO, Anichtchik OV, Eriksson KS et al. (2002). Increased brain histamine levels in Parkinson’s disease but not in multiple system atrophy. J Neurochem 81: 954–960. Rosse RB, Kendrick K, Fay-McCarthy M et al. (1996). An open-label study of the therapeutic efficacy of high-dose famotidine adjuvant pharmacotherapy in schizophrenia: preliminary evidence for treatment efficacy. Clin Neuropharmacol 19: 341–348. Roth T, Rogowski R, Hull S et al. (2007). Efficacy and safety of doxepin 1 mg, 3 mg, and 6 mg in adults with primary insomnia. Sleep 30: 1555–1561. Rothberg MB, Herzig SJ, Pekow PS et al. (2013). Association between sedating medications and delirium in older inpatients. J Am Geriatr Soc 61: 923–930. Sanchez-Lemus E, Arias-Montano JA (2004). Histamine H3 receptor activation inhibits dopamine D1 receptor-induced cAMP accumulation in rat striatal slices. NeurosciLett 364: 179–184. Scharf M, Rogowski R, Hull S et al. (2008). Efficacy and safety of doxepin 1 mg, 3 mg, and 6 mg in elderly patients with primary insomnia: a randomized, double-blind, placebo-controlled crossover study. J Clin Psychiatry 69: 1557–1564. Schayer RW (1957). Histidine decarboxylase of rat stomach and other mammalian tissues. Am J Physiol 189: 533–536. Schlicker E, Fink K, Detzner M et al. (1993). Histamine inhibits dopamine release in the mouse striatum via presynaptic H3 receptors. J Neural Transm Gen Sect 93: 1–10. Schneider EH, Seifert R (2016). The histamine H4-receptor and the central and peripheral nervous system: a critical analysis of the literature. Neuropharmacology 106: 116–128. Selbach O, Brown RE, Haas HL (1997). Long-term increase of hippocampal excitability by histamine and cyclic AMP. Neuropharmacology 36: 1539–1548. Shan L, Dauvilliers Y, Siegel JM (2015). Interactions of the histamine and hypocretin systems in CNS disorders. Nat Rev Neurol 11: 401–413. Shimamura T, Shiroishi M, Weyand S et al. (2011). Structure of the human histamine H1 receptor complex with doxepin. Nature 475: 65–70. Snyder SH, Axelrod J (1965). Sex differences and hormonal control of histamine methyltransferase activity. Biochim Biophys Acta 111: 416–421. Sundvik M, Panula P (2012). The organization of the histaminergic system in adult zebrafish (Danio rerio) brain: neuron number, location and co-transmitters. J Comp Neurol 520: 3827–3845. Traiffort E, Pollard H, Moreau J et al. (1992). Pharmacological characterization and autoradiographic localization of

HISTAMINE RECEPTORS, AGONISTS, AND ANTAGONISTS IN HEALTH AND DISEASE histamine H2 receptors in human brain identified with [125I]iodoaminopotentidine. J Neurochem 59: 290–299. Tuomisto L, Kilpelainen H, Riekkinen P (1983). Histamine and histamine-N-methyltransferase in the CSF of patients with multiple sclerosis. Agents Actions 13: 255–257. Vincent SR, Hokfelt T, Skirboll LR et al. (1983). Hypothalamic gamma-aminobutyric acid neurons project to the neocortex. Science 220: 1309–1311. Wagner U, Segura-Torres P, Weiler T et al. (1993). The tuberomamillary nucleus region as a reinforcement inhibiting substrate: facilitation of ipsihypothalamic self-stimulation by unilateral ibotenic acid lesions. Brain Res 613: 269–274.

387

Watanabe T, Taguchi Y, Shiosaka S et al. (1984). Distribution of the histaminergic neuron system in the central nervous system of rats; a fluorescent immunohistochemical analysis with histidine decarboxylase as a marker. Brain Res 295: 13–25. Yamashita M, Fukui H, Sugama K et al. (1991). Expression cloning of a cDNA encoding the bovine histamine H1 receptor. Proc Natl Acad Sci U S A 88: 11515–11519. Yanai K, WATANABE T, Meguro K et al. (1992). Age-dependent decrease in histamine H1 receptor in human brains revealed by PET. Neuroreport 3: 433–436.

This page intentionally left blank

Handbook of Clinical Neurology, Vol. 180 (3rd series) The Human Hypothalamus: Middle and Posterior Region D.F. Swaab, F. Kreier, P.J. Lucassen, A. Salehi, and R.M. Buijs, Editors https://doi.org/10.1016/B978-0-12-820107-7.00024-0 Copyright © 2021 Elsevier B.V. All rights reserved

Chapter 24

The tuberomamillary nucleus in neuropsychiatric disorders LING SHAN1,2,3*, ROLF FRONCZEK1,2, GERT JAN LAMMERS1,2, AND DICK F. SWAAB3 1

Department of Neurology, Leiden University Medical Centre, Leiden, The Netherlands 2

Sleep Wake Centre SEIN, Heemstede, The Netherlands

3

Department Neuropsychiatric Disorders, Netherlands Institute for Neuroscience, an Institute of the Royal Netherlands Academy of Arts and Sciences, Amsterdam, The Netherlands

Abstract The tuberomamillary nucleus (TMN) is located within the posterior part of the hypothalamus. The histamine neurons in it synthesize histamine by means of the key enzyme histidine decarboxylase (HDC) and from the TMN, innervate a large number of brain areas, such as the cerebral cortex, hippocampus, amygdala as well as the thalamus, hypothalamus, and basal ganglia. Brain histamine is reduced to an inactivated form, tele-methylhistamine (t-MeHA), by histamine N-methyltransferase (HMT). In total, there are four types of histamine receptors (H1–4Rs) in the brain, all of which are G-protein coupled. The histaminergic system controls several basal physiological functions, including the sleep–wake cycle, energy and endocrine homeostasis, sensory and motor functions, and cognitive functions such as attention, learning, and memory. Histaminergic dysfunction may contribute to clinical disorders such as Parkinson’s disease, Alzheimer’s disease, Huntington’s disease, narcolepsy type 1, schizophrenia, Tourette syndrome, and autism spectrum disorder. In the current chapter, we focus on the role of the histaminergic system in these neurological/neuropsychiatric disorders. For each disorder, we first discuss human data, including genetic, postmortem brain, and cerebrospinal fluid studies. Then, we try to interpret the human changes by reviewing related animal studies and end by discussing, if present, recent progress in clinical studies on novel histaminerelated therapeutic strategies.

INTRODUCTION The tuberomamillary nucleus (TMN) is located in the posterior hypothalamus and characterized by large, irregularly bordered, lipofuscin-laden neurons that surround the fornix in its final descending course and the mamillary body (Shan et al., 2015a) (Fig. 24.1A–C). The TMN is the exclusive production site of neuronal histamine that is produced by the key enzyme histidine decarboxylase (HDC). The histaminergic system is currently receiving increased attention as much new knowledge has been gained in the last few years about its physiological (dys-) functions and because of the successful development of

histamine-3 receptor (H3R) antagonists/inverse agonists as treatment for narcolepsy. We will summarize the anatomical and physiological evidence for a role of histaminergic neurons in pathologies like Parkinson’s disease (PD), Alzheimer’s disease (AD), Huntington’s disease (HD), narcolepsy type 1, Tourette syndrome (TS), and autism spectrum disorder.

NEUROANATOMY OF HISTAMINERGIC NEURONS In the TMN, HDC converts the amino acid histidine into histamine, which is then packaged into synaptic vesicles by the vesicular monoamine transporter (VMAT2, also

*Correspondence to: Ling Shan Ph.D., Netherlands Institute for Neuroscience, KNAW, Meibergdreef 47, 1105 BA Amsterdam, The Netherlands. Tel: +31-205665500, Fax: +31-205666121, E-mail: [email protected]

390

L. SHAN ET AL.

Fig. 24.1. Location of the human tuberomamillary nucleus (TMN) and its typical neurons with histidine decarboxylase (HDC) and GABA cotransmission. (A) Scheme of coronal section of the human hypothalamus at the level of the TMN that is highlighted in red. (B and C) Examples of Nissl staining of TMN neurons showing intensely stained endoplasmic reticulum interspersed with the typical irregularities in the cell membrane. (D) Coexpression of histidine decarboxylase (HDC)-mRNA (darkfield illumination) and (E) glutamic acid decarboxylase (GAD)-67-like immunoreactivity in two consecutive sections of the TMN. (F) Specific HDCmRNA radioactive ISH signal was observed in the TMN both on film autoradiograms and (G) after emulsion autoradiography with thionin-counterstaining, respectively, using an HDC antisense probe. Scale bar ¼ 5 mm in (B and C); Scale bar ¼ 50 mm in (D) and (E); In (F and G), Scale bar ¼ 1 mm for the film autoradiograms, ¼ 50 mm for the emulsion autoradiographs; and ¼ 12.5 mm in insertions. Abbreviations: BSTp, bed nucleus of the stria terminalis posterior; DMN, dorsomedial hypothalamic nucleus; fx or f, fornix; INF, infundibular nucleus; LHA, lateral hypothalamus; LV, lateral ventricle; NTL, lateral tuberal nucleus; OT, optic tract; TM, tuberomamillary nucleus; TMv, ventral TM; VMN, ventromedial hypothalamic nucleus; 3V, third ventricle. Panels (D, E) Modified from Trottier S, Chotard C, Traiffort E et al. (2002). Co-localization of histamine with GABA but not with galanin in the human tuberomamillary nucleus. Brain Res 939: 52–64, with permission. Modified from Shan L, Bao AM, Swaab DF (2015a). The human histaminergic system in neuropsychiatric disorders. Trends Neurosci 38: 167–177, with permission.

known as solute carrier family 18 member 2 SlC18A2) and released throughout the brain (Haas et al., 2008). Brain histamine is further inactivated by histamine N-methyltransferase (HMT) with tele-methylhistamine (t-MeHA) being a stable metabolite of histamine (Haas and Panula, 2003).

There are four types of histamine receptors (H1–4Rs) in the brain, all of which are G-protein coupled. Their different functions, signaling, and distribution patterns have been extensively reviewed before (Haas et al., 2008; Panula and Nuutinen, 2013). H3R is an autoreceptor/ heteroreceptor that moderates the release of histamine

THE TUBEROMAMILLARY NUCLEUS IN NEUROPSYCHIATRIC DISORDERS as well as of other neurotransmitters, including acetylcholine, dopamine, GABA, serotonin and neuropeptides, as reviewed elsewhere (Haas et al., 2008; Panula and Nuutinen, 2013). During development, the first histamine neurons are seen on rat embryonic day 13 in the border area between the mesencephalon and metencephalon, while the histamine immunoreactive nerve fibers are not detected until embryonic day 15. On postnatal day 14, an adult-like pattern of neurons and fibers develops (Auvinen and Panula, 1988). Fetal development of the human TMN starts around 16–18 weeks of gestation (Koutcherov et al., 2002).

THE HETEROGENEOUS POPULATION OF TMN NEURONS Neurochemical heterogeneity In the human TMN, the histaminergic neurons are also GABAergic, as shown by their coexpression of the synthesizing enzyme glutamic acid decarboxylase-67 (GAD67/GAD1) (Sherin et al., 1998; Trottier et al., 2002) (Fig. 24.1D and E). In rodents, not all histamine neurons are GABAergic (Williams et al., 2014), about 80% of HDC positive neurons express GABA, as shown by the presence of GAD67 and vesicular GABA transporter (VGAT, also known as solute carrier family 32 member 1, SLC32A1) (Yu et al., 2015). In the human TMN, acetylcholinesterase (Saper and German, 1987), preprodynorphin or preproenkephalin (Sukhov et al., 1995), and monoamine oxidase (Nakamura et al., 1991) coexpression have been observed.

Functionally heterogeneous Multiple lines of evidence suggest that histaminergic neurons are not only neurochemically heterogeneous but also functionally. For instance, a lesion of one of the TMN subregions had anxiolytic-like effects in rats (Frisch et al., 1998). Also, different subpopulations of histaminergic neurons respond differently to the same stressor. This is seen, for example, in terms of their neuronal activity, as determined in rodents by c-Fos immunocytochemistry, or by changes in in situ hybridization signal for HDC-mRNA, between single neurons (Miklos and Kovacs, 2003). Pharmacological studies also hint at independent functions of different subsets of histamine neurons. Histaminergic neurons exhibit, for example a variable sensitivity for GABA agonists, due to differences in their composition of the GABA-A receptor subunits (Sergeeva et al., 2002). In the human histamine system, also region-specific changes were seen, for instance, in Alzheimer’s disease.

391

AD patients demonstrate a significant loss of the TMN neurons in all three subareas (rostral, medial, and caudal) of the TMN, whereas HDC-mRNA expression, a marker for neuronal histamine production, was diminished significantly only in the central part of the AD TMN (Shan et al., 2012b). In depression, including both major depressive disorder and bipolar disorder, no changes were observed in HDC-mRNA (Shan et al., 2013). However, a reduction was found in HMT-mRNA in the anterior cingulate gyrus but not in the PFC region of depressed patients (Shan et al., 2013). In addition, positron emission tomography and [11C]-doxepin ligand showed that the H1R binding was significantly lower in the PFC but not in the temporal cortex region of depression subjects (Kano et al., 2004). It will be interesting to study further changes in these and other measures of the histamine system in the different TMN subdivisions and in relation to these and other neuropsychiatric disorders.

HDC EXPRESSION AND SLEEP–WAKE MODULATION The TMN is active during waking, has a very low level of neuronal activity during sleep, and minimal activity during rapid eye movement (REM) sleep (Lin, 2000). HDC knockout mice show impaired wakefulness during the lights-off period and increased REM sleep (Huang et al., 2006; Parmentier et al., 2002). Interestingly, long-term genetic lesioning of histamine neurons in adult mice induced the same phenotype as HDC knockout mice, while selective short-term chemo-genetic inhibition of histamine neurons resulted in non-REM sleep in mice (Yu et al., 2019). Recent findings showed that TMN neurons are also important for enhancing arousal under conditions like exposure to a novel environment, but that they do not elevate arousal under familiar conditions (Venner et al., 2019). Moreover, mice in which the BMAL1, a key clock gene, was knocked out in their TMN neurons, showed functional alterations in their sleep architecture (Yu et al., 2014). Furthermore, in both humans and rodents, circadian fluctuations of HDCmRNA expression have been reported (Shan et al., 2012c; Yu et al., 2014) (Fig. 24.2). Although there are several steps from the levels of HDC transcripts to final histamine levels in the cerebrospinal fluid (CSF), CSF histamine in the diurnal squirrel monkey reaches acrophase values at 18:00 h (Zeitzer et al., 2012), consistent with the maximum values of HDC-mRNA observed in the human postmortem TMN (i.e., also around 18:00 h) (Shan et al., 2012c) (Fig. 24.2). In addition, a loss of this circadian fluctuation of HDC-mRNA was observed in a number of neurodegenerative diseases,

392

L. SHAN ET AL.

Fig. 24.2. (A) Box plots show the median, 25th–75th percentiles and the total range of radioactivity in arbitrary units. The total amount of radioactivity of histidine decarboxylase (HDC)-mRNA expression is given for control subjects between daytime (08:01–20:00; n ¼ 18) and nighttime (20:01–08:00; n ¼ 15) on the left side, and for neurodegenerative diseases group (NDD, daytime n ¼ 20, nighttime n ¼ 11) on the right side. Note that there is a significant difference (P ¼ 0.004) between daytime and nighttime in control subjects, but not in NDD (P ¼ 0.410). (B and C) Raw data of HDC-mRNA expression plotted along the clock time of death. Nonlinear periodic functions describe the circadian cycles. The model in controls (open dots) reaches an estimated maximum at the end of the afternoon (Tmax ¼ 18:09 h) and a minimum shortly after midnight (Tmin ¼ 1:09 h). The model in NDD group (block dots) reaches an estimated maximum in the morning (Tmax ¼ 8:56 h) and a minimum in the afternoon (Tmin ¼ 14:43 h). The horizontal lines indicate the 24 h mean of HDC-mRNA expression in controls and NDD group, respectively. Source with permission Shan L, Bao AM, Swaab DF (2015a). The human histaminergic system in neuropsychiatric disorders. Trends Neurosci 38: 167–177.

including AD, preclinical PD, PD, and HD (Shan et al., 2012c) that are known to be associated with sleep disorders (Shan et al., 2015b). To restore, the circadian rhythm of HDC-mRNA expression may therefore be a promising target to improve the sleep quality of those patients.

Parkinson’s disease There are conflicting opinions about the nature of alterations of the TMN in Parkinson’s disease (PD). On the basis of the abundant accumulation of Lewy bodies

THE TUBEROMAMILLARY NUCLEUS IN NEUROPSYCHIATRIC DISORDERS and Lewy neuritis in the TMN, it was presumed that this nucleus was severely destroyed in PD (Braak et al., 1996). However, HDC-mRNA levels are unaltered in both the preclinical and clinical PD stages, indicating that neuronal histamine production remains intact (Shan et al., 2012d). This is in line with the observed total number of histaminergic neurons (Nakamura et al., 1996) and the enzymatic activity of HDC (Garbarg et al., 1983) observed in clinical PD. This stability is further supported by the unaltered CSF levels of the main metabolite of histamine, tele-methylhistamine (t-MeHA), in endstage PD patients (Prell et al., 1991). Postmortem studies have revealed significantly increased histamine levels in the putamen, the substantia nigra (SN) and the internal and external globus pallidus in clinical PD compared to controls (Rinne et al., 2002). In agreement with these observations, an increased density of histaminergic fibers and H3R binding was found in the SN of PD patients (Anichtchik et al., 2000, 2001). In addition, several polymorphism studies and also a recent meta-analysis have shown that the lower HMT activity allele, rs11558538, protects against PD (Agundez et al., 2008; Jim
enez-Jim
enez et al., 2016; Ledesma et al., 2008; Palada et al., 2012; Yang et al., 2015). Moreover, a negative correlation was observed between HMTmRNA expression in the SN and the disease duration of PD (Shan et al., 2012a). While the basal gangliarelated changes may be involved in the impairment of motor functions of PD, this implies that the more serious the disease is (and thus the shorter the disease duration), the higher the expression of HMT-mRNA in the SN, and possibly also of histamine, in this region. In vivo animal experiments suggest that such an increased activity of the histaminergic system may accelerate the degeneration of dopaminergic neurons in the SN by an inflammatory process (Rocha et al., 2016; Vizuete et al., 2000). Increased endogenous histamine indeed accelerates dopamine degeneration in the 6-hydroxydopamine (6-OHDA)-lesioned rat, one of the classic PD models (Liu et al., 2007). Furthermore, a decrease of endogenous histamine by injection of a-fluoromethylhistidine, an irreversible inhibitor of HDC, prevented the loss of dopaminergic neurons in the SN in an early stage of the 6-OHDA lesion and strongly reduced the rotation behavior in this rat PD model (Liu et al., 2007). Blocking histamine receptors should then also ameliorate the PD disease process. Also our own experimental data, ranging from postmortem observations to animal models, indicate that inhibiting H4R might indeed be a therapeutic target for PD; for instance, H4R-mRNA levels were robustly (4.2- to 6.3-fold) increased in the basal ganglia of PD patients (Shan et al., 2012a), despite the ongoing neurodegeneration of this brain area in this disorder. H4R-mRNA was also upregulated in the

393

rotenone lesioned rat, a model that closely resembles human PD (Zhou et al., 2019). Three weeks of continuous H4R antagonist JNJ7777120 administration further protected about dopamine neurons from degeneration (Zhou et al., 2019). Surprisingly, the presence of Lewy body-like structures, i.e., a main neuropathological hallmark of PD, was also significantly reduced by this treatment (Zhou et al., 2019). As also microglial markers were significantly increased in the SN of rotenone-treated rats (Sherer et al., 2003), we tested whether H4R can be a potential target for microglia inhibition in the treatment of PD. Indeed, the increased “ionized calcium binding adaptor molecule I” (Ibal 1) immunocytochemistry after rotenone treatment was eliminated by treatment with the H4R antagonist JNJ7777120 (Zhou et al., 2019). A role for histamine H4R in inhibiting the proinflammatory phenotype of microglia in this PD model was further confirmed by reductions of Interleukin 1 beta (IL-1b) and tumor necrosis factor alpha (TNF-a) at both the mRNA and protein level (Zhou et al., 2019). Interestingly, mRNA levels reflecting of a neuroprotective type of microglia, i.e., Arginase 1 (Arg1) and insulin-like growth factor 1(IGF-1), were not affected by rotenone nor by treatment with JNJ7777120. Taken together, these translational results support potential efficacy of an H4R antagonist in treating PD.

Alzheimer’s disease Compared to PD, Alzheimer’s disease (AD) neuropathology appears to affect the TMN already at an earlier stage. Neurofibrillary tangles are, for example, more prominent in the TMN than in its adjacent areas, like the nucleus tuberalis lateralis, and already at a mild cognitive impairment stage (Braak III) (Braak et al., 1993). Also the presence of hyperphosphorylated taucontaining neurites and cell bodies in the TMN of early AD patients is indicative of an early stage of tangle formation (van de Nes et al., 1998). Moreover, b/A4-postive Congo-negative amorphic plaques are found in the TMN of AD patients (van de Nes et al., 1998). In line with neuropathological data, other researchers showed a loss of large histaminergic neurons in the TMN in AD (Nakamura et al., 1993) and diminished histamine levels in hippocampus, frontal, and temporal cortex in AD (Mazurkiewicz-Kwilecki and Nsonwah, 1989; Panula et al., 1997). A significant (57%) loss of TMN neurons in the final stage of AD (Braak VI) was reported in a study with well-matched controls (i.e., for postmortem delay, gender, and age as well as for the diurnal fluctuations) (Shan et al., 2012b). Interestingly, no changes were found in HDC-mRNA expression in the TMN compared with controls (Shan et al., 2012b), which is in agreement with the finding in a large cohort of AD

394

L. SHAN ET AL.

patients where only slightly lower levels of lumbar CSF t-MeHA were found compared with controls (Motawaj et al., 2010), and suggest that histamine production was largely compensated for by the remaining TMN neurons. As they were found to stimulate the release of various neurotransmitters, including histamine and acetylcholine in the rat cortex, H3R antagonists/inverse agonists were proposed as potential treatment for AD (Passani and Blandina, 2011). However, the collective data show that the histaminergic system is relatively unaffected in AD patients, in line with a lack of improvement in cognition in patients with AD in several randomized controlled trials of H3R antagonists/inverse agonists (Egan et al., 2012; Grove et al., 2014; Kubo et al., 2015). Another possible explanation for these disappointing results may be the sex-dependent changes we observed in the histaminergic system in AD. TMN neurons contain nuclear estrogen receptor (ER)-a staining, in both men and women (Kruijver et al., 2002). In men, more cytoplasmic ER-a was observed, while ER-b staining was more intense in women than in men (Kruijver et al., 2003). Interestingly, cytoplasmic ER-b signal was significantly higher in the TMN of AD patients, whereas the cytoplasmic ER-a signal in a trend to be lower in TMN of AD subjects (Ishunina et al., 2003). We also showed that only female H3R-and HMT-mRNA expression was upregulated in the stage of mild cognitive impairment (Braak stage III–IV), and that this upregulation was maintained during the later clinical stages of AD (Braak stage V–VI) (Shan et al., 2012a). Although the protein level information is missing, this data may, at least partly, suggest that only female AD cases will be responsive to H3R antagonist treatment, a possibility that warrants further investigations.

Huntington’s disease Huntington’s disease is a genetic, neurodegenerative condition, associated with a CAG repeat expansion in the gene encoding the protein huntingtin (Phillips et al., 2008). The disease develops when the CAG repeat in the N-terminal of the huntingtin gene translates into a long polyglutamine (poly-Q) tract that exceeds 37 (Kremer et al., 1994). Patients develop characteristic motor, cognitive, and behavioral abnormalities over the course of the disease and chorea as a core manifestation (Phillips et al., 2008). The TMN is one of the hypothalamic nuclei that has a high frequency of cytoplasmic inclusions of the mutant huntingtin protein (Aziz et al., 2008). HDCmRNA expression is significantly increased in the postmortem brains of Huntington’s disease patients, while the number of TMN neurons remains unchanged (Van Wamelen et al., 2011). In the frontal cortex of Huntington’s

disease, HMT-mRNA is significantly upregulated as well (Van Wamelen et al., 2011). These observations are in line with a prior study showing an increased level of histamine metabolites in the CSF of Huntington’s patients (Prell and Green, 1991). The enhanced brain histamine concentrations in the HMT deficient mice exhibited sleep disorders and aggressive behaviors (Naganuma et al., 2017), which at least partly, resembled some of symptoms of HD patients (Phillips et al., 2008).

Narcolepsy type 1 Narcolepsy type 1 is caused by a loss of hypocretin/ orexin (Hcrt) signaling. Narcoleptic phenotypes, including cataplexy, have been observed in mice without Hcrt neuropeptides and in dogs with mutations of the Hcrt type 2 receptor (Lin et al., 1999; Sakurai et al., 1998). The overwhelming majority (86%–98%) of narcolepsy type 1 patients carry the human leukocyte antigen (HLA) subtype DQB1*06:02, which is present in only 20%–30% of the general population (Mignot et al., 2001). In addition, narcolepsy is associated with the T cell receptor alpha (Fontana et al., 2010; Peyron et al., 2000) and autoreactive T cells (Latorre et al., 2018; Pedersen et al., 2019; Schinkelshoek et al., 2019) suggesting an autoimmune-mediated etiology. As in other autoimmune disorders, the development of narcolepsy is sometimes linked to environmental factors such as winter upper airway infections and H1N1 influenza infections (Han et al., 2011). Subsequently, researchers demonstrated low or undetectable Hcrt levels in CSF of narcolepsy type 1 patients (Mignot et al., 2002; Nishino et al., 2000). In line with these CSF observations, a more than 90% depletion of Hcrt neurons was reported in postmortem studies of patients with narcolepsy type 1 (Peyron et al., 2000; Thannickal et al., 2000). Hcrt peptides are produced exclusively by a cluster of neurons in the lateral hypothalamus that are adjacent and intermixed with histaminergic neurons (Shan et al., 2015b). In 2013, two independent groups simultaneously reported a strong increase (64%–94%) in the number of histamine neurons in the TMN of narcolepsy with cataplexy (John et al., 2013; Valko et al., 2013). In contrast, the number of histamine neurons was unchanged in a series of narcoleptic animal models, i.e., Hcrt type 2 receptor mutant dogs, Hcrt knockout mice, ataxin-3-Hcrt mice, and doxycycline-controlled diphtheria toxin A-Hcrt mice (John et al., 2013). This leads to the conclusion that the large increase in histamine neurons in human patients with narcolepsy type 1 does not seem to compensate for the loss of Hcrt neurons. Therefore, this phenomenon is proposed, rather, to be associated with its autoimmune-mediated etiology or possibility

THE TUBEROMAMILLARY NUCLEUS IN NEUROPSYCHIATRIC DISORDERS related to neurogenesis or neurotransmitter respecification (Shan et al., 2015b). On the other hand, the levels of HDC-mRNA should be measured by in situ hybridization (Liu et al., 2010) (Fig. 24.1F and G) to confirm if the HDC-mRNA is also upregulated in narcoleptic human brains. Changes in the human CSF levels of histamine and its main metabolite t-MeHA in narcolepsy are contradictory, and likely not accurate biomarkers for the disorder. Initial studies reported low histamine levels in the CSF of narcolepsy type 1 patients (Bassetti et al., 2010; Kanbayashi et al., 2009; Nishino et al., 2009), but a more recent, larger study found no reduction in histamine or t-MeHA in narcolepsy type 1 patients, a lack of correlation between histamine and subjective (Epworth sleepiness scale) or objective (multiple sleep latency test) measures of sleepiness (Dauvilliers et al., 2012). Recently, another study showed higher CSF histamine levels, lower t-MeHA levels, and lower t-MeHA/histamine ratios in children with narcolepsy type 1 (Franco et al., 2019). Although the number of histamine neurons is increased in narcolepsy type 1, histamine is a labile monoamine that is present in very low concentrations in lumbar CSF (Rye, 2012). Future studies of histamine may, therefore, focus on histamine and t-MeHA measurements in postmortem brain samples. In a knockout rodent model for narcolepsy, the H3R antagonists pitolisant and GSK189254 significantly improved symptoms of narcolepsy, such as excessive daytime sleepiness and abnormally short REM sleep latency at sleep onset (Guo et al., 2009; Lin et al., 2008). Two follow-up, randomized, double-blind placebocontrolled trails confirmed the efficacy and safety of pitolisant in narcolepsy type 1 and 2 (Dauvilliers et al., 2013; Szakacs et al., 2017). Based upon these experimental data, both the European Medicines Agency and the United States Food and Drug Administration approved pitolisant, for the treatment of excessive sleepiness in narcolepsy (Kollb-Sielecka et al., 2017).

Schizophrenia Several findings suggest the possible involvement of the TMN in the pathogenesis of schizophrenia. Both positron emission tomography (PET) and postmortem studies have shown decreased histamine receptor-1 (H1R) binding in the medial and inferior PFC and anterior and posterior cingulate gyrus of schizophrenic patients (Iwabuchi et al., 2005; Nakai et al., 1991), while H3R radioligand binding was found to be increased in the dorsolateral PFC of schizophrenic patients (Jin et al., 2009). The globus pallidus of schizophrenia patients further showed a moderately increased H2R binding (Martinez-Mir et al., 1993).

395

As a result of these findings, H3R antagonist/inverse agonists were proposed to have potential therapeutic effects in schizophrenic preclinical animal models (AquinoMiranda et al., 2019; Sadek et al., 2016). However, a phase II exploratory study failed to show that H3R-antagonist had overall beneficial effects for schizophrenia patients (Jarskog et al., 2015). In agreement with this negative finding, a randomized double-blind placebo-controlled study showed a lack of cognitive improvement in schizophrenia patients by a H3R-antagonist (Haig et al., 2014). However, in open-label clinical trials, the H2Rantagonist famotidine was reported to have antipsychotic effects and to reduce negative schizophrenia symptoms (Kaminsky et al., 1990; Oyewumi et al., 1994; Rosse et al., 1996). A randomized clinical trial further confirmed the antipsychotic effects of famotidine, showing obvious improvements in both positive and negative symptoms of schizophrenia patients (Meskanen et al., 2013). Also a meta-analysis of eight double-blinded randomized placebo-controlled trials, testing the efficacy and tolerability of H2R antagonist as adjunct antipsychotic treatment, reported a lack of improvement in the overall symptoms of schizophrenia (Kishi and Iwata, 2015). Alterations in H1R, H2R or H3R binding were restricted to certain brain regions in schizophrenia patients. However, the antagonists may affect more brain areas. This may, at least partly, explain the discrepancy between clinical trials and human observations.

