The Human Hypothalamus: Anterior Region (Volume 179) (Handbook of Clinical Neurology, Volume 179) 012819975X, 9780128199756

Citation preview

THE HUMAN HYPOTHALAMUS: ANTERIOR REGION

HANDBOOK OF CLINICAL NEUROLOGY Series Editors

MICHAEL J. AMINOFF, FRANÇOIS BOLLER, AND DICK F. SWAAB VOLUME 179

THE HUMAN HYPOTHALAMUS: ANTERIOR 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 179 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: 978-0-12-819975-6 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. Vol. Vol. Vol. Vol. Vol. Vol. Vol. Vol. Vol. Vol. Vol. Vol. Vol. Vol. Vol. Vol. Vol. Vol. Vol. Vol. Vol. Vol. Vol. Vol. Vol. Vol. Vol. Vol. Vol. Vol. Vol. Vol. Vol. Vol. Vol. Vol. Vol. Vol. Vol. Vol. Vol. Vol. Vol. Vol. Vol. Vol. Vol. Vol.

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

M. Angeles-Castellanos Faculty of Medicine, Universidad Nacional Autónoma de M
exico (UNAM), Ciudad de M
exico, Mexico A.-M. Bao Department of Neurobiology and Department of Neurology of the Second Affiliated Hospital, Zhejiang University School of Medicine; NHC and CAMS Key Laboratory of Medical Neurobiology, School of Brain Science and Brain Medicine, Zhejiang University, Hangzhou, China A. Bazer Department of Psychology, Rutgers University, Piscataway, NJ, United States J.U. Blackford Department of Psychiatry and Behavioral Sciences, Vanderbilt University Medical Center; Research Health Scientist, Tennessee Valley HealthCare System, US Department of Veterans Affairs, Nashville, TN, United States K.M. Braas Department of Neurological Sciences, University of Vermont Larner College of Medicine, Burlington, VT, United States G.M. Brown Department of Psychiatry, Centre for Addiction and Mental Health, University of Toronto, Toronto, ON, Canada

V. Carelli IRCCS Istituto delle Scienze Neurologiche di Bologna, UOC Clinica Neurologica; Dipartimento di Scienze Biomediche e Neuromotorie, Università degli Studi di Bologna, Bologna, Italy E. Challet Institute of Cellular and Integrative Neurosciences, CNRS, University of Strasbourg, Strasbourg, France T. De Boer Laboratory for Neurophysiology, Department of Cell and Chemical Biology, Leiden University Medical Center, Leiden, The Netherlands E.R. de Natale Neurodegeneration Imaging Group, University of Exeter Medical School, London, United Kingdom A. De Salles Departments of Neurosurgery and Radiation Oncology, University of California, Los Angeles, CA, United States; Department of Neurosurgery and Radiation Oncology, HCor Neuroscience, São Paulo, Brazil W.S. Dhillo Section of Endocrinology and Investigative Medicine, Imperial College London, London, United Kingdom B. Dudás Neuroendocrine Organization Laboratory, Lake Erie College of Osteopathic Medicine, Erie, PA, United States; Department of Anatomy, Histology and Embryology, University of Szeged, Szeged, Hungary

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

M. Eriksdotter Division of Clinical Geriatrics, Department of Neurobiology, Care Sciences and Society (NVS), Karolinska Institutet, Stockholm; Theme Aging, Karolinska University Hospital, Huddinge, Sweden

D.P. Cardinali Faculty of Medical Sciences, Pontificia Universidad Católica Argentina, Buenos Aires, Argentina

C. Escobar Faculty of Medicine, Universidad Nacional Autónoma de M
exico (UNAM), Ciudad de M
exico, Mexico

xiv

CONTRIBUTORS

A. Fahimi Children's Hospital Los Angeles, University of Southern California, Los Angeles, CA, United States M.-P. Felder-Schmittbuhl Institute of Cellular and Integrative Neurosciences, CNRS, University of Strasbourg, Strasbourg, France K. Fifel International Institute for Integrative Sleep Medicine (WPI-IIIS), University of Tsukuba, Tsukuba, Ibaraki, Japan C.P. Fitzsimons Brain Plasticity Group, Swammerdam Institute for Life Sciences, University of Amsterdam, Amsterdam, The Netherlands F. Gbahou Institut Cochin, Universit
e de Paris, Paris, France S.M. Gentleman Department of Brain Sciences, Faculty of Medicine, Imperial College London, London, United Kingdom S.E. Hammack Department of Psychological Science, University of Vermont, Burlington, VT, United States J. Homolak Department of Pharmacology, and Croatian Institute for Brain Research, University of Zagreb School of Medicine, Zagreb, Croatia A.M. Hulsman Experimental Psychopathology & Treatment, Behavioural Science Institute; Affective Neuroscience, Donders Institute for Brain, Cognition and Behaviour, Centre for Cognitive Neuroimaging, Radboud University, Nijmegen, The Netherlands G. Hurtado Alvarado 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

F. Klumpers Experimental Psychopathology & Treatment, Behavioural Science Institute; Affective Neuroscience, Donders Institute for Brain, Cognition and Behaviour, Centre for Cognitive Neuroimaging, Radboud University, Nijmegen, The Netherlands F. Kreier Department Pediatrics, OLVG Hospitals, Amsterdam, The Netherlands S. Kullmann Institute for Diabetes Research and Metabolic Diseases of the Helmholtz Center Munich at the University of T€ubingen, T€ubingen; German Center for Diabetes Research (DZD e.V.), Neuherberg; Department of Internal Medicine, Division of Endocrinology, Diabetology, and Nephrology, Eberhard Karls University T€ubingen, T€ubingen, Germany C. La Morgia IRCCS Istituto delle Scienze Neurologiche di Bologna, UOC Clinica Neurologica; Dipartimento di Scienze Biomediche e Neuromotorie, Università degli Studi di Bologna, Bologna, Italy J.-J. Lemaire Institut Pascal, Clermont-Ferrand, and Service de Neurochirurgie, Centre Hospitalier et Universitaire, Clermont-Ferrand, France A.S.P. Lim Division of Neurology, Department of Medicine, Sunnybrook Health Sciences Centre, University of Toronto, Toronto, ON, Canada A.K.L. Liu Department of Brain Sciences, Faculty of Medicine, Imperial College London, London, United Kingdom P.J. Lucassen Brain Plasticity Group, Swammerdam Institute for Life Sciences, University of Amsterdam, Amsterdam, The Netherlands

R. Jockers Institut Cochin, Universit
e de Paris, Paris, France

I. Maita Department of Psychology, Rutgers University, Piscataway, NJ, United States

P.S. Jones Department of Neurosurgery, Massachusetts General Hospital; Harvard Medical School, Boston, MA, United States

V. May Department of Neurological Sciences, University of Vermont Larner College of Medicine, Burlington, VT, United States

CONTRIBUTORS S.L. McElroy Lindner Center of HOPE Research Institute, Lindner Center of HOPE, Mason; Department of Psychiatry and Behavioral Neuroscience, University of Cincinnati College of Medicine, Cincinnati, OH, United States M.J. McKinley Florey Institute of Neuroscience and Mental Health; Department of Anatomy and Physiology, University of Melbourne, Parkville, VIC, Australia I. Merchenthaler Department of Epidemiology and Public Health and of Anatomy and Neurobiology, University of Maryland Baltimore, Baltimore, MD, United States S. Mitra Division of Clinical Geriatrics, Department of Neurobiology, Care Sciences and Society (NVS), Karolinska Institutet, Stockholm, Sweden M. Modi Section of Endocrinology and Investigative Medicine, Imperial College London, London, United Kingdom R.J. Nelson Department of Neuroscience, Rockefeller Neuroscience Institute, West Virginia University, Morgantown, WV, United States

xv

M. Politis Neurodegeneration Imaging Group, University of Exeter Medical School, London, United Kingdom K. Roelofs Experimental Psychopathology & Treatment, Behavioural Science Institute; Affective Neuroscience, Donders Institute for Brain, Cognition and Behaviour, Centre for Cognitive Neuroimaging, Radboud University, Nijmegen, The Netherlands A. Rojas-Granados Faculty of Medicine, Universidad Nacional Autónoma de M
exico (UNAM), Ciudad de M
exico, Mexico F. Romo-Nava Lindner Center of HOPE Research Institute, Lindner Center of HOPE, Mason; Department of Psychiatry and Behavioral Neuroscience, University of Cincinnati College of Medicine, Cincinnati, OH, United States P.J. Ryan Florey Institute of Neuroscience and Mental Health, University of Melbourne, Parkville, VIC, Australia A.A. Sadun Department of Ophthalmology, Doheny Eye Institute, University of California, Los Angeles, CA, United States A. Salehi Department of Psychiatry and Behavioral Sciences, Stanford Medical School, Palo Alto, CA, United States

M. Noroozi Cancer Clinical Trials Office, Stanford Medicine, Stanford University School of Medicine, Palo Alto, CA, United States

B.A. Samuels Department of Psychology, Rutgers University, Piscataway, NJ, United States

A. Oishi Institut Cochin, Universit
e de Paris, Paris, France

C.B. Saper Department of Neurology, Beth Israel Deaconess Medical Center, Harvard Medical School, Boston, MA, United States

S.R. Pandi-Perumal Somnogen Canada Inc, Toronto, ON, Canada G.L. Pennington Florey Institute of Neuroscience and Mental Health, University of Melbourne, Parkville, VIC, Australia M. Perosevic Neuroendocrine Unit, Massachusetts General Hospital; Harvard Medical School, Boston, MA, United States P. Pevet Institute of Cellular and Integrative Neurosciences, CNRS, University of Strasbourg, Strasbourg, France

T.W. Schmitz Department of Physiology and Pharmacology, University of Western Ontario, London, ON, Canada A. Sharif Lille Neuroscience & Cognition, University of Lille, Lille, France E.C. Soto Tinoco 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

xvi

CONTRIBUTORS

D.F. Swaab Department Neuropsychiatric Disorders, Netherlands Institute for Neuroscience, an Institute of the Royal Netherlands Academy of Arts and Sciences, Amsterdam, The Netherlands D. Terburg Department of Experimental Psychology, Utrecht University, Utrecht, The Netherlands; Department of Psychiatry and Mental Health, University of Cape Town, Cape Town, South Africa K. Toljan Department of Neurology, Neurological Institute, Cleveland Clinic, Cleveland, OH, United States N.A. Tritos Neuroendocrine Unit, Massachusetts General Hospital; Harvard Medical School, Boston, MA, United States R. Veit Institute for Diabetes Research and Metabolic Diseases of the Helmholtz Center Munich at the University of

T€ubingen, T€ubingen; German Center for Diabetes Research (DZD e.V.), Neuherberg, Germany W.H. Walker II Department of Neuroscience, Rockefeller Neuroscience Institute, West Virginia University, Morgantown, WV, United States J.C. Walton Department of Neuroscience, Rockefeller Neuroscience Institute, West Virginia University, Morgantown, WV, United States H. Wilson Neurodegeneration Imaging Group, University of Exeter Medical School, London, United Kingdom L. Zaborszky Center for Molecular and Behavioral Neuroscience, Rutgers University, Newark, NJ, United States

Contents 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) SECTION 2

141

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

159

175

xviii

CONTENTS

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) 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

189

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

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)

371

SECTION 4

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

CONTENTS 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) 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

xix 403

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

This page intentionally left blank

Contents of related volumes Volume 180 (The Human Hypothalamus: Middle and Posterior Region) Volume 181 (The Human Hypothalamus: Neuroendocrine Disorders) Volume 182 (The Human Hypothalamus: Neuropsychiatric Disorders)

Contents of Volume 180 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)

161

xxii SECTION 7

CONTENTS OF RELATED VOLUMES CONTINUED 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) SECTION 8

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

173

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

CONTENTS OF RELATED VOLUMES CONTINUED SECTION 12

xxiii

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

SECTION 13

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

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

xxiv

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

xxv

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

xxvi

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)

xxvii 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

xxviii 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

Section 1 Introduction

This page intentionally left blank

Handbook of Clinical Neurology, Vol. 179 (3rd series) The Human Hypothalamus: Anterior Region D.F. Swaab, F. Kreier, P.J. Lucassen, A. Salehi, and R.M. Buijs, Editors https://doi.org/10.1016/B978-0-12-819975-6.00014-5 Copyright © 2021 Elsevier B.V. All rights reserved

Chapter 1

Introduction: The anterior hypothalamus DICK F. SWAAB1*, RUUD M. BUIJS2, FELIX KREIER3, PAUL J. LUCASSEN4, AND AHMAD SALEHI5 1

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

Hypothalamic Integration Mechanisms Laboratory, Department of Cellular Biology and Physiology, Instituto de Investigaciones Biomedicas, Universidad Nacional Autónoma de Mexico (UNAM), Ciudad de Mexico, Mexico 3

Department Pediatrics, OLVG Hospitals, Amsterdam, The Netherlands

4

Brain Plasticity Group, Swammerdam Institute for Life Sciences, University of Amsterdam, Amsterdam, The Netherlands 5

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

SECTION 1: INTRODUCTION The first section lays out the basis for the four hypothalamus volumes. It starts with a chapter on the rich scientific neuroendocrine history. This is followed by an outline of the gross anatomy of the hypothalamus, and adjacent areas from rostral to caudal and the microscopic chemical anatomy of the nuclei, the hypothalamic vascular supply, and its afferents and efferents are described. Subsequently, the cellular morphology and distribution of the peptidergic neurotransmitter and neuromodulator systems are shown using immunohistochemistry and high-resolution, three-dimensional mapping. For the fetal development of the human hypothalamic nuclei and an overview of their adult chemical markers, see Swaab (2003, Chapter 1). Subsequently, three chapters are dedicated to recent progress in magnetic resonance imaging (MRI). Although due to the absence of white matter around hypothalamic nuclei, it is still not possible to clearly identify most of these structures, DTI (diffusion tensor imaging) fiber tracking techniques provide a basis for hypothalamic parcellation based on its afferent and efferent tracts. In addition, MRI has become the gold standard for the detection and characterization of tumors, infections, cystic or vascular lesions in the hypothalamus and surrounding regions. Resting-state functional connectivity has recently been employed to study the functional communication pathways of the hypothalamus with other

brain regions in health and to show dysfunctional connections in neurological and psychiatric diseases. The adult human brain harbors specific niches where stem cells undergo substantial plasticity and, in some regions, generate new neurons throughout life. Current knowledge of the adult human hypothalamic niche is, however, based mainly on neuroanatomic studies that rely on the immunodetection of proxy markers of neurogenesis. Additional studies and strategies are discussed to confirm the existence of an active neurogenic niche in the adult human hypothalamus. The last introductory chapter illustrates that the quality of postmortem hypothalamus research strongly depends on thorough clinical documentation of the patient’s disorder and therapies, a professional neuropathological investigation of the entire brain of both, cases and controls, and a careful matching of a considerable number of confounding factors between patients and controls.

SECTION 2: THE BASAL FOREBRAIN CHOLINERGIC SYSTEM The basal forebrain cholinergic system consists of the diagonal band of Broca (DBB) and the nucleus basalis of Meynert (NBM). For detailed maps of the topographical organization of the subdivisions of this system in the human brain and adjacent structures, see Swaab (2003, Chapter 2). Human data are recently not only obtained from postmortem material but also by novel noninvasive

*Correspondence to: Prof. Dr. Dick F. Swaab, Department Neuropsychiatric Disorders, Netherlands Institute for Neuroscience, Meibergdreef 47, 1105 BA Amsterdam, The Netherlands. Tel: +31-20-5665500, Fax: +31-20-5666121, E-mail: [email protected]

4

D.F. SWAAB ET AL.

neuroimaging techniques. Converging evidence indicates a selective vulnerability of cholinergic neurons in Alzheimer’s and Parkinson’s diseases that is expressed along a rostral-caudal topography in the basal forebrain cholinergic system. The DBB sends major projections to the hippocampal CA2 subfield and is particularly involved in memory retrieval. The DBB appears to be especially susceptible to Lewy body pathologies. Dysfunction of the cholinergic system and degeneration of the NBM have been implicated in the pathophysiology of cognitive impairment in Parkinson’s disease. Microstructural alterations within the NBM, revealed by diffusion tensor imaging, have been identified as a predictor for the development of cognitive impairment in patients with this neurodegenerative disorder. The cholinergic system has been the prime target for research and therapy in Alzheimer’s disease. Evidence is presented that points to the effects of endosomal pathology and a mainly unidirectional traffic jam in the cholinergic neurons. This is now leading to novel therapeutic strategies to restore their function. The currently approved treatment for Alzheimer’s disease, the inhibitors of acetylcholine-esterase, has only mild beneficial effects and often side effects. In recent years, approaches to rescue the degenerating cholinergic system in Alzheimer are focusing on nerve growth factor (NGF). Since NGF does not pass the blood–brain barrier, approaches with cerebral injection of genetically modified cells or viral vectors or implantation of encapsulated cells into the cholinergic system in Alzheimer patients have been performed. These attempts have been partially successful.

SECTION 3: THE CIRCADIAN SYSTEM During the evolution of life, the temporal rhythm of our rotating planet was internalized in the form of circadian rhythms. These rhythms allow to anticipate predictable changes in the environment and are precisely controlled by a molecular master clock located within the suprachiasmatic nucleus (SCN) of the hypothalamus. The SCN is also responsible for seasonal rhythms. SCNinduced behavior, such as rhythmic food intake, induces oscillating clock genes, which serve as a transcriptional driving force for numerous cellular processes in all tissues. The oscillating systems continuously change their homeostatic set points over the day–night cycle in order to adjust to external environmental or hormonal changes. If different rhythms in the body are not synchronous for extended periods of times, disease may develop. Although the suprachiasmatic nucleus is present in human from the middle of pregnancy, circadian rhythms only develop gradually during the first postnatal months. The exposure of premature babies to light/dark cycles

results in a more rapid establishment of circadian rhythms. In adulthood there is a strong association between circadian disruption, in the form of shift work, exposure to light at night, jet lag, and social jet lag, and psychiatric illnesses such as anxiety, major depressive disorder, bipolar disorder, and schizophrenia. Changes in diurnal and seasonal rhythms in gene expression are also found in the human brain in aging and Alzheimer’s disease. The resulting nightly restlessness in Alzheimer’s disease can be treated with bright light and melatonin. Impaired circadian rhythms and SCN function have also been observed in patients with neurodegenerative diseases such as Parkinson’s disease, multiple system atrophy, and progressive supranuclear palsy. Bright light exposure improves the quality of life in Parkinson patients too. Circadian disturbances in neurodegenerative disorders are not only due to alterations in the SCN. Melanopsin retinal ganglion cells are responsible for the photoentrainment of circadian rhythms through their projections to the SCN. Their dysfunction or loss may play a crucial role in affecting circadian rhythms and sleep in many neurodegenerative disorders. Melatonin, secreted during the night by the pineal gland, is an important efferent hormonal signal of the SCN. Its role is to reinforce nighttime physiology. Melatonin acts through two high-affinity membrane receptors. The development of new ligands which are highly selectivity for each subtype is thought to be at the root of new therapeutic agents for treating specific pathologies, not only circadian rhythm- and sleep-related disorders but are presumed also to be beneficial for complex diseases like major depression, Alzheimer’s disease, autism, and attention-deficit/hyperactivity disorders. These are examples of chronotherapy, which is meant to optimize medical treatments taking into account the body’s circadian rhythms. Chronotherapy can be accomplished by restoring the sleep–wake rhythms of patients to improve the sequels of several pathologies or to take into account the circadian rhythms of patients to improve therapeutic effects and diminish side effects. A novel observation in this field is that nightly melatonin administration mitigates antipsychotic-induced adverse metabolic effects.

SECTION 4: BED NUCLEUS OF THE STRIA TERMINALIS AND THE FEAR CIRCUIT The bed nucleus of the stria terminalis (BNST) is, as part of the “extended amygdala,” functionally closely linked to the amygdala nuclear complex and is also connected to hypothalamic nuclei and other limbic structures. The BNST plays a critical role in responding to distant or ambiguous threats, while the amygdala is particularly

INTRODUCTION: THE ANTERIOR HYPOTHALAMUS involved in acute threats. The BNST circuits are predominantly GABAergic and more than a dozen neuropeptides may be differentially coexpressed in BNST neurons. Some BNST subregions are highly sexually dimorphic. For the chemical anatomy of the human BNST and the reversed sex differences of the BNST in transsexual people, see Swaab (2003, Chapter 7). The BNST modulates the hypothalamo-pituitary adrenal axis. The heterogeneous subnuclei of the BNST integrate inputs from mood and reward-related areas and send direct inhibitory projections to the hypothalamus. BNST–hypothalamus circuitry is not only implicated in anxiety behaviors but also in motivated behaviors, feeding, and sexual behavior and may underlie various psychiatric diseases, such anorexia, addiction, chronic pain, depression, anxiety-related disorders, and posttraumatic stress disorder.

5

fluid, temperature, sleep, and cardiovascular homeostasis and is characterized by angiotensin binding. During menopause, normal thermoregulatory response is disrupted by enhanced neurokinin B signaling via the neurokinin-3 receptor (NK3R) in the preoptic area, resulting in hot flushes. NK3R antagonists can produce rapid and sustained improvements in hot flushes and are a safe and efficacious alternative to hormone replacement therapy. In the lateral preoptic area, the intermediate nucleus is localized, containing largely galanin-expressing neurons. This nucleus is larger in men than in women and is also known as the interstitial nucleus of the anterior hypothalamus-1, the sexually dimorphic nucleus, and the ventrolateral preoptic nucleus. It is thought to drive sleep behavior.

REFERENCE

SECTION 5: PREOPTIC AREA The median preoptic area occupies the midline anterior wall of the third ventricle. It is a major regulator of

Swaab DF (2003). In: MJ Aminoff, F Boller, DF Swaab (Eds.), The human hypothalamus. Basic and clinical aspects. Part I: nuclei of the hypothalamus. Handbook of clinical neurology. vol. 79. Elsevier, Amsterdam.

This page intentionally left blank

Handbook of Clinical Neurology, Vol. 179 (3rd series) The Human Hypothalamus: Anterior Region D.F. Swaab, F. Kreier, P.J. Lucassen, A. Salehi, and R.M. Buijs, Editors https://doi.org/10.1016/B978-0-12-819975-6.00031-5 Copyright © 2021 Elsevier B.V. All rights reserved

Chapter 2

History of hypothalamic research: “The spring of primitive existence” FELIX KREIER1* AND DICK F. SWAAB2 1

Department Pediatrics, OLVG Hospitals, Amsterdam, The Netherlands

2

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

Abstract The central brain region of interest for neuroendocrinology is the hypothalamus, a name coined by Wilhelm His in 1893. Neuroendocrinology is the discipline that studies hormone production by neurons, the sensitivity of neurons for hormones, as well as the dynamic, bidirectional interactions between neurons and endocrine glands. These interactions do not only occur through hormones, but are also partly accomplished by the autonomic nervous system that is regulated by the hypothalamus and that innervates the endocrine glands. A special characteristic of the hypothalamus is that it contains neuroendocrine neurons projecting either to the neurohypophysis or to the portal vessels of the anterior lobe of the pituitary in the median eminence, where they release their neuropeptides or other neuroactive compounds into the bloodstream, which subsequently act as neurohormones. In the 1970s it was found that vasopressin and oxytocin not only are released as hormones in the circulation but that their neurons project to other neurons within and outside the hypothalamus and function as neurotransmitters or neuromodulators that regulate central functions, including the autonomic innervation of all our body organs. Recently magnocellular oxytocin neurons were shown to send not only an axon to the neurohypophysis, but also axon collaterals of the same neuroendocrine neuron to a multitude of brain areas. In this way, the hypothalamus acts as a central integrator for endocrine, autonomic, and higher brain functions. The history of neuroendocrinology is described in this chapter from the descriptions in De humani corporis fabrica by Vesalius (1537) to the present, with a timeline of the scientists and their findings.

DISCIPLINE OF NEUROENDOCRINOLOGY: CURRENT DEFINITIONS Here in this well-concealed spot, almost to be covered with a thumbnail, lies the very main spring of primitive existence—vegetative, emotional, reproductive—on which with more or less success, man has come to superimpose a cortex of inhibitions. Cushing (1932)

Neuroendocrinology is the discipline that studies hormone production by neurons, the sensitivity of neurons for hormones, as well as the dynamic, bidirectional interactions between neurons and endocrine glands. These interactions do not only occur through hormones but are also partly accomplished by the autonomic system that is regulated by the hypothalamus and that innervates the endocrine glands. In 1909 Karplus and Kreidl showed that stimulation of the hypothalamus

*Correspondence to: Felix Kreier, M.D., Ph.D., Department of Pediatrics, OLVG, Bijlmerdreef 939, Amsterdam 1103 TW, The Netherlands. Tel:+31-161-867-9189, Fax: +31-84-830-62-64, E-mail: [email protected]

8

F. KREIER AND D.F. SWAAB

led to a number of autonomic effects, such as changes in heart rate and salivation (Zuckerman, 1954). The executing branches of the autonomic nervous system form a parasympathetic and a sympathetic branch. These two branches target structures and organs in the body and have an antagonistic function, whereby it is assumed that the parasympathetic branch is involved in anabolic functions and the sympathetic branch in catabolic functions. Via these two branches also information about the functional condition of our organs is transmitted back to the brain. Neuroendocrine processes function in concert to allow maintenance of homeostasis for the organism, and the survival of the individual and of the species, as phrased beautifully by Cushing in the citation mentioned previously. The central brain region of interest for neuroendocrinology is the hypothalamus, a name coined and introduced by Wilhelm His in 1893. Many endocrine cascades and also the autonomic regulation are initiated in this structure and feedback regulation of these systems also takes place in this brain area. The neurons of the hypothalamus contain one or more of all four types of neuroactive substances; acetylcholine, amines, amino acids, and a multitude of neuropeptides. Two central endocrine glands that are involved are the pituitary gland, for which, in 1778, Von Soemering proposed the name “Hypophysis,” and the pineal gland that was considered to be the seat of the soul by Ren
e Descartes (1596–1650, Fig. 2.1).

Fig. 2.1. Ren
e Descartes (1596–1650) who considered the pineal gland to be the seat of the soul. Painting by Frans Hals, Louvre Museum, Paris.

A special characteristic of the hypothalamus is that it contains neuroendocrine neurons projecting either to the neurohypophysis or to the portal vessels of the anterior lobe of the pituitary in the median eminence, where they release their neuropeptides or other neuroactive compounds into the bloodstream, which subsequently act as neurohormones. The strong innervation of the neurohypophysis was known already by Ramón y Cajal, the Nobel Prize laureate for Physiology or Medicine (Cajal y Ramón, 1909, Defelipe, 2010, Fig. 2.2). The neuroendocrine cells form the integrative element between the endocrine glands and the brain. In the 1970s it was found that vasopressin and oxytocin not only are released as hormones in the circulation but that their neurons project to other neurons within and outside the hypothalamus (Buijs et al., 1978), where neuropeptides are released from synapses (Buijs and Swaab, 1979, Fig. 2.3) and function as neurotransmitters or neuromodulators that regulate central functions, including the autonomic innervation of all our body organs (Buijs and Kalsbeek, 2001). Oxytocin neurons were recently shown to send not only an axon to the neurohypophysis but also axon collaterals of the same neuroendocrine neuron to a multitude of brain areas (Zhang et al., 2020). In this way, hypothalamus acts as a central integrator for endocrine, autonomic, and higher brain functions.

NEUROSECRETION AND CENTRAL EFFECTS OF NEUROPEPTIDES The axons of the supraoptic and paraventricular nucleus (SON and PVN) run toward the neurohypophysis. Due to its shape and localization along the optic tract, the SON was formally called the “tangential nucleus (Cajal y Ramón, 1911).” Together, these structures form the hypothalamo-neurohypophysial system (HNS), which represents the classic example of a neuroendocrine system. Vasopressin (¼antidiuretic hormone, ADH) and oxytocin are key hormones produced in the large, magnocellular neurons of the SON and PVN. Vasopressin is involved in the control of waterbalance, circulating volume and antidiuresis, while oxytocin plays a role in labor and lactation. A second type of (smaller) neuroendocrine cell, the parvicellular neuroendocrine neurons, is found for instance in the PVN. These neurons release their peptides into the portal capillaries that transport them to the anterior lobe of the pituitary. Examples are corticotropin-releasing hormone (CRH) and thyrotropin-releasing hormone (TRH). A third type of PVN cells projects to other neurons, where the peptides act as neurotransmitters/ neuromodulators and in this way influence many functions, like sexual behavior, social behavior, the regulation of stress, and metabolism. Such peptidergic neurons, e.g., centrally projecting vasopressin, oxytocin, CRH, and

HISTORY OF HYPOTHALAMIC RESEARCH

9

Fig. 2.2. (A) Santiago Ramo´n y Cajal (1852–1934). Portrait in 1906, by R. Madrazo y Garreta. Madrid Athenaeum. (B) The strong innervation of the neurohypophysis was known already by Ramo´n y Cajal, the Nobel Prize laureate for Physiology or Medicine. From Ramo´n y Cajal S (1909). Histology of the nervous system (Translated from the French). In: N Swanson, LW Swanson (Eds.), History of neuroscience, vol. 1. General principles, spinal cord, spinal ganglia, medulla & pons. Oxford University Press, NY. 1995.

TRH neurons, also regulate the autonomic nervous system. The PVN also regulates the pineal gland, which secretes melatonin, via a polysynaptic autonomous pathway (Buijs and Kalsbeek, 2001). The horseshoe-shaped infundibular nucleus (or in rodents the arcuate nucleus) surrounds the lateral and posterior entrance of the infundibulum and is situated outside the blood–brain barrier. In this way it receives information that is crucial for maintaining body homeostasis. The infundibular nucleus is involved in reproduction, eating behavior and metabolism, thyroid hormone feedback, growth, and dopamine regulation and contains many small neuroendocrine neurons that project to the portal system. The infundibular nucleus is continuous with the stalk/median eminence region containing the

adenohypophysis portal capillaries. The stalk of the pituitary is surrounded by the pars tuberalis of the pituitary. CRH, luteinizing hormone-releasing hormone (LHRH), opiomelanocortins, somatostatin, growth hormonereleasing hormone, galanin, TRH, and substance-P fibers innervate the stalk/median eminence region.

PORTAL SYSTEM The primary plexus of the portal system is made up of two vascular systems that are intricately linked: the surface network (the rete mirabile of Galenus) and the deep network. According to Galenus, the vital spirit changed into the animal spirit in the blood from the heart, at the rete mirabile (Anderson and Haymaker, 1974).

10

F. KREIER AND D.F. SWAAB

Fig. 2.3. Vasopressin positive terminal forming a synapse (black arrow) with an unlabeled dendrite in lateral habenular nucleus. Note DAB deposit around clear vesicle-like structures (arrows). Bar ¼ 0.25 mm. From Buijs RM, Swaab DF (1979). Immunoelectron microscopical demonstration of vasopressin and oxytocin synapses in the limbic system of the rat. Cell Tissue Res 204: 355–365, with permission.

Fig. 2.4. Vesalius was only 22 when, in Padua in 1537, he started to work on his book, De humani corporis fabrica libri septem, containing 300 illustrations. Andreas Vesalius, De Humani Corporis Fabrica, Basilae 1543.

This surface network (or mantel plexus) covers the surface of the median eminence. From this network stem, numerous short capillary loops penetrate into the median eminence, where neurosecretory substances are released into them (Duvernoy, 1972; Swaab, 2004; Chapter 17).

EARLY IDEAS ABOUT PITUITARY AND PINEAL GLAND FUNCTION Vesalius was only 22 when, in Padua in 1537, he started to work on his book, De humani corporis fabrica libri septem (Fig. 2.4), containing 300 illustrations. Although

HISTORY OF HYPOTHALAMIC RESEARCH

11

Fig. 2.4—Cont’d

he was generally very critical of the old ideas, he accepted Galenus’ (131–201 AD) view that the brain excreted waste material (pituita) through the infundibulum into the pituitary (deij), whence it somehow passed into the nasopharynx (Fig. 2.5). He described the pituita as residu, left over, from the ultrarefinement of animal spirit, the substance responsible for sensation and motion, which had reached the brain in the form of vital

spirit from the heart (Anderson and Haymaker, 1974). Galenus’ Rete Mirabileia coincides with the surface capillary network of Duvernoy (1972). The pineal gland or epiphysis began to receive special attention because Ren
e Descartes in the early 1630s constructed a description of the body that did not rely on vegetative or vital souls. Instead, he proposed that the body was made up of particles that obeyed the laws

12

F. KREIER AND D.F. SWAAB

Fig. 2.5. Vesalius infundibulum. Vesalius’ representation of the infundibulum of the pituitary (Anderson and Haymaker, 1974) Vesalius: “In this small figure we have depicted the pelvis or cup (cyathus) set upright by which the pituita of the brain distills into the gland underlying it, and then we have sketched in the four ducts carrying the pituita down from the gland through foramina in the neighborhood of the gland. Therefore A indicates the gland (not shown in this figure) into which the pituita is distilled and B, the pelvis along which it is led. C, D, E, F are the passages provided for the easier exit of the pituita passing down there.” Andreas Vesalius, De Humani Corporis Fabrica, Basilae 1543.

of physics and that it was no different from a machine; neither needed a soul to drive its movements. Descartes envisioned nerves as a system of cords and inflating tubes that mechanically produced involuntary movements: “Finally, I want you to consider all functions of this machine—the intake and digestion of food, respiration, waking and sleeping, the reception by the external sense organs of light, sounds, smells, the retention of these ideas in the memory, the lower impulses of lusts and passions and finally the external movements of all the limbs—I want you to consider all these functions as taking place naturally within this machine, solely as a result of the nature of its organs, exactly like the movements of a clock.” His famous analogy was that of a church organ. He claimed that the air that is pumped into the organ corresponds with the most delicate and most active particles of the blood, the “animal spirits,” blown into the ventricles by hypothetical openings via a system of arteries (what we now call the plexus choroideus). These hollow nerves would direct these animal spirits to the muscles. In this analogy the epiphysis played the role of the organ keyboard, which forced the animal spirits into certain ventricles, in the same way that the keyboard forces the air into particular pipes. This has made Descartes, unintentionally and unjustly, into the father of the dualistic approach in the body–mind discussion. For instance, in a recent

best seller, Damasio argues that Descartes’ “error” was the dualist separation of mind and body, rationality, and emotion (Damasio, 1994). The pineal gland was later for a long time considered to be a mere vestigial remnant of evolution, until the discovery of melatonin in 1958 by Aaron B. Lerner (López-Muñoz et al., 2011). Dissection of the human brain took place in Amsterdam in the 17th century, even though the context then was very different from the present state of affairs. The old city gate on Nieuwmarkt, which is called De Waag (the Weighing House), featured as an anatomical theater, where criminals who had been sentenced to death and who had been hanged were autopsied. After the execution the body was put at the disposal of the guild of surgeons for a public dissection in the “Theatrum Anatomicum” on the first floor of De Waag. The city council allowed one public dissection per year, which had to take place in winter, as it took 3–5 days. The audience generally consisted of a few 100 people who each paid a fee of 20 cents. Heart, liver, and kidneys were handed round for inspection by the public. The justification for this use of the body for the benefit of surgeons in training can still be seen around the inner wall of the old Theatrum Anatomicum where it says (in Latin): “Those who during life were wrongdoers who caused only harm, become useful after they have died.” Rembrandt, in his Dr. Deijman’s Anatomy Lesson of 1656, captured a crucial moment of such a postmortem examination (Fig. 2.6). The praelector and doctor medicinae Deijman is behind the body of the dissected Flemish tailor and thief Joris Fontijn, a.k.a. Black John, who was sentenced to die by hanging on January 27, 1656. The assistant of the praelector, college master Gijsbert Calcoen, patiently waits, skull in hands, for the brain to be deposited there. Dr Deijman meanwhile lifts up the falx cerebri—the membrane that separates the left and right hemispheres—with a pair of tweezers, at the same time also lifting the pineal gland. This was prescribed in the protocol because on the authority of Descartes the epiphysis was considered the seat of the soul, and it was an extra punishment for the soul to see the body destroyed at the end of the dissection. Descartes lived in the Netherlands for some 19 years. In Amsterdam he lived, among other places, in Kalverstreet, a cattle market, where he bought carcasses for research purposes. Rembrandt’s depictions of scenes set in Amsterdam were clearly influenced by him. Dr Deijman’s Anatomy Lesson, now on display at the Amsterdam Historical Museum, is the central part of the original work, 2.5 by 3 m, which hung in De Waag when it was almost completely destroyed in a fire in 1723. A sketch by Rembrandt shows that seven well-known surgeons were present around the central part of the painting. It should be noted that when

HISTORY OF HYPOTHALAMIC RESEARCH

13

Fig. 2.6. The central remaining fragment of Rembrandt’s (1656) painting of Dr. Deyman’s anatomical lecture. Amsterdam Historical Museum. Dr. Deyman’s anatomical lecture lasted three days and was open to the public for the amount of 20 cents. In the middle the praelector and doctor medicinae Jan Deyman is lifting the falx cerebri with a lancet in order to show the soul, which was thought to be localized in the pineal gland of the dissected body. This was an integral part of the punishment. The college master, Gijsbert Calkoen, waits patiently to collect the brain in the skull of the thief Joris Fonteyn (Black John) who had just been executed. In 1723 a large part of this painting was destroyed by a fire in the medieval gate “De Waag,” where the teatrum anatomicum was situated. Amsterdam Museum, with permission.

Rembrandt was painting this anatomical lesson in 1656, the public autopsies were not yet performed in De Waag, but still took place in the Sint Margaretha convent at the Nes. The autopsy hall was there from 1639 to 1691. During this same period, the group led by the Oxford physician Thomas Willis (1621–75) dissected brains of humans, dogs, sheep, and other animals. Willis recorded their work in his 1664 book The Anatomy of the Brain, the first major work only on the brain ever written. Of course his name is still mentioned in relation to the arterial circle at the base of the brain. Over the next 8 years, he relied on his anatomical discoveries and careful bedside observations to write Cerebral Pathology on convulsive disorders and Two Discourses Concerning the Soul of Brutes on neurological and psychological disorders. Willis dismissed Descartes’ notions of the pineal gland and ventricles and demonstrated that these chambers could not house the spirits. Arguing that the brain itself was the site of mental functions, he carried out experiments to show that different functions were located in different regions. Instead of Descartes’ speculative sketch of involuntary movements wrongly involving the pineal gland, Willis offered a far more accurate account of reflexes (Zimmer, 2004). Julius Axelrod (Nobel Prize 1970) played an important role in revealing

the real role of the pineal gland by the investigation of the biosynthesis of its hormone melatonin, the hormone released during the night, that is involved in sleep regulation. However, concerning the function of the infundibulum, in 1664, Thomas Willis stayed close to Vesalius and Galenus when he argued that “the position and structure” of the infundibulum indicate that “some humour out of the ventricles of the cerebrum is carried into the pituitary gland. For that part is so constituted, that a discharge of humours is effected into its aperture from every angle and recess of the interior cerebrum, and its appendage; and while in the various animals the shape and the situation of the ventricles differ, nevertheless in every one of them all the ventricles, of whatever kind they be, have apertures opening in the direction of the infundibulum.” Willis felt that “there was no room for doubting, that serious liquids descend by this way from the cerebrum into the pituitary gland…. If any one takes but a cursory view of the parts which are situated around the ventricles, and if he examines their structure but lightly, he will easily agree with the Ancients, that the excrements of the cerebrum are discharged partly through the infundibulum into the palate underneath, or that in an anterior direction

14

F. KREIER AND D.F. SWAAB

NEUROSECRETION AND THE NEUROHYPOPHYSIS One of my anatomist friends who has made notable contributions to our knowledge of the anterior lobe hormones has challenged me to produce any corresponding clinical or experimental evidence of posterior lobe activity. To this challenge, this and the succeeding papers are a partial answer. Cushing (1932, p. 60)

Fig. 2.7. Van Rijnberk, who stated in 1901, on the basis of his experimental PhD research with Lo Monaco, G.: “the pituitary is a rudimentary organ without any functional meaning.” Photo by Merkelbach, Stadsarchief Amsterdam.

they are cast out through the olfactory bulbs into the nares” (Zuckerman, 1954). The function of the pituitary gland remained an enigma for a long time. In 1901 on the basis of his experimental PhD research with Lo Monaco, G. Van Rijnberk (Fig. 2.7) stated: “the pituitary is a rudimentary organ without any functional meaning” (cited by Prof. C. Winkler) (Winkler, 1934). Van Rijnberk later became professor of physiology at the University of Amsterdam! Around the year 1908, from the occurrence of adipose-genital dystrophy following partial excision of the pituitary, Cushing and his coworkers drew the important conclusion that acromegaly would be due to hyperpituitarism. In 1913 the antidiuretic effect of posterior pituitary extracts was described in patients suffering from diabetes insipidus, and as early as 1909, the oxytocic effects of posterior pituitary extracts were used in labor (see later). By the mid-1930s the pituitary was established as the “master gland,” the conductor of the endocrine orchestra. Selye, after his observation in that period that many stressors could stimulate adrenal cortical activity, formulated the “general adaptation syndrome” (for references, Meites 1992).

For a long time it was believed that the neurohypophysis made a product and that its function was regulated by the innervating nerves from the SON and PVN. In 1908 P.T. Herring first described the “peculiar hyaline bodies” seen in sections of the neurohypophysis and expressed the belief that what we now call “Herring bodies” represented the secretory product of the epithelial investment of the posterior lobe known as the pars intermedia. When Collin (1928) found stained droplets in the hypothalamus of the guinea pig, droplets similar to the secretory material found in the neural lobe; he suggested that this material had been transported from the pituitary to the hypothalamus. Indeed, the “Herring bodies” are accumulations of neurosecretory material. However, Cushing’s description was that the globules appeared to find their way toward the tuber cinereum and “in favorable histological preparations could be seen passing between the bodies of the ependymal cells to enter the infundibular cavity” (Cushing, 1932), an erroneous interpretation. In spite of the suggestions of Cushing (1932) that in those days there was no clinical or experimental evidence for a function for posterior lobe products (see citation mentioned previously), clinically relevant endocrine effects on the kidney and uterus had already been known for years. In 1898 Howell showed the vasoconstrictive action of an extract of the posterior lobe of the pituitary, and Sir Henry Dale (Nobel Prize 1936) demonstrated in 1906, “almost by accident” as he said in 1957, the oxytocic action on the uterus (Dale, 1906). In 1905 Ott and Scott described the milk ejection effect of such an extract (Brooks, 1988). Clinical applications, too, were already known. In 1913 Von den Velden and Farini had described the antidiuretic effect of posterior pituitary extracts in patients suffering from diabetes insipidus, and in 1909 Blair-Bell had reported the oxytocic effects of posterior pituitary extracts in labor. In 1921 Starling and Verney showed that pituitrin, a posterior pituitary extract, caused a strong antidiuretic effect on an isolated kidney and rightly concluded that the antidiuretic effect of pituitrin was a direct one on the kidney and that the hypotonic urine of the isolated kidney preparation was almost

HISTORY OF HYPOTHALAMIC RESEARCH certainly due to the absence of an antidiuretic substance. Later he showed that blood coming from the head inhibited urine flow in his isolated kidney preparation and blood coming from the pelvis or lower legs did not. Moreover, blood from the head lost its power when the pituitary was removed. Verney concluded that “an antidiuretic principle or principles are contributed by the pituitary to the blood” (for references, see Sawin, 2000). One can only speculate on the question whether, in 1932, Harvey Cushing was aware of this evidence on the effects of pituitary extracts (see citation mentioned previously) or whether he did not want to know, the more so since he himself injected a commercially available posterior lobe extract that was called “obstetrical pituitrin” into the cerebral ventricle of patients (Cushing, 1932, see later).

Neuroendocrine neurons The evidence that such cells secrete colloid and are to be considered a ‘diencephalic gland’ is morphological evidence and does not deserve acceptance at this time. H.B. Van Dyke, 1939 cited by Scharrer (1975) Starting with K€ohler in 1886, several clinical and experimental observations have since shown that lesions of the basal hypothalamus or of the pituitary stalk could induce diabetes insipidus (Brooks, 1988). In addition, there was a famous case report telling of a man who, in 1910, received a gunshot wound, resulting in a strong polyuria and “sexual dystrophy.” At autopsy the bullet was found to be lodged in the sella-turcica and to have destroyed the infundibular process and posterior lobe (Vonderahe, 1949; Brooks, 1988). However, no immediate connection was made between the observations revealing the diuresis that occurred following lesions of the hypothalamus and the already known antidiuretic effects of the neurohypophysis. The concept of “neurosecretion,” i.e., hormone secretion by neurons, was first proposed by Carl Casky Speidel in 1917 in his thesis “Gland Cells of internal secretion in the spinal cord of Scates,” which hardly received any attention (Sarnat, 1983). The acceptance of this idea came only in 1949, after a long time of research by Ernst and Berta Scharrer, followed by a crucial observation by Wolfgang Bargmann. Ernst Scharrer concluded in his PhD thesis in 1928 on a teleost fish Phoxinus laevis that some hypothalamic neurons specialize in secretory activity comparable to endocrine gland cells and that they had a relationship to hypophysial functions. In 1933 Ernst Scharrer and Scharrer and Graup described a similar microscopy of neurosecretory cells in a number of species, including the large SON and PVN cells in the

15

human hypothalamus. Scharrer (1933) and Scharrer and Gaupp (1933) mentioned the presence of multiple nuclei in some of the neurons in the human SON and PVN and even a possible relationship between the human SON and PVN and water metabolism and diabetes insipidus. After his marriage to Berta in 1934, they decided to divide the animal kingdom between the two of them. Ernst would continue his studies on vertebrates and Berta would search for comparable phenomena among invertebrates. When the “Aryans” Ernst and Berta arrived at the Institute in Frankfurt after the “Machtergreifung” by Hitler, they used to have briefcases in both hands, preventing them to ta carry out “Heil Hitler” greetings. They decided to leave Germany because of the persecution of their Jewish colleagues (Zeidman, 2020). When in 1937 Ernst was granted a Rockefeller Fellowship, they left for the University of Chicago. There was an almost universal rejection of Scharrer’s observations by the scientific community (Fig. 2.8; Scharrer, 1992). According to the critics of those days, his concept of neurosecretion was based on “nothing more than signs of pathological processes, postmortem changes or fixation artifacts.” In the 1940s “practically everybody vigorously or even viciously” rejected the concept that a neuron could have a glandular function Berta Scharrer later wrote to one of us (DFS). They tried to counter the criticism in the first place by carefully showing that such phenomena were present throughout the animal kingdom, in vertebrates and invertebrates (Scharrer, 1933, 1975; Scharrer and Scharrer, 1940). In the second place, they argued “that in about 200 human cases studied by several investigators, no definite relation was found between the histological picture of the SON and PVN on the one hand and possible confounders of the individual at the time of death on the other. No correlation could be established between age, sex, terminal disease, time of the year, etc., and the histological appearance of the cells of the SON and PVN” (Scharrer and Scharrer, 1945). The initially highly charged rejection of the neurosecretion concept was followed by acceptance only when Wolfgang Bargmann (Fig. 2.9), who had met Ernst Scharrer already as a student in 1933–34 in Frankfurt, demonstrated the same Gomori-positive material in the neurohypophysis and in the neurons of the SON and PVN and their fibers and concluded that the axons from the SON and PVN transport material to the neurohypophysis (Bargmann, 1949). He named the aggregate of fibers “the neurosecretory pathway.” Anderson and Haymaker described this episode as follows: Seeking a new method for revealing neurosecretory material, he (Bargmann) placed sections from a dog’s brain into acid-permanganate-chrome-alum

16

F. KREIER AND D.F. SWAAB

Fig. 2.9. Wolfgang Bargmann (1906–1978). In 1949, following a staining according to Gomori’s method, Bargmann, shouted, waving the cigar that was always in his hand: “Donnerwetter!,” the cells of the supraoptic and paraventricular nuclei and the fibers extending into the infundibulum and reaching the posterior lobe had selectively taken on a blue hue! From Anderson E, Haymaker W (1974). Breakthroughs in hypothalamic and pituitary research. Prog Brain Res 41: 1–60, with permission.

Fig. 2.8. (A) Ernst and Berta Scharrer preparing serial sections of hypothalamus at the Edinger Institute of Neurology, Frankfurt am Main, Germany, 1935, two years before they fled to Chicago. In the 1940s “practically everybody vigorously or even viciously” rejected the concept that a neuron could have a glandular function Berta Scharrer later wrote to DFS. From Scharrer B (1992). The concept of neurosecretion and its place in neurobiology. In: FG Worden, JP Swazey, G Adelman (Eds.), The neurosciences: paths of discovery, I. Birkh€auser, Boston, with permission. (B) Neurons filled with secretory droplets. From Scharrer E (1934). Stammt alles Kolloid im Zwischenhirn aus der Hypophyse? Frankf Z Path 47: 134.

hematoxylin, according to Gomori’s method, and was astonished at what he found (instead of shouting Eureka, Bargmann, waving the cigar that was always in his hand

exclaimed in a booming voice: “Donnerwetter!”), the cells of the supraoptic and paraventricular nuclei and the fibers extending into the infundibulum and reaching the posterior lobe had selectively taken on a blue hue! When, ever after, in papers dealing with hypothalamic neurosecretion “Gomori-positive” and “Gomori-negative” results were cited, Gomori would comment in conversation that he found the terms distasteful but amusing. “Right now” he once said before lunch to a fellow-Hungarian, Jacob Furth: “I feel Gomori-negative (Anderson and Haymaker, 1974)”.

Vasopressin, oxytocin, and the neurophysins In 1955 Du Vigneaud received the Nobel Prize for the elucidation of the chemical nature of vasopressin and oxytocin and the synthesis of vasopressin. The nonapeptides vasopressin and oxytocin are synthesized in the hypothalamus as part of a large precursor that includes a neurophysin for both peptides and a C-terminal glycoprotein for the vasopressin precursor only. The

HISTORY OF HYPOTHALAMIC RESEARCH vasopressin and oxytocin precursor genes are only separated by 12 kb in the human genome, located on the distal short arm of chromosome 20, and are transcribed toward each other (Schmale et al., 1993; Bahnsen et al., 1992, see also Chapter 9 in Volume 182). In 1957 in a paper of 130 pages summarizing 14 years of work, Verney concluded that osmoreceptors regulating the release of ADH (¼vasopressin) are located in the hypothalamus, probably in the anterior or preoptic areas. Connections with the SON and PVN need to be intact to enable hormone release (for references, see Sawin, 2000). When the neurophysins were discovered by Archer et al. (1956), they were considered to be inactive fragments of the precursor, with a higher molecular weight and were proposed to act as “carriers” for vasopressin and oxytocin (Archer et al., 1956). The molecular role of neurophysins is at present considered to be much more important in the light of the knowledge on mutations in the part of the neurophysin that causes diabetes insipidus. Mutations in the neurophysins produce a change in their polymerization and salt bridges and thus in their intracellular trafficking, resulting in an accelerated, aspecific enzymatic degradation of the hormone accumulation in an organelle and subsequently degeneration of the neuron, revealing the clinical symptomatology around 9 years of age. So, rather than being a mere inactive part of the precursor, neurophysins are considered an essential molecular system for carrying and protecting the nonapeptides during their transport through the cell (Legros and Geenen, 1996).

17

flow to the SON and PVN. In 1935 Houssay et al. first observed, in the toad, that the blood flowed downward from the median eminence to the anterior pituitary. On the basis of histological observations, Wislocki and King, in 1936, proposed that the blood flow in the portal vessels of the monkey was also downward. In 1947 and 1949 using a combination of India ink perfusion techniques and direct microscopic observations in the living rat, Green and Harris (Fig. 2.10) subsequently demonstrated that the flow direction was indeed from the hypothalamus to the pituitary. Harris and Campbell showed that these capillaries are of the specialized fenestrated type also found in other secretory and absorptive organs (Harris and Campbell, 1966). This makes the blood–brain barrier permeable to larger molecules. Green and Harris used hypophysectomy, pituitary transplantation, and the placement of a barrier between the pituitary and the median eminence to prove the functional importance of the portal system (for historical references, see Meites, 1992; Raisman, 1997). The story goes that when Geoffrey Harris had succeeded in exposing the portal vessels with the microscopic operation he

BLOOD SUPPLY TO THE HYPOTHALAMUS AND PITUITARY …the central nervous system regulates the activity of the adenohypophysis by means of a humoral relay through the hypophysial portal vessels. Green and Harris (1947)

The portal system Arterial blood supply to the human hypothalamus is derived from terminal branches of the internal carotid and basilar arteries and from the anastomoses between them, which form the arterial circle of Willis or circulus arteriosus cerebri. Sir Thomas Willis (1621–75), who coined the name “Neurology,” never claimed to be the first to describe the circle, but he is still honored every year on St. Martin’s day, the day he died in 1675 (Wolpert, 1997). After describing the origin of the portal vessels in the median eminence in 1930, Popa and Fielding erroneously deduced that blood in the portal vessels flowed upwards, collecting blood from both lobes of the pituitary and sending it along the pituitary stalk to the hypothalamus. They even suggested that the blood would

Fig. 2.10. G.W. Harris (1913–1971). In 1955, he wrote: “… the central nervous system plays a key role in the response to stressful physical and psychological events in the environment, and does so by means of factors that are channeled through the hypothalamo-hypophysial portal vascular system to control the secretion of ACTH by the pituitary gland.” From Anderson E, Haymaker W (1974). Breakthroughs in hypothalamic and pituitary research. Prog Brain Res 41: 1–60, with permission.

18

F. KREIER AND D.F. SWAAB

had developed in a rat under anesthesia, he switched to a higher magnification by adding an extra lens in order to see in what direction the blood was moving. To his amazement and dismay he saw the blood cells moving upwards from the site of the pituitary below. He went to the pub and drank a lot, and it was only the next day that he realized that the extra lens had reversed “up” and “down,” and that he had in fact proven that the portal blood was streaming from the hypothalamus down to the pituitary.

RELEASING FACTORS …the central nervous system plays a key role in the response to stressful physical and psychological events in the environment, and does so by means of factors that are channeled through the hypothalamo-hypophysial portal vascular system to control the secretion of ACTH by the pituitary gland. G.W. Harris (1955) The discovery of the portal system and the acceptance of the concept of neurosecretion led to the “chemotransmitter hypothesis,” which stated that neurons synthesize humoral substances that are transported by the portal vessels and stimulate the pituitary gland to release its hormones. In 1955 Saffran and Schally and Guillemin and Rosenberg showed that a hypothalamic extract was able to release adrenocorticotropic hormone (ACTH) in vitro, and the former investigators named the putative compound corticotrophin releasing factor (CRF). In 1960 McCann reported the first evidence for a luteinizing-releasing factor in hypothalamic extracts and in 1961 Harris succeeded in inducing ovulation in rabbits with an hypothalamic extract. In the same year a thyrotropin-releasing factor (TRF) was reported to be present in hypothalamic extracts by Schreiber et al. (Meites, 1992). In 1968 a year before the chemical nature of TRH was clarified, a highly purified preparation of porcine TRF was injected into three hypothyroid cretins in Mexico and a detectable rise in thyroid-stimulating hormone was seen within 3 min, with a peak between 15 and 30 min (Schally and Gual, 2001). The compound was thus biologically active, also in human. Although it was successfully shown that hypothalamic extracts contained a CRF, the isolation and chemical nature remained elusive. Schally describes that he was exposed to sarcasm, skepticism, and even ridicule by American scientists and physicians in the endocrine field who did not understand the technical problems involved in the effort. The hypothalamic factors were likened to the Loch Ness monster and the Abominable Snowman of the Himalayas and the question arose

whether or not these factors were but a watery illusion. The laboratories of both Schally and Guillemin attacked the problem by arranging to procure the hundreds of thousands of hypothalami of domestic animals necessary for any realistic effort to purify useful quantities of hypothalamic hormones. The two groups, who had both been involved in this long and unsuccessful search for the chemical identity of CRH, independently worked out that TRH was the tripeptide: pGlu-His-Pro-NH2 (Burgus et al., 1969; Schally and Gual, 2001). The second hypothalamic factor to be structurally characterized was LHRH (or GnRH) by the group of Schally. Both Schally (Fig. 2.11) and Guillemin (Fig. 2.12) received the Nobel Prize for Physiology and Medicine in 1977, together with Rosalin Yalow, who received the prize for the discovery of the radioimmunoassay. CRH, the releasing factor that triggered the race, appeared to be a 41 amino acid peptide that

Fig. 2.11. Andrew Schally (1926). Although it was successfully shown that hypothalamic extracts contained a CRF, the isolation and chemical nature remained elusive. Schally describes that he was exposed to sarcasm, skepticism, and even ridicule by American scientists and physicians in the endocrine field who did not understand the technological problems involved in the effort. The hypothalamic factors were likened to the Loch Ness monster and the Abominable Snowman of the Himalayas and the question arose whether or not these factors were but a watery illusion. National Library of Medicine.

HISTORY OF HYPOTHALAMIC RESEARCH

Fig. 2.12. Roger Guillemin (1924) and Schally received the Nobel Prize for Physiology and Medicine in 1977, together with Rosalin Yalow, who received the prize for the discovery of the radioimmunoassay. National Library of Medicine.

was isolated from ovine hypothalamus and characterized until 1981, by Vale and coworkers.

NEUROPEPTIDES AS NEUROTRANSMITTERS OR NEUROMODULATORS: THE CENTRAL PATHWAYS INVOLVED …and evidence will be presented to show that posterior lobe extracts are far more potent when injected into the cerebral ventricles than by any other means of administration. Cushing (1932, p. 21) Traditionally, hypothalamic research has focused on “lower” functions, or, as is clear from the poetically phrased citation of Cushing (1932) at the beginning of the chapter. However, commercially available posterior lobe extract (obstetrical pituitrin), which Cushing (1932, p. 59) injected into the ventricle in 38 instances in 24 subjects convalescing (!) after pituitary adenoma operations, appeared to have a pronounced stimulatory effect, essentially parasympathetic in nature, and, apparently, central in origin (Fig. 2.13). Since intramuscularly injected pituitrin had a reverse effect, and no central effects were noted when the hypothalamus was affected by hydrocephaly or by a tumor, Cushing indeed had

19

Fig. 2.13. Showing the vasodilator and suderific effects, sparing the bone flap of a recent operation, of 2.5 mg of pilocarpine injected into the cerebral ventricles: an intraventricular injection of 1 mL of Pituitrin in susceptible persons gives an equally marked response. From Cushing, H., 1932. Papers relating to the pituitary body, hypothalamus and parasympathetic nervous system. Charles C. Thomas, Springfield, IL, fig. 25, p. 58.

solid evidence to propose a central action (i.e., on the hypothalamus) of neurohypophyseal extracts. These impressive observations passed into oblivion, perhaps because the scientific community chose to fight the “neurosecretion” concept proposed by the Scharrers (see earlier). Also, when the neurosecretion concept was finally accepted when Bargmann (1949) demonstrated the same Gomori-positive material in the neurohypophysis and in the neurons of the SON and PVN, this did not explain central effects of neurohypophysial hormones. In the concept of neurosecretion, Barry’s proposition (1954) concerning the existence of Gomori-positive endings outside the hypothalamus or, as he called them, “des synapses neuros
ecr
etoires,” could not be properly appreciated and was, therefore, eventually forgotten, a process likely to have been expedited by the fact that his articles were in French (Barry et al., 1954, 1961). Barry’s own conclusion was that the synaptic nature of these terminations could only be proven with electron microscopy, which could not be combined with the Gomori technique or electrophysiological techniques (Barry, pers. commun.). In addition, Legait and Legait (1958) and Barry et al. (1954, 1961) independently demonstrated

20

F. KREIER AND D.F. SWAAB

Fig. 2.14. David de Wied (1925–2004). When De Wied showed that posterior lobectomy of the pituitary in the rat resulted in an accelerated extinction of conditioned shuttlebox avoidance response, and, later on, that this behavioral deficiency could be alleviated by peripherally administered vasopressin (De Wied, 1965; Bohus and De Wied, 1966); these effects of vasopressin on memory processes were explained in terms of an endocrine concept. Courtesy of Willem Hendrik Gispen, with permission.

antidiuretic activity in the habenular region of the chicken and oxytocinergic activity in the area of the amygdala in the mouse, respectively, by means of bioassays. However, these excellent studies were also neglected (Barry et al., 1954, 1961; Legait and Legait, 1958). When De Wied (1965, Fig. 2.14) showed that posterior lobectomy of the pituitary in the rat resulted in an accelerated extinction of conditioned shuttle-box avoidance response, and, later on, that this behavioral deficiency could be alleviated by peripherally administered vasopressin (De Wied, 1965; Bohus and De Wied, 1966), these effects of vasopressin on memory processes were explained in terms of an endocrine concept. Because the effect of removal of the endocrine “gland” was abolished by peripheral administration of the hormone, the peripheral circulation (Thompson and de Wied, 1973) or the cerebrospinal fluid was proposed (van Wimersma et al., 1975) as the routes via which vasopressin could reach its central sites of action. In the latter experiment intraventricular but not intravenous administration of vasopressin–antiserum resulted in the prevention of the avoidance response, suggesting a physiological role for endogenous vasopressin in the consolidation of information. High vasopressin levels in the CSF of hypophysectomized rats (Dogterom et al.,

1977) were, however, apparently contradictory with the absence of avoidance response. This suggested that the CSF neurohypophyseal hormone content was not of primary importance for this behavioral paradigm. In the meantime, the application of a new technique, i.e., immunocytochemistry, first by our group, and subsequently by others, had revealed yet another site of production for vasopressin—the suprachiasmatic nucleus (SCN) (Swaab et al., 1975). This paper was awarded the “Citation Classic” status, since it not only showed a new, nonneuroendocrine, site of vasopressin production but also described a procedure for purifying polyclonal antibodies in order to remove cross-reactivity with related compounds. Immunocytochemistry for vasopressin and neurophysin subsequently led to the rediscovery of Barry’s (1954) and Legait and Legait (1957) extrahypothalamic pathways and showed that they indeed contained vasopressin and oxytocin (Legait and Legait, 1957). They also appeared to be much more extensive than the Gomori studies suggested and were found to terminate in various brain structures, ranging from the olfactory bulb to the spinal cord (Swanson, 1977; Buijs et al., 1978; Sofroniew and Weindl, 1978). Subsequently, a preembedding staining technique was developed that enabled the immunoelectron microscopical visualization of the synaptic peptidergic terminations, making a central function of neuropeptides as neurotransmitter or neuromodulator very probable (Buijs and Swaab, 1979). Indeed soon after that observation synaptic release of both vasopressin and oxytocin was demonstrated (Buijs and Van Heerikhuize, 1982). Recently it was indicated that oxytocin neurons might not have only an axon to the neurohypophysis but also different axon collaterals of the same neuroendocrine neuron to different brain areas (Zhang et al., 2020). The main sources of these vasopressinergic and oxytocinergic pathways were thought to be the SCN and the PVN. Bilateral lesions of the rat PVN did not eliminate the entire vasopressin content of the brainstem and the spinal cord but caused a reduction of 50% and 80%, respectively, while oxytocin almost completely disappeared. This suggested that the PVN was the only site of origin of extrahypothalamic oxytocin but not of vasopressin (Lang et al., 1983). Once again, Barry et al. (1973) were the first to use colchicine in order to visualize LHRH perikarya in the guinea pig brain (Barry et al., 1973). This compound prevented axonal transport of the neuropeptides, while their production continued. Since that time, colchicine application has been widely used to detect neuropeptide-synthesizing cell bodies (e.g., Hokfelt et al., 1980) and a number of extrahypothalamic production sites of vasopressin were indeed found (de Vries et al., 1986). Coexistence of different neuroactive substances was found both in the peripheral and the central nervous system. A major type of coexistence

HISTORY OF HYPOTHALAMIC RESEARCH found in the brain was that of a biogenic amine and a peptide (Hokfelt et al., 1986). Improvements of the immunocytochemical techniques also revealed additional cell bodies containing vasopressin or oxytocin in a number of human brain structures (Swaab, 2003). Using tritiated vasopressin, oxytocin, and agonists, high-affinity binding sites were found in a number of human brain areas (Loup et al., 1989, 1991) and electrophysiological studies strengthened the concept that neuropeptides could act as neurotransmitters or neuromodulators (Urban, 1987).

Central effect of neuropeptides Vasopressin and oxytocin: Yin-yang hormones. Legros (2001) Currently hundreds of neuropeptides are known that regulate a large number of very different central processes. Only a few will be mentioned here as examples. The central vasopressinergic fibers may be involved in blood pressure and temperature regulation, regulation of osmolality, and in the stress axis, i.e., in corticosteroid secretion, and may thus influence stress, cognitive functions, aggression, paternal behavior, and social attachment (Legros et al., 1980; de Wied and van Ree, 1982; Buijs et al., 1983; Fliers et al., 1986; Holsboer et al., 1992; Insel, 1997). Intranasal vasopressin differentially affects social communication in men and women (Thompson et al., 2006). It is, therefore, of great interest that polymorphisms in the vasopressin V1a receptor are related to the autism spectrum (Yirmiya et al., 2006; see also Chapter 9 in Volume 182). Oxytocinergic central pathways are involved in reproduction, pairbonding, cognition, tolerance, adaptation, sedation, and in the antistress effects that occur during lactation via the autonomic system in the regulation of cardiovascular and respiratory functions (Uvnas-Moberg, 1997; Gutkowska et al., 2000; Mack et al., 2002). Oxytocin also has central effects on food intake, and oxytocin neurons are considered to be the putative satiety neurons in eating behavior (Swaab et al., 1995; see also Chapter 6 in Volume 180). Intranasal administration of oxytocin improves mind-reading (Domes et al., 2007). Possibly related to this central effect is the finding that polymorphisms of the receptor for oxytocin are involved in the susceptibility for autism (Wu et al., 2005). New neuropeptides keep being found and new functions reported. A recent example in the field of neurology is the involvement of the neuropeptide orexin or hypocretin in cataplexy, a symptom that occurs in narcolepsy. Positioning cloning has established that an autosomal recessive mutation of the hypocretin receptor-2 gene (HCRT2) is responsible for the genetic form in well-established canine models, i.e., Doberman pinschers and Labrador retrievers (Kadotani et al., 1998; Lin et al., 1999). In addition,

21

hypocretin knockout mice exhibit narcolepsy (Chemelli et al., 1999) and modafinil, an antinarcoleptic drug, activates orexin-containing neurons. Hypocretin neurons were found to be located in the perifornical lateral hypothalamic region in the human brain (Peyron et al., 2000), while in human narcolepsy these hypocretin neurons were substantially (85%–95%) reduced (Peyron et al., 2000; Thannickal et al., 2000; Moore et al., 2001), and hypocretin-1 levels have been reported to be dramatically decreased in the CSF of the majority of patients suffering from narcolepsy (Nishino et al., 2000). However, in contrast to animal models, most cases of narcolepsy are not familial (Siegel, 1999). It is unlikely that a high proportion of human narcoleptics have a mutation or a polymorphism responsible for narcolepsy, either in the Hcrt gene or in the HCRT-1 or -2 receptors (Olafsdottir et al., 2001) and an autoimmune destruction of the hypocretin cells in the hypothalamus has been proposed as the cause for cataplexy (see Chapter 12 in Volume 181).

CONTROL OF AUTONOMIC FUNCTIONS BY THE HYPOTHALAMUS The first evidence: An accident or a logical step in a chain of events in the 1900s in Vienna? About 100 years ago, in Vienna, the neurologist and psychiatrist Karplus, together with the physiologist Kreidl, conducted the first experiment on the role of the hypothalamus in the control of the autonomic nervous system (Wang, 1965; Triarhou, 2018). They induced dilatation of the pupils by electrical stimulation of the base of the cat brain in an area caudal to the optic tract and medial to the cerebral peduncle. This effect was still present after removal of the hemispheres but could be abolished by section of the cervical sympathetic trunks (Karplus and Kreidl, 1909, Fig. 2.15). Between 1909 and 1928, Karplus and Kreidl published a series of eight experiments in German under the name “Zwischenhirnbasis und Halssympathicus,” exploring the effect of electrical stimulation of the hypothalamus on the sympathetic nervous system (Karplus and Kreidl, 1909, 1927; Ingram, 1939; Anderson and Haymaker, 1974). They found a series of sympathetic effects: papillary dilatation, lacrimation, sweating, increase in heart rate, inhibition of gut motility, increased amplitude of respiration, contraction of the urinary bladder, and hypertension, the latter even after removal of the pituitary and the adrenals. The hypothalamic control of vital functions via the autonomic nervous system, even independently of the pituitary or the cortex, was proven. We do not know why Karplus and Kreidl performed these experiments. Their rather short papers, the earlier ones even lacking an introduction or any references, and the loss of the archives of the Physiology Department

22

F. KREIER AND D.F. SWAAB

Fig. 2.15. Karplus and Kreidl (1909): Electrical stimulation of the hypothalamus (S) induced dilatation of the pupils. This effect was still present after removal of the hemispheres but could be abolished by section of the cervical sympathetic trunks. Between 1909 and 1928, Karplus and Kreidl published a series of eight experiments in German under the name “Zwischenhirnbasis und Sympathicus,” exploring the effect of electrical stimulation of the hypothalamus on the sympathetic nervous system (Karplus and Kreidl, 1909, 1927; Ingram, 1939; Anderson and Haymaker, 1974). They found a series of sympathetic effects: papillary dilatation, lacrimation, sweating, increase in heart rate, inhibition of gut motility, increased amplitude of respiration, contraction of the urinary bladder and hypertension, the latter even after removal of the pituitary and the adrenals. From Karplus J, Kreidl A (1909). Gehirn und Sympathicus. I. Zwischenhirnbasis und Halssympathicus. Pflugers Arch 129: 138–144.

in Vienna during World War II, leave this question open. It has been suggested that they just hit the hypothalamus by accident while working on the optic nerve (Wang, 1965). However, reviewing the traditions of the scientific community at the beginning of the 20th century and taking into account who the colleagues of Karplus and Kreidl in Vienna were (see later), this “accident” happened at a logical moment.

The “peripheral” nervous system The story of the peripheral nervous system started in France, 100 years before the experiments of Karplus

and Kreidl. In 1800, Bichat (Fig. 2.16) conducted anatomical studies and concluded that the body has two nervous systems: first, the “animal” system originating in the brain and directed to the outer world, formed by education and habit, dealing with intellect and passions (Bichat, 1800; Ackerknecht, 1974), and second, the sympathetic chain of ganglia. Bichat proposed the existence of an “organic” nervous system, directed to our inner body, dominating the viscera by a chain of “little brains” next to the vertebrae as its centers, totally independent from education or habit. Today, this description reminds us of the dualism of body and mind. However, with the

HISTORY OF HYPOTHALAMIC RESEARCH

23

Fig. 2.16. Xavier Bichat (1771–1802). (A) Portrait of Bichat. Pierre-Maximilien Delafontaine, 1799, Palace of Versailles. (B and C) Bichat recognized the existence of an independent nervous system projecting to the internal organs. With the paradigm of body humors and vital spirits that pass through the nervous vessels into the body still present in 1800, the idea of a peripheral nervous system as an entity that is independent from the brain was an innovative perspective. From Bichat F (1800). Recherches physiologiques sur la vie et la mort, Brosson, Gabon et Cie, Paris.

paradigm of body humors and vital spirits that pass through the nervous vessels into the body still present in 1800, the idea of a peripheral nervous system as an entity that is independent from the brain was an innovative perspective. Bichat’s follower Reil named this “organic nervous system” the “vegetative nervous system” (Ackerknecht, 1974). The availability of the microscope in the 19th century allowed scientists a closer look at the nerve endings in the organs. In 1840 Remak described the accumulation of nerve fibers in heart and urinary bladder and discovered peripheral ganglia in target organs (Remak, 1838). Ernst Heinrich and Eduard Weber found, in 1845, that

stimulation of the vagal nerve can induce heart arrest, discovering inhibition as a new concept in neurophysiology (Weber and Weber, 1846; Ackerknecht, 1974). In 1840 Henle stated that nerves control smooth muscles in blood vessels. In 1851 Claude Bernard conducted experiments on heat production and described how cutting the sympathetic nerves induced vasodilation (Bernard, 1851; Montastruc et al., 1996). One year later, in 1852, Brown-Sequard (Fig. 2.17, considered to be the father of neuroendocrinology, Aminoff, 2010) proved, inducing vasoconstriction by stimulating the cut end of sympathetic nerves, that Henle was right (Brown-Sequard, 1852).

24

F. KREIER AND D.F. SWAAB

Fig. 2.17. Brown-Sequard (1817–1894). In 1852, BrownSequard proved, inducing vasoconstriction by stimulating the cut end of sympathetic nerves. Photograph by unknown artist, University of Paris, ca. 1880.

In summary, in the first half of the 19th century, anatomical and physiological studies led to the notion that a vegetative nervous system exists with a stimulating or inhibiting effect on viscera.

Claude Bernard: From the periphery back to the brain Fifty years before Karplus and Kreidl, Claude Bernard (Fig. 2.18; Garrison, 1917) did his famous experiments on the brainstem (Bernard, 1850; Vitturi and Sanvito, 2020). In 1850 he described that after the floor of the fourth ventricle was punctured, his animals developed hyperglycemia, which could be prevented by splanchnic nerve dissection. This milestone in science connected the brain to the periphery and the nervous system to metabolism, which could now be studied due to the invention of new biochemical methods in the first half of the 19th century.

The diencephalon and vital body functions At the same time, clinical observations drew the attention of researchers to the diencephalon. In 1840 von Mohr described the rapid development of obesity in a case of a hypophysial tumor (Mohr, 1840), and in 1842, Rokitansky noticed that lesions at the base of the brain induce

Fig. 2.18. Claude Bernard (1813–1878). Fifty years before Karplus and Kreidl, Claude Bernard did his famous experiments on the brainstem (Bernard, 1850; Vitturi and Sanvito, 2020). In 1850, he described that after the floor of the fourth ventricle was punctured, his animals developed hyperglycemia, which could be prevented by splanchnic nerve dissection. This milestone in science did not only connect the brain to the periphery, but also the nervous system to metabolism, which could now be studied due to the invention of new biochemical methods in the first half of the 19th century. From Garrison FH (1917). An introduction to the history of medicine, WB Saunders Company, Philadelphia.

gastric ulcers (Rokitansky, 1842). Schiff confirmed these clinical findings in 1856 (Schiff, 1856, Fig. 2.19), who published his experiments on lesions of the diencephalon that induced gastric perforation, which could be blocked by vagotomy but not by spinal cord section (Schiff, 1856). It was also Schiff who discovered that the stimulation of the splanchnic nerve induced contraction of abdominal vessels. Some years of research later, in 1867, he proposed a joint antagonistic control of the stomach by the sympathetic and parasympathetic branch and proposed that an unbalance of the two branches could induce gastric ulcers (Schiff, 1867). In 1862 Bezold found a motor heart center in the medulla oblongata, and in 1891, Ott performed experiments with the thermal stimulation of the hypothalamus and found an effect on heat production, breathing, and blood vessel tone. In 1889 20 years after the observations of Schiff, Langley combined all the data of the previous 50 years

HISTORY OF HYPOTHALAMIC RESEARCH

Fig. 2.19. Mortiz Schiff (1823–1896) explored the role of the autonomic nervous system in 1856. He studied lesions of the diencephalon that induced gastric perforation, which could be blocked by vagotomy but not by spinal cord section (Schiff, 1856). It was also Schiff who discovered that the stimulation of the splanchnic nerve induced contraction of abdominal vessels. Some years of research later, in 1867, he proposed a joint antagonistic control of the stomach by the sympathetic and parasympathetic branch, and proposed that an unbalance of the two branches could induce gastric ulcers. Painting by Nicolai Ge 1876.

and formulated a theory of what he called “the autonomic nervous system” (Langley, 1899, 1921, Fig. 2.20). He understood that antagonism of the branches was a core attribute of the system and investigated the combined effect of the sympathetic and parasympathetic system on the heart, stomach, and blood vessels.

The emotional diencephalon In the 1870s neurophysiology was a highly appreciated science, as illustrated by the fact that Freud tried to become a researcher in basic neurosciences in Vienna. Unfortunately, he was unable to finance his dream (Fig. 2.21). By this time, it was recognized that there must be a place in the brain where emotions are produced, and that these emotions are accompanied by a typical reaction pattern, which is the consequence of autonomic activity. Again, this important step toward a comprehension of brain

25

function starts with brilliant clinicians who recognized syndromes in patients. Nothnagel (1879) (Karplus worked with him) described that patients with lesions in the diencephalon display a discrepancy between emotional expression and voluntary movements of facial muscle. In 1887 Bechterew published a paper on the “Bedeutung der Sehh€ugel” (function of the midbrain), which covered experimental and pathological data on the diencephalon. He concluded that in the diencephalon centers are present that influence automatic and reflector functions, which are functioning in different “soul affects” (Bechterew, 1887). Sadly enough, his sharp view on these clinical symptoms probably killed him, when he was summoned to the Kremlin by the depressed Stalin in 1927. Soon after he had diagnosed that Stalin suffered from “grave paranoia,” Bechterev died, of influenza, under suspicious circumstances. When in 1892, Goltz described his 18-month experiment with a dog after removal of the hemispheres, he paid much attention to the unexpected finding of quick and strong affective reactions without clear inducement. Most of his colleagues, however, did not appreciate his findings and tried to refute them (Goltz, 1892). In 1888, almost 40 years after Bernard’s experiments on the brainstem and diabetes, Bromwell observed that a tumor at the base of the brain resulted in lipodystrophy and genital failure, a syndrome he named adiposogenital dystrophy (Bramwell, 1888). Ten years later, in 1898, Probst described a cat that was happily snoring, resting a lot and interested in food after removal of both hemispheres. This animal with a lesion of the posterior hypothalamus rapidly developed obesity. However, Probst was sure that this “side finding” must be coincidental, and decided it was not important (Probst, 1898). Only in 1928 Bard explored this phenomenon further with a removal of the hemispheres in cat (Bard, 1928). He confirmed the findings of Goltz from 1892 and called the unpredictable behavior of the animals “sham rage.” Moreover, he introduced the concept of integration of autonomic functions within the hypothalamus. In 1929 Cannon (Fig. 2.22) described a sympathetic reaction pattern in fear, consisting of dilatation of the pupils, tachycardia, constriction of visceral blood vessels resulting in hypertension, enlargement of the bronchioles, sweating, and hyperglycemia (Cannon, 1929). Thus scientists understood that the diencephalon was capable of creating emotions which were altered after removal of the cortex. These emotions are transported by the ANS, and moreover, are linked to metabolism.

A conflict in Vienna In the year 1900 (100 years after Bichat) Fr€ohlich and Babinski published a syndrome that bore some

26

F. KREIER AND D.F. SWAAB

Fig. 2.20. (A) John N. Langley (1852–1925). Wellcome Collection. (B and C) In 1889, Langley combined all the data of the previous 50 years and formulated a theory of what he called “the autonomic nervous system” (Langley, 1899). He understood that antagonism of the branches was a core attribute of the system and investigated the combined effect of the sympathetic and parasympathetic system on the heart, stomach, and blood vessels. From Langley JN (1921). The autonomic nervous system. Heffer, Cambridge.

resemblance to adiposogenital dystrophy. The so-called Babinski–Fr€ ohlich–Syndrome was characterized by feminine obesity and sexual infantilism, atrophy, or hypoplasia of the gonads and altered secondary sex characteristics (Babinski, 1900; Fr€ ohlich, 1901). Babinski and Fr€ ohlich recognized a tumor at the base of the brain and concluded that a hypophysial tumor was the cause of the symptoms. In 1904, Erdheim (Fig. 2.23), after a careful analysis of the diencephalon in these patients, published a study from which he concluded that it was not the pituitary but the compressed base of the hypothalamus that might be the problem (Erdheim, 1904). Within a few years, this conflict was vigorously discussed in the scientific community, inducing an important impulse for hypothalamic research (Anderson and Haymaker, 1974). The neurosurgical school founded by Cushing studied the role of the pituitary by its surgical removal by a transseptal transsphenoidal approach (Cushing, 1912). As they were confident of their operating skills, they had no doubt that the hypothalamus was not affected by the hypophysectomy

and chose the side of Babinski and Fr€ohlich. However, in 1909 and 1912, Aschner, also from Vienna, published a long series of hypophysectomies in dogs in a 150-page paper and confirmed the findings of Erdheim (Aschner, 1909, 1912). Moreover, he showed that stimulation of the hypothalamus by lesion, electric current, and pharmacological compounds could induce glycosuria, hypertension, and contraction of bladder and uterus (Aschner, 1912). Even more evidence came from Camus and Roussy, who showed, with a different technique, that animals did not automatically exhibit obesity following an hypophysectomy, but only if in addition the hypothalamus had been punctured (Camus and Roussy, 1913). At the same time, in Vienna, Meyer, the head of the pharmacology department, and with Loewi (Fig. 2.24) as his assistant, translated the Langley theory of the antagonistic autonomic nervous system into pharmacological experiments on neurotransmission, with a huge impact on clinical medicine (Pick, 1941). Loewi reinforced his studies on the autonomic nervous system,

HISTORY OF HYPOTHALAMIC RESEARCH

27

Fig. 2.21. (A) Inscription of a book (1891) of Sigmund Freud (1856–1939) for Hermann Nothnagel (1841–1905), who described that patients with lesions in the diencephalon display a discrepancy between emotional expression and voluntary movements of facial muscles. From 1891 to 1903 Johann Paul Karplus (1866–1936), worked with Nothnagel. Karplus and Freud where friends. Illustration of the fact, Karplus and Kreidl’s experiments were not a coincidence but a logical step in the chain of experiments that had been done before in Vienna. University Library of the Medical University Vienna. (B) Hermann Nothnagel (Sport and Salon, 1902). (C) Johann Paul Karplus. Triarhou LC 2018, Photo by Max Schneider, 1908. From Julius Wagner-Jauregg’s Lebenserinnerungen, Springer, 1950, with permission. Continued

28

F. KREIER AND D.F. SWAAB

Fig. 2.21—Cont’d

Fig. 2.22. In 1929, Walter B. Cannon (1871–1945) described a sympathetic reaction pattern in fear, consisting of dilatation of the pupils, tachycardia, constriction of visceral blood vessels resulting in hypertension, enlargement of the bronchioles, sweating, and hyperglycemia (Cannon, 1929). Thus, scientists understood that the diencephalon was capable of creating emotions which were altered after removal of the cortex. These emotions are transported by the ANS, and moreover, are linked to metabolism. Wellcome Collection.

and in 1921 discovered the function of acetylcholine in the neurotransmission of the vagus in the heart (Loewi, 1921). Against this background we propose that Karplus and Kreidl’s experiments were not a coincidence but a logical step in the chain of experiments that had been

done before in Vienna. These experiments were also not the last of their kind in Vienna: only 3 years later, in 1912, the urologist Lichtenstern proved that the urinary bladder contracted when the hypothalamus was stimulated (Lichtenstern, 1912).

HISTORY OF HYPOTHALAMIC RESEARCH

29

The hypothalamus and the autonomic nervous system: From the isolated pupil reflexes to complex body functions

Fig. 2.23. Jakob Erdheim (1874–1937). In 1904, Erdheim, after a careful analysis of patients with a tumor at the base of the brain, concluded that it was not the pituitary but the compressed base of the hypothalamus that might cause the Babinski–Fr€ohlich–Syndrome (Erdheim, 1904). Jakob Erdheim in his institute at Lainz Hospital, ca. 1933. Josephinum – Ethics, Collections and History of Medicine, MedUni Vienna, with permission.

After Karplus and Kreidl and others had proven that the hypothalamus and the autonomic nervous system controlled an isolated function such as pupil reflexes, the next question to be answered was how the hypothalamus coordinated complex body functions. In 1914 after stimulation experiments, Isenschmid concluded that there was a body heat center in the posterior hypothalamus (Isenschmid and Schnitzler, 1914). In this paper the German researcher admitted, rather frustratedly, that he had overlooked the work by Ott from 1891, which had been published in English, in the United States. For us it is, however, astonishing how well the research community generally exchanged ideas and results without the modern means of communication, such as the internet, but managed to communicate efficiently, by traveling and reading papers in multiple languages. The exclamation of Isenschmid illustrates that, in 1914, science had developed to a point where the need for international collaboration was great. However, that same year saw the outbreak of the First World War and for the following 4 years many scientists around the world had other things to worry about than the lab. Scientifically speaking, the world stood still.

Between the wars: The rise of endocrinology Another scientific breakthrough happened 3 years after the First World War by Banting and colleagues (Fig. 2.25), a Canadian army surgeon who had had to amputate gangrenous limbs in Europe between 1914 and 1918.

Fig. 2.24. Otto Loewi (1873–1961): Studied the autonomic nervous system and discovered proofed the effect of acetylcholine in 1921, called “Vagusstoff.” Institute of Pharmacology, Graz.

30

F. KREIER AND D.F. SWAAB

Fig. 2.25. Frederick Banting (right, 1892–1941) and Charles Best (left, 1899–1978). In 1922, they published on the subject of insulin: that it reduced serum glucose levels, both in normoglycemia and hyperglycemia (Banting and Best, 1922). The concept of a small amount of a blood-borne factor that changes the metabolic state of the body profoundly was so revolutionary that not everybody believed it straightaway. To prove that insulin really mattered, Banting and colleagues studied the effect of insulin in well-known animal models of hyperglycemia immediately after the first publication. In the same year (1922) they published the effect of insulin in experimental diabetes. One of the studies combined Claude Bernard’s “piqure” (puncture of the fourth ventricle), i.e., hyperglycemia induced by a central lesion, with injection of insulin, showing that insulin overruled the effect of the piqure (Banting et al., 1922). From Thomas Fisher Rare Book Library, University of Toronto, with permission.

When his attempt to set up an orthopedic practice failed, he became a scientist. He read every article he could get hold of; as no patients showed up he had plenty of time to read anyway. One article he read was a detailed pathological report on necrotic processes in the pancreas in patients with bile stones. It seemed that the exocrine function of the pancreas was the first to degenerate, while the islets of Langerhans were preserved for some weeks (Barron, 1920). Everyone tried to extract insulin in those days, which was impossible because the exocrine enzymes destroyed it during the extraction process. Banting understood that if he first ligated the pancreatic ducts and then waited for a while, he might be able to extract insulin. He and his colleagues were right, and in the end, they saved the lives of millions of people (Fig. 2.26). In 1922 he published on the subject of insulin: that it reduced serum glucose levels, both in normoglycemia and hyperglycemia (Banting and Best, 1922). The concept of a small amount of a blood-borne factor that changes the metabolic state of the body profoundly was so revolutionary that not

everybody believed it straightaway. To prove that insulin really mattered, Banting and colleagues studied the effect of insulin in well-known animal models of hyperglycemia immediately after the first publication. In the same year (1922) they published the effect of insulin in experimental diabetes. One of the studies combined Claude Bernard’s “piqure” (puncture of the fourth ventricle), i.e., hyperglycemia induced by a central lesion, with injection of insulin, showing that insulin overruled the effect of the piqure (Banting et al., 1922). Banting and colleagues’ publication is of significant historical interest for us today for several reasons: it demonstrates that the role of the brain in energy homeostasis was a common scientific idea at the beginning of the 20th century. Brain manipulation to induce hyperglycemia was the golden standard. Moreover, this paper was the first to demonstrate the important role of hormones in metabolism that are capable of overruling the brain. Thereby, this publication preludes the development of endocrinology as a concept of the main regulator of metabolism, superior to the brain. Up to the

HISTORY OF HYPOTHALAMIC RESEARCH

31

Fig. 2.26. Letter to Frederick Banting after the discovery of diabetes. Banting and colleagues’ publication is of significant historical interest for us today for several reasons: it demonstrates that the role of the brain in energy homeostasis was a common scientific idea at the beginning of the 20th century. Brain manipulation to induce hyperglycemia was the golden standard. Moreover, this paper was the first to demonstrate the important role of hormones in metabolism that are capable of overruling the brain. Thereby, this publication preludes the development of endocrinology as a concept of the main regulator of metabolism, superior to the brain. Up to the discovery of leptin in 1994, the brain had no significant role in the concepts of metabolism. For over 70 years endocrinology was to remain the science of brainless metabolism. From Thomas Fisher Rare Book Library, University of Toronto, with permission.

discovery of leptin in 1994, the brain had no significant role in the concepts of metabolism. For over 70 years endocrinology was to remain the science of brainless metabolism.

Lipodystrophy: A neurological disease Thus endocrinologists would become the experts on metabolism in the following century. However, in the 1920s this was not yet the case. During this decade, the neurologists were still the experts on lipodystrophy (a disease with a symmetrical atrophy and hypertrophy of fat tissue). Still, in 1928 a case report article published in Brain about lipodystrophy was written by the neurologist Ziegler (1928, Fig. 2.27). In 1921 the psychiatrist/neurologist Klien presented a case of symmetrical lipodystrophy and proposed a role for the hypothalamus in the development of the syndrome (Klien, 1921). He argued that it was hard to explain a syndrome of disturbed symmetrical body fat distribution by blood-borne factors and supposed a role for the sympathetic nervous system, with a difference in local tone as the mechanism for the local differences in fat mass. He argued that “ohne K€ unstelei” (without being artificial) one would be able to think of regions in the hypothalamus that are specialized in the control of fat tissue in certain body region. If one of these regions in the hypothalamus would be malfunctioning, a disturbed body fat distribution would occur. Moreover, he hypothesized that some regions of the hypothalamus would be

specialized in accumulating fat tissue and other tissue in the mobilization of energy stores. Today, we know that he was right: particular hypothalamic regions and neurons are specialized in anabolic or catabolic functions, and in the control of different fat compartments via the autonomic nervous system (Schwartz et al., 2000; Kreier et al., 2002, 2006). In 1924 Goering studied the relation of the “vegetative” nervous system and fat tissue under supervision of the Internist-Neurologist M€uller (see next paragraph) and published this in the textbook “Die Lebensnerven” (the vital nerves) (Goering, 1924, Fig. 2.28). This impressive publication illustrates how far the clinicians combining the practice of internal medicine and neurology were in their understanding of the regulation of metabolic processes. Goering asked to stop calling fat a passive tissue, since it is an active gland. From a careful review of the literature and observation of lipodystrophy patients from M€ullers’ clinic, she concluded that the sympathetic nervous system plays an important role in lipolysis. However, Goering discovered that in some cases the sympathetic nervous system was not involved in symmetrical lipodystrophy and speculated that also trophic nerves should exist that stimulate lipogenesis. The authors infer that “Fettanordnung” (body fat distribution) should be under control of hypothalamic centers, specialized by function: lipogenesis, lipolysis, and body compartment regulated by specialized hypothalamic regions. Finally, Goering and M€uller recognized the effect of sex hormones on body

32

F. KREIER AND D.F. SWAAB

Fig. 2.27. Ziegler et al., Brain (1928). Publication on lipodystrophy. Illustration of the fact that up to the 1920s neurologists did study lipodystrophy and published it in neurology journals. Later, this field was almost exclusively covered by endocrinologists.

fat distribution and theorized that this effect might be centrally mediated via the hypothalamus and the autonomic nervous system. Nowadays, 80 years later, a number of the ideas of Goering and M€ uller have been confirmed: the sympathetic nervous system does stimulate lipolysis (Bamshad et al., 1998), the trophic nerve is the vagus, inducing lipogenesis (Kreier et al., 2002), and fat tissue in different compartments is indeed differentially innervated up to the hypothalamus (Kreier et al., 2002, 2006).

“Old-fashioned” internist-neurologists and the feeling of being hungry Neurology as a specialty derived from internal medicine in the 19th century. At the beginning of the 20th century, many physicians were both internists and neurologists. The internist M€ uller (Fig. 2.29), for instance, published

a series of neurological papers on the role of the hypothalamus and the ANS in metabolism. M€uller warned that neurology could not survive as a specialty separate from internal medicine (Neund€orfer and Hilz, 1998). He was interested in how we actually feel that we are hungry or thirsty. In two remarkable but almost forgotten papers € from 1915, “Uber die Hungerempfindung” (On the feel€ ing of being hungry), and 1920 “Uber den Durst und die Durstempfindung” (On thirst and the feeling of being thirsty), he concludes that, first, the hypothalamus senses a shortage of energy or volume in the passing blood (M€uller, 1915, 1920, Fig. 2.30). If energy or volume needs to be replaced, the hypothalamus stimulates the vagus nerve, resulting in contractions of the esophagus and stomach, after that the brain registers this as hunger or thirst. Thus the full feeling of being hungry or thirsty is produced by

HISTORY OF HYPOTHALAMIC RESEARCH

33

Fig. 2.28. Goering, “Die Lebensnerven” (The vital nerves), Mueller editor (1924). This paper from a PhD student of Mueller reviews the evidence for an autonomic control of fat tissue. The sympathetic control of fat tissue with a catabolic effect is taken for granted; however, Dora Goering speculates that there should be a “trophic nerve supply” to fat tissue inducing anabolic processes.

visceral contractions, as a secondary process, after a need that was recognized in the brain before. The way M€ uller describes the interaction between brain and body bears a striking resemblance to Damasio’s theories on emotional feedback to the brain (Damasio, 2003).

Further exploration of hypothalamic– autonomic interaction: New methods In 1928 Wang and colleagues confirmed the findings of Karplus and Kreidl that after electrical stimulation of the hypothalamus a galvanic skin response occurs in the footpads, which could be abolished by sectioning the sympathetic nerves to the limb (Wang et al., 1929; Ingram, 1939). Subsequently, Grafe published a study on the relationship of hypothalamic lesions and

metabolism (Grafe and Gr€unthal, 1929). He observed an obesity syndrome with reduced basal metabolism after unilateral lesions of the posterior hypothalamus that was even present following correction for temperature and feeding. In 1930 Himwich observed a rise in blood glucose after electrical stimulation of the hypothalamus (Himwich and Keller, 1930). In the same year, Beattie and colleagues combined neuroanatomy and physiology and showed that nerve tract degeneration occurs after lesions to the posterior hypothalamus (Beattie et al., 1930a,b). Moreover, they showed by acute experiments with electrocardiograms that stimulation of the same area induces adrenergic secretion and extrasystoles by changes in the A–V time. In a next step, they demonstrated that the changes in heart conductance were

34

F. KREIER AND D.F. SWAAB

Fig. 2.29. Neurology as a specialty derived from internal medicine in the 19th century. At the beginning of the 20th century, many physicians were both internists and neurologists. The internist M€uller (Fig. 2.28), for instance, published a series of neurological papers on the role of the hypothalamus and the ANS in metabolism. M€uller warned that neurology could not survive as a specialty separate from internal medicine. From Neund€orfer B, Hilz MJ (1998). Ludwig Robert M€uller (1870-1962)—a pioneer of autonomic nervous system research. Clin Auton Res 8: 1–5, with permission.

present even in adrenalectomized animals. In 1932 Beattie and colleagues stimulated the tuber cinereum and observed stomach peristalsis, gastric secretion, increased urinary bladder tone, and changed heart rate and concluded that the posterior hypothalamus controls the sympathetic nervous system, while the anterior hypothalamus controls the parasympathetic nervous system (Beattie, 1932). In 1930 Hess started to use freely moving animals in his stimulation experiments on circulation and breathing, aiming at more physiological conditions for his experiments (Hess, 1930, 1931). Already in 1908, 1 year before Karplus and Kreidl, the Englishmen Horsley and Clarke had developed a stereotaxic apparatus studying cerebellar function in monkey as “by this means every cubic millimeter of the brain could be studied and recorded” (Horsley and Clark, 1908). It took the scientific community about 20 years to understand the advantages of precisely targeted lesions. In 1928 Lewy was among the first to use this apparatus, studying the effect of electrical stimulation of the diencephalon on papillary dilatation, salivation, and lacrimation (Lewy, 1928; Ingram, 1939). In the following years the Horsley–Clark stereotaxic apparatus was used to explore the hypothalamus systematically and precisely. From 1930 to 1933, Ingram and Ranson further evaluated pupil reflex and found that dilatation could be induced, especially by stimulating the lateral hypothalamus (Ingram et al., 1931; Ranson and Magoun, 1933). In 1934 by “exploring the hypothalamus millimeter by millimeter” with electrical stimulation, they found pupil dilatation, hypertension, increased respiration rate from the lateral hypothalamus, and, in contrast, urinary bladder contraction, hypotension, bradycardia, and decreased respiratory rate by stimulation of the anterior hypothalamus (Ranson et al., 1934). In the same year, Raab explored the effect of pituitrin on the development of a fatty liver after its injection into the lateral ventricles

Fig. 2.30. Mueller, Deutsche Medizinische Wochenschrift (1915): Publication of the clinician Mueller, specialized as many colleagues up to the 1960s in both internal medicine and neurology. His article about “Hungerempfindung” (the feeling of being hungry) states that the hypothalamus feels the need for calories, activates the viscera through the ANS, which is felt by us as hunger.

HISTORY OF HYPOTHALAMIC RESEARCH

35

Fig. 2.31. Walle Nauta (1924–1994) described the anatomy of the hypothalamus and its projections to the ANS by neuronal tracing techniques. From Anderson E, Haymaker W (1974). Breakthroughs in hypothalamic and pituitary research. Prog Brain Res 41: 1–60. Courtesy MIT Museum, with permission.

Fig. 2.32. Johannes (Hans) Arie¨ns Kappers, former director of the Netherlands Brain Institute, explored the sympathetic innervation the pineal: After removal of the sympathetic ganglia the terminal autonomic innervation of the pinealcytes degenerate. Photo by Boogaard, Netherlands Instute for Neuroscience, with permission.

(Raab, 1934). In his animals, this effect could be abolished by spinal cord section or splanchnic nerve section. In 1940 Gellhorn continued the experiments on sham rage that had shown that they are accompanied by hyperglycemia, which turns into hypoglycemia after spinal cord section. He demonstrated that this

hypoglycemic effect could be abolished by vagotomy (Gellhorn et al., 1940). In 1941 Pitts demonstrated the direct link between posterior hypothalamic stimulation and sympathetic chain discharge (Pitts and Bronk, 1941). In 1943 Hetherington found that animals with a hypophysial lesion would become obese only after an

36

F. KREIER AND D.F. SWAAB

Fig. 2.33. Timeline of the scientists and their findings discussed in this chapter.

HISTORY OF HYPOTHALAMIC RESEARCH

Fig. 2.33—Cont’d

37

38

F. KREIER AND D.F. SWAAB

additional lesion in the hypothalamus (Hetherington, 1943). In 1944 Heinbecker induced obesity in dogs after lesions to the paraventricular hypothalamus. The stereotactic apparatus thus enabled further exploration of the relationship between the hypothalamus and the ANS.

Tracing techniques and the antagonistic model of catabolic and anabolic regions in the hypothalamus Using the Horsley–Clarke stereotaxic apparatus, Hetherington demonstrated in 1940 in rat that lesions to the medial hypothalamus induced marked obesity, in contrast to lesions to the lateral hypothalamus, which lead to decreased food intake (Hetherington and Ranson, 1940; Elmquist et al., 1999). When Anand confirmed the findings of Hetherington, the concept of the lateral hypothalamus as a feeding center and the ventromedial nucleus as a satiety center emerged (Hetherington and Ranson, 1940; Anand and Brobeck, 1951a,b; Elmquist et al., 1999). Nauta (Fig. 2.31), Kuypers, and others further described the anatomy of the hypothalamus and its projections to the ANS by neuronal tracing techniques (Nauta and Haymaker, 1969; Kuypers and Maisky, 1975; Swanson, 1999). Hans Ariëns Kappers, former director of the Netherlands Brain Institute, explored the sympathetic innervation the pineal: After removal of the sympathetic ganglia, the terminal autonomic innervation of the pinealcytes degenerate (Ariëns Kappers, 1960, Fig. 2.32). In the 1970s and 1980s studies on the effect of refined lesions in combination with neuronal tracing and peripheral autonomic nerve sections questioned this clear-cut view and demonstrated the involvement of autonomic projections on the effects (Gold, 1973; Opsahl and Powley, 1974; Saper et al., 1976; Inoue and Bray, 1977; Berthoud and Jeanrenaud, 1979; Gold et al., 1980; Swanson and Kuypers, 1980; Cox and Powley, 1981; Sclafani, 1981; Saper, 1995). However, since the discovery of orexigenic and anorexigenic neuropeptides in the 1980s and 90s, the concept reemerged when molecular probes to these compounds detected orexigenic neuropeptides in the lateral hypothalamus and anorexigenic peptides in the ventral medial hypothalamus (Elmquist et al., 1999) (Fig. 2.33).

FUTURE DIRECTIONS The history of the hypothalamus has taught us so far into detail how the hypothalamus communicates with the rest of the body: via hormones and the autonomic nervous system. Though it is difficult to predict the future, it seems now to be a task for scientists and clinicians to focus on the way how the hypothalamus receives

feedback from the body. The integration of that information, together with the information from the rest of the brain to the hypothalamus will lead to a closed regulatory circle that when disturbed by environment or emotions may lead to disease.

REFERENCES Ackerknecht EH (1974). The history of the discovery of the vegetative (autonomic) nervous system. Med Hist 18: 1–8. Aminoff MJ (2010). Brown-Sequard: an improbable genius who transformed medicine, Oxford University Press, Oxford. Anand BK, Brobeck JR (1951a). Hypothalamic control of food intake in rats and cats. Yale J Biol Med 24: 123–140. Anand BK, Brobeck JR (1951b). Localization of a “feeding center” in the hypothalamus of the rat. Proc Soc Exp Biol Med 77: 323–324. Anderson E, Haymaker W (1974). Breakthroughs in hypothalamic and pituitary research. Prog Brain Res 41: 1–60. Archer R, Chauvet J, Olivry G (1956). Sur l’existence eventuelle d’une hormone unique neurohypophysaire. Relation entre l’oxytocine, la vasopressine et la proteine de van Dyke extraites de las neurohypophyse du boeuf. Biochim Biophys Acta 22: 421–427. Arie¨ns Kappers J (1960). The development, topographical relations and innervation of the epiphysis cerebri in the albino rat. Z Zellforsch Mikrosk Anat 52: 163–215. Aschner B (1909). Demonstration von Hunden nach Exptirpation der Hypophyse. Wien Klin Wochenschr 22: 1730–1732. € Aschner B (1912). Uber die Funktion der Hypophyse. Pflugers Arch 146: 1–146. Babinski J (1900). Tumeur du corps pituitaire sans acrom
egalie et avec arr^et de d
eveloppement des organes g
enitaux. Rev Neurol 8: 531–535. Bahnsen U, Oosting P, Swaab DF et al. (1992). A missense mutation in the vasopressin-neurophysin precursor gene cosegregates with human autosomal dominant neurohypophyseal diabetes insipidus. EMBO J 11: 19–23. Bamshad M, Aoki VT, Adkison MG et al. (1998). Central nervous system origins of the sympathetic nervous system outflow to white adipose tissue. Am J Phys 275: R291–R299. Banting F, Best C (1922). The internal secretion of the pancreas. J Lab Clin Med 7: 251–266. Banting F, Best C, Collip J et al. (1922). The effects of insulin on experimental hyperglycemia in rabbits. Am J Phys 62: 559–580. Bard P (1928). A diencephalic mechanism for the expression of rage with special reference to the sympathetic nervous system. Am J Phys 84: 490–515. Bargmann W (1949). Ueber die neurosekretorische Verknupfung von Hypothalamus und Neurohypophyse. Z Zellforsch 34: 610–634.

HISTORY OF HYPOTHALAMIC RESEARCH Barron M (1920). The relation of the islets of Langerhans to diabetes with special reference to cases of pancreatic lithiasis. Surg Gynecol Obstet 31: 437–448. Barry J (1954). Neurocrinie et synapses neuros
ecr
etoires. Arch Anat Micr Morph Exp 43: 310–320. Barry J (1961). Recherches morphologiques et exp
erimentales sur la glande dienc
ephalique et l’appareil hypothalamohypophysaire. Ann Sci Univ Besancon, Zool Physiol Sur 2: 3–133. Barry J, Dubois MP, Poulain P (1973). LRF producing cells of the mammalian hypothalamus. A fluorescent antibody study. Z Zellforsch Mikrosk Anat 146: 351–366. Beattie J (1932). The relation of the tuber cinereum to gastric and cardiac functions. Can Med Assoc J 26: 178. Beattie J, Brow GR, Long CNH (1930a). The hypothalamus and the sympathetic nervous system. In: W Time et al. (Eds.), The vegetative nervous system. Association for research in nervous and mental disease IX. Williams & Wilkins, Baltimore, pp. 249–316. Beattie J, Brow GR, Long CNH (1930b). Physiological and anatomical evidence for the existence of nerve tracts connecting the hypothalamus with spinal sympathetic centres. Proc R Soc London, Ser B 106: 253–275. Bechterew W (1887). Die Bedeutung der Sehfl€ugel auf Grund von experimentellen und pathologischen Daten. Virchows Arch 110: 322–337. Bernard C (1850). Chiens rendu diabetique. C R Soc Biol (Paris) 1: 60. Bernard C (1851). Influence du grand sympathique sur la sensibilite et sur la calorification. C R Soc Biol (Paris) 3: 163–164. Berthoud HR, Jeanrenaud B (1979). Acute hyperinsulinemia and its reversal by vagotomy after lesions of the ventromedial hypothalamus in anesthetized rats. Endocrinology 105: 146–151. Bichat F (1800). Recherches physiologiques sur la vie et la mort, Brosson, Gabon et Cie, Paris. Bohus B, De Wied D (1966). Inhibitory and facilitatory effect of two related peptides on extinction of avoidance behavior. Science 153: 318–320. Bramwell B (1888). Intracranial tumors, Lippincott, Philadelphia. Brooks CM (1988). The history of thought concerning the hypothalamus and its functions. Brain Res Bull 20: 657–667. Brown-Sequard C (1852). Effect of section of the cervical sympathetic. Med Exam 8: 481–503. Buijs RM, Kalsbeek A (2001). Hypothalamic integration of central and peripheral clocks. Nat Rev Neurosci 2: 521–526. 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.

39

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. Buijs RM, De Vries GJ, Van Leeuwen FW et al. (1983). Vasopressin and oxytocin: distribution and putative functions in the brain. Prog Brain Res 60: 115–122. Burgus R, Dunn TF, Desiderio D et al. (1969). Molecular structure of the hypothalamic hypophysiotropic TRF factor of ovine origin: mass spectrometry demonstration of the PCA-His-Pro-NH2 sequence. C R Acad Sci Hebd Seances Acad Sci D 269: 1870–1873. Cajal y Ramo´n S (1909). Histology of the nervous system, Translated from the French by Swanson, N., Swanson, L.W. (Eds.). History of neuroscience, vol 1, General principles, spinal cord, spinal ganglia, medulla & pons. Oxford University Press, NY. 1995. Cajal y Ramo´n S (1911). Histologie du syste`me nerveux de l’homme et les vert
ebr
es, T.2. Maloine, Paris. Camus J, Roussy G (1913). Hypophysectomie et polyurie experimentales. C R Soc Biol (Paris) 75: 483–486. Cannon W (1929). Bodily changes in pain, hunger, rage and fear, D. Appleton and Company, New York. Chemelli RM, Willie JT, Sinton CM et al. (1999). Narcolepsy in orexin knockout mice: molecular genetics of sleep regulation. Cell 98: 437–451.
Collin R (1928). La Neurocrinie hypophysaire. Etude histophysiologique du complex tub
ero-infundibulo-pituitaire. Arch Morph Gen Exp 28: 102. Cox JE, Powley TL (1981). Prior vagotomy blocks VMH obesity in pair-fed rats. Am J Phys 240: E573–E583. Cushing H (1912). The pituitary body and its disorders. Clinical status produced by disorders of the hypophysis cerebri, JB Lippincott, Philadelphia. Cushing H (1932). Papers relating to the pituitary body, hypothalamus and parasympathetic nervous system, Charles C. Thomas, Springfield, IL. Dale H (1906). On some physiological actions of ergot. J Physiol 34: 163–206. Damasio A (1994). Decartes’ error, GP Putnam’s Sons, New York. Damasio A (2003). Mental self: the person within. Nature 423: 227. Defelipe J (2010). Cajal’s butterflies of the soul, Oxford University Press, Oxford. de Vries GJ, Duetz W, Buijs RM et al. (1986). Effects of androgens and estrogens on the vasopressin and oxytocin innervation of the adult rat brain. Brain Res 399: 296–302. De Wied D (1965). The influence of the posterior and intermediate lobe of the pituitary and pituitary peptides on the maintenance of a conditioned avoidance response in rats. Int J Neuropharmacol 4: 157–167. de Wied D, van Ree JM (1982). Neuropeptides, mental performance and aging. Life Sci 31: 709–719. Dogterom J, Van Wimersma Greidanus TB, Swabb DF (1977). Evidence for the release of vasopressin and oxytocin into cerebrospinal fluid: measurements in plasma and CSF of

40

F. KREIER AND D.F. SWAAB

intact and hypophysectomized rats. Neuroendocrinology 24: 108–118. Domes G, Heinrichs M, Michel A et al. (2007). Oxytocin improves “mind-reading” in humans. Biol Psychiatry 61: 731–733. https://doi.org/10.1016/j.biopsych.2006.07.015. Epub 2006 Nov 29, PMID:17137561. Duvernoy H (1972). The vascular architecture of the median eminence. In: Brain-endocrine interaction median eminence: structure and function, Karger, Basel. Elmquist JK, Elias CF, Saper CB (1999). From lesions to leptin: hypothalamic control of food intake and body weight. Neuron 22: 221–232. € Erdheim J (1904). Uber Hypophysenganggeschw€ulste und Hirncholesteatome. S Wien Akad Wiss III: 537–726. 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. Fr€ ohlich A (1901). Ein Fall von Tumor der Hypophysis cerebri ohne Akromegalie. Wien Klin Rundsch 15: 833–836. Garrison FH (1917). An introduction to the history of medicine, WB Saunders Company, Philadelphia. Gellhorn E, Cortell R, Feldman J (1940). The autonomic basis of emotion. Science 92: 288–289. Goering D (1924). Beziehungen des vegetativen Nervensystems zum Fettgewebe. In: LR Mueller (Ed.), Die Lebensnerven. Springer, Berlin. Gold RM (1973). Hypothalamic obesity: the myth of the ventromedial nucleus. Science 182: 488–490. Gold RM, Sawchenko PE, DeLuca C et al. (1980). Vagal mediation of hypothalamic obesity but not of supermarket dietary obesity. Am J Phys 238: R447–R453. Goltz F (1892). Der Hund ohne Grosshirn. Pflugers Arch 51: 570–614. € Grafe E, Gr€unthal E (1929). Uber isolierte Beeinflussung des Gesamtstofwechsels vom Zwischenhirn aus. Klin Wochenschr 8: 1013–1016. Green JD, Harris GW (1947). The neurovascular link between the neurohypophysis and the adenohypophysis. J Endocrinol 5: 136–146. Gutkowska J, Jankowski M, Mukaddam-Daher S et al. (2000). Oxytocin is a cardiovascular hormone. Braz J Med Biol Res 33: 625–633. Harris G, Campbell H (1966). The regulation of the secretion of luteinizing hormone and ovulation. In: G Harris, H Campbell (Eds.), The pituitary gland. Butterworths, London. Hess W (1930). Die Regulierung des Blutkreislaufes, Thieme, Leipzig. Hess W (1931). Die Regulierung der Atmung, Thieme, Leipzig. Hetherington A (1943). The production of hypothalamic obesity in rats already displaying chronic hypopituitarism. Am J Phys 140: 89–92. Hetherington A, Ranson S (1940). Hypothalamic lesions and adiposity in the rat. Anat Rec 78: 149–172. Himwich H, Keller A (1930). Effect of stimulation of the hypothalamus on blood glucose. Am J Phys 93: 658–662.

Hokfelt T, Johansson O, Ljungdahl A et al. (1980). Peptidergic neurones. Nature 284: 515–521. Hokfelt T, Fuxe K, Pernov B (1986). Coexistence of neuronal messengers: a new principle in chemical transmission. In: Progress in brain research, vol. 68. Elsevier, Amsterdam. Holsboer F, Spengler D, Heuser I (1992). The role of corticotropin-releasing hormone in the pathogenesis of Cushing’s disease, anorexia nervosa, alcoholism, affective disorders and dementia. Prog Brain Res 93: 385–417. Horsley V, Clark R (1908). The structure and functions of the cerebellum examined by a new method. Brain 31: 45–124. Ingram W (1939). The hypothalamus: a review of the experimental data. Psychosom Med 1: 48–91. Ingram W, Ranson S, Hannett F (1931). Pupillary dilatation produced by direct stimulation of the tegmentum of the brainstem. Am J Phys 98: 687–691. Inoue S, Bray GA (1977). The effects of subdiaphragmatic vagotomy in rats with ventromedial hypothalamic obesity. Endocrinology 100: 108–114. Insel TR (1997). A neurobiological basis of social attachment. Am J Psychiatry 154: 726–735. Isenschmid R, Schnitzler W (1914). Beitrag zur Lokalisation des der W€armeregulation vorstehenden Zentralapparates im Zwischenhirn. Arch Exp Pathol Pharmakol 76: 202–223. Kadotani H, Faraco J, Mignot E (1998). Genetic studies in the sleep disorder narcolepsy. Genome Res 8: 427–434. Karplus J, Kreidl A (1909). Gehirn und Sympathicus. I. Zwischenhirnbasis und Halssympathicus. Pflugers Arch 129: 138–144. € Karplus J, Kreidl A (1927). Gehirn und Sympathicus. VII. Uber Beziehungen der Hypothalamuszentren zu Blutdruck und innerer Sekretion. Pflugers Arch 215: 667–670. Klien H (1921). Ueber Lipodystrophie nebst Mitteilung eines Falles. Muench Med Wochenschr 7: 206–208. 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. Kreier F, Kap YS, Mettenleiter TC et al. (2006). Tracing from fat tissue, liver, and pancreas: a neuroanatomical framework for the role of the brain in type 2 diabetes. Endocrinology 147: 1140–1147. Kuypers HG, Maisky V (1975). Retrograde axonal transport of horseradish peroxidase from spinal cord to brain stem cell groups in the cat. Neurosci Lett 1: 9–14. Lang RE, Heil J, Ganten D et al. (1983). Effects of lesions in the paraventricular nucleus of the hypothalamus on vasopressin and oxytocin contents in brainstem and spinal cord of rat. Brain Res 260: 326–329. Langley JN (1899). Presidential address, section I—Physiology, Report of British Association881–892. Langley JN (1921). The autonomic nervous system, Heffer, Cambridge. Legait H, Legait E (1957). Les voies extra-hypophysaires des noyaux neurosecretoires hypothalamiques chez les batraciens et les reptiles. Acta Anat 30: 429–443. Legait H, Legait E (1958). Presence of hypothalamo-habenular neurosecretory pathway & demonstration of an

HISTORY OF HYPOTHALAMIC RESEARCH antidiuretic activity at the level of the habe-nular ganglia in hen. C R Seances Soc Biol Fil 152: 828–830. Legros (2001). Inhibitory effect of oxytocin on corticotrope function in humans: are vasopressin and oxytocin ying-yang neurohormones? Psychoneuroendocrinology 26: 649–655. https://doi.org/10.1016/s0306-4530(01)00018-x. PMID: 11500247. Legros JJ, Geenen V (1996). Neurophysins in central diabetes insipidus. Horm Res 45: 182–186. Legros JJ, Gilot P, Schmitz S et al. (1980). Neurohypophyseal peptides and cognitive function: a clinical approach. In: F Brambilla, G Racagni, D De Wied (Eds.), Progress in psychoneuroendocrinology. Elsevier, Amsterdam, pp. 325–337. Lewy F (1928). Reizversuche zur zentralen Pupillinnervation. Deutsche Ztschr Nervenh 102: 89. € Lichtenstern R (1912). Uber die zentrale Blaseninnervation, ein Beitrag zur Physiologie des Zwischenhirns. Wien Klin Wochenschr II: 1248–1249. 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. € € Loewi O (1921). Uber die humorale Ubertragbarheit der Herznebenwirkung. Pflugers Arch 189: 239–242. Lo´pez-Mun˜oz F, Molina JD, Rubio G et al. (2011). An historical view of the pineal gland and mental disorders. J Clin Neurosci 18: 1028–1037. 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: 220–232. Mack SO, Kc P, Wu M et al. (2002). Paraventricular oxytocin neurons are involved in neural modulation of breathing. J Appl Physiol 92: 826–834. Meites J (1992). Short history of neuroendocrinology and the International Society of Neuroendocrinology. Neuroendocrinology 56: 1–10. Mohr U (1840). Hypertrophie der hypophysis cerebri und dadurchbedingter Druck auf die Hirngrundfl€ache, insbesondere auf die Sehnerven, das Chiasma deselben und den linksseitigen Hirnschenkel. Wochenschr ges Heilk 6: 556–571. Montastruc JL, Rascol O, Senard JM (1996). The discovery of vasomotor nerves. Clin Auton Res 6: 183–187. Moore RY, Abrahamson EA, Van Den Pol A (2001). The hypocretin neuron system: an arousal system in the human brain. Arch Ital Biol 139: 195–205. M€ uller L (1915). Ueber die Hungerempfindung. D Med Woch 41: 1297–1301. M€ uller L (1920). Ueber den Durst und ueber die Durstempfindung. D Med Woch 46: 113–116. Nauta W, Haymaker W (1969). Hypothalamic nuclei and fiber connections. In: W Haymaker, E Anderson, W Nauta (Eds.), The hypothalamus. Charles C. Thomas, Springfield.

41

Neund€ orfer B, Hilz MJ (1998). Ludwig Robert M€ uller (1870-1962)—a pioneer of autonomic nervous system research. Clin Auton Res 8: 1–5. Nishino S, Ripley B, Overeem S et al. (2000). Hypocretin (orexin) deficiency in human narcolepsy. Lancet 355: 39–40. Nothnagel C (1879). Topische Diagnostik der Gehirnkrankheiten, Eine klinische Studie, Berlin. Olafsdottir BR, Rye DB, Scammell TE et al. (2001). Polymorphisms in hypocretin/orexin pathway genes and narcolepsy. Neurology 57: 1896–1899. Opsahl CA, Powley TL (1974). Failure of vagotomy to reverse obesity in the genetically obese Zucker rat. Am J Phys 226: 34–38. 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. Pick E (1941). Hans Horst Meyer 1853–1939. Pharm Exp Ther 71: 301–302. Pitts R, Bronk D (1941). Excitablility cycle of the hypothalamussympathetic neurone system. Am J Phys 135: 504–522. Probst M (1898). Experimentelle Untersuchungen € uber das Zwischenhirn und dessen Verbindungen, besonders der sogenannten R€ uckenschleife. J Neurol 13: 384–408. Raab W (1934). Die Beeinflussung des Fettstoffwechsels durch Hypophysenstoffe. Klin Wochenschrift I: 281–285. Raisman G (1997). An urge to explain the incomprehensible: Geoffrey Harris and the discovery of the neural control of the pituitary gland. Annu Rev Neurosci 20: 533–566. Ranson S, Magoun H (1933). Respiratory and pupillary reactions induced by electrical stimulation of the hypothalamus. Arch Neurol Psychiatr 29: 1179–1194. Ranson S, Kabat H, Magoun H (1934). Autonomic reactions induced by electrical stimulation of the hypothalamus. Am J Phys 109: 85–86. Remak R (1838). Dissertationes anatomicae et microscopicae de systematis nervosi structura. Berlin. Rokitansky C (1842). Handbuch der pathologischen Anatomie, Wien. Saper CB (1995). Central autonomic system. In: G Paxinos (Ed.), The rat nervous system. Academic Press, San Diego. Saper CB, Loewy AD, Swanson LW et al. (1976). Direct hypothalamo-autonomic connections. Brain Res 117: 305–312. Sarnat HB (1983). The discovery, proof and reproof of neurosecretion (Speidel, 1917; Scharrer and Scharrer, 1934). Can J Neurol Sci 10: 208–212. Sawin C (2000). Vasopressin is a hormone: the work of Ernest Basil Verney (1894–1967). Endocrinologist 10: 79–82. Schally A, Gual C (2001). Some recollections of early clinical studies on hypothalamic hormones: a tale of a successful international collaboration. Endocrinologist 11: 341–349. Scharrer E (1933). Die Erklaerung der scheinbar pathologischen Zellbilder im Nucleus supraopticus und Nucleus paraventricularis. Z Neurol Psychiat 145: 462–470. Scharrer E (1934). Stammt alles Kolloid im Zwischenhirn aus der Hypophyse? Frankf Z Path 47: 134.

42

F. KREIER AND D.F. SWAAB

Scharrer B (1975). Neurosecretion and its role in neuroendocrine regulation. In: J Meites, B Donovan, SM McCann (Eds.), Pioneers in neuroendocrinology, Plenum Publishing Cooperation, New York. Scharrer B (1992). The concept of neurosecretion and its place in neurobiology. In: FG Worden, JP Swazey, G Adelman (Eds.), The neurosciences: paths of discovery, I. Birkh€auser, Boston. Scharrer E, Gaupp R (1933). Neuere Befunde am Nucleus supraopticus und Nucleus paraventricularis des Menschen. Z Neurol Psychiat 148: 766–772. Scharrer E, Scharrer B (1940). Secretory cells within the hypothalamus. Res Publ Assoc Nerv Ment Dis 20: 170–194. Scharrer E, Scharrer B (1945). Neurosecretion. Physiol Rev 15: 171–181. Schiff M (1856). Neue Versuche €uber den Einfluss der Nerven auf die Gef€asse und die thierische W€arme. Berner Natf Ges 368: 69–72. Schiff M (1867). Lecons sur la physiologie de la digestion, Germer Baillie`re, Paris. Schmale H, Bahnsen U, Richter D (1993). Structure and expression of the vasopressin precursor gene in central diabetes insipidus. Ann N Y Acad Sci 689: 74–82. Schwartz MW, Woods SC, Porte Jr D et al. (2000). Central nervous system control of food intake. Nature 404: 661–671. Sclafani A (1981). The role of hyperinsulinema and the vagus nerve in hypothalamic hyperphagia reexamined. Diabetologia 20: 402–410. Siegel JM (1999). Narcolepsy: a key role for hypocretins (orexins). Cell 98: 409–412. Sofroniew MV, Weindl A (1978). Extrahypothalamic neurophysin-containing perikarya, fiber pathways and fiber clusters in the rat brain. Endocrinology 102: 334–337. Swaab DF (2003). The human hypothalamus. Basic and clinical aspects. Part I: nuclei of the hypothalamus. In: MJ Aminoff, F Boller, DF Swaab (Series Eds.), Handbook of clinical neurology. vol. 79. Elsevier, Amsterdam. Swaab DF (2004). The human hypothalamus. Basic and clinical aspects. Part II: neuropathology of the hypothalamus and adjacent brain structures. In: MJ Aminoff, F Boller, DF Swaab (Series Eds.), Handbook of clinical neurology. vol. 80. Elsevier, Amsterdam. Swaab DF, Pool CW, Nijveldt F (1975). Immunofluorescence of vasopressin and oxytocin in the rat hypothalamoneurohypophypopseal system. J Neural Transm 36: 195–215. 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. https://doi.org/10.1210/jcem.80.2.7852523. PMID: 7852523. 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 (1999). The neuroanatomy revolution of the 1970s and the hypothalamus. Brain Res Bull 50: 397. Swanson LW, Kuypers HG (1980). A direct projection from the ventromedial nucleus and retrochiasmatic area of the hypothalamus to the medulla and spinal cord of the rat. Neurosci Lett 17: 307–312. Thannickal TC, Moore RY, Nienhuis R et al. (2000). Reduced number of hypocretin neurons in human narcolepsy. Neuron 27: 469–474. Thompson EA, de Wied D (1973). The relationship between the antidiuretic activity of rat eye plexus blood and passive avoidance behaviour. Physiol Behav 11: 377–380. Thompson RR, George K, Walton JC et al. (2006). Sexspecific influences of vasopressin on human social communication. Proc Natl Acad Sci U S A 103: 7889–7894. https://doi.org/10.1073/pnas.0600406103. Epub 2006 May 8, PMID: 16682649. Triarhou LC (2018). Johann Paul Karplus (1866–1936). J Neurol 265: 2752–2753. Urban IJ (1987). Brain vasopressin: from electrophysiological effects to neurophysiological function. Prog Brain Res 72: 163–172. Uvnas-Moberg K (1997). Oxytocin linked antistress effects— the relaxation and growth response. Acta Physiol Scand Suppl 640: 38–42. van Wimersma TB, Dogterom J, de Wied D (1975). Intraventricular administration of anti-vasopressin serum inhibits. Life Sci 16: 637–643. Vitturi BK, Sanvito WL (2020). Claude Bernard (1813–1878). J Neurol https://doi.org/10.1007/s00415-020-10091-3. Vonderahe A (1949). Changes in the hypothalamus in organic disease. In: The hypothalamus, The Williams and Wilkins Company, Baltimore. Wang GH (1965). Johann Paul Karplus (1866-1936) and Alois Kreidl (1864-1928): two pioneers in the study of central mechanisms of vegetative function. Bull Hist Med 39: 529–539. Wang G, Pan J, Lu T (1929). The galvanic skin reflex in normal, thalamic, decerebrated and spinal cats under anaesthesia. Chin J Phys 3: 109–122. Weber E, Weber E (1846). Experiences qui prouvent que les nerfs vagues peuvent retarder le mouvement. Arch Gen Med Ser 4: Suppl 12–13. Winkler C (1934). Introduction of "Een greep uit het werk van G. van Rijnberk", Van Dishoeck, Bussum. Wolpert SM (1997). The circle of Willis. Am J Neuroradiol 18: 1033–1034. Wu S, Jia M, Ruan Y et al. (2005). Positive association of the oxytocin receptor gene (OXTR) with autism in the Chinese Han population. Biol Psychiatry 58: 74–77. https://doi.org/ 10.1016/j.biopsych.2005.03.013. PMID: 15992526. Yirmiya N, Rosenberg C, Levi S et al. (2006). Association between the arginine vasopressin 1a receptor (AVPR1a) gene and autism in a family-based study: mediation by socialization skills. Mol Psychiatry 11: 488–494. https://doi.org/10.1038/sj.mp.4001812. PMID: 16520824.

HISTORY OF HYPOTHALAMIC RESEARCH Zeidman LA (2020). Brain science under the swastika, Oxford University Press165. Zhang B, Qiu L, Xiao W et al. (2020). Reconstruction of the hypothalamo-neurohypophysial system and functional dissection of magnocellular oxytocin neurons in the brain. Neuron 10: S0896–S6273.

43

Ziegler L (1928). Lipodystrophies: report of seven cases. Brain 51: 145–167. Zimmer C (2004). Beyond the ivory tower. A distant mirror for the brain. Science 303: 43–44. Zuckerman S (1954). The secretions of the brain I. Lancet 263: 739–788.

This page intentionally left blank

Handbook of Clinical Neurology, Vol. 179 (3rd series) The Human Hypothalamus: Anterior Region D.F. Swaab, F. Kreier, P.J. Lucassen, A. Salehi, and R.M. Buijs, Editors https://doi.org/10.1016/B978-0-12-819975-6.00001-7 Copyright © 2021 Elsevier B.V. All rights reserved

Chapter 3

Anatomy and cytoarchitectonics of the human hypothalamus BERTALAN DUDÁS1,2* 1

Neuroendocrine Organization Laboratory, Lake Erie College of Osteopathic Medicine, Erie, PA, United States 2

Department of Anatomy, Histology and Embryology, University of Szeged, Szeged, Hungary

Abstract Due to the complexity of hypothalamic functions, the organization of the hypothalamus is extremely intricate. This relatively small brain area contains several nuclei, most of them are ill-defined regions without distinct boundaries; these nuclei are often connected with each other and other distant brain regions with similarly indistinct pathways. These hypothalamic centers control numerous key physiological functions including reproduction, growth, food intake, circadian rhythm, behavior, and autonomic balance via neural and endocrine signals. To understand the morphology of the hypothalamus is therefore extremely important, though it remains a stupendous task due to the complex organization of neuronal networks formed by the various neurotransmitter and neuromodulator systems.

GROSS ANATOMY The hypothalamus is part of the diencephalon that is located inferior to the thalamus, hence the name; the Greek words hypo- (ὕpο) means “under” and thalamos (yάlamoB) means “bridal couch,” “nuptial chamber,” or “innermost room.” The word thalamos is used by Homer (Homeros) particularly in the Odyssey and later by the Roman writer Virgil (Vergilius) in a modern latinized version “thalamus.” As an anatomical term, thalamus appears first in the works of Galen of Pergamon (Gal
enos, from the adjective “galZnόB,” “calm”), in the second century AD. The term thalamus has been adopted in the writings of numerous authors including the Flemish Andreas Vesalius, although the term had a broader meaning since it also meant a hidden part (chamber): thalamī p
enis denoted the cavernous spaces of the penis and thalamī cordis the chambers of the heart. After Vesalius, these archaic connotations have faded and the term thalamus was used solely in diencephalic context. It was the 19th century, when the area below the thalamus has been introduced as “regio subthalamica” by Forel.

The term hypothalamus was coined by Wilhelm His in 1893, possibly due to anticipation of the Basle Nomina Anatomica that has been assembled 2 years later, and which was the first attempt of international standardization of anatomical terminology. Nevertheless, the prefixes “hypo” (ὕpο-) in Greek and “sub” in Latin have exactly the same meaning, similarly to the older “under” and the more recent “below” terms in English. The modern use of subthalamus denoting an area adjacent to the “hypothalamus” made the terminology and the subsequent morphology rather complicated. Human hypothalamus is a relatively minor part of the diencephalon occupying around 4 cm3, roughly the size of a walnut (Hofman and Swaab, 1992; Dudas, 2013). The borders of the hypothalamus are easily discernable from the base of the brain, where it extends from the anterior border of the optic chiasm to the posterior border of the mamillary bodies (Fig. 3.1). These boundaries are even more apparent when the hypothalamus is split through the median sagittal plain; the anterior border is defined by the lamina terminalis attached to the optic chiasm, and the posterior border is demarcated by the

*Correspondence to: Bertalan Dudás, M.D., Ph.D., M.S. (Med. Ed.), Professor of Anatomy, Assistant Dean of Research and Scholarship, Anatomy, Lake Erie College of Osteopathic Medicine (LECOM), 1858 West Grandview Blvd, Erie, PA 16509, United States. Tel: +1-814-866-8142, Fax: +1-814-866-8411, E-mail: [email protected]

Fig. 3.1. Gross anatomy of the hypothalamus and the adjoining structures. Inferior view of the hypothalamus with (A) and after (B) the removal of the vessels forming the circulus arteriosus Willisii. The severed hypophyseal stalk is denoted by asterisk. (C) Median sagittal section of the brain depicting the hypothalamus and the neighboring diencephalic structures. In this particular case, thalamic adhesion is missing. Infundibular and optic recesses of the third ventricle are indicated by the five-pointed and sixpointed stars, respectively. Abbreviations: AC, anterior commissure; Bas, basilar artery; V4, fourth ventricle; CA, cerebral aqueduct; CC, corpus callosum; Ch, choroid plexus of the third ventricle; CP, cerebral peduncle; Fx, fornix; HS, hypothalamic sulcus; IC, internal carotid a.; ICo, inferior colliculus; Inf, infundibulum; IV, interventricular foramen; MB, mamillary body; Oc, oculomotor nerve; OCh, optic chiasm; Ol, olfactory tract, ON, optic nerve; OT, optic tract; PB, pineal body; PCe, posterior cerebral artery; PCo, posterior communicating artery; Pons, pons; Pt, paraterminal gyrus; SCo, superior colliculus; and Un, uncus. Arrowheads point to the lamina terminalis.

coronal plane running parallel with the lamina terminalis at the posterior border of the mamillary bodies. Superiorly, the hypothalamus extends anteriorly to the horizontal plane through the anterior commissure and more

posteriorly to the hypothalamic sulcus with the periventricular zone slightly extending beyond it; this sulcus marks the boundary between the thalamus and the underlying hypothalamus, and it is well detectable from the

ANATOMY AND CYTOARCHITECTONICS OF THE HUMAN HYPOTHALAMUS medial aspect of the third ventricle. Laterally, the hypothalamus is surrounded by the basal nuclei; thus, the hypothalamic border laterally is less exact extending approximately to the paramedian-sagittal plane running through the olfactory tracts (Figs. 3.2–3.7). The anterior border of the third ventricle is the lamina terminalis, which is a thin vascularized plate extending

47

from the anterior commissure to the optic chiasm, where it forms the optic recess with the superior chiasmatic surface. Its precise name is lamina terminalis cinerea (ash gray) indicating that this region is the most anterior portion of the prosencephalon, which stayed in its original place. The basal, funnel-like part of the hypothalamus, the infundibulum, connects the hypothalamus to the

Fig. 3.2. Plastinated coronal section of the hypothalamus and the bordering diencephalic structures. Abbreviations: AC, anterior commissure; CC, corpus callosum; CN, caudate nucleus; eGP, external zone of the globus pallidus; fx, fornix; IC, internal capsule; LV, lateral ventricle; NAc, nucleus accumbens; OCh, optic chiasm; Pt, putamen; S, septum pellucidum; and SN, septal nuclei. Section thickness: 5 mm.

48

B. DUDÁS

Fig. 3.3. Plastinated coronal section of the hypothalamus and the bordering diencephalic structures. Abbreviations: AC, anterior commissure; CC, corpus callosum; CN, caudate nucleus; eGP, external zone of the globus pallidus; fx, fornix; IC, internal capsule; inf, infundibulum; LV, lateral ventricle; OCh, optic chiasm; po, preoptic area; Pt, putamen; S, septum pellucidum; and SN, septal nuclei. Section thickness: 5 mm.

hypophysis by the hypophyseal stalk that contains the hypothalamo-hypophyseal tract projecting mostly from the paraventricular and supraoptic nuclei to the neurohypophysis. The cavity of infundibulum is the deepest region of the third ventricle, the infundibular recess

that, in contrast to Galen’s description, does not reach the hypophysis. The infundibulum causes a grayish protrusion at the hypothalamic base that is called tuber cinereum after its color (cinereus translates to ash gray). Posteriorly, this protrusion is particularly well

ANATOMY AND CYTOARCHITECTONICS OF THE HUMAN HYPOTHALAMUS

49

Fig. 3.4. Plastinated coronal section of the hypothalamus and the bordering diencephalic structures. Abbreviations: 3V, third ventricle; AC, anterior commissure; CC, corpus callosum; CN, caudate nucleus; Fx, fornix; eGP, external zone of the globus pallidus; IC, internal capsule; Inf, infundibulum; LV, lateral ventricle; OT, optic tract; and Pt, putamen. Section thickness: 5 mm.

observable forming the median eminence in the midline extending almost to the mamillary region. The infundibulum and the overlaying hypothalamic areas define the infundibular or tuberal region that has a particular importance, since it is highly vascularized; via the portal vessels, releasing and inhibiting hormones reach the anterior

lobe of the hypophysis and regulate the secretion of the hypophyseal tropic hormones. The mamillary bodies are perhaps the most distinctive structures of the hypothalamus; with their large, spherical surface, they define the posterior hypothalamic area rostral to the posterior perforated substance of the

50

B. DUDÁS

Fig. 3.5. Plastinated coronal section of the hypothalamus and the bordering diencephalic structures. Abbreviations: Am, amygdala; 3V, third ventricle; AC, anterior commissure; CC, corpus callosum; CN, caudate nucleus; Fx, fornix; eGP, external zone of the globus pallidus; iGP, globus pallidus, internal zone; IC, internal capsule; inf, infundibulum; LV, lateral ventricle; OT, optic tract; and Pt, putamen. Section thickness: 5 mm.

interpeduncular fossa. The distended lateral mamillary nucleus may form an additional set of accessory mamillary bodies laterally to the proper ones that appear to be the normal size (Fig. 3.14); this variation is rather unique since it has been described only once in the literature (Corso et al., 2019).

VASCULAR SUPPLY The blood supply of the hypothalamus is provided by multiple arteries that approach the diencephalon from the base of the brain. Since there is some discrepancy regarding the anatomy of these vessels when described

ANATOMY AND CYTOARCHITECTONICS OF THE HUMAN HYPOTHALAMUS

51

Fig. 3.6. Plastinated coronal section of the hypothalamus and the bordering diencephalic structures. Abbreviations: 3V, third ventricle; CC, corpus callosum; CN, caudate nucleus; Fx, fornix; eGP, globus pallidus, external zone; iGP, globus pallidus, internal zone; Hc, hippocampus; IC, internal capsule; LV, lateral ventricle; MB, mamillary body; MM, medial mamillary nucleus of the mamillary body; LM, lateral mamillary nucleus of the mamillary body; OT, optic tract; PMB, principal mamillary bundle; Pt, putamen; and va, ventral anterior thalamic nucleus. Asterisks denote the white matter plate of alveus hippocampi. Section thickness: 5 mm.

by multiple authors, the present description of the blood supply is based on previously published literature that provided comprehensive description of the vessels (Haymaker et al., 1969; Swaab, 2004; Dudas, 2013).

In the second century, Galenus has described the rete mirabile, a rich vascular web at the base of the diencephalon; however, this network of vessels is characteristic for ungulates and completely missing in humans. The

52

B. DUDÁS

Fig. 3.7. Plastinated coronal section of the hypothalamus and the bordering diencephalic structures. Abbreviations: 3V, third ventricleap, anteroprincipal thalamic nucleus; CC, corpus callosum; CN, caudate nucleus; CP, cerebral peduncle; Fx, fornix; eGP, globus pallidus, external zone; iGP, globus pallidus, internal zone; Hc, hippocampus; IC, internal capsule; ld, laterodorsal thalamic nucleus; LV, lateral ventricle; MB, mamillary body; md, medial dorsal thalamic nucleus; mte, mamillotegmental tract; mth, mamillothalamic tract; Pons, pons; Pt, putamen; SN, substantia nigra; st, subthalamic nucleus; vl, ventrolateral thalamic nucleus; and zi, zona incerta. Asterisks denote the white matter plate of alveus hippocampi. Section thickness: 5 mm.

arterial supply of the hypothalamus is provided by the circulus arteriosus Willisii, an anastomotic circle formed by the internal carotid and basilar arteries as well as connecting vessels. This circle is termed after Thomas Willis, an

English physician, but has been formerly described by other authors (Figs. 3.1A and 3.8), and it is rather variable; the type that is described by most of the textbooks exists in less than half of the cases only (Alpers et al., 1959; Swaab,

ANATOMY AND CYTOARCHITECTONICS OF THE HUMAN HYPOTHALAMUS

53

Fig. 3.8. Vascular supply of the hypothalamus depicting arteries in red and veins in blue. Left inset: basal view with the circulus arteriosus Willisii; right inset: median sagittal aspect. Diencephalic commissural pathways are denoted in green. The vascular nomenclature is based on the works of Haymaker et al. (1969), Swaab (2004), and Dudas (2013). Abbreviations: ACe, anterior cerebral a.; AC, anterior commissure; ACv, anterior cerebral v.; AP, anterior perforate aa.; CC, corpus callosum; Co, anterior communicating a.; Fx, fornix; HCm, habenular commissure; Hyp, hypophysis; IC, internal carotid a.; Inf, infundibulum; M, mamillary aa.; MB, mamillary bodies; MCe, middle cerebral a.; MCo, median commissural a.; OCh, optic chiasm; PCe, posterior cerebral a.; PCo, posterior communicating a.; PCm, posterior commissure; PMv, premamillary v.; PO, midline, medial, and lateral preoptic aa.; POv, postoptic v.; PP, posterior perforate aa.; PPv, posterior perforate v.; PV, paraventricular a.; SC, suprachiasmatic a.; SH, superior hypophyseal a.; SO, supraoptic a.; TC, medial and lateral tuber cinereal aa.; TP, thalamoperforate a.; and TT, thalamotuberal a.