Tourette syndrome A series of crucial data supported that changes in the key enzyme involved in the production of neuronal histamine, i.e., HDC, could cause a rare familial case of Tourette syndrome (Ercan-Sencicek et al., 2010). A mutation (Trp317X) in the HDC gene truncated fulllength HDC protein to deplete the enzymatic activity of HDC (Ercan-Sencicek et al., 2010). Subsequent studies in HDC knockout mice have confirmed that this mutation causes tic-like behavioral and neurochemical abnormalities, which is in line with symptoms in patients and thus at least partly validates the HDC knockout mice as a model of tic pathophysiology (Castellan Baldan et al., 2014; Rapanelli et al., 2017). Furthermore, histamine deficiency lead to distinct Tourette syndromelike behavioral alterations strongly associated with a dopamine disbalance on the level of the striatum (Abdurakhmanova et al., 2019). Similar to what was found in HDC knockout mice, Tourette syndrome patients also suffer from sleep–wake impairments (Shan et al., 2015b). One Tourette syndrome patient had a substantial decrease in daytime sleepiness after being treated with pitolisant, whereas tic scores remained unaltered (Hartmann et al., 2012).

396

L. SHAN ET AL.

Both the efficacy of H3R antagonist/inverse agonists on tic symptoms and sleep–wake disorders in Tourette syndrome patients should be evaluated in a larger clinical trial.

AUTISM SPECTRUM DISORDER Literature presents scarce data about the histaminergic system in autism spectrum disorder. There is overlap in the genetic risks shared between Tourette syndrome and autism spectrum disorder in as far as the histaminergic system is involved, in particular for H1R- and H2R-mediated signaling (Fernandez et al., 2012). A recent postmortem investigation showed that the HMT-, H1R-, H2R-, and H3R-mRNA were significantly increased in the dorsolateral prefrontal cortex of autism spectrum disorder patients, which was replicated in an independent postmortem data set with 28 controls and 19 patients (Wright et al., 2017). Furthermore, there are some preliminary data on histamine receptor inverse agonist/antagonists in the treatment of autism spectrum disorder. Some symptoms such as irritability, hyperactivity, and the atypical pattern of eye contact were ameliorated after treatment with famotidine (Linday, 1997; Linday et al., 2001). Niaprazine, a H1R antagonist, showed positive effects in 52% of 25 autistic patients, in particular on hyperkinesia, unstable attention, and resistance to change frustration, mild anxiety signs, aggressiveness, and sleep disorders (Pipitone and D’Agata, 1999). The administration of the H3R antagonists DL77 or ciproxifan attenuates autistic-like behaviors in an autistic animal model, i.e., the prenatal valproic acid mouse model (Baronio et al., 2015; Eissa et al., 2018). This evidence provides a starting point for a more systematic investigation of the histaminergic system in the pathogenesis of autism spectrum disorder.

CONCLUSIONS The TMN is a highly heterogeneous structure, both in its neurochemical components and functional activities. The histaminergic system of the TMN contributes to neuropsychiatric disorders like PD, AD, HD, narcolepsy, schizophrenia, Tourette syndrome, and autism spectrum disorder. As diurnal fluctuations of HDC-mRNA and HDC immunohistochemistry have been reported and a loss of this HDC-mRNA fluctuation was observed in preclinical PD, PD, AD, and HD cases, the histamine system may be linked to sleep disorders in those patients. Collectively, the data indicate that the TMN subdivisions and their specific functional circuits deserve further systematic studies in human brain. In the human postmortem PD brain, endogenous histamine production remains unaltered, despite the accumulation of Lewy bodies and Lewy neuritis in the

TMN. Also the histamine level and histamine fibers were increased in the basal ganglia of PD. The H4R-mRNA is robustly (4.2- to 6.3-fold) increased in the basal ganglia of PD patients, and a preclinical study demonstrated this to be a promising antineuroinflammation target (Zhou et al., 2019). AD neuropathology, i.e., neurofibrillary tangles and amyloid plaques, are prominent in the TMN. The diminished histamine production in the middle part of the TMN appears to be largely compensated for by the remaining TMN neurons in AD. In females, but not in males, H3R- and HMT-mRNA expressions are upregulated from the mild cognitive impairment stages (Braak stages III–IV) and maintained into the clinical stages of AD (Braak stages V–VI). Confounding factors such as sex may contribute to the negative results of several recent randomized controlled trials with H3R antagonists/inverse agonists, which failed to show improvements in cognition in AD. Patients with HD have an increased activity of the histaminergic system, which is presumed to be associated with their sleep problems and aggressive behavior. The increased number of HDC marked histamine neurons in narcoleptic patients is further hypothesized to be associated with the presumed autoimmune etiology of the disorder. Both the European Medicines Agency and the United States Food and Drug Administration have approved the H3R-antagonist/inverse agonist pitolisant for the treatment of excessive daytime sleepiness in narcolepsy, as examples of how specific histaminergic compounds can have therapeutic effects in neuropsychiatric disorders.

ACKNOWLEDGMENTS Dr. Ling Shan has received funding from the European Union’s Horizon 2020 research and innovation program under the Marie Skłodowska-Curie grant (Agreement No. 707404). The authors are grateful for the valuable discussion with Profs. Ai-Min Bao (Zhejiang University), Paul J. Lucassen (University of Amsterdam), and Ruud Buijs (Universidad Nacional Autonoma de Mexico) on an earlier version of this manuscript, and to W.T.P. Verweij for secretarial assistance.

CONFLICTS OF INTEREST We declare that we have no conflicts of interest.

REFERENCES Abdurakhmanova S, Semenova S, Piepponen TP et al. (2019). Abnormal behavior, striatal dopamine turnover and opioid peptide gene expression in histamine-deficient mice. Genes Brain Behav 18: e12595.

THE TUBEROMAMILLARY NUCLEUS IN NEUROPSYCHIATRIC DISORDERS Agundez JA, Luengo A, Herraez O et al. (2008). Nonsynonymous polymorphisms of histamine-metabolising enzymes in patients with Parkinson’s disease. Neuromolecular Med 10: 10–16. Anichtchik OV, Huotari M, Peitsaro N et al. (2000). Modulation of histamine H3 receptors in the brain of 6-hydroxydopamine-lesioned rats. Eur J Neurosci 12: 3823–3832. Anichtchik OV, Peitsaro N, Rinne JO et al. (2001). Distribution and modulation of histamine H(3) receptors in basal ganglia and frontal cortex of healthy controls and patients with Parkinson’s disease. Neurobiol Dis 8: 707–716. Aquino-Miranda G, Rivera-Ramı´rez N, Ma´rquez-Go´mez R et al. (2019). Histamine H3 receptor activation reduces the impairment in prepulse inhibition (PPI)of the acoustic startle response and Akt phosphorylation induced by MK-801 (dizocilpine), antagonist at N-methyl-d-aspartate (NMDA)receptors. Prog Neuro-Psychopharmacol Biol Psychiatry 94: 109653. Auvinen S, Panula P (1988). Development of histamineimmunoreactive neurons in the rat brain. J Comp Neurol 276: 289–303. Aziz A, Fronczek R, Maat-Schieman M et al. (2008). Hypocretin and melanin-concentrating hormone in patients with Huntington disease. Brain Pathol 18: 474–483. Baronio D, Castro K, Gonchoroski T et al. (2015). Effects of an H3R antagonist on the animal model of autism induced by prenatal exposure to valproic acid. PLoS One 10: e0116363. Bassetti CL, Baumann CR, Dauvilliers Y et al. (2010). Cerebrospinal fluid histamine levels are decreased in patients with narcolepsy and excessive daytime sleepiness of other origin. J Sleep Res 19: 620–623. Braak H, Braak E, Bohl J (1993). Staging of Alzheimer-related cortical destruction. Eur Neurol 33: 403–408. Braak H, Braak E, Yilmazer D et al. (1996). Pattern of brain destruction in Parkinson’s and Alzheimer’s diseases. J Neural Transm 103: 455–490. Castellan Baldan L, Williams KA, Gallezot JD et al. (2014). Histidine decarboxylase deficiency causes Tourette syndrome: parallel findings in humans and mice. Neuron 81: 77–90. Dauvilliers Y, Bassetti C, Lammers GJ et al. (2013). Pitolisant versus placebo or modafinil in patients with narcolepsy: a double-blind, randomised trial. Lancet Neurol 12: 1068–1075. Dauvilliers Y, Delallee N, Jaussent I et al. (2012). Normal cerebrospinal fluid histamine and tele-methylhistamine levels in hypersomnia conditions. Sleep 35: 1359–1366. Egan M, Yaari R, Liu L et al. (2012). Pilot randomized controlled study of a histamine receptor inverse agonist in the symptomatic treatment of AD. Curr Alzheimer Res 9: 481–490. Eissa N, Jayaprakash P, Azimullah S et al. (2018). The histamine H3R antagonist DL77 attenuates autistic behaviors in a prenatal valproic acid-induced mouse model of autism. Sci. Rep 8: 13077. Ercan-Sencicek AG, Stillman AA, Ghosh AK et al. (2010). L-histidine decarboxylase and Tourette’s syndrome. N Engl J Med 362: 1901–1908.

397

Fernandez TV, Sanders SJ, Yurkiewicz IR et al. (2012). Rare copy number variants in tourette syndrome disrupt genes in histaminergic pathways and overlap with autism. Biol Psychiatry 71: 392–402. Fontana A, Gast H, Reith W et al. (2010). Narcolepsy: autoimmunity, effector T cell activation due to infection, or T cell independent, major histocompatibility complex class II induced neuronal loss? Brain 133: 1300–1311. Franco P, Dauvilliers Y, Inocente CO et al. (2019). Impaired histaminergic neurotransmission in children with narcolepsy type 1. CNS Neurosci Ther 25: 386–395. Frisch C, Hasen€ ohrl RU, Krauth J et al. (1998). Anxiolytic-like behavior after lesion of the tuberomamillary nucleus E2-region. Exp Brain Res 119: 260–264. Garbarg M, Javoy-Agid F, Schwartz JC et al. (1983). Brain histidine decarboxylase activity in Parkinson’s disease. Lancet 1: 74–75. Grove RA, Harrington CM, Mahler A et al. (2014). A randomized, double-blind, placebo-controlled, 16-week study of the H3 receptor antagonist, GSK239512 as a monotherapy in subjects with mild-to-moderate Alzheimer’s disease. Curr Alzheimer Res 11: 47–58. Guo RX, Anaclet C, Roberts JC et al. (2009). Differential effects of acute and repeat dosing with the H3 antagonist GSK189254 on the sleep-wake cycle and narcoleptic episodes in ox / mice. Br J Pharmacol 157: 104–117. Haas H, Panula P (2003). The role of histamine and the tuberomamillary nucleus in the nervous system. Nat Rev Neurosci 4: 121–130. Haas HL, Sergeeva OA, Selbach O (2008). Histamine in the nervous system. Physiol Rev 88: 1183–1241. Haig GM, Bain E, Robieson W et al. (2014). A randomized trial of the efficacy and safety of the H3 antagonist ABT288 in cognitive impairment associated with schizophrenia. Schizophr Bull 40: 1433–1442. Han F, Lin L, Warby SC et al. (2011). Narcolepsy onset is seasonal and increased following the 2009 H1N1 pandemic in China. Ann Neurol 70: 410–417. Hartmann A, Worbe Y, Arnulf I (2012). Increasing histamine neurotransmission in Gilles de la Tourette syndrome. J Neurol 259: 375–376. Huang ZL, Mochizuki T, Qu WM et al. (2006). Altered sleepwake characteristics and lack of arousal response to H3 receptor antagonist in histamine H1 receptor knockout mice. Proc Natl Acad Sci U S A 103: 4687–4692. Ishunina TA, van Heerikhuize JJ, Ravid R et al. (2003). Estrogen receptors and metabolic activity in the human tuberomamillary nucleus: changes in relation to sex, aging and Alzheimer’s disease. Brain Res 988: 84–96. Iwabuchi K, Ito C, Tashiro M et al. (2005). Histamine H1 receptors in schizophrenic patients measured by positron emission tomography. Eur Neuropsychopharmacol 15: 185–191. Jarskog LF, Lowy MT, Grove RA et al. (2015). A phase II study of a histamine H(3) receptor antagonist GSK239512 for cognitive impairment in stable schizophrenia subjects on antipsychotic therapy. Schizophr Res 164: 136–142.

398

L. SHAN ET AL.

Jim
enez-Jim
enez FJ, Alonso-Navarro H, Garcı´a-Martı´n E et al. (2016). Thr105Ile (rs11558538) polymorphism in the histamine N -methyltransferase (HNMT) gene and risk for Parkinson disease. Med. (United States) 95: e4147. Jin CY, Anichtchik O, Panula P (2009). Altered histamine H3 receptor radioligand binding in post-mortem brain samples from subjects with psychiatric diseases. Br J Pharmacol 157: 118–129. John J, Thannickal TC, McGregor R et al. (2013). Greatly increased numbers of histamine cells in human narcolepsy with cataplexy. Ann Neurol 74: 786–793. Kaminsky R, Moriarty TM, Bodine J et al. (1990). Effect of famotidine on deficit symptoms of schizophrenia. Lancet 335: 1351–1352. Kanbayashi T, Kodama T, Kondo H et al. (2009). CSF histamine contents in narcolepsy, idiopathic hypersomnia and obstructive sleep apnea syndrome. Sleep 32: 181–187. Kano M, Fukudo S, Tashiro A et al. (2004). Decreased histamine H1 receptor binding in the brain of depressed patients. Eur J Neurosci 20: 803–810. Kishi T, Iwata N (2015). Efficacy and tolerability of Histamine-2 receptor antagonist adjunction of antipsychotic treatment in schizophrenia: a meta-analysis of randomized placebo-controlled trials. Pharmacopsychiatry 48: 30–36. Kollb-Sielecka M, Demolis P, Emmerich J et al. (2017). The European medicines agency review of pitolisant for treatment of narcolepsy: summary of the scientific assessment by the Committee for Medicinal Products for human use. Sleep Med 33: 125–129. Koutcherov Y, Mai JK, Ashwell KWS et al. (2002). Organization of human hypothalamus in fetal development. J Comp Neurol 446: 301–324. Kremer B, Goldberg P, Andrew SE et al. (1994). A worldwide study of the Huntington’s disease mutation. The sensitivity and specificity of measuring CAG repeats. N Engl J Med 330: 1401–1406. Kruijver FP, Balesar R, Espila AM et al. (2003). Estrogenreceptor-beta distribution in the human hypothalamus: similarities and differences with ER alpha distribution. J Comp Neurol 466: 251–277. Kruijver FP, Balesar R, Espila AM et al. (2002). Estrogen receptor-alpha distribution in the human hypothalamus in relation to sex and endocrine status. J Comp Neurol 454: 115–139. Kubo M, Kishi T, Matsunaga S et al. (2015). Histamine H3 receptor antagonists for Alzheimer’s disease: a systematic review and meta-analysis of randomized placebocontrolled trials. J Alzheimers Dis 48: 667–671. Latorre D, Kallweit U, Armentani E et al. (2018). T cells in patients with narcolepsy target self-antigens of hypocretin neurons. Nature 562: 63–68. Ledesma MC, Garcia-Martin E, Alonso-Navarro H et al. (2008). The nonsynonymous Thr105Ile polymorphism of the histamine N-methyltransferase is associated to the risk of developing essential tremor. Neuromolecular Med 10: 356–361.

Lin JS (2000). Brain structures and mechanisms involved in the control of cortical activation and wakefulness, with emphasis on the posterior hypothalamus and histaminergic neurons. Sleep Med Rev 4: 471–503. Lin JS, Dauvilliers Y, Arnulf I et al. (2008). An inverse agonist of the histamine H(3) receptor improves wakefulness in narcolepsy: studies in orexin / mice and patients. Neurobiol Dis 30: 74–83. Lin L, Faraco J, Li R et al. (1999). The sleep disorder canine narcolepsy is caused by a mutation in the hypocretin (orexin) receptor 2 gene. Cell 98: 365–376. Linday LA (1997). Oral famotidine: a potential treatment for children with autism. Med Hypotheses 48: 381–386. Linday LA, Tsiouris JA, Cohen IL et al. (2001). Famotidine treatment of children with autistic spectrum disorders: pilot research using single subject research design. J Neural Transm 108: 593–611. Liu CQ, Chen Z, Liu FX et al. (2007). Involvement of brain endogenous histamine in the degeneration of dopaminergic neurons in 6-hydroxydopamine-lesioned rats. Neuropharmacology 53: 832–841. Liu CQ, Shan L, Balesar R et al. (2010). A quantitative in situ hybridization protocol for formalin-fixed paraffin-embedded archival post-mortem human brain tissue. Methods 52: 359–366. Martinez-Mir MI, Pollard H, Moreau J et al. (1993). Loss of striatal histamine H2 receptors in Huntington’s chorea but not in Parkinson’s disease: comparison with animal models. Synapse 15: 209–220. Mazurkiewicz-Kwilecki IM, Nsonwah S (1989). Changes in the regional brain histamine and histidine levels in postmortem brains of Alzheimer patients. Can J Physiol Pharmacol 67: 75–78. Meskanen K, Ekelund H, Laitinen J et al. (2013). A randomized clinical trial of histamine 2 receptor antagonism in treatment-resistant schizophrenia. J Clin Psychopharmacol 33: 472–478. Mignot E, Lammers GJ, Ripley B et al. (2002). The role of cerebrospinal fluid hypocretin measurement in the diagnosis of narcolepsy and other hypersomnias. Arch Neurol 59: 1553–1562. Mignot E, Lin L, Rogers W et al. (2001). Complex HLA-DR and -DQ interactions confer risk of narcolepsy-cataplexy in three ethnic groups. Am J Hum Genet 68: 686–699. Miklos IH, Kovacs KJ (2003). Functional heterogeneity of the responses of histaminergic neuron subpopulations to various stress challenges. Eur J Neurosci 18: 3069–3079. Motawaj M, Peoc’H K, Callebert J et al. (2010). CSF levels of the histamine metabolite tele-methylhistamine are only slightly decreased in Alzheimer’s disease. J Alzheimers Dis 22: 861–871. Naganuma F, Nakamura T, Yoshikawa T et al. (2017). Histamine N-methyltransferase regulates aggression and the sleep-wake cycle. Sci Rep 7. Nakai T, Kitamura N, Hashimoto T et al. (1991). Decreased histamine H1 receptors in the frontal cortex of brains from patients with chronic schizophrenia. Biol Psychiatry 30: 349–356.

THE TUBEROMAMILLARY NUCLEUS IN NEUROPSYCHIATRIC DISORDERS Nakamura S, Kawamata T, Yasuhara O et al. (1991). The histochemical demonstration of monoamine oxidase-containing neurons in the human hypothalamus. Neuroscience 44: 457–463. Nakamura S, Ohnishi K, Nishimura M et al. (1996). Large neurons in the tuberomamillary nucleus in patients with Parkinson’s disease and multiple system atrophy. Neurology 46: 1693–1696. Nakamura S, Takemura M, Ohnishi K et al. (1993). Loss of large neurons and occurrence of neurofibrillary tangles in the tuberomamillary nucleus of patients with Alzheimer’s disease. Neurosci Lett 151: 196–199. Nishino S, Ripley B, Overeem S et al. (2000). Hypocretin (orexin) deficiency in human narcolepsy. Lancet 355: 39–40. Nishino S, Sakurai E, Nevsimalova S et al. (2009). Decreased CSF histamine in narcolepsy with and without low CSF hypocretin-1 in comparison to healthy controls. Sleep 32: 175–180. Oyewumi LK, Vollick D, Merskey H et al. (1994). Famotidine as an adjunct treatment of resistant schizophrenia. J Psychiatry Neurosci 19: 145–150. Palada V, Terzic J, Mazzulli J et al. (2012). Histamine N-methyltransferase Thr105Ile polymorphism is associated with Parkinson’s disease. Neurobiol Aging 33: 836. e1–836.e3. Panula P, Nuutinen S (2013). The histaminergic network in the brain: basic organization and role in disease. Nat Rev Neurosci 14: 472–487. Panula P, Rinne J, Kuokkanen K et al. (1997). Neuronal histamine deficit in Alzheimer’s disease. Neuroscience 82: 993–997. Parmentier R, Ohtsu H, Djebbara-Hannas Z et al. (2002). Anatomical, physiological, and pharmacological characteristics of histidine decarboxylase knock-out mice: evidence for the role of brain histamine in behavioral and sleep-wake control. J Neurosci 22: 7695–7711. Passani MB, Blandina P (2011). Histamine receptors in the CNS as targets for therapeutic intervention. Trends Pharmacol Sci 32: 242–249. Pedersen NW, Holm A, Kristensen NP et al. (2019). CD8 + T cells from patients with narcolepsy and healthy controls recognize hypocretin neuron-specific antigens. Nat Commun 10: 837. Peyron C, Faraco J, Rogers W et al. (2000). A mutation in a case of early onset narcolepsy and a generalized absence of hypocretin peptides in human narcoleptic brains. Nat Med 6: 991–997. Phillips W, Shannon KM, Barker RA (2008). The current clinical management of Huntington’s disease. Mov Disord 23: 1491–1504. Pipitone E, D’Agata M (1999). Niaprazine in the treatment of autistic disorder. J Child Neurol 14: 547–550. Prell GD, Green JP (1991). Histamine metabolites and prosmethylimidazoleacetic acid in human cerebrospinal fluid. Agents Actions Suppl 33: 343–363. Prell GD, Khandelwal JK, Burns RS et al. (1991). Levels of pros-methylimidazoleacetic acid: correlation with severity

399

of Parkinson’s disease in CSF of patients and with the depletion of striatal dopamine and its metabolites in MPTP-treated mice. J Neural Transm Park Dis Dement Sect 3: 109–125. Rapanelli M, Frick L, Bito H et al. (2017). Histamine modulation of the basal ganglia circuitry in the development of pathological grooming. Proc Natl Acad Sci U S A 114: 6599–6604. Rinne JO, Anichtchik OV, Eriksson KS et al. (2002). Increased brain histamine levels in Parkinson’s disease but not in multiple system atrophy. J Neurochem 81: 954–960. Rocha SM, Saraiva T, Cristovao AC et al. (2016). Histamine induces microglia activation and dopaminergic neuronal toxicity via H1 receptor activation. J Neuroinflammation 13: 137. Rosse RB, Kendrick K, Fay-McCarthy M et al. (1996). An open-label study of the therapeutic efficacy of high-dose famotidine adjuvant pharmacotherapy in schizophrenia: preliminary evidence for treatment efficacy. Clin Neuropharmacol 19: 341–348. Rye DB (2012). Inability to replicate cerebrospinal fluid histamine deficits in the primary hypersomnias: a back to the drawing board moment. Sleep 35: 1315–1317. Sadek B, Saad A, Sadeq A et al. (2016). Histamine H3 receptor as a potential target for cognitive symptoms in neuropsychiatric diseases. Behav Brain Res 312: 415–430. Sakurai T, Amemiya A, Ishii M et al. (1998). Orexins and orexin receptors: a family of hypothalamic neuropeptides and G protein-coupled receptors that regulate feeding behavior. Cell 92: 573–585. Saper CB, German DC (1987). Hypothalamic pathology in Alzheimer’s disease. Neurosci Lett 74: 364–370. Schinkelshoek MS, Fronczek R, Kooy-Winkelaar EMC et al. (2019). H1N1 hemagglutinin-specific HLA-DQ6-restricted CD4+ T cells can be readily detected in narcolepsy type 1 patients and healthy controls. J Neuroimmunol 332: 167–175. Sergeeva OA, Eriksson KS, Sharonova IN et al. (2002). GABA (A) receptor heterogeneity in histaminergic neurons. Eur J Neurosci 16: 1472–1482. Shan L, Bao AM, Swaab DF (2015a). The human histaminergic system in neuropsychiatric disorders. Trends Neurosci 38: 167–177. Shan L, Bossers K, Luchetti S et al. (2012a). Alterations in the histaminergic system in the substantia nigra and striatum of Parkinson’s patients: a postmortem study. Neurobiol Aging 33: 1488 e1–13. Shan L, Bossers K, Unmehopa U et al. (2012b). Alterations in the histaminergic system in Alzheimer’s disease: a postmortem study. Neurobiol Aging 33: 2585–2598. Shan L, Dauvilliers Y, Siegel JM (2015b). Interactions of the histamine and hypocretin systems in CNS disorders. Nat Rev Neurol 11: 401–413. Shan L, Hofman MA, van Wamelen DJ et al. (2012c). Diurnal fluctuation in histidine decarboxylase expression, the rate

400

L. SHAN ET AL.

limiting enzyme for histamine production, and its disorder in neurodegenerative diseases. Sleep 35: 713–715. Shan L, Liu CQ, Balesar R et al. (2012d). Neuronal histamine production remains unaltered in Parkinson’s disease despite the accumulation of Lewy bodies and Lewy neurites in the tuberomamillary nucleus. Neurobiol Aging 33: 1343–1344. Shan L, Qi XR, Balesar R et al. (2013). Unaltered histaminergic system in depression: a postmortem study. J Affect Disord 146: 220–223. Sherer TB, Betarbet R, Kim JH et al. (2003). Selective microglial activation in the rat rotenone model of Parkinson’s disease. Neurosci Lett 341: 87–90. Sherin JE, Elmquist JK, Torrealba F et al. (1998). Innervation of histaminergic tuberomamillary neurons by GABAergic and galaninergic neurons in the ventrolateral preoptic nucleus of the rat. J Neurosci 18: 4705–4721. Sukhov RR, Walker LC, Rance NE et al. (1995). Opioid precursor gene expression in the human hypothalamus. J Comp Neurol. Szakacs Z, Dauvilliers Y, Mikhaylov V et al. (2017). Safety and efficacy of pitolisant on cataplexy in patients with narcolepsy: a randomised, double-blind, placebo-controlled trial. Lancet Neurol 16: 200–207. Thannickal TC, Moore RY, Nienhuis R et al. (2000). Reduced number of hypocretin neurons in human narcolepsy. Neuron 27: 469–474. Trottier S, Chotard C, Traiffort E et al. (2002). Colocalization of histamine with GABA but not with galanin in the human tuberomamillary nucleus. Brain Res 939: 52–64. Valko PO, Gavrilov YV, Yamamoto M et al. (2013). Increase of histaminergic tuberomamillary neurons in narcolepsy. Ann Neurol 74: 794–804. van de Nes JA, Kamphorst W, Ravid R et al. (1998). Comparison of beta-protein/A4 deposits and Alz-50stained cytoskeletal changes in the hypothalamus and adjoining areas of Alzheimer’s disease patients: amorphic plaques and cytoskeletal changes occur independently. Acta Neuropathol 96: 129–138.

Van Wamelen DJ, Shan L, Aziz NA et al. (2011). Functional increase of brain histaminergic signaling in Huntington’s disease. Brain Pathol 21: 419–427. Venner A, Mochizuki T, De Luca R et al. (2019). Reassessing the role of histaminergic tuberomamillary neurons in arousal control. J Neurosci 1032–19. Vizuete ML, Merino M, Venero JL et al. (2000). Histamine infusion induces a selective dopaminergic neuronal death along with an inflammatory reaction in rat substantia nigra. J Neurochem 75: 540–552. Williams RH, Chee MJ, Kroeger D et al. (2014). Optogeneticmediated release of histamine reveals distal and autoregulatory mechanisms for controlling arousal. J Neurosci 34: 6023–6029. Wright C, Shin JH, Rajpurohit A et al. (2017). Altered expression of histamine signaling genes in autism spectrum disorder Transl Psychiatry 7. Yang X, Liu C, Zhang J et al. (2015). Association of Histamine N-methyltransferase Thr105Ile polymorphism with Parkinson’s disease and schizophrenia in Han Chinese: a case-control study. PLoS One 10: e0119692. Yu X, Ma Y, Harding EC et al. (2019). Genetic lesioning of histamine neurons increases sleep-wake fragmentation and reveals their contribution to modafinil-induced wakefulness Sleep 42. Yu X, Ye Z, Houston CM et al. (2015). Wakefulness is governed by GABA and histamine Cotransmission. Neuron 87: 164–178. Yu X, Zecharia A, Zhang Z et al. (2014). Circadian factor BMAL1 in histaminergic neurons regulates sleep architecture. Curr Biol 24: 2838–2844. Zeitzer JM, Kodama T, Buckmaster CL et al. (2012). Timecourse of cerebrospinal fluid histamine in the wakeconsolidated squirrel monkey. J Sleep Res 21: 189–194. Zhou P, Homberg JR, Fang Q et al. (2019). Histamine-4 receptor antagonist JNJ7777120 inhibits pro-inflammatory microglia and prevents the progression of Parkinson-like pathology and behaviour in a rat model. Brain Behav Immun 76: 61–73.

Section 13 Subthalamic nucleus

This page intentionally left blank

Handbook of Clinical Neurology, Vol. 180 (3rd series) The Human Hypothalamus: Middle and Posterior Region D.F. Swaab, F. Kreier, P.J. Lucassen, A. Salehi, and R.M. Buijs, Editors https://doi.org/10.1016/B978-0-12-820107-7.00025-2 Copyright © 2021 Elsevier B.V. All rights reserved

Chapter 25

Imaging of the human subthalamic nucleus ANNEKE ALKEMADE AND BIRTE U. FORSTMANN* Integrative Model-Based Cognitive Neuroscience Research Unit, University of Amsterdam, Amsterdam, The Netherlands

Abstract The human subthalamic nucleus (STN) is a small lens shaped iron rich nucleus, which has gained substantial interest as a target for deep brain stimulation surgery for a variety of movement disorders. The internal anatomy of the human STN has not been fully elucidated, and an intensive debate, discussing the level of overlap between putative limbic, associative, and motor zones within the STN is still ongoing. In this chapter, we have summarized anatomical information obtained using different neuroimaging modalities focusing on the anatomy of the STN. Additionally, we have highlighted a number of major challenges faced when using magnetic resonance imaging (MRI) approaches for the visualization of small iron rich deep brain structures such as the STN. In vivo MRI and postmortem microscopy efforts provide valuable complementary information on the internal structure of the STN, although the results are not always fully aligned. Finally, we provide an outlook on future efforts that could contribute to the development of an integrative research approach that will help with the reconciliation of seemingly divergent results across research approaches.

INTRODUCTION The human subthalamic nucleus (STN) is a biconvexshaped structure located above the cerebral peduncle, which contains approximately 550,000 neurons (Parent et al., 1996). The STN is part of the basal ganglia network which, together with cortex, controls the execution of planned motivated behavior involving motor, cognitive, and limbic circuits (Haber, 2003). The seminal work by Bergman et al. (1990), showing that lesioning of the STN reversed experimental 1-methyl-4-phenyl-1,2,3,6tetrahydropyridine (MPTP) parkinsonism in monkeys has formed the scientific basis for the development of current deep brain stimulation (DBS) procedures targeting the STN in a variety of neuromotor diseases. The development of DBS has sparked a strong scientific and clinical interest in the structure and function of the STN. Together with the increasing interest, imaging efforts to visualize the human STN in vivo have become increasingly successful due to the development of

ultra-high-resolution 7 Tesla (T) magnetic resonance imaging (MRI) systems, as well as the adaptation of MRI sequences to accommodate the tissue properties of the STN and its surrounding tissue. Unfortunately, despite the dramatic improvements that have been made in the visualization of the STN, in vivo neuroimaging approaches continue to pose challenges. To date, there is no consensus on which neuroimaging technique is best to visualize the STN for surgical planning (Brunenberg et al., 2011). As a result, methods vary greatly between centers, and some groups apply indirect visualization techniques using anatomical landmarks and/or atlases that incorporate anatomical and functional data. The alternative approach includes direct visualization and is reported using a variety of MRI contrasts (Brunenberg et al., 2011). In general, neuroimaging approaches and the associated compromises are chosen to match the clinical or basic research question at hand. Based on the purpose

*Correspondence to: Birte U Forstmann, Ph.D., Integrative Model-based Cognitive Neuroscience (IMCN) research unit, University of Amsterdam, Nieuwe Achtergracht 129B j Room G3.06, PO Box 15926 j 1001 NK Amsterdam, The Netherlands. Tel: + 31-6-15324988, E-mail: [email protected]

404

A. ALKEMADE AND B.U. FORSTMANN

for which the STN is imaged, a number of factors affecting the imaging procedure are weighed and together determine the imaging strategy. In this chapter, we will discuss different approaches for the neuroimaging of the STN for clinical and research purposes, as well as the underlying functional neuroanatomy of the nucleus including the challenges associated with fMRI studies of the STN. Finally, we will discuss to what extent in vivo neuroimaging data can be merged with the underlying functional neuroanatomy of the STN.

IN VIVO NEUROIMAGING OF THE STN For in vivo neuroimaging of the human STN, MRI represents the method of choice (Forstmann et al., 2017b). MRI techniques are based on the imaging of the magnetic behavior of atomic nuclei with a net positive charge (Grover et al., 2015). Even though MRI techniques are based on the same principle, the obtained images vary as a result of differences in the used hardware, applied scan sequences, as well as data postprocessing and contrast calculations. Together, this has resulted in a plethora of different types of MRI images of the brain. STN MRI is notoriously challenging due to the physical location of the nucleus, which is tucked away deep in the crowded subcortex, at a large, nearly equal distance from the MR receiver coils. Additionally, the STN directly borders the substantia nigra, which, like the STN, contains high concentrations of iron (de Hollander et al., 2014). MRI acquisition strategies for imaging of the STN reported in literature wildly vary, and the applied methods are dependent on the availability of hardware, implemented scan sequences, as well as the purpose for which the STN is imaged. In this chapter, we do not intend to provide an overview of all in vivo STN neuroimaging efforts that are available in the literature. We will discuss a number of crucial factors in the MRI imaging protocol that can be optimized to achieve the best account possible of the internal structure of the human STN, as well as comparisons across groups.

Benefits of high MRI field strength An increasing number of 7 T MRI scanners are available for brain imaging purposes (Forstmann et al., 2017a; Keuken et al., 2018; Ladd et al., 2018). Theoretically, higher field strengths allow for better signal-to-noise and contrast-to-noise ratios, and following that line of reasoning, higher field strengths can deliver better brain images (Kraff et al., 2015; Trattnig et al., 2018). Additionally, a more powerful gradient system, available in a subset of MRI systems, will further benefit the obtained image quality. The use of 7 T MRI is still largely confined to research protocols, whereas for clinical imaging purposes, 1.5 or 3 T is commonly applied. In DBS,

neurosurgeons are also guided by other information, such as brain atlases, and information that is obtained intraoperatively, including electrophysiological assessments, and the direct clinical effect induced by the insertion of the electrode (Lozano et al., 2019). These assessments together allow for a reliable positioning of the electrode for surgical purposes. Although 7 T MRI systems with powerful gradients are best suited for direct imaging of the STN, visualization using a 7 T scanner with a less powerful gradient system, or a more commonly available clinically approved 3 T scanner can also be achieved in a reliable fashion (Fig. 25.1). For each MRI system, image quality can be optimized by investing in the development of tailored scan acquisition protocols (de Hollander et al., 2017). Additionally, the averaging of multiple scan repetitions will benefit the MRI contrast, although this will increase scan times. Surprisingly, clinically implemented scan sequences are usually not optimized for STN imaging (Alkemade et al., 2017). Finally, the optimal contrast may differ between field strengths, which further complicates the interpretation of literature reports, and further complicates the choice for MRI contrasts based on available studies (Keuken et al., 2016; Bot et al., 2019).

Isotropy A logical MRI parameter to optimize for the visualization of the three-dimensional (D) STN is the voxel size. In general, smaller voxel volumes are preferred. However, not only voxel volume but also voxel shape should be considered (Mulder et al., 2019). Isotropic voxels are voxels with the same dimension in every direction. Nonisotropic voxels are voxels that provide a high in-plane resolution combined with a relatively larger slice thickness, resulting in rod-shaped voxels. The application of high in-plane resolution in combination with a larger slice thickness allows for a clear identification of smaller structures such as the STN in a single plane, and often a single slice. For studies focused on larger brain structures, size and shape of the individual voxels will have limited effects on the anatomical delineation and size estimates of a structure. The effects for smaller nuclei in the brain can, however, be detrimental. To quantify the effects of voxel size on STN imaging in a systematic fashion, we performed simulation studies in which we varied voxel size, shape, and imaging orientation on an ellipsoid representative in size for the STN (Mulder et al., 2019). We found that larger slice thickness, reflecting a stronger anisotropy resulted in a more substantial overestimation of the structure volume (Fig. 25.2). Importantly, volume estimates consistently deviated more than 50% from their predefined volume when slice thickness had a threefold anisotropy for a resolution of

IMAGING OF THE HUMAN SUBTHALAMIC NUCLEUS

405

Fig. 25.1. Comparisons of MRI contrasts for the visualization of the healthy human subthalamic nucleus (STN). The upper panels show STN visibility using a 3 Telsa (T) T2-weighted image with anisotropic voxels, representative for clinical MRI imaging, which does not allow discerning of the STN in the sagittal and coronal view. The second row of panels shows the STN in QSM contrasts of the same subject, using an optimized 3 T scan. The bottom row shows QSM contrasts of the same subject in an optimized 7 T image. Note the improved STN-SN border visibility in the coronal 7 T contrast. The red nucleus (RN) and substantia nigra (SN) are indicated for anatomical orientation.

1.0  1.0 mm in-plane voxel size (Mulder et al., 2019). Given that the average voxel volume used for the visualization of subcortical structures on 7 T MRI reported in the scientific literature is 1.09 mm3 (Keuken et al., 2018), and voxels >1 mm isotropic are common, it is clear that further optimization of voxel size and shape represents a step forward.

(Quantitative) MRI characteristics of the STN As compared to cortical brain areas, the STN and other nuclei of the basal ganglia have high iron concentrations (Schafer et al., 2012; Deistung et al., 2013; de Hollander et al., 2014; Birkl et al., 2016; Alkemade et al., 2017). The high iron content affects MRI parameters, causing a substantial shortening of the T1 and T2 relaxation times. This means that for optimal imaging of the STN, echo times of the scan sequence need to be substantially shorter than echo times commonly used to image the neocortex (Schafer et al., 2012; Forstmann et al., 2017b). Interestingly, the iron concentrations can also be used to improve imaging quality, and they can serve as input for quantitative MRI studies.

Quantitative MRI refers to the calculation and analyses of maps that project meaningful physical or chemical variables that can be expressed in physical units and compared between and within tissue regions among individuals (Pierpaoli, 2010). The read out of multiple echo’s allows the detection of small susceptibility changes, as well as correction for multiexponential T2*-decay (St€uber et al., 2014). These parameters can be used for the calculation of iron concentrations in the STN, by means of quantitative susceptibility mapping (QSM) (St€uber et al., 2014). Detailed analyses of QSM contrasts of the STN have demonstrated an inhomogeneous distribution of iron within the nucleus, more specifically, an iron gradient, with highest iron concentrations in the ventromedial part of the nucleus (de Hollander et al., 2014). The demonstration of the iron gradient illustrates the potential value of MRI and more specifically QSM for studies targeting the internal structure of the STN in vivo. Quantitative comparisons of delineations made independently on QSM and T2* contrasts in volunteers with and without Parkinson’s disease revealed that QSM contrasts give a higher interrater agreement as compared to T2* (Alkemade et al., 2017) (Fig. 25.3). This can be interpreted as an improved

406

A. ALKEMADE AND B.U. FORSTMANN

Fig. 25.2. Simulations of the effects of voxel shape and size on the visualization of the human STN, which is represented as an ellipsoid. Note the deformations resulting from the increase in slice thickness. Reproduced with permission from Mulder MJ et al. (2019). Size and shape matter: the impact of voxel geometry on the identification of small nuclei. PLoS One 14: e0215382. Bergsland N (Ed.). doi: 10.1371/journal.pone.0215382. Public Library of Science.

visibility of the STN (Alkemade et al., 2017). Interestingly, particularly volunteers with Parkinson’s disease benefitted from the use of iron sensitive contrasts, stressing the potential benefit of sequence optimization and the implementation of iron sensitive contrasts for clinical purposes (Fig. 25.3) (Alkemade et al., 2017). It is important to note that QSM, which is calculated using the phase images, can vary with the use of different calculation protocols. The use of various methods of calculation

complicate direct comparisons between studies available from scientific literature.

Interindividual variation of STN volume and location In vivo MRI measurements of the STN allow the study of a larger number of individuals as compared to postmortem analyses to obtain estimations of the variability in

IMAGING OF THE HUMAN SUBTHALAMIC NUCLEUS

407

Fig. 25.3. STN visualization in T2*-weighted and QSM contrasts calculated from a single MRI scan. The red and blue outlines represent the independent delineations by two raters. Note the increased agreement in the QSM images, which was particularly higher in volunteers with Parkinson’s disease (PD). Figure was reproduced from Alkemade A et al. (2017). Comparison of T2*weighted and QSM contrasts in Parkinson’s disease to visualize the STN with MRI. PLoS One 12: e0176130. Jiang Q (Ed.). doi: 10.1371/journal.pone.0176130. Public Library of Science, with permission. Table 25.1 Volume estimations of the human subthalamic nucleus STN volume (mm3)

Applied method

References

64 120 175 141 106 174 100.5 157 109 180 99 131 Average: 129.7 (SD 36.5)

Microscopy Microscopy Microscopy Microscopy MRI 9.4 T Microscopy MRI 7.0 T Microscopy MRI 7.0 T Microscopy MRI 3.0 T Microscopy

Fussenich (1967) Hardman et al. (2002) Levesque and Parent (2005) Lange et al. (1976) Massey et al. (2012) Nowinski et al. (2005) Plantinga et al. (2016) von Bonin and Shariff (1951) Weiss et al. (2015) Yelnik and Percheron (1979) Zwirner et al. (2017) Zwirner et al. (2017)

Adapted from Mulder MJ et al. (2019). Size and shape matter: the impact of voxel geometry on the identification of small nuclei. PLoS One 14: e0215382. Bergsland N (Ed.). doi: 10.1371/journal.pone. 0215382. Public Library of Science.

STN volume, as well as location. As described previously, MRI results are strongly dependent on the MRI acquisition protocol, as well as on the contrast calculations and data analyses. Size estimates of the postmortem STN are available from literature and are summarized in Table 25.1. On average the size of the STN is 129.7 mm3. Numbers obtained from in vivo MRI are often based on conjunct volumes, which only include voxels included in STN delineations by two independent researchers (Keuken et al., 2013; Keuken and Forstmann, 2015; Alkemade et al., 2017). As a consequence of the chosen method of analysis, these numbers represent a conservative estimation of STN size. We initially reported an average

conjunct STN volume in healthy volunteers of 69.2 mm3 when the STN was delineated on T2*weighted images. Using the QSM contrast, we found that the observed STN volume became substantially larger (82.34 mm3) (Alkemade et al., 2017). Similar findings were present in volunteers with PD. The observed increase in STN volume could be explained by improved visibility of the STN in QSM images, which was evidenced by higher interrater agreement scores (Dice and Dice, 1945; Alkemade et al., 2017). Interestingly, size estimates were not affected by age, but the location of the STN shifts in lateral directions with increasing age. This effect could not be explained by widening of the ventricular cavities as a result of aging, and warrants replication as well as further investigation (Keuken et al., 2013).

Diffusion-weighted imaging Resolving the connectivity profile of the STN could represent another approach to understand the internal structure of the human STN. Lambert et al. (2012) have made the first step in using probabilistic tractography to estimate the spatial distribution of white matter pathways. This study confirmed the presence of white matter connections to other limbic, associative, and motor brain regions. A large proportion of the STN voxels contained mixed connection profiles corresponding to limbic, associative, and motor functions of the STN. The question rises whether this could be attributed to partial volume effects, or whether connection profiles are highly intermixed within the STN (Alkemade and Forstmann, 2014). Further development of DWI techniques is needed to resolve this question.

Functional (f )MRI The STN is implicated in action selection, cognitive control, and response inhibition (Alexander and Crutcher, 1990; Parent and Hazrati, 1995; Redgrave et al., 1999;

408

A. ALKEMADE AND B.U. FORSTMANN

Middleton and Strick, 2000; Frank, 2006; Aron, 2011). Measurements of blood oxygenation level-dependent MRI allow assessment of brain activity (Ogawa et al., 1990), and to target questions on STN function, development of fMRI approaches are a logical step. These have proven to be technically challenging, since the size of 3 T fMRI voxels is commonly in the range of 3 mm isotropic (de Hollander et al., 2017). Given the size of the STN, this results in fewer than 5 voxels per STN, which are likely to suffer from partial voluming effects, and therefore may also incorporate signal from adjoining structures including the substantia nigra, and the lateral hypothalamus (de Hollander et al., 2015, 2017). These challenges are further aggravated in fMRI studies using standardized analysis techniques for which the individual brain scans are registered to a standard space, which does not accommodate the interindividual variation in the anatomy of the basal ganglia (de Hollander et al., 2017). Despite these challenges, several groups have made efforts to target the STN region using functional MRI. Pathways between the STN and the inferior frontal cortex, as well as the presupplementary motor area were investigated (Aron et al., 2007), followed by investigations on the connectivity between the STN and the presupplementary motor area, primary motor cortex, anterior cingulate cortex inferior frontal gyrus, and the striatum (Forstmann et al., 2010). STN regions of interest were tested for their connections to motor, associative, and limbic areas in the brain (Brunenberg et al., 2011). A recent careful comparison of 3 and 7 T fMRI protocols has revealed increased STN activation in failed stop trials as compared with successful stop and go trials (de Hollander et al., 2017). These studies were performed using a robust stop-signal paradigm (Logan et al., 1984; Aron and Poldrack, 2006). Interestingly, not only the STN, but also the substantia nigra, the red nucleus, and the globus pallidus (internal and external segment) showed this activation pattern. The coactivation of these individual basal ganglia nuclei together with the partial voluming effects in fMRI call for caution in the interpretation of available fMRI studies on the STN, and further technical developments are needed to address these challenges.

THE MICROSCOPIC ANATOMY OF THE STN Despite the strong interest in the STN, the number of publications that show a detailed investigation of protein expression in the human STN is limited (Alkemade et al., 2015). Studies addressing the internal anatomy of the STN at a cyto- and immunocytochemical level only represent a minor fraction of the available body of literature on the human STN and predominantly consists of older

studies. An additional limited body of literature is available on the internal structure of the STN in nonhuman primates (Alkemade et al., 2015). Like any other brain structure studied in the human postmortem brain, studies of the human STN are limited by the low number of available tissue specimens. Putative confounding factors in studies on human postmortem brain specimens include sex, age, brain weight, agonal state, seasonal and circadian variation, lateralization, as well as tissue treatment, including postmortem delay, fixation duration, and storage time. Available literature almost exclusively consists of qualitative assessments of distribution patterns of individual proteins or mRNA transcripts. Differences in antibody characteristics, as well as staining protocols including staining amplification, preclude quantitative comparisons between immunoreactivity distribution patterns for different protein markers, even within the same specimens. Comparisons across studies reported literature are therefore usually limited to qualitative assessments.

Immunoreactivity in the human STN Calcium-binding proteins have been used to identify individual populations of interneurons in the basal ganglia including the STN (Parent et al., 1996). Calretinin (CALR) positive neurons were reported to be concentrated in the ventromedial part of the STN, whereas parvalbumin (PARV) positive neurons showed a more dense distribution in the dorsolateral part of the STN (Parent et al., 1996; Augood et al., 1999; Morel et al., 2002). Additional studies from our group investigating 10 STNs from nondemented controls showed slightly different results for PARV (Fig. 25.4). We replicated the labeling of PARV neurons in the ventromedial region of the STN and found that their staining was more abundant in the dorsal parts of the STN. However, highest intensities were observed in the medial parts of the dorsal extent of the STN, not the lateral areas (Alkemade et al., 2019). It remains unknown what causes this difference. In these studies we also found that CALR showed strongest neuronal labeling in the ventromedial tip of the human STN, thereby replicating earlier findings (Parent and Hazrati, 1995; Alkemade et al., 2019). Together these findings suggest regional differences in calcium signaling within the human STN. The dorsal part of the STN tip appears to be predominantly dependent on PARV, and to a lesser extent on CALR, which showed highest expression levels in the ventromedial part of the STN. Two studies are available on the expression of neurofilament (SMI32) staining in the human STN (Morel et al., 2002; Alkemade et al., 2019). Morel et al. (2002) described the distribution of SMI32 as similar to that of

IMAGING OF THE HUMAN SUBTHALAMIC NUCLEUS

409

Fig. 25.4. Example of immunoreactivity in a single specimen for serotonin transporter (SERT), calretinin (CALR), parvalbumin (PARV), tyrosine hydroxylase (TH), synaptophysin (SYN), transferrin (TF), glutamic acid decarboxylase (GAD65/67), neurofilament H (SMI32), ferritin (FERR), GABA receptor subunit A3 (GABRA3), vesicular glutamate transporter 1 (VGLUT1), and myelin basic protein (MBP). Comb, comb system; SN, substantia nigra; STN, subthalamic nucleus. Image was reproduced from Alkemade A, de Hollander G, Miletic S, Keuken Max C et al. (2019). The functional microscopic neuroanatomy of the human subthalamic nucleus. Brain Struct Funct 224: 3213–3227. doi: 10.1007/s00429-019-01960-3. Springer Berlin Heidelberg, with permission.

CALR with relatively weak staining in the neuropil and intense staining resembling Golgi stains in neurons distributed along the border of the STN. Most intense staining was observed in the ventromedial part of the STN.

These observations differed from our later quantitative analyses, which revealed a relatively low expression of SMI32 in the ventromedial parts of the STN (Alkemade et al., 2019). Staining clear labeling of the neuronal cell

410

A. ALKEMADE AND B.U. FORSTMANN

bodies with a gradual decrease in intensity in the ventromedial direction, where weaker neuronal labeling was observed. The effect was most pronounced in the anterior STN. In the caudal regions of the STN, staining was more uniform. Weak fiber staining was also observed. Long thin fibers were occasionally stained in the dorsolateral part of the STN. Although less pronounced, a similar gradual decrease in fiber staining was observed in the ventromedial direction. It is also possible that differences in SMI-32 expression are associated with different axonal lengths and projection sites, or that the discrepancy between Alkemade et al. (2019) and the study by Morel et al. (2002) is the result of differences in the sampling procedure. The quantitative results we presented in our studies were obtained from a systematic 300 mm sampling interval, whereas the studies by Morel et al. (2002) followed a more descriptive approach. For synaptophysin punctate, staining was scattered throughout the nucleus. Staining density of the presynaptic terminals or boutons appeared higher in the ventromedial and dorsolateral parts of the nucleus indicating intense communication with cells in this area. Interestingly, punctate staining extended beyond the dorsolateral tip of the nucleus in the shape of a cap. At the caudal level, the staining was more homogeneous. Furthermore, neuronal somata surrounded by puncta were detected, scattered throughout the nucleus (Alkemade et al., 2019). There are several candidate neurotransmitters for the dense input in the ventromedial part of the STN. Markers for proteins involved in GABA-ergic, glutamatergic, serotonergic, and dopaminergic signaling are all expressed in the STN. The strong glutamatergic output of the STN, which projects to the main basal ganglia output structures, is reflected by the presence of glutamate, and the expression of metabotropic glutamate receptors in the monkey STN (Smith and Parent, 1988; Parent and Hazrati, 1995; Kuwajima et al., 2004) as well as punctate vesicular glutamate transporter 1 staining in the human STN (Alkemade et al., 2019). A minority of STN cells express glutamate decarboxylase, which is indicative of local glutamate to GABA conversion (Levesque and Parent, 2005; Alkemade et al., 2019). The expression of the GABA transporter (GAT)1, which removes GABA from the synaptic cleft, shows a more general distribution in the human STN (Smith and Parent, 1988; Augood et al., 1999; Hirunsatit et al., 2009). GABA-A and B receptors have been identified both in human and monkey STN, through the use of immunocytochemistry and in situ hybridization, respectively (Kultas-Ilinsky et al., 1998; Charara et al., 2000; Alkemade et al., 2019). In humans GABA receptor type A, alpha 3 subunit showed a higher staining intensity in the ventromedial part of the nucleus

(Alkemade et al., 2019). Additionally, a minority of the neurons displayed Glutamate decarboxylase (GAD65/67) immunoreactivity, again with a higher staining intensity in the ventromedial part of the nucleus, indicating that GABA-ergic signaling was not homogeneously distributed throughout the STN (Alkemade et al., 2019). The expression of type 1 and 2 dopamine receptors (D1R and D2R) has been described by two independent studies using radioactive in situ hybridization studies (Augood et al., 2000; Hurd et al., 2001). Both studies were able to detect D1R expression, but only Hurd et al. (2001), and not Augood et al. (2000) reported the detection of D2R mRNA. It is feasible that D2R mRNA expression levels did not reach the detection threshold in the studies by Augood et al. (2000) and these studies warrant replication using more sensitive techniques. Expression of tyrosine hydroxylase (TH), the rate-limiting enzyme in catecholamine synthesis, and thereby crucial for dopamine production was described originally by Hedreen (1999). In these studies, the authors showed that the majority of immunoreactive axons passed over and through the STN, with occasional branching in the STN, which was interpreted as indicative of STN innervation (Hedreen, 1999). Later studies by our group confirmed TH innervation, demonstrating both thick long and thin punctate fibers, with a clear gradient in the medial-lateral direction with highest fiber density in the ventromedial part of the STN (Alkemade et al., 2019). Serotonergic innervation of the monkey STN was reported by Parent et al. (2011), who described a clear topological organization of serotonin transporter (SERT) immunoreactivity within the STN, with a stronger innervation of its anterior half. The human STN shows comparable SERT distribution, with clear SERT fiber staining showing a graded density decreasing in the lateral direction, with highest densities observed in the ventromedial tip. Fiber densities in the dorsolateral two-thirds of the nucleus were low. This effect was most pronounced in anterior and central parts of the STN, and a general decrease in staining intensity was observed in the caudal extent of the nucleus (Alkemade et al., 2019). These findings are in line with serotonin (5HT) distribution patterns observed in monkeys (Mori et al., 1985). 5HT density was highest in the more medial and ventral parts of the STN, with rostral scattered fiber tracts, and fiber segments oriented toward the lateral margin of the STN (Mori et al., 1985). Older studies described the expression of endogenous opioid receptors using RNA blotting (Raynor et al., 1995), revealing transcripts in the STN, and preproenkephalin B was reported in the STN of monkeys, including an increased expression during levodopa treatment in

IMAGING OF THE HUMAN SUBTHALAMIC NUCLEUS experimental Parkinson’s disease (Aubert et al., 2007). As discussed previously, only a limited number of studies investigating the internal structure of the STN have been published. In addition to valuable descriptions of the distribution of immunoreactivity targeting proteins that are expressed in the STN, it is possible that more studies on the STN have been conducted, yielding negative results. These studies may not have made their way into the scientific literature due to a publication bias (Ioannidis, 2005). We therefore consider it also of importance to report on proteins targeted in immunocytochemical studies, and which did not reveal clear staining patterns in the STN (Alkemade et al., 2019). NPY and CRH did not show expression in the STN, and Orexin A only showed sporadic immunopositive fibers as did vasoactive intestinal polypeptide and somatostatin. Aromatase labeled a few STN neurons in a single subject, and ChAT staining revealed sporadic immunoreactive boutons (Alkemade et al., 2019).

Comparing microscopy studies to MRI Additionally, we would like to mention the histological and immunocytochemical staining of molecules that are of potential interest in the translation across modalities used for investigations of the STN. From MRI studies it is clear that iron is of great interest to study the internal structure of the STN. Perl staining was used for validation of MRI observations confirming that hyperintense areas on QSM and hypointense areas on T2*-weighted images corresponded iron rich brain structures (Dormont et al., 2004; Sun et al., 2015). In the STN immunoreactivity for transferrin, which controls the level of free iron available in the blood, was present in numerous blood vessels and oligodendrocytes. The oligodendrocytes displayed a rounded shape and were arranged in rows between fibers in the white matter regions. Signal was present in the rim of the cytoplasm. Neuronal staining was also observed, although less abundant as compared to oligodendrocytes. Neuronal labeling was cytoplasmic, and fiber staining was occasionally observed. Transferrin staining showed substantial background staining, which varied in intensity between subjects. This fits with transferrin labeling in the extracellular matrix. Transferrin also showed a lower staining intensity in the ventromedial part of the STN. Transferrin plays an important role in the delivery of iron to brain cells and is expressed in neurons as well as oligodendrocytes (Alkemade et al., 2019). Numerous ferritin positive oligodendrocytes were detected across the STN. Ferritin plays an important role in iron storage. The distribution of ferritin positive oligodendrocytes was quite uniform across the STN, as well as across subjects.

411

The question rises whether protein distribution patterns can be related to regional functional differences and distinct neuronal populations within the STN. In nonhuman primates, neurons located in the dorsolateral part of the STN are connected to the globus pallidus externa, whereas neurons connected to the globus pallidus interna, substantia nigra pars reticulate, and caudate nucleus are largely confined to the ventromedial parts of the STN (Nauta and Cole, 1978; Smith et al., 1990). In addition, neurons projecting to the ventral globus pallidus are located in the medial STN (Nauta and Cole, 1978; Smith et al., 1990). The ventromedial part of the human STN contains smaller neurons and shows a higher cell density as compared to more lateral regions (Fussenich, 1967). However, despite consistent distribution patterns across specimens, we observed considerable overlap in the staining patterns of markers, as well as considerable differences. Some markers extended beyond the borders of the STN, which appears to be at odds with older reports that the STN is a closed nucleus (Rafols and Fox, 1976). It is difficult to speculate on the importance of the observed topographical variations in the mechanism underlying deep brain stimulation and their relation to functional subdivisions. According to the literature (Greenhouse et al., 2013), STN electrodes inserted to treat a number of motor and other disorders are aimed at the dorsolateral part of the STN. None of the tested markers was confined to the dorsolateral STN. Overlapping protein distribution patterns are in line with those of Haynes and Haber (2013), who showed significant overlap in projection patterns within the primate STN and the description of topographically arranged transition zones within the STN (Lambert et al., 2015).

COMPARING AND INTEGRATING IN VIVO MRI AND POSTMORTEM STUDIES High-quality neuroanatomical research in humans is contingent on a constructive dialogue between those engaged in postmortem research defining underlying architectural properties and those attempting to model and capture these features using in vivo MRI (Lambert et al., 2015). To facilitate the dialogue between researchers performing postmortem studies and those involved in in vivo imaging of the STN, it is crucial to understand to what extent the applied techniques are comparable and results can be translated across modalities (Forstmann et al., 2017b). Size measurements reported based on postmortem investigations using either MRI or microscopy approaches provide an estimation of the average STN size of 129.7 mm3, which is substantially larger than that

412

A. ALKEMADE AND B.U. FORSTMANN

observed using in vivo MRI techniques. In vivo MRI thus appears to provide an underestimation of the size of the STN. Postmortem estimations are based on a very low number of observations, and it is unclear whether this could have biased the results. Additionally, the majority of histological approaches do not provide a 3D account of the structure, which hampers volume estimations. At the same time, estimations from MRI images are dependent on the used MRI parameters and the calculated contrast. Given the inherent limitations associated with the individual research approaches, the separate research fields should be considered of complimentary value, and results from these individual fields would be expected to converge. Together the findings should contribute to an improved understanding of the structure and function of small human brain nuclei, including the STN. Further integration of the research approaches is expected to contribute to the resolving of potential discrepancies, and integration can aid in the cross validation of the individual imaging techniques. Furthermore the integration of research fields can provide a crucial contribution to resolve ongoing discussions on the structure and function of the human STN, which can be attributed, at least to some extent to the difficulties associated with the translation of results across research fields (Lambert et al., 2012; Alkemade and Forstmann, 2014). Such comparisons across modalities are gaining interest within the scientific community and often used for qualitative comparison in smaller tissue blocks, in some cases in combination with histological approaches (e.g., B€ urgel et al., 1999; Castellanos et al., 2008; Makris et al., 2013; Adler et al., 2014; Annese et al., 2014; Augustinack et al., 2014; Plantinga et al., 2016). Practical limitations of postmortem MRI investigations of the human brain include the impracticalities associated with the scanning of unfixed human brain tissue. Such experiments are dependent on the availability of fresh donor material and therefore difficult to plan. Scan duration is limited as a result of the tissue degradation after the demise of the donor. Therefore, often formalin fixed tissues are used for detailed MRI investigations, solving the problem of tissue degradation, but at the same time introducing other challenges. Formalin fixation affects the shape and MR characteristics of the tissue (Chu et al., 2005; Schmierer et al., 2008; van Duijn et al., 2011; St€uber et al., 2014). Additionally, often postmortem studies involve small tissue samples, which are then imaged using ultra-high-resolution MRI. Through adjustment of the MR characteristics, it is possible to visualize the STN in such tissue specimens with excellent detail (Massey et al., 2012). Remaining challenges include coregistration between specimens, and with MRI standard space, given the limited number

of landmarks present in smaller samples for registration purposes. We have created a multistage approach, which allowed to coregister detailed anatomical scans of the STN to MNI space (Weiss et al., 2015), although registration of whole brain specimens would be preferred in view of the larger amount of shared information available for registration. A logical, but challenging next step is to reconstruct whole human brains in 3D, for which impressive proof of concept was provided by Amunts et al. (2013) through the creation of BigBrain. BigBrain is a high-resolution 3D reconstruction of a whole human brain of a 64-year-old male, which was sliced in 20 mm sections, which were histologically processed and digitally coregistered (Amunts et al., 2013). The creation of BigBrain is exciting, showing that a whole human brain processed for microscopy purposes can be reconstructed. Unfortunately, the MRI data collected for BigBrain was limited in quality and therefore does not allow reliable identification of the STN on the MRI images. Exciting expected future developments that can be anticipated include the combination of state-of-the-art postmortem multimodal MRI imaging using ultra-high-field scanners, with subsequent coregistration of the data with 3D reconstructions of histological data using comparable techniques as used for the creation of BigBrain (Amunts et al., 2013). Such approaches will be invaluable for the cross validation of imaging techniques and for future studies that will contribute to the reconciliation of differences across research fields investigating the internal structure and function of the human STN.