2004). The circulus arteriosus is located at the base of the diencephalon, and with the exception of the basilar artery that runs on the pons, all of the constituting arteries provide branches to the hypothalamus. According to the origin of these branches, the arterial blood supply of the hypothalamus can be subdivided into three distinct regions. Septal and anterior hypothalamic areas including the optic chiasm, lamina terminalis, and the preoptic area are perfused by the branches of the anterior cerebral arteries; the tuberal/infundibular region is supplied by the middle cerebral arteries, and the anterior part of the posterior communicating arteries while the posterior cerebral arteries and the posterior part of the posterior communicating arteries provide the vascular supply of the posterior hypothalamus with the mamillary bodies. The two anterior cerebral arteries are connected by the anterior communicating artery that also gives

suprachiasmatic arteries supplying the basal zone of the preoptic area around the optic recess as well as the median commissural artery for the anterior commissure and adjacent preoptic and septal regions. Slightly laterally to these branches, the median, medial, and lateral preoptic arteries originate from the medial part of the anterior cerebral arteries and supply the preoptic region accordingly. Finally, the lateral part of the anterior cerebral arteries gives the supraoptic and anterior perforating arteries that perfuse the septal area and the lateral hypothalamus; the recurrent arteries of Heubner (Swaab, 2004) that pierce the anterior perforated area originate from the same part of the anterior cerebral arteries. The periventricular and medial hypothalamic area of the tuberal region that includes the paraventricular nucleus is supplied by the paraventricular arteries from the internal carotids that also give the superior and

54

B. DUDÁS

inferior hypophyseal arteries that form a rich capillary network for the hypophyseal stalk. The blood from these capillaries are collected by the portal vessels in the stalk that form a second capillary bed for the anterior lobe of the hypophysis through which releasing and inhibiting hormones are delivered to the cells producing the tropic hormones. In addition to contributing to the capillary network in the hypophyseal stalk, inferior hypophyseal arteries also provide blood supply to the neurohypophysis. In contrast, adenohypophysis has no direct arterial blood supply and receives blood only by portal circulation. The posterior part of the infundibulum that includes the tuber cinereum is supplied by the medial and lateral tuber cinereal arteries from the anterior part of the posterior communicating arteries that also provide the thalamotuberal arteries entering the hypothalamus more laterally. The posterior part of the posterior communicating arteries give the mamillary arteries for the posterior surface of the mamillary bodies, while the anterior surface is supplied by the tuber cinereal arteries. In addition to the posterior communicating arteries, the posterior cerebral arteries also provide blood supply to the posterior hypothalamus via the thalamoperforate arteries, and numerous posterior perforate arteries ramifying caudal to the mamillary bodies. The venous supply of the hypothalamus is significantly less complicated; the anterior cerebral, postoptic, premamillary, and posterior perforate veins drain into the basal vein.

CYTOARCHITECTURE Hypothalamus contains a number of well-defined nuclei; however, most of the structures described as nuclei are rather areas without distinct boundaries when observed with hematoxylin-eosin or Nissl stainings. In addition to this cytoarchitectonic indistinctness, the outlines of these hypothalamic regions are also ambiguous due to the variability of the shape of the hypothalamus in different individuals that can be influenced by the improper fixation of the brain and inaccuracy during sectioning. To make the descriptive morphology of the hypothalamus even more difficult, terminology that is used in the literature to describe these nuclei is challenging as there is some disagreement among the authors regarding the hypothalamic nomenclature; therefore, the terminus technicus used in this book is mostly based on the work of Saper (2004), Mai et al. (2008), and Swaab (2003). Despite the elusiveness of the hypothalamic cytoarchitecture, most authors describe hypothalamic morphology using a system that subdivides the hypothalamus into three compartments both anteroposteriorly and mediolaterally.

When using the sagittal, anteroposterior approach, the three discrete parts of the hypothalamus can be distinguished by structures at the base of the diencephalon and separated from each other by coronal planes that run parallel with the lamina terminalis. The most anterior part is the preoptic or chiasmatic area defined by the basally located optic chiasm. Posteriorly to the preoptic region is the infundibular area that is superior to the basally located infundibulum. Finally, the paired mamillary bodies with the diencephalic area above them define the posterior hypothalamic region. In addition to the anteroposterior subdivisions, hypothalamus can be subdivided into three shell-like regions mediolaterally. The most medial shell is the periventricular region that is an approximately 1-mm-thick layer covered by the ependymal lining of the third ventricle. Moving laterally to the priventricular region, the medial and lateral hypothalamus is separated by fornix that descends toward the mamillary bodies. These three mediolateral and anteroposterior regions define nine compartments that are used in the following to describe the position of the hypothalamic nuclei and pathways. According to this compartmentalized view, paraventricular nucleus occupies primarily the preoptic region extending into the tuberal zone where it tapers off (Figs. 3.11 and 3.12). The superior border of the nucleus is the hypothalamic sulcus that also marks the border between the thalamus and the hypothalamus. Mediolaterally, the paraventricular nucleus is in the medial hypothalamic area and the periventricular zone. In human, the former region is filled mostly by darkly stained magnocellular neurons that are easily detectable with Nissl staining, while the latter area is populated primarily with smaller, less intensely stained parvocellular cells. Magnocellular neurosecretory system is responsible for the production of oxytocin and vasopressin; the axons of these neurons form the hypothalamo-hypophyseal tract running through the hypophyseal stalk and projecting to the neurohypophysis, where oxytocin and vasopressin are eventually stored in nerve terminals (Herring bodies). The neurons that secrete oxytocin and vasopressin inhabit different zones in the magnocellular system; vasopressin-secreting perikarya form a dense cluster populating the ventrolateral zone of the paraventricular nucleus, while oxytocin-producing cells tend to avoid this zone and instead they are more scattered throughout the nucleus (Saper, 2004). Neurons composing the supraoptic nucleus have similar morphology to the cells populating the magnocellular part of the paraventricular nucleus; they have large, intensely stained perikarya populating the corner of lateral hypothalamus between the basal surface and the optic tract (Figs. 3.10–3.12). Magnocellular neurons also form additional cell clusters scattered along an

ANATOMY AND CYTOARCHITECTONICS OF THE HUMAN HYPOTHALAMUS arched line between the supraoptic and paraventricular nuclei as well as along the medial edge of the optic tract. Similar to the rest of the magnocellular cells, these perikarya are oxytocin- or vasopressin immunoreactive and project to the neurohypophysis. Both supraoptic and paraventricular nuclei are densely vascularized, and magnocellular neurons are often associated with blood vessels.

55

Suprachiasmatic nucleus consists of small, mediumstained neurons that are situated above the optic chiasm, in the periventricular zone of the preoptic region, adjacent to the optic recess of the third ventricle (Figs. 3.9 and 3.10). Suprachiasmatic nucleus plays a fundamental role in the modulation of circadian rhythm; it is targeted by direct afferents from the ganglion cells of the retina that approach the nucleus via the optic chiasm

Fig. 3.9. Coronal section of the anterior part of the hypothalamus illustrating the most observable structures with Nissl staining. Abbreviations: DBB, diagonal band of Broca; Inf, infundibulum; LT, lamina terminalis; MPO, medial preoptic area; NDB, nucleus of diagonal band of Broca; OCh, optic chiasm; and SCN, suprachiasmatic nucleus.

56

B. DUDÁS

Fig. 3.10. Coronal section of the preoptic part of the hypothalamus illustrating the most observable structures with Nissl staining. Abbreviations: DBB, diagonal band of Broca; Inf, infundibulum; MPO, medial preoptic area; NBM, nucleus Basalis of Meynert; OCh, optic chiasm; PVNa, paraventricular nucleus, anterior parvocellular part; SCN, suprachiasmatic nucleus; and SON, supraoptic nucleus.

(Sadun et al., 1984; Moore, 1993; Dai et al., 1998). In addition, the nucleus appears to be sexually dimorphic, since it tends to be elongated in females and more spherical in males (Hofman and Swaab, 1991). Halfway between the supraoptic and paraventricular nuclei at the transition of the preoptic and tuberal areas,

a relatively well-defined cluster of medium-sized and medium-stained perikarya is located that has been first described by Brockhaus as intermediate nucleus (Brockhaus, 1942) and verified by other authors (Braak and Braak, 1987; Saper, 2004). Since the nucleus encloses considerably more neurons in

ANATOMY AND CYTOARCHITECTONICS OF THE HUMAN HYPOTHALAMUS

57

Fig. 3.11. Coronal section of the anterior infundibular region of the hypothalamus illustrating the most observable structures with Nissl staining. Abbreviations: AC, anterior commissure; Fx, fornix; Inf, infundibulum; LHA, lateral hypothalamic area; MPO, medial preoptic area; OT, optic tract; PVNd, paraventricular nucleus, dorsal part; PVNm, paraventricular nucleus, magnocellular part; SI, substantia innominata; and SON, supraoptic nucleus.

men than in women, it is believed to be identical with the sexually dimorphic nucleus described by Swaab and Fliers (1985). After this seminal study, several authors reported on the sexually dimorphic character of hypothalamic areas (Allen et al., 1989; Levay, 1991; Byne et al., 2000), although the published results appear to be contradictory to some extent.

Ventromedial and dorsomedial nuclei are large, indistinct nuclei without well-defined boundaries in human, occupying primarily the medial hypothalamic area at the tuberal region (Fig. 3.12). Ventromedial nucleus forms ventrolateral and dorsomedial subclusters; the latter extends between the paraventricular and arcuate nuclei while the former is located between the arcuate nucleus

58

B. DUDÁS

Fig. 3.12. Coronal section through the infundibular recess of the hypothalamus illustrating the most observable structures with Nissl staining. Abbreviations: Am, accessory magnocellular neurons; AN, arcuate nucleus (infundibular nucleus); DM, dorsomedial nucleus; Fx, fornix; LHA, lateral hypothalamic area; LT, lateral tuberal nucleus; OT, optic tract; PVNd, paraventricular nucleus, dorsal part; PVNm, paraventricular nucleus, magnocellular part; SON, supraoptic nucleus; VMd, ventromedial nucleus, dorsomedial subdivision; and VMv, ventromedial nucleus, ventrolateral subdivision.

and the optic tract. Dorsomedial nucleus is bordered by the fornix laterally and stretches from the paraventricular nucleus to the dorsomedial subdivision of ventromedial nucleus (Saper, 2004). Dorsomedial nucleus is somewhat more discernible than the ventromedial one, since its compact part embedded into the main

body of the nucleus is populated by densely packed, strongly stained perikarya when examined in Nisslstained sections. The mediobasal part of the infundibular region is occupied by a relatively well-circumscribed cluster of cells that form the arcuate nucleus, also referred as

ANATOMY AND CYTOARCHITECTONICS OF THE HUMAN HYPOTHALAMUS infundibular nucleus. These cells are located periventricularly extending to the medial hypothalamus (Fig. 3.12). This part of the infundibulum is highly vascularized, and perikarya and fibers are in close juxtaposition with these vessels. Neurons in the arcuate nucleus produce releasing and inhibiting hormones that enter into the hypophyseal portal circulatory system, and the substances released from capillaries in the anterior lobe regulate the release of the tropic (luteinizing hormone, LH; follicle stimulating hormone, FSH; thyroid stimulating hormone, TSH; adrenocorticotropic hormone, ACTH) and non-tropic hormones (growth hormone, GH; prolactin) from the adenohypophysis. The basolateral region of the tuber cinereum contains the lateral tuberal nucleus (Fig. 3.12) that extends caudally and tapers down at the lateral part of the mamillary complex (Fig. 3.13); it is commonly manifested as eminences on the lateral side of the tuber cinereum clearly distinguishable when examining the base of the brain. The lateral tuberal nucleus contains a dense network of somatostatinergic neurons whose axonal varicosities frequently terminate on other perikarya whithin the nucleus, forming basket-like structures. This somatostatinergic cell population appears to be involved in the pathology of various neurodegenerative diseases; somatostatinergic immunoreactivity in the nucleus is diminished in Huntington’s disease (Kremer et al., 1991), and these cells exhibit cytoskeletal pathological changes in patients suffering in the early stages of Alzheimer’s disease (Timmers et al., 1996). Mamillary bodies are spherical structures at the base of the brain; together with the overlaying hypothalamic areas, they define the posterior hypothalamus. In human, the largest nucleus of the mamillary complex is the medial mamillary nucleus dwarfing the dorsolaterally located lateral mamillary nucleus (LeGros Clark, 1936; Saper, 2004; Mai et al., 2008) that is separated from it by a thin white matter shell (Figs. 3.6 and 3.13). The morphology and staining of the neurons in the medial and lateral nuclei are virtually indistinguishable, apart from a thin shell of intensely stained cells at the lateral margin of the lateral nucleus; therefore, previous authors rejected the morphological and functional distinction between them (Saper, 2004). Indeed, the lateral mamillary nucleus in rats does not seem to be analogous with the same nucleus in human, since it contains larger and more darkly stained neurons when compared to humans (Saper, 2004). However, a recent study described a unique condition of accessory mamillary bodies of considerable size (Fig. 3.14) that are formed by the greatly enlarged lateral mamillary nuclei at the lateral side of the primary mamillary bodies that appear to be normal (Corso et al., 2019). This interesting morphological finding suggests that the lateral

59

mamillary nucleus is an individual complex and not merely the lateral shell of the medial mamillary nucleus and that extrapolating animal morphology to human should be handled with extreme caution. The medial and lateral mamillary nuclei are surrounded with a shell of large, darkly stained neurons that form the tuberomamillary nucleus (Fig. 3.13) that also extends into the zone located basally between the medial and lateral mamillary nuclei; these cells exhibit similar morphology as the neurons of the magnocellular system in the supraoptic and paraventricular nuclei, but they are immunoreactive (IR) for galanin, histamine, and melanin-concentrating hormone (MCH) (Gai et al., 1990; Panula et al., 1990; Airaksinen et al., 1991; Mouri et al., 1993) instead of secreting oxytocin and vasopressin. The supramamillary nucleus is composed of neurons with similar morphology (Fig. 3.13), while perikarya populating the lateral tuberal nucleus are slightly smaller and less intensely stained (Fig. 3.13). Lateral to the tuberomamillary nucleus, adjacent to the emerging cerebral peduncles, lies the caudal part of the previously described lateral tuberal nucleus (Fig. 3.13). Mamillary lesions often lead to memory impairment. Destruction of the medial mamillary nucleus results in spatial working memory deficits, and therefore, it has been suggested that the mamillary complex may play a crucial role in the short-term storage of proprioceptive information essential for the consecutive execution of behavioral choices as well as maintaining the state of arousal of the animal (Field et al., 1978). Moreover, it has been theoretized that the lateral and medial mamillary nuclei work together in memory formation in a synergistic fashion modulating head direction information and theta rhythm, respectively (for review, see Vann and Aggleton, 2004). Most of the afferents and efferents of the mamillary nuclei run in the fornix, in the mamillothalamic and mamillotegmental tracts as well as in the mamillary peduncle and in the medial forebrain bundle (Fig. 3.15). These pathways are discussed in detail later in the present chapter. Lateral to the sagittal plane defined by the postcommissural fornix descending and terminating in the mamillary bodies is the extensive lateral hypothalamic area populated with neurons of diverse morphology (Figs. 3.10–3.12), including scattered large, intensely stained perikarya between the rostrocaudally arranged fibers of the medial forebrain bundle. There are only few well-defined nuclei located in this broad region: the previously described supraoptic nucleus is located basally, at the lateral border of the optic tracts; in addition, the lateral tuberal nucleus occupies the lateral side of the infundibulum and the tuberomamillary nucleus extends in the posterior hypothalamic region,

60

B. DUDÁS

Fig. 3.13. Coronal section of the posterior, mamillary region of the hypothalamus illustrating the most observable structures with Nissl staining. Abbreviations: Fx, fornix; LHA, lateral hypothalamic area; LT, lateral tuberal nucleus; MM, medial mamillary nucleus; OT, optic tract; PHA, posterior hypothalamic area; PMB, principal mamillary bundle; SM, supramamillary nucleus; and TM, tuberomamillary nucleus.

sometimes forming small eminences on the basal hypothalamic surface (LeGros Clark, 1936).

HYPOTHALAMIC PATHWAYS One of the most well-defined tracts, the fornix, is conveying fibers from the hippocampus, subiculum, and

presubiculum; it is also the most significant afferent pathway of the hypothalamus (Fig. 3.15). Hippocampal fibers initially form the alveus, a thin layer of white matter on the ventricular surface of the hippocampus (Figs. 3.6 and 3.7). Fibers composing the alveus are accumulated in the fimbria hippocampi ascending posterosuperorly and eventually transitioning into the fornix that,

ANATOMY AND CYTOARCHITECTONICS OF THE HUMAN HYPOTHALAMUS

61

Fig. 3.14. Accessory mamillary bodies (asterisks, inset B) are formed by the unusually massive lateral mamillary nuclei (LM; inset C). Normal anatomy is depicted on inset A (control). Abbreviations: Fx, fornix; OCh, optic chiasm; OT, optic tract, LM, lateral mamillary nucleus, MM, medial mamillary nucleus; and TM, tuberomamillary nucleus. Arrows denote the hypophyseal stalk. Magnification of the micrograph C: 25x.

Fig. 3.15. Hypothalamic afferents and efferents, the thickness of the arrows corresponds to the estimated size of the pathways. Nomenclature is based on the work of Nieuwenhuys et al. (1988) and Dudas (2013). Fornix (Fx) is the most distinct hypothalamic pathway containing primarily afferents composed of axons of the hippocampal (Hc) pyramidal neurons. Anterior diencephalic areas are connected with the brain stem and the spinal cord via the medial forebrain bundle (MFB) that is a bidirectional tract running through the lateral hypothalamic area. Dorsal longitudinal fasciculus is a predominantly descending pathway between the hypothalamus and structures of the brain stem and spinal cord. Ventral amygdalofugal (AF) and amygdalopetal (AP) pathways as well as the stria terminalis (ST) connect the nuclei of amygdala (Am) with the hypothalamus. MB project to the anterior thalamic nuclei (ant) by mamillothalamic tract (MTh) and to the tegmentum by the mamillotegmental tract (MTe) and receive inputs from the brain stem via the mamillary peduncle (MP). SC that is responsible for the regulation of circadian rhythm receives retinohypothalamic afferents (RH) from the retina (Ret) via the optic chiasm (OCh). Axons of neurons in the magnocellular system located primarily in the paraventricular and supraoptic nuclei project to the posterior lobe of hypophysis (Hyp) forming the hypothalamo-hypophyseal tract (HH). Stria medullaris contains fibers connecting the anterior hypothalamus and septal region with the epithalamus, including the pineal body. Additional abbreviations: AC, anterior commissure; CC, corpus callosum; and dm, dorsomedial thalamic nucleus.

after an anterior turn, continues anteromedially. Below the corpus callosum, fibers composing the left and right fornices form a decussation. Approaching the hypothalamus, fornix curves around the interventricular foramen

contacting the posterior part of the anterior commissure and forms the columna fornicis that decline toward the mamillary bodies, where most of the fibers eventually terminate. Medial and lateral hypothalamic areas are

62

B. DUDÁS

separated by the descending columna fornicis in coronal sections, where fornix is easily identifiable due to its content of robustly myelinated axons. Fibers of the fornix terminate mostly in the mamillary nuclei, medial, and lateral hypothalamus and in septal area in rats (Guillery, 1955; Powell et al., 1957; Swanson and Cowan, 1977; Canteras and Swanson, 1992; Kishi et al., 2000) and monkeys (Valenstein and Nauta, 1959). Fibers deriving from the hippocampus target mostly the septal area in rats (Saper, 2004), while mamillary bodies receive afferents from the subiculum and partially from the presubiculum in rats and primates (Swanson and Cowan, 1975; Krayniak et al., 1979). In addition to hypothalamic afferent fibers forming the vast majority of the fornix, some neurons, predominantly from the preoptic area, terminate in the hippocampal formation via the fornix (Saper, 2004). The medial forebrain bundle can be considered as a misnomer; it is in fact rather laterally located in the human diencephalon and it is not a bundle, but a loose, ill-defined organization of unmyelinated, rostrocaudally oriented fibers perforating the lateral hypothalamus. The bundle interconnects several regions of the brain with the hypothalamus (Fig. 3.15). Corticohypothalamic projections originating from diverse locations including the olfactory cortex, septal, lateral frontal, insular, prelimbic, infralimbic areas reach the hypothalamus via the anterior part of the medial forebrain bundle and innervate various hypothalamic areas including the perifornical area, anterior hypothalamic region, ventromedial nucleus, and the lateral hypothalamus in the rat (Kita and Oomura, 1982; Saper, 1982; Sesack et al., 1989) and monkeys (Ongur et al., 1998; Rempel-Clower and Barbas, 1998; Barbas, 2000; Freedman et al., 2000). In turn, hypothalamic efferents from the tuberal and posterior lateral regions also project to occipital, parietal, and frontal cortical areas in monkeys (Kievit and Kuypers, 1975; Porrino and Goldman-Rakic, 1982; Mesulam et al., 1983; Tigges et al., 1983; Rempel-Clower and Barbas, 1998). These hypothalamo-cortical projections, running most likely via the medial forebrain bundle, contain orexinergic, histaminergic, and MCH- IR fibers (Panula et al., 1990; Elias et al., 1998). Similarly to the anterior part, the caudal part of the medial forebrain bundle contains both ascending and descending fiber systems. Descending hypothalamic projections, deriving primarily from the retrochiasmatic area, lateral hypothalamus, and from the paraventricular, dorsomedial, and arcuate nuclei terminate in the substantia nigra, the tegmentum, and the midbrain reticular formation via the medial forebrain bundle (Saper et al., 1978; Saper et al., 1979). Some of these fibers also reach the spinal cord where they innervate somatosensory and autonomic nuclei including parabrachial nucleus, dorsal

horn neurons, nucleus tractus solitarii, ventrolateral reticular formation, spinal intermediate gray matter as well as parasympathetic and sympathetic spinal and medullary preganglionic neurons (Saper et al., 1976; Swanson and Kuypers, 1980; Cechetto and Saper, 1988; Jansen et al., 1992, 1997; Elias et al., 2000; Zhang et al., 2000; Saper, 2004). Indeed, hypothalamic lesion may lead to ipsilateral sympathetic deficit (Nathan and Smith, 1986). In addition to the descending brain stem projections, various cortical areas receive fibers via the medial forebrain bundle from the tuberal and posterior lateral hypothalamic areas, from GABAergic, galaninergic, and histaminergic systems at the tuberomamillary region, as well as from the field of Forel in rat (Saper, 1984). Via ascending projections of medial forebrain bundle, monoaminergic pathways from the brain stem innervate hypothalamic structures; these fibers also reach the diencephalon as via the periventricular fiber system. Medial forebrain bundle also conveys sensory input; lateral hypothalamus receives olfactory afferents and fibers from the parabrachial nucleus and from the nucleus tractus solitarii; the latter two also projects to the preoptic region, and to the paraventricular, dorsomedial, and ventromedial nuclei in rats (Ricardo and Koh, 1978; Saper and Loewy, 1980; Fulwiler and Saper, 1984, 1985; Bester et al., 1997) and monkeys (Beckstead et al., 1980; Pritchard et al., 2000). Moreover, hypothalamus receives viscerosensory input via direct solitary tract projections in rat; since this finding was not confirmed in monkeys, viscerosensory information may reach the hypothalamus indirectly in primates, possibly via the periaqueductal gray, reticular formation, or parabrachial nucleus. Rostrocaudally arranged fibers running on the dorsomedial surface of the thalamus compose the stria medullaris serving for the attachment of the lateral border of the choroid plexus of the third ventricle that is also called taenia thalami, and it can be easily observed after the removal of the stria. Fibers running in the stria medullaris originate posterior to the anterior commissure and connect chiefly the habenular nuclei to the lateral preoptic area (Fig. 3.15). Stria terminalis is one of the major efferents of the amygdala; it is an anteroposteriorly located white matter band covering the surface of the terminal vein that runs in the groove created between the caudate nucleus and the thalamus. It also serves for the origination of the lateral edge of the choroid plexus of the lateral ventricles that is also named as the taenia choroidea after the removal of the stria. Fibers in stria terminalis project from the amygdala toward the anterior hypothalamus; before the anterior commissure (Fig. 3.15), these fibers split into precommissural, postcommissural, and

ANATOMY AND CYTOARCHITECTONICS OF THE HUMAN HYPOTHALAMUS commissural components, the latter merging with the fibers composing the anterior commissure (Nieuwenhuys et al., 1988). Afferent fibers running in the stria terminalis modulate hypothalamic functions either directly or indirectly via the bed nucleus of stria terminalis, that is a band of neurons along the stria itself, where the majority of the fibers terminate; the remnant of the fibers project primarily to the anterior hypothalamus (Price and Amaral, 1981). In addition, the stria terminalis contains efferent amygdalopetal fibers that project from the ventromedial nucleus to the bed nucleus of stria terminalis, amygdala, substantia innominata in monkey (Saper et al., 1979). In addition to the stria terminalis, the ventral amydalofugal pathway is a major efferent of the amygdala that interconnects it with diencephalic structures by a shorter route than the stria terminalis. The pathway also contains afferent amygdalopetal components (ventral amygdalopetal fibers) intermixed with the efferent ones (Fig. 3.15). The ventral amygdalofugal tract projects medially from the basolateral and corticomedial cell groups of the amygdala passing through the substantia innominata and the substantia perforata anterior (Willis and Haines, 2018). Here, the fibers fan out and join the diagonal band of Broca, the thalamic peduncle, posteriorly the medial forebrain bundle (Nieuwenhuys et al., 1988). Among hypothalamic structures, these fibers target the lateral preoptic area over the optic tract; fibers also innervate the thalamus, the mediofrontal cortex, and structures in the brain stem (Nieuwenhuys et al., 1988). The periventricular hypothalamic area sends fibers to the brain stem; vica versa, fibers from the brain stem autonomic control nuclei target periventricular areas. This complex and rather elusive pathway has been named previously as dorsal longitudinal fasciculus of Sh€ utz, although the term periventricular fiber system is undoubtedly a more accurate terminology, and appropriately, it has been preferred by numerous authors (Sutin, 1966; Saper, 2004). In all regards, the periventricular fiber system interconnects the periaqueductal gray matter with the periventricular area and medial hypothalamus, and it is composed of both ascending and descending fibers (Fig. 3.15). Pathways, including monoaminergic systems, reach the hypothalamus via the periventricular fiber system or through the medial forebrain bundle. Similarly, descending fibers from the periventricular zone and medial hypothalamus target the periaqueductal zone and additional, more distal locations in monkey and rats (Saper et al., 1978, 1979) and were termed as hypothalamomedullary or hypothalamospinal fibers, respectively; the former targeting cranial nerve nuclei and the latter terminating on the intermediolateral cell column, hence having a direct influence on the autonomic nervous system. Dorsal tegmental nucleus or the central gray matter of the midbrain can serve as a hub, where

63

these ascending and descending fibers composing the periventricular fiber system can be synaptically interrupted in addition to the axons that directly interconnect the forebrain with autonomic centers (Hancock, 1976; Saper et al., 1976; Ricardo and Koh, 1978). Supraoptic decussation pathways are commissures associated with the optic chiasm that are described by Gudden and Meynert in the 19th century. These tracts are composed of myelinated fibers that do not derive from the optic nerves, but instead, they have been suggested to connect the tectum, subthalamus, geniculate bodies, and the globus pallidus with the identical structures contralaterally (Palkovits and Zaborsky, 1979; Weindl, 2007). The supraoptic decussation pathways are often subdivided to dorsal supraoptic decussation, pars dorsalis (Ganseri), and pars ventralis (Meynerti) as well as ventral supraoptic decussation (Guddeni) that is also named inferior commissure of Gudden’s (Palkovits and Zaborsky, 1979; Weindl, 2007). From hypothalamic aspect, the latter one is the most important; ventral supraoptic decussation occupies the posterior part of the optic chiasm and fibers interconnecting the lateral hypothalamic areas run in this bundle (Nauta and Haymaker, 1969; Saper, 2004). In addition, sensory systems originating from mechanoreceptors and thermoreceptors from the brain stem reach hypothalamic areas along the medial border of the optic tract (Cliffer et al., 1991; Saper, 2004). These fibers project into the contralateral lateral hypothalamic areas via the ventral supraoptic commissure in rats and monkeys (Burstein et al., 1987). The commissural fibers of the optic tract form the optic chiasm below the optic recess. Via the optic chiasm, these fibers from the ganglion cells of the retina project into the suprachiasmatic nucleus (Fig. 3.15) (Sadun et al., 1984; Moore, 1993; Dai et al., 1998) and form the major input to the structures regulating the circadian rhythm. Some of these retinohypothalamic fibers distribute to the supraoptic nucleus and the anteroventral hypothalamus (Dai et al., 1998). Mamillary peduncle or pedunculus corporis mamillaris is primarily an ascending pathway originating mainly from the dorsal tegmental nucleus of Gudden and from the mesencephalic reticular formation (Cowan et al., 1964). In the midbrain, the pathway runs medially in close proximity to the surface of the interpeduncular fossa following its contours while turning rostrally (Fig. 3.15). Here, the peduncle is located immediately laterally to the interpeduncular nucleus that occupies the midline; then it passes between the fibers that form the oculomotor nerve (Nieuwenhuys et al., 1988) and targets primarily the lateral mamillary nucleus (Saper, 2004). In addition, some fibers continue rostrally in the medial forebrain bundle into the lateral supramamillary

64

B. DUDÁS

region and lateral hypothalamus, and a few reach anterior hypothalamic areas (Cowan et al., 1964). The mamillary peduncle also contains efferent fibers from the mamillary bodies terminating in the tegmental nuclei and reticular formation (Shen, 1983). The principal mamillary bundle is a compact, welldefined bundle originating from the medial part of mamillary bodies (Nieuwenhuys et al., 1988), and after a short course, it splits into the ascending mamillothalamic and descending mamillotegmental tracts (Figs. 3.7 and 3.15). The mamillothalamic tract (Fig. 3.15) is primarily composed of efferents of the medial mamillary nuclei that target the anterior dorsal, anterior ventral, and anterior medial thalamic nuclei in numerous species (Shen, 1983; Alpeeva and Makarenko, 2009). Mamillothalamic tract contributes to the memory formation as part of the circuit of Papez; indeed, damage of the mamillothalamic tract can result in memory impairment (Field et al., 1978). It has been also reported that electrical stimulation mamillothalamic tract raises seizure threshold in animals, and it has been used in the therapy of refractory epilepsy in humans (Balak et al., 2018). The mamillotegmental tract (Fig. 3.15), similarly to the mamillothalamic tract, originates from the principal mamillary bundle (Nieuwenhuys et al., 1988) turns around the superior surface of the nucleus ruber and descends to the brain stem following a curved course. The tract contains efferents originating primarily from the medial mamillary nucleus and projecting into the tegmentum. The mamillotegmental tract projects to the pontine tegmental reticular nucleus that serves as a source of pontocerebellar fibers as well as to the ventral and dorsal tegmental nuclei of Gudden (Cruce, 1977; Ricardo, 1983; Shen, 1983; Nieuwenhuys et al., 1988; Hayakawa and Zyo, 1989). In addition to the mamillothalamic tract, numerous hypothalamic areas project to the thalamus in rats, primarily via the inferior thalamic peduncle that is a large fiber bundle connecting the thalamus with the temporal cortical areas (Nieuwenhuys et al., 1988; Saper, 2004). The peduncle also contains amygdalothalamic fibers from the ventral amygdalofugal pathway as well as fibers connecting the orbitofrontal, insular, and temporal cortices (Nieuwenhuys et al., 1988). Axons of the magnocellular neurons populating the paraventricular and supraoptic nuclei and the accessory nuclei between form a pathway that targets the posterior lobe of the hypophysis (neurohypophysis) via the hypophyseal stalk, forming the hypothalamohypophyseal tract (Fig. 3.15). Oxytocin and vasopressin secreted by these neurons is stored in the neurohypophysis.

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. Allen LS, Hines M, Shryne JE et al. (1989). Two sexually dimorphic cell groups in the human brain. J Neurosci 9: 497–506. Alpeeva EV, Makarenko IG (2009). Perinatal development of the mammillothalamic tract and innervation of the anterior thalamic nuclei. Brain Res 1248: 1–13. Alpers BJ, Berry RG, Paddison RM (1959). Anatomical studies of the circle of Willis in normal brain. AMA Arch Neurol Psychiatry 81: 409–418. Balak N, Balkuv E, Karadag A et al. (2018). Mammillothalamic and mammillotegmental tracts as new targets for dementia and epilepsy treatment. World Neurosurg 110: 133–144. Barbas H (2000). Connections underlying the synthesis of cognition, memory, and emotion in primate prefrontal cortices. Brain Res Bull 52: 319–330. Beckstead RM, Morse JR, Norgren R (1980). The nucleus of the solitary tract in the monkey: projections to the thalamus and brain stem nuclei. J Comp Neurol 190: 259–282. Bester H, Besson JM, Bernard JF (1997). Organization of efferent projections from the parabrachial area to the hypothalamus: a Phaseolus vulgaris-leucoagglutinin study in the rat. J Comp Neurol 383: 245–281. Braak H, Braak E (1987). The hypothalamus of the human adult: chiasmatic region. Anat Embryol 175: 315–330. Brockhaus H (1942). Beitrag zur normalen Anatomie des Hypothalamus und der Zona inzerta bei Menschen. J Psychol Neurol 51: 96–196. Burstein R, Cliffer KD, Giesler Jr GJ (1987). Direct somatosensory projections from the spinal cord to the hypothalamus and telencephalon. J Neurosci 7: 4159–4164. Byne W, Lasco MS, Kemether E et al. (2000). The interstitial nuclei of the human anterior hypothalamus: an investigation of sexual variation in volume and cell size, number and density. Brain Res 856: 254–258. Canteras NS, Swanson LW (1992). Projections of the ventral subiculum to the amygdala, septum, and hypothalamus: a PHAL anterograde tract-tracing study in the rat. J Comp Neurol 324: 180–194. Cechetto DF, Saper CB (1988). Neurochemical organization of the hypothalamic projection to the spinal cord in the rat. J Comp Neurol 272: 579–604. Cliffer KD, Burstein R, Giesler Jr GJ (1991). Distributions of spinothalamic, spinohypothalamic, and spinotelencephalic fibers revealed by anterograde transport of PHA-L in rats. J Neurosci 11: 852–868. Corso T, Grignol G, Kulesza R et al. (2019). Accessory mammillary bodies formed by the enlarged lateral mammillary nuclei: cytoarchitecture. Brain Struct Funct 224: 1971–1974. Cowan WM, Guillery RW, Powell TP (1964). The origin of the mammillary peduncle and other hypothalamic connexions from the midbrain. J Anat 98: 345–363.

ANATOMY AND CYTOARCHITECTONICS OF THE HUMAN HYPOTHALAMUS Cruce JA (1977). An autoradiographic study of the descending connections of the mammillary nuclei of the rat. J Comp Neurol 176: 631–644. Dai J, Van DV, Swaab DF et al. (1998). Human retinohypothalamic tract as revealed by in vitro postmortem tracing. J Comp Neurol 397: 357–370. Dudas B (2013). Anatomy of the human hypothalamus. In: The human hypothalamus: anatomy, functions and disorders, Nova Scientific Publishers, New York. Elias CF, Saper CB, Maratos-Flier E et al. (1998). Chemically defined projections linking the mediobasal hypothalamus and the lateral hypothalamic area. J Comp Neurol 402: 442–459. Elias CF, Kelly JF, Lee CE et al. (2000). Chemical characterization of leptin-activated neurons in the rat brain. J Comp Neurol 423: 261–281. Field TD, Rosenstock J, King EC et al. (1978). Behavioral role of the mammillary efferent system. Brain Res Bull 3: 451–456. Freedman LJ, Insel TR, Smith Y (2000). Subcortical projections of area 25 (subgenual cortex) of the macaque monkey. J Comp Neurol 421: 172–188. Fulwiler CE, Saper CB (1984). Subnuclear organization of the efferent connections of the parabrachial nucleus in the rat. Brain Res 319: 229–259. Fulwiler CE, Saper CB (1985). Cholecystokinin-immunoreactive innervation of the ventromedial hypothalamus in the rat: possible substrate for autonomic regulation of feeding. Neurosci Lett 53: 289–296. 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. Guillery RW (1955). A quantitative study of the mamillary bodies and their connexions. J Anat 89: 19–32. Hancock MB (1976). Cells of origin of hypothalamo-spinal projections in the rat. Neurosci Lett 3: 179–184. Hayakawa T, Zyo K (1989). Retrograde double-labeling study of the mammillothalamic and the mammillotegmental projections in the rat. J Comp Neurol 284: 1–11. Haymaker W, Anderson E, Nauta WJH (1969). The hypothalamus, Charles C Thomas Publisher, Springfield, IL. Hofman MA, Swaab DF (1991). Sexual dimorphism of the human brain: myth and reality. Exp Clin Endocrinol 98: 161–170. Hofman MA, Swaab DF (1992). The human hypothalamus: comparative morphometry and photoperiodic influences. In: DF Swaab, MA Hofman, M Mirmiran, R Ravid, FW Van Leeuwen (Eds.), The human hypothalamus in health and disease. Elsevier, Amsterdam. Jansen AS, Ter Horst GJ, Mettenleiter TC et al. (1992). CNS cell groups projecting to the submandibular parasympathetic preganglionic neurons in the rat: a retrograde transneuronal viral cell body labeling study. Brain Res 572: 253–260. 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.

65

Kievit J, Kuypers HG (1975). Basal forebrain and hypothalamic connection to frontal and parietal cortex in the rhesus monkey. Science 187: 660–662. Kishi T, Tsumori T, Ono K et al. (2000). Topographical organization of projections from the subiculum to the hypothalamus in the rat. J Comp Neurol 419: 205–222. Kita H, Oomura Y (1982). An HRP study of the afferent connections to rat lateral hypothalamic region. Brain Res Bull 8: 63–71. Krayniak PF, Siegel A, Meibach RC et al. (1979). Origin of the fornix system in the squirrel monkey. Brain Res 160: 401–411. Kremer B, Swaab D, Bots G et al. (1991). The hypothalamic lateral tuberal nucleus in Alzheimer’s disease. Ann Neurol 29: 279–284. LeGros Clark WE (1936). The topography and homologies of the hypothalamic nuclei in man. J Anat 70: 203–216. Levay S (1991). A difference in hypothalamic structure between heterosexual and homosexual men. Science 253: 1034–1037. Mai JK, Assheuer J, Paxinos G (2008). Atlas of the human brain, third edn. Academic Press. Mesulam MM, Mufson EJ, Levey AI et al. (1983). Cholinergic innervation of cortex by the basal forebrain: cytochemistry and cortical connections of the septal area, diagonal band nuclei, nucleus basalis (substantia innominata), and hypothalamus in the rhesus monkey. J Comp Neurol 214: 170–197. Moore RY (1993). Organization of the primate circadian system. J Biol Rhythms 8 Suppl: S3–S9. Mouri T, Takahashi K, Kawauchi H et al. (1993). Melaninconcentrating hormone in the human brain. Peptides 14: 643–646. Nathan PW, Smith MC (1986). The location of descending fibres to sympathetic neurons supplying the eye and sudomotor neurons supplying the head and neck. J Neurol Neurosurg Psychiatry 49: 187–194. Nauta WJH, Haymaker W (1969). Hypothalamic nuclei and fiber connections. In: W Haymaker, E Anderson, WJH Nauta (Eds.), The hypothalamus. Charles C. Thomas, Springfield, IL, pp. 136–209. Nieuwenhuys R, Voogd J, van Huijzen C (1988). The human central nervous system, third edn. Sringer-Verlag. Ongur D, An X, Price JL (1998). Prefrontal cortical projections to the hypothalamus in macaque monkeys. J Comp Neurol 401: 480–505. Palkovits M, Zaborsky L (1979). Neural connections of the hypothalamus. In: PJ Morgane, J Panksepp (Eds.), Handbook of the hypothalamus. Marcel Dekker, Basel New York, pp. 379–509. Panula P, Airaksinen MS, Pirvola U et al. (1990). A histaminecontaining neuronal system in human brain. Neuroscience 34: 127–132. Porrino LJ, Goldman-Rakic PS (1982). Brainstem innervation of prefrontal and anterior cingulate cortex in the rhesus monkey revealed by retrograde transport of HRP. J Comp Neurol 205: 63–76.

66

B. DUDÁS

Powell TP, Guillery RW, Cowan WM (1957). A quantitative study of the fornixmamillo-thalamic system. J Anat 91: 419–437. Price JL, Amaral DG (1981). An autoradiographic study of the projections of the central nucleus of the monkey amygdala. J Neurosci 1: 1242–1259. Pritchard TC, Hamilton RB, Norgren R (2000). Projections of the parabrachial nucleus in the old world monkey. Exp Neurol 165: 101–117. Rempel-Clower NL, Barbas H (1998). Topographic organization of connections between the hypothalamus and prefrontal cortex in the rhesus monkey. J Comp Neurol 398: 393–419. Ricardo JA (1983). Hypothalamic pathways involved in metabolic regulatory functions, as identified by track-tracing methods. Adv Metab Disord 10: 1–30. Ricardo JA, Koh ET (1978). Anatomical evidence of direct projections from the nucleus of the solitary tract to the hypothalamus, amygdala, and other forebrain structures in the rat. Brain Res 153: 1–26. Sadun AA, Schaechter JD, Smith LE (1984). A retinohypothalamic pathway in man: light mediation of circadian rhythms. Brain Res 302: 371–377. Saper CB (1982). Convergence of autonomic and limbic connections in the insular cortex of the rat. J Comp Neurol 210: 163–173. Saper CB (1984). Organization of cerebral cortical afferent systems in the rat. II Magnocellular basal nucleus. J Comp Neurol 222: 313–342. Saper CB (2004). Hypothalamus. In: G Paxinos, JK Mai (Eds.), The human nervous system. Elsevier 513–550. Saper CB, Loewy AD (1980). Efferent connections of the parabrachial nucleus in the rat. Brain Res 197: 291–317. Saper CB, Loewy AD, Swanson LW et al. (1976). Direct hypothalamo-autonomic connections. Brain Res 117: 305–312. Saper CB, Swanson LW, Cowan WM (1978). The efferent connections of the anterior hypothalamic area of the rat, cat and monkey. J Comp Neurol 182: 575–599. Saper CB, Swanson LW, Cowan WM (1979). Some efferent connections of the rostral hypothalamus in the squirrel monkey (Saimiri sciureus) and cat. J Comp Neurol 184: 205–241. Sesack SR, Deutch AY, Roth RH et al. (1989). Topographical organization of the efferent projections of the medial prefrontal cortex in the rat: an anterograde tract-tracing study with Phaseolus vulgaris leucoagglutinin. J Comp Neurol 290: 213–242.

Shen CL (1983). Efferent projections from the mammillary complex of the guinea pig: an autoradiographic study. Brain Res Bull 11: 43–59. Sutin J (1966). The periventricular stratum of the hypothalamus. Int Rev Neurobiol 9: 263–300. Swaab DF (2003). The human hypothalamus: basic and clinical aspects. Part 1: nuclei of the human hypothalamus, first edn Elsevier. Swaab DF (2004). The human hypothalamus: basic and clinical aspects. Part 2: neuropathology of the human hypothalamus and adjacent structures, first edn Elsevier. Swaab DF, Fliers E (1985). A sexually dimorphic nucleus in the human brain. Science 228: 1112–1115. Swanson LW, Cowan WM (1975). Hippocampo-hypothalamic connections: origin in subicular cortex, not ammon’s horn. Science 189: 303–304. Swanson LW, Cowan WM (1977). An autoradiographic study of the organization of the efferent connections of the hippocampal formation in the rat. J Comp Neurol 172: 49–84. Swanson LW, Kuypers HG (1980). A direct projection from the ventromedial nucleus and retrochiasmatic area of the hypothalamus to the medulla and spinal cord of the rat. Neurosci Lett 17: 307–312. Tigges J, Walker LC, Tigges M (1983). Subcortical projections to the occipital and parietal lobes of the chimpanzee brain. J Comp Neurol 220: 106–115. Timmers HJ, Swaab DF, van de Nes JA et al. (1996). Somatostatin 1-12 immunoreactivity is decreased in the hypothalamic lateral tuberal nucleus of Huntington’s disease patients. Brain Res 728: 141–148. Valenstein ES, Nauta WJ (1959). A comparison of the distribution of the fornix system in the rat, guinea pig, cat, and monkey. J Comp Neurol 113: 337–363. Vann SD, Aggleton JP (2004). The mammillary bodies: two memory systems in one? Nat Rev Neurosci 5: 35–44. Weindl A (2007). Commissura supraoptica ventralis, Nucleus tegmentalis dorsalis und Nucleus interpedencularis: drei von Bernhard von Gudden beschriebene neuroanatomische Strukturen—einst und jetzt. In: H Hippius, R Steinberg (Eds.), Bernhard von Gudden. Springer, Berlin, Heidelberg, pp. 59–65. Willis MA, Haines DE (2018). The limbic system. In: DE Haines, GA Mihailoff (Eds.), Fundamental neuroscience for basic and clinical applications. Elsevier, New York. Zhang YH, Lu J, Elmquist JK et al. (2000). Lipopolysaccharide activates specific populations of hypothalamic and brainstem neurons that project to the spinal cord. J Neurosci 20: 6578–6586.

Handbook of Clinical Neurology, Vol. 179 (3rd series) The Human Hypothalamus: Anterior Region D.F. Swaab, F. Kreier, P.J. Lucassen, A. Salehi, and R.M. Buijs, Editors https://doi.org/10.1016/B978-0-12-819975-6.00002-9 Copyright © 2021 Elsevier B.V. All rights reserved

Chapter 4

Morphology and distribution of hypothalamic peptidergic systems BERTALAN DUDÁS1,2* AND ISTVÁN MERCHENTHALER3 1

Neuroendocrine Organization Laboratory, Lake Erie College of Osteopathic Medicine, Erie, PA, United States 2

Department of Anatomy, Histology and Embryology, University of Szeged, Szeged, Hungary

3

Department of Epidemiology and Public Health and of Anatomy and Neurobiology, University of Maryland Baltimore, Baltimore, MD, United States

Abstract Neuropeptides participate in the regulation of numerous hypothalamic functions that are aimed for sustaining the homeostasis of the organism. These neuropeptides can act in two different levels. They can influence the release of hormones from the adenohypophysis via the portal circulation; in addition, they can act as neurotransmitters/neuromodulators modulating the functioning of numerous hypothalamic neurotransmitter systems. Indeed, most of these peptidergic systems form a complex network in the infundibular and periventricular nuclei of the human hypothalamus, communicating with each other by synaptic connections that may control fundamental physiologic functions. In the present chapter, we provide an overview of the distribution of neuropeptides in the human hypothalamus using immunohistochemistry and high-resolution, three-dimensional mapping.

INTRODUCTION The hypothalamus, as the main autonomic center in the brain, plays a pivotal role in the regulation of endocrine functions and maintaining homeostasis. The two major hypothalamic cell groups are the magno- and parvicellular neuronal systems. They do not only have a different size of perikarya but also different projection sites. The cell bodies of the hypothalamic magnocellular system project to the posterior pituitary where among others, oxytocin and vasopressin are released from the axon terminals into the general circulation (Moller, 2021). The parvicellular cell groups form associations with the hypophysial portal vessels in the median eminence and release the neuropeptides into the hypophysial portal circulation. These neurons are called the “hypohysiotropic”

neurons as they target the anterior pituitary. The peptides that these parvicellular neurons synthesize and release into the hypophysial portal system are the releasing and inhibiting hormones as they, respectively, either stimulate or inhibit the release of hormones of the anterior pituitary that in turn regulate the functioning of peripheral endocrine organs. In addition to releasing peptides into the hypophysial portal circulation, hypothalamic neurons also utilize these substances as neurotransmitters or neuromodulators to communicate with other hypothalamic or extrahypothalmic neuronal systems. In this chapter, we provide a current status of the distribution of neuropeptides in the human hypothalamus using immunohistochemistry as well as high-resolution, three-dimensional mapping.

*Correspondence to: Bertalan Dudás, M.D., Ph.D., M.S. (Med. Ed.), Professor of Anatomy, Assistant Dean of Research and Scholarship, Anatomy, Lake Erie College of Osteopathic Medicine (LECOM), 1858 West Grandview Blvd, Erie, PA 16509, United States. Tel: +1-814-866-8142, Fax: +1-814-866-8411, E-mail: [email protected]

68

B. DUDÁS AND I. MERCHENTHALER

HYPOTHALAMIC PEPTIDERGIC SYSTEMS IN HUMAN Corticotropin-releasing hormone system Corticotropin-releasing hormone (CRH), also called corticoliberin, is a 41 amino acid peptide (Vale et al., 1981) involved in stress response by stimulating the release of adrenocortocotropin hormone (ACTH) as well as other biologically active substances (e.g., b-endorphin) from the anterior pituitary. CRH is the major hypothalamic activator of the hypothalamic–pituitary–adrenal axis. In addition, CRH is an important mediator of the stress response, which integrates endocrine, autonomic, immunologic, and behavioral reflexes. Beyond its role as an endocrine hormone, increasing evidence suggests a role for CRH in the regulation of gastrointestinal and reproductive functions. In addition to being a hypophysiotropic hormone stimulating the release of ACTH, CRH is widely distributed in the central nervous system (CNS) and as a neurotransmitter/neuromodulator plays a role in numerous physiological (Valentino, 1989; Dedic et al., 2018) and pathophysiological functions (Raadsheer et al., 1995; Goncharuk et al., 2007; Holsboer and Ising, 2021). Morphologically, CRH-immunoreactive (IR) perikarya are mainly fusiform-shaped (Fig. 4.1A); however, few multipolar cells are also present in periventricular zone of the hypothalamus (Fig. 4.1B). The vast majority of CRH-IR perikarya are distributed periventricularly in the preoptic region, the paraventricular nucleus (PVN), tuberal, and posterior hypothalamic regions (Fig. 4.2). Numerous CRH-IR neurons are located around the anterior commissure and the fornix. A small number of CRH-IR perikarya populate the periventricular region of the posterior hypothalamus and the zone around the

mamillary body. Several CRH-IR perikarya receive abutting CRH-IR fiber varicosities forming multiple contacts while passing by (Fig. 4.1A and B). CRH-IR fibers are arranged into numerous bundles. CRH-IR axons emerging from the infundibulum project laterally, running parallel to the basal surface of the hypothalamus and arch over the optic tract. From the PVN, CRH-IR fibers project horizontally, running laterally through the substantia innominata or basally toward the infundibulum/median eminence, surrounding the fornix. In addition, numerous CRH-IR fibers occupy the periventricular zone and surround blood vessels in the PVN (Fig. 4.1C) as well as in the periventricular and tuberal regions (Fig. 4.1D). Occasional CRH-IR fibers are scattered around the anterior commissure and in the posterior hypothalamus, surrounding the mamillary bodies.

Gonadotropin-releasing hormone system Gonadotropin-releasing hormone (GnRH, also called luteinizing hormone-releasing hormone [LHRH]) is a decapeptide representing the final common pathway of a neuronal network that integrates multiple external and internal factors to control reproduction. Its primary function is to stimulate the release of luteinizing hormone (LH) and follicle-stimulating hormone (FSH) from the anterior pituitary. As such, GnRH is the central regulator of the menstrual cycle and ovulation. A unique feature of the GnRH neuronal system is that the neurons originate from the nasal placode and migrate into the brain (Cariboni and Balasubramanian, 2021) where they occupy a relatively restricted area in the infundibulum, the preoptic area, and medial septum. The majority of GnRH perikarya in primates are in the infundibulum (Barry, 1976; King and Anthony, 1984; Rance et al., 1994; Dudas et al., 2000),

Fig. 4.1. Corticotropin-releasing hormone (CRH)-immunoreactive (IR) elements (black) in the human hypothalamus. CRH-IR fibers contact CRH-IR fusiform (A) and multipolar (B) neurons in the paraventricular nucleus (some juxtapositions are marked with arrows). CRH-IR neuronal elements often surround vessels in the paraventricular nucleus (C) and in the median eminence (D). The vessel lumens are marked by an asterisk. Magnification: 400x (A, B) and 200x (C, D).

MORPHOLOGY AND DISTRIBUTION OF HYPOTHALAMIC PEPTIDERGIC SYSTEMS

69

Fig. 4.2. Stereoscopic images of the human hypothalamus reconstituted from 30-mm-thick sections, illustrating the distribution of the corticotropin-releasing hormone (CRH)-immunoreactive (IR) perikarya. Stereoscopic images can be seen using U or parallel vision. The eyes are relaxed to look into the distance until the pair of the images fuse and then refocused by the brain. With this technique, a 3D hypothalamus can be seen on the figure, floating in front of the paper, with the immunolabeled perikarya in it at different depth marked by the dots. The optimal viewing distance is about 50 cm from the paper surface (average reading distance). A stereoscopic viewer, or stereoscope, is suggested to readers unfamiliar with U or parallel vision (a stereoscope is an inexpensive device for viewing a stereoscopic pair of separate images, depicting left-eye and right-eye views of the same scene, as a single three-dimensional image). Abbreviations: AC, anterior commissure; Inf, infundibulum; MB, mamillary body; and OCh, optic chiasm.

Fig. 4.3. Gonadotropin-releasing hormone (GnRH) elements in the lamina terminalis. Most of the GnRH neurons are fusiform in shape.

while in rodents most of them are in the preoptic area, diagonal band of Broca, and medial septum (Merchenthaler et al., 1980). The secretion of GnRH into the hypophysial portal blood is pulsatile. Low-frequency GnRH pulses lead to FSH release and follicle maturation in the ovaries, whereas high-frequency GnRH pulses stimulate LH release and subsequently ovulation (Hurley et al., 1983; Miller et al., 1983; Sullivan et al., 1999). As all the other hypophysiotropic peptides, GnRH also functions as a neurotransmitter/neuromodulator and regulates the activity of several neuronal systems (Hsueh and Schaeffer, 1985). GnRH-IR perikarya are fusiform with thin cell bodies and two processes on the opposite poles of the perikarya (Fig. 4.3). In primates, numerous multipolar neurons with triangle-shaped or rounded cell bodies can also be detected, often populating the preoptic region. Diencephalic GnRH-IR neurons are not clustered into nuclei;

instead, they are scattered throughout the human hypothalamus forming a loose network (Fig. 4.4). The density of GnRH-IR neuronal elements decreases in the mediolateral and rostrocaudal direction. The vast majority of the immunolabeled neurons in humans are localized in the basal part of the infundibular region, but they also populate the medial parts of the periventricular zone of the hypothalamus, occupying primarily the preoptic area. The number of labeled cells gradually declines in the mediolateral direction and approximately 20 mm laterally from the ependymal surface of the third ventricle virtually no GnRH-IR perikarya can be observed. Numerous immunolabeled neurons are present in the septal area along the diagonal band of Broca and in the lamina terminalis cinerea. Few GnRH neurons can also be detected in the paraventricular and supraoptic nuclei and in the posterior hypothalamus around the mamillary bodies. A similar pattern of GnRH neurons in the human diencephalon has also been reported by others (King et al., 1985; Kuljis and Advis, 1989; Bloch et al., 1992; Kordon et al., 1994; Rance et al., 1994; Silverman et al., 1994; Dudas et al., 2000). GnRH-IR fibers with axon varicosities form loose bundles in the diagonal band of Broca and above the dorsal surface of the optic chiasm running through the medial preoptic area toward the tuberal region. Numerous fibers can be observed periventricularly under the ependymal surface of the third ventricle and running parallel to the basal surface of the tuberal region, where they were reported to form occasional juxtapositions with GnRH perikarya (Dudas et al., 2000). Occasional

70

B. DUDÁS AND I. MERCHENTHALER

Fig. 4.4. Stereoscopic images of the human hypothalamus reconstituted from 30-mm-thick sections, illustrating the distribution of gonadotropin-releasing hormone (GnRH), also called luteinizing hormone-releasing hormone (LHRH) (dots). Stereoscopic images can be seen using U or parallel vision as described for Fig. 4.2. Abbreviations: AC, anterior commissure; Inf, infundibulum; MB, mamillary body; and OCh, optic chiasm.

GnRH-IR axons occupy the lateral hypothalamic zone and the posterior hypothalamus, primarily around the mamillary bodies.

Growth hormone-releasing hormone system Growth hormone-releasing hormone (GHRH) is a 44 amino acid peptide and it is the most confined hypophysiotropic system in the brain. GHRH stimulates the release of growth hormone (GH) from the anterior pituitary, which is required for normal postnatal growth and participates in protein, carbohydrate, and lipid metabolism. Acting as a neurotransmitter/neuromodulator, GHRH also promotes slow-wave sleep by modulating GABA-ergic transmission in the ventrolateral and median preoptic nuclei (Obal Jr. et al., 2001; Obal Jr. and Krueger, 2001). GHRH-immunoreactivity first appears in the human hypothalamus between 18 and 29 weeks of gestation, which corresponds to the start of production of GH and other somatotropes in the fetus (Aubert et al., 1977; Bugnon et al., 1977; Ishikawa et al., 1986; Charnay et al., 1987; Kedzia et al., 2009). The actions of GHRH are opposed by somatostatin (growth hormone release-inhibiting hormone). Somatostatin is also released into the hypothalamo-hypophysial portal circulation where it inhibits GH secretion. Somatostatin (SS) and GHRH are secreted in alternation, giving rise to the markedly pulsatile secretion of GH (Muller, 1987; Veldhuis, 2008). GHRH-IR perikarya are confined to the basal hypothalamus, located almost exclusively in the basal part of the infundibular region (Merchenthaler et al., 1984; Deltondo et al., 2008; Korf and Moller, 2021) (Figs. 4.5 and 4.6). Here, the perikarya are clustered into four well-defined subdivisions: (1) The majority of the GHRH-IR cell bodies are in the infundibulum/median

eminence and (2) in the basal part of the periventricular zone. (3) A group of neurons occupies the dorsomedial subdivision of the ventromedial nucleus and (4) the basal perifornical area of the tuberal region (Figs. 4.5 and 4.6). GHRH-IR perikarya can occasionally be observed in the medial preoptic area and in the posterior hypothalamus but not in the lateral hypothalamus. GHRH-IR fibers are present in the basal part of the infundibulum and the basal periventricular area. The number of immunoreactive fibers decreases dorsally. Numerous axons are present in the basal part of the medial hypothalamus, oriented parallel with the pial surface. Immunoreactive fibers can also be seen perifornically in the preoptic and tuberal regions. Few labeled axon varicosities are located in the basal zone of the lateral hypothalamus, the lamina terminalis cinerea, around the medial part of the mamillary bodies, and in the medial and lateral zones of the paraventricular nuclei, while virtually no GHRH-IR fibers can be found in the supraoptic nuclei.

Somatostatinergic system Somatostatin or somatotropin release-inhibiting hormone or GH-inhibiting hormone is a 14–28 amino acid peptide that inhibits GH release from the pituitary. Contrary to GHRH, whose location is restricted to the infundibulum, SS is distributed not only in multiple sites of the human brain but in several peripheral tissues including the gastrointestinal system and the endocrine pancreas. As such, SS not only regulates GH secretion from the pituitary but also affects neurotransmission and cell proliferation acting on G protein-coupled receptors (GPCRs) and inhibits the release of numerous secondary hormones, e.g., insulin, glucagon, gastrin, and histamine (Thorner et al., 1990; Herzig et al., 1994;

MORPHOLOGY AND DISTRIBUTION OF HYPOTHALAMIC PEPTIDERGIC SYSTEMS

71

Fig. 4.5. Growth hormone-releasing hormone (GHRH)-immunoreactive (IR) elements in the human hypothalamus. GHRH-IR neurons are most abundant in the infundibulum/median eminence (A) and in the basal part of the periventricular area (B). Perikarya and axonal varicosities are scarcer in the dorsomedial subdivision of the ventromedial nucleus (C) and the basal perifornical area (D). Six-pointed asterisk marks the third ventricle, while the cross section of a portal vessel is denoted by a five-pointed asterisk. Magnification: 100x.

Liguz-Lecznar et al., 2016). Somatostatinergic neurons in the periventricular nucleus mediate the negative feedback effects of GH on its own release. Somatostatinergic neurons respond to the high-circulating concentrations of GH and somatomedins by increasing the release of somatostatin, which in turn reduces the rate of GH secretion. Somatostatin-IR perikarya are located in the infundibulum/median eminence (Korf and Moller, 2021) and in the periventricular area of the preoptic regions (Figs. 4.7 and 4.8). Numerous SS-IR neurons populate the suprachiasmatic and ventromedial nuclei and the nucleus of the diagonal band of Broca. Cell bodies can also be observed around the mamillary nuclei and in the supramamillary nuclei. Few SS-IR perikarya are scattered perifornically at the tuberal region (Fig. 4.7C). Several somatostatinergic perikarya can be found in the paraventricular nuclei and in the lateral hypothalamus, predominantly at the infundibular and posterior hypothalamic regions. The lamina terminalis cinerea contains few SS-IR perikarya with scattered fiber varicosities. Somatostatin-IR fibers are abundant in the infundibulum, periventricular area of the preoptic and infundibular regions, and in the nucleus tuberalis lateralis (Timmers et al., 1996), while the medial hypothalamic regions contain only a small number of somatostatin-IR fibers. In rodents, the hypophysiotropic SS neurons are located in the anterior periventricular nucleus (Merchenthaler et al., 1989; Merchenthaler, 1990; Fodor et al., 2006), but no data are available on the location of hypophysiotropic SS neurons in humans. The wide distribution of SS in the hypothalamus and many extrahypothalamic brain

areas suggest that like other hypophysiotropic peptides, it also functions as a neurotransmitter/neuromodulator.

Thyrotropin-releasing hormone system Thyrotropin-releasing hormone (TRH), also termed thyroliberin, a tripeptide amide, (Boler et al., 1969; Burgus et al., 1969) has a critical role in the central regulation of thyroid hormone-stimulating hormone (TSH) secretion from the anterior pituitary, and subsequently, thyroid hormone secretion from the thyroid gland (Lechan et al., 2009; Boelen et al., 2021). The hypothalamic–pituitary–thyroid axis primarily functions to maintain normal, circulating levels of thyroid hormone that is essential for the biological function of many tissues, including, among others, brain development, regulation of cardiovascular, bone, and liver function, food intake, and energy expenditure (Lechan and Fekete, 2006; Lechan et al., 2009). In addition to its role in the regulation of TSH secretion, TRH also serves as a potent regulator of prolactin (PRL) secretion from the anterior pituitary (Freeman et al., 2000; Grattan, 2015) by stimulating PRL secretion either directly from lactotrophs or indirectly via its action on tuberoinfundibular dopamine neurons. In rodents, the hypophysiotropic TRH neurons are located in the medial subdivision of the PVN (Ishikawa et al., 1988; Lechan and Segerson, 1989; Kawano et al., 1991; Merchenthaler and Liposits, 1994; Kadar et al., 2010). In humans, the PVN also contains a large population of TRH neurons, especially in its medial part, but the exact location of hypophysiotropic TRH neurons has not been elucidated yet (Fliers et al., 1994; Mihaly et al., 2001).

72

B. DUDÁS AND I. MERCHENTHALER

Fig. 4.6. Stereoscopic images of the human hypothalamus reconstituted from 30-mm-thick sections, illustrating the distribution of growth hormone-releasing hormone (GHRH)-immunoreactive (IR) perikarya (dots). Stereoscopic images can be seen using U or parallel vision as described for Fig. 4.2. Abbreviations: AC, anterior commissure; Inf, infundibulum; MB, mamillary body; and OCh, optic chiasm.

Fig. 4.7. Somatostatinergic neurons in the human hypothalamus. The infundibulum (A) and paraventricular nucleus (B) contains dense network of perikarya and axonal varicosities. Five-pointed asterisk denotes the pial surface, while six-pointed asterisk marks the third ventricle. Perifornical neurons are characteristically fusiform (C). Supramamillary nucleus contains somatostatinergic perikarya (D). At the periventricular region, somatostatinergic fibers occasionally surround neurons that are not immunoreactive for somatostatin (E). Magnification: 400x (A, B) and 200x (C–E).

Fig. 4.8. Stereoscopic images of the human hypothalamus reconstituted from 30-mm-thick sections, illustrating the distribution of somatostatinergic perikarya. Stereoscopic images can be seen using U or parallel vision as described for Fig. 4.2. Abbreviations: AC, anterior commissure; Inf, infundibulum; MB, mamillary body; and OCh, optic chiasm.

MORPHOLOGY AND DISTRIBUTION OF HYPOTHALAMIC PEPTIDERGIC SYSTEMS The activity of hypophysiotropic TRH neurons is regulated by the negative feedback effects of thyroid hormone and by neuronal inputs from several brain areas operating with multiple neurotransmitters (for a review, see Fekete and Lechan, 2014). Like other hypophysiotropic peptides, TRH also functions as a neuromodulator/neurotransmitter (Hokfelt et al., 1980; Swanson and Sawchenko, 1980; Landgraf and Neumann, 2004; Russell, 2018), and as such, it regulates a wide variety of physiological and behavioral functions (for reviews, see Russell, 2018; Frohlich and Wahl, 2019). TRH-IR perikarya in humans form a loose network of bipolar (Fig. 4.9A) and multipolar (Fig. 4.9B) neurons that are arranged in several clusters. Periventricularly arranged cells are located primarily in the rostral preoptic area and their number declines in caudal direction. TRHIR neurons can be found in the parvicellular parts of the PVN, primarily in the periventricular zone (Fig. 4.10). At the preoptic region, cells are typically arranged above the optic chiasm forming a well-circumscribed group. This subpopulation extends superiorly and occupies the medial hypothalamus extending to the lateral hypothalamic zone as well as to the infundibular region (Fig. 4.10). A subgroup of these neurons follows the basal surface of the brain above the optic tracts. The suprachiasmatic nucleus contains a small number of TRH-IR neurons. Numerous, mostly bipolar neurons are located perifornically and follow the curvature of the fornix with their processes (Fig. 4.9). At the infundibular region, the ventrolateral and dorsomedial subdivisions of the ventromedial nucleus contain relatively well-defined subgroups of TRH-IR

73

neurons (Fig. 4.10). Numerous cells can also be observed in the dorsomedial nucleus. In the infundibular nucleus, cell bodies are sparse. The posterior hypothalamus contains a relatively well-circumscribed cell group located in the supramamillary nucleus and occasional perikarya in the posterior medial hypothalamic area (Fig. 4.10). TRH-IR fibers often form a loose network of terminals or well-defined fiber baskets surrounding unlabeled neurons. These baskets are typically arranged periventricularly (Fig. 4.2C) and perifornically (Fig. 4.9D), but several of them can be seen in the medial and occasionally in the lateral hypothalamic zones (Fig. 4.9E). TRHIR axonal varicosities can also be seen in close proximity of vessels, primarily in the medial hypothalamic and infundibular areas. The infundibulum contains a dense network of TRH-IR fibers apparently descending toward the median eminence. In the septal region, few axonal varicosities can be observed along the diagonal band of Broca, around the contours of the anterior commissure, and in the lamina terminalis cinerea. In the posterior hypothalamus, TRH-IR fibers form a relatively dense network in the medial hypothalamic area above the medial mamillary nuclei.

b-endorphin-system b-endorphin is a 31 amino acid peptide, deriving from processing of the precursor proopiomelanocortin (POMC) by prohormone convertases (Li et al., 1976). POMC processing also gives rise to other peptide hormones, including adrenocorticotropic hormone, as well a- and g-melanocyte stimulating hormone. b-endorphin belongs to the endogenous opiates family, which also

Fig. 4.9. Thyrotropin-releasing hormone (TRH)-immunoreactive (IR) elements in the human hypothalamus. Periventricular and perifornical TRH neurons are typically bipolar (A). Multipolar cells can be often observed in the lateral hypothalamus (B). TRH-IR axon varicosities often form fiber baskets surrounding neurons that are obviously not TH or TRH immunoreactive in the periventricular (C) and perifornical region (D). Occasionally, these fiber baskets can also be observed in the lateral hypothalamic area (E). Magnification: 400x.

74

B. DUDÁS AND I. MERCHENTHALER

Fig. 4.10. Stereoscopic images of the human hypothalamus reconstituted from 30-mm-thick sections, illustrating the distribution of TRH-IR perikarya (dots). Stereoscopic images can be seen using U or parallel vision as described for Fig. 4.2. Abbreviations: AC, anterior commissure; Inf, infundibulum; MB, mamillary body; and OCh, optic chiasm.

Fig. 4.11. b-endorphin-immunoreactive (IR) elements in the human hypothalamus. (A) Fusiform and (B) multipolar b-endorphinIR neurons contact b-endorphin-IR axon varicosities (arrowheads) in the infundibulum. (C) b-endorphin-IR fibers are in close proximity of portal vessels in the median eminence. The lumen of the vessel is denoted by asterisk. Scale bar: 10 mm (A, B); 40 mm (C). Reprinted with permission from Elsevier. Dudas B, Merchenthaler I (2004). Topography and associations of b-endorphin and luteinizing hormone-releasing hormone neuronal systems in the human diencephalon. Neuroscience 124: 221–229.

includes met-enkephalin, leu-enkephalin, and dynorphin. b-endorphin binds primarily to m-receptors with high affinity but also binds to d- and k-receptors with lower affinity. b-endorphin provides analgesia and a feeling of well-being. As an important neurotransmitter/neuromodulator, it also regulates multiple functions in the brain and periphery (Simantov and Snyder, 1976). b-endorphin-IR neurons are fusiform in shape with two processes emanating from the opposite poles of the cells (Fig. 4.11A), but a few multipolar cells located in the infundibular nucleus (Fig. 4.11B) are also present. The b-endorphin-IR neurons are located in a single, welldefined cell cluster in the infundibulum/median eminence of the human diencephalon (Sukhov et al., 1995) (Fig. 4.12), where axons are often found in close proximity to portal vessels (Dudas and Merchenthaler, 2004, 2006) (Fig. 4.11C), and often form juxtapositions with b-endorphin-IR perikarya. These potential connections with portal vessels have not been seen in rodents in retrograde labeling experiments (Merchenthaler et al., 1989; Merchenthaler, 1990) suggesting species

differences in endorphin functions. A similar pattern of POMC expression has been previously reported by Sukhov et al. (1995) in the human hypothalamus (Korf and Moller, 2021). b-endorphin-IR fibers are in the periventricular zone of the hypothalamus, where they form a loose network of immunoreactive axon varicosities. Such axons can also be observed along the diagonal band of Broca, at the basal part of the lamina terminalis cinerea and around the anterior commissure. At the medial preoptic region, most of the b-endorphin-IR fibers are located periventricularly, but immunoreactive axons can also be seen in the dorsomedial subdivision of the ventromedial nucleus. A delicate b-endorphin-IR fiber network is associated with the portal vessels in the infundibulum/median eminence (Fig. 4.11C), projecting laterally from the infundibulum at the base of the diencephalon. The lateral hypothalamic zone contains only few fibers. In the posterior hypothalamus, scattered b-endorphin-IR axons populate the area around the mamillary bodies and the tuberomamillary nucleus.

MORPHOLOGY AND DISTRIBUTION OF HYPOTHALAMIC PEPTIDERGIC SYSTEMS

75

Fig. 4.12. Stereoscopic images of the human hypothalamus reconstituted from 30-mm-thick sections, illustrating the distribution of beta-endorphin-immunoreactive (IR) perikarya (dots). Stereoscopic images can be seen using U or parallel vision as described for Fig. 4.2. Abbreviations: AC, anterior commissure; Inf, infundibulum; MB, mamillary body; and OCh, optic chiasm.

Enkephalin system Enkephalins (leu- and met-enkephalin) are pentapeptides belonging to the endogenous opioid peptides family (Noda et al., 1982). Both are products of the proenkephalin gene. The amino acid sequence of Met-enkephalin is Tyr-Gly-Gly-Phe-Met and of Leu-enkephalin is Tyr-GlyGly-Phe-Leu. Both bind preferentially to delta opioid receptors. Enkephalins are widely distributed in the CNS and they regulate a large variety of CNS functions including, among others nociception, mood, movement, and neuroendocrine functions (see Cullen and Cascella, 2020 for review). Morphologically, the majority of the leu-enkephalinIR perikarya are fusiform in shape and they are characteristically oriented with the axis of the fusiform cell bodies running parallel to the ependymal surface of the third ventricle (Fig. 4.13A). The vast majority of the leu-enkephalin-IR perikarya (82%) are located in the periventricular area of the tuberal region (Fig. 4.14) with only few immunoreactive cells in the periventricular zone of the medial preoptic area (Fig. 4.14). Leu-enkephalin-IR fibers are generally oriented periventricularly along the medial surface of the hypothalamus (Fig. 4.13A). An additional fiber network appears to project from the infundibulum toward the lateral hypothalamic area. Leu-enkephalin-IR fibers are also located at the lateral part of the anterior commissure, around the fornix, and in the dorsal part of the lateral hypothalamic area. In the infundibulum, axon varicosities can also be detected in close proximity to portal vessels (Fig. 4.13D). These potential connections with portal vessels have not been seen in rodents in retrograde labeling experiments (Merchenthaler, 1992), suggesting species differences in enkephalin functions. Numerous leu-enkephalin-IR fiber baskets can be observed surrounding fusiform unidentified neurons and covering

the majority of their surface in the periventricular area of the tuberal and preoptic regions (Fig. 4.13B and C).

Kisspeptin/neorkinin B/dynorphin system Hypogonadotropic hypogonadism is a disorder characterized by low gonadotropin levels leading to gonadal dysfunctions. In 2003, two independent groups discovered almost simultaneously that idiopathic hypogonadotropic hypogonadism was caused by disabling mutations of a GPCR, GPR54 (de Roux et al., 2003; Seminara et al., 2003). Although GPR54 shares a modest sequence homology with the known galanin receptors, galanin apparently does not bind specifically to this receptor (Lee et al., 1999), and the natural ligand of GPR54 was unknown at that time. In 2001, three groups discovered that the peptide metastatin was the natural ligand of the previously orphaned receptor (Kotani et al., 2001; Muir et al., 2001; Ohtaki et al., 2001). Metastatin suppresses the metastasis of melanomas (Lee et al., 1996), it is derived from a larger protein called kisspeptin (KP), and it is the product of the Kiss1 gene that was originally isolated as a tumor metastasis gene. The original observations were corroborated by studies of mice bearing targeted deletions of GPR54, where it was noted that the only remarkable phenotypic anomaly was reproductive dysfunction (Funes et al., 2003; Seminara et al., 2003). Thus, KP-GPR54 signaling appears to be essential to initiate gonadotropin secretion at puberty and support reproductive function in the adult. In addition to innervating GnRH neurons, KP neurons in the hypothalamus also project to several limbic structures by which KP in humans can integrate sexual and emotional brain processing with reproduction (Hrabovszky, 2014, 2021). Kisspeptin (KiSS mRNA)-expressing neurons in the human and rodent brain are localized in two discrete regions: (i) one in the anteroventral periventricular

76

B. DUDÁS AND I. MERCHENTHALER

Fig. 4.13. Leu-enkephalin-immunoreactive (IR) elements in the human hypothalamus. (A) Leu-enkephalin-IR axon varicosities contact a fusiform, leu-enkephalin-IR neuron (arrowheads) in the periventricular area of the tuberal region. Leu-enkephalin-IR fibers occasionally form fiber baskets around fusiform neurons (asterisks) that are apparently not immunoreactive for leuenkephalin in the (B) dorsal part of the tuberal periventricular region and (C) in the medial preoptic area. (D) Leu-enkephalinIR fibers are in close proximity of the portal vessels (V) in the median eminence. Scale bar: 10 mm (A–C); 50 mm (D).

Fig. 4.14. Stereoscopic images of the human hypothalamus reconstituted from 30-mm-thick sections, illustrating the distribution of leu-enkaphalin (ENK)-immunoreactive (IR) perikarya (dots). Stereoscopic images can be seen using U or parallel vision as described for Fig. 4.2. Abbreviations: AC, anterior commissure; Inf, infundibulum; MB, mamillary body; and OCh, optic chiasm.

nucleus (AVPV), the anterior periventricular nucleus, the anterodorsal preoptic nucleus and (ii) the other in the infundibulum/arcuate nucleus (Muir et al., 2001; Gottsch et al., 2004; Smith and Clarke, 2007). Kisspeptin-IR fibers have been shown to project into regions including the arcuate (/infundibular) and dorsomedial nuclei, the preoptic area, the retrochiasmatic area, and the zona incerta (Brailoiu et al., 2005). Steroid regulation of kisspeptin (KiSS-1) expression in the strategically two important regions of the hypothalamus, i.e., the anteroventral periventricular nucleus and the arcuate/ infundibular nuclei, where KP neurons are located (Hrabovszky et al., 2010; Lehman et al., 2010) is different. In the AVPV, where androgen receptors and both estrogen receptor-alpha (ERa) and ERb are expressed (Simerly et al., 1990; Hagihara et al., 1992; Shughrue et al., 1997), gonadectomy decreases while sex steroid hormone replacement increases KiSS-1 expression. In

the arcuate/infundibular nucleus, however, the changes are the opposite, i.e., gonadectomy increases while hormone replacement decreases the expression of KiSS-1 (Smith et al., 2005a,b). Thus, the activity of KP neurons in the arcuate nucleus is stimulated by gonadectomy and inhibited by sex steroids. If KP neurons in the arcuate nucleus provide tonic inhibitory input to GnRH neurons, it seems plausible that KP neurons could mediate the negative feedback effects of steroids on GnRH secretion by inhibiting GnRH neurons when sex steroids levels rise. On the contrary, the KP neurons in the AVPV clearly behave differently. Since the AVPV has been implicated in the generation of the preovulatory GnRH/LH surge in the female (Gu and Simerly, 1997) and estrogen upregulates KiSS-1 expression in this region, KP neurons in the AVPV should participate in the generation of the preovulatory GnRH/LH surge (for a recent review, see Moore et al., 2018).

MORPHOLOGY AND DISTRIBUTION OF HYPOTHALAMIC PEPTIDERGIC SYSTEMS

77

Kisspeptin is colocalized with a range of neuropeptides. Unlike the preoptic KP neuronal population, KP neurons in the medial basal hypothalamus colocalize with the tachykinin peptide neurokinin B (NKB) in monkeys (Ramaswamy et al., 2010) and humans (Hrabovszky et al., 2010). The degree of colocalization is age and sex (estrogen) dependent. The extent of KP colocalization in NKB neurons is higher in aged than young individuals (Molnar et al., 2012; Yang et al., 2012), and it is particularly higher in postmenopausal than cycling women (Molnar et al., 2012). Unlike in rodents (Burke et al., 2006; Ciofi et al., 2006; Navarro et al., 2011), a small percentage of KP neurons colocalize dynorphin. The number of these KNDy (KP, NKB, dynorphin) neurons in the human hypothalamus is low compared with rodents (Hrabovszky et al., 2012). In addition to their roles in GnRH pulse generation and steroid negative feedback, the past few years have also seen increasing evidence of the involvement of KNDy cells in a number of reproductive and other physiological functions. These include the role of KNDy cells in seasonal breeding, puberty, the effects of stress on the reproductive axis, the interaction between metabolic cues and reproduction, the influence of gonadal steroids on prolactin secretion and of prolactin on reproduction, and the control of thermoregulation (for a recent review, see Moore et al., 2018). The discovery of the role of KNDy cells and NKB in thermoregulation has led to recent success in the use of NK3R receptor antagonists in randomized, double-blind trials for the treatment of postmenopausal hot flushes (Prague et al., 2017, 2018). The distribution pattern of NKB neurons is similar to those expressing substance P (SP) (Rance and Young III, 1991). This observation raised the possibility that the two tachykinin peptides derived from different genes might be coexpressed in a subset of KP neurons. Indeed, the results of triple-immunofluorescent studies by Hrabovszky (2014) indicate that 25.1% of NKB-IR and 30.6% of KP-IR perikarya contain SP in the infundibulum of postmenopausal women. Furthermore, 16.5% of all immunolabeled cell bodies are triple-labeled (KP/ NKB/SP-IR) in this human model (Hrabovszky et al., 2013). A quantitative analysis of SP-IR cell numbers in the infundibulum of postmenopausal women also revealed significantly more SP-IR neurons in postmenopausal women versus either age-matched or young men (Hrabovszky et al., 2013).

with three GPCRs: GalR1, GalR2, and GalR3 (Hokfelt, 2010). Galanin is implicated in the control of feeding, alcohol intake, seizure threshold, cognitive performance and mood, pain, neurogenesis, and neuroprotection (Hokfelt, 2010; Merchenthaler, 2010). Galanin is one of the most inducible neuropeptides. Its expression is induced by estrogen (Vrontakis et al., 1987), peripheral axotomy (Hokfelt et al., 1987), and seizure activity (Mazarati, 2004). Galanin is also a neuroprotective peptide most likely acting by suppressing the differentiated activity of the neurons and letting them concentrate on reparation (Holmes et al., 2000). Galanin also promotes neurogenesis (Shen et al., 2003). Within the hypothalamus and brainstem, galanin regulates many autonomic functions (for reviews, see Merchenthaler, 2010) and the recently published book by Hokfelt (2010). Most of the Galanin-IR perikarya are fusiform (Fig. 4.15A), but numerous multipolar galanin-IR cells can also be observed in the hypothalamus (Fig. 4.15B). GalaninIR neurons are present in the periventricular zone of the preoptic and tuberal regions, the PVN, and the infundibulum/median eminence (Fig. 4.16), where densely packed galanin-IR cell bodies are closely associated with the portal vessels (Fig. 4.15C). In the preoptic region, galanin neurons characterize the intermediate nucleus (Garcia-Falgueras et al., 2011), also known as sexually dimorphic nucleus, interstitial nucleus of the anterior hypothalamus-1 (INAH-1), or ventrolateral preoptic nucleus (Saper, 2021). Few perikarya can be seen along the diagonal band of Broca and in the lamina terminalis cinerea. In the posterior hypothalamus, galanin-IR neurons are scattered in the periventricular area and in the tuberomamillary nucleus (Fig. 4.16). A similar distribution of galanin-IR neurons in the human hypothalamus has been reported by Gentleman et al. (1989). Galanin-IR fibers form a dense network in the infundibulum/median eminence, often surrounding portal vessels (Fig. 4.16C). An additional subpopulation of axonal varicosities can be detected in the periventricular zone of the chiasmatic and tuberal regions. Few galanin-IR fibers are located along the diagonal band of Broca and around the anterior commissure and fornix, as well as in the lamina terminalis cinerea. In the posterior hypothalamus, a small number of fibers occupy the lateral zones of the hypothalamus. Similarly to many other peptidergic systems, numerous galanin-IR perikarya receive contacting galanin-IR fiber varicosities (Fig. 4.15).

Galanin system

Galanin-like peptide system

Galanin is a 29 amino acid neuropeptide widely expressed in the central and peripheral nervous systems. It has been shown to regulate numerous physiological and pathophysiological processes through interactions

Galanin-like peptide (GALP), a 60 amino acid neuropeptide originally isolated from the porcine hypothalamus, has been shown to bind galanin receptor subtypes 1 and 2 (GalR1, GalR2), although its binding affinity for

78

B. DUDÁS AND I. MERCHENTHALER

Fig. 4.15. Galanin-immunoreactive (IR) elements in the human hypothalamus. Galanin-IR fibers contact with a fusiform galaninIR neuron in the periventricular area of the tuberal region (A) and with a multipolar galanin-IR neuron in the infundibular nucleus (B). Galanin-IR axon varicosities abut on a galanin-IR cell body in a close proximity of a portal vessel (asterisk) in the median eminence (C). The lumen of the vessel is marked by asterisk. Some of the contacting fibers are denoted by arrows. Scale bar: 20 mm. Reprinted with permission from Elsevier. Dudas B, Merchenthaler I (2004). Bidirectional associations between galanin and luteinizing hormone-releasing hormone neuronal systems in the human diencephalon. Neuroscience 127: 695–707.

Fig. 4.16. Stereoscopic images of the human hypothalamus reconstituted from 30-mm-thick sections, illustrating the distribution of galanin-immunoreactive (IR) perikarya (dots). Stereoscopic images can be seen using U or parallel vision as described for Fig. 4.2. Abbreviations: AC, anterior commissure; Inf, infundibulum; MB, mamillary body; and OCh, optic chiasm.

GalR2 is higher than that for GALR1 (Ohtaki et al., 1999). GALP and galanin are coded by distinct genes, which are located on separate chromosomes in humans (chromosomes 19 and 11, respectively) but on the same chromosome in rats (chromosome 1) (Cunningham et al., 2002). Using GalR1 or GalR2 knockout mice, Krasnow et al. (2004) have demonstrated that neither GalR1 nor GalR2 is essential for mediating the effects of GALP on feeding, body weight or LH secretion. In addition, GALP stimulates the release of GnRH in vitro from the GT1–7 immortalized GnRH cell line, although these cells do not express any of the galanin receptor subtypes (Seth et al., 2004). Taken together, these results suggest that GALP signals are transduced via a yet unidentified GALP-specific receptor in vivo.

Subsequently, GALP-IR perikarya were shown to be distributed in the arcuate nucleus and posterior pituitary of rats (Takatsu et al., 2001; Fujiwara et al., 2002), mice (Jureus et al., 2001), and macaques (Cunningham et al., 2002). These immunocytochemical observations have been confirmed with in situ hybridization histochemistry in macaques (Cunningham et al., 2002) and rats (Jureus et al., 2000; Kerr et al., 2000; Larm and Gundlach, 2000). Currently, there is no data on the distribution of GALP in the human hypothalamus. GALP-IR fibers are located in the arcuate and paraventricular nuclei, in the bed nucleus of stria terminalis, medial preoptic area, lateral septal nucleus (Takatsu et al., 2001), and in the lateral hypothalamus near the fornix of rats (Takenoya et al., 2005). GALP-IR fibers appear to project from the arcuate to the PVN, to the

MORPHOLOGY AND DISTRIBUTION OF HYPOTHALAMIC PEPTIDERGIC SYSTEMS medial preoptic area, bed nucleus of stria terminalis, and lateral septal nucleus (Wodowska and Ciosek, 2015).

Neuropeptide Y system Neuropeptide Y (NPY) is a 36 amino acid peptide widely distributed in the human hypothalamus with broad range of cardiovascular, respiratory, immune, and reproductive functions (Tatemoto, 1982; Tatemoto et al., 1982; Adrian et al., 1983; Reichmann and Holzer, 2016; Korf and Moller, 2021). The NPY-IR neurons are morphologically fusiform or multipolar (Fig. 4.17A and B). The density of NPY-IR perikarya increases from lateral to medial direction in the human diencephalon. The majority of the NPY-IR cell bodies can be observed in the septal region, the medial and lateral preoptic and periventricular areas, and the infundibulum, while NPY-IR cell bodies are less

79

numerous in the posterior hypothalamus (Fig. 4.18). While in the medial preoptic area, NPY-IR neurons are located further away from the ventricular ependymal surface, in the tuberal region and posterior hypothalamus, the immunoreactive perikarya are often located periventricularly, close to the ependymal surface (Fig. 4.18). The distribution of the NPY neuronal system in the human diencephalon has also been confirmed by Escobar et al. (2004). Similarly to the density of the perikarya, the density of the NPY-IR fibers is gradually decreasing mediolaterally in the human hypothalamus, with the exception of the lateral zones of the diagonal band of Broca at the septal region, where the fibers delineate a dense bundle. NPYIR fibers form axonal varicosities in the infundibular nucleus, in the periventricular area of the preoptic and tuberal regions, and along the diagonal band of Broca.

Fig. 4.17. NPY-immunoreactive (IR) neural elements in the human hypothalamus. (A) NPY-IR axonal varicosities often abut on fusiform perikarya in the infundibulum, forming multiple contacts (arrows). (B) Multipolar NPY-IR neuron in the suprachiasmatic nucleus. (C) NPY-IR fibers frequently form dense terminal fields around fusiform neurons that are apparently not NPY-IR in the lateral and (D) medial hypothalamic areas. Scale bar: 10 mm.

Fig. 4.18. Stereoscopic images of the human hypothalamus reconstituted from 30-mm-thick sections, illustrating the distribution of neuropeptide Y (NPY)-immunoreactive (IR) perikarya (dots). Stereoscopic images can be seen using U or parallel vision as described for Fig. 4.2. Abbreviations: AC, anterior commissure; Inf, infundibulum; MB, mamillary body; and OCh, optic chiasm.

80

B. DUDÁS AND I. MERCHENTHALER

NPY-IR fibers are also present in the dorsomedial, ventromedial, and the paraventricular nuclei. Similarly to other neuropeptides, NPY-IR axonal varicosities often abut on NPY-IR neurons forming multiple contacts that appear to be synapses (Fig. 4.17A), whereby NPY may modulate the activity of neuronal circuits operating with NPY. NPY-IR fibers frequently form dense terminal fields around fusiform neurons that are apparently not NPY-IR in the lateral and medial hypothalamic areas (Fig. 4.17C and D).

Substance P system Substance P is an 11 amino acid neuropeptide (von Euler and Gaddum, 1931) that functions as a neurotransmitter and as a neuromodulator. It belongs to the tachykinin neuropeptide family. Substance P and its closely related neuropeptide, neurokinin A, are produced from a polyprotein precursor after differential splicing of the preprotachykinin A gene. Substance P binds to neurokinin 1 receptor that belongs to the tachykinin receptor subfamily of GPCRs (Gerard et al., 1991; Maggi, 1995). Although SP is primarily involved in nociception (Zubrzycka and Janecka, 2000), it also has other neuromodulatory functions throughout the brain, including effects in mood, anxiety, stress (Ebner and Singewald, 2006), reinforcement (Huston et al., 1993), neurogenesis (Park et al., 2007), respiratory rhythm (Bonham, 1995), neurotoxicity and nausea/emesis (Hesketh, 2001), and reproduction (Kalil et al., 2016). In addition, SP is an active vasodilator. Morphologically, most of the SP-IR perikarya are fusiform shaped (Fig. 4.19A), but the dorsomedial

subdivision of the ventromedial nucleus also contains numerous multipolar cells (Fig. 4.19B). SP-IR perikarya are confined almost exclusively to the tuberal region (Korf and Moller, 2021), but a few scattered SP-IR neurons can also be observed at the periventricular zone of the preoptic area and in the basal part of the posterior hypothalamus (Fig. 4.20). In the tuberal region, the SP-IR neurons are arranged in several clusters; subgroups of neurons can be found in the infundibular nucleus/median eminence, the dorsomedial subdivision of the ventromedial nucleus, the basal part of the periventricular area, and the basal perifornical area of the tuberal region (Fig. 4.20). In the infundibulum, SP is colocalized with KP, NKB, and Dyn as described in this chapter (Kisspeptin/Neurokinin B/Dynorphin) previously. Similar distribution of the SP neuronal system in the human hypothalamus has been previously described by Chawla et al. (1997). The majority of SP-IR fibers are located in the infundibulum/hypophyseal stalk, where they surround the hypophysial portal vessels (Fig. 4.19D). In the preoptic and tuberal regions, SP-IR fibers are often periventricularly arranged. On the basal region of the diencephalon, SP-IR fibers are also located laterally passing over the optic tract. Few SP-IR axon varicosities can be observed around the anterior commissure, in the PVN, and along the diagonal band of Broca. SP-IR fibers are also detected around the fornix at the preoptic area and tuberal region. As described for most of the hypothalamic peptides, SP-IR perikarya frequently receive contacting SP-IR fiber varicosities (Fig. 4.19A) providing the basis for autofeedback regulation and/or for synchronous activity of certain SP neuronal systems. In the periventricular

Fig. 4.19. Substance P (SP)-immunoreactive (IR) neural elements in the human hypothalamus. (A) Fusiform SP-IR perikaryon receives contacting SP-IR fiber varicosities in the infundibular nucleus (arrows). (B) Multipolar neuron in the dorsomedial subdivision of the ventromedial nucleus. (C) In the periventricular area, SP-IR fibers often form fiber baskets around fusiform cells that are apparently not immunoreactive for SP. (D) Portal vessels in the infundibular region are surrounded by SP-IR fiber varicosities. Magnification: 400x (A–C), 200x (D).

MORPHOLOGY AND DISTRIBUTION OF HYPOTHALAMIC PEPTIDERGIC SYSTEMS

81

Fig. 4.20. Stereoscopic images of the human hypothalamus reconstituted from 30-mm-thick sections, illustrating the distribution of substance P (SP)-immunoreactive (IR) perikarya (dots). Stereoscopic images can be seen using U or parallel vision as described for Fig. 4.2. Abbreviations: AC, anterior commissure; Inf, infundibulum; MB, mamillary body; and OCh, optic chiasm.

area, SP-IR fibers often form fiber baskets around fusiform cells that are apparently not immunoreactive for SP (Fig. 4.19C).

CONCLUDING REMARKS Hypothalamic neuropeptides play a pivotal role in the regulation of several physiological processes including stress response, growth, reproduction, feeding, metabolism, thermoregulation, and many others that are responsible for maintaining the homeostasis of the organism. Most of these peptides occupy the basal area of the infundibulum/median eminence as well as the periventricular and medial hypothalamic regions regulating the release of various hormones from the anterior pituitary via the hypophysial portal circulation, and thus, they control the functioning of the hypothalamohypophysial-endocrine glands axis. In addition, these hypothalamic peptides also act as neurotransmitters/ neuromodulators orchestrating intimate interactions between other peptidergic and nonpeptidergic (e.g., dopamine) neuronal systems within and beyond the hypothalamus. It appears to be a distinctive phenomenon that most of the hypothalamic peptides communicate with each other via axosomatic or axodendritic connections forming an intricate web that controls multiple vital functions.

REFERENCES Adrian TE, Allen JM, Bloom SR et al. (1983). Neuropeptide Y distribution in human brain. Nature 306: 584–586. Aubert ML, Grumbach MM, Kaplan SL (1977). The ontogenesis of human fetal hormones. IV. Somatostatin, luteinizing hormone releasing factor, and thyrotropin releasing factor in hypothalamus and cerebral cortex of human fetuses 10–22 weeks of age. J Clin Endocrinol Metab 44: 1130–1141.

Barry J (1976). Characterization and topography of LH-RH neurons in the human brain. Neurosci Lett 3: 287–291. Bloch B, Gaillard RC, Culler MD et al. (1992). Immunohistochemical detection of proluteinizing hormone-releasing hormone peptides in neurons in the human hypothalamus. J Clin Endocrinol Metab 74: 135–138. Boelen A, von Trotsenburg P, Fliers E (2021). Congenital isolated central hypothyroidism: novel mutations and their functional implications. In: DF Swaab, PJ Lucassen, RM Buijs (Eds.), The human hypothalamus: Middle and posterior hypothalamus. Elsevier. Boler J, Enzmann F, Folkers K et al. (1969). The identity of chemical and hormonal properties of the thyrotropin releasing hormone and pyroglutamyl-histidyl-proline amide. Biochem Biophys Res Commun 37: 705–710. Bonham AC (1995). Neurotransmitters in the CNS control of breathing. Respir Physiol 101: 219–230. Brailoiu GC, Dun SL, Ohsawa M et al. (2005). KiSS-1 expression and metastin-like immunoreactivity in the rat brain. J Comp Neurol 481: 314–329. Bugnon C, Fellmann D, Bloch B (1977). Immunocytochemical study of the ontogenesis of the hypothalamic somatostatin-containing neurons in the human fetus. Cell Tissue Res 183: 319–328. Burgus R, Dunn TF, Desiderio D et al. (1969). Molecular structure of the hypothalamic hypophysiotropic TRF factor of ovine origin: mass spectrometry demonstration of the PCA-His-Pro-NH2 sequence. C R Acad Hebd Seances Acad Sci D 269: 1870–1873. Burke MC, Letts PA, Krajewski SJ et al. (2006). Coexpression of dynorphin and neurokinin B immunoreactivity in the rat hypothalamus: morphologic evidence of interrelated function within the arcuate nucleus. J Comp Neurol 498: 712–726. Cariboni A, Balasubramanian R (2021). Kallman syndrome: semaphorin and other mutations and GnRH migration. In: DF Swaab, PJ Lucassen, RM Buijs (Eds.), The human hypothalamus: Hypothalamus and neuropsychiatric disorders. Elsevier.

82

B. DUDÁS AND I. MERCHENTHALER

Charnay Y, Chayvialle JA, Pradayrol L et al. (1987). Ontogeny of somatostatin-like immunoreactivity in the human fetus and infant spinal cord. Brain Res 433: 63–73. Chawla MK, Gutierrez GM, Young 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. Ciofi P, Leroy D, Tramu G (2006). Sexual dimorphism in the organization of the rat hypothalamic infundibular area. Neuroscience 141: 1731–1745. Cullen JM, Cascella M (2020). Physiology, Enkephalin. StatPearls Publishing LLC. Cunningham MJ, Scarlett JM, Steiner RA (2002). Cloning and distribution of galanin-like peptide mRNA in the hypothalamus and pituitary of the macaque. Endocrinology 143: 755–763. 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. Dedic N, Chen A, Deussing JM (2018). The CRF family of neuropeptides and their receptors—mediators of the central stress response. Curr Mol Pharmacol 11: 4–31. Deltondo J, Por I, Hu W et al. (2008). Associations between the human growth hormone-releasing hormone- and neuropeptide-Y-immunoreactive systems in the human diencephalon: a possible morphological substrate of the impact of stress on growth. Neuroscience 153: 1146–1152. Dudas B, Merchenthaler I (2004). Topography and associations of b-endorphin and luteinizing hormone-releasing hormone neuronal systems in the human diencephalon. Neuroscience 124: 221–229. Dudas B, Merchenthaler I (2006). Three dimensional representation of the neurotransmitter systems of the human hypothalamus: inputs of the luteinizing hormonereleasing hormone neuronal system. J Neuroendocrinol 18: 79–95. Dudas B, Mihaly A, Merchenthaler I (2000). Topography and associations of luteinizing hormone-releasing hormone and neuropeptide Y-immunoreactive neuronal systems in the human diencephalon. J Comp Neurol 427: 593–603. Ebner K, Singewald N (2006). The role of substance P in stress and anxiety responses. Amino Acids 31: 251–272. Escobar CM, Krajewski SJ, Sandoval-Guzman T et al. (2004). Neuropeptide Y gene expression is increased in the hypothalamus of older women. J Clin Endocrinol Metab 89: 2338–2343. Fekete C, Lechan RM (2014). Central regulation of hypothalamic-pituitary-thyroid axis under physiological and pathophysiological conditions. Endocr Rev 35: 159–194. Fliers E, Noppen NW, Wiersinga WM et al. (1994). Distribution of thyrotropin-releasing hormone (TRH)containing cells and fibers in the human hypothalamus. J Comp Neurol 350: 311–323. Fodor M, Kordon C, Epelbaum J (2006). Anatomy of the hypophysiotropic somatostatinergic and growth hormonereleasing hormone system minireview. Neurochem Res 31: 137–143.

Freeman ME, Kanyicska B, Lerant A et al. (2000). Prolactin: structure, function, and regulation of secretion. Physiol Rev 80: 1523–1631. Frohlich E, Wahl R (2019). The forgotten effects of thyrotropin-releasing hormone: metabolic functions and medical applications. Front Neuroendocrinol 52: 29–43. Fujiwara K, Adachi S, Usui K et al. (2002). Immunocytochemical localization of a galanin-like peptide (GALP) in pituicytes of the rat posterior pituitary gland. Neurosci Lett 317: 65–68. Funes S, Hedrick JA, Vassileva G et al. (2003). The KiSS-1 receptor GPR54 is essential for the development of the murine reproductive system. Biochem Biophys Res Commun 312: 1357–1363. Garcia-Falgueras A, Ligtenberg L, Kruijver FP et al. (2011). Galanin neurons in the intermediate nucleus (InM) of the human hypothalamus in relation to sex, age, and gender identity. J Comp Neurol 519: 3061–3084. Gentleman SM, Falkai P, Bogerts B et al. (1989). Distribution of galanin-like immunoreactivity in the human brain. Brain Res 505: 311–315. Gerard NP, Garraway LA, Eddy Jr. RL et al. (1991). Human substance P receptor (NK-1): organization of the gene, chromosome localization, and functional expression of cDNA clones. Biochemistry 30: 10640–10646. Goncharuk VD, Buijs RM, Swaab DF (2007). Corticotropinreleasing hormone neurons in hypertensive patients are activated in the hypothalamus but not in the brainstem. J Comp Neurol 503: 148–168. Gottsch ML, Cunningham MJ, Smith JT et al. (2004). A role for kisspeptins in the regulation of gonadotropin secretion in the mouse. Endocrinology 145: 4073–4077. Grattan DR (2015). 60 Years of neuroendocrinology: the hypothalamo-prolactin axis. J Endocrinol 226: T101–T122. Gu GB, Simerly RB (1997). Projections of the sexually dimorphic anteroventral periventricular nucleus in the female rat. J Comp Neurol 384: 142–164. Hagihara K, Hirata S, Osada T et al. (1992). Distribution of cells containing progesterone receptor mRNA in the female rat di- and telencephalon: an in situ hybridization study. Brain Res Mol Brain Res 14: 239–249. Herzig KH, Louie DS, Owyang C (1994). Somatostatin inhibits CCK release by inhibiting secretion and action of CCK-releasing peptide. Am J Physiol 266: G1156–G1161. Hesketh PJ (2001). Potential role of the NK1 receptor antagonists in chemotherapy-induced nausea and vomiting. Support Care Cancer 9: 350–354. Hokfelt T (2010). Galanin, Springer. Hokfelt T, Johansson O, Ljungdahl A et al. (1980). Peptidergic neurones. Nature 284: 515–521. Hokfelt T, Wiesenfeld-Hallin Z, Villar M et al. (1987). Increase of galanin-like immunoreactivity in rat dorsal root ganglion cells after peripheral axotomy. Neurosci Lett 83: 217–220. Holmes FE, Mahoney S, King VR et al. (2000). Targeted disruption of the galanin gene reduces the number of sensory neurons and their regenerative capacity. Proc Natl Acad Sci U S A 97: 11563–11568.

MORPHOLOGY AND DISTRIBUTION OF HYPOTHALAMIC PEPTIDERGIC SYSTEMS Holsboer F, Ising M (2021). The hypothalamic stress systems in major depression disorder, bipolar disorder and suicide. In: DF Swaab, PJ Lucassen, RM Buijs (Eds.), The human hypothalamus: Hypothalamus and neuropsychiatric disorders. Elsevier. Hrabovszky E (2014). Neuroanatomy of the human hypothalamic kisspeptin system. Neuroendocrinology 99: 33–48. Hrabovszky E (2021). The human hypothalamic kisspeptin system. Functional neuroanatomy and clinical perspectives. In: DF Swaab, PJ Lucassen, RM Buijs (Eds.), The human hypothalamus: Middle and posterior hypothalamus. Elsevier. 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 neurons. Eur J Neurosci 31: 1984–1998. 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. Hsueh AJ, Schaeffer JM (1985). Gonadotropin-releasing hormone as a paracrine hormone and neurotransmitter in extrapituitary sites. J Steroid Biochem 23: 757–764. Hurley DM, Brian RJ, Burger HG (1983). Ovulation induction with subcutaneous pulsatile gonadotropin-releasing hormone: singleton pregnancies in patients with previous multiple pregnancies after gonadotropin therapy. Fertil Steril 40: 575–579. Huston JP, Hasenohrl RU, Boix F et al. (1993). Sequencespecificeffects of neurokinin substance P on memory, reinforcement, and brain dopamine activity. Psychopharmacology (Berl) 112: 147–162. Ishikawa K, Katakami H, Jansson JO et al. (1986). Ontogenesis of growth hormone-releasing hormone neurons in the rat hypothalamus. Neuroendocrinology 43: 537–542. Ishikawa K, Taniguchi Y, Inoue K et al. (1988). Immunocytochemical delineation of thyrotrophic area: origin of thyrotropin-releasing hormone in the median eminence. Neuroendocrinology 47: 384–388. Jureus A, Cunningham MJ, McClain ME et al. (2000). Galanin-like peptide (GALP) is a target for regulation by leptin in the hypothalamus of the rat. Endocrinology 141: 2703–2706. Jureus A, Cunningham MJ, Li D et al. (2001). Distribution and regulation of galanin-like peptide (GALP) in the hypothalamus of the mouse. Endocrinology 142: 5140–5144. Kadar A, Sanchez E, Wittmann G et al. (2010). Distribution of hypophysiotropic thyrotropin-releasing hormone (TRH)synthesizing neurons in the hypothalamic paraventricular nucleus of the mouse. J Comp Neurol 518: 3948–3961. Kalil B, Ramaswamy S, Plant TM (2016). The distribution of substance P and kisspeptin in the mediobasal hypothalamus of the male rhesus monkey and a comparison of intravenous

83

administration of these peptides to release GnRH as reflected by LH secretion. Neuroendocrinology 103: 711–723. Kawano H, Tsuruo Y, Bando H et al. (1991). Hypophysiotrophic TRH-producing neurons identified by combining immunohistochemistry for pro-TRH and retrograde tracing. J Comp Neurol 307: 531–538. Kedzia A, Obara-Moszynska M, Chmielnicka-Kopaczyk M (2009). Assessment of ghrelin, GHS-R, GH, and neurohormones in human fetal pituitary glands and central nervous system: an immunohistochemical study. Folia Histochem Cytobiol 47: 505–510. Kerr NC, Holmes FE, Wynick D (2000). Galanin-like peptide (GALP) is expressed in rat hypothalamus and pituitary, but not in DRG. Neuroreport 11: 3909–3913. King JC, Anthony EL (1984). LHRH neurons and their projections in humans and other mammals: species comparisons. Peptides 5: 195–207. King JC, Anthony EL, Fitzgerald DM et al. (1985). Luteinizing hormone-releasing hormone neurons in human preoptic/ hypothalamus: differential intraneuronal localization of immunoreactive forms. J Clin Endocrinol Metab 60: 88–97. Kordon C, Drouva SV, Martinez de la Escalera G et al. (1994). Role of classic and peptide neuromediators in the neuroendocrine regulation of luteinizing hormone and prolactin. In: E Knobil, JD Neill (Eds.), The physiology of reproduction. Raven Press, New York, pp. 1621–1681. Korf HW, Moller M (2021). The median eminence and infundibular nucleus; window for endocrine and metabolic communication. In: DF Swaab, PJ Lucassen, RM Buijs (Eds.), The human hypothalamus: Middle and posterior hypothalamus. Elsevier. Kotani M, Detheux M, Vandenbogaerde A et al. (2001). The metastasis suppressor gene KiSS-1 encodes kisspeptins, the natural ligands of the orphan G protein-coupled receptor GPR54. J Biol Chem 276: 34631–34636. Krasnow SM, Hohmann JG, Gragerov A et al. (2004). Analysis of the contribution of galanin receptors 1 and 2 to the central actions of galanin-like peptide. Neuroendocrinology 79: 268–277. Kuljis RO, Advis JP (1989). Immunocytochemical and physiological evidence of a synapse between dopamine- and luteinizing hormone releasing hormone-containing neurons in the ewe median eminence. Endocrinology 124: 1579–1581. 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. Larm JA, Gundlach AL (2000). Galanin-like peptide (GALP) mRNA expression is restricted to arcuate nucleus of hypothalamus in adult male rat brain. Neuroendocrinology 72: 67–71. Lechan RM, Fekete C (2006). The TRH neuron: a hypothalamic integrator of energy metabolism. Prog Brain Res 153: 209–235. Lechan RM, Segerson TP (1989). Pro-TRH gene expression and precursor peptides in rat brain. Observations by

84

B. DUDÁS AND I. MERCHENTHALER

hybridization analysis and immunocytochemistry. Ann N Y Acad Sci 553: 29–59. Lechan RM, Hollenberg A, Fekete C (2009). Hypothalamicpituitary-thyroid axis: organization, neural/endocrine control of TRH. In: LR Squire (Ed.), Encyclopedia of neuroscience. Elsevier 75–87. Lee JH, Miele ME, Hicks DJ et al. (1996). KiSS-1, a novel human malignant melanoma metastasis-suppressor gene. J Natl Cancer Inst 88: 1731–1737. Lee DK, Nguyen T, O’Neill GP et al. (1999). Discovery of a receptor related to the galanin receptors. FEBS Lett 446: 103–107. Lehman MN, Merkley CM, Coolen LM et al. (2010). Anatomy of the kisspeptin neural network in mammals. Brain Res 1364: 90–102. Li CH, Chung D, Doneen BA (1976). Isolation, characterization and opiate activity of beta-endorphin from human pituitary glands. Biochem Biophys Res Commun 72: 1542–1547. Liguz-Lecznar M, Urban-Ciecko J, Kossut M (2016). Somatostatin and somatostatin-containing neurons in shaping neuronal activity and plasticity. Front Neural Circuits 10: 48. Maggi CA (1995). The mammalian tachykinin receptors. Gen Pharmacol 26: 911–944. Mazarati AM (2004). Galanin and galanin receptors in epilepsy. Neuropeptides 38: 331–343. Merchenthaler I (1990). Retrograde labeling of hypophysiotropic neurons by local injection of wheat germ agglutinin (WGA) into the median eminence or peripheral administration of fluoro-gold. Mol Cell Neurosci 1: 93–106. Merchenthaler I (1992). Enkephalin-immunoreactive neurons in the parvicellular subdivisions of the paraventricular nucleus project to the external zone of the median eminence. J Comp Neurol 326: 112–120. Merchenthaler I (2010). Galanin and the neuroendocrine axes. In: T Hokfelt (Ed.), Galanin. Springer. Merchenthaler I, Liposits Z (1994). Mapping of thyrotropinreleasing hormone (TRH) neuronal systems of rat forebrain projecting to the median eminence and the OVLT. Immunocytochemistry combined with retrograde labeling at the light and electron microscopic levels. Acta Biol Hung 45: 361–374. Merchenthaler I, Kovacs G, Lavasz G et al. (1980). The preoptico-infundibular LH-RH tract of the rat. Brain Res 198: 63–74. Merchenthaler I, Thomas CR, Arimura A (1984). Immunocytochemical localization of growth hormone releasing factor (GHRF)-containing structures in the rat brain using anti-rat GHRF serum. Peptides 5: 1071–1075. Merchenthaler I, Setalo G, Csontos C et al. (1989). Combined retrograde tracing and immunocytochemical identification of luteinizing hormone-releasing hormone- and somatostatin-containing neurons projecting to the median eminence of the rat. Endocrinology 125: 2812–2821. Mihaly E, Fekete C, Legradi G et al. (2001). Hypothalamic dorsomedial nucleus neurons innervate thyrotropinreleasing hormone-synthesizing neurons in the paraventricular nucleus. Brain Res 891: 20–31.

Miller DS, Reid RR, Cetel NS et al. (1983). Pulsatile administration of low-dose gonadotropin-releasing hormone. Ovulation and pregnancy in women with hypothalamic amenorrhea. JAMA 250: 2937–2941. Moller M (2021). Vasopression and oxytocin beyond the pituitary in the human brain. In: DF Swaab, PJ Lucassen, RM Buijs (Eds.), The human hypothalamus: Middle and posterior hypothalamus. Elsevier. 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. Muller EE (1987). Neural control of somatotropic function. Physiol Rev 67: 962–1053. Navarro VM, Gottsch ML, Wu M et al. (2011). Regulation of NKB pathways and their roles in the control of Kiss1 neurons in the arcuate nucleus of the male mouse. Endocrinology 152: 4265–4275. Noda M, Teranishi Y, Takahashi H et al. (1982). Isolation and structural organization of the human preproenkephalin gene. Nature 297: 431–434. Obal Jr. F, Krueger JM (2001). The somatotropic axis and sleep. Rev Neurol (Paris) 157: S12–S15. Obal Jr. F, Fang J, Taishi P et al. (2001). Deficiency of growth hormone-releasing hormone signaling is associated with sleep alterations in the dwarf rat. J Neurosci 21: 2912–2918. Ohtaki T, Kumano S, Ishibashi Y et al. (1999). Isolation and cDNA cloning of a novel galanin-like peptide (GALP) from porcine hypothalamus. J Biol Chem 274: 37041–37045. Ohtaki T, Shintani Y, Honda S et al. (2001). Metastasis suppressor gene KiSS-1 encodes peptide ligand of a G-protein-coupled receptor. Nature 411: 613–617. Park SW, Yan YP, Satriotomo I et al. (2007). Substance P is a promoter of adult neural progenitor cell proliferation under normal and ischemic conditions. J Neurosurg 107: 593–599. Prague JK, Roberts RE, Comninos AN et al. (2017). Neurokinin 3 receptor antagonism as a novel treatment for menopausal hot flushes: a phase 2, randomised, double-blind, placebocontrolled trial. Lancet 389: 1809–1820. Prague JK, Roberts RE, Comninos AN et al. (2018). Neurokinin 3 receptor antagonism rapidly improves vasomotor symptoms with sustained duration of action. Menopause 25: 862–869. 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. Ramaswamy S, Seminara SB, Ali B et al. (2010). Neurokinin B stimulates GnRH release in the male monkey (Macaca mulatta) and is colocalized with kisspeptin in the arcuate nucleus. Endocrinology 151: 4494–4503.

MORPHOLOGY AND DISTRIBUTION OF HYPOTHALAMIC PEPTIDERGIC SYSTEMS Rance NE, Young III WS (1991). Hypertrophy and increased gene expression of neurons containing neurokinin-B and substance-P messenger ribonucleic acids in the hypothalami of postmenopausal women. Endocrinology 128: 2239–2247. Rance NE, Young III WS, McMullen NT (1994). Topography of neurons expressing luteinizing hormone-releasing hormone gene transcripts in the human hypothalamus and basal forebrain. J Comp Neurol 339: 573–586. Reichmann F, Holzer P (2016). Neuropeptide Y: a stressful review. Neuropeptides 55: 99–109. Russell JA (2018). Fifty years of advances in neuroendocrinology. Brain Neurosci Adv 2: 2398212818812014. Saper CB (2021). INAH1 (intermediate nucleus): cytoarchitecture, chemoarchitecture: relation to sleep and Alzheimer. In: DF Swaab, PJ Lucassen, RM Buijs (Eds.), The human hypothalamus: Anterior hypothalamus. Elsevier. Seminara SB, Messager S, Chatzidaki EE et al. (2003). The GPR54 gene as a regulator of puberty. N Engl J Med 349: 1614–1627. Seth A, Stanley S, Jethwa P et al. (2004). Galanin-like peptide stimulates the release of gonadotropin-releasing hormone in vitro and may mediate the effects of leptin on the hypothalamo-pituitary-gonadal axis. Endocrinology 145: 743–750. Shen PJ, Larm JA, Gundlach AL (2003). Expression and plasticity of galanin systems in cortical neurons, oligodendrocyte progenitors and proliferative zones in normal brain and after spreading depression. Eur J Neurosci 18: 1362–1376. Shughrue PJ, Lane MV, Merchenthaler I (1997). Comparative distribution of estrogen receptor-alpha and -beta mRNA in the rat central nervous system. J Comp Neurol 388: 507–525. Silverman AJ, Livne I, Witkin JW (1994). The gonadotropinreleasing hormone (GnRH) neuronal systems: immunocytochemistry and in situ hybridization. In: E Knobil, JD Neill (Eds.), The physiology of reproduction. Raven Press, New York, pp. 1683–1709. Simantov R, Snyder SH (1976). Morphine-like peptides in mammalian brain: isolation, structure elucidation, and interactions with the opiate receptor. Proc Natl Acad Sci U S A 73: 2515–2519. Simerly RB, Chang C, Muramatsu M et al. (1990). Distribution of androgen and estrogen receptor mRNA-containing cells in the rat brain: an in situ hybridization study. J Comp Neurol 294: 76–95. Smith JT, Clarke IJ (2007). Kisspeptin expression in the brain: catalyst for the initiation of puberty. Rev Endocr Metab Disord 8: 1–9. 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.

85

Sukhov RR, Walker LC, Rance NE et al. (1995). Opioid precursor gene expression in the human hypothalamus. J Comp Neurol 353: 604–622. Sullivan MW, Stewart-Akers A, Krasnow JS et al. (1999). Ovarian responses in women to recombinant folliclestimulating hormone and luteinizing hormone (LH): a role for LH in the final stages of follicular maturation. J Clin Endocrinol Metab 84: 228–232. Swanson LW, Sawchenko PE (1980). Paraventricular nucleus: a site for the integration of neuroendocrine and autonomic mechanisms. Neuroendocrinology 31: 410–417. Takatsu Y, Matsumoto H, Ohtaki T et al. (2001). Distribution of galanin-like peptide in the rat brain. Endocrinology 142: 1626–1634. Takenoya F, Hirayama M, Kageyama H et al. (2005). Neuronal interactions between galanin-like-peptide- and orexinor melanin-concentrating hormone-containing neurons. Regul Pept 126: 79–83. Tatemoto K (1982). Neuropeptide Y: complete amino acid sequence of the brain peptide. Proc Natl Acad Sci U S A 79: 5485–5489. Tatemoto K, Carlquist M, Mutt V (1982). Neuropeptide Y—a novel brain peptide with structural similarities to peptide YY and pancreatic polypeptide. Nature 296: 659–660. Thorner MO, Vance ML, Hartman ML et al. (1990). Physiological role of somatostatin on growth hormone regulation in humans. Metabolism 39: 40–42. Timmers HJ, Swaab DF, van de Nes JA et al. (1996). Somatostatin 1–12 immunoreactivity is decreased in the hypothalamic lateral tuberal nucleus of Huntington’s disease patients. Brain Res 728: 141–148. Vale W, Spiess J, Rivier C et al. (1981). Characterization of a 41-residue ovine hypothalamic peptide that stimulates secretion of corticotropin and beta-endorphin. Science 213: 1394–1397. Valentino RJ (1989). Corticotropin-releasing factor: putative neurotransmitter in the noradrenergic nucleus locus ceruleus. Psychopharmacol Bull 25: 306–311. Veldhuis JD (2008). Pulsatile hormone secretion: mechanism, significance and evolution. In: D Lloyd, E Rossi (Eds.), Ultradian rhythms from molecule to mind. Springer p. 228. von Euler US, Gaddum JH (1931). An unidentified depressor substance in certain tissue extracts. J Physiol 72: 74–87. Vrontakis ME, Peden LM, Duckworth ML et al. (1987). Isolation and characterization of a complementary DNA (galanin) clone from estrogen-induced pituitary tumor messenger RNA. J Biol Chem 262: 16755–16758. Wodowska J, Ciosek J (2015). Galanin-like peptide (GALP): localization, receptors and biological function. Postepy Hig Med Dosw (Online) 69: 1067–1076. Yang JJ, Caligioni CS, Chan YM et al. (2012). Uncovering novel reproductive defects in neurokinin B receptor null mice: closing the gap between mice and men. Endocrinology 153: 1498–1508. Zubrzycka M, Janecka A (2000). Substance P: transmitter of nociception (Minireview). Endocr Regul 34: 195–201.

This page intentionally left blank

Handbook of Clinical Neurology, Vol. 179 (3rd series) The Human Hypothalamus: Anterior Region D.F. Swaab, F. Kreier, P.J. Lucassen, A. Salehi, and R.M. Buijs, Editors https://doi.org/10.1016/B978-0-12-819975-6.00003-0 Copyright © 2021 Elsevier B.V. All rights reserved

Chapter 5

MRI maps, segregation, and white matter connectivity of the human hypothalamus in health JEAN-JACQUES LEMAIRE1* AND ANTONIO DE SALLES2,3 1

Institut Pascal, Clermont-Ferrand, and Service de Neurochirurgie, Centre Hospitalier et Universitaire, Clermont-Ferrand, France 2

Departments of Neurosurgery and Radiation Oncology, University of California, Los Angeles, CA, United States 3

Department of Neurosurgery and Radiation Oncology, HCor Neuroscience, São Paulo, Brazil

Abstract The human hypothalamus is composed of several gray matter territories, forming 10 different structures mainly referred to as nuclei: the preoptic, suprachiasmatic, supraoptic, infundibular, paraventricular, dorsomedial, ventromedial, posterior (dorsal; dorsal hypothalamic area), and tuberomamillary nuclei, and the lateral hypothalamic area. The macroconnectivity, described since the middle of the 19th century, is currently probed using MRI methods, notably those relying on diffusion techniques. The structural connections can be grouped as follows: connections with the olfactory system; stria terminalis connections; stria medullaris connections; ansa lenticularis connections; subthalamus connections; optic tract connections; intrahypothalamic connections; hypothalamo-hypophysis connections; hypothalamic commissures; cortex connections.

INTRODUCTION The human hypothalamus is coarsely visible on clinical 1.5-Tesla and 3-Tesla magnetic resonance imaging (MRI) machines using currently available sequences (Lemaire et al., 2011a,b; Baroncini et al., 2012; Lemaire et al., 2013). Nevertheless, it is not yet possible to clearly identify most of the hypothalamic nuclei directly, merely because of the absence of capsule-like white matter around most of them. Moreover, the complex architecture of this symmetrical diencephalic structure (Federative International Programme for Anatomical Terminology, 2017) hampers its detailed mastering. Some French names are provided giving insights into, often left in oblivion, pertinent data. More recently, the availability of diffusion tensor imaging (DTI) sequences has enabled a large-scale analysis of the main white matter fascicles supporting the macroconnectivity of the hypothalamus. Tracing of these fascicles using DTI fiber tracking

(DTI-FT) techniques provided the foundation for the hypothalamus parcellation based on its afferent and efferent tracts (Lemaire et al., 2011a,b).

MRI MAPPING AND FIBER TRACKING OF THE HYPOTHALAMUS The wealth of histologic images of the hypothalamus contrasts sharply with the limited work on stereotactic segmentation of the hypothalamus. Stereotactic segmentation is important for determining the spatial location and organization of hypothalamic sites in relation to cortical and subcortical territories. This enables the identification and therapeutic targeting of these elements using exquisite medical images, currently provided mainly by MRI. This is accomplished by direct visualizing or indirectly through proportional grid systems plotted on these images, as previously proposed when structures were not promptly visualized as they are today by MRI.

*Correspondence to: J.J. Lemaire, Service de Neurochirurgie, CHU de Clermont-Ferrand, H^ opital Gabriel-Montpied, 58 rue Montalembert, Clermont-Ferrand 63000, France. Tel: +33-4-73-752-163 (off ), Fax: + 33-4-73-752-166, E-mail: [email protected]

88

J.-J. LEMAIRE AND A. DE SALLES

Historically, Talairach et al. (1957) proposed a segmentation using brain landmarks, a method known worldwide and still employed: the anterior (AC) and posterior (PC) white matter commissures of the diencephalon, used when the best estimations of structure locations were based on the shadow of these structures visualized with ventriculography. From these anatomic landmarks, Talairach et al. used a proportional grid system allowing the determination of a probabilistic location of, among others, hypothalamic nuclei. We must bear in mind that only ventriculographic imaging, frontal and lateral radiographic views, were available during this pioneering work. Schaltenbrand and Bailey (1959) further proposed a similar approach coupled with stereotaxic histological images. More recently, we proposed to map the human hypothalamus coupling stereotactic segmentation and structural connectivity inferred from DTI-FT (Lemaire et al., 2011a,b). Nonstereotactic segmentations of the human hypothalamus have also been explored with MRI, using rough simplified systematizations of nuclei coregistered with resting-state functional MRI (Kullmann et al., 2014) or structural 1.5-Tesla MRI (Makris et al., 2013).

Fontaine et al., 2010; Lemaire et al., 2011a,b); the latter is also called the nucleus pallido-hypothalamicus (Wahren, 1959). The hypothalamus is made up of several gray matter territories, forming 10 different structures commonly referred to as nuclei: the preoptic, suprachiasmatic, supraoptic, infundibular, paraventricular, dorsomedial, ventromedial, posterior (dorsal; dorsal hypothalamic area), and tuberomamillary nuclei, and the lateral hypothalamic area. The lateral hypothalamic area shares a territory with the tuberomamillary nucleus, the lateral intermediate hypothalamic area (see Lemaire et al., 2019). The hypothalamus is crossed by axonal fibers, some organized into white matter fascicles and tracts, assuming that a fascicle refers to a microscopically limited group of bundle(s) of fibers, whereas a tract refers to fibers carrying out a specific function (Riley, 1953). Structural maps of the hypothalamus and of its close neighborhood, built from high-resolution coronal MRI slices (Lemaire et al., 2019), enable a detailing of this region of the deep brain (Fig. 5.2).

MACROCONNECTIVITY 3D AND 2D HYPOTHALAMIC ARCHITECTURE The human hypothalamus has a complex 3D shape (Fig. 5.1). It is placed anteriorly (rostrally) and inferiorly to the thalamus or dorsal thalamus (or couche optique (Dejerine, 1901)), and to the subthalamus or ventral thalamus (or region sous-optique (Dejerine, 1901)) or prethalamus (Federative International Programme for Anatomical Terminology, 2017). The hypothalamus forms a portion of the walls of the third ventricle of the diencephalon. The tuber cinereum, which forms part of the floor of the third ventricle, is located between the mamillary bodies and the infundibulum (Mettler, 1948; Laget, 1973; Duvernoy et al., 1992). The hypothalamus is nearly contained in a volume limited anteriorly (rostrally) by the frontal plane going through the lamina terminalis of the third ventricle, posteriorly (caudally) by the posterior limit of the mamillary body, superiorly by a horizontal plane going through the AC, and laterally by a vertical plane aligned with the optic tract (Lemaire et al., 2011a,b). The lamina supraoptic (gray matter) of the optic chiasma is in continuity with the lamina terminalis (Riley, 1953). The posterior limit of the hypothalamus merges with the ventral tegmental area, which includes the area densa and the nucleus ento(endo) peduncular or nucleus of the ansa lenticularis (see

The macroconnectivity of the hypothalamus relies on known white matter fascicles and fibers (e.g., Fig. 5.3). They were listed by Wahren (1959) and Riley (1953) following pioneering work in laboratory animals and human brain samples, using dissections and tracing techniques such as horseradish peroxidase. The nerve fibers (myelinated or not) are linked with different regions: the olfactory system or rhinencephalon comprising the olfactory bulb and the olfactory tubercle (tuberculum olfactorium, region of anterior perforated substance, substantia perforata anterior), frontal and temporal cortices, and the olfactory amygdala (Nieuwenhuys et al., 2008; Federative International Programme for Anatomical Terminology, 2017); neocortices; optic system; basal ganglia; hypophysis; and intrahypothalamic connections between hypothalamic nuclei. The olfactory cortices are made up of the following (Dejerine, 1901; Duvernoy et al., 1992; Nieuwenhuys et al., 2008; Federative International Programme for Anatomical Terminology, 2017): on the frontal side, the subcallosal gyrus (carrefour olfactif de Broca) and the olfactory tubercle; on the temporal side, the piriform lobe (cortex) and the entorhinal cortex, thus the uncus (circonvolution du crochet) and the hippocampal gyrus (circonvolution de l’hippocampe), which correspond to the anterior part of the parahippocampal gyrus (or T5, or 2ème circonvolution limbique).

MRI MAPS, SEGREGATION, AND WHITE MATTER CONNECTIVITY

89

Fig. 5.1. 3D views of the hypothalamus: (A) projection of the hypothalamus on the diencephalon; (B) medial, (C) lateral, (D) anterior, (E) posterior, (F) inferior, and (G) superior views; dm, dorsomedial nucleus; fo, fornix; if, infundibular nucleus; ipf, interpeduncular fossa; itha, interthalamic adhesion; la, lateral hypothalamic area; lia, lateral intermediate hypothalamic area; lt, lamina terminalis; mb, mamillary body; mes, mesencephalon; ot, optic tract; pv, paraventricular nucleus; p, posterior (dorsal) nucleus; po, preoptic nucleus; soc, suprachiasmatic nucleus and supraoptic nucleus; so-t, supraoptic tract; tc, tuber cinereum; tm, tuberomamillary nucleus; vm, ventromedial nucleus; V3, third ventricle.

The hypothalamus is connected with the olfactory system by two main fascicles, the medial or median forebrain bundle and the fornix, and two small fascicles, one to the mamillary body, the rhinencephalo-mamillaris fascicle (fasciculus ad corpus mamilare), and one to the

tuber cinereum and infundibulum, the infundibular or tuber cinereum fascicle (fasciculus ad infundibulum or fasciculus ad tuber cinereum). The medial forebrain bundle (fasciculus basalis olfactorius, fasciculus telencephalicus medialis, fasci-

Fig. 5.2. Coronal maps going through the hypothalamus, from anterior to posterior: (A) locations of slices going through po (1), pv and if (2), la (3; dm and vm are hidden), dm and vm (4), p (5) and mb (6); (B) coronal maps; AC, the anterior white commissure; ALEN, the ansa lenticularis; ALEN-nu, the nucleus of ansa lenticularis; AL-nu-TH, the anterolateral nucleus of thalamus; ALEN-TIPP-fa, the ansa lenticularis, the tip-pallidal fascicle of Talairach; AMD-nu-TH, the anteromedial dorsal nucleus of thalamus; AMV-nu-TH, the anteromedial ventral nucleus of thalamus; DL-nu-TH, the dorsolateral nucleus of thalamus; DMnu-HYPOT, the dorsomedial nucleus; E-Lam-TH, the external lamina of thalamus; FO, the fornix; GPi, the internal, medial, globus pallidus; I-HYPOT-a, the lateral intermediate hypothalamic area; INF-nu-HYPOT, the infundibular nucleus; INNO-sub, the innominate substance; I-TH-ped, the inferior thalamic peduncle; LAO-nu-TH, the laminar oral nucleus of thalamus; LEN-fa, the lenticular fascicle; L-HYPOT-a, the lateral hypothalamic area; MB, the mamillary body; M-nu-TH, the medial nucleus of thalamus; MT-fa, the mamillothalamic fascicle; OT, the optic tract; PO-nu-HYPOT, the preoptic nucleus; POST-nu-HYPOT, the posterior (dorsal; dorsal hypothalamic area) nucleus; PTH-ret-z, the prethalamic reticularoid zone; PV-nu-HYPOT, the paraventricular nucleus; SFL-TH, the superficial lateral thalamus; SO-nu-HYPOT, the suprachiasmatic nucleus and the supraoptic nucleus; ST-nu, the subthalamic nucleus; STR-MED-TH, the stria medullaris of thalamus; STR-ter, the stria terminalis; SN, the substantia nigra; SUOT, the supraoptic tract; TH-fa, the thalamic fascicle; TM-nu-HYPOT, the tuberomamillary nucleus; VM-nu-HYPOT, the ventromedial nucleus; ZI, the zona incerta.

MRI MAPS, SEGREGATION, AND WHITE MATTER CONNECTIVITY

91

Fig. 5.3. Color-coded diffusion maps of the hypothalamus (color code of DTI: red-transversal, green-sagittal, and blue-vertical directions; 60 directions) merged with inversion-recovery MRI maps (3T-MRI; voxels resampled ¼ 0.5 mm3). (A) Sagittal (parallel to ACPC; laterality, 1/12th of ACPC length) map going through the point of penetration of the fornix (fo; inlay 3D view of the fornix projected on the section) in the hypothalamus and the tuberomamillary nucleus (tm); ac, anterior commissure; cc, corpus callosum; tha, thalamus. (B) Axial (parallel to ACPC; 1/12th of ACPC length, below ACPC) map going through the superior colliculus (sc); ac, anterior commissure; al, ansa lenticularis; fo, fornix; GPi, globus pallidum intern; ic, internal capsule; mtf, mamillothalamic fascicle; stf, subthalamic tegmental field.

culus longitudinalis basalis, fasciculus prosencephali medialis), also called the deep olfactive radiations (radiations olfactives profondes (Dejerine, 1901)), a loose fascicle connecting olfactory (notably the basal nucleus of Ganser or olfactory tubercle, tuberculum olfactorium), septal, striatal (fundus), sublenticular, and tegmental structures, passes through the internal capsule and the hypothalamus, where it connects mainly with lateral and tuberal nuclei (Mettler, 1948; Riley, 1953; Wahren, 1959; Nauta and Haymaker, 1969; Laget, 1973; Palkovits and Zaborszky, 1979; Nieuwenhuys et al., 2008; Lemaire et al., 2011a,b; Van Hartevelt and Kringelbach, 2012; Federative International Programme for Anatomical Terminology, 2017). The fornix (fasciculus olfactorius fornicis, tractus corticomamillaris), also called cerebral trigome (Dejerine, 1901), a dense and complex fascicle originating in the hippocampal region, crosses the hypothalamus accompanied by perifornical hypothalamic nuclei (Wahren, 1959), receiving fibers from anterior and intermediate hypothalamic regions. It goes to the mamillary body and the tuberomamillary nucleus, and a portion may continue in the fasciculus olfactorius (hippocampi), or diagonal band of Broca. It also connects subthalamic tegmental structures, the amygdala, limbic structures such as septal nuclei, the cingulate, and the innominate substance with the basal nucleus of Meynert (Dejerine, 1901; Riley, 1953). The principal mamillary tract (fasciculus mamillaris princeps) divides into the mamillotegmental fascicle (or tract, fasciculus mamillaris ad tegmentum, fasciculus mamillo-tegmentalis, fasciculus tegmento-mamillaris, fasciculus pedunculo-mamillaris; likely also fibrae hypothalamo-tegmentales) and the mamillothalamic

fascicle (or bundle, tract, or fascicle of Vicq d’Azyr; mamillo-thalamicus, thalamo-mamillaris). The mamillotegmental fascicle ends in the tegmentum and connects with the dorsal longitudinal fascicle; the mamillothalamic fascicle ends in the anterior thalamus, and relays to the frontal, cingulate, and hippocampal cortices (Riley, 1953; Wahren, 1959). The mamillopeduncular fascicle (fasciculus mamillo-peduncularis, pedunculus corpus mamillaris) connects the mamillary body to the brainstem, from the mesencephalon and the tegmentum, such as the substantia nigra and the red nucleus to the bulb (Riley, 1953; Wahren, 1959). The stria terminalis or semicircularis (or cornea) contains fibers from the preoptic area and the hypothalamus, and connects the amygdala (Riley, 1953; Nauta and Haymaker, 1969; Palkovits and Zaborszky, 1979); it also continues with the diagonal band of Broca, the medial forebrain bundle, and the stria medullaris (Wahren, 1959). The stria medullaris (stria habenularis, taenia thalami) connects the hypothalamus with the habenula, and the habenula with septal and olfactory structures (Riley, 1953; Wahren, 1959; Palkovits and Zaborszky, 1979). The ansa lenticularis connects the amygdalo-piriform complex to the lateral hypothalamus; it forms the ansa peduncularis, along with the inferior thalamic peduncle, which connects with the temporal and frontoorbital lobes (Riley, 1953; Nauta and Haymaker, 1969; Nieuwenhuys et al., 2008). Together, the medial forebrain bundle and the ansa lenticularis form the basal telencephalic bundle (Riley, 1953). The ansa lenticularis can also be identified using DTI-FT (Lemaire et al., 2011a,b). Another fascicle, the fasciculus pallido-hypothalamicus (tractus pallido-hypothalamicus,

92

J.-J. LEMAIRE AND A. DE SALLES

tractus hypothalamo-tegmentalis), could connect the pallidum to the ventromedial nucleus of the hypothalamus via the subthalamic nucleus (Riley, 1953). The fasciculus tuberis cinerei connects the sublenticular region with the tuber cinereum ipsi and contra (via the dorsal supraoptic commissure) (Riley, 1953). Subthalamic structures connect with the hypothalamus using the thalamic fascicle (fasciculus thalamicus hypothalamic, area tegmentalis pars dorsali, Forel field H1, fasciculus thalami; also likely fibrae subthalamohypothalamicae) and the lenticular fascicle (fasciculus lenticularis hypothalamic, area tegmenti pars ventralis, Forel field H2, fasciculus pedunculi) (Riley, 1953). Thalamic nuclei, such as the median, connect with the hypothalamus via thalamohypothalamic fibers (fibrae thalamo-hypothalamicae) (Riley, 1953). The optic tract (fasciculus opticus, tractus opticus) connects with the hypothalamus through fascicles (Riley, 1953; Wahren, 1959): the fasciculus retinosupraopticus (optico-thalamicus, retino-thalamicus, dorsal hypothalamic root of the chiasma opticus) with the supraoptic and suprachiasmatic nuclei, ipsi- and contralaterally, and the geniculate body, ipsi- and contralaterally via the supraoptic commissures; and the fasciculus optico-tuberis with the tuber cinereum. Intrahypothalamic and hypothalamo-hypophysis fascicles are also described (Riley, 1953): the fasciculus inter-paraventriculo-tangentialis, between the paraventricular and the supraoptic nuclei and also with the hypophysis; the fasciculus hypothalamicus lateroperiventriculus, between the lateral area and the paraventricular nucleus; the tractus hypothalamo-mamillaris, between the mamillary body and the tuber cinereum; the tractus supraoptico-hypophyseus (fasciculus tuber fornicis, tractus hypothalamico-hypophyseus, fasciculus hypophyseus) from the paraventricular (tractus paraventriculo-hypophyseus) and supraoptic nuclei to the neurohypophysis; the tractus paraventricularis cinereus (tractus hypothalamo-filiformis), between the paraventricular nucleus, the posterior hypothalamus, and the tuber cinereum; and fibers go to the hypophysis (tractus tuberohypophyseus) and the brainstem. The dorsal longitudinal fascicle (of Sch€ utz; fasciculus longitudinalis dorsalis), emblematic of the periventricular fiber system embedded within the subependymal region, connects hypothalamic (notably via the tractus paraventricularis cinereus), thalamic, and tectal structures (Wahren, 1959). All hypothalamic nuclei are known to be interconnected (Wahren, 1959). The hypothalamic commissures are located within the tuber cinereum, the supraoptic, and the supramamillary regions. The dorsal supraoptic commissure (of Ganser, fibrae ansulatae, decussation hypothalamic[a] anterior; combining two historical pars, dorsalis of Ganser and ventralis of Gudden) primarily connects the pallidum,

via the ansa lenticularis, with medial subventricular nuclei, such as the ventromedial nucleus of hypothalamus (Riley, 1953; Wahren, 1959). The ventral supraoptic commissure (of Meynert) connects the basal nucleus of Meynert, the pallidum, the subthalamic nucleus, the tuber cinereum, and tecto-mesencephalic structures (Riley, 1953; Wahren, 1959). The supramamillary commissure (commissura supramamillaris, decussation hypothalamica posterior, subthalamic posterior, retroinfundibularis), also called the commissure of Forel (Dejerine, 1901), connects contra subthalamic structures (subthalamic nucleus, pallidum, red nucleus, tegmental area) and the lateral geniculate body, and the mamillary body with the habenula via the retroflexus fascicle (habenulo-interpeduncular fascicle, fascicle of Meynert) (Dejerine, 1901; Riley, 1953). The cortical connections are numerous and direct with frontal, temporal, and limbic (cingulate) cortices and indirect to the cortex via the thalamus (Wahren, 1959). These findings were recently confirmed using DTI-FT and from numerous data observed using invasive techniques, showing the wide cortical connectivity of the hypothalamus (Lemaire et al., 2011a,b) with some features: the preoptic, the anteroventral, the lateral, and the posterior compartments are the most connected with the cortex; the ventromedial compartment, i.e., the ventromedial nucleus and the adjacent tuberomamillary nucleus, connects predominantly with the prefrontal cortex; the preoptic region connects preferentially with the septal region, the innominate substance (of Reichert), and the anterior perforate region; the anterodorsal compartment, i.e., the dorsal intermediate and medial parts of the hypothalamus, connects preferentially with the midline gray substance and the medial thalamus; the supraoptic compartment connects with the infundibulooptic region and the insula; interestingly, a rightward hemispheric connectivity was observed for the anterior and lateral (preoptic, anteroventral, and lateral) hypothalamus. The routes of these connections used known fascicles (Lemaire et al., 2011a,b), such as the large medial forebrain fascicle and ansa lenticularis, and others fallen into oblivion, such as the radiate system and thalamic radiations (Dejerine, 1901). The retromamillary region containing the ventral tegmental area is linked functionally with the substantia nigra compacta and the retrorubral nucleus (retrorubral field), the so-called VTA-nigral complex in rodents, which is connected with the supramamillary nucleus (not defined in human; http://braininfo.rprc.washington.edu/) and the lateral hypothalamus (Ferreira et al., 2008). The connectivity of the ventral tegmental area is wide and complex (Riley, 1953; Nieuwenhuys et al., 2008; Haber et al., 2012; Kwon and Jang, 2014; Zhou et al., 2019). The 3D organization of the macroconnectivity of the hypothalamus is partially depicted in Fig. 5.4.

MRI MAPS, SEGREGATION, AND WHITE MATTER CONNECTIVITY

93

Fig. 5.4. Connectivity of the hypothalamus: olfactory cortices (A; 1 subcallosal gyrus or carrefour olfactif de Broca, 2 olfactive tubercle, 3 uncus, and 4 parahippocampal gyrus [anterior segment]), limbic cortices (including olfactory cortices), and amygdalohippocampal complex (B), 3D views (anteromedial, C; posteromedial, D) with coronal section going through the optical tract (perpendicular to ACPC); ac, anterior commissure; al, ansa lenticularis; am, amygdala; dbd, diagonal band of Broca; fo, fornix; ha, habenula; hi, hippocampus; hyp, hypophysis; is, innominate substance; itp, inferior thalamic peduncle; lgb, lateral geniculate body; ln, lenticular nucleus; oa, olfactory area; ot, optical tract; pc, posterior commissure; rf, retroflexus fascicle; sm, stria medullaris; sn, substantia nigra; st, stria terminalis; stn, subthalamic nucleus; tha, thalamus; vta, ventral tegmental area; zi, zona incerta.

CONCLUSION Further works using mixed methods with high geometric resolution of structural and functional image data sets should facilitate in vivo deciphering of the functional connectomics of the human hypothalamus. The understanding of the hypothalamus’ connections and their related functions may lead to new and effective targeted treatments for psychiatric and endocrine diseases. The wide hypothalamic connections demonstrated here, based on the literature and our own findings with tractography, bring to light the hypothalamus as the crossroads of the most instinctive behaviors for survival. It embodies emotions, memories, and perceptions of the

world around us. It integrates the most diverse regions of the brain. Centered between the hindbrain and the forebrain, the hypothalamus integrates the most primitive brain with the most developed areas that make up the intelligent human being.

REFERENCES Baroncini M, Jissendi P, Balland E et al. (2012). MRI atlas of the human hypothalamus. Neuroimage 59: 168–180. https://doi.org/10.1016/j.neuroimage.2011.07.013. Dejerine J (1901). Anatomie des centres nerveux (Tomes 1 and 2), Rueff et Cie, Paris. Duvernoy H, Cabanis E-A, Iba-Zizen M-T et al. (1992). Le cerveau humain: surfaces, coupes seriees tridimentionelles

94

J.-J. LEMAIRE AND A. DE SALLES

et IRM, Springer-Verlag, Paris Berlin Heidelberg New York Londres Tokyo Hong Kong Barcelone Budapest. Federative International Programme for Anatomical Terminology, FIPAT (2017). Terminologia neuroanatomica. http://fipat.library.dal.ca/wp-content/uploads/2017/02/ FIPAT-TNA-Ch1.pdf. Ferreira JGP, Del-Fava F, Hasue RH et al. (2008). Organization of ventral tegmental area projections to the ventral tegmental area-nigral complex in the rat. Neuroscience 153: 196–213. https://doi.org/10.1016/j. neuroscience.2008.02.003. Fontaine D, Lanteri-Minet M, Ouchchane L et al. (2010). Anatomical location of effective deep brain stimulation electrodes in chronic cluster headache. Brain 133: 1214–1223. https://doi.org/10.1093/brain/awq041. Haber SN, Adler A, Bergman H et al. (2012). The basal ganglia. In: The human nervous system, third edn. Amsterdam, Boston, Heidelberg, London, New York, Oxford, Paris, San Diego, San Francisco, Singapore, Sydney, Tokyo, pp. 678–738. Kullmann S, Heni M, Linder K et al. (2014). Resting-state functional connectivity of the human hypothalamus. Hum Brain Mapp 35: 6088–6096. https://doi.org/10.1002/ hbm.22607. Kwon HG, Jang SH (2014). Differences in neural connectivity between the substantia nigra and ventral tegmental area in the human brain. Front Hum Neurosci 8: 41, 1–4. https:// doi.org/10.3389/fnhum.2014.00041. Laget P (1973). Elements de neuro-anatomie fonctionnelle, Masson, Paris. Lemaire J-J, Cosnard G, Sakka L et al. (2011a). White matter anatomy of the human deep brain revisited with high resolution DTI fibre tracking. Neurochirurgie 57: 52–67. https://doi.org/10.1016/j.neuchi.2011.04.001. Lemaire J, Frew AJ, McArthur D et al. (2011b). White matter connectivity of human hypothalamus. Brain Res 1371: 43–64. https://doi.org/10.1016/j.brainres.2010.11.072. Lemaire J, Nezzar H, Sakka L et al. (2013). Maps of the adult human hypothalamus. Surg Neurol Int 4: S156–S163. https://doi.org/10.4103/2152-7806.110667.

Lemaire J-J, De Salles A, Coll G et al. (2019). MRI atlas of the human deep brain. Front Neurol 10: 851, 1–8. https://doi. org/10.3389/fneur.2019.00851. Makris N, Swaab DF, van der Kouwe A et al. (2013). Volumetric parcellation methodology of the human hypothalamus in neuroimaging: normative data and sex differences. Neuroimage 69: 1–10. https://doi.org/10.1016/j. neuroimage.2012.12.008. Mettler F (1948). Neuroanatomy, C.V. Mosby, St. Louis. Nauta WJH, Haymaker W (1969). Hypothalamic nuclei and fiber connections. In: The hypothalamus, Charles C. Thomas, Springfield, Illinois, USA, pp. 136–200. Nieuwenhuys R, Voogd J, Huijzen C (2008). The human central nervous system, fourth. edn. Springer-Verlag, Berlin, Heidelberg, New York. Palkovits M, Zaborszky L (1979). Neural connections of the hypothalamus. In: Handbook of the hypothalamus, Anatomy of the hypothalamus, Marcel Dekker, New York and Basel, pp. 379–509. Riley H (1953). An atlas of the basal ganglia, brain stem and spinal cord, Williams & Wilkins, Baltimore. Schaltenbrand G, Bailey P (1959). Introduction to stereotaxis with an atlas of the human brain, Georg Thieme Verlag, Stuttgart. Talairach J, David M, Tournoux P et al. (1957). Atlas d’anatomie stereotaxique. Reperage radiologique indirect des noyaux gris centraux des regions mesencephalo-sousoptiques et hypothalamiques de l’homme, Masson et Cie, Paris. Van Hartevelt TJ, Kringelbach ML (2012). The olfactory system. In: JK Mai, G Paxinos (Eds.), The human nervous system. Amsterdam, Boston, Heidelberg, London, New York, Oxford, Paris, San Diego, San Francisco, Singapore, Sydney, Tokyo, pp. 1219–1238. Wahren W (1959). Anatomy of the hypothalamus. In: Introduction to stereotaxis with an atlas of the human brain, Georg Thieme Verlag Stuttgart. 119–151. Zhou G, Lane G, Cooper SL et al. (2019). Characterizing functional pathways of the human olfactory system. Elife 8: 1–27. https://doi.org/10.7554/elife47177.

Handbook of Clinical Neurology, Vol. 179 (3rd series) The Human Hypothalamus: Anterior Region D.F. Swaab, F. Kreier, P.J. Lucassen, A. Salehi, and R.M. Buijs, Editors https://doi.org/10.1016/B978-0-12-819975-6.00004-2 Copyright © 2021 Elsevier B.V. All rights reserved

Chapter 6

Magnetic resonance imaging of the hypothalamo–pituitary region MILICA PEROSEVIC1,3*, PAMELA S. JONES2,3, AND NICHOLAS A. TRITOS1,3 1

Neuroendocrine Unit, Massachusetts General Hospital, Boston, MA, United States

2

Department of Neurosurgery, Massachusetts General Hospital, Boston, MA, United States 3

Harvard Medical School, Boston, MA, United States

Abstract The diagnosis and management of mass lesions in the sellar and parasellar areas remain challenging. When approaching patients with possible sellar or hypothalamic masses, it is important not only to focus on imaging but also detect possible pituitary hormone deficits or excess, in order to establish an appropriate diagnosis and initiate treatment. The imaging modalities used to characterize hypothalamic and pituitary lesions have significantly evolved over the course of the past several years. Computed tomography (CT) and CT angiography play a major role in detecting various sellar lesions, especially in patients who have contraindications to magnetic resonance imaging (MRI) and can also yield important information for surgical planning. However, MRI has become the gold standard for the detection and characterization of hypothalamic and pituitary tumors, infections, cystic, or vascular lesions. Indeed, the imaging characteristics of hypothalamic and sellar lesions can help narrow down the differential diagnosis preoperatively. In addition, MRI can help establish the relationship of mass lesions to surrounding structures. A pituitary MRI examination should be obtained if there is concern for mass effect (including visual loss, ophthalmoplegia, headache) or if there is clinical suspicion and laboratory evidence of either hypopituitarism or pituitary hormone excess. The information obtained from MRI images also provides us with assistance in planning surgery. Using intraoperative MRI can be very helpful in assessing the adequacy of tumor resection. In addition, MRI images yield reliable data that allow for noninvasive monitoring of patients postoperatively.

INTRODUCTION The base of the skull is a very complex area containing critical neurovascular structures, which may serve as a source of many pathologies that can arise from the pituitary gland, infundibular stalk, leptomeninges, cavernous sinuses and their contents (internal carotid arteries or cranial nerves III, IV, V, or VI), or the skull base itself. The aim of this chapter is to review the most relevant imaging characteristics of the pituitary–hypothalamic region, indications for imaging, characteristics of sellar,

parasellar, and hypothalamic lesions, indications for surgery and discuss intraoperative magnetic resonance imaging (MRI). The focus will be on MRI imaging, as it is the gold standard for imaging of the pituitary and hypothalamus nowadays. The comprehensive care of patients with pituitary and hypothalamic lesions generally requires a multidisciplinary team, including endocrinologists, neurosurgeons, radiologists, neurooncologists, and neuroophthalmologists, in order to establish the diagnosis, guide the management, determine the surgical approach, and establish appropriate follow up. Accurate

*Correspondence to: Milica Perosevic, M.D., Massachusetts General Hospital, Neuroendocrine Unit, 100 Blossom Street #140, Boston, MA, 02114, United States. Tel: +1-617-726-7948, Fax: +16-177-261-241, E-mail: [email protected]

96

M. PEROSEVIC ET AL.

imaging of the hypothalamic–pituitary region is essential in implementing appropriate and effective patient evaluation and treatment.

HISTORY OF PITUITARY AND HYPOTHALAMIC IMAGING Imaging modalities of sellar, parasellar, and hypothalamic lesions have significantly evolved over the course of several years, helping to visualize lesions not only in the pituitary gland but also identifying involvement of surrounding structures, such as the cavernous sinuses and optic chiasm. Indirect imaging modalities include plain radiography, pneumoencephalography, and angiography, while direct imaging technologies are computed tomography (CT) and MRI. Plain radiography was the first imaging method that became available for imaging the sella turcica, which mainly evaluated the shape, size, contour, and density of the bony sella (Maya and Pressman, 2017). Lateral imaging of the skull may adequately identify an enlarged or empty sella turcica, recognizable by a double floor of the sella. However, plain radiography is quite a nonspecific and nonsensitive imaging modality for visualizing the pituitary gland, sellar, parasellar lesions, or the hypothalamus (Tekiner et al., 2015). Pneumoencephalography included an intraspinal injection of air into the subarachnoid space, which replaced some cerebrospinal fluid (CSF) and provided tissue contrast, leading to a better diagnostic outline of intracranial lesions. However, its use was limited by significant side effects such as headaches, neck stiffness, tachycardia, and vomiting; there was also a risk of brainstem herniation due to increased intracranial pressure and, per some study reports, the mortality was as high as 30% (Ishaquex et al., 2017). Angiography has been performed since the 1950s and was mostly used as an indirect imaging method in the diagnosis of large pituitary lesions. If there is lateral enlargement of a sellar mass, it may cause lateral displacement of the intracavernous sinus carotid artery. On the other hand, if there is an expansion of the pituitary gland toward the suprasellar cistern, it may result in displacement of the anterior cerebral or anterior communicating arteries, which can be visualized angiographically. It lacks sensitivity for the diagnosis of pituitary microadenomas (Maya and Pressman, 2017). Angiography is used during aneurysm coiling or pipeline stenting. Computed tomography was introduced in the 1970s and provides details regarding bony structures, pathologic fat, and calcifications. It may precisely detect bone lesions and may help in planning transsphenoidal

surgery. With the addition of intravenous iodine contrast and due to the absence of blood–brain barrier, other structures such as the pituitary gland, infundibulum, and cavernous sinuses can be better visualized. Nowadays, CT has been replaced to a large extent by MRI, which will be discussed in detail further in this chapter. CT mostly provides information regarding osseous processes, including either primary processes in the bone or secondary involvement (metastasis, calcifications). However, CT should be avoided in certain patients (those with allergy to iodinated contrast, impaired renal function, multiple myeloma). A CT scan may be helpful in emergency settings to exclude acute pituitary hemorrhage. Calcifications of craniopharyngioma, acute pituitary hemorrhage (apoplexy), bone erosions, or sclerosis are better visualized on CT, compared to MRI (Fig. 6.1). A CT angiogram is particularly useful in identifying aneurysms or vascular malformations (Fig. 6.2).

Fig. 6.1. CT with calcified lesion in the posterior sella, possible calcified cyst, sagittal view.

Fig. 6.2. CT angiogram showing a partially thrombosed basilar apex aneurysm (arrow). Artifacts from prior aneurysmal pipeline stenting are also seen.

MRI OF THE SELLAR REGION

97

TECHNIQUE AND METHODOLOGY OF MAGNETIC RESONANCE IMAGING MRI is superior in the examination of soft tissues and the visualization of small solid, cystic, fatty, or hemorrhagic lesions in the sella and hypothalamus (Chin et al., 2012). A standard MRI protocol includes thin sections (2–3 mm) in the coronal and sagittal plane using T1-weighted imaging both before and after the administration of gadolinium-containing contrast agents, supplemented by T2-weighted imaging and T2/FLAIR weighted imaging. A “generic” brain MRI examination barely shows the pituitary gland, but an MRI scan performed with a pituitary protocol has a field of view that is large enough to show the sellar contents and hypothalamus in detail. MRI slices should not be thicker than 2 mm, as even a 1 mm gap may miss microlesions, which are sometimes as small as 1–2 mm. Optimal MR imaging parameters yield high spatial resolution and excellent signal-to-noise ratio, which is achieved by utilizing a magnetic field strength of at least 1.5 Tesla and ideally 3.0 Tesla (Pressman, 2017).

NORMAL MRI OF THE PITUITARY GLAND AND HYPOTHALAMIC REGION The dimensions of the pituitary gland depend on patient’s age and gender and may be affected by pregnancy or menopause in females (Figs. 6.3 and 6.4). In both males and females until the age of 11 years and above the age of 50 years, the pituitary gland height is less than 5 mm; however, a smaller size can be found in pathologic conditions such as congenital hypopituitarism. In females of childbearing age, the pituitary height reaches 9 mm and is usually superiorly flat. During pregnancy and postpartum, the height is up to 12 mm and may have an upward convexity.

Fig. 6.4. Normal pituitary gland on MRI, T1-weighted noncontrast image, sagittal view. Note the posterior bright spot (corresponding to the posterior pituitary).

On T1-weighted MRI imaging, the normal anterior lobe of the pituitary (adenohypophysis) is isointense to gray matter, while the posterior lobe (neurohypophysis) is hyperintense, which is commonly recognized as the “bright spot of the pituitary” (Fenstermaker and Abad, 2016). For many years, it was believed that the “brightness” of the posterior lobe comes from fat, but today, we know that it is due to phospholipid vesicles containing vasopressin (ADH). Patients with diabetes insipidus often exhibit loss of the “bright spot” in the posterior pituitary. Interestingly, in infants up to the first few months of life, both the anterior and posterior pituitary lobes were found to have hyperintensity on T1-weighted imaging, but later in life, the adenohypophysis loses its hyperintensity and becomes isointense (Rennert and Doerfler, 2007). Dynamic imaging is obtained in coronal view and includes thin slices just before and after intravenous gadolinium-containing contrast administration (every 30 s for 3 min). With this fast scanning technique, the aim is to evaluate small lesions, such as microadenomas, which are largely missed during regular MRI imaging. It is useful in differentiating normal pituitary tissue, which avidly enhances after contrast administration, from microadenomas, which have delayed contrast enhancement (Zamora and Castillo, 2017). Earlier enhancement is noticed in normal posterior lobe compared to the anterior, due to its direct blood supply via branches of the internal carotid artery.

INDICATIONS FOR PITUITARY MRI

Fig. 6.3. Normal pituitary gland on MRI (arrow). Postcontrast T1-weighted image, coronal view.

The most common indications for pituitary and hypothalamic imaging are: mass effect (exerted by a sellar mass to surrounding structures), hormonal imbalance (decreased or excessive pituitary hormone production), or follow-up on incidentalomas (see later and Table 6.1).

98

M. PEROSEVIC ET AL.

Table 6.1 a

Indications for pituitary and hypothalamic imaging Mass effect

● Ophthalmoplegia, vision loss

(central or peripheral) ● Changes in mental status

(suspected hydrocephalus) ● Ataxia, hemiparesis (suspected

brain tumor) ● Gelastic (laughing) seizures

(suspected hypothalamic hamartomas) ● Severe headaches with focal neurologic deficits Deficiency of pituitary hormone production

● Anterior pituitary deficits

Excess of pituitary hormone production

● Hyperprolactinemia (in the

(central hypogonadism, central hypothyroidism, central hypoadrenalism, GH deficiency) ● Posterior pituitary deficits (diabetes insipidus)

● ● ● ●

Incidentaloma follow up

absence of pregnancy, medication effect, or primary hypothyroidism) Cushing syndrome (ACTH dependent) Growth hormone excess (gigantism or acromegaly) Central (secondary) hyperthyroidism (rare) Precocious puberty

● Microincidentalomas: MRI in

1 year after initial imaging, then every 1–2 years for the next 3–4 years if no increase in size, then less frequently ● Macroincidentalomas: MRI in 6 months after initial imaging, then every 1 year for the next 3–4 years; if no increase in size, then less frequently

Microincidentalomas are lesions 20 mm diameter, abundant Nissl substance >30 mm diameter, visible nucleus, abundant Nissl substance >20 mm diameter, nucleated nerve cells >20 mm diameter, visible nucleolus, abundant Nissl substance Nissl-stained, visible nucleolus >20 mm diameter, visible nucleus, abundant Nissl substance

AD

14

60.8

67.50%

63.71%

—

AD

5

60.8

—

—

62.50%

AD

9

78.9

70% (mean); 83.56% — (max)

—

AD

7

91.1

—

—

54.20%

AD

10

79.3

41.09% (mean); 41.42% (max)

—

—

AD

10

84.5

—

—

No loss

AD

15

65

—

—


78%a (vs adult control);
64.5% (vs elderly control)

AD

7

77

—

—

16.7% (n.s.)

AD

4

89.5

56%

53%

55%

AD

8 (from 24) Unknown for the 8
60%a (ApoE4 selected cases negative);
80%a (ApoE4 positive) 4 83.5 — 2 83.0 11.2% (n.s.)

—

—

— —

n.s. —

3

60

n.s.

—

—

2

71.5

44.80%

—

—

NGFR immunoreactive cell soma ChAT-positive cell body (displaying either strong or light immunostaining) AChE-positive, >30 mm, abundant Nissl substance on adjacent section ChAT-immunopositive cells ChAT-immunopositive cells

2006 Fujishiro et al. (2006) AD AD 2017 Liu (2017) Lewy body disorders All neurons using Nissl stain 1983 Whitehouse et al. (1983) >30 mm diameter, abundant PD Nissl substance PDD

1983 Arendt et al. (1983) 1995 Arendt et al. (1995a)

Cholinergic neurons 2006 Fujishiro et al. (2006) 2014 Hall et al. (2014) 2017 Liu (2017)

>20 mm diameter, abundant PD Nissl substance PD >20 mm diameter, visible nucleus, abundant Nissl substance

5

58.5

76.60%

69.80%

—

6

58.2

—

—


75%a (vs adult control);
59.7% (vs elderly control)

ChAT-immunopositive cells DLB ChAT-immunopositive cells PD PDD ChAT-immunopositive cells PD PD-MCI PDD DLB

8 5 6 8 5 16 2

75 60.4 60.2 73.5 81.4 78.2 74.5

— — — 32.0% (n.s.) 45.5% (n.s.) 52.1% 68.9% (n.s.)

— — — — — — —


45%a n.s. n.s. — — — —

Postencephalitic 1 Parkinsonism Korsakoff’s disease 3

50.3

85.16% increase

—

—

43.3

56.67%

40.33%

—

CJD

1

47

—

—

Infantile autism

1

29

11% (n.s.)

54% increase

43% (right hemisphere); 39% (left hemisphere) —

Aging

11

78.1

—

—


38%a (vs adult control)

39.5

—

—

54.8

74%

72%


68%a (vs adult control);
48.4% (vs elderly control) —

Other neurological conditions All neurons using Nissl stain 1983 Whitehouse et al. (1983) >30 mm diameter, abundant Nissl substance >20 mm diameter, abundant 1983 Arendt et al. (1983) Nissl substance 1984 Arendt et al. (1984) >20 mm diameter, abundant Nissl substance 1985 Bauman and Kemper Visible nucleoli (1985) 1995 Arendt et al. (1995a) >20 mm diameter, visible nucleus, abundant Nissl substance

2013 R€ub et al. (2013)

Korsakoff’s disease 6 with Wernicke’s encephalopathy All countable Nissl-stained SCA2 4 neurons

Table adapted with permission from Liu, A.K.L., Lim, E.J., Ahmed, I., et al., 2018. Review: revisiting the human cholinergic nucleus of the diagonal band of Broca. Neuropathol Appl Neurobiol 44, 647–662. doi: 10.1111/nan.12513. a Approximation from graphs within the study. Abbreviations: AChE, acetylcholinesterase; AD, Alzheimer’s disease; ChAT, choline acetyltransferase; CJD, Creutzfeldt–Jakob disease; DLB, dementia with Lewy bodies; MSN, medial septal nucleus; NGFR, nerve growth factor receptor; n.s., no significance; nvlDBB, nucleus of the vertical limb of diagonal band of Broca; PD, Parkinson’s disease; PDD, Parkinson’s disease dementia; PD-MCI, Parkinson’s disease with mild cognitive impairment; and SCA2, spinocerebellar ataxia type 2.

182

A.K.L. LIU AND S.M. GENTLEMAN

mean neuronal cross sectional area (Leh
ericy et al., 1993), whereas a more contemporary study found neuronal shrinkage but no significant loss of Ch1/2 neurons in four AD cases (Fujishiro et al., 2006). The disparities in these results may be due to differences in the sensitivity of the anti-ChAT antibodies or the staining amplification techniques used, but, nevertheless, there is a general consensus that the rostral cholinergic basal forebrain is relatively spared in AD.

PARKINSON’S DISEASE (PD) AND LEWY BODY DEMENTIAS (LBD) There have been relatively few studies investigating neuronal loss in the nvlDBB in Lewy body disorders (Table 11.1). Whitehouse and colleagues, using Nissl staining, reported 44.8% loss of nvlDBB neurons in two PD with dementia (PDD) cases but no significant change in three PD cases with no cognitive impairment (Whitehouse et al., 1983). Subsequent studies, using a >20 mm neuronal size exclusion, reported 76.6% loss of nvlDBB neurons in five cases (Arendt et al., 1985) and another, with six PD cases, reported 59.7% loss of nvlDBB and MSN neurons (Arendt et al., 1995a). In a recent comparative study, it has been reported that there is significant Ch1/2 loss in dementia with Lewy body (DLB) cases relative to age-matched controls and AD cases (Fujishiro et al., 2006). Another study using postmortem tissues from a prospective, and clinically well-characterized, cohort found no significant ChATpositive neuronal loss in the Ch1/2 region of five PD and six PDD cases, although there was a trend toward a

decrease in neurons associated with the progression from PD to PDD (Hall et al., 2014). Both studies have linked the change in nvlDBB neuronal count in LBD cases to the well-recognized increase in ubiquitin-positive, alphasynuclein Lewy neurite pathology in the CA2 subsector of the hippocampus (Dickson et al., 1991, 1994; Irwin et al., 2012). From the Parkinson’s UK Tissue Bank, we selected 33 cases (2 AD, 8 PD, 5 PD with mild cognitive impairment, 16 PDD, and 2 DLB) and six age-matched controls with basal forebrain sections containing the nvlDBB and quantified cholinergic neuronal loss using ChAT immunohistochemistry (Fig. 11.3). A significant loss of nvlDBB ChAT-positive neurons was found in Lewy body disorders with cognitive impairment (52.1%) but not in AD or PD without cognitive impairment (Liu, 2017). This further supports the suggestion that the cholinergic nvlDBB is more severely affected in LBD as compared to AD, although similar studies in a larger cohort will be needed to confirm this hypothesis.

OTHER NEUROLOGICAL CONDITIONS AND AGING Degeneration of the nvlDBB is not exclusive to neurodegenerative diseases, with a 38% cell loss having been reported in aged brains when compared to younger adult controls (Arendt et al., 1995a). However, there is remarkably little in the literature concerning other neurological conditions (Table 11.1). In Korsakoff’s disease, there have been a number of studies suggesting a 50% cell loss (Arendt et al., 1983, 1995a) and then there are single-case

Fig. 11.3. Photomicrographs of ChAT-immunostained basal forebrain sections comparing cholinergic neuronal loss in the nvlDBB between control, AD, DLB, PD, PD-MCI, and PDD cases. Magnification 40x.

THE DIAGONAL BAND OF BROCA IN HEALTH AND DISEASE studies of postencephalitic Parkinsonism (Whitehouse et al., 1983), Creutzfeldt–Jakob disease (Arendt et al., 1984), and infantile autism (Bauman and Kemper, 1985), which have reported varying degrees of neuronal loss. Interestingly, severe basal forebrain cholinergic degeneration has been reported in spinocerebellar ataxia types 1 (SCA1) and 2 (SCA2) with relatively well-preserved cognitive function (R€ ub et al., 2012, 2013). In one of these studies, a 74% loss of nvlDBB neurons was observed in SCA2 cases (n ¼ 4). In these cases, there was no significant correlation with tau pathology (R€ub et al., 2013), indicating a possible causal role of polyglutamine expansion in the death of basal forebrain cholinergic neurons.

FUNCTIONAL CORRELATES OF nvlDBB CELL LOSS Functional connections of the nvlDBB in the human brain are presumed to be similar to those seen in nonhuman primates but this remains to be confirmed. However, case studies from specific vascular lesions have provided some insight. The nvlDBB is supplied by perforator arteries of the anterior communicating artery (Román and Kalaria, 2006). Ruptured anterior communicating artery aneurysms tend to have widespread effects on multiple basal forebrain nuclei and fiber tracts (Lindqvist and Norl
en, 1966; Talland et al., 1967; Logue et al., 1968; Gade, 1982; Volpe and Hirst, 1983; Alexander and Freedman, 1984; Vilkki, 1985; Phillips et al., 1987; Parkin et al., 1988). Nevertheless, there have been two cases reported with relatively discrete basal forebrain lesions (Damasio et al., 1985; Abe et al., 1998). In one of the cases, surgical clipping of an aneurysm at the A1 and A2 junction of the left anterior cerebral artery was performed (Damasio et al., 1985). The anatomical regions affected by this lesion were similar to a case reported by Abe and colleagues showing an isolated lesion in the right nvlDBB and anterior hypothalamus (Abe et al., 1998). In both cases, the patients exhibited anterograde and retrograde amnesia. However, there was a benefit with cueing that indicated a deficit in retrieval rather than encoding of memory. A similar amnestic profile was reported by Morris and colleagues on a case with surgical resection of a low-grade astrocytoma in the right anterior basal ganglia region containing the nvlDBB (Morris et al., 1992). From these cases, it appears that the nvlDBB is involved in memory retrieval. Memory impairment in PD involves deficits in memory retrieval rather than the storage/encoding deficits seen in AD (Emre et al., 2007). This would suggest that nvlDBB may be more significantly affected in LBD than in AD. This is supported by a postmortem study reporting a significant decrease in ChAT-positive neurons in

183

the nvlDBB of DLB cases relative to AD and agematched controls (Fujishiro et al., 2006). However, a recent imaging study with 11 AD and 11 DLB patients failed to demonstrate difference in MSN/nvlDBB atrophy on structural MRI scans (Grothe et al., 2014). This discrepancy may be due to the fact that MRI-based volumetric changes cannot distinguish between neuronal loss and cell atrophy. In rodent studies, low-frequency theta rhythm in the hippocampus is generated in the MSN-nvlDBB, mainly through GABAergic output neurons, and contributes to spatial learning and memory (Freund and Antal, 1988; Colgin, 2016). GABAergic output neurons from the MSN/nvlDBB form synaptic contacts with hippocampal interneurons and act as pacemaker cells for theta rhythm with disinhibition of hippocampal pyramidal neurons (Freund and Antal, 1988). Recent studies also showed glutamatergic neurons modulate theta rhythm via local modulation of GABAergic neurons within MSN/ nvlDBB (Robinson et al., 2016). Since MSN/nvlDBB neuronal loss was initially reported in AD cases with little or no significant loss in the ChAT-immunopositive component, it could be hypothesized that a noncholinergic GABAergic and glutamatergic neuronal loss is more significant in AD. As with the nvlDBB, the hippocampal CA2 subfield is a relatively unexplored region in the human brain. Reciprocal connections between the MSN-nvlDBB and the CA2 have been identified in the mouse brain (Cui et al., 2013). In LBD, Lewy neurites are largely confined to the CA2 subfield in the hippocampus (Dickson et al., 1991, 1994; Irwin et al., 2012; Hall et al., 2014), in contrast to the preferential deposition of neurofibrillary tangles in the CA1 region in AD (Hyman et al., 1984). Coincidentally, this is the subregion where the highest density of ChAT-positive fibers can be identified in the human hippocampus (Ransmayr et al., 1989). Although the presence of hippocampal dopaminergic innervation has been extensively reported in rodents (Milner and Bacon, 1989), evidence of such projection in the human brain was lacking with one study failing to demonstrate the presence of monoaminergic fibers in the hippocampus of DLB and aged control brains using immunostaining with antityrosine hydroxylase antibodies (Dickson et al., 1994). On the other hand, significant hippocampal cholinergic depletion was found in PDD cases when compared with PD and control (Hall et al., 2014). The subfield specific protein aggregation pathology and cholinergic deficits in the hippocampus of Lewy body disorders appeared to be associated with the severe nvlDBB cholinergic depletion described in the sections mentioned previously. In a recent study, we found that in PD, only cases with dementia have a significantly greater Lewy pathology burden in the hippocampal

184

A.K.L. LIU AND S.M. GENTLEMAN

Fig. 11.4. Proposed schema for the neurodegenerative changes within cholinergic and noncholinergic populations of the nvlDBB in Lewy body disorders and Alzheimer’s disease with possible clinicopathological correlates. Abbreviations: AD, Alzheimer’s disease; CA, Cornu ammonis; Ch2, cholinergic population of the vertical limb of the diagonal band of Broca; DG, dentate gyrus; GABA, g-aminobutyric acid; and LBD, Lewy body disorder. Figure reproduced with permission from Liu, A.K.L., Lim, E.J., Ahmed, I., et al., 2018. Review: revisiting the human cholinergic nucleus of the diagonal band of Broca. Neuropathol Appl Neurobiol 44, 647–662. doi: 10.1111/nan.12513.

CA2 subfield, whereas cholinergic fiber depletion was already evident in PD cases with mild cognitive impairment and this was significantly correlated with loss of cholinergic neurons in nvlDBB (Liu et al., 2019). As a result, it can be hypothesized that specific Lewy pathology targeting the cholinergic system within the CA2 subfield is associated with cholinergic degeneration in the nvlDBB, which may contribute to the unique memory retrieval deficit seen in patients with Lewy body disorders, while a predominant noncholinergic neuronal loss of the nvlDBB may contribute to the encoding and storage memory deficits in AD (Fig. 11.4).

CONCLUSION The functional importance of the diagonal band in human cognition has long been recognized due to its anatomical position in the limbic loop. However, the significance of pathology in the nvlDBB in neurodegenerative conditions has not been studied in detail for the following reasons: (i) absence of clear anatomical boundaries to delineate nvlDBB from surrounding magnocellular basal forebrain nuclei; (ii) application of size criteria for neuronal quantification, discounting the possibility of cell atrophy; and (iii) most studies attributing magnocellular

neuronal loss in the nvlDBB to cholinergic nerve cells without use of specific cholinergic markers. In this chapter, we have reestablished the anatomical delineation of the nvlDBB to facilitate reliable sampling for future clinicopathological studies. In addition, we recommend the use of specific cholinergic immunohistochemical markers such as ChAT for the identification and quantification of Ch2 (cholinergic component of nvlDBB) neurons. Through the study of cases with discrete vascular or other lesions affecting the rostral basal forebrain, it can be deduced that the nvlDBB plays a discrete role in retrieval memory function, a cognitive domain that is severely affected in Lewy body disorders. Furthermore, extrapolating from existing clinicopathological studies, an anatomical and functional connection between Ch2 and the CA2 subfield in the hippocampus may be especially vulnerable to Lewy pathologies.

REFERENCES Abe K, Inokawa M, Kashiwagi A et al. (1998). Amnesia after a discrete basal forebrain lesion. J Neurol Neurosurg Psychiatry 65: 126–130. https://doi.org/10.1136/jnnp. 65.1.126.

THE DIAGONAL BAND OF BROCA IN HEALTH AND DISEASE Alexander MP, Freedman M (1984). Amnesia after anterior communicating artery aneurysm rupture. Neurology 34: 752. Alexandris A, Liu AKL, Chang RC-C et al. (2015). Differential expression of galanin in the cholinergic basal forebrain of patients with Lewy body disorders. Acta Neuropathol Commun 3: 77. https://doi.org/10.1186/ s40478-015-0249-4. Amunts K, Zilles K (2015). Architectonic mapping of the human brain beyond Brodmann. Neuron 88: 1086–1107. https://doi.org/10.1016/j.neuron.2015.12.001. Andy OJ, Stephan H (1968). The septum in the human brain. J Comp Neurol 133: 383–410. https://doi.org/10.1002/ cne.901330308. Arendt T, Bigl V, Arendt A et al. (1983). Loss of neurons in the nucleus basalis of Meynert in Alzheimer’s disease, paralysis agitans and Korsakoff’s disease. Acta Neuropathol 61: 101–108. Arendt T, Bigl V, Arendt A (1984). Neurone loss in the nucleus basalis of Meynert in Creutzfeldt-Jakob disease. Acta Neuropathol 65: 85–88. Arendt T, Bigl V, Tennstedt A et al. (1985). Neuronal loss in different parts of the nucleus basalis is related to neuritic plaque formation in cortical target areas in Alzheimer’s disease. Neuroscience 14: 1–14. Arendt T, Br€uckner MK, Bigl V et al. (1995a). Dendritic reorganisation in the basal forebrain under degenerative conditions and its defects in Alzheimer’s disease. II. Ageing, Korsakoff’s disease, Parkinson’s disease, and Alzheimer’s disease. J Comp Neurol 351: 189–222. Arendt T, Marcova L, Bigl V et al. (1995b). Dendritic reorganisation in the basal forebrain under degenerative conditions and its defects in Alzheimer’s disease. I. Dendritic organisation of the normal human basal forebrain. J Comp Neurol 351: 169–188. Bartus RT, Dean RL, Beer B et al. (1982). The cholinergic hypothesis of geriatric memory dysfunction. Science 217: 408–414. Bauman M, Kemper TL (1985). Histoanatomic observations of the brain in early infantile autism. Neurology 35: 866. Broca P (1877). M
emoires d’anthropologie, Reinwald. Broca P (1878). Anatomie comparee des circonvolutions cerebrales: Le grand lobe limbique et la scissure limbique dans la serie des mammiferes. Broca P, Pozzi S (1888). M
emoires sur le cerveau de l’homme et des primates, C. Reinwald. Brockhaus H (1942). Vergleichend-anatomische Untersuchungen €uber den Basalkernkomplex. J Psychol Neurol 51: 57–95. Chan-Palay V (1988). Galanin hyperinnervates surviving neurons of the human basal nucleus of Meynert in dementias of Alzheimer’s and Parkinson’s disease: a hypothesis for the role of galanin in accentuating cholinergic dysfunction in dementia. J Comp Neurol 273: 543–557. https://doi. org/10.1002/cne.902730409. Colgin LL (2016). Rhythms of the hippocampal network. Nat Rev Neurosci 17: 239–249. https://doi.org/10.1038/ nrn.2016.21. Colom LV, Castaneda MT, Reyna T et al. (2005). Characterization of medial septal glutamatergic neurons

185

and their projection to the hippocampus. Synapse 58: 151–164. https://doi.org/10.1002/syn.20184. Crosby EC (1917). The forebrain of Alligator mississippiensis. J Comp Neurol 27: 325–402. Cui Z, Gerfen CR, Young WS (2013). Hypothalamic and other connections with dorsal CA2 area of the mouse hippocampus. J Comp Neurol 521: 1844–1866. https://doi.org/ 10.1002/cne.23263. Daitz HM, Powell TP (1954). Studies of the connexions of the fornix system. J Neurol Neurosurg Psychiatry 17: 75–82. Damasio AR, Graff-Radford NR, Eslinger PJ et al. (1985). Amnesia following basal forebrain lesions. Arch Neurol 42: 263–271. https://doi.org/10.1001/archneur.1985. 04060030081013. Dickson DW, Ruan D, Crystal H et al. (1991). Hippocampal degeneration differentiates diffuse Lewy body disease (DLBD) from Alzheimer’s disease: light and electron microscopic immunocytochemistry of CA2-3 neurites specific to DLBD. Neurology 41: 1402–1409. Dickson DW, Schmidt ML, Lee VM et al. (1994). Immunoreactivity profile of hippocampal CA2/3 neurites in diffuse Lewy body disease. Acta Neuropathol 87: 269–276. https://doi.org/10.1007/BF00296742. Emmers R, Akert K (1963). A stereotaxic atlas of the brain of the squirrel monkey Saimiri Sciureus; Emmers, Raimond, and Konrad Akert; With a Forew. by Clinton N. Woolsey, University of Wisconsin Press. Emre M, Aarsland D, Brown R et al. (2007). Clinical diagnostic criteria for dementia associated with Parkinson’s disease. Mov Disord 22: 1689–1707. https://doi.org/ 10.1002/mds.21507. Etienne P, Robitaille Y, Wood P et al. (1986). Nucleus basalis neuronal loss, neuritic plaques and choline acetyltransferase activity in advanced Alzheimer’s disease. Neuroscience 19: 1279–1291. Freund TF, Antal M (1988). GABA-containing neurons in the septum control inhibitory interneurons in the hippocampus. Nature 336: 170–173. https://doi.org/10.1038/336170a0. Fujishiro H, Umegaki H, Isojima D et al. (2006). Depletion of cholinergic neurons in the nucleus of the medial septum and the vertical limb of the diagonal band in dementia with Lewy bodies. Acta Neuropathol 111: 109–114. https://doi. org/10.1007/s00401-005-0004-1. Gade A (1982). Amnesia after operations on aneurysms of the anterior communicating artery. Surg Neurol 18: 46–49. Gertz HJ, Cervos-Navarro J, Ewald V (1987). The septohippocampal pathway in patients suffering from senile dementia of Alzheimer’s type. Evidence for neuronal plasticity? Neurosci Lett 76: 228–232. https://doi.org/ 10.1016/0304-3940(87)90720-8. Gouras GK, Rance NE, Scott Young W et al. (1992). Tyrosinehydroxylase-containing neurons in the primate basal forebrain magnocellular complex. Brain Res 584: 287–293. https://doi.org/10.1016/0006-8993(92)90907-Q. Gray PA (1924). The cortical lamination pattern of the opossum, Didelphys virginiana. J Comp Neurol 37: 221–263. Grothe MJ, Schuster C, Bauer F et al. (2014). Atrophy of the cholinergic basal forebrain in dementia with Lewy bodies

186

A.K.L. LIU AND S.M. GENTLEMAN

and Alzheimer’s disease dementia. J Neurol 261: 1939–1948. https://doi.org/10.1007/s00415-014-7439-z. Gurdjian ES (1928). The corpus striatum of the rat. Studies on the brain of the rat. No. 3. J Comp Neurol 45: 249–281. Hall H, Reyes S, Landeck N et al. (2014). Hippocampal Lewy pathology and cholinergic dysfunction are associated with dementia in Parkinson’s disease. Brain 137: 2493–2508. https://doi.org/10.1093/brain/awu193. Higgins GA, Mufson EJ (1989). NGF receptor gene expression is decreased in the nucleus basalis in Alzheimer’s disease. Exp Neurol 106: 222–236. https://doi.org/10.1016/00144886(89)90155-6. Hyman B, Van Hoesen G, Damasio A et al. (1984). Alzheimer’s disease: cell-specific pathology isolates the hippocampal formation. Science 225: 1168–1170. https:// doi.org/10.1126/science.6474172. Irwin DJ, White MT, Toledo JB et al. (2012). Neuropathologic substrates of Parkinson disease dementia. Ann Neurol 72: 587–598. https://doi.org/10.1002/ana.23659. Johnston JB (1915). The cell masses in the forebrain of the turtle, Cistudo carolina. J Comp Neurol 25: 393–468. Johnston JB (1923). Further contributions to the study of the evolution of the forebrain. J Comp Neurol 35: 337–481. Leh
ericy S, Hirsch EC, Cervera-Pi
erot P et al. (1993). Heterogeneity and selectivity of the degeneration of cholinergic neurons in the basal forebrain of patients with Alzheimer’s disease. J Comp Neurol 330: 15–31. https:// doi.org/10.1002/cne.903300103. Lindqvist G, Norl
en G (1966). Korsakoff’s syndrome after operation on ruptured aneurysm of the anterior communicating artery. Acta Psychiatr Scand 42: 24–34. Liu AKL (2017). Clinicopathological investigations of the cholinergic basal forebrain in Lewy body disorders and ageing, Imperial College, London. Liu AKL, Chang RC-C, Pearce RKB et al. (2015). Nucleus basalis of Meynert revisited: anatomy, history and differential involvement in Alzheimer’s and Parkinson’s disease. Acta Neuropathol 129: 527–540. https://doi.org/10.1007/ s00401-015-1392-5. Liu AKL, Chau TW, Lim EJ et al. (2019). Hippocampal CA2 Lewy pathology is associated with cholinergic degeneration in Parkinson’s disease with cognitive decline. Acta Neuropathol Commun 7: 61. https://doi.org/10.1186/ s40478-019-0717-3. Logue V, Durward M, Pratt RT et al. (1968). The quality of survival after rupture of an anterior cerebral aneurysm. Br J Psychiatry 114: 137–160. Mai JK, Majtanik M, Paxinos G (2016). Atlas of the human brain, Academic Press. McGeer PL, McGeer EG, Suzuki J et al. (1984). Aging, Alzheimer’s disease, and the cholinergic system of the basal forebrain. Neurology 34: 741–745. Mesulam M-M (2013). Cholinergic circuitry of the human nucleus basalis and its fate in Alzheimer’s disease. J Comp Neurol 521: 4124–4144. https://doi.org/10.1002/ cne.23415. Mesulam MM, Geula C (1988). Nucleus basalis (Ch4) and cortical cholinergic innervation in the human brain:

observations based on the distribution of acetylcholinesterase and choline acetyltransferase. J Comp Neurol 275: 216–240. https://doi.org/10.1002/cne.902750205. Mesulam MM, Mufson EJ, Levey AI et al. (1983). Cholinergic innervation of cortex by the basal forebrain: cytochemistry and cortical connections of the septal area, diagonal band nuclei, nucleus basalis (substantia innominata), and hypothalamus in the rhesus monkey. J Comp Neurol 214: 170–197. https://doi.org/10.1002/cne.902140206. Milner TA, Bacon CE (1989). Ultrastructural localization of tyrosine hydroxylase-like immunoreactivity in the rat hippocampal formation. J Comp Neurol 281: 479–495. https:// doi.org/10.1002/cne.902810311. Morris MK, Bowers D, Chatterjee A et al. (1992). Amnesia following a discrete basal forebrain lesion. Brain 115: 1827–1847. https://doi.org/10.1093/brain/115. 6.1827. Mufson EJ, Bothwell M, Kordower JH (1989). Loss of nerve growth factor receptor-containing neurons in Alzheimer’s disease: a quantitative analysis across subregions of the basal forebrain. Exp Neurol 105: 221–232. https://doi. org/10.1016/0014-4886(89)90124-6. Mufson EJ, Cochran E, Benzing W et al. (1993). Galaninergic innervation of the cholinergic vertical limb of the diagonal band (Ch2) and bed nucleus of the stria terminalis in aging, Alzheimer’s disease and Down’s syndrome. Dementia 4: 237–250. Nieuwenhuys R, Voogd J, Van Huijzen C (2007). The human central nervous system: A synopsis and atlas, Springer Science & Business Media. Papez JW, Aronson LR (1934). Thalamic nuclei of Pithecus (Macacus) Rhesus. Arch Neurol Psychiatry 32: 1–26. https://doi.org/10.1001/archneurpsyc.1934.02250070007001. Parkin AJ, Leng NRC, Stanhope N et al. (1988). Memory impairment following ruptured aneurysm of the anterior communicating artery. Brain Cogn 7: 231–243. Phillips S, Sangalang V, Sterns G (1987). Basal forebrain infarction. A clinicopathologic correlation. Arch Neurol 44: 1134–1138. Poirier J, Delisle MC, Quirion R et al. (1995). Apolipoprotein E4 allele as a predictor of cholinergic deficits and treatment outcome in Alzheimer disease. Proc Natl Acad Sci U S A 92: 12260–12264. https://doi.org/10.1073/pnas.92. 26.12260. Ransmayr G, Cervera P, Hirsch E et al. (1989). Choline acetyltransferase-like immunoreactivity in the hippocampal formation of control subjects and patients with Alzheimer’s disease. Neuroscience 32: 701–714. https:// doi.org/10.1016/0306-4522(89)90291-1. Robinson J, Manseau F, Ducharme G et al. (2016). Optogenetic activation of septal glutamatergic neurons drive hippocampal theta rhythms. J Neurosci 36: 3016–3023. Roma´n GC, Kalaria RN (2006). Vascular determinants of cholinergic deficits in Alzheimer disease and vascular dementia. Neurobiol Aging 27: 1769–1785. https://doi.org/ 10.1016/j.neurobiolaging.2005.10.004. R€ ub U, B€ urk K, Timmann D et al. (2012). Spinocerebellar ataxia type 1 (SCA1): new pathoanatomical and

THE DIAGONAL BAND OF BROCA IN HEALTH AND DISEASE clinico-pathological insights. Neuropathol Appl Neurobiol 38: 665–680. https://doi.org/10.1111/j.1365-2990.2012. 01259.x. R€ ub U, Farrag K, Seidel K et al. (2013). Involvement of the cholinergic basal forebrain nuclei in spinocerebellar ataxia type 2 (SCA2). Neuropathol Appl Neurobiol 39: 634–643. https://doi.org/10.1111/nan.12025. Saper CB, Chelimsky TC (1984). A cytoarchitectonic and histochemical study of nucleus basalis and associated cell groups in the normal human brain. Neuroscience 13: 1023–1037. Schaltenbrand G, Bailey P (1959). Einf€uhrung in die Stereotakischen Operationen Mit Einem Atlas des Menschlichen Gehirns, Introduction to stereotaxis with an atlas of the human brainvols. 1–3. Georg Thieme Verlag, Stuttgart. Semba K (2004). Phylogenetic and ontogenetic aspects of the basal forebrain cholinergic neurons and their innervation of the cerebral cortex. Prog Brain Res 145: 3–43. Smith GE (1897). The relation of the fornix to the margin of the cerebral cortex. J Anat Physiol 32: 23–58. Talland GA, Sweet WH, Ballantine HT (1967). Amnesic syndrome with anterior communicating artery aneurysm. J Nerv Ment Dis 145: 179–192.

187

Vilkki J (1985). Amnesic syndromes after surgery of anterior communicating artery aneurysms. Cortex 21: 431–444. Vogels OJ, Broere CA, ter Laak HJ et al. (1990). Cell loss and shrinkage in the nucleus basalis Meynert complex in Alzheimer’s disease. Neurobiol Aging 11: 3–13. https:// doi.org/10.1016/0197-4580(90)90056-6. Volpe BT, Hirst W (1983). Amnesia following the rupture and repair of an anterior communicating artery aneurysm. J Neurol Neurosurg Psychiatry 46: 704–709. Walker LC, Price DL, Young WS (1989). GABAergic neurons in the primate basal forebrain magnocellular complex. Brain Res 499: 188–192. https://doi.org/10.1016/00068993(92)90907-Q. Whitehouse PJ, Hedreen JC, White CL et al. (1983). Basal forebrain neurons in the dementia of Parkinson disease. Ann Neurol 13: 243–248. https://doi.org/10.1002/ana. 410130304. Wilcock GK, Esiri MM, Bowen DM et al. (1988). The differential involvement of subcortical nuclei in senile dementia of Alzheimer’s type. J Neurol Neurosurg Psychiatry 51: 842–849. Zaborszky L, Hoemke L, Mohlberg H et al. (2008). Stereotaxic probabilistic maps of the magnocellular cell groups in human basal forebrain. Neuroimage 42: 1127–1141. https://doi.org/10.1016/j.neuroimage.2008.05.055.

This page intentionally left blank

Handbook of Clinical Neurology, Vol. 179 (3rd series) The Human Hypothalamus: Anterior Region D.F. Swaab, F. Kreier, P.J. Lucassen, A. Salehi, and R.M. Buijs, Editors https://doi.org/10.1016/B978-0-12-819975-6.00010-8 Copyright © 2021 Elsevier B.V. All rights reserved

Chapter 12

Nucleus basalis of Meynert degeneration predicts cognitive impairment in Parkinson’s disease HEATHER WILSON, EDOARDO ROSARIO DE NATALE, AND MARIOS POLITIS* Neurodegeneration Imaging Group, University of Exeter Medical School, London, United Kingdom

Abstract Cognitive impairment affects approximately 20%–50% of patients with Parkinson’s disease (PD), with a higher prevalence as the disease advances. The nucleus basalis of Meynert (NBM) provides the majority of cholinergic innervations to the cerebral cortex. Dysfunction of the cholinergic system and degeneration of the NBM have been implicated in the pathophysiology of cognitive impairment in neurodegenerative disorders including PD. Several studies have aimed to identify risk factors associated with cognitive decline in order to construct models to predict future cognitive impairment in PD. Given the link between cholinergic dysfunction and the pathogenesis of cognitive decline in PD, a number of studies have focused on the role of the NBM underlying cognitive performance. Recently, microstructural alterations within the NBM, detected using diffusion tensor imaging, have been identified as a strong predictor for the development of cognitive impairment in patients with PD. These microstructural changes in NBM have been shown to precede structural gray matter volumetric loss and may present with an early marker to predict cognitive decline in patients with PD. Longitudinal studies are warranted to provide insights into the potential utility of cholinergic positron emission tomography imaging to predict the development of cognitive impairment in PD and other neurodegenerative disorders. Provided the urgent need for disease modifying therapies aiming to slow and ultimately halt the progression of cognitive impairment, neuromodulation of NBM, and treatments targeting the cholinergic system may hold a promising potential. In this review, we discuss the link between NBM pathology and clinical symptomatology of cognitive impairment in PD with a focus on the use of in vivo imaging techniques and potential therapeutic interventions.

INTRODUCTION The nucleus basalis of Meynert (NBM) is a predominantly cholinergic nucleus in the basal forebrain, which provides the majority of cholinergic innervations to the cerebral cortex. Dysfunction of the cholinergic system and degeneration of the NBM have been implicated to play a role in the pathophysiology of cognitive impairment in neurodegenerative disorders including Parkinson’s disease (PD) and Alzheimer’s disease (AD) (Candy et al., 1983; Gratwicke et al., 2013). PD is a common progressive neurodegenerative disorder classified by the presence of cardinal motor

symptoms of bradykinesia, rigidity, and tremor as well as a host of nonmotor symptoms including autonomic dysfunction, sleep disturbances, mood changes, and cognitive impairment (Politis et al., 2010). From a pathological viewpoint, PD is classified by the presence of Lewy body inclusions of alpha-synuclein (Braak et al., 2003). Cognitive impairment affects the spectrum of Lewy body disorders with varying degrees of severity, time of onset, and progression as is observed in idiopathic and familial forms of PD, Parkinson’s disease dementia (PDD) and dementia with Lewy bodies (DLB) (Litvan et al., 2012; Geurtsen et al., 2014; Goldman et al., 2014; Aarsland et al., 2017; Liu et al., 2017). Cognitive

*Correspondence to: Professor Marios Politis, Neurodegeneration Imaging Group, University of Exeter Medical School, London, United Kingdom. E-mail: [email protected]

190

H. WILSON ET AL.

impairment is one of the most common and debilitating nonmotor symptoms of PD and has a major impact on the patient’s quality of life and has a high economic cost (Politis et al., 2010; Vossius et al., 2011; Leroi et al., 2012). Multiple studies have presented models, which include clinical, biological (blood and cerebrospinal fluid (CSF)), and imaging variables, in an effort to identify a reliable model to predict the onset of cognitive impairment in PD and to help stratify patients for future clinical trials (Foltynie et al., 2004; Mollenhauer et al., 2006; Compta et al., 2009; Williams-Gray et al., 2009, 2013; Aarsland et al., 2010; Alves et al., 2010; Hu et al., 2014; Liu et al., 2017; Wilson et al., 2020). However, given the link between cholinergic dysfunction with the pathogenesis of cognitive decline in PD (Candy et al., 1983), a number of studies have recently turned to focus on the role of the NBM (Ray et al., 2018; Schulz et al., 2018). These studies aimed to identify a noninvasive biomarker for the NBM and cholinergic dysfunction to incorporate into these predictive models. In vivo imaging techniques, such as positron emission tomography (PET) molecular imaging and magnetic resonance imaging (MRI), provide valuable tools to study cognitive impairment in PD from the earliest stages of the disease through to advanced disease stages (Politis, 2014; Yousaf et al., 2017). Technological advances over the last decade have enabled these noninvasive imaging techniques to be employed to study the NBM in vivo to help elucidate its role in the onset and progression of cognitive impairment in PD (Ray et al., 2018; Schulz et al., 2018). This review will discuss the link between NBM pathology and clinical symptomatology of cognitive impairment in PD touching on potential therapeutic interventions.

COGNITIVE IMPAIRMENT IN PARKINSON’S DISEASE Cognitive impairment affects approximately 20%–50% of PD patients; although in more advanced stages of the disease, with more than 20 years disease duration, the presence of cognitive impairment has been shown to affect more than 80% of PD patients (Foltynie et al., 2004; Litvan et al., 2012). PD patients with mild cognitive impairment (MCI) have an increased risk of developing dementia, with an annual rate of about 11% of PD-MCI patients developing PDD (Hobson and Meara, 2015) and approximately 42% of PD-MCI patients converting to PDD over within 6–8 years (Galtier et al., 2016). The CamPaIGN study reported a 33% prevalence rate of cognitive impairment in PD patients at the time of diagnosis with the prevalence increasing to 57% after 3–5 years, at which time approximately 10% of the

patients develop dementia (Williams-Gray et al., 2009), and to 46% prevalence after 10 years (Williams-Gray et al., 2013). The presentation of cognitive impairment is heterogeneous with a range of cognitive domains being affected and conflicting views on the precise cognitive profile of PD-MCI (Watson and Leverenz, 2010). The cognitive profile of PD-MCI has been shown to vary from that of patients with MCI due to AD. The presentation of memory deficits and difficulties remembering autographical events are typically more common in MCI due to AD, while in PD-MCI executive dysfunction, which is thought to be due to dopaminergic dysfunction in the fronto-striatal circuit, visuospatial deficits are more frequent (Albert et al., 2011; McKhann et al., 2011; Moustafa and Poletti, 2013). Amnestic cognitive deficits are most typical for AD and AD-MCI (Knopman and Petersen, 2014). In PD-MCI, some studies indicate that nonamnestic symptoms are more common than amnestic cognitive deficits (Aarsland et al., 2009b, 2010; Goldman et al., 2012; Poletti et al., 2012; Kalbe et al., 2016); however, other studies indicate that amnestic dysfunction is more frequent (Sollinger et al., 2010). Cognitive domains affected in PD-MCI can include attention/working memory, executive function, language, memory, and visuospatial function (Watson and Leverenz, 2010; Litvan et al., 2012). However, it is unclear if multiple-domain impairment (Poletti et al., 2012; Cholerton et al., 2014) or single-domain impairment (Aarsland et al., 2009b, 2010; Goldman et al., 2012; Yu et al., 2012) is more common in PD-MCI. Cognitive impairment can present across the spectrum of Parkinsonism disorders including in PD-MCI, PDD, and DLB. It remains to be elucidated whether PDD and DLB truly represent two distinct diseases or rather represent a spectrum of the same disease (Aarsland et al., 2004a; Jellinger and Korczyn, 2018). The term Lewy body dementias has been proposed as an umbrella term to encapsulate these disorders (Walker et al., 2015). The current diagnostic criteria for PDD and DLB are based on the timing of motor and cognitive symptom onset (McKeith et al., 2017). According to the current diagnostic criteria, DLB is diagnosed when cognitive impairment precedes or occurs within 5 years, of Parkinsonism symptoms, while PDD is diagnosed when cognitive impairment develops over 1 year after Parkinsonism onset. Atypical forms of Parkinsonism, such as multiple system atrophy, progressive supranuclear palsy, and corticobasal syndrome, can also present with varying profiles of cognitive impairment (Sulena et al., 2017). However, accurately distinguishing between these disorders currently remains a challenge, especially in early disease stages and before the manifestation of the full spectrum of clinical symptoms (Mok et al., 2004).

NUCLEUS BASALIS OF MEYNERT DEGENERATION PREDICTS COGNITIVE IMPAIRMENT Therefore, care should be taken when interrupting findings of cognitive impairment across the spectrum of Parkinsonism and atypical Parkinsonism disorders. Further research to identify distinct cognitive profiles, and the relationship with underlying disease pathology, across these disorders is warranted and could aid differential diagnosis. The application of different definitions and classification methods to define cognitive impairment in PD varies between studies adding another challenging layer to disentangling cognitive impairment in PD and the identification of reliable predictive markers. To help address this, the Movement Disorder Society published guidelines for the diagnostic criteria of PD-MCI based on two levels (Litvan et al., 2012; Geurtsen et al., 2014). Level 1 is defined with an abbreviated assessment such as the Mini-Metal State Examination or the Montreal Cognitive Assessment (MoCA). Level 2 involves a more comprehensive assessment requiring formal neuropsychological testing with at least two tests in each of the five cognitive domains to be affected. The criteria for determining impairment in a cognitive measure are usually set at 1.5 standard deviations from the normative mean (Schinka et al., 2010). Several studies have aimed to identify risk factors associated with cognitive decline in order to construct predictive models of cognitive decline in PD. However, the predictors identified are not always consistent across studies. Cognitive impairment in PD has been associated with older age of onset (Meireles and Massano, 2012), motor symptom severity (Aarsland et al., 2010), the presence of an akinetic-rigid phenotype (Wojtala et al., 2019), diabetes mellitus (Ashraghi et al., 2016; Pagano et al., 2018), olfactory dysfunction (Baba et al., 2012) (Bohnen et al., 2010), and CSF levels of tau, hyperphosphorylated-tau and beta-amyloid (Mollenhauer et al., 2006; Compta et al., 2009; Alves et al., 2010). The CamPaIGN study reported that baseline predictors of dementia, at 10-year follow-up, were older age, semantic fluency deficit, impaired ability to copy an intersecting pentagons figure, and a score of 25 on the Unified Parkinson’s disease Rating Scale Part-III for motor symptoms (Williams-Gray et al., 2013). The Oxford Discovery cohort reported similar findings with older age and more advanced H&Y staging as predictors of lower MoCA scores over an 18-month follow-up period (Hu et al., 2014). Recently, our group reported a profile in early drug-naïve PD patients of increased age at disease onset and worse performance on the semantic fluency and symbol digit modalities tests, at a 36-month follow-up, which gave a 64% risk for cognitive decline at a 36-month follow-up (Wilson et al., 2020). A clinical-genetic risk score has been proposed by Liu and colleagues, based on the b-glucocerebrosidase

191

(GBA) genotype, to predict cognitive impairment within 10 years of PD diagnosis (Liu et al., 2017). Furthermore, the apolipoprotein E4 (APoE4) (Morley et al., 2012), microtubule-associated protein tau H1/H1 genotype (Williams-Gray et al., 2013), catechol-O-methyltransferase Val/Val (Williams-Gray et al., 2008), and brain-derived neurotrophic factor Val/Val genotypes (Foltynie et al., 2005) have been shown to be influence dementia and cognitive decline. Therefore, genetic risk factors are likely to play a role in the development of cognitive decline and should be taken into considering in future predictive models and to help stratify patients. Neuroimaging and biological variables, including blood and CSF markers, have also been included within predictive models of cognitive decline. We recently reported a model combining reduced CSFAb42, increased CSF total tau, and reduced caudate dopamine transporter (DAT) uptake, on [123I]FP-CIT single-photon emission computed tomography, had a 65% risk of developing cognitive impairment (Yousaf et al., 2019). Other studies have reported similar findings, linking lower CSF Ab42, higher CSF total tau, and decreased striatal DAT uptake with cognitive decline (Alves et al., 2010; CaspellGarcia et al., 2017; Schrag et al., 2017). Changes in MRI markers, including volumetric and microstructural changes, have also been linked to cognitive decline and incorporated into predictive models as outlined later (Caspell-Garcia et al., 2017; Ray et al., 2018; Schulz et al., 2018). The association between these biological and neuroimaging variables with cognitive impairment suggests a possible pathophysiological link between total tau and amyloid accumulation and striatal dopaminergic dysfunction with cognitive decline in PD. Efforts to identify the earliest signs of cognitive impairment led to the classification of subjective cognitive impairment, which could predict future development of MCI in PD (Erro et al., 2014; Hong et al., 2014). Longitudinal studies in healthy controls have shown that the presence of subjective cognitive impairment was a predictor of the dementia, with about 10% of individuals developing MCI over 3 years (Jessen et al., 2010). In cognitively normal PD patients who reported subjective cognitive impairment at baseline who more rapid decline in semantic fluency and visuospatial memory tasks than cognitively normal PD patients who did not reported subjective cognitive impairment at baseline (Hong et al., 2014). Therefore, subjective cognitive impairment in cognitively normal PD patients could act as a predictor for future cognitive decline and progression to MCI. Moreover, subjective cognitive impairment opens new opportunities to study the earliest changes in vivo to better understand whether distinct pathologies may contribute to subjective cognitive impairment associated with future cognitive decline.

192

H. WILSON ET AL.

THE NUCLEUS BASALIS OF MEYNERT Structure and function of the nucleus basalis of Meynert The NBM was first described by Theodor Meynert in 1872 who described a group of magnocellular hyperchromic neurons in the basal forebrain. Early anatomical studies illustrated the NBM as a structure extending from the olfactory tubercle anteriorly to the level of the uncal hippocampus, reaching 13–14 mm in the sagittal plane (Mesulam and Geula, 1988). The NBM cholinergic neurons are located in the substantia innominate and along the ventral extent of the basal forebrain. Other cholinergic nuclei in the basal forebrain include the medial septum and vertical and horizontal limbs of the diagonal band of Broca. The function and connectivity of the NBM was developed in the 1970s with the proposal of the cholinergic hypothesis in AD (Bartus et al., 1982). At this time, the NBM was identified as the cholinergic center in the brain with neurons providing cholinergic inputs to the neocortex (Divac, 1975; Kievit and Kuypers, 1975; Mesulam et al., 1983; Mesulam and Geula, 1988). The basis of the cholinergic hypothesis is that neuronal loss within the NBM decreases the levels of cortical acetylcholine resulting in cognitive impairments. Mesulam and colleagues used histochemical and immunohistochemical labeling for acetylcholinesterase (AChE) and choline acetyltransferase to describe the four cholinergic cell groups, Ch1–Ch4, in the basal forebrain (Mesulam et al., 1983, 1984). The Ch1 is located in the medial septal nucleus and the Ch2 in the vertical limb of the diagonal band nucleus; both Ch1 and Ch2 project to the hippocampal complex. The Ch3 is located in the horizontal limb of the diagonal band nucleus with projections innervating the olfactory bulb. The Ch4, which can be further subdivided in six sectors, is located in the NBM with projections innervating the cortex and amygdala. Over 90% of the magnocellular neurons in the NBM are cholinergic with Ch4 making up the largest cholinergic groups in the basal forebrain. Other small neuronal populations that also exist in the NBM include GABAergic and galaninergic neurons (Mesulam and Geula, 1988; Gritti et al., 1993; Mufson et al., 2003). Based on topographical and connectivity properties, the anterior, intermediated, and posterior NBM can be subdivided into six sectors: the posterior sector (Ch4p), the intermediate subsector (Ch4i) which is separated into the inferior/ventral subsector (Ch4iv) and the superior/ dorsal subsector (Ch4id), the anterointermediate sector (Ch4ai), and the anterior sector (Ch4a) which is separated into the anteromedial subsector (Ch4am) and the anterolateral subsector (Ch4al). The posterior NBM is made up of the Ch4i and the Ch4p; the intermediate

NBM is made up of the Ch4al and the Ch4i; and the anterior NBM is composed of the Ch4am and the Ch4al (Liu et al., 2015). Projections from the anterior NBM mainly innervate the frontal and cingulate cortex with potential roles in executive dysfunction, while projections from the posterior NBM mainly innervate the temporal pole with potential roles in memory impairment. The intermediate NBM projects to the parietal and occipital cortex with possible implications for visuospatial dysfunction and visual hallucinations (Liu et al., 2015).

Pathological studies in Parkinson’s disease Lewy body pathology and neuronal loss was first established in the NBM of postmortem brain tissue from PD patients in 1913 by Friedrich Lewy (Schiller, 2000). Up to 80% neuronal loss in NBM has been reported in PD, with even greater loss in PDD brains (Candy et al., 1983; Perry et al., 1985; Chui et al., 1986). The level of neuronal loss in the NBM observed in PD is similar to or even greater than the NBM neuronal loss observed in AD postmortem data (Rogers et al., 1985). Neuronal loss observed in the NBM could reflect a combination of neuronal shrinkage due to atrophy as well as cell loss (Rinne et al., 1987; Swaab et al., 2002). In line with the early cholinergic hypothesis, neuronal loss in the NBM correlates strongly with cortical cholinergic deficits and the degree of cognitive impairment in both AD and PDD (Perry et al., 1985; Etienne et al., 1986; Gilmor et al., 1999). While both AD and PDD present with gradual loss of cholinergic tracts of the NBM, in PDD, this is mainly due to neuronal cell loss, while in AD, this is predominately due to axonal die back (Candy et al., 1983; Perry et al., 1985). AD-related pathology, including tau neurofibrillary tangles and b-amyloid plaques, can often coexist with Lewy bodies in PDD and DLB (Jellinger and Attems, 2015). The pathological interplay between these three pathologies has been suggested to play a role in the development of dementia (Compta et al., 2011; Irwin et al., 2013). However, the loss of cell bodies in the NBM has been reported to be in PD in the absence of coexisting tau neurofibrillary tangles in the cortex, hippocampus, or NBM (Candy et al., 1983; Gaspar and Gray, 1984; Nakano and Hirano, 1984). Therefore, alpha-synuclein pathology is likely to play a prominent role in NBM neuronal loss in PD in the absence of AD-related pathology. Braak’s histopathological staging of PD suggests that basal forebrain pathology occurs at the same stage as nigral pathology (Braak et al., 2003). The posterior NBM in PD appears to be relatively spared, indicating that pathology in the NBM may start in the anterior portion and progresses caudally to the posterior NBM as the

NUCLEUS BASALIS OF MEYNERT DEGENERATION PREDICTS COGNITIVE IMPAIRMENT disease advances with most prominent involvement of the posterior NBM in PDD (Liu et al., 2015). It has been hypothesized that the spread of pathology from the anterior NBM could explain the higher prevalence of a cognitive profile with executive dysfunction in PD, while memory problems are more pronounced in PDD. Furthermore, studies indicate that there is a slightly greater deficit in the intermediate NBM in PDD than in AD (Liu et al., 2015). The pathological mechanisms underlying cognitive decline in PD are likely complex and changes in the NBM should also be considered in the context of other affected systems and nuclei. Neuronal loss in the NBM has also been shown to occur in parallel with dopaminergic neuronal loss in the substantial nigra and ventral tegmental area and with noradrenergic neuronal loss in the locus coeruleus (Kehagia et al., 2010).

NEUROIMAGING TECHNIQUES TO STUDY COGNITIVE IMPAIRMENT AND THE ROLE OF THE NUCLEUS BASALIS OF MEYNERT IN PARKINSON’S DISEASE Positron emission tomography Molecular PET imaging is a noninvasive technique that uses positron emission to detect and quantify the distribution of a radioligand in vivo. A radioisotope is attached to a chemical compound of interest, which is known to bind to a specific target in the brain, to form a biological radioligand that can be injected intravenously into the subject (Townsend, 2008). The PET scanner detects the annihilation of photon pairs that are reconstructed to form a quantifiable image of the distribution of the target within the brain (Phelps, 2000; Innis et al., 2007; Gunn et al., 2011). Radioligands have been developed to quantify enzymes, receptors, and misfolded proteins within the brain (Perani et al., 2019). In PD, PET imaging has been employed to study the neurotransmitter system, enzymatic expression, neuroinflammation, glucose metabolism, and the accumulation of misfolded proteins (Politis, 2014; Yousaf et al., 2017; Wilson et al., 2019c). In relation to cognitive impairment in PD, PET studies have highlighted the role of enzymes such as phosphodiesterase 4 using [11C]Rolipram (Niccolini et al., 2017); glucose metabolism using [18F]FDG (Edison et al., 2013); microglia activation using [11C]PK11195 (Edison et al., 2013); astroglia activation using [11C] BU99008 (Wilson et al., 2019a); b-amyloid accumulation using [11C]PIB (Gomperts et al., 2012) and [18F] Florbetapir (Akhtar et al., 2017); tau accumulation using [18F]AV1451 (Gomperts et al., 2016); presynaptic dopaminergic dysfunction using [18F]DOPA (Bruck et al., 2001; Jokinen et al., 2009); postsynaptic dopaminergic dysfunction using [11C]FLB-457 (Christopher et al., 2014); and serotonergic dysfunction with [11C]DASB (Kotagal

193

et al., 2018). Cholinergic PET imaging targeting the cholinergic system has also been employed to study cognitive impairment in PD (Roy et al., 2016). This review focuses on the role of the NBM in predicting cognitive impairment in PD and discusses findings from PET studies of the cholinergic system and glucose metabolism within the context of the NBM.

GLUCOSE METABOLIC PET MOLECULAR IMAGING [18F]FDG PET has been employed to investigate the relationship between glucose metabolism and functional connectivity with cognitive profiles in PD. Hypometabolism in the occipito-parietal region, with lesser involvement of the frontal cortex, has been reported in PDD compared to controls (Gonzalez-Redondo et al., 2014). [18F]FDG PET in PD revealed positive correlation between hypometabolism in the frontal cortex with worse executive dysfunction, while visuospatial function impairment correlated with hypometabolism in the occipito-parietal region (Garcia-Garcia et al., 2012). Extrapolating these findings within the context of postmortem NBM studies, it could be suggested that executive dysfunction in PDD is related to Ch4a due to frontal and limbic cortical innervation from the anterior Ch4 area. Moreover, limbic cortical cholinergic denervation has been associated with anosmia which could be related to Ch4a pathology (Bohnen et al., 2010), and visuospatial impairment in PD could be related to neuronal loss in Ch4i. With improved segmentation of the NBM in vivo, multimodal imaging studies are warranted to further investigate the mechanistic interplay and the relationship between degeneration of the NBM subsectors and the presentation of specific cognitive profiles.

CHOLINERGIC PET MOLECULAR IMAGING Cholinergic PET imaging studies using radioligands to target the presynaptic acetylcholinesterase, vesicular acetylcholine transporter, and the postsynaptic nicotinic acetylcholine receptors have been conducted over the last 20 years to assess the role of cholinergic dysfunction across the spectrum of dementia disorders (Roy et al., 2016; Wilson et al., 2019c). Cholinergic PET studies, using radioligands targeting acetylcholinesterase, have reported significant cortical cholinergic deficits in PD and PDD patients suggesting that, in addition to the dopaminergic system, cholinergic dysfunction also contributes to cognitive impairment (Kuhl et al., 1996; Shinotoh et al., 1999; Hilker et al., 2005; Bohnen et al., 2006; Shimada et al., 2009). Therefore, these findings support the dual syndrome hypothesis in PD suggesting that dopaminergic deficits could play a more prominent role in executive dysfunction, while cholinergic deficits likely contribute toward visuospatial impairment

194

H. WILSON ET AL.

(Kehagia et al., 2013). Cholinergic projections to the cortex and thalamus have been found significantly reduced in a PET study using [11C]MP4A in PD patients and in thalamus in PDD compared to AD (Shinotoh et al., 1999; Kotagal et al., 2012). Bohnen and colleagues employed [11C]PMP PET imaging to compare cholinergic activity across PD, PDD, and mild AD (Bohnen et al., 2003). The mean cortical AChE activity was the lowest in PDD compared to cognitively normal PD and mild AD. Greatest loss of AChE activity in PDD has been reported in the temporo-parietal, left precentral region, and posterior cingulate (Hilker et al., 2005). The posterior–anterior gradient of cortical cholinergic deficits in PDD could be due to extensive cell loss in the Ch4i of the NBM affecting the cholinergic innervation to occipital–parietal cortical regions (Klein et al., 2010). Cholinergic dysfunction has also been reported in PD patients without dementia suggesting that cholinergic denervation may occur early in selected PD patients (Bohnen et al., 2012). Shimada and colleagues demonstrated that progressive cholinergic denervation is associated with dementia in PD (Shimada et al., 2009). Moreover, the presence of cognitive impairment in PD is associated with more severe cholinergic dysfunction compared to cognitively normal PD patients (Kuhl et al., 1996; Shinotoh et al., 1999; Bohnen et al., 2003; Hilker et al., 2005). This data supports postmortem studies providing evidence that basal forebrain degeneration starts early and progresses in line with the development of cognitive impairment (Ruberg et al., 1986). Longitudinal studies are warranted to provide insights into the utility of cholinergic PET to predict future development of cognitive impairment in PD.

Magnetic resonance imaging Structural and functional MRI techniques have been employed to investigate pathophysiological mechanisms underlying cognitive impairment in PD as well as in the identification of biomarkers to predict cognitive decline. Structural T1-weighted MRI sequences are utilized to study macrostructural changes related to gray matter subcortical volumetric and cortical thickness alterations (Sterling et al., 2016). Diffusion-weighted imaging (DWI) enables microstructural changes to be studied in vivo investigating the integrity of neurons (gray matter) and neural tracts (white matter) through measures of water diffusion (Basser et al., 1994). Diffusion tensor imaging (DTI) uses a technique that models the displacement of water molecules as a rotationally invariant tensor. Diffusion measures can be expressed as fractional anisotropy which describes the spatial integrity capturing the level of restriction imposed by the microstructural

environment and mean diffusivity which describes the overall molecular displacement (Soares et al., 2013). Both measures are based on the assumption that water molecules are less restricted due to disease processes such as axonal loss. Resting-state functional MRI is a technique that can be used to study brain connectivity and alterations of functional networks (Prodoehl et al., 2014). Arterial spin labeling is a measurement of tissue perfusion using with magnetically labeled arterial blood water protons as flow tracer, which allows for the cerebral blood flow to be calculated (Haller et al., 2016). MRI techniques such as susceptibility-weighted imaging and quantitative susceptibility mapping allow the detection and quantification of iron levels within the brain (Acosta-Cabronero et al., 2017). These MRI techniques have been employed to study pathological mechanisms involved in PD and utilized to investigate cognitive impairment in PD searching for noninvasive imaging markers to predict cognitive decline (Lanskey et al., 2018). This review focuses on the utility of structural T1-weighted MRI and DTI to study the role of the NBM in cognitive impairment in PD.

STRUCTURAL GRAY MATTER VOLUMETRIC MRI The association between gray matter volumetric loss, measured with structural T1-weighted MRI, and cognitive impairment was identified more than 20 years ago (Laakso et al., 1996). Localized cortical thinning is detectable starting from early stages of the disease with more widespread atrophy and subcortical volume loss as the disease advances (Wilson et al., 2019b). Moreover, cortical thinning in frontal, temporal, parietal and cingulate cortices, and hippocampal volumetric loss have been associated with cognitive decline in PD (Burton et al., 2004; Jubault et al., 2011; Ibarretxe-Bilbao et al., 2012; Pereira et al., 2012; Hanganu et al., 2013; Zarei et al., 2013; Segura et al., 2014; Yoon et al., 2014; Mak et al., 2015; Yildiz et al., 2015; Wilson et al., 2019b). Early onset of cognitive impairment has been correlated with faster progression of cortical atrophy in the supplementary motor area and temporal, occipital, and parietal regions, as well as atrophy in the limbic subcortical circuit in PD-MCI patients (Hanganu et al., 2014). PD-MCI compared with cognitively normal PD has been characterized by a higher degree of atrophy in the right middle frontal region; as cognitive decline progresses to dementia and a diagnosis of PDD, the pattern of cortical involvement extends to the parietal, temporal, insular, and prefrontal regions (Song et al., 2011). The severity of atrophy and cortical changes associated with the progression of cognitive decline has been proposed as a stepwise process with hypometabolism and atrophy representing consecutive stages

NUCLEUS BASALIS OF MEYNERT DEGENERATION PREDICTS COGNITIVE IMPAIRMENT (Gonzalez-Redondo et al., 2014). Taken together these studies suggest that the progression of cortical and subcortical atrophy is associated with cognitive decline and linked to underlying pathological mechanisms. Furthermore, structural changes have also been investigated as biomarkers to predict the onset and progression of cognitive impairment in PD (Caspell-Garcia et al., 2017). Structural MRI has been utilized to investigate changes in the NBM and the relationship with cognitive impairment in vivo with early studies focusing on AD and MCI and more recent studies focusing on changes in PD-related dementias. One challenge of in vivo imaging is the accurate segmentation of the NBM. Zaborszky and colleagues developed a probabilistic, cytoarchitectonic map of the Ch4 NBM in MNI standard space, which has been utilized to segment the NBM (Zaborszky et al., 2008). Combined postmortem histological sections with anatomical maps from postmortem proton density MRI identified reduced signal intensity in the NBM, likely reflecting degeneration of cholinergic neurons, in AD and MCI, which correlated with gray matter atrophy in the bilateral prefrontal cortex, inferior parietal cortex, and cingulate gyrus (Teipel et al., 2005). Using the probabilistic NBM map in MNI space (Zaborszky et al., 2008), significant atrophy of the NBM in MCI patients at high risk of developing AD has been reported (Grothe et al., 2010). Moreover, volume reductions in the posterior compartment of NBM and temporal lobe were associated with impaired delayed recall in MCI patients. In AD, reduced volume has been observed in the anteromedial, lateral, and posterior substantia innominate using structural MRI, which corresponds to the localization of cholinergic nuclei Ch2, Ch3, Ch4ai, Ch4i, and Ch4p using postmortem histological section analysis (Teipel et al., 2011). While the degree of atrophy was less pronounced in MCI, the regions affected were similar. Comparing MCI with AD, revealed atrophy of the Ch4al was mainly driven by the AD patients. Voxel-based morphometry studies have demonstrated reduction of the substantia innominate, reflecting degeneration of the NBM, in both AD and PDD patients (Hanyu et al., 2002, 2007). Furthermore, cognitively normal subjects with atrophy of the basal forebrain at baseline showed an increased risk to develop dementia over a 4-year follow-up period (Hely et al., 2008). Taken together these findings suggest that localized atrophy within the NBM occurs early and could precede the onset of dementia in at risk individuals (Schmitz et al., 2016). Early study investigated changes in the substantia innominate of the basal forebrain in PD used outcomes measures of length to demonstrate associations between substantia innominate atrophy and global cognition (Hanyu et al., 2002; Oikawa et al., 2004). Choi and

195

colleagues demonstrated reductions in the volume of the substantia innominate in PD with cognitive impairment with even greater decreases in PDD patients (Choi et al., 2012). Furthermore, volumetric loss of the substantia innominate was associated with global cognitive impairment as well as with deficits in attention, frontal executive dysfunction, language and visuospatial deficits, and verbal and visual memory impairments. Advances in stereotactic mapping have improved the ability to delineate the NBM among the other nuclei within the basal forebrain (Zaborszky et al., 2008; Kilimann et al., 2014). These stereotactic maps were derived from combined postmortem MRI with histological data. Ray and colleagues demonstrated atrophy of the posterior NBM subsector Ch4p, in drug-naïve PD patients at baseline predicted decline in global cognition over a 2-year follow-up (Ray et al., 2018). Atrophy in other nuclei of the basal forebrain, such as the Ch1 and Ch2, were not predictors of future cognitive decline. Patients with atrophy of the NBM as a whole at baseline had a 3.5-fold increased risk of development either MCI or PDD within 5 years compared to those without NBM atrophy at baseline. Furthermore, patients with NBM atrophy at baseline showed faster progression of decline on cognitive tests assessing immediate and delayed verbal recall and recognition, as well as previously learnt factual information and language. These findings support the hypothesis that a cognitive profile of memory and semantic processing deficits, which have been shown as predictors of later dementia (Williams-Gray et al., 2009), in PD arise from degeneration of the NBM (Gratwicke et al., 2015). The fact that the posterior NBM was associated with early global cognitive decline could suggest that pathology starts in the posterior NBM and spread to the whole nucleus as the disease progresses (Ray et al., 2018). However, this is not in line with postmortem analysis which suggests that the posterior NBM is relatively spared in PD (Liu et al., 2015). Therefore, further work is required to fully elucidate the spread of pathology within the subsector of the NBM. Moreover, the inclusion of additional clinical, imaging, biological, and genetic variables could help to refine the predictive model for future cognitive decline in PD.

MICROSTRUCTURAL WHITE MATTER DIFFUSION MRI White matter microstructural changes, assessed using DTI, have been described across the spectrum of cognitive impairment in PD (Hattori et al., 2012; Deng et al., 2013; Kamagata et al., 2013; Melzer et al., 2013; Agosta et al., 2014; Zheng et al., 2014; Duncan et al., 2016). White matter changes including reduced fractional anisotropy and increased mean diffusivity have been

196

H. WILSON ET AL.

shown to be present in PD with cognitive impairments prior to the detection of gray matter volumetric changes (Hattori et al., 2012; Agosta et al., 2014). PD patients with impaired semantic fluency, which is linked with dementia risk, had higher mean diffusivity of parietal and frontal subcortical tracts (Duncan et al., 2016). Moreover, increased white matter microstructural changes have been associated with the development of PD-related dementia (Gorges et al., 2019). Taken together, these findings suggest that white matter microstructural changes could be an earlier marker of cognitive involvement in PD, compared to gray matter atrophy measures, with mean diffusivity potentially serving as a predictive marker of cognitive impairment. With improved delineation of the NBM in vivo (Zaborszky et al., 2008; Kilimann et al., 2014), DTI has recently been employed to study microstructural changes in the NBM. We recently demonstrated that increased mean diffusivity, alongside gray matter volumetric loss, is present in the NBM of PD with cognitive impairment compared to cognitively normal PD patients (Fig. 12.1) (Schulz et al., 2018). PD patients with cognitive impairment also had increased mean diffusivity within the entorhinal cortex, which has prominent cholinergic innervation, in the absence of gray matter volumetric changes. Therefore, both structural gray matter volumetric changes and microstructural white matter alterations in cholinergic structures could contribute to cognitive impairment in PD. Given the importance of the NBM in cognitive function and findings that volumetric gray matter changes could hold predictive value for cognitive decline, we wanted to investigate whether microstructural damage observed in the NBM could be used as a predictive marker of future cognitive impairment (Schulz et al., 2018). Cognitively intact PD patients at baseline were followed for 36 months to identify variables at baseline which could be predictive of cognitive decline. Both structural volumetric gray matter loss and microstructural changes, increased mean diffusivity, in the NBM were predictive of cognitive decline in PD with degeneration of the NBM occurring prior to the development of cognitive impairment (Schulz et al., 2018). Given the influence of clinical and nonclinical variables as predictors for the development of cognitive impairment in PD (Williams-Gray et al., 2013; Schrag et al., 2017), we repeated the multivariate Cox survival analysis adjusting for age, depression, postural instability, APOE genotype group, olfactory dysfunction, motor symptom severity, the presence of rapid eye movement sleep behavior disorder, and the b-amyloid to tau CSF ratio. After controlled of these variables, mean diffusivity in the NBM remained a significant predictor of the development of cognitive impairment. Therefore, our findings provide evidence that microstructural

Fig. 12.1. Gray matter volume loss and microstructural changes in the nucleus basalis of Meynert in Parkinson’s disease. The anatomical location of the nucleus basalis of Meynert is highlighted in the coronal plane on a T1-weighted MRI brain in standard space using yellow crosshairs (top panel). Gray matter volumetric loss (middle panel) using voxelbased morphometry and increased mean diffusivity (bottom panel) using diffusion tensor imaging within the nucleus basalis of Meynert (blue color map) has been shown to predict the development of cognitive impairment in cognitively intact Parkinson’s disease patients (Schulz et al., 2018). Therefore, it highlights the potential utility of these in vivo techniques to predict cognitive decline through measurements of degeneration within the nucleus basalis of Meynert.

alterations within the NBM, occurring prior to the onset of cognitive impairment, is a strong predictor of the future development of cognitive impairment, which is independent on other clinical and nonclinical variables assessed in this study (Schulz et al., 2018). Furthermore, microstructural changes looked to precede structural volumetric loss of gray matter in the NBM; therefore, they acting as a potentially earlier biomarker in PD. Further work is required to elucidate how early damage occurs in the NBM.

NUCLEUS BASALIS OF MEYNERT DEGENERATION PREDICTS COGNITIVE IMPAIRMENT DTI has the potential to provide in vivo tractography maps of the NBM cortical connectivity. In AD, atrophy of the substantia innominate correlated with decreased integrity of intracortical projecting fiber tracts (Teipel et al., 2011). Hepp and colleagues found significant changes in mean diffusivity in the parieto-occipital projections of the NBM in PD patients with hallucinations (Hepp et al., 2017). Furthermore, tract-based spatial statistic methods have been utilized to assess the effects of basal forebrain atrophy on fiber tracts using high-resolution DTI (Teipel et al., 2011). Future highresolution DWI studies employing advanced tractography analysis could help to investigate the role of different subsectors of the NBM in vivo, their projecting tracts, and topographical innervation. Accurate subdivision of the NBM in vivo could also help to disentangle the spread of pathology from the anterior to posterior portion of the NBM and which subsector is affected first.

THERAPEUTIC INTERVENTIONS Currently, there is no disease modifying treatment for dementia or cognitive impairment available (Ballard et al., 2011). Limited therapeutic treatments, such as cholinesterase inhibitor and NMDA-receptor antagonists, are available which provide moderate symptomatic relief for cognitive deficits (Qaseem et al., 2008; Aarsland et al., 2009a; Emre et al., 2010). Given the degeneration observed in PDD, anticholinesterase medication, such as rivastigmine and galantamine, are used in clinical practice to help treat cognitive and neuropsychiatric symptoms in PD (Aarsland et al., 2004b). However, anticholinesterase medication often provides only modest benefits to patients with cognitive impairment (Raina et al., 2008; Bond et al., 2012). In individuals with subjective memory impairment, physical exercise has shown to be beneficial (Lautenschlager et al., 2008); however, further work is required to fully understand the mechanism underlying nonpharmacologic interventions (Daviglus et al., 2011). There is a need for a disease modifying therapy to slow and ultimate halt the progression of cognitive impairment. With the identification of reliable biomarkers to predict patients who will develop cognitive impairment, patients can be better stratified in clinical trials aiming to prevent the progression of PD to PDD. The potential for neuromodulatory treatments targeting the NBM, such as deep brain stimulation (DBS), is actively being investigated.

Deep brain stimulation DBS is an established treatment approach, to help alleviate symptoms, utilized across the spectrum of movement

197

disorders including PD, essential tremor, and dystonia (Deuschl et al., 2006; Williams et al., 2010; Vidailhet et al., 2013). The precise mechanisms underlying DBS are not well characterized. DBS is often classified as inhibitory or excitatory stimulation; however, it can have different effects depending on the neural tissue, for example, the soma is inhibited, while axons are excited (McIntyre et al., 2004). Therefore, it is likely that the overall network effect is the most critical (McIntyre and Hahn, 2010). In PD, DBS is currently performed for the subthalamic nuclei (STN) and the globus pallidus internal (GPi). While cognitive decline is widely reported in patients with STN DBS and GPi DBS, the presentation of cognitive decline and the cognitive profile after DBS varies across studies (Cernera et al., 2019). Studies directing comparing these two targets reported that the frequency of cognitive decline is higher in STN DBS compared with GPi DBS (Follett et al., 2010; Weaver et al., 2012; Odekerken et al., 2015; Boel et al., 2016). However, the precise pathways and pathophysiology underlying cognitive decline after DBS remain unclear. The NBM has been proposed as a next target for neuromodulation using DBS hoping to improve cognitive functions. Kurosawa and colleagues demonstrated that lowfrequency stimulation of the neurons in the NBM has an activating effect which could be utilized for therapeutic effects (Kurosawa et al., 1989). Bilateral artificial stimulation of the remaining efferent projections could enhance their cholinergic stimulation of cortical targets aiming to improve cognitive functions (Gratwicke et al., 2013). The investigation of the potential for lowfrequency stimulation of the neurons in the NBM to provide symptomatic relief in early animal models was restricted due to the lack of animal models for AD and PDD presenting with cholinergic deficits and NBM cell loss. The development of new transgenic models of AD and PDD showing both cholinergic pathology and cognitive deficits opened the way for animal studies investigating the potential use of low-frequency stimulation in dementia (Magen et al., 2012; Laursen et al., 2013). One study implanted a DBS system in the NBM of rats with and without cognitive deficits at baseline and provided unilateral low-frequency stimulations during a two-way active avoidance task, which assesses associative memory (Montero-Pastor et al., 2001, 2004). Lowfrequency stimulations of the NBM resulted in faster learning of the task, during the pretraining, which suggests improved memory acquisition, and enhanced performance on a test of retention, during posttraining, suggesting improved memory consolidation. In this study, memory consolidation was observed to be associated with the duration and amplitude of the DBS of the NBM. Furthermore, greatest improvements from

198

H. WILSON ET AL.

the NBM DBS were observed in rats who had cognitive deficits at baseline compared to those who were cognitively normal. The first application of NBM DBS in humans was in 1984. A 74-year-old man with moderate AD was implanted with a flexible electrode into the left NBM with low-frequency stimulations developed in cycles of 15 s followed by 12 min off (Turnbull et al., 1985). While no clinical improvement in cognitive function was observed, over a 6-month period, the stimulated side shows no progression in the decline of cortical metabolic activity compared to the unstimulated hemisphere. Therefore, it suggests that chronic stimulation of the NBM reversed depressed cortical glucose metabolism (Laxton and Lozano, 2013). In 2009, bilateral lowfrequency stimulation of the NBM with DBS was performed on a 71-year-old male patient with severe PDD who presented with significant impairment in immediate episodic memory and learning, long-term memory, and visual perceptual difficulties at baseline (Freund et al., 2009). DBS electrodes were implanted into the Ch4i subsector of the NBM. The Ch4i subsector was selected because it is the largest subsector of the NBM and has the most widespread cortical projections, therefore giving the potential to affect more cognitive domains. NBM DBS resulted in sustained improvement, over 2 months of constant stimulation, in immediate episodic memory, long-term memory, and visual perceptual abilities (Freund et al., 2009; Barnikol et al., 2010). Cognitive benefits were observed to be dependent on DBS of the NBM. Therefore, it indicates that the DBS of the NBM is safe in patients with advanced dementia and has the potential to improve cognitive abilities and patient’s quality of life. However, careful thought needs to be given to the ethical considerations and utility of NBM DBS in patients with advanced dementia, especially in patients lacking capacity to give informed consent. Moreover, the underlying mechanisms are still unclear and the long-term effects and efficacy of NBM DBS need to be better established. In PD patients, the possible combination of NBM DBS with DBS of the basal ganglia was recently investigated by Gratwicke and colleagues in a blinded randomized controlled trail of NBM stimulation (Gratwicke et al., 2018). Six PDD patients were implanted with bilateral DBS leads spanning the motor globus pallidus internus and the Ch4i subsector of the NBM. Patients showed improvements only in dyskinesias and the neuropsychiatric inventory, which was mostly driven by improvement in hallucinations, with stimulation on. Clinical trials in single-case studies are ongoing to establish the clinical efficacy of low-frequency NBM DBS in dementia and to

establish the specific cognitive profile and patient populations in which NBM DBS might show greater therapeutic benefits. Moreover, further work is required to identify the optimal time windows for intervention with NBM DBS before the neuronal loss is too severe. Reliable tools to monitor and predict the progression of NBM pathology and associated cognitive decline are warranted to help advance this technique. Given the subsectors of the NBM which efferent connections to specific targets, stimulation of distinct NBM projections will likely improve specific cognitive profiles (Gratwicke et al., 2013). Better methodology to dissect the anatomy of the NBM and its projections in vivo is warranted to help optimize and target NBM DBS as well as elucidate the underlying mechanisms. The lack of an immediate cognitive response to help guide programming of NBM DBS poses an additional challenge since cognitive effects are often more subtle and take longer to manifest. To this end, DWI and quantitative electroencephalography (QEEG) are being utilities to help identify detailed understanding of the NBM tracts in vivo and the physiological response of the NBM to stimulation during programming in real time, respectively (Kumbhare et al., 2018). Tractography maps from DWI could provide key advances in DBS of the NBM by providing more accurate segmentation of NBM subsectors connected to cortical targets with specific cognitive functions. Greater knowledge of the specific cortical area targeted by NBM DBS can also aid the selection of an appropriate outcome measure tailored to the corresponding cognitive function of the target. Furthermore, the absence of NBM cortical projections, identified using DWI, as a result of advanced neuronal loss could help to identify regions with insufficient levels of cholinergic cells for successful DBS outcomes (Kumbhare et al., 2018). While DWI-targeted guidance for the placement of the DBS could improve the anatomical targeting of the NBM, physiological confirmation for the correction placement of the DBS is still required. NBM-induced changes in EEG have been observed across a number of studies which correlated with specific cognitive functions (Metherate et al., 1992; Duque et al., 2000; McLin 3rd et al., 2002; Montero-Pastor et al., 2004). Therefore, QEEG has been proposed as a physiological biomarker to guide targeting, programming as well as serving as a secondary outcome measure of NBM DBS (Kumbhare et al., 2018). Endogenous NBM activity plays a key role in arousal, attention, and sleep cycle transitions (Yamakawa et al., 2016). The role of the NBM in the stages of sleep and the role of the stages of sleep on memory need to be carefully considered in further interventions modulating the NBM.

NUCLEUS BASALIS OF MEYNERT DEGENERATION PREDICTS COGNITIVE IMPAIRMENT

CONCLUSION AND FUTURE DIRECTIONS There is strong evidence to support the role of the NBM in the development of cognitive impairment in PD. Establishing better clinic-pathological associations especially in relation to the different subsectors of the NBM will aid our understanding of the different forms of cognitive decline and the role of forebrain cholinergic mechanisms in normal cognition and dementia. While a number of studies have proposed models to predict the risk of cognitive impairment in PD, they often present different predictor factors making the identification of a feasible and reliable model challenging. Recently, microstructural alterations within the NBM, which occur prior to the onset of cognitive impairment, has been identified using noninvasive DTI MRI, as a strong predictor of the future development of cognitive impairment in PD. Microstructural alterations within the NBM was independent of other clinical and nonclinical variables. Furthermore, microstructural changes looked to precede structural volumetric loss of gray matter in the NBM; therefore, it acts as a potentially earlier biomarker to predict cognitive decline. Further work is required to elucidate how early these microstructural changes occur in the NBM and to validate the utility of DTI to accurately predict PD patients who will progress to develop cognitive impairment. It would be interesting to explore whether changes in the NBM are present in cognitively normal PD patients with subjective cognitive impairment, which could help to identify early distinct patterns of pathologies associated with future cognitive decline. Longitudinal studies are warranted to provide insights into the utility of cholinergic PET imaging to predict the future development of cognitive impairment in PD. Moreover, subjective cognitive impairment opens new opportunities to study the earliest changes in vivo to better understand whether distinct pathologies may contribute to subjective cognitive impairment associated with future cognitive decline. The identification of an early biomarker of NBM degeneration in drug-naïve PD patients will improve stratification for clinical trials aimed at preventing progression to PD-MCI and PDD. There is a need for a disease modifying therapy to slow and ultimate halt the progression of cognitive impairment. While DBS of the NBM requires further work to fully understand the therapeutic effects in humans, the results over many decades support the use of NBM DBS as a potential neuromodulatory treatment for dementia with the potential to restore functionality of the NBM network.

199

REFERENCES Aarsland D, Ballard CG, Halliday G (2004a). Are Parkinson’s disease with dementia and dementia with Lewy bodies the same entity? J Geriatr Psychiatry Neurol 17: 137–145. Aarsland D, Mosimann UP, McKeith IG (2004b). Role of cholinesterase inhibitors in Parkinson’s disease and dementia with Lewy bodies. J Geriatr Psychiatry Neurol 17: 164–171. Aarsland D, Ballard C, Walker Z et al. (2009a). Memantine in patients with Parkinson’s disease dementia or dementia with Lewy bodies: a double-blind, placebo-controlled, multicentre trial. Lancet Neurol 8: 613–618. Aarsland D, Bronnick K, Larsen JP et al. (2009b). Cognitive impairment in incident, untreated Parkinson disease: the Norwegian ParkWest study. Neurology 72: 1121–1126. Aarsland D, Bronnick K, Williams-Gray C et al. (2010). Mild cognitive impairment in Parkinson disease: a multicenter pooled analysis. Neurology 75: 1062–1069. Aarsland D, Creese B, Politis M et al. (2017). Cognitive decline in Parkinson disease. Nat Rev Neurol 13: 217–231. Acosta-Cabronero J, Cardenas-Blanco A, Betts MJ et al. (2017). The whole-brain pattern of magnetic susceptibility perturbations in Parkinson’s disease. Brain 140: 118–131. Agosta F, Canu E, Stefanova E et al. (2014). Mild cognitive impairment in Parkinson’s disease is associated with a distributed pattern of brain white matter damage. Hum Brain Mapp 35: 1921–1929. Akhtar RS, Xie SX, Chen YJ et al. (2017). Regional brain amyloid-beta accumulation associates with domainspecific cognitive performance in Parkinson disease without dementia. PLoS One 12: e0177924. Albert MS, DeKosky ST, Dickson D et al. (2011). The diagnosis of mild cognitive impairment due to Alzheimer’s disease: recommendations from the National Institute on Aging-Alzheimer’s Association workgroups on diagnostic guidelines for Alzheimer’s disease. Alzheimers Dement 7: 270–279. Alves G, Bronnick K, Aarsland D et al. (2010). CSF amyloid-beta and tau proteins, and cognitive performance, in early and untreated Parkinson’s disease: the Norwegian ParkWest study. J Neurol Neurosurg Psychiatry 81: 1080–1086. Ashraghi MR, Pagano G, Polychronis S et al. (2016). Parkinson’s disease, diabetes and cognitive impairment. Recent Pat Endocr Metab Immune Drug Discovery 10: 11–21. Baba T, Kikuchi A, Hirayama K et al. (2012). Severe olfactory dysfunction is a prodromal symptom of dementia associated with Parkinson’s disease: a 3 year longitudinal study. Brain 135: 161–169. Ballard C, Gauthier S, Corbett A et al. (2011). Alzheimer’s disease. Lancet 377: 1019–1031. Barnikol TT, Pawelczyk NB, Barnikol UB et al. (2010). Changes in apraxia after deep brain stimulation of the nucleus basalis Meynert in a patient with Parkinson dementia syndrome. Mov Disord 25: 1519–1520.

200

H. WILSON ET AL.

Bartus RT, Dean 3rd RL, Beer B et al. (1982). The cholinergic hypothesis of geriatric memory dysfunction. Science 217: 408–414. Basser PJ, Mattiello J, LeBihan D (1994). MR diffusion tensor spectroscopy and imaging. Biophys J 66: 259–267. Boel JA, Odekerken VJ, Schmand BA et al. (2016). Cognitive and psychiatric outcome 3 years after globus pallidus pars interna or subthalamic nucleus deep brain stimulation for Parkinson’s disease. Parkinsonism Relat Disord 33: 90–95. Bohnen NI, Kaufer DI, Ivanco LS et al. (2003). Cortical cholinergic function is more severely affected in parkinsonian dementia than in Alzheimer disease: an in vivo positron emission tomographic study. Arch Neurol 60: 1745–1748. Bohnen NI, Kaufer DI, Hendrickson R et al. (2006). Cognitive correlates of cortical cholinergic denervation in Parkinson’s disease and parkinsonian dementia. J Neurol 253: 242–247. Bohnen NI, Muller ML, Kotagal V et al. (2010). Olfactory dysfunction, central cholinergic integrity and cognitive impairment in Parkinson’s disease. Brain 133: 1747–1754. Bohnen NI, Muller ML, Kotagal V et al. (2012). Heterogeneity of cholinergic denervation in Parkinson’s disease without dementia. J Cereb Blood Flow Metab 32: 1609–1617. Bond M, Rogers G, Peters J et al. (2012). The effectiveness and cost-effectiveness of donepezil, galantamine, rivastigmine and memantine for the treatment of Alzheimer’s disease (review of Technology Appraisal No. 111): a systematic review and economic model. Health Technol Assess 16: 1–470. Braak H, Del Tredici K, Rub U et al. (2003). Staging of brain pathology related to sporadic Parkinson’s disease. Neurobiol Aging 24: 197–211. Bruck A, Portin R, Lindell A et al. (2001). Positron emission tomography shows that impaired frontal lobe functioning in Parkinson’s disease is related to dopaminergic hypofunction in the caudate nucleus. Neurosci Lett 311: 81–84. Burton EJ, McKeith IG, Burn DJ et al. (2004). Cerebral atrophy in Parkinson’s disease with and without dementia: a comparison with Alzheimer’s disease, dementia with Lewy bodies and controls. Brain 127: 791–800. Candy JM, Perry RH, Perry EK et al. (1983). Pathological changes in the nucleus of Meynert in Alzheimer’s and Parkinson’s diseases. J Neurol Sci 59: 277–289. Caspell-Garcia C, Simuni T, Tosun-Turgut D et al. (2017). Multiple modality biomarker prediction of cognitive impairment in prospectively followed de novo Parkinson disease. PLoS One 12: e0175674. Cernera S, Okun MS, Gunduz AA (2019). Review of cognitive outcomes across movement disorder patients undergoing deep brain stimulation. Front Neurol 10: 419. Choi SH, Jung TM, Lee JE et al. (2012). Volumetric analysis of the substantia innominata in patients with Parkinson’s disease according to cognitive status. Neurobiol Aging 33: 1265–1272. Cholerton BA, Zabetian CP, Wan JY et al. (2014). Evaluation of mild cognitive impairment subtypes in Parkinson’s disease. Mov Disord 29: 756–764.

Christopher L, Marras C, Duff-Canning S et al. (2014). Combined insular and striatal dopamine dysfunction are associated with executive deficits in Parkinson’s disease with mild cognitive impairment. Brain 137: 565–575. Chui HC, Mortimer JA, Slager U et al. (1986). Pathologic correlates of dementia in Parkinson’s disease. Arch Neurol 43: 991–995. Compta Y, Marti MJ, Ibarretxe-Bilbao N et al. (2009). Cerebrospinal tau, phospho-tau, and beta-amyloid and neuropsychological functions in Parkinson’s disease. Mov Disord 24: 2203–2210. Compta Y, Parkkinen L, O’Sullivan SS et al. (2011). Lewy- and Alzheimer-type pathologies in Parkinson’s disease dementia: which is more important? Brain 134: 1493–1505. Daviglus ML, Plassman BL, Pirzada A et al. (2011). Risk factors and preventive interventions for Alzheimer disease: state of the science. Arch Neurol 68: 1185–1190. Deng B, Zhang Y, Wang L et al. (2013). Diffusion tensor imaging reveals white matter changes associated with cognitive status in patients with Parkinson’s disease. Am J Alzheimers Dis Other Demen 28: 154–164. Deuschl G, Schade-Brittinger C, Krack P et al. (2006). A randomized trial of deep-brain stimulation for Parkinson’s disease. N Engl J Med 355: 896–908. Divac I (1975). Magnocellular nuclei of the basal forebrain project to neocortex, brain stem, and olfactory bulb. Review of some functional correlates. Brain Res 93: 385–398. Duncan GW, Firbank MJ, Yarnall AJ et al. (2016). Gray and white matter imaging: a biomarker for cognitive impairment in early Parkinson’s disease? Mov Disord 31: 103–110. Duque A, Balatoni B, Detari L et al. (2000). EEG correlation of the discharge properties of identified neurons in the basal forebrain. J Neurophysiol 84: 1627–1635. Edison P, Ahmed I, Fan Z et al. (2013). Microglia, amyloid, and glucose metabolism in Parkinson’s disease with and without dementia. Neuropsychopharmacology 38: 938–949. Emre M, Tsolaki M, Bonuccelli U et al. (2010). Memantine for patients with Parkinson’s disease dementia or dementia with Lewy bodies: a randomised, double-blind, placebocontrolled trial. Lancet Neurol 9: 969–977. Erro R, Santangelo G, Barone P et al. (2014). Do subjective memory complaints herald the onset of mild cognitive impairment in Parkinson disease? J Geriatr Psychiatry Neurol 27: 276–281. Etienne P, Robitaille Y, Wood P et al. (1986). Nucleus basalis neuronal loss, neuritic plaques and choline acetyltransferase activity in advanced Alzheimer’s disease. Neuroscience 19: 1279–1291. Follett KA, Weaver FM, Stern M et al. (2010). Pallidal versus subthalamic deep-brain stimulation for Parkinson’s disease. N Engl J Med 362: 2077–2091. Foltynie T, Brayne CE, Robbins TW et al. (2004). The cognitive ability of an incident cohort of Parkinson’s patients in the UK. The CamPaIGN study. Brain 127: 550–560.

NUCLEUS BASALIS OF MEYNERT DEGENERATION PREDICTS COGNITIVE IMPAIRMENT Foltynie T, Lewis SG, Goldberg TE et al. (2005). The BDNF Val66Met polymorphism has a gender specific influence on planning ability in Parkinson’s disease. J Neurol 252: 833–838. Freund HJ, Kuhn J, Lenartz D et al. (2009). Cognitive functions in a patient with Parkinson-dementia syndrome undergoing deep brain stimulation. Arch Neurol 66: 781–785. Galtier I, Nieto A, Lorenzo JN et al. (2016). Mild cognitive impairment in Parkinson’s disease: diagnosis and progression to dementia. J Clin Exp Neuropsychol 38: 40–50. Garcia-Garcia D, Clavero P, Gasca Salas C et al. (2012). Posterior parietooccipital hypometabolism may differentiate mild cognitive impairment from dementia in Parkinson’s disease. Eur J Nucl Med Mol Imaging 39: 1767–1777. Gaspar P, Gray F (1984). Dementia in idiopathic Parkinson’s disease. A neuropathological study of 32 cases. Acta Neuropathol 64: 43–52. Geurtsen GJ, Hoogland J, Goldman JG et al. (2014). Parkinson’s disease mild cognitive impairment: application and validation of the criteria. J Parkinsons Dis 4: 131–137. Gilmor ML, Erickson JD, Varoqui H et al. (1999). Preservation of nucleus basalis neurons containing choline acetyltransferase and the vesicular acetylcholine transporter in the elderly with mild cognitive impairment and early Alzheimer’s disease. J Comp Neurol 411: 693–704. Goldman JG, Weis H, Stebbins G et al. (2012). Clinical differences among mild cognitive impairment subtypes in Parkinson’s disease. Mov Disord 27: 1129–1136. Goldman JG, Williams-Gray C, Barker RA et al. (2014). The spectrum of cognitive impairment in Lewy body diseases. Mov Disord 29: 608–621. Gomperts SN, Locascio JJ, Marquie M et al. (2012). Brain amyloid and cognition in Lewy body diseases. Mov Disord 27: 965–973. Gomperts SN, Locascio JJ, Makaretz SJ et al. (2016). Tau positron emission tomographic imaging in the Lewy body diseases. JAMA Neurol 73: 1334–1341. Gonzalez-Redondo R, Garcia-Garcia D, Clavero P et al. (2014). Grey matter hypometabolism and atrophy in Parkinson’s disease with cognitive impairment: a two-step process. Brain 137: 2356–2367. Gorges M, Muller HP, Liepelt-Scarfone I et al. (2019). Structural brain signature of cognitive decline in Parkinson’s disease: DTI-based evidence from the LANDSCAPE study. Ther Adv Neurol Disord 12: 1756286419843447. Gratwicke J, Kahan J, Zrinzo L et al. (2013). The nucleus basalis of Meynert: a new target for deep brain stimulation in dementia? Neurosci Biobehav Rev 37: 2676–2688. Gratwicke J, Jahanshahi M, Foltynie T (2015). Parkinson’s disease dementia: a neural networks perspective. Brain 138: 1454–1476. Gratwicke J, Zrinzo L, Kahan J et al. (2018). Bilateral deep brain stimulation of the nucleus basalis of meynert for Parkinson disease dementia: a randomized clinical trial. JAMA Neurol 75: 169–178.

201

Gritti I, Mainville L, Jones BE (1993). Codistribution of GABA- with acetylcholine-synthesizing neurons in the basal forebrain of the rat. J Comp Neurol 329: 438–457. Grothe M, Zaborszky L, Atienza M et al. (2010). Reduction of basal forebrain cholinergic system parallels cognitive impairment in patients at high risk of developing Alzheimer’s disease. Cereb Cortex 20: 1685–1695. Gunn RN, Guo Q, Salinas CA et al. (2011). Advances in biomathematical modeling for PET neuroreceptor imaging. Drug Discov Today Technol 8: e45–e51. Haller S, Zaharchuk G, Thomas DL et al. (2016). Arterial spin labeling perfusion of the brain: emerging clinical applications. Radiology 281: 337–356. Hanganu A, Bedetti C, Jubault T et al. (2013). Mild cognitive impairment in patients with Parkinson’s disease is associated with increased cortical degeneration. Mov Disord 28: 1360–1369. Hanganu A, Bedetti C, Degroot C et al. (2014). Mild cognitive impairment is linked with faster rate of cortical thinning in patients with Parkinson’s disease longitudinally. Brain 137: 1120–1129. Hanyu H, Asano T, Sakurai H et al. (2002). MR analysis of the substantia innominata in normal aging, Alzheimer disease, and other types of dementia. Am J Neuroradiol 23: 27–32. Hanyu H, Shimizu S, Tanaka Y et al. (2007). MR features of the substantia innominata and therapeutic implications in dementias. Neurobiol Aging 28: 548–554. Hattori T, Orimo S, Aoki S et al. (2012). Cognitive status correlates with white matter alteration in Parkinson’s disease. Hum Brain Mapp 33: 727–739. Hely MA, Reid WG, Adena MA et al. (2008). The Sydney multicenter study of Parkinson’s disease: the inevitability of dementia at 20 years. Mov Disord 23: 837–844. Hepp DH, Foncke EMJ, Berendse HW et al. (2017). Damaged fiber tracts of the nucleus basalis of Meynert in Parkinson’s disease patients with visual hallucinations. Sci Rep 7: 10112. Hilker R, Thomas AV, Klein JC et al. (2005). Dementia in Parkinson disease: functional imaging of cholinergic and dopaminergic pathways. Neurology 65: 1716–1722. Hobson P, Meara J (2015). Mild cognitive impairment in Parkinson’s disease and its progression onto dementia: a 16-year outcome evaluation of the Denbighshire cohort. Int J Geriatr Psychiatry 30: 1048–1055. Hong JY, Sunwoo MK, Chung SJ et al. (2014). Subjective cognitive decline predicts future deterioration in cognitively normal patients with Parkinson’s disease. Neurobiol Aging 35: 1739–1743. Hu MT, Szewczyk-Krolikowski K, Tomlinson P et al. (2014). Predictors of cognitive impairment in an early stage Parkinson’s disease cohort. Mov Disord 29: 351–359. Ibarretxe-Bilbao N, Junque C, Segura B et al. (2012). Progression of cortical thinning in early Parkinson’s disease. Mov Disord 27: 1746–1753. Innis RB, Cunningham VJ, Delforge J et al. (2007). Consensus nomenclature for in vivo imaging of reversibly binding radioligands. J Cereb Blood Flow Metab 27: 1533–1539.

202

H. WILSON ET AL.

Irwin DJ, Lee VM, Trojanowski JQ (2013). Parkinson’s disease dementia: convergence of alpha-synuclein, tau and amyloid-beta pathologies. Nat Rev Neurosci 14: 626–636. Jellinger KA, Attems J (2015). Challenges of multimorbidity of the aging brain: a critical update. J Neural Transm (Vienna) 122: 505–521. Jellinger KA, Korczyn AD (2018). Are dementia with Lewy bodies and Parkinson’s disease dementia the same disease? BMC Med 16: 34. Jessen F, Wiese B, Bachmann C et al. (2010). Prediction of dementia by subjective memory impairment: effects of severity and temporal association with cognitive impairment. Arch Gen Psychiatry 67: 414–422. Jokinen P, Bruck A, Aalto S et al. (2009). Impaired cognitive performance in Parkinson’s disease is related to caudate dopaminergic hypofunction and hippocampal atrophy. Parkinsonism Relat Disord 15: 88–93. Jubault T, Gagnon JF, Karama S et al. (2011). Patterns of cortical thickness and surface area in early Parkinson’s disease. Neuroimage 55: 462–467. Kalbe E, Rehberg SP, Heber I et al. (2016). Subtypes of mild cognitive impairment in patients with Parkinson’s disease: evidence from the LANDSCAPE study. J Neurol Neurosurg Psychiatry 87: 1099–1105. Kamagata K, Motoi Y, Tomiyama H et al. (2013). Relationship between cognitive impairment and white-matter alteration in Parkinson’s disease with dementia: tract-based spatial statistics and tract-specific analysis. Eur Radiol 23: 1946–1955. Kehagia AA, Barker RA, Robbins TW (2010). Neuropsychological and clinical heterogeneity of cognitive impairment and dementia in patients with Parkinson’s disease. Lancet Neurol 9: 1200–1213. Kehagia AA, Barker RA, Robbins TW (2013). Cognitive impairment in Parkinson’s disease: the dual syndrome hypothesis. Neurodegener Dis 11: 79–92. Kievit J, Kuypers HG (1975). Basal forebrain and hypothalamic connection to frontal and parietal cortex in the Rhesus monkey. Science 187: 660–662. Kilimann I, Grothe M, Heinsen H et al. (2014). Subregional basal forebrain atrophy in Alzheimer’s disease: a multicenter study. J Alzheimers Dis 40: 687–700. Klein JC, Eggers C, Kalbe E et al. (2010). Neurotransmitter changes in dementia with Lewy bodies and Parkinson disease dementia in vivo. Neurology 74: 885–892. Knopman DS, Petersen RC (2014). Mild cognitive impairment and mild dementia: a clinical perspective. Mayo Clin Proc 89: 1452–1459. Kotagal V, Muller ML, Kaufer DI et al. (2012). Thalamic cholinergic innervation is spared in Alzheimer disease compared to parkinsonian disorders. Neurosci Lett 514: 169–172. Kotagal V, Spino C, Bohnen NI et al. (2018). Serotonin, betaamyloid, and cognition in Parkinson disease. Ann Neurol 83: 994–1002. Kuhl DE, Minoshima S, Fessler JA et al. (1996). In vivo mapping of cholinergic terminals in normal aging, Alzheimer’s disease, and Parkinson’s disease. Ann Neurol 40: 399–410.

Kumbhare D, Palys V, Toms J et al. (2018). Nucleus basalis of meynert stimulation for dementia: theoretical and technical considerations. Front Neurosci 12: 614. Kurosawa M, Sato A, Sato Y (1989). Stimulation of the nucleus basalis of Meynert increases acetylcholine release in the cerebral cortex in rats. Neurosci Lett 98: 45–50. Laakso MP, Partanen K, Riekkinen P et al. (1996). Hippocampal volumes in Alzheimer’s disease, Parkinson’s disease with and without dementia, and in vascular dementia: an MRI study. Neurology 46: 678–681. Lanskey JH, McColgan P, Schrag AE et al. (2018). Can neuroimaging predict dementia in Parkinson’s disease? Brain 141: 2545–2560. Laursen B, Mork A, Plath N et al. (2013). Cholinergic degeneration is associated with increased plaque deposition and cognitive impairment in APPswe/PS1dE9 mice. Behav Brain Res 240: 146–152. Lautenschlager NT, Cox KL, Flicker L et al. (2008). Effect of physical activity on cognitive function in older adults at risk for Alzheimer disease: a randomized trial. JAMA 300: 1027–1037. Laxton AW, Lozano AM (2013). Deep brain stimulation for the treatment of Alzheimer disease and dementias. World Neurosurg 80: S28.e1–S28.e8. Leroi I, McDonald K, Pantula H et al. (2012). Cognitive impairment in Parkinson disease: impact on quality of life, disability, and caregiver burden. J Geriatr Psychiatry Neurol 25: 208–214. Litvan I, Goldman JG, Troster AI et al. (2012). Diagnostic criteria for mild cognitive impairment in Parkinson’s disease: Movement Disorder Society Task Force guidelines. Mov Disord 27: 349–356. Liu AK, Chang RC, Pearce RK et al. (2015). Nucleus basalis of Meynert revisited: anatomy, history and differential involvement in Alzheimer’s and Parkinson’s disease. Acta Neuropathol 129: 527–540. Liu G, Locascio JJ, Corvol JC et al. (2017). Prediction of cognition in Parkinson’s disease with a clinical-genetic score: a longitudinal analysis of nine cohorts. Lancet Neurol 16: 620–629. Magen I, Fleming SM, Zhu C et al. (2012). Cognitive deficits in a mouse model of pre-manifest Parkinson’s disease. Eur J Neurosci 35: 870–882. Mak E, Su L, Williams GB et al. (2015). Baseline and longitudinal grey matter changes in newly diagnosed Parkinson’s disease: ICICLE-PD study. Brain 138: 2974–2986. McIntyre CC, Hahn PJ (2010). Network perspectives on the mechanisms of deep brain stimulation. Neurobiol Dis 38: 329–337. McIntyre CC, Grill WM, Sherman DL et al. (2004). Cellulareffects of deep brain stimulation: model-based analysis of activation and inhibition. J Neurophysiol 91: 1457–1469. McKeith IG, Boeve BF, Dickson DW et al. (2017). Diagnosis and management of dementia with Lewy bodies: fourth consensus report of the DLB Consortium. Neurology 89: 88–100.

NUCLEUS BASALIS OF MEYNERT DEGENERATION PREDICTS COGNITIVE IMPAIRMENT McKhann GM, Knopman DS, Chertkow H et al. (2011). The diagnosis of dementia due to Alzheimer’s disease: recommendations from the National Institute on AgingAlzheimer’s Association workgroups on diagnostic guidelines for Alzheimer’s disease. Alzheimers Dement 7: 263–269. McLin 3rd DE, Miasnikov AA, Weinberger NM (2002). The effects of electrical stimulation of the nucleus basalis on the electroencephalogram, heart rate, and respiration. Behav Neurosci 116: 795–806. Meireles J, Massano J (2012). Cognitive impairment and dementia in Parkinson’s disease: clinical features, diagnosis, and management. Front Neurol 3: 88. Melzer TR, Watts R, MacAskill MR et al. (2013). White matter microstructure deteriorates across cognitive stages in Parkinson disease. Neurology 80: 1841–1849. Mesulam MM, Geula C (1988). Nucleus basalis (Ch4) and cortical cholinergic innervation in the human brain: observations based on the distribution of acetylcholinesterase and choline acetyltransferase. J Comp Neurol 275: 216–240. Mesulam MM, Mufson EJ, Levey AI et al. (1983). Cholinergic innervation of cortex by the basal forebrain: cytochemistry and cortical connections of the septal area, diagonal band nuclei, nucleus basalis (substantia innominata), and hypothalamus in the rhesus monkey. J Comp Neurol 214: 170–197. Mesulam MM, Mufson EJ, Levey AI et al. (1984). Atlas of cholinergic neurons in the forebrain and upper brainstem of the macaque based on monoclonal choline acetyltransferase immunohistochemistry and acetylcholinesterase histochemistry. Neuroscience 12: 669–686. Metherate R, Cox CL, Ashe JH (1992). Cellular bases of neocortical activation: modulation of neural oscillations by the nucleus basalis and endogenous acetylcholine. J Neurosci 12: 4701–4711. Mok W, Chow TW, Zheng L et al. (2004). Clinicopathological concordance of dementia diagnoses by community versus tertiary care clinicians. Am J Alzheimers Dis Other Demen 19: 161–165. Mollenhauer B, Trenkwalder C, von Ahsen N et al. (2006). Beta-amlyoid 1–42 and tau-protein in cerebrospinal fluid of patients with Parkinson’s disease dementia. Dement Geriatr Cogn Disord 22: 200–208. Montero-Pastor A, Vale-Martinez A, Guillazo-Blanch G et al. (2001). Nucleus basalis magnocellularis electrical stimulation facilitates two-way active avoidance retention, in rats. Brain Res 900: 337–341. Montero-Pastor A, Vale-Martinez A, Guillazo-Blanch G et al. (2004). Effects of electrical stimulation of the nucleus basalis on two-way active avoidance acquisition, retention, and retrieval. Behav Brain Res 154: 41–54. Morley JF, Xie SX, Hurtig HI et al. (2012). Genetic influences on cognitive decline in Parkinson’s disease. Mov Disord 27: 512–518. Moustafa AA, Poletti M (2013). Neural and behavioral substrates of subtypes of Parkinson’s disease. Front Syst Neurosci 7: 117.

203

Mufson EJ, Ginsberg SD, Ikonomovic MD et al. (2003). Human cholinergic basal forebrain: chemoanatomy and neurologic dysfunction. J Chem Neuroanat 26: 233–242. Nakano I, Hirano A (1984). Parkinson’s disease: neuron loss in the nucleus basalis without concomitant Alzheimer’s disease. Ann Neurol 15: 415–418. Niccolini F, Wilson H, Pagano G et al. (2017). Loss of phosphodiesterase 4 in Parkinson disease: relevance to cognitive deficits. Neurology 89: 586–593. Odekerken VJ, Boel JA, Geurtsen GJ et al. (2015). Neuropsychological outcome after deep brain stimulation for Parkinson disease. Neurology 84: 1355–1361. Oikawa H, Sasaki M, Ehara S et al. (2004). Substantia innominata: MR findings in Parkinson’s disease. Neuroradiology 46: 817–821. Pagano G, Polychronis S, Wilson H et al. (2018). Diabetes mellitus and Parkinson disease. Neurology 90: e1654–e1662. Perani D, Iaccarino L, Jacobs AH et al. (2019). Application of advanced brain positron emission tomography-based molecular imaging for a biological framework in neurodegenerative proteinopathies. Alzheimers Dement (Amst) 11: 327–332. Pereira JB, Ibarretxe-Bilbao N, Marti MJ et al. (2012). Assessment of cortical degeneration in patients with Parkinson’s disease by voxel-based morphometry, cortical folding, and cortical thickness. Hum Brain Mapp 33: 2521–2534. Perry EK, Curtis M, Dick DJ et al. (1985). Cholinergic correlates of cognitive impairment in Parkinson’s disease: comparisons with Alzheimer’s disease. J Neurol Neurosurg Psychiatry 48: 413–421. Phelps ME (2000). Positron emission tomography provides molecular imaging of biological processes. Proc Natl Acad Sci U S A 97: 9226–9233. Poletti M, Frosini D, Pagni C et al. (2012). Mild cognitive impairment and cognitive-motor relationships in newly diagnosed drug-naive patients with Parkinson’s disease. J Neurol Neurosurg Psychiatry 83: 601–606. Politis M (2014). Neuroimaging in Parkinson disease: from research setting to clinical practice. Nat Rev Neurol 10: 708–722. Politis M, Wu K, Molloy S et al. (2010). Parkinson’s disease symptoms: the patient’s perspective. Mov Disord 25: 1646–1651. Prodoehl J, Burciu RG, Vaillancourt DE (2014). Resting state functional magnetic resonance imaging in Parkinson’s disease. Curr Neurol Neurosci Rep 14: 448. Qaseem A, Snow V, Cross Jr. JT et al. (2008). Current pharmacologic treatment of dementia: a clinical practice guideline from the American College of Physicians and the American Academy of Family Physicians. Ann Intern Med 148: 370–378. Raina P, Santaguida P, Ismaila A et al. (2008). Effectiveness of cholinesterase inhibitors and memantine for treating dementia: evidence review for a clinical practice guideline. Ann Intern Med 148: 379–397.

204

H. WILSON ET AL.

Ray NJ, Bradburn S, Murgatroyd C et al. (2018). In vivo cholinergic basal forebrain atrophy predicts cognitive decline in de novo Parkinson’s disease. Brain 141: 165–176. Rinne JO, Paljarvi L, Rinne UK (1987). Neuronal size and density in the nucleus basalis of Meynert in Alzheimer’s disease. J Neurol Sci 79: 67–76. Rogers JD, Brogan D, Mirra SS (1985). The nucleus basalis of Meynert in neurological disease: a quantitative morphological study. Ann Neurol 17: 163–170. Roy R, Niccolini F, Pagano G et al. (2016). Cholinergic imaging in dementia spectrum disorders. Eur J Nucl Med Mol Imaging 43: 1376–1386. Ruberg M, Rieger F, Villageois A et al. (1986). Acetylcholinesterase and butyrylcholinesterase in frontal cortex and cerebrospinal fluid of demented and nondemented patients with Parkinson’s disease. Brain Res 362: 83–91. Schiller F (2000). Fritz Lewy and his bodies. J Hist Neurosci 9: 148–151. Schinka JA, Loewenstein DA, Raj A et al. (2010). Defining mild cognitive impairment: impact of varying decision criteria on neuropsychological diagnostic frequencies and correlates. Am J Geriatr Psychiatry 18: 684–691. Schmitz TW, Nathan Spreng R, Alzheimer’s Disease Neuroimaging Initiative (2016). Basal forebrain degeneration precedes and predicts the cortical spread of Alzheimer’s pathology. Nat Commun 7: 13249. Schrag A, Siddiqui UF, Anastasiou Z et al. (2017). Clinical variables and biomarkers in prediction of cognitive impairment in patients with newly diagnosed Parkinson’s disease: a cohort study. Lancet Neurol 16: 66–75. Schulz J, Pagano G, Fernandez Bonfante JA et al. (2018). Nucleus basalis of Meynert degeneration precedes and predicts cognitive impairment in Parkinson’s disease. Brain 141: 1501–1516. Segura B, Baggio HC, Marti MJ et al. (2014). Cortical thinning associated with mild cognitive impairment in Parkinson’s disease. Mov Disord 29: 1495–1503. Shimada H, Hirano S, Shinotoh H et al. (2009). Mapping of brain acetylcholinesterase alterations in Lewy body disease by PET. Neurology 73: 273–278. Shinotoh H, Namba H, Yamaguchi M et al. (1999). Positron emission tomographic measurement of acetylcholinesterase activity reveals differential loss of ascending cholinergic systems in Parkinson’s disease and progressive supranuclear palsy. Ann Neurol 46: 62–69. Soares JM, Marques P, Alves V et al. (2013). A hitchhiker’s guide to diffusion tensor imaging. Front Neurosci 7: 31. Sollinger AB, Goldstein FC, Lah JJ et al. (2010). Mild cognitive impairment in Parkinson’s disease: subtypes and motor characteristics. Parkinsonism Relat Disord 16: 177–180. Song SK, Lee JE, Park HJ et al. (2011). The pattern of cortical atrophy in patients with Parkinson’s disease according to cognitive status. Mov Disord 26: 289–296. Sterling NW, Lewis MM, Du G et al. (2016). Structural imaging and Parkinson’s disease: moving toward quantitative markers of disease progression. J Parkinsons Dis 6: 557–567.

Sulena GD, Sharma AK, Singh B (2017). Clinical profile of dysphagia in patients with Parkinson’s disease, progressive supranuclear palsy and multiple system atrophy. J Assoc Physicians India 65: 32–37. Swaab DF, Dubelaar EJ, Hofman MA et al. (2002). Brain aging and Alzheimer’s disease; use it or lose it. Prog Brain Res 138: 343–373. Teipel SJ, Flatz WH, Heinsen H et al. (2005). Measurement of basal forebrain atrophy in Alzheimer’s disease using MRI. Brain 128: 2626–2644. Teipel SJ, Meindl T, Grinberg L et al. (2011). The cholinergic system in mild cognitive impairment and Alzheimer’s disease: an in vivo MRI and DTI study. Hum Brain Mapp 32: 1349–1362. Townsend DW (2008). Positron emission tomography/computed tomography. Semin Nucl Med 38: 152–166. Turnbull IM, McGeer PL, Beattie L et al. (1985). Stimulation of the basal nucleus of Meynert in senile dementia of Alzheimer’s type. A preliminary report. Appl Neurophysiol 48: 216–221. Vidailhet M, Jutras MF, Grabli D et al. (2013). Deep brain stimulation for dystonia. J Neurol Neurosurg Psychiatry 84: 1029–1042. Vossius C, Larsen JP, Janvin C et al. (2011). The economic impact of cognitive impairment in Parkinson’s disease. Mov Disord 26: 1541–1544. Walker Z, Possin KL, Boeve BF et al. (2015). Lewy body dementias. Lancet 386: 1683–1697. Watson GS, Leverenz JB (2010). Profile of cognitive impairment in Parkinson’s disease. Brain Pathol 20: 640–645. Weaver FM, Follett KA, Stern M et al. (2012). Randomized trial of deep brain stimulation for Parkinson disease: thirty-six-month outcomes. Neurology 79: 55–65. 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. Williams-Gray CH, Hampshire A, Barker RA et al. (2008). Attentional control in Parkinson’s disease is dependent on COMT val 158 met genotype. Brain 131: 397–408. Williams-Gray CH, Evans JR, Goris A et al. (2009). The distinct cognitive syndromes of Parkinson’s disease: 5 year follow-up of the CamPaIGN cohort. Brain 132: 2958–2969. Williams-Gray CH, Mason SL, Evans JR et al. (2013). The CamPaIGN study of Parkinson’s disease: 10-year outlook in an incident population-based cohort. J Neurol Neurosurg Psychiatry 84: 1258–1264. Wilson H, Dervenoulas G, Pagano G et al. (2019a). Imidazoline 2 binding sites reflecting astroglia pathology in Parkinson’s disease: an in vivo11C-BU99008 PET study. Brain 142: 3116–3128. Wilson H, Niccolini F, Pellicano C et al. (2019b). Cortical thinning across Parkinson’s disease stages and clinical correlates. J Neurol Sci 398: 31–38.

NUCLEUS BASALIS OF MEYNERT DEGENERATION PREDICTS COGNITIVE IMPAIRMENT Wilson H, Pagano G, Politis M (2019c). Dementia spectrum disorders: lessons learnt from decades with PET research. J Neural Transm (Vienna) 126: 233–251. Wilson H, Pagano G, Yousaf T et al. (2020). Predict cognitive decline with clinical markers in Parkinson’s disease (PRECODE-1). J Neural Transm (Vienna) 127: 51–59. Wojtala J, Heber IA, Neuser P et al. (2019). Cognitive decline in Parkinson’s disease: the impact of the motor phenotype on cognition. J Neurol Neurosurg Psychiatry 90: 171–179. Yamakawa GR, Basu P, Cortese F et al. (2016). The cholinergic forebrain arousal system acts directly on the circadian pacemaker. Proc Natl Acad Sci U S A 113: 13498–13503. Yildiz D, Erer S, Zarifoglu M et al. (2015). Impaired cognitive performance and hippocampal atrophy in Parkinson disease. Turk J Med Sci 45: 1173–1177. Yoon JH, Lee JE, Yong SW et al. (2014). The mild cognitive impairment stage of dementia with Lewy bodies and Parkinson disease: a comparison of cognitive profiles. Alzheimer Dis Assoc Disord 28: 151–155.

205

Yousaf T, Wilson H, Politis M (2017). Imaging the nonmotor symptoms in Parkinson’s disease. Int Rev Neurobiol 133: 179–257. Yousaf T, Pagano G, Niccolini F et al. (2019). Predicting cognitive decline with non-clinical markers in Parkinson’s disease (PRECODE-2). J Neurol 266: 1203–1210. Yu RL, Wu RM, Tai CH et al. (2012). Neuropsychological profile in patients with early stage of Parkinson’s disease in Taiwan. Parkinsonism Relat Disord 18: 1067–1072. Zaborszky L, Hoemke L, Mohlberg H et al. (2008). Stereotaxic probabilistic maps of the magnocellular cell groups in human basal forebrain. Neuroimage 42: 1127–1141. Zarei M, Ibarretxe-Bilbao N, Compta Y et al. (2013). Cortical thinning is associated with disease stages and dementia in Parkinson’s disease. J Neurol Neurosurg Psychiatry 84: 875–881. Zheng Z, Shemmassian S, Wijekoon C et al. (2014). DTI correlates of distinct cognitive impairments in Parkinson’s disease. Hum Brain Mapp 35: 1325–1333.

This page intentionally left blank

Handbook of Clinical Neurology, Vol. 179 (3rd series) The Human Hypothalamus: Anterior Region D.F. Swaab, F. Kreier, P.J. Lucassen, A. Salehi, and R.M. Buijs, Editors https://doi.org/10.1016/B978-0-12-819975-6.00011-X Copyright © 2021 Elsevier B.V. All rights reserved

Chapter 13

Enlargement of early endosomes and traffic jam in basal forebrain cholinergic neurons in Alzheimer’s disease ATOOSSA FAHIMI1, MAHJABIN NOROOZI2, AND AHMAD SALEHI3* 1

Children’s Hospital Los Angeles, University of Southern California, Los Angeles, CA, United States

2

Cancer Clinical Trials Office, Stanford Medicine, Stanford University School of Medicine, Palo Alto, CA, United States 3

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

Abstract While a handful of neurotransmitter systems including cholinergic, norepinephrinergic, and serotonergic undergo significant degeneration in Alzheimer’s disease, the cholinergic system has been the prime target for research and therapy. The cholinergic system in the basal forebrain is strategically located to impose significant modulatory effects on vast cortical and subcortical regions of the brain. Numerous studies have established a strong link between neurotrophin signaling and basal forebrain cholinergic neuron degeneration in several neurodegenerative disorders. Evidence presented during the last few years points to the effects of endosomal pathology and primarily unidirectional traffic jam. Hence, formulating new therapies, e.g., to reduce local production of b C-terminal fragments and preventing changes in endosomal morphology have become attractive potential therapeutic strategies to restore cholinergic neurons and their neuromodulatory function. While it is not expected that restoring the cholinergic system function will fully mitigate cognitive dysfunction in Alzheimer’s disease, pivotal aspects of cognition including attentiondeficit during the prodromal stages might well be at disposal for corrective measures.

INTRODUCTION The cholinergic signaling plays a pivotal role in both structural and functional aspects of spatial memory, i.e., synaptic plasticity and long-term potentiation (LTP) (Hampel et al., 2018). Both cortical and subcortical regions of the brain receive massive cholinergic inputs from basal forebrain cholinergic neurons (BFCNs). These neurons are primarily aggregated in the medial septum, vertical and horizontal limbs of the diagonal band of Broca (DBB), and the nucleus basalis of Meynert (NBM). While the medial septum and DBB project primarily to the hippocampus, parahippocampus, and the olfactory bulb (Knox and Keller, 2016), NBM neurons are the source of a majority of neuromodulatory inputs to the cortex. The massive collateral branching of BFCNs

enables them to impose strong modulatory influence on their projected regions. In fact, a single human BFCN could generate more than 100 m of axonal branching (Wu et al., 2014). While extensive projections by BFCNs equip them with the ability to induce modulatory influence on vast cortical and subcortical regions, which comes with a cost, i.e., high vulnerability to neurodegeneration. As a result, BFCNs undergo significant degeneration in age-related neurodegenerative disorders, particularly Alzheimer’s disease (AD).

BASAL FOREBRAIN CHOLINERGIC NEURONS DEGENERATION While BFCN degeneration is not unique to AD, almost all cases of AD show some degree of cholinergic

*Correspondence to: Ahmad Salehi, MD, PhD, Adjunct Prof, Psychiatry and Behavioral Sciences, Stanford Medical School, Stanford, CA, United States. Tel: +1-650-743-8126, Fax: +1-925-480-3101, E-mail: [email protected]

208

A. FAHIMI ET AL.

degeneration (Salehi et al., 2004) detected even in the prodromal stages of the disease (Grothe et al., 2012). Accordingly, a direct positive correlation has been detected between cognitive functions, assessed by the mini-mental state examination (MMSE) and BFCN volume in individuals with mild cognitive impairment (MCI) (Drachman and Leavitt, 1974). One might question what aspects of cognitive function are modulated by the cholinergic system. Impaired attention is considered a crucial aspect of AD symptomatology. Importantly, cholinesterase inhibitors that reduce acetylcholine (ACh) breakdown, stabilize attention performance in AD even though memory performance continues to decline (Bracco et al., 2014). Understanding the pathogenesis of BFCN degeneration in AD has been the subject of intense study for the past many years (see Salehi et al., 2007). Several factors including poor myelination, long axons, failed trophic factor signaling, high metabolic rate, and the extensive connections with brain regions with abundant AD pathology have all been suggested to play a role in the heightened vulnerability of BFCNs in AD. Here, we will focus on possible extrinsic (e.g., iatrogenic) and intrinsic (e.g., failed nerve growth factor (NGF)) elements causing BFCN degeneration in AD (Fig. 13.1).

POSSIBLE IATROGENIC MECHANISMS OF CHOLINERGIC DEGENERATION High vulnerability of BFCNs to neurodegeneration has been compounded by the fact that a number of relatively popular drugs demonstrate anticholinergic (AC) properties. Several first-generation antihistamines, tricyclic antidepressants, analgesics, inhaled bronchodilators, muscle relaxants, and psychotropic drugs have all shown to alter cholinergic transmission. Ironically, AC drugs are commonly prescribed for the treatment of conditions like overactive bladder, affecting millions of older persons worldwide. The popularity of AC drugs is shown by the fact that in 2017, they were prescribed more than 25 million times in the United States. The long-term use of AC drugs has been linked to several AD-related abnormalities, including cortical thinning in the posterior cingulate gyrus and middle frontal gyrus, elevated amyloid-beta (Ab) levels, leading to an increased risk of dementia (Gray and Hanlon, 2018). A number of benzodiazepines display AC activity. It has been shown that these drugs can increase the risk of dementia by 50% (odds ratio; 1:51; Billioti de Gage et al., 2014). In a large-scale study in more than 100,000 individuals, a significant correlation was found between the use of AC drugs and the risk of dementia (Richardson et al., 2018). Unfortunately, the use of drugs with AC activity in the elderly, possibly suffering from dementia, is common. For instance, Sura

et al. (2013) reported that 1 out of 5 patients with dementia is likely to use drugs with significant AC activity for symptoms including mood disorders and urinary incontinence. In a large-population study in people 65 years and older, those taking drugs with AC properties seem to be at increased risk for cognitive decline and dementia (Carrière et al., 2009). The central nervous system (CNS)-related adverse effects associated with the prescription of AC drugs in older adults include memory impairment, confusion, and hallucinations. Importantly, it appears that the increased risk of developing dementia of AD type is a dose-dependent phenomenon (Gray et al., 2015). In addition, drugs that directly target the cholinergic system also induce AD-like symptomatology and prevent hippocampal activation. For instance, scopolamine, i.e., a competitive inhibitor of muscarinic receptors used for prophylactic treatment of motion sickness can cause confusion and/or delirium. Importantly, this drug has been linked to brain atrophy, reducing dendritic span, and Ab accumulation (Joseph et al., 2020). The question of whether drugs with AC activity (directly or indirectly) can induce AD-linked abnormalities or they just elevate BFCN vulnerability to insults is yet to be fully addressed. It appears that the mechanism of action of AC drugs is divergent from those causing increased Ab levels. For instance, no synergic effects have been detected between the use of AC drugs and inheriting APOE E4 alleles. In fact, the risk of cognitive impairment among AC users is increased if they are not carriers of the APOE E4 alleles (Campbell et al., 2010).