CONCLUSION The continuously growing interest in the structure and function of the STN, together with exciting technical advances in the field of MRI, have substantially increased our understanding of the structure and function of STN. The use of submillimeter isotropic voxels and iron sensitive contrasts are key to obtain reliable images in vivo, although functional MRI data should still be interpreted with caution in view of partial voluming effects, as well as interindividual variation. Ideally, postmortem studies would provide complementary and converging information providing the spatial resolution required to resolve the internal structure of the STN. However, discrepancies between findings from these research fields continue to exist. To understand the discrepancies between research fields and to potentially mitigate the shortcomings of the individual techniques, the integration of MRI and histological and immunocytochemical approaches is of great importance.

IMAGING OF THE HUMAN SUBTHALAMIC NUCLEUS

REFERENCES Adler DH et al. (2014). Histology-derived volumetric annotation of the human hippocampal subfields in postmortem MRI. NeuroImage. Academic Press. 84: 505–523. https:// doi.org/10.1016/J.NEUROIMAGE.2013.08.067. Alexander GE, Crutcher MD (1990). Functional architecture of basal ganglia circuits: neural substrates of parallel processing. Trends Neurosci 13: 266–271. Available at http://www.ncbi.nlm.nih.gov/pubmed/1695401. Alkemade A, Forstmann BU (2014). Do we need to revise the tripartite subdivision hypothesis of the human subthalamic nucleus (STN)? NeuroImage 95: 326–329, Elsevier Inc. https://doi.org/10.1016/j.neuroimage.2014. 03.010. Alkemade A, Schnitzler A, Forstmann BU (2015). Topographic organization of the human and non-human primate subthalamic nucleus. Brain Struct Funct 220: 3075–3086, Springer Berlin Heidelberg. https://doi.org/10.1007/s00429015-1047-2. Alkemade A et al. (2017). Comparison of T2*-weighted and QSM contrasts in Parkinson’s disease to visualize the STN with MRI. PLoS One. Edited by Q. Jiang. Public library of science. 12: e0176130. https://doi.org/10.1371/ journal.pone.0176130. Alkemade A, de Hollander G, Miletic S et al. (2019). The functional microscopic neuroanatomy of the human subthalamic nucleus. Brain Struct Funct 224: 3213–3227. https:// doi.org/10.1007/s00429-019-01960-3. Amunts K et al. (2013). BigBrain: an ultrahigh-resolution 3D human brain model. Science 2013/06/22, 340: 1472–1475. https://doi.org/10.1126/science.1235381. Annese J et al. (2014). Postmortem examination of patient H.M.’s brain based on histological sectioning and digital 3D reconstruction. Nat Commun 5: 3122, Nature Publishing Group. https://doi.org/10.1038/ncomms4122. Aron AR (2011). From reactive to proactive and selective control: developing a richer model for stopping inappropriate responses. Biol Psychiatry 69: e55–e68. https://doi.org/ 10.1016/j.biopsych.2010.07.024. Aron AR, Poldrack RA (2006). Cortical and subcortical contributions to stop signal response inhibition: role of the subthalamic nucleus. J Neurosci 26: 2424–2433. https://doi. org/10.1523/JNEUROSCI.4682-05.2006. Aron AR et al. (2007). Triangulating a cognitive control network using diffusion-weighted magnetic resonance imaging (MRI) and functional MRI. J Neurosci 27: 3743–3752. https://doi.org/10.1523/JNEUROSCI.0519-07.2007. Aubert I et al. (2007). Enhanced preproenkephalin-Bderived opioid transmission in striatum and subthalamic nucleus converges upon globus pallidus internalis in L-3,4-dihydroxyphenylalanine-induced dyskinesia. Biol Psychiatry 61: 836–844. https://doi.org/10.1016/j.biopsych. 2006.06.038. Augood SJ et al. (1999). Localization of calcium-binding proteins and GABA transporter (GAT-1) messenger RNA in the human subthalamic nucleus. Neuroscience 88: 521–534. Available at http://www.ncbi.nlm.nih.gov/ pubmed/10197772.

413

Augood SJ et al. (2000). Localization of dopaminergic markers in the human subthalamic nucleus. J Comp Neurol 421: 247–255. Available at http://www.ncbi.nlm. nih.gov/pubmed/10813785. Augustinack JC et al. (2014). MRI parcellation of ex vivo medial temporal lobe. NeuroImage Academic Press. 93: 252–259. https://doi.org/10.1016/J.NEUROIMAGE.2013. 05.053. Bergman H, Wichmann T, DeLong MR (1990). Reversal of experimental parkinsonism by lesions of the subthalamic nucleus. Science 249: 1436–1438, American Association for the Advancement of Science. https://doi.org/10.1126/ science.2402638. Birkl C et al. (2016). Effects of formalin fixation and temperature on MR relaxation times in the human brain. NMR Biomed 29: 458–465. https://doi.org/10.1002/nbm.3477. Bot M, Verhagen O, Caan M et al. (2019). Defining the dorsal STN border using 7.0-T MRI: a comparison to microelectrode recordings and lower field strength MRI. Stereotact Funct Neurosurg 97: 153–159. https://doi.org/10.1159/0005 00109. Brunenberg EJL et al. (2011). Magnetic resonance imaging techniques for visualization of the subthalamic nucleus. J Neurosurg 115: 971–984. https://doi.org/10.3171/2011. 6.JNS101571. B€ urgel U et al. (1999). Mapping of histologically identified long fiber tracts in human cerebral hemispheres to the MRI volume of a reference brain: position and spatial variability of the optic radiation. NeuroImage 10: 489–499, Academic Press. https://doi.org/10.1006/NIMG. 1999.0497. Castellanos FX et al. (2008). Cingulate-precuneus interactions: a new locus of dysfunction in adult attention-deficit/ hyperactivity disorder. Biol Psychiatry 63: 332–337. https://doi.org/10.1016/j.biopsych.2007.06.025. Charara A et al. (2000). Pre- and postsynaptic localization of GABA(B) receptors in the basal ganglia in monkeys. Neuroscience 95: 127–140. Available at http://www.ncbi. nlm.nih.gov/pubmed/10619469. Chu W-S et al. (2005). Ultrasound-accelerated formalin fixation of tissue improves morphology, antigen and mRNA preservation. Mod Pathol 18: 850–863, Nature Publishing Group. https://doi.org/10.1038/modpathol.3800354. de Hollander G et al. (2014). A gradual increase of iron toward the medial-inferior tip of the subthalamic nucleus. Hum Brain Mapp 35: 4440–4449. https://doi.org/10.1002/hbm. 22485. de Hollander G, Keuken MC, Forstmann BU (2015). The subcortical cocktail problem; mixed signals from the subthalamic nucleus and substantia nigra. PLoS One 10: e0120572. https://doi.org/10.1371/journal.pone.0120572. de Hollander G, Keuken MC, van der Zwaag W et al. (2017). Comparing functional MRI protocols for small, iron-rich basal ganglia nuclei such as the subthalamic nucleus at 7 T and 3 T. Hum Brain Mapp 38: 3226–3248. https://doi.org/ 10.1002/hbm.23586. Deistung A et al. (2013). Toward in vivo histology: a comparison of quantitative susceptibility mapping (QSM) with magnitude-, phase-, and R2*-imaging at ultra-high magnetic

414

A. ALKEMADE AND B.U. FORSTMANN

field strength. Neuroimage 2012/10/06, 65: 299–314. https:// doi.org/10.1016/j.neuroimage.2012.09.055. Dice LR, Dice (1945). Measurements of amount of ecologic association between species. Ecology 26: 297–302, Ecological Society of America. https://doi.org/10.2307/1932409. Dormont D et al. (2004). Is the subthalamic nucleus hypointense on T2-weighted images? A correlation study using MR imaging and stereotactic atlas data. AJNR Am J Neuroradiol 25: 1516–1523. 2004/10/27. Available at http://www.ncbi.nlm.nih.gov/pubmed/15502130. Forstmann BU et al. (2010). Cortico-striatal connections predict control over speed and accuracy in perceptual decision making. Proc Natl Acad Sci U S A 107: 15916–15920. https://doi.org/10.1073/pnas.1004932107. Forstmann BU, Isaacs BR, Temel Y (2017a). Ultra high field MRI-guided deep brain stimulation. Trends Biotechnol 35: 904–907, Elsevier Ltd. https://doi.org/10.1016/j.tibtech. 2017.06.010. Forstmann BU et al. (2017b). Towards a mechanistic understanding of the human subcortex. Nat Rev Neurosci 18: 57–65, Nature Publishing Group. https://doi.org/10.1038/ nrn.2016.163. Frank MJ (2006). Hold your horses: a dynamic computational role for the subthalamic nucleus in decision making. Neural Netw 19: 1120–1136. https://doi.org/10.1016/j. neunet.2006.03.006. Fussenich M (1967). Vergleichend anatomische studien uber den nucleus subthalamicus (corpus Luys) bei primaten, Ph.D. thesis, University of Freiburg. Greenhouse I et al. (2013). Stimulation of contacts in ventral but not dorsal subthalamic nucleus normalizes response switching in Parkinson’s disease. Neuropsychologia 51: 1302–1309. https://doi.org/10.1016/j.neuropsychologia.2013. 03.008. Grover VPB et al. (2015). Magnetic resonance imaging: principles and techniques: lessons for clinicians. J Clin Exp Hepatol 5: 246–255, Elsevier. https://doi.org/10.1016/ j.jceh.2015.08.001. Haber SN (2003). The primate basal ganglia: parallel and integrative networks. J Chem Neuroanat 26: 317–330. Available at: http://www.ncbi.nlm.nih.gov/pubmed/14729134. Hardman CD et al. (2002). Comparison of the basal ganglia in rats, marmosets, macaques, baboons, and humans: volume and neuronal number for the output, internal relay, and striatal modulating nuclei. J Comp Neurol Wiley Subscription Services, Inc., A Wiley Company. 445: 238–255. https:// doi.org/10.1002/cne.10165. Haynes WIA, Haber SN (2013). The organization of prefrontal-subthalamic inputs in primates provides an anatomical substrate for both functional specificity and integration: implications for basal ganglia models and deep brain stimulation. J Neurosci Society for Neuroscience. 33: 4804–4814. https://doi.org/10.1523/JNEUROSCI.467412.2013. Hedreen JC (1999). Tyrosine hydroxylase-immunoreactive elements in the human globus pallidus and subthalamic nucleus. J Comp Neurol 409: 400–410. Available at http://www.ncbi.nlm.nih.gov/pubmed/10379826.

Hirunsatit R et al. (2009). Twenty-one-base-pair insertion polymorphism creates an enhancer element and potentiates SLC6A1 GABA transporter promoter activity. Pharmacogenet Genomics 19: 53–65. https://doi.org/ 10.1097/FPC.0b013e328318b21a. Hurd YL, Suzuki M, Sedvall GC (2001). D1 and D2 dopamine receptor mRNA expression in whole hemisphere sections of the human brain. J Chem Neuroanat 22: 127–137. Available at http://www.ncbi.nlm.nih.gov/pubmed/ 11470560. Ioannidis JPA (2005). Why most published research findings are false. PLoS Med 2: e124, Public Library of Science. https://doi.org/10.1371/journal.pmed.0020124. Keuken MC, Forstmann BU (2015). A probabilistic atlas of the basal ganglia using 7 T MRI. Data Brief 4: 577–582. https://doi.org/10.1016/j.dib.2015.07.028. Keuken MC et al. (2013). Ultra-high 7T MRI of structural agerelated changes of the subthalamic nucleus. J Neurosci 33: 4896–4900. https://doi.org/10.1523/JNEUROSCI.324112.2013. Keuken MC, Sch€afer A, Forstmann BU (2016). Can we rely on susceptibility-weighted imaging (SWI) for subthalamic nucleus identification in deep brain stimulation surgery? Neurosurgery 79: e945–e946. https://doi.org/10.1227/NEU. 0000000000001395. Keuken MC, Isaacs BR, Trampel R et al. (2018). Visualizing the human subcortex using ultra-high field magnetic resonance imaging. Brain Topogr 31: 513–545. https://doi.org/ 10.1007/s10548-018-0638-7. Kraff O et al. (2015). MRI at 7 tesla and above: demonstrated and potential capabilities. J Magn Reson Imaging 41: 13–33. https://doi.org/10.1002/jmri.24573. Kultas-Ilinsky K, Leontiev V, Whiting PJ (1998). Expression of 10 GABA(A) receptor subunit messenger RNAs in the motor-related thalamic nuclei and basal ganglia of Macaca mulatta studied with in situ hybridization histochemistry. Neuroscience 85: 179–204. Available at http:// www.ncbi.nlm.nih.gov/pubmed/9607711. Kuwajima M et al. (2004). Subcellular and subsynaptic localization of group I metabotropic glutamate receptors in the monkey subthalamic nucleus. J Comp Neurol 474: 589–602. https://doi.org/10.1002/cne.20158. Ladd ME et al. (2018). Pros and cons of ultra-high-field MRI/MRS for human application. Prog Nucl Magn Reson Spectrosc 109: 1–50, The Authors. https://doi.org/10.1016/ j.pnmrs.2018.06.001. Lambert C et al. (2012). Confirmation of functional zones within the human subthalamic nucleus: patterns of connectivity and sub-parcellation using diffusion weighted imaging. Neuroimage 60: 83–94. https://doi.org/10.1016/ j.neuroimage.2011.11.082. Lambert C et al. (2015). Do we need to revise the tripartite subdivision hypothesis of the human subthalamic nucleus (STN)? Response to Alkemade and Forstmann. Neuroimage, 2015/01/27. 110: 1–2. https://doi.org/10.1016/j.neuroimage. 2015.01.038. Lange H et al. (1976). Morphometric studies of the neuropathological changes in choreatic diseases. J Neurol

IMAGING OF THE HUMAN SUBTHALAMIC NUCLEUS Sci 28: 401–425. https://doi.org/10.1016/0022-510X(76) 90114-3. Levesque JC, Parent A (2005). GABAergic interneurons in human subthalamic nucleus. Mov Disord 20: 574–584. https://doi.org/10.1002/mds.20374. Logan GD, Cowan WB, Davis KA (1984). On the ability to inhibit simple and choice reaction time responses: a model and a method. J Exp Psychol Hum Percept Perform 10: 276–291. https://doi.org/10.1037/0096-1523.10.2.276. Lozano AM et al. (2019). Deep brain stimulation: current challenges and future directions. Nat Rev Neurol 15: 148–160, Nature Publishing Group. https://doi.org/10.1038/s41582018-0128-2. Makris N et al. (2013). Volumetric parcellation methodology of the human hypothalamus in neuroimaging: normative data and sex differences. Neuroimage, 2012/12/19. 69: 1–10. https://doi.org/10.1016/j.neuroimage.2012.12.008. Massey LA et al. (2012). High resolution MR anatomy of the subthalamic nucleus: imaging at 9.4 T with histological validation. Neuroimage, 2011/11/01. 59: 2035–2044. https://doi.org/10.1016/j.neuroimage.2011.10.016. Middleton FA, Strick PL (2000). Basal ganglia and cerebellar loops: motor and cognitive circuits. Brain Res Brain Res Rev 31: 236–250. Available at http://www.ncbi.nlm.nih. gov/pubmed/10719151. Morel A et al. (2002). Neurochemical organization of the human basal ganglia: anatomofunctional territories defined by the distributions of calcium-binding proteins and SMI32. J Comp Neurol 443: 86–103, John Wiley & Sons, Ltd. https://doi.org/10.1002/cne.10096. Mori S et al. (1985). Immunohistochemical demonstration of serotonin nerve fibers in the subthalamic nucleus of the rat, cat and monkey. Neurosci Lett 62: 305–309. Available at http://www.ncbi.nlm.nih.gov/pubmed/2419795. Mulder MJ et al. (2019). Size and shape matter: the impact of voxel geometry on the identification of small nuclei. PLoS One 14: e0215382. https://doi.org/10.1371/journal.pone. 0215382. Edited by N. Bergsland. Public Library of Science. Nauta HJ, Cole M (1978). Efferent projections of the subthalamic nucleus: an autoradiographic study in monkey and cat. J Comp Neurol 180: 1–16. https://doi.org/10.1002/ cne.901800102. Nowinski WL et al. (2005). Statistical analysis of 168 bilateral subthalamic nucleus implantations by means of the probabilistic functional atlas. Neurosurgery 57: 319–330. https:// doi.org/10.1227/01.NEU.0000180960.75347.11. Ogawa S et al. (1990). Brain magnetic resonance imaging with contrast dependent on blood oxygenation. Proc Natl Acad Sci U S A 87: 9868–9872, National Academy of Sciences. https://doi.org/10.1073/pnas.87.24.9868. Parent A, Hazrati LN (1995). Functional anatomy of the basal ganglia. I. The cortico-basal ganglia-thalamo-cortical loop. Brain Res Brain Res Rev 20: 91–127. Available at http:// www.ncbi.nlm.nih.gov/pubmed/7711769. Parent A et al. (1996). Calcium-binding proteins in primate basal ganglia. Neurosci Res 25: 309–334. Available at http://www.ncbi.nlm.nih.gov/pubmed/8866512.

415

Parent M et al. (2011). Serotonin innervation of basal ganglia in monkeys and humans. J Chem Neuroanat 41: 256–265. https://doi.org/10.1016/j.jchemneu.2011.04.005. Pierpaoli C (2010). Quantitative brain MRI. Top Magn Reson Imaging 21: 63, NIH Public Access. https://doi.org/ 10.1097/RMR.0b013e31821e56f8. Plantinga BR et al. (2016). Ultra-high field MRI post mortem structural connectivity of the human subthalamic nucleus, substantia nigra, and globus pallidus. Front Neuroanat Frontiers Media SA, 10: 66. https://doi.org/10.3389/fnana. 2016.00066. Rafols JA, Fox CA (1976). The neurons in the primate subthalamic nucleus: a Golgi and electron microscopic study. J Comp Neurol 168: 75–111. https://doi.org/10.1002/ cne.901680105. Raynor K, Kong H, Mestek A et al. (1995). Characterization of the cloned human mu opioid receptor. J Pharmacol Exp Ther 272: 423–428. Available at http://www.ncbi.nlm.nih. gov/pubmed/7815359. Redgrave P, Prescott TJ, Gurney K (1999). The basal ganglia: a vertebrate solution to the selection problem? Neuroscience 89: 1009–1023. Available at http://www. ncbi.nlm.nih.gov/pubmed/10362291. Schafer A et al. (2012). Direct visualization of the subthalamic nucleus and its iron distribution using high-resolution susceptibility mapping. Hum Brain Mapp 33: 2831–2842. https://doi.org/10.1002/hbm.21404. Schmierer K et al. (2008). Quantitative magnetic resonance of postmortem multiple sclerosis brain before and after fixation. Magn Reson Med 59: 268–277, Wiley Subscription Services, Inc., A Wiley Company, 59: 268–277. https:// doi.org/10.1002/mrm.21487. Smith Y, Parent A (1988). Neurons of the subthalamic nucleus in primates display glutamate but not GABA immunoreactivity. Brain Res 453: 353–356. Available at: http://www. ncbi.nlm.nih.gov/pubmed/2900056. Smith Y, Hazrati LN, Parent A (1990). Efferent projections of the subthalamic nucleus in the squirrel monkey as studied by the PHA-L anterograde tracing method. J Comp Neurol 294: 306–323. https://doi.org/10.1002/cne.902940213. St€ uber C et al. (2014). Myelin and iron concentration in the human brain: a quantitative study of MRI contrast. Neuroimage 93: 95–106. https://doi.org/10.1016/j.neuroimage.2014. 02.026. Sun H et al. (2015). Validation of quantitative susceptibility mapping with Perls’ iron staining for subcortical gray matter. NeuroImage 105: 486–492, Academic Press. https:// doi.org/10.1016/J.NEUROIMAGE.2014.11.010. Trattnig S et al. (2018). Key clinical benefits of neuroimaging at 7 T. Neuroimage 168: 477–489. https://doi.org/10.1016/ j.neuroimage.2016.11.031. van Duijn S et al. (2011). MRI artifacts in human brain tissue after prolonged formalin storage. Magn Reson Med 65: 1750–1758. https://doi.org/10.1002/mrm.22758. von Bonin G, Shariff GA (1951). Exteapyramidal nuclei among mammals. A quantitative study. J Comp Neurol 94: 427–438, John Wiley & Sons, Ltd. https://doi.org/ 10.1002/cne.900940306.

416

A. ALKEMADE AND B.U. FORSTMANN

Weiss M et al. (2015). Spatial normalization of ultrahigh resolution 7 T magnetic resonance imaging data of the postmortem human subthalamic nucleus: a multistage approach. Brain Struct Funct 220: 1695–1703, Springer Berlin Heidelberg. https://doi.org/10.1007/s00429-014-0754-4. Yelnik J, Percheron G (1979). Subthalamic neurons in primates: a quantitative and comparative analysis.

Neuroscience 4: 1717–1743. https://doi.org/10.1016/03064522(79)90030-7. Zwirner J et al. (2017). Subthalamic nucleus volumes are highly consistent but decrease age-dependently-a combined magnetic resonance imaging and stereology approach in humans. Hum Brain Mapp John Wiley & Sons, Ltd 38: 909–922. https://doi.org/10.1002/hbm.23427.

Handbook of Clinical Neurology, Vol. 180 (3rd series) The Human Hypothalamus: Middle and Posterior Region D.F. Swaab, F. Kreier, P.J. Lucassen, A. Salehi, and R.M. Buijs, Editors https://doi.org/10.1016/B978-0-12-820107-7.00026-4 Copyright © 2021 Elsevier B.V. All rights reserved

Chapter 26

Neuropsychiatric effects of subthalamic deep brain stimulation PHILIP E. MOSLEY1* AND HARITH AKRAM2 1

Systems Neuroscience Laboratory, QIMR Berghofer Medical Research Institute, Brisbane, QLD, Australia

2

Unit of Functional Neurosurgery, UCL Queen Square Institute of Neurology & The National Hospital for Neurology and Neurosurgery, London, United Kingdom

Abstract The subthalamic nucleus (STN) is a subcortical, glutamatergic, excitatory, relay nucleus that increases the inhibitory drive of the basal ganglia and suppresses action. It is of central relevance to the neuropsychological construct of inhibition, as well as the pathophysiology of Parkinson’s disease (PD). Deep brain stimulation (DBS) of the STN (STN-DBS) is an established surgical treatment for PD that can be complicated by adverse neuropsychiatric side effects, most commonly characterized by impulsivity and mood elevation, although depression, anxiety, apathy, and cognitive changes have also been reported. Notwithstanding these adverse neuropsychiatric effects in PD, STN-DBS may also have a role in the treatment of refractory psychiatric disorders, as more is understood about the physiology of this nucleus and techniques in neuromodulation are refined. In this chapter, we link neuropsychiatric symptoms after STN-DBS for PD to the biological effects of electrode implantation, neurostimulation, and adjustments to dopaminergic medication, in the setting of neurodegeneration affecting cortico-striatal connectivity. We then provide an overview of clinical trials that have employed STN-DBS to treat obsessive–compulsive disorder and discuss future directions for subthalamic neuromodulation in psychiatry.

INTRODUCTION The subthalamic nucleus (STN) is a subcortical, glutamatergic, excitatory, relay nucleus that increases the inhibitory drive of the basal ganglia and suppresses action. It is of central relevance to the neuropsychological construct of inhibition, as well as the pathophysiology of Parkinson’s disease (PD). Deep brain stimulation (DBS) of the STN (STN-DBS) is an established surgical treatment for PD that can be complicated by adverse neuropsychiatric side effects, most commonly characterized by impulsivity and mood elevation, although depression,

anxiety, apathy, and cognitive changes have also been reported (Volkmann et al., 2010). Notwithstanding these adverse neuropsychiatric effects in PD, STN-DBS may also have a role in the treatment of refractory psychiatric disorders, as more is understood about the physiology of this nucleus and techniques in neuromodulation are refined. In this chapter, we discuss the neuropsychiatric symptoms that may present following STN-DBS for PD. We explicate these in view of DBS lead implantation effects, neurostimulation, and adjustments to medications, all in the setting of brain network dysfunction caused by neurodegeneration. We then provide an overview of

*Correspondence to: Dr. Philip E. Mosley, M.A. (Oxon.), B.M.B.Ch. (Oxon.), Ph.D., F.R.A.N.Z.C.P., Neuropsychiatrist and Clinical Research Fellow, Systems Neuroscience Laboratory, QIMR Berghofer Medical Research Institute, Brisbane, QLD, Australia. Tel: +61-7-3839-3688, Fax: +61-7-3839-3588, E-mail: [email protected]

418

P.E. MOSLEY AND H. AKRAM

clinical trials of STN-DBS for obsessive–compulsive disorder (OCD). Finally, we consider future directions for subthalamic neuromodulation in psychiatry.

The subthalamic nucleus The STN is a small, almond-shaped structure approximately 3
5
13 mm in dimension. It is located inferior to the zona incerta and superior to the substantia nigra in the diencephalon (Coenen et al., 2008). The STN is a nucleus of the basal ganglia: a group of subcortical structures underpinning movement, in addition to cognition and emotion. Nuclei in the basal ganglia receive their primary input from the cerebral cortex and send their output back to the prefrontal and motor cortices (and brainstem), via the thalamus, in corticosubcortical reentrant circuits. The output nuclei of the basal ganglia tonically inhibit their targets in the thalamus, with their activity modulated by two parallel pathways originating in the striatum: the direct and indirect pathways. The STN forms part of the indirect pathway, which passes from the striatum to the external pallidal segment (GPe) and then to the STN, before reaching the output nuclei. The glutamatergic projection from the STN to the output nuclei is the only excitatory connection of the basal ganglia. Therefore, STN activity results in a net increase in cortical inhibition and suppresses action. In addition to this pathway from the striatum, the STN is also a second input station of the basal ganglia. It receives direct cortical projections from the frontal lobe, forming a “hyperdirect” pathway through which the cerebral cortex can access the output nuclei directly. Activity in the hyperdirect pathway rapidly and powerfully increases inhibitory basal ganglia output (Nambu et al., 2000). Coordinated motor sequencing is hypothesized to recruit the hyperdirect pathway, which globally suppresses competing motor programs, allowing a targeted balance of the direct and indirect pathways in fine motor execution and termination (Nambu et al., 2002). The STN has an overlapping functional topography with its cortical connections, which has been delineated with histological and imaging techniques. A motor subregion in the superolateral aspect of the nucleus transitions into cognitive–associative and affective subregions in the inferiomedial plane (Lambert et al., 2012; Haynes and Haber, 2013; Accolla et al., 2014; Ewert et al., 2018). However, efferents from the STN to the basal ganglia output nuclei arborize widely on their target neurons, facilitating a widespread inhibitory effect on action initiation (Hazrati and Parent, 1992). Therefore, hyperdirect associative and affective prefrontal projections to the STN are an anatomical substrate through which top-down cognitive and emotional control can be exerted over all behavioral programs transiting through the basal ganglia. The overlapping boundaries between

functional territories of the STN also facilitate this integration (Yelnik, 2008). From a computational perspective, STN activity has been considered to implement a “delay” on cognitive–associative circuits in the basal ganglia, allowing more information to be gathered to guide the most appropriate behavioral policy, suppressing impulsive and potentially error-prone responding (Frank et al., 2007; Cavanagh et al., 2011). Functional and structural brain imaging supports the role of the STN as a key node in motor inhibition, in a “stopping network” involving the inferior frontal gyrus and the presupplementary motor area (Aron et al., 2007; Rae et al., 2015).

Parkinson’s disease PD is a progressive, incurable, neurodegenerative disorder characterized by motor symptoms: tremor, rigidity, and bradykinesia, as well as by pronounced non-motor symptoms and inflammatory changes, that vary considerably between patients (Doorn et al., 2012, 2014). The neuropathology of PD involves the intracellular aggregation of protein inclusions containing misfolded alphasynuclein (Lewy bodies). This is most pronounced in the substantia nigra pars compacta of the brain, leading to the degeneration of dopaminergic neurons in the basal ganglia. The loss of cells in the substantia nigra is well established before the first manifestation of motor symptoms and neurodegeneration continues after diagnosis in other regions of the brainstem and cerebral cortex. Therefore, although in its early stages PD may cause only mild functional disability, in later stages sufferers become profoundly dependent. The median time from onset to death from PD is approximately 9–15 years (de Lau et al., 2005; Forsaa et al., 2010). Dopamine replacement therapy provides substantial relief from the motor symptoms of PD. These drugs increase dopamine concentrations or directly stimulate dopamine receptors in the brain. They include levodopa, dopamine agonists, monoamine oxidase inhibitors, and catechol-O-methyltransferase inhibitors. Unfortunately, for many patients, the window of therapeutic benefit gradually narrows and individuals fluctuate between phases where motor control is good (“on”) and periods without alleviation of symptoms (“off”) in a dose-dependent fashion. Performance during “on” periods may also be limited by abnormal involuntary choreiform or dystonic movements (dyskinesias) that associate in a peak or biphasic pattern with levels of levodopa. After 5 years of treatment with dopamine replacement therapy, 40% of patients suffer from such complications (Ahlskog and Muenter, 2001).