INTRINSIC MECHANISM: FAILED NGF SIGNALING Due to their polarized morphology, neurons are very dependent on extensive axonal transport machinery (Coleman, 2011). In fact, thin diameters of axons, along with abundant and large cargoes, generate a condition extremely vulnerable to intrinsic and extrinsic abnormalities (see Delcroix et al., 2004; Encalada and Goldstein, 2014). Depending on the nature of cargoes, they are required to be actively transported in anterograde or retrograde directions. While the axonal machinery employs distinct types of motor proteins for movement in each direction, recent evidence suggests that these movements along the axonal path include numerous starts and stops in either direction. The bidirectional axonal highway is characterized by a very narrow and crowded path and large cargoes to be transported over a very long distance. In addition, transport machinery is very energy demanding. In the axonal transport system, microtubules (MTs) serve as tracks; motor proteins represent the engines and vehicles; and vesicles and organelles are the loads carried by these vehicles. Moreover, microtubule-associated

ENLARGEMENT OF EARLY ENDOSOMES IN ALZHEIMER'S DISEASE

209

Fig. 13.1. Schematic representation of a basal forebrain cholinergic neuron (BFCN) projecting from the basal forebrain to the cortex/hippocampus. Following binding to tropomyosin receptor kinase A (TrkA) receptors, nerve growth factor (NGF) is internalized and packaged in early endosomes and transported back to the cell body. Possible mechanisms of failed NGF retrograde transport are illustrated. MVB; multivesicular bodies.

proteins (MAPs) maintain the integrity of the roads and oddly sometimes cause traffic jam. Among neurotrophins, NGF has been particularly implicated in phenotypic support of BFCNs including the extent of dendritic arbors, axonal length, cell body size, and synaptogenesis (Garofalo et al., 1992). In addition, a

strong link has been established between the transcription of cholinergic markers and NGF tropomyosin receptor kinase A (TrkA) signaling. Dysfunction of NGF and its TrkA receptors is believed to contribute to the selective degeneration of BFCN associated with the progressive cognitive decline in AD. We previously described

210

A. FAHIMI ET AL.

significant atrophy of BFCNs along with decreased levels of TrkA receptors in the basal forebrain of AD patients (Salehi et al., 2004, 2006). A number of studies have suggested that BFCN degeneration can be linked to failed NGF signaling caused by impaired axonal transport. Axonal transport defects may occur at an early stage in AD pathogenesis. Indeed, investigating trisomic mice overexpressing App (Cooper et al., 2001; Salehi et al., 2006) along with transgenic mice overexpressing mutant APP we investigated the pathogenesis of BFCN neurodegeneration in AD focusing on failed trophic factor signaling as the primary pathogenic mechanism (Salehi et al., 2003, 2004, 2007). The question is by what mechanism, failure in NGF retrograde transport occurs in AD. Our previous work in both mouse models of AD and Down syndrome (DS) has shown that failed NGF signaling can hardly be linked to either reduced NGF production, binding, or internalization. However, we found a drastic reduction in NGF retrograde axonal transport in these mice (Cooper et al., 2001; Salehi et al., 2003). Here we review pathogenic mechanisms in AD that can potentially hamper retrograde axonal transport. Logically, the first step in developing new therapies is to better understand the underpinning molecular mechanism by which, NGF retrograde axonal transport fails in AD. As mentioned previously, while a significant reduction of NGF signaling occurs in BFCNs, the main transport machinery remains intact in the Ts65Dn mouse model of DS (Salehi et al., 2006), suggesting that the failure in retrograde axonal transport does not occur indiscriminately and is specific to NGF-TrkA-packages. Here, we study the possible mechanisms by which failure in the transport of NGF-NGF-receptor cargoes occurs and drastically diminishes the amount of trophic factors that reach the nucleus. Through retrograde axonal transport, early endosomes (EEs) carrying NGF-NGF-receptor complexes are transported to the neuronal cell body to induce gene expression required for survival and differentiation of neurons (Salehi et al., 2004). In addition to trophic factors, retrograde axonal transport plays a role in the removal of unwanted proteins from the nerve terminal region. Identification of several mutations in genes linked to axonal transport machinery in neurodegenerative disorders puts forward evidence that the failure in axonal transport is a key player in the pathogenesis of these diseases including hereditary spastic paraplegia, Charcot–Marie–Tooth disease (CMT), spinal and bulbar muscular atrophy (SBMA), and Perry syndrome (Beijer et al., 2019). A mutation in the p150 subunit of dynactin subunit 1 (DCTN1), which is essential for the binding of dynein to cargo has been linked to SBMA (Puls et al., 2005). Similarly, mutations in mitofusin 2 (Mfn2) and

ras-related protein (Rab7), which encode Mfn2 and Rab7 protein involved in endosomal sorting, have been identified in CMT type 2 (Ponomareva et al., 2016). All these findings support the notion that minor changes in axonal transport machinery could lead to drastic phenotypic consequences in humans. Here, we review several possible mechanisms that could potentially result in failure of retrograde axonal transport in AD (Fig. 13.1).

Blocking the path There has been a longstanding notion that axonal transport is disrupted in AD, which stems from structural evidence for the failure of MTs in the AD brain (Dai et al., 2002; Vicario-Orri et al., 2015) and MT instability due to tau hyperphosphorylation and dissociation from MTs (Iqbal et al., 2010). As mentioned previously, the axonal transport machinery occurs on narrow and crowded axonal paths and similar to our daily street traffic in urban areas, is very prone to a halt at any time. The primary components of the axonal path include MTs and MT-associated proteins. MTs are cytoskeletal elements that facilitate axonal transport in both minus- and plusend directions. A number of studies have indicated that hyperphosphorylation of MAPs including tau disrupts its binding ability to MTs, thereby affecting their stability (Green et al., 2008). Tau undergoes several posttranslational modifications including phosphorylation, methylation, and acetylation. In the AD brain, tau is 2–3 folds more phosphorylated compared to that of a healthy brain (Iqbal et al., 2010). The phosphorylated tau has a higher propensity to detachment from MTs, thus leading to their destabilization. In addition, the detached and phosphorylated tau has the propensity for aggregation leading to rearrangement of tracks, partial road closure, and thus transport deficiency. Human tau is encoded by a single gene on chromosome 17, which can be spliced into six main isoforms in the CNS. Tau has a very hydrophilic composition that makes it highly soluble. Phosphorylation of certain motifs in the repeats has the effect of detaching tau from MTs, therefore contributing to MT instability. Neurofibrillary tangles (NFTs) are considered as one of the primary neuropathological hallmarks of AD, and their frequency correlates well with cognitive decline. The hyperphosphorylation of tau on its serine or threonine residues increases its intracellular deposition that participates in the formation of NFTs (Graham et al., 2017). NFTs are generally formed in axons and then move to the perikaryon. These large aggregated structures can potentially block the axonal transport and cause NGFTrkA cargo traffic jam. If indeed, NFTs lead to mechanical blocking of the axonal path, the block should not be biased toward any specific directions and prevent the transport in both anterograde and retrograde directions.

ENLARGEMENT OF EARLY ENDOSOMES IN ALZHEIMER'S DISEASE

Abnormal tracks Axonal transport takes place on MTs that run the length of the axon, which can be up to 1 m long. These tubular polymers are composed of a- and b-tubulin heterodimers. As mentioned previously, abnormal phosphorylation of tau weakens its association with MTs, form insoluble aggregates that can fill the entire intracellular space of a neuron within neuronal soma and dendrites and compromises its stabilizing function (Chesser et al., 2013). Hyperphosphorylated tau proteins have a reduced affinity for MTs and can sequester normal tau protein into filaments and thereby promote destabilization and disassembly of MTs (Alonso et al., 1996). MT-binding properties of the tau protein can affect axonal formation, maintenance, and transport. A multitude of evidence suggests that the accumulation of normal or abnormal tau with or without disassembled MT leads to failed axonal transport and thus traffic jam (Butzlaff et al., 2015). Several drugs including colchicine and nocodazole that disrupt the polymerization of tubulin monomers have shown to lead to a significant delay in axonal transport in both directions (Melemedjian et al., 2014). Theoretically, abnormal tracks should lead to a bidirectional failure in axonal transport. For these reasons, similar to blocking the path, abnormal tracks might not be the primary cause of failed NGF-TrkA retrograde transport.

Loss of motors Dyneins are a family of primary motor proteins that hydrolyze ATP to generate a mechanical force needed for retrograde transport in neurons. These complex and large molecules consist of dynein heavy chains (DHC), dynein light chains (DLC), and dynein intermediate chains (DIC). While DHC binds to MTs, DLCs are involved in binding to cargo EEs. The activity of dynein is regulated by a number of factors including dynactin, i.e., a multi-subunit protein complex that plays several roles including dynein recruitment to MTs and cargo recognition. Both dynein and DCNT1 bind directly to each other via the DIC and dynactin p150 glued subunit. The failure in the retrograde axonal transport could occur due to defects in these molecular motors. The dynein/dynactin complex mediates retrograde transport (Kardon and Vale, 2009). Accordingly, silencing dynein/ dynactin complex members causes major pathological changes in the axonal compartment. Several neurodegenerative disorders with possible failure in transport machinery including Perry syndrome, amyotrophic lateral sclerosis (ALS), and CMT have all been linked to mutations in genes encoding DCTN1 and dynein cytoplasmic 1 heavy chain 1 (DYNC1H1). Lis-1 is another important regulator of dynein-mediated processes.

211

Mutations in the gene have been linked to lissencephaly, a rare brain disorder characterized by abnormal cortex lamination and thickness (see Lipka et al., 2013). The question on the role of motor protein abnormalities in failed axonal transport in cholinergic neurons remains to be addressed. We quantified dynein levels in the Ts65Dn mouse model of DS with a significant diminishment in axonal transport and found no evidence supporting the role of reduced levels of motor proteins in failed NGF retrograde transport (Salehi et al., 2006).

Destruction of the signal Most macromolecules including proteins, glycosaminoglycans, and lipids are degraded by lysosomes (Saftig and Klumperman, 2009). These membrane-bound organelles also play a significant role in NGF-TrkA complexes. Hyperactivation of lysosomal activity could increase the rate of premature NGF-TrkA signals destruction (Xu et al., 2016). Indeed, lysosomal activity has been shown to be increased in the AD brain (Nixon, 2016). A number of neurodegenerative disorders with failure in retrograde transport including frontotemporal lobar degeneration, i.e., a common form of dementia characterized by behavior and language impairments in ALS have been linked to failed lysosomal degradation. Several studies have linked lysosomal dysfunction to genetic mutations in AD including vesicular trafficking. For instance, dystrophic neurites found in AD brains contain an accumulation of autophagic vacuoles (AVs) (Nixon et al., 2005). Local accumulation of these degradative organelles in axons causes failure in retrograde axonal transport. The focal swellings contain accumulation of AVs and endosomes, which are almost exclusively retrograde cargoes. Numerous disease-causing mutations affect lysosomal function. An example is charged multivesicular body protein 2B (CHMP2B), a component of the endosomal sorting complex required for endosomal sorting complexes required for transport. Mutations in CHMP2B induce enlargement of late endosomes, which is considered as pathological change specific to a subtype of frontotemporal dementia (Urwin et al., 2010). In order to be activated, dynein requires simultaneous binding to an appropriate cargo adaptor and dynactin (McKenney et al., 2014). Direct binding of DIC to snapin mediates dynein recruitment to late endosomes in neurons. Another dynein subunit, dynein intermediate chain, interacts with the scaffold protein JNK-interacting protein (JIP3) (Arimoto et al., 2011). The loss of JIP3 function results in the accumulation of late endosomes in dystrophic axons (Gowrishankar et al., 2017). Dynactin is recruited via the interaction between the dynactin p150 subunit and the Rab7 effector, Rab-interacting lysosomal protein

212

A. FAHIMI ET AL.

(RILP) (Johansson et al., 2007). Following p150-RILP binding, another Rab7 effector known as oxysterolbinding protein-related protein 1L (ORP1L) participates in a chain of events that result in the attachment of dynactin to bIII spectrin on late endosomes (Johansson et al., 2007). The deletion of the dynactin-interaction domain in RILP results in a loss of endolysosomal association with dynactin. In adult 5xFAD transgenic mouse model that coexpresses five familial AD mutations (three in APP and two in PSEN1), endolysosomes were found to accumulate at focal axonal swellings where amyloid deposits are accumulated, suggesting a local block in the trafficking and early maturation of lysosomes which could be detrimental to the translocation of cell body-bound cargoes (Gowrishankar et al., 2015). Snapin, a SNARE-associated protein originally implicated in vesicle fusion and synaptic transmission, has been shown to play a crucial role in regulating late endosome retrograde transport (Cai et al., 2010). Indeed, snapin deletion causes accumulation of late endosomes and impairs lysosomal function and failure in BDNFTrkB retrograde transport (Zhou et al., 2012). Recent evidence indicates that amphisome retrograde transport is impaired in the mossy fiber tracts of the dentate gyrus in young 3xTg-AD mice, compared to wild type. Additionally, these axons appeared swollen and dystrophic (Tammineni and Cai, 2017). Amphisomes have been shown to be key organelle for the degradation of b-CTF via the lysosomal pathway. Several genetic risk factors for AD that are functionally related to autophagy/lysosomal flux might become more vulnerable during aging (Van Acker et al., 2019). Extensive accumulation of misfolded and ubiquitinated proteins in the brain of DS individuals (Tramutola et al., 2016) suggests that global defects in protein quality control ultimately disrupt lysosomal clearance. There is also evidence for an early transcriptional upregulation of genes promoting autophagy and downregulation of negative regulators of autophagy flux in CA1 pyramidal neurons (Bordi et al., 2016). In the context of lysosomal dysfunction, the attention toward APP is well justified. Numerous lines of evidence have linked APP and its products to lysosomal dysfunction in AD (Nixon, 2007; Xu et al., 2016), and the triplication of the APP gene in DS suggests that APP-mediated lysosomal defects occur in DS in a similar fashion. Further, the amelioration of learning and memory defects in conjunction with reduced amyloid via the rescue of lysosomal function in the CRDN8 mice model support the role of lysosomal dysfunction in the progression of APP-associated disease phenotypes (Nardiello et al., 2018). Additional strong pathological evidence for autophagic-lysosomal dysfunction in APP mutant mouse models has also been reported (Lauritzen et al., 2016).

Endosomal enlargement Eukaryotic cells take advantage of endosomes for intracellular transport. A number of small GTPases play the role of binding platforms between motor proteins and these membrane-bound vesicles. Among GTPases, Rab5 is considered as a master regulator of transport between the plasma membrane and EEs. In fact, overexpression of wild-type Rab5 accelerates the process of receptormediated endocytosis. A number of studies have also suggested that Rab5 regulates EEs fusion. Accordingly, overexpression of Rab5 in baby hamster kidney cells led to the enlargement of EEs (Bucci et al., 1992). Structural alterations of cargoes, particularly those leading to cargo expansion, may increase the energy required for their transport and cause transport deficits. Endosome slowing has been correlated with their abnormal enlargement caused by Rab5 overactivation in the presence of excess APP b-CTF (Kim et al., 2016). EEs play a pivotal role in the transport of NGF-TrkA complexes to BFCNs’ cell bodies (see Salehi et al., 2004), and a number of studies have put a spotlight on the endosomal abnormalities in AD pathogenesis. Genome-wide association studies have identified variants in genes encoding endocytic trafficking factors to be significantly associated with AD occurrence (Karch and Goate, 2015). Due to the presence of g-secretase primarily in endosomal structures, disrupted transport of b-CTF-containing endosomes results in their accumulation in axon terminals. This may contribute to axonal swelling prior to amyloid deposition and neuritic dystrophy similar to that observed at the early stages of AD (Stokin et al., 2005). Increased expression of basal forebrain Rab5 has shown to correlate well with cognitive decline in both individuals with MCI and AD (Ginsberg et al., 2011). We have shown that triplication of the App gene, which encodes the precursor of Ab, results in enlarged early endosomes in cholinergic axons and defective retrograde transport of NGF (Salehi et al., 2006). The EE pathology is associated with elevated activation of the Rab5 GTPase, and this activation results from elevated levels of either the full-length APP protein or the APP Cterminal fragment. This may provide a mechanism for the selective vulnerability of cholinergic neurons in AD. It must be noted that EE abnormalities detected in AD brains appear to precede Ab accumulation (Woodruff et al., 2016). While there is some evidence on Abdependent endocytic trafficking defects (Treusch et al., 2011), EE abnormalities may not necessarily be driven by Ab but by b-CTFs (Xu et al., 2016). Kwart et al. (2019) used CRISPR/Cas9 to generate a panel of isogenic human-induced pluripotent stem cells (iPSCs) differentiating them into cortical neurons expressing

ENLARGEMENT OF EARLY ENDOSOMES IN ALZHEIMER'S DISEASE APP and PSEN1 at endogenous levels, revealed abnormalities in a number of AD-implicated genes as well as in endocytosis-associated genes. EE enlargement could be rescued by inhibition of b-site amyloid precursor protein cleaving enzyme 1 (BACE1). Numerous studies have linked b-CTFs and EEs (Takasugi et al., 2018). It has been shown that in membranes of EEs, BACE1 cleaves APP and produces transmembrane b-CTFs. This is followed by cleaving the product and generating Ab peptides. A number of studies have identified functional g-secretase in endosomes that produce b-CTF (Kaether et al., 2006). As described before, Kwart et al. (2019) shed some light on this mechanism. RNA sequencing of these cell lines generated in this study revealed alterations in 406 genes from which several have been linked to endocytic vesicle pathways including APOE, CLU, SORL1, and LRP2. Furthermore, they observed increases in the size of EEs in the APP and PSEN1 mutant lines. Importantly, neurons carrying two mutated copies of both APP and PSEN1 had the largest endosomes. The link between b-CTFs and EEs size is emphasized by the fact that neurons homozygous for APP-A673T, a protective mutation, did not have enlarged endosomes. If there is a link between b-CTFs and EEs size, altering APP processing should lead to EE size normalization. Indeed, treating dAP neurons with a BACE inhibitor and blocking APP cleavage and shifting the production of b-CTFs led to the normalization of endosomal size. An APP knockout iPSC-derived neuron line used as a control had smaller endosomes than wild-type neurons did, suggesting that APP affects the size of endosomes. Additionally, treating wild-type iPSC neurons with g-secretase inhibitors reduced Ab production and increased both b- and a-CTFs. Then b-CTFs, not Ab peptides or a-CTFs, can be linked to the endosomal phenotype. This led to normalization of EEs in these cells. Importantly, blocking APP expression in iPSC-derived neurons led to a generation of smaller EEs compared to wild-type neurons. Inhibiting g-secretase activity in iPSC neurons blocks Ab production and leads to an accumulation of both b- and a-CTFs along with EE enlargement. Adding a BACE inhibitor to the g-secretase inhibitor treatment blocked all but a-secretase processing of APP. These data link b-CTFs and not Ab peptides or a-CTFs, to EEs phenotypic changes. The question on the nature of the effects of b-CTFs on EEs was convincingly addressed by Wu and his colleagues, showing that the overexpression of b-CTF activates Rab5, leading to endosomal enlargement and disrupted retrograde axonal trafficking of NGF-TrkA cargos (Wu et al., 2014). New evidence points to an adaptor protein, phosphotyrosine, interacting with the PH domain and leucine

213

zipper (APPL1) as the mediator. APPL1 contains a domain that binds to b-CTF and leads to the recruitment of APPL1 to Rab5-positive endosomes. APPL1 is localized mainly in EEs membranes and it mediates Rab5 activation caused by elevated levels of b-CTF in DS and AD. Accordingly, knocking down APPL1 reverses b-CTF-induced Rab5 activation and prevents endosomal enlargement. It appears that b-CTF, through acting on APPL1, increases Rab5 activity on endosomes, leading to EE enlargement and thus impaired axonal transport of endosomes in neurons. Rab5 is a small GTPase and modulates early steps of endocytosis, leading to endosomal fusion among several other physiological functions. EEs are a major site of APP processing by b-secretase to yield the b-CTFs. Rab5 plays a critical role in the docking of endosomal membranes. Effector proteins bind to an activated form of Rab5 fusion of the EEs. Adult mice overexpressing Rab5 display enlarged EEs. The phenotypes observed in these mice included upregulated endocytosis, enlarged EEs, enhanced long-term depression, reduced LTP, shorter dendritic spines in the hippocampus, and tau hyperphosphorylation. More importantly, the mice showed BFCN degeneration later in life (Kim et al., 2016). It has been suggested that Rab5 exchange factors mediate the effects of b-CTF on EE morphology. This is supported by the fact that the recruitment of a Rab5 guanine nucleotide exchange factor to EEs increases Rab5 levels in these organelles and leads to their enlargement (see Xu et al., 2016). The data obtained from mutant iPSCs support that both Ab and b-CTF interfere with the endolysosomal system (Willen et al., 2017). These data are consistent with our previous studies in mouse models of AD and DS. The adult Ts65Dn mouse model of DS shows a significant degeneration of BFCN along with decreased levels of NGF retrograde axonal transport in these neurons. Importantly, deleting an extra copy of App in these mice led to the normalization of axonal transport and morphology in BFCNs. In addition, a strong correlation was found between b-CTF levels and failed axonal transport and EE enlargement (Salehi et al., 2006, 2009). Additionally, Cotman and colleagues also found that the retrograde transport deficit was reversed using g-secretase inhibitors. Also, Ab oligomers caused a threefold reduction in axonal transport suggesting that increased Ab levels and not total APP could be blamed for transport deficits (Poon et al., 2011). There are several lines of evidence that link pathogenic factors of familial AD to impaired dynein-mediated retrograde transport of endosomes. Intracellular Ab oligomers interact with dynein in the neurites of mutant App transgenic mouse neurons and hamper axonal transport (Tammineni and Cai, 2017).

214

A. FAHIMI ET AL.

We reported that App triplication in DS mice was found to disrupt retrograde axonal transport of NGFcontaining endosomes, which was normalized by deleting one copy of the App gene (Salehi et al., 2006). APP triplication in DS, which encodes the precursor of Ab, results in enlarged early endosomes in cholinergic axons and defective retrograde transport of NGF (Salehi et al., 2006; Wu et al., 2014). Excess b-CTF over activates Rab5, which in turn causes the enlargement of the endosomes (Kim et al., 2016). Prior to amyloid accumulation and the occurrence of dystrophic neuropathy, b-CTF-containing endosomes tend to accumulate in axons, which causes enlargement of EEs and the disruption of axonal transport. Indeed, recent studies in a cohort of Icelanders indicate that the coding mutation (A673T) in APP was linked to protection against AD and the lack of cognitive decline in the elderly carriers. Importantly, the substitution mapped to a location adjacent to the aspartyl protease b-site causes a drastic reduction in soluble APPb levels. It is crucial to test the status of EEs in the BFCNs of the elderly carrying this protective mutation (Jonsson et al., 2012).

THERAPEUTIC STRATEGIES TARGETING THE CHOLINERGIC SYSTEM Several strategies have been employed in order to restore the structure and function of degenerating BFCNs in AD. Here we focus on two approaches, which directly improve cholinergic neurotransmission and those improving trophic factor signaling in BFCNs.

Improving ACh neurotransmission Currently, three of four marketed drugs for AD act by inhibiting the ACh degrading enzyme, acetylcholinesterase (AChE). While the primary target of these drugs is AChE, it has been shown that they may also increase the biosynthesis of ACh by improving choline acetyltransferase (ChAT) activity. For instance, galantamine is a rapidly reversible inhibitor of AChE used in the symptomatic treatment of dementia of AD type. It has been shown that a 3-month period of treatment with this drug improved ChAT activity and the Cholinergic index (the ratio of ChAT to AChE) in the cerebrospinal fluid (Karami et al., 2019) along with improvement in MMSE compared to baseline. While cholinergic drugs induce positive effects in AD, their effects on cognitive function are modest and transient (see Adlimoghaddam et al., 2018).

Improving NGF signaling There is little doubt that NGF plays a vital role in reducing the loss of cholinergic transmission in AD. However,

its precise delivery to the brain, its sustained and longterm production, and limiting its action to the brain regions affected remain significant technical challenges. Novel methods to deliver NGF on degradable carriers to achieve continuous release have solved some of these technical issues. Despite some adverse effects, a number of strategies used for direct delivery, including CNS infusion, intranasal administration, gene therapy approach along with cell-based delivery, provide attractive methods to efficiently deliver NGF to the brain (Tuszynski et al., 2015). A novel delivery system called encapsulated cell bio-delivery (ECB) provides an opportunity to improve continuous and localized NGF delivery. The implant is a catheter-like device containing a genetically engineered human NGF-secreting cell line, placed at the tip of a semipermeable fiber membrane. The encapsulation of the tip allows the protection of the allogeneic cell line from the host immune system and lets locally secreted NGF molecules diffuse to the brain parenchyma. Four patients with mild to moderate AD were recruited for an open-label, 6-month-long study. Each patient underwent stereotactic implant surgery, and four NGF-ECB implants were placed near the cholinergic basal forebrain. Changes from baseline values of cholinergic markers in CSF correlated with cortical nicotinic receptor expression and the MMSE score. While the method appears to be relatively safe, all patients showed declines on the MMSE (2–3 points) and increased on the Alzheimer’s disease assessment scale-cognitive subscale (ADAS-Cog) scores (3–6 points) during the study. In addition, the follow-up patients continued to show declines in the MMSE score until an additional 20 months after implant removal. As an alternative method of NGF delivery in AD, NGF gene therapy has also been tested (Rafii et al., 2014). NGF was delivered via CERE-110, an adenoassociated virus (AAV)-based gene delivery vector encoding for human NGF, which showed sustained NGF expression. In a multicenter phase II clinical trial, 49 middle age and old participants with mild to moderate AD receive stereotactically guided bilateral injections into the NBM. The change from baseline at 24 months in ADAS-Cog, Clinical Dementia Rating–Sum of Boxes, MMSE, [18F]-fluorodeoxyglucose (FDG)-PET scans were used to assess brain metabolic activity in the bilateral posterior cingulate gyri at baseline and at months 6, 12, and 24. T1 volumetric MRI analysis was performed to assess the whole-brain, hippocampus, and lateral ventricle volumes. No significant differences were detected between the treatment and placebo groups in ADAS-Cog after 24 months. Oddly, the treatment group tended to perform worse on all other clinical outcome measures. The postmortem examination of brain tissue found AAV-NGF expression persisting for years

ENLARGEMENT OF EARLY ENDOSOMES IN ALZHEIMER'S DISEASE after surgery and at levels sufficient to induce new cholinergic afferents in neurons. While no significant positive effects were found, the method of delivery was found to be well tolerated. These rather disappointing results could be linked to low number of subjects, the types of outcome measures selected, selection of nonstratified AD subjects, or whether a sufficient amount of NFG was delivered to NBM neurons. An advantage of the ECB technology over other methods is that the NGF delivery can be stopped by removing the device that contains genetically modified NGF-producing cells. Obtaining a more predictable and stable long-term targeted NGF delivery to the brain in patients with AD with a larger number of individuals along with a longer timeline would be the next logical step. Developing a subset of cognitive measures that heavily focus on the assessment of attention deficits might also be beneficial.

CONCLUSIONS While BFCN degeneration is a common occurrence in AD, several other neurotransmitter systems, including b-adrenergic, histaminergic, and serotonergic systems, also undergo significant degeneration in AD and its mouse models (see Das et al., 2014). This could certainly explain the lack of strong positive effects of cholinergic drugs in AD. To be effective, it seems that there is a need for drugs that restore the function of more than one system. For instance, in addition to cholinergic deficiency, AD is characterized by a reduced density of sigma-1 receptor (Sig-1R). This receptor is a molecular chaperone involved in calcium homeostasis and has been reported to be reduced in cortical regions in AD (Mishina et al., 2008). Accordingly, deleting Sig-1R in primary hippocampal neurons negatively affected the survival of these cells (Hedskog et al., 2013). An example of a drug acting on both muscarinic and sigma receptors is ANAVEX2-73 (tetrahydro-N,Ndimethyl-2,2-diphenyl-3-furanmethanamine hydrochloride), which has shown to block tau hyperphosphorylation and Ab1-42 production in AD mouse models (Lahmy et al., 2013). A commonly prescribed drug donepezil has also been demonstrated to protect memory function synergistically with Sig-1R agonists PRE-084 or ANAVEX2-73 in mice treated with Ab25-35 (Maurice, 2016). An example of a multitarget small molecule for AD is Contilisant. The drug has an affinity for histamine and Sig-1 receptors with anticholinesterase properties. Contilisant improves the cognitive deficit induced by Ab1-42 oligomers in the radial maze assay in young mice (Bautista-Aguilera et al., 2018). One rational therapeutic strategy for preventing EEs morphological abnormalities is to reduce APP-CTF production. g-secretase is a high-molecular-weight complex

215

composed of four components: PS, nicastrin, anterior pharynx defective 1, and presenilin enhancer 2 and is able to cleave b-CTF. Numerous studies have provided evidence that PSEN mutations reduce the efficiency of these cleavages by the g-secretase complex and lead to the release of more Ab42. While due to significant side effects, the initial use of g-secretase inhibitors (GSIs) has been disappointing, g-secretase modulators (GSMs) seem to be associated with fewer side effects including those related to Notch1 cleavage inhibition. Importantly, second-generation GSMs bind at distinct sites in the PS and provide the possibility of causing mutant-specific protection. As evidence for the efficacy of GSMs in vivo, APP/PS1 double transgenic mice showed reduced Ab plaque load and improvement in memory following treatments with a GSM drug. While a number of these drugs in humans caused adverse effects including liver abnormalities, others caused robust Ab40/42 reduction along with reasonable safety measures. A Phase III study of a GSM (tarenflurbil) in AD showed no positive effect on cognition possibly due to its low brain penetrance (Green et al., 2009). The hope is that the clinical trials with GSMs will not have the same destiny as GSIs due to better specificity and minimal adverse effects. The availability of more potent and specific GSMs and their local administration in BFCNs could potentially prevent or reverse endocytic morphological changes and improve NGF retrograde transport in these cells. Recent studies have been able to link endosomal abnormalities, Ab accumulation, and cognitive dysfunction (Xie et al., 2020) and suggest targeting Rab5 as a therapeutic strategy. This could be achieved either by directly targeting Rab5 or by acting on its effectors like GDP dissociation factor. (Yuan and Song, 2020). The fact that Rab5 activation has been linked to several neoplastic conditions in humans suggests that we are not that far from developing effective agents targeting Rab5 and generating an opportunity for repurposing drugs to correct EE morphology and restore BFCN’s function in AD.

ACKNOWLEDGMENT We would like to thank Dr. Sarah Moghadam for her critical reading of the text and constructive comments.

REFERENCES Adlimoghaddam A, Neuendorff M, Roy B et al. (2018). A review of clinical treatment considerations of donepezil in severe Alzheimer’s disease. CNS Neurosci Ther 24: 876–888. Alonso AC, Grundke-Iqbal I, Iqbal K (1996). Alzheimer’s disease hyperphosphorylated tau sequesters normal tau into tangles of filaments and disassembles microtubules. Nat Med 2: 783–787.

216

A. FAHIMI ET AL.

Arimoto M, Koushika SP, Choudhary BC et al. (2011). The Caenorhabditis elegans JIP3 protein UNC-16 functions as an adaptor to link kinesin-1 with cytoplasmic dynein. J Neurosci 31: 2216–2224. Bautista-Aguilera OM, Budni J, Mina F et al. (2018). Contilisant, a tetratarget small molecule for Alzheimer’s disease therapy combining cholinesterase, monoamine oxidase inhibition, and H3R antagonism with S1R agonism profile. J Med Chem 61: 6937–6943. Beijer D, Sisto A, Van Lent J et al. (2019). Defects in axonal transport in inherited neuropathies. J Neuromuscul Dis 6: 401–419. Billioti de Gage S, Moride Y, Ducruet T et al. (2014). Benzodiazepine use and risk of Alzheimer’s disease: case-control study. BMJ 349: g5205. Bordi M, Berg MJ, Mohan PS et al. (2016). Autophagy flux in CA1 neurons of Alzheimer hippocampus: increased induction overburdens failing lysosomes to propel neuritic dystrophy. Autophagy 12: 2467–2483. Bracco L, Bessi V, Padiglioni S et al. (2014). Do cholinesterase inhibitors act primarily on attention deficit? A naturalistic study in Alzheimer’s disease patients. J Alzheimers Dis 40: 737–742. Bucci C, Parton RG, Mather IH et al. (1992). The small GTPase rab5 functions as a regulatory factor in the early endocytic pathway. Cell 70: 715–728. Butzlaff M, Hannan SB, Karsten P et al. (2015). Impaired retrograde transport by the Dynein/Dynactin complex contributes to Tau-induced toxicity. Hum Mol Genet 24: 3623–3637. Cai Y, Xiong K, Zhang XM et al. (2010). Beta-Secretase-1 elevation in aged monkey and Alzheimer’s disease human cerebral cortex occurs around the vasculature in partnership with multisystem axon terminal pathogenesis and beta-amyloid accumulation. Eur J Neurosci 32: 1223–1238. Campbell NL, Boustani MA, Lane KA et al. (2010). Use of anticholinergics and the risk of cognitive impairment in an African American population. Neurology 75: 152–159. Carrie`re I, Fourrier-Reglat A, Dartigues JF et al. (2009). Drugs with anticholinergic properties, cognitive decline, and dementia in an elderly general population, the 3-city study. Arch Intern Med 169: 1317–1324. Chesser AS, Pritchard SM, Johnson GV (2013). Tau clearance mechanisms and their possible role in the pathogenesis of Alzheimer disease. Front Neurol 4: 122. Coleman M (2011). Molecular signaling how do axons die? Adv Genet 73: 185–217. Cooper JD, Salehi A, Delcroix JD et al. (2001). Failed retrograde transport of NGF in a mouse model of Down’s syndrome: reversal of cholinergic neurodegenerative phenotypes following NGF infusion. Proc Natl Acad Sci U S A 98: 10439–10444. Dai J, Buijs RM, Kamphorst W et al. (2002). Impaired axonal transport of cortical neurons in Alzheimer’s disease is associated with neuropathological changes. Brain Res 948: 138–144.

Das D, Phillips C, Hsieh W et al. (2014). Neurotransmitterbased strategies for the treatment of cognitive dysfunction in Down syndrome. Prog Neuropsychopharmacol Biol Psychiatry 3: 140–148. Delcroix JD, Valletta J, Wu C et al. (2004). Trafficking the NGF signal: implications for normal and degenerating neurons. Prog Brain Res 146: 3–23. Drachman DA, Leavitt J (1974). Human memory and the cholinergic system. A relationship to aging? Arch Neurol 30: 113–121. Encalada SE, Goldstein LS (2014). Biophysical challenges to axonal transport: motor-cargo deficiencies and neurodegeneration. Annu Rev Biophys 43: 141–169. Garofalo L, Ribeiro-da-Silva A, Cuello AC (1992). Nerve growth factor-induced synaptogenesis and hypertrophy of cortical cholinergic terminals. Proc Natl Acad Sci U S A 89: 2639–2643. Ginsberg SD, Mufson EJ, Alldred MJ et al. (2011). Upregulation of select rab GTPases in cholinergic basal forebrain neurons in mild cognitive impairment and Alzheimer’s disease. J Chem Neuroanat 42: 102–110. Gowrishankar S, Yuan P, Wu Y et al. (2015). Massive accumulation of luminal protease-deficient axonal lysosomes at Alzheimer’s disease amyloid plaques. Proc Natl Acad Sci U S A 112: E3699–E3708. Gowrishankar S, Wu Y, Ferguson SM (2017). Impaired JIP3dependent axonal lysosome transport promotes amyloid plaque pathology. J Cell Biol 216: 3291–3305. Graham WV, Bonito-Oliva A, Sakmar TP (2017). Update on Alzheimer’s disease therapy and prevention strategies. Annu Rev Med 68: 413–430. Gray SL, Hanlon JT (2018). Anticholinergic drugs and dementia in older adults. BMJ 361: k1722. Gray SL, Anderson ML, Dublin S et al. (2015). Cumulative use of strong anticholinergics and incident dementia: a prospective cohort study. JAMA Intern Med 175: 401–407. Green KN, Steffan JS, Martinez-Coria H et al. (2008). Nicotinamide restores cognition in Alzheimer’s disease transgenic mice via a mechanism involving sirtuin inhibition and selective reduction of Thr231-phospho tau. J Neurosci 28: 11500–11510. Green RC, Schneider LS, Amato DA et al. (2009). Effect of tarenflurbil on cognitive decline and activities of daily living in patients with mild Alzheimer disease: a randomized controlled trial. JAMA 302: 2557–2564. Grothe M, Heinsen H, Teipel SJ (2012). Atrophy of the cholinergic basal forebrain over the adult age range and in early stages of Alzheimer’s disease. Biol Psychiatry 71: 805–813. Hampel H, Mesulam MM, Cuello AC et al. (2018). The cholinergic system in the pathophysiology and treatment of Alzheimer’s disease. Brain 1: 1917–1933. Hedskog L, Pinho CM, Filadi R et al. (2013). Modulation of the endoplasmic reticulum-mitochondria interface in Alzheimer’s disease and related models. Proc Natl Acad Sci U S A 110: 7916–7921.

ENLARGEMENT OF EARLY ENDOSOMES IN ALZHEIMER'S DISEASE Iqbal K, Liu F, Gong CX et al. (2010). Tau in Alzheimer disease and related tauopathies. Curr Alzheimer Res 7: 656–664. Johansson AS, Bergquist J, Volbracht C et al. (2007). Attenuated amyloid-beta aggregation and neurotoxicity owing to methionine oxidation. Neuroreport 18: 559–563. Jonsson T, Atwal JK, Steinberg S et al. (2012). A mutation in APP protects against Alzheimer’s disease and age-related cognitive decline. Nature 488: 96–99. Joseph E, Villalobos-Acosta DMA, Torres-Ramos MA et al. (2020). Neuroprotective effects of apocynin and galantamine during the chronic administration of scopolamine in an Alzheimer’s disease model. J Mol Neurosci 70: 180–193. Kaether C, Haass C, Steiner H (2006). Assembly, trafficking and function of gamma-secretase. Neurodegener Dis 3: 275–283. Karami A, Eriksdotter M, Kadir A et al. (2019). CSF cholinergic index, a new biomeasure of treatment effect in patients with Alzheimer’s disease. Front Mol Neurosci 12: 239. Karch CM, Goate AM (2015). Alzheimer’s disease risk genes and mechanisms of disease pathogenesis. Biol Psychiatry 77: 43–51. Kardon JR, Vale RD (2009). Regulators of the cytoplasmic dynein motor. Nat Rev Mol Cell Biol 10: 854–865. Kim S, Sato Y, Mohan PS et al. (2016). Evidence that the rab5 effector APPL1 mediates APP-betaCTF-induced dysfunction of endosomes in down syndrome and Alzheimer’s disease. Mol Psychiatry 21: 707–716. Knox D, Keller SM (2016). Cholinergic neuronal lesions in the medial septum and vertical limbs of the diagonal band of Broca induce contextual fear memory generalization and impair acquisition of fear extinction. Hippocampus 26: 718–726. Kwart D, Gregg A, Scheckel C et al. (2019). A large panel of isogenic APP and PSEN1 mutant human iPSC neurons reveals shared endosomal abnormalities mediated by APP beta-CTFs, not Abeta. Neuron 104 (256–270): e5. Lahmy V, Meunier J, Malmstr€om S et al. (2013). Blockade of tau hyperphosphorylation and Abeta1-42 generation by the aminotetrahydrofuran derivative ANAVEX2-73, a mixed muscarinic ands₁receptor agonist, in a non transgenic mouse model of Alzheimer’s disease. Neuropsychopharmacology 38: 1706–1723. Lauritzen I, Pardossi-Piquard R, Bourgeois A et al. (2016). Intraneuronal aggregation of the beta-CTF fragment of APP (C99) induces Abeta-independent lysosomal-autophagic pathology. Acta Neuropathol 132: 257–276. Lipka J, Kuijpers M, Jaworski J et al. (2013). Mutations in cytoplasmic dynein and its regulators cause malformations of cortical development and neurodegenerative diseases. Biochem Soc Trans 41: 1605–1612. Maurice T (2016). Protection by sigma-1 receptor agonists is synergic with donepezil, but not with memantine, in a mouse model of amyloid-induced memory impairments. Behav Brain Res 296: 270–278.

217

McKenney RJ, Huynh W, Tanenbaum ME et al. (2014). Activation of cytoplasmic dynein motility by dynactincargo adapter complexes. Science 345: 337–341. Melemedjian OK, Tillu DV, Moy JK et al. (2014). Local translation and retrograde axonal transport of CREB regulates IL-6-induced nociceptive plasticity. Mol Pain 10: 45. Mishina M, Ohyama M, Ishii K et al. (2008). Low density of sigma1 receptors in early Alzheimer’s disease. Ann Nucl Med 22: 151–156. Nardiello P, Pantano D, Lapucci A et al. (2018). Diet supplementation with hydroxytyrosol ameliorates brain pathology and restores cognitive functions in a mouse model of amyloid-b deposition. Alzheimers Dis 63: 1161–1172. Nixon RA (2007). Autophagy, amyloidogenesis and Alzheimer disease. J Cell Sci 120 (pt 23): 4081–4091. Nixon RA (2016). The lysosome in aging-related neurodegenerative diseases. Lysosome: biology, disease, and therapeutics, John Wiley & Sons, Inc, Hoboken, NJ, USA, pp. 137–179. Chapter 9. Nixon RA, Wegiel J, Kumar A et al. (2005). Extensive involvement of autophagy in Alzheimer disease: an immuno-electron microscopy study. J Neuropathol Exp Neurol 64: 113–122. Ponomareva OY, Eliceiri KW, Halloran MC (2016). CharcotMarie-Tooth 2b associated Rab7 mutations cause axon growth and guidance defects during vertebrate sensory neuron development. Neural Dev 11: 2. Poon WW, Blurton-Jones M, Tu CH et al. (2011). b-Amyloid impairs axonal BDNF retrograde trafficking. Neurobiol Aging 32: 821–833. Puls I, Oh SJ, Sumner CJ et al. (2005). Distal spinal and bulbar muscular atrophy caused by dynactin mutation. Ann Neurol 57: 687–694. Rafii MS, Baumann TL, Bakay RA et al. (2014). A phase1 study of stereotactic gene delivery of AAV2-NGF for Alzheimer’s disease. Alzheimers Dement 10: 571–581. Richardson K, Fox C, Maidment I et al. (2018). Anticholinergic drugs and risk of dementia: case-control study. BMJ 361: k1315. Saftig P, Klumperman J (2009). Lysosome biogenesis and lysosomal membrane proteins: trafficking meets function. Nat Rev Mol Cell Biol 10: 623–635. Salehi A, Delcroix JD, Mobley WC (2003). Traffic at the intersection of neurotrophic factor signaling and neurodegeneration. Trends Neurosci 26: 73–80. Salehi A, Delcroix JD, Swaab DF (2004). Alzheimer’s disease and NGF signaling. J Neural Transm 111: 323–345. Salehi A, Delcroix JD, Belichenko PV et al. (2006). Increased App expression in a mouse model of Down’s syndrome disrupts NGF transport and causes cholinergic neuron degeneration. Neuron 51: 29–42. Salehi A, Kleshevnikov A, Mobley WC (2007). Cholinergic neurodegeneration in Alzheimer’s disease: basis for nerve growth factor therapy. In: Pharmacological mechanisms in Alzheimer’s therapeutics, Springer, New York. NY64–104.

218

A. FAHIMI ET AL.

Salehi A, Wu C, Zhan K et al. (2009). Axonal transport of neurotrophic signals: an Achilles’ Heel for neurodegeneration? In: Intracellular traffic and neurodegenerative disorders, 87–101. Stokin GB, Lillo C, Falzone TL et al. (2005). Axonopathy and transport deficits early in the pathogenesis of Alzheimer’s disease. Science 307: 1282–1288. Sura SD, Carnahan RM, Chen H et al. (2013). Prevalence and determinants of anticholinergic medication use in elderly dementia patients. Drugs Aging 30: 837–844. Takasugi N, Araya R, Kamikubo Y et al. (2018). TMEM30A is a candidate interacting partner for the beta-carboxylterminal fragment of amyloid-beta precursor protein in endosomes. PLoS One 13: e0200988. Tammineni P, Cai Q (2017). Defective retrograde transport impairs autophagic clearance in Alzheimer disease neurons. Autophagy 13: 982–984. Tramutola A, Pupo G, Di Domenico F et al. (2016). Activation of p53 in down syndrome and in the Ts65Dn mouse brain is associated with a pro-apoptotic phenotype. J Alzheimers Dis 52: 359–371. Treusch S, Hamamichi S, Goodman JL et al. (2011). Functional links between Abeta toxicity, endocytic trafficking, and Alzheimer’s disease risk factors in yeast. Science 334: 1241–1245. Tuszynski MH, Yang JH, Barba D et al. (2015). Nerve growth factor gene therapy: activation of neuronal responses in Alzheimer disease. JAMA Neurol 72: 1139–1147. Urwin H, Josephs KA, Rohrer JD et al. (2010). FUS pathology defines the majority of tau- and TDP-43-negative frontotemporal lobar degeneration. Acta Neuropathol 120: 33–41. Van Acker ZP, Bretou M, Annaert W (2019). Endo-lysosomal dysregulations and late-onset Alzheimer’s disease: impact of genetic risk factors. Mol Neurodegener 14: 20. Vicario-Orri E, Opazo CM, Munoz FJ (2015). The pathophysiology of axonal transport in Alzheimer’s disease. J Alzheimers Dis 43: 1097–1113.

Willen K, Edgar JR, Hasegawa T et al. (2017). Abeta accumulation causes MVB enlargement and is modelled by dominant negative VPS4A. Mol Neurodegener 12: 61. Woodruff G, Reyna SM, Dunlap M et al. (2016). Defective transcytosis of APP and lipoproteins in human iPSCderived neurons with familial Alzheimer’s disease mutations. Cell Rep 17: 759–773. Wu H, Williams J, Nathans J (2014). Complete morphologies of basal forebrain cholinergic neurons in the mouse. Elife 3: e02444. Xie Y, Ba L, Wang M et al. (2020). Chronic sleep fragmentation shares similar pathogenesis with neurodegenerative diseases: endosome-autophagosome-lysosome pathway dysfunction and microglia-mediated neuroinflammation. CNS Neurosci Ther 26: 215–227. Xu TH, Yan Y, Kang Y et al. (2016). Alzheimer’s diseaseassociated mutations increase amyloid precursor protein resistance to gamma-secretase cleavage and the Abeta42/ Abeta40 ratio. Cell Discov 2: 16026. Yuan W, Song C (2020). The emerging role of Rab5 in membrane receptor trafficking and signaling pathways. Biochem Res Int 11: 4186308. Zhou B, Cai Q, Xie Y et al. (2012). Snapin recruits dynein to BDNF-TrkB signaling endosomes for retrograde axonal transport and is essential for dendrite growth of cortical neurons. Cell Rep 2: 42–51.

FURTHER READING Phillips C, Baktir MA, Srivatsan M et al. (2014). Neuroprotective effects of physical activity on the brain: a closer look at trophic factor signaling. Front Cell Neurosci 8: 170. Salehi A, Verhaagen J, Dijkhuizen PA et al. (1996). Co-localization of high-affinity neurotrophin receptors in nucleus basalis of Meynert neurons and their differential reduction in Alzheimer’s disease. Neuroscience 75: 373–387.

Handbook of Clinical Neurology, Vol. 179 (3rd series) The Human Hypothalamus: Anterior Region D.F. Swaab, F. Kreier, P.J. Lucassen, A. Salehi, and R.M. Buijs, Editors https://doi.org/10.1016/B978-0-12-819975-6.00012-1 Copyright © 2021 Elsevier B.V. All rights reserved

Chapter 14

Gene and cell therapy for the nucleus basalis of Meynert with NGF in Alzheimer’s disease MARIA ERIKSDOTTER1,2* AND SUMONTO MITRA1 1

Division of Clinical Geriatrics, Department of Neurobiology, Care Sciences and Society (NVS), Karolinska Institutet, Stockholm, Sweden 2

Theme Aging, Karolinska University Hospital, Huddinge, Sweden

Abstract There is currently no effective treatment for the most common of the dementia disorders, Alzheimer’s disease (AD). It has been known for decades that the central cholinergic system is important for memory. The cholinergic neurons in the basal forebrain with its cortical and hippocampal projections degenerate in AD and thus contribute to the cognitive decline characteristic of AD. This knowledge led to the development of the currently approved treatment for AD, with inhibitors of acetylcholine-esterase targeting the cholinergic system with beneficial but mild effects. In recent years, other approaches to influence the degenerating cholinergic system in AD focusing on nerve growth factor (NGF) have been undertaken. NGF is required for the survival and function of the basal forebrain cholinergic neurons, the most important being the nucleus basalis of Meynert (nbM). Since there is a lack of NGF and its receptors in the AD forebrain, the hypothesis is that local delivery of NGF to the nbM could revive the cholinergic circuitry and thereby restore cognitive functions. Since NGF does not pass through the blood–brain barrier, approaches involving cerebral injections of genetically modified cells or viral vectors or implantation of encapsulated cells in the nbM in AD patients have been used. These attempts have been partially successful but also have limitations, which are presented and discussed here. In conclusion, these trials point to the importance of further development of NGF-related therapies in AD.

INTRODUCTION The population of the Western world is aging rapidly. Disorders with cognitive disturbances increase with age and dementia diseases are among the most common. According to the World Health Organization (WHO), there are nearly 10 million new dementia cases annually and currently about 50 million patients with dementia, forecasted to increase to 152 million in 2050 (World Health Organization, 2019). Dementia leads to a progressive cognitive and functional decline over time and is one of the leading causes of disability and dependency worldwide.

Alzheimer’s disease is the most common form of dementia, accounting for about two-thirds of total dementia cases. Alzheimer’s disease (AD) is a devastating neurodegenerative disorder associated with memory decline and deterioration of other cognitive functions, abnormal protein modification, and inflammation. The current AD treatments are drugs that stabilize acetylcholine (ACh) metabolism (ChEIs) or NMDA receptor signaling (memantine), having mainly symptomatic effects, although evidence for biological effects has been demonstrated (Hampel et al., 2018; Atri, 2019). The limited success thus far in the development of a disease-modifying therapy is linked to the complexity

*Correspondence to: Maria Eriksdotter, Department of Neurobiology, Care Sciences and Society, Karolinska Institutet, Neo, Blickagången 16, Plan 7, 141 83 Huddinge, Sweden. Tel: +46-8-524-86479, E-mail: [email protected]

220

M. ERIKSDOTTER AND S. MITRA

and poor understanding of the disease pathogenesis. AD is characterized by intercellular plaques with an amyloid-b protein core and intracellular tau tangles; a great deal of evidence suggests that Ab amyloidosis has an important role in AD pathogenesis. For the past two decades, the amyloid pathway in AD has been extensively targeted therapeutically, but hitherto has failed to show therapeutic efficacy (Gold, 2017; Mehta et al., 2017; Servick, 2019). Recent negative first results of two antiamyloid drug trials in persons with familial AD with known mutations in the APP gene have again dampened hopes (Washington University Press release, 2020). Interestingly, the biotechnology company Biogen Inc. has announced positive effects with high doses of the b-amyloid-targeting monoclonal antibody aducanumab; although its effects have been debated (Schneider, 2020), marketing approval by the USFDA is pending. Regardless of potential approval of aducanumab and other future antiamyloid substances, it is increasingly clear that therapies addressing nonamyloid targets as disease-modifying therapies for AD are needed.

THE CHOLINERGIC SYSTEM AND THE BASAL FOREBRAIN Cholinergic neurons are essential in fine-tuning brain function (for review, see (Bonsi et al., 2011)) and maintaining the excitation-inhibition balance within neural circuits (Zhou et al., 2017). The central cholinergic

system of the human brain plays a central role in memory in health and disease (Ballinger et al., 2016; Gielow and Zaborszky, 2017). Alterations to the cholinergic system can lead to severe dysfunction of neuronal circuits. For example, cholinergic neuron loss in the forebrain leads to cognitive deficits associated with Parkinson’s disease (PD; for review, see (Pepeu and Grazia Giovannini, 2017)) and Alzheimer’s disease (AD; for review, see (Hampel et al., 2018)). Although cholinergic neurons are distributed in discrete regions, they can project to almost all parts of the brain (Dautan et al., 2016) and release acetylcholine (ACh). A major population is found in the striatum, which contains the highest levels of ACh in the brain. Cholinergic neurons are also found in the basal forebrain and brainstem and smaller cholinergic populations are located in the cerebral cortex, hypothalamus, and olfactory bulb (Ahmed et al., 2019). The basal forebrain cholinergic neurons (BFCNs) are located in four nuclei: the medial septal nucleus, the vertical and horizontal limbs of the diagonal band of Broca, and the nucleus basalis of Meynert (nbM) (Fig. 14.1). The cholinergic neurons in the medial septum and vertical diagonal band of Broca (vdB, including the cholinergic component called Ch2) project to the hippocampus (Teles-Grilo Ruivo and Mellor, 2013). The most important nucleus is the nbM, containing most of the BFCNs. Ch4 denotes the group of cholinergic neurons within the nbM. The nbM is located in the corticoid-limbic belt of the brain.

Fig. 14.1. Brain anatomy and the basal forebrain nuclei—A schematic representation of the brain where the highlighted structures in green indicate the nuclei of the basal forebrain. The dark shaded (green) region indicates the brain structures containing the medial septal nucleus and the vertical diagonal band of Broca (vdB, including the cholinergic neuron group Ch2). Similarly, the lighter shaded green area contains the brain structures of the horizontal diagonal band of Broca (hdB) and the nucleus basalis of Meynert (nbM, including the cholinergic neuron group Ch4). Cholinergic projections originating from the nbM project to the cortex and from dvB to the hippocampus, two important structures of memory and cognition. The scale of the representation is arbitrary.

GENE AND CELL THERAPY The cholinergic neurons in the basal forebrain play an important role in cognition through acetylcholine innervation to the neocortex, hippocampus, olfactory bulb, and amygdala (Mesulam et al., 1983; Woolf, 1991) involved in attention, planning, and memory, which has been confirmed using advanced imaging techniques to track cholinergic white matter pathways (Nemy et al., 2020). Cognitive decline in AD has been associated with central cholinergic dysfunction and neurodegeneration in the basal forebrain (Bowen et al., 1976; Davies and Maloney, 1976; Mesulam, 1976; Whitehouse et al., 1981), including a profound cholinergic neuron loss in the nbM (Whitehouse et al., 1981, 1982). The demonstration that cholinergic antagonists impair and agonists improve memory (Drachman and Leavitt, 1974), along with several other supporting reports, led to the proposal of the cholinergic hypothesis, which for the first time proposed the importance of cholinergic neuron degeneration in AD associated with cognitive impairment (Bartus et al., 1982). Later it was shown that abnormally phosphorylated tau in the form of neurofibrillary tangles in the BFCNs was found in preclinical and early stages of the disease and correlated with memory dysfunction (Mesulam, 2004). Reports also demonstrated a direct correlation between reduced basal forebrain volume and cognitive decline in predementia patients lacking overt brain atrophy (Grothe et al., 2010; Kerbler et al., 2015). Interestingly, the decrement in the basal forebrain volume precedes any major impact on hippocampal volume and predicts the cortical spread of AD pathology, thereby implying the early susceptibility of basal forebrain during AD development (Schmitz et al., 2016; Teipel et al., 2018).

TREATMENT WITH ChEIs IN ALZHEIMER’S DISEASE Based on the cholinergic hypothesis, cholinesterase inhibitors (ChEIs) were developed in the 1990s to enhance ACh signaling. Despite three decades of intensive research, ChEIs are still the main antidementia drugs currently in clinical use for AD, and a wealth of clinical studies support the benefit of ChEIs in AD (Giacobini, 2002; Jelic and Winblad, 2016; Grossberg et al., 2019). ChEIs may also provide benefits in AD pathology. A French randomized trial of 12-month treatment with the ChEI donepezil in suspected prodromal AD patients showed a 45% reduction of the rate of hippocampal atrophy in the donepezil group when compared with placebo (Dubois et al., 2015). Although a treatment response on hippocampal atrophy was observed, further studies showed that the volumes of hippocampal and basal forebrain could not predict treatment response on cognition (Teipel et al., 2016).

221

Interestingly, increasing evidence links ChEI use to cardio- and cerebrovascular protection. A recent metaanalysis of nine cohort studies found a 37% reduction in cardiovascular events, including stroke, myocardial infarction, acute coronary syndrome, and cardiovascular mortality (Isik et al., 2018). Using data from the Swedish Dementia registry, SveDem, (Religa et al., 2015), it was shown that ChEIs are associated with reduced mortality in AD (Secnik et al., 2020), reduced risk for myocardial infarction (Nordstrom et al., 2013) and stroke (Tan et al., 2018), and delay in antipsychotic and anxiolytic initiation in AD (Tan et al., 2020), thereby reducing polypharmacy. These effects may be due to the known antiinflammatory effects of the ChEIs with attenuation of macrophage and microglial response to cytokines (Pavlov et al., 2003; Shytle et al., 2004). At present, ChEIs are part of the therapeutic arsenal in AD, and emerging therapies will therefore in many cases be tested as add-ons to the current ChEI treatment.

RATIONALE FOR NERVE GROWTH FACTOR IN ALZHEIMER’S DISEASE Neurotrophins are a family of structurally related peptides of which nerve growth factor (NGF) was the first to be discovered, in 1951 (Levi-Montalcini and Hamburger, 1951); Rita Levi-Montalcini won the 1986 Nobel Prize, along with Stanley Cohen, for their work on NGF. In both central and peripheral nervous systems, NGF regulates differentiation, growth, survival, and plasticity of certain cell types, including the central cholinergic neurons (Levi-Montalcini, 2004; Niewiadomska et al., 2011). The BFCNs depend on NGF for their survival and function and express the two types of NGF receptors, the tropomyosin receptor kinase A (TrkA, the high affinity receptor) and the p75 neurotrophin receptor (p75NTR or LNGFR, the low-affinity receptor) (Niewiadomska et al., 2011; Isaev et al., 2017). Endogenous NGF is synthesized by postsynaptic cortical and hippocampal neurons and taken up by the presynaptic cholinergic neuronal projections through the TrkA/p75 receptors (Segal, 2003; Biane et al., 2014); it is then internalized and retrogradely transported to the cholinergic cell bodies in the basal forebrain (Schwab et al., 1979), where further signaling cascades are initiated, including cell survival and release of acetylcholine through the cortico-hippocampal projections (DiStefano et al., 1992; Campenot and MacInnis, 2004). NGF is secreted in a precursor form (proNGF; Fahnestock et al., 2001), which is converted by the protease plasmin to mature NGF (mNGF) and finally degraded by matrix metalloproteinase 9 (MMP9; Mroczko et al., 2013). Both the precursor and mature forms of NGF are biologically active (Iulita and Cuello, 2014). ProNGF alone is capable

222

M. ERIKSDOTTER AND S. MITRA

of inducing apoptotic signaling through its interaction with the p75-receptor, while both proNGF and mNGF can induce neurotrophic effects through the TrkA receptor, although the mature form is more potent than the pro- form (Al-Shawi et al., 2007; Bradshaw et al., 2015; Ioannou and Fahnestock, 2017). Studies several decades ago demonstrated that NGF has regenerative and neuroprotective effects on basal forebrain cholinergic neurons in rats and monkeys (Hefti, 1986; Eriksdotter-Nilsson et al., 1989; Kordower et al., 1994) and improved spatial memory in aged rats (Fischer et al., 1987). The NGF metabolism is reported to be hampered in AD (Iulita and Cuello, 2014; Isaev et al., 2017; Mitra et al., 2019) with a gradual dysregulation of NGF during AD development (Peng et al., 2004; Budni et al., 2015). Several studies reported decreased levels of NGF and NGF receptors in the basal forebrain in individuals with AD, while NGF levels in hippocampus and cortex are unchanged or increased. Impaired signaling of NGF through its receptors is another important factor contributing to the development of AD (Latina et al., 2017). The mechanisms behind the reduced neurotrophic support in AD are not completely understood, but disturbances in the retrograde transport and/or signaling of NGF, as well as an imbalance in the ratio of NGF and its precursor proNGF, play a role (Capsoni et al., 2010; Tiveron et al., 2013; Iulita and Cuello, 2016). The apoptotic signaling induced by proNGF is favored by the decreased ratio of TrkA/p75 found in AD brains (Counts et al., 2004; Ginsberg et al., 2006; Masoudi et al., 2009). Higher proNGF levels not only induce proapoptotic signaling but are also found to affect the receptor binding of mNGF and its axonal trafficking and result in atrophy of cholinergic neurons in the basal forebrain (Sobottka et al., 2008; Ioannou and Fahnestock, 2017; Cuello et al., 2019). These changes have also been demonstrated in AD-like mice models (Tiveron et al., 2013). Shortage of mNGF has been shown in a mice model that expresses anti-mNGF antibodies (the AD11 mice model) to induce inflammatory responses and Alzheimer’s neurodegeneration (Capsoni et al., 2011). All in all, in AD, the cholinergic system is affected by altered NGF maturation, skewed TrkA/p75 receptor ratio, inefficient axonal transport and signaling, Abinduced modulation of NGF receptors, and suboptimal acetylcholine innervation-induced inflammatory response (Mitra et al., 2019). Although the NGF neurotrophic effects are diverse, considerable experimental evidence demonstrates the ability of NGF to restore cognitive alterations (Hampel et al., 2018) (Mitra et al., 2019), suggesting the potential of NGF as a therapeutic substance for AD by rescuing the cholinergic neurons in the basal forebrain. However, the delivery of NGF to the BFCN is a challenge, since one essential caveat for many molecules

and compounds such as NGF is insufficient diffusion across the blood–brain barrier (Poduslo and Curran, 1996). Thus these molecules, need to be delivered locally into the brain.

INTRACEREBROVENTRICULAR INFUSION OF NGF One possible administration route is by intracerebroventricular (icv) infusion by small infusion pumps, such as the Alzet osmotic pumps. The pump is connected to a metal catheter, with a plastic platform that can be customized for targeting different brain regions or can be adjusted in size. In this way a constant delivery of compounds or proteins into the brain regions of interest can be ensured. The challenges are to define what doses are effective and tolerable, and also to remain in the target location long enough to elicit an effect. Intracerebral infusion is a challenging route that has been reviewed in clinical trials involving AD patients (Hagg, 2007). In the 1990s our research group performed a small clinical study in 3 AD patients infusing mouse NGF into the cerebral lateral ventricle (Olson et al., 1992; Eriksdotter Jonhagen et al., 1998). In this study, nicotinic receptors and regional cerebral blood flow were both increased in the neocortex following NGF infusion, as analyzed by positron emission tomography (PET) imaging. This effect was maintained for several months after the NGF infusion ended. In addition, EEG activity showed signs of normalizing in these patients. However, due to adverse events, including centrally induced neuropathic pain and weight loss, the icv NGF infusion study was stopped prematurely (Eriksdotter Jonhagen et al., 1998). It was later shown in animal studies that when NGF was infused directly into the brain parenchyma, no pain-related effects were found (Hao et al., 2000). The NGF infusion into the cerebral lateral ventricle aimed to counter cognitive decline by reaching areas with cholinergic nerve cell bodies and the vast cholinergic termination areas in cortex and hippocampus. The cerebrospinal fluid (CSF) moves along the perivascular channels to the subarachnoid space. NGF infused into the ventricular fluid may therefore be absorbed by tissues in brain regions such as the cerebral cortex (Greitz and Hannerz, 1996) and affect the basal forebrain cholinergic projections to the cortex. The adverse effect of pain described here was probably due to the fact that NGF delivery indirectly affected the release of pain mediators, including histamine and prostaglandins from mast cells, capable of releasing NGF sensitizing adjacent nociceptive neurons (Kawamoto et al., 2002). Moreover, direct icv NGF administration has been shown to cause pain in animals by affecting the dorsal root ganglion and weight loss by affecting the hypothalamus function

GENE AND CELL THERAPY (McKelvey et al., 2013). In contrast, when NGF was infused directly into the rat brain parenchyma (Hao et al., 2000) or primate brain tissue (Tuszynski et al., 1990), no pain-related effects were observed. These painful side effects of NGF elicited by the icv administration route led to the abandonment of development of this route for direct NGF delivery to the brain (Hagg, 2007; Mitra et al., 2019).

Painless NGF Another route forward could involve the use of “painless NGF” (hNGFp), a double mutant form of NGF (hNGFp61S/R100E) based on the human genetic disease hereditary sensory autonomic neuropathy type V (Cattaneo and Capsoni, 2019). The painless NGF activates trkA and has similar neurotrophic properties to wild-type NGF, but a markedly lower pain sensitivity (Capsoni et al., 2011). Intraparenchymal delivery of hNGF to the nbM in transgenic mouse models for AD has been reported to decrease Ab plaque load but to stimulate cholinergic sprouting. Interestingly, using intranasal delivery, a decrease in plaque load in several brain regions was shown, at least partly through hNGFp actions on astrocytes and microglia (Capsoni et al., 2017). The procholinergic activity in addition to neuroprotection mediated by microglia in combination with reduced pain-sensitizing activity make hNGFp an interesting candidate for further therapeutical development in AD, and preclinical enabling studies are ongoing (Cattaneo and Capsoni, 2019).

LOCAL DELIVERY OF NGF Since NGF cannot pass through the blood–brain barrier and the icv administrative route appeared not to be feasible, interest has turned to the development of other delivery routes. In addition, the delivered substance must not elicit any adverse effects. With this said, it is clear that delivery of NGF to the BFCN to restore cholinergic function in AD would need to be local, precise, and safe. This represents, however, a clinical and technical challenge. To enhance delivery without complication from NGF-associated side effects, different approaches have been undertaken. Both local injections of cells or viral particles transfected with NGF-producing vectors and implantation of encapsulated cells transfected with NGF are discussed in the following sections.

NGF delivery via genetically modified cells or viral vectors to the nbM Using viral vector-mediated gene transfer, a therapeutic gene can be expressed and can release potent proteins for prolonged periods. Autologous cells, immune-compatible

223

with the host and transfected with a defined gene for release of a specific protein to the target, can be injected. Another approach is to inject viral particles to target multiple cell types at the injection site and transduce them with the sought-after therapeutic gene coding vector to induce secretion locally. The advantage is no requirement for refilling or replacement, while the disadvantage is these cells or viral particles, once injected, cannot be retrieved. In the late 1980s a mouse fibroblast cell line was transfected with rat-NGF-secreting vectors and implanted into rat brains (Ernfors et al., 1989). The basic concept of gene therapy with NGF-producing cells was developed in subsequent years, essentially consisting of cells being genetically modified to produce NGF, tested for release properties in vitro, and then grafted to the medial septal or nucleus basalis region (Chen and Gage, 1995; Klein et al., 2000; Tuszynski and Blesch, 2004). Continuous NGF delivery over several months by implanted genetically modified central nervous system-derived neural stem cells to aged rat basal forebrain led to functional recovery in spatial memory (Martinez-Serrano et al., 1996; Martinez-Serrano and Bjorklund, 1998). Autologous fibroblasts harboring NGF-expressing vectors implanted in the basal forebrain of primates long term did not significantly increase Ab deposition (Tuszynski et al., 1998). This knowledge paved the way for the first human study using gene therapy in 8 patients with AD, published in 2005 (Tuszynski et al., 2005). Data from this 2-year, open-label, phase I study in 8 Alzheimer’s patients, following in vivo NGF gene delivery using retroviral vectors encoding NGF in autologous fibroblasts implanted in the nbM, showed good tolerability, a significant increase of glucose uptake in several cortical regions, and stabilized cognitive function or cognitive decline at a slower rate than expected (Tuszynski et al., 2005); no pain side effects were reported. Analyses of brains from these patients up to 10 years later showed persisting neuronal sprouting toward the implanted fibroblasts, indicating long-lasting effects of the NGF administration (Tuszynski et al., 2015). Further preclinical development led to the emergence of adeno-associated virus serotype 2 (AAV2) vectors to monkey or human brain as safe and long-lasting (Bankiewicz et al., 2000; Janson et al., 2002; Eberling et al., 2008). A dose-escalating phase I study on 10 mild to moderate AD patients using AAV2-NGF delivery to the nbM showed that the delivery method was feasible in humans, well tolerated, safe, and resulted in long-term NGF expression (Rafii et al., 2014; Tuszynski et al., 2015). These encouraging results led to a multicenter sham placebo-controlled phase II clinical trial on 40 AD patients receiving 2 (AAV2) vectors expressing human

224

M. ERIKSDOTTER AND S. MITRA

NGF to the nbM (Rafii et al., 2018). This trial, however, did not report any effect on cognition in up to 2 years of follow-up, in part because the vector placement in the brain was unsatisfactory. Further analysis of postmortem brains from the phase I dose-escalation study (Tuszynski et al., 2015) revealed that, although AAV2-NGF vectors were capable of inducing NGF signaling for almost 7 years in the injected sites, those sites were off target, i.e., not within the cholinergic nucleus basalis of Meynert, nbM, intended to be activated by the therapy and, in addition, the mean distance of AAV2-NGF spread was disappointingly small (Castle et al., 2020). Due to the limited spread of the injected AAV2-NGF vectors coupled with incorrect stereotactic targeting, this multicenter clinical trial was not able to induce the cholinergic pathways needed for the improvement of cognition in AD patients and thus the efficacy of AAV2 therapy with NGF is still unresolved. The limited spread of NGF reported had not previously been seen in preclinical animal studies on rat (Rosenberg et al., 1988; Bishop et al., 2008) and monkey (Hadaczek et al., 2006). One cause of the limited spread in humans as opposed to animals may have been the noncontinuous manual injections used in human studies. At present it is thus not possible to predict whether vectors expressing NGF delivered to the nbM will have clinical efficacy. Improved methods to enable target precision

(for example, using MRI guidance) and to support spread over longer distances (for example, using continuous vector infusion and convection-enhanced delivery) will be the next step. Future approaches may also include the use of icv infusion of AAV9, where gene expression has been found in the basal forebrain and in the cortex (Samaranch et al., 2012; Gray et al., 2013), although risk of adverse effects must be taken into account. Another approach under development involves intravenous AAV infusion using ultrasound to transiently open up the blood–brain barrier and deliver AAV to targeted brain regions (Noroozian et al., 2019; Weber-Adrian et al., 2019).

Encapsulated cell biodelivery of NGF to the nbM Another approach employs the encapsulated cell biodelivery (ECB) method, wherein neurotrophin-releasing modified cells are ensheathed inside a polymer-based capsule (Wahlberg et al., 2012) (Fig. 14.2A). These capsules can be implanted in specific brain locations and can also be removed when needed, a clear advantage to ex vivo or in vivo gene therapy described previously. Along with our industry partner (Gloriana Therapeutics, Inc., United States), we have utilized these ECB capsules

Fig. 14.2. Preparation of an encapsulated cell biodelivery (ECB) device and its implantation—(A) The ECB devices are hollow fiber semipermeable membranes harboring 3D structures supporting the growth of genetically modified cells in the “active part.” These cells are genetically modified to express high levels of nerve growth factor (NGF). The semipermeable membrane allows the outward diffusion of the released NGF from the cells and allows the uptake of oxygen, glucose, and other metabolites needed for their survival. The active part is attached to a long “tether” used to hold the active part and guide it through the brain tissue during the implantation. The tether is also the region of the device that is mechanically fixed to the skull to keep the active part in its desired place during the duration of the therapy and is also utilized to explant the active part after the end of therapy. (B) Schematic representation showing the placement of the ECB in a human brain, within the basal forebrain structure nbM (including the Ch4 with the cholinergic neurons) shown in light green. The other end of the tether is fixed to the skull to maintain the position of the active region within the basal forebrain. The scale of the representation is arbitrary.

GENE AND CELL THERAPY containing cells releasing NGF implanted in the basal forebrain in 10 AD patients (Fig. 14.2B). The implant is a catheter-like device containing a genetically engineered NGF-secreting human cell line (ARPE-19 human retinal pigment epithelial cell line) placed within the tip of a semipermeable hollow fiber membrane (Wahlberg et al., 2012). The allogenic cell line is thus protected from the host immune system but at the same time the diffusion of NGF, oxygen, and nutrients across the semipermeable membrane is ensured. Implants were placed with stereotactic neurosurgery bilaterally in the nbM (Ch4) nuclei in all patients and also in the vertical limb of the diagonal band of Broca (Ch2) in 7 patients. The surgery was preceded by a stereotactic MRI to define trajectories and localize targets precisely. An image fusion technique was used to document the position of the implants. In the first 6 patients with mild to moderate AD, encapsulated NGF-producing cell devices were implanted in the basal forebrain and remained in place for 12 months (Eriksdotter-Jonhagen et al., 2012; Wahlberg et al., 2012). The feasibility and safety of neurotrophin (NGF) delivery using ECB-NGF devices were clearly demonstrated, but cell survival and NGF release from the encapsulated cells were low when the retrieved ECB devices were analyzed after their removal from the brain 12 months postimplantation (EriksdotterJonhagen et al., 2012). Therefore the last 4 patients received implants with an improved version of the device secreting significantly higher amounts of NGF using the Sleeping Beauty transposon technology, a nonviral DNA transfer tool, in which cells were cotransfected with separate plasmids coding for human NGF and a Sleeping Beauty transposase (Izsvak et al., 2009; Fjord-Larsen et al., 2012). The ECB devices were implanted in both the nbM with the cholinergic Ch4 region and the vbB including the cholinergic Ch2 region of the basal forebrain bilaterally, and they remained in place for 6 months (Eyjolfsdottir et al., 2016). The improved encapsulated cells delivered 10 times more NGF (Fjord-Larsen et al., 2012) than the encapsulated cells used in the first 6 patients (Eriksdotter-Jonhagen et al., 2012). These devices were successfully removed after 6 months. The procedure demonstrated good safety and tolerability over time and the side effects of pain and weight loss previously observed with icv delivery were not found (Eriksdotter-Jonhagen et al., 2012; Eyjolfsdottir et al., 2016). This was actually the first time neurosurgeons successfully placed implants both in the nbM (including the Ch4 region) and in the vdB, including the Ch2 region of the basal forebrain. Half the patients responded to the NGF treatment with increased cholinergic markers (ChAT activity) in CSF, correlating with improved cognition and brain glucose metabolism (Karami et al., 2015), less overall brain

225

atrophy (Ferreira et al., 2014), and normalization of the EEG pattern (Eriksdotter-Jonhagen et al., 2012). A recent report showed that the NGF treatment to the basal forebrain seemed to slow the rates of atrophy in the hippocampus and part of the parietal cortex, but not the primary somatosensory cortex nor the basal forebrain itself (Machado et al., 2020). Despite the encouraging results from the ECB-NGF therapeutic approach, variations were observed in the levels of NGF release and cell survival between capsules. A careful characterization of the transplanted genetically engineered cells and of the quality and quantity of their produced therapeutic molecule is critical. To further develop the therapeutic platform, multifunctional approaches are needed to achieve optimal efficacy of the NGF-producing cells by investigating what factors are involved in influencing NGF production. The surgical procedure when implanting the ECB device may cause small tissue damage to small blood vessels, neurons, and glia (Hovens et al., 2013). Microglia and astrocytes in the surrounding brain tissue may be activated and respond by releasing various cytokines and free radicals (Feng et al., 2017), which in turn could diffuse through the capsule membrane and affect the NGF release and cell survival. Indeed, exposure of NGFproducing cells to CSF from AD patients significantly reduced the NGF release, compared with CSF from nondemented patients, indicating a negative influence from factors in AD CSF (Eriksdotter et al., 2018). Using in vitro methods, the proinflammatory cytokine IL-1b was shown to negatively affect the NGF-releasing cells, suggesting that inflammation can affect the ECB cells (Eriksdotter et al., 2018). On the other hand, at physiological concentrations, neither Ab40 nor Ab42 had any major impact on cell viability or NGF production (Eriksdotter et al., 2018), which is encouraging since Ab peptides represent an obvious factor that might affect survival of the NGF-producing cells. Thus, not unexpectedly, the disease conditions of the human brain per se play an important role in the function of encapsulated cells, and more knowledge about factors affecting the transplanted genetically engineered cells and about the quality and quantity of their produced therapeutic molecule is critical to optimize and streamline the procedures. These promising pilot data need to be confirmed in a large multicenter study. Limitations include the invasiveness of the procedure (although safety and tolerability have been shown to be very good) and the variability of the NGF release from the encapsulated cells. These may be overcome by ongoing work to improve and stabilize cell survival and release, as well as by improvements in the surgical procedure.

226

M. ERIKSDOTTER AND S. MITRA

CONCLUSION The cholinergic hypothesis was the first theory to explain AD symptomatology and it is still valid. After decades of intense research, the best drug treatments available for AD are still the ChEIs, enhancing acetylcholine transmission. The cholinergic neurons in the basal forebrain are particularly vulnerable to degeneration in AD. Therefore restoration of the signaling from the cholinergic neurons in this region is particularly important. One intervention avenue involves local supplementation of NGF to the basal forebrain in AD patients. Such approaches, using gene therapy with injections of cells or viruses transfected with the NGF gene or cell therapy using encapsulated cells genetically modified to release NGF, have been employed and have shown good safety and tolerability with positive effects on the cholinergic system. Although a robust clinical effect has not been clearly shown, these trials demonstrated the potential of NGF-related therapies for AD. They will pave the way for future therapies involving NGF targeting of the cholinergic system in the basal forebrain and its projections, aiming at different mechanisms of cholinergic signaling in combination with, for example, lifestyle factor interventions. All in all, therapies targeting the cholinergic system will be part of the future therapeutic arsenal for AD and are here to stay.

REFERENCES Ahmed NY, Knowles R, Dehorter N (2019). New insights into cholinergic neuron diversity. Front Mol Neurosci 12: 204. Al-Shawi R, Hafner A, Chun S et al. (2007). ProNGF, sortilin, and age-related neurodegeneration. Ann N Y Acad Sci 1119: 208–215. Atri A (2019). Current and future treatments in Alzheimer’s disease. Semin Neurol 39: 227–240. Ballinger EC, Ananth M, Talmage DA et al. (2016). Basal forebrain cholinergic circuits and signaling in cognition and cognitive decline. Neuron 91: 1199–1218. Bankiewicz KS, Eberling JL, Kohutnicka M et al. (2000). Convection-enhanced delivery of AAV vector in parkinsonian monkeys; in vivo detection of gene expression and restoration of dopaminergic function using pro-drug approach. Exp Neurol 164: 2–14. Bartus RT, Dean 3rd RL, Beer B et al. (1982). The cholinergic hypothesis of geriatric memory dysfunction. Science 217: 408–414. Biane J, Conner JM, Tuszynski MH (2014). Nerve growth factor is primarily produced by GABAergic neurons of the adult rat cortex. Front Cell Neurosci 8: 220. Bishop KM, Hofer EK, Mehta A et al. (2008). Therapeutic potential of CERE-110 (AAV2-NGF): targeted, stable, and sustained NGF delivery and trophic activity on rodent basal forebrain cholinergic neurons. Exp Neurol 211: 574–584.

Bonsi P, Cuomo D, Martella G et al. (2011). Centrality of striatal cholinergic transmission in basal ganglia function. Front Neuroanat 5: 6. Bowen DM, Smith CB, White P et al. (1976). Neurotransmitter-related enzymes and indices of hypoxia in senile dementia and other abiotrophies. Brain 99: 459–496. Bradshaw RA, Pundavela J, Biarc J et al. (2015). NGF and ProNGF: regulation of neuronal and neoplastic responses through receptor signaling. Adv Biol Regul 58: 16–27. Budni J, Bellettini-Santos T, Mina F et al. (2015). The involvement of BDNF, NGF and GDNF in aging and Alzheimer’s disease. Aging Dis 6: 331–341. Campenot RB, MacInnis BL (2004). Retrograde transport of neurotrophins: fact and function. J Neurobiol 58: 217–229. Capsoni S, Tiveron C, Vignone D et al. (2010). Dissecting the involvement of tropomyosin-related kinase a and p75 neurotrophin receptor signaling in NGF deficit-induced neurodegeneration. Proc Natl Acad Sci U S A 107: 12299–12304. Capsoni S, Brandi R, Arisi I et al. (2011). A dual mechanism linking NGF/proNGF imbalance and early inflammation to Alzheimer’s disease neurodegeneration in the AD11 antiNGF mouse model. CNS Neurol Disord Drug Targets 10: 635–647. Capsoni S, Malerba F, Carucci NM et al. (2017). The chemokine CXCL12 mediates the anti-amyloidogenic action of painless human nerve growth factor. Brain 140: 201–217. Castle MJ, Baltanas FC, Kovacs I et al. (2020). Postmortem analysis in a clinical trial of AAV2-NGF gene therapy for Alzheimer’s disease identifies a need for improved vector delivery. Hum Gene Ther 31: 415–422. Cattaneo A, Capsoni S (2019). Painless nerve growth factor: a TrkA biased agonist mediating a broad neuroprotection via its actions on microglia cells. Pharmacol Res 139: 17–25. Chen KS, Gage FH (1995). Somatic gene transfer of NGF to the aged brain: behavioral and morphological amelioration. J Neurosci 15: 2819–2825. Counts SE, Nadeem M, Wuu J et al. (2004). Reduction of cortical TrkA but not p75(NTR) protein in early-stage Alzheimer’s disease. Ann Neurol 56: 520–531. Cuello AC, Pentz R, Hall H (2019). The brain NGF metabolic pathway in health and in Alzheimer’s pathology. Front Neurosci 13: 62. Dautan D, Hacioglu Bay H, Bolam JP et al. (2016). Extrinsic sources of cholinergic innervation of the striatal complex: a whole-brain mapping analysis. Front Neuroanat 10: 1. Davies P, Maloney AJ (1976). Selective loss of central cholinergic neurons in Alzheimer’s disease. Lancet 2: 1403. DiStefano PS, Friedman B, Radziejewski C et al. (1992). The neurotrophins BDNF, NT-3, and NGF display distinct patterns of retrograde axonal transport in peripheral and central neurons. Neuron 8: 983–993. Drachman DA, Leavitt J (1974). Human memory and the cholinergic system. A relationship to aging? Arch Neurol 30: 113–121. Dubois B, Chupin M, Hampel H et al. (2015). Donepezil decreases annual rate of hippocampal atrophy in suspected prodromal Alzheimer’s disease. Alzheimers Dement 11: 1041–1049.

GENE AND CELL THERAPY Eberling JL, Jagust WJ, Christine CW et al. (2008). Results from a phase I safety trial of hAADC gene therapy for Parkinson disease. Neurology 70: 1980–1983. Eriksdotter Jonhagen M, Nordberg A, Amberla K et al. (1998). Intracerebroventricular infusion of nerve growth factor in three patients with Alzheimer’s disease. Dement Geriatr Cogn Disord 9: 246–257. Eriksdotter M, Navarro-Oviedo M, Mitra S et al. (2018). Cerebrospinal fluid from Alzheimer patients affects cell-mediated nerve growth factor production and cell survival in vitro. Exp Cell Res 371: 175–184. Eriksdotter-Jonhagen M, Linderoth B, Lind G et al. (2012). Encapsulated cell biodelivery of nerve growth factor to the basal forebrain in patients with Alzheimer’s disease. Dement Geriatr Cogn Disord 33: 18–28. Eriksdotter-Nilsson M, Skirboll S, Ebendal T et al. (1989). NGF treatment promotes development of basal forebrain tissue grafts in the anterior chamber of the eye. Exp Brain Res 74: 89–98. Ernfors P, Ebendal T, Olson L et al. (1989). A cell line producing recombinant nerve growth factor evokes growth responses in intrinsic and grafted central cholinergic neurons. Proc Natl Acad Sci U S A 86: 4756–4760. Eyjolfsdottir H, Eriksdotter M, Linderoth B et al. (2016). Targeted delivery of nerve growth factor to the cholinergic basal forebrain of Alzheimer’s disease patients: application of a second-generation encapsulated cell biodelivery device. Alzheimers Res Ther 8: 30. Fahnestock M, Michalski B, Xu B et al. (2001). The precursor pro-nerve growth factor is the predominant form of nerve growth factor in brain and is increased in Alzheimer’s disease. Mol Cell Neurosci 18: 210–220. Feng X, Valdearcos M, Uchida Y et al. (2017). Microglia mediate postoperative hippocampal inflammation and cognitive decline in mice. JCI Insight 2: e91229. Ferreira D, Perestelo-Perez L, Westman E et al. (2014). Metareview of CSF Core biomarkers in Alzheimer’s disease: the state-of-the-art after the new revised diagnostic criteria. Front Aging Neurosci 6: 47. Fischer W, Wictorin K, Bjorklund A et al. (1987). Amelioration of cholinergic neuron atrophy and spatial memory impairment in aged rats by nerve growth factor. Nature 329: 65–68. Fjord-Larsen L, Kusk P, Emerich DF et al. (2012). Increased encapsulated cell biodelivery of nerve growth factor in the brain by transposon-mediated gene transfer. Gene Ther 19: 1010–1017. Giacobini E (2002). Long-term stabilizing effect of cholinesterase inhibitors in the therapy of Alzheimer’ disease. J Neural Transm Suppl 181–187. Gielow MR, Zaborszky L (2017). The input-output relationship of the cholinergic basal forebrain. Cell Rep 18: 1817–1830. Ginsberg SD, Che S, Wuu J et al. (2006). Down regulation of trk but not p75NTR gene expression in single cholinergic basal forebrain neurons mark the progression of Alzheimer’s disease. J Neurochem 97: 475–487.

227

Gold M (2017). Phase II clinical trials of anti-amyloid beta antibodies: when is enough, enough? Alzheimers Dement (N Y) 3: 402–409. Gray SJ, Nagabhushan Kalburgi S, McCown TJ et al. (2013). Global CNS gene delivery and evasion of anti-AAVneutralizing antibodies by intrathecal AAV administration in non-human primates. Gene Ther 20: 450–459. Greitz D, Hannerz J (1996). A proposed model of cerebrospinal fluid circulation: observations with radionuclide cisternography. AJNR Am J Neuroradiol 17: 431–438. Grossberg GT, Tong G, Burke AD et al. (2019). Present algorithms and future treatments for Alzheimer’s disease. J Alzheimers Dis 67: 1157–1171. Grothe M, Zaborszky L, Atienza M et al. (2010). Reduction of basal forebrain cholinergic system parallels cognitive impairment in patients at high risk of developing Alzheimer’s disease. Cereb Cortex 20: 1685–1695. Hadaczek P, Kohutnicka M, Krauze MT et al. (2006). Convection-enhanced delivery of adeno-associated virus type 2 (AAV2) into the striatum and transport of AAV2 within monkey brain. Hum Gene Ther 17: 291–302. Hagg T (2007). Intracerebral infusion of neurotrophic factors. Methods Mol Biol 399: 167–180. Hampel H, Mesulam MM, Cuello AC et al. (2018). The cholinergic system in the pathophysiology and treatment of Alzheimer’s disease. Brain 141: 1917–1933. Hao J, Ebendal T, Xu X et al. (2000). Intracerebroventricular infusion of nerve growth factor induces pain-like response in rats. Neurosci Lett 286: 208–212. Hefti F (1986). Nerve growth factor promotes survival of septal cholinergic neurons after fimbrial transections. J Neurosci 6: 2155–2162. Hovens IB, Schoemaker RG, van der Zee EA et al. (2013). Surgery-induced behavioral changes in aged rats. Exp Gerontol 48: 1204–1211. Ioannou MS, Fahnestock M (2017). ProNGF, but not NGF, switches from neurotrophic to apoptotic activity in response to reductions in TrkA receptor levels. Int J Mol Sci 18: 599–611. Isaev NK, Stelmashook EV, Genrikhs EE (2017). Role of nerve growth factor in plasticity of forebrain cholinergic neurons. Biochemistry (Mosc) 82: 291–300. Isik AT, Soysal P, Stubbs B et al. (2018). Cardiovascular outcomes of cholinesterase inhibitors in individuals with dementia: a meta-analysis and systematic review. J Am Geriatr Soc 66: 1805–1811. Iulita MF, Cuello AC (2014). Nerve growth factor metabolic dysfunction in Alzheimer’s disease and Down syndrome. Trends Pharmacol Sci 35: 338–348. Iulita MF, Cuello AC (2016). The NGF metabolic pathway in the CNS and its dysregulation in Down syndrome and Alzheimer’s disease. Curr Alzheimer Res 13: 53–67. Izsvak Z, Chuah MK, Vandendriessche T et al. (2009). Efficient stable gene transfer into human cells by the sleeping beauty transposon vectors. Methods 49: 287–297. Janson C, McPhee S, Bilaniuk L et al. (2002). Clinical protocol. Gene therapy of Canavan disease: AAV-2 vector for neurosurgical delivery of aspartoacylase gene (ASPA) to the human brain. Hum Gene Ther 13: 1391–1412.

228

M. ERIKSDOTTER AND S. MITRA

Jelic V, Winblad B (2016). Alzheimer disease. Donepezil and nursing home placement—benefits and costs. Nat Rev Neurol 12: 11–13. Karami A, Eyjolfsdottir H, Vijayaraghavan S et al. (2015). Changes in CSF cholinergic biomarkers in response to cell therapy with NGF in patients with Alzheimer’s disease. Alzheimers Dement 11: 1316–1328. Kawamoto K, Aoki J, Tanaka A et al. (2002). Nerve growth factor activates mast cells through the collaborative interaction with lysophosphatidylserine expressed on the membrane surface of activated platelets. J Immunol 168: 6412–6419. Kerbler GM, Fripp J, Rowe CC et al. (2015). Basal forebrain atrophy correlates with amyloid beta burden in Alzheimer’s disease. Neuroimage Clin 7: 105–113. Klein RL, Hirko AC, Meyers CA et al. (2000). NGF gene transfer to intrinsic basal forebrain neurons increases cholinergic cell size and protects from age-related, spatial memory deficits in middle-aged rats. Brain Res 875: 144–151. Kordower JH, Winn SR, Liu YT et al. (1994). The aged monkey basal forebrain: rescue and sprouting of axotomized basal forebrain neurons after grafts of encapsulated cells secreting human nerve growth factor. Proc Natl Acad Sci U S A 91: 10898–10902. Latina V, Caioli S, Zona C et al. (2017). Impaired NGF/TrkA Signaling causes early AD-linked presynaptic dysfunction in cholinergic primary neurons. Front Cell Neurosci 11: 68. Levi-Montalcini R (2004). The nerve growth factor and the neuroscience chess board. Prog Brain Res 146: 525–527. Levi-Montalcini R, Hamburger V (1951). Selective growth stimulating effects of mouse sarcoma on the sensory and sympathetic nervous system of the chick embryo. J Exp Zool 116: 321–361. Machado A, Ferreira D, Grothe MJ et al. (2020). The cholinergic system in subtypes of Alzheimer’s disease: an in vivo longitudinal MRI study. Alzheimers Res Ther 12: 51. Martinez-Serrano A, Bjorklund A (1998). Ex vivo nerve growth factor gene transfer to the basal forebrain in presymptomatic middle-aged rats prevents the development of cholinergic neuron atrophy and cognitive impairment during aging. Proc Natl Acad Sci U S A 95: 1858–1863. Martinez-Serrano A, Fischer W, Soderstrom S et al. (1996). Long-term functional recovery from age-induced spatial memory impairments by nerve growth factor gene transfer to the rat basal forebrain. Proc Natl Acad Sci U S A 93: 6355–6360. Masoudi R, Ioannou MS, Coughlin MD et al. (2009). Biological activity of nerve growth factor precursor is dependent upon relative levels of its receptors. J Biol Chem 284: 18424–18433. McKelvey L, Shorten GD, O’Keeffe GW (2013). Nerve growth factor-mediated regulation of pain signalling and proposed new intervention strategies in clinical pain management. J Neurochem 124: 276–289. Mehta D, Jackson R, Paul G et al. (2017). Why do trials for Alzheimer’s disease drugs keep failing? A discontinued drug perspective for 2010-2015. Expert Opin Investig Drugs 26: 735–739.

Mesulam M (1976). A horseradish peroxidase method for the identification of the efferents of acetyl cholinesterasecontaining neurons. J Histochem Cytochem 24: 1281–1285. Mesulam M (2004). The cholinergic lesion of Alzheimer’s disease: pivotal factor or side show? Learn Mem 11: 43–49. Mesulam MM, Mufson EJ, Levey AI et al. (1983). Cholinergic innervation of cortex by the basal forebrain: cytochemistry and cortical connections of the septal area, diagonal band nuclei, nucleus basalis (substantia innominata), and hypothalamus in the rhesus monkey. J Comp Neurol 214: 170–197. Mitra S, Behbahani H, Eriksdotter M (2019). Innovative therapy for Alzheimer’s disease-with focus on biodelivery of NGF. Front Neurosci 13: 38. Mroczko B, Groblewska M, Barcikowska M (2013). The role of matrix metalloproteinases and tissue inhibitors of metalloproteinases in the pathophysiology of neurodegeneration: a literature study. J Alzheimers Dis 37: 273–283. Nemy M, Cedres N, Grothe MJ et al. (2020). Cholinergic white matter pathways make a stronger contribution to attention and memory in normal aging than cerebrovascular health and nucleus basalis of Meynert. Neuroimage 211: 116607. Niewiadomska G, Mietelska-Porowska A, Mazurkiewicz M (2011). The cholinergic system, nerve growth factor and the cytoskeleton. Behav Brain Res 221: 515–526. Nordstrom P, Religa D, Wimo A et al. (2013). The use of cholinesterase inhibitors and the risk of myocardial infarction and death: a nationwide cohort study in subjects with Alzheimer’s disease. Eur Heart J 34: 2585–2591. Noroozian Z, Xhima K, Huang Y et al. (2019). MRI-guided focused ultrasound for targeted delivery of rAAV to the brain. Methods Mol Biol 1950: 177–197. Olson L, Nordberg A, von Holst H et al. (1992). Nerve growth factor affects 11C-nicotine binding, blood flow, EEG, and verbal episodic memory in an Alzheimer patient (case report). J Neural Transm Park Dis Dement Sect 4: 79–95. Pavlov VA, Wang H, Czura CJ et al. (2003). The cholinergic anti-inflammatory pathway: a missing link in neuroimmunomodulation. Mol Med 9: 125–134. Peng S, Wuu J, Mufson EJ et al. (2004). Increased proNGF levels in subjects with mild cognitive impairment and mild Alzheimer disease. J Neuropathol Exp Neurol 63: 641–649. Pepeu G, Grazia Giovannini M (2017). The fate of the brain cholinergic neurons in neurodegenerative diseases. Brain Res 1670: 173–184. Poduslo JF, Curran GL (1996). Permeability at the blood-brain and blood-nerve barriers of the neurotrophic factors: NGF, CNTF, NT-3, BDNF. Brain Res Mol Brain Res 36: 280–286. Rafii MS, Baumann TL, Bakay RA et al. (2014). A phase1 study of stereotactic gene delivery of AAV2-NGF for Alzheimer’s disease. Alzheimers Dement 10: 571–581. Rafii MS, Tuszynski MH, Thomas RG et al. (2018). Adenoassociated viral vector (serotype 2)-nerve growth factor for patients with Alzheimer disease: a randomized clinical trial. JAMA Neurol 75: 834–841. Religa D, Fereshtehnejad SM, Cermakova P et al. (2015). SveDem, the Swedish dementia registry—a tool for

GENE AND CELL THERAPY improving the quality of diagnostics, treatment and care of dementia patients in clinical practice. PLoS One 10: e0116538. Rosenberg MB, Friedmann T, Robertson RC et al. (1988). Grafting genetically modified cells to the damaged brain: restorative effects of NGF expression. Science 242: 1575–1578. Samaranch L, Salegio EA, San Sebastian W et al. (2012). Adeno-associated virus serotype 9 transduction in the central nervous system of nonhuman primates. Hum Gene Ther 23: 382–389. Schmitz TW, Nathan Spreng R, Alzheimer’s Disease Neuroimaging I (2016). Basal forebrain degeneration precedes and predicts the cortical spread of Alzheimer’s pathology. Nat Commun 7: 13249. Schneider L (2020). A resurrection of aducanumab for Alzheimer’s disease. Lancet Neurol 19: 111–112. Schwab ME, Suda K, Thoenen H (1979). Selective retrograde transsynaptic transfer of a protein, tetanus toxin, subsequent to its retrograde axonal transport. J Cell Biol 82: 798–810. Secnik J, Schwertner E, Alvarsson M et al. (2020). Cholinesterase inhibitors in patients with diabetes mellitus and dementia: an open-cohort study of 23,000 patients from the Swedish Dementia Registry. BMJ Open Diabetes Res Care 8: e000833. https://doi.org/10.1136/ bmjdrc-2019-000833. Segal RA (2003). Selectivity in neurotrophin signaling: theme and variations. Annu Rev Neurosci 26: 299–330. Servick K (2019). Doubts persist for claimed Alzheimer’s drug. Science 366: 1298. Shytle RD, Mori T, Townsend K et al. (2004). Cholinergic modulation of microglial activation by alpha 7 nicotinic receptors. J Neurochem 89: 337–343. Sobottka B, Reinhardt D, Brockhaus M et al. (2008). ProNGF inhibits NGF-mediated TrkA activation in PC12 cells. J Neurochem 107: 1294–1303. Tan ECK, Johnell K, Garcia-Ptacek S et al. (2018). Acetylcholinesterase inhibitors and risk of stroke and death in people with dementia. Alzheimers Dement 14: 944–951. Tan ECK, Johnell K, Bell JS et al. (2020). Do Acetylcholinesterase inhibitors prevent or delay psychotropic prescribing in people with dementia? Analyses of the Swedish Dementia Registry. Am J Geriatr Psychiatry 28: 108–117. Teipel SJ, Cavedo E, Grothe MJ et al. (2016). Predictors of cognitive decline and treatment response in a clinical trial on suspected prodromal Alzheimer’s disease. Neuropharmacology 108: 128–135. Teipel SJ, Cavedo E, Hampel H et al. (2018). Basal forebrain volume, but not hippocampal volume, is a predictor of global cognitive decline in patients with Alzheimer’s disease treated with cholinesterase inhibitors. Front Neurol 9: 642.

229

Teles-Grilo Ruivo LM, Mellor JR (2013). Cholinergic modulation of hippocampal network function. Front Synaptic Neurosci 5: 2. Tiveron C, Fasulo L, Capsoni S et al. (2013). ProNGF\NGF imbalance triggers learning and memory deficits, neurodegeneration and spontaneous epileptic-like discharges in transgenic mice. Cell Death Differ 20: 1017–1030. Tuszynski MH, Blesch A (2004). Nerve growth factor: from animal models of cholinergic neuronal degeneration to gene therapy in Alzheimer’s disease. Prog Brain Res 146: 441–449. Tuszynski MH, U HS, Amaral DG et al. (1990). Nerve growth factor infusion in the primate brain reduces lesioninduced cholinergic neuronal degeneration. J Neurosci 10: 3604–3614. Tuszynski MH, Smith DE, Roberts J et al. (1998). Targeted intraparenchymal delivery of human NGF by gene transfer to the primate basal forebrain for 3 months does not accelerate beta-amyloid plaque deposition. Exp Neurol 154: 573–582. Tuszynski MH, Thal L, Pay M et al. (2005). A phase 1 clinical trial of nerve growth factor gene therapy for Alzheimer disease. Nat Med 11: 551–555. Tuszynski MH, Yang JH, Barba D et al. (2015). Nerve growth factor gene therapy: activation of neuronal responses in Alzheimer disease. JAMA Neurol 72: 1139–1147. Wahlberg LU, Lind G, Almqvist PM et al. (2012). Targeted delivery of nerve growth factor via encapsulated cell biodelivery in Alzheimer disease: a technology platform for restorative neurosurgery. J Neurosurg 117: 340–347. Washington University Press release (2020). Investigational drugs didn’t slow memory loss, cognitive decline in rare, inherited Alzheimer’s, initial analysis indicates. Weber-Adrian D, Kofoed RH, Chan JWY et al. (2019). Strategy to enhance transgene expression in proximity of amyloid plaques in a mouse model of Alzheimer’s disease. Theranostics 9: 8127–8137. Whitehouse PJ, Price DL, Clark AW et al. (1981). Alzheimer disease: evidence for selective loss of cholinergic neurons in the nucleus basalis. Ann Neurol 10: 122–126. Whitehouse PJ, Price DL, Struble RG et al. (1982). Alzheimer’s disease and senile dementia: loss of neurons in the basal forebrain. Science 215: 1237–1239. Woolf NJ (1991). Cholinergic systems in mammalian brain and spinal cord. Prog Neurobiol 37: 475–524. World Health Organization (2019). Dementia [online]. Switzerland: World Health Organization. Available: http:// www.who.int/mediacentre/factsheets/fs362/en/. [Accessed May 14 2019]. Zhou K, Cherra 3rd SJ, Goncharov A et al. (2017). Asynchronous cholinergic drive correlates with excitationinhibition imbalance via a neuronal Ca(2+) sensor protein. Cell Rep 19: 1117–1129.

This page intentionally left blank

Section 3 The circadian system

This page intentionally left blank

Handbook of Clinical Neurology, Vol. 179 (3rd series) The Human Hypothalamus: Anterior Region D.F. Swaab, F. Kreier, P.J. Lucassen, A. Salehi, and R.M. Buijs, Editors https://doi.org/10.1016/B978-0-12-819975-6.00013-3 Copyright © 2021 Elsevier B.V. All rights reserved

Chapter 15

The circadian system: From clocks to physiology RUUD M. BUIJS1*, EVA C. SOTO TINOCO1, GABRIELA HURTADO ALVARADO1, AND CAROLINA ESCOBAR2 1

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 2

Faculty of Medicine, Universidad Nacional Autónoma de M
exico (UNAM), Ciudad de M
exico, Mexico

Abstract The circadian system, composed of the central autonomous clock, the suprachiasmatic nucleus (SCN), and systems of the body that follow the signals of the SCN, continuously change the homeostatic set points of the body over the day–night cycle. Changes in the body’s physiological state that do not agree with the time of the day feedback to the hypothalamus, and provide input to the SCN to adjust the condition, thus reaching another set point required by the changed conditions. This allows the adjustment of the set points to another level when environmental conditions change, which is thought to promote adaptation and survival. In fasting, the body temperature drops to a lower level only at the beginning of the sleep phase. Stressful conditions raise blood pressure relatively more during the active period than during the rest phase. Extensive, mostly reciprocal SCN interactions, with hypothalamic networks, induce these physiological adjustments by hormonal and autonomic control of the body’s organs. More importantly, in addition to SCN’s hormonal and autonomic influences, SCN induced behavior, such as rhythmic food intake, induces the oscillation of many genes in all tissues, including the so-called clock genes, which have an essential role as a transcriptional driving force for numerous cellular processes. Consequently, the light–dark cycle, the rhythm of the SCN, and the resulting rhythm in behavior need to be perfectly synchronized, especially where it involves synchronizing food intake with the activity phase. If these rhythms are not synchronous for extended periods of times, such as during shift work, light exposure at night, or frequent night eating, disease may develop. As such, our circadian system is a perfect illustration of how hypothalamic-driven processes depend on and interact with each other and need to be in seamless synchrony with the body’s physiology.

INTRODUCTION The need to synchronize our activities with the light–dark cycle Light on earth is essential for life; it is necessary to produce energy, e.g., by photosynthesis or the generation of warmth. Therefore, the development of life under the eternal light–dark cycle induced by the rotation of the earth has promoted the development of biological systems that can use light input as a synchronizing cue to

organize activity, inactivity, and energy consumption. These light susceptible systems allowed organisms to be active during a specific phase of the light–dark period and help anticipate the sun’s rising and setting, coupling it to the moment of energy consumption and rest. This arrangement of light input, connected to behavior or directly to locomotor systems, has allowed organisms to determine the most favorable moment to be active, to forage for food, and to avoid predators. A simple example of this is the swimming behavior of a marine zooplankton;

*Correspondence to: Ruud M. Buijs, Ph.D., Departamento de Biología Celular y Fisiología, Instituto de Investigaciones Biom
edicas, UNAM. Circuito Mario de la Cueva s/n, Ciudad Universitaria, Coyoacán, CP 04510 Ciudad de M
exico, Mexico, Tel: +152-55-56228958, E-mail: [email protected]

234

R.M. BUIJS ET AL.

these tiny shrimp-like creatures need to feed near the surface of the sea, where most food is available. However, when the sun is shining, the UV light of the sun will damage it. Therefore, the animal’s movement is inhibited by light, resulting in the sinking of the animal to lower levels where light is less or absent. The mechanism of this light-induced inhibition of activity is as follows: In the absence of light, the animal starts producing melatonin (also our “dark” hormone), which initiates muscle contraction driving the animal up until the animal gets exposed to light. This stops the production of melatonin, followed by the arrest of muscle contraction, inducing the animal to sink into the deeper regions where darkness prevails, and the whole cycle starts again (Tosches et al., 2014). Later in evolution, light could no longer penetrate whole complex organisms. Consequently, specific mechanisms, i.e., clock genes, developed, allowing organisms to still anticipate the sunset and sunrise. Consequently, now all organisms can anticipate the moment when they need to become active and eat, and when they need to sleep. This predictive capacity was so crucial for survival that evolutionary pressure has resulted in the development of an entire circadian system, composed of cellular and molecular mechanisms that enable the organism to arrange its activity according to the light–dark cycle, even when those changes in light are absent. This rhythmic anticipatory process is made possible by the presence of clock genes that through the interaction of their transcriptional–translational feedback loops have a periodicity of 24 h. Ultimately, their protein products can steer functional cyclic processes within the cell with a cycle of 24 h (Beytebiere et al., 2019). The rhythmicity of these clock genes depends on two things: the moment of energy use by and energy availability for the organs; clock genes prepare cells for the moments when the energy enters, or is needed. This becomes clear by the observation that besides being modulated by the light–dark cycle, a significant part of the clock genes is driven by metabolic cues (Dibner and Schibler, 2015). The circadian system’s importance for many fundamental functions in the body was recognized by awarding the Nobel Prize in Physiology and Medicine to three investigators (Jeffrey C. Hall, Michael Rosbash, and Michael W. Young) who were crucial for the discovery of the molecular mechanism of the circadian clock. As the committee indicated in the justification of their Prize in Physiology and Medicine: “Chronic misalignment between our lifestyle and the rhythm dictated by our inner timekeeper is associated with increased risk for various diseases.” Therefore, in the present review, most attention will be given to the importance of timing for our physiology and health in general and less to the more detailed molecular mechanisms.

The central clock as a coordinator of physiology The circadian system with all its components can drive rhythmicity, synchronization, and anticipation. For the sake of clarity and to focus on the importance of human physiology, we will only pay attention to the circadian system in mammals. In mammals, the circadian system consists of a central biological clock, the suprachiasmatic nucleus (SCN), and other brain and body structures that follow its signals. The location of the SCN in the central nervous system, at the base of the hypothalamus, is fundamental for the circadian system’s functioning. This strategic location allows the SCN to integrate light dark information and interact with a wide variety of hypothalamic nuclei, thus influencing all aspects of body homeostasis by synchronizing behavior, via autonomic and hormonal outputs, with the functionality of the organs (Fig. 15.1). There is another functional advantage of the location of the SCN. As its name indicates, it is located above the optic chiasm, receiving direct light input from the retina, allowing synchronization of its neuronal activity with the light–dark cycle, and imposing this synchrony to target structures (Jones et al., 2018). The importance of the circadian system for the homeostasis of the organism is illustrated by the development of the SCN as a central autonomous clock, that is able to synchronize physiology, in particular hormone secretion and behavior, even in conditions without a clear light–dark cycle (Kalsbeek et al., 1996; Buijs and Kalsbeek, 2001; Perreau-Lenz et al., 2003, 2004).

CLOCK GENES: THE DRIVING FORCE BEHIND THE RHYTHM IN THE SCN AND OF THE METABOLISM IN PERIPHERAL CELLS As explained previously, clock genes and the specifically timed expression pattern of their clock gene proteins are essential for the organization of circadian rhythms and SCN neuronal activity. Although clock genes are also expressed in other neurons, these neurons do not have an endogenous rhythm. The specific organization of SCN neurons and their glia cells (Herzog et al., 2017; Jones et al., 2018; Brancaccio et al., 2019) makes the SCN a unique brain region with autonomous rhythm capacity. Neurons and astrocytes of the SCN have an activity rhythm of approximately 24 h, and that neuronal activity is synchronized on a daily basis by light input from the retina (Brancaccio et al., 2019). The circadian rhythm of the electrical activity of SCN neurons promotes the rhythmic release of SCN neurotransmitters, which ultimately drives the activity of target nuclei in the hypothalamus. Specific clock gene knockout (KO)

Fig. 15.1. Diagram of the rat brain showing the pathways leading to melatonin and corticosterone secretion. The suprachiasmatic nucleus (SCN) sends out signals of vasopressin (AVP), vasoactive intestinal peptide (VIP), GABA and glutamate (Glut) to the paraventricular nucleus (PVN) to stimulate or inhibit distinct preautonomic neurons that project to the intermedio lateral column (IML) of the spinal cord where sympathetic motor neurons project to the superior cervical ganglion (SCG), which neurons release noradrenalin and stimulate melatonin secretion at night. Or IML neurons project directly to the adrenal where they induce the secretion of corticosterone just before the active period. In addition, the circadian rhythm of corticosterone secretion needs the presence of adrenocorticotrophic hormone (ACTH) action on the adrenal cortex. ACTH secretion is induced by the influence of corticotrophin-releasing hormone (CRH) in the median eminence but is hardly rhythmic.

236

R.M. BUIJS ET AL.

mouse models have demonstrated the importance of clock-protein products for the rhythmicity of the SCN (van der Horst et al., 1999; McDearmon et al., 2006). By eliminating clock genes, the SCN completely loses its endogenous rhythm in electrical activity and can no longer impose a rhythm on its target brain areas or target tissues in constant darkness. Nevertheless, even without clock genes, the SCN, as an aggregate of neurons with the same connections and output, still receives light input from the retina. It remains functionally responsive to light and transmits this light signal to hypothalamic target areas, ultimately resulting in a light-induced inhibition of activity. This is illustrated by the observation that when both the Cry1,2 clock genes are lost, these knock out animals lose their rhythm in constant dark conditions. However, when a Cry1–2 double KO animal is exposed to light, it immediately stops its activity. Consequently, a light–dark cycle induces a perfect sleep-activity rhythm in the animal. These data demonstrate that without clock genes, the SCN still responds to the light input and can inhibit the animal’s locomotor activity via its output to other brain areas (van der Horst et al., 1999). When the SCN is lesioned or dysfunctional, the animal cannot respond to light anymore and becomes arrhythmic, irrespective of the light conditions (Buijs et al., 1993), demonstrating that an intact SCN is needed to suppress locomotor activity by light. Surprisingly, some other clock mutant animals do not show the same behavior as the Cry1,2 double KO. For example, Bmal1 KO animals, the most widely used clock mutant animal, hardly responds to light and does not inhibit its activity when exposed a normal light–dark cycle. This happens even when Bmal1 is removed from the forebrain only (Bunger et al., 2000; McDearmon et al., 2006; Rakai et al., 2014), indicating that the SCN cannot respond to the light input from the retina anymore. This hampered response to light indicates that in these Bmal1 KO animals, the functionality of the SCN neurons is severely impaired. Consequently, Bmal1 knockouts in other neurons or organs may disrupt the functionality of the organ, not because the clock cannot work anymore but rather because a critical regulatory gene is removed. Despite this drawback, organ-specific Bmal1 KOs are widely used to study the clock’s possible functions in organs, tissues, or brain areas. Indeed, often the targeted organs of these tissue-specific Bmal1 KO’s do not function properly anymore (Lamia et al., 2008; Paschos and FitzGerald, 2010; Paschos et al., 2012; Orozco-solis et al., 2016). The interpretation that the Bmal1 KO induced-dysfunction is due to a missing clock mechanism is therefore doubtful. Instead, the adequate interpretation should be that an essential transcription factor in the cells is missing resulting in dysfunctional tissue. Simultaneously, these studies also indicate that

the rhythmicity of these genes provides an essential function in these cells, possibly contributing to a specific activity that is rhythmic. Consequently, clock genes’ essential function is to synchronize certain activities in the organ’s cells, thus allowing the organ to anticipate the approach of its activity. As a consequence, clock genes are sensitive to stimuli that demand energy from the cell. Under normal conditions, all these functions are synchronized by the SCN and the light–dark cycle, resulting in synchronized rhythms in food intake, locomotor activity, temperature, and hormone secretion.

CIRCADIAN RHYTHMS IN PHYSIOLOGY: CONTINUOUSLY CHANGING HOMEOSTASIS Already some years ago, Nicholas Mrosovsky coined the term rheostasis for naturally occurring, e.g., seasonal and daily time-induced changes in homeostasis (Mrosovsky, 1990). It is clear now that homeostasis is continuously changing under the influence of the SCN. However, how can the SCN in interaction with other brain areas drive the temporal organization of our homeostasis? For instance, glucose levels that are fine in the morning are an indication of hyperglycemia if the same levels appear before we go to sleep. Another example is the circadian rhythm in temperature, whereby a temperature increase anticipates the activity period and a temperature decrease anticipates the sleep phase. The decrease in temperature is under a strong influence of the metabolic condition of the individual. The temperature dip before and at the beginning of the sleep phase is more pronounced in fasting conditions than in the fed state, and simultaneously, it is also associated with a more pronounced decrease in heart rate and blood glucose levels (Yoda et al., 2000; La Fleur et al., 2001; Scheer et al., 2005). In contrast, at the end of the sleep phase, the temperature increases until its normal preactivity levels, even when the individual continues to fast. Only recently, insight has been gained into how the SCN can be involved in these processes. The SCN has extensive reciprocal contacts with several brain areas; these reciprocal connections serve to inform the SCN about the actual situation of the physiology. This feedback results in an adaptation of the output of the SCN to accommodate the physiology to the current needs of the body at that time of the day (Buijs et al., 2017, 2019). For example, the SCN has a reciprocal interaction with the arcuate nucleus, also located in the hypothalamus. The arcuate nucleus, together with the median eminence (ME), wis a circumventricular organ (CVO), meaning that it is located near the cerebral ventricles and has a less strict blood–brain barrier, whereby the ME contains a lot of fenestrated capillaries. This

THE CIRCADIAN SYSTEM: FROM CLOCKS TO PHYSIOLOGY characteristic places the arcuate nucleus in a privileged position, allowing it to continuously monitor circulating metabolic information (Gropp et al., 2005; Dietrich et al., 2015; Buijs et al., 2017; see Chapter 15 in Volume 180). We have shown that the interaction between the SCN and arcuate nucleus is essential for the organization of the rhythm in temperature, which is highly associated with the animal’s metabolic conditions. On the one hand, the SCN imposes a rhythm on the activity of the arcuate a-MSH neurons, which activities are essential for maintaining a high temperature at the end of the activity phase. On the other hand, vasopressin projections from the SCN to one of the primary brain areas involved in temperature regulation, i.e., the medial preoptic area (MPA), are essential for the decrease in temperature at the beginning of the sleep phase (Guzmán-Ruiz et al., 2014; Guzman-Ruiz et al., 2015).

RHYTHMIC SECRETION OFHORMONES: A REFLECTION OF SCN ACTIVITY Through its connections with other hypothalamic nuclei, the SCN has vast possibilities to influence the secretion of multiple hormones produced by hypothalamic neurons. Other hormone rhythms, such as blood insulin or vasopressin rhythm, are generated by behavior modulated by the SCN, such as food or water intake, or general activity. Many hormones show a precise circadian rhythm directly driven by the SCN. For example, melatonin secretion is induced by neuronal activity of glutamatergic SCN neurons to the preautonomic neurons in the paraventricular nucleus (PVN), which, via autonomic sympathetic output, drives the melatonin secretion from the pineal gland (Teclemariam Mesbah et al., 1999; Perreau-Lenz et al., 2004; see also Chapter 4 in Volume 180). Surprisingly, these glutamatergic SCN neurons are always active and give a constant stimulus for melatonin secretion, which raises the question as to how the strong influence of light on the inhibition of melatonin secretion is executed. In a series of studies, Perreau-Lenz et al. (2004, 2005) demonstrated that these preautonomic PVN neurons also receive input from GABA-ergic neurons of the SCN, and it is their inhibitory activity that prevents the activation of the sympathetic output to the pineal during the day. This GABA-ergic activity is essential for the timing of melatonin secretion, that is, low during daytime and high at night (Fig. 15.1). This “simple” control of melatonin secretion might be related to the fact that in all organisms, melatonin is the “dark hormone” and is not (or only very little) influenced by other hormones or behaviors. The feedback from physiological systems permits the SCN to adjust hormone levels and other physiological

237

processes to the body’s current condition; this feedback also allows it to influence the level of hormone secretion acutely. To illustrate this and the role of the SCN in this pattern, two examples will be given, i.e., the control of cortisol or corticosterone secretion and of luteinizing hormone (LH) secretion.

Cortisol secretion: Associated with the onset of activity or with food intake The secretion of cortisol in humans and corticosterone in rodents show a precise circadian rhythm with higher circulating levels anticipating the activity period. Several studies indicate that this circadian peak in cortisol is not driven by adrenocorticotropic hormone (ACTH), but rather by the sympathetic innervation of the adrenal (Engeland and Arnhold, 2005). Already the early studies of Berson and Yalow (1968) have demonstrated that the blood levels of ACTH in humans hardly show a rhythm, in contrast to cortisol, which shows a high-amplitude rhythm. In rodents where also ACTH does not show a pronounced rhythm, the corticosterone peak is driven by SCN neurotransmitters influencing PVN preautonomic neurons, projecting to sympathetic autonomic neurons that innervate the adrenal gland (Kalsbeek et al., 1996; Buijs et al., 1999; Ishida et al., 2005). The lack of ACTH rhythm indicates that also in humans, the direct sympathetic drive to the adrenal is responsible for the circadian peak in cortisol (Kudielka et al., 2007; see also Chapter 4 in Volume 180), making ACTH a permissive factor (Fig. 15.1). Also, the stress response is under circadian control, with lower cortisol responses to stressors at the beginning of the activity period and higher responses before sleep (Kudielka et al., 2004). In rodents, the same was shown (Buijs et al., 1993). Superimposed on the circadian rhythm, a pulsatile rhythm of ultradian, near-hourly glucocorticoid oscillations induce cyclic glucocorticoid receptor (GR)-mediated transcriptional regulation, or gene pulsing, in a.o. the hippocampus. These oscillations represent an additional level of adaptation to a changing environment and have been implicated in cortical plasticity, anxiety, and hippocampal neurogenesis (Conway-Campbell et al., 2010; Fitzsimons et al., 2016; Schouten et al., 2020). There is further a close relationship between the circadian secretion of cortisol and the secretion of glucose. Both show the same timing in rhythmicity; both are often released after the same stimuli illustrating the sympathetic control of both cortisol and glucose secretion. An interesting, but little understood example of the interaction of other systems with the rhythmic release of cortisol is the influence on metabolism. In obese people, the cortisol rhythm shows a low amplitude, whereby the cortisol levels at all time points are low compared to

238

R.M. BUIJS ET AL.

those observed in nonobese people. These low cortisol levels contrast with increased ACTH blood levels, that just as in nonobese persons, do not show a clear rhythm (Jessop et al., 2001). These authors attribute this discrepancy to a lower sensitivity to mineralocorticoid feedback in obese people. However, another possibility is that in obese people, the sympathetic neurons projecting to the abdominal cavity, where the adrenal gland is located, are less active, and ultimately resulting in less cortisol secretion (Kreier et al., 2002, 2003). Not all parts of the body receive autonomic input from the same autonomic neurons (Kreier et al., 2002, 2003, 2005). Thus, while the adrenal might receive a lower sympathetic input, other organs may receive a higher, compensatory, sympathetic input. Indeed, obese people have higher blood pressure (BP) (Bjorntorp et al., 1999), indicating a higher sympathetic output to the thoracic cavity. Also related to metabolism is the observation that the cortisol response to meal consumption shows a circadian pattern, with higher responses in the evening than in the morning (Van Cauter et al., 1992). This illustrates that cortisol may increase considerably in response to a stimulus, in spite of the fact that the circadian system sets the basal levels to the lowest amount. The importance of the arcuate nucleus as a sensor of metabolic information (see Chapter 15 in Volume 180), together with the mealinduced changes in cortisol secretion, draws attention to the arcuate nucleus as an area of importance also for the control of cortisol secretion.

Circumventricular organs: Monitoring the circulation for cortisol and metabolites To adjust the circulating level of cortisol, it is essential to precisely monitor its concentration and transmit this information to the areas in the brain involved in the release of ACTH and cortisol. How circulating cortisol may enter the brain is still under discussion. In the blood–brain barrier, multidrug resistance P-glycoprotein (MDR) is supposed to play a role in the transport of cortisol and corticosterone into the brain (Karssen et al., 2001). However, these MDRs are also proposed to play a role in removing corticosteroids from the brain (Uhr et al., 2002). Also, up till recently, the negative feedback of cortisol was supposed to occur at the PVN level where corticotrophin-releasing hormone (CRH) neurons control the secretion of ACTH from the pituitary. These CRH neurons express GRs and diminish their activity and production of CRH under the influence of glucocorticoids (Erkut et al., 1998). However, as was described previously, glucocorticoid secretion is controlled mainly by preautonomic neurons in the PVN projecting to the adrenal. These neurons do not express GR and thus are not directly sensitive to glucocorticoid feedback (Leon-Mercado et al., 2017).

Moreover, since glucocorticoids do not easily penetrate the BBB, another question arises: If there is a fast release of cortisol, is there also fast feedback? To answer this question, attention must be focused on the four sensory CVOs in the rat brain: the organum vasculosum of the lamina terminalis (OVLT), the subfornical organ (SFO), the ME-arcuate nucleus complex (ARC), and the area postrema (AP). Since they are in direct contact with circulating blood, the function of these four structures is largely associated with the sensing of metabolites. The OVLT and SFO are mainly involved in the detection of the mineral balance of the body (Mimee et al., 2013; Gizowski et al., 2016; Gizowski and Bourque, 2020), while the ME-ARC and AP are important for the detection of changes in metabolic measures (Larsen et al., 1997; Langlet et al., 2013). Two of the CVOs, the OVLT and ME-ARC, have extensive reciprocal interaction with the SCN, while for the SFO and AP, this has not been demonstrated, but all have extensive contacts with the PVN. Of all these CVOs, the ARC has a high concentration of GR, making it the logical candidate for corticosterone’s fast negative feedback. Using microdialysis probes inside the ARC and infusing GR and MR agonists and antagonists at different times of the day, we have demonstrated that the MR is essential for the negative feedback when corticosterone is low. In contrast, the GR is essential for the negative feedback when corticosterone is high. Notably, the increase or suppression in circulating corticosterone levels by MR or GR (Ant)agonists took place without any change in circulating ACTH (Leon-Mercado et al., 2017), confirming that the hypothalamic output to the adrenal executed those changes. This observation illustrates that the brain’s CVOs play a crucial role in the sensing of circulating molecules and the signaling of those levels to regulatory centers in the brain stem and hypothalamus.

LH secretion Probably no other hormone is under the influence of so many factors as the secretion of LH in females. In rodents, the experimental evidence showing that the SCN drives the preovulatory LH release is overwhelming. The pioneering study of Everett and Sawyer evidenced for the very first time the circadian control of ovulation (Everett and Sawyer, 1950), showing that an injection of Nembutal at a crucial moment before ovulation postponed the LH surge by 24 h in rats. Furthermore, ovariectomized and estrogen-treated animals show a daily surge in LH (Legan and Karsch, 1975), providing an excellent model to study how the SCN influences LH secretion. In the following years, more evidence accumulated. Apart from the gonadotrophin-releasing hormone (GnRH) system and the circadian system,

THE CIRCADIAN SYSTEM: FROM CLOCKS TO PHYSIOLOGY several systems were shown to be involved in the circadian control of LH secretion. Although in humans, ovulation only takes place once every 28 days, similar control mechanisms exist. Spontaneous initiation of the preovulatory LH surge in women generally occurs in the morning together with the cortisol peak, indicating the importance of the SCN in the timing of human ovulation (Cahill et al., 1998; Kerdelhue et al., 2002). The involvement of the SCN in the control of human ovulation was challenged by the observation that amplitude and frequency of pulsatile LH secretion did not vary over a 24 h period in premenopausal women when studied under constant laboratory conditions (Klingman et al., 2011). However, very little can be concluded from that study since there was no inclusion of women under normal LD conditions. As indicated before, the menstrual cycle is influenced by many factors. In particular, the conditions used in that study, including constant light and constant activity for 32 h, could be enough to disrupt a cycle that also depends on melatonin (Scarinci et al., 2019). The complexity of the LH surge timing becomes evident when we examine the contribution of different SCN neurons to the oestrus cycle (Fig. 15.2). SCN vasopressin

239

neurons project to the MPA where (even in SCN lesioned animals) vasopressin infusion can induce an LH surge (Palm et al., 1999). Moreover, in SCN-intact, but ovariectomized, estradiol-treated animals, vasopressin could only induce this LH surge within a specific time window (Palm et al., 2001). Both studies indicate the importance of vasopressin stimulation for the LH surge. The SCN vasopressin projections to Kisspeptin neurons in the MPA (Vida et al., 2010) probably underlie these effects of vasopressin injection on LH secretion. Also, the SCN has direct vasoactive intestinal peptide (VIP) projections to GnRH neurons located in the MPA (Van Der Beek et al., 1997). These VIP neuronal terminals appose preferential GnRH neurons that show activation by c-Fos during an LH surge (Van Der Beek et al., 1994). In agreement with this, the LH surge is diminished or prevented by an injection of VIP antiserum (which neutralizes the effects of VIP) (van der Beek et al., 1999). Therefore, also the VIP projections from the SCN seem to serve to stimulate the GnRH neurons. Less is known about the SCN prokineticin neurons, but also these SCN neurons might be involved in the (in)direct control of GnRH neurons, since receptors for this peptide are present on estradiol-activated neurons in the MPA (Xiao et al., 2014).

Fig. 15.2. Diagram of the rat brain showing the pathways leading to the rhythm in luteinizing hormone (LH) secretion. The suprachiasmatic nucleus (SCN) sends out signals of vasoactive intestinal peptide (VIP) to gonadotrophin-releasing hormone (GnRH) neurons in the preoptic area (POA), where VIP is a stimulating factor. The SCN also projects with vasopressin (AVP), containing fibers to the Kisspeptin (Kiss) neurons in the POA or arcuate nucleus (ARC), where AVP stimulation of Kiss in turn induces GnRH neuron activation. The SCN also sends VIP projections to the dorsomedial hypothalamus (DMH), where VIP inhibits the RFRP3 neurons that inhibit the GnRH secretion. These actions of the SCN are perfectly timed by the light dark cycle and are synchronized such that at the right moment just before the activity period GnRH is released to stimulate LH secretion. In this synchronization, the sudden drop in estrogen before the LH surge also plays an essential role, since its presence stimulates the production of Kisspeptin in the POA while it inhibits the Kisspeptin in the ARC. Therefore, the drop in Estrogen stimulates Kisspeptin production in the ARC just in time that the peptide can liberate GnRH in the median eminence.

240

R.M. BUIJS ET AL.

Apart from daily rhythms, there are also other essential influences on the LH surge, such as seasonal or metabolic ones. The seasonal influence on reproduction will hardly play a role in most humans, except when we consider the shortage of food, which may be strongly seasonal in some cultures. Several studies have provided insight into how metabolic conditions may play an important role in the organization of the reproductive cycle. Again, we encounter the arcuate nucleus as an essential nucleus for the control of ovulation due to its possibility to monitor the metabolic state of the animal. Arcuate nucleus Kisspeptin neurons, agouty-related peptide, and pro-opioid melanocortin neurons, all project to the MPA, to the dorsomedial nucleus of the hypothalamus (DMH), and to the PVN (see also Chapter 17 in Volume 180). All these structures are involved in the processing of reproductive and metabolic information. As described previously, the arcuate has bidirectional connections with the SCN that are essential for the organization of many circadian rhythms (Yi et al., 2006; Buijs et al., 2017). Such reciprocal connections also exist between the DMH and the SCN (Acosta-Galvan et al., 2011), demonstrating the importance of the interaction between time and metabolism. Furthermore, the extensive projections of SCN, arcuate and DMH to the MPA, emphasize the importance of these areas for controlling reproduction and temperature regulation. Arcuate Kisspeptin-Neurokinin B neurons are under a strong negative influence of estrogen. This is important for temperature regulation, which is especially evident in menopausal women. During menopause, when estrogen levels diminish, arcuate Kisspeptin-Neurokinin B neurons become more active and induce in the MPA, an excess release of Neurokinin B, leading to the activation of the parasympathetic outflow to the blood vessels of the skin resulting in vasodilatation and the feeling of hot flushes (Rometo et al., 2007; Mittelman-Smith et al., 2012a,b; Padilla et al., 2018). Interestingly, the same neurons are also important for the control of metabolism (Padilla et al., 2019), again evidencing the coupling between temperature, reproduction, and metabolism. A similar interaction occurs between the SCN and the DMH, which is also an area where circadian, metabolic, and temperature information is integrated. Here, another population of Kisspeptin-related neurons, RFRP-3 (RFamide-related peptide 3), regulates GnRH neuron activity and gonadotropin secretion. RFRP-3 is known to exert an inhibitory role over the GnRH signaling, although that depends on the species studied. In female Syrian hamsters (Mesocricetus auratus), RFRP-3 neurons have close appositions with SCNderived VP and VIP fibers (Russo et al., 2015), suggesting that the SCN could also be involved in coordinating the inhibition of RFRP-3 neurons, allowing the LH surge

by removing their inhibitory influence. Indeed, only when injected in the evening, VIP suppresses RFRP-3 neuronal activity. This data indicates that the SCN can stimulate GnRH secretion by direct projections to the GnRH neurons and indirectly through inhibition of RFRP-3 neurons, both actions through VIP projections. These examples illustrate that the SCN orchestrates an optimal timing of such an important event as ovulation via multiple targets. Moreover, the SCN VIP neurons receive a dense input (feedback) from the DMH RFRP-3 neurons (AcostaGalvan et al., 2011), that is, essential for the organization of locomotor activity of the animal, which is another behavior preceding ovulation in many animal species. Since there are multisynaptic connections from the SCN to the ovary (Gerendai et al., 2000) and disruption of the autonomic output to the ovaries disrupts the onset of ovulation (Ramírez et al., 2017), it is likely that the SCN is involved in additional aspects related to ovulation. These examples of cortisol and LH regulation reveal that the SCN incorporated in several neuronal networks, synchronizing hormonal rhythms that are essential for metabolism, cardiovascular regulation, and reproduction.

THE SCN CONTROLS THE RHYTHMIC PHYSIOLOGY OF ALL THE ORGANS IN THE BODY In addition to the variation in hormonal regulations, as discussed previously for melatonin, cortisol, and LH, the SCN also organizes the functionality of our organs mainly by setting, via the autonomic nervous system, a level of response, activity, and rest associated with the time of the day. Our basal heart rate, for example, is lower at night than during the day, even under the same activity conditions (Scheer et al., 2004). The direct influence of the SCN on the activity of our organs is probably best illustrated by the observations that light exposure induces critical functional changes in different organs. Heart rate increases when we are exposed to light at night and decreases when we shut our eyes during daytime (Scheer et al., 2003, 2004); the reverse happens in rats, which are nocturnal animals. This influence of light on autonomic output also extends to the control of hormone secretion by different tissues. An obvious example is the pineal, where light via the retinal projections to the SCN removes the SCN stimulatory influence on melatonin secretion (Perreau-Lenz et al., 2003). However, this regulation by light also extends to other organs. As previously mentioned, the adrenal receives light input via the SCN and corticosterone secretion is directly influenced by light (Buijs et al., 1999; Ishida et al., 2005). The liver also receives autonomic input, and here light influences glucose production (Opperhuizen et al., 2019).

THE CIRCADIAN SYSTEM: FROM CLOCKS TO PHYSIOLOGY Finally, light can directly influence the functioning and homeostasis of the skin by modulating clock genes (Buhr et al., 2019) and via the SCN and the autonomic nervous system the physiology of the skin (Fan et al., 2018). These changes may serve to protect the skin from the damage of UV light.

DISTURBANCES IN SCN ACTIVITY TRIGGER PATHOLOGIES Considering, as shown previously, that the biological clock plays an essential role in determining the set points of several aspects of our physiology, any long-term disturbance in the activity of the SCN may have severe repercussions for our health. For instance, it has been shown that older people and people with Alzheimer’s disease have a diminished melatonin secretion (Mirmiran et al., 1989; Uchida et al., 1996), indicating a lower activity of the glutamatergic neurons of the SCN. In this regard, it is interesting that melatonin secretion is also diminished in people with hypertension (Brugger et al., 1995; Zeman et al., 2005), indicating that hypertension may also interfere with SCN neuronal activity. This idea was corroborated in a study that evaluated postmortem tissue from hypertensive people, where substantial reductions in SCN neuronal activity were observed (Goncharuk et al., 2001). In search of an explanation for those observations, we demonstrated in rodents that the SCN is sensitive to increases in BP (Buijs et al., 2014; Romo-Nava et al., 2017; Yilmaz et al., 2018, 2019) and observed that the nucleus of tractus solitarius transmits information about BP increases directly to the SCN. This feedback serves to bring down the BP to normal levels, determined by the time of the day. A clear illustration of the importance of the SCN in the control of BP is that when a stressful stimulus is given to an SCN lesioned animal, there is an exacerbated increase in BP when compared to a sham-operated animal (Buijs et al., 2014; Romo-Nava et al., 2017). The observations that hypertensive and obese people have a more disturbed sleep–wake pattern than nonhypertensive people (Gangwisch et al., 2005, 2006) provides further support for the hypothesis that alterations in our biological clock might be at the core of the recent surge in diabetes and hypertension (Kreier et al., 2003). Furthermore, it was recently shown that the postmortem brain of diabetes type 2 patients show a diminished activity of the SCN (Hogenboom et al., 2019). These observations raise the question of whether these SCN changes in the postmortem hypertensive or diabetic human brain are a cause or consequence of hypertension and diabetes. Recent shifts in human behavior, such as being active and eating during the night, together with the abovedetailed observations that the SCN is sensitive for

241

feedback, suggests that our behavior and consequently changed physiology may be responsible for the observed alterations in the SCN. In the next paragraph, attention is given to the observations that changes in our behavior that are not compatible with the signals of the SCN may result in disease, it is also possible that the use of medicines via their action on the SCN may induce important pathologies as side effect. For example, the use of second-generation antipsychotics (SGA), such as olanzapine, are associated with adverse cardiometabolic side effects contributing to premature mortality in patients (Lieberman et al., 2005). Recently, animal studies have further demonstrated that olanzapine not only activated limbic system areas but also the SCN and through the SCN, parasympathetic output of the hypothalamus (Romo-Nava et al., 2017). As is indicated previously, the SCN selectively coordinates the balance of the autonomic nervous system and influences metabolism in different parts of the body, so alteration of this output may in time promote the development of the metabolic syndrome (Kreier et al., 2003). Parasympathetic activation induces adiposity and an increase in circulating adiponectin, which is also observed with even short-term treatment with olanzapine (Togo et al., 2004). With time, the increased parasympathetic activity induced by olanzapine favors the appearance of cardiometabolic adverse effects like obesity, as well as lipid, insulin, and glucose disturbances (Lieberman et al., 2005) similar to those observed in the metabolic syndrome, whereby in the long term, a compensatory increased sympathetic cardiovascular tone is reported. This observed activation of the SCN by olanzapine in rats could be prevented by the treatment of these animals with melatonin known to reduce the activity of the SCN (Romo-Nava et al., 2017). These observations were done as a follow-up study to explain the observations that melatonin attenuates SGA-induced cardiometabolic effects in patients, particularly those with bipolar disorder without affecting the psychopathological outcome (Romo-Nava et al., 2014). In two other studies, melatonin also mitigated olanzapine-induced cardiometabolic effects in patients diagnosed with schizophrenia and bipolar disorder (Modabbernia et al., 2014; Mostafavi et al., 2014). Taken together, these studies indicate that disturbances in SCN neuronal activity may in the long-term result in important deviations of the normal physiology.

DESYNCHRONY OF BEHAVIOR WITH THE LIGHT–DARK CYCLE: A DEVIATION IN MODERN LIFE LEADS TO A WIDE VARIATION OF PATHOLOGIES The influence of the SCN over physiology is not an immediate matter of life and death, disease, or health.

242

R.M. BUIJS ET AL.

However, every day, it is there setting the thresholds of nearly all systems in a subtle but undeniable way. Consequently, ignoring the signals of our biological clock in a regular way like we do when we do shift work, eat or simply, are in front of a cellular screen at night, makes us prone to develop diseases like obesity, cardiovascular, or immune disorders (Fig. 15.3) (Gangwisch et al., 2006, 2007). In experimental conditions (or in humans driven, e.g., by the demands of our current society, jet lag or shift work, late TV dinner), when food intake shifts to the sleep phase of the individual, such aberrant feeding schedule alters the rhythm of clock genes in the liver and in many other tissues, but not in the SCN (Damiola et al., 2000; Salgado-Delgado et al., 2013; Sabath et al., 2014; Oishi et al., 2017). Therefore, when behavior loses its synchrony with the light–dark cycle and thus deviates from the rhythm of the central biological clock, tissues lose their synchrony with each other and with the SCN. In such conditions, even brain areas can lose their synchrony with the SCN and follow the aberrant rhythm of food (Mukherji et al., 2015).

However, when food intake is restricted to the active phase, these metabolic disturbances do not occur (Salgado-Delgado et al., 2010), even when the animal is forced to be active during its normal resting phase. Therefore, the metabolic effects caused by the aberrant activity pattern of the individual can be overridden, as long as food intake is still in phase with the signals from the SCN. Since the rhythms in the organs follow the food rhythm, this paradigm allows the organs to be in synchrony with the SCN, which continues to follow the LD cycle. These observations in laboratory animals have led to studies in people with the metabolic syndrome; they restricted their food intake to 8–10 h during the day, followed by a night fasting period of at least 14 h. This eating paradigm resulted in improved glucose tolerance and lower BP (Hutchison et al., 2019; Wilkinson et al., 2020). Therefore, the strengthening of the clock signals with a strong rhythm of food in synchrony with the light–dark cycle can revert the metabolic syndrome.

Fig. 15.3. Line drawings represent the circadian rhythms of food intake and activity in rodents (blue) and men (red). Rodents have the activity phase in the dark period, while men have it in the light period. Since circadian clocks in brain and body are synchronized, respectively, by light and food, these clocks are in synchrony when rodents eat in the dark phase and people in the light phase (upper panel). When people or rodents change their activity and/or food intake pattern outside this period, their clocks become desynchronized which results in “modern” diseases like obesity hypertension and diabetes.

THE CIRCADIAN SYSTEM: FROM CLOCKS TO PHYSIOLOGY

CONCLUSION The intertwining of most of our body’s physiological processes, with daily activity and food intake, has resulted in a high evolutionary pressure on the development of the circadian system. This evolutionary pressure culminated in the development of a central biological clock and additional molecular clock mechanisms in all the cells of the body, allowing synchronization of their function to accommodate the needs of the organs. The function of the central biological clock is to synchronize physiology and behavior with the light–dark cycle. The function of the peripheral clock genes is to synchronize the organs with energy uptake and demand. These two rhythms need to be perfectly synchronized, which is the main task of the SCN. When this intricate balance between central and peripheral clock timing is disturbed by behavior or even by light exposure that is not in synchrony with the time of the day, the emanating desynchronization process will result in disease (Fig. 15.3.). Changing the desynchronizing behavior (Salgado-Delgado et al., 2010; Hutchison et al., 2019; Wilkinson et al., 2020), or enforcing the strength of the lost rhythm, i.e., via elevation of the night signal with melatonin (Scheer et al., 2003; Cagnacci et al., 2005; Romo-Nava et al., 2014, 2017), is often sufficient to prevent or reverse such pathologies.

REFERENCES Acosta-Galvan G, Yi C-X, Van Der Vliet J et al. (2011). Interaction between hypothalamic dorsomedial nucleus and the suprachiasmatic nucleus determines intensity of food anticipatory behavior. Proc Natl Acad Sci U S A 108: 5813–5818. https://doi.org/10.1073/pnas.1015551108. Berson SA, Yalow RS (1968). Radioimmunoassay of ACTH in plasma. J Clin Invest 47: 2725–2751. https://doi.org/ 10.1172/JCI105955. Beytebiere JR, Trott AJ, Greenwell BJ et al. (2019). Tissuespecific BMAL1 cistromes reveal that rhythmic transcription is associated with rhythmic enhancer-enhancer interactions. Genes Dev 33: 294–309. https://doi.org/10.1101/ gad.322198.118. Bjorntorp P, Holm G, Rosmond R (1999). Hypothalamic arousal, insulin resistance and type 2 diabetes mellitus. Diabet Med 16: 373–383. Brancaccio M, Edwards MD, Patton AP et al. (2019). Cellautonomous clock of astrocytes drives circadian behavior in mammals. Science 363: 187–192. https://doi.org/10.1126/ science.aat4104. Brugger P, Marktl W, Herold M (1995). Impaired nocturnal secretion of melatonin in coronary heart disease. Lancet 345: 1408. Buhr ED, Vemaraju S, Diaz N et al. (2019). Neuropsin (OPN5) mediates local light-dependent induction of circadian clock genes and circadian Photoentrainment in exposed murine skin. Proc Natl Acad Sci U S A 115: 3478–3487.e4. https://doi.org/10.1073/pnas.1719548115.

243

Buijs RM, Kalsbeek A (2001). Hypothalamic integration of central and peripheral clocks. Nat Rev Neurosci 2: 521–526. https://doi.org/10.1038/35081582. Buijs RM, Kalsbeek A, Van der Woude TP et al. (1993). Suprachiasmatic nucleus lesion increases corticosterone secretion. Am J Physiol Regul Integr Comp Physiol 264: 33–36. https://doi.org/10.1152/ajpregu.1993.264.6.r1186. 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 FN, Cazarez F, Basualdo MC et al. (2014). The suprachiasmatic nucleus is part of a neural feedback circuit adapting blood pressure response. Neuroscience 266: 197–207. https://doi.org/10.1016/j.neuroscience.2014.02.018. Buijs FN, Guzma´n-Ruiz M, Leo´n-Mercado L et al. (2017). Suprachiasmatic nucleus interaction with the arcuate nucleus; Essential for organizing physiological rhythms. eNeuro 4. 1–14. https://doi.org/10.1523/ENEURO.002817.2017. Buijs RM, Guzma´n Ruiz MA, M
endez Herna´ndez R et al. (2019). The suprachiasmatic nucleus; a responsive clock regulating homeostasis by daily changing the setpoints of physiological parameters. Auton Neurosci Basic Clin 218: 43–50. https://doi.org/10.1016/j.autneu.2019.02.001. Bunger MK, Wilsbacher LD, Moran SM et al. (2000). Mop3 is an essential component of the master circadian pacemaker in mammals. Cell 103: 1009–1017. https://doi.org/10.1016/ s0092-8674(00)00205-1. Cagnacci A, Cannoletta M, Renzi A et al. (2005). Prolonged melatonin administration decreases nocturnal blood pressure in women. Am J Hypertens 18: 1614–1618. Cahill DJ, Wardle PG, Harlow CR et al. (1998). Onset of the preovulatory luteinizing hormone surge: diurnal timing and critical follicular prerequisites. Fertil Steril 70: 56–59. https://doi.org/10.1016/s0015-0282(98)00113-7. Conway-Campbell BL, Sarabdjitsingh RA, McKenna MA et al. (2010). Glucocorticoid ultradian rhythmicity directs cyclical gene pulsing of the clock gene period 1 in rat hippocampus. J Neuroendocrinol 22: 1093–1100. https://doi. org/10.1111/j.1365-2826.2010.02051.x. Damiola F, Le Minli N, Preitner N et al. (2000). Restricted feeding uncouples circadian oscillators in peripheral tissues from the central pacemaker in the suprachiasmatic nucleus. Genes Dev 14: 2950–2961. https://doi.org/10.1101/gad. 183500. Dibner C, Schibler U (2015). Circadian timing of metabolism in animal models and humans. J Intern Med 277: 513–527. Dietrich MO, Zimmer MR, Bober J et al. (2015). Hypothalamic Agrp neurons drive stereotypic behaviors beyond feeding. Cell 160: 1222–1232. Engeland WC, Arnhold MM (2005). Neural circuitry in the regulation of adrenal corticosterone rhythmicity. Endocrine 28: 325–332. https://doi.org/10.1385/ENDO:28:3:325. Erkut ZA, Pool C, Swaab DF (1998). Glucocorticoids suppress corticotropin-releasing hormone and vasopressin expression in human hypothalamic neurons. J Clin Endocrinol Metab 83: 2066–2073. https://doi.org/10.1210/jcem.83.6.4881. Everett JW, Sawyer CH (1950). A 24-hour periodicity in the “LH-release apparatus” of female rats, disclosed by

244

R.M. BUIJS ET AL.

barbiturate sedation. Endocrinology 71: 198–218. https:// doi.org/10.1210/endo-47-3-198. Fan SM-Y, Chang Y-T, Chen C-L et al. (2018). External light activates hair follicle stem cells through eyes via an ipRGC-SCN-sympathetic neural pathway. Proc Natl Acad Sci U S A 115: E6880–E6889. https://doi.org/ 10.1073/pnas.1719548115. Fitzsimons CP, Herbert J, Schouten M et al. (2016). Circadian and ultradian glucocorticoid rhythmicity: implications for the effects of glucocorticoids on neural stem cells and adult hippocampal neurogenesis. Front Neuroendocrinol 41: 44–58. https://doi.org/10.1016/j.yfrne.2016.05.001. Gangwisch JE, Malaspina D, Boden-Albala B et al. (2005). Inadequate sleep as a risk factor for obesity: analyses of the NHANES I. Sleep 28: 1289–1296. Gangwisch JE, Heymsfield SB, Boden-Albala B et al. (2006). Short sleep duration as a risk factor for hypertension: analyses of the first National Health and nutrition examination survey. Hypertension 47: 833–839. https://doi.org/ 10.1161/01.HYP.0000217362.34748.e0. Gangwisch JE, Heymsfield SB, Boden-Albala B et al. (2007). Sleep duration as a risk factor for diabetes incidence in a large US sample. Sleep 30: 1667–1673. https://doi.org/ 10.1093/sleep/30.12.1667. Gerendai I, Toth IE, Boldogkoi Z et al. (2000). CNS structures presumably involved in vagal control of ovarian function. J Auton Nerv Syst 80: 40–45. Gizowski C, Bourque CW (2020). Sodium regulates clock time and output via an excitatory GABAergic pathway. Nature 583: 421–424. https://doi.org/10.1038/s41586020-2471-x. Gizowski C, Zaelzer C, Bourque CW (2016). Clock-driven vasopressin neurotransmission mediates anticipatory thirst prior to sleep. Nature 537: 685–688. Goncharuk VD, Van Heerikhuize J, Dai JP et al. (2001). Neuropeptide changes in the suprachiasmatic nucleus in primary hypertension indicate functional impairment of the biological clock. J Comp Neurol 431: 320–330. https://doi.org/10.1002/1096-9861(20010312)431:3< 320:: AID-CNE1073>3.0.CO;2-2. Gropp E, Shanabrough M, Borok E et al. (2005). Agoutirelated peptide-expressing neurons are mandatory for feeding. Nat Neurosci 8: 1289–1291. https://doi.org/10.1038/ nn1548. Guzma´n-Ruiz M, Saderi N, Cazarez-Ma´rquez F et al. (2014). The suprachiasmatic nucleus changes the daily activity of the arcuate nucleus a-MSH neurons in male rats. Endocrinology 155: 525–535. https://doi.org/10.1210/en. 2013-1604. Guzman-Ruiz MA, Ramirez-Corona A, Guerrero-Vargas NN et al. (2015). Role of the suprachiasmatic and arcuate nuclei in diurnal temperature regulation in the rat. J Neurosci 35: 15419–15429. Herzog ED, Hermanstyne T, Smyllie NJ et al. (2017). Regulating the suprachiasmatic nucleus (SCN) circadian clockwork: interplay between cell-autonomous and circuit-level mechanisms. Cold Spring Harb Perspect Biol 9: a027706.

Hogenboom R, Kalsbeek MJ, Korpel NL et al. (2019). Loss of arginine vasopressin- and vasoactive intestinal polypeptide-containing neurons and glial cells in the suprachiasmatic nucleus of individuals with type 2 diabetes. Diabetologia 62: 2088–2093. https://doi.org/10.1007/ s00125-019-4953-7. Hutchison AT, Regmi P, Manoogian ENC et al. (2019). Timerestricted feeding improves glucose tolerance in men at risk for type 2 diabetes: a randomized crossover trial. Obesity (Silver Spring) 27: 724–732. https://doi.org/10.1002/ oby.22449. 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. https://doi.org/ 10.1016/j.cmet.2005.09.009. Jessop DS, Dallman MF, Fleming D et al. (2001). Resistance to glucocorticoid feedback in obesity. J Clin Endocrinol Metab 86: 4109–4114. https://doi.org/10.1210/jcem.86. 9.7826. Jones JR, Simon T, Lones L et al. (2018). SCN VIP neurons are essential for normal light-mediated resetting of the circadian system. J Neurosci 38: 7986–7995. Kalsbeek A, Van Heerikhuize JJ, Wortel J et al. (1996). A diurnal rhythm of stimulatory input to the hypothalamo-pituitary-adrenal system as revealed by timed intrahypothalamic administration of the vasopressin V1 antagonist. J Neurosci 16: 5555–5565. Karssen AM, Meijer OC, van der Sandt IC et al. (2001). Multidrug resistance P-glycoprotein hampers the access of cortisol but not of corticosterone to mouse and human brain. Endocrinology 142: 2686–2694. Kerdelhue B, Brown S, Lenoir V et al. (2002). Timing of initiation of the preovulatory luteinizing hormone surge and its relationship with the circadian cortisol rhythm in the human. Neuroendocrinology 75: 158–163. Klingman KM, Marsh EE, Klerman EB et al. (2011). Absence of circadian rhythms of gonadotropin secretion in women. J Clin Endocrinol Metab 96: 1456–1461. https://doi.org/ 10.1210/jc.2010-2739. 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. https://doi.org/10.1172/JCI0215736. Kreier F, Yilmaz A, Kalsbeek A et al. (2003). Hypothesis: shifting the equilibrium from activity to food leads to autonomic unbalance and the metabolic syndrome. Diabetes 52: 2652–2656. https://doi.org/10.2337/diabetes.52.11.2652. Kreier F, Kap YS, Mettenleiter TC et al. (2005). Tracing from fat tissue, liver and pancreas: a neuroanatomical framework for the role of the brain in type 2 diabetes. Endocrinology 147: 1140–1147. Kudielka BM, Schommer NC, Hellhammer DH et al. (2004). Acute HPA axis responses, heart rate, and mood changes to psychosocial stress (TSST) in humans at different times of day. Psychoneuroendocrinology 29: 983–992. https:// doi.org/10.1016/j.psyneuen.2003.08.009. Kudielka BM, Buchtal J, Uhde A et al. (2007). Circadian cortisol profiles and psychological self-reports in shift

THE CIRCADIAN SYSTEM: FROM CLOCKS TO PHYSIOLOGY workers with and without recent change in the shift rotation system. Biol Psychol 74: 92–103. https://doi.org/10.1016/j. biopsycho.2006.08.008. La Fleur SE, Kalsbeek A, Wortel J et al. (2001). A daily rhythm in glucose tolerance: a role for the suprachiasmatic nucleus. Diabetes 50: 1237–1243. Lamia KA, Storch K-F, Weitz CJ (2008). Physiological significance of a peripheral tissue circadian clock. Proc Natl Acad Sci U S A 105: 15172–15177. https://doi.org/ 10.1073/pnas.0806717105. Langlet F, Levin BE, Luquet S et al. (2013). Tanycytic VEGF-A boosts blood-hypothalamus barrier plasticity and access of metabolic signals to the arcuate nucleus in response to fasting. Cell Metab 17: 607–617. 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. Legan SJ, Karsch FJ (1975). A daily signal for the LH surge in the rat. Endocrinology 96: 57–62. https://doi.org/10.1210/ endo-96-1-57. Leon-Mercado L, Chao DHM, Basualdo MC et al. (2017). The arcuate nucleus: a site of fast negative feedback for corticosterone secretion in male rats. eNeuro 4. 1–14. https:// doi.org/10.1523/ENEURO.0350-16.2017. Lieberman JA, Stroup TS, McEvoy JP et al. (2005). Effectiveness of antipsychotic drugs in patients with chronic schizophrenia. N Engl J Med 353: 1209–1223. https://doi.org/10.1056/NEJMoa051688. McDearmon EL, Patel KN, Ko CH et al. (2006). Dissecting the functions of the mammalian clock protein BMAL1 by tissue-specific rescue in mice. Science 314: 1304–1308. Mimee A, Smith PM, Ferguson AV (2013). Circumventricular organs: targets for integration of circulating fluid and energy balance signals? Physiol Behav 121: 96–102. https://doi.org/ 10.1016/j.physbeh.2013.02.012. Mirmiran M, Swaab DF, Witting W et al. (1989). Biological clocks in development, aging and Alzheimer’s disease. Brain Dysfunct 2: 57–66. Mittelman-Smith MA, Williams H, Krajewski-Hall SJ et al. (2012a). Arcuate kisspeptin/neurokinin B/dynorphin (KNDy) neurons mediate the estrogen suppression of gonadotropin secretion and body weight. Endocrinology 153: 2800–2812. Mittelman-Smith MA, Williams H, Krajewski-Hall SJ et al. (2012b). Role for kisspeptin/neurokinin B/dynorphin (KNDy) neurons in cutaneous vasodilatation and the estrogen modulation of body temperature. Proc Natl Acad Sci USA 109: 19846–19851. Modabbernia A, Heidari P, Soleimani R et al. (2014). Melatonin for prevention of metabolic side-effects of olanzapine in patients with first-episode schizophrenia: randomized double-blind placebo-controlled study. J Psychiatr Res 53: 133–140. https://doi.org/10.1016/j.jpsychires.2014. 02.013. Mostafavi A, Solhi M, Mohammadi M-R et al. (2014). Melatonin decreases olanzapine induced metabolic sideeffects in adolescents with bipolar disorder: a randomized

245

double-blind placebo-controlled trial. Acta Med Iran 52: 734–739. Mrosovsky N (1990). Rheostasis: the physiology of change, Oxford University Press. Mukherji A, Kobiita A, Chambon P (2015). Shifting the feeding of mice to the rest phase creates metabolic alterations, which, on their own, shift the peripheral circadian clocks by 12 hours. Proc Natl Acad Sci USA 112: E6683–E6690. Oishi K, Yasumoto Y, Higo-Yamamoto S et al. (2017). Feeding cycle-dependent circulating insulin fluctuation is not a dominant Zeitgeber for mouse peripheral clocks except in the liver: differences between endogenous and exogenous insulin effects. Biochem Biophys Res Commun 483: 165–170. https://doi.org/10.1016/j.bbrc.2016.12.173. Opperhuizen AL, Foppen E, Jonker M et al. (2019). Effects of light-at-night on the rat liver—a role for the autonomic nervous system. Front Neurosci 13: 1–14. https://doi.org/ 10.3389/fnins.2019.00647. Orozco-solis R, Aguilar-arnal L, Murakami M et al. (2016). The circadian clock in the ventromedial hypothalamus controls cyclic energy expenditure. Cell Metab 23: 467–478. Padilla SL, Johnson CW, Barker FD et al. (2018). Report a neural circuit underlying. Cell Rep 24: 271–277. https:// doi.org/10.1016/j.celrep.2018.06.037. Padilla SL, Perez JG, Ben-Hamo M et al. (2019). Kisspeptin neurons in the arcuate nucleus of the hypothalamus orchestrate circadian rhythms and metabolism. Curr Biol 29: 592–604.e4. https://doi.org/10.1016/j.cub.2019.01.022. Palm IF, Van Der Beek EM, Wiegant VM et al. (1999). Vasopressin induces a luteinizing hormone surge in ovariectomized, estradiol-treated rats with lesions of the suprachiasmatic nucleus. Neuroscience 93: 659–666. https://doi. org/10.1016/S0306-4522(99)00106-2. Palm IF, Van Der Beek EM, Wiegant VM et al. (2001). The stimulatory effect of vasopressin on the luteinizing hormone surge in ovariectomized, estradiol-treated rats is time-dependent. Brain Res 901: 109–116. Paschos GK, FitzGerald GA (2010). Circadian clocks and vascular function. Circ Res 106: 833–841. https://doi.org/ 10.1161/CIRCRESAHA.109.211706. Paschos GK, Ibrahim S, Song W-L et al. (2012). Obesity in mice with adipocyte-specific deletion of clock component Arntl. Nat Med 18: 1768–1777. https://doi.org/10.1038/ nm.2979. 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. https://doi.org/10.1046/j.1460-9568.2003. 02442.x. 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. https://doi.org/ 10.1111/j.0953-816X.2003.03132.x. Perreau-Lenz S, Kalsbeek A, Van Der Vliet J et al. (2005). In vivo evidence for a controlled offset of melatonin synthesis at dawn by the suprachiasmatic nucleus in the rat. Neuroscience 130: 797–803. https://doi.org/10.1016/ j.neuroscience.2004.10.014.

246

R.M. BUIJS ET AL.

Rakai BD, Chrusch MJ, Spanswick SC et al. (2014). Survival of adult generated hippocampal neurons is altered in circadian arrhythmic mice. PLoS One 9: e99527. https://doi.org/ 10.1371/journal.pone.0099527. Ramı´rez DA, Vieyra E, Gonza´lez AI et al. (2017). Both the suprachiasmatic nucleus and the superior ovarian nerve contribute to the processes of ovulation and steroid hormone secretion on Proestrus. Reprod Sci 24: 844–855. https://doi.org/10.1177/1933719116670307. 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. Romo-Nava F, Alvarez-Icaza Gonza´lez D, Fresa´n-Orellana A et al. (2014). Melatonin attenuates antipsychotic metabolic effects: an eight-week randomized, double-blind, parallelgroup, placebo-controlled clinical trial. Bipolar Disord 16: 410–421. https://doi.org/10.1111/bdi.12196. Romo-Nava F, Buijs FN, Vald
es-Tovar M et al. (2017). Olanzapine-induced early cardiovascular effects are mediated by the biological clock and prevented by melatonin. J Pineal Res 62: 1–11. https://doi.org/10.1111/jpi.12402. Russo KA, La JL, Stephens SB et al. (2015). Circadian control of the female reproductive axis through gated responsiveness of the RFRP-3 system to VIP Signaling. Endocrinology 156: 2608–2618, en20141762. Sabath E, Salgado-Delgado R, Guerrero-Vargas NN et al. (2014). Food entrains clock genes but not metabolic genes in the liver of suprachiasmatic nucleus lesioned rats. FEBS Lett 588: 3104–3110. https://doi.org/10.1016/j.febslet.2014. 06.045. Salgado-Delgado R, Angeles-Castellanos M, Saderi N et al. (2010). Food intake during the normal activity phase prevents obesity and circadian desynchrony in a rat model of night work. Endocrinology 151: 1019–1029. https:// doi.org/10.1210/en.2009-0864. Salgado-Delgado RC, Saderi N, Basualdo MDC et al. (2013). Shift work or food intake during the rest phase promotes metabolic disruption and desynchrony of liver genes in male rats. PLoS One 8. 1–11. https://doi.org/10.1371/ journal.pone.0060052. Scarinci E, Tropea A, Notaristefano G et al. (2019). “Hormone of darkness” and human reproductive process: direct regulatory role of melatonin in human corpus luteum. J Endocrinol Invest 42: 1191–1197. https://doi.org/ 10.1007/s40618-019-01036-3. Scheer FA, Kalsbeek A, Buijs RM (2003). Cardiovascular control by the suprachiasmatic nucleus: neural and neuroendocrine mechanisms in human and rat. Biol Chem 384: 697–709. Scheer FAJL, Van Doornen LJP, Buijs RM (2004). Light and diurnal cycle affect autonomic cardiac balance in human; possible role for the biological clock. Auton Neurosci Basic Clin 110: 44–48. https://doi.org/10.1016/ j.autneu.2003.03.001. Scheer FAJL, Pirovano C, Van Someren EJW et al. (2005). Environmental light and suprachiasmatic nucleus interact in

the regulation of body temperature. Neuroscience 132: 465–477. https://doi.org/10.1016/j.neuroscience.2004.12.012. Schouten M, Bielefeld P, Garcia-Corzo L et al. (2020). Circadian glucocorticoid oscillations preserve a population of adult hippocampal neural stem cells in the aging brain. Mol Psychiatry 25: 1382–1405. https://doi.org/10.1038/ s41380-019-0440-2. Teclemariam Mesbah R, Ter Horst GJ, Postema F et al. (1999). Anatomical demonstration of the suprachiasmatic nucleuspineal pathway. J Comp Neurol 406: 171–182. Togo T, Kojima K, Shoji M et al. (2004). Serum adiponectin concentrations during treatment with olanzapine or risperidone: a pilot study. Int Clin Psychopharmacol 19: 37–40. https://doi.org/10.1097/00004850-200401000-00007. Tosches MA, Bucher D, Vopalensky P et al. (2014). Melatonin signaling controls circadian swimming behavior in marine zooplankton. Cell 159: 46–57. Uchida K, Okamoto N, Ohara K et al. (1996). Daily rhythm of serum melatonin in patients with dementia of the degenerate type. Brain Res 717: 154–159. Uhr M, Holsboer F, M€ uller MB (2002). Penetration of endogenous steroid hormones corticosterone, cortisol, aldosterone and progesterone into the brain is enhanced in mice deficient for both mdr1a and mdr1b P-glycoproteins. J Neuroendocrinol 14: 753–759. Van Cauter E, Shapiro ET, Tillil H et al. (1992). Circadian modulation of glucose and insulin responses to meals— relationship to cortisol rhythm. Am J Physiol 262: E467–E475. Van Der Beek EM, Van Oudheusden HJC, Buijs RM et al. (1994). Preferential induction of c-fos immunoreactivity in vasoactive intestinal polypeptide-innervated gonadotropinreleasing hormone neurons during a steroid-induced luteinizing hormone surge in the female rat. Endocrinology 134: 2636–2644. https://doi.org/10.1210/endo.134.6.8194489. 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. van der Beek EM, Swarts HJ, Wiegant VM (1999). Central administration of antiserum to vasoactive intestinal peptide delays and reduces luteinizing hormone and prolactin surges in ovariectomized, estrogen-treated rats. Neuroendocrinology 69: 227–237. https://doi.org/10.1159/ 000054423. van der Horst GT, Muijtjens M, Kobayashi K et al. (1999). Mammalian Cry1 and Cry2 are essential for maintenance of circadian rhythms [see comments]. Nature 398: 627–630. Vida B, Deli L, Hrabovszky E et al. (2010). Evidence for suprachiasmatic vasopressin neurones innervating kisspeptin neurones in the rostral periventricular area of the mouse brain: regulation by oestrogen. J Neuroendocrinol 22: 1032–1039. Wilkinson MJ, Manoogian ENC, Zadourian A et al. (2020). Ten-hour time-restricted eating reduces weight, blood pressure, and atherogenic lipids in patients with metabolic

THE CIRCADIAN SYSTEM: FROM CLOCKS TO PHYSIOLOGY syndrome. Cell Metab 31: 92–104.e5. https://doi.org/ 10.1016/j.cmet.2019.11.004. Xiao L, Zhang C, Li X et al. (2014). Signaling role of prokineticin 2 on the estrous cycle of female mice. PLoS One 9: e90860. https://doi.org/10.1371/journal.pone.0090860. Yi C-X, Van Der Vliet J, Dai J et al. (2006). Ventromedial arcuate nucleus communicates peripheral metabolic information to the suprachiasmatic nucleus. Endocrinology 147: 283–294. https://doi.org/10.1210/en.2005-1051. Yilmaz A, Kalsbeek A, Buijs RM (2018). Functional changes of the SCN in spontaneous hypertension but not after the induction of hypertension. Chronobiol Int 35: 1–15. https://doi.org/10.1080/07420528.2018.1469035.

247

Yilmaz A, Buijs FN, Kalsbeek A et al. (2019). Neuropeptide changes in the suprachiasmatic nucleus are associated with the development of hypertension. Chronobiol Int 36: 1072–1087. https://doi.org/10.1080/ 07420528.2019.1613424. Yoda T, Crawshaw LI, Yoshida K et al. (2000). Effects of food deprivation on daily changes in body temperature and behavioral thermoregulation in rats. Am J Physiol Regul Integr Comp Physiol 278: R134–R139. Zeman M, Dulkova´ K, Bada V et al. (2005). Plasma melatonin concentrations in hypertensive patients with the dipping and non-dipping blood pressure profile. Life Sci 76: 1795–1803. https://doi.org/10.1016/j.lfs.2004.08.034.

This page intentionally left blank

Handbook of Clinical Neurology, Vol. 179 (3rd series) The Human Hypothalamus: Anterior Region D.F. Swaab, F. Kreier, P.J. Lucassen, A. Salehi, and R.M. Buijs, Editors https://doi.org/10.1016/B978-0-12-819975-6.00015-7 Copyright © 2021 Elsevier B.V. All rights reserved

Chapter 16

Development of the circadian system and relevance of periodic signals for neonatal development CAROLINA ESCOBAR, ADELINA ROJAS-GRANADOS, AND MANUEL ANGELES-CASTELLANOS* Faculty of Medicine, Universidad Nacional Autónoma de M
exico (UNAM), Ciudad de M
exico, Mexico

Abstract Circadian rhythms are generated endogenously with a period of approximately 24 h. Studies carried out during the last decade indicate that the circadian system develops before birth, and that the suprachiasmatic nucleus, a structure that is considered the mammalian circadian clock, is present in primates from the middle of pregnancy. Recent evidence shows that the infants’ circadian system is sensitive to light from very early stages of development; it has also been proposed that low-intensity lighting can regulate the developing clock. After birth there is a progressive maturation of the outputs of the circadian system with marked rhythms in sleep–wake phenomena and hormone secretion. These facts express the importance of circadian photic regulation in infants. Thus, the exposure of premature babies to light/dark cycles results in a rapid establishment of activity/rest patterns, which are in the light–dark cycle. With the continuous study of the development of the circadian system and the influence on human physiology and disease, it is anticipated that the application of circadian biology will become an increasingly important component in the perinatal care.

INTRODUCTION The physiology of mammals is organized around a series of cyclic environmental events, such as the day/ night cycle, temperature variations, feeding time, among others. They represent a challenge to which individuals must adapt and respond efficiently in order to keep homeostasis and survival. This requires physiological mechanisms capable of detecting and anticipating predictable environmental changes that will signal possible alterations of the homeostatic regulation system (Dibner et al., 2010). Anticipation is one of the main functions of the circadian system. The circadian system monitors the passage of time especially of the day–night cycle and transmits time signals to all functions in the organism. The circadian system is built up by a biological clock, the

suprachiasmatic nucleus (SCN), and multiple structures in the brain and peripheral organs that exhibit temporal oscillations in a coupled manner (Mohawk et al., 2012). The SCN communicates with all brain and peripheral oscillators via multiple pathways, which can be neural or humoral (Buijs et al., 2003; Hastings et al., 2007). In this way, the SCN transmits the timing and keeps a temporal order in physiological and behavioral processes, allowing to anticipate environmental changes. Moreover, the SCN receives sensory input from several sensory allowing the synchrony between the external and internal cycles. The main synchronization pathway is the retina, which provides the light information through the retinohypothalamic tract to the SCN. The synchrony between the internal and external cycles is fundamental for an optimal response to the environmental challenges and for adaptation.

*Correspondence to: Manuel Ángeles-Castellanos, Departamento de Anatomía, Edificio B 4 Piso, Faculty of Medicine UNAM, M
exico, DF 04510, Mexico. Tel: +5255-5623-2422, Fax: +5255-5623-2424, E-mail: [email protected]

250

C. ESCOBAR ET AL.

Development of the circadian system In primates, the fetal SCN is functional during the last trimester of pregnancy. At that time, the fetal SCN is able to generate circadian rhythms and synchronizes to the mother’s circadian signals. (Reppert and Schwartz, 1984; Rivkees, 1997). In primates (Rhesus Monkey), the neurogenesis of the SCN occurs very early in development. It is present as early as days 27–48 of pregnancy (van Eerdenburg and Rakic, 1994). Rivkees (2007) mentions that the embryonic development of the SCN in monkeys and humans may be very similar during the first 100 days of gestation; therefore, it is suggested that neurons of the human SCN are probably also formed early in pregnancy. In studies using in vitro autoradiography with 125I-labeled melatonin, which allowed to visualize melatonin binding sites in the primate hypothalamus, the SCN was visualized on week 18 of gestation (Reppert et al., 1988a). Oscillations associated with the day–night cycle are already detectable in utero (Rivkees and Hao, 2000), suggesting that the human SCN is prenatally functional.

MATERNAL INFLUENCE IN THE DEVELOPMENT OF THE CIRCADIAN SYSTEM DURING GESTATION

During pregnancy the fetus functions as another peripheral oscillator inside the mother. The fetus receives humoral circadian signals from the mother that pass through the placenta and keep it synchronized (Fig. 16.1). These signals are considered nonphotic temporal stimuli for the fetus and include nutritional and hormonal signals that cross the placenta (Seron-Ferre et al., 2007). For the fetus, the temporal metabolic fluctuations depend on the maternal feeding schedules (Ohta et al., 2008) while hormonal signals depend on endogeneous clock-driven oscillations of melatonin (Torres-Farfan et al., 2006) and cortisol (Guardino et al., 2016). Experimental studies reported that keeping pregnant rats under fixed feeding schedules drove circadian rhythms in fetuses even when the mother was arrhythmic due to a dysfunctional SCN (Reppert et al., 1988b; Nováková et al., 2010). In rodents, fixed feeding schedules are a powerful nonphotic signal of synchronization for metabolic circadian rhythms and neuronal activity (Escobar et al., 1998; Angeles-Castellanos et al., 2007, 2010). However, the composition of the diet can play a relevant role in the process of synchronization. Exposing pregnant rodents to a high-fat diet altered the fetal expression of clock genes in peripheral tissues such as the liver (Suter et al., 2011). Another signal of circadian importance is undoubtedly maternal melatonin, which can directly cross the placental barrier. Melatonin is rapidly transferred from

Fig. 16.1. The maternal–fetus relationship. The mother’s synchronization to the light–dark cycle is transmitted by metabolic and hormonal signals (such as melatonin) with circadian periods, which serve as synchronization signals to the fetus through the placenta.

maternal to fetal circulation (Okatani et al., 1998; Reiter et al., 2014), resulting in similar serum levels between the fetus and the mother (Kennaway et al., 1996). The synthesis of melatonin in humans begins in the postnatal stage (Kennaway et al., 1992); therefore, maternal melatonin is necessary for circadian rhythms of this hormone in the fetus (Yellon and Longo, 1988). As previously mentioned, in the human fetus, the SCN and other parts of the brain have functional melatonin receptors (Weaver and Reppert, 1996; Thomas et al., 2002); therefore, maternal melatonin plays an important role in programming the development of the fetal SCN. It may also be necessary for protecting or reducing oxidative stress, in the fetus. It is now known that circadian oscillations of clock genes occur in the placenta and in the uterus. Especially, the per1 gene maintains stable circadian oscillations during pregnancy in the uterus and placental decidua, making it possible for the fetus to be provided with another circadian path of temporal information (Akiyama et al., 2010).

DEVELOPMENT OF THE CIRCADIAN SYSTEM Glucocorticoid secretion is necessary for the maturation of essential organs such as the lung and intestine. Blood levels of maternal glucocorticoids increase during the late night and cross the placenta thanks to the saturation of the enzyme 11-b hydroxylase. This maternal hormonal signal is of the utmost importance since glucocorticoids in the fetal circulation are necessary for adequate pulmonary maturation. Therefore, it is considered that the chronobiological regulation of the hypothalamic–pituitary–adrenal axis may be an essential mechanism for the adequate regulation of fetal growth, as well as in the development and maturation of the circadian rhythms (Kivlighan et al., 2008; Ivars et al., 2015). In an ideal system, where the mother provides nutrients and temporal signals from the environment to the developing fetus, it is logical to think that any circumstance that directly affects the circadian order in the mother can impact the circadian signals to the fetus in utero (Aagaard-Tillery et al., 2008). Therefore, it is important to assess the adverse conditions that can affect the maternal–fetal development. It is expected that a healthy fetus of 16–20 weeks exhibits a rhythm of locomotor activation and heart rate coupled to the sleep–wake rhythm of the mother. Thus, the absence of a clear rhythm between weeks 20th and 24th of pregnancy indicates delayed development and delayed growth of the fetus (Kintraia et al., 2005). Regarding the maternal influence on the newborn, most of the studies converge by pointing out the negative effect of clinical problems during pregnancy, including metabolic disease, genetic background, vascular, autoimmune disorders, and infections, which affect fetal growth and have an impact on the baby weight at birth. A disrupted sleep–wake cycle and/or chronic stress of the mother impact on the circadian rhythms of hormones such as melatonin and cortisol (El-Hennamy et al., 2008), and this exerts a negative influence on the fetal development. This is now an emerging problem that will require preventive campaigns.

BIOLOGICAL RHYTHMS IN THE NEWBORN At birth, when the umbilical cord is cut, the direct maternal–fetal communication is ended and environmental stimuli become more relevant for the infant. In the newborn during the first 28 days of life, a series of adaptive events occur, among which are the self-adjustment of the physiological variables essential for survival in an external environment hostile to the newborn. For example, the baby needs to have the ability to regulate body temperature and maintain blood levels of energy metabolites such as glucose and insulin, which were regulated by the mother when the baby was in the womb. Similarly, the newborn must cope with and adapt to cyclic variations

251

in the environment, such as environmental temperature variations and the light–dark cycle (LD). The suprachiasmatic nucleus of baby is fully developed (Seron-Ferre et al., 2001), and it is considered necessary for proper adjustment to the environment; therefore, we propose it necessary that the baby is exposed to environmental cyclic signals such as the light–dark cycle. During the first week of life, part of the rhythmicity found in the baby can still partly represent the previous influence of the mother. With the interruption of rhythmic maternal factors transmitted through the placenta, some rhythms that were detected in the fetus are decreased in the newborn (Haus and Touitou, 1997). The SCN maturation and adjustments continue after birth (Swaab, 1995). In full-term newborns, the number of vasopressinergic neurons in the SCN is 20% of the number present in adults, and it is only in the first year of age that babies and adults have a comparable number of vasopressin neurons as adults (Swaab et al., 1990). The development and maturation of other components of the circadian system occur later and gradually, i.e., the retina begins its development from the 12th week of gestation; however, only at the age of 5 years, it is considered fully mature (Nag and Wadhwa, 2001). This does not rule out, however, that from the moment of birth it can already participate as one of the main pathways of synchronization. Rhythms with ultradian frequencies are predominant in the newborn, and the development of a recognizable circadian periodicity in the baby occurs gradually during the first month of extrauterine life. For some variables, the appearance of circadian rhythms can be extended up to the first 2 years of life. It is now known that ultradian rhythms are more marked in preterm infants than in term infants (Waterhouse and DeCoursey, 2004). Direct observation of infants during the first 6 months of life evidenced a change from ultradian patterns to longer cycles with periods of around 25 h at 8 weeks of age. Synchronized patterns to 24 h were observed in the sleep–wake cycle around 16 weeks of age (Henderson et al., 2010). In newborns, the sleep–wake patterns have been evaluated, as well as heart rate, and temperature (Rivkees and Hao, 2000). Some studies indicate that at birth rhythms with periods of less than 24 h (ultradian) are present (Anders and Keener, 1985; Glotzbach et al., 1995). Human milk contains several components that show circadian variations such as melatonin and cortisol. Melatonin shows high levels in human milk at night and undetectable levels during the day (Illnerová et al., 1993; White, 2017), suggesting that melatonin could be a possible entrainment factor for newborns. However, the ultradian rhythm of feeding (each 3 h) produces interruptions of the sleep–wake cycle and increased

252

C. ESCOBAR ET AL.

locomotor activity of the newborns; consequently, the evaluation of endogenous circadian rhythms in newborns is complicated because possible endogenous circadian rhythms are masked. Therefore, in this field of neonatal chronobiological examination, there is a constant dilemma about the presence of circadian rhythms at birth and neonatal life. The introduction of more sophisticated analysis systems has allowed the description of circadian rhythmicity in cardiovascular and pulmonary functions both in the fetus and in the newborn, allowing to use this circadian function as an index of a healthy newborn and the lack or low expression of circadian rhythms as a high-risk condition for the newborn. Therefore, several features about early circadian rhythms are relevant: (1) different physiological functions develop their circadian rhythmicity independently, (2) the circadian rhythm of different functions manifests itself at different times after birth, and (3) during development, circadian periodicity shows an increase in the range of oscillation, and this occurs in all physiological functions (Rivkees, 2004; Henderson et al., 2010).

Factors influencing early circadian development In full-term and healthy newborns, the environmental conditions to which they are exposed immediately after birth can favor a rapid and adequate response to the cycles of the external environment or delay their adaptation. A clear example of the relevance of environmental conditions was reported in a study, where feeding restricted to the day period in newborns reduced the time required for adaptation to a night–day sleep–wake schedule. This study was carried out with two twin brothers, where one of them was fasted at night from the 5th week after birth, having only access to food during the day, while the other twin brother was nursed at free demand and night feeding. After 4 weeks in the restricted feeding schedules, the appearance of a stable circadian rhythm of sleep–wake was favored. Contrasting, in the second baby, a delay in the circadian organization of the sleep–wake cycle was observed. The study concluded that the introduction of feeding schedules coupled to the light–dark cycle is essential for the circadian organization (L€ orh and Siegmund, 1999). Thus, it is clear that temporal signals are of vital importance for the development and synchronization of the circadian system. The alternating light–dark cycles are also a powerful and important temporary signal for newborns. In a clinical study, we demonstrated that healthy preterm infants kept in conditions of constant dim light during the first 10 postnatal days, required after hospital discharge, approximately 20 days to show a circadian rhythmic

organization in sleep–wake behavior. Contrasting, preterm newborns who were placed under a LD cycle (12 h of light/12 h of darkness), during the first 10 postnatal days, exhibited a day–night organization of the sleep–wake rhythm immediately after hospital discharge (Rivkees, 2004). The results of these studies reveal that the circadian system after birth is fully functional and has the capacity to respond and, above all, to synchronize to temporarily regular and predictable environmental events, as in the example of food, and light. In this sense, light exerts strong effects on health and development, so that a constant light environment is related to stress in infants, manifesting an increase in locomotor activity, decreased sleep hours, producing arrhythmicity and bradycardia (Mirmiran and Ariagno, 2000). These findings have been replicated experimentally in nocturnal rodents, where the presence of constant light induces the gradual loss of various circadian rhythms and, in some cases, the appearance of ultradian rhythms (Depr
es-Brummer et al., 1995; Tapia-Osorio et al., 2013).

Relevance of a LD in the neonatal intensive care unit Findings mentioned previously indicate that postnatal physiological adaptation is compromised by disruption of circadian maternal function, and by disrupted cycling stimuli after birth. Moreover, studies performed with newborn babies indicate that premature newborns have a delay in the development of circadian rhythms compared to full-term newborns (Rivkees and Hao, 2000; Rivkees, 2003; Jackson et al., 2004). Many premature babies are hospitalized for prolonged periods in intensive care units (NICU), which has facilitated the chronobiological study in these patients (Rivkees, 1997), with limitations to measure circadian patterns in different variables, from preestablished aspects in hospital policies such as lighting conditions, to ethical aspects due to invasive measures in premature patients (D’Souza et al., 1992). No day–night differences in the sleep–wake patterns have been detected (Anders et al., 1985); only rhythms with ultradian periods have been revealed but no clear circadian rhythms (Glotzbach et al., 1995). It is suggested that feeding schedules and physical contact during preterm care influence an infant’s temperature, heart rate, and activity patterns, and thus mask circadian rhythms. The findings in studies with premature infants depend mainly on the NICU conditions in which premature newborns are staying. Often, they are deprived of the synchronization signals normally experienced during early life, such as regular LD cycles, the variation of the environmental temperature and feeding schedules, rather

DEVELOPMENT OF THE CIRCADIAN SYSTEM they encounter environmental factors such as noise, medical interventions, continuous enteral or parenteral feeding, and ventilatory support. These patients, instead of predictable temporal signals, are exposed to chaotic temporal signals, since most NICU are generally illuminated both day and night, without cyclic variations in intensity, that interfere with the synchronization of a normal circadian rhythm (McKenna and Reiss, 2018). Surprisingly, the light levels in the NICU to which premature newborns are exposed are not controlled (Robinson et al., 1990). Babies who receive treatment in most NICU, including preterm infants, are hospitalized for prolonged periods and are kept in similar light exposure conditions. The constant lighting conditions can be constant bright light or constant dim light; the main point is that there is no significant alternation in the intensity of illumination between a phase of light and one of darkness. As mentioned previously, it is very important to expose newborns to external cyclic events to favor and stimulate the development of the circadian system. However, due to the needs and protocols of in-hospital surveillance of the NICU, it is difficult to implement LD lighting in the neonatal intensive care. Rivkees (2004) reported that the exposure of preterm infants to an LD cycle for 2 weeks or more before hospital discharge induced in premature newborns sleep–wake patterns synchronized with the LD cycle. In this study, infants who were exposed to constant dim illumination in the NICU were not able to show sleep–wake patterns as observed in babies who had spent their first 20 days of life at home. Contrasting, an experimental group of preterm babies exposed to cyclic illumination before discharge exhibited differences in day–night activity during the first week after discharge (Rivkees, 2003). Unfortunately, due to the difficulty in detecting and recording circadian rhythmicity in preterm infants, the mechanisms by which light can influence the growth of premature babies are not clear (Rivkees and Hao, 2000). Moreover, since the rhythmicity is not apparent in preterm infants until 6 weeks after birth, these conclusions have been based on continuous rectal temperature records for 48 h and on continuous sleep–wake recordings (Glotzbach et al., 1995; McGraw et al., 1999; Mirmiran and Ariagno, 2000). On the other hand, there are reports indicating that circadian rhythms can be evidenced in preterm infants who were exposed to LD cycles from the moment of birth using activity sensors connected to a monitor that records sleep–wake rhythms (Hao and Rivkees, 1999; Rivkees, 2003). Constant light generates potential damage to the eyes of newborns. This was confirmed with experimental newborn rats exposed to constant light from the first day of birth. At weaning (postnatal day 23), we observed damage to the cellular organization of the retina

253

Fig. 16.2. Photomicrographs of histological sections of rat retina that was kept under three different lighting conditions from birth to 45 postnatal days. Histological analysis in the three groups shows a different cellular organization. (A) Control group maintained in a light–dark cycle. In this group, the distribution, shape, and cell size in each of the layers is observed morphologically normal. (B) Group of rats maintained in constant darkness conditions, the morphology of the retinas, show less heterogeneity and a thickening in the photoreceptor layer compared to the control group; greater intercellular space and a loss of cell alignment are observed; a decrease in the number of ganglion cells was evident. (C) Group of rats maintained in constant light conditions; the retinas show more degeneration and cellular disorganization; a notable loss of the photoreceptor layer is observed;, the outer nuclear layer shows abnormal thickness, vacuolization, and fewer cells, probably edema; in addition to loss of nuclei as well as karyorrhexis in addition to gliosis most likely at the expense of microglia, finally the ganglion cells reflect degeneration and pyknosis. GCL: ganglion cell layer; INL: inner nuclear layer; IPL: inner plexiform layer; ONL: outer nuclear layer; OPL: outer plexiform layer; RCL: rods and cones layer. Hematoxylin–eosin staining 20X (Unpublished data).

(Unpublished data; see Fig. 16.2). It is known that a constant light environment can contribute to retinal diseases in premature infants (Glass et al., 1985) producing retinal degeneration. The retina is the only pathway of photic

254

C. ESCOBAR ET AL.

Fig. 16.3. Photograph showing the lighting conditions of babies in the studio. (A) A newborn of the control group is shown to remain under constant lighting conditions, commonly in the Neonatal Intensive Care Unit (light: 249
11 lux), and (B) two newborns of the experimental group were under a light–dark cycle of 12 h light (249
11 lux from 7:00 a.m. to 07:00 p.m.) and 12 h darkness (27
0.8 lux from 7:00 p.m. to 7:00 a.m., with the infant’s head covered by a cloth).

synchronization for the mammalian circadian system; thus, constant light will continuously stimulate the SCN. It is known that constant light generates a phenomenon of loss of circadian rhythm in experimental animals (Tapia-Osorio et al., 2013) and possibly affect the development of the circadian system. Importantly, the use of constant light conditions in most hospitalization services does not have a solid scientific basis (Ariagno and Mirmiran, 2001). A proposed intervention program for the NICU based on the knowledge that the environment in the uterus is dark suggests that especially premature infants can develop better in an environment of darkness rather than in one of constant light (Als et al., 1994). However, the infant is prenatally exposed to the mother’s rhythms in phase with the light–dark cycle, which leads that indirectly the newborn is synchronized with this LD signal (Reppert et al., 1988b). Consequently, this indicates that the use of LD cycles immediately after birth might be the best option for the infant. The lack of more experimental clinical studies in this area have generated a lack of consensus whether or not the LD cycle should be used as a routine in the NICU. Our group, convinced by the hypothesis that “establishing an LD cycle for prematurely born infants in NICUs helps to improve their clinical development and evolution,” performed a clinical-experimental study (Vásquez-Ruiz et al., 2014). Thirty-eight healthy premature infants with an average birth age of 31.73
0.31 were studied and divided in two groups: (1) A control group (N ¼ 19) that was kept in the usual constant lighting conditions in the NICU. (2) An experimental group (N ¼ 19) that was exposed to a light–dark cycle. In order to produce the dark phase in despite a constantly

illuminated environment, acrylic helmets covered with cloth were used, as shown in Fig. 16.3. This measure allowed us to establish a significant difference in light exposure to the eyes during the 24-h. Assessment of vital signs as well as food consumption and bodyweight gain were performed daily from the first day of birth until discharge from the NICU. As expected, in these newborns, we could not clearly identify a circadian rhythm in any of the parameters evaluated, which coincides with previous reports, and confirms that during the first postnatal days babies only show ultradian rhythms (Rivkees and Hao, 2000). However, when analyzing the same variables day by day, we observed that they improved their evolution from the third and fifth day after exposure to an LD cycle; heart rate stabilized, oxygen saturation in the periphery was improved (Fig. 16.4). In addition, the LD cycle promoted better tolerance to food, this induced accelerated bodyweight increase, resulting in an earlier hospital discharge of patients maintained under the LD cycle (Fig. 16.5). Babies in the experimental and control groups were similar in birth weight and gestational age. Therefore, we concluded that the LD cycle promoted adaptive responses in the baby, while the constant light environment had detrimental effects that caused heart rate instability and decreased peripheral oxygen saturation, important physiological parameters during this critical stage of development (Merritt et al., 2003). There is still a controversy over whether an LD cycle is necessary at an early age and whether it has any effect on development. Kennedy et al. (2001) did not observe benefits or harmful effects on weight gain by reducing the intensity of light that reaches the eyes of premature babies by using glasses that reduced light 97% during the entire hospital stay of

DEVELOPMENT OF THE CIRCADIAN SYSTEM

255

Fig. 16.4. (A) Heart rate of the control group maintained in constant illumination (yellow) exhibits an unstable pattern and the experimental group (black) maintained under a light/dark cycle exhibited a stable heart rate from day 3 of manipulation. (B) Pulse oximetry in the experimental group (black) showed better peripheral oxygen saturation from the third day under the light/dark cycle, compared to the control group. Graphs modified with permission from Va´squez-Ruiz S, Maya-Barrios JA, Torres-Narva´ez P, et al. (2014). A light/dark cycle in the NICU accelerates body weight gain and shortens time to discharge in preterm infants. Early Hum Dev 90: 535–540.

Fig. 16.5. (A) The dynamics of hospital discharge, showing that infants under the light–dark cycle (LD) (black) cycle were discharged from the hospital before those who remained in LL (yellow). (B) Bar graph showing the average days of hospitalization in the control group (yellow) was longer than that of the experimental group (black) who was discharged 20.5 days earlier.

babies (4–36 weeks). The design of this study decreased the light intensity in a constant manner, and we believe that the positive results of our study are mainly because of the temporal signal of the alternation produced by the LD cycle. Beneficial effects on the development of babies similar to this study have been reported by other groups. Mann et al. (1986) reported that exposing babies to LD cycles favored weight gain and increased sleep episodes within 24 h, which did not occur in infants under constant lighting conditions. Brandon et al. (2002) also reported

bodyweight gain, a reduction of the hospital stays, and reduction of respiratory complications in infants, especially when exposure to the LD cycle began before 36 weeks of age. The development of melatonin production in infants has been studied by the detection of 6-sulfaxymethyltonin, which is a urinary metabolite of melatonin. In full-term infants, the circadian rhythm of melatonin is observed after 8 weeks of age, while premature babies only express a rhythmic production of melatonin after 12 weeks (Kennaway et al., 1992, 1996). It is important to mention

256

C. ESCOBAR ET AL.

that urine samples for the determination of melatonin were obtained while the babies were in the hospital, under conditions of relatively constant fluorescent lighting throughout the day with nocturnal intensities ranging between 400 and 700 lux. This intensity inhibits the production of melatonin in adults (Skene and Arendt, 2006). In our report, infants maintained in constant light conditions (249
11 lux), melatonin levels remained low and constant, while infants exposed to a cycle of illumination of 249
11 lux during the day and 27
0.8 lux at night exhibited a daily rhythmicity of salivary melatonin. These observations are supported by a previous study in which high levels of melatonin and rhythm disturbance were observed in newborns admitted to the NICU in constant light conditions (Marseglia et al., 2013). This finding indicates that a well-established alternation of an LD cycle condition is beneficial for the maturation of the baby’s physiology and for the circadian system.

FINAL COMMENTS Many of the studies reviewed in this chapter focus on the endogenous functionality of the circadian system in the early stages of development. It is clear that research has focused on two important periods. One is the prenatal stage where fetal rhythms are influenced by the maternal rhythms synchronized to the LD cycle. The second moment is the postnatal stage, where the newborn interacts with the environmental signals. In this postnatal stage, the circadian system is still in the process of maturation, and it responds with the generation of ultradian rhythms in both full-term and preterm babies. Not observing circadian rhythms in early stages of birth does not rule out the potential importance of a LD cycle in the hospital environment that can be relatively easy to achieve and at a low cost. There are enough data (reviewed previously) to reject continuing a noncircadian chaotic environmental approach in the NICU. The study and characterization of the effect of the LD cycle on the development and maturation of homeostatic regulation systems should be extended. More clinical studies should explore other variables as an integral part of newborn. The exposure of premature babies to an LD cycle improved physiological development, promoting weight gain and especially a decrease in hospital stay. The latter implies benefits for the baby who joins the family quickly; moreover, leaving the hospital reduces the risk of exposure to nosocomial diseases. In addition, direct and indirect economic costs for health care are also reduced. Finally, it is important to carry out rigorous experimental designs in this field that allow demonstrating the contributions of chronobiology in perinatal medical care.

ACKNOWLEDGMENT Manuel Angeles-Castellanos received support from the FOSISS-CONACYT-project 272596.

REFERENCES Aagaard-Tillery KM, Grove K, Bishop J et al. (2008). Developmental origins of disease and determinants of chromatin structure: maternal diet modifies the primate fetal epigenome. J Mol Endocrinol 41: 91–102. Akiyama S, Ohta H, Watanabe S et al. (2010). The uterus sustains stable biological clock during pregnancy. Tohoku J Exp Med 221: 287–298. Als H, Lawhon G, Duffy FH et al. (1994). Individualized developmental care for the very low birth weight preterm infant. Medical and neurofunctional effects. JAMA 272: 853–858. Anders TF, Keener M (1985). Developmental course of nighttime sleep-wake patterns in fullterm and premature infants during the first year of life. I. Sleep 8: 173–192. Anders TF, Keener MA, Kraemer H (1985). Sleep-wake state organization, neonatal assessment and development in premature infants during the first year of life. II. Sleep 8: 193–206. Angeles-Castellanos M, Mendoza J, Escobar C (2007). Restricted feeding schedules phase shift daily rhythms of c-Fos and protein Per1 immunoreactivity in corticolimbic regions in rats. Neuroscience 144: 344–355. Angeles-Castellanos M, Salgado-Delgado R, Rodriguez K et al. (2010). The suprachiasmatic nucleus participates in food entrainment: a lesion study. Neuroscience 165: 1115–1126. Ariagno RL, Mirmiran M (2001). Shedding light on the very low birth weight infant. J Pediatr 139: 476–477. Brandon DH, Holditch-Davis D, Belyea M (2002). Preterm infants born at less than 31 weeks’ gestation have improved growth in cycled light compared with continuous near darkness. J Pediatr 140: 192–199. 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. D’Souza SW, Tenreiro S, Minors D et al. (1992). Skin temperature and heart rate rhythms in infants of extreme prematurity. Arch Dis Child 67: 784–788. Depr
es-Brummer P, L
evi F, Metzger G et al. (1995). Lightinduced suppression of the rat circadian system. Am J Physiol 268: R1111–R1116. Dibner C, Schibler U, Albrecht U (2010). The mammalian circadian timing system: organization and coordination of central and peripheral clocks. Annu Rev Physiol 72: 517–549. El-Hennamy R, Mateju K, Bendova´ Z et al. (2008). Maternal control of the fetal and neonatal rat suprachiasmatic nucleus. J Biol Rhythms 3: 435–444. Escobar C, Diaz-Munoz M, Encinas F et al. (1998). Persistence of metabolic rhythmicity during fasting and its entrainment by restricted feeding schedules in rats. Am J Physiol Regul Integr Comp Physiol 274: R1309–R1316.

DEVELOPMENT OF THE CIRCADIAN SYSTEM Glass P, Avery GB, Subramanian KN et al. (1985). Effect of bright light in the hospital nursery on the incidence of retinopathy of prematurity. N Engl J Med 313: 401–404. Glotzbach SF, Edgar DM, Ariagno RL (1995). Biological rhythmicity in preterm infants prior to discharge from neonatal intensive care. Pediatrics 95: 231–237. Guardino CM, Schetter CD, Saxbe DE et al. (2016). Diurnal salivary cortisol patterns prior to pregnancy predict infant birth weight. Health Psychol 35: 625–633. Hao H, Rivkees SA (1999). The biological clock of very premature primate infants is responsive to light. Proc Natl Acad Sci U S A 96: 2426–2429. Hastings M, O’Neill JS, Maywood ES (2007). Circadian clocks: regulators of endocrine and metabolic rhythms. J Endocrinol 195: 187–198. Haus E, Touitou Y (1997). Chronobiology of development and aging. In: PH Redfern, B Lemmer (Eds.), Physiology and pharmacology of biological rhythms. Springer-Verlag, Berlin, pp. 95–122. Henderson JM, France KG, Owens JL et al. (2010). Sleeping through the night: the consolidation of self-regulated sleep across the first year of life. Pediatrics 126: e1081–e1087. Illnerova´ H, Buresova´ M, Presl J (1993). Melatonin rhythm in human milk. J Clin Endocrinol Metabol 77: 838–841. Ivars K, Nelson N, Theodorsson A et al. (2015). Development of salivary cortisol circadian rhythm and reference intervals in full-term infants. PLoS One 10: e0129502. Jackson JA, Wailoo MP, Thompson JR et al. (2004). Early physiological development of infants with intrauterine growth retardation. Arch Dis Child Fetal Neonatal Ed 89: 46–50. Kennaway DJ, Stamp GE, Goble FC (1992). Development of melatonin production in infants and the impact of prematurity. J Clin Endocrinol Metab 75: 367–379. Kennaway DJ, Goble FC, Stamp GE (1996). Factors influencing the development of melatonin rhythmicity in humans. J Clin Endocrinol Metab 81: 1525–1532. Kennedy KA, Fielder AR, Hardy RJ et al. (2001). Reduced lighting does not improve medical outcomes in very low birth weight infants. J Pediatr 139: 527–531. Kintraia PI, Zarnadze MG, Kintraia NP et al. (2005). Development of daily rhythmicity in heart rate and locomotor activity in the human fetus. J Circadian Rhythms 3: e5. Kivlighan KT, DiPietro JA, Costigan KA et al. (2008). Diurnal rhythm of cortisol during late pregnancy: associations with maternal psychological wellbeing and fetal growth. Psychoneuroendocrinology 33: 1225–1235. L€ orh B, Siegmund R (1999). Ultradian and circadian rhythms behavior during early infancy of sleep-wake and foodintake. Chronobiol Int 16: 129–148. Mann NP, Haddow R, Stokes L et al. (1986). Effect of night and day on preterm infants in a newborn nursery: randomised trial. Br Med J (Clin Res Ed) 293: 1265–1267. Marseglia L, Aversa S, Barberi I et al. (2013). High endogenous melatonin levels in critically ill children: a pilot study. J Pediatr 162: 357–360. McGraw K, Hoffmann R, Harker C et al. (1999). The development of circadian rhythms in a human infant. Sleep 22: 303–310.

257

McKenna H, Reiss IKM (2018). The case for a chronobiological approach to neonatal care. Early Hum Dev 126: 1–5. Merritt TA, Pillers D, Prows SL (2003). Early NICU discharge of very low birth weight infants: a critical review and analysis. Semin Neonatol 8: 95–115. Mirmiran M, Ariagno RL (2000). Influence of light in the NICU on the development of circadian rhythms in preterm infants. Semin Perinatol 24: 247–257. Mohawk JA, Green CB, Takahashi JS (2012). Central and peripheral circadian clocks in mammals. Annu Rev Neurosci 35: 445–462. Nag TC, Wadhwa S (2001). Differential expression of syntaxin-1 and synaptophysin in the developing and adult human retina. J Biosci 26: 179–191. Nova´kova´ M, Sla´dek M, Sumova´ A (2010). Exposure of pregnant rats to restricted feeding schedule synchronizes the SCN clocks of their fetuses under constant light but not under a light-dark regime. J Biol Rhythms 25: 350–360. Ohta H, Xu S, Moriya T et al. (2008). Maternal feeding controls fetal biological clock. PLoS One 3: e2601. Okatani Y, Okamoto K, Hayashi K et al. (1998). Maternal-fetal transfer of melatonin in pregnant women near term. J Pineal Res 25: 129–134. Reiter RJ, Tamura H, Tan DX et al. (2014). Melatonin and the circadian system: contributions to successful female reproduction. Fertil Steril 102: 321–328. Reppert SM, Schwartz WJ (1984). Functional activity of the suprachiasmatic nucleus in the fetal primate. Neurosci Lett 46: 145–149. Reppert SM, Weaver DR, Rivkees SA (1988a). Maternal communication of circadian phase to the developing mammal. Psychoneuroendocrinology 13: 63–78. Reppert SM, Weaver DR, Rivkees SA et al. (1988b). Putative melatonin receptors in a human biological clock. Science 242: 78–81. Rivkees SA (1997). Developing circadian rhythmicity. Basic and clinical aspects. Pediatr Clin North Am 44: 467–487. Rivkees SA (2003). Developing circadian rhythmicity in infants. Pediatrics 112: 373–381. Rivkees SA (2004). Emergence and influences of circadian rhythmicity in infants. Clin Perinatol 31: 217–228. Rivkees SA (2007). The development of circadian rhythms: from animals to humans. Sleep Med Clin 1: 331–334. Rivkees SA, Hao H (2000). Developing circadian rhythmicity. Semin Perinatol 24: 232–242. Robinson J, Moseley MJ, Fielder AR (1990). Illuminance of neonatal units. Arch Dis Child 65: 679–682. Seron-Ferre M, Torres-Farfan C, Forcelledo ML et al. (2001). The development of circadian rhythms in the fetus and neonate. Semin Perinatol 25: 363–370. Seron-Ferre M, Valenzuela GJ, Torres-Farfan C (2007). Circadian clocks during embryonic and fetal development. Birth Defects Res C Embryo Today 81: 204–214. Skene DJ, Arendt J (2006). Human circadian rhythms: physiological and therapeutic relevance of light and melatonin. Ann Clin Biochem 43: 344–353. Suter M, Bocock P, Showalter L et al. (2011). Epigenomics: maternal high-fat diet exposure in utero disrupts peripheral

258

C. ESCOBAR ET AL.

circadian gene expression in nonhuman primates. FASEB J 25: 714–726. Swaab DF (1995). Development of the human hypothalamus. Neurochem Res 20: 509–519. Swaab DF, Hofman MA, Honnebier MB (1990). Development of vasopressin neurons in the human suprachiasmatic nucleus in relation to birth. Brain Res Dev Brain Res 52: 289–293. Tapia-Osorio A, Salgado-Delgado R, Angeles-Castellanos M et al. (2013). Disruption of circadian rhythms due to chronic constant light leads to depressive and anxiety-like behaviors in the rat. Behav Brain Res 252: 1–9. Thomas L, Purvis CC, Drew JE et al. (2002). Melatonin receptors in human fetal brain: 2-[(125)I]iodomelatonin binding and MT1 gene expression. J Pineal Res 33: 218–224. Torres-Farfan C, Valenzuela FJ, Germain AM et al. (2006). Maternal melatonin stimulates growth and prevents maturation of the capuchin monkey fetal adrenal gland. J Pineal Res 41: 58–66. van Eerdenburg FJ, Rakic P (1994). Early neurogenesis in the anterior hypothalamus of the rhesus monkey. Brain Res Dev Brain Res 79: 290–296. Va´squez-Ruiz S, Maya-Barrios JA, Torres-Narva´ez P et al. (2014). A light / dark cycle in the NICU accelerates body weight gain and shortens time to discharge in preterm infants. Early Hum Dev 90: 535–540. Waterhouse JM, DeCoursey PJ (2004). Human circadian organization. In: JC Dumlap, JJ Loros, PJ DeCorsey (Eds.), Chonobiology: biological timekeeping. Sinauer Asocietes, Massachusetts, pp. 291–323.

Weaver DR, Reppert SM (1996). The Mel1a melatonin receptor gene is expressed in human suprachiasmatic nuclei. Neuroreport 8: 109–112. White RD (2017). Circadian variation of breast milk components and implications for care. Breastfeed Med 12: 398–400. Yellon SM, Longo LD (1988). Effect of maternal pinealectomy and reverse photoperiod on the circadian melatonin rhythm in the sheep and fetus during the last trimester of pregnancy. Biol Reprod 39: 1093–1099.

FURTHER READING Aguilar R, Drucker R, Moore RY (1992a). Behavioral and morphological studies of fetal neural transplants into SCN-lesioned rats. Chronobiol Int 9: 278–296. Aguilar R, Shibata S, Speh JC et al. (1992b). Morphological and functional development of the suprachiasmatic nucleus in transplanted fetal hypothalamus. Brain Res 580: 288–296. Drucker CR, Aguilar RR, Garcı´a HF et al. (1984). Fetal suprachiasmatic nucleus transplants: diurnal rhythm recovery of lesioned rats. Brain Res 311: 353–357. Rivkees SA, Hofman PL, Fortman J (1997). Newborn primate infants are entrained by low intensity lighting. Proc Natl Acad Sci U S A 94: 292–297. Swaab DF, Fliers E, Partiman TS (1985). The suprachiasmatic nucleus of the humans brain in relation to sex, age, and dementia. Brain Res 342: 37–44.

Handbook of Clinical Neurology, Vol. 179 (3rd series) The Human Hypothalamus: Anterior Region D.F. Swaab, F. Kreier, P.J. Lucassen, A. Salehi, and R.M. Buijs, Editors https://doi.org/10.1016/B978-0-12-819975-6.00016-9 Copyright © 2021 Elsevier B.V. All rights reserved

Chapter 17

Disrupted circadian rhythms and mental health WILLIAM H. WALKER II*, JAMES C. WALTON, AND RANDY J. NELSON Department of Neuroscience, Rockefeller Neuroscience Institute, West Virginia University, Morgantown, WV, United States

Abstract During the evolution of life, the temporal rhythm of our rotating planet was internalized in the form of circadian rhythms. Circadian rhythms are
24 h internal manifestations that drive daily patterns of physiology and behavior. These rhythms are entrained (synchronized) to the external environment, primarily by the light–dark cycle, and precisely controlled via molecular clocks located within the suprachiasmatic nucleus of the hypothalamus. Misalignment and/or disruption of circadian rhythms can have detrimental consequences for human health. Indeed, studies suggest strong associations between mental health and circadian rhythms. However, direct interactions between mood regulation and the circadian system are just beginning to be uncovered and appreciated. This chapter examines the relationship between disruption of circadian rhythms and mental health. The primary focus will be outlining the association between circadian disruption, in the form of night shift work, exposure to light at night, jet lag, and social jet lag, and psychiatric illness (i.e., anxiety, major depressive disorder, bipolar disorder, and schizophrenia). Additionally, we review animal models of disrupted circadian rhythms, which provide further evidence in support of a strong association between circadian disruption and affective responses. Finally, we discuss future directions for the field and suggest areas of study that require further investigation.

INTRODUCTION Outside of the highest latitudes, life on this planet evolved under daily light–dark cycles. Early during the evolution of life, the temporal rhythm of our rotating planet was internalized. Virtually, all organisms on the planet now display self-sustaining, internal biological clocks that drive daily rhythms of physiology and behavior. These so-called circadian (circa ¼ about; diem ¼ day) rhythms have periods approximating a solar day and are set to precisely 24 h by exposure to bright light during the day. In plants, these clocks likely evolved to forecast light and dark to efficiently orchestrate photosynthesis or counteract harmful redox reactions on a daily basis. As complexity of organisms increased, these internal clocks regulated not only metabolism but additional functions as well. In humans, virtually every aspect of our physiology and behavior, ranging from hormone secretion to

body temperature regulation, to food intake, to sleep, and to mood is mediated by our internal circadian clocks (Bedrosian and Nelson, 2013; Walker et al., 2020b). Functional circadian rhythms rely on precise entrainment (synchronization) to the environment, primarily via light information. Since the invention and widespread use of electric lights, however, individuals are more likely to entrain their activities to artificial light schedules, as well as social, school, and work schedules (Depner et al., 2018; McHill and Wright, 2019). Most North Americans and Europeans are exposed to artificial light at night (LAN) (Navara and Nelson, 2007) and approximately 20% of those populations work night shifts, which negatively affect sleep and other aspects of circadian organization (Boivin and Boudreau, 2014; Books et al., 2017). Exposure to light at night may hinder typical entrainment of circadian rhythms resulting in

*Correspondence to: William H. Walker II, PhD, Department of Neuroscience, Rockefeller Neuroscience Institute, West Virginia University, Morgantown, WV 26506 United States. Tel: +1-304-581-1760, Fax: +1-304-293-7182, E-mail: william.walker2@ hsc.wvu.edu

260

W.H. WALKER II ET AL.

misalignment of physiological and behavioral processes with the solar day (Bedrosian and Nelson, 2013; Walker et al., 2020a,b). Other aspects of modern life including jet travel across time zones produces a well-known desynchronization of circadian rhythms termed jet lag. Indeed, even small temporal changes such as the switch between standard and daylight saving time impairs circadian organization and biological function (Kantermann et al., 2007; Roenneberg et al., 2019). Most of us do not travel extensively or frequently across multiple time zones; however, most of us regularly experience what is termed, “social jet lag” (Roenneberg et al., 2003; Wittmann et al., 2006). Social jet lag is caused by the shift in sleep–wake schedules between school/work days and weekends/holidays; it is defined as the difference between the midpoint of sleep on school or workdays compared to the midpoint of sleep on free days reflecting the misalignment of individuals’ circadian rhythms and social constraints (Roenneberg et al., 2003). Disrupted circadian rhythms by exposure to light and social jet lag are the most common factors responsible for temporal physiological and behavioral misalignment. Temporal misalignment among hormones, neuropeptides, neurotransmitters, and their receptors may contribute to behavioral and affective disorders (Walker et al., 2020a,b). As described later, all forms of disrupted circadian rhythms, exposure to light at night, night shift work, jet lag, and social jet lag affect mood.

CIRCADIAN SYSTEM Virtually all cells in the body display circadian rhythms, but the mammalian master clock, located in the suprachiasmatic nuclei (SCN) of the hypothalamus, directs the daily cycles in physiology and behavior throughout the body (Hastings et al., 2018). Within SCN neurons, a transcription–translation feedback loop generates endogenous rhythms of approximately 24 h; the gene and protein components of this cycle are primarily entrained to the daily 24 h temporal environment via light information sent directly to the SCN via the retinohypothalamic tract (RHT). Correctly timed light information is crucial to precise biological timekeeping because without light and dark input, the endogenous clock rapidly goes out of phase with the external environment. The retina is the sole mechanism of light detection in mammals (Nelson et al., 1981; Lockley et al., 1999), comprised of image-forming photoreceptors (i.e., rods and cones) and nonimage forming, intrinsically photosensitive retinal ganglion cells (ipRGCs) (Lucas, 2013). IpRGCs are depolarized in response to light and are principally responsible for circadian photoentrainment. The ipRGCs, a small fraction of the total retinal ganglion cells, express melanopsin, a photopigment that

displays peak sensitivity to blue light (
480 nm); in contrast, melanopsin displays minimal sensitive to red light (>600 nm) (Brainard et al., 2001; Berson et al., 2002). The spectral composition of sunlight varies across the day, with enriched short wavelengths (blue) during twilight hours around dusk and dawn, and enriched longer wavelengths (red) peaking around midday (Hut et al., 2000). The spectrum of sensitivity for melanopsin may be an adaptation to the natural solar cycle, allowing for ipRGCs to differentiate twilight from midday and permitting precise entrainment of circadian rhythms (Bedrosian and Nelson, 2017). When stimulated, ipRGCs send neural signals via the RHT directly to the SCN. This monosynaptic pathway sends light information through glutamate release resulting in Ca2+ influx and activation of intracellular signaling cascades that affect the gene expression of canonical clock genes within SCN neurons (see later) (Partch et al., 2014). IpRGCs also project both directly and indirectly to additional targets within the brain, encompassing multiple mood-related structures (e.g., habenula and amygdala) (Hattar et al., 2006). The molecular mechanism of the cellular clock reflects an autoregulatory transcriptional/translational feedback loop, generated via a specific class of genes, including genes encoding brain and muscle ARNT-like protein 1 (BMAL1), circadian locomotor output cycles kaput (CLOCK), Cryptochrome (CRY 1,2), Period (PER 1,2,3), and others. For a detailed review of the molecular clock mechanism, see Takahashi (2017). Briefly, CLOCK and BMAL1 proteins increasingly form heterodimers at the beginning of the circadian day which function as a transcription factor to stimulate Cry (1,2), Per (1,2,3), and additional clock gene expression. PER and CRY protein products accumulate over the day and upon reaching a critical threshold, feed back to the nucleus, thus repressing transcription of CLOCK and BMAL1 (Takahashi, 2016). Notably, this feedback cycle requires
24h, hence driving circadian rhythms. Several other regulatory loops, in addition to this core feedback loop, are involved in the strict generation of circadian rhythms. Additional mechanisms, including posttranslational modifications, appear crucial for precise molecular clock function; e.g., methylation, phosphorylation, sumoylation, and additional modifications determine the localization, activity, and degradation of essential components within the molecular timing loop (Partch et al., 2014). The molecular clock of the SCN, in the absence of environmental signals, will continue to generate circadian rhythms. However, light typically serves as the primary environmental entraining cue, thus precisely sustaining internal synchrony with the environment. Exposure to light at night phase shifts the clock by rapidly inducing expression of Per1 or Per2, depending on whether the light occurs during the early or late night

DISRUPTED CIRCADIAN RHYTHMS AND MENTAL HEALTH (Challet et al., 2003; Ikeno and Yan, 2016). Although minor phase shifts can be advantageous for adapting to slight day length changes across the seasons, abrupt phase shifts due to night-time light exposure or transmeridian jet travel can be problematic for maintaining temporal integration. For instance, electronics use during the night can unintentionally phase shift circadian rhythms, uncoupling them from the environmental light– dark cycles (e.g., Chang et al., 2015) and potentially affecting physiology, behavior, and mood (Bedrosian and Nelson, 2017; Walker et al., 2020b). All hormones display circadian secretory and functional rhythms, but glucocorticoids and melatonin are the prominent markers of circadian function. Melatonin is secreted exclusively at night and exquisitely sensitive to light; > 3 lux of light at night can suppress the onset of melatonin secretion and shorten melatonin secretion duration in humans (Gooley et al., 2011). In contrast, glucocorticoid concentrations (cortisol in humans) tend to peak in humans just prior to awakening and decrease throughout the day (Son et al., 2011). These two types of hormones are important in several behavioral health conditions; indeed dysregulated cortisol and melatonin have been associated with multiple psychiatric illnesses (e.g., Srinivasan et al., 2006; Watson and Mackin, 2006; Dedovic and Ngiam, 2015; Caumo et al., 2019). The preclinical and clinical effects of disrupted circadian rhythms are reviewed later.

DISRUPTION OF CIRCADIAN RHYTHMS AND ANXIETY Anxiety disorders are the most common psychiatric illnesses in the world, with current prevalence rates estimated between 1% and 28% and lifetime prevalence rates as high as 30% (Wittchen et al., 2011; Kessler et al., 2012; Baxter et al., 2013; Bandelow et al., 2017). Anxiety disorders represent a group of psychiatric illnesses that include generalized anxiety disorder, specific phobias, panic disorder, social anxiety disorder, selective mutism, and others (Bandelow et al., 2017). Anxiety is a common healthy emotion in which one experiences temporary worry or fear in anticipation of a future threat. However, anxiety disorders differ from “healthy” anxiety due to an excessive or persistent worry or fear that typically interferes with daily functions (American Psychiatric Association, 2013). Anxiety disorders demonstrated high comorbidity with a secondary psychiatric illness, particularly depressive disorders, and a prominent sex difference in their diagnosis (i.e., women are more likely than men to be diagnosed with an anxiety disorder) (Maeng and Milad, 2015; Thibaut, 2017). Clinical studies examining the relationship between disrupted circadian rhythms and anxiety have produced

261

modest evidence of an association. For example, night shift and rotating shift work have been associated with a development or worsening of symptoms of anxiety (Flo et al., 2012; Chang et al., 2014; Kalmbach et al., 2015; Aburuz and Hayeah, 2017; Booker et al., 2020). Indeed, nurses working permanent nightshifts demonstrated higher levels of anxiety, as measured by the hospital anxiety and depression scale (HADS), relative to their daytime counterparts (Aburuz and Hayeah, 2017). However, studies suggest that elevated anxiety may reflect alterations in sleep, rather than disruptions in circadian rhythms per se. For example, day shift workers without prior sleep disturbances who transitioned to rotating shift work reported elevated anxiety along with altered sleep (Kalmbach et al., 2015). Additionally, nurses who develop shift work disorder, a circadian rhythm sleep disorder characterized by excessive sleepiness and/or insomnia due to the work schedule, report higher anxiety levels on the HADS (Flo et al., 2012; Waage et al., 2014). Some studies, however, failed to demonstrate an association between shift work and anxiety (Øyane et al., 2013). Studies examining the association among jet lag, social jet lag, and anxiety are sparse; nonetheless, these few studies have suggested a possible link (Montange et al., 1981; Mathew et al., 2019; Zhang et al., 2020). Among adolescents social jet lag is positively associated with the symptoms of anxiety (Mathew et al., 2019). Additionally, participants experiencing jetlag via a long-haul flight across six time zones demonstrated higher anxiety scores relative to a 50-day follow-up assessment (Zhang et al., 2020). Further, jet lag via a 7-h westward time shift by jet and, 1 month later, a 7-h eastward shift was associated with disrupted sleep and elevated anxiety and depression scores, particularly in eastward travel (Montange et al., 1981). Additional support for an association between circadian rhythm disruptions and anxiety disorders comes from the fact that studies have reported circadian fluctuations in anxiety symptoms, anxiety symptoms tended to be more severe in the afternoon or evening than in the morning (Cameron et al., 1986), and that most successful treatments of anxiety, SSRIs, SNIRs, and benzodiazepines can affect circadian rhythms (Buxton et al., 2000; Mcclung, 2011; Walker et al., 2020b). Rodent studies have demonstrated associations among anxiety-like behaviors and the circadian system. For example, CLOCK mutant mice (D19 mutation) display reduced anxiety-like behavior on the elevated plus maze and during the open field test (Roybal et al., 2007). Expression of functional CLOCK protein via viral injection into the ventral tegmental area rescued the reduced anxiety-like behavior (Roybal et al., 2007). Additionally, mutant mice lacking both functional Per1 and Per2 display elevated anxiety-like behavior

262

W.H. WALKER II ET AL.

(Spencer et al., 2013). Anxiety-like behavior was unaltered, however, in mice that lack either Per1 or Per2. Knockdown of both Per1 and Per2 expression in the nucleus accumbens (NAc) similarly increased anxiety-like behavior as seen in the mutant animals, suggesting a causal role for these core clock genes in the NAc for regulating anxiety (Spencer et al., 2013). More recent studies have demonstrated that optogenetic stimulation of the SCN induces anxiety-like behavior; particularly SCN-mediated dampening of rhythms was directly correlated with increased anxiety-like behavior (Vadnie et al., 2020). Additional rodent studies have investigated how circadian rhythm disruption alters anxiety-like behavior. Many studies have examined the effects of bright light or dim light at night on anxiety-like responses (Castro et al., 2005; Ashkenazy et al., 2009; Fonken et al., 2009; Borniger et al., 2014; Ciss
e et al., 2016; Ikeno and Yan, 2016; Walker et al., 2020a). However, the effects of light on anxiety-like responses are inconsistent across studies, which may reflect the species studied, time-of-day of behavioral assessments, type of light, duration of light exposure, intensity of light, as well as the developmental time window in which circadian disruption occurs. For example, disrupted circadian rhythms in rats via constant light at night increase anxiety-like behavior (Tapia-Osorio et al., 2013), although constant light exposure in mice either reduces or does not affect anxiety-like behavior (Castro et al., 2005; Fonken et al., 2009). Additionally, exposure to dim light at night during early development increases adult mouse anxiety-like responses (Borniger et al., 2014; Ciss
e et al., 2016). Other studies have reported that exposure of adult mice to light at night either reduces anxiety-like responses or does not alter anxiety-like behavior (Fonken et al., 2009; Walker et al., 2020a). Additional circadian rhythm disruption paradigms demonstrate a possible relationship between circadian disruption and anxiety-like behavior. Indeed, mice housed in 20-h light–dark cycles reduced complexity and dendritic length in neurons of the prelimbic prefrontal cortex, with a concurrent anxiolytic response (Karatsoreos et al., 2011). Further, simulating chronic jet lag in rats increased anxiety-like behavior (Horsey et al., 2020). Together these data provide modest evidence in support of an association between disruption of circadian rhythms and anxiety.

DISRUPTION OF CIRCADIAN RHYTHMS AND MAJOR DEPRESSIVE DISORDER Major depressive disorder (MDD) affects vast numbers of people worldwide. Indeed, it is estimated that MDD affects approximately 6% of the adult population each

year (Otte et al., 2016). The incidence of MDD worldwide continues to rise; diagnoses of depression increased by
18% from 2005 to 2015 (Walker et al., 2020b). Notably, the increased incidence of MDD correlate with the modernization of society (Hidaka, 2012), which may be explained by reduced stigmatization of MDD, better diagnostic tests. However, the increased disruption of circadian rhythms as society modernizes (i.e., night shift work, light at night, social jet lag, and jet lag) may contribute to the increases in MDD. MDD is typified by changes in mood, particularly increased sadness or irritability that is accompanied with psychophysiological symptoms (e.g., inability to experience pleasure, crying, suicidal thoughts, alterations in sleep, sexual desire, or appetite, and slowing of speech or actions) (Belmaker and Agam, 2008). Diagnosis of MDD requires these symptoms to interfere with normal daily functions and persist for at least 2 weeks (Belmaker and Agam, 2008). Similar to anxiety disorders, diagnosis of MDD occurs approximately twice as often in women as in men (Otte et al., 2016). Human studies have demonstrated strong associations between MDD and the circadian system. Indeed, disruption of biological rhythms underlie hallmarks of MDD. For example, patients with MDD display alterations in hormone rhythms, body temperature rhythms, and sleep/wake states (Vadnie and Mcclung, 2017). Postmortem examination within the brains of MDD patients demonstrates alterations in circadian patterns of gene expression, specifically, altered phase relationship between genes, reduced amplitude in gene expression, and shifted peaks in core clock genes (Li et al., 2013). Similar to anxiety, studies have reported circadian fluctuations in MDD symptoms, with patients exhibiting symptoms in a morning-worse or eveningworse pattern (Rusting and Larsen, 1998). MDD patients who display more severe symptoms in the morning, typically experience a more severe form of depression (Rusting and Larsen, 1998). Further, the degree of misalignment of circadian rhythms is correlated with the severity of MDD (Emens et al., 2009). Additional support for an association between circadian rhythm disruptions and MDD comes from the observation that treatments of depression, antidepressants (SSRIs, SNRIS, and agomelatine), bright light therapy, social therapy, and wake therapy directly affect circadian rhythms (Germain and Kupfer, 2008). Indeed, administration of antidepressants or morning bright light therapy result in a phase advance of circadian rhythms (Terman et al., 2001; Mcclung, 2011; Robillard et al., 2018). The degree of phase advancement due to morning bright light therapy or antidepressant treatment is correlated with the reduction of depressive symptoms (Terman et al., 2001; Robillard et al., 2018). Together, diurnal variation

DISRUPTED CIRCADIAN RHYTHMS AND MENTAL HEALTH in symptoms as well as successful treatment of MDD with chronotherapies provides evidence for an association between MDD and the circadian system. Most human studies combine all forms of depression when examining associations between circadian rhythm disruption and MDD. The few studies that have assessed MDD specifically, however, have reported inconsistent results (Ohayon and Hong, 2006; Murcia et al., 2013; Oenning et al., 2018). Indeed, one such study, examined
4000 South Koreans and demonstrated that the prevalence of MDD was significantly higher in night shift workers relative to daytime workers (Ohayon and Hong, 2006). In a large Brazilian cohort of
36,000 workers, night shift work was significantly associated with MDD only in women (Oenning et al., 2018). Additionally, in a French study, the authors reported no association between MDD and shift work (Murcia et al., 2013). A clear association between night shift work and depression appears when all forms of depression are combined (Moon et al., 2015; Lee et al., 2016, 2017; Booker et al., 2020). A meta-analysis containing 11 studies reported that night shift workers are
40% more likely to develop depression compared to daytime workers (Lee et al., 2017). Even fewer human studies have examined the effects of social jet lag or jet lag on MDD (Knapen et al., 2018). One such study examined the amount of social jet lag in patients with MDD and healthy individuals and concluded that there was no association between social jet lag and severity of depressive symptoms (Knapen et al., 2018). However, similar to shift work, combining all types of depression results in a strong association among jet lag, social jet lag, and depression (Young, 1995; Katz et al., 2001; Srinivasan et al., 2010; Levandovski et al., 2011; McNeely et al., 2018). Rodent studies have provided ample evidence to support a clear association between disruption of circadian rhythms and depressive-like behavior. Indeed, rats housed in constant light lose diurnal rhythms in activity, melatonin, and corticosterone with a concurrent increase in depressive-like behavior (Tapia-Osorio et al., 2013). Agomelatine administration to rats housed in constant light restored diurnal corticosterone and melatonin rhythms and eliminated the increase in depressive-like behavior (Tchekalarova et al., 2018, 2019). Numerous studies have examined the effects of dim light at night on depressive-like behavior in multiple species of rodents (Fonken et al., 2012; Bedrosian et al., 2013; Fonken and Nelson, 2013; Walker et al., 2020a). For example, female Siberian hamsters housed in 5 lux of dim light at night for 4 weeks display reduced dendritic spine density within the hippocampus with concurrent increases in neuroinflammation and depressive-like behavior (Bedrosian et al., 2013). Additionally, diurnal

263

rats housed in 5 lux of dim light at night for 3 weeks demonstrate reduced dendritic length within CA1 and dentate gyrus and increased depressive-like behavior (Fonken et al., 2012). Dim light at night can rapidly alter depressive-like responses. Indeed, housing mice in 5 lux of dim light at night for 3 nights increased depressivelike behavioral responses (Walker et al., 2020a). Studies in C57Bl/6 mice have not reported any association between light at night and affective responses suggesting a potential strain specific effect of LAN (Martynhak et al., 2017; Cleary-Gaffney and Coogan, 2018). Taken together, there is a sizable amount of clinical and preclinical evidence to support a link between disruptions of circadian rhythms and MDD/depression.

CIRCADIAN RHYTHM DISRUPTION AND BIPOLAR DISORDER Bipolar disorder (BD), previously called manic depression or manic-depressive disorder, is marked by divergent atypical episodes of extreme mood swings. These episodes, which can last for days or weeks, cycle between depression and mania, with intervening periods of normal affect. Onset of BD usually occurs between 20 and 30 years of age, and BD has an estimated lifetime prevalence of about 2.4%. Current diagnosis of BD is divided into three main categories of increasing severity: Cyclothymic, characterized by episodes of hypomania and depressive symptoms over a period of at least 2 years, Bipolar II, characterized by at least one major depressive episode and at least one hypomanic episode, and Bipolar I, characterized by a manic episode that may or may not be followed by a hypomanic or major depressive episodes. Individual symptoms vary by severity during episodes, but usually involve concomitant changes in energy, activity, and sleep. Although not traditionally thought of as a circadian disorder, there is emerging evidence that the circadian system is strongly implicated in this disorder. Genetic studies have reported almost 90% heritability of BD (McGuffin et al., 2003; Etain et al., 2011) and revealed that multiple components of the molecular circadian clock are associated with BD (Le-Niculescu et al., 2009; Etain et al., 2011). Furthermore, gene polymorphisms in, and dysregulation of, the molecular circadian clock contribute to both susceptibility to develop BD and to relapse into bipolar episodes (Bellivier et al., 2015). Although more than 40 clinical studies have investigated and reported disruption of circadian rhythms in BD, these studies have not been able to determine the cause–effect relationships due to their cross-sectional designs (Melo et al., 2017). Further clinical studies properly designed will be necessary to determine whether disruption of circadian rhythms in BD is secondary to other factors, or if it is a primary

264

W.H. WALKER II ET AL.

pathophysiology of the disease. Regardless, circadian markers, such as cortisol rhythms and buccal cell molecular circadian clock rhythms, have been proposed as biomarkers of mania (phase-advanced circadian rhythms) and depression (phase-delayed circadian rhythms) in BD patients (Moon et al., 2016). As a therapeutic approach, stabilizing and normalizing circadian rhythmicity has proven effective against both symptoms and relapses in BD over time (Pinho et al., 2016; Gold and Kinrys, 2019). Preliminary circadian studies in BD patients reported that they suffered from chronic circadian rhythm disruption due to fast running circadian clocks; thus, treatments that slowed the molecular circadian clock, such as lithium, stabilized circadian rhythms and helped to resolve BD symptoms (Kripke et al., 1978). More recently, circadian rhythm normalization via mid-day bright light therapy has proven effective as a treatment for bipolar depression in some patients; however, bright light therapy in the morning should be avoided as it can induce mixed states (Sit et al., 2007, 2018). Similarly, manipulation of the photic environment to normalize circadian rhythms via enforced darkness or blue light blocking glasses can help alleviate mania in some BD patients (Barbini et al., 2005; Henriksen et al., 2016). On the flip side of the coin, disruption of circadian rhythms appears to be able to induce BD episodes. For example, major social disruptive events have been reported to induce episodes of mania, but not depression (Malkoff-Schwartz et al., 2000). Additionally, jet lag can induce bipolar episodes in a longitudinally specific fashion: phase advances in rhythms from flying across time zones west to east can induce mania, whereas phase delays in rhythms induced by traveling east to west can induce episodes of depression (Jauhar and Weller, 1982; Young, 1995; Katz et al., 2002). Taken together, these clinical data suggest that dysregulated circadian rhythmicity is both a trait and a state marker of BD. Dysregulation arising from both environmental (jet lag, social jet lag) and internal (genetic/ molecular) factors can induce and predispose individuals to bipolar episodes; the type of which (mania or depression) is dependent upon the phase relationship of the internal circadian rhythm with the environmental circadian rhythm. Chronotherapeutic approaches are effective in treatment and prevention of BD, providing strong support that BD has a significant circadian rhythm component. However, advances in the treatment and understanding of the underlying mechanisms of BD have been hampered by the lack of valid animal models of this disease in which to conduct translational research (Machado-Vieira et al., 2004; Malkesman et al., 2009; Nestler and Hyman, 2010). The majority of animal models probe the specific states of BD, viz., depression and mania. Animal models

of circadian rhythm disruption in depression are discussed previously in the MDD section. For bipolar mania, converging preclinical and clinical data have implicated dysregulation of protein kinase c (PKC) signaling, potentially by environmental disruption of circadian rhythms (Saxena et al., 2017). Although there are few preclinical models of circadian rhythm disruption and mania in rodents, transient mania-like states can be induced in mice via sleep deprivation (Benedetti et al., 2008). Furthermore, mice that fail to recover from sleep disruption (i.e., do not reentrain to their original circadian rhythm) induced by a 72-h exposure to inverted environmental light dark cycles also display mania-like behavior (hyperactivity) in the absence of increases in depressive-like behavior, which is associated with dysregulated PKC signaling in frontal and limbic brain regions (Jung et al., 2014; Moon et al., 2018). In addition to PKC, dysregulated central dopaminergic signaling has also been implicated in circadian rhythm disruption in bipolar mania. Genetic disruption of the molecular circadian clock via deletion of ClockD19 induces mania-like behavior in mice that are normalized by either lithium treatment or by restoring a functional molecular circadian clock to the ventral tegmental area to normalize dopaminergic neuronal activity (Roybal et al., 2007). Although the ClockD19 mouse model of mania recapitulates much of the pathophysiology of bipolar mania in humans, the fundamentally disrupted molecular circadian clock precludes it from studying external (environmental) influences on circadian rhythm disruption in BD. Among the animal models of mania and depression discussed previously, none have successfully recapitulated the state switching between mania and depression that is prevalent in BD (Logan and McClung, 2016). Recently, a mouse model of state switching was proposed in mice with reduced dopamine transporter expression in which mania-like and depressive-like behaviors were displayed dependent upon environmental day length (Young et al., 2018); however, these results have not been replicated nor validated as a model of state switching in BD (Rosenthal and McCarty, 2019). Regardless, current evidence strongly supports the association between disrupted circadian rhythms and the onset and severity of BD, and this disorder is in critical need of both preclinical and clinical research with circadian rhythms as a key biological variable.

CIRCADIAN RHYTHM DISRUPTION AND SCHIZOPHRENIA Schizophrenia (SZ) has a low lifetime prevalence (
0.5%), yet it is an extremely disabling mental disorder with onset typically occurring around 20–30 years of age (McGrath et al., 2008; Simeone et al., 2015). SZ is

DISRUPTED CIRCADIAN RHYTHMS AND MENTAL HEALTH characterized by cognitive impairments, negative symptoms (impaired avolition and sociality, alogia, and anhedonia), and positive symptoms (movement disorders, disorganized speech, hallucinations, and delusions). Heritability of
80% strongly suggests a genetic component to SZ; however, environmental or epigenetic factors significantly influence risk as demonstrated by much lower concordance rates (40%–65%) found in monozygotic twin studies of SZ (Cardno and Gottesman, 2000; Bromundt et al., 2011). Disrupted circadian rhythms have been identified as a prodrome of SZ, and symptom severity has been associated with levels of circadian rhythm and sleep disruption (Bromundt et al., 2011). In common with other psychiatric disorders discussed previously in this chapter, several case studies indicate that circadian rhythm disruption via jet lag appears to be able to induce or cause relapses in SZ (Oyewumi, 1998; Katz et al., 1999). Several genes that have been implicated in SZ are also involved in circadian organization. Blunted expression of CRY1, PER1, and CLOCK was reported in mononuclear blood cells collected from SZ patients after their first psychotic episode (Johansson et al., 2016). Similarly, skin fibroblasts collected and cultured from chronic SZ patients lacked rhythmic expression of PER2 and CRY1, yet circadian expression of CLOCK, PER1, CRY2, DBP, REV-ERBa, and BMAL1 did not differ from healthy controls (Johansson et al., 2016). One study investigating SNPs in eight circadian genes reported modest associations with SNPs on PER3 and TIMELESS in SZ patients (Mansour et al., 2006). A followup study of 276 SNPs on 21 circadian genes identified significant associations of eight individual SNPs on NPAS2, PER2, PER3, and RORB in SZ samples; however, these associations did not confer a substantial risk for SZ (OR >1.5) (Mansour et al., 2009). In addition to these core clock gene SNP associations with SZ, copy number variants also enhance risk for SZ. Several studies have reported a strong association of vasoactive polypeptide receptor (VIPR2) duplication with SZ in humans (Vacic et al., 2011). Vasoactive polypeptide (VIP) neurons in the SCN core and neurons expressing its receptor (VIPR2) in the SCN shell are critical for circadian rhythmicity and entrainment (Vosko et al., 2007). Indeed, in mice, deletion of VIPR2 (VPAC2) not only affects cognition in a manner similar to SZ (Chaudhury et al., 2008) it also disrupts adrenal clock genes which in turn disrupts circadian rhythms in glucocorticoid secretion (Fahrenkrug et al., 2012). Of note, there is an abundance of VIPR2 in the cortex and the knock out of the VIP receptor will also affect cortical transmission. In addition to the aforementioned altered circadian genes, altered circadian rhythms in the neuroendocrine system, specifically in two prominent transducers of

265

circadian synchronization discussed previously— cortisol and melatonin, also contribute to the etiology of SZ. Circadian cortisol rhythms generally remain intact in SZ (Rao et al., 1995; Sun et al., 2016), yet elevated cortisol levels and hyperreactivity of the HPA axis are common in SZ and in people at risk for developing SZ (Ryan et al., 2004; Sun et al., 2016). Additionally, these altered circadian patterns of cortisol secretion are associated with increased severity of SZ symptoms (Kaneko et al., 1992; Ho et al., 2016); however, the antipsychotic olanzapine, when used to treat negative symptoms, can blunt the elevated cortisol levels found in SZ (Mann et al., 2006). Among unmedicated individuals with SZ, melatonin rhythms are blunted and peak concentrations are reduced when compared to healthy individuals (Ferrier et al., 1982; Monteleone et al., 1992; Viganò et al., 2001). Additionally, melatonin acrophase is advanced in unmedicated individuals with SZ (Rao et al., 1994). However, in SZ patients who are medicated, melatonin rhythms can be similarly blunted without or with a phase shift (Bromundt et al., 2011), or display either a phase advance or a phase delay (Wirz-Justice et al., 1997; Wulff et al., 2012). The alteration in neuroendocrine circadian rhythms and sleep in SZ has been attributed to both compromised responses to the effects of melatonin on sleep and to social withdrawal resulting in reduced zeitgeber exposure (Afonso et al., 2011; Bromundt et al., 2011). Regardless of direction of phase shift or change in levels in circulation, dysregulation of neuroendocrine circadian rhythms represents an uncoupling of internal circadian rhythms from environmental rhythms, resulting in desynchronization of behavioral and physiological process which contribute to susceptibility and severity of psychiatric diseases such as SZ.

CONCLUSIONS AND FUTURE DIRECTIONS Taken together, this chapter supports a strong association between disruption of circadian rhythms and psychiatric illness. As evidenced previously, disruptions of the circadian system and/or misalignment of circadian rhythms can have detrimental consequences for mental health. However, the relationship between the pathophysiology of psychiatric illness and circadian rhythm disruption remains poorly defined. The bulk of clinical evidence suggesting a relationship between psychiatric illness and circadian rhythm disruption are correlational. Thus, causation cannot definitively be assigned. The current association between misalignment of circadian rhythms and mental health may reflect any of the following: (1) disruption of circadian rhythms drives psychiatric illness, (2) psychiatric illness leads to circadian rhythm disruption, or (3) there is no causative relationship

266

W.H. WALKER II ET AL.

between psychiatric illness and misalignment of circadian rhythms. However, careful examination of the rodent literature provides strong support that disruptions of circadian rhythms can alter affective responses. Further, targeted resynchronization of circadian rhythms in rodents can alleviate alteration in the symptomology of mood disorders. Future studies are needed to expand both the clinical and basic science literature in relation to circadian misalignment and mental health. Focus should be placed on examining the effects of jet lag and social jet lag on various psychiatric illnesses, as numerous clinical studies have typically focused on the effects of shift work on psychiatric health. Particular focus should be placed on uncovering associations among BD, SZ, and circadian rhythm disruptions, as relative to anxiety disorders and depressive disorders; these areas remain vastly understudied. In sum, although our current knowledge does not allow for assignment of a causative role in circadian disruption eliciting psychiatric illness, it does support a strong association between the two.

ACKNOWLEDGMENTS Preparation of this chapter was supported by National Institutes of Health grants; award number R01NS092388 from the National Institute of Neurological Disorders and Stroke and 5U54GM104942-03 from the National Institute of General Medical Sciences. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

REFERENCES Aburuz ME, Hayeah HMA (2017). Insomnia induced by night shift work is associated with anxiety, depression, and fatigue, among critical care nurses. Adv Stud Biol 9: 137–156. Afonso P, Figueira ML, Paiva T (2011). Sleep-promoting action of the endogenous melatonin in schizophrenia compared to healthy controls. Int J Psychiatry Clin Pract 15: 311–315. American Psychiatric Association (2013). Anxiety disorders, in: diagnostic and statistical manual of mental disorders, American Psychiatric Association pp. 1–992. Ashkenazy T, Einat H, Kronfeld-Schor N (2009). We are in the dark here: induction of depression- and anxiety-like behaviours in the diurnal fat sand rat, by short daylight or melatonin injections. Int J Neuropsychopharmacol 12: 83. Bandelow B, Michaelis S, Wedekind D (2017). Treatment of anxiety disorders. Dialogues Clin Neurosci 19: 93–106. Barbini B, Benedetti F, Colombo C et al. (2005). Dark therapy for mania: a pilot study. Bipolar Disord 7: 98–101. Baxter AJ, Scott KM, Vos T et al. (2013). Global prevalence of anxiety disorders: a systematic review and meta-regression. Psychol Med 5: 897–910.

Bedrosian TA, Nelson RJ (2013). Influence of the modern light environment on mood. Mol Psychiatry 18: 751–757. Bedrosian TA, Nelson RJ (2017). Timing of light exposure affects mood and brain circuits. Transl Psychiatry 7: e1017. Bedrosian TA, Weil ZM, Nelson RJ (2013). Chronic dim light at night provokes reversible depression-like phenotype: possible role for TNF. Mol Psychiatry 18: 930–936. Bellivier F, Geoffroy PA, Etain B et al. (2015). Sleep- and circadian rhythm-associated pathways as therapeutic targets in bipolar disorder. Expert Opin Ther Targets 6: 747–763. Belmaker RH, Agam G (2008). Major depressive disorder. N Engl J Med 358: 55–68. Benedetti F, Fresi F, Maccioni P et al. (2008). Behavioural sensitization to repeated sleep deprivation in a mice model of mania. Behav Brain Res 187: 221–227. Berson DM, Dunn FA, Takao M (2002). Phototransduction by retinal ganglion cells that set the circadian clock. Science 295: 1070–1073. Boivin DB, Boudreau P (2014). Impacts of shift work on sleep and circadian rhythms. Pathol Biol 62: 292–301. Booker LA, Sletten TL, Alvaro PK et al. (2020). Exploring the associations between shift work disorder, depression, anxiety and sick leave taken amongst nurses. J Sleep Res 29: e12872. Books C, Coody LC, Kauffman R et al. (2017). Night shift work and its health effects on nurses. Health Care Manag 36: 347–353. Borniger JC, McHenry ZD, Abi Salloum BA et al. (2014). Exposure to dim light at night during early development increases adult anxiety-like responses. Physiol Behav 133: 99–106. Brainard GC, Hanifin JR, Greeson JM et al. (2001). Action spectrum for melatonin regulation in humans: evidence for a novel circadian photoreceptor. J Neurosci 21: 6405–6412. Bromundt V, K€ oster M, Georgiev-Kill A et al. (2011). Sleep— wake cycles and cognitive functioning in schizophrenia. Br J Psychiatry 198: 269–276. Buxton OM, Copinschi G, Van Onderbergen A et al. (2000). A benzodiazepine hypnotic facilitates adaptation of circadian rhythms and sleep-wake homeostasis to an eight hour delay shift simulating westward jet lag. Sleep 23: 915–927. Cameron OG, Lee MA, Kotun J et al. (1986). Circadian symptom fluctuations in people with anxiety disorders. J Affect Disord 11: 213–218. Cardno AG, Gottesman II (2000). Twin studies of schizophrenia: from bow-and-arrow concordances to Star Wars Mx and functional genomics. Am J Med Genet 97: 12–17. Castro JPMV, Frussa-Filho R, Fukushiro DF et al. (2005). Effects of long-term continuous exposure to light on memory and anxiety in mice. Physiol Behav 86: 218–223. Caumo W, Hidalgo MP, Souza A et al. (2019). Melatonin is a biomarker of circadian dysregulation and is correlated with major depression and fibromyalgia symptom severity. J Pain Res 12: 545–556. Challet E, Caldelas I, Graff C et al. (2003). Synchronization of the molecular clockwork by light- and food-related cues in mammals. Biol Chem 384: 711–719.

DISRUPTED CIRCADIAN RHYTHMS AND MENTAL HEALTH Chang YS, Chen HL, Wu YH et al. (2014). Rotating night shifts too quickly may cause anxiety and decreased attentional performance, and impact prolactin levels during the subsequent day: a case control study. BMC Psychiatry 14: 1–10. Chang AM, Aeschbach D, Duffy JF et al. (2015). Evening use of light-emitting eReaders negatively affects sleep, circadian timing, and next-morning alertness. Proc Natl Acad Sci U S A 112: 1232–1237. Chaudhury D, Loh DH, Dragich JM et al. (2008). Select cognitive deficits in vasoactive intestinal peptide deficient mice. BMC Neurosci 9: 1–12. Ciss
e YM, Peng J, Nelson RJ (2016). Dim light at night prior to adolescence increases adult anxiety-like behaviors. Chronobiol Int 33: 1473–1480. Cleary-Gaffney M, Coogan AN (2018). Limited evidence for affective and diurnal rhythm responses to dim light-atnight in male and female C57Bl/6 mice. Physiol Behav 189: 78–85. Dedovic K, Ngiam J (2015). The cortisol awakening response and major depression: examining the evidence. Neuropsychiatr Dis Treat 11: 1181–1189. Depner CM, Melanson EL, McHill AW et al. (2018). Mistimed food intake and sleep alters 24-hour time-of-day patterns of the human plasma proteome. Proc Natl Acad Sci U S A 115: E5390–E5399. Emens J, Lewy A, Kinzie JM et al. (2009). Circadian misalignment in major depressive disorder. Psychiatry Res 168: 259–261. Etain B, Milhiet V, Bellivier F et al. (2011). Genetics of circadian rhythms and mood spectrum disorders. Eur Neuropsychopharmacol 4: S676–S682. Fahrenkrug J, Georg B, Hannibal J et al. (2012). Altered rhythm of adrenal clock genes, StAR and serum corticosterone in VIP receptor 2-deficient mice. J Mol Neurosci 48: 584–596. Ferrier IN, Arendt J, Johnstone EC et al. (1982). Reduced nocturnal melatonin secretion in chronic schizophrenia: relationship to body weight. Clin Endocrinol (Oxf ) 17: 181–187. Flo E, Pallesen S, Magerøy N et al. (2012). Shift work disorder in nurses—assessment, prevalence and related health problems. PLoS One 7: e33981. Fonken LK, Nelson RJ (2013). Dim light at night increases depressive-like responses in male C3H/HeNHsd mice. Behav Brain Res 243: 74–78. Fonken LK, Finy MS, Walton JC et al. (2009). Influence of light at night on murine anxiety- and depressive-like responses. Behav Brain Res 205: 349–354. Fonken LK, Kitsmiller E, Smale L et al. (2012). Dim nighttime light impairs cognition and provokes depressivelike responses in a diurnal rodent. J Biol Rhythms 27: 319–327. Germain A, Kupfer DJ (2008). Circadian rhythm disturbances in depression. Hum Psychopharmacol 23: 571–585. Gold AK, Kinrys G (2019). Treating circadian rhythm disruption in bipolar disorder. Curr Psychiatry Rep 21: 14. Gooley JJ, Chamberlain K, Smith KA et al. (2011). Exposure to room light before bedtime suppresses melatonin onset

267

and shortens melatonin duration in humans. J Clin Endocrinol Metab 96: E463–E472. Hastings MH, Maywood ES, Brancaccio M (2018). Generation of circadian rhythms in the suprachiasmatic nucleus. Nat Rev Neurosci 19: 453–469. Hattar S, Kumar M, Park A et al. (2006). Central projections of melanopsin-expressing retinal ganglion cells in the mouse. J Comp Neurol 497: 326–349. Henriksen TE, Skrede S, Fasmer OB et al. (2016). Blue-blocking glasses as additive treatment for mania: a randomized placebo-controlled trial. Bipolar Disord 18: 221–232. Hidaka BH (2012). Depression as a disease of modernity: explanations for increasing prevalence. J Affect Disord 140: 205–214. Ho RH, Fong TT, Wan AY et al. (2016). Associations between diurnal cortisol patterns and lifestyle factors, psychotic symptoms, and neurological deficits: a longitudinal study on patients with chronic schizophrenia. J Psychiatr Res 81: 16–22. Horsey EA, Maletta T, Turner H et al. (2020). Chronic jet lag simulation decreases hippocampal neurogenesis and enhances depressive behaviors and cognitive deficits in adult male rats. Front Behav Neurosci 13: 272. Hut RA, Scheper A, Daan S (2000). Can the circadian system of a diurnal and a nocturnal rodent entrain to ultraviolet light? J Comp Physiol A 186: 707–715. Ikeno T, Yan L (2016). Chronic light exposure in the middle of the night disturbs the circadian system and emotional regulation. J Biol Rhythms 31: 352–364. Jauhar P, Weller MPI (1982). Psychiatric morbidity and time zone changes: a study of patients from Heathrow Airport. Br J Psychiatry 140: 231–235. Johansson AS, Owe-Larsson B, Hetta J et al. (2016). Altered circadian clock gene expression in patients with schizophrenia. Schizophr Res 174: 17–23. Jung SH, Park JM, Moon E et al. (2014). Delay in the recovery of normal sleep-wake cycle after disruption of the lightdark cycle in mice: a bipolar disorder-prone animal model? Psychiatry Investig 11: 487–491. Kalmbach DA, Pillai V, Cheng P et al. (2015). Shift work disorder, depression, and anxiety in the transition to rotating shifts: the role of sleep reactivity. Sleep Med 16: 1532–1538. Kaneko M, Yokoyama F, Hoshino Y et al. (1992). Hypothalamic-pituitary-adrenal axis function in chronic schizophrenia: association with clinical features. Neuropsychobiology 25: 1–7. Kantermann T, Juda M, Merrow M et al. (2007). The human circadian clock’s seasonal adjustment is disrupted by daylight saving time. Curr Biol 17: 1996–2000. Karatsoreos IN, Bhagat S, Bloss EB et al. (2011). Disruption of circadian clocks has ramifications for metabolism, brain, and behavior. Proc Natl Acad Sci U S A 108: 1657–1662. Katz G, Durst R, Barel YC et al. (1999). De novo jet-lag psychosis. Br J Psychiatry 174: 558–559. Katz G, Durst R, Zislin Y et al. (2001). Psychiatric aspects of jet lag: review and hypothesis. Med Hypotheses 56: 20–23.

268

W.H. WALKER II ET AL.

Katz G, Knobler HY, Laibel Z et al. (2002). Time zone change and major psychiatric morbidity: the results of a 6-year study in Jerusalem. Compr Psychiatry 43: 37–40. Kessler RC, Petukhova M, Sampson NA et al. (2012). Twelvemonth and lifetime prevalence and lifetime morbid risk of anxiety and mood disorders in the United States. Int J Methods Psychiatr Res 21: 169–184. Knapen SE, Riemersma-van der Lek RF, Antypa N et al. (2018). Social jetlag and depression status: results obtained from the Netherlands study of depression and anxiety. Chronobiol Int 35: 1–7. Kripke D, Mullaney D, Atkinson M et al. (1978). Circadian rhythm disorders in manic-depressives. Biol Psychiatry 13: 335–351. Lee HY, Kim MS, Kim O et al. (2016). Association between shift work and severity of depressive symptoms among female nurses: the Korea nurses’ health study. J Nurs Manag 24: 192–200. Lee A, Myung SK, Cho JJ et al. (2017). Night shift work and risk of depression: meta-analysis of observational studies. J Korean Med Sci 32: 1091–1096. Le-Niculescu H, Patel SD, Bhat M et al. (2009). Convergent functional genomics of genome-wide association data for bipolar disorder: comprehensive identification of candidate genes, pathways and mechanisms. Am J Med Genet Part B Neuropsychiatr Genet 150: 155–181. Levandovski R, Dantas G, Fernandes LC et al. (2011). Depression scores associate with chronotype and social jetlag in a rural population. Chronobiol Int 28: 771–778. Li JZ, Bunney BG, Meng F et al. (2013). Circadian patterns of gene expression in the human brain and disruption in major depressive disorder. Proc Natl Acad Sci U S A 110: 9950–9955. Lockley SW, Skene DJ, Thapan K et al. (1999). Extraocular light exposure does not suppress plasma melatonin in humans. In: MF Holick, EG Jung (Eds.), Biologic effects of light 1998. Springer, US, pp. 403–406. Logan RW, McClung CA (2016). Animal models of bipolar mania: the past, present and future. Neuroscience 321: 163–188. Lucas RJ (2013). Mammalian inner retinal photoreception. Curr Biol 23: R125–R133. Machado-Vieira R, Kapczinski F, Soares JC (2004). Perspectives for the development of animal models of bipolar disorder. Prog Neuro-Psychopharmacology Biol Psychiatry 28: 209–224. Maeng LY, Milad MR (2015). Sex differences in anxiety disorders: interactions between fear, stress, and gonadal hormones. Horm Behav 76: 106–117. Malkesman O, Austin DR, Chen G et al. (2009). Reverse translational strategies for developing animal models of bipolar disorder. DMM Dis Model Mech 2: 238–245. Malkoff-Schwartz S, Frank E, Anderson BP et al. (2000). Social rhythm disruption and stressful life events in the onset of bipolar and unipolar episodes. Psychol Med 30: 1005–1016. Mann K, Rossbach W, M€uller MJ et al. (2006). Nocturnal hormone profiles in patients with schizophrenia treated with olanzapine. Psychoneuroendocrinology 31: 256–264.

Mansour HA, Wood J, Logue T et al. (2006). Association study of eight circadian genes with bipolar I disorder, schizoaffective disorder and schizophrenia. Genes Brain Behav 5: 150–157. Mansour HA, Talkowski ME, Wood J et al. (2009). Association study of 21 circadian genes with bipolar I disorder, schizoaffective disorder, and schizophrenia. Bipolar Disord 11: 701–710. Martynhak BJ, Hogben AL, Zanos P et al. (2017). Transient anhedonia phenotype and altered circadian timing of behaviour during night-time dim light exposure in Per3/ mice, but not wildtype mice. Sci Rep 7: 1–9. Mathew GM, Li X, Hale L et al. (2019). Sleep duration and social jetlag are independently associated with anxious symptoms in adolescents. Chronobiol Int 36: 461–469. Mcclung CA (2011). Circadian rhythms and mood regulation: insights from pre-clinical models. Eur Neuropsychopharmacol Suppl. 4: S683–S693. McGrath J, Saha S, Chant D et al. (2008). Schizophrenia: a concise overview of incidence, prevalence, and mortality. Epidemiol Rev 30: 67–76. McGuffin P, Rijsdijk F, Andrew M et al. (2003). The heritability of bipolar affective disorder and the genetic relationship to unipolar depression. Arch Gen Psychiatry 60: 497–502. McHill AW, Wright KP (2019). Cognitive impairments during the transition to working at night and on subsequent night shifts. J Biol Rhythms 34: 432–446. McNeely E, Mordukhovich I, Tideman S et al. (2018). Estimating the health consequences of flight attendant work: comparing flight attendant health to the general population in a cross-sectional study. BMC Public Health 18: 1–11. Melo MCA, Abreu RLC, Linhares Neto VB et al. (2017). Chronotype and circadian rhythm in bipolar disorder: a systematic review. Sleep Med Rev 34: 46–58. Montange MF, Van Cauter E, Refetoff S et al. (1981). Effects of “jet lag” on hormonal patterns. II adaptation of melatonin circadian periodicity. J Clin Endocrinol Metab 52: 642–649. Monteleone P, Maj M, Fusco M et al. (1992). Depressed nocturnal plasma melatonin levels in drug-free paranoid schizophrenics. Schizophr Res 7: 77–84. Moon HJ, Lee SH, Lee HS et al. (2015). The association between shift work and depression in hotel workers. Ann Occup Environ Med 27: 29. Moon JH, Cho CH, Son GH et al. (2016). Advanced circadian phase in mania and delayed circadian phase in mixed mania and depression returned to normal after treatment of bipolar disorder. EBioMedicine 11: 285–295. Moon E, Choe BM, Park JM et al. (2018). Protein kinase C activity and delayed recovery of sleep-wake cycle in mouse model of bipolar disorder. Psychiatry Investig 15: 907–913. Murcia M, Chastang JF, Niedhammer I (2013). Psychosocial work factors, major depressive and generalised anxiety disorders: results from the French national SIP study. J Affect Disord 146: 319–327.

DISRUPTED CIRCADIAN RHYTHMS AND MENTAL HEALTH Navara KJ, Nelson RJ (2007). The dark side of light at night: physiological, epidemiological, and ecological consequences. J Pineal Res 43: 215–224. Nelson RJ, Zucker I, Nelson RJ et al. (1981). Absence of extraocular photoreception in diurnal and nocturnal rodents exposed to direct sunlight. Comp Biochem Physiol A Physiol 69: 145–148. Nestler EJ, Hyman SE (2010). Animal models of neuropsychiatric disorders. Nat Neurosci 7: 39–61. Oenning NSX, Ziegelmann PK, de Goulart BNG et al. (2018). Occupational factors associated with major depressive disorder: a Brazilian population-based study. J Affect Disord 240: 48–56. Ohayon MM, Hong SC (2006). Prevalence of major depressive disorder in the general population of South Korea. J Psychiatr Res 40: 30–36. Otte C, Gold SM, Penninx BW et al. (2016). Major depressive disorder. Nat Rev Dis Prim 2: 16065. Øyane NMF, Pallesen S, Moen BE et al. (2013). Associations between night work and anxiety, depression, insomnia, sleepiness and fatigue in a sample of Norwegian nurses. PLoS One 8: e70228. Oyewumi LK (1998). Jet lag and relapse of schizoaffective psychosis despite maintenance clozapine treatment. Br J Psychiatry 173: 268. Partch CL, Green CB, Takahashi JS (2014). Molecular architecture of the mammalian circadian clock. Trends Cell Biol 24: 90–99. Pinho M, Sehmbi M, Cudney LE et al. (2016). The association between biological rhythms, depression, and functioning in bipolar disorder: a large multi-center study. Acta Psychiatr Scand 133: 102–108. Rao ML, Gross G, Strebel B et al. (1994). Circadian rhythm of tryptophan, serotonin, melatonin, and pituitary hormones in schizophrenia. Biol Psychiatry 35: 151–163. Rao ML, Strebel B, Halaris A et al. (1995). Circadian rhythm of vital signs, norepinephrine, epinephrine, thyroid hormones, and cortisol in schizophrenia. Psychiatry Res 57: 21–39. Robillard R, Carpenter JS, Feilds K-L et al. (2018). Parallel changes in mood and melatonin rhythm following an adjunctive multimodal chronobiological intervention with agomelatine in people with depression: a proof of concept open label study. Front Psych 9: 624. Roenneberg T, Wirz-Justice A, Merrow M (2003). Life between clocks: daily temporal patterns of human chronotypes. J Biol Rhythms 18: 80–90. Roenneberg T, Wirz-Justice A, Skene DJ et al. (2019). Why should we abolish daylight saving time? J Biol Rhythms 34: 227–230. Rosenthal SJ, McCarty R (2019). Switching winter and summer photoperiods in an animal model of bipolar disorder. Neuropsychopharmacology 44: 1677–1678. Roybal K, Theobold D, Graham A et al. (2007). Mania-like behavior induced by disruption of CLOCK. Proc Natl Acad Sci U S A 104: 6406–6411. Rusting CL, Larsen RJ (1998). Diurnal patterns of unpleasant mood: associations with neuroticism, depression, and anxiety. J Pers 66: 85–103.

269

Ryan MCM, Sharifi N, Condren R et al. (2004). Evidence of basal pituitary-adrenal overactivity in first episode, drug naı¨ve patients with schizophrenia. Psychoneuroendocrinology 29: 1065–1070. Saxena A, Scaini G, Bavaresco DV et al. (2017). Role of protein kinase c in bipolar disorder: a review of the current literature. Mol Neuropsychiatry 3: 108–124. Simeone JC, Ward AJ, Rotella P et al. (2015). An evaluation of variation in published estimates of schizophrenia prevalence from 1990-2013: a systematic literature review. BMC Psychiatry 15: 1–14. Sit D, Wisner KL, Hanusa BH et al. (2007). Light therapy for bipolar disorder: a case series in women. Bipolar Disord 9: 918–927. Sit DK, McGowan J, Wiltrout C et al. (2018). Adjunctive bright light therapy for bipolar depression: a randomized double-blind placebo-controlled trial. Am J Psychiatry 175: 131–139. Son GH, Chung S, Kim K (2011). The adrenal peripheral clock: glucocorticoid and the circadian timing system. Front Neuroendocrinol 32: 451–465. Spencer S, Falcon E, Kumar J et al. (2013). Circadian genes period 1 and period 2 in the nucleus accumbens regulate anxiety-related behavior. Eur J Neurosci 37: 242–250. Srinivasan V, Smits M, Spence W et al. (2006). Melatonin in mood disorders. World J Biol Psychiatry 7: 138–151. Srinivasan V, Singh J, Pandi-Perumal SR et al. (2010). Jet lag, circadian rhythm sleep disturbances, and depression: the role of melatonin and its analogs. Adv Ther 27: 796–813. Sun HQ, Li SX, Chen FB et al. (2016). Diurnal neurobiological alterations after exposure to clozapine in first-episode schizophrenia patients. Psychoneuroendocrinology 64: 108–116. Takahashi JS (2016). Molecular architecture of the circadian clock in mammals. Res Perspect Endocr Interact 1: 13–24. Takahashi JS (2017). Transcriptional architecture of the mammalian circadian clock. Nat Rev Genet 18: 164–179. Tapia-Osorio A, Salgado-Delgado R, Angeles-Castellanos M et al. (2013). Disruption of circadian rhythms due to chronic constant light leads to depressive and anxiety-like behaviors in the rat. Behav Brain Res 252: 1–9. Tchekalarova J, Stoynova T, Ilieva K et al. (2018). Agomelatine treatment corrects symptoms of depression and anxiety by restoring the disrupted melatonin circadian rhythms of rats exposed to chronic constant light. Pharmacol Biochem Behav 171: 1–9. Tchekalarova J, Stoyanova T, Gesheva R et al. (2019). Agomelatine treatment corrects depressive-like behaviour induced by chronic constant light exposure through modulation of circadian rhythm of corticosterone release. C R Acad Bulg Sci 72: 537–544. Terman JS, Terman M, Lo ES et al. (2001). Circadian time of morning light administration and therapeutic response in winter depression. Arch Gen Psychiatry 58: 69–75. Thibaut F (2017). Anxiety disorders: a review of current literature. Dialogues Clin Neurosci 19: 87–88. Vacic V, McCarthy S, Malhotra D et al. (2011). Duplications of the neuropeptide receptor gene VIPR2 confer significant risk for schizophrenia. Nature 471: 499–503.

270

W.H. WALKER II ET AL.

Vadnie CA, Mcclung CA (2017). Circadian rhythm disturbances in mood disorders: insights into the role of the suprachiasmatic nucleus. Neural Plast 2017: 1504507. Vadnie CA, Eberhardt LA, Hildebrand MA et al. (2020). Suprachiasmatic nucleus regulates anxiety-like behavior in mice. bioRxiv. https://doi.org/10.1101/2020.05.08.085217. Vigano` D, Lissoni P, Rovelli F et al. (2001). A study of light/ dark rhythm of melatonin in relation to cortisol and prolactin secretion in schizophrenia. Neuroendocrinol Lett 22: 137–141. Vosko AM, Schroeder A, Loh DH et al. (2007). Vasoactive intestinal peptide and the mammalian circadian system. Gen Comp Endocrinol 10: 542. Waage S, Pallesen S, Moen BE et al. (2014). Predictors of shift work disorder among nurses: a longitudinal study. Sleep Med 15: 1449–1455. Walker WH, Borniger JC, Gaudier-Diaz MM et al. (2020a). Acute exposure to low-level light at night is sufficient to induce neurological changes and depressive-like behavior. Mol Psychiatry 25: 1080–1093. Walker WH, Walton JC, DeVries AC et al. (2020b). Circadian rhythm disruption and mental health. Transl Psychiatry 10: 28.

Watson S, Mackin P (2006). HPA axis function in mood disorders. Psychiatry 5: 166–170. Wirz-Justice A, Cajochen C, Nussbaum P (1997). A schizophrenic patient with an arrhythmic circadian rest-activity cycle. Psychiatry Res 73: 83–90. Wittchen HU, Jacobi F, Rehm J et al. (2011). The size and burden of mental disorders and other disorders of the brain in Europe 2010. Eur Neuropsychopharmacol 21: 655–679. Wittmann M, Dinich J, Merrow M et al. (2006). Social jetlag: misalignment of biological and social time. Chronobiol Int 23: 497–509. Wulff K, Dijk DJ, Middleton B et al. (2012). Sleep and circadian rhythm disruption in schizophrenia. Br J Psychiatry 200: 308–316. Young DM (1995). Psychiatric morbidity in travelers to Honolulu, Hawaii. Compr Psychiatry 36: 224–228. Young JW, Cope ZA, Romoli B et al. (2018). Mice with reduced DAT levels recreate seasonal-induced switching between states in bipolar disorder. Neuropsychopharmacology 43: 1721–1731. Zhang F, Li W, Li H et al. (2020). The effect of jet lag on the human brain: a neuroimaging study. Hum Brain Mapp 41: 2281–2291.

Handbook of Clinical Neurology, Vol. 179 (3rd series) The Human Hypothalamus: Anterior Region D.F. Swaab, F. Kreier, P.J. Lucassen, A. Salehi, and R.M. Buijs, Editors https://doi.org/10.1016/B978-0-12-819975-6.00017-0 Copyright © 2021 Elsevier B.V. All rights reserved

Chapter 18

Diurnal and seasonal molecular rhythms in the human brain and their relation to Alzheimer disease ANDREW S.P. LIM* Division of Neurology, Department of Medicine, Sunnybrook Health Sciences Centre, University of Toronto, Toronto, ON, Canada

Abstract Diurnal and seasonal rhythms influence many aspects of human physiology including brain function. Moreover, altered diurnal and seasonal behavioral and physiological rhythms have been linked to Alzheimer’s disease and related dementias (ADRD). Understanding the molecular basis for these links may lead to identification of novel targets to mitigate the negative impact of normal and abnormal diurnal and seasonal rhythms on ADRD or to alleviate the adverse consequences of ADRD on normal diurnal and seasonal rhythms. Diurnally and seasonally rhythmic gene expression and epigenetic modification in the human neocortex may be a key mechanism underlying these links. This chapter will first review the observed epidemiological links between normal and abnormal diurnal and seasonal rhythmicity, cognitive impairment, and ADRD. Then it will review normal diurnal and seasonal rhythms of brain epigenetic modification and gene expression in model organisms. Finally, it will review evidence for diurnal and seasonal rhythms of epigenetic modification and gene expression the human brain in aging, Alzheimer’s disease, and other brain disorders.

DIURNAL RHYTHMS AND DEMENTIA IN OLDER ADULTS Diurnal rhythms and normal human brain function Circadian rhythms are near 24-h biological rhythms driven by an internal biological clock that persists even in the absence of cyclical environmental time cues. The capacity to generate circadian rhythms is found in organisms as diverse as cyanobacteria, fungi, plants, flies, mice, and humans, and in these organisms, circadian rhythms influence a broad range of biological processes ranging from metabolism, to cell cycle control, to complex behaviors like the sleep–wake cycle. The phylogenetic ubiquity of circadian rhythms suggests that such rhythms must confer an evolutionary advantage. Indeed, decreased fitness has been demonstrated in

cyanobacteria (Ouyang et al., 1998), plants (Green et al., 2002), and flies (Beaver et al., 2002) with mutations resulting in abnormal circadian rhythms. There are at least two general mechanisms for this. First, the capacity to generate circadian rhythms allows an organism to anticipate rather than react to cyclic environmental cues such as the day– night cycle. Second, internal circadian rhythms provide a central temporal framework with which to sequence internal biological events relative to one another. In humans, circadian rhythms influence nearly every major organ system. Among these is the brain. The propensity for sleep and wake are among the most obvious of major brain processes to show circadian regulation, with clear circadian rhythmicity of sleep propensity and alertness, and disruption of sleep when out of synchrony with internal circadian rhythms (Czeisler et al., 1980; Wyatt et al., 1999).

*Correspondence to: Andrew S.P. Lim, Division of Neurology, Department of Medicine, Sunnybrook Health Sciences Centre, University of Toronto, 2075 Bayview Ave—Room M1-600, Toronto, ON M4N 3M5, Canada. Tel: +1-416-480-6100 x2461, Fax: +1-416-480-6092, E-mail: [email protected]

272

A.S.P. LIM

Cognition, a key brain function, also shows considerable circadian rhythmicity. Investigators reported prominent 24-h variation in cognitive performance (Kleitman, 1933) as early as the 1930s, an observation that was replicated by others in later studies (Folkard, 1975). With the widespread application of human experimental protocols such as constant routine and forced desynchrony protocols that allow for true circadian effects to be separated from the effects of environmental and behavioral factors, including sleep and wake, a host of studies have demonstrated robust circadian rhythms in cognitive processes as diverse as addition (Wyatt et al., 1999; Darwent et al., 2010), digit symbol substitution (Wyatt et al., 1999; Darwent et al., 2010), cued recall (Wyatt et al., 1999), language processing (Rosenberg et al., 2009), target tracking (Jasper et al., 2010), visual search (Pomplun et al., 2012), inhibitory motor control (Sagaspe et al., 2012), and simulated driving (Matthews et al., 2012). Moreover, experimental chronic circadian disruption is associated with declines in cognitive performance (Lee et al., 2009b; Cohen et al., 2010; Silva et al., 2010).

Altered diurnal rhythms, cognitive impairment, and dementia in older adults Whereas “circadian” refers specifically to 24-h physiological or behavioral rhythms that persist in the absence of external time cues, most community-based human studies are done with participants exposed to cyclical environmental time cues (e.g., the light–dark cycle), and thus the rhythms observed in these studies are more properly referred to as “diurnal” rather than “circadian.” Several studies have identified links between altered behavioral diurnal rhythmicity and ADRD. Compared to adults without AD, adults with AD have decreased stability and increased fragmentation of behavioral diurnal rhythms measured by actigraphy (Witting et al., 1990). Longitudinal studies suggest that in older community-dwelling adults, lower amplitude or less stable diurnal activity rhythms may be associated with a more rapid subsequent cognitive decline and higher likelihood of subsequently developing dementia (Tranah et al., 2011; Rogers-Soeder et al., 2018). Moreover, imaging studies suggest that increased fragmentation of diurnal activity rhythms is associated with a higher burden of cortical amyloid-beta as measured by positron emission tomography, even in individuals without clinical AD (Musiek et al., 2018). A later phase of diurnal rest activity rhythms has also been associated with a higher likelihood of developing dementia (Tranah et al., 2011; Walsh et al., 2014; Bokenberger et al., 2017; Suh et al., 2018).

SEASONAL RHYTHMS AND DEMENTIA IN OLDER ADULTS There is emerging evidence that season may also be associated with variation in human cognition. Seasonal rhythms modulate fMRI brain responses to cognitive tasks (Meyer et al., 2016), and several studies suggest that season may modulate cognition in younger adults (Brennen et al., 1999; Rajajarvi et al., 2010), although this is not a universal finding (Lacny et al., 2011; Afsar and Kirkpantur, 2013; Meyer et al., 2016). One recent study found an association between season and cognitive function in multiple independent cohorts of older adults, with cognitive function, higher in the summer and fall compared to winter and spring with a 30% higher odds of meeting clinical criteria for MCI or dementia in the winter or spring (Lim et al., 2018).

DIURNAL BRAIN MOLECULAR RHYTHMS IN MODEL ORGANISMS Organization of the circadian timing system The key structure responsible for generating and maintaining circadian rhythms in rodents and other mammals is the suprachiasmatic nucleus in the anterior hypothalamus. Lesions of the suprachiasmatic nucleus abolish circadian rhythms of locomotor activity, feeding, and cortisol secretion in rats (Moore and Eichler, 1972; Stephan and Zucker, 1972), and neural suprachiasmatic nucleus grafts from adult or fetal donors can restore locomotor behavioral circadian rhythms to hamsters with suprachiasmatic nucleus ablation (Ralph et al., 1990) or to genetically arrhythmic mice (Sujino et al., 2003). Moreover, in mice that are genetically arrhythmic due to deletion of the clock gene BMAL1 (see later), anatomically selective restoration of BMAL1 in the suprachiasmatic nucleus is sufficient to restore circadian rhythms of locomotor behavior (Fuller et al., 2008). Isolated suprachiasmatic nucleus neurons have self-sustained circadian rhythms of electrical activity when maintained in culture (Welsh et al., 1995; Liu et al., 1997; Herzog et al., 1998). However, individual cells exhibit a wide range of circadian periods (Welsh et al., 1995; Liu et al., 1997; Ko et al., 2010) and intercellular coupling synchronizes neurons (Herzog et al., 2004), confers precision to the population output of suprachiasmatic nucleus neurons (Herzog et al., 2004), and confers robustness against genetic perturbation (Liu et al., 2007a). This coupling is dependent on neuronal firing (Yamaguchi et al., 2003) and on the vasoactive intestinal polypeptide neurotransmitter system (Aton et al., 2005; Maywood et al., 2006; Brown et al., 2007; Hughes et al., 2008). Under normal conditions, circadian rhythms generated by the SCN are ultimately synchronized or entrained

DIURNAL AND SEASONAL MOLECULAR RHYTHMS IN THE HUMAN BRAIN to the external environment. The most important entraining signal is light. Environmental light cues are ultimately sensed by intrinsically photosensitive retinal ganglion cells expressing the photopigment melanopsin, which project to the SCN via the retinohypothalamic tract, and are necessary for photic entrainment (Gooley et al., 2001; Berson et al., 2002; Hattar et al., 2002, 2003; Panda et al., 2003, 2005; Goz et al., 2008; Hatori et al., 2008). These retinal ganglion cells utilize both glutamate and pituitary adenylate cyclase activating polypeptide as neurotransmitters (Ding et al., 1994; Ebling, 1996; Hannibal et al., 2004; Michel et al., 2006) and ultimately trigger calcium influx in suprachiasmatic nucleus neurons, activation of protein kinase pathways, and phosphorylation of Ca2 +/AMP response element binding protein (CREB) inducing expression of the period genes (see later) (Albrecht et al., 1997, 2001; Shearman et al., 1997; Zhang et al., 2005; Golombek and Rosenstein, 2010). Circadian rhythms generated in the suprachiasmatic nucleus are conveyed to other organs by a variety of signals. Locally, suprachiasmatic nucleus neurons project directly or indirectly to a number of key hypothalamic structures including the dorsal and ventral subparaventricular zones, the dorsomedial hypothalamus, the paraventricular nucleus, the lateral hypothalamic area, the ventrolateral preoptic nucleus, and the preoptic area (Swanson and Cowan, 1975; Leak and Moore, 2001; Lu et al., 2001; Chou et al., 2003; Saper et al., 2005). Key neuropeptides include transforming growth factor alpha (Kramer et al., 2001), vasopressin, and Prokineticin 2 (Cheng et al., 2002, 2005). These local hypothalamic projections drive circadian rhythms of body temperature (preoptic area) (Brown et al., 2002; Buhr et al., 2010; Saini et al., 2012; Guzman-Ruiz et al., 2015), cortisol secretion (paraventricular nucleus) (Kalsbeek et al., 1996; Balsalobre et al., 2000; Yamamoto et al., 2005; Reddy et al., 2007; Fujihara et al., 2014), and autonomic tone (paraventricular nucleus) (Buijs et al., 1999; Ueyama et al., 1999; Cailotto et al., 2005, 2009; Ishida et al., 2005; Vujovic et al., 2008; Kalsbeek et al., 2010) that in turn entrain circadian rhythms in organs outside the suprachiasmatic nucleus. An evolutionarily conserved transcriptional negative feedback loop lies at the core of the molecular clock present in neurons of the SCN and indeed in nucleated cells of most other tissues (Yamazaki et al., 2000; Yamamoto et al., 2004; Yoo et al., 2004). At the core of the molecular clock are the homologous period 1 (Per1) and period 2 (Per2) (Zheng et al., 1999, 2001; Bae et al., 2001) and the homologous cryptochrome 1 (Cry1) and cryptochrome 2 (Cry2) (Griffin et al., 1999; Vitaterna et al., 1999). The transcription of Per1/Per2 and Cry1/Cry2 is activated by the binding of a complex containing the transcription factors brain muscle arnt-like 1 (BMAL1)

273

and circadian locomotor output cycles kaput (CLOCK) to E-box motifs in their promoter regions (King et al., 1997; Gekakis et al., 1998; Hogenesch et al., 1998; Jin et al., 1999). PER1/PER2 protein and CRY1/CRY2 protein accumulate in the cytosol where they form PER-CRY heterodimers (Kume et al., 1999). These heterodimers then translocate to the nucleus where they bind BMAL1-CLOCK complexes and inhibit further transcription of Per1/Per2 and Cry1/Cry2 (Kume et al., 1999; Shearman et al., 2000; Lee et al., 2001; Sato et al., 2006) leading to falling PER1/PER2 and CRY1/ CRY2, falling PER-CRY heterodimer levels, and thus release of the inhibition of BMAL1-CLOCK and resumption of Per1/Per2 and Cry1/Cry2 transcription. This cycle takes
24 h. Several other genes and proteins participate in the feedback loop described previously. Basic helix-loophelix family member e41 and Basic helix-loop-helix family member e40 are transcription factors that are found at E-box elements and repress CLOCK-BMAL1induced transcription (Honma et al., 2002). Thyroid hormone receptor-associated protein 150 binds to the CLOCK-MAL1 complex and stimulates CLOCKBMAL1-associated transcription (Lande-Diner et al., 2013). Metastasis-associated protein 1 (MTA1) also binds to the CLOCK-BMAL1 complex and facilitates CRY1-mediates repression of CLOCK-BMAL1 transcriptional activity (Li et al., 2013a). Circadianassociated repressor of transcription (CHRONO) is another protein that binds the CLOCK-BMAL1 complex and represses CLOCK-BMAL1 transcriptional activity (Goriki et al., 2014). In addition to the previously mentioned, there are some tissue-dependent variations. For instance, in the mouse forebrain, neuronal PAS domain containing protein 2 (NPAS2) forms heterodimers with BMAL1 in a similar manner as CLOCK and appears to play a nonredundant role in regulating forebrain transcriptional rhythms (Reick et al., 2001). BMAL1 participates in a second interlocking negative feedback loop involving two other transcription factors RAR-related orphan receptor alpha (ROR-alpha) and REV-ERB-alpha. In this loop, the transcription of Bmal1 is activated by the binding of ROR-alpha to ROR response elements (ROREs) in its promoter (Sato et al., 2004; Akashi and Takumi, 2005). The BMAL1: CLOCK complex then activates the transcription of Rev-Erb-alpha by binding to E-boxes in its promoter region (Preitner et al., 2002). REV-ERB-alpha then binds to ROREs in the Bmal1 promoter (Ueda et al., 2002), resulting in repression of Bmal1 transcription and therefore decreasing levels of BMAL1 (Preitner et al., 2002), decreased E-box-mediated Rev-Erb-alpha transcription, and disinhibition of RORE-mediated Bmal1 transcription, completing the cycle. In addition to the previously

274

A.S.P. LIM

mentioned, ROR-alpha is itself cyclic due to E-boxes in its promoter region (Sato et al., 2004). Moreover, a homolog of REV-ERB-alpha, REV-ERB-beta is largely redundant with it (Liu et al., 2008). Light entrainment of the molecular clock is achieved in part by induction of Per1I and Per2 transcription (Albrecht et al., 1997, 2001; Shearman et al., 1997). Calcium influx leads to phosphorylation of cAMP response element binding protein, which in turn binds to cAMP response elements in Per1 and Per2 promoters (Zhang et al., 2005; Golombek and Rosenstein, 2010). Transcriptional activation by the BMAL1-CLOCK complex, which lies at the core of the molecular clock, is mediated in large part by dynamic histone acetylation and methylation. For instance, rhythmic BMAL1CLOCK transcriptional activation is accompanied by rhythmic H3K and H4K acetylation paralleling transcription of target genes (Etchegaray et al., 2003; Ripperger and Schibler, 2006). This may be due to several mechanisms including recruitment of E1A binding protein p300 (EP300) and CREB-binding protein (CBP) which are both histone acetyltransferases (Takahata et al., 2000; Etchegaray et al., 2003; Curtis et al., 2004). As well, CLOCK itself possesses intrinsic histone acetyltransferase activity, an activity central to its function, and is capable of directly acetylating H3 and H4, especially H3K14 and histone 3 lysine 9 (H3K9) (Doi et al., 2006). In addition to this, the BMAL1-CLOCK complex recruits JumonjiC and ARID domain containing histone lysine demethylase 1a (JARID1a), which among other functions is an inhibitor of histone deacetylase 1 (HDAC1) (DiTacchio et al., 2011). Transcriptional activation by BMAL1-CLOCK may also depend in part on histone methylation. The BMAL1CLOCK complex recruits the histone methyltransferase mixed lineage leukemia 1 (MLL1) to target gene promoters with associated local trimethylation of H3K4 (Katada and Sassone-Corsi, 2010). Interestingly, the histone demethylase function of JARID1a does not seem to play a role in its impact on BMAL1-CLOCK-mediated transcriptional activation (DiTacchio et al., 2011). Given the importance of histone acetylation and methylation to the transcriptional activation function of the BMAL1-CLOCK complex, it is not surprising that rhythmic inhibition of BMAL1-CLOCK transcriptional activity is also achieved in part by modifying histone acetylation and methylation. CRY1 and PER1 are associated with PTB-associated splicing factor and SIN3 transcription regulator family member A (SIN3A) which in turn are associated with HDAC1, resulting in decreased acetylation of H3K9 and H4K5 in the promoter regions of BMAL1-CLOCK target genes, leading to transcriptional repression (Naruse et al., 2004; Duong et al., 2011). BMAL1-CLOCK transcriptional activation

and H3K9 acetylation is also repressed by a recently discovered clock protein circadian-associated repressor of transcription (CHRONO) which decreases H3K9 acetylation by abrogating BMAL1 binding to CBP (Anafi et al., 2014) and also binding HDAC1 (Goriki et al., 2014). In addition to binding the SIN3-HDAC1 complex, PER1 also binds suppressor of variegation 39 homolog 1 (SUV39H1), a histone methyltransferase which promotes dimethylation and trimethylation at histone H3K9, which is associated with transcriptional repression (Duong and Weitz, 2014). CRY1-mediated BMAL1-CLOCK transcriptional repression is enhanced by the presence of the enhancer of zeste 2 polycomb repressive complex 2 subunit (EZH2), a histone methyltransferase, which methylates H3K27, a repressive mark (Etchegaray et al., 2006). In addition to this, CRY1 interacts with type II protein arginine methyltransferase 4 (PRMT5), a histone methyltransferase that promotes H4R dimethylation which is a repressive histone mark (Na et al., 2012). Finally, sirtuin 1 binds to CLOCK and rhythmically deacetylates H3K9, H3K14, and H4K16, leading to transcriptional repression (Asher et al., 2008; Nakahata et al., 2008). Histone modifications also play a role in mediating transcriptional activation/repression at ROREs by ROR-alpha and REV-ERB-alpha. ROR-alpha binds to peroxisome proliferator-activate receptor gamma, coactivator 1 alpha (PGC-1alpha), which in turn binds P300 and general control of amino-acid synthesis protein 5-like 2 (GCN5) which are both histone acetyltransferases, and which lead to H3K9 acetylation at the Bmal1 promoter (Liu et al., 2007b). Meanwhile, REV-ERBalpha recruits nuclear receptor co-repressor 1 histone deacetylase 3 complexes to ROREs, resulting in histone H3K9 deacetylation and transcriptional repression (Yin and Lazar, 2005; Yin et al., 2007; Alenghat et al., 2008; Feng et al., 2011). An additional layer of regulation of the core molecular clock is provided by a number of posttranscriptional mechanisms. miRNAs collectively contribute to a delay in the translation of some core clock genes, as evidenced by faster translation in Dicer-deficient mice, who lack miRNA (Chen et al., 2013). A specific miRNA, miRNA-132 appears to play a specific role in modulating the circadian entrainment in the SCN by light pulses (Cheng et al., 2007; Alvarez-Saavedra et al., 2011). In addition, a number of RNA binding protein have been identified that bind the mRNAs of core clock genes and either modulate their stability or enhance their translation (Kojima et al., 2007; Woo et al., 2009, 2010; Morf et al., 2012; Preussner et al., 2014). Posttranslational mechanisms, especially phosphorylation, play a key role in regulating the molecular clock,

DIURNAL AND SEASONAL MOLECULAR RHYTHMS IN THE HUMAN BRAIN largely by regulating the proteolytic degradation and subcellular compartmentalization of clock proteins. For instance, phosphorylation of PER1 and PER2 by casein kinase I epsilon promotes ubiquitination of PER1 and PER2 by Skp1-Cul1-F-box (SCF) protein ubiquitin ligase complexes containing the F-box protein betatransducing repeat containing protein 1 (Eide et al., 2005; Shirogane et al., 2005). PER2 is also phosphorylated by casein kinase I delta (Xu et al., 2005), glycogen synthase kinase 3 beta (GSK3-beta) (Iitaka et al., 2005), and casein kinase 2 (CK2) (Maier et al., 2009). While phosphorylation at some sites appears to promote proteolytic degradation as described previously, phosphorylation at other sites seems to be necessary for nuclear localization (Vanselow et al., 2006; Lee et al., 2009a). The cryptochrome proteins also undergo phosphorylation by GSK3-beta (Harada et al., 2005) and 50 AMPactivated protein kinase (Lamia et al., 2009), leading to their ubiquitination by an SCF ubiquitin protein ligase complex containing the protein F-box/LRR-repeat protein 3 (FBXL3) and subsequent proteolytic degradation (Busino et al., 2007; Godinho et al., 2007; Siepka et al., 2007). CLOCK is phosphorylated by GSK3-beta targeting it for degradation (Spengler et al., 2009). Meanwhile, BMAL is phosphorylated by GSK3-beta (Sahar et al., 2010), CK2 (Tamaru et al., 2003), and protein kinase C alpha (PKC-alpha) (Robles et al., 2010), with resulting regulation of nucleocytoplasmic localization (Tamaru et al., 2003), proteolytic degradation (Sahar et al., 2010), and transcriptional activity (Robles et al., 2010). Finally, REV-ERB-alpha is also ubiquitinated by SCF complexes containing FBXL3 (Shi et al., 2013) and is phosphorylated by GSK3-beta, which contributes to its stability (Yin et al., 2006).