Deep brain stimulation DBS involves the surgical implantation of brain leads that deliver a small electrical current through electrodes

NEUROPSYCHIATRIC EFFECTS OF SUBTHALAMIC DBS placed within deep brain nuclei. DBS may exert its function by modulating activity in distributed brain networks passing through the target structures (Horn et al., 2017). The STN arose as a target for DBS in PD based on the careful study of animal models, which implicated pathological STN activity as a key downstream consequence of dopaminergic denervation (Bergman et al., 1990, 1994; Aziz et al., 1991; Benabid et al., 1994). In PD, STN neurons display abnormal patterns of burst firing (Vila et al., 2000) and low-frequency synchronization of local field potentials with cortical regions (Brown et al., 2001; Eusebio et al., 2009). STN-DBS for PD is now thought to work by interrupting these pathological synchronous oscillations (Eusebio et al., 2011; Shimamoto et al., 2013). STN-DBS reduces tremor, rigidity, and bradykinesia, improves quality of life, and permits reduction or cessation of dopaminergic therapies (Krack et al., 2003; Williams et al., 2010; Schuepbach et al., 2013). The techniques used for DBS device implantation vary slightly between surgical centers. At the center in Brisbane (Australia), the STN is first visualized using a magnetic resonance imaging (MRI) brain scan, which allows the neurologist and neurosurgeon to identify the subcortical targets for neuromodulation. Although the onset of motor symptoms in PD is typically unilateral, almost all surgical candidates now have bilateral symptoms and thus one electrode is placed in each hemisphere of the brain. The neurosurgeon attaches a stereotactic frame to the skull of the patient and fuses a computed tomography (CT) brain scan (with the frame attached) to the existing MRI scan. This fused image is used to calculate the precise three-dimensional intracranial coordinates of the surgical target. Under general anesthesia, the neurosurgeon drills a small burr hole on each side of the patient’s skull and passes a recording electrode along a predetermined trajectory to the target structure. Accurate placement in the STN is confirmed using intraoperative microelectrode recording (MER) of local field potentials, and clinical evaluation once the patient has been roused from anesthesia. At this time, the neurologist can assess the effect of intraoperative stimulation on the patient’s motor symptoms and screen for any unwanted motor or sensory effects (such as facial pulling, gaze deviation, or paresthesia). Once accurate placement is verified, permanent DBS leads are implanted and their final position is verified using another CT brain scan. Other neurosurgical centers employ an MRI-guided and MRI-verified approach under general anesthesia and without relying on MER or on fusing different imaging modalities. Dedicated stereotactic MRI sequences are acquired with the frame on. The DBS leads are then implanted in a “sweet spot” within the STN to maximize

419

efficacy and minimize side effects. Once the leads are implanted, the MRI is repeated to confirm lead placement (Foltynie et al., 2011). In the rare instance (2%) when the lead is not in the desired target (within 1.5 mm), the lead can be repositioned intraoperatively. The implanted leads are routed subcutaneously and connected to an implantable pulse generator (IPG) through extension cables. The IPG is generally inserted in the pectoral or abdominal pocket. Each lead comprises between four to eight contacts, any number of which can be activated to deliver a square wave pulse of charge in the local neural elements. In the electrical circuit, the charge is delivered by means of at least one positive (anodal) and one negative (cathodal) terminal. Initially, the device is programmed with one or more contacts as the cathode and the pulse generator as the anode, a configuration known as “monopolar” (the polarity of this system can also be reversed in some experimental settings). This gives rise to an approximately spherical field of charge with a relatively wide distribution within the local tissue. A configuration with both anode and cathode on the same electrode is known as “bipolar” and delivers a more focused stimulation field. Higher amplitudes are required to produce an equivalent distribution of charge in the bipolar configuration. Some stimulating electrodes are manufactured with “segmented” contacts, which allows the clinician to activate only a portion of the ring of the selected contact, causing the electrical charge to be “directed” toward the chosen target structure (referred to as current steering). The amount of charge delivered is a function of the stimulation amplitude, frequency, pulse width, and the impedance of the neural substrate. The amplitude can be modulated postoperatively and noninvasively using a clinician programmer that communicates wirelessly with the pulse generator. Stimulation is titrated in gradual increments over the weeks to months subsequent to DBS device implantation as dopaminergic therapies are gradually withdrawn. Increases in stimulation are guided by residual motor symptoms and the emergence of side effects. Motor side effects may include dyskinesia (if stimulation titration is too rapid), worsening of dysarthria, blepharospasm, and gaze palsy. Inadequate response to treatment at an adequate amplitude, or the emergence of significant side effects, may prompt a switch to a bipolar stimulation configuration or selection of an alternative contact on the electrode. It may take 6–12 months to find the optimal stimulation parameters, with the slow accrual of motor benefits during this time. For DBS to have clinically significant effects, it is likely that a certain minimal volume of neural tissue must be modulated. Preliminary work has approximated this to a radius of 2.5 mm around the active contact (Maks et al., 2009; Madler and Coenen, 2012).

420

P.E. MOSLEY AND H. AKRAM

NEUROPSYCHIATRIC EFFECTS OF SUBTHALAMIC DEEP BRAIN STIMULATION IN PARKINSON’S DISEASE PD is a complex neurological disorder. Although diagnosed on the basis of motor symptoms, comorbid neuropsychiatric symptoms are frequently present and may be the principal source of disability and caregiver burden (Aarsland et al., 1999; Carter et al., 2008, 2012; Shin et al., 2012). These “nonmotor” symptoms may include depression, anxiety, apathy, impulse-control behaviors (ICBs), psychosis, and cognitive impairment (Weintraub and Burn, 2011). Among this comorbidity, ICBs such as pathological gambling, hypersexuality, binge eating, and compulsive shopping are paradigmatic of how neurodegeneration interacts with dopaminergic therapy to give rise to neuropsychiatric symptoms. ICBs in nonsurgical cohorts have been associated with differential patterns of neurodegeneration and dopamine receptor expression within prefrontal networks, mediating differential neural and behavioral responses to exogenous dopamine (van Eimeren et al., 2010; Voon et al., 2011; Antonelli et al., 2014; Stark et al., 2018; Mosley et al., 2019a). In an analogous manner, STN-DBS and the attendant postoperative alterations in dopaminergic therapy may disrupt a compensated but brittle homeostatic system regulating movement, mood, and cognition, already anatomically and physiologically disrupted prior to surgery. There may arise a postoperative dissociation between motor and nonmotor outcomes as stimulation parameters and medication are optimized to restore degenerated motor circuits, with a resultant imbalance in topographically distinct functional networks regulating mood, motivation, and impulse control. For most patients, STN-DBS is a safe therapy. The rate of serious psychiatric complications following DBS for PD is comparable to patients treated with dopaminergic therapies alone (Witt et al., 2008). However, understanding why a proportion of individuals experience neuropsychiatric symptoms attributable to DBS is a clinically significant question, for at present, there is no established method of identifying “at-risk” surgical candidates.

Mania, hypomania, and impulsivity The STN is a key node in the inhibitory control of complex, motivated behavior; thus, disruption of its physiological functioning (even in the pathological setting of PD) may induce behavioral disinhibition (Castrioto et al., 2014). A transient period of euphoria may be observed in the immediate aftermath of STN-DBS and is commonly attributed to a postimplantation stun effect. However, in up to 15% of patients with PD, chronic STN

stimulation may induce a sustained phase of clinically significant impulsivity and mood elevation (Romito et al., 2002; Daniele et al., 2003; Hershey et al., 2004, 2010; Voon et al., 2006; Appleby et al., 2007; Mallet et al., 2007; Welter et al., 2014). Although these symptoms can be ameliorated with stimulation reprogramming, they may nonetheless be associated with lasting social, financial, and legal harms (Mosley et al., 2019b) and caregiver burden (Mosley et al., 2018a). The most common features of this presentation are an elevated or irritable mood, agitation, distractibility, impulsivity, and an increase in goal-directed activity. Insomnia and grandiosity may also be present. Stimulation effects can be distinguished from postimplantation stun effects by their persistence and temporal association with DBS adjustment. Some cases manifest attenuated symptoms, making the diagnosis more difficult to capture and potentially accounting for the discrepancy in reported incidence. Close family members such as spouses are nonetheless likely to detect these changes and commonly complain that the patient is no longer “himself.” Impulsivity is frequently present and may be a common endophenotype in this cohort, defined informally as the tendency to act recklessly and prematurely, without foresight. The preoperative identification of surgical candidates susceptible to such complications has been elusive. ICBs display the greatest phenomenological overlap with these stimulation-induced neuropsychiatric symptoms, but their presence is not predictive and their absence is not protective: patients with pre-DBS ICBs may remit after STN-DBS following a reduction in dopaminergic medication (Lhommee et al., 2012; Eusebio et al., 2013), while those with no history of ICBs may develop such behaviors after STN-DBS (Lim et al., 2009; Moum et al., 2012). Previous reports have attributed the onset of mood changes to the amplitude of stimulation (Chopra et al., 2011) or the position of the active electrode contact within the STN (Raucher-Chene et al., 2008). Remediation of mania has been described through reducing the amplitude of stimulation (Chopra et al., 2011), using a different contact (Mandat et al., 2006; Raucher-Chene et al., 2008), delivering a more focused stimulation field by using the contact as the anode (Mandat et al., 2006) or the addition of mood-stabilizing medication (Herzog et al., 2003). Previously, we reported a case of mania with psychotic symptoms in a patient with PD without a personal or family history of psychiatric symptoms or ICBs (Mosley et al., 2018b). This case also described the phenomenon of a sustained mood change that was positively evaluated by the patient and his family, with emergence of clinically significant manic symptoms only after activation of an additional stimulating contact 9 months after

NEUROPSYCHIATRIC EFFECTS OF SUBTHALAMIC DBS device implantation. We employed a novel methodology for localizing and quantifying the field of stimulation in the STN, supporting the role of ventromedial (limbic and associative) subthalamic stimulation in the pathogenesis of postoperative mania. At a subsyndromal level, patients with PD may develop impulsive biases after STN-DBS without meeting the criteria for clinically significant impairment. These may include failures of motor inhibition (Hershey et al., 2004), action cancellation, (Obeso et al., 2013), as well as weakened prepotent verbal inhibition (Witt et al., 2004; Thobois et al., 2007). Furthermore, when faced with a difficult choice, patients with PD may speed rather than slow their decision-making after STN-DBS (Frank et al., 2007; Cavanagh et al., 2011), where taking more time would be an optimal response in order to make an accurate decision. STN-DBS also increases risk taking, reversing the tendency to risk aversion observed in participants with PD “off” medication and in a state of dopaminergic deficit (Irmen et al., 2019). These findings support the status of impulsivity as a unifying construct within a spectrum of neuropsychiatric effects attributable to STN-DBS, although factors responsible for the observed variability in degree and harm require elaboration. The distribution of stimulation within the STN may be a key factor moderating this variability. Stimulating electrodes are targeted to the dorsolateral sensorimotor subregion, with stimulation in this region associated with optimal motor outcomes (Wodarg et al., 2012; Akram et al., 2017). However, the small size of the STN means that current diffusion from a stimulating contact in the sensorimotor region could modulate subthalamic regions with greater connectivity to fronto-striatal networks implicated in mood, decision-making, and reward. The more ventral and medial the stimulating contact, the more likely these circuits are to be affected by DBS. In prior work, we systematically investigated whether the distribution of subthalamic stimulation moderated the evolution of postoperative neuropsychiatric symptoms in a sizeable cohort of patients with PD assessed longitudinally before and after STN-DBS (Mosley et al., 2018c). Tractographic parcellation of the STN into motor, associative, and limbic subregions furnished precise information on the position of each electrode contact relative to this topography. We simulated a volume of activated tissue for each hemisphere, for each participant, at each follow up. This enabled us to estimate the dispersion of charge within each STN subregion at a given time and allowed us to evaluate the contribution of stimulation parameters to emergent symptoms during titration. We found that the dispersion of charge within the associative subregion of the right STN was associated with both a neuropsychological index of disinhibition and clinically significant changes in behavior characterized by

421

impulsivity and mood elevation. The right-lateralization of our findings was intriguing, given previous work suggesting that executive control of inhibition is primarily a right-lateralized process (Aron et al., 2004; Possin et al., 2009; D’Alberto et al., 2017). Across the whole STN, we were also able to delineate distinct clusters of voxels associated with this inhibitory impairment, as well as the likelihood of clinically significant neuropsychiatric symptoms (Fig. 26.1). Machine-learning classification methods applied to our data were able to classify these “cases” (of clinically significant neuropsychiatric impairment) with an accuracy of 80%. In summary, our findings confirmed that stimulation diffusion within more ventromedial aspects of this nucleus does have an important role in mediating neuropsychiatric outcomes. However, a mechanistic association between modulation of frontostriatal networks and the emergence of post-DBS neuropsychiatric symptoms has not been established beyond the local effects of DBS. Moreover, stimulation diffusion into associative and limbic subregions may not be the only mediators of adverse neuropsychiatric side effects, which may also include modulation of white matter tracts traversing adjacent to the nucleus, such as the medial forebrain bundle (Coenen et al., 2009, 2012). In order to address this mechanistic question, we employed a battery of neuropsychiatric instruments and a naturalistic gambling task (a virtual casino, described in Paliwal et al. (2019)) in a further longitudinal study of patients with PD undertaking STN-DBS (Mosley et al., 2020). We found that impulsivity and gambling behavior covaried with the structural connectivity between the field of stimulation and fronto-striatal networks underpinning the constructs of reward evaluation and response inhibition. Notably, structural connectivity of these networks at baseline (prior to DBS) did not associate with postoperative behavior, suggesting a primary role for the locus and distribution of stimulation in explaining this variance, rather than a preexisting anatomical vulnerability. In particular, betting behavior in the virtual casino associated with the recruitment of bilateral tracts connecting the site of stimulation with the orbitofrontal cortex (OFC), a finding of interest given the role of this region in predicting outcomes after behavioral choices (Rudebeck and Murray, 2014) and contributing to prediction error signaling in ascending dopaminergic projections from the ventral tegmental area (Takahashi et al., 2011). This also lends further support to the theory that spread of electrical stimulation into ventromedial regions of the STN (regions more densely connected to the OFC) can quantitatively influence impulsivity. In a further analysis, fibers associated with betting behavior were isolated from a normative connectome and compared between participants who did and did not develop clinically significant neuropsychiatric symptoms

422

P.E. MOSLEY AND H. AKRAM

Fig. 26.1. STN clusters associated with high and low likelihood of developing clinically significant neuropsychiatric symptoms. In Mosley et al. (2018c), patients who developed clinically significant (with impairment or distress) neuropsychiatric symptoms comprising mood elevation, impulsivity, disinhibition, and reckless decision-making were defined as “cases.” Clusters of subthalamic voxels associated with these symptoms were identified using threshold-free cluster-enhancement (Smith and Nichols, 2009). Within each STN limbic ¼ yellow, associative ¼ blue, and motor ¼ maroon subregions. Green: a cluster of family wise error (FWE) corrected voxels significantly associated with a low likelihood of neuropsychiatric impairment can be identified in the dorsolateral aspect of the right STN. Red: a cluster of FWE-corrected voxels significantly associated with impairment is identified in the ventromedial aspect of both the right and left STN. These findings correspond with the known anatomy of this nucleus, with motor representations in the dorsolateral aspect of this nucleus and cognitive–associative circuits in the ventromedial region. (A) Axial view, (B) coronal view, and (C) oblique view. Findings are overlaid on the BigBrain atlas (Amunts et al., 2013).

attributable to neurostimulation (Fig. 26.2). This analysis identified a tract between the ventromedial STN and right OFC that did not extend into the diencephalon, conceivably representing direct cortical-STN connections that drive STN output and behavioral inhibition. This tract was associated with both dysregulated gambling plus pathological mood elevation and impulsivity. These findings converged with our previous work (outlined previously) demonstrating that ventromedial dispersion of the stimulation field within the right STN was more likely to be associated with disinhibition and the development of clinically significant hypomania and harmful impulsivity (Mosley et al., 2018c).

Depression and anxiety If postoperative disinhibition after STN-DBS for PD is related to stimulation of affective and associative networks connected to the STN, then underactivation of these same nonmotor circuits also produces characteristic neuropsychiatric sequelae. Substantial reduction of dopaminergic medication following STN-DBS for PD is a common practice as electrical stimulation in motor regions of the STN predominates. It may, however,

unmask symptoms such as depression and anxiety in up to 50% of patients (Thobois et al., 2010). For these individuals, dopaminergic medication has a psychotropic effect and withdrawal symptoms arise on discontinuation (Giovannoni et al., 2000), particularly in those treated with dopamine agonists (Rabinak and Nirenberg, 2010). The symptoms of this “dopamine agonist withdrawal syndrome” (DAWS) include anxiety, agitation, depression, irritability, insomnia, and suicidal ideation. These symptoms respond to dopamine agonist repletion. DAWS, therefore, may also be a driver of postoperative neuropsychiatric complications after STN-DBS (Nirenberg, 2010). As medication withdrawal is not a problem for all individuals, the disease-related denervation pattern may explain this hypodopaminergic vulnerability, with afflicted patients having greater mesocorticolimbic dopaminergic denervation than those without symptoms (Thobois et al., 2010). Direct stimulation-induced depressive symptoms are rare (Bejjani et al., 1999). At a cohort level, optimally sited stimulation appears to have a modest moodelevating effect (Wolz et al., 2012). However, the issue of suicide following STN-DBS for PD was highlighted by a retrospective case series that found it to be a

NEUROPSYCHIATRIC EFFECTS OF SUBTHALAMIC DBS

423

Fig. 26.2. Fiber tracts weighted most strongly in the association of connectivity with gambling behavior after STN-DBS. In Mosley et al. (2020), we employed a normative connectome derived from individuals with Parkinson’s disease to visualize white matter fibers connected to the stimulation field (methods described in more detail in Ewert et al. (2018) and Horn et al. (2017)). We selected fibers predictive of bet size and compared these between participants who developed clinically significant neuropsychiatric symptoms attributable to STN-DBS (case positive) and those that did not (case negative) participants. This facilitated identification of fibers associated with gambling behavior that were also associated with pathological behavior of a clinically relevant nature. In case negative participants, streamlines predictive of increased bet size passed from the diencephalon lateral to the STN (shown in orange in the lowest panel) onward to the right ventromedial prefrontal cortex (vmPFC) and orbitofrontal cortex (OFC). In the right hemisphere, a portion of these streamlines traversed the ventral tegmental area (shown in blue in the lowest panel). However, in case positive participants, right-hemispheric fibers predominantly involved the OFC rather than vmPFC. Moreover, these fibers were situated medial to the right STN and appeared to terminate/originate in the STN rather than passing into the midbrain. The middle panel in the lower row visualizes the relationship between connectivity of a fronto-striatal network underpinning reward evaluation and bet size, differentiated by caseness. Case negative participants demonstrated a scaling of connectivity with bet size, while case positive participants did not, which replicates prior work by Mosley et al. (2019a) and Haagensen et al. (2020).

predominant cause of mortality in the first postoperative year (Voon et al., 2008), although the quality of this data was diminished by its retrospective nature. The only randomized, controlled trial to examine this issue did not find an increased rate of suicide in the DBS group at 6 months postoperatively (Weintraub et al., 2013), although this study was underpowered to detect a significant difference in such a rare event. A history of depression, the dopamine dysregulation syndrome, or an ICB was associated with a risk of attempted suicide post-DBS in the aforementioned retrospective series (Voon et al., 2008). This may be a marker of both impulsivity and a liability to develop a DAWS-induced depressive state, as comorbidity of DAWS with ICBs is common (Lhommee et al., 2012).

Initiation and apathy Aside from the effects on neuropsychological constructs of impulsivity, STN-DBS has also been associated with deficits in initiation (Voon et al., 2006) although as with depression and anxiety, this may relate less to the direct effects of stimulation and more to the role of neurodegeneration and dopaminergic deficit. Initiation is an executive function supported by the frontal cortex, which complements the role of inhibition underpinned by the STN (Fuster, 2015). An adaptive agent balances these processes, such that stimulus-response associations rapidly generate behavior when appropriate, but automatic responses are suppressed when alternative actions with greater value are available. In PD, problems with the

424

P.E. MOSLEY AND H. AKRAM

initiation of action give rise to cardinal motor symptoms such as rigidity and bradykinesia. In the nonmotor domain, initiation deficits are associated with apathy; a cognitive and emotional syndrome characterized by diminished goal-directed behavior and diminished interest (Starkstein et al., 1992; Starkstein and Brockman, 2011). Initiation can be measured using the construct of energization, the attentional process that underpins the initiation and maintenance of a response (Barker et al., 2018). Timed word (verbal) fluency tasks are commonly used to assess energization, during which verbal responses must be initiated and maintained after a starting cue. If the process of energization is defective, responding slows over time, akin to “running out of steam” (Robinson et al., 2015). In PD, dopaminergic depletion may impair energization resulting in a quantitative reduction in effortful activity (McAuley, 2003). Deficits in word fluency have been reported from the early stages of PD (Lees and Smith, 1983). Performance in both semantic and phonemic verbal fluency may worsen after subthalamic DBS (Parsons et al., 2006; Witt et al., 2008; Combs et al., 2015). However, this postoperative impairment may relate to surgeryindependent factors such as aging and the progression of neurodegeneration (Daniels et al., 2010), postoperative reduction of dopaminergic therapies unmasking apathy (Thobois et al., 2010), or to penetration of the caudate nucleus during lead implantation and stimulation outside of the optimal motor subregion of the STN (Witt et al., 2013). For most patients, STN-DBS has no cognitive sequelae, but many clinical studies exclude individuals aged over 70 and those with significant preexisting cognitive impairment. Risk factors for the progression of cognitive impairment postoperatively may include older age, total presurgical dopaminergic medication requirement and axial symptoms, potentially reflecting advanced disease (Daniels et al., 2010). However, controlled prospective trials are required to establish whether STNDBS has any effect on the longitudinal risk of developing PD dementia (Limousin and Foltynie, 2019).

SUBTHALAMIC DBS FOR PSYCHIATRIC DISORDERS The successful use of DBS for neurological disorders has raised the prospect of the use of DBS in the treatment of intractable psychiatric conditions. This has been driven by an underlying biological, neurocircuit-based model of severe mental disorder (Haber and Knutson, 2010), but also by the history of ablative neurosurgery for psychiatric disorders including treatment-resistant depression and OCD (Christmas et al., 2004). Presently,

STN-DBS has only been employed in the treatment of OCD, but preclinical models also suggest a potential role in the treatment of addiction. In these conditions, the anteromedial STN is deliberately targeted with the goal of modulating affective and cognitive circuits connected to this region of the nucleus. Whereas in PD such stimulation may induce harmful disinhibition, in these psychiatric conditions this disinhibition is harnessed to break engrained ritualistic behaviors (OCD) or the habitual patterns of substance use (addiction).

Obsessive–compulsive disorder OCD has an estimated lifetime prevalence of between 1% and 2% (Kessler et al., 2005). It is characterized by the intrusion of ego-dystonic, anxiety-provoking thoughts (obsessions). These are accompanied by mental acts or behaviors (compulsions), which must be carried out to neutralize the obsessions or to mitigate anxiety associated with them (American Psychiatric Association, 2013). The phenomenology of these obsessions is broad and often incomprehensible. Sufferers may be excessively concerned with germs (contamination), preoccupied with symmetry or disturbed by intrusive violent, sexual, or religious thoughts. Compulsions such as cleaning, ordering, checking, and repeating may consume waking hours. Secondary anxiety and depressive disorders are common (Torres et al., 2006). The onset of OCD is typically in early adulthood, but with a considerable delay (over 10 years) between the onset of symptoms and first treatment (Cullen et al., 2008), often reflecting attempts to conceal or adapt to the disorder. Pharmacological treatment is often unsatisfactory; approximately 50% of patients respond to serotonergic antidepressants at high dose (Erzegovesi et al., 2001), but remission of symptoms is rare. Augmentation strategies may be helpful for approximately a third of the remainder (Bloch et al., 2006), but a considerable proportion remains treatment resistant or impaired by residual symptoms. Psychological treatment is often unacceptable; many individuals cannot tolerate challenging their obsessive symptoms during exposureoriented cognitive behavioral therapy (Issakidis and Andrews, 2002). These factors mean that OCD is a chronic disorder with a detrimental effect on functioning across the lifespan, making it a leading neuropsychiatric cause of global disability (Lopez, 2006). Neuropsychiatric correlates of OCD include hyperactivity of fronto-striatal regions at baseline and in response to symptom provocation (Adler et al., 2000). Abnormal patterns of metabolic activity predict response to medication (Rauch et al., 2002) and normalize after successful treatment (Baxter et al., 1992). Biological models of OCD implicate the OFC, the nucleus accumbens (NAcc), the anterior cingulate cortex (ACC),

NEUROPSYCHIATRIC EFFECTS OF SUBTHALAMIC DBS dorsolateral prefrontal cortex (DLPFC), and the amygdala, giving rise to a pathogenic cognitive profile characterized by cognitive inflexibility, impaired reward processing, heightened error sensitivity, and enhanced fear conditioning (Milad and Rauch, 2012). Animal models of OCD also suggest a role for abnormal patterns of connectivity in fronto-striatal circuits (Ahmari et al., 2013), and electrophysiological activity in the STN correlates with the severity of OCD symptoms (Welter et al., 2011), consistent with a role for this nucleus in the pathophysiology of the disorder. Following the improvement of comorbid OCD in 2 patients with PD treated with STN-DBS (Mallet et al., 2002), 16 patients with primary, treatment-refractory, severe OCD took part in a 3-month on-stimulation/3month off-stimulation crossover trial (Mallet et al., 2008). An anteromedial site of stimulation targeted the cognitive and affective portions of this nucleus. A significant difference between active and sham stimulation was noted, with a mean difference of nine points on the Yale-Brown Obsessive–Compulsive Scale (YBOCS), the gold-standard instrument for the assessment of OCD. Typically, a decrease in total score of 35% is taken to be significant on this scale, with a maximum score of 40 and scores above 24 taken to reflect a severe disorder. Therefore, these findings were extremely promising. However, serious adverse events arose in a number of participants, including one intracerebral hemorrhage, two infections necessitating device explantation, three cases of hypomania, and one case of disabling dyskinetic movements. In recent work, six participants with treatmentrefractory, severe OCD were implanted with bilateral electrodes in both the ventromedial/anteromedial STN and the ventral capsule/ventral striatum (VC/VS) to determine the optimal stimulation site for effective treatment and define anatomical networks associated with treatment response (Tyagi et al., 2019). Participants received a 3-month period of stimulation at both sites, in a blinded, counterbalanced, crossover design. DBS at both sites led to a significant reduction in OCD symptoms (a mean reduction of 16 points with STN-DBS). Side effects included hypomanic symptoms in two participants during STN-DBS, which resolved with device reprogramming. Although there was no difference by surgical target in the effect on the YBOCS score, there was an interesting dissociation in the effect on mood and cognitive flexibility: VC/VS-DBS reduced depressive symptoms, while STN-DBS remediated cognitive inflexibility in a neuropsychological measure of reversal learning. This is consistent with a role for aberrant STN activity in OCD in mediating cognitive rigidity, which may be a mechanism linking obsessive thoughts to their associated compulsions. Tractography seeded from activated electrode contacts suggested that these dissociated

425

effects were related to modulation of distinct brain networks connected to the stimulation site. In the case of STN-DBS, streamlines were visualized between the site of stimulation and lateral OFC, ACC, and DLPFC (Fig. 26.3). Further work will be necessary to establish if these dissociable network findings and distinct effects on affective and cognitive symptom clusters are reproducible in larger cohorts. It has been suggested that the beneficial effect of DBS at both sites on core OCD symptoms may be related to modulation of a distinct frontothalamic tract connecting both regions to the medial and lateral prefrontal cortex (Baldermann et al., 2019; Li et al., 2019). However, an alternative explanation may simply be that these two targets lie on a common network that mediates OCD symptomatology, but each also has distinct connectivity to other regions that modulate these divergent effects on affect and cognition.

Addiction Substance use disorders are characterized by persisting use of intoxicating, psychoactive compounds despite psychological, medical, social, financial, and legal harms. In Australia, there has been a growing concern about the impact of crystal methamphetamine (“ice”), which has been associated with high levels of socioeconomic instability and mental illness (Australian Institute of Health and Welfare, 2017), with estimated annual social and economic costs of AUD 5 billion (Whetton et al., 2017). In animal models of addiction, there is preliminary evidence that STN-DBS reduces the incentive salience of amphetamines, while also preventing reinstatement of compulsive amphetamine use after a period of abstinence, suggesting that this may be a potential target for treatment of refractory cases. Contemporary perspectives of addiction propose an initial phase of action-outcome (model-based) learning, in which drug use is positively reinforced (Everitt and Robbins, 2005). This is mediated by the NAcc, a region of the ventral striatum implicated in neuronal coding of appetitive rewards and hence with reinforcement learning (Schultz et al., 1997; Abler et al., 2006; Daw et al., 2006; Tanaka et al., 2008; Wittmann et al., 2008; Basar et al., 2010; Haber and Knutson, 2010; de Wit et al., 2012; Kishida et al., 2016; Hampton et al., 2017). However, extended training leads to the formation of stimulus-response (habit or model-free) associations that sustain pathological behavior, even when addicted individuals want to stop using drugs (Voon et al., 2015). These habits now impair behavioral flexibility and are maintained by the enhanced significance of drug-associated conditioned reinforcers, which act as stimuli for continued drug-seeking behavior. For example, relapses in drug-taking behavior commonly occur

426

P.E. MOSLEY AND H. AKRAM

Fig. 26.3. Group-average tractography from anteromedial STN-DBS in treatment-refractory severe OCD. Average streamlines generated from the simulated stimulation volumes of participants treated with STN-DBS for OCD. Streamlines were connected to the lateral orbitofrontal cortex (LAT OFC), dorsal anterior cingulate cortex (DACC), dorsolateral prefrontal cortex (DLPFC), and medial forebrain bundle (MFB). STN-DBS was associated with improved cognitive flexibility as well as reduced OCD symptoms (Tyagi et al., 2019). VC, ventral capsule; VTA, volume of tissue activation.

after reexposure to a drug-taking environment with salient drug-taking cues. This is because the model-free updating scheme is slow to learn new contingencies when the consequences of actions and their values change after a long period of training: termed “devaluation insensitivity” in animal models, such as when a rat maintains a lever press despite satiation and thus devaluation of its food reward (Balleine and Dickinson, 1998). This habitual behavior is underpinned by neuronal activity in the dorsal, rather than ventral striatum. As behavior becomes stereotyped, activity in the ventral striatum decreases and dorsal striatal activity increases (Thorn et al., 2010). This transition from a goaldirected action to a stimulus-response habit depends on connectivity of the striatum with the midbrain. When these recurrent circuits are interrupted in animal models, it reduces the acquisition of compulsive drug-seeking habits (Belin and Everitt, 2008). Due to the central place of the STN within these subcortical pathways, it may be a therapeutic target for interrupting the acquisition of compulsive consumption or preventing the persistence of compulsive behaviors once established (Pelloux and Baunez, 2019). An important feature of drug addiction is the narrowing of the user’s behavioral repertoire toward activities

associated with drug use at the expense of other social and interpersonal goals. In a rat model of amphetamine dependence, STN-DBS reduced the motivation to work for a cocaine reward while increasing the motivation to work for a natural reward (food) (Rouaud et al., 2010). This effect on incentive salience was independent of a direct effect on the reinforcing properties of the drug. In addition, STN-DBS reduced the conditioned preference for a place previously associated with a cocaine reward, while increasing the preference for an environment associated with food. This dissociation of motivation has therapeutic relevance: neuromodulation for addiction should preferentially affect conditioned behaviors associated with the drug of dependence, rather than provoking a global blunting of motivation for all “naturally” rewarding stimuli. Another key feature of addiction is the rapid reinstatement of dependent patterns of substance use if relapse occurs after a period of abstinence. In a similar rat model of amphetamine dependence, STN-DBS not only reduced the initial escalation of cocaine use when rats had long periods of access to this substance, but it also limited reescalation of cocaine use after a protracted

NEUROPSYCHIATRIC EFFECTS OF SUBTHALAMIC DBS period of abstinence (Pelloux et al., 2018). A similar effect was observed in a rat model of opioid dependence (Wade et al., 2017). Despite this promising preclinical work, one of the main barriers to implementing DBS of any target nucleus for addiction in human participants is the difficulty in recruiting potential candidates (Luigjes et al., 2015). Previous European trials have been limited by low referral and high dropout rates, which may relate to perceptions of addiction as a “moral” rather than “biological” phenomenon, the fluctuant relapsing and remitting nature of addiction and the socioeconomic and interpersonal instability of many with substance use disorders.