Diurnal rhythms of brain gene expression Cell autonomous molecular clocks are also present in many tissues outside the suprachiasmatic nucleus with the latter being synchronized with the former by means of physiological (body temperature, autonomic innervation) and hormonal (cortisol, melatonin) signals. Moreover, peripheral clocks can also be entrained independent of the SCN by peripheral signals such as feeding (Damiola et al., 2000). Peripheral clocks in turn influence tissue physiology by regulating the expression of sets of genes that mediate tissue function, referred to as clock-controlled genes. Indeed, in mice, it is estimated that between 5% and 20% of the transcriptome in suprachiasmatic nucleus (Panda et al., 2002; Ueda et al., 2002), liver (Panda et al., 2002; Storch et al., 2002; Ueda et al., 2002; Zvonic et al., 2006; Miller et al., 2007; Koike et al., 2012; Le Martelot et al., 2012; Vollmers et al., 2012;

275

Zhang et al., 2014), myocardium (Storch et al., 2002; Zhang et al., 2014), adrenal (Oster et al., 2006; Zhang et al., 2014), brown adipose tissue (Zvonic et al., 2006; Zhang et al., 2014), white adipose tissue (Zvonic et al., 2006, Zhang et al., 2014), skeletal muscle (Miller et al., 2007; Zhang et al., 2014), retina (Storch et al., 2007), kidney (Zuber et al., 2009; Zhang et al., 2014), lungs (Zhang et al., 2014), and aorta (Zhang et al., 2014) show 24-h rhythms of abundance. Cross-tissue comparisons (Akhtar et al., 2002; Panda et al., 2002; Storch et al., 2002; Ueda et al., 2002; Ptitsyn et al., 2006; Miller et al., 2007; Yan et al., 2008; Zhang et al., 2014) reveal several features of the circadian transcriptome. First, when many tissues are considered, a large minority (estimated at 43% in one study (Zhang et al., 2014)) of transcripts are rhythmic in at least one organ. However, the overlap between organs in minimal, and indeed only a small minority of transcripts, mostly core clock genes, are rhythmic in a majority of organs (Yan et al., 2008; Zhang et al., 2014). Second, even where a transcript is rhythmic in more than one organ, the timing of expression may vary dramatically between organs. In one study of 12 tissues, >1400 genes were phase shifted with respect to themselves by at least 6 h between two organs, and some genes were expressed at different times in many organs (Zhang et al., 2014). These observations suggest that the circadian transcriptome is highly tissue-specific (outside of core clock genes) and limit the capacity to extrapolate from one tissue to another. Considering the mammalian cerebral cortex specifically, it is estimated that 5%–15% of cerebral cortical genes show diurnal variation in expression, including 10% in rats (Cirelli et al., 2004), 8%–10% in mice (Maret et al., 2007; Yang et al., 2007), and 12% in baboons (Mure et al., 2018). One important distinction relevant to cerebral cortex gene expression is to what extent these gene expression differences are driven directly by the circadian clock, and to what extent they are primarily reflective of sleep–wake state. Several studies suggest that in the case of the cerebral cortex specifically, sleep–wake state drives the diurnal variation in a majority of diurnally rhythmic genes (Cirelli et al., 2004; Maret et al., 2007). One recent study modeled the expression of individual mouse cerebral cortical genes as a function of sleep effects and circadian effects with or without amplitude modulation by sleep deprivation (Hor et al., 2019). It was inferred that sleep–wake history was a more important driver of overall transcriptome changes than was circadian rhythmicity per se. A majority of genes identified as diurnally rhythmic were found to be primarily sleep related and diurnally rhythmic genes that were primarily or partially sleep related

276

A.S.P. LIM

included many core clock genes including CLOCK, NPAS2, PER1, PER2, and DBP. Another important feature of the cerebral cortical transcriptome is the importance of cellular compartment. For instance, a recent study found that whereas only 6% of mouse cerebral cortex transcripts were found to be rhythmic when bulk tissue was considered, 67% of synaptoneurosome mRNAs were diurnally rhythmic, the vast majority of which were rhythmic in the synaptoneurosome only (Noya et al., 2019).

Diurnal rhythms of brain epigenetic modification With the advent of technologies allowing genome-wide assessment of chromatin modification, a number of studies have examined rhythms of H3K9ac, H3K4me3, and other chromatin modifications in mouse liver on a genome-wide scale (Koike et al., 2012; Le Martelot et al., 2012; Vollmers et al., 2012). Indeed, in mouse liver, there are circadian rhythms of many of these histone modifications throughout the genome, with both global and site-specific rhythms, and overlapping with ROR-alpha/REV-ERBalpha and BMAL1-CLOCK binding sites. There is also evidence for diurnal rhythms of DNA methylation in several tissues. In mouse liver, methylcytosine levels peak in the middle of the light period, several hours after peak hepatic levels of DNMT3b (Maekawa et al., 2012). Moreover, a recent study examined cytosine modifications in a single chromosome in mouse lungs and the liver and found that 8% of epigenetically variable cytosines in the liver and 35% in the lungs were diurnally rhythmic (Oh et al., 2018). There are relatively fewer data concerning diurnal rhythms of cerebral cortical histone modification and DNA methylation in model organisms. However, one study found several hundred cytosines differentially methylated under conditions of sleep deprivation in the mouse, suggesting an important sleep modulation of cerebral cortical DNA methylation (Massart et al., 2014).

SEASONAL BRAIN MOLECULAR RHYTHMS IN MODEL ORGANISMS There are relatively fewer data concerning seasonal molecular rhythms in the mammalian brain, much of which has come from the study of hibernating species. For instance, there is differential expression of a number of genes in the hypothalamus of Siberian hamsters sacrificed while living in long vs. short photoperiods, many of which are related to hormone secretion and neuropeptide signaling (Bao et al., 2019). There is also differential gene expression in the hypothalamus of dromedary camels in summer vs. winter (Alim et al., 2019). There

are less data on seasonal rhythms of gene expression in the mammalian cortex, although one study found that roughly 1000 genes are differentially expressed in the cortex of the ground squirrel between prehibernation (the October active period), torpor, interbout arousals, and posthibernation (the April active period) states (Schwartz et al., 2013). At least in the hypothalamus, some of these differences in gene expression may be driven by differences in DNA methylation. In the Siberian hamster, the level of DNA Methyltransferase 3a is lower under short photoperiod conditions than long (Stevenson, 2017), and this is associated with decreased methylation of the Dio3 promoter and increased DIO3 levels compared to exposure to long photoperiods (Stevenson and Prendergast, 2013).

DIURNAL AND SEASONAL MOLECULAR RHYTHMS IN THE HUMAN NEOCORTEX Diurnal molecular rhythms Diurnal rhythms of gene expression and epigenetic modification in the human neocortex, if present, may represent a key mechanism underlying diurnal rhythms of human brain processes including cognition. Moreover, alterations in diurnal rhythmicity of gene expression and epigenetic modification in brain diseases such as Alzheimer’s disease, may represent a key mechanism mediating the bidirectional links between Alzheimer’s disease and altered diurnal rhythms. Thus, identification of genes and epigenetic loci whose rhythmicity is altered in Alzheimer’s disease may identify targets for therapeutic interventions to prevent the negative effects of altered diurnal rhythmicity on Alzheimer’s disease and the negative effects of Alzheimer’s disease on diurnally rhythmic processes in the human brain. Because it is impossible to serially sample human neocortical tissue from living human subjects, or to control the timing or circumstances of death, inferences about the presence and nature of human neocortical diurnal molecular rhythms have largely come from observational postmortem studies, of which there have been several. All have modeled measurements of gene expression and epigenomic modification as a function of recorded time of death, making the implicit assumption that study participants were in phase at the time of death. Using this approach, diurnal rhythms of clock gene expression can be inferred, with the relative timing of clock gene expression remarkably consistent across datasets (Li et al., 2013b; Chen et al., 2015; Lim et al., 2017). Moreover, when the neocortical transcriptome as a whole is considered, there is also evidence of diurnal rhythmicity, with a cluster of transcripts peaking at sunset, and a separate cluster peaking at sunrise (Lim et al., 2017).

DIURNAL AND SEASONAL MOLECULAR RHYTHMS IN THE HUMAN BRAIN Depending on the size of the study, and the statistical threshold used, somewhere between 1% and 10% of the human neocortical transcriptome appears to be diurnally rhythmic, with substantial overlap in the genes identified as rhythmic between datasets (Li et al., 2013b; Chen et al., 2015; Lim et al., 2017; Seney et al., 2019). Moreover, the timing and amplitude of these rhythms is different in brains from older vs. younger individuals and from individuals with and without neurological and psychiatric disease. For instance, the transcriptome as a whole is relatively phase advanced in brains with a pathological diagnosis of Alzheimer’s disease vs. those without (Lim et al., 2017). When brains from older individuals, or individuals with depression or schizophrenia are examined, several hundred transcripts have evidence of less robust rhythmicity in older vs. younger brains, or in brains from individuals with compared to without depression or schizophrenia (Li et al., 2013b; Chen et al., 2015; Seney et al., 2019). However, interestingly there are also many transcripts that show more robust rhythmicity with age or with schizophrenia (Chen et al., 2015; Seney et al., 2019). Controlled manipulation of the rhythmicity of these genes in appropriate animal models of aging, depression, schizophrenia, and Alzheimer’s disease are needed to determine whether these alternations in rhythmicity have functional consequences visà-vis diurnal rhythms of brain function and behavior, and vis-à-vis the progression of underlying disease. Evidence for diurnal rhythms of epigenetic modification in the human neocortex can also be inferred from postmortem studies. The cytosine methylome and histone 3 lysine 9 acetylome as a whole both show evidence of diurnal rhythmicity, with clusters of H3K9 acetylation and cytosine methylation sites showing peak H3K9 acetylation or nadir cytosine methylation around sunrise and sunset (Lim et al., 2014, 2017). Interestingly, there is an association between the timing of the diurnal rhythmicity of a cytosine methylation or H3K9 acetylation site and its physical position relative to nearby active transcription start sites, with H3K9 acetylation peaks within 2 kb of active transcription start sites relatively more likely show peak acetylation near sunset, and cytosine methylation sites within 2 kb of active transcription start sites relatively more likely to show nadir methylation near sunrise (Lim et al., 2017). A linkage between rhythms of diurnal rhythms of epigenetic modification and those of gene expression is also supported by the observation that the 2 kb proximate to the transcription start site of evening-peaking genes is relatively enriched in evening-peaking H3K9 acetylation sites and evening nadir DNA methylation sites (Lim et al., 2017), with, on average, a
2 h interval between nadir DNA methylation and peak gene abundance (Lim et al., 2014). On balance,

277

considering the cytosine methylome as a whole, the amplitude of diurnal rhythmicity may be modestly attenuated in brains with a pathological diagnosis of Alzheimer’s disease than those without, but there is considerable variation between individual methylation sites. There were not significant differences in the phase of diurnal rhythms of epigenetic modification between brains with and without a pathological diagnosis of AD, when adjusted for seasonal rhythmicity (see later), suggesting that the observed differences in the phase of gene expression may not be primarily due to epigenetic differences (Lim et al., 2017).

Seasonal molecular rhythms Similar to diurnal rhythms, seasonal molecular rhythms can also be inferred from the human postmortem data, by considering gene expression and epigenetic modification as a function of date rather than (or in addition to) time of death. One study using this approach found significant seasonal rhythms of canonical clock genes in the human prefrontal cortex. Moreover, there was evidence for seasonal rhythmicity when the transcriptome as a whole was considered (Lim et al., 2017). There was also evidence of seasonal rhythmicity in the H3K9 acetylome and cytosine methylome considered as a whole (Lim et al., 2017). As with diurnal rhythms, there was, on average, an association between the timing of the seasonal nadir of methylation and the timing of peak transcript expression of nearby genes. However, a similar association was not demonstrated for the H3K9 acetylome (Lim et al., 2017). Interestingly, there appeared to be an association between the seasonal peak of a transcript expression and its diurnal peak, such that sunset-peaking transcripts tended to peak in the spring and sunrise-peaking transcripts tended to peak in the fall. Similar relationships were seen at the level of H3K9 acetylation, but not DNA methylation. Moreover, transcription factor binding sites whose presence favored spring-peaking rhythmicity also favored evening-peaking rhythmicity. Together, these suggest that shared regulatory mechanisms may regulate the diurnal and seasonal rhythmicity of genes (Lim et al., 2017). When specific gene coexpression systems were examined, several gene coexpression systems strongly associated with cognition were also seasonally rhythmic, with systems positively associated with cognition peaking in the late summer and early fall, coincident with the seasonal peak of cognition, and systems negative associated with cognition peaking in the late winter and early spring, coincident with the seasonal nadir of cognition (Lim et al., 2018). The rhythmicity of these coexpression systems persisted in the context of a pathological

278

A.S.P. LIM

diagnosis of Alzheimer’s disease (Lim et al., 2018). However, when the transcriptome as a whole is considered, on average, there is a greater amplitude and delayed phase of seasonal gene expression rhythms (Lim et al., 2017).

Limitations and opportunities Taken together, these data suggest that significant diurnal and seasonal rhythms of gene expression, related in part to parallel rhythms of epigenetic modification, are present in the human neocortex and may be influenced by aging and key brain disorders like Alzheimer’s disease, depression, and schizophrenia. However, these studies share common limitations and leave unanswered several key questions. With rare exceptions, the human neocortical transcriptome and epigenome can only be assessed at a single point in time—at death. Thus estimating diurnal and seasonal rhythmicity relies on pooling data from many individuals and making the implicit assumption that the samples were in phase at the time of death. For experiments on groups of model organisms in controlled environments, this is a reasonable assumption. However, the individuals donating neocortical tissue for human postmortem studies would have all been living in different environments in the days or weeks leading up to death. This makes it highly likely that they are in fact out of phase at the time of death. The presence of a period of terminal decline leading to death, and the fact that some individual die at home, while others die in the hospital or other nonhabitual environments, only accentuates this problem. One way of partially accounting for this might be direct measurement of the environment (e.g., light exposure) and behavior (e.g., sleep–wake cycles) for donors in the weeks or days leading up to death. Since the date of death is unpredictable, this would in effect require continuous monitoring of large numbers of individuals for long periods of time. Another way to address this would be to directly assess circadian rhythmicity at time of death, for instance by sampling and maintenance in culture of cells from hair follicles (Akashi et al., 2010) or other sources. Another important problem created by a reliance on tissue from free-living human subjects is that of masking by environmental (e.g., temperature, light) or behavioral (e.g., sleep/wake state, physical activity). Thus from these sort of observational data, it can be difficult to impossible to distinguish true endogenous circadian/ seasonal effects from the effects of behaviors or environmental factors showing diurnal/seasonal rhythmicity. An important case in point is the effect of sleep. As noted previously, in model organisms, a large preponderance of diurnally rhythmic neocortical transcripts is more strongly related to sleep/wake state than to endogenous

circadian rhythmicity per se (Maret et al., 2007; Hor et al., 2019), a distinction that can only be made with experimental designs that dissociate sleep from circadian rhythmicity. From a clinical translational perspective, this question is of great importance as there are many therapeutics that may modify sleep independent of the circadian pacemaker and vice versa. In principle, one way to address this in observational studies would be careful measurement of potential masking factors (e.g., sleep, physical activity, temperature, light) in the hours leading to death and leveraging spontaneously occurring instances of dissociation of sleep, physical activity, and other factors from underlying circadian and seasonal rhythmicity. However, as previously mentioned, because of the unpredictable nature of death, this would require nearly continuous monitoring of the environment and physiology in large numbers of individuals over years. The reliance on postmortem tissue also introduces cause of death as an important confounding factor— specific causes of death may be more common at particular times of the day (e.g., cardiovascular disease in the morning) and times of the year (e.g., pneumonia in the winter), which may in turn differentially affect the neocortical epigenome or transcriptome. Careful ascertainment of the cause of death and time course of agonal illness is necessary to account or this. Postmortem studies of the human SCN have revealed both circadian and seasonal rhythms of key genes such as vasopressin that vary with age (Hofman and Swaab, 1994, 1995), Alzheimer’s disease (Liu et al., 2000) and depression (Liu et al., 2000; Zhou et al., 2001). Thus far, studies examining diurnal and seasonal molecular rhythms in the human neocortex have examined the neocortex independent of subcortical structures such as the SCN. Thus the extent to which differences in diurnal or seasonal neocortical molecular rhythms reflect hypothalamic and other subcortical influences is uncertain. This is an important question because if disease effects on cortical diurnal and seasonal rhythmicity are purely “downstream” of the hypothalamus, then interventions targeting the hypothalamus (e.g., interventions targeting primarily the central circadian pacemaker) may not address these effects. Studies simultaneously assessing diurnal and seasonal rhythmicity in the hypothalamus and multiple cortical and subcortical regions in human tissue are needed to clarify these relationships. Finally, human postmortem diurnal and seasonal brain tissue studies thus far have examined only bulk neocortical tissue, containing a mixture of neurons, microglia, astrocytes, oligodendrocytes, and others. However, it is highly likely that different genes cycle in different cell types, and they do in different tissues (Zhang et al., 2014). Moreover, as noted previously, recent studies suggest that diurnal rhythmicity may differ

DIURNAL AND SEASONAL MOLECULAR RHYTHMS IN THE HUMAN BRAIN even by cellular compartment—e.g., synaptoneurosome (Noya et al., 2019). Determining in greater detail which gene cycle in which cells in which compartments is of clinical translational significance since many therapeutics target specific brain cell types.

REFERENCES Afsar B, Kirkpantur A (2013). Are there any seasonal changes of cognitive impairment, depression, sleep disorders and quality of life in hemodialysis patients? Gen Hosp Psychiatry 35: 28–32. Akashi M, Takumi T (2005). The orphan nuclear receptor RORalpha regulates circadian transcription of the mammalian core-clock Bmal1. Nat Struct Mol Biol 12: 441–448. Akashi M, Soma H, Yamamoto T et al. (2010). Noninvasive method for assessing the human circadian clock using hair follicle cells. Proc Natl Acad Sci U S A 107: 15643–15648. 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. Albrecht U, Sun ZS, Eichele G et al. (1997). A differential response of two putative mammalian circadian regulators, mper1 and mper2, to light. Cell 91: 1055–1064. Albrecht U, Zheng B, Larkin D et al. (2001). MPer1 and mper2 are essential for normal resetting of the circadian clock. J Biol Rhythms 16: 100–104. Alenghat T, Meyers K, Mullican SE et al. (2008). Nuclear receptor corepressor and histone deacetylase 3 govern circadian metabolic physiology. Nature 456: 997–1000. Alim FZD, Romanova EV, Tay YL et al. (2019). Seasonal adaptations of the hypothalamo-neurohypophyseal system of the dromedary camel. PLoS One 14: e0216679. Alvarez-Saavedra M, Antoun G, Yanagiya A et al. (2011). miRNA-132 orchestrates chromatin remodeling and translational control of the circadian clock. Hum Mol Genet 20: 731–751. Anafi RC, Lee Y, Sato TK et al. (2014). Machine learning helps identify CHRONO as a circadian clock component. PLoS Biol 12: e1001840. Asher G, Gatfield D, Stratmann M et al. (2008). SIRT1 regulates circadian clock gene expression through PER2 deacetylation. Cell 134: 317–328. Aton SJ, Colwell CS, Harmar AJ et al. (2005). Vasoactive intestinal polypeptide mediates circadian rhythmicity and synchrony in mammalian clock neurons. Nat Neurosci 8: 476–483. Bae K, Jin X, Maywood ES et al. (2001). Differential functions of mPer1, mPer2, and mPer3 in the SCN circadian clock. Neuron 30: 525–536. Balsalobre A, Brown SA, Marcacci L et al. (2000). Resetting of circadian time in peripheral tissues by glucocorticoid signaling. Science 289: 2344–2347. Bao R, Onishi KG, Tolla E et al. (2019). Genome sequencing and transcriptome analyses of the Siberian hamster hypothalamus identify mechanisms for seasonal energy balance. Proc Natl Acad Sci U S A 116: 13116–13121.

279

Beaver LM, Gvakharia BO, Vollintine TS et al. (2002). Loss of circadian clock function decreases reproductive fitness in males of Drosophila melanogaster. Proc Natl Acad Sci U S A 99: 2134–2139. Berson DM, Dunn FA, Takao M (2002). Phototransduction by retinal ganglion cells that set the circadian clock. Science 295: 1070–1073. Bokenberger K, Strom P, Dahl Aslan AK et al. (2017). Association between sleep characteristics and incident dementia accounting for baseline cognitive status: a prospective population-based study. J Gerontol A Biol Sci Med Sci 72: 134–139. Brennen T, Martinussen M, Hansen BO et al. (1999). Arctic cognition: a study of cognitive performance in summer and winter at 69 degrees N. Appl Cogn Psychol 13: 561–580. Brown SA, Zumbrunn G, Fleury-Olela F et al. (2002). Rhythms of mammalian body temperature can sustain peripheral circadian clocks. Curr Biol 12: 1574–1583. Brown TM, Colwell CS, Waschek JA et al. (2007). Disruptedneuronal activity rhythms in the suprachiasmatic nuclei of vasoactive intestinal polypeptide-deficient mice. J Neurophysiol 97: 2553–2558. Buhr ED, Yoo SH, Takahashi JS (2010). Temperature as a universal resetting cue for mammalian circadian oscillators. Science 330: 379–385. 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. Busino L, Bassermann F, Maiolica A et al. (2007). SCFFbxl3 controls the oscillation of the circadian clock by directing the degradation of cryptochrome proteins. Science 316: 900–904. Cailotto C, La Fleur SE, Van Heijningen C et al. (2005). The suprachiasmatic nucleus controls the daily variation of plasma glucose via the autonomic output to the liver: are the clock genes involved? Eur J Neurosci 22: 2531–2540. Cailotto C, Lei J, Van Der Vliet J et al. (2009). Effects of nocturnal light on (clock) gene expression in peripheral organs: a role for the autonomic innervation of the liver. PLoS One 4: e5650. Chen R, D’alessandro M, Lee C (2013). miRNAs are required for generating a time delay critical for the circadian oscillator. Curr Biol 23: 1959–1968. Chen CY, Logan RW, Ma T et al. (2015). Effects of aging on circadian patterns of gene expression in the human prefrontal cortex. Proc Natl Acad Sci U S A 113: 206–211. Cheng MY, Bullock CM, Li C et al. (2002). Prokineticin 2 transmits the behavioural circadian rhythm of the suprachiasmatic nucleus. Nature 417: 405–410. Cheng MY, Bittman EL, Hattar S et al. (2005). Regulation of prokineticin 2 expression by light and the circadian clock. BMC Neurosci 6: 17. Cheng HY, Papp JW, Varlamova O et al. (2007). microRNA modulation of circadian-clock period and entrainment. Neuron 54: 813–829. Chou TC, Scammell TE, Gooley JJ et al. (2003). Critical role of dorsomedial hypothalamic nucleus in a wide range of

280

A.S.P. LIM

behavioral circadian rhythms. J Neurosci 23: 10691–10702. Cirelli C, Gutierrez CM, Tononi G (2004). Extensive and divergent effects of sleep and wakefulness on brain gene expression. Neuron 41: 35–43. Cohen DA, Wang W, Wyatt JK et al. (2010). Uncovering residual effects of chronic sleep loss on human performance. Sci Transl Med 2: 14ra3. Curtis AM, Seo SB, Westgate EJ et al. (2004). Histone acetyltransferase-dependent chromatin remodeling and the vascular clock. J Biol Chem 279: 7091–7097. Czeisler CA, Weitzman E, Moore-Ede MC et al. (1980). Human sleep: its duration and organization depend on its circadian phase. Science 210: 1264–1267. Damiola F, Le Minh N, Preitner N et al. (2000). Restricted feeding uncouples circadian oscillators in peripheral tissues from the central pacemaker in the suprachiasmatic nucleus. Genes Dev 14: 2950–2961. Darwent D, Ferguson SA, Sargent C et al. (2010). Contribution of core body temperature, prior wake time, and sleep stages to cognitive throughput performance during forced desynchrony. Chronobiol Int 27: 898–910. Ding JM, Chen D, Weber ET et al. (1994). Resetting the biological clock: mediation of nocturnal circadian shifts by glutamate and NO. Science 266: 1713–1717. Ditacchio L, Le HD, Vollmers C et al. (2011). Histone lysine demethylase JARID1a activates CLOCK-BMAL1 and influences the circadian clock. Science 333: 1881–1885. Doi M, Hirayama J, Sassone-Corsi P (2006). Circadian regulator CLOCK is a histone acetyltransferase. Cell 125: 497–508. Duong HA, Weitz CJ (2014). Temporal orchestration of repressive chromatin modifiers by circadian clock Period complexes. Nat Struct Mol Biol 21: 126–132. Duong HA, Robles MS, Knutti D et al. (2011). A molecular mechanism for circadian clock negative feedback. Science 332: 1436–1439. Ebling FJ (1996). The role of glutamate in the photic regulation of the suprachiasmatic nucleus. Prog Neurobiol 50: 109–132. Eide EJ, Woolf MF, Kang H et al. (2005). Control of mammalian circadian rhythm by CKIepsilon-regulated proteasomemediated PER2 degradation. Mol Cell Biol 25: 2795–2807. Etchegaray JP, Lee C, Wade PA et al. (2003). Rhythmic histone acetylation underlies transcription in the mammalian circadian clock. Nature 421: 177–182. 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. Feng D, Liu T, Sun Z et al. (2011). A circadian rhythm orchestrated by histone deacetylase 3 controls hepatic lipid metabolism. Science 331: 1315–1319. Folkard S (1975). Diurnal variation in logical reasoning. Br J Psychol 66: 1–8. Fujihara Y, Kondo H, Noguchi T et al. (2014). Glucocorticoids mediate circadian timing in peripheral osteoclasts resulting in the circadian expression rhythm of osteoclast-related genes. Bone 61: 1–9.

Fuller PM, Lu J, Saper CB (2008). Differential rescue of light- and food-entrainable circadian rhythms. Science 320: 1074–1077. Gekakis N, Staknis D, Nguyen HB et al. (1998). Role of the CLOCK protein in the mammalian circadian mechanism. Science 280: 1564–1569. Godinho SI, Maywood ES, Shaw L et al. (2007). The afterhours mutant reveals a role for Fbxl3 in determining mammalian circadian period. Science 316: 897–900. Golombek DA, Rosenstein RE (2010). Physiology of circadian entrainment. Physiol Rev 90: 1063–1102. Gooley JJ, Lu J, Chou TC et al. (2001). Melanopsin in cells of origin of the retinohypothalamic tract. Nat Neurosci 4: 1165. Goriki A, Hatanaka F, Myung J et al. (2014). A novel protein, CHRONO, functions as a core component of the mammalian circadian clock. PLoS Biol 12: e1001839. Goz D, Studholme K, Lappi DA et al. (2008). Targeted destruction of photosensitive retinal ganglion cells with a saporin conjugate alters the effects of light on mouse circadian rhythms. PLoS One 3: e3153. Green RM, Tingay S, Wang ZY et al. (2002). Circadian rhythms confer a higher level of fitness to Arabidopsis plants. Plant Physiol 129: 576–584. Griffin Jr EA, Staknis D, Weitz CJ (1999). Light-independent role of CRY1 and CRY2 in the mammalian circadian clock. Science 286: 768–771. Guzman-Ruiz MA, Ramirez-Corona A, Guerrero-Vargas NN et al. (2015). Role of the suprachiasmatic and arcuate nuclei in diurnal temperature regulation in the rat. J Neurosci 35: 15419–15429. Hannibal J, Hindersson P, Ostergaard J et al. (2004). Melanopsin is expressed in PACAP-containing retinal ganglion cells of the human retinohypothalamic tract. Invest Ophthalmol Vis Sci 45: 4202–4209. Harada Y, Sakai M, Kurabayashi N et al. (2005). Ser-557phosphorylated mCRY2 is degraded upon synergistic phosphorylation by glycogen synthase kinase-3 beta. J Biol Chem 280: 31714–31721. Hatori M, Le H, Vollmers C et al. (2008). Inducible ablation of melanopsin-expressing retinal ganglion cells reveals their central role in non-image forming visual responses. PLoS One 3: e2451. Hattar S, Liao HW, Takao M et al. (2002). Melanopsincontaining retinal ganglion cells: architecture, projections, and intrinsic photosensitivity. Science 295: 1065–1070. Hattar S, Lucas RJ, Mrosovsky N et al. (2003). Melanopsin and rod-cone photoreceptive systems account for all major accessory visual functions in mice. Nature 424: 76–81. Herzog ED, Takahashi JS, Block GD (1998). Clock controls circadian period in isolated suprachiasmatic nucleus neurons. Nat Neurosci 1: 708–713. Herzog ED, Aton SJ, Numano R et al. (2004). Temporal precision in the mammalian circadian system: a reliable clock from less reliable neurons. J Biol Rhythms 19: 35–46. 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.

DIURNAL AND SEASONAL MOLECULAR RHYTHMS IN THE HUMAN BRAIN Hofman MA, Swaab DF (1995). Influence of aging on the seasonal rhythm of the vasopressin-expressing neurons in the human suprachiasmatic nucleus. Neurobiol Aging 16: 965–971. Hogenesch JB, Gu YZ, Jain S et al. (1998). The basic-helixloop-helix-PAS orphan MOP3 forms transcriptionally active complexes with circadian and hypoxia factors. Proc Natl Acad Sci U S A 95: 5474–5479. Honma S, Kawamoto T, Takagi Y et al. (2002). Dec1 and Dec2 are regulators of the mammalian molecular clock. Nature 419: 841–844. Hor CN, Yeung J, Jan M et al. (2019). Sleep-wake-driven and circadian contributions to daily rhythms in gene expression and chromatin accessibility in the murine cortex. Proc Natl Acad Sci U S A 116: 25773–25783. Hughes AT, Guilding C, Lennox L et al. (2008). Live imaging of altered period1 expression in the suprachiasmatic nuclei of Vipr2-/- mice. J Neurochem 106: 1646–1657. Iitaka C, Miyazaki K, Akaike T et al. (2005). A role for glycogen synthase kinase-3beta in the mammalian circadian clock. J Biol Chem 280: 29397–29402. 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. Jasper I, Roenneberg T, Haussler A et al. (2010). Circadian rhythm in force tracking and in dual task costs. Chronobiol Int 27: 653–673. Jin X, Shearman LP, Weaver DR et al. (1999). A molecular mechanism regulating rhythmic output from the suprachiasmatic circadian clock. Cell 96: 57–68. Kalsbeek A, Van Heerikhuize JJ, Wortel J et al. (1996). A diurnal rhythm of stimulatory input to the hypothalamo-pituitaryadrenal system as revealed by timed intrahypothalamic administration of the vasopressin V1 antagonist. J Neurosci 16: 5555–5565. Kalsbeek A, Bruinstroop E, Yi CX et al. (2010). Hypothalamic control of energy metabolism via the autonomic nervous system. Ann N Y Acad Sci 1212: 114–129. Katada S, Sassone-Corsi P (2010). The histone methyltransferase MLL1 permits the oscillation of circadian gene expression. Nat Struct Mol Biol 17: 1414–1421. King DP, Zhao Y, Sangoram AM et al. (1997). Positional cloning of the mouse circadian clock gene. Cell 89: 641–653. Kleitman N (1933). Studies on the physiology of sleep: VIII. Diurnal variation in performance. Am J Physiol 104: 449–556. Ko CH, Yamada YR, Welsh DK et al. (2010). Emergence of noise-induced oscillations in the central circadian pacemaker. PLoS Biol 8: e1000513. Koike N, Yoo SH, Huang HC et al. (2012). Transcriptional architecture and chromatin landscape of the core circadian clock in mammals. Science 338: 349–354. Kojima S, Matsumoto K, Hirose M et al. (2007). LARK activates posttranscriptional expression of an essential mammalian clock protein, PERIOD1. Proc Natl Acad Sci U S A 104: 1859–1864. Kramer A, Yang FC, Snodgrass P et al. (2001). Regulation of daily locomotor activity and sleep by hypothalamic EGF receptor signaling. Science 294: 2511–2515.

281

Kume K, Zylka MJ, Sriram S et al. (1999). mCRY1 and mCRY2 are essential components of the negative limb of the circadian clock feedback loop. Cell 98: 193–205. Lacny C, Kirk A, Morgan DG et al. (2011). Does day length affect cognitive performance in memory clinic patients? Can J Neurol Sci 38: 461–464. Lamia KA, Sachdeva UM, Ditacchio L et al. (2009). AMPK regulates the circadian clock by cryptochrome phosphorylation and degradation. Science 326: 437–440. Lande-Diner L, Boyault C, Kim JY et al. (2013). A positive feedback loop links circadian clock factor CLOCKBMAL1 to the basic transcriptional machinery. Proc Natl Acad Sci U S A 110: 16021–16026. Le Martelot G, Canella D, Symul L et al. (2012). Genomewide RNA polymerase II profiles and RNA accumulation reveal kinetics of transcription and associated epigenetic changes during diurnal cycles. PLoS Biol 10: e1001442. Leak RK, Moore RY (2001). Topographic organization of suprachiasmatic nucleus projection neurons. J Comp Neurol 433: 312–334. Lee C, Etchegaray JP, Cagampang FR et al. (2001). Posttranslational mechanisms regulate the mammalian circadian clock. Cell 107: 855–867. Lee H, Chen R, Lee Y et al. (2009a). Essential roles of CKIdelta and CKIepsilon in the mammalian circadian clock. Proc Natl Acad Sci U S A 106: 21359–21364. Lee JH, Wang W, Silva EJ et al. (2009b). Neurobehavioral performance in young adults living on a 28-h day for 6 weeks. Sleep 32: 905–913. Li DQ, Pakala SB, Reddy SD et al. (2013a). Metastasisassociated protein 1 is an integral component of the circadian molecular machinery. Nat Commun 4: 2545. Li JZ, Bunney BG, Meng F et al. (2013b). Circadian patterns of gene expression in the human brain and disruption in major depressive disorder. Proc Natl Acad Sci U S A 110: 9950–9955. Lim AS, Srivastava GP, Yu L et al. (2014). 24-hour rhythms of DNA methylation and their relation with rhythms of RNA expression in the human dorsolateral prefrontal cortex. PLoS Genet 10: e1004792. Lim AS, Klein HU, Yu L et al. (2017). Diurnal and seasonal molecular rhythms in human neocortex and their relation to Alzheimer’s disease. Nat Commun 8: 14931. Lim ASP, Gaiteri C, Yu L et al. (2018). Seasonal plasticity of cognition and related biological measures in adults with and without Alzheimer disease: analysis of multiple cohorts. PLoS Med 15: e1002647. Liu C, Weaver DR, Strogatz SH et al. (1997). Cellular construction of a circadian clock: period determination in the suprachiasmatic nuclei. Cell 91: 855–860. Liu RY, Zhou JN, Hoogendijk WJ et al. (2000). Decreased vasopressin gene expression in the biological clock of Alzheimer disease patients with and without depression. J Neuropathol Exp Neurol 59: 314–322. Liu AC, Welsh DK, Ko CH et al. (2007a). Intercellular coupling confers robustness against mutations in the SCN circadian clock network. Cell 129: 605–616.

282

A.S.P. LIM

Liu C, Li S, Liu T et al. (2007b). Transcriptional coactivator PGC-1alpha integrates the mammalian clock and energy metabolism. Nature 447: 477–481. Liu AC, Tran HG, Zhang EE et al. (2008). Redundant function of REV-ERBalpha and beta and non-essential role for Bmal1 cycling in transcriptional regulation of intracellular circadian rhythms. PLoS Genet 4: e1000023. Lu J, Zhang YH, Chou TC et al. (2001). Contrasting effects of ibotenate lesions of the paraventricular nucleus and subparaventricular zone on sleep-wake cycle and temperature regulation. J Neurosci 21: 4864–4874. Maekawa F, Shimba S, Takumi S et al. (2012). Diurnal expression of Dnmt3b mRNA in mouse liver is regulated by feeding and hepatic clockwork. Epigenetics 7: 1046–1056. Maier B, Wendt S, Vanselow JT et al. (2009). A large-scale functional RNAi screen reveals a role for CK2 in the mammalian circadian clock. Genes Dev 23: 708–718. Maret S, Dorsaz S, Gurcel L et al. (2007). Homer1a is a core brain molecular correlate of sleep loss. Proc Natl Acad Sci U S A 104: 20090–20095. Massart R, Freyburger M, Suderman M et al. (2014). The genome-wide landscape of DNA methylation and hydroxymethylation in response to sleep deprivation impacts on synaptic plasticity genes. Transl Psychiatry 4: e347. Matthews RW, Ferguson SA, Zhou X et al. (2012). Simulated driving under the influence of extended wake, time of day and sleep restriction. Accid Anal Prev 45 (Suppl): 55–61. Maywood ES, Reddy AB, Wong GK et al. (2006). Synchronization and maintenance of timekeeping in suprachiasmatic circadian clock cells by neuropeptidergic signaling. Curr Biol 16: 599–605. Meyer C, Muto V, Jaspar M et al. (2016). Seasonality in human cognitive brain responses. Proc Natl Acad Sci U S A 113: 3066–3071. Michel S, Itri J, Han JH et al. (2006). Regulation of glutamatergic signalling by PACAP in the mammalian suprachiasmatic nucleus. BMC Neurosci 7: 15. Miller BH, Mcdearmon EL, Panda S et al. (2007). Circadian and CLOCK-controlled regulation of the mouse transcriptome and cell proliferation. Proc Natl Acad Sci U S A 104: 3342–3347. Moore RY, Eichler VB (1972). Loss of a circadian adrenal corticosterone rhythm following suprachiasmatic lesions in the rat. Brain Res 42: 201–206. Morf J, Rey G, Schneider K et al. (2012). Cold-inducible RNA-binding protein modulates circadian gene expression posttranscriptionally. Science 338: 379–383. Mure LS, Le HD, Benegiamo G et al. (2018). Diurnal transcriptome atlas of a primate across major neural and peripheral tissues. Science 359: eaao0318. Musiek ES, Bhimasani M, Zangrilli MA et al. (2018). Circadian rest-activity pattern changes in aging and preclinical alzheimer disease. JAMA Neurol 75: 582–590. Na J, Lee K, Kim HG et al. (2012). Role of type II protein arginine methyltransferase 5 in the regulation of Circadian Per1 gene. PLoS One 7: e48152. Nakahata Y, Kaluzova M, Grimaldi B et al. (2008). The NAD+dependent deacetylase SIRT1 modulates CLOCK-mediated

chromatin remodeling and circadian control. Cell 134: 329–340. Naruse Y, Oh-Hashi K, Iijima N et al. (2004). Circadian and light-induced transcription of clock gene Per1 depends on histone acetylation and deacetylation. Mol Cell Biol 24: 6278–6287. Noya SB, Colameo D, Bruning F et al. (2019). The forebrain synaptic transcriptome is organized by clocks but its proteome is driven by sleep. Science 366: eaav2642. Oh G, Ebrahimi S, Carlucci M et al. (2018). Cytosine modifications exhibit circadian oscillations that are involved in epigenetic diversity and aging. Nat Commun 9: 644. Oster H, Damerow S, Hut RA et al. (2006). Transcriptional profiling in the adrenal gland reveals circadian regulation of hormone biosynthesis genes and nucleosome assembly genes. J Biol Rhythms 21: 350–361. Ouyang Y, Andersson CR, Kondo T et al. (1998). Resonating circadian clocks enhance fitness in cyanobacteria. Proc Natl Acad Sci U S A 95: 8660–8664. Panda S, Antoch MP, Miller BH et al. (2002). Coordinated transcription of key pathways in the mouse by the circadian clock. Cell 109: 307–320. Panda S, Provencio I, Tu DC et al. (2003). Melanopsin is required for non-image-forming photic responses in blind mice. Science 301: 525–527. Panda S, Nayak SK, Campo B et al. (2005). Illumination of the melanopsin signaling pathway. Science 307: 600–604. Pomplun M, Silva EJ, Ronda JM et al. (2012). The effects of circadian phase, time awake, and imposed sleep restriction on performing complex visual tasks: evidence from comparative visual search. J Vis 12: 14. Preitner N, Damiola F, Lopez-Molina L et al. (2002). The orphan nuclear receptor REV-ERBalpha controls circadian transcription within the positive limb of the mammalian circadian oscillator. Cell 110: 251–260. Preussner M, Wilhelmi I, Schultz AS et al. (2014). Rhythmic U2af26 alternative splicing controls PERIOD1 stability and the circadian clock in mice. Mol Cell 54: 651–662. Ptitsyn AA, Zvonic S, Conrad SA et al. (2006). Circadian clocks are resounding in peripheral tissues. PLoS Comput Biol 2: e16. Rajajarvi E, Antila M, Kieseppa T et al. (2010). The effect of seasons and seasonal variation on neuropsychological test performance in patients with bipolar I disorder and their first-degree relatives. J Affect Disord 127: 58–65. Ralph MR, Foster RG, Davis FC et al. (1990). Transplanted suprachiasmatic nucleus determines circadian period. Science 247: 975–978. Reddy AB, Maywood ES, Karp NA et al. (2007). Glucocorticoid signaling synchronizes the liver circadian transcriptome. Hepatology 45: 1478–1488. Reick M, Garcia JA, Dudley C et al. (2001). NPAS2: an analog of clock operative in the mammalian forebrain. Science 293: 506–509. Ripperger JA, Schibler U (2006). Rhythmic CLOCK-BMAL1 binding to multiple E-box motifs drives circadian Dbp transcription and chromatin transitions. Nat Genet 38: 369–374.

DIURNAL AND SEASONAL MOLECULAR RHYTHMS IN THE HUMAN BRAIN Robles MS, Boyault C, Knutti D et al. (2010). Identification of RACK1 and protein kinase Calpha as integral components of the mammalian circadian clock. Science 327: 463–466. Rogers-Soeder TS, Blackwell T, Yaffe K et al. (2018). Restactivity rhythms and cognitive decline in older men: the osteoporotic fractures in men sleep study. J Am Geriatr Soc 66: 2136–2143. Rosenberg J, Pusch K, Dietrich R et al. (2009). The tick-tock of language: is language processing sensitive to circadian rhythmicity and elevated sleep pressure? Chronobiol Int 26: 974–991. Sagaspe P, Taillard J, Amieva H et al. (2012). Influence of age, circadian and homeostatic processes on inhibitory motor control: a Go/Nogo task study. PLoS One 7: e39410. Sahar S, Zocchi L, Kinoshita C et al. (2010). Regulation of BMAL1 protein stability and circadian function by GSK3beta-mediated phosphorylation. PLoS One 5: e8561. Saini C, Morf J, Stratmann M et al. (2012). Simulated body temperature rhythms reveal the phase-shifting behavior and plasticity of mammalian circadian oscillators. Genes Dev 26: 567–580. Saper CB, Scammell TE, Lu J (2005). Hypothalamic regulation of sleep and circadian rhythms. Nature 437: 1257–1263. Sato TK, Panda S, Miraglia LJ et al. (2004). A functional genomics strategy reveals Rora as a component of the mammalian circadian clock. Neuron 43: 527–537. Sato TK, Yamada RG, Ukai H et al. (2006). Feedback repression is required for mammalian circadian clock function. Nat Genet 38: 312–319. Schwartz C, Hampton M, Andrews MT (2013). Seasonal and regional differences in gene expression in the brain of a hibernating mammal. PLoS One 8: e58427. Seney ML, Cahill K, Enwright 3rd JF et al. (2019). Diurnal rhythms in gene expression in the prefrontal cortex in schizophrenia. Nat Commun 10: 3355. Shearman LP, Zylka MJ, Weaver DR et al. (1997). Two period homologs: circadian expression and photic regulation in the suprachiasmatic nuclei. Neuron 19: 1261–1269. Shearman LP, Sriram S, Weaver DR et al. (2000). Interacting molecular loops in the mammalian circadian clock. Science 288: 1013–1019. Shi G, Xing L, Liu Z et al. (2013). Dual roles of FBXL3 in the mammalian circadian feedback loops are important for period determination and robustness of the clock. Proc Natl Acad Sci U S A 110: 4750–4755. Shirogane T, Jin J, Ang XL et al. (2005). SCFbeta-TRCP controls clock-dependent transcription via casein kinase 1-dependent degradation of the mammalian period-1 (Per1) protein. J Biol Chem 280: 26863–26872. Siepka SM, Yoo SH, Park J et al. (2007). Circadian mutant Overtime reveals F-box protein FBXL3 regulation of cryptochrome and period gene expression. Cell 129: 1011–1023. Silva EJ, Wang W, Ronda JM et al. (2010). Circadian and wake-dependent influences on subjective sleepiness, cognitive throughput, and reaction time performance in older and young adults. Sleep 33: 481–490.

283

Spengler ML, Kuropatwinski KK, Schumer M et al. (2009). A serine cluster mediates BMAL1-dependent CLOCK phosphorylation and degradation. Cell Cycle 8: 4138–4146. Stephan FK, Zucker I (1972). Circadian rhythms in drinking behavior and locomotor activity of rats are eliminated by hypothalamic lesions. Proc Natl Acad Sci U S A 69: 1583–1586. Stevenson TJ (2017). Circannual and circadian rhythms of hypothalamic DNA methyltransferase and histone deacetylase expression in male Siberian hamsters (Phodopus sungorus). Gen Comp Endocrinol 243: 130–137. Stevenson TJ, Prendergast BJ (2013). Reversible DNA methylation regulates seasonal photoperiodic time measurement. Proc Natl Acad Sci U S A 110: 16651–16656. Storch KF, Lipan O, Leykin I et al. (2002). Extensive and divergent circadian gene expression in liver and heart. Nature 417: 78–83. Storch KF, Paz C, Signorovitch J et al. (2007). Physiological importance of a circadian clock outside the suprachiasmatic nucleus. Cold Spring Harb Symp Quant Biol 72: 307–318. Suh SW, Han JW, Lee JR et al. (2018). Sleep and cognitive decline: a prospective nondemented elderly cohort study. Ann Neurol 83: 472–482. Sujino M, Masumoto KH, Yamaguchi S et al. (2003). Suprachiasmatic nucleus grafts restore circadian behavioral rhythms of genetically arrhythmic mice. Curr Biol 13: 664–668. Swanson LW, Cowan WM (1975). The efferent connections of the suprachiasmatic nucleus of the hypothalamus. J Comp Neurol 160: 1–12. Takahata S, Ozaki T, Mimura J et al. (2000). Transactivation mechanisms of mouse clock transcription factors, mClock and mArnt3. Genes Cells 5: 739–747. Tamaru T, Isojima Y, Van Der Horst GT et al. (2003). Nucleocytoplasmic shuttling and phosphorylation of BMAL1 are regulated by circadian clock in cultured fibroblasts. Genes Cells 8: 973–983. Tranah GJ, Blackwell T, Stone KL et al. (2011). Circadian activity rhythms and risk of incident dementia and MCI in older women. Ann Neurol 70: 722–732. Ueda HR, Chen W, Adachi A et al. (2002). A transcription factor response element for gene expression during circadian night. Nature 418: 534–539. Ueyama T, Krout KE, Nguyen XV et al. (1999). Suprachiasmatic nucleus: a central autonomic clock. Nat Neurosci 2: 1051–1053. Vanselow K, Vanselow JT, Westermark PO et al. (2006). Differential effects of PER2 phosphorylation: molecular basis for the human familial advanced sleep phase syndrome (FASPS). Genes Dev 20: 2660–2672. Vitaterna MH, Selby CP, Todo T et al. (1999). Differential regulation of mammalian period genes and circadian rhythmicity by cryptochromes 1 and 2. Proc Natl Acad Sci U S A 96: 12114–12119. Vollmers C, Schmitz RJ, Nathanson J et al. (2012). Circadian oscillations of protein-coding and regulatory RNAs in a highly dynamic mammalian liver epigenome. Cell Metab 16: 833–845.

284

A.S.P. LIM

Vujovic N, Davidson AJ, Menaker M (2008). Sympathetic input modulates, but does not determine, phase of peripheral circadian oscillators. Am J Physiol Regul Integr Comp Physiol 295: R355–R360. Walsh CM, Blackwell T, Tranah GJ et al. (2014). Weaker circadian activity rhythms are associated with poorer executive function in older women. Sleep 37: 2009–2016. Welsh DK, Logothetis DE, Meister M et al. (1995). Individual neurons dissociated from rat suprachiasmatic nucleus express independently phased circadian firing rhythms. Neuron 14: 697–706. Witting W, Kwa IH, Eikelenboom P et al. (1990). Alterations in the circadian rest-activity rhythm in aging and Alzheimer’s disease. Biol Psychiatry 27: 563–572. Woo KC, Kim TD, Lee KH et al. (2009). Mouse period 2 mRNA circadian oscillation is modulated by PTB-mediated rhythmic mRNA degradation. Nucleic Acids Res 37: 26–37. Woo KC, Ha DC, Lee KH et al. (2010). Circadian amplitude of cryptochrome 1 is modulated by mRNA stability regulation via cytoplasmic hnRNP D oscillation. Mol Cell Biol 30: 197–205. Wyatt JK, Ritz-De Cecco A, Czeisler CA et al. (1999). Circadian temperature and melatonin rhythms, sleep, and neurobehavioral function in humans living on a 20-h day. Am J Physiol 277: R1152–R1163. Xu Y, Padiath QS, Shapiro RE et al. (2005). Functional consequences of a CKIdelta mutation causing familial advanced sleep phase syndrome. Nature 434: 640–644. Yamaguchi S, Isejima H, Matsuo T et al. (2003). Synchronization of cellular clocks in the suprachiasmatic nucleus. Science 302: 1408–1412. Yamamoto T, Nakahata Y, Soma H et al. (2004). Transcriptional oscillation of canonical clock genes in mouse peripheral tissues. BMC Mol Biol 5: 18. Yamamoto T, Nakahata Y, Tanaka M et al. (2005). Acute physical stress elevates mouse period1 mRNA expression in mouse peripheral tissues via a glucocorticoid-responsive element. J Biol Chem 280: 42036–42043. Yamazaki S, Numano R, Abe M et al. (2000). Resetting central and peripheral circadian oscillators in transgenic rats. Science 288: 682–685. Yan J, Wang H, Liu Y et al. (2008). Analysis of gene regulatory networks in the mammalian circadian rhythm. PLoS Comput Biol 4: e1000193.

Yang S, Wang K, Valladares O et al. (2007). Genome-wide expression profiling and bioinformatics analysis of diurnally regulated genes in the mouse prefrontal cortex. Genome Biol 8: R247. Yin L, Lazar MA (2005). The orphan nuclear receptor Rev-erbalpha recruits the N-CoR/histone deacetylase 3 corepressor to regulate the circadian Bmal1 gene. Mol Endocrinol 19: 1452–1459. Yin L, Wang J, Klein PS et al. (2006). Nuclear receptor Reverbalpha is a critical lithium-sensitive component of the circadian clock. Science 311: 1002–1005. Yin L, Wu N, Curtin JC et al. (2007). Rev-erbalpha, a heme sensor that coordinates metabolic and circadian pathways. Science 318: 1786–1789. Yoo SH, Yamazaki S, Lowrey PL et al. (2004). PERIOD2:: LUCIFERASE real-time reporting of circadian dynamics reveals persistent circadian oscillations in mouse peripheral tissues. Proc Natl Acad Sci U S A 101: 5339–5346. Zhang X, Odom DT, Koo SH et al. (2005). Genome-wide analysis of cAMP-response element binding protein occupancy, phosphorylation, and target gene activation in human tissues. Proc Natl Acad Sci U S A 102: 4459–4464. Zhang R, Lahens NF, Ballance HI et al. (2014). A circadian gene expression atlas in mammals: implications for biology and medicine. Proc Natl Acad Sci U S A 111: 16219–16224. Zheng B, Larkin DW, Albrecht U et al. (1999). The mPer2 gene encodes a functional component of the mammalian circadian clock. Nature 400: 169–173. Zheng B, Albrecht U, Kaasik K et al. (2001). Nonredundant roles of the mPer1 and mPer2 genes in the mammalian circadian clock. Cell 105: 683–694. Zhou JN, Riemersma RF, Unmehopa UA et al. (2001). Alterations in arginine vasopressin neurons in the suprachiasmatic nucleus in depression. Arch Gen Psychiatry 58: 655–662. Zuber AM, Centeno G, Pradervand S et al. (2009). Molecular clock is involved in predictive circadian adjustment of renal function. Proc Natl Acad Sci U S A 106: 16523–16528. Zvonic S, Ptitsyn AA, Conrad SA et al. (2006). Characterization of peripheral circadian clocks in adipose tissues. Diabetes 55: 962–970.

Handbook of Clinical Neurology, Vol. 179 (3rd series) The Human Hypothalamus: Anterior Region D.F. Swaab, F. Kreier, P.J. Lucassen, A. Salehi, and R.M. Buijs, Editors https://doi.org/10.1016/B978-0-12-819975-6.00018-2 Copyright © 2021 Elsevier B.V. All rights reserved

Chapter 19

Circadian changes in Alzheimer’s disease: Neurobiology, clinical problems, and therapeutic opportunities KARLO TOLJAN1,∗ AND JAN HOMOLAK2 1

Department of Neurology, Neurological Institute, Cleveland Clinic, Cleveland, OH, United States

2

Department of Pharmacology, and Croatian Institute for Brain Research, University of Zagreb School of Medicine, Zagreb, Croatia

Abstract The understanding of Alzheimer’s disease (AD) pathophysiology is an active area of research, and the traditional focus on hippocampus, amyloid and tau protein, and memory impairment has been expanded with components like neuroinflammation, insulin resistance, and circadian rhythm alterations. The bidirectional vicious cycle of neuroinflammation and neurodegeneration on a molecular level may cause functional deficits already long before the appearance of overt clinical symptoms. Located at the crossroads of metabolic, circadian, and hormonal signaling, the hypothalamus has been identified as another brain region affected by AD pathophysiology. Current findings on hypothalamic dysfunction open a broader horizon for studying AD pathogenesis and offer new opportunities for diagnosis and therapy. While treatments with cholinomimetics and memantine form a first line of pharmacological treatment, additional innovative research is pursued toward the development of antiinflammatory, growth factor, or antidiabetic types of medication. Following recent epidemiological data showing associations of AD incidence with modern societal and “life-style”-related risk factors, also nonpharmacological interventions, including sleep optimization, are being developed and some have been shown to be beneficial. Circadian aspects in AD are relevant from a pathophysiological standpoint, but they can also have an important role in pharmacologic and nonpharmacologic interventions, and appropriate timing of sleep, meals, and medication may boost therapeutic efficacy.

INTRODUCTION The calculated cumulative incidence of Alzheimer’s disease (AD) starting at age 45 is around 20% for women and 10% for men (Ch^ene et al., 2015). Based on anticipated increases in the absolute global population of people older than 65 years, estimates indicate that relative to 1995, the incidence of AD will have doubled by 2050, with an expected total of around 1 million cases (Hebert et al., 2001). Besides memory impairment, other aspects of AD have also gained attention. Circadian changes in AD have considerable pathophysiological and clinical

∗

relevance. Neuropsychiatric symptoms in AD do not exclusively arise from hippocampal dysfunction, but other brain regions as well. Hypothalamus is one of the key regions of interest for the study of circadian changes in AD, as it contains the biological clock of the brain, i.e., the suprachiasmatic nucleus (SCN), and orchestrates crucial interactions between the nervous and endocrine system in a rhythmic pattern. With an increasing focus on hypothalamic aspects of AD, also broader implications of the disease become recognized.

Correspondence to: Karlo Toljan, Department of Neurology, Neurological Institute, Cleveland Clinic, Cleveland, OH 44195, United States. Tel: +1-216-444-2200, Fax: +1-216-445-9754, E-mail: [email protected]

286

K. TOLJAN AND J. HOMOLAK

Focusing on those aspects may expand our pathophysiological understanding of AD symptoms and thereby help to uncover novel therapeutic targets and approaches.

PATHOGENESIS OF AD: BEATEN TRACKS AND NOVEL EMERGING CONCEPTS Although a vast amount of resources have been directed toward research into AD, its pathogenesis remains largely unknown. There are at least two types of AD, i.e., the familial form (fAD), accounting for 1%–5% of all cases, and the sporadic form (sAD), responsible for the remaining 95%–99% of the patients (Homolak et al., 2018). The familial form of the disease is mainly passed down through autosomal dominant genes involved in processing amyloid b (PSEN1, PSEN2, and APP), and the symptoms usually present around the third or fourth life decade. In contrast, sAD usually starts after the sixth decade and is therefore commonly referred to as the late onset AD. Significant efforts were also put into understanding the genetic background of the sporadic form of the disease with results suggestive of the importance of heritability (Gatz et al., 2006; Barber, 2012; Van Cauwenberghe et al., 2016). Nevertheless, the pathophysiological course of the disease is defined by complex interactions of multiple genetic and environmental components that are still not well understood. In semantic terms, even though fAD and sAD are two forms of the same disease, it might be better to classify fAD as a disease and sAD as a syndrome; “a recognizable complex of symptoms and physical findings which indicate a specific condition for which a direct cause is not necessarily understood” (Calvo et al., 2003). The distinction is crucial in understanding AD etiopathogenesis as most of the research over the years focused on sAD and fAD interchangeably. This is probably most evident in animal research where genetic manipulation is by far the most common method for studying sAD etiopathogenesis even though it is more appropriate for understanding fAD. Cautious interpretation of the mentioned animal research is necessary when it is used to explain related human pathophysiological mechanisms. Although a complex pathophysiological interplay of factors likely orchestrates the neurodegeneration in sAD, preclinical studies have so far focused on a rather narrow “amyloidocentric” approach. We hypothesize that these conservative amyloidocentric and “hippocampocentric” approaches have led us astray from a better understanding of the etiopathogenetic mechanisms present in AD. So far, classic models of the disease remain prevalent in preclinical literature despite the fact that novel therapeutics remain largely ineffective (Karran et al., 2011).

Recent findings have proposed some overlooked molecular and anatomical patterns that appear related to hypothalamic nuclei at least in pathophysiological terms. In this chapter, we will therefore focus on the role of hypothalamus in AD and discuss alterations from a molecular and clinical perspective and in the context of new diagnostic and therapeutic opportunities (Fig. 19.1).

HYPOTHALAMUS AT THE CROSSROADS OF EARLY CIRCADIAN AND METABOLIC DISRUPTIONS IN AD: EVIDENCE FROM ANIMAL MODELS Most of the research on AD, both in humans and animal models, is focused on understanding neuropathologic changes in two brain areas—cerebral cortex and hippocampus, the latter being an especially popular area of interest in animal models of AD due to its well established role in memory and cognition. However, studies suggests that focusing on hippocampus-driven research only may hinder the study of other etiopathogenetic aspects of AD, and changes in other brain areas may, e.g., occur earlier or show a greater correlation with clinical progression (Grinberg et al., 2009; Simic et al., 2009). In this context, the hypothalamus is particularly interesting as both human and animal studies have reported hypothalamic areas to undergo neurodegenerative changes in AD (Ishii and Iadecola, 2015; Zheng et al., 2018). A unique function of the hypothalamus relates to the physiological integration of metabolism, sleep, reproduction, and autonomic homeostasis (Swaab, 1997). This may provide at least some of the explanations for sleep disorders, insulin-resistant brain state, metabolic syndrome, and hormonal abnormalities, which are all dysfunctions reported in AD patients (Neth and Craft, 2017; Brzecka et al., 2018; Zhang et al., 2019). Hypothalamic atrophy was noted in early clinical stages of the disease, and pathognomonic neuropathologic changes such as plaques and tangles have been described in hypothalamic nuclei of the deceased patients (Swaab et al., 1992; Ishii and Iadecola, 2015). Interestingly, hypothalamic deposits were described even in the well-known paper on neuropathologic AD staging by Braak and Braak, with the most pronounced changes occurring in the late stages of the disease (Braak and Braak, 1991). Animal studies support hypothalamic involvement in AD pathogenesis. Some findings suggest that hypothalamic changes occur prior to the appearance of neuropathological findings and cognitive dysfunction, both in transgenic and nontransgenic models of the disease. In the study by Zheng et al. (2018), NMR-based

CIRCADIAN CHANGES IN ALZHEIMER'S DISEASE

287

Fig. 19.1. A schematic representation of therapeutic opportunities targeting hypothalamic dysfunction in Alzheimer’s disease. (A) Restoring circadian rhythmicity by reducing the burden of environmental circadian disruptors, improving sleep hygiene, and optimizing endogenous rhythms by melatonin and modulation of pharmacotherapy. (B) Reducing systemic and neuroinflammation by targeting etiologic factors driving systemic low-grade inflammation. (C) Increasing neuronal resilience by stimulation of neuronal growth factor signaling. (D) Improving vascular health by behavioral and pharmacological targeting of hypertension, dyslipidemia, and insulin resistance. (E) Improving metabolic homeostasis in the brain and in peripheral organs by pharmacological and nonpharmacological interventions.

metabolomics was used to examine a metabolic fingerprint at different stages of cognitive decline and in different brain regions of APP/PS1 transgenic mice. Considering metabolic networks as a reflection of cellular function and cellular response to noxious signals, it has been suggested that a robust spatiotemporal metabolomics dissection might provide indispensable information to identify early drivers of AD pathophysiology (Mapstone et al., 2014; Toledo et al., 2017; Varma et al., 2018; Low et al., 2019), since recent data shows links

between metabolic dysfunction and neurodegeneration (Barilar et al., 2020). While taking into account genetic background  age interaction effects, metabolomics analysis of the mouse cortex, cerebellum, hippocampus, hypothalamus, midbrain, and striatum at 1, 5, and 10 months of age revealed that the hypothalamus is a brain region undergoing the most pronounced metabolic perturbation in the process of neurodegeneration (Zheng et al., 2018). Metabolic profiling detected greatest discrepancies in the

288

K. TOLJAN AND J. HOMOLAK

hypothalamus at 5 months of age, with metabolic changes being suggestive of hypermetabolism, while cognitive function remained unchanged until the age of 10 months, a period characterized by normalization of hypothalamic hypermetabolism. The exact molecular process responsible for the observed metabolic perturbations in hypothalamus, as well as mechanism of apparent normalization remain to be elucidated; however, loss of hypothalamic functional capacity to cope with the allostatic load, and subsequent pathophysiological processes driving neurodegeneration and cognitive dysfunction, provides one compelling explanation that should be further explored. Human studies showed similar changes in nucleus basalis of Meynert (Dubelaar et al., 2006) and prefrontal cortex (Bossers et al., 2010), suggestive of hypermetabolism being a more general phenomenon related to AD pathogenesis. Findings from nontransgenic animal models of AD also support an early involvement of the hypothalamus in the development of cognitive and noncognitive pathological changes. In one of the most commonly used nontransgenic rat models of AD, which is based on the development of an insulin-resistant brain state following intraventricular administration of diabetogenic toxin streptozotocin (STZ), the hypothalamus is one of the brain regions showing the most pronounced degeneration early after the induction procedure. The current concept of STZ toxicity is explained by its selective uptake by the low-affinity glucose transporter 2 (GLUT2) that is expressed abundantly in rodent insulin-producing pancreatic beta cells, liver, and kidney and thereby enables the relatively selective damage in this rodent model of diabetes. In the brain, the GLUT2 is found to be expressed in the hypothalamus and in the circumventricular organs, where the transporter is involved in the signaling mechanisms closely related to nutrient sensing (Grieb, 2016). An early involvement of the hypothalamus in the neurodegeneration occurring after intracerebroventricular STZ (STZ-icv) administration has been reported by several groups. Oliveira Santos et al. reported pronounced hypothalamic Fluoro-Jade C signal 24 h after STZ administration (Santos et al., 2012), and Knezovic et al. (2017) have shown effects of STZ administration can be observed in the ependymal lining of the third ventricle as early as 1 h after the administration. In the same study by Santos et al. (2012), but in a separate cohort of rats used for evaluation of long-term effects at the 30-day time point, hypothalamic expression of Ab demonstrated a 86.6% increase relative to controls, while changes in the hippocampal and cortical regions were less pronounced and failed to reach statistical significance. Interestingly, on the same time point, 1 month after STZ administration, metabolic changes assessed by means of 18FDG-PET

suggest that the hypothalamus is the most affected region of the brain and indicated the involvement of glucose hypometabolism (Knezovic et al., 2018). Others have observed pronounced changes in the hypothalamus and also in adjacent regions of the brain. For example, Shoham et al. reported an enlargement of the third ventricle by 100%–150% accompanied by loss of ependymal cells and damage to hypothalamic periventricular myelin in STZ-treated rats (Shoham et al., 2003), an intriguing finding considering the importance of myelinated axons adjacent to the affected region. Finally, hypothalamic involvement in the STZ-icv model, as well as in other animal models exploiting intracerebroventricular administration of different toxins (e.g., intracerebroventricular Ab; Kim et al., 2016) should be considered in the context of methodological background, as the process of ventricular cannulation itself might contribute to the development of neuroinflammation and insults in surrounding tissue could act as a “second hit.” The latter is especially interesting in the context of rapid administration procedures where significant ventricular distension is possible as rapid increment of ventricular volume has been proposed as one of the alternative etiologic models of AD mediated by axonal stretch accompanied by a separation of trans-synaptic proteins (Schiel, 2018).

FUNCTIONAL CONSEQUENCES OF AN EARLY HYPOTHALAMIC INVOLVEMENT IN AD PATHOGENESIS Functional consequences of the involvement of hypothalamus in the process of neurodegeneration are well documented in the literature; however, they are often described separately and are rarely reviewed in the context of the pathophysiological role of hypothalamus in AD. Among these, circadian rhythmicity and sleep and metabolic misalignment stand out both in AD patients and animal models of the disease.

Circadian rhythmicity and sleep Dysregulation of sleep and wakefulness patterns is present in various rodent models of AD. In Tg2576, one of the most widely used mouse models of AD with accumulation of Ab and progressive age-dependent cognitive deterioration, a number of sleep and circadian abnormalities have been reported. The extensive study by Wisor et al. (2005) has shown that Tg2576 display altered patterns of wheel running rhythms and higher encephalographic frequencies during nonrapid eye movement (NREM) sleep. Furthermore, transgenic animals failed to show increased encephalographic delta waves (1–4 Hz) during NREM sleep following sleep deprivation, and the wake-promoting effect of donepezil was less effective

CIRCADIAN CHANGES IN ALZHEIMER'S DISEASE when compared to the one observed in the control group (Wisor et al., 2005). Interestingly, circadian disruption was present in transgenic mice at all ages studied, with the first test trial being conducted at the age of 5 months. For contextual purposes, cognitive deficit, accumulation of amyloid plaques, and increased microglial density in Tg2576 usually occur around the age of 12 months (Alzforum Research Models Repository: Tg2576, n.d.). The exact cause for circadian misalignment in these transgenic animals is still speculative; however, degeneration of basal forebrain cholinergic nuclei and dysfunction of SCN may provide possible explanations for the observed effects (Wisor et al., 2005; Roy et al., 2019). Circadian dysfunction has also been reported in nontransgenic models of AD. In the STZ-icv rat model of sAD, a significant increase in wakefulness, as well as decrease in both NREM and REM sleep, has been reported 2 weeks after the induction procedure (Cui et al., 2018). Further analyses revealed that the observed findings were paralleled by a reduced GABAergic tone in the ventrolateral preoptic nucleus and in the parabrachial nucleus involved in the functional maintenance of the waking state (Cui et al., 2018). A dysfunctional circadian rhythm is a well-known feature of AD (Wu and Swaab, 2007; Ju et al., 2014; Homolak et al., 2018). In contrast to the previous perception of sleep disturbances as consequences of the disease process, accumulating findings support an early involvement of circadian dysfunction in AD pathogenesis (Hahn et al., 2014) with a bidirectional link between circadian dysrhythmia and disease progression and a tendency to enter a pathophysiological positive feedback loop (Ju et al., 2014; Homolak et al., 2018). The most common sleep problems in AD patients are frequent daytime napping, difficulty falling asleep, nocturnal sleep fragmentation, and early awakening (Musiek et al., 2015), often accompanied by loss of slow-wave sleep (stage three of non-REM) and REM sleep (Homolak et al., 2018). Electroencephalographic findings support the hypotheses of early non-REM involvement in AD pathogenesis as a decreased density of K-complexes is present in patients when compared to findings from patients diagnosed with MCI or healthy controls (De Gennaro et al., 2017).

Metabolic dysregulation Metabolic dysregulation has been reported in transgenic and in nontransgenic models of AD; however, its involvement in pathogenesis of AD-like phenotype in transgenic models has only recently been examined more thoroughly. An impairment of glucose tolerance, such as increased fasting plasma insulin and diminished response to insulin in the intraperitoneal glucose tolerance

289

test, has been reported in APP/PS1 transgenic mice prior to accumulation of Ab or development of cognitive deficits, with metabolic changes being present already at the second month of age (Macklin et al., 2017). Furthermore, reduced glucose tolerance does not seem to be the only finding of peripheral metabolic dysfunction in transgenic models of AD as pancreatic (Liu et al., 2019a,b), kidney, and liver (González-Domínguez et al., 2015) metabolic profiles of APP/PS1 differ from that of the control mice. Metabolic dysregulation in transgenic models is also evident at the level of central regulation as rodent models of AD exhibit different patterns of feeding behavior, impaired satiation, and hypermetabolism (Adebakin et al., 2012; Knight et al., 2012). Altered peripheral metabolism is also evident in nontransgenic models. For example, Bloch and colleagues described a number of peripheral metabolic changes in Lewis rats after STZ-icv (Bloch et al., 2017). In the first 2 weeks following induction procedure, usually considered as the period of pronounced acute neuroinflammation, STZicv-treated rats lost weight; however, in the subsequent weeks, accelerated weight gain, liver fat accumulation, hypertrophy of pancreatic islets, and elevated blood insulin, adiponectin, and leptin were reported. Interestingly, peripheral glucose levels were within normal reference range, suggesting allostatic load was still relatively compensated during the experiment (Bloch et al., 2017). The pronounced metabolic changes that are not associated with a significant disruption of plasma glucose homeostasis also suggest that peripheral metabolic changes should be more closely investigated in AD patients even when the glucose profile remains inside the reference range. AD is often accompanied by peripheral metabolic dysfunctions (Cai et al., 2012). The exact role of metabolic changes in AD is still a matter of debate; however, literature suggests a causative role as metabolic dysfunction, both cerebral and peripheral, often precedes neurocognitive impairment. Excess body weight, obesity, and metabolic syndrome during middle-age have all been described as risk factors for development of AD (Cai et al., 2012). Type 2 diabetes mellitus (T2DM) is another factor associated with an increased risk for AD as shown, e.g., in the Rotterdam study, where a twofold greater risk to develop the disease was found in diabetic patients when compared to patients without T2DM, and a fourfold increase in the ones using exogenous insulin (Ott et al., 1999). Differences in concentration or signaling pathways of metabolic hormones have been described in AD patients. Insulin signaling is disrupted in AD brains with molecular findings suggestive of an insulin-resistant brain state and now considered an important early pathogenic factor and possible pharmacologic target.

290

K. TOLJAN AND J. HOMOLAK

The importance of metabolic hypothesis of AD and the involvement of insulin in the pathophysiologic cascade is probably best reflected in the term “diabetes mellitus type 3” proposed for AD by de la Monte, due to a number of common pathobiological mechanisms shared by AD and diabetes (Steen et al., 2005). Other metabolic hormones such as leptin, adiponectin, ghrelin, and glucagon-like peptide 1 are also affected in AD, suggesting a general metabolic dysregulation to be part of the pathologic process (Cai et al., 2012).

CLINICAL CHALLENGES: FROM BENCH TO BEDSIDE Basic science models provide insight into certain aspects of AD pathogenesis, but the ultimate step is translation to a clinical setting. Several challenges stand in the way of global standardization of care for the population affected by AD. Also, there is a tremendous socioeconomic burden for patient’s caregivers, especially nonhealthcare-associated caretakers. Diagnostic methods are steadily improving as pathophysiology is more understood, but capacities are rather limited. An improved awareness combined with suitable biomarkers that would allow an earlier recognition would provide a better chance to halt or postpone a detrimental process.

Diagnosis Symptoms and signs of AD are underappreciated. Consequently, the AD spectrum is underdiagnosed. Although in a sample group of individuals aged 45 years or older, many reported subjective cognitive decline, less than half sought professional assessment (Taylor, 2018). Conversely, less than half of US primary care physicians formally assess their older patients’ cognitive status (2020 Alzheimer’s Disease Facts and Figures, 2020). Due to considerable disability and morbidity associated with AD, disease burden is overwhelming and caregiver burnout is not uncommon (2020 Alzheimer’s Disease Facts and Figures, 2020). Developing an effective framework is work in progress as attention is also dedicated to noncognitive symptoms or to comorbidities commonly accompanying AD. An earlier recognition or slowing of pathophysiologic processes should improve the quality of life for patients and caregivers. Despite the development of neuroimaging and CSF biomarkers, trained professionals are required to establish an appropriate diagnosis with careful assessment of history, risk factors, and clinical features. Diagnostic uncertainty may be resolved with CSF Ab1–42 and tau levels, which are decreased and increased in AD, respectively (Niemantsverdriet et al., 2017). Newer CSF

biomarkers, such as SNAP-25 and chromogranin, may even point to pre- and postsynaptic dysfunction (Moya-Alvarado et al., 2016). The ultimate gold standard would be histopathological brain analysis. An exemplary finding is that brain glucose metabolism is pathologically changed for more than a decade before florid clinical symptoms appear (Mosconi et al., 2009). Whereas normal aging is associated with the decrease of glucose uptake mostly in medial frontal areas, individuals who develop AD already have prominently lower glucose utilization in their parietotemporal and posterior cingulate cortex initially (Mosconi, 2013). Besides such central metabolic changes, peripheral metabolic dysfunction may provide valuable diagnostic information, and exclusion of metabolic factors from the current diagnostic criteria has been criticized by some. However, clear diagnostic parameters related to peripheral metabolic changes are still not reported (Cai et al., 2012). Alongside metabolic parameters, recent attention is directed at other symptoms of hypothalamic dysfunction, such as compromise of circadian rhythms reflected in sleep fragmentation, alteration in daily thermodynamics, and hormonal and metabolic dysregulation (Ishii and Iadecola, 2015). An illustrative quality example is a prospective study following a cohort of 214 senior citizens without baseline dementia (Hahn et al., 2014), which found sleep reduction of more than 2 h daily was associated with subsequent development of AD. Although symptoms of depression have been shown as a confounder, patients developing AD had a tendency for sleep disruption as a stronger associated risk factor. Depression overlaps with dementia by symptomatology and pathophysiology, thus pointing to certain probable shared underlying pathogenetic mechanisms and clinical features, which may not be a pure confounding association but rather a coexisting process (Herbert and Lucassen, 2016). This has been demonstrated by a metaanalysis that showed strong association between depression and subsequent risk for AD (Ownby et al., 2006).

Comorbidities As the etiopathogenesis of AD is multifactorial, implicated disease processes have usually already affected other organ systems in a related manner. Indeed, cardiovascular and metabolic diseases are common with AD, and in a larger US random sample, at least a third of patients were reported to have coronary artery disease or heart failure, diabetes, or chronic kidney disease (2020 Alzheimer’s Disease Facts and Figures, 2020). A quarter of them had five or more chronic conditions, which is six times as much as controls without AD or other forms of dementia. Anxiety, depression, and pain syndromes were also noted in almost a third of patients

CIRCADIAN CHANGES IN ALZHEIMER'S DISEASE in a larger UK sample (Nelis et al., 2019). Notably, hypertension was by far the most prevalent comorbidity, also a known AD risk factor, and present in 40% of patients with AD (Nelis et al., 2019). These facts point to an evident need to optimize the general medical condition as a prerequisite for getting most from interventions targeting AD in specific. Regarding AD comorbidities in relation to hypothalamic dysfunctions, also hypothyroidism (Choi et al., 2017), hypogonadism (Tan and Pu, 2003; Brinton, 2004), obesity (Pegueroles et al., 2018), or low weight with hypoleptinemia (Lee, 2011), and sleep fragmentation may reflect a disturbed circadian rhythm (Lim et al., 2013) and have hence been considered as both contributing factors and consequences of AD pathology.

Bidirectional vicious cycle The urban lifestyle with prominent “social jet-lag,” random or overall extended meal timing, significant exposure to artificial light during nighttime, or shift work, leads to mistimed exposure to zeitgebers, which disrupt the physiologic regulation of the central and peripheral circadian clocks (Homolak et al., 2018). Additionally, stress, which is considered to be increasingly prevalent in the modern way of life, disrupts circadian rhythm and amplifies the effects of its disruption on normal functioning of the organism (Koch et al., 2017). Circadian rhythm dysfunction is associated with systemic pathophysiological alterations as reflected in its connection with metabolic syndrome features, (neuro)inflammation, and ultimately neurodegeneration—all recognized risk factors or contributors to AD pathogenesis (Rojas-Gutierrez et al., 2017). Conversely, AD affects hypothalamic nuclei, including the SCN (Ishii and Iadecola, 2015), which leads to impairments in central circadian clock regulation. Prominent clinical features in such cases, i.e., the sleep– wake cycle alterations and sleep fragmentation, are accompanied by reduced waste clearing via glymphatic mechanisms (Lucey et al., 2018), which otherwise help to physiologically reduce the amyloid burden. Once unleashed, the bidirectional process of circadian rhythm dysfunction, neuroinflammation, and neurodegeneration may become a pathophysiologic vicious cycle (Homolak et al., 2018). On top of the molecular interconnection, both processes overlap in association with clinical features. Common examples include sleep–wake cycle disruption with changes in sleep quality and quantity (Zhu and Zee, 2012), insulin resistance (Stenvers et al., 2018), obesity, hypertension (Douma and Gumz, 2018), and obstructive sleep apnea (Andrade et al., 2018). Such overlap is also present on a larger scale, and the aforementioned pathologies are major public health

291

problems with an impact exceeding that for the individual (Roenneberg and Merrow, 2016, 2020 Alzheimer’s Disease Facts and Figures, 2020). Nonpharmacological interventions as tools to ameliorate pathogenetic mechanisms and clinical symptoms are also a shared link. Such a complex interplay of factors necessitates a motivating clinical environment with considerable efforts in order to gain most from the therapeutic interventions. Caregiver burnout, insurance coverage, social issues, and ageism stand as possible challenges along the way (2020 Alzheimer’s Disease Facts and Figures, 2020).

Neurodegeneration Accumulation of amyloid and tau are neuropathologic hallmarks of AD, and in vivo experiments have shown that sleep is an essential process for facilitating the clearance of those substrates through glymphatic activity (Xie et al., 2013). Neuroinflammation, accompanying synaptic dysfunction, and neurodegeneration are the pathogenetic drivers of neurocognitive AD features (Moya-Alvarado et al., 2016). By the time the disease process has clinically manifested, alterations had been present for a considerable time. The identification of ongoing neurogenesis in the adult hippocampus has, after extensive discussions in the field (Kempermann et al., 2018; Lucassen et al., 2020a,b), offered some hope that neurorestoration could a viable option (Moreno-Jimenez et al., 2019; Tobin et al., 2019). Also, selective modification of neurogenesis was shown to interfere with, and often rescue (Richetin et al., 2015), cognitive deficits in AD models (Hu et al., 2010; Lazarov and Hollands, 2016; Hollands et al., 2017; Choi et al., 2018). However, neurodegeneration is an irreversible process and any neural repair may result in altered plasticity. The complex ontogenesis of cerebral cortex and neural networks remains a challenge when considering full reversal of AD pathology (Reisberg et al., 1999). Although neuroplasticity offers many theoretical possibilities for neurorehabilitation, involved neurons are highly dependent on metabolic milieu governed by glial cells. As the understanding of the salience of central nervous system (CNS) insulin signaling and peripheral-tocentral immune cell trafficking for AD pathogenesis is expanded, it is clear that the CNS immunometabolic environment is intertwined with the systemic or “peripheral” one. As long as pharmacotherapy for dementia remains symptomatic, without causal targets, addressing medical comorbidities remains a great priority in order to minimize contribution to disease burden. Beyond the common focus on hippocampal and cortical involvement in AD, there are substantial reports of hypothalamus being

292

K. TOLJAN AND J. HOMOLAK

affected by AD-related neurodegeneration, including consequential functional compromise, notably of SCN as the important circadian rhythm hub (Ishii and Iadecola, 2015). Current understanding implies that pharmacologic and nonpharmacologic modalities addressing circadian disruption ought to ameliorate AD-related features. Ideally, such well-timed interventions could slow-down AD progression even in a preclinical phase.

THERAPEUTIC OPPORTUNITIES Currently established and specific pharmacotherapy for AD consists of cholinomimetics rivastigmine, galantamine, donepezil, and an NMDA-antagonist, memantine (2020 Alzheimer’s Disease Facts and Figures, 2020). The rationale behind these medications is to bolster cholinergic neurotransmission and enhance neuroplasticity

in a favorable direction, as both processes are severely impaired in AD. Antidepressants and antipsychotics are used to manage dementia-associated symptoms such as mood changes and agitation. Melatonin as a pharmacologic treatment or in combination with bright light (Riemersma-van der Lek et al., 2008) is an attempt to replace the deficiency of this endogenous hormone, a noted feature of hypothalamic dysfunction associated with AD (Wu and Swaab, 2007), with an aim to hopefully relieve some of the circadian abnormalities which are clinically reflected as evening agitation (sundowning) and sleep fragmentation. Nonpharmacologic modalities such as cognitive stimulation training and supportive psychosocial interventions improve cognitive and behavioral performance and quality of life, though a substantial support system is a prerequisite (Berg-Weger and Stewart, 2017) (Fig. 19.2).

Fig. 19.2. A bidirectional association between neurodegeneration and functional hypothalamic dysfunction reflected by circadian and peripheral metabolic dyshomeostasis. (A) Neurodegeneration of hypothalamic nuclei, especially suprachiasmatic nucleus (SCN), induces circadian dysrhythmia. (B) Circadian misalignment and dysrhythmia disrupts peripheral metabolic homeostasis. (C) Dysfunctional peripheral metabolism potentiates neurodegeneration through disruption of cerebral energy homeostasis, vascular health, and (neuro)inflammation. (D) Circadian dysrhythmia affects cerebral homeostasis by affecting amyloid clearance and tau homeostasis, and by potentiating inflammation and oxidative stress. (E) Metabolic dysregulation affects circadian rhythmicity and sleep through direct hormonal regulation, nutrient availability, autonomic activation, and inflammation. (F) Neurodegeneration of associated or regulatory hypothalamic nuclei affects behavioral activity patterns, autonomic system function and satiety, and affects peripheral metabolism directly through neurovisceral and hormonal homeostatic modulation.

CIRCADIAN CHANGES IN ALZHEIMER'S DISEASE Multiple therapies aimed at AD pathologic substrates, namely Ab and tau protein, have so far failed to reach the necessary clinical trial goals. Considering the AD pathophysiology on a molecular level, ongoing neurodegenerative and neuroinflammatory mechanisms are highly interconnected in an overlapping signaling network. Clinically, this means addressing multiple pathogenetic factors with a single intervention is more likely with a less target-specific treatment than with a highly selective antibody. The underlying pathophysiological and clinical complexity, with multiple cooccurring pathologies, warrants a multimodal approach.

Restoring circadian rhythmicity Targets for restoring circadian rhythmicity may be central or peripheral cellular clocks. For the former, appropriate exposure to light and dark is the key. Daytime should be associated with adequate exposure to properly timed (sun)light, whereas chronological nighttime should be associated with darkness. Environmental disruptors during those phases, such as artificial light and noise, and behavioral habits, such as frequent or prolonged daytime napping and late bedtime, should be minimized. Also, pharmacological agents that alter sleep architecture such as benzodiazepines, other sedative hypnotics, or alcohol should not be used. These steps should aid in keeping a proper sleep hygiene. Physiologically, due to aging-associated homeostatic changes, some central rhythmicity is lost, which is evident as loss of nighttime sleep duration and a greater propensity for afternoon daytime naps (Schmidt et al., 2012). Additionally, the urge for nighttime sleep occurs earlier, the diurnal temperature oscillations are less pronounced, cortisol secretion shows a phase advance and lower amplitude, and melatonin synthesis is decreased (Hood and Amir, 2017). The latter may also be compromised by exposure to (artificial) blue light. Longitudinal research over a period of 10 years has shown a preserved melatonin secretion pattern in middle aged or older men, but not women (Kin et al., 2004). However, an underlying neurodegenerative process is often linked with reductions in melatonin secretion. Since AD pathology is definitely associated with the SCN degeneration (Swaab et al., 1985; Wu and Swaab, 2007), interventions that represent otherwise healthy circadian patterns may improve behavioral symptoms. Light therapy and melatonin supplementation, one timed during daytime, the other during evening, helps prevent sundowning and improves quality of sleep. Exposure to light with an intensity of at least 1000 lx for a few hours in the morning, or during chronological day, and dimmer lights in the evening (200–300 lx) increases daytime wakefulness and reduces daytime napping (Hanford and Figueiro, 2013).

293

In smaller trials with less than 100 participants at a time, melatonin in a 3–10 mg oral dose at bedtime was shown to be effective for increasing sleep duration or even improving scores on cognitive tests, but larger randomized trials have failed to replicate such findings to confirm robust effects (Cardinali et al., 2010). Nevertheless, melatonin should primarily be considered a chronobiotic instead of a hypnotic, and supplementation in AD is reasonable given neurodegenerative changes of SCN and reduced melatonin production in such pathology. Higher therapeutic doses may be needed given the number of SCN neurons expressing MT1 melatonin receptors is reduced in aging and AD (Wu and Swaab, 2007). Entrainers of peripheral clocks, which ultimately affect the central one, are physical activity and meal timing. Morning seems to be the optimal time for exercise in case of AD, and aside from aiding in circadian resynchronization with improvement in sleep quality, it also showed benefits for cognitive performance (Baldacchino et al., 2018). Timing meals less frequently, yet at a similar schedule, can induce food anticipatory behavior with promoted activity, but which then also ultimately presents as a zeitgeber for the central clock (Kent, 2014). These features could be clinically used to enhance daytime activity and provide physiologic entrainment signals when circadian asynchrony is present, with AD being one of those conditions. Standard pharmacologic agents and treatment strategies should also be reexamined in the context of circadian rhythmicity. For example, a positive effect of donepezil seems to be pronounced during the diurnal period; however, prolonged (>24 h) increment of ACh disrupts physiological dip of ACh during the slow-wave sleep, the deepest stage of NREM sleep thought to be important for consolidation of memory traces into neocortical networks. In order to prevent potential potentiation of sleep disorders through ACh modulation pharmacokinetics, type of drug and treatment timing should be carefully considered and optimized individually (Van Erum et al., 2019). Novel pharmacologic agents, e.g., acting as orexin receptor antagonists, are being investigated as effective sleeping aids with less unwanted side effects for which sedative hypnotics are known (Janto et al., 2018). In a randomized trial involving 285 patients with mild-tomoderate AD, the orexin antagonist suvorexant enabled half an hour of extra sleep time as compared to placebo (Herring et al., 2020).

Reducing inflammation and neuroinflammation Chronic low-grade inflammation is a reflection of (immuno)metabolic changes present in chronic conditions

294

K. TOLJAN AND J. HOMOLAK

and AD is not an exception. Besides being associated with metabolic syndrome and related pathophysiology, neuroinflammation is parallelly ongoing in the AD-affected CNS. Glia is key to this process, since glial cells determine the metabolic and immune environment for the neurons, as well as govern the glymphatic clearance. Therefore, therapeutic targets can be broader, in terms of reducing overall systemic inflammatory processes, or narrower in targeting the CNS itself. This division still implies mutual dependence as there is direct and indirect communication between the CNS and the peripheral tissues. Given the available antiinflammatories, NSAIDs and steroids were studied for AD prevention or treatment, without encouraging end results (Aisen et al., 2000; Ali et al., 2019). One of the larger interventions with NSAIDs was the ADAPT trial with more than 2000 participants older than 70 years. After a 10-year follow up from initial intervention (1–3 years of NSAID use), there were no significant benefits for cognitive health as compared to placebo (ADAPT-FS Research Group, 2015). While dietary interventions have to some extent been effective in adult rodent models in rescuing cognitive deficits induced by early life stress (Naninck et al., 2017; Liu et al., 2019a,b; Wang et al., 2019a,b, 2020; Yam et al., 2019), lifestyle interventions combined with polyunsaturated omega-3 acids or vitamin supplements have been studied in smaller human trials, with disappointing results, but warranting firmer conclusions with higher power (Kivipelto et al., 2018). Microglia and astrocytes have both protective and detrimental roles for AD pathogenesis, depending on the level of neuroinflammation present. With advanced vicious cycles, glia contributes to further neurodegeneration as its protective mechanisms have been exhausted (Fakhoury, 2018). In vitro experiments have shown benefits of NSAIDs targeting glial cells in AD models, but clinical trials failed to support their use (Ali et al., 2019). Active areas of basic and clinical research are glial modulators such as minocycline (Metz et al., 2017) (Plane et al., 2010), ibudilast (Fox et al., 2018), and naltrexone (Toljan and Vrooman, 2018). They are clinically studied in neurologic conditions such as multiple sclerosis and Parkinson’s disease as neuroinflammation attenuators; however, they still not clinically for AD spectrum. In vitro and animal studies provide a mechanistic rationale for benefits in a clinical scenario, and it was demonstrated that those compounds act as Toll-like receptor-4 antagonists, a receptor upregulated following microglial activation (Fiebich et al., 2018). Their efficacy varies in regards to clinical metrics or symptoms, but their safety profile is excellent with adverse effects comparable to placebo. The latter feature should make them attractive for future investigations.

Other emerging fundamental understanding of CNS (patho)physiology implies that waste clearing via the glymphatic system could be enhanced with adequate sleep, as animal experiment showed that the interstitial glymphatic space and bulk flow increases during sleep and low arousal (Hauglung et al., 2020). Interventions that improve circadian rhythmicity would cover this aspect as well. Finally, there is great interest in modulating the autonomic nervous system, which is the direct channel between the viscera and the brain, to enhance the antiinflammatory signaling and decrease the proinflammatory, primarily by targeting the vagal nerve transmission (Treatment of Mild Cognitive Impairment With Transcutaneous Vagal Nerve Stimulation, 2020). In an extended scope, microbiota is also a factor that could be utilized for optimizing autonomic nervous system activation and ultimately decreasing neuroinflammatory cascades (Angelucci et al., 2019). Treatments with antibiotics such as doxycycline, rifampin, and D-cycloserine showed initial positive results, but repeated studies did not redemonstrate the benefits. Ongoing clinical trials are investigating the potential of probiotics as an intervention (Probiotics in dementia, 2020).

Increasing neuronal resilience Decreasing or halting neurodegeneration is a daunting task, especially with advanced disease stages on a cellular level. Addressing the underlying risk factors, comorbidities, and lifestyle modifications remains the cornerstone of enabling a more favorable environment for neurorestoration. As such, the substrate underlying neuronal resilience remains poorly understood (Lesuis et al., 2018). Specific neurorepair therapies focus, for example, on stem cells (Wang et al., 2019a,b), neural growth factors (Mitra et al., 2019), and autophagy enhancers (Liu and Li, 2019). The outcomes of currently ongoing clinical trials on the use of stem cells for AD are still awaited. Devices for reliable biodelivery of neural growth factors are tested (Mitra et al., 2019), but some encouraging results are reported with the use of peptide preparation of cerebrolysin as an adjuvant to standard treatment (Allegri and Guekht, 2012). Importantly, the pleiotropic effects of physical activity on cognition, so far mainly in elderly (Castells-Sánchez et al., 2019; Stillman et al., 2020), include associated increase in levels of BDNF, and possibly increased hippocampal blood flow and neurogenesis, the latter two shown in animal experiments for now only (Marlatt et al., 2012, 2013; Liu and Nusslock, 2018). Autophagy enhancement is seen as a key therapeutic approach to age-associated degeneration or cellular malfunction, including for CNS-related tissues. Nonhuman experimental findings point to possible use of autophagy boosters based on trehalose,

CIRCADIAN CHANGES IN ALZHEIMER'S DISEASE lithium, and rapamycin, as options for future clinical studies of a wide spectrum of neurodegenerative conditions (Liu and Nusslock, 2018).

Optimizing vascular health The overlap between traditional cerebrovascular and AD risk factors, namely hypertension, type 2 diabetes mellitus, dyslipidemia, atherosclerosis, and atrial fibrillation, indicates that appropriate management of those stands as a salient preventative and therapeutic task (Cechetto et al., 2008). Worse vascular health is associated with neurodegeneration but not increased amyloid deposition per se (Vemuri et al., 2017). Application of nonpharmacologic lifestyle interventions that have shown benefits for cerebrovascular health is recommended in regards to AD as well. Furthermore, medications such as statins and ACE inhibitors may bring additional benefits. Lipophilic statins such as lovastatin and atorvastatin were possibly associated with worsening of cognitive health, but other analysis actually showed benefits with statin use. Ultimate analysis points to a more favorable effect of statins for cognitive health, though for a smaller group of especially vulnerable patients, those medications could be associated with cognitive decline (Schultz et al., 2018). ACE inhibitors may exert their observed positive clinical effects indirectly by controlling hypertension, and directly by acting on CNS angiotensin metabolism which reduces neuroinflammatory cascades (Rygiel, 2016). Compromised cerebral perfusion is associated with increased risk for cognitive decline (Wolters et al., 2017), and the vast majority of AD coexists with cerebrovascular disease (Attems and Jellinger, 2014). With regards to commonly used heart failure medications, beta blockers are known to delay the functional decline in AD, whereas diuretics were associated with the opposite trend (Rosenberg et al., 2008). Additionally, insulin resistance and hyperglycemia are key detrimental factors for endothelial health, and the use of antidiabetic drugs provides indirect benefits in regards to AD (Rizvi et al., 2015).

Increasing insulin sensitivity Insulin sensitivity and peripheral metabolic homeostasis are closely related to processes driving neurodegeneration, and a number of indicators of metabolic health such as glucostasis, BMI, and obesity have been recognized as important risk factors for AD (Cai et al., 2012). Consequently, both pharmacological and nonpharmacological interventions targeting metabolism and insulin signaling have been proposed in the context of prevention and treatment of AD. A number of modulators of insulin signaling are being tested as a potential therapy for AD both in the preclinical and clinical

295

setting (Ohyagi and Takei, 2020). A group led by Suzzane Craft has shown that intranasal insulin is able to improve memory in AD/MCI with regular insulin being superior to long-acting insulin detemir (Craft et al., 2012, 2017) and that insulin treatment might be more effective in males and carriers of APO-E4 (Claxton et al., 2013, 2015). Other antidiabetics have also shown some promising effects in clinical trials. Pioglitazone and rosiglitazone, agonists of peroxisome proliferator-activated receptors g that are currently used in therapy of T2DM, have been investigated as a possible therapy for AD due to their dual antiinflammatory and insulin sensitizing effect (Cai et al., 2012). A clinical trial suggested that rosiglitazone might be able to decrease the progression of cognitive dysfunction in AD/MCI (Watson et al., 2005), and pioglitazone has been shown to improve cognitive function in healthy individuals (Knodt et al., 2019), as well as in patients with mild AD accompanied with DM, in whom pioglitazone also increased parietal lobe blood flow and insulin sensitivity (Sato et al., 2011). Sitagliptin, an inhibitor of dipeptidyl peptidase-4, has shown some promising effects since a 6-month treatment was able to increase MMSE scores both in elderly with T2DM and T2DM and AD, in comparison to metformin (Isik et al., 2017). Agonists of GLP-1 receptor are also potential candidate drugs for AD as Gejl et al. demonstrated 6-month treatment with liraglutide was able to recover glucose transport across the blood–brain barrier (Gejl et al., 2017). Nonpharmacologic efforts to increase insulin sensitivity have been explored in the context of protective effects on cognition. Aerobic exercise and physical activity are well known for their protective effects on cognitive performance and metabolic health, and there is evidence that implementation of physical activity might be a good strategy to postpone cognitive decline and reduce the effect of AD risk factors such as glucose intolerance (Baker et al., 2010; Castells-Sánchez et al., 2019; Stillman et al., 2020). Exercise affects both central and peripheral circadian clocks, and correct implementation of physical activity might help reduce harmful consequences of circadian dysrhythmia (Tahara et al., 2017). Another important nonpharmacological treatment modality targeting metabolism in AD is optimization of nutrient intake. Numerous nutritional deficiencies were reported in the elderly and multinutritional correction of deficient polyunsaturated fatty acids, B vitamins, and antioxidants were proposed both for preventive and therapeutic purposes (Kamphuis and Scheltens, 2010; Soininen et al., 2017; Doorduijn et al., 2020). Furthermore, meal timing should be optimized in AD as it might help restore circadian rhythmicity and enable proper nutrient extraction (Homolak et al., 2018).

296

K. TOLJAN AND J. HOMOLAK

CONCLUSION/SUMMARY AD is a clinical dementia syndrome with increasing epidemiologic importance as the general population ages. Traditional focus on memory problems as the hallmark has been expanded with the recognition of accompanying symptoms and common comorbidities that also point to hypothalamic, neuroendocrine, and neurovascular dysregulation, such as decreased sleep quality and quantity and features of metabolic syndrome. Disrupted circadian rhythms, notably by environmental factors or behavior, and impaired brain glucose metabolism, were found to be associated with later development of AD. On a molecular level, a widespread vicious cycle between neurodegeneration and neuroinflammation in the CNS is evident in AD, and hypothalamus, including the SCN, is also affected by these destructive processes. Complex multidirectional pathophysiologic interactions ensue as circadian rhythm disruption and inflammatory processes related to metabolic changes promote further neurodegeneration and neuroinflammation. Ultimately, it leads to a common pathway characterized by CNS amyloid and tau protein buildup. With still evolving understanding of the full AD pathophysiology, therapeutic arsenal remains limited. Cholinomimetics and memantine as approved treatment yield humble benefits and cause clinically relevant side effects (Leonard, 2004), whereas recent attempts with monoclonal antibodies targeting the pathologic substrates seen in AD have not lived up to their expectations. By deeper understanding of the underlying disease processes, attention is slowly being directed at addressing the neuroinflammation and brain metabolic alterations, as these steps precede the generation of microscopically visible pathologic end products associated with AD. Compounds with potential to promote lowering of brain insulin resistance (insulin-based and other antidiabetic drugs), decrease neuronal and glial inflammation (traditional antiinflammatories and glial modulators) and promote neuronal resilience (various growth factors), which are being investigated as possible pharmacologic treatments effective for AD. Nonpharmacologic interventions that improve sleep and circadian rhythmicity, decrease cerebrovascular risk factors, and improve overall metabolic aspects of health have been shown to be beneficial for AD and represent readily available therapeutic modalities. However, the epidemiologic context is broadened with the impact of caregiver burnout and indirect disease burden for the society. Ideally, by detecting subclinical changes that are associated with greater risk for developing AD, timely interventions may slow the disease progression and impact the quality and quantity of life for the patients and their caregivers. A greater understanding of AD etiopathogenesis should yield more

effective or new therapies. Considering AD beyond hippocampus and memory problems, one possible route to achieve the aforementioned is by expanding it with the hypothalamic perspective.

REFERENCES 2020 Alzheimer’s Disease Facts and Figures (2020). 2020 Alzheimer’s disease facts and figures. Alzheimers Dement 16: 391–460. https://doi.org/10.1002/alz.12068. ADAPT-FS Research Group (2015). Follow-up evaluation of cognitive function in the randomized Alzheimer’s Disease Anti-inflammatory Prevention Trial (ADAPT) and its Follow-up Study (ADAPT-FS). Alzheimers Dement 11: 216. NIH Public Access. Adebakin A et al. (2012). Impaired satiation and increased feeding behaviour in the triple-transgenic Alzheimer’s disease mouse model. PLoS One 7: e45179. Aisen PS et al. (2000). A randomized controlled trial of prednisone in Alzheimer’s disease. Alzheimer’s disease cooperative study. Neurology 54: 588–593. https://doi.org/ 10.1212/wnl.54.3.588. Ali MM et al. (2019) Recommendations for anti-inflammatory treatments in Alzheimer’s disease: a comprehensive review of the literature. Cureus 11: e4620 . https://doi.org/10.7759/ cureus.4620. Cureus Inc. Allegri RF, Guekht A (2012). Cerebrolysin improves symptoms and delays progression in patients with Alzheimer’s disease and vascular dementia. Drugs Today (Barc) 48: 25–41. https://doi.org/10.1358/dot.2012.48(Suppl.A). 1739721. Drugs Today (Barc). Alzforum Research Models Repository: Tg2576. n.d. Available at: https://www.alzforum.org/research-models/ tg2576 (Accessed: 9 August 2020). Andrade AG et al. (2018). The relationship between obstructive sleep apnea and Alzheimer’s disease. J Alzheimers Dis 64: S255–S270. IOS Press. Angelucci F et al. (2019). Antibiotics, gut microbiota, and Alzheimer’s disease. J Neuroinflammation 16: 1–10. BioMed Central. Attems J, Jellinger KA (2014). The overlap between vascular disease and Alzheimer’s disease—lessons from pathology. BMC Med 12: 206. https://doi.org/10.1186/s12916-0140206-2. BioMed Central. Baker LD et al. (2010). Aerobic exercise improves cognition for older adults with glucose intolerance, a risk factor for Alzheimer’s disease. J Alzheimers Dis 22: 569–579. Baldacchino FV, Pedrinolla A, Venturelli M (2018). Exerciseinduced adaptations in patients with Alzheimer’s disease: the role of circadian scheduling. Sport Sci Health 14: 227–234. Springer. Barber RC (2012). The genetics of Alzheimer’s disease. Scientifica 2012: 246210. Barilar JO et al. (2020). Shared cerebral metabolic pathology in non-transgenic animal models of Alzheimer’s and Parkinson’s disease. J Neural Transm 127: 231–250. Berg-Weger M, Stewart DB (2017). Non-pharmacologic interventions for persons with dementia. MO Med

CIRCADIAN CHANGES IN ALZHEIMER'S DISEASE 114: Available at: https://pubmed.ncbi.nlm.nih.gov/30228557/ (Accessed: 5 July 2020). Bloch K et al. (2017). Intracerebroventricular streptozotocin induces obesity and dementia in Lewis rats. J Alzheimers Dis 60: 121–136. Bossers K et al. (2010). Concerted changes in transcripts in the prefrontal cortex precede neuropathology in Alzheimer’s disease. Brain 133: 3699–3723. Braak H, Braak E (1991). Neuropathological stageing of Alzheimer-related changes. Acta Neuropathol 82: 239–259. Brinton RD (2004). Impact of estrogen therapy on Alzheimer’s disease: a fork in the road? CNS Drugs 18: 405–422. Brzecka A et al. (2018). Sleep disorders associated with Alzheimer’s disease: a perspective. Front Neurosci 12: 330. Cai H et al. (2012). Metabolic dysfunction in Alzheimer’s disease and related neurodegenerative disorders. Curr Alzheimer Res 9: 5–17. Calvo F et al. (2003). Diagnoses, syndromes, and diseases: a knowledge representation problem. In: AMIA ... Annual Symposium Proceedings/AMIA Symposium, AMIA Symposium, p. 802. Cardinali DP, Furio AM, Brusco LI (2010). Clinical aspects of melatonin intervention in Alzheimer’s disease progression. Curr Neuropharmacol 8: 218. Bentham Science Publishers. Castells-Sa´nchez A et al. (2019). Effects and mechanisms of cognitive, aerobic exercise, and combined training on cognition, health, and brain outcomes in physically inactive older adults: The Projecte Moviment Protocol. Front Aging Neurosci 11: 216. Cechetto DF, Hachinski V, Whitehead SN (2008). Vascular risk factors and Alzheimer’s disease. Expert Rev Neurother 8: 743–750. https://doi.org/10.1586/14737175. 8.5.743. Ch^ene G et al. (2015). Gender and incidence of dementia in the Framingham Heart Study from mid-adult life. Alzheimers Dement 11: 310–320. Choi HJ et al. (2017). Associations of thyroid hormone serum levels with in-vivo Alzheimer’s disease pathologies. Alzheimers Res Ther 9: 1–9. BioMed Central. Choi SH et al. (2018). Combined adult neurogenesis and BDNF mimic exercise effects on cognition in an Alzheimer’s mouse model. Science 361: eaan8821. https://doi.org/ 10.1126/science.aan8821. Claxton A et al. (2013). Sex and ApoE genotype differences in treatment response to two doses of intranasal insulin in adults with mild cognitive impairment or Alzheimer’s disease. J Alzheimers Dis 35: 789–797. Claxton A et al. (2015). Long-acting intranasal insulin detemir improves cognition for adults with mild cognitive impairment or early-stage Alzheimer’s disease dementia. J Alzheimers Dis 44: 897–906. Craft S et al. (2012). Intranasal insulin therapy for Alzheimer disease and amnestic mild cognitive impairment: a pilot clinical trial. Arch Neurol 69: 29–38. Craft S et al. (2017). Effects of regular and long-acting insulin on cognition and Alzheimer’s disease biomarkers: a pilot clinical trial. J Alzheimers Dis 57: 1325–1334. Cui S-Y et al. (2018). Intracerebroventricular streptozotocininduced Alzheimer’s disease-like sleep disorders in rats:

297

role of the GABAergic system in the parabrachial complex. CNS Neurosci Ther 24: 1241–1252. De Gennaro L et al. (2017). The fall of sleep K-complex in Alzheimer disease. Sci Rep 7: 39688. Doorduijn AS, de van der Schueren MAE, van de Rest O et al. (2020). Nutritional status is associated with clinical progression in Alzheimer’s disease. the NUDAD project, J Am Med Dir Assoc S1525-8610(20)30895-1. https:// doi.org/10.1016/j.jamda.2020.10.020. Epub ahead of print. PMID: 33239240. Douma LG, Gumz ML (2018). Circadian clock-mediated regulation of blood pressure. Free Radic Biol Med 119: 108–114. Dubelaar EJG et al. (2006). Increased metabolic activity in nucleus basalis of Meynert neurons in elderly individuals with mild cognitive impairment as indicated by the size of the Golgi apparatus. J Neuropathol Exp Neurol 65: 257–266. Fakhoury M (2018). Microglia and astrocytes in Alzheimer’s disease: implications for therapy. Curr Neuropharmacol 16: 508. Bentham Science Publishers. Fiebich BL, Batista CRA, Saliba SW et al. (2018). Role of microglia TLRs in neurodegeneration. Front Cell Neurosci 12: 329. https://doi.org/10.3389/fncel.2018.00329. Fox RJ et al. (2018). Phase 2 trial of ibudilast in progressive multiple sclerosis. N Engl J Med 379: 846–855. Gatz M et al. (2006). Role of genes and environments for explaining Alzheimer disease. Arch Gen Psychiatry 63: 168–174. Gejl M et al. (2017). Blood-brain glucose transfer in Alzheimer’s disease: effect of GLP-1 analog treatment. Sci Rep 7: 17490. Gonza´lez-Domı´nguez R et al. (2015). Metabolomic investigation of systemic manifestations associated with Alzheimer’s disease in the APP/PS1 transgenic mouse model. Mol Biosyst 11: 2429–2440. Grieb P (2016). Intracerebroventricular streptozotocin injections as a model of Alzheimer’s disease: in search of a relevant mechanism. Mol Neurobiol 53: 1741–1752. Grinberg LT et al. (2009). The dorsal raphe nucleus shows phospho-tau neurofibrillary changes before the transentorhinal region in Alzheimer’s disease. A precocious onset? Neuropathol Appl Neurobiol 35: 406–416. Hahn EA et al. (2014). A change in sleep pattern may predict Alzheimer disease. Am J Geriatr Psychiatry 22: 1262–1271. Hanford N, Figueiro M (2013). Light therapy and Alzheimer’s disease and related dementia: past, present, and future. J Alzheimers Dis 33: 913. NIH Public Access. Hauglung N, Pavan C, Nedergaard M (2020). Cleaning the sleeping brain—the potential restorative function of the glymphatic system. Curr Opin Physio 15: 1–6. Elsevier. Hebert LE et al. (2001). Annual incidence of Alzheimer disease in the United States projected to the years 2000 through 2050. Alzheimer Dis Assoc Disord 15: 169–173. Herbert J, Lucassen PJ (2016). Depression as a risk factor for Alzheimer’s disease: genes, steroids, cytokines and neurogenesis—what do we need to know? Front Neuroendocrinol 41: 153–171.

298

K. TOLJAN AND J. HOMOLAK

Herring WJ et al. (2020). Polysomnographic assessment of suvorexant in patients with probable Alzheimer’s disease dementia and insomnia: a randomized trial. Alzheimers Dement 16: 541–551. https://doi.org/10.1002/alz.12035. Hollands C et al. (2017). Depletion of adult neurogenesis exacerbates cognitive deficits in Alzheimer’s disease by compromising hippocampal inhibition. Mol Neurodegener 12: 64. Homolak J et al. (2018). Circadian rhythm and Alzheimer’s disease. Med Sci (Basel, Switzerland) 6: 52. https://doi. org/10.3390/medsci6030052. Hood S, Amir S (2017). The aging clock: circadian rhythms and later life. J Clin Invest 127: 437–446. American Society for Clinical Investigation. Hu Y-S et al. (2010). Complex environment experience rescues impaired neurogenesis, enhances synaptic plasticity, and attenuates neuropathology in familial Alzheimer’s disease-linked APPswe/PS1DeltaE9 mice. FASEB J 24: 1667–1681. 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. Isik AT et al. (2017). The effects of sitagliptin, a DPP-4 inhibitor, on cognitive functions in elderly diabetic patients with or without Alzheimer’s disease. Diabetes Res Clin Pract 123: 192–198. Janto K, Prichard JR, Pusalavidyasagar S (2018). An update on dual orexin receptor antagonists and their potential role in insomnia therapeutics. J Clin Sleep Med 14: 1399–1408. Ju Y-ES, Lucey BP, Holtzman DM (2014). Sleep and Alzheimer disease pathology—a bidirectional relationship. Nat Rev Neurol 10: 115–119. Kamphuis PJGH, Scheltens P (2010). Can nutrients prevent or delay onset of Alzheimer’s disease? J Alzheimers Dis 20: 765–775. Karran E, Mercken M, De Strooper B (2011). The amyloid cascade hypothesis for Alzheimer’s disease: an appraisal for the development of therapeutics. Nat Rev Drug Discov 10: 698–712. Kempermann G et al. (2018). Human adult neurogenesis: evidence and remaining questions. Cell Stem Cell 23: 25–30. Kent BA (2014). Synchronizing an aging brain: can entraining circadian clocks by food slow Alzheimer’s disease? Front Aging Neurosci 6: 234. https://doi.org/10.3389/fnagi. 2014.00234. Frontiers Media SA. Kim HY et al. (2016). Intracerebroventricular injection of amyloid-b peptides in normal mice to acutely induce Alzheimer-like cognitive deficits. J Vis Exp 53308. https://doi.org/10.3791/53308. Kin NM et al. (2004). Secretion of melatonin in healthy elderly subjects: a longitudinal study. Ann N Y Acad Sci 1019: 326–329. https://doi.org/10.1196/annals.1297.055. Kivipelto M, Mangialasche F, Ngandu T (2018). Lifestyle interventions to prevent cognitive impairment, dementia and Alzheimer disease. Nat Rev Neurol 14: 653–666. Nature Publishing Group. Knezovic A et al. (2017). Rat brain glucose transporter-2, insulin receptor and glial expression are acute targets of

intracerebroventricular streptozotocin: risk factors for sporadic Alzheimer’s disease? J Neural Transm 124: 695–708. Knezovic A et al. (2018). Glucagon-like peptide-1 mediates effects of oral galactose in streptozotocin-induced rat model of sporadic Alzheimer’s disease. Neuropharmacology 135: 48–62. Knight EM et al. (2012). Hypermetabolism in a tripletransgenic mouse model of Alzheimer’s disease. Neurobiol Aging 33: 187–193. Knodt AR et al. (2019). Effects of pioglitazone on mnemonic hippocampal function: a blood oxygen level-dependent functional magnetic resonance imaging study in elderly adults. Alzheimers Dement 5: 254–263. Koch CE et al. (2017). Interaction between circadian rhythms and stress. Neurobiol Stress 6: 57–67. Lazarov O, Hollands C (2016). Hippocampal neurogenesis: learning to remember. Prog Neurobiol 138–140: 1–18. Lee EB (2011). Obesity, leptin, and Alzheimer’s disease. Ann N Y Acad Sci 1243: 15. NIH Public Access. Leonard BE (2004). Pharmacotherapy in the treatment of Alzheimer’s disease: an update. World Psychiatry 3: 84–88. Lesuis SL, Hoeijmakers L, Korosi A et al. (2018). Vulnerability and resilience to Alzheimer’s disease: early life conditions modulate neuropathology and determine cognitive reserve. Alzheimers Res Ther 10: 95. https://doi.org/10.1186/s13195018-0422-7. Lim ASP et al. (2013). Sleep fragmentation and the risk of incident Alzheimer’s disease and cognitive decline in older persons. Sleep 36: 1027–1032. Oxford Academic. Liu J, Li L (2019). Targeting autophagy for the treatment of Alzheimer’s disease: challenges and opportunities. Front Mol Neurosci 12: 203. https://doi.org/10.3389/fnmol. 2019.00203. Liu PZ, Nusslock R (2018). Exercise-mediated neurogenesis in the hippocampus via BDNF. Front Neurosci12: 52. https://doi.org/10.3389/fnins.2018.00052. Frontiers. Liu X et al. (2019a). Interplay between Alzheimer’s disease and global glucose metabolism revealed by the metabolic profile alterations of pancreatic tissue and serum in APP/ PS1 transgenic mice. Acta Pharmacol Sin 40: 1259–1268. Liu Y et al. (2019b). SIRT3 mediates hippocampal synaptic adaptations to intermittent fasting and ameliorates deficits in APP mutant mice. Nat Commun 10: 1886. Low DY et al. (2019). Diet-related metabolites associated with cognitive decline revealed by untargeted metabolomics in a prospective cohort. Mol Nutr Food Res 63: e1900177. Lucassen PJ, Toni N, Kempermann G et al. (2000a) Limits to human neurogenesis – really? Mol Psychiatry 25: 2207–2209. https://doi.org/10.1038/s41380-018-0337-5. Lucassen PJ et al. (2020b). Adult neurogenesis, human after all (again): classic, optimized, and future approaches. Behav Brain Res 381: 112458. Lucey BP et al. (2018). Effect of sleep on overnight cerebrospinal fluid amyloid b kinetics. Ann Neurol 83: 197–204. Macklin L et al. (2017). Glucose tolerance and insulin sensitivity are impaired in APP/PS1 transgenic mice prior to amyloid plaque pathogenesis and cognitive decline. Exp Gerontol 88: 9–18.

CIRCADIAN CHANGES IN ALZHEIMER'S DISEASE Mapstone M et al. (2014). Plasma phospholipids identify antecedent memory impairment in older adults. Nat Med 20: 415–418. Marlatt MW et al. (2012). Running throughout middle-age improves memory function, hippocampal neurogenesis, and BDNF levels in female C57BL/6J mice. Dev Neurobiol 72: 943–952. Marlatt MW et al. (2013). Prolonged running, not fluoxetine treatment, increases neurogenesis, but does not alter neuropathology, in the 3xTg mouse model of Alzheimer’s disease. Curr Top Behav Neurosci 15: 313–340. Metz LM et al. (2017). Trial of minocycline in a clinically isolated syndrome of multiple sclerosis. N Engl J Med 376: 2122–2133. Mitra S, Behbahani H, Eriksdotter M (2019). Innovative therapy for Alzheimer’s disease-with focus on biodelivery of NGF. Front Neurosci 13: 38. https://doi.org/10.3389/ fnins.2019.00038. Moreno-Jimenez EP et al. (2019). Adult hippocampal neurogenesis is abundant in neurologically healthy subjects and drops sharply in patients with Alzheimer’s disease. Nat Med 25: 554–560. Nature Publishing Group. Mosconi L (2013). Glucose metabolism in normal aging and Alzheimer’s disease: methodological and physiological considerations for PET studies. Clin Transl Imaging 1: 217–233. https://doi.org/10.1007/s40336-013-0026-y. Mosconi L et al. (2009). FDG-PET changes in brain glucose metabolism from normal cognition to pathologically verified Alzheimer’s disease. Eur J Nucl Med Mol Imaging 36: 811–822. Moya-Alvarado G et al. (2016). Neurodegeneration and Alzheimer’s disease (AD). What can proteomics tell us about the Alzheimer’s brain? Mol Cell Proteomics 15: 409–425. American Society for Biochemistry and Molecular Biology. Musiek ES, Xiong DD, Holtzman DM (2015). Sleep, circadian rhythms, and the pathogenesis of Alzheimer disease. Exp Mol Med 47: e148. Naninck EFG et al. (2017). Early micronutrient supplementation protects against early stress-induced cognitive impairments. FASEB J 31: 505–518. Nelis SM et al. (2019). The impact of co-morbidity on the quality of life of people with dementia: findings from the IDEAL study. Age Ageing 48: 361–367. Oxford Academic. Neth BJ, Craft S (2017). Insulin resistance and Alzheimer’s disease: bioenergetic linkages. Front Aging Neurosci 9: 345. Niemantsverdriet E et al. (2017). Alzheimer’s disease CSF biomarkers: clinical indications and rational use. Acta Neurol Belg 117: 591. Springer. Ohyagi Y, Takei SI (2020). Insulin signaling as a therapeutic target in Alzheimer’s disease: efficacy of apomorphine. Neurol Clin Neurosci 8: 146–154. Ott A et al. (1999). Diabetes mellitus and the risk of dementia: the Rotterdam study. Neurology 53: 1937–1942. Ownby RL et al. (2006). Depression and risk for Alzheimer disease: systematic review, meta-analysis, and metaregression analysis. Arch Gen Psychiatry 63: 530–538. American Medical Association.

299

Pegueroles J et al. (2018). Obesity and Alzheimer’s disease, does the obesity paradox really exist? A magnetic resonance imaging study. Oncotarget 9: 34691–34698. Impact Journals. Plane JM et al. (2010). Prospects for minocycline neuroprotection. Arch Neurol 67: 1442–1448. American Medical Association. Probiotics in dementia. Available at: https://clinicaltrials.gov/ ct2/show/NCT03847714 (Accessed: 8 August 2020). Reisberg B et al. (1999). Retrogenesis: clinical, physiologic, and pathologic mechanisms in brain aging, Alzheimer’s and other dementing processes. Eur Arch Psychiatry Clin Neurosci 249: S28–S36. Springer. Richetin K et al. (2015). Genetic manipulation of adult-born hippocampal neurons rescues memory in a mouse model of Alzheimer’s disease. Brain 138: 440–455. Riemersma-van der Lek RF et al. (2008). Effect of bright light and melatonin on cognitive and noncognitive function in elderly residents of group care facilities: a randomized controlled trial. JAMA 299: 2642–2655. Rizvi SMD et al. (2015). Role of anti-diabetic drugs as therapeutic agents in Alzheimer’s disease. EXCLI J 14: 684. Leibniz Research Centre for Working Environment and Human Factors. Roenneberg T, Merrow M (2016). The circadian clock and human health. Curr Biol 26: R432–R443. Cell Press. Rojas-Gutierrez E et al. (2017). Alzheimer’s disease and metabolic syndrome: a link from oxidative stress and inflammation to neurodegeneration. Synapse 71: e21990. Rosenberg PB et al. (2008). Effects of cardiovascular medications on rate of functional decline in Alzheimer disease. Am J Geriatr Psychiatry 16: 883. NIH Public Access. Roy U et al. (2019). Degeneration of the suprachiasmatic nucleus in an Alzheimer’s disease mouse model monitored by in vivo magnetic resonance relaxation measurements and immunohistochemistry. J Alzheimers Dis 69: 363–375. Rygiel K (2016). Can angiotensin-converting enzyme inhibitors impact cognitive decline in early stages of Alzheimer’s disease? An overview of research evidence in the elderly patient population. J Postgrad Med 62: 242. Wolters Kluwer—Medknow Publications. Santos TO et al. (2012). Early and late neurodegeneration and memory disruption after intracerebroventricular streptozotocin. Physiol Behav 107: 401–413. Sato T et al. (2011). Efficacy of PPAR-g agonist pioglitazone in mild Alzheimer disease. Neurobiol Aging 32: 1626–1633. Schiel KA (2018). A new etiologic model for Alzheimer’s disease. Med Hypotheses 111: 27–35. Schmidt C, Peigneux P, Cajochen C (2012). Age-related changes in sleep and circadian rhythms: impact on cognitive performance and underlying neuroanatomical networks. Front Neurol 3: 118. https://doi.org/10.3389/fneur. 2012.00118. Frontiers. Schultz BG, Patten DK, Berlau DJ (2018). The role of statins in both cognitive impairment and protection against dementia: a tale of two mechanisms. Transl Neurodegener 7: 5. https://doi.org/10.1186/s40035-018-0110-3. BioMed Central.

300

K. TOLJAN AND J. HOMOLAK

Shoham S et al. (2003). Intracerebroventricular injection of streptozotocin causes neurotoxicity to myelin that contributes to spatial memory deficits in rats. Exp Neurol 184: 1043–1052. Simic G et al. (2009). Does Alzheimer’s disease begin in the brainstem? Neuropathol Appl Neurobiol 35: 532–554. Soininen H, Solomon A, Visser PJ et al. (2017). 24-month intervention with a specific multinutrient in people with prodromal Alzheimer’s disease (LipiDiDiet): a randomised, double-blind, controlled trial. Lancet Neurol 16: 965–975. https://doi.org/10.1016/S1474-4422(17)30332-0. Epub 2017 Oct 30. PMID: 29097166; PMCID: PMC569 7936. Steen E et al. (2005). Impaired insulin and insulin-like growth factor expression and signaling mechanisms in Alzheimer’s disease—is this type 3 diabetes? J Alzheimers Dis 7: 63–80. Stenvers DJ et al. (2018). Circadian clocks and insulin resistance. Nat Rev Endocrinol 15: 75–89. Nature Publishing Group. Stillman CM et al. (2020). Effects of exercise on brain and cognition across age groups and health states. Trends Neurosci 43: 533–543. Swaab DF (1997). Chapter II. Neurobiology and neuropathology of the human hypothalamus. In: The primate nervous system, part I, Elsevier. 39–137. Handbook of Chemical Neuroanatomy. 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. Elsevier. 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. Tahara Y, Aoyama S, Shibata S (2017). The mammalian circadian clock and its entrainment by stress and exercise. J Physiol Sci 67: 1–10. Tan RS, Pu SJ (2003). A pilot study on the effects of testosterone in hypogonadal aging male patients with Alzheimer’s disease. Aging Male 6: 13–17. Aging Male. Available at: https://pubmed.ncbi.nlm.nih.gov/12809076/ (Accessed: 13 June 2020). Taylor CA (2018). Subjective cognitive decline among adults aged 45 years—United States, 2015–2016. Morb Mortal Wkly Rep 67: 753–757. https://doi.org/10.15585/mmwr. mm6727a1. Tobin MK et al. (2019). Human hippocampal neurogenesis persists in aged adults and Alzheimer’s disease patients. Cell Stem Cell 24: 974–982.e3. Toledo JB et al. (2017). Metabolic network failures in Alzheimer’s disease: a biochemical road map. Alzheimers Dement 13: 965–984. Toljan K, Vrooman B (2018). Low-dose naltrexone (LDN)— review of therapeutic utilization. Med Sci 6: 82. Multidisciplinary Digital Publishing Institute. Treatment of Mild Cognitive Impairment With Transcutaneous Vagal Nerve Stimulation (2020). Treatment of Mild

Cognitive Impairment With Transcutaneous Vagal Nerve Stimulation. Available at: https://clinicaltrials.gov/ct2/ show/NCT03359902 (Accessed: 8 August 2020). Van Cauwenberghe C, Van Broeckhoven C, Sleegers K (2016). The genetic landscape of Alzheimer disease: clinical implications and perspectives. Genet Med 18: 421–430. Van Erum J, Van Dam D, De Deyn PP (2019). Alzheimer’s disease: neurotransmitters of the sleep-wake cycle. Neurosci Biobehav Rev 105: 72–80. Varma VR et al. (2018). Brain and blood metabolite signatures of pathology and progression in Alzheimer disease: a targeted metabolomics study. PLoS Med 15: e1002482. Vemuri P et al. (2017). Age, vascular health, and Alzheimer’s disease biomarkers in an elderly sample. Ann Neurol 82: 706. NIH Public Access. Wang S-M, Lee C-U, Lim HK (2019a). Stem cell therapies for Alzheimer’s disease: is it time? Curr Opin Psychiatry 32: 105–116. Wang Y et al. (2019b). Choline supplementation ameliorates behavioral deficits and Alzheimer’s disease-like pathology in transgenic APP/PS1 mice. Mol Nutr Food Res 63: e1801407. Wang J et al. (2020). Capsaicin consumption reduces brain amyloid-beta generation and attenuates Alzheimer’s disease-type pathology and cognitive deficits in APP/PS1 mice. Transl Psychiatry 10: 230. Watson GS et al. (2005). Preserved cognition in patients with early Alzheimer disease and amnestic mild cognitive impairment during treatment with rosiglitazone: a preliminary study. Am J Geriatr Psychiatry 13: 950–958. Wisor JP et al. (2005). Sleep and circadian abnormalities in a transgenic mouse model of Alzheimer’s disease: a role for cholinergic transmission. Neuroscience 131: 375–385. Wolters FJ et al. (2017). Cerebral perfusion and the risk of dementia: a population-based study. Circulation 136: 719–728. Wu Y-H, Swaab DF (2007). Disturbance and strategies for reactivation of the circadian rhythm system in aging and Alzheimer’s disease. Sleep Med 8: 623–636. Xie L et al. (2013). Sleep drives metabolite clearance from the adult brain. Science 342: 373–377. American Association for the Advancement of Science. Yam K-Y et al. (2019). Increasing availability of o-3 fatty acid in the early-life diet prevents the early-life stress-induced cognitive impairments without affecting metabolic alterations. FASEB J 33: 5729–5740. Zhang F et al. (2019). Alteration in sleep architecture and electroencephalogram as an early sign of Alzheimer’s disease preceding the disease pathology and cognitive decline. Alzheimers Dement 15: 590–597. Zheng H et al. (2018). The hypothalamus as the primary brain region of metabolic abnormalities in APP/PS1 transgenic mouse model of Alzheimer’s disease. Biochim Biophys Acta 1864: 263–273. Molecular Basis of Disease. Zhu L, Zee PC (2012). Circadian rhythm sleep disorders. Neurol Clin 30: 1167–1191. Elsevier.

Handbook of Clinical Neurology, Vol. 179 (3rd series) The Human Hypothalamus: Anterior Region D.F. Swaab, F. Kreier, P.J. Lucassen, A. Salehi, and R.M. Buijs, Editors https://doi.org/10.1016/B978-0-12-819975-6.00019-4 Copyright © 2021 Elsevier B.V. All rights reserved

Chapter 20

The circadian system in Parkinson’s disease, multiple system atrophy, and progressive supranuclear palsy KARIM FIFEL1* AND TOM DE BOER2 1

International Institute for Integrative Sleep Medicine (WPI-IIIS), University of Tsukuba, Tsukuba, Ibaraki, Japan 2

Laboratory for Neurophysiology, Department of Cell and Chemical Biology, Leiden University Medical Center, Leiden, The Netherlands

Abstract Circadian organization of physiology and behavior is an important biologic process that allows organisms to anticipate and prepare for predictable changes in the environment. Circadian disruptions are associated with a wide range of health issues. In patients with neurodegenerative diseases, alterations of circadian rhythms are among the most common and debilitating symptoms. Although a growing awareness of these symptoms has occurred during the last decade, their underlying neuropathophysiologic circuitry remains poorly understood and, consequently, no effective therapeutic strategies are available to alleviate these health issues. Recent studies have examined the neuropathologic status of the different neural components of the circuitry governing the generation of circadian rhythms in neurodegenerative diseases. In this review, we will dissect the potential contribution of dysfunctions in the different nodes of this circuitry to circadian alterations in patients with parkinsonism-linked neurodegenerative diseases (namely, Parkinson’s disease, multiple system atrophy, and progressive supranuclear palsy). A deeper understanding of these mechanisms will provide not only a better understanding of disease neuropathophysiology but also holds promise for the development of more effective and mechanisms-based therapies.

INTRODUCTION In the course of aging, the circadian organization of a wide range of behavioral and physiologic health parameters progressively deteriorates (Mattis and Sehgal, 2016). Among the most established and easily recognized signs of circadian alterations in the elderly is the progressive worsening of sleep/wake behavior (Mattis and Sehgal, 2016). In neurodegenerative disorders, these alterations are exacerbated in degree and pace of progression (van Someren et al., 1996; Hu et al., 2009; Coogan et al., 2013; Videnovic and Golombek, 2013, 2017; Videnovic et al., 2014a,b; Mattis and Sehgal, 2016; Musiek and Holtzman, 2016; Videnovic and Willis, 2016; De Pablo-Fernández et al., 2017; Fifel, 2017;

Leng et al., 2019). Circadian alterations have a significant debilitating impact on both patients and a burdening impact on family members. Indeed, the challenge imposed by sleep and circadian disturbances to family caregivers is the main cause of institutionalization in the elderly (Pollak and Perlick, 1991). According to a report on the world population aging by the United Nations in 2017, the number of people aged 60 or over is expected to more than double worldwide, reaching 2.1 billion people by 2050. Due to life expectancy improvements, the societal and economic burden associated with age-related neurodegenerative disorders is going to be challenging. Currently, there are no effective treatments or cures for neurodegenerative diseases.

*Correspondence to: Karim Fifel, International Institute for Integrative Sleep Medicine (WPI-IIIS), University of Tsukuba, 1-1-1 Tennodai, Tsukuba, Ibaraki 305-8575, Japan. E-mail: [email protected]

302

K. FIFEL AND T. DE BOER

Consequently, growing lines of research are aiming to develop palliative strategies (Riemersma-van der Lek et al., 2008; Fifel and Videnovic, 2018, 2019) alongside novel neural circuit-inspired therapeutic interventions (Canter et al., 2016). Along these lines, and given the reciprocal functional links between circadian rhythms and health quality (Bass and Lazar, 2016), an increasing number of studies are aimed at investigating and understanding neuronal mechanisms underlying the alterations of circadian rhythms in patients with neurodegenerative disorders. Insights from these investigations are expected to benefit the current attempts to develop efficient clinical interventions that aim to restore robust circadian rhythms, to positively impact global health of the patients. In this review, we describe the main circadian alterations in three age-related neurodegenerative diseases: Parkinson’s disease (PD), multiple system atrophy (MSA), and progressive supranuclear palsy (PSP). These three neurologic disorders have in common the development of parkinsonism as a cardinal motor phenotype over the course of disease progression (Fanciulli and Wenning, 2015; Poewe et al., 2017; Stamelou et al., 2019a,b). We additionally dissect the pathologic status of the different components of the neuronal networks governing circadian rhythms. The potential of this neuronal network-based approach in understanding circadian alterations in these disorders as well as in the development of efficient therapeutic interventions is also highlighted.

CIRCADIAN ALTERATIONS IN PD, MAS, AND PSP Circadian rhythms are 24-h oscillating biologic changes that enable an organism to anticipate and successfully adapt to the external daily changes in the environment induced by the earth’s rotation on its axes (Bass and Lazar, 2016). Circadian rhythms manifest at every level of organization, from gene expression to interorgan functional coordination that is essential for physiologic homeostasis (Bass and Lazar, 2016). The underlying molecular basis of circadian rhythms consists of transcriptional and translational autoregulatory feedback loops that are fine-tuned and modulated by several posttranslational biochemical reactions (Takahashi et al., 2008; Bass and Lazar, 2016). The ubiquitous expression of clock genes in virtually all organismal cells necessitates a well-controlled hierarchic coordination in order to ensure an optimal temporal function at the organismal level (Hastings et al., 2003; Dibner et al., 2010; Albrecht, 2012; Schibler et al., 2015). Indeed, chronic temporal misalignment between central and peripheral clocks as a result of the imperatives of modern society (i.e., electric

lights, jet travel, shift work, 24/7 food availability) is causally associated with serious debilitating health disorders (Takahashi et al., 2008; Bass and Lazar, 2016). During aging, clock function declines progressively as evidenced by a blunted and fragmented rest/activity rhythm and by impaired phase entrainment of several physiologic processes, such as melatonin release, cortisol secretion, and body temperature (Van Someren, 2000; Welz and Benitah, 2019; Zhao et al., 2019). It is therefore not surprising that patients afflicted with agerelated neurodegenerative disorders complain of alterations in a variety of circadian rhythms (Coogan et al., 2013; Schroeder and Colwell, 2013; Videnovic and Golombek, 2013, 2017; Videnovic et al., 2014a; Mattis and Sehgal, 2016; Musiek and Holtzman, 2016; Videnovic and Willis, 2016; De Pablo-Fernández et al., 2017; Fifel, 2017; Leng et al., 2019). In this chapter, we discuss in detail the nature as well as the underlying neuropathophysiology of circadian alterations in PD, MSA, and PSP. For other neurodegenerative disorders, we refer readers to several excellent reviews (Musiek and Holtzman, 2016; Fifel, 2017; Leng et al., 2019; Long and Holtzman, 2019). PD, MSA, and PSP are multisystem and autonomic system disorders involving several subcortical nuclei, cortical areas, and spinal cord structures; portions of the peripheral and enteric nervous system also become involved (Fanciulli and Wenning, 2015; Poewe et al., 2017; Stamelou et al., 2019a,b). PD is the second most prevalent neurodegenerative disease that is clinically diagnosed by parkinsonism, consisting of a tetrad of motor symptoms: bradykinesia, resting tremor, rigidity, and postural instability (Poewe et al., 2017). MSA is characterized by a progressive autonomic failure, parkinsonian features, and cerebellar and pyramidal features in different combinations (Fanciulli and Wenning, 2015). It is classified as a parkinsonian subtype when parkinsonism is the predominant symptom, as a cerebellar subtype when cerebellar features predominate, and as Shy-Drager syndrome when the autonomic features are dominant (Fanciulli and Wenning, 2015). PSP is a neurodegenerative disease characterized by loss of voluntary control of eye movements, axial rigidity, postural instability, bradykinesia, and subcortical dementia (Stamelou et al., 2019a,b). As mentioned before, all three neurologic disorders invariably display the cardinal parkinsonian motor dysfunctions. In fact, MSA and PSP patients are usually misdiagnosed as PD patients, especially in the early stages of the disease or in cases showing high clinical variability. Patients afflicted with PD, MSA, or PSP also suffer from a wide range of nonmotor symptoms that worsen over disease progression. Alterations of circadian rhythms

THE CIRCADIAN SYSTEM IN PARKINSON'S DISEASE are among the most debilitating of these nonmotor symptoms. These circadian dysfunctions can be classified into three categories: behavioral, physiologic, and molecular alterations.

Parkinson’s disease Relative to MSA and PSP, PD is the most investigated disorder in terms of circadian rhythm disorders. The most manifest circadian alteration at the behavioral level in PD patients is the progressive deterioration of qualitative and quantitative aspects of sleep/wake cycles (Gros and Videnovic, 2020). These sleep alterations consist mainly of insomnia (experienced as difficulty in falling asleep and maintaining normal uninterrupted periods of sleep), excessive daytime sleepiness (EDS), fragmentation of sleep/wake cycle, and rapid eye movement (REM) sleep behavioral disorder (RBD). Worsening of sleep symptoms during disease advancement can lead to a dramatic reduction in the amplitude of rest/activity cycles (Gros and Videnovic, 2020). In advanced stages, the degree of sleep/wake alterations becomes so severe that patients may display quasiarrhythmia of their rest/activity cycles (Niwa et al., 2011). Like other nonmotor symptoms of PD (such as constipation, olfactory deficit, and depression), strong evidence currently suggests that some sleep alterations in PD patients, namely EDS and RBD, predate the emergence of the cardinal motor symptoms by 10–15 years (Maetzler et al., 2009). This clinical feature has prompted the still nascent idea of using these early symptoms as biomarkers in a battery of behavioral and physiologic tests to identify people with high risk of developing PD. The development of such prodromal markers should benefit new therapeutic interventions aiming to prevent and/or delay the progression of PD pathophysiology (Chen-Plotkin, 2014). The main limitation of this goal, however, is the high degree of variability in the overall motor and nonmotor symptoms experienced by PD patients. Another limitation is the fact that many of the potential biomarkers are not exclusively found in PD patients (Chen-Plotkin, 2014). In fact, patients afflicted with other neurodegenerative disorders also display several of these biomarkers as part of their clinical diagnosis (Chen-Plotkin, 2014). The temporal organization and quality of sleep/ wake behavior are tightly regulated by, respectively, the circadian clock and the sleep-homeostatic process (Achermann and Borbely, 2011). Alterations of sleep/ wake behavior are precipitated from dysfunctions in one or both processes. In PD patients, several neural centers involved in sleep regulation show either signs of neurodegeneration or abnormal deposition of Lewy body pathology (Fifel et al., 2016). As discussed in detail

303

in the following paragraphs, the involvement of circadian clock dysfunction is well-supported by our current knowledge of PD pathophysiology (Fifel, 2017; Fifel and Videnovic, 2018, 2019). Next to changes in sleep/wake behavior, the second aspect of circadian alterations in PD patients is evident from the circadian alteration of several physiologic parameters. These dysfunctions include a reduced amplitude of the 24-h rhythm of cortisol secretion (Hartmann et al., 1997), reversal or even a full arrhythmia of blood pressure and heart rate variability (Ejaz et al., 2006; Stuebner et al., 2013; Vetrano et al., 2015), dampening or reversal of the daily pattern of urine excretion (Hineno et al., 1994), and an impaired body temperature rhythm (Suzuki et al., 2007). In addition, the daily cycle of melatonin secretion is affected (Bordet et al., 2003; Bolitho et al., 2014; Breen et al., 2014; Videnovic et al., 2014a,b). The melatonin cycle is considered a reliable marker of the endogenous central clock because the SCN directly controls the rhythmic pattern of melatonin secretion through a polysynaptic connection (Arendt and Skene, 2005). In PD patients, both its amplitude and phase of entrainment relative to the light–dark (LD) cycle are altered (Bordet et al., 2003; Bolitho et al., 2014; Breen et al., 2014; Fifel and DeBoer, 2014; Videnovic et al., 2014a,b). Recently, these melatonin rhythm alterations have been linked to the deterioration of several nonmotor symptoms, including cardiovascular symptoms, sleep disorders, and gastrointestinal dysfunctions (Li et al., 2020). The most recent body of evidence linking dysfunctional circadian timing with PD pathology comes from molecular studies. By investigating the 24-h expression of some clock genes in peripheral tissues (i.e., blood mononuclear cells) sampled from PD patients, these studies revealed either a dampening (Cai et al., 2010) or a complete loss of the rhythmic expression of BMAL1, a core clock gene (Breen et al., 2014). The circadian expression of other clock genes (PER2 and REV-ERBa) was also affected (Ding et al., 2011; Breen et al., 2014). Taken together, there is now a growing body of research that confidently implicates a dysfunctional circadian system in PD pathology. These alterations causally impact several health parameters that contribute to the deterioration of the quality of life in PD patients (Fifel, 2017).

Multiple system atrophy As with PD patients, sleep disorders are common clinical manifestations in patients with MSA (Ferini-Strambi and Marelli, 2012). They also exteriorize as fragmented sleep/wake cycle, EDS, insomnia, and RBD (FeriniStrambi and Marelli, 2012). RBD is the most common

304

K. FIFEL AND T. DE BOER

symptom, affecting up to 90%–100% of patients with MSA, and is considered, as in PD patients, to be a prodromal sign of MSA (Ferini-Strambi and Marelli, 2012). In addition to these sleep alterations, and because of the autonomic failure characterizing MSA, most of the investigated circadian alterations in MSA are linked with autonomic nervous system dysfunctions. The best characterized of these are the cardiovascular autonomic neuropathologies. Using 24-h ambulatory blood pressure and heart-rate monitoring, several studies have described disruptions of circadian variations of blood pressure, characterized by the loss of the nocturnal decline of blood pressure (nondipping pattern) or even blood pressure nocturnal increase (reverse dipping pattern) (Fanciulli et al., 2014; Pilleri et al., 2014; Indelicato et al., 2015; Franzen et al., 2017; Vichayanrat et al., 2017; Vallelonga et al., 2019). Pilleri et al. (2014) have shown that the circadian profile of heart rate can discriminate between MSA and PD patients. However, two recent studies have failed to replicate these results and showed similar circadian blood pressure patterns in both MSA and PD patients (Vichayanrat et al., 2017; Vallelonga et al., 2019). The 24-h rhythm of core body temperature is also affected in MSA patients (Pierangeli et al., 2001). As expected, these body temperature alterations were worse in MSA compared to PD patients, as evidenced by significantly lower amplitudes (Pierangeli et al., 2001). Constipation is another nonmotor symptom in both PD and MSA patients (Suzuki et al., 2005). This disorder is also linked with the autonomic dysfunctions intrinsic to the diseases. By using noninvasive cutaneous electrogastrography (EGC), Suzuki et al. measured the circadian pattern of gastric myoelectrical activity (Suzuki et al., 2005). In healthy subjects, gastric mobility follows a circadian rhythm that matches the rhythm of food intake, with increased motility during the daytime, which subsequently wanes and decreases during the sleep phase. In MSA patients, however, this circadian variation was severely blunted, indicating that the temporal dimension of gastric mobility is also impaired in MSA patients (Suzuki et al., 2005). Nocturnal polyuria is another autonomic dysfunction in patients with MSA (Fanciulli and Wenning, 2015). As discussed in the next section, the underlying pathophysiology of nocturia involves failure of both peripheral organs and central control. In healthy individuals, urine excretion follows a rhythmic pattern, with maximal urine flow in the early afternoon and minimal flow at midnight (Hineno et al., 1994). In contrast, MSA patients excrete a large volume of their urine at night (Ozawa et al., 1993; Sakakibara et al., 2003). Consequently, the increased frequency of nocturnal urination contributes significantly to the fragmentation and worsening of sleep.

Other, more reliable physiologic markers of the endogenous circadian clock, such as melatonin rhythm or clock gene expression, have not been investigated in patients with MSA. Only a single study of 20 MSA patients measured the profile of plasma cortisol concentrations and showed that morning cortisol concentrations were significantly lower in patients with MSA relative to controls (Ozawa et al., 2001). Although all these studies suggest altered circadian rhythmicity in MSA patients, they should be interpreted with caution, given the small sample sizes and study designs that reflect influences of both the internal circadian clock and external entraining factors (such as light/dark cycles, physical activity, food intake, and social interactions) on health parameters of MSA patients. These exogenous factors might mask the true nature of circadian insults in patients (Johnson et al., 2003). Additionally, given the very few markers used to investigate circadian alterations in MSA patients, additional investigations using several behavioral, physiologic, and molecular markers are required to delineate a comprehensive picture of the extent of circadian desynchrony in MSA patients.

Progressive supranuclear palsy Unfortunately, very few studies have attempted to investigate circadian rhythms in patients with PSP. This lack of data could be attributed to the relatively low incidence of PSP in the general population, as well as the intrinsic methodologic challenges of putting patients who exhibit debilitating motor, cognitive, and psychiatric symptoms through intense and long protocols for circadian rhythm and sleep research. Additionally, sleep problems in PSP tend to go undiagnosed by physicians and are underreported by patients as a result of a perceived lower burden of the sleep problems relative to the motor and cognitive features that are recognized as cardinal symptoms of this disorder. Despite these limitations, available clinical studies indicate that PSP patients suffer from serious sleep and circadian alterations (Stamelou et al., 2019a,b). The first objective sleep study using polysomnography in 6 PSP patients was published in 1997 (Montplaisir et al., 1997). This study described several qualitative and quantitative insults to sleep/wake behavior that were exteriorized as decreased sleep efficiency, reduced amount of REM sleep as evidenced by a significant decrease of both episode number and duration, decreased REM density, a drastic reduction in sleep spindles, and a general slowing of frontal EEG during wakefulness. Muscle atonia and phasic EMG during REM sleep were not affected. This last feature could explain the relatively low incidence of RBD in PSP patients relative to PD patients reported by Nomura et al. (2012). However, other polysomnography studies,

THE CIRCADIAN SYSTEM IN PARKINSON'S DISEASE while confirming most of these sleep alterations, failed to replicate the low incidence of RBD in PSP relative to PD patients (Arnulf et al., 2005; Sixel-D€ oring et al., 2009; Walsh et al., 2017). These differences could reflect variable neuropathologic profiles among PSP patients (Stamelou et al., 2019a,b). Actigraphy is another reliable technique that can be used to assess characteristics of the rest/activity cycle (Fekedulegn et al., 2020). The only study that used actigraphy to specifically investigate circadian features of rest/activity rhythms in PSP patients found evidence of weak circadian rhythms as evidenced by a decreased amplitude and mesor of the rest/activity rhythm (Walsh et al., 2016). By using nonparametric circadian rhythm analysis, the authors also found increased intradaily variability, which reflects increased fragmentation of the rest/activity cycle. PSP patients also displayed irregular rest/activity cycles as evidenced by decreased interdaily stability (Walsh et al., 2016). All of these alterations were associated with increasing disease severity. The only other circadian physiologic overt rhythms investigated in PSP patients are cardiovascular functions and core body temperature (Schmidt et al., 2009; Suzuki et al., 2009). Both blood pressure and heart rate change in a circadian fashion with periodic drops in blood pressure at night. Schmidt et al. showed that 40% of PSP patients did not have this nocturnal drop in blood pressure (Schmidt et al., 2009). Similarly, the amplitude of core body temperature was reduced, and the nocturnal drop was smaller in PSP relative to PD patients (Suzuki et al., 2009). Collectively, all these studies show that circadian disturbances occur in PSP patients. As in MSA, however, the paucity of circadian overt rhythms investigated so far precludes any confident conclusion regarding the real extent of circadian abnormalities in PSP patients.

NEUROPATHOLOGIC AND FUNCTIONAL DISSECTION OF THE NEURONAL CIRCUITRY GOVERNING CIRCADIAN RHYTHMS IN PD, MSA, AND PSP A key discovery in the field of circadian biology was the widespread expression of clock genes in almost all cells of the body (Zhang et al., 2014). This discovery brought into prominence the concept of hierarchy and synchronization between organismal circadian clocks (Dibner et al., 2010; Turek, 2016). Indeed, the critical role of this hierarchic synchronization in health is now well established and the breakdown of the overall 24-h temporal order has been associated with many pathologic conditions (Hastings et al., 2003; Bass and Lazar, 2016). According to this concept, the clock in the suprachiasmatic nucleus (SCN) is regarded as the main driver and coordinator, rather than a generator, of peripheral

305

clocks (Hastings et al., 2003; Dibner et al., 2010). To achieve this, the SCN uses both neuronal and humoral signals to imprint a rhythmic temporal signal on many central and peripheral tissues (Hastings et al., 2003; Bass and Lazar, 2016). These output pathways of the SCN are responsible for the generation of an optimal and wellcoordinated 24-h temporal order between all peripheral organs (Hastings et al., 2003; Bass and Lazar, 2016). Although the SCN continues to generate highly precise circadian rhythms when isolated from all external cues (i.e., constant conditions), under normal conditions where environmental stimuli such as the light/dark cycles alternate periodically, the clock in the SCN is continuously reset so that its period matches the period of the external driving factor (called a zeitgeber, like the light/dark cycles). This feature of the clock is termed “entrainment” and its impairment can also yield pathologic conditions (Johnson et al., 2003). By far the most potent entraining factor for the central SCN clock is the light/dark cycle. The SCN, among other brain structures, receives environmental photic information via a photic nonvisual neural pathway from the retina, called the retinal hypothalamic tract. This pathway originates from a subset of intrinsically photosensitive retinal ganglion cells (ipRGCs) that uses melanopsin as a photopigment to sense the irradiance level in the surrounding environment. In addition to ipRGCs, studies have implicated rods and cones in circadian photoentrainment as well (Altimus et al., 2008; Guler et al., 2008). However, ipRGCs remain the principal neuronal route by which rods and cones influence the nonimage-forming properties of the retina (Guler et al., 2008). As discussed in following text, impairments at the level of the eye as a result of either normal aging or neuropathologic processes related to neurodegenerative disease contribute as an important causal factor in the weakening of the circadian system. Collectively, this network organization of the circadian system (Hastings et al., 2003) implies that circadian dysfunctions may emerge as a consequence of a dysfunction in one or multiple hubs of this network, in their connecting pathways, or in the combination of some or all of these. In this section, we dissect the neuropathologic status of the different parts of the circadian network in PD, MSA, and PSP and follow its evolution over disease progression. A recent review detailed such an investigation in PD (Fifel, 2017). Consequently, only a summary and an update will be provided here for PD.

PATHOLOGIC STATUS OF THE AFFERENT PATHWAYS TO THE CLOCK The alternation of the light/dark cycles remains the most potent zeitgeber for the circadian system (Friborg et al., 2014; LeGates et al., 2014). Age-related as well as

306

K. FIFEL AND T. DE BOER

neurodegenerative-related impairments in the visual system are therefore expected to contribute to the deterioration in circadian rhythms. Aging is indeed associated with several pathologies in different components of the eye (Esquiva et al., 2017; Lax et al., 2019). These dysfunctions include an increase in crystalline lens light absorption, a decrease in pupil area that causes a decreased retinal illumination (Turner and Mainster, 2008), and a progressive decline in the capacity of the lens to transmit short-wavelength light (Lutze and Bresnick, 1991) that is critical for the optimal stimulation of retinal ipRGCs (Lax et al., 2019). Recently, an agerelated decline in the density of ipRGCs was found (Esquiva et al., 2017). Importantly, this study also reported a significant atrophy in the dendritic arborization in the remaining ipRGCs. From 50 years old onwards, the complexity of the ipRGCs plexus decreased dramatically, and after the age of 70 dendrites showed little overlapping with few contacts between the few remaining ipRGCs (Esquiva et al., 2017). In general, these age-related impairments are exacerbated in neurodegenerative diseases. In PD, neuropathologic deposition of toxic misfolded proteins has been found in retinas obtained postmortem from PD patients (Beach et al., 2014). Furthermore, recently a 33% reduction in the density and plexus complexity of ipRGCs was found in PD patients compared to age-matched controls (Ortuño-Lizarán et al., 2018a,b). At the functional level, and in line with these histopathologic findings, the pupillary light reflex—a nonimage-forming autonomic function mediated by ipRGCs—is impaired in PD patients (Joyce et al., 2018). The neuropathologic status as well as the functional integrity of the eye in MSA and PSP are still poorly understood. A recent meta-analysis showed that retinal abnormalities in patients with MSA show common features distinct from PD patients (Mendoza-Santiesteban et al., 2017a). In MSA patients, no retinal deposits of a-syn have been identified despite a severe reduction in retinal ganglion cell density (Mendoza-Santiesteban et al., 2017b). In PSP, the thickness of the retinal nerve fiber layer is significantly reduced (Albrecht et al., 2012; Beach et al., 2014; Schneider et al., 2014; Stemplewitz et al., 2017; Gulmez Sevim et al., 2018). This retinal degeneration was independent of disease duration or severity. These retinal ganglion abnormalities in MSA and PSP patients may arise as a consequence of the degeneration of ganglion cells belonging to the visual pathway of the retina. Future anatomic and functional studies investigating whether ipRGCs are also affected are needed. Collectively, all these studies suggest that retinal impairments are potential causal factors in the alterations of circadian rhythms by contributing to the weakening of photoentrainment in patients.

In addition to light/dark cycles, the central and peripheral clocks are also entrained by several nonphotic zeitgebers (Hastings et al., 2003; Takahashi et al., 2008; Bass and Lazar, 2016). These include sleep/wake cycles (Borbely et al., 2016), rest/exercise cycles (Gabriel and Zierath, 2019), feeding/fasting rhythms (Panda, 2016), and body temperature cycles (Hastings et al., 2003; Takahashi et al., 2008). Given that the circadian pattern of all these synchronizing factors is fragmented in patients with PD, MSA, and PSP, their alterations are expected to contribute to the weakening and progressive worsening of the circadian system in patients. The mechanism(s) employed by these secondary zeitgebers to negatively feed back to the circadian system is still not completely understood. A direct impact on the central clock (in the case of sleep/wake cycles and rest/activity cycles (Borbely et al., 2016; Gabriel and Zierath, 2019)) and/or on the peripheral clocks (in the case of the other zeitgebers (Hastings et al., 2003)) is possible. Impairments of the autonomic nervous system in neurodegenerative diseases might underlie at least some misalignments and amplitude reduction in circadian rhythms of peripheral organs (Hastings et al., 2003). Although still not thoroughly investigated, this mechanism is more pertinent in patients with MSA, where the autonomic dysfunctions prevail (Fanciulli and Wenning, 2015). Mechanistically, the contribution of impaired autonomic control to circadian abnormalities is supported by studies revealing the potential of the autonomic nervous system in driving or resetting peripheral oscillators. For example, sympathetic denervation of the pineal gland or liver in vivo abolishes local circadian oscillation of Per gene expression (Takekida et al., 2000; Terazono et al., 2003), which can be restored using adrenergic agonists (Takekida et al., 2000; Terazono et al., 2003). Future studies designed to investigate the mechanisms and the extent of alterations of nonphotic entraining cues are of fundamental importance in order to develop efficient therapeutic and management strategies for patients with neurodegenerative diseases.

PATHOLOGIC STATUS OF THE CENTRAL SCN CLOCK Is a dysfunctional SCN-central clock involved in the pathophysiology of circadian alterations experienced by patients with PD, MSA, and PSP? And if so, when, over the course of the disease, does the SCN clock go awry? A prominent feature of the master clock in the SCN is the fact that the generation of the circadian time is not an emergent property of functional interactions within the SCN neuronal network. Rather, every SCN neuron is capable of generating its own, self-sustained circadian rhythm. To generate a coherent circadian

THE CIRCADIAN SYSTEM IN PARKINSON'S DISEASE output at the organismal level, the SCN uses several synchronizing signals such as GABA, arginine vasopressin (AVP), vasointestinal polypeptide (VIP), transforming growth factor alpha (TGFa), and prokineticin-2 (PK2) (Hastings et al., 2018). Using these molecules as markers of SCN function, a few histopathologic studies have examined the integrity of the SCN using postmortem brain samples obtained from patients with a history of a neurodegenerative disorder. In addition, SCN (dys-) function may be inferred from the extent of the pathologic depositions within SCN neurons of the hallmark cytoplasmic inclusions that characterize neurodegenerative diseases (i.e., a-syn immunoreactive aggregates in the case of PD, MSA, and PSP). In PD, these depositions aggregate to form distinct structures called Lewy neurites and Lewy bodies. An early histopathologic study failed to detect Lewy neurites and Lewy bodies in the SCN, although several hypothalamic structures were affected (Langston and Forno, 1978). More recently, however, De Pablo-Fernandez et al. examined 13 SCNs from brains of patients deceased at advanced stages of PD (mean disease duration ¼ 14.3 years). In the majority of patients (9/13 cases), the authors found abnormal Lewy body inclusion with mild to moderate severity in the SCN (De Pablo-Fernández et al., 2017; Fifel, 2019). The same study by De Pablo-Fernandez et al. also examined SCNs from MSA (n ¼ 5) and PSP (n ¼ 5) patients (De Pablo-Fernández et al., 2017). The authors found no neuropathologic changes in SCNs of MSA patients while all SCNs from PSP patients showed pathologic tau-inclusions (De Pablo-Fernández et al., 2017). Given the small sample size, the negative results in the MSA cases should be interpreted cautiously. In fact, two earlier studies reported significant functional impairments within the SCN of MSA patients (Ozawa et al., 1998; Benarroch et al., 2006). In a case report, Ozawa et al. (1998) showed a significant decrease in the number of AVP-positive neurons in the SCN of a patient with MSA. Neural atrophy and signs of gliosis were also present in the SCN (Ozawa et al., 1998). These neuropathologic findings were confirmed by Benarroch et al. (2006). The authors of these studies linked these pathologies to abnormal circadian fluctuations of plasma AVP, and nocturnal polyuria reported in MSA (Ozawa et al., 1993). However, given that these neurodegenerative processes occur in the central SCN clock, they may contribute to a wider impairment of circadian regulation of endocrine and autonomic functions in MSA. In summary, although reports regarding the functional integrity of the central clock in PD, MSA, and PSP patients are still rare, there is convincing evidence implicating a dysfunctional SCN clock as a significant causal factor in precipitating circadian alterations in patients suffering from these neurodegenerative disorders.

307

When, over the disease progression, is the SCN impaired? The answer to this question is crucial if we want to develop efficient, specific, and mechanismbased therapies. Additionally, this kind of knowledge will inform as to when to administer these therapies. As mentioned earlier, histopathologic studies cannot provide information about the time of SCN dysfunction because they usually involve brain samples obtained from patients that died at late stages of the disease. Prospective studies that follow changes in overt circadian rhythms over the progression of neurodegenerative diseases are lacking. Alternatively, cross-sectional studies that examine one or more circadian outputs over different stages of the disease could inform us about the functional evolution of the SCN clock. Currently, however, such studies are absent in the case of MSA and PSP. We will therefore investigate this question in PD, for which a few studies have been published. To track the start and progression of circadian clock impairment over time in patients, assessing circadian functioning during prodromal phases would be ideal. Currently, however, this approach is impossible for PD, given the absence of reliable preclinical markers of the disease (Chen-Plotkin, 2014). Nonetheless, insights from studies comparing circadian profiles in de novo vs. advanced PD patients are instrumental in overcoming this limitation. In a few studies, the 24-h circadian profile of several biologic rhythms (i.e., melatonin, growth hormone, thyrotropin, prolactin, leptin) in de novo, drug-naïve PD patients was examined. No significant alterations in either entrainment or amplitude were found (Aziz et al., 2009a,b; Fifel, 2017). These findings suggest that up to clinical diagnosis, the central SCN clock is still able to generate normal circadian rhythms. Following diagnosis, several factors such as L-DOPA intake and the emergence and deterioration of motor and several nonmotor symptoms will negatively impact clock function and precipitate circadian rhythm alterations (Fifel, 2017). In PD, therefore, it seems that circadian alterations are not caused by the intrinsic neurodegenerative processes driving the disease but are rather emerging secondary to health deterioration following diagnosis. Similar studies should be conducted for MSA and PSP patients to assess the relative contribution of central clock dysfunction to the circadian alterations experienced by these patients.

PATHOLOGIC STATUS OF THE EFFERENT PATHWAYS OF THE CLOCK The central clock in the SCN integrates photic and several nonphotic entraining inputs in order to optimally orchestrate the circadian timing of physiology and behavior (Saper et al., 2005; Dibner et al., 2010).

308

K. FIFEL AND T. DE BOER

This distribution of the circadian signal from the SCN to the rest of the organism is mediated by both humoral signals and dense neural projections to several central and peripheral tissues (Dibner et al., 2010). Except for a few hypothalamic nuclei that receive direct SCN inputs, most output structures integrate SCN information through polysynaptic connections from a few hypothalamic relay centers (Dai et al., 1997, 1998a; Saper et al., 2005). Given the widespread deposition of toxic aggregates in the brains of PD, MSA, and PSP patients, and their well-established negative impact on neuronal activity and neural network dynamics (Fanciulli and Wenning, 2015; Poewe et al., 2017; Stamelou et al., 2019a,b), all affected areas could potentially display a dysfunctional circadian clock and hence lead to abnormal circadian phenotype of their corresponding circadian overt rhythm. The paucity of available studies does not permit an exhaustive examination of the functional characteristics of the circadian clock in all affected SCN-output structures in PD, MSA, and PSP. We therefore highlight a few examples. Several brain areas downstream of the SCN are involved in the regulation of sleep/wake behavior (Saper et al., 2005). The dorsomedial hypothalamus (DMH) is the primary hypothalamic nucleus that replays SCN signals to several sleep/wake centers in order to generate a circadian sleep/wake behavior (Dai et al., 1998b; Saper et al., 2005). The SCN inputs to the DMH are relayed through the ventral subparaventricular zone (vSPZ), which sends dense glutamatergic innervation to the lateral hypothalamus (Saper, 2013). These excitatory inputs form direct synaptic contacts with orexinergic neurons in LH (Abrahamson et al., 2001). Though sparse, direct projections from the SCN to LH have also been reported (Abrahamson et al., 2001). In all three neurodegenerative disorders covered by this review, orexinergic neurons undergo progressive degeneration over the course of the disease (Ozawa, 2007; Hauw et al., 2011; Fifel et al., 2014, 2016). Given the role of orexin neurons in the consolidation of wake and sleep behavior in day and night, respectively (Sakurai, 2007), their loss is expected to account, at least partially, for the fragmentation of sleep/wake behavior seen in PD, MSA, and PSP patients. Several other key wake- or sleep-promoting centers that have direct or indirect polysynaptic connections with the SCN show signs of neurodegeneration and/or abnormal neuropathologic depositions (Ozawa, 2007; Hauw et al., 2011; Fifel et al., 2016). The cumulative impact of dysfunctions in these centers underlies the qualitative and quantitative alterations of sleep/wake behavior in PD, MSA, and PSP patients. Another output structure that shows dysfunctional circadian timing in neurodegenerative disease is the

pineal gland. The pineal gland is connected to the SCN via a polysynaptic pathway that drives the nocturnal synthesis and release of melatonin in the blood circulation (Arendt and Skene, 2005). As mentioned earlier, evidence of circadian alterations of melatonin release is clear, at least in PD (Fifel, 2017). Therefore, alterations in clock gene expression in the pineal of PD patients might account for the dysfunctional endogenous circadian clock in pinealocytes and/or a dysfunctional connection between these and the central SCN clock. In PD, pathologic depositions of Lewy bodies and Lewy neurites have also been recently found in the pineal gland (De Pablo-Fernández et al., 2017). Additionally, L-DOPA intake (the standard firstoption treatment of parkinsonism in PD) is known to increase the dopamine content of the pineal in both experimental rats (Hyyppa et al., 1971) and PD patients (Ghaemi et al., 2001). This increase was positively correlated with the severity of the disease (Ghaemi et al., 2001), suggesting a pathologic link. Although we currently don’t know whether melatonin rhythm is also affected in MSA and PSP patients, neuropathologic deposition of glial and tau-immunopositive cytoplasmic aggregate in the pineal glands of, respectively, MSA and PSP patients have been documented (De Pablo-Fernández et al., 2017). Collectively, these findings suggest a potential causal role of a dysfunctional pineal gland in the circadian alterations in patients with PD, MSA, and PSP. Direct evidence for this is, however, still missing. Future studies should aim to correlate the in vivo patterns of melatonin release in patients with postmortem profiles of clock gene expression in the pineal gland. Such analysis has conclusively implicated the endogenous circadian clock of the pineal gland in the pathophysiology of circadian rhythms in Alzheimer’s disease (Wu et al., 2006; Fifel and Videnovic, 2020). As we discussed previously, the failure of the autonomic nervous system is a prominent clinical feature that might precede motor symptoms in MSA patients (Ozawa, 2007). This failure encompasses several vital functions such as blood pressure regulation, body temperature, respiration, glycemia, stress responses, and regulation of body fluid volume (Ozawa, 2007; Fanciulli and Wenning, 2015). The circadian regulation of many of these autonomic responses is impaired (see earlier). Unsurprisingly, and given its vital role, the autonomic system is under tight control by the SCN (Buijs and Kalsbeek, 2001; Hastings et al., 2003). This regulation is mediated by both AVP and VIP neurons of the SCN. Polysynaptic connections through the paraventricular nucleus (PVN) of the hypothalamus carry circadian output signals to the adrenocorticotropic axis and to autonomic ganglia that innervate the viscera (Buijs and Kalsbeek, 2001). Several components of this network are affected in MSA patients. AVP neurons in the

THE CIRCADIAN SYSTEM IN PARKINSON'S DISEASE posterior subnucleus of the PVN show significant degeneration in MSA patients (Benarroch et al., 2006). Additionally, the functional interactions between the SCN and PVN are impaired as a result of the loss of AVP containing GABAergic neurons in the SCN (Benarroch et al., 2006). This SCN-PVN pathway is responsible for the circadian regulation of sympathetic and parasympathetic outputs to peripheral organs (Kreier et al., 2006). In addition to these alterations, several other circadian output pathways and structures are affected in MSA patients (Ozawa, 2007). These include the intermediolateral column of the spinal cord, the ventrolateral portion of the intermediate reticular formation, the dorsal motor nucleus of the nervous vagus, ambiguous nucleus, raphe nuclei, arcuate nucleus, locus coeruleus, sympathetic innervation of the heart, histamine neurons in the tuberomamillary nucleus, catecholamine neurons in the hypothalamus, and the growth hormone release by the anterior pituitary gland (Ozawa, 2007). Collectively, all these pathologies at the level of SCN outputs contribute to an abnormal imprinting of SCN signals and hence could explain the wide range of circadian alterations of the autonomic nervous system in MSA patients. Another pathologic mechanism that contributes to the exacerbation of circadian alterations over disease progression in patients with neurodegenerative diseases is the negative feedback of dysfunctions in the peripheral physiology to the central clock. Indeed, several outputs of the clock play a mediating role between the SCN and the periphery in order to coordinate and insure an optimal circadian typing at the organismal level (Hastings et al., 2003; Dibner et al., 2010). Prime examples of these messenger outputs are sleep/wake cycles, body temperature, and melatonin. These overt physiologic rhythms are of paramount importance in orchestrating circadian physiology in the periphery (Hastings et al., 2003; Dibner et al., 2010). These dysfunctions will therefore precipitate self-reinforcing vicious cycles leading to the rapid deterioration of both motor and nonmotor symptoms in patients with neurodegenerative diseases. The ultimate consequence of these vicious cycles is the induction of a state of general circadian desynchrony in which several physiologic parameters operate outside of the optimal phase relationship to each other. Evidence for this negative feedback is strong in the case of PD patients (Fifel, 2017) and in other neurodegenerative diseases (i.e., Alzheimer’s disease) (Fifel and Videnovic, 2020). Again, in the case of MSA and PSP patients, our current knowledge precludes any reliable conclusion. Future studies should investigate the pathophysiology of disease progression in MSA and PSP patients and sort out the interconnectedness of different circadian symptoms in driving the progressive worsening of health parameters and the quality of life of patients in general.

309

CONCLUSIONS In summary, we have shown in this review that, in addition to their cardinal clinical symptoms, patients with PD, MSA, and PSP suffer from a wide range of debilitating circadian disorders. These disorders dominate the clinical picture of patients in their advanced stages and will inevitably contribute to the deterioration of the quality of life of both patients and their caregivers. Although the pathophysiology of these circadian alterations is still not fully understood, it evidently involves a complex interaction between the intrinsic neuropathology and secondary negative feedback of a plethora of motor and nonmotor symptoms to the molecular clock. By dissecting the neuropathology and functional status of the neural circuitry controlling circadian rhythms, we have shown that the functioning of the central SCN clock becomes impaired—and worsens over disease progression—after the emergence of motor symptoms in PD (Fifel, 2017). In MSA and PSP patients, histopathologic evidence also implicates the SCN as a potential cause of circadian alterations experienced by these patients. However, when in the course of disease progression the SCN goes awry is unknown. Future studies using in vivo imaging techniques should examine in detail the functional integrity of the different nodes of the neuronal network underlying the generation of circadian rhythms. Regardless of the intricacy of the neuropathophysiology, a common clinical outcome in patients with PD, MSA, and PSP is a mistiming of several central and peripheral biologic rhythms. This state of desynchrony has a major deleterious impact on global health, even compared to the consequence of a dysfunctional SCN clock (Fernandez et al., 2014). Attempts to mitigate the impact of circadian dysfunctions in patients with neurodegenerative disease should therefore be at the forefront of our clinical interests. Currently, pharmacologic approaches aimed at alleviating circadian alterations are still limited (Schroeder and Colwell, 2013). Recently, chronobiologic applications of bright-light exposure have shown promising results in improving the quality of life in PD patients (Fifel and Videnovic, 2018, 2019). Further development of mechanisms-based protocols of light therapy holds promising benefits for the wide range of motor and nonmotor symptoms in parkinsonism-related neurologic disorders (Fifel and Videnovic, 2018, 2019). (See Chapter 4 in Volume 182.)

REFERENCES Abrahamson EE, Leak RK, Moore RY (2001). The suprachiasmatic nucleus projects to posterior hypothalamic arousal systems. Neuroreport 12: 435–440. Achermann P, Borbely AA (2011). Sleep homeostasis and models of sleep regulation. In: MH Kryger, T Roth,

310

K. FIFEL AND T. DE BOER

WC Dement (Eds.), Principles and practice of sleep medicine, fifth edn. Elsevier Saunders, Philadelphia, pp. 431–444. Albrecht U (2012). Timing to perfection: the biology of central and peripheral circadian clocks. Neuron 74: 246–260. Albrecht P, M€uller AK, S€udmeyer M et al. (2012). Optical coherence tomography in parkinsonian syndromes. PLoS One 7: e34891. Altimus CM, Guler AD, Villa KL et al. (2008). Rods-cones and melanopsin detect light and dark to modulate sleep independent of image formation. Proc Natl Acad Sci U S A 105: 19998–20003. Arendt J, Skene DJ (2005). Melatonin as a chronobiotic. Sleep Med Rev 9: 25–39. Arnulf I, Merino-Andreu M, Bloch F et al. (2005). REM sleep behavior disorder and REM sleep without atonia in patients with progressive supranuclear palsy. Sleep 28: 349–354. Aziz NA, Pijl H, Fr€olich M et al. (2009a). Increased hypothalamic-pituitary-adrenal axis activity in Huntington’s disease. J Clin Endocrinol Metab 94: 1223–1228. Aziz NA, Pijl H, Fr€olich M et al. (2009b). Delayed onset of the diurnal melatonin rise in patients with Huntington’s disease. J Neurol 256: 1961–1965. Bass J, Lazar MA (2016). Circadian time signatures of fitness and disease. Science 354: 994–999. Beach TG, Carew J, Serrano G et al. (2014). Phosphorylated a-synuclein-immunoreactive retinal neuronal elements in Parkinson’s disease subjects. Neurosci Lett 571: 34–38. Benarroch EE, Schmeichel AM, Sandroni P et al. (2006). Differential involvement of hypothalamic vasopressin neurons in multiple system atrophy. Brain 129: 2688–2696. Bolitho SJ, Naismith SL, Rajaratnam SM et al. (2014). Disturbances in melatonin secretion and circadian sleepwake regulation in Parkinson disease. Sleep Med 15: 342–347. Borbely AA, Daan S, Wirz-Justice A et al. (2016). The twoprocess model of sleep regulation: a reappraisal. J Sleep Res 25: 131–143. Bordet R, Devos D, Brique S et al. (2003). Study of circadian melatonin secretion pattern at different stages of Parkinson’s disease. Clin Neuropharmacol 26: 65–72. Breen DP, Vuono R, Nawarathna U et al. (2014). Sleep and circadian rhythm regulation in early Parkinson disease. JAMA Neurol 71: 589–595. Buijs RM, Kalsbeek A (2001). Hypothalamic integration of central and peripheral clocks. Nat Rev Neurosci 2: 521–526. Cai Y, Liu S, Sothern RB et al. (2010). Expression of clock genes Per1 and Bmal1 in total leukocytes in health and Parkinson’s disease. Eur J Neurol 17: 550–554. Canter RG, Penney J, Tsai LH (2016). The road to restoring neural circuits for the treatment of Alzheimer’s disease. Nature 539: 187–196. Chen-Plotkin AS (2014). Unbiased approaches to biomarker discovery in neurodegenerative diseases. Neuron 84: 594–607. Coogan AN, Schutova´ B, Husung S et al. (2013). The circadian system in Alzheimer’s disease: disturbances, mechanisms, and opportunities. Biol Psychiatry 74: 333–339. 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. (1998a). Postmortem tracing reveals the organization of hypothalamic projections of the suprachiasmatic nucleus in the human brain. J Comp Neurol 400: 87–102. Dai J, Van Der Vliet J, Swaab DF et al. (1998b). Postmortem anterograde tracing of intrahypothalamic projections of the human dorsomedial nucleus of the hypothalamus. J Comp Neurol 401: 16–33. 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. Dibner C, Schibler U, Albrecht U (2010). The mammalian circadian timing system: organization and coordination of central and peripheral clocks. Annu Rev Physiol 72: 517–549. Ding H, Liu S, Yuan Y et al. (2011). Decreased expression of Bmal2 in patients with Parkinson’s disease. Neurosci Lett 499: 186–188. Ejaz AA, Sekhon IS, Munjal S (2006). Characteristic findings on 24-h ambulatory blood pressure monitoring in a series of patients with Parkinson’s disease. Eur J Intern Med 17: 417–420. Esquiva G, Lax P, Perez-Santonja JJ et al. (2017). Loss of melanopsin-expressing ganglion cell subtypes and dendritic degeneration in the aging human retina. Front Aging Neurosci 9: 79. Fanciulli A, Wenning GK (2015). Multiple-system atrophy. N Engl J Med 372: 249–263. Fanciulli A, Strano S, Ndayisaba JP et al. (2014). Detecting nocturnal hypertension in Parkinson’s disease and multiple system atrophy: proposal of a decision-support algorithm. J Neurol 261: 1291–1299. Fekedulegn D, Andrew ME, Shi M et al. (2020). Actigraphybased assessment of sleep parameters. Ann Work Expo Health 64: 350–367. Ferini-Strambi L, Marelli S (2012). Sleep dysfunction in multiple system atrophy. Curr Treat Options Neurol 14: 464–473. Fernandez F, Lu D, Ha P et al. (2014). Circadian rhythm. Dysrhythmia in the suprachiasmatic nucleus inhibits memory processing. Science 346: 854–857. Fifel K (2017). Alterations of the circadian system in Parkinson’s disease patients. Mov Disord 32: 682–692. Fifel K (2019). Neuropathology of circadian alterations in Parkinson disease. JAMA Neurol 76: 115. Fifel K, DeBoer T (2014). The central clock in patients with Parkinson disease. JAMA Neurol 71: 1455–1456. Fifel K, Videnovic A (2018). Light therapy in Parkinson’s disease: towards mechanism-based protocols. Trends Neurosci 41: 252–254. Fifel K, Videnovic A (2019). Chronotherapies for Parkinson’s disease. Prog Neurobiol 174: 16–27. Fifel K, Videnovic A (2020). Circadian alterations in patients with neurodegenerative diseases: neuropathological basis of underlying network mechanisms. Neurobiol Dis 144: 105029.

THE CIRCADIAN SYSTEM IN PARKINSON'S DISEASE Fifel K, Vezoli J, Dzahini K et al. (2014). Alteration of daily and circadian rhythms following dopamine depletion in MPTP treated non-human primates. PLoS One 9: e86240. Fifel K, Piggins H, Deboer T (2016). Modeling sleep alterations in Parkinson’s disease: how close are we to valid translational animal models? Sleep Med Rev 25: 95–111. Franzen K, Fliegen S, Koester J et al. (2017). Central hemodynamics and arterial stiffness in idiopathic and multiple system atrophy. J Neurol 264: 327–332. Friborg O, Rosenvinge JH, Wynn R et al. (2014). Sleep timing, chronotype, mood, and behavior at an Arctic latitude (69° N). Sleep Med 15: 798–807. Gabriel BM, Zierath JR (2019). Circadian rhythms and exercise—re-setting the clock in metabolic disease. Nat Rev Endocrinol 15: 197–206. Ghaemi M, Rudolf J, Hilker R et al. (2001). Increased pineal Fdopa uptake is related to severity of Parkinson’s disease-a PET study. J Pineal Res 30: 213–219. Gros P, Videnovic A (2020). Overview of sleep and circadian rhythm disorders in Parkinson disease. Clin Geriatr Med 36: 119–130. Guler AD, Ecker JL, Lall GS et al. (2008). Melanopsin cells are the principal conduits for rod-cone input to non-imageforming vision. Nature 453: 102–105. Gulmez Sevim D, Unlu M, Gultekin M et al. (2018). Evaluation of retinal changes in progressive supranuclear palsy and Parkinson disease. J Neuroophthalmol 38: 151–155. Hartmann A, Veldhuis JD, Deuschle M et al. (1997). Twenty-Four hour cortisol release profiles in patients with Alzheimer’s and Parkinson’s Disease compared to normal controls: ultradian secretory pulsatility and diurnal variation. Neurobiol Aging 18: 285–289. Hastings MH, Reddy AB, Maywood ES (2003). A clockwork web: circadian timing in brain and periphery, in health and disease. Nat Rev Neurosci 4: 649–661. Hastings MH, Maywood ES, Brancaccio M (2018). Generation of circadian rhythms in the suprachiasmatic nucleus. Nat Rev Neurosci 19: 453–469. Hauw JJ, Hausser-Hauw C, De Girolami U et al. (2011). Neuropathology of sleep disorders: a review. J Neuropathol Exp Neurol 70: 243–252. Hineno T, Mizobuchi M, Nishimatsu O et al. (1994). Daynight variation of urine volume in Parkinson’s disease. Jpn J Psychiatry Neurol 48: 583–587. Hu K, Van Someren EJ, Shea SA et al. (2009). Reduction of scale invariance of activity fluctuations with aging and Alzheimer’s disease: involvement of the circadian pacemaker. Proc Natl Acad Sci U S A 106: 2490–2494. Hyyppa M, Lehtinen P, Rinne UK (1971). Effect of L-dopa on the hypothalamic, pineal and striatal monoamines and on the sexual behaviour of the rat. Brain Res 30: 265–272. Indelicato E, Fanciulli A, Poewe W et al. (2015). Cerebral autoregulation and white matter lesions in Parkinson’s disease and multiple system atrophy. Parkinsonism Relat Disord 21: 1393–1397. Johnson CH, Elliott JA, Foster R (2003). Entrainment of circadian programs. Chronobiol Int 20: 741–774.

311

Joyce DS, Feigl B, Kerr G et al. (2018). Melanopsin-mediated pupil function is impaired in Parkinson’s disease. Sci Rep 178: 7796. Kreier F, Kap YS, Mettenleiter TC et al. (2006). Tracing from fat tissue, liver, and pancreas: a neuroanatomical framework for the role of the brain in type 2 diabetes. Endocrinology 147: 1140–1147. Langston JW, Forno LS (1978). The hypothalamus in Parkinson disease. Ann Neurol 3: 129–133. Lax P, Ortun˜o-Lizara´n I, Maneu V et al. (2019). Photosensitive melanopsin-containing retinal ganglion cells in health and disease: implications for circadian rhythms. Int J Mol Sci 20: 3164. LeGates TA, Fernandez DC, Hattar S (2014). Light as a central modulator of circadian rhythms, sleep and affect. Nat Rev Neurosci 15: 443–454. Leng Y, Musiek ES, Hu K et al. (2019). Association between circadian rhythms and neurodegenerative diseases. Lancet Neurol 18: 307–318. Li L, Zhao Z, Ma J et al. (2020). Elevated plasma melatonin levels are correlated with the non-motor symptoms in Parkinson’s disease: a cross-sectional study. Front Neurosci 14: 505. Long JM, Holtzman DM (2019). Alzheimer disease: an update on pathobiology and treatment strategies. Cell 179: 312–339. Lutze M, Bresnick GH (1991). Lenses of diabetic patients “yellow” at an accelerated rate similar to older normals. Invest Ophthalmol Vis Sci 32: 194–199. Maetzler W, Liepelt I, Berg D (2009). Progression of Parkinson’s disease in the clinical phase: potential markers. Lancet Neurol 8: 1158–1171. Mattis J, Sehgal A (2016). Circadian rhythms, sleep, and disorders of aging. Trends Endocrinol Metab 27: 192–203. Mendoza-Santiesteban CE, Gabilondo I, Palma JA et al. (2017a). The retina in multiple system atrophy: systematic review and meta-analysis. Front Neurol 8: 206. Mendoza-Santiesteban CE, Palma JA, Ortun˜o-Lizara´n I et al. (2017b). Pathologic confirmation of retinal ganglion cell loss in multiple system atrophy. Neurology 88: 2233–2235. Montplaisir J, Petit D, Decary A et al. (1997). Sleep and quantitative EEG in patients with progressive supranuclear palsy. Neurology 49: 999–1003. Musiek ES, Holtzman DM (2016). Mechanisms linking circadian clocks, sleep, and neurodegeneration. Science 354: 1004–1008. Niwa F, Kuriyama N, Nakagawa M et al. (2011). Circadian rhythm of rest activity and autonomic nervous system activity at different stages in Parkinson’s disease. Auton Neurosci 165: 195–200. Nomura T, Inoue Y, Takigawa H et al. (2012). Comparison of REM sleep behaviour disorder variables between patients with progressive supranuclear palsy and those with Parkinson’s disease. Parkinsonism Relat Disord 18: 394–396. Ortun˜o-Lizara´n I, Beach TG, Serrano GE et al. (2018a). Phosphorylated a-synuclein in the retina is a biomarker of Parkinson’s disease pathology severity. Mov Disord 33: 1315–1324.

312

K. FIFEL AND T. DE BOER

Ortun˜o-Lizara´n I, Esquiva G, Beach TG et al. (2018b). Degeneration of human photosensitive retinal ganglion cells may explain sleep and circadian rhythms disorders in Parkinson’s disease. Acta Neuropathol Commun 10: 90. Ozawa T (2007). Morphological substrate of autonomic failure and neurohormonal dysfunction in multiple system atrophy: impact on determining phenotype spectrum. Acta Neuropathol 114: 201–211. Ozawa T, Tanaka H, Miyatake T et al. (1993). Shy-Drager syndrome with abnormal circadian rhythm of plasma antidiuretic hormone secretion and urinary excretion. Intern Med 32: 225–227. Ozawa T, Oyanagi K, Tanaka H et al. (1998). Suprachiasmatic nucleus in a patient with multiple system atrophy with abnormal circadian rhythm of arginine-vasopressin secretion into plasma. J Neurol Sci 154: 116–121. Ozawa T, Soma Y, Yoshimura N et al. (2001). Reduced morning cortisol secretion in patients with multiple system atrophy. Clin Auton Res 11: 271–272. Panda S (2016). Circadian physiology of metabolism. Science 354: 1008–1015. Pierangeli G, Provini F, Maltoni P et al. (2001). Nocturnal body core temperature falls in Parkinson’s disease but not in Multiple-System Atrophy. Mov Disord 16: 226–232. Pilleri M, Levedianos G, Weis L et al. (2014). Heart rate circadian profile in the differential diagnosis between Parkinson disease and multiple system atrophy. Parkinsonism Relat Disord 20: 217–221. Poewe W, Seppi K, Tanner CM et al. (2017). Parkinson disease. Nat Rev Dis Primers 23: 17013. Pollak CP, Perlick D (1991). Sleep problems and institutionalization of the elderly. J Geriatr Psychiatry Neurol 4: 204–210. Riemersma-van der Lek RF, 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: a randomized controlled trial. JAMA 299: 2642–2655. Sakakibara R, Matsuda S, Uchiyama T et al. (2003). The effect of intranasal desmopressin on nocturnal waking in urination in multiple system atrophy patients with nocturnal polyuria. Clin Auton Res 13: 106–108. Sakurai T (2007). The neural circuit of orexin (hypocretin): maintaining sleep and wakefulness. Nat Rev Neurosci 8: 171–181. Saper CB (2013). The central circadian timing system. Curr Opin Neurobiol 23: 747–751. Saper CB, Scammell TE, Lu J (2005). Hypothalamic regulation of sleep and circadian rhythms. Nature 437: 1257–1263. Schibler U, Gotic I, Saini C et al. (2015). Clock-talk: interactions between central and peripheral circadian oscillators in mammals. Cold Spring Harb Symp Quant Biol 80: 223–232. Schmidt C, Berg D, Herting et al. (2009). Loss of nocturnal blood pressure fall in various extrapyramidal syndromes. Mov Disord 24: 2136–2142. Schneider M, M€uller HP, Lauda F et al. (2014). Retinal singlelayer analysis in Parkinsonian syndromes: an optical

coherence tomography study. J Neural Transm (Vienna) 121: 41–47. Schroeder AM, Colwell CS (2013). How to fix a broken clock. Trends Pharmacol Sci 34: 605–619. Sixel-D€ oring F, Schweitzer M, Mollenhauer B et al. (2009). Polysomnographic findings, video-based sleep analysis and sleep perception in progressive supranuclear palsy. Sleep Med 10: 407–415. Stamelou M, Giagkou N, H€ oglinger GU (2019a). One decade ago, one decade ahead in progressive supranuclear palsy. Mov Disord 34: 1284–1293. Stamelou M, Giagkou N, H€ oglinger GU (2019b). One decade ago, one decade ahead in progressive supranuclear palsy. Mov Disord 34: 1284–1293. Stemplewitz B, Kromer R, Vettorazzi E et al. (2017). Retinal degeneration in progressive supranuclear palsy measured by optical coherence tomography and scanning laser polarimetry. Sci Rep 7: 5357. Stuebner E, Vichayanrat E, Low DA et al. (2013). Twenty-four hour non-invasive ambulatory blood pressure and heart rate monitoring in Parkinson’s disease. Front Neurol 49: 1–14. Suzuki A, Asahina M, Ishikawa C et al. (2005). Impaired circadian rhythm of gastric myoelectrical activity in patients with multiple system atrophy. Clin Auton Res 15: 368–372. Suzuki K, Miyamoto T, Miyamoto M et al. (2007). Circadian variation of core body temperature in Parkinson disease patients with depression: a potential biological marker for depression in Parkinson disease. Neuropsychobiology 56: 172–179. Suzuki K, Miyamoto T, Miyamoto M et al. (2009). The core body temperature rhythm is altered in progressive supranuclear palsy. Clin Auton Res 19: 65–68. Takahashi JS, Hong HK, Ko CH et al. (2008). The genetics of mammalian circadian order and disorder: implications for physiology and disease. Nat Rev Genet 9: 764–775. Takekida S, Yan L, Maywood ES et al. (2000). Differential adrenergic regulation of the circadian expression of the clock genes Period1 and Period2 in the rat pineal gland. Eur J Neurosci 12: 4557–4561. Terazono H, Mutoh T, Yamaguchi S et al. (2003). Adrenergic regulation of clock gene expression in mouse liver. Proc Natl Acad Sci U S A 100: 6795–6800. Turek FW (2016). Circadian clocks: not your grandfather’s clock. Science 354: 992–993. Turner PL, Mainster MA (2008). Circadian photoreception: ageing and the eye’s important role in systemic health. Br J Ophthalmol 92: 1439–1444. Vallelonga F, Di Stefano C, Merola A et al. (2019). Blood pressure circadian rhythm alterations in alpha-synucleinopathies. J Neurol 266: 1141–1152. Van Someren EJ (2000). Circadian and sleep disturbances in the elderly. Exp Gerontol 35: 1229–1237. Van Someren EJ, Hagebeuk EE, Lijzenga C et al. (1996). Circadian rest-activity rhythm disturbances in Alzheimer’s disease. Biol Psychiatry 40: 259–270. Vetrano DL, Pisciotta MS, Lo Monaco MR et al. (2015). Association of depressive symptoms with circadian blood pressure alterations in Parkinson’s disease. J Neurol 262: 2564–2571.

THE CIRCADIAN SYSTEM IN PARKINSON'S DISEASE Vichayanrat E, Low DA, Iodice V et al. (2017). Twenty-fourhour ambulatory blood pressure and heart rate profiles in diagnosing orthostatic hypotension in Parkinson’s disease and multiple system atrophy. Eur J Neurol 24: 90–97. Videnovic A, Golombek D (2013). Circadian and sleep disorders in Parkinson’s disease. Exp Neurol 243: 45–56. Videnovic A, Golombek D (2017). Circadian dysregulation in Parkinson’s disease. Neurobiol Sleep Circadian Rhythms 2: 53–58. Videnovic A, Willis GL (2016). Circadian system—a novel diagnostic and therapeutic target in Parkinson’s disease? Mov Disord 31: 260–269. Videnovic A, Lazar AS, Barker RA et al. (2014a). ‘The clocks that time us’—circadian rhythms in neurodegenerative disorders. Nat Rev Neurol 10: 683–693. Videnovic A, Noble C, Reid KJ et al. (2014b). Circadian melatonin rhythm and excessive daytime sleepiness in Parkinson disease. JAMA Neurol 71: 463–469.

313

Walsh CM, Ruoff L, Varbel J et al. (2016). Rest-activity rhythm disruption in progressive supranuclear palsy. Sleep Med 22: 50–56. Walsh CM, Ruoff L, Walker K et al. (2017). Sleepless night and day, the plight of progressive supranuclear palsy. Sleep 40: zsx154. Welz PS, Benitah SA (2019). Molecular connections between circadian clocks and aging. J Mol Biol 432: 3661–3679. Wu YH, Fischer DF, Kalsbeek A et al. (2006). Pineal clock gene oscillation is disturbed in Alzheimer’s disease, due to functional disconnection from the “master clock”. FASEB J 20: 1874–1876. Zhang R, Lahens NF, Balance HI et al. (2014). A circadian gene expression atlas in mammals: implications for biology and medicine. Proc Natl Acad Sci U S A 111: 16219–16224. Zhao J, Warman GR, Cheeseman JF (2019). The functional changes of the circadian system organization in aging. Ageing Res Rev 52: 64–71.

This page intentionally left blank

Handbook of Clinical Neurology, Vol. 179 (3rd series) The Human Hypothalamus: Anterior Region D.F. Swaab, F. Kreier, P.J. Lucassen, A. Salehi, and R.M. Buijs, Editors https://doi.org/10.1016/B978-0-12-819975-6.00020-0 Copyright © 2021 Elsevier B.V. All rights reserved

Chapter 21

Retina and melanopsin neurons CHIARA LA MORGIA1,2*, VALERIO CARELLI1,2, AND ALFREDO A. SADUN3 1

IRCCS Istituto delle Scienze Neurologiche di Bologna, UOC Clinica Neurologica, Bologna, Italy

2

Dipartimento di Scienze Biomediche e Neuromotorie, Università degli Studi di Bologna, Bologna, Italy

3

Department of Ophthalmology, Doheny Eye Institute, University of California, Los Angeles, CA, United States

Abstract Melanopsin retinal ganglion cells (mRGCs) are the third class of retinal photoreceptors with unique anatomical, electrophysiological, and biological features. There are different mRGC subtypes with differential projections to the brain. These cells contribute to many nonimage-forming functions of the eye, the most relevant being the photoentrainment of circadian rhythms through the projections to the suprachiasmatic nucleus of the hypothalamus. Other relevant biological functions include the regulation of the pupillary light reflex, mood, alertness, and sleep, as well as a possible role in formed vision. The relevance of the mRGC-related pathways in the brain is highlighted by the role that the dysfunction and/or loss of these cells may play in affecting circadian rhythms and sleep in many neurodegenerative disorders including Alzheimer’s, Parkinson’s and Huntington’s disease and in aging. Moreover, the occurrence of circadian dysfunction is a known risk factor for dementia. In this chapter, the anatomy, physiology, and functions of these cells as well as their resistance to neurodegeneration in mitochondrial optic neuropathies or their predilection to be lost in other neurodegenerative disorders will be discussed.

MELANOPSIN RETINAL GANGLION CELLS: ANATOMY, PHYSIOLOGY, AND FUNCTIONS Melanopsin retinal ganglion cells: Discovery and anatomy Melanopsin retinal ganglion cells (mRGCs) are intrinsically photosensitive photoreceptors located in the inner retina of humans and other mammals. Their discovery came from the observation that mice with extensive loss of rods and cones (rd/rd and rd/rd cl) were still able to entrain their circadian rhythms to the light–dark cycle (Foster et al., 1991; Freedman et al., 1999; Lucas et al., 1999) and humans without light perception were still able to suppress melatonin secretion (Czeisler et al., 1995).

Provencio and colleagues identified melanopsin (MPS) as a photopigment independent from the other opsins, encoded by the OPN4 gene (Provencio et al., 1998, 2000). This established the existence of a third (i.e., not rod and not cone) class of photoreceptors in the retina. In 2002, it was finally proven that the intrinsically photosensitive cells expressing MPS are a subset of retinal ganglion cells (RGCs), which project to the suprachiasmatic nucleus (SCN) of the hypothalamus contributing to the photoentrainment of circadian rhythms (Berson et al., 2002; Hattar et al., 2002). Melanopsin RGCs are maximally sensitive to blue light and are responsive even when isolated from the surrounding retinal tissue (Hattar et al., 2002). These cells project to the SCN through the retinohypothalamic tract (RHT)

*Correspondence to: Chiara La Morgia, M.D., Ph.D., IRCCS Istituto delle Scienze Neurologiche di Bologna, UOC Clinica Neurologica, Bologna, Italy; Dipartimento di Scienze Biomediche e Neuromotorie, Università degli Studi di Bologna, Bologna, Italy. Tel: +39-051-4966112, Fax: +39-051-4966082, E-mail: [email protected]

316

C. LA MORGIA ET AL.

(Gooley et al., 2001; Hannibal et al., 2001, 2002; Hannibal, 2002; Hattar et al., 2002; Berson et al., 2002), a well-known anatomical pathway connecting the eye to the SCN (Moore and Lenn, 1972; Sadun et al., 1984; Moore et al., 1995). Melanopsin RGCs represent about 1% of the whole population of RGCs in humans (Hannibal et al., 2004; La Morgia et al., 2010, 2016). Their concentration is higher in the parafoveal and far nasal regions of the human retina (Hannibal et al., 2004, 2017; Dacey et al., 2005; La Morgia et al., 2010). They are characterized by a large soma (15–20 mm) with a central nucleus and the soma is located in about equal proportions in the retinal ganglion cell and inner nuclear layers (INL) of the human retina (Hannibal et al., 2004) (Fig. 21.1A and B). An example of regular RGCs is provided as comparison (Fig. 21.1C). Melanopsin RGCs are classified in subtypes, based on the levels of melanopsin expression, dendritic field arborization, physiology, and central projections (Sand et al., 2012; Schmidt et al., 2011; Lee and Schmidt, 2018). Another distinctive feature discriminating mRGC subpopulations is if they express or not the Brn3b

transcription factor (Chen et al., 2011). The mouse retina is thought to contain six mRGC subtypes, characterized by specific properties (Lee and Schmidt, 2018; Do, 2019). M1 mRGCs stratify in the OFF sublamina of the inner plexiform layer (IPL), whereas M2, M4, and M5 mRGCs stratify in the ON sublamina, and M3 and M6 mRGCs bistratify in both the ON and OFF sublaminae. The M4 (which are overlapping with the ON alpha RGCs) and M5 are characterized by a cellular soma bigger than the M1-M2-M3 subtypes (Lee and Schmidt, 2018). The M1, M2, and M3 are characterized by the highest MPS level expression (Schmidt et al., 2008). The recently described M6 cell, characterized by low levels of MPS expression, highly branched dendritic arborization, and small receptive fields, projects mainly to the dorsal lateral geniculate nucleus (dLGN), thus contributing to pattern vision, and to the pretectum (Quattrochi et al., 2019). The MPS photopigment is primarily expressed in the membrane of the soma, in the dendrites and even in the axons running within the retinal nerve fiber layer (RNFL) (Fig. 21.2A–C). The majority of dendrites stratifies either in the “off” or in the “on” layers of the IPL,

Fig. 21.1. Light micrographs of paraffin-embedded human retinas, immunoperoxidase stained for melanopsin with diaminobenzidine (brown color) showing examples of melanopsin retinal ganglion cells (mRGCs) in control subjects. (A) Shows one example of mRGC with the soma located in the retinal ganglion cell layer in a 74-year-old subject, whereas (B) shows one mRGC with the soma in the inner nuclear layer in an 80-year-old subject. Examples of regular RGCs are shown in (C) for a 105-year-old subject (for details on the methods of immunostaining, see La Morgia et al., 2010, 2016). From original source.

Fig. 21.2. Light micrographs of paraffin-embedded retinas, immunoperoxidase stained for melanopsin with diaminobenzidine (brown color) showing melanopsin retinal ganglion cell (mRGC) dendrites in control human retinas running in the inner nuclear layer (A) in a 105-year-old subject and in the retinal ganglion cell layer (B) in a 74-year-old subject. A mRGC axon running in the retinal nerve fiber layer is also shown in (C) in an 80-year-old subject (for details on the methods of immunostaining, see La Morgia et al., 2010, 2016). From original source.

RETINA AND MELANOPSIN NEURONS

317

M1, M1d (displaced), M2, and M3 with M1, M2, and M3 characterized by the GCL location of the soma, whereas the M1d (the most common subtype in human retinas) had their soma in the INL and their dendrites in S1. The M1 stratify in the S1 plexus of the IPL, the M2 in the S5, and the M3 in both (Ortuño-Lizarán et al., 2018a,b; Esquiva et al., 2017).

Melanopsin retinal ganglion cells: Physiology Fig. 21.3. The photoreceptive net constituted by melanopsin retinal ganglion cells in a flat-mount preparation of a control human retina is shown. Image provided by courtesy of Prof. Jens Hannibal, Department of Clinical Biochemistry, Bispebjerg Frederiksberg Hospital, University of Copenhagen, Copenhagen, Denmark. From original source.

thus providing large and arborized dendritic fields that constitute a photoreceptive net in the retina (Fig. 21.3) (Provencio et al., 2002; La Morgia et al., 2010; Hannibal et al., 2017). Most of the available information on the anatomy and retinal distribution of mRGCs is derived from rodent studies, less from primates (Dacey et al., 2005) and even less from human retinas (Hannibal et al., 2004, 2017; La Morgia et al., 2010, 2016; Esquiva et al., 2017; Ortuño-Lizarán et al., 2018a; Esquiva and Hannibal, 2019). Hannibal and coauthors recently provided a comprehensive contribution reporting in detail subtypes, distribution and intraretinal connections of human mRGCs (Hannibal et al., 2017). By using antibodies targeting the terminal part of human melanopsin, confocal microscopy and 3D reconstruction of melanopsin-immunoreactive RGCs, and applying the same criteria used in mouse, they identified M1, displaced M1, M2, and M4 mRGCs in human retinas. M1 cells are the majority of mRGCs, and the density of M2 and M4 cells is higher in the nasal retina (Hannibal et al., 2017; Esquiva and Hannibal, 2019). Two other subtypes of mRGCs named “gigantic M1” and “gigantic displaced M1” are described (Hannibal et al., 2017). Furthermore, only a few M3 cells and none of the M5 or M6 subtypes were identified. The total number of mRGCs compared to the total RGCs was 0.63%– 0.75%. Finally, they were able to demonstrate contacts between the AII amacrine cells and mRGCs, and the direct contact of rod bipolar cells, via ribbon synapses, with mRGCs in the innermost ON layer of the IPL as well as of dopaminergic amacrine cells and GABAergic processes with mRGCs in the outermost OFF layer of the IPL (Hannibal et al., 2017). More recently, Ortuño-Lizarán and coauthors described the morphology and distribution of mRGCs in retinas from controls and patients with Parkinson’s disease (PD), reporting four types of mRGCs:

The mRGCs are characterized by unique physiological properties due to the presence of the MPS photopigment (Berson et al., 2002; Hattar et al., 2002). These cells are maximally sensitive to short wavelength blue light (peak response at 460–470 nm), having a sustained response to light that persists after the light is switched off (Foster, 2005; Do and Yau, 2010; Wong, 2012; Emanuel and Do, 2015). These unique electrophysiological properties explain why mRGCs produce a sustained pupillary light reflex (PLR), i.e., postillumination pupillary response (PIPR) (Kankipati et al., 2010), and why they can function as irradiance (Lall et al., 2010) and brightness detectors (Brown et al., 2012). The different mRGC subtypes have also distinct electrophysiological features of the intrinsic responses, which are at lower threshold, higher amplitude, and faster for M1 cells than those of M2–M5 subtypes (Zhao et al., 2014). Moreover, the M2–M5 mRGCs had centersurround-organized receptive fields, making them able to detect spatial contrast, whereas the receptive fields of M1 cells lacked this center-surround antagonism (Zhao et al., 2014). Another unique feature of mRGCs is the tristable nature of the photopigment. This refers to the ability of light to independently regenerate 11-cis retinaldehyde in photobleached opsin without depending on an independent isomerase, unlike classical photoreceptors, i.e., rods and cones (Koyanagi et al., 2005; Matsuyama et al., 2012; Emanuel and Do, 2015). Moreover, mRGCs use a G-protein-coupled invertebrate-like rhabdomeric transduction of the light stimulus, depolarizing in response to light, opposite to rods and cones, which are hyperpolarized in response to light (Graham et al., 2008) and it has been recently shown that mRGCs use both the rhabdomeric and ciliary transduction systems (Jiang et al., 2018). Melanopsin is expressed in two isoforms in mice: the short (OPN4s), mainly expressed in the M1 subtype, and the long (OPN4l), mainly represented in the non-M1 subtypes, suggesting the possibility of different functional roles of the two isoforms (Jagannath et al., 2015). Interestingly, MPS expression itself follows a circadian cycle, peaking at the end of the day and decreasing at night (Hannibal et al., 2005).

318

C. LA MORGIA ET AL.

MPS expression profile has been focused on mRGCs in the retina and in a few other neuronal subtypes such as trigeminal sensory ganglia in mice (Matynia et al., 2016). A recent study systemically investigated MPS expression in the whole human brain suggesting that, even if with variable magnitudes, a much wider MPS expression occurs than previously believed (Nissil€a et al., 2017). Although mRGCs can function independently as photoreceptors, they also receive inputs from rods and cones through amacrine and bipolar cells (Viney et al., 2007). This circuitry implicates an input from the other retinal photoreceptors, i.e., rods and cones, allowing the system to compare inputs and function as an “irradiancedetector” in addition to serving formed vision (Viney et al., 2007; Jusuf et al., 2007; Schmidt and Kofuji, 2009, 2010; Lall et al., 2010). Melanopsin RGCs primarily use glutamate as neurotransmitter, but they also express the pituitary adenylate cyclase activating peptide (PACAP) (Hannibal et al., 2002), which is uniquely expressed by mRGCs and is a marker for the RHT (Hannibal et al., 2002, 2004).

Melanopsin retinal ganglion cells: Functions Melanopsin RGCs may have several important biological functions in the body. We know that mRGCs contribute to nonimage-forming functions of the eye and they measure overall irradiance, but not so much visual resolution or contrast sensitivity (Do, 2019). The threshold and physiology in terms of irradiance and other electrophysiological parameters of the different nonimage-forming functions are variable (Do, 2019), and the various mRGC subtypes contribute differently to diverse nonimage-forming functions according to their distinct melanopsin content, electrophysiological properties, and projections to the brain (Fig. 21.4). M1 cells, given their peculiar physiological properties, are reputed to encode irradiance and contribute to nonimage-forming functions of the eye, in particular to circadian photoentrainment through the projections to the SCN, and to other brain centers for other nonimageforming functions. These include in mice projections to the shell of olivary pretectal nucleus (OPN), the anterior

Fig. 21.4. Major brain targets of mouse melanopsin retinal ganglion cells A sample of mRGC brain targets is depicted in a quasisagittal schematic of the mouse brain. Below is a plot of innervation densities across mRGC types, drawn after Berson and colleagues (Quattrochi et al., 2019) and incorporating additional information (Hattar et al., 2006; Ecker et al., 2010; Morin and Studholme, 2014; Zhao et al., 2014; Huang et al., 2019;). Each blue dot indicates the approximate density of innervation by its size, a white dot indicates undetectable innervation, and lack of a dot indicates an absence of information. M5s and M6s are pooled because their projections were examined together for technical reasons. AH, anterior hypothalamus; BST, bed nucleus of the stria terminalis; dLGN, dorsal lateral geniculate nucleus; IGL, intergeniculate leaflet; LH, lateral hypothalamus; MA, medial amygdala; OPN, olivary pretectal nucleus (with shell, s, and core, c, regions); PA, preoptic area, which includes the VLPO (ventrolateral preoptic area); PAG, periaqueductal gray; PHb, perihabenular zone; pSON, perisupraoptic nucleus; SC, superior colliculus; SCN, suprachiasmatic nucleus; sPa, subparaventricular zone; and vLGN, ventral lateral geniculate nucleus. From Do MTH (2019). Melanopsin and the intrinsically photosensitive retinal ganglion cells: biophysics to behavior. Neuron 104: 205–226.

RETINA AND MELANOPSIN NEURONS hypothalamus (AH); lateral hypothalamus (LH); medial amygdala (MA); preoptic area (PA), which includes the ventrolateral pre-optic (VLPO); periaqueductal gray (PAG); bed nucleus of the stria terminalis (BST), perihabenular zone (PHb); subparaventricular zone (sPa), and the perisupraoptic nucleus (pSON) (Hattar et al., 2006; Do, 2019). In particular, in mice, the M1 Brn3b-negative project to the SCN and mediate the circadian photoentrainment, whereas the M1 Brn3b-positive constitute the major contributors of the projections to the shell of the OPN regulating the pupillary light reflex, non-SCN hypothalamic targets, such as the ventral medial hypothalamus, the pSON and VLPO area, and other thalamic and midbrain targets including the posterior limitans, lateral habenula, superior colliculus (SC), and rostral dorsal lateral geniculate nucleus (dLGN) (Chen et al., 2011; Li and Schmidt, 2018). In contradistinction, the non-M1 are involved in the regulation of the PLR (mainly the M2 projecting to the core of the OPN), but especially in visual-forming functions, by projecting to the SC and the dLGN and vLGN where they convey center-surround receptive fields, as do the regular RGCs (Hattar et al., 2006; Morin and Studholme, 2014; Do, 2019; Huang et al., 2019). Perhaps the most important function of mRGCs is to synchronize circadian biological rhythms to the light– dark cycle, through the retinofugal projections to the SCN. Dysfunction of this visual input to the hypothalamic master clock leads to circadian misalignment and all the pathological consequences related to this (Blackeman et al., 2016; Maury, 2019; Leng et al., 2019). Melanopsin RGCs also contribute, through retinofugal projections to the OPN, to the control of the PLR, and in particular of its sustained component (PIPR) (Gooley et al., 2012) maintaining the tonic and sustained part of the PLR. In fact, they are characterized by a response that persists for the whole duration of the stimulus and is particularly relevant in bright light, whereas rods and cones contribute to its transient phase at low irradiance (Munch and Kawasaki 2013; Keenan et al., 2016; La Morgia et al., 2018; Kelbsch et al., 2019). Melanopsin RGCs also regulate melatonin secretion and its suppression by light exposure (Prayag et al., 2019). Moreover, a direct role of mRGCs on sleep is postulated, based on the projections to the VLPO of the hypothalamus. The acute induction of sleep mediated by light is abolished in mice without mRGCs (Altimus et al., 2008; Lupi et al., 2008) and OPN4 / mice show abnormal homeostatic responses to sleep deprivation (Tsai et al., 2009). A behavioral arousal has been demonstrated in mice in response to blue light, corresponding to c-fos induction in the SCN, whereas green light promotes sleep through the activation of the VLPO and both these

319

responses are abnormal in melanopsin-deficient mice (Pilorz et al., 2016). The authors hypothesized that the different effects may be explained by projections to distinct brain areas (VLPO for sleep-inducing effect and SCN for the alerting effect), possibly mediated by specific mRGC subtypes (Pilorz et al., 2016). Interestingly, Rupp and colleagues showed that a molecularly defined mRGC subtype (the Brn3b + cells projecting mainly to non-SCN brain regions), distinct from those responsible for the circadian effects of light (the Brn3b cells projecting to the SCN), mediated the acute effects on sleep and temperature in mice (Rupp et al., 2019). The brain areas potentially implicated in the acute effects of light on sleep and temperature are the sPa, pretectum, SC, PO areas including MPO and VLPO even though the exact brain areas implicated in these acute effects are still unknown (Rupp et al., 2019). At difference with rodents, which are nocturnal animals for whom an acute induction of sleep by light is reported (Altimus et al., 2008; Lupi et al., 2008), in humans the light exerts an alerting effect (Vandewalle et al., 2007). This alerting effect of light has been demonstrated also by functional MRI studies, corresponding to the activation of the brainstem, in particular of the locus coeruleus (Vandewalle et al., 2007). Light can also modulate cognition in humans through its alerting effect and activation of prefrontal and frontoparietal circuits (Vandewalle et al., 2011a), as well as emotional brain responses by activating temporal cortex and amygdala (Vandewalle et al., 2010). Remarkably, an abnormal hypothalamic response to blue light has been found in seasonal affective disorder (SAD) patients (Vandewalle et al., 2011b). In SAD, there is a specific vulnerability to the absence of sufficient white or blue light stimulation, especially in northern latitudes and in wintertime, and it has been demonstrated the possible beneficial effect of light therapy (Pjrek et al., 2020). Congruently, polymorphic genetic variants of the OPN4 gene have been associated with SAD suggesting a possible role in the susceptibility to this disorder (Roecklein et al., 2009). More recently, a possible contribution of mRGCs to image-forming functions has also been postulated (Brown et al., 2010; Ecker et al., 2010; Sonoda and Schmidt, 2016; Quattrochi et al., 2019). In fact, mRGCs may also contribute to conscious (formed) vision as inferred from the observation that in individuals blinded by severe outer retina degeneration there remains the perception of intense blue light (Zaidi et al., 2007). Functional MRI studies, using the technique of the silent substitution to selectively target one photopigment, specifically melanopsin, demonstrated that its specific stimulation with light led to the activation of the primary visual cortex in association with the distinct and

320

C. LA MORGIA ET AL.

conscious perception of brightness (Spitschan et al., 2017; Spitschan, 2019). Moreover, a specific activation of dLGN, vLGN, and SC neurons mediated by the activation of mRGCs has been described in mice (Allen et al., 2014, 2017), as well as the observed direct contacts of mRGCs with amacrine cells might be the anatomical basis for the potential role of mRGCs in image-forming vision. It must be emphasized that most of the information available to understand mRGC functions comes from mice studies, whereas in humans, there are studies focusing mainly on PLR (Park et al., 2011; La Morgia et al., 2018; Rukmini et al., 2019). It should be reminded that the regulation of PLR by mRGCs has distinct peculiarities in primates and mice (Spitschan et al., 2014; Hayter and Brown, 2018). A further developing story, which enlarges the landscape of mRGC functions concerns their role in the exacerbation of pain elicited by light (Noseda et al., 2010) and photophobia (Matynia et al., 2012). These latter functions seem mediated by the MPS expression in the trigeminal sensory ganglia (Matynia et al., 2016) and in the cornea (Delwig et al., 2018). Intriguingly, melanopsin is expressed in mice embryos before rods and cones photopigments (Tarttelin et al., 2003), and mRGCs are functional before eye opening in neonatal mice (Sekaran et al., 2005), highlighting the earlier evolutionary role of this cell type and recapitulation in ontogeny and development of the eye and its retinal connections.

MELANOPSIN RETINAL GANGLION CELLS: RESISTANCE TO NEURODEGENERATION IN MITOCHONDRIAL OPTIC NEUROPATHIES Mice models and observations in humans with no light perception were pivotal to identify the existence of mRGCs, the third class of photoreceptors in the eye, explaining the preservation of circadian photoentrainment in blind people with destructive retinal and optic nerve disorders (Czeisler et al., 1995). Inherited optic neuropathies are disorders where RGCs are selectively targeted by the neurodegenerative process leading to atrophy of the optic nerve and profound blindness, but with intact outer retina, i.e., rods/cones and retinal pigmented epithelium (Yu-Wai-Man et al., 2016). The large majority of inherited optic neuropathies are due to mitochondrial dysfunction related to mutations affecting either the small circular and multicopy mitochondrial genome (mtDNA), as for Leber’s Hereditary Optic Neuropathy (LHON) (Leber, 1871), or the nuclear genome (nDNA), which encodes about 1500 mitochondrial proteins (Carelli et al., 2004). In this latter case, the most

frequent disease is Dominant Optic Atrophy (DOA) (Kjer, 1959), which, in over 60% of cases, is due to mutations in the optic atrophy 1 (OPA1) gene (Delettre et al., 2000). Both disorders present similar clinical features characterized by profound loss of central vision with optic atrophy due to the selective loss of RGCs, as such possibly including the mRGC subpopulation (Carelli et al., 2004; Yu-Wai-Man et al., 2016). However, particularly for LHON, there has been a longstanding question about the puzzling observation of a relatively spared PLR notwithstanding severe visual impairment (Wakakura and Yokoe, 1995; Bremner et al., 1999). Interestingly, spared axons leaving the optic chiasm and projecting to the hypothalamus, along the RHT, were also documented in a postmortem study of a single LHON patient (Bose et al., 2005). For these reasons, the model of LHON and DOA has been investigated for the possibility that both disorders might have selective sparing of mRGCs. The results of this study, evaluating LHON and DOA postmortem retinas, revealed the substantial preservation of mRGCs, in contrast to neurodegeneration that severely ablated the general RGC population (La Morgia et al., 2010). Patients with LHON and DOA maintained the suppression of melatonin secretion after light exposure and did not suffer obvious circadian or sleep disturbances, similar to controls. Furthermore, mRGCs and axonal counts on postmortem specimens revealed the selective sparing of mRGCs in both diseases (La Morgia et al., 2010; Fig. 21.5). This finding, which also explains the maintenance of the PLR in these patients, was further confirmed by other studies investigating in vivo the mRGC system by the pupillometry-based assessment of the PLR and PIPR in LHON patients (Kawasaki et al., 2010; Moura et al., 2013). The observation of the unique metabolic

Fig. 21.5. Light micrographs of paraffin-embedded retinas, immunoperoxidase stained for melanopsin with diaminobenzidine (brown color) shows at low-mag two mRGCs, one located in the inner nuclear layer and one in the retinal ganglion cell layer in a postmortem retina of a 52-year-old patient severely affected by Leber’s Hereditary Optic Neuropathy (disease onset at 27 years) (for details on the methods of immunostaining, see La Morgia et al., 2016). From original source.

RETINA AND MELANOPSIN NEURONS resistance of mRGCs to mitochondrial dysfunction is still not fully understood, but the robustness of this photoreceptive cell system goes beyond axonal degeneration from mitochondrial injury. There are many other examples including optic nerve resection models of RGC injury, as well as excitotoxicity induced by N-methylD-aspartate, confirming the peculiar resilience of mRGCs (Georg et al., 2017; Sánchez-Migallón et al., 2018; Wang et al., 2018). Currently, there are only a few studies investigating the metabolic features of mRGCs, which are now known for having an abundant mitochondrial mass and a distinctive activity in calcium handling that probably relates to their photoreceptive functions (Hartwick et al., 2007; Georg et al., 2017). These properties may be of importance for the intrinsic robustness of mRGCs to various categories of injuries.

MELANOPSIN RETINAL GANGLION CELLS IN AGE-RELATED NEURODEGENERATIVE DISORDERS Alzheimer’s disease Alzheimer’s disease (AD) is the most common age-related dementia, whose prevalence is growing with the increase of life expectancy (Lane et al., 2018). The neuropathological brain hallmarks of AD are amyloid plaques and neurofibrillary tangles (Serrano-Pozo et al., 2011). It is now well established by both histological and optical coherence tomography (OCT) studies that AD patients present with age-related optic nerve pathology, more evident in the superior quadrant (Hinton et al., 1986; Sadun and Bassi, 1990; La Morgia et al., 2016; Chan et al., 2019; Asanad et al., 2019a). Moreover, the occurrence of amyloid pathology in AD human retinas has been demonstrated by using in vivo

321

curcumin-based noninvasive optical imaging (KoronyoHamaoui et al., 2011; Koronyo et al., 2017; Hart et al., 2016;) and in postmortem AD retinas (La Morgia et al., 2016; Koronyo et al., 2017). Recent evidences also point to the presence of a retinal vascular pathology in AD retinas (Asanad et al., 2019b). AD is characterized, even in the early stages of the disease, by the occurrence of sleep and circadian disturbances (Peter-Derex et al., 2015; Musiek and Holtzman, 2016; La Morgia et al., 2016, 2017a; Leng et al., 2019), and circadian dysfunction is considered a major predictor for the development of cognitive impairment (Tranah et al., 2011). The circadian disturbances in AD impact on many biological functions including rest-activity, body temperature, and melatonin rhythms and are relevant for the occurrence of sundowning (Volicer et al., 2001; Skene and Swaab, 2003; Hooghiemstra et al., 2015; La Morgia et al., 2016, 2017a; Musiek et al., 2018). Such disturbances may appear even in the presymptomatic stage of the disease, and fragmentation of the rest-activity rhythm, i.e., intradaily variability, is correlated with phosphorylated tau/Ab42 ratio (Musiek et al., 2018). These abnormalities have been related to the presence of SCN and SCN-pineal axis neurodegenerative changes including classical amyloid pathology (Swaab et al., 1985, 1987, 1988; Witting et al., 1990; Stopa et al., 1999; Wu and Swaab, 2005; Wu et al., 2007; Harper et al., 2008; Wang et al., 2015), a dysfunctional clock genes machinery (Bellanti et al., 2017), and retinal abnormalities (La Morgia et al., 2016). In particular, the presence of mRGC loss and the occurrence of amyloid pathology specifically affecting mRGCs, including neurodegenerative changes affecting dendritic morphology, have been demonstrated in postmortem human AD retinas (La Morgia et al., 2016) (Fig. 21.6A–C).

Fig. 21.6. Light micrographs of paraffin-embedded retinas, immunoperoxidase stained for melanopsin with diaminobenzidine (brown color). In (A) is shown one example of a melanopsin retinal ganglion cell (mRGC) dendrite in a retina from a 95-year-old patient affected by Alzheimer’s disease (AD), characterized by focal attenuation of the melanopsin staining and varicosity. In (B) and (C), two examples of flat-mounted retinas with mRGCs and their dendritic harborization are shown for a 60-year-old control individual (B) and an 80-year-old AD patient (C). For these images, a three-dimensional reconstruction was performed followed by analysis of the processes using the filament trace module in Imaris (Bitplane) (for details on methods, see La Morgia et al., 2016). From original source (Panels B and C courtesy of Prof. Jens Hannibal, University of Copenaghen).

322

C. LA MORGIA ET AL.

Loss of mRGCs in AD in this cohort of cases is age-independent pointing to a specific AD pathology affecting these cells, whereas the loss of other optic nerve axons remains mostly an age-related phenomenon (La Morgia et al., 2016). The evidence of pupil abnormalities in AD, even at presymptomatic stages, further supports these findings (Chougule et al., 2019; Oh et al., 2019; Romagnoli et al., 2020).

Parkinson’s disease Parkinson’s disease is a common age-related neurodegenerative disorder characterized by bradykinesia, rigidity, and resting tremor (Ascherio and Schwarzschild, 2016; Postuma et al., 2015). Visual problems in PD include dry eyes, dyschromatopsia, impairment of contrast sensitivity, visual hallucinations, and oculomotor abnormalities including dysmetria, slower saccadic eye movements, cogwheeling smooth pursuits, and convergence insufficiency (Weil et al., 2016). Contrast sensitivity dysfunction is probably related to dopamine depletion in PD retinas (Bodis-Wollner, 1990; Harnois and Di Paolo, 1990). OCT and histological studies in PD patients demonstrated RGC loss and axonal depletion in the optic nerve, especially in the eye contralateral to the more affected body side and more evident in the temporal quadrant (i.e., the papillomacular bundle) (La Morgia et al., 2013). This last feature is distinct from AD where RGC loss is more prominent in the superior quadrant (La Morgia et al., 2017a,b). Furthermore, recent studies documented a-synuclein deposition in the inner retina of PD patients, which was correlated with disease severity and Lewy Body brain pathology (Beach et al., 2014; Bodis-Wollner et al., 2014; Ortuño-Lizarán et al., 2018b; Veys et al., 2019). Sleep problems are frequent in PD patients including insomnia, excessive daytime sleepiness, sleep-related breathing disorders, restless leg syndrome, and REM behavior disorder (Chahine et al., 2017). There are studies demonstrating the presence of abnormal melatonin, rest-activity, blood pressure, and heart rate circadian rhythms in PD (Ejaz et al., 2006; Niwa et al., 2011; Bolitho et al., 2014; Leng et al., 2019; Vallelonga et al., 2019). Moreover, Lewy body pathology has been demonstrated in the SCN of PD patients (De PabloFernández et al., 2018), even though the SCN is otherwise unaffected in PD until late stages of the disease (Fifel, 2017). Remarkably, Ortuño-Lizarán and colleagues recently documented mRGC loss in six postmortem PD retinas, compared to controls, mostly affecting the M1d and M2 subtypes (Ortuño-Lizarán et al., 2018a). Furthermore, these authors also highlighted degenerative features of the remaining mRGCs, in particular a lower dendrite complexity, reduced contact

between cells, reduced dendritic beads, and lower MPS staining, with an overall lower cell complexity and ramifications of mRGCs, similar to what has been found in AD (La Morgia et al., 2016; Ortuño-Lizarán et al., 2018a). The authors hypothesized that mRGC pathology, in particular of the M1d subtype, may be driven by the dopamine depletion affecting the synaptic interaction between amacrine dopaminergic cells and mRGCs in PD retinas (Ortuño-Lizarán et al., 2018a). These findings support the concept that the dysfunction of mRGCs may contribute to sleep and circadian abnormalities in PD, as it has been reported for AD (La Morgia et al., 2016, 2017a). The loss of mRGCs in PD patients is further confirmed by the recent observation, using chromatic pupillometry and comparing short and long wavelength stimulation in 17 PD and 12 controls, of a reduced PIPR and peak amplitude of the PLR with blue light in PD patients, suggesting that also in PD patients there is a melanopsin-mediated pupil dysfunction (Joyce et al., 2018).

Other neurodegenerative disorders The possible role of mRGCs in the pathogenesis of circadian dysfunction has also been investigated in other neurodegenerative disorders, such as Huntington’s disease (HD), in which the presence of circadian dysfunction is a prominent and well-documented feature (Morton, 2013). In HD, there is evidence of delayed phase and abnormal day–night ratio of rest-activity and melatonin rhythms and consistent sleep fragmentation (Morton, 2013). OCT studies in HD patients evidenced optic nerve degeneration, especially in the temporal RNFL with a pattern similar to PD and mitochondrial optic neuropathies, which correlated with disease duration (Kersten et al., 2015; Andrade et al., 2016). Furthermore, PLR dysfunction is described in R6/2 and Q175 mouse models with a prevalent contribution of mRGCs in old mice (Ouk et al., 2016). However, even if a reduction of MPS expression is evident in both mouse models, mRGCs were morphologically intact with no evidence of huntingtin aggregation (Ouk et al., 2016). Recently, Lin and coauthors by studying two HD mouse models (R6/2 and N171-82Q) reported that not only there was a reduced MPS expression before the onset of motor symptoms but also a loss of M1 mRGCs in R6/2 male mice due to apoptosis (Lin et al., 2019). This loss led to a reduced light-induced c-fos and vasoactive intestinal peptide (VIP) expression in the SCN, providing a basis for the defective circadian photoentrainment in HD. However, in these mice models, there was a relative preservation of the non-M1 ipRGC subtypes (Lin et al., 2019). These results are in line with the observation of

RETINA AND MELANOPSIN NEURONS

323

an abnormal photic entrainment in R6/2 mice (Ouk et al., 2019), and, interestingly, treatment with blue light improves circadian dysfunction as well as motor symptoms in BACHD and Q175 mice (Wang et al., 2017). The possible role of the mRGC system in the pathogenesis of circadian dysfunction in HD is also supported by the evidence of SCN degeneration in HD, as demonstrated by the presence in the SCN of HD postmortem tissues of early neuronal loss in the central pacemaker in the Q175 mouse model (Kuljis et al., 2018), loss of VIP, and arginine vasopressin neurons as well as of mutant huntingtin inclusions (van Wamelen et al., 2013, 2014). Finally, a (11)C-raclopride (RAC) and (11)C-(R)-PK11195 (PK) PET study reported significant D(2) receptor loss and microglia activation in the hypothalamus of HD patients in the presymptomatic stage of the disease, suggesting an early hypothalamic pathology in this disease (Politis et al., 2008).

stimulating effect of blue light on cognitive brain functions (Daneault et al., 2014, 2018) and reduces the acute and phase-advancing responses to light (Sletten et al., 2009), as well leads to a weakening of body temperature and locomotor activity circadian rhythms robustness in a rodent model of retinitis pigmentosa (Lax et al., 2016). Pupillometry studies evaluating mRGC function in relation to aging failed to reveal a clear decline of mRGC-mediated PIPR responses (Daneault et al., 2012; Adhikari et al., 2015; Rukmini et al., 2017). In one study, paradoxically enhanced PIPR amplitude was reported with age (Herbst et al., 2012). The lack of abnormalities in pupil response in relation to age may be possibly explained by the inclusion in these studies of people younger than 70 years.

MELANOPSIN RETINAL GANGLION CELLS IN AGING

Melanopsin RGCs represent the third class of photoreceptor in the eye contributing to many biological relevant functions including nonimage-forming functions such as the photoentrainment of circadian rhythms, regulation of the PLR, sleep and melatonin synthesis, and its suppression but also to image-forming functions. Different subtypes of mRGCs have been described in rodent and human retinas with distinct projections to the brain. While a great deal of studies are available on animals, the investigation of mRGC anatomy, physiology, and biological role in humans remains at still initial stages, mostly due to objective difficulties in exploring this system in vivo. However, given the relevance of these cells in chronobiology, the investigation of neurodegenerative age-related disorders with clear-cut evidences of circadian and sleep dysfunction and eye pathology highlighted mRGC-SCN axis as relevant contributor to circadian and sleep disturbances. A deeper understanding of how this circuitry may impact on aging and possibly longevity is the next challenge, as well the possibility to use light therapy as a potential treatment for circadian misalignment in neurodegenerative disorders and aging.

Normal aging is characterized by the occurrence of sleep and circadian dysfunction. In particular, there is evidence of abnormal synchronization of circadian rhythms to the light–dark cycle with phase advance, reduced amplitude, reduced responsiveness to phase-shifting signals, and reduced robustness of circadian rhythms with aging (Zhang et al., 1996; Harper et al., 2005; Cajochen et al., 2006). These changes may have a multifactorial etiology. In fact, with aging it becomes evident a reduction of the strength of the SCN rhythms in terms of electrical activity of SCN neurons and output signals (Satinoff et al., 1993; Aujard et al., 2001; Nakamura et al., 2011; Farajnia et al., 2012). Moreover, age-related decrease of vasopressin mRNA expression in the SCN is reported in humans (Liu et al., 2000) and in rodents (Krajnak et al., 1998; Kalló et al., 2004; Duncan et al., 2010), as well as loss of VIP neurons (Zhou et al., 1995). With aging, there is also evidence of dampening or disruption of clock gene expression rhythms in the SCN (Duncan et al., 2013; Kolker et al., 2003), which all contribute to circadian dysfunction. The presence of mRGC loss with aging is reported in young and aged rodless/coneless mice (Semo et al., 2003) and more recently in human retinas (La Morgia et al., 2010, 2016; Esquiva et al., 2017). Esquiva and coauthors, in fact, evaluated 24 postmortem retinas in subjects with age ranging from 10 to 81 years showing a decrease of mRGC count after age 70 (about 30% less than individuals aged between 30 and 50 years). Moreover, after 50 years of age, a decrease in the dendritic area of mRGCs is evident (Esquiva et al., 2017; Esquiva and Hannibal, 2019; La Morgia et al., 2010). In accordance with these findings, aging reduces the

CONCLUSIONS

ACKNOWLEDGMENTS We deeply thank Fred Ross-Cisneros (Doheny Eye Institute, Los Angeles, CA, USA) who performed all the immunohistochemical analyses on postmortem retinas and Prof. Jens Hannibal (Department of Clinical Biochemistry, Bispebjerg Frederiksberg Hospital, University of Copenhagen, Copenhagen, Denmark) who provided the antibodies for melanopsin staining and figures for the current chapter.

324

C. LA MORGIA ET AL.

FUNDING This work was supported by the Ministry of Health Young Researcher project (GR-2013-02358026) (to C.L.M.), the Italian Ministry of Health and of Research, the Gino Galletti Foundation (to V.C. and C.L.M.), and Research to Prevent Blindness (RPB to A.A.S.).

REFERENCES Adhikari P, Pearson CA, Anderson AM et al. (2015). Effect of age and refractive error on the Melanopsin mediated postillumination pupil response (PIPR). Sci Rep 5: 17610. Allen AE, Storchi R, Martial FP et al. (2014). Melanopsindriven light adaptation in mouse vision. Curr Biol 24: 2481–2490. Allen AE, Storchi R, Martial FP et al. (2017). Melanopsin contributions to the representation of images in the early visual system. Curr Biol 27: 1623–1632.e4. Altimus CM, G€uler AD, Villa KL et al. (2008). Rods-cones and melanopsin detect light and dark to modulate sleep independent of image formation. Proc Natl Acad Sci U S A 16; 105: 19998–20003. Andrade C, Beato J, Monteiro A et al. (2016). Spectral-domain optical coherence tomography as a potential biomarker in Huntington’s disease. Mov Disord 31: 377–383. Asanad S, Ross-Cisneros FN, Nassisi M et al. (2019a). The retina in Alzheimer’s disease: histomorphometric analysis of an ophthalmologic biomarker. Invest Ophthalmol Vis Sci 60: 1491–1500. Asanad S, Ross-Cisneros FN, Barron E et al. (2019b). The retinal choroid as an oculovascular biomarker for Alzheimer’s dementia: a histopathological study in severe disease. Alzheimers Dement 11: 775–783. Ascherio A, Schwarzschild MA (2016). The epidemiology of Parkinson’s disease: risk factors and prevention. Lancet Neurol 15: 1257–1272. Aujard F, Herzog ED, Block GD (2001). Circadian rhythms in firing rate of individual suprachiasmatic nucleus neurons from adult and middle-aged mice. Neuroscience 106: 255–261. Beach TG, Carew J, Serrano G et al. (2014). Arizona Parkinson’s disease consortium. Phosphorylated a-synucleinimmunoreactive retinal neuronal elements in Parkinson’s disease subjects. Neurosci Lett 571: 34–38. Bellanti F, Iannelli G, Blonda M et al. (2017). Alterations of clock gene RNA expression in brain regions of a triple transgenic model of Alzheimer’s disease. J Alzheimers Dis 59: 615–631. Berson DM, Dunn FA, Takao M (2002). Phototransduction by retinal ganglion cells that set the circadian clock. Science 295: 1070–1073. Blakeman V, Williams JL, Meng QJ et al. (2016). Circadian clocks and breast cancer. Breast Cancer Res 18: 89. Bodis-Wollner I (1990). Visual deficits related to dopamine deficiency in experimental animals and Parkinson’s disease patients. Trends Neurosci 13: 296–302. Bodis-Wollner I, Kozlowski PB, Glazman S et al. (2014). a-Synuclein in the inner retina in Parkinson disease. Ann Neurol 75: 964–966.

Bolitho SJ, Naismith SL, Rajaratnam SM et al. (2014). Disturbances in melatonin secretion and circadian sleepwake regulation in Parkinson disease. Sleep Med 15: 342–347. Bose S, Dhillon N, Ross-Cisneros FN et al. (2005). Relative post-mortem sparing of afferent pupil fibers in a patient with 3460 Leber’s hereditary optic neuropathy. Graefes Arch Clin Exp Ophthalmol 243: 1175–1179. Bremner FD, Shallo-Hoffmann J, Riordan-Eva P et al. (1999). Comparing pupil function with visual function in patients with Leber’s hereditary optic neuropathy. Invest Ophthalmol Vis Sci 40: 2528–2534. Brown TM, Gias C, Hatori M et al. (2010). Melanopsin contributions to irradiance coding in the thalamo-cortical visual system. PLoS Biol 8: e1000558. Brown TM, Tsujimura S, Allen AE et al. (2012). Melanopsinbased brightness discrimination in mice and humans. Curr Biol 2: 1134–1141. Cajochen C, M€ unch M, Knoblauch V et al. (2006). Age-related changes in the circadian and homeostatic regulation of human sleep. Chronobiol Int 23: 461–474. Carelli V, Ross-Cisneros FN, Sadun AA (2004). Mitochondrial dysfunction as a cause of optic neuropathies. Prog Retin Eye Res 23: 53–89. Chahine LM, Amara AW, Videnovic A (2017). A systematic review of the literature on disorders of sleep and wakefulness in Parkinson’s disease from 2005 to 2015. Sleep Med Rev Sleep 35: 33–50. Chan VTT, Sun Z, Tang S et al. (2019). Spectral-domain OCT measurements in Alzheimer’s disease: a systematic review and meta-analysis. Ophthalmology 126: 497–510. Chen SK, Badea TC, Hattar S (2011). Photoentrainment and pupillary light reflex are mediated by distinct populations of ipRGCs. Nature 476: 92–95. Chougule PS, Najjar RP, Finkelstein MT et al. (2019). Lightinduced pupillary responses in Alzheimer’s disease. Front Neurol 10: 360. Czeisler CA, Shanahan TL, Klerman EB et al. (1995). Suppression of melatonin secretion in some blind patients by exposure to bright light. N Engl J Med 332: 6–11. Dacey DM, Liao HW, Peterson BB et al. (2005). Melanopsinexpressing ganglion cells in primate retina signal colour and irradiance and project to the LGN. Nature 433: 749–754. Daneault V, Vandewalle G, Hebert M et al. (2012). Does pupil constriction under blue and green monochromatic light exposure change with age? J Biol Rhythms 27: 257–264. Daneault V, Hebert M, Albouy G et al. (2014). Aging reduces the stimulating effect of blue light on cognitive brain functions. Sleep 37: 85–96.  et al. (2018). Plasticity in Daneault V, Dumont M, Masse E the sensitivity to light in aging: decreased non-visual impact of light on cognitive brain activity in older individuals but no impact of lens replacement. Front Physiol 9: 1557. De Pablo-Ferna´ndez E, Courtney R, Warner TT et al. (2018). A histologic study of the circadian system in Parkinson disease, multiple system atrophy, and progressive Supranuclear palsy. JAMA Neurol 75: 1008–1012.

RETINA AND MELANOPSIN NEURONS Delettre C, Lenaers G, Griffoin JM et al. (2000). Nuclear gene OPA1, encoding a mitochondrial dynamin-related protein, is mutated in dominant optic atrophy. Nat Genet 26: 207–210. Delwig A, Chaney SY, Bertke AS et al. (2018). Melanopsin expression in the cornea. Vis Neurosci 35: E004. Do MTH (2019). Melanopsin and the intrinsically photosensitive retinal ganglion cells: biophysics to behavior. Neuron 104: 205–22623. Do MTH, Yau KW (2010). Intrinsically photosensitive retinal ganglion cells. Physiol Rev 90: 1547–1581. Duncan MJ, Hester JM, Hopper JA et al. (2010). The effects of aging and chronic fluoxetine treatment on circadian rhythms and suprachiasmatic nucleus expression of neuropeptide genes and 5-HT1B receptors. Eur J Neurosci 31: 1646–1654. Duncan MJ, Prochot JR, Cook DH et al. (2013). Influence of aging on Bmal1 and Per2 expression in extra-SCN oscillators in hamster brain. Brain Res 1491: 44–5323. Ecker JL, Dumitrescu ON, Wong KY et al. (2010). Melanopsin-expressing retinal ganglion-cell photoreceptors: cellular diversity and role in pattern vision. Neuron 67: 49–60. Ejaz AA, Sekhon IS, Munjal S (2006). Characteristic findings on 24-h ambulatory blood pressure monitoring in a series of patients with Parkinson’s disease. Eur J Intern Med 17: 417–420. Emanuel AJ, Do MT (2015). Melanopsin tristability for sustained and broadband phototransduction. Neuron 85: 1043–1055. Esquiva G, Hannibal J (2019). Melanopsin-expressing retinal ganglion cells in aging and disease. Histol Histopathol 34: 1299–1311. Esquiva G, Lax P, Perez-Santonja JJ et al. (2017). Loss of Melanopsin-expressing ganglion cell subtypes and dendritic degeneration in the aging human retina. Front Aging Neurosci 9: 79. Farajnia S, Michel S, Deboer T et al. (2012). Evidence for neuronal desynchrony in the aged suprachiasmatic nucleus clock. J Neurosci 32: 5891–5899. Fifel K (2017). Alterations of the circadian system in Parkinson’s disease patients. Mov Disord 32: 682–692. Foster RG (2005). Neurobiology: bright blue times. Nature 433: 698–699. Foster RG, Provencio I, Hudson D et al. (1991). Circadian photoreception in the retinally degenerate mouse (rd/rd). J Comp Physiol A 169: 39–50. Freedman MS, Lucas RJ, Soni B et al. (1999). Regulation of mammalian circadian behavior by non-rod, non-cone, ocular photoreceptors. Science 284: 502–504. Georg B, Ghelli A, Giordano C et al. (2017). Melanopsinexpressing retinal ganglion cells are resistant to cell injury, but not always. Mitochondrion 36: 77–84. Gooley JJ, Lu J, Chou TC et al. (2001). Melanopsin in cells of origin of the retinohypothalamic tract. Nat Neurosci 4: 1165. Gooley JJ, Ho Mien I, St Hilaire MA et al. (2012). Melanopsin and rod-cone photoreceptors play different roles in mediating pupillary light responses during exposure to continuous light in humans. J Neurosci 32: 14242–14253.

325

Graham DM, Wong KY, Shapiro P et al. (2008). Melanopsin ganglion cells use a membrane-associated rhabdomeric phototransduction cascade. J Neurophysiol 99: 2522–2532. Hannibal J (2002). Neurotransmitters of the retinohypothalamic tract. Cell Tissue Res 309: 73–88. Hannibal J, Vrang N, Card JP et al. (2001). Light-dependent induction of cFos during subjective day and night in PACAP-containing ganglion cells of the retinohypothalamic tract. J Biol Rhythms 16: 457–470. Hannibal J, Hindersson P, Knudsen SM et al. (2002). The photopigment melanopsin is exclusively present in PACAP containing retinal ganglion cells of the retinohypothalamic tract. J Neurosci 22: 1–7. Hannibal J, Hindersson P, Ostergaard J et al. (2004). Melanopsin is expressed in PACAP-containing retinal ganglion cells of the human retinohypothalamic tract. Invest Ophthalmol Vis Sci 45: 4202–4209. Hannibal J, Georg B, Hindersson P et al. (2005). Light and darkness regulate melanopsin in the retinal ganglion cells of the albino Wistar rat. J Mol Neurosci 27: 147–155. Hannibal J, Christiansen AT, Heegaard S et al. (2017). Melanopsin expressing human retinal ganglion cells: subtypes, distribution, and intraretinal connectivity. J Comp Neurol 525: 1934–1961. Harnois C, Di Paolo T (1990). Decreased dopamine in the retinas of patients with Parkinson’s disease. Invest Ophthalmol Vis Sci 31: 2473–2475. Harper DG, Volicer L, Stopa EG et al. (2005). Disturbance of endogenous circadian rhythm in aging and Alzheimer disease. Am J Geriatr Psychiatry 13: 359–368. Harper DG, Stopa EG, Kuo-Leblanc V et al. (2008). Dorsomedial SCN neuronal subpopulations subserve different functions in human dementia. Brain 131: 1609–1617. Hart NJ, Koronyo Y, Black KL et al. (2016). Ocular indicators of Alzheimer’s: exploring disease in the retina. Acta Neuropathol 132: 767–787. Hartwick AT, Bramley JR, Yu J et al. (2007). Light-evoked calcium responses of isolated melanopsin-expressing retinal ganglion cells. J Neurosci 5; 27: 13468–13480. Hattar S, Liao HW, Takao M et al. (2002). Melanopsin-containing retinal ganglion cells: architecture, projections, and intrinsic photosensitivity. Science 295: 1065–1070. Hattar S, Kumar M, Park A et al. (2006). Central projections of melanopsin-expressing retinal ganglion cells in the mouse. J Comp Neurol 497: 326–349. Hayter EA, Brown TM (2018). Additive contributions of melanopsin and both cone types provide broadband sensitivity to mouse pupil control. BMC Biol 16: 83. Herbst K, Sander B, Lund-Andersen H et al. (2012). Intrinsically photosensitive retinal ganglion cell function in relation to age: a pupillometric study in humans with special reference to the age-related optic properties of the lens. BMC Ophthalmol 12: 4. Hinton DR, Sadun AA, Blanks JC et al. (1986). Optic-nerve degeneration in Alzheimer’s disease. N Engl J Med 315: 485–487.

326

C. LA MORGIA ET AL.

Hooghiemstra AM, Eggermont LH, Scheltens P et al. (2015). The rest-activity rhythm and physical activity in earlyonset dementia. Alzheimer Dis Assoc Disord 29: 45–49. Huang L, Xi Y, Peng Y et al. (2019). A visual circuit related to Habenula underlies the antidepressive effects of light therapy. Neuron 102: 128–142.e8. Jagannath A, Hughes S, Abdelgany A et al. (2015). Isoforms of Melanopsin mediate different behavioral responses to light. Curr Biol 25: 2430–2434. Jiang Z, Yue WWS, Chen L et al. (2018). Cyclic-nucleotideand HCN-channel-mediated phototransduction in intrinsically photosensitive retinal ganglion cells. Cell 175: 652–664.e12. Joyce DS, Feigl B, Kerr G et al. (2018). Melanopsin-mediated pupil function is impaired in Parkinson’s disease. Sci Rep 8: 7796. Jusuf PR, Lee SC, Hannibal J et al. (2007). Characterization and synaptic connectivity of melanopsin-containing ganglion cells in the primate retina. Eur J Neurosci 26: 2906–2921. Kallo´ I, Kalamatianos T, Piggins HD et al. (2004). Ageing and the diurnal expression of mRNAs for vasoactive intestinal peptide and for the VPAC2 and PAC1 receptors in the suprachiasmatic nucleus of male rats. J Neuroendocrinol 16: 758–766. Kankipati L, Girkin CA, Gamlin PD (2010). Post-illumination pupil response in subjects without ocular disease. Invest Ophthalmol Vis Sci 51: 2764–2769. Kawasaki A, Herbst K, Sander B et al. (2010). Selective wavelength pupillometry in Leber hereditary optic neuropathy. Clin Experiment Ophthalmol 38: 322–324. Keenan WT, Rupp AC, Ross RA et al. (2016). A visual circuit uses complementary mechanisms to support transient and sustained pupil constriction. Elife 5: e15392. Kelbsch C, Strasser T, Chen Y et al. (2019). Standards in pupillography. Front Neurol 10: 129. Kersten HM, Danesh-Meyer HV, Kilfoyle DH et al. (2015). Optical coherence tomography findings in Huntington’s disease: a potential biomarker of disease progression. J Neurol 262: 2457–2465. Kjer P (1959). Infantile optic atrophy with dominant mode of inheritance: a clinical and genetic study of 19 Danish families. Acta Ophthalmol Suppl 164: 1–147. Kolker DE, Fukuyama H, Huang DS et al. (2003). Aging alters circadian and light-induced expression of clock genes in golden hamsters. J Biol Rhythms 18: 159–169. Koronyo Y, Biggs D, Barron E et al. (2017). Retinal amyloid pathology and proof-of-concept imaging trial in Alzheimer’s disease. JCI Insight 2: e93621. Koronyo-Hamaoui M, Koronyo Y, Ljubimov AV et al. (2011). Identification of amyloid plaques in retinas from Alzheimer’s patients and noninvasive in vivo optical imaging of retinal plaques in a mouse model. Neuroimage 1: S204–S217. Koyanagi M, Kubokawa K, Tsukamoto H et al. (2005). Cephalochordate melanopsin: evolutionary linkage between invertebrate visual cells and vertebrate photosensitive retinal ganglion cells. Curr Biol 15: 1065–1069.

Krajnak K, Kashon ML, Rosewell KL et al. (1998). Aging alters the rhythmic expression of vasoactive intestinal polypeptide mRNA but not arginine vasopressin mRNA in the suprachiasmatic nuclei of female rats. J Neurosci 18: 4767–4774. Kuljis D, Kudo T, Tahara Y et al. (2018). Pathophysiology in the suprachiasmatic nucleus in mouse models of Huntington’s disease. J Neurosci Res 96: 1862–1875. La Morgia C, Ross-Cisneros FN, Sadun AA et al. (2010). Melanopsin retinal ganglion cells are resistant to neurodegeneration in mitochondrial optic neuropathies. Brain 133: 2426–2438. La Morgia C, Barboni P, Rizzo G et al. (2013). Loss of temporal retinal nerve fibers in Parkinson disease: a mitochondrial pattern? Eur J Neurol 20: 198–201. La Morgia C, Ross-Cisneros FN, Koronyo Y et al. (2016). Melanopsin retinal ganglion cell loss in Alzheimer disease. Ann Neurol 79: 90–109. La Morgia C, Ross-Cisneros FN, Sadun AA et al. (2017a). Retinal ganglion cells and circadian rhythms in Alzheimer’s disease, Parkinson’s disease, and beyond. Front Neurol 8: 162. La Morgia C, Di Vito L, Carelli V et al. (2017b). Patterns of retinal ganglion cell damage in neurodegenerative disorders: parvocellular vs magnocellular degeneration in optical coherence tomography studies. Front Neurol 8: 710. La Morgia C, Carelli V, Carbonelli M (2018). Melanopsin retinal ganglion cells and pupil: clinical implications for neuro-ophthalmology. Front Neurol 9: 1047. Lall GS, Revell VL, Momiji H et al. (2010). Distinct contributions of rod, cone, and melanopsin photoreceptors to encoding irradiance. Neuron 66: 417–428. Lane CA, Hardy J, Schott JM (2018). Alzheimer’s disease. Eur J Neurol 25: 59–70. Lax P, Esquiva G, Fuentes-Broto L et al. (2016). Agerelated changes in photosensitive melanopsin-expressing retinal ganglion cells correlate with circadian rhythm impairments in sighted and blind rats. Chronobiol Int 33: 374–391. Leber T (1871). Ueber hereditaere und congenital angelegte sehnervenleiden. Graefes Arch Clin Exp Ophthalmol 17: 249–291. Lee SK, Schmidt TM (2018). Morphological identification of Melanopsin-expressing retinal ganglion cell subtypes in mice. Methods Mol Biol 1753: 275–287. Leng Y, Musiek ES, Hu K et al. (2019). Association between circadian rhythms and neurodegenerative diseases. Lancet Neurol 18: 307–318. Li JY, Schmidt TM (2018). Divergent projection patterns of M1 ipRGC subtypes. J Comp Neurol 526: 2010–2018. Lin MS, Liao PY, Chen HM et al. (2019). Degeneration of ipRGCs in mouse models of Huntington’s disease disrupts non-image-forming Behaviors before motor impairment. J Neurosci 39: 1505–1524. Liu RY, Zhou JN, Hoogendijk WJ et al. (2000). Decreased vasopressin gene expression in the biological clock of Alzheimer disease patients with and without depression. J Neuropathol Exp Neurol 59: 314–322.

RETINA AND MELANOPSIN NEURONS Lucas RJ, Freedman MS, Mun˜oz M et al. (1999). Regulation of the mammalian pineal by non-rod, non-cone, ocular photoreceptors. Science 284: 505–507. Lupi D, Oster H, Thompson S et al. (2008). The acute lightinduction of sleep is mediated by OPN4-based photoreception. Nat Neurosci 11: 1068–1073. Matsuyama T, Yamashita T, Imamoto Y et al. (2012). Photochemical properties of mammalian melanopsin. Biochemistry 51: 5454–5462. Matynia A, Parikh S, Chen B et al. (2012). Intrinsically photosensitive retinal ganglion cells are the primary but not exclusive circuit for light aversion. Exp Eye Res 105: 60–69. Matynia A, Nguyen E, Sun X et al. (2016). Peripheral sensory neurons expressing Melanopsin respond to light. Front Neural Circuits 10: 60. Maury E (2019). Off the clock: from circadian disruption to metabolic disease. Int J Mol Sci 20: Pii: E1597. Moore RY, Lenn NJ (1972). A retinohypothalamic projection in the rat. J Comp Neurol 146: 1–14. Moore RY, Speh JC, Card JP (1995). The retinohypothalamic tract originates from adistinct subset of retinal ganglion cells. J Comp Neurol 352: 351–366. Morin LP, Studholme KM (2014). Retinofugal projections in the mouse. J Comp Neurol 522: 3733–3753. Morton AJ (2013). Circadian and sleep disorder in Huntington’s disease. Exp Neurol 243: 34–44. Moura AL, Nagy BV, La Morgia C et al. (2013). The pupil light reflex in Leber’s hereditary optic neuropathy: evidence for preservation of melanopsin-expressing retinal ganglion cells. Invest Ophthalmol Vis Sci 54: 4471–4477. M€ unch M, Kawasaki A (2013). Intrinsically photosensitive retinal ganglion cells: classification, function and clinical implications. Curr Opin Neurol 26: 45–51. Musiek ES, Holtzman DM (2016). Mechanisms linking circadian clocks, sleep, and neurodegeneration. Science 354: 1004–1008. Musiek ES, Bhimasani M, Zangrilli MA et al. (2018). Circadian rest-activity pattern changes in aging and preclinical Alzheimer disease. JAMA Neurol 75: 582–590. Nakamura TJ, Nakamura W, Yamazaki S et al. (2011). Age-related decline in circadian output. J Neurosci 31: 10201–10205. Nissil€a JS, M€antt€ari SK, S€arkioja TT et al. (2017). The distribution of melanopsin (OPN4) protein in the human brain. Chronobiol Int 34: 37–44. Niwa F, Kuriyama N, Nakagawa M et al. (2011). Circadian rhythm of rest activity and autonomic nervous system activity at different stages in Parkinson’s disease. Auton Neurosci 165: 195–200. Noseda R, Kainz V, Jakubowski M et al. (2010). A neural mechanism for exacerbation of headache by light. Nat Neurosci 13: 239–245. Oh AJ, Amore G, Sultan W et al. (2019). Pupillometry evaluation of melanopsin retinal ganglion cell function and sleep-wake activity in pre-symptomatic Alzheimer’s disease. PLoS One 14: e0226197.

327

Ortun˜o-Lizara´n I, Esquiva G, Beach TG et al. (2018a). Degeneration of human photosensitive retinal ganglion cells may explain sleep and circadian rhythms disorders in Parkinson’s disease. Acta Neuropathol Commun 6: 90. Ortun˜o-Lizara´n I, Beach TG, Serrano GE et al. (2018b). Phosphorylated a-synuclein in the retina is a biomarker of Parkinson’s disease pathology severity. Mov Disord 33: 1315–1324. Ouk K, Hughes S, Pothecary CA et al. (2016). Attenuated pupillary light responses and downregulation of opsin expression parallel decline in circadian disruption in two different mouse models of Huntington’s disease. Hum Mol Genet 25: 5418–5432. Ouk K, Aungier J, Ware M et al. (2019). Abnormal photic entrainment to phase-delaying stimuli in the R6/2 mouse model of Huntington’s disease, despite retinal responsiveness to light. eNeuro 6: ENEURO.0088-19.2019. Park JC, Moura AL, Raza AS et al. (2011). Toward a clinical protocol for assessing rod, cone, and melanopsin contributions to the human pupil response. Invest Ophthalmol Vis Sci 52: 6624–6635. Peter-Derex L, Yammine P, Bastuji H et al. (2015). Sleep and Alzheimer’s disease. Sleep Med Rev 19: 29–38. Pilorz V, Tam SK, Hughes S et al. (2016). Melanopsin regulates both sleep-promoting and arousal-promoting responses to light. PLoS Biol 14: e1002482. Pjrek E, Friedrich ME, Cambioli L et al. (2020). The efficacy of light therapy in the treatment of seasonal affective disorder: a meta-analysis of randomized controlled trials. Psychother Psychosom 89: 17–24. Politis M, Pavese N, Tai YF et al. (2008). Hypothalamic involvement in Huntington’s disease: an in vivo PET study. Brain 131: 2860–2869. Postuma RB, Berg D, Stern M et al. (2015). MDS clinical diagnostic criteria for Parkinson’s disease. Mov Disord 30: 1591–1601. Prayag AS, Najjar RP, Gronfier C (2019). Melatonin suppression is exquisitely sensitive to light and primarily driven by melanopsin in humans. J Pineal Res 66: e12562. Provencio I, Jiang G, De Grip WJ et al. (1998). Melanopsin: an opsin in melanophores, brain, and eye. Proc Natl Acad Sci U S A 95: 340–345. Provencio I, Rodriguez IR, Jiang G et al. (2000). A novel human opsin in the inner retina. J Neurosci 20: 600–605. Provencio I, Rollag MD, Castrucci AM (2002). Photoreceptive net in the mammalian retina. This mesh of cells may explain how some blind mice can still tell day from night. Nature 415: 493. Quattrochi LE, Stabio ME, Kim I et al. (2019). The M6 cell: a small-field bistratified photosensitive retinal ganglion cell. J Comp Neurol 527: 297–311. Roecklein KA, Rohan KJ, Duncan WC et al. (2009). A missense variant (P10L) of the melanopsin (OPN4) gene in seasonal affective disorder. J Affect Disord 114: 279–285. Romagnoli M, Stanzani Maserati M, De Matteis M et al. (2020). Chromatic pupillometry findings in Alzheimer’s disease. Front Neurosci 14: 780.

328

C. LA MORGIA ET AL.

Rukmini AV, Milea D, Aung T et al. (2017). Pupillary responses to short-wavelength light are preserved in aging. Sci Rep 7: 43832. Rukmini AV, Milea D, Gooley JJ (2019). Chromatic Pupillometry methods for assessing photoreceptor health in retinal and optic nerve diseases. Front Neurol 10: 76. Rupp AC, Ren M, Altimus CM et al. (2019). Distinct ipRGC subpopulations mediate light’s acute and circadian effects on body temperature and sleep. Elife 8: Pii: e44358. Sadun AA, Bassi CJ (1990). Optic nerve damage in Alzheimer’s disease. Ophthalmology 97: 9–17. Sadun AA, Schaechter JD, Smith LE (1984). A retinohypothalamic pathway in man: light mediation of circadian rhythms. Brain Res 302: 371–377. Sa´nchez-Migallo´n MC, Valiente-Soriano FJ, Nadal-Nicola´s FM et al. (2018). Survival of melanopsin expressing retinal ganglion cells long term after optic nerve trauma in mice. Exp Eye Res 174: 93–97. Sand A, Schmidt TM, Kofuji P (2012). Diverse types of ganglion cell photoreceptors in the mammalian retina. Prog Retin Eye Res 31: 287–302. Satinoff E, Li H, Tcheng TK et al. (1993). Do the suprachiasmatic nuclei oscillate in old rats as they do in young ones? Am J Physiol 265: R1216–R1222. Schmidt TM, Kofuji P (2009). Functional and morphological differences among intrinsically photosensitive retinal ganglion cells. J Neurosci 29: 476–482. Schmidt TM, Kofuji P (2010). Differential cone pathway influence on intrinsically photosensitive retinal ganglion cell subtypes. J Neurosci 30: 16262–16271. Schmidt TM, Taniguchi K, Kofuji P (2008). Intrinsic and extrinsic light responses in melanopsin-expressing ganglion cells during mouse development. J Neurophysiol 100: 371–384. Schmidt TM, Do MT, Dacey D et al. (2011). Melanopsinpositive intrinsically photosensitive retinal ganglion cells: from form to function. J Neurosci 31: 16094–16101. Sekaran S, Lupi D, Jones SL et al. (2005). Melanopsindependent photoreception provides earliest light detection in the mammalian retina. Curr Biol 15: 1099–1107. Semo M, Lupi D, Peirson SN et al. (2003). Light-induced c-fos in melanopsin retinal ganglion cells of young and aged rodless/coneless (rd/rd cl) mice. Eur J Neurosci 18: 3007–3017. Serrano-Pozo A, Frosch MP, Masliah E et al. (2011). Neuropathological alterations in Alzheimer disease. Cold Spring Harb Perspect Med 1: a006189. Skene DJ, Swaab DF (2003). Melatonin rhythmicity: effect of age and Alzheimer’s disease. Exp Gerontol 38: 199–206. Sletten TL, Revell VL, Middleton B et al. (2009). Age-related changes in acute and phase-advancing responses to monochromatic light. J Biol Rhythms 24: 73–84. Sonoda T, Schmidt TM (2016). Re-evaluating the role of intrinsically photosensitive retinal ganglion cells: new roles in image-forming functions. Integr Comp Biol 56: 834–841. Spitschan M (2019). Melanopsin contributions to non-visual and visual function. Curr Opin Behav Sci 30: 67–72.

Spitschan M, Jain S, Brainard DH et al. (2014). Opponent melanopsin and S-cone signals in the human pupillary light response. Proc Natl Acad Sci U S A 111: 15568–15572. Spitschan M, Bock AS, Ryan J et al. (2017). The human visual cortex response to melanopsin-directed stimulation is accompanied by a distinct perceptual experience. Proc Natl Acad Sci U S A 114: 12291–12296. Stopa EG, Volicer L, Kuo-Leblanc V et al. (1999). Pathologic evaluation of the human suprachiasmatic nucleus in severe dementia. J Neuropathol Exp Neurol 58: 29–39. 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). (1987). Suprachiasmatic nucleus in aging, Alzheimer’s disease, transsexuality and Prader-Willi syndrome. Prog Brain Res 72: 301–310. Swaab DF, Fisser B, Kamphorst W et al. (1988). The human suprachiasmatic nucleus; neuropeptide changes in senium and Alzheimer’s disease. Basic Appl Histochem 32: 43–54. Tarttelin EE, Bellingham J, Bibb LC et al. (2003). Expression of opsin genes early in ocular development of humans and mice. Exp Eye Res 76: 393–396. Tranah GJ, Blackwell T, Stone KL et al. (2011). Circadian activity rhythms and risk of incident dementia and mild cognitive impairment in older women. Ann Neurol 70: 722–732. Tsai JW, Hannibal J, Hagiwara G et al. (2009). Melanopsin as a sleep modulator: circadian gating of the direct effects of light on sleep and altered sleep homeostasis in Opn4 ( / ) mice. PLoS Biol 7: e1000125. Vallelonga F, Di Stefano C, Merola A et al. (2019). Blood pressure circadian rhythm alterations in alpha-synucleinopathies. J Neurol 266: 1141–1152. van Wamelen DJ, Aziz NA, Anink JJ et al. (2013). Suprachiasmatic nucleus neuropeptide expression in patients with Huntington’s disease. Sleep 36: 117–125. van Wamelen DJ, Aziz NA, Roos RA et al. (2014). Hypothalamic alterations in Huntington’s disease patients: comparison with genetic rodent models. J Neuroendocrinol 26: 761–775. Vandewalle G, Schmidt C, Albouy G et al. (2007). Brain responses to violet, blue, and green monochromatic light exposures in humans: prominent role of blue light and the brainstem. PLoS One 2: e1247. Vandewalle G, Schwartz S, Grandjean D et al. (2010). Spectral quality of light modulates emotional brain responses in humans. Proc Natl Acad Sci U S A 107: 19549–19554. Vandewalle G, Archer SN, Wuillaume C et al. (2011a). Effects of light on cognitive brain responses depend on circadian phase and sleep homeostasis. J Biol Rhythms 26: 249–259. Vandewalle G, Hebert M, Beaulieu C et al. (2011b). Abnormal hypothalamic response to light in seasonal affective disorder. Biol Psychiatry 70: 954–961. Veys L, Vandenabeele M, Ortun˜o-Lizara´n I et al. (2019). Retinal a-synuclein deposits in Parkinson’s disease

RETINA AND MELANOPSIN NEURONS patients and animal models. Acta Neuropathol 137: 379–395. Viney TJ, Balint K, Hillier D et al. (2007). Local retinal circuits of melanopsin-containing ganglion cells identified by transsynaptic viral tracing. Curr Biol 17: 981–988. Volicer L, Harper DG, Manning BC et al. (2001). Sundowning and circadian rhythms in Alzheimer’s disease. Am J Psychiatry 158: 704–711. Wakakura M, Yokoe J (1995). Evidence for preserved direct pupillary light response in Leber’s hereditary optic neuropathy. Br J Ophthalmol 79: 442–446. Wang JL, Lim AS, Chiang WY et al. (2015). Suprachiasmatic neuron numbers and rest-activity circadian rhythms in older humans. Ann Neurol 78: 317–322. Wang HB, Whittaker DS, Truong D et al. (2017). Blue light therapy improves circadian dysfunction as well as motor symptoms in two mouse models of Huntington’s disease. Neurobiol Sleep Circadian Rhythms 2: 39–52. Wang S, Gu D, Zhang P et al. (2018). Melanopsin-expressing retinal ganglion cells are relatively resistant to excitotoxicity induced by N-methyl-d-aspartate. Neurosci Lett 662: 368–373. Weil RS, Schrag AE, Warren JD et al. (2016). Visual dysfunction in Parkinson’s disease. Brain 139: 2827–2843. Witting W, Kwa IH, Eikelenboom P et al. (1990). Alterations in the circadian rest-activity rhythm in aging and Alzheimer’s disease. Biol Psychiatry 27: 563–572.

329

Wong KY (2012). A retinal ganglion cell that can signal irradiance continuously for 10 hours. J Neurosci 32: 11478–11485. Wu YH, Swaab DF (2005). The human pineal gland and melatonin in aging and Alzheimer’s disease. J Pineal Res 38: 145–152. Wu YH, Zhou JN, Van Heerikhuize J et al. (2007). Decreased MT1 melatonin receptor expression in the suprachiasmatic nucleus in aging and Alzheimer’s disease. Neurobiol Aging 28: 1239–1247. Yu-Wai-Man P, Votruba M, Burte F et al. (2016). A neurodegenerative perspective on mitochondrial optic neuropathies. Acta Neuropathol 132: 789–806. Zaidi FH, Hull JT, Peirson SN et al. (2007). Short-wavelength light sensitivity of circadian, pupillary, and visual awareness in humans lacking an outer retina. Curr Biol 17: 2122–2128. Zhang Y, Kornhauser JM, Zee PC et al. (1996). Effects of aging on light-induced phase-shifting of circadian behavioral rhythms, fos expression and CREB phosphorylation in the hamster suprachiasmatic nucleus. Neuroscience 70: 951–961. Zhao X, Stafford BK, Godin AL et al. (2014). Photoresponse diversity among the five types of intrinsically photosensitive retinal ganglion cells. J Physiol 592: 1619–1636. Zhou JN, Hofman MA, Swaab DF (1995). VIP neurons in the human SCN in relation to sex, age, and Alzheimer’s disease. Neurobiol Aging 16: 571–576.

This page intentionally left blank

Handbook of Clinical Neurology, Vol. 179 (3rd series) The Human Hypothalamus: Anterior Region D.F. Swaab, F. Kreier, P.J. Lucassen, A. Salehi, and R.M. Buijs, Editors https://doi.org/10.1016/B978-0-12-819975-6.00021-2 Copyright © 2021 Elsevier B.V. All rights reserved

Chapter 22

Melatonin and the circadian system: Keys for health with a focus on sleep PAUL PEVET*, ETIENNE CHALLET, AND MARIE-PAULE FELDER-SCHMITTBUHL Institute of Cellular and Integrative Neurosciences, CNRS, University of Strasbourg, Strasbourg, France

Abstract Melatonin (MLT), secreted during the night by the pineal gland, is an efferent hormonal signal of the master circadian clock located in the suprachiasmatic nucleus (SCN). Consequently, it is a reliable phase marker of the SCN clock. If one defines as “chronobiotic,” a drug able to influence the phase and/or the period of the circadian clock, MLT is a very potent one. The most convincing data obtained so far come from studies on totally blind individuals. Exogenous MLT administered daily entrains the sleep–wake cycle of these individuals to a 24-h cycle. MLT, however, is not essential to sleep. In nocturnally, active mammals, MLT is released during the night concomitantly with the daily period of wakefulness. Therefore, MLT cannot be simply considered as a sleep hormone, but rather as a signal of darkness. Its role in the circadian system is to reinforce nighttime physiology, including timing of the sleep–wake cycle and other circadian rhythms. MLT exerts its effects on the sleep cycle especially by a direct action on the master circadian clock. The sleep–wake cycle is depending not only on the circadian clock but also on an orchestrated network of different centers in the brain. Thus, the control of sleep–wake rhythm might be explained by a parallel and concomitant action of MLT on the master clock (chronobiotic effect) and on sleep-related structures within the brain. MLT acts through two high-affinity membrane receptors (MT1 and MT2) with striking differences in their distribution pattern. MLT is a powerful synchronizer of human circadian rhythms, thus justifying the use of MLT and MLT agonists in clinical medicine as pharmacological tools to manipulate the sleep–wake cycle, and to treat sleep disorders and other circadian disorders. Available MLT analogs/drugs are all nonspecific MT1/MT2 agonists. The development of new ligands which are highly selectivity for each subtype is clearly a new challenge for the field and will be at the root of new therapeutic agents for curing specific pathologies, including sleep disorders.

INTRODUCTION The importance of circadian and seasonal rhythmicity for human health and welfare is becoming increasingly recognized (Schernhammer et al., 2006; Zelinski et al., 2014; Stenvers et al., 2019). Developing strategies to treat or prevent circadian disturbances is a new challenge for science and medicine. The mechanisms used for the daily coordination of physiological and cellular functions are far from being well understood. We know, however, that daily rhythms in physiological and behavioral

processes are controlled by a network of hierarchically organized clocks. In mammals, at the top of the network is a master clock located in the suprachiasmatic nuclei (SCN) of the hypothalamus, mainly reset by the daynight cycle. The SCN clock signals are forwarded through nervous, humoral, and hormonal pathways to reach specialized target structures in the brain and peripheral tissues (Buijs and Kalsbeek, 2001; Dibner et al., 2010). One important structure in this network is the pineal gland. In all mammals, both diurnal and nocturnal species, this gland synthesizes and secretes the

*Correspondence to: Paul Pevet, Institute of Cellular and Integrative Neurosciences, CNRS, University of Strasbourg, Strasbourg, France. Tel: (+33)-06-32-98-25-71, E-mail: [email protected]

332

P. PEVET ET AL.

hormone melatonin (MLT) at night. The finding that MLT is both a SCN clock output and an internal time giver in the circadian network (Pevet and Challet, 2011) strongly suggests that a major function of MLT is to maintain proper internal synchronization within the multioscillatory circadian network. In the present review, we will focus on the role and mechanisms of action of MLT in this context.

MELATONIN: IDENTIFICATION AND SYNTHESIS The chemical nature of the hormone MLT, isolated from bovine pineal gland, was identified as N-acetyl5-methoxytryptamine (Lerner et al., 1958). MLT is synthesized from the essential aminoacid tryptophan which is first converted into 5-hydroxytryptophan by tryptophan-5-hydroxylase before being decarboxylated into serotonin (5-hydroxytryptamine, 5-HT). Subsequently, two major enzymatic steps are involved. The first one is N-acetylation by arylalkylamine N-acetyltransferase (AA-NAT) to yield N-acetylserotonin. In rodents, the regulation of AA-NAT, with its sharp increase in activity at night, is the major regulatory step for rhythmic MLT synthesis. The second step is the transfer of a methyl group to the 5-hydroxy group of N-acetylserotonin, catalyzed by N-acetylserotonin-O-methyltransferase (ASMT) to yield MLT. Due to its amphiphilic nature, MLT is not stored inside pineal cells but is immediately released into the general circulation. Therefore, circulating MLT concentrations precisely reflects pineal synthesis. In 1965, the “melatonin hypothesis” proposed that the pineal gland acts as a transducer of changes in ambient light exposure/duration by adjusting its rates of MLT synthesis and release (Wurtman and Axelrod, 1965). Although the role of sympathetic pineal innervation in the synthesis of MLT was clearly demonstrated (Arendt, 2005), it was only in recent decades that the whole pathway was mapped. In short, the synthesis of MLT is driven directly by the SCN through a multisynaptic neural pathway, which successively includes preautonomic neurons of the paraventricular nucleus of the hypothalamus (PVN), sympathetic preganglionic neurons of the intermediolateral cell column of the spinal cord, and noradrenergic sympathetic neurons of the superior cervical ganglion. The respective role of each relay station, the identification of the different neurotransmitters used in each step as well as their specific daily pattern of release are now well defined, at least in the rat (Kalsbeek et al., 2006). To control the daily rhythm of MLT synthesis, the SCN uses a combination of rhythmic daytime inhibitory and constant stimulatory signals in the PVN-pineal pathway. GABA is responsible for the daytime inhibitory

message, whereas glutamate provides the nighttime stimulatory message (Perreau-Lenz et al., 2003). Pineal MLT synthesis is timed every day by the SCN. Accordingly, it is an efferent hormonal signal from the master clock which provides an endocrine message to the organism through the systemic circulation (Arendt, 2005). The duration of the nocturnal peak of MLT also follows the length of the night. This unique trait turns melatonin into an internal synchronizer that correctly matches the organism’s physiology to the daily and seasonal demands (Dardente et al., 2019). MLT plays thus an important role as a time-cue for the circadian system. The present chapter will focus on this feature. It should be noted that the pineal gland is not the only source of MLT. In mammals, MLT synthesis has also been described in several brain and nonneural structures, including the nucleus gracilis, pons, medulla oblongata, cerebellum, retina, adrenal gland, Harderian gland, lacrymal gland, thyroid gland, ovary, carotic body, gastrointestinal tract (ileum, colon, biliary cells, cholangiocytes), testis, skin, liver, kidney, adrenal pancreas, thymus, airway epithelium, choroid plexus, spleen, cochlea, lens, mast cells red blood cells, platelets, mononuclear cells, and endothelial cells (Acuna-Castroviejo et al., 2014). Recently, it has even been suggested that MLT is synthesized in large amounts within the mitochondrial matrix of the mouse brain cells, thereby activating a specific mitochondrial signal-transduction pathway (Suofu et al., 2017). Although most of these data need further validation, the physiological role of nonpineal MTL remains unknown. In most of these structures, MLT is not rhythmically synthesized and may not be released into the systemic circulation, at least under physiological conditions. Further research is required to determine the exact role of this nonpineal MLT, but this is beyond the scope of this review.

MELATONIN: ROLE, MECHANISMS OF ACTION, AND CLINICAL PERSPECTIVES Since its discovery, MLT is described as a hormone or a pharmacological agent with a plethora of functions (Reiter et al., 2010).The hormone’s high amphipathicity enables it to penetrate all peripheral organs or tissues throughout the body, all structures within the brain, as well as all compartments within cells. Actions of MLT have been reported to be mediated via interactions with specific intracellular proteins such as calmodulin, calreticulin, and tubulin. This opens a lot of physiological and clinical perspectives, but further investigation is needed to define the exact role of these interactions. The wide distribution of MLT synthesizing organs/structures also raises the question of autocrine or paracrine effects of the hormone. MLT might act locally or close to sites where it is synthesized. This is probable, especially when

MELATONIN AND THE CIRCADIAN SYSTEM extrapineal sources are considered. Local effects have been invoked with regard to metabolism, immune function, gut function, inflammation, membrane fluidity, mitochondrial function, or apoptosis. The best studied example is retinal MTL, which is rhythmically produced and inhibits nocturnal release of retinal dopamine (Dubocovich, 1983). Conversely, stimulation of retinal dopamine receptors suppresses the nocturnal increase of AA-NAT enzyme activity, thus inhibiting melatonin synthesis in the retina. Another example of potential clinical interest is mitochondria. There is evidence (which requires confirmation) that MLT is produced in high amounts in the mitochondrial matrix, diffuses out, and interacts with MT1 receptors on the outer membrane of the mitochondria. The term “automitocrine” was created to define this process (Suofu et al., 2017). High concentrations of MLT (mM and above) act as potent free radical scavenger (Reiter et al., 2007). The physiological significance of these antioxidant properties still remains to be established. Nevertheless, they have gathered increasing attention for the potential use of MLT as a pharmacological drug to treat various pathologies (Reiter et al., 2010). Accordingly, therapeutic protective effects are reported with regard to various neural, oncological, and cardiovascular systems. Binding of MLT to quinone reductase (QR2) shows that antioxidant activity can also result indirectly from increased activity of endogenous antioxidant enzymes (Boutin, 2015). Even if these antioxidant mechanisms of action are poorly understood, they can partly explain the multiple functions described in the literature for high doses of exogenously administered MLT. As indicated previously, the well-established role of MLT is its action as a hormonal transducer of photic information, and this effect is mediated through activation of MLT receptors. We will now concentrate our analysis on this mechanism.

Role of endogenous melatonin in the circadian system Pineal MLT synthesis is timed on a daily basis by the SCN and is thus an efferent endocrine output of the master clock. Directly and easily measurable in plasma and saliva (or indirectly in urine through 6-sulfatoxy-melatonin, the main hepatic metabolite of MLT), circulating rhythms of MLT provide a reliable phase markers of the SCN clock. This is of special interest in clinical research. Today, this feature is extensively used in humans to investigate the characteristics of circadian rhythms and to evaluate the circadian phase of a subject’s master clock (Arendt, 2005). The characteristics of MLT secretion in normal healthy volunteers have been studied for many years. Abnormalities in MLT rhythm have been related to various

333

pathologies (Arendt, 2019). Correct timing of treatment with MLT for disrupted rhythms is therefore possible. The important question of optimal circadian timing for drug treatment has become crucial in personalized medicine and individualization of treatment regimes, especially for cancer therapy. Correlation with a given pathology is, however, difficult, since low MLT levels do not necessarily constitute a specific symptom, but rather reflect a rhythm status. MLT profiling is extensively used in medical research to provide a way of normalizing experimental subjects with diverse angles of entrainment (see Arendt, 2019). The nocturnal peak of MLT also plays a role as an internal time giver on a daily basis (Pevet and Challet, 2011). The daily rhythm of plasma MLT distributes a circadian signal generated by the master clock throughout the body, thus playing a role in coupling between the master and secondary clocks (Pevet and Challet, 2011). For example, circadian rhythmicity of pups during gestation and weaning is synchronized by maternal MLT through the placenta and milk (Torres-Farfan et al., 2008). In adults, MLT (or MLT agonist) can inhibit spontaneous and light-evoked activity of cells in the intergeniculate leaflets of the thalamus (Ying et al., 1993), stimulate splenic lymphocyte proliferation, inhibit leucocyte rolling and adhesion in rat blood capillaries, influence response of Leydig and luteal cells, induce vasoconstriction of cerebral and tail arteries (Lotufo et al., 2001), glucagon secretion by pancreatic cells in vivo (B€ahr et al., 2011), rhythmic protein synthesis in hepatocytes in vivo (Alonso-Vale et al., 2005), circadian modulation of sodium–potassium-ATPase and sodium–proton exchangers in human erythrocytes (Chakravarty and Rizvi, 2011), and a phase-dependent increased amplitude of clock gene oscillations within the cultured skin fibroblasts (Sandu et al., 2015). Pineal MLT secreted at night can also impact daily rhythmicity of metabolic hormones and glucose through a complex physiological regulation. In vivo MLT is a time giver for the rhythmic secretion of leptin in hamsters (Chakir et al., 2015). Indeed, without circulating MLT at night (e.g., after pinealectomy), the daily rhythm of plasma leptin is blunted. In that study, it was noticed that pinealectomy also affects daily rhythms of plasma cortisol and insulin. The mechanism involved however may not be a direct one. Most if not all peripheral organs are sensitive to internal time givers, hormonal, or not (e.g., MLT, glucocorticoids, rest/activity cycle, feeding behavior, body temperature rhythm). This does not mean, however, that these organs are all sensitive to all circadian cues. Some appear to be more specifically sensitive to one-time giver than others. In the mouse liver for instance, glucocorticoids play a strong synchronizing effect, as shown by arhythmicity in metabolic gene expression after

334

P. PEVET ET AL.

adrenalectomy (Oishi et al., 2005), whereas MLT cues do not markedly affect hepatic timing (Houdek et al., 2015). In the mouse brain, the daily variation of dopamine in the striatum is controlled by rhythmic MLT, as evidenced by disappearance of rhythmicity after pineal ablation and restoration with daily injections of MLT, while adrenalectomy does not impair timing of these dopaminergic variations (Khaldy et al., 2002). This is similar to the daily variations of serotonin in rodent brain, which are controlled by rhythmic glucocorticoids (Malek et al., 2007) but are not affected by pinealectomy or daily injections of MLT (Khaldy et al., 2002). On the other hand, in primary isolated adipocytes, both glucocorticoids and MLT can modulate timing of daily profiles of metabolic gene expression (Alonso-Vale et al., 2005; Su et al., 2014) or hormonal output such as leptin (Alonso-Vale et al., 2005; Chakir et al., 2015). Other in vitro studies on isolated organs are in accordance with this interpretation (Peschke and Peschke, 1998; Bering et al., 2020). All these data favor the hypothesis that rhythmic MLT acts, directly or indirectly, via other internal cues, as an internal circadian synchronizer. The exact functional implications of these specific temporal regulations for physiology and health remain to be investigated. To further dissect the circadian role of MLT, we need to identify and to study structures in which the temporal organization of a response is exclusively dependent on the MLT signal. In the context of the multioscillatory nature of the circadian system, two possibilities have to be considered: the MLT signal either directly drives a rhythm (see the example reported on the Pars tuberalis) or entrains central and/or peripheral oscillators. As shown by the disappearance of clock gene oscillations after pinealectomy, nocturnal MLT appears to directly control clock gene rhythm in some structures such as striatum, spleen, and pars tuberalis of the hypophysis (Messager et al., 2001). Melatoninergic cues may even feedback on the master clock in the SCN in which they affect not only clock gene expression (Agez et al., 2007) but also firing rate (Rusak and Yu, 1993). In addition, MLT also induces a shift in the 24-h oscillatory expression of two clock genes, Bmal1 and Per2, in cultured fetal adrenal gland (TorresFarfan et al., 2011). Clock gene expression is widespread in mammalian tissues, but these tissues do not always exhibit cell-autonomous self-sustaining rhythmicity. It appears that cyclic expression in some of the peripheral structures might be driven by the SCN through the MLT nocturnal peak. The pars tuberalis is the best studied example of that mechanism. Removal of the pineal abolishes clock gene oscillations which are also undetectable in MLT-deficient strains of mice (Stehle et al., 2002). Using an acute injection paradigm, Dardente et al. (2003) observed that expression of the clock gene

Cry1 is strongly but transiently induced by MLT, independently of the administration time. In nontreated animals, a peak of expression occurs during the dark phase (i.e., when MLT is present in the blood stream), indicating that MLT gates Cry1 expression. On the other hand, Per1 mRNA levels peak early in the day, when blood plasma MLT levels are at very low levels. Thus, Per1 expression in the pars tuberalis appears to be linked to the offset of MLT secretion. This dual effect of MLT not only explains the rhythmic expression of this molecular machinery but, considering the photoperioddependent pattern of MLT secretion, may also help to understand how the pars tuberalis is involved in seasonal control of photoperiodic functions (Dardente et al., 2010). The pars tuberalis can thus be defined as a MLT-driven oscillator. However, it cannot be fully excluded that MLT is actually needed to synchronize the individual cellular oscillators within the pars tuberalis. It is still an open question whether this hormonal drive of daily rhythmicity in a tissue devoid of direct neuronal connections with the SCN clock is specific to the pars tuberalis or is a more general mechanism also present in other structures. The example of the pars tuberalis indicates that it would be appropriate to systematically analyze the expression of Cry1, Per1, and other clock genes in additional structures containing MLT receptors in pinealectomized rodents or in MLT-deficient mice. Clearly further fundamental research is warranted to delineate the synchronizing effects of MLT on cerebral and peripheral oscillators. Most of the results described previously concern the role of endogenous MLT. In regard to the potential therapeutic use of MLT or MLT derivatives, the effect of exogenous MLT must also be considered.