CONCLUSIONS AND FUTURE DIRECTIONS The STN is a key node in the control of voluntary movement, cognition, and emotion that functions as a “handbrake” on basal ganglia output. It has been implicated in behavioral inhibition, cognitive control, and the monitoring of cognitive conflict. Using focused electrical stimulation facilitated by DBS, manipulating STN activity relieves the motor symptoms of PD, but may give rise to harmful neuropsychiatric symptoms as a corollary of the small size and dense connectivity of this brain region. Future work will employ increasingly sophisticated electrical targeting to identify “sweet spots” that optimize motor symptoms without neuropsychiatric sequelae. In contrast, modulating these nonmotor circuits may have an emerging role in the treatment of neuropsychiatric disorders such as OCD and potentially, addiction. Future work will carefully delineate the contribution of the STN from adjacent regions that could also plausibly be modulated by anteriomedial STN stimulation, such as the medial forebrain bundle and interpeduncular– habenula tracts.

REFERENCES Aarsland D, Larsen JP, Karlsen K et al. (1999). Mental symptoms in Parkinson’s disease are important contributors to caregiver distress. Int J Geriatr Psychiatry 14: 866–874. Abler B, Walter H, Erk S et al. (2006). Prediction error as a linear function of reward probability is coded in human nucleus accumbens. Neuroimage 31: 790–795. Accolla EA, Dukart J, Helms G et al. (2014). Brain tissue properties differentiate between motor and limbic basal ganglia circuits. Hum Brain Mapp 35: 5083–5092. Adler CM, McDonough-Ryan P, Sax KW et al. (2000). fMRI of neuronal activation with symptom provocation in unmedicated patients with obsessive compulsive disorder. J Psychiatr Res 34: 317–324.

427

Ahlskog JE, Muenter MD (2001). Frequency of levodoparelated dyskinesias and motor fluctuations as estimated from the cumulative literature. Mov Disord 16: 448–458. Ahmari SE, Spellman T, Douglass NL et al. (2013). Repeated cortico-striatal stimulation generates persistent OCD-like behavior. Science 340: 1234–1239. Akram H, Sotiropoulos SN, Jbabdi S et al. (2017). Subthalamic deep brain stimulation sweet spots and hyperdirect cortical connectivity in Parkinson’s disease. Neuroimage 158: 332–345. American Psychiatric Association (2013). Diagnostic and statistical manual of mental disorders: DSM-5, American Psychiatric Publishing, Washington, DC. Amunts K, Lepage C, Borgeat L et al. (2013). BigBrain: an ultrahigh-resolution 3D human brain model. Science 340: 1472–1475. Antonelli F, Ko JH, Miyasaki J et al. (2014). Dopamineagonists and impulsivity in Parkinson’s disease: impulsive choices vs. impulsive actions. Hum Brain Mapp 35: 2499–2506. Appleby BS, Duggan PS, Regenberg A et al. (2007). Psychiatric and neuropsychiatric adverse events associated with deep brain stimulation: a meta-analysis of ten years’ experience. Mov Disord 22: 1722–1728. Aron AR, Robbins TW, Poldrack RA (2004). Inhibition and the right inferior frontal cortex. Trends Cogn Sci 8: 170–177. Aron AR, Behrens TE, Smith S et al. (2007). Triangulating a cognitive control network using diffusion-weighted magnetic resonance imaging (MRI) and functional MRI. J Neurosci 27: 3743–3752. Australian Institute of Health and Welfare (2017). National drug strategy household survey 2016: detailed findings. Drug statistics series no. 31. Cat. no. PHE 214, Canberra. Aziz TZ, Peggs D, Sambrook MA et al. (1991). Lesion of the subthalamic nucleus for the alleviation of 1-methyl-4phenyl-1,2,3,6-tetrahydropyridine (MPTP)-induced parkinsonism in the primate. Mov Disord 6: 288–292. Baldermann JC, Melzer C, Zapf A et al. (2019). Connectivity profile predictive of effective deep brain stimulation in obsessive-compulsive disorder. Biol Psychiatry 85: 735–743. Balleine BW, Dickinson A (1998). Goal-directed instrumental action: contingency and incentive learning and their cortical substrates. Neuropharmacology 37: 407–419. Barker MS, Nelson NL, O’Sullivan JD et al. (2018). Energization and spoken language production: evidence from progressive supranuclear palsy. Neuropsychologia 119: 349–362. Basar K, Sesia T, Groenewegen H et al. (2010). Nucleus accumbens and impulsivity. Prog Neurobiol 92: 533–557. Baxter Jr LR, Schwartz JM, Bergman KS et al. (1992). Caudate glucose metabolic rate changes with both drug and behavior therapy for obsessive-compulsive disorder. Arch Gen Psychiatry 49: 681–689. Bejjani BP, Damier P, Arnulf I et al. (1999). Transient acute depression induced by high-frequency deep-brain stimulation. N Engl J Med 340: 1476–1480.

428

P.E. MOSLEY AND H. AKRAM

Belin D, Everitt BJ (2008). Cocaine seeking habits depend upon dopamine-dependent serial connectivity linking the ventral with the dorsal striatum. Neuron 57: 432–441. Benabid AL, Pollak P, Gross C et al. (1994). Acute and longterm effects of subthalamic nucleus stimulation in Parkinson’s disease. Stereotact Funct Neurosurg 62: 76–84. Bergman H, Wichmann T, DeLong MR (1990). Reversal of experimental parkinsonism by lesions of the subthalamic nucleus. Science 249: 1436–1438. Bergman H, Wichmann T, Karmon B et al. (1994). The primate subthalamic nucleus. II. Neuronal activity in the MPTP model of parkinsonism. J Neurophysiol 72: 507–520. Bloch MH, Landeros-Weisenberger A, Kelmendi B et al. (2006). A systematic review: antipsychotic augmentation with treatment refractory obsessive-compulsive disorder. Mol Psychiatry 11: 622–632. Brown P, Oliviero A, Mazzone P et al. (2001). Dopamine dependency of oscillations between subthalamic nucleus and pallidum in Parkinson’s disease. J Neurosci 21: 1033–1038. Carter JH, Stewart BJ, Lyons KS et al. (2008). Do motor and nonmotor symptoms in PD patients predict caregiver strain and depression? Mov Disord 23: 1211–1216. Carter JH, Lyons KS, Lindauer A et al. (2012). Pre-death grief in Parkinson’s caregivers: a pilot survey-based study. Parkinsonism Relat Disord 18: S15–S18. Castrioto A, Lhommee E, Moro E et al. (2014). Mood and behavioural effects of subthalamic stimulation in Parkinson’s disease. Lancet Neurol 13: 287–305. Cavanagh JF, Wiecki TV, Cohen MX et al. (2011). Subthalamic nucleus stimulation reverses mediofrontal influence over decision threshold. Nat Neurosci 14: 1462–1467. Chopra A, Tye SJ, Lee KH et al. (2011). Voltage-dependent mania after subthalamic nucleus deep brain stimulation in Parkinson’s disease: a case report. Biol Psychiatry 70: e5–e7. Christmas D, Matthews K, Eljamel MS (2004). Neurosurgery for mental disorder. Br J Psychiatry 185: 173–174; author reply 174. Coenen VA, Prescher A, Schmidt T et al. (2008). What is dorso-lateral in the subthalamic nucleus (STN)?—a topographic and anatomical consideration on the ambiguous description of today’s primary target for deep brain stimulation (DBS) surgery. Acta Neurochir 150: 1163–1165. Coenen VA, Honey CR, Hurwitz T et al. (2009). Medial forebrain bundle stimulation as a pathophysiological mechanism for hypomania in subthalamic nucleus deep brain stimulation for Parkinson’s disease. Neurosurgery 64: 1106–1114discussion 1114–1115. Coenen VA, Panksepp J, Hurwitz TA et al. (2012). Human medial forebrain bundle (MFB) and anterior thalamic radiation (ATR): imaging of two major subcortical pathways and the dynamic balance of opposite affects in understanding depression. J Neuropsychiatry Clin Neurosci 24: 223–236.

Combs HL, Folley BS, Berry DT et al. (2015). Cognition and depression following deep brain stimulation of the subthalamic nucleus and globus pallidus pars internus in Parkinson’s disease: a meta-analysis. Neuropsychol Rev 25: 439–454. Cullen B, Samuels JF, Pinto A et al. (2008). Demographic and clinical characteristics associated with treatment status in family members with obsessive-compulsive disorder. Depress Anxiety 25: 218–224. D’Alberto N, Funnell M, Potter A et al. (2017). A split-brain case study on the hemispheric lateralization of inhibitory control. Neuropsychologia 99: 24–29. Daniele A, Albanese A, Contarino MF et al. (2003). Cognitive and behavioural effects of chronic stimulation of the subthalamic nucleus in patients with Parkinson’s disease. J Neurol Neurosurg Psychiatry 74: 175–182. Daniels C, Krack P, Volkmann J et al. (2010). Risk factors for executive dysfunction after subthalamic nucleus stimulation in Parkinson’s disease. Mov Disord 25: 1583–1589. Daw ND, O’Doherty JP, Dayan P et al. (2006). Cortical substrates for exploratory decisions in humans. Nature 441: 876–879. de Lau LM, Schipper CM, Hofman A et al. (2005). Prognosis of Parkinson disease: risk of dementia and mortality: the Rotterdam Study. Arch Neurol 62: 1265–1269. de Wit S, Watson P, Harsay HA et al. (2012). Corticostriatal connectivity underlies individual differences in the balance between habitual and goal-directed action control. J Neurosci 32: 12066–12075. Doorn KJ, Lucassen PJ, Boddeke HW et al. (2012). Emerging roles of microglial activation and non-motor symptoms in Parkinson’s disease. Prog Neurobiol 98: 222–238. Doorn KJ, Moors T, Drukarch B et al. (2014). Microglial phenotypes and toll-like receptor 2 in the substantia nigra and hippocampus of incidental Lewy body disease cases and Parkinson’s disease patients. Acta Neuropathol Commun 2: 90. Erzegovesi S, Cavallini MC, Cavedini P et al. (2001). Clinical predictors of drug response in obsessive-compulsive disorder. J Clin Psychopharmacol 21: 488–492. Eusebio A, Pogosyan A, Wang S et al. (2009). Resonance in subthalamo-cortical circuits in Parkinson’s disease. Brain 132: 2139–2150. Eusebio A, Thevathasan W, Doyle Gaynor L et al. (2011). Deep brain stimulation can suppress pathological synchronisation in parkinsonian patients. J Neurol Neurosurg Psychiatry 82: 569–573. Eusebio A, Witjas T, Cohen J et al. (2013). Subthalamic nucleus stimulation and compulsive use of dopaminergic medication in Parkinson’s disease. J Neurol Neurosurg Psychiatry 84: 868–874. Everitt BJ, Robbins TW (2005). Neural systems of reinforcement for drug addiction: from actions to habits to compulsion. Nat Neurosci 8: 1481–1489. Ewert S, Plettig P, Li N et al. (2018). Toward defining deep brain stimulation targets in MNI space: a subcortical atlas based on multimodal MRI, histology and structural connectivity. Neuroimage 170: 271–282.

NEUROPSYCHIATRIC EFFECTS OF SUBTHALAMIC DBS Foltynie T, Zrinzo L, Martinez-Torres I et al. (2011). MRI-guided STN DBS in Parkinson’s disease without microelectrode recording: efficacy and safety. J Neurol Neurosurg Psychiatry 82: 358. Forsaa EB, Larsen JP, Wentzel-Larsen T et al. (2010). What predicts mortality in Parkinson disease?: a prospective population-based long-term study. Neurology 75: 1270–1276. Frank MJ, Samanta J, Moustafa AA et al. (2007). Hold your horses: impulsivity, deep brain stimulation, and medication in parkinsonism. Science 318: 1309–1312. Fuster JM (2015). Chapter 8—Overview of prefrontal functions: E pluribus unum—coordinating new sequences of purposeful action. In: JM Fuster (Ed.), The prefrontal cortex, fifth edn. Academic Press, San Diego, pp. 375–425. Giovannoni G, O’Sullivan JD, Turner K et al. (2000). Hedonistic homeostatic dysregulation in patients with Parkinson’s disease on dopamine replacement therapies. J Neurol Neurosurg Psychiatry 68: 423–428. Haagensen BN, Herz DM, Meder D et al. (2020). Linking brain activity during sequential gambling to impulse control in Parkinson’s disease. Neuroimage Clin 27: 102330. Haber SN, Knutson B (2010). The reward circuit: linking primate anatomy and human imaging. Neuropsychopharmacology 35: 4–26. Hampton WH, Alm KH, Venkatraman V et al. (2017). Dissociable frontostriatal white matter connectivity underlies reward and motor impulsivity. Neuroimage 150: 336–343. Haynes WI, Haber SN (2013). The organization of prefrontalsubthalamic inputs in primates provides an anatomical substrate for both functional specificity and integration: implications for basal ganglia models and deep brain stimulation. J Neurosci 33: 4804–4814. Hazrati LN, Parent A (1992). Convergence of subthalamic and striatal efferents at pallidal level in primates: an anterograde double-labeling study with biocytin and PHA-L. Brain Res 569: 336–340. Hershey T, Revilla FJ, Wernle A et al. (2004). Stimulation of STN impairs aspects of cognitive control in PD. Neurology 62: 1110–1114. Hershey T, Campbell MC, Videen TO et al. (2010). Mapping Go-No-Go performance within the subthalamic nucleus region. Brain 133: 3625–3634. Herzog J, Reiff J, Krack P et al. (2003). Manic episode with psychotic symptoms induced by subthalamic nucleus stimulation in a patient with Parkinson’s disease. Mov Disord 18: 1382–1384. Horn A, Reich M, Vorwerk J et al. (2017). Connectivity predicts deep brain stimulation outcome in Parkinson disease. Ann Neurol 82: 67–78. Irmen F, Horn A, Meder D et al. (2019). Sensorimotor subthalamic stimulation restores risk-reward trade-off in Parkinson’s disease. Mov Disord 34: 366–376. Issakidis C, Andrews G (2002). Service utilisation for anxiety in an Australian community sample. Soc Psychiatry Psychiatr Epidemiol 37: 153–163.

429

Kessler RC, Berglund P, Demler O et al. (2005). Lifetime prevalence and age-of-onset distributions of DSM-IV disorders in the National Comorbidity Survey Replication. Arch Gen Psychiatry 62: 593–602. Kishida KT, Saez I, Lohrenz T et al. (2016). Subsecond dopamine fluctuations in human striatum encode superposed error signals about actual and counterfactual reward. Proc Natl Acad Sci USA 113: 200–205. Krack P, Batir A, Van Blercom N et al. (2003). Five-year follow-up of bilateral stimulation of the subthalamic nucleus in advanced Parkinson’s disease. N Engl J Med 349: 1925–1934. Lambert C, Zrinzo L, Nagy Z et al. (2012). Confirmation of functional zones within the human subthalamic nucleus: patterns of connectivity and sub-parcellation using diffusion weighted imaging. Neuroimage 60: 83–94. Lees AJ, Smith E (1983). Cognitive deficits in the early stages of Parkinson’s disease. Brain 106: 257–270. Lhommee E, Klinger H, Thobois S et al. (2012). Subthalamic stimulation in Parkinson’s disease: restoring the balance of motivated behaviours. Brain 135: 1463–1477. Li N, Baldermann JC, Kibleur A et al. (2019). Toward a unified connectomic target for deep brain stimulation in obsessive-compulsive disorder. Nat Commun 11: 3364. Lim SY, O’Sullivan SS, Kotschet K et al. (2009). Dopamine dysregulation syndrome, impulse control disorders and punding after deep brain stimulation surgery for Parkinson’s disease. J Clin Neurosci 16: 1148–1152. Limousin P, Foltynie T (2019). Long-term outcomes of deep brain stimulation in Parkinson disease. Nat Rev Neurol 15: 234–242. Lopez AD (2006). Global burden of disease and risk factors, Oxford University Press; World Bank, New York, NY; Washington, DC. Luigjes J, van den Brink W, Schuurman PR et al. (2015). Is deep brain stimulation a treatment option for addiction? Addiction 110: 547–548. Madler B, Coenen VA (2012). Explaining clinical effects of deep brain stimulation through simplified target-specific modeling of the volume of activated tissue. AJNR Am J Neuroradiol 33: 1072–1080. Maks CB, Butson CR, Walter BL et al. (2009). Deep brain stimulation activation volumes and their association with neurophysiological mapping and therapeutic outcomes. J Neurol Neurosurg Psychiatry 80: 659–666. Mallet L, Mesnage V, Houeto JL et al. (2002). Compulsions, Parkinson’s disease, and stimulation. Lancet 360: 1302–1304. Mallet L, Schupbach M, N’Diaye K et al. (2007). Stimulation of subterritories of the subthalamic nucleus reveals its role in the integration of the emotional and motor aspects of behavior. Proc Natl Acad Sci USA 104: 10661–10666. Mallet L, Polosan M, Jaafari N et al. (2008). Subthalamic nucleus stimulation in severe obsessive-compulsive disorder. N Engl J Med 359: 2121–2134. Mandat TS, Hurwitz T, Honey CR (2006). Hypomania as an adverse effect of subthalamic nucleus stimulation:

430

P.E. MOSLEY AND H. AKRAM

report of two cases. Acta Neurochir 148: 895–897. discussion 898. McAuley JH (2003). The physiological basis of clinical deficits in Parkinson’s disease. Prog Neurobiol 69: 27–48. Milad MR, Rauch SL (2012). Obsessive-compulsive disorder: beyond segregated cortico-striatal pathways. Trends Cogn Sci 16: 43–51. Mosley PE, Breakspear M, Coyne T et al. (2018a). Caregiver burden and caregiver appraisal of psychiatric symptoms are not modulated by subthalamic deep brain stimulation for Parkinson’s disease. NPJ Parkinsons Dis 4: 12. Mosley PE, Marsh R, Perry A et al. (2018b). Persistence of mania after cessation of stimulation following subthalamic deep brain stimulation. J Neuropsychiatry Clin Neurosci 30: 246–249. Mosley PE, Smith D, Coyne T et al. (2018c). The site of stimulation moderates neuropsychiatric symptoms after subthalamic deep brain stimulation for Parkinson’s disease. Neuroimage Clin 18: 996–1006. Mosley PE, Paliwal S, Robinson K et al. (2019a). The structural connectivity of discrete networks underlies impulsivity and gambling in Parkinson’s disease. Brain 142: 3917–3935. Mosley PE, Robinson K, Coyne T et al. (2019b). ‘Woe betides anybody who tries to turn me down.’ A qualitative analysis of neuropsychiatric symptoms following subthalamic deep brain stimulation for Parkinson’s disease. Neuroethics. https://doi.org/10.1007/s12152-019-09410-x. Mosley PE, Paliwal S, Robinson K et al. (2020). The structural connectivity of subthalamic deep brain stimulation correlates with impulsivity in Parkinson’s disease. Brain 143: 2235–2254. Moum SJ, Price CC, Limotai N et al. (2012). Effects of STN and GPi deep brain stimulation on impulse control disorders and dopamine dysregulation syndrome. PLoS One 7: e29768. Nambu A, Tokuno H, Hamada I et al. (2000). Excitatory cortical inputs to pallidal neurons via the subthalamic nucleus in the monkey. J Neurophysiol 84: 289–300. Nambu A, Tokuno H, Takada M (2002). Functional significance of the cortico-subthalamo-pallidal ‘hyperdirect’ pathway. Neurosci Res 43: 111–117. Nirenberg MJ (2010). Dopamine agonist withdrawal syndrome and non-motor symptoms after Parkinson’s disease surgery. Brain 133: e155; author reply e156. Obeso I, Wilkinson L, Rodriguez-Oroz MC et al. (2013). Bilateral stimulation of the subthalamic nucleus has differential effects on reactive and proactive inhibition and conflict-induced slowing in Parkinson’s disease. Exp Brain Res 226: 451–462. Paliwal S, Mosley PE, Breakspear M et al. (2019). Subjective estimates of uncertainty during gambling and impulsivity after subthalamic deep brain stimulation for Parkinson’s disease. Sci Rep 9: 14795. Parsons TD, Rogers SA, Braaten AJ et al. (2006). Cognitive sequelae of subthalamic nucleus deep brain stimulation in Parkinson’s disease: a meta-analysis. Lancet Neurol 5: 578–588.

Pelloux Y, Baunez C (2019). Chapter 19—Harnessing circuits for the treatment of addictive disorders. In: M Torregrossa (Ed.), Neural mechanisms of addiction. Academic Press, pp. 271–285. Pelloux Y, Degoulet M, Tiran-Cappello A et al. (2018). Subthalamic nucleus high frequency stimulation prevents and reverses escalated cocaine use. Mol Psychiatry 23: 2266–2276. Possin KL, Brambati SM, Rosen HJ et al. (2009). Rule violation errors are associated with right lateral prefrontal cortex atrophy in neurodegenerative disease. J Int Neuropsychol Soc 15: 354–364. Rabinak CA, Nirenberg MJ (2010). Dopamine agonist withdrawal syndrome in Parkinson disease. Arch Neurol 67: 58–63. Rae CL, Hughes LE, Anderson MC et al. (2015). The prefrontal cortex achieves inhibitory control by facilitating subcortical motor pathway connectivity. J Neurosci 35: 786–794. Rauch SL, Shin LM, Dougherty DD et al. (2002). Predictors of fluvoxamine response in contamination-related obsessive compulsive disorder: a PET symptom provocation study. Neuropsychopharmacology 27: 782–791. Raucher-Chene D, Charrel CL, de Maindreville AD et al. (2008). Manic episode with psychotic symptoms in a patient with Parkinson’s disease treated by subthalamic nucleus stimulation: improvement on switching the target. J Neurol Sci 273: 116–117. Robinson GA, Spooner D, Harrison WJ (2015). Frontal dynamic aphasia in progressive supranuclear palsy: distinguishing between generation and fluent sequencing of novel thoughts. Neuropsychologia 77: 62–75. Romito LM, Raja M, Daniele A et al. (2002). Transient mania with hypersexuality after surgery for high frequency stimulation of the subthalamic nucleus in Parkinson’s disease. Mov Disord 17: 1371–1374. Rouaud T, Lardeux S, Panayotis N et al. (2010). Reducing the desire for cocaine with subthalamic nucleus deep brain stimulation. Proc Natl Acad Sci USA 107: 1196–1200. Rudebeck PH, Murray EA (2014). The orbitofrontal oracle: cortical mechanisms for the prediction and evaluation of specific behavioral outcomes. Neuron 84: 1143–1156. Schuepbach WM, Rau J, Knudsen K et al. (2013). Neurostimulation for Parkinson’s disease with early motor complications. N Engl J Med 368: 610–622. Schultz W, Dayan P, Montague PR (1997). A neural substrate of prediction and reward. Science 275: 1593–1599. Shimamoto SA, Ryapolova-Webb ES, Ostrem JL et al. (2013). Subthalamic nucleus neurons are synchronized to primary motor cortex local field potentials in Parkinson’s disease. J Neurosci 33: 7220–7233. Shin H, Lee JY, Youn J et al. (2012). Factors contributing to spousal and offspring caregiver burden in Parkinson’s disease. Eur Neurol 67: 292–296. Smith SM, Nichols TE (2009). Threshold-free cluster enhancement: addressing problems of smoothing, threshold dependence and localisation in cluster inference. Neuroimage 44: 83–98.

NEUROPSYCHIATRIC EFFECTS OF SUBTHALAMIC DBS Stark AJ, Smith CT, Lin Y-C et al. (2018). Nigrostriatal and mesolimbic D2/3 receptor expression in Parkinson’s disease patients with compulsive reward-driven behaviors. J Neurosci 38: 3230–3239. Starkstein SE, Brockman S (2011). Apathy and Parkinson’s disease. Curr Treat Options Neurol 13: 267–273. Starkstein SE, Mayberg HS, Preziosi TJ et al. (1992). Reliability, validity, and clinical correlates of apathy in Parkinson’s disease. J Neuropsychiatry Clin Neurosci 4: 134. Takahashi YK, Roesch MR, Wilson RC et al. (2011). Expectancy-related changes in firing of dopamine neurons depend on orbitofrontal cortex. Nat Neurosci 14: 1590–1597. Tanaka SC, Balleine BW, O’Doherty JP (2008). Calculating consequences: brain systems that encode the causal effects of actions. J Neurosci 28: 6750–6755. Thobois S, Hotton GR, Pinto S et al. (2007). STN stimulation alters pallidal-frontal coupling during response selection under competition. J Cereb Blood Flow Metab 27: 1173–1184. Thobois S, Ardouin C, Lhommee E et al. (2010). Nonmotor dopamine withdrawal syndrome after surgery for Parkinson’s disease: predictors and underlying mesolimbic denervation. Brain 133: 1111–1127. Thorn CA, Atallah H, Howe M et al. (2010). Differential dynamics of activity changes in dorsolateral and dorsomedial striatal loops during learning. Neuron 66: 781–795. Torres AR, Prince MJ, Bebbington PE et al. (2006). Obsessivecompulsive disorder: prevalence, comorbidity, impact, and help-seeking in the British National Psychiatric Morbidity Survey of 2000. Am J Psychiatry 163: 1978–1985. Tyagi H, Apergis-Schoute AM, Akram H et al. (2019). A randomized trial directly comparing ventral capsule and anteromedial subthalamic nucleus stimulation in obsessive-compulsive disorder: clinical and imaging evidence for dissociable effects. Biol Psychiatry 85: 726–734. van Eimeren T, Pellecchia G, Cilia R et al. (2010). Drug-induced deactivation of inhibitory networks predicts pathological gambling in PD. Neurology 75: 1711–1716. Vila M, Perier C, Feger J et al. (2000). Evolution of changes in neuronal activity in the subthalamic nucleus of rats with unilateral lesion of the substantia nigra assessed by metabolic and electrophysiological measurements. Eur J Neurosci 12: 337–344. Volkmann J, Daniels C, Witt K (2010). Neuropsychiatric effects of subthalamic neurostimulation in Parkinson disease. Nat Rev Neurol 6: 487–498. Voon V, Kubu C, Krack P et al. (2006). Deep brain stimulation: neuropsychological and neuropsychiatric issues. Mov Disord 21: S305–S327. Voon V, Krack P, Lang AE et al. (2008). A multicentre study on suicide outcomes following subthalamic stimulation for Parkinson’s disease. Brain 131: 2720–2728. Voon V, Gao J, Brezing C et al. (2011). Dopamine agonists and risk: impulse control disorders in Parkinson’s disease. Brain 134: 1438–1446.

431

Voon V, Derbyshire K, Ruck C et al. (2015). Disorders of compulsivity: a common bias towards learning habits. Mol Psychiatry 20: 345–352. Wade CL, Kallupi M, Hernandez DO et al. (2017). Highfrequency stimulation of the subthalamic nucleus blocks compulsive-like re-escalation of heroin taking in rats. Neuropsychopharmacology 42: 1850–1859. Weintraub D, Burn DJ (2011). Parkinson’s disease: the quintessential neuropsychiatric disorder. Mov Disord 26: 1022–1031. Weintraub D, Duda JE, Carlson K et al. (2013). Suicide ideation and behaviours after STN and GPi DBS surgery for Parkinson’s disease: results from a randomised, controlled trial. J Neurol Neurosurg Psychiatry 84: 1113–1118. Welter ML, Burbaud P, Fernandez-Vidal S et al. (2011). Basal ganglia dysfunction in OCD: subthalamic neuronal activity correlates with symptoms severity and predicts high-frequency stimulation efficacy. Transl Psychiatry 1: e5. Welter ML, Schupbach M, Czernecki V et al. (2014). Optimal target localization for subthalamic stimulation in patients with Parkinson disease. Neurology 82: 1352–1361. Whetton S, Shanahan M, Cartwright K et al. (2017). The social costs of methamphetamine in Australia 2013/14, National Drug Research Institute, Curtin University, Perth. Williams A, Gill S, Varma T et al. (2010). Deep brain stimulation plus best medical therapy versus best medical therapy alone for advanced Parkinson’s disease (PD SURG trial): a randomised, open-label trial. Lancet Neurol 9: 581–591. Witt K, Pulkowski U, Herzog J et al. (2004). Deep brain stimulation of the subthalamic nucleus improves cognitive flexibility but impairs response inhibition in Parkinson disease. Arch Neurol 61: 697–700. Witt K, Daniels C, Reiff J et al. (2008). Neuropsychological and psychiatric changes after deep brain stimulation for Parkinson’s disease: a randomised, multicentre study. Lancet Neurol 7: 605–614. Witt K, Granert O, Daniels C et al. (2013). Relation of lead trajectory and electrode position to neuropsychological outcomes of subthalamic neurostimulation in Parkinson’s disease: results from a randomized trial. Brain 136: 2109–2119. Wittmann BC, Daw ND, Seymour B et al. (2008). Striatal activity underlies novelty-based choice in humans. Neuron 58: 967–973. Wodarg F, Herzog J, Reese R et al. (2012). Stimulation site within the MRI-defined STN predicts postoperative motor outcome. Mov Disord 27: 874–879. Wolz M, Hauschild J, Fauser M et al. (2012). Immediate effects of deep brain stimulation of the subthalamic nucleus on nonmotor symptoms in Parkinson’s disease. Parkinsonism Relat Disord 18: 994–997. Yelnik J (2008). Modeling the organization of the basal ganglia. Rev Neurol (Paris) 164: 969–976.

This page intentionally left blank

Handbook of Clinical Neurology, Vol. 180 (3rd series) The Human Hypothalamus: Middle and Posterior Region D.F. Swaab, F. Kreier, P.J. Lucassen, A. Salehi, and R.M. Buijs, Editors https://doi.org/10.1016/B978-0-12-820107-7.00027-6 Copyright © 2021 Elsevier B.V. All rights reserved

Chapter 27

The subthalamic nucleus and the placebo effect in Parkinson’s disease ELISA FRISALDI1, DENISA ADINA ZAMFIRA1, AND FABRIZIO BENEDETTI1,2* 1

Department of Neuroscience, University of Turin Medical School, Turin, Italy 2

Medicine and Physiology of Hypoxia, Plateau Rosà, Switzerland

Abstract The study of the placebo effect, or response, is related to the investigation of the psychologic component of different therapeutic rituals. The high rate of placebo responses in Parkinson’s disease clinical trials provided the impetus for investigating the underlying mechanisms. Ruling out spontaneous remission and regression to the mean through the appropriate experimental designs, genuine psychologic placebo effects have been identified, in which both patients’ expectations of therapeutic benefit and learning processes are involved. Specifically, placebo effects are associated with dopamine release in the striatum and changes in neuronal activity in the subthalamic nucleus, substantia nigra pars reticulata, and motor thalamus in Parkinson’s disease, as assessed through positron emission tomography and single-neuron recording during deep brain stimulation, respectively. Conversely, verbal suggestions of clinical worsening or drug dose reduction induce nocebo responses in Parkinson’s disease, which have been detected at both behavioral and electrophysiologic level. Important implications and applications emerge from this new knowledge. These include better clinical trial designs, whereby patients’ expectations should always be assessed, as well as better drug dosage in order to reduce drug intake.

INTRODUCTION There is today general agreement on the interaction between biologic mechanisms and psychosocial influences, although many of these interactions are still little understood or completely unknown. Indeed, emerging experimental evidence in modern medicine indicates powerful influences of the mind over the body, whereby the patients’ psychologic state and the social context around them are all involved in both the pathophysiology and the treatment outcomes of a given disease. A placebo is usually defined as an inert substance with no pharmacologic action or as a sham physical intervention, although this definition is not complete, as placebos are made of many things, such as words, rituals, symbols, and meanings. A placebo effect, or placebo response, might follow the administration of an inert treatment,

be it pharmacologic or not, and it derives from a positive psychosocial context around the patient and the therapy (Benedetti, 2013; Benedetti, 2014a,b). Moreover, what is emerging today is that not a single placebo effect but many placebo effects exist, which are elicited by different mechanisms across different conditions and different systems (Benedetti, 2013; Benedetti, 2014a,b). The placebo responsiveness of a person, which has determined the distinction between “placebo responder” and “placebo non-responder,” may depend on many factors and on the specific mechanisms involved (Benedetti, 2014b; Benedetti and Frisaldi, 2014; Frisaldi et al., 2018). It is worth noting that not all improvements observed after placebo administration are attributable to real psychobiologic phenomena. In fact, many improvements can be due to different factors, such as the natural history

*Correspondence to: Fabrizio Benedetti, Department of Neuroscience, University of Turin Medical School, Corso Raffaello 30, 10125 Turin, Italy. Tel: +0039-011-6708492, Fax: +39-011-6708174, E-mail: [email protected]

434

E. FRISALDI ET AL.

of the disease, regression to the mean, biases by experimenters and patients, as well as unidentified cointerventions (Benedetti, 2013). Therefore, the placebo effect is approached differently by the clinical trialist and the neuroscientist, because the former is not interested in the cause of the improvement following the administration of the inert substance, whereas the latter is interested only in the psychologic component of different therapeutic rituals which lead to the improvement. Alongside the placebo effect there is also the nocebo effect, its “bad side.” Indeed, a negative context may induce negative expectations of clinical worsening that, in turn, are anxiogenic and lead to negative outcomes. While there has been less research on the nocebo effect historically, mainly due to ethical constraints, the nocebo response is a good model to understand anxiety, particularly anticipatory anxiety, and this may help minimize the harmful effects it can induce (Benedetti, 2013; Benedetti, 2014a,b). The efforts to understand the mechanisms of the interaction between psychosocial and biologic factors are shown by the recent explosion of placebo research. By using many sophisticated neurobiologic approaches, several mechanisms have been identified and many new concepts have emerged (Enck et al., 2008; Price et al., 2008; Finniss et al., 2010; Benedetti, 2013, Benedetti, 2014a,b; Colagiuri et al., 2015; Shaibani et al., 2017). Most of our knowledge about the neurobiologic mechanisms of the placebo effect comes from the field of pain and analgesia (Benedetti, 2013; Benedetti, 2014a,b). It is well established that different endogenous neuronal networks are responsible for the modulation of pain by placebos. In particular, the activation of the descending pain-modulating network from the cerebral cortex to the brainstem and spinal cord has been described, with the involvement of opioid, cannabinoid, cholecystokinin, and dopamine systems (Tracey, 2010; Benedetti, 2014a,b; Carlino et al., 2014). Parkinson’s disease (PD) has also emerged as an excellent model to study the placebo mechanisms for at least three reasons, as extensively described in this chapter. First, the placebo responses in PD have been widely observed and are robust and substantial. Second, the recording from single neurons during implantation of electrodes for deep brain stimulation (DBS) allows us to investigate placebo responses at the level of single neurons. Third, a placebo treatment induces the release of dopamine in the striatum, and this release can be quantified. This neuroscientific approach to placebo studies is giving credibility and compelling evidence to old concepts about the psychosocial influence on the therapeutic outcome. Moreover, by using this neurobiologic approach, the placebo effect represents today an

interesting model to understand how the human brain works (Benedetti, 2014a,b; Wager and Atlas, 2015) and may have profound implications for both medical practice and clinical trials (Benedetti, 2014a,b; Benedetti et al., 2016).

PLACEBO RESPONSE IN PD CLINICAL TRIALS PD is a multisystem disorder characterized by a core of three motor symptoms, which are resting tremor, rigidity, and bradykinesia, and several nonmotor symptoms, including neuropsychiatric problems, cognitive impairment, sleep disturbances, and autonomic dysfunction (Postuma et al., 2015). As for the core symptoms, tremor is at rest and involves mainly the upper limbs, although other body parts may be subject to tremor, such as the chin. Rigidity involves all muscles, with a global impairment of movements and gait. Bradykinesia means that movements slow down, so that any action is performed very slowly and with difficulty (Postuma et al., 2015). The disruption of dopamine function in the neural pathway from the substantia nigra pars compacta to the dorsal striatum (putamen and caudate nucleus) represents the pathophysiologic substrate of PD. The primary deficit involves the selective degeneration of the nigrostriatal dopamine-producing neurons, although at later stages of the disease other dopamine projections and other neurotransmitters may also be affected. There are in fact different dopaminergic pathways in the brain, such as the mesocortical, the mesolimbic, and the tuberoinfundibular. These pathways are associated with volition and emotional responsiveness, desire and reward, sensory processes, and maternal behavior (Obeso et al., 2008; Frisaldi et al., 2014). Dopamine has a critical role in the modulation of the basal ganglia functioning (Alexander et al., 1986; Haber, 2003) and its depletion results in difficulties initiating movement (akinesia), slowness of movement (bradykinesia), rigidity, tremor at rest, and postural instability. The pharmacologic treatment of PD is aimed at replacing the lost dopamine by either dopamine precursors or synthetic agonists acting at dopamine receptors. Substantial improvements in parkinsonian symptoms have been reported in the placebo groups of many clinical trials that assess pharmacologic treatments for PD. In a 24-week-long double-blind trial of dopamineagonist pergolide, significant improvements were seen in both the drug-treated group (30%) and the placebo group (23%) (Diamond et al., 1985). In a review of 36 studies, Shetty et al. (1999) found that 12 of them reported a 9%–59% improvement in motor symptoms following placebo treatment. Goetz et al. (2000) reported that 14% of patients enrolled in a 6-month randomized

THE SUBTHALAMIC NUCLEUS AND THE PLACEBO EFFECT IN PARKINSON'S DISEASE placebo-controlled clinical trial of ropinirole, a dopamineagonist drug currently used for the treatment of PD, achieved a 50% improvement in motor function while on placebo treatment. All domains of parkinsonian disability were subject to the placebo effect, but bradykinesia and rigidity were more susceptible than tremor, gait, or balance. In a trial of deprenyl (selegiline) and tocopherol antioxidative therapy, 21% of patients showed objective improvement in motor function during placebo therapy over 6 months (Goetz et al., 2002). In another study, Goetz et al. (2008 b) examined rates and timing of placebo effects to identify patient- and study-based characteristics predicting positive placebo response in several clinical trials. The authors collected individual patient data from the placebo groups of 11 medical and surgical treatment trials involving PD patients with differing disease severities and placeboassignment probabilities. A positive placebo response was defined as larger than or equal to 50% improvement in total unified Parkinson’s disease rating scale motor (UPDRSm) score or as a decrease by more than or equal to two points on at least two UPDRSm items compared to baseline. Positive placebo response rates were calculated at early (3–7 weeks), mid (8–18 weeks), and late (23–35 weeks) follow-up during the study. A total of 858 patients on placebo groups met inclusion criteria for analysis. The overall placebo response rate was 16% (range: 0–55) and placebo responses were temporally distributed similarly during early, mid, and late phases of follow-up. Age, gender, PD duration, and baseline Hoehn and Yahr stage did not predict positive placebo effects. Conversely, patients with higher baseline UPDRSm scores and studies that focused on PD with motor fluctuations, surgical interventions, or those with a higher probability of placebo assignment showed increased odds of positive placebo response. All these trials were aimed at comparing active treatments with placebos, thus most of them did not include a no-treatment group. Therefore, clinical improvements observed in the placebo groups cannot be attributed to real placebo effects only; spontaneous remission and regression to the mean were likely to take part in the reduction of the symptoms in many cases (Benedetti, 2013, Benedetti, 2014b). None the less, the high rate of placebo effects in PD clinical trials provided the impetus to investigate the placebo responses of parkinsonian patients in more detail and under strictly controlled conditions. In addition, it should be noted that a placebo response in drug trials may reflect not only patients’ expectations, but physicians’ expectations as well. Although little is known in PD, physicians’ expectations are able to influence placebo efficacy well, as documented in pain studies (Gracely et al., 1985).

435

THE SUBTHALAMIC–NIGRAL– THALAMIC CIRCUIT INVOLVED IN THE PLACEBO RESPONSE Surgical therapy of PD, and specifically DBS, has revealed to be a useful clinical setting to investigate placebo effects at the single-neuron level. The subthalamic nucleus (STN) has a central role in basal ganglia functioning and is a major target in the surgical therapy of PD (Limousin et al., 1998). The activity of single neurons in the STN as well as in the surrounding regions can be recorded in different conditions, for example, after the administration of a pharmacologic agent or a placebo. The basal ganglia are a group of subcortical nuclei involved in a variety of processes including motor, associative, cognitive, and mnemonic functions. The dorsal division of the basal ganglia consists of the striatum (caudate-putamen), the external globus pallidus (GPe), the internal segment of globus pallidus (GPi), the subthalamic nucleus, and the substantia nigra. The latter structure is divided into two main parts, the dorsal pars compacta (SNc) in which the dopaminergic nigrostriatal neurons are located and the more ventral pars reticulata (SNr). In addition to these structures, which are associated with motor and associative functions, there is a ventral division of the basal ganglia (ventral striatum or nucleus accumbens, ventral pallidum and ventral tegmental area) that is associated with limbic functions (Bolam et al., 2000; Obeso et al., 2008; Frisaldi et al., 2014). The major input to the basal ganglia is derived from the cerebral cortex. Virtually the whole of the cortical mantle projects the striatum; this cortical information is “processed” within the striatum and passed via the so-called direct and indirect pathways. In the first case, the information is transmitted directly from the striatum to the output nuclei of the basal ganglia, which are the SNr and GPi. In the second case, the information is transmitted indirectly to SNr and GPi, via the complex network interconnecting the STN and GPe. The basal ganglia influence behavior by the projections of these output nuclei to the thalamus and then back to the cortex (Bolam et al., 2000; Obeso et al., 2008; Frisaldi et al., 2014). This is due to the neurochemical nature of neurons in the pathways and their basal activity. Indeed, striatal projection neurons are gamma-aminobutyric acid-ergic (GABAergic) and quiescent under resting conditions; basal ganglia output neurons are also GABAergic but have a high discharge rate, tonically inhibiting the neurons in the ventral thalamus. When the system is activated by the firing of corticostriatal glutamatergic neurons, striatal neurons discharge, which in turn causes inhibition of SNr and GPi. This reduction in firing of

436

E. FRISALDI ET AL.

basal ganglia output neurons leads to release from inhibition, namely disinhibition, of neurons in the thalamus. Conversely, activation of the indirect pathway leads to the opposite physiologic effect, i.e., increased firing of SNr and GPi and increased inhibition of thalamus (Bolam et al., 2000; Obeso et al., 2008; Frisaldi et al., 2014). At the neurophysiologic level, depletion of dopamine in the striatum induces both hyperactivity (high firing rate) (Blandini et al., 2000) and bursting activity (Bergman et al., 1994; Levy et al., 2001) of STN which, in turn, results in an increased neuronal activity in the SNr and GPi that leads to excessive inhibition of the thalamocortical and brainstem motor systems (Obeso et al., 2008; Frisaldi et al., 2014; Fig. 27.1). Several studies have reported that the anti-Parkinson agent, apomorphine, induced a reduction of the STN firing pattern in patients with PD (Lozano et al., 2000; Levy et al., 2001; Stefani et al., 2002). In particular, Stefani et al. (2002) reported that the administration of apomorphine is invariably followed by a reduction of firing rate from about 40 Hz in medication “off” state to about 27 Hz in medication “on” state. Moreover, although the mean firing rate of the STN neurons is a good parameter for

assessing activity of this surgical target, bursting and oscillatory patterns have also been described in PD and related to motor symptoms and to apomorphine effects (Bergman et al., 1994; Levy et al., 2001). The first study looking for a possible placebo effect at the single-neuron level of PD patients was performed in 2004 by Benedetti et al. (2004). The authors performed a double-blind study in which they recorded the activity from single neurons in the STN before and after placebo administration to assess whether neuronal changes were associated with the clinical placebo response. In order to make the placebo response stronger, the placebo was administered in the operating room after several preoperative administrations of the anti-parkinsonian drug apomorphine. As shown in Fig. 27.1, the activity of neurons was recorded from one STN prior to implantation of the first electrode and used as a control. After placebo administration, which consisted of a subcutaneous injection of saline solution along with the verbal suggestion of motor improvement, neuronal activity was recorded from neurons prior to implanting the second electrode into the other subthalamic nucleus. A placebo response was defined as the decrease of arm rigidity of at least one

excitatory

Low firing rate

Neuronal firing increase inhibitory

electrode track

electrode track

Thalamus VA VLa

1s

Thalamus VA VLa

High firing rate STN

Neuronal firing decrease STN

STN single neuron activity

1s

STN single neuron activity

SNr High firing rate

Before placebo administration

SNr Neuronal firing decrease

After placebo administration

Fig. 27.1. Single-neuron placebo effect in Parkinson’s disease at both subthalamic and thalamic levels. The left side shows the classic pathophysiologic mechanisms underlying Parkinson’s disease. The dopamine depletion in the striatum induces hyperactivity and an abnormal bursting activity in the subthalamic nucleus (STN), causing the dysregulation of the basal ganglia circuitry (thick arrows). After placebo administration—right side of the figure—the STN neurons both decrease in firing rate and change from a pattern of bursting activity to a pattern of nonbursting discharge. Restoration of the STN activity leads to normal neuronal firing at the level of both substantia nigra pars reticulata (SNr) and motor thalamus (VA, ventral anterior; VLa, anterior ventral lateral) (thin arrows). Data from Benedetti F, Lanotte M, Colloca L et al. (2009). Electrophysiological properties of thalamic, subthalamic and nigral neurons during the anti-parkinsonian placebo response. J Physiol 587: 3869–3883.

THE SUBTHALAMIC NUCLEUS AND THE PLACEBO EFFECT IN PARKINSON'S DISEASE point on the clinical evaluation scale. Patients who showed a straightforward clinical placebo response, assessed by means of arm rigidity and subjective report of well-being, also showed a significant decrease of firing rate compared to the preplacebo subthalamic nucleus. In order to rule out the possibility that the difference in firing rate between the pre- and postplacebo subthalamic nuclei was independent of the placebo treatment itself, a no-treatment group (natural history) was studied. The patients of this no-treatment group did not undergo any placebo treatment between the implantation of the first and second electrode. All these patients showed no significant differences between the neuronal firing rates of the two subthalamic nuclei, which indicated that the difference between the first and second side of implantation in the placebo group was due to the placebo intervention per se. The same authors (Benedetti et al., 2004) also analyzed the bursting activity of the STN neurons before and after placebo administration in order to evaluate whether, beside the frequency decrease of STN firing rate, there was also a change in the pattern of discharge. They found that the STN neurons of all placebo responders shifted significantly from a pattern of bursting activity to a pattern of nonbursting discharge. None of the placebo nonresponders showed any difference in the number of bursting neurons before and after placebo administration. Likewise, the no-treatment group showed no significant difference in bursting activity between the first and second STN. In the study by Benedetti et al. (2004) there was a clear-cut correlation between subjective reports of the patients’ clinical responses and neurophysiologic responses. In fact, a decrease in firing rate as well as a change from bursting to nonbursting activity were correlated with both the patients’ subjective reports of wellbeing and the muscle rigidity reduction at the wrist, as assessed by a blinded neurologist. Although it is tempting to speculate that these neuronal changes represent a downstream effect of dopamine release in the striatum (see later in the next section), the dopamine release in the striatum (De la Fuente-Fernández et al., 2001) and the single-neuron changes (Benedetti et al., 2004) were observed in two different studies, thus no definitive conclusion can be drawn. Nonetheless, on the basis of our knowledge about the basal ganglia circuitry, it is plausible that placebo-induced release of dopamine acting on the inhibitory D2 receptors disinhibits the gammaaminobutyric acid neurons of the GPe, which, in turn, increase their inhibition onto the STN. In a subsequent study, Benedetti et al. (2009) extended the previous results on single STN neurons to two thalamic nuclei (VA, ventral anterior and VLa, anterior ventral lateral) and the SNr. It is worth remembering that the

437

STN receives inputs from both the cortex and the GPe, and sends excitatory output pathways to both GPi and SNr. Considering the effect of placebo administration on the STN (Benedetti et al., 2004), a significant placebo effect should also be expected in its output regions. Indeed, in parkinsonian patients who exhibited a clinical placebo response, the decrease in firing rate in the STN was associated with a decrease in the SNr and an increase in the thalamic nuclei (Fig. 27.1). Conversely, placebo nonresponders showed either no changes or partial changes in the STN only. Thus the whole subthalamicnigral-thalamic circuit appears to be involved for a clinical placebo response to occur (Frisaldi et al., 2014).

DOPAMINE RELEASE IN VENTRAL AND DORSAL STRIATUM In 2001 De la Fuente-Fernández et al. (2001) conducted the first brain imaging study of the placebo effect, by means of positron emission tomography. In this study, patients were aware that they would be receiving an injection of either active drug (apomorphine, a dopamine receptor agonist) or placebo (an inert substance that the patient believed to be apomorphine), according to classic clinical trials methodology. These researchers assessed the release of endogenous dopamine before and after placebo administration by using the radiotracer raclopride, which binds to dopamine D2 and D3 receptors and competes with endogenous dopamine. After placebo administration it was found that dopamine was released in the striatum, corresponding to a change of 200% or more in extracellular dopamine concentration, comparable to the response to amphetamine in subjects with an intact dopamine system. In addition, the release of dopamine in the motor striatum (putamen and dorsal caudate) was greater in those patients who reported clinical improvement. In the studies by De la Fuente-Fernández et al. (2001, 2002a), all patients showed dopamine placebo responses, yet only half the patients reported motor improvement. These patients also released larger amounts of dopamine in the dorsal motor striatum, suggesting a relationship between the amount of dorsal striatal dopamine release and clinical benefit. This relationship was not present in the ventral striatum, in which all patients showed increased dopamine release, irrespective of whether they perceived any improvement. Compared to the dorsal motor striatum, the ventral striatum (nucleus accumbens) is involved in motivation and reward anticipation (Ikemoto and Panksepp, 1999; Schultz et al., 2000; Schultz, 2002; Knutson and Cooper, 2005). Accordingly, the investigators proposed that the dopamine released in the ventral striatum was associated with patient expectation of improvement in symptoms, which could in turn be considered a form of reward.

438

E. FRISALDI ET AL.

Strafella et al. (2006) confirmed these findings using sham transcranial magnetic stimulation as a placebo. Patients were told they had a 50% chance of receiving either a real or sham treatment, when instead they always received the sham treatment. These authors found that changes in 11C-raclopride binding were greater in the hemisphere contralateral to the more affected side, particularly in the putamen. Although the patients who perceived clinical benefit had a slightly higher amount of dopamine release in the dorsal and ventral striatum, this difference failed to reach statistical significance. In order to determine how the strength of expectation of clinical improvement influences the degree of striatal dopamine release in response to placebo in patients with PD, Lidstone et al. (2010) manipulated patients’ expectations by telling them that they had a probability of 25%, 50%, 75%, or 100% of receiving active medication when they in fact received placebo. Significant dopamine release occurred when the declared probability of receiving active medication was 75%, but not at other probabilities. Placebo-induced dopamine release in all regions of the striatum was also highly correlated with the dopaminergic response to open administration of active medication. Whereas response to prior medication was the major determinant of placebo-induced dopamine release in the motor striatum, expectation of clinical improvement was additionally required to drive dopamine release in the ventral striatum. Therefore, the strength of belief of improvement can directly modulate dopamine release in PD patients, and this emphasizes the importance of uncertainty and/or salience both in clinical practice and in the design of clinical trials. As discussed in the previous section, these changes in dopamine release after placebo administrations are likely to be the cause of the neuronal firing in STN, SNr, and motor thalamus, although it must be recognized that the dopamine studies and the single-neuron studies have been conducted separately and by different authors, thus no definitive conclusion can be drawn on the direct cause–effect relationship between striatum dopamine and STN neuronal activity.

THE ROLE OF PATIENTS’ EXPECTATIONS AND LEARNING By studying PD patients under strictly controlled conditions in order to analyze the very nature of the placebo effect, expectation of clinical benefit has been found to play a key role. In a typical placebo procedure, a placebo is administered along with verbal suggestions of motor improvement. Therefore, patients expect an improvement of their motor symptoms, such as tremor, muscle rigidity, and bradykinesia.

In one study (Pollo et al., 2002), the velocity of movements was analyzed in PD patients who had been implanted with electrodes in the STN for DBS surgery. They were tested in two opposite conditions. In the first, they expected a good motor performance; in the second, they expected a bad motor performance. The effect of STN DBS on the velocity of movement of the right hand was analyzed by means of a movement analyzer. Patients performed a visual directional-choice task on a rectangular surface, with their right index finger positioned on a central sensor with a green light. After a random interval of a few seconds, a red light turned on randomly in one of three sensors placed 10 cm away from the green light sensor. Patients were instructed to move their hand as quickly as possible in order to reach the target red light sensor. The hand movement was found to be faster when patients expected a good motor performance than when they expected a bad performance. Interestingly, all these effects occurred within minutes, which indicates that expectations induce neural changes very quickly. In another study by Benedetti et al. (2003), patients implanted for DBS were tested for the velocity of movement of their right hand according to a double-blind experimental design in which neither the patient nor the experimenter knew if the stimulator was turned off. The velocity of hand movement was assessed with a movement analyzer. The stimulator was turned off several times (at 4 and 2 weeks) before the test session. Each time, the velocity of movement was measured just before the stimulator was turned off and 30 min later. On the day of the experimental session, the stimulator was kept on, but patients were told it had been turned off, so as to induce negative expectations of motor performance worsening (nocebo procedure). Although the stimulator was on, motor performance worsened and mimicked the worsening of the previous days. This nocebo bradykinesia could be completely prevented by verbal suggestions of good motor performance (placebo procedure). Therefore, motor performance can be modulated in two opposite directions by placebos and nocebos, and this modulation occurs on the basis of positive and negative expectations. These findings have been reproduced by Mercado et al. (2006), who also found different effects of expectations for tremor, rigidity, and bradykinesia; in fact, these authors found significant effects for bradykinesia, but not for tremor and rigidity. On the basis of the work done by Pollo et al. (2002), Benedetti et al. (2003), and Mercado et al. (2006), bradykinesia seems to be a symptom that is more sensitive to verbal suggestions than tremor or rigidity. Interestingly, these results are similar to those obtained by Goetz et al. (2000) who reported that all domains of parkinsonism were subject to the placebo effect, but bradykinesia and rigidity were more susceptible than tremor, gait, or balance.

THE SUBTHALAMIC NUCLEUS AND THE PLACEBO EFFECT IN PARKINSON'S DISEASE Expectations of motor improvement can also make a big difference in clinical trials. In a clinical trial of human fetal mesencephalic transplantation, the investigators studied the effect of this treatment compared with placebo for 12 months. They also assessed the patient’s perceived assignment to either the active (fetal tissue implant) or placebo treatment (sham surgery). There were no differences between the transplant and sham surgery groups on several outcome measures, such as physical and quality of life scores. However, the perceived assignment of treatment group had a beneficial impact on the overall outcome and this difference was still present 12 months after surgery. Patients who believed they received transplanted tissue had significant improvements in both their quality of life and motor outcomes, regardless of whether they received sham surgery or fetal tissue implantations (McRae et al., 2004). Expectations have also been found to change cortical excitability in parkinsonian patients. Lou et al. (2013) randomized 26 PD patients to one of three groups: 0%, 50%, and 100% expectation of receiving levodopa. All subjects received placebo regardless of expectation group. Cortical excitability was measured by amplitude of the motor-evoked potentials induced by transcranial magnetic stimulation. The degree of expectation had a significant effect on motor-evoked potential response: subjects in the 50% and 100% expectation groups responded with a decrease in motor-evoked potentials, whereas those in the 0% expectation group responded with an increase. Besides motor functioning, expectation affects cognitive functions as well. Keitel et al. (2013) investigated how expectation modulates the pattern of motor improvement in PD patients treated with DBS of the STN and its interaction with verbal fluency. Expectations of 24 hypokinetic-rigid parkinsonian patients about the impact of DBS on motor symptoms were manipulated by positive (placebo), negative (nocebo), and neutral (control) verbal suggestions. It was found that expectations significantly affected proximal but not distal movements resulting in better performance in the placebo than in the nocebo condition. Placebo responders with improvement larger than or equal to 25% were characterized by a trend for impaired lexical verbal fluency, a frequent side effect of the STN stimulation. Therefore, in this study positive motor expectations exerted both motor placebo and cognitive nocebo responses.

TEACHING NEURONS TO RESPOND TO PLACEBOS Although patients’ expectations are today recognized as a major mediator of placebo responses, recent research suggests that learning is even more important. In 2016,

439

Benedetti et al. (2016) showed that placebos given for the first time to naïve patients induce neither clinical nor neuronal improvement in PD patients who underwent the implantation of electrodes for DBS. However, this lack of placebo responsiveness could be turned into substantial placebo responses following previous exposure to repeated administrations of the anti-Parkinson agent apomorphine. As the number of apomorphine administrations increased from one through four, both the clinical response and the neuronal activity in two of the motor thalamic nuclei (ventral anterior, VA, and anterior ventral lateral, VLa) increased. In fact, after four apomorphine exposures, placebo administration induced clinical responses that were as large as those to apomorphine, along with long-lasting neuronal changes. These clinical placebo responses following four apomorphine administrations were again elicited after a reexposure to a placebo 24 h postsurgery, but not after 48 h (Fig. 27.2). As described for muscle rigidity (Benedetti et al., 2016), in 2017 Frisaldi et al. (2017) investigated the effects of prior exposure to apomorphine on the placebo response for bradykinesia by means of a movement time analyzer. No placebo response was found if the placebo was given for the first time, whereas the placebo response was substantial after prior pharmacologic conditioning with apomorphine. Overall, these findings indicate that prior exposure to drugs is a critical factor in the occurrence and magnitude of placebo effects. This may shed light on a possible source of high placebo responses in PD clinical trials, which can often be related to learning effects, that is, to previous exposure to pharmacologic agents. A conditioning procedure in which a drug– drug-placebo-drug–drug-placebo protocol is used may be exploited to reduce drug intake. On the other hand, within the clinical trial setting, enriched designs, which entail previous exposure to drugs, should be avoided because of the possible induction of high placebo responses.

THE ROLE OF VERBAL INSTRUCTIONS IN THE MODULATION OF MOTOR PERFORMANCE AND FATIGUE Changing the dose of a drug is common practice in the clinical setting. Usually, drug doses can be either increased, if the therapeutic outcome is deemed to be unsatisfactory, or decreased, if one wants to reduce adverse events. For example, chronic use of L-dopa for the treatment of PD may lead to motor complications such as dyskinesia, which can worsen the quality of life of Parkinson patients and can also undermine the L-dopa therapy itself. Classic strategies to decrease these adverse events may imply the reduction of the single dose and/or

Timeline (days) SURGERY DAYS UPDRS

–1

0.5 0

2.5 2 1.5 1

PLACEBO

0.5 0

2.5 2 1.5 1

min

min

min

min

APO

APO

APO

APO

0.5 0 0

5 15 30 45 60

min

0

5 15 30 45 60

min

0 5 15 30 45 60 0

min

5 15 30 45 60

min

5 15 30 45 60

0

min

Improvement (rigidity score)

Improvement (rigidity score)

0

150 120 90 60 30 0

PLACEBO

2.5 2 1.5 1 0.5 0 0

5 15 30 45 60

min

2.5 2 1.5 1

PLACEBO

0

min

150 120 90 60 30 0 0

5 15 30 45 60

min

2

PLACEBO

0.5 0

5 15 30 45 60

5 15 30 45 60 0

min

Improvement (rigidity score)

2.5 2 1.5 1

1

SNR

Change in neuron activity (%)

–2

Change in neuron activity (%)

–3

Improvement (rigidity score)

Improvement (rigidity score)

–4

POST-SURGERY DAYS

Improvement (rigidity score)

PRE-SURGERY DAYS

2.5 2 1.5 1

PLACEBO

5 15 30 45 60

min

PLACEBO

0.5 0 0

5 15 30 45 60 0

min

5 15 30 45 60

min

Fig. 27.2. Clinical and electrophysiologic placebo responses induced by apomorphine preconditioning. When a placebo is given for the first time, it induces neither clinical improvements nor neuronal changes in motor thalamic neurons (VA, ventral anterior; VLa, anterior ventral lateral) of parkinsonian patients who undergo deep brain stimulation. Conversely, a placebo administration preceded by four apomorphine preconditioning trials is able to elicit clinical responses similar to those induced by apomorphine, along with neuronal changes, and clinical improvements lasting up to 24 h after surgery. APO, apomorphine; SNR, single neuron recording from motor thalamus; UPDRS, unified Parkinson’s disease rating scale. *P < 0.05. Data from Benedetti F, Frisaldi E, Carlino E et al. (2016). Teaching neurons to respond to placebos. J Physiol 594: 5647–5660.

Get full dose Told full dose

UPDRS-III

Get half dose Told half dose

Increase

Number of flexions Readiness potential Fatigue

Decrease

UPDRS-III Fatigue Readiness potential

Increase

Number of flexions

Decrease

Decrease

Increase

THE SUBTHALAMIC NUCLEUS AND THE PLACEBO EFFECT IN PARKINSON'S DISEASE

441

Number of flexions

UPDRS-III Fatigue Readiness potential

Get half dose Told full dose

Fig. 27.3. Verbal communication about drug dosage modulates motor performance and fatigue in Parkinson’s disease patients. Halving L-dopa along with correct information about its reduction (Get half dose/Told half dose) leads to a small clinical improvement (UPDRS-III). Halving L-dopa deceptively (Get half dose/Told full dose) produces significant improvements in clinical (UPDRS-III), motor (number of flexions and perception of fatigue), and electrophysiologic (readiness potential) assessments, which are undistinguishable from those elicited by the full standard dose (Get full dose/Told full dose). UPDRS-III, unified Parkinson’s disease rating scale, part III. Data from Carlino E, Piedimonte A, Romagnolo A et al. (2019). Verbal communication about drug dosage balances drug reduction in Parkinson’s disease: behavioral and electrophysiological evidences. Parkinsonism Relat Disord 65: 184–189.

the frequency of administration of L-dopa, the combination with other pharmacological agents, or even the use of advanced surgical treatments. On the basis of the important role of expectations in the therapeutic outcome, the effects of verbal instructions on dose change of the widely used anti-Parkinson agent, L-dopa, were recently assessed. In fact, Carlino et al. (2019) performed clinical (UPDRS), motor (number of finger flexions and perceived fatigue), and electrophysiological measurements (readiness potential, RP) in Parkinson patients during their medication-off and medication-on conditions in three groups. Whereas the first group got a full dose of L-dopa and was told it was a full dose, the second group got half dose and was told it was half dose, as in routine clinical practice. However, the third group got half dose, but it was told it was a full standard dose, thus the patients of this group did not know that the L-dopa dose had been halved. Whereas, as expected, the overt half dose was less effective than the full dose for clinical improvement, motor performance, as well as the RP

(Fig. 27.3), the covert half dose (patients get half dose but are told it is a full dose) induced clinical improvement, motor performance, and RP (Fig. 27.3) that were not different from the full dose. These findings indicate that verbal communication about dose change can be as powerful as half dose of L-dopa. In other words, in this study, positive verbal communication, actually a deception, compensated the 50% dose change itself. This psychologic effect involves the supplementary motor area (SMA), the main source of the RP. It is important to remember that RP is a movement-related negative potential that is recorded over the human scalp about 2 s before a self-paced motor act (Shibasaki and Hallett, 2006). This slow potential is mainly generated by areas linked to motor preparation, such as SMA, and motor execution, such as the primary motor cortex (Deecke, 1996; Shibasaki and Hallett, 2006). In the absence of fatigue, its amplitude is related to the amount of voluntary force and perceived effort; in the presence of fatigue, its amplitude increases along with the increase in fatigue (Freude and Ullsperger, 1987).

442

E. FRISALDI ET AL.

Three important implications emerge from this study. First, the study of RP may represent an interesting approach to better understand the neural networks, besides STN and the basal ganglia, that are involved in the placebo effect in PD. Second, verbal communication in pharmacotherapy may have an important impact on the therapeutic outcome. Third, an ethical discussion is needed in order to understand whether this deceptive administration can be used to the patient’s advantage.

CONCLUSIONS AND FUTURE DIRECTIONS Differently from the past, placebos are no longer a nuisance in clinical research, but they represent an interesting target of scientific inquiry. Besides their understanding as a psychologic and neuroscientific phenomenon, which helps us clarify different brain functions, the substantial neurobiologic advances of the placebo effect are essential for disentangling drug effects from placebo effects and may allow clinicians to maximize therapeutic outcomes. In addition, the understanding of the neural network that is involved in PD, particularly the STN circuitry, helps us understand how the human brain works both in health and in pathologic conditions.

REFERENCES Alexander GE, DeLong MR, Strick PL (1986). Parallel organization of functionally segregated circuits linking basal ganglia and cortex. Annu Rev Neurosci 9: 357–381. Benedetti F (2013). Placebo and the new physiology of the doctor-patient relationship. Physiol Rev 93: 1207–1246. Benedetti F (2014a). Placebo effects: from the neurobiological paradigm to translational implications. Neuron 84: 623–637. Benedetti F (2014b). Placebo effects, second edn. Oxford University Press, Oxford. Benedetti F, Frisaldi E (2014). Creating placebo responders and nonresponders in the laboratory: boons and banes. Pain Manag 4: 165–167. Benedetti F, Pollo A, Lopiano L (2003). Conscious expectation and unconscious conditioning in analgesic, motor, and hormonal placebo/nocebo responses. J Neurosci 23: 4315–4323. Benedetti F, Colloca L, Torre E et al. (2004). Placeboresponsive Parkinson patients show decreased activity in single neurons of subthalamic nucleus. Nat Neurosci 7: 587–588. Benedetti F, Lanotte M, Colloca L et al. (2009). Electrophysiological properties of thalamic, subthalamic and nigral neurons during the anti-parkinsonian placebo response. J Physiol 587: 3869–3883. Benedetti F, Frisaldi E, Carlino E et al. (2016). Teaching neurons to respond to placebos. J Physiol 594: 5647–5660. Bergman H, Wichmann T, Karmon B et al. (1994). The primate subthalamic nucleus. II Neuronal activity in the MPTP model of parkinsonism. J Neurophysiol 72: 507–520.

Blandini F, Nappi G, Tassorelli C et al. (2000). Functional changes of the basal ganglia circuitry in Parkinson’s disease. Prog Neurobiol 62: 63–88. Bolam JP, Hanley JJ, Booth PA et al. (2000). Synaptic organisation of the basal ganglia. J Anat 196: 527–542. Carlino E, Frisaldi E, Benedetti F (2014). Pain and the context. Nat Rev Rheumatol 10: 348–355. Carlino E, Piedimonte A, Romagnolo A et al. (2019). Verbal communication about drug dosage balances drug reduction in Parkinson’s disease: behavioral and electrophysiological evidences. Parkinsonism Relat Disord 65: 184–189. Colagiuri B, Schenk LA, Kessler MD et al. (2015). The placebo effect: from concepts to genes. Neuroscience 307: 171–190. De la Fuente-Ferna´ndez R, Ruth TJ, Sossi V et al. (2001). Expectation and dopamine release: mechanism of the placebo effect in Parkinson’s disease. Science 293: 1164–1166. De la Fuente-Ferna´ndez R, Phillips AG, Zamburlini M et al. (2002a). Dopamine release in human ventral striatum and expectation of reward. Behav Brain Res 136: 359–363. Deecke L (1996). Planning, preparation, execution, and imagery of volitional action. Cogn Brain Res 3: 59–64. Diamond SG, Markham CH, Treciokas LJ (1985). Doubleblind trial of pergolide for Parkinson’s disease. Neurology 35: 291–295. Enck P, Benedetti F, Schedlowski M (2008). New insights into the placebo and nocebo responses. Neuron 59: 195–206. Finniss DG, Kaptchuk TJ, Miller F et al. (2010). Biological, clinical, and ethical advances of placebo effects. Lancet 375: 686–695. Freude G, Ullsperger P (1987). Changes in Bereitschaftspotential during fatiguing and non-fatiguing hand movements. Eur J Appl Physiol Occup Physiol 56: 105–108. Frisaldi E, Carlino E, Lanotte M et al. (2014). Characterization of the thalamic-subthalamic circuit involved in the placebo response through single-neuron recording in Parkinson patients. Cortex 60: 3–9. Frisaldi E, Carlino E, Zibetti M et al. (2017). The placebo effect on bradykinesia in Parkinson’s disease with and without prior drug conditioning. Mov Disord 32: 1474–1478. Frisaldi E, Shaibani A, Benedetti F (2018). Placebo responders and nonresponders: what’s new? Pain Manag 8: 405–408. Goetz CG, Leurgans S, Raman R et al. (2000). Objective changes in motor function during placebo treatment in PD. Neurology 54: 710–714. Goetz CG, Leurgans S, Raman R (2002). Placebo-associated improvements in motor function: comparison of subjective and objective sections of the UPDRS in early Parkinson’s disease. Mov Disord 17: 283–288. Goetz CG, Wuu J, McDermott M et al. (2008b). Placebo response in Parkinson’s disease: comparisons among 11 trials covering medical and surgical interventions. Mov Disord 23: 690–699. Gracely RH, Dubner R, Deeter WD et al. (1985). Clinicians’ expectations influence placebo analgesia. Lancet 1: 43.

THE SUBTHALAMIC NUCLEUS AND THE PLACEBO EFFECT IN PARKINSON'S DISEASE Haber SN (2003). The primate basal ganglia: parallel and integrative networks. J Chem Neuroanat 26: 317–330. Ikemoto S, Panksepp J (1999). The role of nucleus accumbens dopamine in motivated behavior: a unifying interpretation with special reference to reward-seeking. Brain Res Rev 31: 6–41. Keitel A, Wojtecki L, Hirschmann J et al. (2013). Motor and cognitive placebo /nocebo-responses in Parkinson’s disease patients with deep brain stimulation. Behav Brain Res 250: 199–205. Knutson B, Cooper JC (2005). Functional magnetic resonance imaging of reward prediction. Curr Opin Neurol 18: 411–417. Levy R, Dostrovskym JO, Lang AE et al. (2001). Effects of apomorphine on subthalamic nucleus and globus pallidus internus neurons in patients with Parkinson’s disease. J Neurophysiol 86: 249–260. Lidstone SC, Schulzer M, Dinelle K et al. (2010). Effects of expectation on placebo-induced dopamine release in Parkinson disease. Arch Gen Psychiatry 67: 857–865. Limousin P, Krack P, Pollak P et al. (1998). Electrical stimulation of the subthalamic nucleus in advanced Parkinson’s disease. N Engl J Med 339: 1105–1111. Lou JS, Dimitrova DM, Hammerschlag R et al. (2013). Effect of expectancy and personality on cortical excitability in Parkinson’s disease. Mov Disord 28: 1257–1262. Lozano AM, Lang AE, Levy R et al. (2000). Neuronal recordings in Parkinson’s disease patients with dyskinesias induced by apomorphine. Ann Neurol 47: S141–S146. McRae C, Cherin E, Yamazaki TG et al. (2004). Effects of perceived treatment on quality of life and medical outcomes in a double-blind placebo surgery trial. Arch Gen Psychiatry 61: 412–420. Mercado R, Constantoyannis C, Mandat T et al. (2006). Expectation and the placebo effect in Parkinson’s disease patients with subthalamic nucleus deep brain stimulation. Mov Disord 21: 1457–1461. Obeso JA, Rodrı´guez-Oroz MC, Benitez-Temino B et al. (2008). Functional organization of the basal ganglia: therapeutic implications for Parkinson’s disease. Mov Disord 23: S548–S559. Pollo A, Torre E, Lopiano L et al. (2002). Expectation modulates the response to subthalamic nucleus stimulation in parkinsonian patients. Neuroreport 13: 1383–1386. Postuma RB, Berg D, Stern M et al. (2015). MDS clinical diagnostic criteria for Parkinson’s disease. Mov Disord 30: 1591–1601. Price DD, Finniss DG, Benedetti F (2008). A comprehensive review of the placebo effect: recent advances and current thought. Annu Rev Psychol 59: 565–590. Schultz W (2002). Getting formal with dopamine and reward. Neuron 36: 241–263. Schultz W, Tremblay L, Hollerman JR (2000). Reward processing in primate orbitofrontal cortex and basal ganglia. Cereb Cortex 10: 272–284. Shaibani A, Frisaldi E, Benedetti F (2017). Placebo response in pain, fatigue, and performance: possible implications for neuromuscular disorders. Muscle Nerve 56: 358–367.

443

Shetty N, Friedman JH, Kieburtz K et al. (1999). The placebo response in Parkinson’s disease. Parkinson Study Group. Clin Neuropharmacol 22: 207–212. Shibasaki H, Hallett M (2006). What is the Bereitschaftspotential? Clin Neurophysiol 117: 2341–2356. Stefani A, Bassi A, Mazzone P et al. (2002). Subdyskinetic apomorphine responses in globus pallidus and subthalamus of parkinsonian patients: lack of clear evidence for the “indirect pathway”. Clin Neurophysiol 113: 91–100. Strafella AP, Ko JH, Monchi O (2006). Therapeutic application of transcranial magnetic stimulation in Parkinson’s disease: the contribution of expectation. Neuroimage 31: 1666–1672. Tracey I (2010). Getting the pain you expect: mechanisms of placebo, nocebo and reappraisal effects in humans. Nat Med 16: 1277–1283. Wager TD, Atlas LY (2015). The neuroscience of placebo effects: connecting context, learning and health. Nat Rev Neurosci 16: 403–418.

FURTHER READING Benedetti F, Amanzio M, Vighetti S et al. (2006). The biochemical and neuroendocrine bases of the hyperalgesic nocebo effect. J Neurosci 26: 12014–12022. Benedetti F, Amanzio M, Rosato R et al. (2011a). Nonopioid placebo analgesia is mediated by CB1 cannabinoid receptors. Nat Med 17: 1228–1230. Benedetti F, Carlino E, Pollo A (2011b). Hidden administration of drugs. Clin Pharmacol Ther 90: 651–661. Benedetti F, Carlino E, Pollo A (2011c). How placebos change the patient’s brain. Neuropsychopharmacology 36: 339–354. Brody H (2000). The placebo response, Harper Collins, New York. Colloca L, Benedetti F (2005). Placebos and painkillers: is mind as real as matter? Nat Rev Neurosci 6: 545–552. Colloca L, Lopiano L, Lanotte M et al. (2004). Overt versus covert treatment for pain, anxiety, and Parkinson’s disease. Lancet Neurol 3: 679–684. De la Fuente-Ferna´ndez R, Schulzer M, Stoessl AJ (2002b). The placebo effect in neurological disorders. Lancet Neurol 1: 85–91. Diederich NJ, Goetz CG (2008). The placebo treatments in neurosciences: new insights from clinical and neuroimaging studies. Neurology 71: 677–684. Galpern WR, Corrigan-Curay J, Lang AE et al. (2012). Sham neurosurgical procedures in clinical trials for neurodegenerative diseases: scientific and ethical considerations. Lancet Neurol 11: 643–650. Gill SS, Patel NK, Hotton GR et al. (2003). Direct brain infusion of glial cell line-derived neurotrophic factor in Parkinson disease. Nat Med 9: 589–595. Goetz CG, Laska E, Hicking C et al. (2008a). Placebo influences on dyskinesia in Parkinson’s disease. Mov Disord 23: 700–707. Gross RE, Watts RL, Hauser RA et al. (2011). Intrastriatal transplantation of microcarrier-bound human retinal pigment epithelial cells versus sham surgery in patients with

444

E. FRISALDI ET AL.

advanced Parkinson’s disease: a double-blind, randomised, controlled trial. Lancet Neurol 10: 509–519. Hutchison WD, Allan RJ, Opitz H et al. (1998). Neurophysiological identification of the subthalamic nucleus in surgery for Parkinson’s disease. Ann Neurol 44: 622–628. Lang AE, Gill S, Patel NK et al. (2006). Randomized controlled trial of intraputamenal glial cell line-derived neurotrophic factor infusion in Parkinson disease. Ann Neurol 59: 459–466. Limousin P, Greene J, Pollak P et al. (1997). Changes in cerebral activity pattern due to subthalamic nucleus or internal pallidum stimulation in Parkinson’s disease. Ann Neurol 42: 283–291. Marks Jr WJ, Ostrem JL, Verhagen L et al. (2008). Safety and tolerability of intraputaminal delivery of CERE-120 (adeno-associated virus serotype 2eneurturin) to patients with idiopathic Parkinson’s disease: an open-label, phase I trial. Lancet Neurol 7: 400–408. Marks Jr WJ, Bartus RT, Siffert J et al. (2010). Gene delivery of AAV2-neurturin for Parkinson’s disease: a double-blind, randomised, controlled trial. Lancet Neurol 9: 1164–1172. Olanow CW, Goetz CG, Kordower JH et al. (2003). A doubleblind controlled trial of bilateral fetal nigral transplantation in Parkinson’s disease. Ann Neurol 54: 403–414. Parent A, Hazrati LN (1995). Functional anatomy of the basal ganglia. II. The place of subthalamic nucleus and external

pallidum in basal ganglia circuitry. Brain Res Rev 20: 128–154. P
eron J, Fr€ uhholz S, V
erin M et al. (2013). Subthalamic nucleus: a key structure for emotional component synchronization in humans. Neurosci Biobehav Rev 37: 358–373. Petrovic P, Dietrich T, Fransson P et al. (2005). Placebo in emotional processing e induced expectations of anxiety relief activate a generalized modulatory network. Neuron 46: 957–969. Scherer KR, Schorr A, Johnstone T (2001). Appraisal processes in emotion: theory, methods, research, Oxford University Press, New York. Scott DJ, Stohler CS, Egnatuk CM et al. (2007). Individual differences in reward responding explain placebo-induced expectations and effects. Neuron 55: 325–336. Slevin JT, Gerhardt GA, Smith CD et al. (2005). Improvement of bilateral motor functions in patients with Parkinson disease through the unilateral intraputaminal infusion of glial cell line-derived neurotrophic factor. J Neurosurg 102: 216–222. Stover NP, Bakay RAE, Subramanian T et al. (2005). Intrastriatal implantation of human retinal pigment epithelial cells attached to microcarriers in advanced Parkinson disease. Arch Neurol 62: 1833–1837. Vase L, Robinson ME, Verne GN et al. (2005). Increased placebo analgesia over time in irritable bowel syndrome (IBS) patients is associated with desire and expectation but not endogenous opioid mechanisms. Pain 115: 338–347.

Section 14 Corpora mamillaria, fornix, and mamillothalamic tract

This page intentionally left blank

Handbook of Clinical Neurology, Vol. 180 (3rd series) The Human Hypothalamus: Middle and Posterior Region D.F. Swaab, F. Kreier, P.J. Lucassen, A. Salehi, and R.M. Buijs, Editors https://doi.org/10.1016/B978-0-12-820107-7.00028-8 Copyright © 2021 Elsevier B.V. All rights reserved

Chapter 28

Electrical stimulation of the fornix for the treatment of brain diseases SARAH HESCHAM* AND YASIN TEMEL Department of Neurosurgery, Maastricht University Medical Center, Maastricht, The Netherlands

Abstract Deep brain stimulation (DBS) has proven to be safe and effective for both hypo- and hyperkinetic movement disorders of basal ganglia origin, while its application to other neural pathways such as the circuit of Papez is under investigation. In particular, the fornix has gained interest as potential DBS target to decrease rates of cognitive decline, enhance memory, aid visuospatial memorization, and improve verbal recollection. While the exact mechanisms of action of fornix DBS are not completely understood, studies found enhanced hippocampal acetylcholine release, synaptic plasticity, and decreased inflammatory responses in cortex and hippocampus. Nevertheless, it is still premature to conclude that fornix DBS can be used in the treatment of cognitive disorders, and the field needs sound, preclinically tested, and disease-specific a posteriori hypotheses.

INTRODUCTION The use of deep brain stimulation (DBS) to modulate pathological neural circuits has changed the way that brain disorders are treated and understood. Over 160,000 patients worldwide have undergone DBS surgery and the numbers are increasing each year (Lozano and Lipsman, 2013). In particular, DBS within the basal ganglia network has proven to be safe and effective for movement disorders, whereas the application of DBS to modulate different neural pathways such as the circuit of Papez is considered experimental. The circuit of Papez is one of the major pathways of the limbic system and is primarily involved in emotional expression, neurovegetative function, and memory (Papez, 1995). The classic circuit consists of the hippocampal formation, fornix, mamillary bodies, mamillothalamic tract, anterior thalamic nucleus, cingulum, and the entorhinal cortex (Papez, 1937). Damage to structures within the circuit of Papez can cause

anterograde amnesia in patients, i.e., an inability to create new episodic memories (Clarke et al., 1994; Harding et al., 2000; Hildebrandt et al., 2001; McDonald et al., 2001; Aggleton et al., 2005). While DBS applied to limbic targets has been evaluated for patients with treatment-resistant depression (Lozano et al., 2008; Malone et al., 2009; Bewernick et al., 2010) and obsessive–compulsive disorder (Denys et al., 2010), recently studies have begun to explore the applicability of DBS in a widening array of conditions such as Alzheimer’s disease (AD) and temporal lobe epilepsy (TLE). DBS of the fornix is thought to facilitate neuromodulation within memoryassociated areas of the Papez circuitry through anterograde and retrograde axonal stimulation. In this light, fornix DBS has shown to impact long-term structural plasticity as well as neurotransmitter release in memory-related brain areas (Gondard et al., 2015; Sankar et al., 2015; Hescham et al., 2016).

*Correspondence to: Sarah Hescham, Ph.D., Department of Neurosurgery, Maastricht University, Universiteitssingel 50, Maastricht 6229 ER, The Netherlands. Tel: +31-43-388-1263, Fax: +31-43-387-6038. E-mail: [email protected]

448

S. HESCHAM AND Y. TEMEL

In this chapter, we focus on the use of fornix DBS for the treatment of brain diseases. For this, we discuss preclinical and clinical studies and point out their potential within future research.

WHAT IS DBS? DBS is a neurosurgical procedure in which stimulation electrodes are stereotactically implanted into specific brain targets and deliver constant or intermittent electric pulses from an implanted battery source, the pacemaker. The pacemaker and the settings can be accessed externally with a wireless connected controller. As a clinical tool, DBS offers several advantages over other surgical approaches for neuromodulation, such as an ablative procedure. Advantages include the reversible nature of DBS, the possibility to adjust stimulation parameters to maximize therapeutic benefit with no or little side effects, and the capacity to directly target circuitopathies with relatively high precision (Lozano et al., 2019). With regard to the latter, DBS has contributed to circuit theories of brain dysfunction by demonstrating that a highly focused site of neuromodulation can have profound influences on brain-wide networks (Lozano et al., 2019). From a scientific point of view, DBS has helped us to understand more of the physiological causes of brain diseases, which in turn enhance clinical outcomes and help to drive technological innovation (Hamani and Temel, 2012). Notwithstanding its advantages, DBS remains an invasive surgical intervention and severe adverse effects such as intracerebral hemorrhages occur in 1%–2% of patients, while less severe or reversible events such as infections, lead, and pulse generator problems occur in a vast minority of the patients (Lozano and Lipsman, 2013).

WHY CONSIDER FORNIX DBS FOR BRAIN DISEASES? The fornix constitutes the main efferent pathway of the hippocampus, connecting the hippocampal formation to the mamillary bodies, anterior thalamic nuclei, and prefrontal cortex. A thin layer of efferent fibers known as the alveus that mainly ascend from the pyramidal cells of the hippocampus form a fringe of fibers known as the fimbria, which represents the initial part of the fornix. Underneath the splenium of the corpus callosum, the white matter of the fimbria separates from the hippocampal formation and becomes the crus of the fornix. The left and right crura then unite in the midline below the trunk of the corpus callosum to form the body of the fornix. At the level of the foramen of Monro, the fornix divides again laterally forming the columns of the fornix. Most of the fibers of each column constitute

the postcommissural fornix (Christiansen et al., 2016). These fibers originate from the subiculum of the hippocampus and pass posterior to the anterior commissure to reach the mamillary bodies. The precommissural fornix encompasses the remaining fibers, which mainly originate from the pyramidal cell layer of the hippocampus but also from the entorhinal cortex and subiculum (Christiansen et al., 2016). The precommissural fornix passes anterior to the commissure and reaches the septal nuclei and basal forebrain, the ventral striatum, and the prefrontal cortex. It also contains projection fibers from the septum that reach the hippocampus and, therefore, the fornix embraces both projection and commissural fibers (Thomas et al., 2011). An anatomic illustration of the fornix can be found in Fig. 28.1. The fornix is an integral part of the classic Papez circuit and is thus imperative to the formation and consolidation of memory in rodents and primates (Thomas et al., 2011). It is known that lesions of the fornix lead to various amnestic syndromes (Sankar et al., 2014). As studies in the field of Parkinson’s disease have found, DBS produces downstream circuit effects on signaling in connected structures (Hashimoto et al., 2003). It is therefore believed that by targeting the fornix, the broader circuitry of memory integration may be modulated thereby improving memory and cognitive symptomatology.

STUDIES ON FORNIX DBS We will discuss selected studies that examined the effects of fornix DBS in two brain diseases, AD and TLE.

Alzheimer’s disease AD is the most prevalent form of dementia. It is characterized by various pathological processes including regionally specific and sequential brain atrophy, amyloid deposition, neurofibrillary tangles, synaptic dysfunction, and neuronal cell death, which are often accompanied by psychiatric symptomatology, deterioration of functional ability, personality changes, and a general decline of quality of life (Ballard et al., 2011; Hescham et al., 2013a). The inability to acquire new memories indicates the early stages of the disease, whereas in later stages, patients suffer from agnosia, aphasia, and apraxia and long-term memory loss (Thies and Bleiler, 2013). Disease-modifying treatment strategies for AD are still under extensive research. For the typical AD patient, current symptomatic therapies (acetylcholinesterase inhibitors and memantine) demonstrate only minimal to modest symptomatic benefit that is not sustained (Rosini et al., 2008; Alzheimer’s Association, 2011). Motivated by this reality, researchers are currently exploring new neuromodulatory techniques like DBS.

FORNIX DBS FOR THE TREATMENT OF BRAIN DISEASES

449

Fig. 28.1. Simplified illustration of the anatomical targeting for fornix deep brain stimulation in clinical studies. The fornix (F) and the hippocampus (H) are depicted in yellow. Efferent fibers of the hippocampus known as the alveus join together to form the fimbria. Beneath the splenium of the corpus callosum, the fimbria separates from the hippocampus and becomes the crus of the fornix. The left and right crura then converge to form the body of the fornix. The body of the fornix travels anteriorly and divides again near the anterior commissure. The left and right parts separate into the anterior pillars, but there is also an anterior/ posterior divergence. The posterior fibers (called the postcommissural fornix) of each side continue through the hypothalamus to the mamillary bodies. The anterior fibers (precommissural fornix) end at the septal nuclei and nucleus accumbens of each hemisphere. (A) Sagittal view of fornix DBS electrode location. (B) Frontal view of fornix DBS electrode location in one hemisphere. Adapted from Liu H, Temel Y, Boonstra J et al. (2020). The effect of fornix deep brain stimulation in brain diseases. Cell Mol Life Sci 77: 3279–3291.

In AD, pathological changes can be found throughout the brain but with a predilection for the neuronal sites involved in memory and cognition. For this reason, the various elements of the circuit of Papez and in particular

the fornix have been targeted with DBS in both preclinical and clinical studies (Hescham et al., 2013a; Laxton and Lozano, 2013). The rationale for using fornix DBS in AD is to counteract the dysfunctional integrated

450

S. HESCHAM AND Y. TEMEL

pathways linking cortical and subcortical sites, especially those serving aspects of memory and cognition (Mirzadeh et al., 2016). In the very first reports about DBS for AD, it seems that the serendipitous finding of episodic memory enhancement following fornix DBS in one morbid obese patient (Hamani et al., 2008) was straightforwardly transformed into a phase I clinical trial with the optimistic hypothesis that it might be possible to use fornix DBS to drive its activity and to modulate memory circuits in patients with mild AD (Laxton et al., 2010). Subjects were stimulated with 3.0–3.5 V, 130 Hz, and 90 ms pulse width. The principle outcomes were that 4 out of 6 patients showed an improvement in their Alzheimer’s Disease Assessment Scale-Cognitive Subscale (ADASCog) scores 6 months after surgery, and 5 out of 6 patients showed a reduced decline in their Mini-Mental State Examination (MMSE) 12 months after surgery. Moreover, it was shown in structural magnetic resonance imaging that fornix DBS not only decreased the mean hippocampal atrophy but also increased the hippocampal volume in 2 patients 1 year after treatment, indicating the possibility for long-term structural plasticity driven by fornix DBS (Sankar et al., 2015). In a phase II trial of a yearlong, randomized, doubleblind trial of fornix DBS in 42 mild AD patients, a multivariate regression analysis showed that age and treatment interacted significantly. Twenty-one patients served as a sham control group (no stimulation) and 21 patients were stimulated with 130 Hz, between 3.0 and 3.5 V and 90 ms pulse width for a period of 12 months. Preoperative Positron Emission Tomography (PET) scans revealed characteristic lower metabolism in temporal and parietal areas in all AD patients. After 6 months of stimulation, PET imaging outcomes demonstrated a reversal of the impaired glucose metabolism in several brain regions in patients older than 65 years (n ¼ 30), but not in patients younger than 65 years (n ¼ 12). Nonetheless, this increase appeared unsustained at 12 months of chronic stimulation. Cognitive worsening was noted in all age groups; however, younger patients (