Role of exogenous melatonin: From hormone to chronobiotic drug In most mammals, MLT receptors are present in the SCN, allowing circulating MLT to feedback onto the master clock. This feedback is thought to play an important role in the long-term functioning of the circadian system (e.g., aging). To date, however, its physiological importance has not been experimentally established. Of note, the presence of MLT receptors within the SCN indicates that exogenous MLT can affect circadian regulation. It has long been demonstrated that daily administration of MLT acts on activity rhythms in rodents For example, MLT administered daily by subcutaneous injection (Redman et al., 1983), infusion (Pitrosky et al., 1999), or via drinking water (Slotten et al., 1999) at the subjective dusk entrains the free-running rhythm of locomotor activity in rats housed in constant darkness.

MELATONIN AND THE CIRCADIAN SYSTEM If one defines as “chronobiotic,” a drug able to influence the phase and/or the period of the circadian clock and adjust directly or indirectly the timing of internal rhythms; MLT is a very potent chronobiotic drug (Pevet et al., 2002; Arendt and Skene, 2005). Actually, MLT is one of the most powerful synchronizers of human circadian rhythms. In addition, its lack of reported toxicity or addictive properties and its efficacy (Arendt, 2019) justify the use of MLT and MLT agonists in clinical medicine as pharmacological tools to manipulate the sleep– wake cycle, sleep disorders (e.g., delayed sleep phase syndrome, advanced sleep phase syndrome), and other circadian disorders. For example, in aged individuals with lower-amplitude MLT rhythms, evening supplementation with MLT improves sleep, morning alertness, cognitive performance, and quality of life (Skene et al., 1990; Skene and Swaab, 2003; Riemersma-van der Lek et al., 2008; Emens and Burgess, 2015). The same benefit may apply to patients with neurodegenerative or psychiatric disorders, cancers, cardiovascular, digestive, or immune dysfunctions, which all display troubles in the sleep–wake cycle (Wu and Swaab, 2007; Wu et al., 2013). The most convincing chronobiotic data obtained so far come from studies on totally blind individuals. Such individuals experience sleep disturbances because their activity rhythms can fall out of phase with social cues for sleeping and wakefulness. Exogenous MLT administered daily entrains the sleep–wake cycle of these individuals to a 24-h cycle (Skene and Arendt, 2007).

Melatonin and sleep MLT is often referred to as the sleep hormone. Even if exogenous MLT has somnogenic properties, humans sleep better when rhythmic circulating melatonin is in phase with the circadian system (Arendt, 2019). MLT is not essential to sleep. In pinealectomized humans (e.g., after pineal tumor ablation), no effects on sleep were described (Slawik et al., 2016). More importantly, nocturnally active mammals (most of the rodents used in laboratories) produce MLT during the night, the hormone release being concomitant with the daily period of wakefulness. Therefore, MLT cannot be a sleep hormone but rather signals darkness. The role of MLT in the circadian system is to reinforce nighttime physiology, including timing of the sleep–wake cycle and other circadian rhythms (Pevet, 2016; Arendt, 2019). In response to timed daily MLT treatments, phase shifts and entrainment of the circadian clock were shown in nocturnal and diurnal mammals. MLT treatment in the late afternoon in humans advances circadian phase and consequently sleep onset, whereas treatment in early morning causes phase delays (Arendt, 2019).

335

In accordance with its effect on sleep onset latency, but not on sleep efficiency, MLT exerts its effects by a direct resetting of the master clock rather than via a possible action on somnogenic structures within the brain. Sleep plays a major role in the restoration of functions critical for physical and mental health, and cognitive processes, such as brain energy, facilitation of the plasticity of cerebral changes that underlie learning, memory consolidation, and extinction (Arendt, 2005). MLT, by its action on the SCN, is therefore an important physiological sleep regulator in mammals, including humans. The sharp increase in sleep propensity at night usually occurs 2 h after the onset of endogenous MLT production in humans (Zisapel, 2007). This explains why in clinical settings it is recommended administering MLT during late afternoon or beginning of the evening. It should be pointed out that in the clinic MLT treatment has often found to be effective in problems/ diseases associated with sleep–wake cycle disorders (e.g., children with autism spectrum disorders and neurogenetic neurodevelopmental disorders, insomnia in adults, insomnia in the elderly, depression, schizophrenia, Alzheimer’s disease, jet-lag, and shift work) (Wu et al., 2013; Elrod and Hood, 2015; Arendt, 2019). MLT treatment may also be used in the case of metabolic disorders (obesity, diabetes, hypercholesterolemia), as well as cardiovascular disease and cancer (Schernhammer et al., 2006; Sookoian et al., 2007; Dochi et al., 2009). For example, in a controlled experiment, exogenous MLT was able to shift heart rate variability together with major circadian rhythms of cortisol, core body temperature, and TSH. This corresponds well to an effect that would be mediated by the master circadian clock (Scheer et al., 2012). Conversely, certain drugs (e.g., b-blockers, naloxone, and nonsteroidal antiinflammatory drugs) that affect the nocturnal production of MLT are also associated with impaired sleep (Waldhauser et al., 1998). There are now many reviews and meta-analyses on the effects of exogenous MLT on pathologies in humans and animal models. Thus it is possible to clearly identify areas in which a consensus exists. MLT exerts its effects on sleep especially by a direct action on the master circadian clock. This conclusion is supported by the observation that the effect of MLT on sleep onset latency is not associated with a change in the amount of slow-wave sleep, a marker of homeostatic sleep pressure (Zisapel, 2007; Arbon et al., 2015). This justifies the use of MLT or MLT agonists for treating chronic disorders of sleep–wake balance. Currently there are expanding possibilities for therapeutic applications to all disturbances of sleep–wake cycle-associated pathologies (De Leersnyder et al., 2006; Cortesi et al., 2012; Tordjman et al., 2013; Cuomo et al., 2017).

336

P. PEVET ET AL.

MLT is available without prescription in many countries especially for the treatment of some sleep disorders and depression. It can be administered, orally and sublingually, as capsules, pills, and transdermal patches. The effects obtained differ depending on whether the delivery is rapid or involving slow release during the night (see Zisapel, 2018). Because most data favor an action of MLT on sleep through the chronobiotic effect, it is necessary to determine the sites and mechanisms of action of the hormone under these conditions.

Sites and mechanisms of action mediating the chronobiotic effects of melatonin With regard to its chronobiotic effect, several lines of evidence support the view that the primary site of action of MLT is the SCN. It is indeed affected by exogenous MLT in vivo and in vitro, as shown by the phase-shifting effects of the firing rate of SCN neurons (Gillette and McArthur, 1996). SCN-lesioned hamsters whose rhythmicity is restored with fetal hypothalamic grafts are also entrained by daily MLT injections (Grosse and Davis, 1988). Furthermore, MLT accelerates the reentrainment of circadian rhythms in rodents subjected to a shift in the light/dark cycle. If the chronobiotic effect of MLT results from a direct action on the SCN clock, an effect on SCN clock outputs is to be expected. Pineal MLT synthesis is one of such outputs. When exogenous MLT is administered in vivo directly within the SCN, a phase advance of the endogenous nocturnal MLT peak is observed measured by transpineal microdialysis performed over four consecutive days. Similar findings are obtained with a MLT agonist, agomelatine. Interestingly, a significant increase in the amplitude of the MLT peak, which persists for at least 3 days, is also observed, suggesting a direct action of MLT on the amplitude of clock oscillations (Bothorel et al., 2002; Castano et al., 2014). Similar observations have been made in humans (Zaidan et al., 1994). Chronobiotic effects of MLT imply a direct action of the hormone on high-affinity MLT receptors located within the SCN. There is a strong correlation between the density of MLT receptors within the SCN and the ability of daily MLT administration to entrain the free-running activity rhythm in mammals. Unlike the rat and Djungarian hamster, which can be entrained by daily MLT and in which a high density of MLT receptors within the SCN is observed, the mink does not appear to have specific MLT receptors within the SCN and cannot be entrained by MLT. Furthermore, newborn Syrian hamsters express MLT receptors in the SCN, but shortly after birth, the number of MLT receptors decreases. Young hamsters are entrainable by daily acute MLT administration (Grosse and

Davis, 1988), whereas MLT does not entrain circadian rhythms in adult Syrian hamsters. Which MLT receptor subtypes are involved? Siberian hamsters lack functional MT2 receptors but still show circadian responses to MLT. Similarly, the robust entraining response to MLT, i.e., synchronization of developing circadian clocks in Syrian hamsters by MLT injections occurs in the absence of a functional MT2 receptor within the SCN. This strongly suggests the involvement of MT1 receptors. However, in vitro experiments in animal models which possess both subtypes (rats and mice) demonstrate that the mechanisms involved are more complex. For example, two distinct effects of MLT have been described: an acute inhibitory effect on neuronal firing and a phase-shifting effect in rhythm of electrical activity. In mice with a targeted deletion of the MT1 receptor, the acute inhibitory effect of MLT is abolished, while the phase-shifting effect remains intact. In addition, the phase-shift effect also disappears when the MT2 selective antagonist 4P-PDOT is provided. Therefore, MT2 receptors in the SCN were long considered to be responsible for the phase-shifting effects of MLT, a conclusion supported by the observation that in MT2 receptor-deleted mice, phase shifts in circadian electrical activity are greatly decreased. In vivo MLT treatment led to phase-shifted circadian rhythms of locomotor activity in the intact and MT2-KO mice but not in MT1-KO or MT1/MT2-KO mice, suggesting that in vivo activation of MT1 receptors triggers phase advances of circadian rhythms (Dubocovich and Markowska, 2005). Together, these results in vivo demonstrate that the MLT receptor involved in phase-shifting circadian rhythms is the MT1 subtype. Other data also support this interpretation. For example, in the mouse, MT1 and MT2 receptor mRNA and protein have been reported in the SCN, but MT2 receptor protein seems to be expressed at such low levels that 2-iodomelatonin binding is undetectable. The expression of MT2 receptor mRNA has also not been detected in vivo by in situ hybridization (Poirel et al., 2003). When expressed in transfected cells, MT1 and MT2 receptor subtypes exhibit both the pharmacological and functional characteristics of 2-iodomelatonin binding sites. However, targeted disruption of the MT1 receptor results in the total disappearance of 2-iodomelatonin binding sites in the SCN as well as in other brain tissues (Liu et al., 1997). By contrast, such a disappearance is not observed in mice following disruption of the MT2 receptors. Furthermore, in situ hybridization has confirmed that MT1 receptor mRNA is present within the SCN and numerous other brain and peripheral structures of most rodents. Although it is difficult to detect MT2 receptors in the SCN, either through pharmacological binding profiles or mRNA expression, the two presently known antagonists of this

MELATONIN AND THE CIRCADIAN SYSTEM receptor subtype (4P-PDOT and 4P-ADOT) appear to have functional effects on the SCN circadian clock and inhibit, at least in vitro, the effect of MLT. This suggests the involvement of a third receptor subtype that would be expressed in the SCN and would bind MLT and 4-PDOT/4P-ADOTwith high affinity, but 2-iodomelatonin with low affinity. Because the phase-shifting response in MT1-deficient mice is blocked in vivo by pertussis toxin (Hablitz et al., 2015), this receptor subtype is a G-protein-coupled receptor. The recent identification of G-protein-coupled inwardly rectifying potassium (GIRK) channels and the observation that GIRK2 knockout mice failed to phase advance wheel-running behavior in response to injection of MLT (Hablitz et al., 2015) might explain the data attributed to the MT2 subtype. In vitro phase resetting of the SCN circadian rhythm by MLT is blocked by coadministration of a GIRK channel antagonist, tertiapin-q, supporting this conclusion. The putative MT2 receptors in the SCN might in fact correspond at least partly to GIRK channels (Hablitz et al., 2015). From all these data, it appears that the chronobiotic effect of MLT on the SCN depends mainly on the MT1 subtype and possibly GIRK channels. The molecular mechanisms underlying the chronobiotic effect of MLT have also been investigated in the rat. It is known that the phase-shifting effects of light and some nonphotic cues are mediated by clock genes, through a rapid action on Per1 and Per2 gene transcription (Challet and Pevet, 2003). Other clock molecules that interact to generate circadian oscillations are also putative targets. RORa and REV-ERBa are known to activate and repress, respectively, Bmal1 expression. Acute MLT injection induces an immediate phase advance of the rhythmic expression of Rev-erba, suggesting that this transcription factor may be the initial molecular target involved in the chronobiotic effect of MLT (Agez et al., 2007). Posttranslational changes of clock proteins and especially phosphorylation processes also participate in the normal functioning of circadian clocks (Duguay and Cermakian, 2009). Whether the chronobiotic effects of MLT implicate posttranslational regulation of clock proteins remains to be investigated. The master clock itself is the site of action for the chronobiotic effect of MLT. Can this chronobiotic effect explain the majority of reported effects of MLT on sleep, circadian rhythms, and circadian disorders? In fact what is described is not exactly an effect on sleep per se but more exactly a phase advance in the sleep–wake cycle. Sleep and even sleep rhythms are not pure manifestations of the circadian system. Sleep is a complex physiological process, involving not only circadian clocks but also an orchestrated neurochemical network implicating different centers in the brain that

337

regulate sleep and wake homeostasis. The neural circuits that generate the various vigilance states, such as wakefulness, rapid-eye movement (REM) sleep and nonREM (NREM) sleep that alternate through the night in a cyclical fashion, remain to be completely elucidated. Humans are diurnal mammals, with a master circadian clock that promotes wakefulness during the day. Sleep timing is phase locked to intrinsic clock-controlled temperature rhythm and extrinsic light and dark conditions (Emens and Burgess, 2015). Different categories of sleep disorders have been defined by the American Academy of Sleep Medicine: (1) Disorders of initiating and maintaining sleep (insomnias); (2) sleep-related breathing disorders (sleep apnea); (3) disorders of excessive somnolence (hypersomnias); (4) dysfunctions associated with sleep, sleep stages, or partial arousals (parasomnias); and (5) disorders of the sleep–wake cycle (circadian rhythm sleep disorders), the last one being the disorder most susceptible to be treated by MLT (see Arendt, 2019). The basic mechanism by which MLT acts on sleep in humans is unclear. It may involve a phase shift of the master circadian clock (chronobiotic effect), a reduction in core body temperature (directly or throughout an effect on the SCN clock), or a direct action on somnogenic structures within the brain. MLT receptors have been described by immunocytochemistry in some structures of the neural circuits that generate arousal and sleep (both NREM and REM) (see later). In order to progress on sleep research at both fundamental and clinical levels, detailed analysis of the MLT receptor distribution should be considered specifically in the context of the multioscillatory circadian network.

Melatonin receptors and their localization In the recent decades, the widespread experimental use of 2-[125I]-iodomelatonin as a radio ligand has allowed the detection of MLT binding sites in a large number of different tissues and organs. In mammals, MLT receptors are present in numerous cerebral structures and peripheral organs, with great variability in localization and density between structures in a given species and in the same structure between species (Masson-Pevet et al., 1994; Dubocovich and Markowska, 2005; Lacoste et al., 2015; Klosen et al., 2019). As explained previously, different subtypes of mammalian MLT receptors have been cloned and characterized pharmacologically (Reppert et al., 1994; Jockers et al., 2016). The MT1 and MT2 subtypes exhibit subnanomolar affinity for MLT. The two cloned MLT receptor subtypes in mammals possess distinct signaling transduction pathways and tissue distribution (see Jockers et al., 2016). To note, a third subtype of high-affinity MLT receptor

338

P. PEVET ET AL.

was cloned from a chicken brain library and termed the Mel1c. It is however not present in mammals. A receptor structurally related to the MLT receptors (the orphan receptor GPR 50) but which does not bind MLT has also been identified. The receptor subtype called MT3 corresponds to a low-affinity binding to a quinone reductase QR2 (Boutin, 2015). Today, it is also believed that MLT receptor subtypes might form heterodimers in situ (Jockers et al., 2008). This heterodimerization would change the functional pharmacology of the receptor complex (Ferre et al., 2014). It is also believed that MLT receptors form heterodimers with other types of seven transmembrane domain G-protein-coupled receptors (e.g., 5-HT2C) (Kamal et al., 2015). This concept of heterodimerization absolutely requires that subtypes be expressed in the same cells. Presently, no clear data in the literature really support such a fact, and even the most carefully performed immunohistochemical analysis of MT1 and MT2 receptors distribution does not support such possibilities. Understanding the links between specific target sites for MLT, identified MLT receptor subtypes, and particular physiological actions is still a great scientific and clinical challenge. In situ hybridization and reverse transcription polymerase chain reaction analyses have also permitted to identify expression sites for MT receptor mRNA in the SCN, other brain structures, and peripheral organs (Weaver and Reppert, 1996). Immunohistochemistry is a better approach for cellular identification, and numerous studies have already been published (Wu et al., 2006; Lacoste et al., 2015; Waly and Hallworth, 2015). However, for the moment, the poor specificity of the MT1 and MT2 antibodies used has hampered collection of convincing data, such that our knowledge of sites of MLT action within the brain remains poor. Using antibodies raised against the human MT1 and MT2 proteins, a group (Lacoste et al., 2015) has made a global analysis in the rat brain. MT1 and MT2 were observed in numerous regions of the rat telencephalon, diencephalon, and mesencephalon. Overall, large differences were observed in the anatomical distribution pattern of MT1 and MT2 proteins. This distinctive localization of each receptor implicates a compartmentalization of MLT effects. They showed specifically that MT2 receptors are located in critical areas for sleep functions. Strong MT2 immunoreactivity of neuronal cell bodies and proximal dendrites was consistently observed, for example, in the reticular nucleus of the thalamus suggesting that the MT2 receptor subtype is probably the one mainly involved in the regulation of NREM sleep (Ochoa-Sanchez et al., 2011). The MT1 receptors observed in the locus coeruleus and lateral hypothalamus would be mainly implicated in the regulation of rapid-eye

movement sleep and arousal and may counterbalance the MT2 mediated effect (Gobbi and Comai, 2019). These two receptors seem to have not only a very specialized function in sleep, but they may also present opposing effects (Gobbi and Comai, 2019). This hypothesis is also supported by the fact that knockout mice for both MT1 and MT2 receptors and pinealectomized rats (Dominguez-Lopez et al., 2012) do not show impairments of NREM and REM sleep duration. To go further, it is necessary to take into account that sleep timing and regulation are different in diurnal vs nocturnal species. A detailed comparative analysis of MT1 and MT2 distribution is thus a perequisite to understand the exact role of MLT in sleep regulation. The direct effect of MLT on the temperature rhythm in humans, but not in nocturnal rodents, has, for example, to be considered. MLT in humans, probably by a direct action on the SCN, reinforces the nocturnal decrease of core temperature, an event which facilitates sleep propensity. Using a “knock-in” strategy replacing MT1 or MT2 coding sequences with a lacZ reporter, the precise localization of structures containing MT1 or MT2 has become possible, at least in mice. Cell populations containing the MT1 subtype are distinct from those containing the MT2 subtype (Klosen et al., 2019). Strong MT1-LacZ reporter expression is detected in only a few structures: the SCN, the Pars tuberalis, and the paraventricular nucleus of the hypothalamus. On the contrary, MT2-LacZ expression in the mouse brain is far more widespread. MT2-LacZ could be seen from the olfactory bulb down to the brain stem (see list in Klosen et al., 2019). Again, nocturnal endogenous MLT is an output used by the master clock to distribute a circadian signal throughout the general circulation. MLT has been described as a time-giving hormone (Pevet and Challet, 2011) to affect almost all main physiological functions of the body and to be useful in a vast array of experimental models reproducing different pathological conditions. All the structures containing MLT receptors are part of the multioscillatory circadian system. The presence of MT2 receptors in critical brain areas for sleep functions (the septum, CA2 layers of the hippocampus, preoptic nucleus, reticular nucleus of the thalamus, red nucleus, substantia nigra pars reticulata, oculomotor nuclei, and ventral tegmental nucleus (Gobbi and Comai, 2019)) leads to speculate that MLT is not a hormone acting per se on sleep, but rather a clock cue influencing circadian rhythms among which is the circadian regulation of sleep and activity in both diurnal and nocturnal animals. Most information concerns MLT and sleep disorders, but it is clear that the other identified MT1- or MT2-containing structures are also involved in circadian-related functions and pathologies. It is now possible, based on the

MELATONIN AND THE CIRCADIAN SYSTEM knowledge of distribution of structures with MLT receptors, to suggest some lines of research in which MLT could be used as a potential treatment. For example, the presence of MT2 in the paraventricular nucleus of the hypothalamus is restricted to a subpopulation of dorsolateral neurons (Klosen et al., 2019). This region contains CRH. Thus stress-related disorders, at least in circadian functioning, would be directly involved. MT2 is also present in large amount in the hippocampus specifically in the CA2 layer. Timing of hippocampal mnemonic processes (acquisition, consolidation, and retrieval of longterm memory [LTM]) and long-term potentiation (LTP) is known to be dependent on the circadian system (Jilg et al., 2019) by still unknown mechanisms. The role of MLT as a circadian time-cue for hippocampal signaling and memory formation is clearly established and the study of mechanisms involved will be facilitated by phenotyping MLT receptors containing cells. We also found MT1 receptors at the level of 5-HT neurons in the dorsal raphe. This speaks in favor of a possible mechanism in the well-established correlations between circadian rhythms, MLT, and several psychiatric diseases (Arendt, 2019). On the other hand, the presence of MT2 receptors in the arcuate nucleus points to relationships between circadian function and metabolism. This suggestion is reinforced by the genetic polymorphism identified within MLT membrane receptors Investigation on these polymorphisms in MT2receptors is presently ongoing in human subjects with various diseases (notably Type 2 diabetes). It is clear that such possibilities open numerous clinical perspectives. Based on the data reported previously, hypotheses, and knowledge about MLT receptors, several pharmacological approaches have been developed and drugs are now available. Preclinical studies have shown great promise, and development of new agents able to interact selectively with MLT receptors is becoming an area of great interest.

Melatonin receptor agonists/antagonists and drugs Since the discovery of MLT receptors, a great number of structurally different MLT receptor ligands have been developed (Dubocovich et al., 1997; Mor et al., 2010; Zlotos, 2012; Rivara et al., 2015). These drugs target the SCN for improving circadian timing (chronobiotic properties) or act indirectly at downstream level(s) of the circadian network (MLT as an internal time giver) to restore or maintain proper internal synchronization. The drugs have been tested in different species especially for potential use in circadian, psychiatric, and sleep disorders. Recent years have seen the introduction of

339

approved MLT receptor agonists. It is worthwhile to note that these MLT analogues/drugs are all nonspecific MT1/ MT2 agonists acting directly on the SCN. Ramelteon is effective in improving sleep latency, sleep efficiency, and subjective total sleep time in individuals with perturbation of the sleep/wake cycle (e.g., insomnia) (Mini et al., 2007). Tasimelteon has been approved for non24-h sleep–wake disorders, especially in totally blind people (Dhillon and Clarke, 2014). Agomelatine, also a nonselective MT1/MT2 agonist with chronobiotic properties (Pitrosky et al., 1999) and acting as an antagonist of 5-HT2C receptors, was approved for the treatment of depression. A prolonged-release MLT formulation (Circadin) has also been approved for insomnia therapy (Zisapel, 2018). These melatoninergic drugs are promising, but more precise analysis on the mechanisms involved and larger trials on different pathological situations are still needed. As mentioned earlier, most of the agonists and drugs currently under clinical or pharmacological evaluation (Mor et al., 2010; Carocci et al., 2014) are nonspecific MT1/MT2 agonists. Even if such nonspecific agonists are useful in modulating clock activities in vivo, their use appears to be limited. The two MLT receptors having different, divergent, or opposite roles (at least when sleep is concerned), mixed MT1–MT2 receptors ligands might not be clinically recommended (Gobbi and Comai, 2019). With respect to the present article, one important message is that the MLT receptor subtype to be targeted is MT1 for the SCN and MT2 for non-SCN brain structures. Consequently, pharmacological/chemical investigations should aim to develop highly selective agonists (and antagonists) for both subtypes. Presently, in vivo as well as in all recombinant systems used, no MTL receptor (MT1) agonist with a degree of selectivity over 500 (vs MT2 subtypes) is currently available (Zlotos et al., 2014). However, the MT2 receptor antagonists 4P-PDOT and 4P-ADOT are reference compounds exhibiting good binding affinity for the human cloned MT2 receptor (pKi ¼ 8.8) and a selectivity for the MT2 receptor at least 100-fold that of the MT1 subtype (Zlotos et al., 2014). The recent identification of a highly potent and selective MT2 agonist (UCM1014) with picomolar MT2 binding affinity more than 10,000-fold selectivity compared to the MT1 (Spadoni et al., 2015) demonstrates that such approaches in medical chemistry yield important breakthroughs. The identification of UCM1014 will now partly permit to differentiate between the MT1 and MT2 receptor-mediated functional responses that will enlarge our knowledge on the physiological role of MLT. In short, MLT receptors are widely distributed in the central nervous system and in the periphery and

340

P. PEVET ET AL.

selectivity for a single receptor could conceivably provide a therapeutic advantage. The development of new highly specific agonists/antagonists for both subtypes of MLT receptors, the identification of the link between MLT target sites within different parts of the brain or the body, and the association of specific MLT receptor subtypes with particular physiological effects will revive long-standing interests regarding many possible therapeutic applications for MTL. Knowledge of the cell types that contain MT1 or MT2 receptors will also aid in better defining the different roles of these two receptor subtypes.

CONCLUSIONS Virtually all aspects of physiology are rhythmic, from cells to systems. The correct timing of different functions in an organism relative to the timing of each organ should be precisely coordinated (e.g., daily timing of the sleep– wake cycle with the light/dark cycle) and depends on a complex circadian network. The growing number of individuals who experience frequent circadian perturbations is important in terms of health care. Restoring a poorly functioning circadian system will indeed not only be a means of improving the quality of daily lives by reestablishing proper sleep/wake cycle but also a step in preventing the development of more serious pathologies including depression, metabolic syndrome, cancer, and inflammation, which all have been associated with poor and/or insufficient sleep, especially during aging. Contrary to other SCN endocrine outputs, the MLT rhythm is a very stable circadian signal depending solely on the SCN clock and the light/dark cycle. However, as detailed previously, the MTL signal is more than just a “clock hand” of the SCN. Due to the presence of MTL receptors within the SCN itself, exogenous MTL has clear and firmy established chronobiotic effects (i.e., shifting effects on the master clock with consequences on all clock outputs), and due to the presence of MLT receptors in numerous peripheral structures/clocks, MLT has a clear internal time-giver role regulating physiological and neural functions (Amaral et al., 2019). MLT is constantly associated with biological rhythms. It acts via well-characterized membrane receptors. Through its chronobiotic properties, most of the experimental and clinical data reported previously, especially when sleep is concerned, can be explained fully. In view of the large distribution of MLT receptors (especially MT2, Klosen et al., 2019) in brain structures, the internal time-giver role of MLT is important to consider. As first as sleep is concerned, we agree with Gobbi and Comai (2019) that the control of the sleep–wake cycle might be explained by a parallel and concomitant action of MLT on the master clock (chronobiotic effect) and on sleep-related

structures within the brain (see also Zisapel, 2007, 2018). Exogenous MLT is the best-known and most studied chronobiotic drug, and its role in regulating clock-controlled circadian rhythms has been firmly established. Exogenous MLT, via the activation of its two membrane receptors (MT1 and MT2), has thus great therapeutic potential. In humans, even if randomized and controlled studies are still rare, some clinical trials have confirmed the efficacy of MLT in circadian rhythm disorders (especially circadian sleep disorders but also psychiatric diseases, aging). Such concomitant actions might also well explain the effect of MLT in other disorders. For example, successful therapeutic strategies designed to combat metabolic syndrome associated with circadian disruption will require targeting of both circadian and metabolic dysfunctions. Because the two MLT receptor subtypes present in mammals, including humans, have different/divergent functions, only the development of a new family of ligands with high selectivity for each subtype will provide therapeutic advantages. Clearly this is a new challenge in the field and will be at the root of new therapeutic agents for curing specific pathologies, including sleep disorders. We are aware of the numerous published data which claim a protective role (antioxidant) of MLT at least when administered at high doses (Boga et al., 2019). Even if the therapeutic potential of this effect should not be neglected, the physiological significance of these antioxidant properties of MLT still remains to be settled in in vivo conditions. They are unlikely to be related to the “chronobiotic” and “time-giver” actions of the hormone discussed in this review. Continuously high circulating levels of MLT would indeed compromise/prevent the proper diurnal functioning of MLT receptors necessary for mediating the chronobiological effects of the hormone.

ACKNOWLEDGMENTS The authors are grateful to Dr. D. Hicks for discussions, suggestions, and for language corrections.

REFERENCES Arbon EL, Knurowska M, Dijk DJ et al. (2015). Randomised clinical trial of the effects of prolonged-release melatonin, temazepam and zolpidem on slow-wave activity during sleep in healthy people. J Psychopharmacol 29: 764–776. https://doi.org/10.1177/0269881115581963. Acuna-Castroviejo D, Escames G, Venegas C et al. (2014). Extrapineal melatonin: sources, regulation, and potential functions. Cell Mol Life Sci 71: 2997–3025. Agez L, Laurent V, Pevet P et al. (2007). Melatonin affects nuclear orphan receptors mRNA in the rat suprachiasmatic nuclei. Neuroscience 144: 522–530.

MELATONIN AND THE CIRCADIAN SYSTEM Alonso-Vale MI, Andreotti S, Peres SB et al. (2005). Melatonin enhances leptin expression by rat adipocytes in the presence of insulin. Am J Physiol Endocrinol Metab 288: E805–E812. Amaral FGD, Andrade-Silva J, Kuwabara WMT et al. (2019). New insights into the function of melatonin and its role in metabolic disturbances. Expert Rev Endocrinol Metab 14: 293–300. Arendt J (2005). Melatonin: characteristics, concerns, and prospects. J Biol Rhythms 20: 291–303. Arendt J (2019). Melatonin: countering chaotic time cues. Front Endocrinol 10: 391400. Arendt J, Skene DJ (2005). Melatonin as a chronobiotic. Sleep Med Rev 9: 25–39. B€ahr I, M€uhlbauer E, Schucht H et al. (2011). Melatonin stimulates glucagon secretion in vitro and in vivo. J Pineal Res 50: 336–344. Bering T, Hertz H, Rath M (2020). Rhythmic release of corticosterone induces circadian clock gene expression in the cerebellum. Neuroendocrinology 110: 604–615. https://doi.org/ 10.1159/000503720. Boga JA, Caballero B, Potes Y et al. (2019). Therapeutic potential of melatonin related to its role as an autophagy regulator: a review. J Pineal Res 66: e12534. https://doi. org/10.1111/jpi.12534. Bothorel B, Barassin S, Saboureau M et al. (2002). Exogenous melatonin increases the amplitude of the pineal melatonin secretion by a direct action on the circadian clock in the rat. Eur J Neurosci 16: 1090–1098. Boutin JA (2015). Quinone reductase 2 as a promising target of melatonin therapeutic actions. Expert Opin Ther Targets 20: 1–15. Buijs RM, Kalsbeek A (2001). Hypothalamic integration of central and peripheral clocks. Nat Rev Neurosci 2: 521–526. Carocci A, Catalano A, Sinicropi MS (2014). Melatonergic drugs in development. Clin Pharmacol 6: 127–137. Castano A, Bothorel B, Seguin L et al. (2014). Like melatonine, Agomelatine (S20098) increases the amplitude of oscillations on two clock outputs: melatonin and temperature rhythms. Chronobiol Int 31: 371–381. Chakir I, Dumont S, Pevet P et al. (2015). Pineal melatonin is a circadian time-giver for leptin in Syrian hamster. Front Neurosci 9: 190–195. Chakravarty S, Rizvi SI (2011). Circadian modulation of sodium-potassium ATPase and sodium - proton exchanger in human erythrocytes: in vitro effect of melatonin. Cell Mol Biol 57: 80–86. Challet E, Pevet P (2003). Interactions between photic and nonphotic stimuli to synchronize the master circadian clock in mammals. Front Biosci 6: s246–s257. Cortesi F, Giannotti F, Sebastiani T et al. (2012). Controlledrelease melatonin, singly and combined with cognitive behavioural therapy, for persistent insomnia in children with autism spectrum disorders: a randomized placebocontrolled trial. J Sleep Res 21: 700–709. Cuomo BM, Vaz S, Lee EAL et al. (2017). Effectiveness of sleep-based interventions for children with autism spectrum disorder: a meta-synthesis. Pharmacotherapy 37: 555–578.

341

Dardente H, Menet JS, Poirel VJ et al. (2003). Melatonin induces Cry1 expression in the pars tuberalis of the rat. Mol Brain Res 114: 91–182. Dardente H, Wyse CA, Birnie MJ et al. (2010). A molecular switch for photoperiod responsiveness in mammals. Curr Biol 20: 2193–2198. Dardente H, Wood S, Ebling F et al. (2019). An integrative view of mammalian seasonal neuroendocrinology. J Neuroendocrinol 31: e12729. https://doi.org/10.1111/ jne.12729. De Leersnyder H, Claustrat B, Munnich A et al. (2006). Circadian rhythm disorder in a rare disease: smithMagenis syndrome. Mol Cell Endocrinol 252: 88–91. Dhillon S, Clarke M (2014). Tasimelteon: first global approval. Drugs 74: 505–511. Dibner C, Schibler U, Albrecht U (2010). The mammalian circadian timing system: organization and coordination of central and peripheral clocks. Annu Rev Physiol 72: 517–549. Dochi M, Suwazono Y, Sakata K et al. (2009). Shift work is a risk factor for increased total cholesterol level: a 14-year prospective cohort study in 6886 male workers. Occup Environ Med 66: 592–597. Dominguez-Lopez S, Maha I, Bambico FR et al. (2012). Short-term effects of melatonin and pinealectomy on serotonergic neuronal activity across the light-dark cycle. J Psychopharmacol 26: 830–844. Dubocovich ML (1983). Melatonin is a potent modulator of dopamine release in the retina. Nature 306: 782–784. Dubocovich ML, Markowska M (2005). Functional MT1 and MT2 melatonin receptors in mammals. Endocrine 27: 101–110. Dubocovich ML, Masana MI, Iacob S et al. (1997). Melatonin receptor antagonists that differentiate between the human Mel1a and Mel1b recombinant subtypes are used to assess the pharmacological profile of the rabbit retina ML1 presynaptic heteroreceptor. Naunyn Schmiedebergs Arch Pharmacol 355: 365–375. Duguay D, Cermakian N (2009). The crosstalk between physiology and circadian clock proteins. Chronobiol Int 26: 1479–1513. Elrod MG, Hood BSS (2015). Sleep differences among children with autism spectrum disorders and typically developing peers: a meta-analysis. J Dev Behav Pediatr 36: 166–177. Emens JS, Burgess HJ (2015). Effect of light and melatonin and other melatonin receptor agonists on human circadian physiology. Sleep Med Clin 10: 435–453. Ferre S, Casado V, Devi LA et al. (2014). G protein-coupled receptor oligomerization revisited: functional and pharmacological perspectives. Pharmacol Rev 66: 413–434. Gillette MU, McArthur AJ (1996). Circadian actions of melatonin at the suprachiasmatic nucleus. Behav Brain Res 73: 135–139. Gobbi G, Comai S (2019). Sleep well. Untangling the role of melatonin MT1 and MT2 receptors in sleep. J Pineal Res 66: e12544. https://doi.org/10.1111/jpi.12544. Epub 2019 Jan 21. Grosse J, Davis FC (1988). Melatonin entrains the restored circadian activity rhythms of Syrian hamsters bearing fetal suprachiasmatic nucleus grafts. J Neurosci 18: 8032–8037.

342

P. PEVET ET AL.

Hablitz LM, Molzof HE, Abrahamsson KE et al. (2015). GIRK channels mediate the non-photic effects of exogenous melatonin. J Neurosci 35: 14957–14965. Houdek P, Polidarova´ L, Nova´kova´ M et al. (2015). Melatonin administered during the fetal stage affects circadian clock in the suprachiasmatic nucleus but not in the liver. Dev Neurobiol 75: 131–144. Jilg A, Bechstein P, Saade A et al. (2019). Melatonin modulates daytime-dependent synaptic plasticity and learning efficiency. J Pineal Res 66: e12553. https://doi.org/10.1111/ jpi.12553. Jockers R, Maurice P, Boutin J et al. (2008). Melatonin receptors, heterodimerization, signal transduction and binding sites: what’s new? Br J Pharmacol 154: 1182–1195. Jockers R, Delagrange P, Dubocovich ML et al. (2016). Update on melatonin receptors: IUPHAR review 20. Br J Pharmacol 173: 2702–2725. Kalsbeek A, Palm IF, La Fleur SE et al. (2006). SCN outputs and the hypothalamic balance of life. J Biol Rhythms 21: 458–469. Kamal M, Ghahou F, Daulat AM et al. (2015). Convergence of melatonin and serotonin (5-HT) signaling at MT2/5-HT2C receptor heteromers. J Biol Chem 290: 11537–11546. Khaldy H, Leo´n J, Escames G et al. (2002). Circadian rhythms of dopamine and dihydroxyphenyl acetic acid in the mouse striatum: effects of pinealectomy and of melatonin treatment. Neuroendocrinology 75: 201–218. Klosen P, Lapmanee S, Schuster C et al. (2019). MT1 and MT2 melatonin receptors are expressed in nonoverlapping neuronal populations. J Pineal Res 67: e12575. https:// doi.org/10.1111/jpi.12575. Lacoste B, Angeloni D, Dominguez-Lopez S et al. (2015). Anatomical and cellular localization of melatonin MT1 and MT2 receptors in the adult rat brain. J Pineal Res 58: 397–417. Lerner AB, Case JD, Takahashi Y et al. (1958). Isolation of melatonin: the pineal gland factor that lightens melanocytes. J Am Chem Soc 80: 2587–2589. Liu C, Weaver DR, Jin X et al. (1997). Molecular dissection of two distinct actions of melatonin on the suprachiasmatic circadian clock. Neuron 19: 91–102. Lotufo CM, Lopes C, Dubocovich ML et al. (2001). Melatonin and N-acetylserotonin inhibit leukocyte rolling and adhesion to rat microcirculation. Eur J Pharmacol 430: 351–357. Malek ZS, Sage D, Pevet P et al. (2007). Daily rhythm of tryptophan hydroxylase-2 messenger ribonucleic acid within raphe neurons is induced by corticoid daily surge and modulated by enhanced locomotor activity. Endocrinology 148: 5165–5172. Masson-Pevet M, George D, Kalsbeek A et al. (1994). An melatonin binding sites with rhythmic functions: a study in five hibernator species. Cell Tissue Res 278: 97–106. Messager S, Garabette ML, Hastings MH et al. (2001). Tissuespecific abolition of Per1 expression in the pars tuberalis by pinealectomy in the Syrian hamster. Neuroreport 12: 579–582. Mini J, Wang-Weigand S, Zhang J (2007). Self-reported efficacy and tolerability of ramelteon 8 mg in older adults

experiencing severe sleep onset difficulty. Am J Geriatr Pharmacother 5: 1777–1784. Mor M, Rivara S, Pala D et al. (2010). Recent advances in the development of melatonin MT1 and MT2 receptor agonists. Expert Opin Ther Pat 20: 1059–1073. Ochoa-Sanchez R, Comai S, Lacoste B et al. (2011). Promotion of non-rapid eye movement sleep and activation of reticular thalamic neurons by a novel MT2 melatonin receptor ligand. J Neurosci 31: 18439–18452. Oishi K, Amagai N, Shirai H et al. (2005). Genome-wide expression analysis reveals 100 adrenal gland-dependent circadian genes in the mouse liver. DNA Res 12: 191–202. 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. Peschke E, Peschke D (1998). Evidence for a circadian rhythm of insulin release from perifused rat pancreatic islets. Diabetologia 41: 1085–1092. Pevet P (2016). Melatonin receptors as therapeutic targets in the suprachiasmatic nucleus. Exper Opin Ther Targets 20: 1209–1218. Pevet P, Challet E (2011). Melatonin: both master clock output and internal time-giver in the circadian clocks network. J Physiol Paris 105: 170–182. Pevet P, Bothorel B, Slotten H et al. (2002). The chronobiotic properties of melatonin. Cell Tissue Res 309: 183–191. Pitrosky B, Kirsch R, Malan A et al. (1999). Organization of rat circadian rhythms during daily infusion of melatonin or S20098, a melatonin agonist. Am J Physiol 277: R812–R828. Poirel VJ, Boggio V, Dardente H et al. (2003). Clock gene mRNA expressions are differentially regulated in the rat suprachismatic by acute melatonin injection. Neuroscience 120: 745–755. Redman JR, Armstrong S, Ng KT (1983). Free-running activity rhythms in the rat: entrainment by melatonin. Science 219: 1089–1091. Reiter RJ, Tan DX, Terron MP et al. (2007). Melatonin and its metabolites: new findings regarding their production and their radical scavenging actions. Acta Biochim Pol 54: 1–9. Reiter RJ, Tan DX, Fuentes-Broto L (2010). Melatonin: a multitasking molecule. Prog Brain Res 181: 127–251. Reppert SM, Weaver DR, Ebisawa T (1994). Cloning and characterization of a mammalian melatonin receptor that mediates reproductive and circadian responses. Neuron 13: 1177–1185. Riemersma-van der Lek RF, 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: a randomized controlled trial. JAMA 299: 2642–2655. Rivara S, Pala D, Bedini A et al. (2015). Therapeutic uses of melatonin and melatonin derivatives: a patent review (2012–2014). Expert Opin Ther Pat 25: 425–441. Rusak B, Yu GD (1993). Regulation of melatonin-sensitivity and firing-rate rhythms of hamster suprachiasmatic nucleus neurons: pinealectomy effects. Brain Res 602: 200–204.

MELATONIN AND THE CIRCADIAN SYSTEM Sandu C, Liu T, Malan A et al. (2015). Circadian clocks in rat skin and dermal fibroblasts: differential effects of aging, temperature and melatonin. Cell Mol Life Sci 72: 2237–2248. Scheer FA, Morris CJ, Garcia JI et al. (2012). Repeated melatonin supplementation improves sleep in hypertensive patients treated with beta-blockers: a randomized controlled trial. Sleep 35: 1395–1402. Schernhammer ES, Kroenke CH, Laden F et al. (2006). Night work and risk of breast cancer. Epidemiology 17: 108–111. Skene DJ, Arendt J (2007). Circadian rhythms disorders in the blind and their treatment with melatonin. Sleep Med 8: 651–655. Skene DJ, Swaab DF (2003). Melatonin rhythmicity: effect of age and Alzheimer’s disease. Exp Gerontol 38: 199–206. Skene DJ, Vivien-Roels B, Sparks DL et al. (1990). Daily variation in the concentration of melatonin and 5-methoxytryptophol in the human pineal gland: effect of age and Alzheimer’s disease. Brain Res 528: 170–174. Slawik H, Stoffel M, Riedl L et al. (2016). Prospective study on salivary evening melatonin and sleep before and after Pinealectomy in humans. J Biol Rhythms 31: 82–93. Slotten HA, Pitrosky B, Pevet P (1999). Entrainment of rat circadian rhythms by daily administration of melatonin: influence of the role of administration. J Biol Rhythms 14: 347–353. Sookoian S, Gemma C, Ferna´ndez Gianotti T et al. (2007). Effects of rotating shift work on biomarkers of metabolic syndrome and inflammation. J Intern Med 261: 285–292. Spadoni G, Bedini A, Lucarini S et al. (2015). Highly potent and selective MT2 melatonin receptor full agonists from conformational analysis of 1-benzyl-2-acylaminomethyltetrahydroquinolines. J Med Chem 58: 7512–7525. Stehle JH, von Gall C, Korf HW (2002). Organisation of the circadian system in melatonin-proficient C3H and melatonin-deficient C57BL mice: a comparative investigation. Cell Tissue Res 309: 173–182. Stenvers DJ, Scheer FAJL, Schrauwen P et al. (2019). Circadian clocks and insulin resistance. Nat Rev Endocrinol 15: 75–89. Su Y, van der Spek R, Foppen E et al. (2014). Effects of adrenalectomy on daily gene expression rhythms in the rat suprachiasmatic and paraventricular hypothalamic nuclei and in white adipose tissue. Chronobiol Int 32: 211–224. Suofu V, Li W, Jean-Alphonse FG et al. (2017). Dual role of mitochondria in producing melatonin and driving GPCR signaling to block cytochrome c release. Proc Natl Acad Sci U S A 114: E7997–E8006. Tordjman S, Najjar I, Bellissant E et al. (2013). Advances in the research of melatonin in autism spectrum disorders: literature review and new perspectives. Int J Mol Sci 14: 20508–20542.

343

Torres-Farfan C, Valenzuela FJ, Mondaca M et al. (2008). Evidence of a role for melatonin in fetal sheep physiology: direct actions of melatonin on fetal cerebral artery, brown adipose tissue and adrenal gland. J Physiol 586: 4017–4027. Torres-Farfan C, Mendez N, Abarzua-Catalan L et al. (2011). A circadian clock entrained by melatonin is ticking in the rat fetal adrenal. Endocrinology 152: 1891–1900. Waldhauser F, Kova´cs J, Reiter E (1998). Age-related changes in melatonin levels in humans and its potential consequences for sleep disorders. Exp Gerontol 33: 759–772. Waly NE, Hallworth R (2015). Circadian pattern of melatonin MT1 and MT2 receptor localization in the rat suprachiasmatic nucleus. J Circadian Rhythms 13: 1–7. Weaver DR, Reppert SM (1996). The Mel1a melatonin receptor gene is expressed in human suprachiasmatic nuclei. Neuroreport 8: 109–112. Wu YH, Swaab D (2007). Sturbance and strategies for reactivation of the circadian rhythm system in aging and Alzheimer’s disease. Sleep Med 8: 623–636. Wu YH, Zhou JN, Balesar R et al. (2006). Distribution of MT1 melatonin receptor immunoreactivity in the human hypothalamus and pituitary gland: colocalization of MT1 with vasopressin, oxytocin, and corticotropin-releasing hormone. J Comp Neurol 499: 897–910. Wu YH, Ursinus J, Zhou JN et al. (2013). Alterations of melatonin receptors MT1 and MT2 in the hypothalamic suprachiasmatic nucleus during depression. J Affect Disord 148: 357–367. Wurtman RJ, Axelrod J (1965). The formation, metabolism, and physiologic effects of melatonin in mammals. Prog Brain Res 10: 520–529. Ying SW, Zhang DX, Rusak B (1993). Effects of serotonin agonists and melatonin on photic responses of hamster intergeniculate leaflet neurons. Brain Res 628: 8–16. Zaidan R, Geoffriau M, Brun J et al. (1994). Melatonin is able to influence its secretion in humans: description of a phaseresponse curve. Neuroendocrinology 60: 105–112. Zelinski EL, Deibel SH, McDonald RJ (2014). The trouble with circadian clock dysfunction: multiple deleterious effects on the brain and body. Neurosci Biobehav Rev 40: 80–101. Zisapel N (2007). Sleep and sleep disturbances: biological basis and clinical implications. Cell Mol Life Sci 64: 1174–1186. Zisapel N (2018). New perspectives on the role of melatonin in human sleep, circadian rhythms and their regulation. Br J Pharmacol 175: 3190–3199. Zlotos D (2012). Recent progress in the development of agonists and antagonists for melatonin receptors. Curr Med Chem 19: 3532–3549. Zlotos DP, Jockers R, Cecon E et al. (2014). MT1 and MT2 melatonin receptors: ligands, models, oligomers, and therapeutic potential. J Med Chem 57: 3161–3185.

This page intentionally left blank

Handbook of Clinical Neurology, Vol. 179 (3rd series) The Human Hypothalamus: Anterior Region D.F. Swaab, F. Kreier, P.J. Lucassen, A. Salehi, and R.M. Buijs, Editors https://doi.org/10.1016/B978-0-12-819975-6.00022-4 Copyright © 2021 Elsevier B.V. All rights reserved

Chapter 23

Melatonin receptors, brain functions, and therapies ATSURO OISHI, FLORENCE GBAHOU, AND RALF JOCKERS* Institut Cochin, Universite de Paris, Paris, France

Abstract In mammals, including humans, the neurohormone melatonin is mainly secreted from the pineal gland at night and acts on two high-affinity G protein-coupled receptors, the melatonin MT1 and MT2 receptors. Major functions of melatonin receptors in the brain are the regulation of circadian rhythms and sleep. Correspondingly, the main indications of the currently available drugs for these receptors indicate this as targets. Yet these drugs may not only improve circadian rhythm- and sleep-related disorders but may also be beneficial for complex diseases like major depression, Alzheimer’s disease, autism, and attention-deficit/hyperactivity disorders. Here, we will focus on the hypothalamic functions of melatonin receptors by updating our knowledge on their hypothalamic expression pattern at normal, aged, and disease states, by discussing their capacity to regulate circadian rhythms and sleep and by presenting the clinical applications of the melatonin receptor-targeting drugs ramelteon, tasimelteon, and agomelatine or of prolonged-release melatonin formulations. Finally, we speculate about future trends in the field of melatonin receptor drugs.

INTRODUCTION Melatonin receptors constitute a subfamily of the G protein-coupled receptor (GPCR) superfamily, composed of approximately 800 members in humans (400 odorant and 400 nonodorant receptors). The melatonin receptor subfamily contains three members (Oishi et al., 2018), the melatonin binding MT1 (Reppert et al., 1994) and MT2 receptors (Reppert et al., 1995), and GPR50, which has high sequence homology with MT1 and MT2 but does not bind to melatonin or any other known ligand (Reppert et al., 1996). Recently, other members have been proposed to belong to the melatonin receptor subfamily but this was not validated (Oishi et al., 2017). Melatonin and its receptors are an evolutionary conserved ligand–receptor pair regulating many physiological functions including the regulation of circadian and seasonal rhythms, sleep onset, mother–fetus synchronization, retinal physiology, glucose homeostasis, and immune functions (Dubocovich et al., 2010; Jockers

et al., 2016). In vertebrates, melatonin is secreted from the pineal gland at night under the control of the hypothalamic suprachiasmatic nuclei (SCN) (Fig. 23.1A). This nocturnal secretion pattern is common between diurnal and nocturnal animals; thus melatonin is also called the hormone of darkness. This review will focus on melatonin receptor functions related to the hypothalamus and report mainly on evidence from humans, complemented by data from rodents and other animal models. We will first describe the three main expression sites of melatonin receptors in the hypothalamus and those in the pituitary functionally connected to the hypothalamus, then discuss known and suspected functions of hypothalamic melatonin receptors and finish with the description of currently marketed drugs targeting these receptors as well as consider future developments and fields of application. For more specialized aspects in terms of melatonin receptor– ligand chemistry and therapeutic potential (Zlotos et al., 2014; Liu et al., 2016), mouse models and human diseases (Tosini et al., 2014) or pharmacological aspects

*Correspondence to: Ralf Jockers, Universite de Paris, Institut Cochin, 22 rue Mechain, F-75014 Paris, France. Tel.: +33-1-40-51-64-34, Fax: +33-1-40-51-64-73, E-mail [email protected]

346

A. OISHI ET AL.

Fig. 23.1. Flow of photoperiodic information through melatonin receptors in the hypothalamus and pituitary. (A) Photoperiodic information is received by the retina in the eye, then transmitted through the retinohypothalamic tract to the SCN which then triggers melatonin production by the pineal gland during the night. In return, melatonin activates melatonin receptors in the SCN and in other central and peripheral tissues. (B) Melatonin secreted from pineal gland activates MT1 receptors in the pars tuberalis of the pituitary-stimulating TSH secretion. TSH receptor activation in ependymal cells of the hypothalamus induces Dio2 expression converting thyroid hormone T4 into T3. In seasonal breeders, T3 stimulates GnRH secretion and seasonal reproduction. 3rd V, third ventricle.

(Cecon et al., 2018, 2019; Boutin and Legros, 2020), the reader is invited to consult the corresponding expert review articles.

EXPRESSION OF MELATONIN RECEPTORS IN THE HUMAN HYPOTHALAMUS Suprachiasmatic nucleus (SCN) The prominent regulatory role of melatonin on the biological master clock in the hypothalamic SCN prompted research very early in detecting the expression of melatonin receptors in this region of the hypothalamus. Initial studies demonstrated their expression in the SCN using 2-[125I]-iodomelatonin (2-[125I]-IMLT) (Weaver et al., 1989, 1993; Williams et al., 1995). Later on, these binding sites were attributed to MT1 since they were completely lost in MT1 knockout mice (Liu et al., 1997). The mRNA for MT1 was detected in SCN neurons

by in situ hybridization in rodents (Reppert et al., 1994) and in humans (Weaver and Reppert, 1996; Thomas et al., 1998). The detection of the MT2 mRNA by in situ hybridization turned out to be more difficult but was eventually achieved (Dubocovich et al., 1998). Subsequently, the distribution of melatonin receptors in the SCN was confirmed by immunocytochemistry for MT1 (Wu et al., 2006, 2007, 2013; van Wamelen et al., 2013) and MT2 (van Wamelen et al., 2013; Wu et al., 2013). In conclusion, both melatonin receptors are expressed in the SCN with MT1 being more readily detectable than MT2.

SCN SUBREGIONS AND CELL POPULATIONS In mice, the SCN is roughly composed of 20,000 cells. Single-cell RNA-sequencing identified eight major cell types, each displaying a specific circadian gene expression pattern (Wen et al., 2020). Neurons represent the

MELATONIN RECEPTORS, BRAIN FUNCTIONS, AND THERAPIES majority of the SCN cells and can be divided into five neuronal subpopulations (Wen et al., 2020). The precise distribution of MT1 and MT2 among these cell types remains to be determined. In humans, the distribution of melatonin receptors has been studied by immunohistochemistry in brain slices. MT1 and MT2-immuno-positive cells were widely distributed over the anterior hypothalamus. Staining was observed in neurons, not in glia cells (Wu et al., 2007, 2013; van Wamelen et al., 2013). Colocalization of MT1 staining was observed with some arginine vasopressin (AVP) neurons (Wu et al., 2006), suggesting that the observed inhibitory effect of melatonin on spontaneous and stimulated AVP release from the SCN could be directly mediated through melatonin receptors expressed in these cells (Watanabe et al., 1998; Vanecek and Watanabe, 1999). A recent immunohistochemistry study in Sapajus apella, a diurnal primate model, revealed immunoreactivity (IR) for MT1 and MT2 that was more prominent in the dorsomedial region of the SCN as compared to the ventrolateral region, and this distribution changed between night and day (Pinato et al., 2017). Colocalization between melatonin receptor IR and vasoactive intestinal peptide (VIP) staining was observed in some cells (Pinato et al., 2017) suggesting that coexpressed melatonin receptors could regulate VIP secretion, which was shown to affect the circadian rhythm of AVP release in vitro (Watanabe et al., 2000). This melatonin–VIP pathway was suggested to relate to the previously mentioned melatonin-induced AVP inhibition (Watanabe et al., 1998). Similar results were obtained in mice using a knock-in model expressing the LacZ reporter gene instead of the MTNR1A and MTNR1B genes coding for MT1 and MT2 (Klosen et al., 2019). Both MT1-LacZ and MT2-LacZ reporter gene expressions were detected in SCN neurons but not in astrocytes. MT1-LacZ expression was higher than MT2-LacZ expression and delineated the boundaries of the SCN. MT2-LacZ expression was more diffuse in the anterior hypothalamic area. MT1-LacZ expression colocalized with a subpopulation of VIP expressing neurons and with some calretinin positive cells, a marker of SCN core cells in mice. In conclusion, MT1 and MT2 IR in the SCN seem to be restricted to neurons, suggesting that functions of MT1 and/or MT2 in the SCN are mediated through neurons. Whether both receptors are coexpressed in these neurons remains to be determined. Some melatonin receptor positive cells colocalize with cells expressing VIP or AVP in humans. The identity of the majority of melatonin receptor expressing cells remains however unknown. This question can now be addressed using single-cell RNA-sequencing together with the recently defined cell types composing the mouse SCN (Wen et al., 2020).

347

CIRCADIAN VARIATION Animal studies show a diurnal rhythm of the density of 2-[125I]-IMLT binding sites in the hypothalamus (Laudon et al., 1988; Masana et al., 2000), MT1 receptor mRNA (Neu and Niles, 1997; Masana et al., 2000; Poirel et al., 2002), and protein in the SCN (Waly and Hallworth, 2015). Melatonin has been suspected to be responsible for the circadian expression profile of MT1. Indeed, melatonin was shown to inhibit the expression of its MT1 receptor at the mRNA level creating a negative feed-back loop (Barrett et al., 1996). The circadian expression profile of MT2 has not been reported mainly due to difficulties to detect its mRNA and protein in rodents. In humans no significant effect of the time of death of people was found on the number of MT1 positive cells in the central SCN section. The number of MT2 positive cells was significantly associated with time of death with low levels around midnight (1 a.m.) and high levels around noon (1 p.m.) (Wu et al., 2013). These observations suggest a circadian variation of MT2 expression in the human SCN in agreement with the negative feedback loop hypothesis. In the diurnal Sapajus apella primate, a day/night rhythm of MT1 and MT2 IR was observed (Pinato et al., 2017). Intriguingly, night levels were higher than daytime levels, which are opposite to humans and would not fit with the melatonin feed-back hypothesis.

VARIATION WITH DISEASE The expression of melatonin receptors has been determined in the human SCN of patients with several diseases, it varies with age. Comparison of young subjects (aged 19–40 years old) and aged subjects (61–85 years old) without any psychiatric and neurological disease showed that the number and density of MT1-immunoreactive neurons in the SCN were decreased by half in aged controls (Wu et al., 2007). Similar levels were observed in early-stage Alzheimer disease (AD) patients and a further decrease in age-matched late stage AD patients in the same study (Wu et al., 2007). This indicates an age- and AD-related decline in MT1 expression in the SCN possibly contributing to the observed decline of robustness of the circadian activity in aged subjects and AD patients. A similar modulation of melatonin receptor expression in AD patients was observed outside of the SCN, in the hippocampus (Savaskan et al., 2001, 2002, 2005) and the retina (Savaskan et al., 2005). In depressed patients, the number of MT1immunoreactive cells was increased in the central SCN (Wu et al., 2013). The number of MT2-immunoreactive cells was unaffected (Wu et al., 2013). The negative correlation of MT1 expressing cells with disease duration

348

A. OISHI ET AL.

suggests that the increase of MT1 expression happens progressively during the course of the disease. Another case–control study accessing MT1 and MT2 receptor expression by immunohistochemistry in Huntington disease patients showed no changes in the SCN (van Wamelen et al., 2013). In contrast, decreased expression of MT1 mRNAwas reported in the striatum of Huntington disease patients (Wang et al., 2011).

Supraoptic nucleus (SON) and paraventricular nucleus (PVN) Apart from the SCN, melatonin receptor expression was observed in two other hypothalamic nuclei in humans, the paraventricular nucleus (PVN), and the supraoptic nucleus (SON) (Wu et al., 2006, 2013). The PVN is one of the most important autonomic control centers in the forebrain (Ferguson et al., 2008) and the SON contains osmoresponsive neurons (Leng and Russell, 2019). The intensity of the MT1 immunostaining is stronger in the SON and PVN than in the SCN. The MT2 staining intensity was similar in the SON, PVN, and SCN. MT1 staining colocalized with some parvocellular and magnocellular AVP and oxytocine (OXT) neurons in the SON and PVN and with some parvocellular corticotropin-releasing hormone (CRH) neurons in the PVN (Wu et al., 2006). MT1 and MT2 staining was also detected in the SON and PVN of the Sapajus apella primate with a predominant staining during the subjective day (Pinato et al., 2017). MT2 IR was also observed in the rat SON (Lacoste et al., 2015) and PVN (Waly and Hallworth, 2015) and in the LacZ reporter mouse where a clear colocalization with CRH was seen (Klosen et al., 2019). MT1 expression in the SON and PVN was not reported in rodents (Lacoste et al., 2015; Klosen et al., 2019). Expression of melatonin receptors in the SON and PVN suggest a regulatory role of melatonin in the principal functions of these two nuclei such as water balance, blood pressure, reproduction, and the activity of the hypothalamus–pituitary–adrenal (HPA) axis (for review, see Swaab, 2003). The colocalization of MT1 and CRH suggests that melatonin might directly modulate the HPA axis in the PVN, which may have implications for stress conditions such as depression. Whereas hypothalamic MT2 expression appears to be limited to the three aforementioned nuclei (SCN, SON, and PVN), MT1 expression was observed in a number of additional human nuclei (periventricular nucleus, sexually dimorphic nucleus, the diagonal band of Broca, the nucleus basalis of Meynert, infundibular nucleus, ventromedial and dorsomedial nucleus, tuberomamillary nucleus, mamillary body, and paraventricular thalamic nucleus) (Wu et al., 2006, 2013) suggesting

further regulatory roles of melatonin that remain to be explored in future studies. In conclusion, important progress has been made in the last 10 years on the localization of melatonin receptors in the hypothalamus shaping a first landscape of MT1 and MT2 expression including regions which are consistent with known functions of melatonin and regions that might hint to previously unsuspected functions that need to be explored. Further studies will be necessary to confirm the currently proposed locations of melatonin receptor expression and their absence in other regions. Indeed, false positive and false negative staining is an inherent limitation of antibody-based technique in terms of the specificity and cross-reactivity of antibodies. Similarly, transgenic animal models might not fully reflect the expression pattern of wild-type animals and speciesspecific differences in expression patterns and functions might limit the extrapolation to humans (Stankov et al., 1991). More specifically, in the case of melatonin, its physiological functions might differ between diurnal humans and nocturnal rodents.

FUNCTION OF MELATONIN RECEPTORS IN HYPOTHALAMUS On the cellular level, MT1 and MT2 couple mainly to Gi/o proteins and under certain circumstances to Gq/11 proteins (Brydon et al., 1999). G protein coupling depends also on the formation of specific heteromeric complexes either between MT1 and MT2 or with serotonin 5-HT2C receptors (Baba et al., 2013; Kamal et al., 2015). Typical effectors are adenylyl cyclases, ion channels, MAPKinases, and phospholipase C-beta (Dubocovich et al., 2010; Jockers et al., 2016). Both receptors recruit b-arrestins to assist receptor internalization (Cecon et al., 2018).

Circadian rhythm and melatonin receptors in the hypothalamus In mammals, the regulatory role of melatonin on the biological master clock located in the SCN has been extensively studied. This effect is observed at subnanomolar concentrations of melatonin indicating the implication of the high-affinity targets of melatonin, MT1, and MT2 receptors (Pevet, 2014; Liu et al., 2019). On the molecular level, Gi/o proteins followed by activation of G protein-coupled inwardly rectifying potassium (GIRK) channels (Hablitz et al., 2015) and Gq/11 proteins followed by activation of protein kinase C (Mc Arthur et al., 1997) are involved in the effects of melatonin in the SCN. Gene deletion studies of MT1 and/or MT2 demonstrated their respective roles in the mouse SCN (von Gall et al., 2000, 2002; Jin et al., 2003; Dubocovich et al., 2005). Circadian phase-shift experiments in MT1

MELATONIN RECEPTORS, BRAIN FUNCTIONS, AND THERAPIES and MT2 knockout mice with melatonin together with studies in wild-type mice treated with MT1- and MT2-selective compounds indicate that both receptors are necessary for the melatonin-mediated circadian regulation (Fig. 23.1A) (Liu et al., 1997; Dubocovich et al., 1998, 2005; Jin et al., 2003; Stein et al., 2020). In sighted people, living under normal light/dark conditions, the circadian rhythm is entrained to the 24 h day by light, which is perceived by the retina through melanopsin receptors and transmitted to the SCN through the retinohypothalamic tract. Among the different cues able to entrain the circadian rhythm, light is recognized as the strongest entrainment cue (Arendt, 2019). In the absence of light, an endogenous, free-running rhythm is revealed that is slightly longer than 24 h. In humans, this free-running rhythm is 24.18 h as determined under constant darkness conditions or in totally blind people. Consequently, in the absence of regular resetting to the 24 h rhythm, the free-running rhythm tends to drift, leading to progressively later bedtimes and wake times and abnormal sleep patterns (Czeisler et al., 1999). Apart from light, other factors regulate the hypothalamic master clock. Among them glucocorticoids, which play a major role in entraining/synchronizing peripheral clocks (Cuesta et al., 2015). Entrainment properties of melatonin were first demonstrated when administrated exogenously. Daily administration of melatonin to totally blind people, just 1 h before preferred bedtime, entrained them to a 24 h rhythm with improved sleep quality (Sack et al., 2000). Similar results were obtained with the melatonin receptor agonist tasimelteon (see chapter 4.3, Lockley et al., 2015). The effectiveness of exogenous melatonin has been shown in different clinical trials reporting the prevention or reduction of jet lag in healthy volunteers (Suhner et al., 1998; Arendt, 2019). The strongest evidence of the efficiency of exogenous melatonin comes from diseases where the circadian rhythm is affected like in blindness, the delayed sleep phase syndrome or in children with the Smith–Magenis syndrome, a mental retardation syndrome primarily ascribed to an inverted release pattern of melatonin (De Leersnyder, 2006; Arendt, 2019; Schroder et al., 2019). In those subjects, exogenous melatonin restores the endogenous circadian melatonin rhythm and at the same time improves sleep disorders. The same beneficial effects of melatonin have been reported for the sleep disorder symptoms associated with some neurodegenerative diseases such as Alzheimer or Parkinson disease (Sanchez-Barcelo et al., 2010; Cardinali, 2019). High doses (up to 10 mg) and long-duration trials reported no evidence of the development of tolerance or the toxicologic effect of melatonin use. The effects on sleep quality were also constant for any duration or dose used (Arendt, 1997; Seabra et al., 2000). The beneficial effects of melatonin on sleep are

349

not recapitulated by other even stronger sleep-promoting medications like benzodiazepines that do not act on melatonin receptors. Melatonin through its chronobiologic effect amplifies day–night differences in alertness and sleep quality in contrast to benzodiazepines (Cardinali and Golombek, 2009). Despite this clear evidence that exogenous melatonin entrains circadian rhythms, endogenous melatonin seems not to be essential for the entrainment of the daily rhythm as illustrated by patients with pinealectomy, as they do not show abnormal circadian rhythm length like totally blind people (Pfeffer et al., 2018). Similar studies in individuals with impaired melatonin receptor function, as reported recently for MT2 (Bonnefond et al., 2012), have not been performed yet. Along the same line, the absence of melatonin production observed in many mouse strains or the absence of melatonin receptors in melatonin producing mice does not compromise the 24 h daily rhythm of these animals (Tosini et al., 2014). Accumulating evidence on the deleterious effects of abruptly changing time cues in, for example, shift work and jet lag lead to the suggestion that one function of endogenous melatonin is to protect against abrupt short-term changes of phase by maintenance of the circadian status quo (Arendt, 2019). Taken together, melatonin, light, and glucocorticoids could provide a most efficient realignment of the central clock located in the SCN.

The pituitary–hypothalamic connection Evidence from animal studies indicate that melatonin and its receptors have a direct effect on the pituitary– hypothalamic connection. In seasonal breeders, such as hamster (Revel et al., 2006; Watanabe et al., 2007; Yasuo et al., 2007) and sheep (Hanon et al., 2008), melatonin regulate seasonal reproduction through the gonadotropin-releasing hormone (GnRH) (Korf, 2018; Nakayama and Yoshimura, 2018). In these animals, melatonin decodes the length of the day to trigger reproduction at the right time. This occurs through the activation of MT1 receptors in the pars tuberalis (PT) of the pituitary triggering the secretion of the thyroidstimulating hormone (TSH). Subsequent activation of TSH receptors in the ependymal cell layer, surrounding the third ventricle at the level of the medio-basal hypothalamus, increases the expression of type 2 deiodinase (Dio2), a key enzyme involved in the conversion of the thyroid hormone T4 into T3, followed by the production of GnRH to trigger reproduction (Korf, 2018; Nakayama and Yoshimura, 2018). Interestingly, in nonseasonal animals like mice strains that are melatoninproficient, part of this pathway is conserved. Indeed, in these mice, melatonin also increases Dio2 expression and T4 to T3 conversion in ependymal cells through

350

A. OISHI ET AL.

melatonin receptors located in the PT (Ono et al., 2008; Yasuo et al., 2009). This effect is mediated by MT1, as absent in MT1 knockout mice and preserved in MT2 knockout mice, demonstrating that MT1 plays central roles in the regulation of the photoperiodic response (Yasuo et al., 2009). The fact that the Siberian hamster, naturally lacking a functional MT2 receptor due to two nonsense mutations, showed strong seasonal reproduction further supports the importance of MT1 in this pathway (Weaver et al., 1996). In humans, the involvement of melatonin and its receptors in the hypothalamus and the pituitary remains unclear. Available evidence suggests a rather indirect effect of melatonin through the regulation of the circadian rhythm. This is most likely the case for the entrainment effect of melatonin on cortisol secretion, the final outcome of the HPA axis, and rhythms in totally blind free-running people (Lockley et al., 2000). Expression of melatonin receptors in CRH positive cells of the PVN (Wu et al., 2006; Klosen et al., 2019) might hint toward a direct participation of melatonin receptors in the regulation of CRH secretion, but functional evidence for this hypothesis is missing in humans.

DRUGS TARGETING MELATONIN RECEPTORS The short half-life of approximately 30 min of melatonin in the plasma limits the therapeutic use of melatonin itself.

Different nonselective agonists for MT1 or MT2 receptors with prolonged plasma half-life have been developed and are clinically used. Currently, four marketed drugs are available to treat pathologies related to circadian dysfunctions such as insomnia, sleep disorders, or depression (Liu et al., 2016) (Table 23.1).

Prolonged-release melatonin The endogenous melatonin production typically declines with age, although important interindividual differences exist (Zhdanova et al., 1998). A prolonged-release melatonin (PRM) formulation (commercialized as Circadin®) has been specifically developed for patients over 55 who suffer from primary insomnia. PRM is available in Europe and several other countries as a monotherapy for the short-term treatment of primary insomnia characterized by poor quality of sleep (Lemoine and Zisapel, 2012). On healthy middle-aged or elder participants, PRM (2 mg) did not affect nocturnal sleep electroencephalograms compared to placebo, contrary to benzodiazepine (Temezapam (20 mg)) and nonbenzodiazepine hypnotics (Zolpidem (10 mg)) (Arbon et al., 2015). Other clinical studies, in patients with primary insomnia (from 18 to 80 years old), revealed in short- (2 weeks) and long-term (6–12 month) trials that PRM increases sleep quality, total sleep time, and decreases sleep latency vs placebo (Luthringer et al., 2009; Wade et al., 2010; Lemoine et al., 2011). The same observations have been made in

Table 23.1 Features of clinically available drugs targeting melatonin receptors Affinity (Ki, nM) Drug Melatonin Prolonged-release melatonin Circadin® (Neurin) Ramelteon Rozerem® (Takeda) Tasimelteon Hetlioz® (Vanda) Agomelatine Valdoxan® (Servier)

MT1

MT2 5-HT2C

TMax (h)

t1/2 (h)

Indications

0.5–0.75 3.5–4.0

Jet lag insomnia Primary insomnia in elderly patients

2.45 0.08

0.50 0.38

No effect 0.25–2.0 No effect 0.75–3.0

0.014 0.30

0.11 0.07

No effect No effect

0.5–1.5 0.5–3.0

1.0–2.6 0.6–3.0

0.01

0.12

708

1.0–2.0

0.9–2.3

Primary insomnia Non-24-h sleep–wake disorder in adults Major depressive disorder in adults (15%) of enkephalinergic neurons in the posterior BNST may be glutaminergic. Some studies have suggested that CRF and met-enkephalin immunoreactivities could be colocalized in BNST neurons (Sakanaka et al., 1989), but as the two peptidergic systems engage opposing signaling mechanisms, the implications of those observations are unclear. VIPergic fibers have also been described to synapse and regulate BNST met-enkephalin neurons (Kozicz et al., 1998). Interestingly, the two opioid systems may have opposing functional roles. Dynorphin release and signaling within the BNST have been shown to produce dysphoria via KOR mechanisms, which could be blocked with KOR antagonists (McLaughlin et al., 2003; Land et al., 2008). In this regards, the effects of dynorphin appear concerted with those for CRF and PACAP in facilitating anxiogenic responses. Indeed, some observations suggested that the dynorphin system may be downstream of CRF signaling. In contrast to CRF/PACAP-mediated GPCR/Gas signaling in anxiety, BNST enkephalin GPCR/Gai signaling has been suggested to facilitate anxiolytic responses (Ragnauth et al., 2001; Wilson et al., 2003; Hebb et al., 2005). The lateral BNST enkephalinergic neurons project to the CeA and enkephalin release/signaling in the amygdala can diminish anxiety-like behaviors. These interpretations have been supported in general in knockout and peptide overexpression studies and appear consistent with other data implicating the BNST in modulating anxiety behavioral levels.

Calcitonin gene-related peptide The calcitonin gene can undergo tissue-specific alternative splicing to generate either calcitonin transcripts in thyroid follicular cells for calcium homeostasis, or CGRP mRNAs in neural tissues for sensory signaling. As for all bioactive peptides, calcitonin and CGRP are ligands for GPCRs, but unlike many other peptide receptor systems, the specificity of the receptors is dependent on the cell coexpression of receptor-associated membrane proteins (RAMP1, RAMP2, and RAMP3). The abilities of specific RAMPs to partner with the calcitonin receptor (CT) or calcitonin related-like receptor generate GPCR specificity to calcitonin, CGRP, amylin, and adrenomedullin. Among bioactive peptides, CGRP appears particularly sensitive to colchicine-induced expression. But even from in situ hybridization data, CGRP expression in the

392

S.E. HAMMACK ET AL.

CNS appears extensive in the diencephalon and brainstem structures. CGRP expression was high in the lateral hypothalamic areas, arcuate nucleus, posterior and peripeduncular thalamic nuclei, lateral olfactory tract, parabrachial nucleus, and cranial motor nuclei (Kresse et al., 1995). CGRP transcript expression was not apparent in the different amygdaloid nuclei; very few CGRP neurons were identified in the anterolateral and anteromedial BNST. However, there are significant levels of CGRPimmunoreactive fibers and terminals in the BNST and as described previously, the majority of the fibers coexpressed PACAP which represented dense afferent projections from the parabrachial nucleus (Missig et al., 2014, 2017). Accordingly, consistent with current appreciations of BNST PACAP/CRF activities, recent studies have also implicated BNST CGRP signaling in stress-related anxiogenic responses. Bilateral CGRP BNST infusions increased baseline startle and decreased open arm entries and times on elevated plus maze tests; the responses could be attenuated with the CGRP8–37 receptor antagonist or with the CRF receptor antagonist GSK876008, suggesting that coordinate PACAP/CGRP signaling is upstream of BNST CRF neurons (Sink et al., 2011, 2013a,b). These interpretations are supported by the observation of apparent CGRP-immunoreactive fibers synapsing onto CRF and met-enkephalin neurons (Kozicz and Arimura, 2001). Parabrachial PACAP and CGRP peptides can carry a variety of sensory signals including nociceptive information and aversive taste responses from illness and hence integrate sensory inputs within limbic structures to inform behavioral responses (Missig et al., 2017; Chen et al., 2018).

Neuropeptide Y The a-amidated 36 amino acid neuropeptide Y has been well studied in the sympathetic autonomic nervous system as a potent vasoconstrictor coexpressed in many catecholaminergic neurons. NPY binds to at least four Gai-coupled GPCRs designated as Y1, Y2, Y4, and Y5; hence, the receptors inhibit adenylyl cyclase and cAMP signaling. Y3 has not been identified definitively; Y6 is not expressed in rats and appears to be a pseudogene in humans. In the CNS, NPY is widely distributed in many regions including cortex, hippocampus, hypothalamus, brainstem nuclei, and limbic structures. Both NPY neurons and fibers were identified in the CeA, BLA, and BNST (Allen et al., 1984), and notably, the BNST and CeA/BLA appeared to have reciprocal NPY circuit connections. Some of the heavy NPYergic fibers in the BLA appeared to represent afferents from the BNST and the amygdalostriatal area, and conversely, lesions of the stria terminalis diminished BNST NPY fiber projections from the amygdaloid complex (Allen

et al., 1984; Leitermann et al., 2016). The roles of NPY in the behavioral consequences of stress, however, are still enigmatic as differential responses have been described based on the actions of postsynaptic Y1 versus predominantly presynaptic Y2 receptor actions. Whereas NPY signaling at the Y1 has been shown to be anxiolytic and buffer the behavioral consequences of stress, Y2 receptor signaling has been associated with augmented anxiety-related responses. NPY Y1 agonist injections into the amygdala or cerebral ventricles produced anxiolytic responses, and in coherence, Y1 receptor knockdown increased anxiety-like behaviors (Heilig et al., 1993; Wahlestedt et al., 1993; Broqua et al., 1995). This contrasted with intracerebroventricular infusions with Y2 receptor antagonists or Y2 receptor knockdown, which resulted in anxiolytic responses (Tschenett et al., 2003; Bacchi et al., 2006; Tasan et al., 2010); however, these Y2 effects were attributed to signaling in the CeA and BLA and not the BNST.

Somatostatin The 14- and 28-amino acid forms of somatostatin bind to GPCRs that are coupled to Gai to potently inhibit adenylyl cyclase/cAMP signaling. From immunocytochemical, in situ hybridization and immunoassay data, Sst is widely distributed in the CNS (Finley et al., 1981; Palkovitz et al., 1982; Kiyama and Emson, 1990). High levels of Sst neurons and fibers were identified in the piriform and neocortex (especially in layers V and VI), hippocampal hilar and CA3 areas, and hypothalamic nuclei and median eminence; notably, some of the highest expression levels in the telencephalon were found in the amygdaloid nuclear complex, lateral septum, nucleus accumbens and BNST, and in fiber tracts such as the diagonal band of Broca, stria terminalis, and the amygdalofugal pathways. In the brainstem, the localization of Sst neurons and fibers appeared extensive in many areas including the periaqueductal gray, locus coeruleus, lateral parabrachial nucleus, superior/inferior colliculi, nucleus of the lateral lemniscus, nucleus ambiguus, nucleus of the solitary tract, spinal trigeminal nucleus, and regions within the reticular formation. The expression and function of BNST and amygdala Sst have been examined in many studies although its roles in stress and behavior are not fully understood. Variably, approximately 5%–25% of the neurons in the all subdivisions of the BNST (including the oval nucleus) and the lateral/capsular divisions of the CeA have been described to express Sst; as for other peptides, BNST and CeA Sst neurons are GABAergic and some populations (up to 40%–70% of Sst neurons) show apparent CRF or NPY coexpression. The afferent inputs to the BNST Sst neurons have not been identified, but interestingly,

CHEMOARCHITECTURE OF THE BED NUCLEUS OF THE STRIA TERMINALIS the heavy lateral parabrachial CGRP (and by inference PACAP) projections to the BNST do not innervate the Sst neurons (Ye and Veinante, 2019). As in the lateral CeA, the BNST Sst neuronal fibers are likely contributory to inhibitory local microcircuits and the longrange efferent projections to the parabrachial nucleus and periaqueductal gray/dorsal raphe. The roles and mechanisms of Sst signaling in stress, mood, emotion, and behavioral abnormalities have not been fully elucidated, and some of the studies appear contradictory. Some studies suggest that the peptide is anxiolytic, antistress and antidepressive; Sst knockout mice displayed increased anxiety-like behaviors and emotionality scores, high-basal corticosterone levels, and downregulated CNS BDNF and GAD67 gene expression profiles that appeared consistent with patterns associated with depression (Lin and Sibille, 2015). Although the studies did not utilize conditional/tissue-selective but instead global Sst knockouts in mice, intracerebroventricular Sst receptor agonist infusions were in coherence and decreased stress-related hormone levels and specific behavioral responses. However, rather than anxiolytic, studies have also shown that Sst signaling is associated with fear and fear-conditioning freezing responses, which could be dissociated from defensive flight behaviors (Fadok et al., 2017). Optogenetic activation of CeA Sst neurons heightened freezing and decreased flight, whereas CRF neurons had the opposing effects of inhibiting freezing and increasing flight. The dichotomous nature of the responses appeared unique and suggested that Sst and CRF neurons form reciprocal inhibitory local circuits. Although these studies were in the CeA, similarities in circuit organizations may suggest similar mechanisms in the BNST.

Oxytocin and vasopressin Oxytocin and vasopressin (arginine vasopressin; Avp) are related a-amidated nonapeptides that differ only by two amino acids. GPCR Oxt receptor (OxtR) signaling through Gai/Gao/Gaq is often related to uterus smooth muscle relaxation in childbirth and lactation processes; Avp V1 and V2 receptor signaling through Gaq and Gas, respectively, have been best studied with respect to homeostatic water balance and blood pressure control mechanisms. In the CNS, both Oxt and Avp are now understood to have wide-ranging behavioral effects (Caldwell et al., 2008; Jurek and Neumann, 2018). Oxt and Avp, with their respective carrier proteins neurophysin 1 and neurophysin 2, have been mapped extensively in the brain and the highest levels of expression are found in the supraoptic, paraventricular and suprachiasmatic nuclei of the hypothalamus. In colchicine-treated preparations, a large number of Avp-immunoreactive neurons

393

was identified in the BNST; by contrast, only a few scattered Oxt immunoreactive neurons were apparent in the BNST which appeared to be confined largely to the posterior division associated with reproductive activities (Sofroniew, 1985; Bingham and Viau, 2008). These observations were largely corroborated by in situ hybridization studies for Oxt and Avp transcripts although the radiolabeled grain densities for BNST Oxt neurons were adjudged to be very low (Caldwell et al., 1989; Hallbeck et al., 1999). However, in situ hybridization data for Oxt and Avp receptor transcripts revealed moderate to high levels of OxtR and variant V1aR expression in the BNST (Yoshimura et al., 1993; Ostrowski, 1998); Avp V1b and V2 receptors were not apparent in this region. The high levels of receptors suggest that Oxt and Avp may be able to regulate BNST function, yet as endogenous BNST Oxt expression levels are very low, these observations may implicate long-distance Oxt fiber projections from other regions, especially the hypothalamus or CeA, as mechanisms of BNST OxtR activation. The roles of BNST Oxt and Avp receptor signaling are still unclear. The BNST is sexually dimorphic and some of the complexity may arise from gonadal steroid hormone regulation of Oxt and OxtR expression (Caldwell et al., 1989; Ostrowski, 1998); estrogen, for example, can upregulate Oxt/OxtR levels. Broadly, in concert with its general prosocial, empathy, coping, and antistress behavioral actions (Jurek and Neumann, 2018), Oxt/OxtR signaling in the BNST appears to be anxiolytic in pharmacological animal and human studies. However, these responses contrast with BNST Oxt/OxtR mechanisms that facilitate fear to acute discrete cues. Fear learning can be complex from changes in neurotransmitter/neuropeptide expression and function from stress experience and brain structural plasticity, and one interpretation of these apparent divergent BNST Oxt actions is that Oxt may heighten acute fear responses as a rapid means of threat detection for survival and preservation but ameliorate responses to diffuse, undefined and long-term threats associated with contextual fear and anxiety-related behaviors (Moaddab and Dabrowska, 2017; Janecek and Dabrowska, 2019). The roles of BNST Avp signaling are similarly complex. Similar to gonadal steroid regulation of Oxt, Avp, and Avp receptors are regulated by androgens. Testosterone priming in neonates, for example, can regulate the number of Avp expressing neurons in the BNST (Bingham and Viau, 2008) and gonadectomy can decrease BNST Avp fiber innervation of distal targets (de Vries et al., 1985). Of note, the Avp neurons in the BNST and amygdala have several parallels. Unlike Avp neurons in other CNS regions, the BNST and amygdala Avp cells are gonadal hormone sensitive and share distal projections. Not only were the central part of the dlBNST (BNSTc) and the CeA shown to have common

394

S.E. HAMMACK ET AL.

projections, but in duality, the medial BNST and the medial amygdala Avp neurons had comparable innervation patterns. In these studies, including lesion experiments, the Avp target projections included the lateral septum, lateral habenula, the diagonal band of Broca, ventral hippocampus, medial BNST, and medial amygdala (de Vries and Buijs, 1983; de Vries et al., 1985; de Vries, 2008). In aggregate, these observations not only suggested common BNST/amygdala Avp functions but reciprocal BNST and amygdala Avp circuit interactions for mutual modulation. Avp has been best studied with respect to facilitating offensive aggression, especially between males, but overlapping with Oxt effects, Avp can have prosocial/affiliative learning and memory, and anxiolytic behaviors (Caldwell et al., 2008). However, Avp-mediated effects can be variable highlighting again that the responses may be dependent on environment, experience, and resulting neuroplasticity.

BNST SEXUAL DIMORPHISM AND CONTROL OF SEXUAL BEHAVIORS The BNST is sexually dimorphic with respect to structure and neurophenotypic expression to differentially modulate sexual and stress-related behaviors. For example, males and females have differential responses to stress, and for many reasons including stress-specific maladaptations, females are twice as likely than males to suffer from psychopathologies, including depression and PTSD (Hammack and May, 2015). Relatedly, these differences extend to male vs female social interactive behaviors which include social communication, aggression and defensive behaviors, precopulatory and mating behaviors and parental care (Lebow and Chen, 2016). In the processing of olfactory cues, for example, the BNST has roles in opposite-sex solicitation. The BNST is important for male sexual motivation (Hull and Dominguez, 2019); in studies with male Syrian hamsters, for example, the BNST facilitated opposite-sex odor preference (Martinez and Petrulis, 2011). However, in females, the BNST was also shown to have sexual behavioral roles in vaginal markings in response to male odors (Martinez and Petrulis, 2011). How BNST dimorphism may be contributory to male vs female sexual behaviors is not well understood. Although there may be no apparent differences in overall BNST area and cell density between male and female rats (Stefanova and Ovtscharoff, 2000), many BNST subnuclei, especially the principal nucleus in the posterior BNST (BNSTp), demonstrate sex differences across species, including humans (Hines et al., 1985, 1992; Allen and Gorski, 1990; Swaab, 2003). The BNSTp in males is larger and has more neurons than females from cell survival during development (Forger, 2009). The BNST is responsive to gonadal steroids from androgen and estrogen receptors

(ER) expression but notably between ERa/b subtypes, the ERa levels are high in the BNSTp of females but absent in males (Kelly et al., 2013). Further, the BNSTp has high levels of BNST aromatase activity, which catalyzes the conversion of testosterone to estrogen, and in aggregate with ER expression and signaling, may contribute to the differences in male vs female sexual and reproductive responses. Although the studies have been limited, the differences in BNST sex hormone effects may contribute to some of the differences in BNST cytoarchitecture, neurophenotypic pattern and epigenetic marker expression and responses to stress between the sexes. The immunocytochemical staining density for synaptophysin, an integral synaptic vesicle protein typically used as a presynaptic marker, was greater in males than females in the posterior BNST (Carvalho-Netto et al., 2011); notably, chronic stress decreased synaptophysin staining only in females and not males. Importantly, as in stress-induced neuroplasticity, the expression of the Arc activity-dependent immediate early gene, which regulates excitatory synaptic function, was increased in male rats following copulation, suggesting that sexual activity can impact BNST synaptic plasticity (Turner et al., 2019). As described previously, CRF is a primary stress peptide and although some studies have failed to show sex differences in BNST CRF mRNA levels or peptide staining, others have shown that the number of BNSTov CRF neurons is greater in females than males (Uchida et al., 2019). Acute stress increased BNSTov CRF transcripts in males and not females; by contrast, acute stress augmented CRF fiber staining in the BNST fusiform nucleus in females only. Surprisingly, many of the measures for CRF were unchanged from control levels after chronic stress paradigms for either sex (Sterrenburg et al., 2011, 2012). How these differential CRF expression patterns may facilitate stress-induced anxiety-related abnormalities in females is still unclear. Although BNST Oxt expression levels are relatively insignificant and do not show dimorphism, Oxt receptor signaling appeared to mediate vaginal marking behaviors (Lebow and Chen, 2016). Avp neuronal and fiber staining was greater in males than females in the anterior and posterior BNST (DiBenedictis et al., 2017); BNST Avp levels were elevated in rodents when presented with potential or existing mates (Lebow and Chen, 2016) and appeared important for male social investigations and urine markings in the presence of females (Rigney et al., 2019). By contrast, GABAergic fibers and leu-enkephalin staining levels were higher in females than males, which may be consequences of androgen inhibition (Stefanova and Ovtscharoff, 2000). There were also differential sex responses in cfos/FosB immunoreactivity and epigenetic DNA methylation and acetylation states following acute or chronic stress (Sterrenburg et al., 2011, 2012). While the implications of these differences between the sexes

CHEMOARCHITECTURE OF THE BED NUCLEUS OF THE STRIA TERMINALIS are not understood, they can impact downstream targets including the dorsal raphe, periaqueductal gray, nucleus accumbens, and medial prefrontal cortex, suggesting that males and females may preferentially activate different segments of neural stress and behavior circuits (Salvatore et al., 2018).

HUMAN BNST The human and rodent BNST share neuroanatomical and neurochemical similarities. As in rodents, the human BNST is bounded by the nucleus accumbens, anterior thalamus, inferior lateral ventricles, and the internal capsule; it is approximately 190 mm3 or the approximate size of a sunflower seed (Avery et al., 2016) and hence not well resolved spatially even with the best MRI scanners (Shackman and Fox, 2016). The structure is also subdivided, but rather than the two broad dorsolateral/ dorsomedial BNST regions described for rodents, the dorsal BNST in humans has been parsed into lateral, central, and medial sectors (Lesur et al., 1989). In analogy to animal data, the human BNST is interconnected with the amygdala through the stria terminalis and ventral amygdalofugal pathways, and parcellated comparable to rodent subdivisions. For obvious reasons, the human BNST circuits cannot be mapped in fine structural detail using tracing techniques, but from recent neuroimaging techniques, the human and rodent BNST appear to share broad similarities in structural and functional connectivity. Diffusion tensor MRI and probabilistic tractography demonstrated that the human BNST, as in rodents, is structurally connected to other limbic structures including the centromedial amygdala and hippocampus, the nucleus accumbens, caudate, putamen, and related regions of the basal ganglia, and the thalamus; a probable connection to the limen insulae of the insular cortex appears distinct to humans. Functional connectivity from resting state fMRI, which can be different from structural connectivity in that monosynaptic fiber tracts between regions may be absent, revealed comparable results in BNST connections with limbic, thalamic, and basal ganglia structures. However, the paracingulate gyrus, which is distinct from the rodent medial prefrontal cortex and not identified as a BNST structural connection, was found to be a functional BNST-associated target in these studies. Although the human studies require additional work, these results suggest that the human and rodent BNST neurocircuits have broad parallels. From these structural and functional similarities, the human BNST may be anticipated to demonstrate neuropeptide expression patterns comparable to those found in rodents. The immunocytochemical processing of human brain sections, however, presents unique challenges related to the use of tissues of variable postmortem

395

duration. Hence, for some studies, the human brain staining patterns can be complicated by protein/peptide degradation and artifacts from poor tissue preservation. Nevertheless, many of the peptides identified in rodent BNST have been found in humans, though not as comprehensively characterized as in rodents. Apparent Sst immunoreactivity, for example, was identified in scattered neurons and dense fiber/varicosity networks in the dBNST/vBNST of human and nonhuman primates (Bennett-Clarke and Joseph, 1986; Lesur et al., 1989; Walter et al., 1991; Kruijver et al., 2000; Kovner et al., 2019); as described previously, a substantial fraction of the Sst neurons may coexpress CRF. Similarly, NPY-, enkephalin-, neurotensin-, VIP-, and substance P-immunoreactive fibers were found in the human BNST; notably, in departure from patterns observed in rodents, VIP-, somatostatin-, enkephalin-, and neurotensin-immunoreactive neurons appeared in a discrete cluster within the central BNST (Lesur et al., 1989; Walter et al., 1991; Zhou et al., 1995). Also, in contrast to rodents, Avp fibers and neurons were observed in the human and nonhuman primate BNST; no Oxt immunoreactive structures were apparent (Fliers et al., 1986; Caffe et al., 1989). Human BNST PACAP levels were measured only by radioimmunoassay and found to be among the highest in the brain (Palkovitz et al., 1995), consistent with rodent data described previously. As in rodents, the human BNST is sexually dimorphic in structure. The male posterior dorsomedial BNST (adjacent to the fornix) and central BNST were shown to be larger approximately 2.5- and 1.5-fold, respectively, compared to females, and correspondingly, the levels of VIP- and somatostatin- immunoreactive material in the female central BNST were diminished compared to males (Allen and Gorski, 1990; Zhou et al., 1995; Kruijver et al., 2000). Critically, these dimorphic differences extend into gender identity. Although BNST size between heterosexual and homosexual males was not different, male-to-female transsexuals presented small BNSTs comparable in size to heterosexual females (Zhou et al., 1995; Kruijver et al., 2000). Even though these observations suggested that gender identity is determined developmentally, interestingly, male and female BNST size did not appear different until adulthood (Chung et al., 2002). These results appear at odds with how transsexuals can feel at childhood of being born into the wrong sex. While transgender issues may be unrelated to BNST sexual differentiation, the cellular, molecular, and/or pubertal hormonal-dependent changes in the BNST function related to gender identity may precede the overt changes in BNST structure. Hence, our understanding of the chemoarchitecture and neurocircuitry of the human BNST is still very limited and requires additional study.

396

S.E. HAMMACK ET AL.

OVERVIEW AND CONCLUSIONS The BNST is a small and complex sexually dimorphic ventral forebrain structure important in the integration of many sensory and regulatory inputs to coordinate physiological and behavioral responses, especially to stress challenges. It is structurally and functionally associated with the amygdala nuclear complex but rather than the mediation of fear responses alone, the BNST also appears to respond to long-duration and/or temporally unpredictable threats, associated with anxiety. Hence, many studies have suggested that sustained BNST activation from chronic stress may result in maladaptive neuroplasticity within the structure to facilitate psychopathologies such as PTSD. The BNST has also been well studied with sexual responses, and clearly, there are intersections between stress-mediated anxiety responses with social and sexual motivational behaviors (Bastida et al., 2014). Understanding BNST neurocircuits driving the responses continues to be a challenge. The BNST

is largely GABAergic with a small population of glutaminergic neurons, but the high and heterogenous coexpression of many neuropeptides in BNST neurons (Fig. 26.2) obscures how these circuits actually work. The CRF/GABA neurons appear to represent the major output of the BNST, but how the intrinsic CRF/GABA interneurons within the BNST local circuits impact the long-distance output neurons is unclear. Hence, there are a number of outstanding questions. A diverse set of stimulatory and inhibitory peptides has been identified in the BNST; are they all GABAergic also or are some coexpressed in glutaminergic neurons? A similar set of peptides is found in afferents to the BNST; how are these peptides and transmitters integrated into the BNST circuits? And in particular, what does output from the same stimulatory CRF but inhibitory GABAergic neuron mean to the downstream neuronal target? In one possible mechanism, the inhibitory peptides and transmitters directly innervate the CRF/GABA output neurons, whereas stimulatory afferents target

Fig. 26.2. BNSTov phenotypic heterogeneity. Diverse peptides in BNSTov neurons and fibers have been identified. The peptidergic fibers include afferents from extrinsic peptidergic neurons, and intrinsic neuronal networks and projection fibers to other BNST subdivisions or long-distance CNS targets. The peptides can be coupled variably to Gas, Gaq, Gai, and/or G protein dependent- or independent signaling cascades. Using the PACAP and CRF neurons to illustrate potential network mechanisms (bold font), the extrinsic PACAP/CGRP projections (i.e., from the lateral parabrachial nucleus) may synapse onto a BNSTov CRF/GABA interneuron to inhibit an output inhibitory CRF/GABA neuron (1). The activation of a local inhibitory neuron may inhibit GABAergic output of the BNST, leading to downstream disinhibition to net a facilitatory stress signal. Stimulatory extrinsic PACAP afferents may also directly stimulate intrinsic CRF and/or glutaminergic neurons, or BNSTov PACAP may have direct output projections (2) to affect stress responses. Mechanistic refinements and alternatives may be apparent from future work. Peptide abbreviations as in text.

CHEMOARCHITECTURE OF THE BED NUCLEUS OF THE STRIA TERMINALIS intermediary inhibitory interneurons which in turn innervate the inhibitory output cells. In this model, the networks disinhibit inhibitory neurons resulting in positive outputs from the BNST (Fig. 26.2). Alternatively, the neuropeptides have dynamic functional roles in BNST function. Neuropeptide vs transmitter release is dependent on stimulation frequency. Peptidergic neurotransmission via dense core vesicle release at the terminals requires high-frequency stimulation or bursting activity and one heuristic interpretation of CRF/GABA signaling may be that GABAergic signaling prevails under basal nonstress conditions and that high or chronic stress stimulatory conditions from peptide Gas/ Gaq presynaptic or postsynaptic neuronal activation (such as PACAP) may drive CRF signaling (Fig. 26.2), to shift the circuit from a anxiolytic to anxiogenic state. The interacting peptide/interneuron circuits within the BNST may fine-tune the inputs to the CRF/GABA neurons so that the appropriate anxiolytic or anxiogenic responses may be generated to specific stressors or cues. In this model, chronic stress-induced peptidergic activation and neuroplasticity may also alter BNST neuronal cytoarchitecture, phenotypic expression, circuit organization, or synaptic function to facilitate maladaptations that promote behavioral abnormalities. Variations in these and other models may become apparent from ongoing high-resolution studies and from the many recent reviews, clear understandings of BNST circuits and mechanisms may provide insights for novel therapeutics to some rather intractable and devastating pain and behavioral disorders.

REFERENCES Agarwal A, Halvorson LM, Legradi G (2005). Pituitary adenylate cyclase-activating polypeptide (PACAP) mimics neuroendocrine and behavioral manifestations of stress: evidence for PKA-mediated expression of the corticotropin-releasing hormone (CRH) gene. Brain Res Mol Brain Res 138: 45–57. 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 YS, Roberts GW, Bloom SR et al. (1984). Neuropeptide Y in the stria terminalis: evidence for an amygdalofugal projection. Brain Res 321: 357–362. Arimura A, Somogyva´ri-Vigh A, Miyata A et al. (1991). Tissue distribution of PACAP as determined by RIA: highly abundant in the rat brain and testes. Endocrinology 129: 2787–2789. Avery SN, Clauss JA, Blackford JU (2016). The human BNST: functional roles in anxiety and addiction. Neuropsychopharmacol Rev 41: 126–141. Bacchi F, Mathe AA, Jimenez P et al. (2006). Anxiolytic-like effect of the selective neuropeptide Y-Y2 receptor antagonist BIIE0246 in the elevated plus-maze. Peptides 27: 3202–3207.

397

Bale TL, Vale WW (2003). Increased depression-like behaviors in corticotropin-releasing factor receptor-2-deficient mice: sexually dichotomous responses. J Neurosci 23: 5295–5301. Bastida CC, Puga F, Gonzalez-Lima F et al. (2014). Chronic social stress in puberty alters appetitive male sexual behavior and neural metabolic activity. Horm Behav 66: 220–227. Bennett-Clarke CA, Joseph SA (1986). Immunocytochemical localization of somatostatin in human brain. Peptides 7: 877–884. Bingham B, Viau V (2008). Neonatal gonadectomy and adult testosterone replacement suggest an involvement of limbic arginine vasopressin and androgen receptors in the organization of the hypothalamic-pituitary-adrenal axis. Endocrinology 149: 3581–3591. Blechman J, Levkowitz G (2013). Alternative splicing of the pituitary adenylate cyclase-activating polypeptide receptor PAC1: mechanisms of fine tuning of brain activity. Front Endocrinol 4: 55. Broqua P, Wettstein JG, Rocher MN et al. (1995). Behavioral effects of neuropeptide Y receptor agonists in the elevated plus-maze and fear-potentiated startle procedures. Behav Pharmacol 6: 215–222. Caffe AR, Van Ryen PC, Van de Woude TP et al. (1989). Vasopressin and oxytocin systems in the brain and upper spinal cord of Macaque fascicularis. J Comp Neurol 287: 302–325. Caldwell JD, Brooks PJ, Jirikowski GF et al. (1989). Estrogen alters oxytocin mRNA levels in the preoptic area. J Neuroendocrinol 1: 273–278. Caldwell HK, Lee J-J, Macbeth AH et al. (2008). Vasopressin: behavioral roles of an “original” neuropeptide. Prog Neurobiol 84: 1–24. Carraway R, Leeman SE (1973). The isolation of a new hypotensive peptide neurotensin from bovine hypothalami. J Biol Chem 248: 6854–6861. Carraway R, Leeman SE (1975). The amino acid sequence of a hypothamlamic peptide neurotensin. J Biol Chem 250: 1907–1911. Carvalho-Netto EF, Myers B, Jones K et al. (2011). Sex differences in synaptic plasticity in stress-responsive brain regions following chronic variable stress. Physiol Behav 104: 242–247. Chen JY, Campos CA, Jarvie BC et al. (2018). Parabrachial CGRP neurons establish and sustain aversive taste memories. Neuron 100: 891–899. Choi DC, Nguyen MMN, Tamashiro KLK et al. (2006). Chronic social stress in the visible burrow system modulates stress-related gene expression in the bed nucleus of the stria terminalis. Physiol Behav 89: 301–310. Chung WCJ, 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. Condro MC, Matynia A, Foster NN et al. (2016). High-resolution characterization of a PACAP-EGFP transgenic mouse model for mapping PACAP-expressing neurons. J Comp Neurol 524: 3827–3848.

398

S.E. HAMMACK ET AL.

Cortes R, Ceccatelli S, Schalling M et al. (1990). Differential effects of intracerebroventricular colchicine administration on the expression of mRNAs for neuropeptides and neurotransmitter enzymes, with special emphasis on galanin: an in situ hybridization study. Synapse 6: 369–391. Cummings S, Elde R, Ells J et al. (1983). Corticotropinreleasing factor immunoreactivity is widely distributed within the central nervous system of the rat: an immunohistochemical study. J Neurosci 3: 1355–1368. Dabrowska J, Martinon D, Moaddab M et al. (2016). Targeting corticotropin-releasing factor projections from the oval nucleus of the bed nucleus of the stria terminalis using cell-type specific neuronal tracing studies in mouse and rat brain. J Neuroendocrinol 28: 10.111/jne 12442. https:// doi.org/10.1111/jne.12442. Daniel SE, Rainnie DG (2016). Stress modulation of opposing circuits in the bed nucleus of the stria terminalis. Neuropsychopharmacology 41: 103–125. Day HE, Curran EJ, Watson SJJ et al. (1999). Distinct neurochemical populations in the rat central nucleus of the amygdala and bed nucleus of the stria terminalis: evidence for their selective activation by interleukin-1beta. J Comp Neurol 413: 113–128. de Vries GJ (2008). Sex differences in vasopressin and oxytocin innervation of the brain. Prog Brain Res 170: 17–27. 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. de Vries 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. Dedic N, Chen A, Deussing JM (2018). The CRF family of neuropeptides and their receptors—mediators of the central stress response. Curr Mol Pharmacol 11: 4–31. Deussing JM, Chen A (2018). The corticotropin-releasing factor family: physiology of the stress response. Physiol Rev 98: 2225–2286. DiBenedictis BT, Naussbaum 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. Dong HW, Swanson LW (2004). Organization of axonal projections from the anterolateral area of the bed nuclei of the stria terminalis. J Comp Neurol 468: 277–298. Dong HW, Petrovich GD, Swanson LW (2001). Basic organization of prejections from the oval and fusiform nuclei of the bed nuclei of the stria terminalis in adult rat brain. J Comp Neurol 436: 430–455. Fadok JP, Krabbe S, Markovic M et al. (2017). A competitive inhibitory circuit for selection of active and passive fear responses. Nature 542: 96–100. Finley JCW, Maderdrut JL, Roger LJ et al. (1981). The immunocytochemical localization of somatostatin-containing neurons in the rat central nervous system. Neuroscience 6: 2173–2192. 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. Forger NG (2009). Control of cell number in the sexually dimorphic brain and spinal cord. J Neuroendocrinol 21: 393–399. Girard BA, Lelievre V, Braas KM et al. (2006). Noncompensation in peptide/receptor gene expression and distinct behavioral phenotypes in VIP- and PACAPdeficient mice. J Neurochem 99: 499–513. Goode TD, Maren S (2017). Role of the bed nucleus of the stria terminalis in aversive learning and memory. Learn Mem 24: 480–491. Hallbeck M, Hermanson O, Blomqvist A (1999). Distribution of preprovasopressin mRNA in the rat central nervous system. J Comp Neurol 411: 181–200. Hammack SE, May V (2015). Pituitary adenylate cyclase activating polypeptide in stress-related disorders: data convergence from animal and human studies. Biol Psychiatry 78: 167–177. Hammack SE, Richey KJ, Schmid MJ et al. (2002). The role of corticotropin-releasing hormone in the dorsal raphe nucleus in mediating the behavioral consequences of uncontrollable stress. J Neurosci 22: 1020–1026. Hammack SE, Schmid MJ, LoPresti ML et al. (2003). Corticotropin releasing hormone type 2 receptors in the dorsal raphe nucleus mediate the behavioral consequences of uncontrollable stress. J Neurosci 23: 1019–1025. Hammack SE, Cheung J, Rhodes KM et al. (2009a). Chronic stress increases pituitary adenylate cyclase-activating peptide (PACAP) and brain-derived neurotrophic factor (BDNF) mRNA expression in the bed nucleus of the stria terminalis (BNST): roles for PACAP in anxiety-like behavior. Psychoneuroendocrinology 34: 833–843. Hammack SE, Guo J-D, Hazra R et al. (2009b). The response of neurons in the bed nucleus of the stria terminalis to serotonin: implications for anxiety. Prog Neuropsychopharmacol Biol Psychiatry 33: 130–1320. Hammack SE, Todd TP, Kocho-Schellenberg M et al. (2015). Role of the bed nucleus of the stria terminalis in the acquisition of contextual fear at long or short context-shock intervals. Behav Neurosci 129: 673–678. Hannibal J (2002). Pituitary adenylate cyclase activating peptide in the rat central nervous system: an immunocytochemical and in situ hybridization study. J Comp Neurol 453: 389–417. Hannibal J, Mikkelsen JD, Fahrenkrug J et al. (1995). Pituitary adenylate cyclase-activating peptide gene expression in corticotropin-releasing factor-containing parvicellular neurons of the rat hypothalamic paraventricular nucleus is induced by colchicine, but not by adrenalectomy, acute osmotic, ether, or restraint stress. Endocrinology 136: 4116–4124. Harmar AJ, Fahrenkrug J, Gozes I et al. (2012). Pharmacology and functions of receptors for vasoactive intestinal peptide and pituitary adenylate cyclase-activating polypeptide: IUPHAR review 1. Br J Pharmacol 166: 4–17. Hashimoto H, Nogi H, Mori K et al. (1996). Distribution of the mRNA for a pituitary adenylate cyclase-activating

CHEMOARCHITECTURE OF THE BED NUCLEUS OF THE STRIA TERMINALIS polypeptide receptor in the rat brain: an in situ hybridization study. J Comp Neurol 371: 567–577. Hashimoto H, Shintani N, Tanaka K et al. (2001). Altered psychomotor behaviors in mice lacking pituitary adenylate cyclase-activating polypeptide (PACAP). Proc Natl Acad Sci U S A 98: 13355–13360. Hebb AL, Poulin J-F, Roach SP et al. (2005). Cholecystokinin and endogenous opioid peptides: interactive influence on pain, congnition and emotion. Prog Neuropsychopharmacol Biol Psychiatry 29: 1225–1238. Heilig M, McLeod S, Brot M et al. (1993). Anxiolytic-like actions of neuropeptide Y: mediation by Y1 receptors in amygdala, and dissociation from food intake effects. Neuropsychopharmacology 8: 357–363. Henckens MJAG, Printz Y, Shamgar U et al. (2017). CRF receptor type 2 neurons in the posterior bed nucleus of the stria terminalis critically contribute to stress recovery. Mol Psychiatry 22: 1691–1700. Hill JM, Agoston DV, Gressens P et al. (1994). Distribution of VIP mRNA and two distinct VIP binding sites in the developing rat brain. Relation to ontogenic events. J Comp Neurol 342: 186–205. Hines M, Davis FC, Coquelin A et al. (1985). Sexually dimorphic regions in the medial preoptic area and the bed nucleus of the stria terminalis of the guinea pig brain: a description and an investigation of their relationship to gonadal steroids in adulthood. J Neurosci 5: 40–47. Hines M, Allen LS, Gorski RA (1992). Sex differences in the subregions of the medial nucleus of the amygdala and the bed nucleus of the stria terminalis of the rat. Brain Res 579: 321–326. Hu P, Maita I, Kwok C et al. (2020a). Chronic stress induces anxiety-associated negative valence behavior by activating a corticotropin-releasing hormone-related molecular signature in the oval bed nucleus of the stria terminalis. J Neurosci 40: 2519–2537. Hu P, Maita I, Phan ML et al. (2020b). Early-life stress alters affective behaviors in adult mice through persistent activation of CRH-BDNF signaling in the oval bed nucleus of the stria terminalis. Transl Psychiatry 10: 396. https://doi.org/ 10.1038/s41398-020-01070-3. Hull EM, Dominguez JM (2019). Neuroendocrine regulation of male sexual behavior. Compr Physiol 9: 1383–1410. Janecek M, Dabrowska J (2019). Oxytocin facilitates adaptive fear and attenuates anxiety responses in animal models and human studies-potential interaction with the corticotropinreleasing factor (CRF) system in the bed nucleus of the stria terminalis (BNST). Cell Tissue Res 375: 143–172. Jennes L, Stumpf WE, Kalivas PW (1982). Neurotensin: a topographical distribution in rat brain by immunocytochemistry. J Comp Neurol 210: 211–224. Ju G, Swanson LW (1989). Studies on the cellular architecture of the bed nuclei of the stria terminalis in the rat: I. cytoarchitecture. J Comp Neurol 280: 587–602. Ju G, Swanson LW, Simerly RB (1989). Studies on the cellular architecture of the bed nuclei of the stria terminalis in the rat: II. Chemoarchitecture. J Comp Neurol 280: 603–621.

399

Jurek B, Neumann ID (2018). The oxytocin receptor: from intracellular signaling to behavior. Physiol Rev 98: 1805–1908. Kelly DA, Varnum MM, Krentzel AA et al. (2013). Differential control of sex differences in estrogen receptor alpha in the bed nucleus of the stria terminalis and anteroventral periventricular nucleus. Neuroendocrinology 154: 3836–3846. Kimura C, Ohkubo S, Ogi K et al. (1990). A novel peptide which stimulates adenylate cyclase: molecular cloning and characterization of the ovine and human cDNAs. Biochem Biophys Res Commun 166: 81–89. Kiyama H, Emson PC (1990). Distribution of somatostatin mRNA in the rat nervous system as visualized by a novel non-radioactive in situ hybridization histochemistry procedure. Neuroscience 38: 223–244. Kono J, Konno K, Talukder AH et al. (2017). Distribution of corticotropin-releasing factor neurons in the mouse brain: a study using corticotropin-releasing factor-modified yellow fluorescent protein knock-in mouse. Brain Struct Funct 222: 1705–1732. Kovner R, Fox AS, French DA et al. (2019). Somatostatin gene and protein expression in the non-human primate central extended amygdala. Neuroscience 400: 157–168. Kozicz T, Arimura A (2001). Axon terminals containing CGRP-immunoreactivity form synapses with CRF- and met-enkephalin-immunopositive neurons in the laterodorsal division of the bed nucleus of the stria terminalis in the rat. Brain Res 893: 11–20. Kozicz T, Vigh S, Arimura A (1997). Axon terminals containing PACAP- and VIP-immunoreactivity form synapses with CRF-immunoreactive neurons in the dorsolateral division of the bed nucleus of the stria terminalis. Brain Res 767: 109–119. Kozicz T, Vigh S, Arimura A (1998). Immunocytochemical evidence for PACAP and VIP interaction with metenkephalin and CRF containing neurons in the bed nucleus of the stria terminalis. Ann N Y Acad Sci 11: 523–528. Kresse A, Jaobowitz DM, Skofitsch G (1995). Detailed mapping of CGRP mRNA expression in the rat central nervous system: comparison with previous immunocytochemical findings. Brain Res Bull 36: 261–274. 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. Land BB, Bruchas MR, Lemos JC et al. (2008). The dysphoric component of stress is encoded by activation of the dynorphin kappa-opioid system. J Neurosci 28: 407–414. Larriva-Sahd J (2006). Histological and cytological study of the bed nucleus of the stria terminalis in adult rat. II oval nucleus: extrinsic inputs, cell types, neuropil and neuronal modules. J Comp Neurol 497: 772–807. Lebow MA, Chen A (2016). Overshadowed by the amygdala: the bed nucleus of the stria terminalis emerges as a key to psychiatric disorders. Mol Psychiatry 21: 450–463. Lebow M, Neufeld-Cohen A, Kuperman Y et al. (2012). Susceptibility to PTSD-like behavior is mediated by corticotropin-releasing factor receptor type 2 levels in the

400

S.E. HAMMACK ET AL.

bed nucleus of the stria terminalis. J Neurosci 32: 6906–6916. Leitermann RJ, Rostkowski AB, Urban JH (2016). Neuropeptide Y input t other rat basolateral amygdala complex and modulation by conditioned fear. J Comp Neurol 524: 2418–2439. Lesur A, Gaspar P, Alvarez C et al. (1989). Chemoanatomic compartments in the human bed nucleus of the stria terminalis. Neuroscience 32: 181–194. Lewis K, Li C, Perrin MH et al. (2001). Identification of urocortin III, an additional member of the corticotropin-releasing factor (CRF) family with high affinity for the CRF2 receptor. Proc Natl Acad Sci U S A 98: 7570–7575. Lezak KR, Roman CW, Braas KM et al. (2014). Regulation of bed nucleus of the stria terminalis PACAP expression by stress and corticosterone. J Mol Neurosci 54: 477–484. Li C, Vaughan J, Sawchenko PE et al. (2002). Urocortin IIIimmunoreactive projections in rat brain: partial overlap with sites of type 2 corticotropin-releasing factor receptor expression. J Neurosci 22: 991–1001. Lin LC, Sibille E (2015). Somatostatin, neuronal vulnerability and behavioral emotionality. Mol Psychiatry 20: 377–387. Martinez LA, Petrulis A (2011). The bed nucleus of the stria terminalis is critical for sexual solicitation, but not for opposite-sex odor preference, in female Syrian hamsters. Horm Behav 60: 651–659. McLaughlin JP, Marton-Popovich M, Chavkin C (2003). Kappa opioid receptor antagonism and prodynorphin gene disruption block stress-induced behavioral responses. J Neurosci 23: 5674–5683. Miles OW, Maren S (2019). Role of the bed nucleus of the stria terminalis in PTSD: insights from preclinical models. Front Behav Neurosci 13: 1–14. Missig G, Roman CW, Vizzard MA et al. (2014). Parabrachial nucleus (PBn) pituitary adenylate cyclase activating polypeptide (PACAP) signaling in the amygdala: implication for the sensory and behavioral effects of pain. Neuropharmacology 86: 38–48. Missig G, Mei L, Vizzard MA et al. (2017). Parabrachial pituitary adenylate cyclase-activating polypeptide activation of amygdala endosomal extracellular signal-regulated kinase signaling regulates the emotional component of pain. Biol Psychiatry 81: 671–682. Miyata A, Jiang L, Dahl RR et al. (1990). Isolation of a neuropeptide corresponding to the N-terminal 27 residues of the pituitary adenylate cyclase activating polypeptide with 38 residues (PACAP38). Biochem Biophys Res Commun 170: 643–648. Moaddab M, Dabrowska J (2017). Oxytocin receptor neurotransmission in the dorsolateral bed nucleus of the stria terminalis facilitates the acquisition of cued fear in the fear-potentiated startle paradigm in rats. Neuropharmacology 121: 130–139. Moga MM, Saper CB, Gray TS (1989). Bed nucleus of the stria terminalis: cytoarchitecture, immunohistochemistry, and

projection to the parabrachial nucleus in the rat. J Comp Neurol 283: 315–332. Normandeau CP, Ventura-Silva AP, Hawken ER et al. (2018). A key role for neurotensin in chronic-stress induced anxiety-like behavior in rats. Neuropsychopharmacology 43: 285–293. Ostrowski NL (1998). Oxytocin receptor mRNA expression in rat brain: implications for behavioral integration and reproductive success. Psychoneuroendocrinology 23: 989–1004. Otto C, Martin M, Wolfer DP et al. (2001). Altered emotional behavior in PACAP-type-I-receptor-deficient mice. Brain Res Mol Brain Res 92: 78–84. Palkovitz M, Tapia-Arancibia L, Kordon C et al. (1982). Somatostatin connections between the hypothalamus and the limbic system of the rat brain. Brain Res 250: 223–228. Palkovitz M, Somogyvari-Vigh A, Arimura A (1995). Concentrations of pituitary adenylate cyclase activating polypeptide (PACAP) in human brain nuclei. Brain Res 699: 116–120. Petit JM, Luppi PH, Peyron C et al. (1995). VIP-like immunoreactivity projections from the dorsal raphe and caudal linear raphe nuclei to the bed nucleus of the stria terminalis demonstrated by a double immunohistochemical method in the rat. Neurosci Lett 193: 77–80. Pomrenze MB, Millan EZ, Hopf FW et al. (2015). A transgenic rat for investigating the anatomy and function of corticotrophin releasing factor circuits. Front Neurosci 9: 487. Poulin J-F, Arbour D, Laforest S et al. (2009). Neuroanatomical characterization of endogenous opioids in the bed nucleus of the stria terminalis. Prog Neuropsychopharmacol Biol Psychiatry 33: 1356–1365. Radley JJ, Sawchenko PE (2011). A common substrate for prefrontal and hippocampal inhibition of the neuroendocrine stress response. J Neurosci 31: 9683–9695. Ragnauth A, Schuller A, Morgan M et al. (2001). Female preproenkephalin-knockout mice display altered emotional responses. Proc Natl Acad Sci U S A 98: 1958–1963. Rigney N, Whylings J, Mieda M et al. (2019). Sexually dimorphic vasopressin cells modulate social investigation and communication in sex-specific ways. eNeuro 6: ENEURO. 0415-0418.2019. Roman CW, Lezak KR, Hartsock MJ et al. (2014). PAC1 receptor antagonism in the bed nucleus of the stria terminalis (BNST) attenuates the endocrine and behavioral consequences of chronic stress. Psychoneuroendocrinology 47: 151–165. Said SI, Mutt V (1972). Isolation from porcine-intestinal wall of a vasocative octacosapeptide related to secretin and to glucagon. Eur J Biochem 28: 199–204. Said SI, Mutt V (1974). Structure of the porcine vasoactive intestinal octacosapeptide. The amino acid sequence Use of kallikrein in its determination. Eur J Biochem 42: 581–589. Sakanaka M, Shibasaki T, Lederis K (1986). Distribution and efferent projections of corticotropin-releasing factor-like

CHEMOARCHITECTURE OF THE BED NUCLEUS OF THE STRIA TERMINALIS immunoreactivity in the rat amygdaloid complex. Brain Res 382: 213–238. Sakanaka M, Shibasaki T, LederisK (1987).Corticotropin releasing factor-like immunoreactivity in the rat brain as revealed by a modified cobalt-glucose oxidase-diaminobenzidine method. J Comp Neurol 260: 256–298. Sakanaka M, Magari S, Shibasaki T et al. (1989). Colocalization of corticotropin-releasing factor- and enkephalin-like immunoreactivities in nerve cells of the rat hypothalamus and adjacent areas. Brain Res 487: 357–362. Salvatore M, Wiersielis KR, Luz S et al. (2018). Sex differences in circuits activated by corticotropin releasing factor in rats. Horm Behav 97: 145–153. Segovia S, Guillamo´n A (1993). Sexual dimorphism in the vomeronasal pathway and sex differences in reproductive behaviors. Brain Res Brain Res Rev 18: 51–74. Shackman AJ, Fox AS (2016). Contribution of the central extended amygdala to fear and anxiety. J Neurosci 36: 8050–8063. Shepard JD, Schulkin J, Myers DA (2006). Chronically elevated corticosterone in the amygdala increases corticotropin releasing factor mRNA in the dorsolateral bed nucleus of stria terminalis following duress. Behav Brain Res 174: 193–196. Sherwood NM, Krueckl SL, McRory JE (2000). The origin and function of the pituitary adenylate cyclase-activating polypeptide (PACAP)/glucagon superfamily. Endocr Rev 21: 619–670. Shimada S, Inagaki S, Kubota Y et al. (1989). Coexistence of peptides (corticotropin releasing factor/neurotensin and substance P/somatostatin) in the bed nucleus of the stria terminalis and central amygdaloid nucleus of the rat. Neuroscience 30: 377–383. Simerly RB (2002). Wired for reproduction: organization and development of sexually dimorphic circuits in the mammalian forebrain. Annu Rev Neurosci 25: 507–536. Sink KS, Walker DL, Yang Y et al. (2011). Calcitonin generelated peptide in the bed nucleus of the stria terminalis produces an anxiety-like pattern of behavior and increases neural activation of anxiety-related structures. J Neurosci 31: 1802–1810. Sink KS, Chung A, Ressler KJ et al. (2013a). Anxiogenic effects of CGRP within the BNST may be mediated by CRF acting at BNST CRFR1 receptors. Behav Brain Res 243: 286–293. Sink KS, Davis M, Walker DL (2013b). CGRP antagonist infused into the bed nucleus of the stria terminalis impairs the acquisition and expression of context but not discretely cued fear. Learn Mem 20: 730–739. Sofroniew MV (1985). Vasopressin- and neurophysinimmunoreactive neurons in the septal region, medial amygdala and locus coeruleus in colchicine-treated rats. Neuroscience 15: 347–358. Spiess J, Rivier J, Rivier C et al. (1981). Primary structure of corticotropin-releasing factor from ovine hypothalamus. Proc Natl Acad Sci U S A 78: 6517–6521.

401

Stefanova N, Ovtscharoff W (2000). Sexual dimorphism of the bed nucleus of the stria terminalis and the amygdala. Adv Anat Embryol Cell Biol 158: 1–78. 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. Sterrenburg L, Gaszner B, Boerrigter J et al. (2012). Sexdependent and differential responses to acute restraint stress of corticotropin-releasing factor-producing neurons in the rat paraventricular nucleus, central amygdala and bed nucleus of the stria terminalis. J Neurosci Res 90: 179–192. Stroth N, Eiden LE (2010). Stress hormone synthesis in mouse hypothalamus and adrenal gland triggered by restraint is dependent on pituitary adenylate cyclaseactivating polypeptide signaling. Neuroscience 165: 1025–1030. Stroth N, Liu Y, Aguilera G et al. (2011). Pituitary adenylate cyclase-activating polypeptide controls stimulus-transcription coupling in the hypothalamic-pituitary-adrenal axis to mediate sustained hormone secretion during stress. J Neuroendocrinol 23: 944–955. Swaab DF (2003). Bed nucleus of the stria terminalis (BST) and the septum. The human hypothalamus: basic and clinical aspects. In: Handbook of clinical neurology, vol. 79. 149–162. Tasan RO, Nguyen NK, Weger S et al. (2010). The central and basolateral amygdala are critical sites of neuropeptide Y/Y2 receptor-mediated regulation of anxiety and depression. J Neurosci 30: 6282–6290. Tschenett A, Singewald N, Carli M et al. (2003). Reduced anxiety and improved stress coping ability in mice lacking NPY-Y2 receptors. Eur J Neurosci 18: 143–148. Turner JM, Will RG, Harvey EA et al. (2019). Copulation induces expression of the immediate early gene arc in mating-relevant brain regions of the male rat. Behav Brain Res 372: 112006. Uchida K, Otsuka H, Morishita M et al. (2019). Female-biased sexual dimorphism of corticotropin-releasing factor neurons in the bed nucleus of the stria terminalis. Biol Sex Differ 10: 6. Uhl GR, Synder SH (1979). Neurotensin: a neuronal pathway projecting from the amygdala through the stria terminalis. Brain Res 161: 522–526. Vale W, Spiess J, Rivier C et al. (1981). Characterization of a 41-residue ovine hypothalamic peptide that stimulates secretion of corticotropin and beta-endorphin. Science 213: 1394–1397. Vaudry D, Falluel-Morel A, Bourgault S et al. (2009). Pituitary adenylate cyclase-activating polypeptide and its receptors: 20 years after the discovery. Pharmacol Rev 61: 283–357. Venihaki M, Skihara S, Subramanian S et al. (2004). Urocortin III, a brain neuropeptide of corticotropin-releasing hormone family: modulation by stress and attenuation of some anxiety-like behaviors. J Neuroendocrinol 16: 411–422.

402

S.E. HAMMACK ET AL.

Waddell J, Morris RW, Bouton ME (2006). Effects of bed nucleus of the stria terminalis lesions on conditioned anxiety: aversive conditioning with long-duration conditional stimuli and reinstatement of extinguished fear. Behav Neurosci 120: 324–336. Wahlestedt C, Pich EM, Koob GF et al. (1993). Modulation of anxiety and neuropeptide Y-Y1 receptors by antisense oligodeoxynucleotides. Science 259: 528–531. Walker DL, Miles LA, Davis M (2009). Selective participation of the bed nucleus of the stria terminalis and CRF in sustained anxiety-like versus phasic fear-like responses. Prog Neuropsychopharmacol Biol Psychiatry 33: 1291–1308. Walter A, Mai JK, Lanta L et al. (1991). Differential distribution of immunohistochemical markers in the bed nucleus of the stria terminalis in the human brain. J Chem Neuroanat 4: 281–298.

Wilson MA, Burghardt PR, Lugo JNJ et al. (2003). Effect of amygdalar opioids on the anxiolytic properties of ethanol. Ann N Y Acad Sci 985: 472–475. Wittmann G, Fuzesi T, Liposits Z et al. (2009). Distribution and axonal projections of neurons coexpressing thyrotropinreleasing hormone and urocortin 3 in the rat brain. J Comp Neurol 517: 825–840. Ye J, Veinante P (2019). Cell-type specific parallel circuits in the bed nucleus of the stria terminalis and the central nucleus of the amygdala of the mouse. Brain Struct Funct 224: 1067–1095. Yoshimura R, Kiyama H, Kimura T et al. (1993). Localization of oxytocin receptor messenger ribonucleic acid in the rat brain. Endocrinology 133: 1239–1246. 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.

Handbook of Clinical Neurology, Vol. 179 (3rd series) The Human Hypothalamus: Anterior Region D.F. Swaab, F. Kreier, P.J. Lucassen, A. Salehi, and R.M. Buijs, Editors https://doi.org/10.1016/B978-0-12-819975-6.00026-1 Copyright © 2021 Elsevier B.V. All rights reserved

Chapter 27

Functional anatomy of the bed nucleus of the stria terminalis–hypothalamus neural circuitry: Implications for valence surveillance, addiction, feeding, and social behaviors ISABELLA MAITA1, ALLYSON BAZER1, JENNIFER URBANO BLACKFORD2,3, AND BENJAMIN ADAM SAMUELS1* 1

Department of Psychology, Rutgers University, Piscataway, NJ, United States

2

Department of Psychiatry and Behavioral Sciences, Vanderbilt University Medical Center, Nashville, TN, United States

3

Research Health Scientist, Tennessee Valley HealthCare System, US Department of Veterans Affairs, Nashville, TN, United States

Abstract The bed nucleus of the stria terminalis (BNST) is a medial basal forebrain structure that modulates the hypothalamo–pituitary–adrenal (HPA) axis. The heterogeneous subnuclei of the BNST integrate inputs from mood and reward-related areas and send direct inhibitory projections to the hypothalamus. The connections between the BNST and hypothalamus are conserved across species, promote activation of the HPA axis, and can increase avoidance of aversive environments, which is historically associated with anxiety behaviors. However, BNST–hypothalamus circuitry is also implicated in motivated behaviors, drug seeking, feeding, and sexual behavior. These complex and diverse roles, as well its sexual dimorphism, indicate that the BNST–hypothalamus circuitry is an essential component of the neural circuitry that may underlie various psychiatric diseases, ranging from anorexia to anxiety to addiction. The following review is a cross-species exploration of BNST–hypothalamus circuitry. First, we describe the BNST subnuclei, microcircuitry and complex reciprocal connections with the hypothalamus. We will then discuss the behavioral functions of BNST–hypothalamus circuitry, including valence surveillance, addiction, feeding, and social behavior. Finally, we will address sex differences in morphology and function of the BNST and hypothalamus.

INTRODUCTION The bed nucleus of the stria terminalis (BNST) is a highly heterogeneous structure that is implicated in many homeostatic and behavioral functions. Sometimes referred to as the extended amygdala, the BNST is located just ventral and anterior to the amygdala proper in rodents. In humans, the BNST is a medial basal forebrain structure approximately the size of a sunflower seed at 190 mm3 (Avery et al., 2016). The BNST is bordered laterally by the internal capsule, supracapsular pallidal tissue, and the basal forebrain region and bordered ventromedially by the

preoptic area. It lies posterior to the nucleus accumbens, and anterior to the thalamus. Like the BNST of rodents, the human BNST is inferior to the lateral ventricles and mainly superior to the anterior commissure, though it extends below the anterior commissure in lateral regions. The structural and functional connectivity of the BNST is thought to be similar in humans, nonhuman primates, and rodents (Avery et al., 2016). In rodents, amygdalar projections to the BNST originate in both the central amygdala (CeA) and the medial amygdala (MeA). The BNST is also directly connected

*Correspondence to: Benjamin Adam Samuels︎ BA, PhD, Assistant Professor, Psychology, Rutgers University, 152 Frelinghuysen Rd, Room 215, Piscataway, NJ 08854, United States. Tel: +1-848-445-8933, E-mail: [email protected]

404

I. MAITA ET AL.

to several other limbic areas, including the hypothalamus, hippocampus, lateral septum, ventral tegmental area, and nucleus accumbens. Sensory inputs, like those from the olfactory bulbs, also project directly to the BNST. BNST nuclei are mostly populated by GABAergic neurons that disinhibit the BNST itself and its hypothalamic connections (Gungor and Pare, 2016), including those to the lateral hypothalamus (LH), paraventricular nucleus (PVN), and preoptic area. Human imaging studies have found anatomic and functional connectivity that is similar to rodents and nonhuman primates. The human BNST has dense connections to the central amygdala (Oler et al., 2012; Avery et al., 2014; Torrisi et al., 2015) and other limbic areas, including the hippocampus, and subcallosal cortex. The structural connections between the human BNST and the thalamus, the insula, and basal ganglia areas, including the accumbens, caudate, putamen, and pallidum, are similar to those found in rodents (Avery et al., 2014; Flook et al., 2020). The evolutionarily conserved connections between the BNST and the hypothalamus, limbic area, basal ganglia, and forebrain areas permit the BNST to play an integratory and regulatory role in diverse processes including osmotic regulation, appetite, sleep, arousal, sustained fear states, sexual behavior, addiction, and the stress response.

The BNST itself is a complex heterogeneous structure consisting of several subnuclei, as well as a few lesser defined regions. Characterization of the BNST subnuclei remains a work in progress, and there are a variety of designations and terms used to describe the divisions of the BNST. In humans, lateral, central, and medial subdivisions can be identified by chemoarchitectural differences (Walter et al., 1991) that correlate with rodent subnuclei. In rodents, 12–18 subnuclei were mainly defined by anatomical experiments with rats. However, the anatomy in mice is somewhat more dispersed and heterogeneous. Therefore, most work on the mouse BNST mainly uses conservative language to appropriately describe the discordant designations of BNST regions and subnuclei. Studies often target the “anterodorsolateral” subgroups (Giardino et al., 2018) or the “dorsal region that includes the oval nucleus” (Russell et al., 2020) rather than specific subnuclei. Here we will use subnuclei terminology as defined by the seminal tracing experiments of Dong et al. (2001), Dong and Swanson (2006), but acknowledge that these demarcations are insufficient to fully encompass BNST heterogeneity. Cytoarchitectural and functional differences segregate the anterior BNST and posterior BNST in rodents. The anterior BNST (Fig. 27.1) is responsible for

Fig. 27.1. Coronal diagram of the mouse anterior BNST and its subnuclei, shown in green, and surrounding regions, based on the Allen Brain Atlas. Anterior subnuclei of the BNST consist of (A) the oval (ov), rhomboid (rh), anterolateral (al), anteromedial (am), dorsomedial (dm), magnocellular (mg), ventral (v) (B) juxtacapsular (jx), and fusiform (fu) nuclei. AC, anterior commissure; BAC, bed nucleus of the anterior commissure; CP, caudoputamen; FS, fundus of striatum; fx, fornix; HYP, hypothalamus; I, internal capsule; LS, lateral septum; LSV, lateral septal nucleus; MS, medial septal nucleus; SE, strial extension; SI, substantia innominata; ST, stria terminalis; TRS, triangular nucleus of septum; V3, third ventricle.

FUNCTIONAL ANATOMY OF BNST–HYPOTHALAMUS CIRCUITRY Hypothalamo–Pituitary–Adrenal (HPA) axis activation and includes the oval, anteromedial, dorsomedial, magnocellular, anterolateral, anteroventral/ventral, rhomboid, juxtacapsular, and fusiform nuclei (Allen Mouse Brain Atlas, 2004). The posterior BNST (Fig. 27.2) is implicated in regulation of reproductive and social defensive behaviors, along with indirect inhibition of the HPA axis (Choi et al., 2007, 2008b), and consists of the principal, intrafascicular, and transverse nuclei (Allen Mouse Brain Atlas, 2004). The human BNST is less divided and

405

includes the lateral, medial, central, and ventral divisions (Fig. 27.3) (Walter et al., 1991; Swaab, 2003). Complexity in the BNST is a product of both diversity in the cell types within BNST nuclei and diversity in nuclei connectivity. Tracing studies in rodents have illuminated the heterogeneity in the anterior and posterior regions, and the microcircuitry of the BNST; however, less research has been conducted in humans because neuroimaging methods lack sufficient resolution to study BNST subnuclei.

Fig. 27.2. Coronal diagram of the posterior bed nucleus of the stria terminalis and surrounding regions, based on the Allen Brain Atlas. The posterior regions are depicted in shades of green: the principal nucleus (prBNST), interfascicular nucleus (ifBNST), transverse nucleus (trBNST), and strial extension (se). The anterior regions are depicted in blue: anterolateral nucleus (al) and ventral nucleus (v). HYP, hypothalamus; TH, thalamus; V3, third ventricle; SI, substantia innominata.

406

I. MAITA ET AL.

Fig. 27.3. Coronal diagram of the human BNST and its subdivisions, shown in shades of green, and surrounding areas. The human BNST can be divided into the central (c), lateral (l), and medial (m) divisions. AC, anterior commissure; B, blood vessel; CN, caudate nucleus; F, fornix; GP, globus pallidus; HYP, hypothalamus; IC, internal capsule; LV, lateral ventricle; SA, septal area; ST, stria terminalis; SI, substantia innominata.

ANTERIOR BNST CIRCUITRY The most thoroughly studied BNST subnucleus is the oval nucleus (ovBNST) because of its important role in the stress response and more homogeneous cell population relative to other subnuclei. The ovBNST is marked by a majority-population of corticotropin releasing factor (CRF, or corticotropin releasing hormone, CRH)expressing GABAergic neurons. CRF + neurons are activated by viscerosensory inputs from the insula and brainstem, as well as periaqueductal gray (PAG) dopaminergic and glutamatergic neurons. CRF inputs also come from both the CeA and PVN. The ovBNST is a major output subnucleus, and CRF neurons in the ovBNST project to and inhibit areas involved in avoidance, reward behaviors, and the stress response, including the hypothalamus, ventral tegmental area, and nucleus accumbens (Taha and Fields, 2005; Gungor and Pare, 2016). GABAergic ovBNST neurons are mainly CRF+ (Forray and Gysling, 2004), but some neurons also express either protein kinase C delta (PKCd) or somatostatin (SOM). PKCd + neurons receive most of the projections into the ovBNST, while SOM+ neurons function as

outputs to the lateral parabrachial nucleus and periaqueductal gray (Ye and Veinante, 2019). These neurons project reciprocally to GABAergic CeA neurons and to the lateral hypothalamus and act as inhibitory interneurons for other ovBNST neurons expressing CRF receptor 1 (CRF1R). OvBNST projections seem to act in a positive feedback system, in which CeA neurons disinhibit ovBNST projections to the hypothalamus, which ultimately increases glucocorticoid release (Dong et al., 2001). The presence of CRF amplifies GABA-A receptor conducted inhibitory postsynaptic currents (Gungor et al., 2018). For the most part, like the ovBNST, other BNST subnuclei are populated by GABAergic neurons that disinhibit one another and their hypothalamic connections. However, the anteroventral BNST (avBNST) has a minority population of glutamatergic neurons projecting to the paraventricular nucleus and ventral tegmental area (Gungor and Pare, 2016). This relatively small number of glutamatergic neurons have a disproportionately large influence on the avBNST’s role in mood regulation. Along with noradrenergic inputs, the avBNST is controlled by GABAergic and glutamatergic inputs. Central amygdalar

FUNCTIONAL ANATOMY OF BNST–HYPOTHALAMUS CIRCUITRY GABAergic inputs into avBNST induce inhibitory postsynaptic potentials that are amplified by CRF. Glutamatergic inputs arrive at the avBNST from the mPFC and subiculum (Gungor and Pare, 2016). The avBNST plays a major role in HPA axis activation through projections to the amygdala, hypothalamus, and brainstem areas that regulate cardiovascular stress response (Choi et al., 2008a). The anteromedial BNST (amBNST) receives most of the inputs into the BNST, including those from the basomedial amygdala (BMA). Glutamatergic inputs from the hippocampus and BMA, and GABAergic inputs from the MeA, are implicated in fear conditioning (Haufler et al., 2013). Most amBNST projections are within the BNST, synapsing on the anterolateral BNST (alBNST) and ovBNST. The dorsomedial BNST (dmBNST) receives inputs from several amygdalar areas, including the BMA, MeA, and CeA. Importantly, the dmBNST is the main connection between the BNST and PVN, with a massive GABAergic output (Dong and Swanson, 2006). The magnocellular BNST is spatially and functionally similar to the dmBNST, with massive outputs to the PVN. Both the magnocellular BNST and dmBNST act as interneurons, receiving GABAergic input from the principal BNST. The anterolateral BNST receives a variety of inputs, including glutamatergic projections from the ventral subiculum, serotonergic inputs from the dorsal raphe, and GABAergic inputs from the CeL, vagus nerve, and BL (Dong et al., 2001; Gungor and Pare, 2016). AlBNST neurons project to a variety of regions, including regions within the somatomotor system, the central autonomic system, the thalamocortical feedback system, and neuroendocrine system, including the hypothalamus. Some neurons express CRF1R receptors and receive CRF inputs from other BNST subnuclei including the ovBNST (Dong and Swanson, 2004a,b). Opposing circuitry in the alBNST helps regulate the stress response (Gungor and Pare, 2016). Several less well-defined areas are also present in the anterior BNST. Some researchers group the rhomboid BNST, juxtacapsular BNST, and fusiform BNST subnuclei with larger, more understood nuclei. For example, the rhomboid BNST is located within the anterolateral BNST but varies from other subnuclei in receptor expression and circuitry. The rhomboid BNST and juxtacapsular BNST neurons are GABAergic and act as interneurons within the BNST, mainly projecting to ovBNST and to CeA neurons containing CRF-1 receptors. The fusiform BNST projects to other subnuclei and several extraBNST areas and is cellularly similar to the ovBNST, with high expression of CRF mRNA. CRF is released from the fusiform BNST into the CeA, PVN, nucleus accumbens, and PAG. The inputs that activate this release are different from the ovBNST, with mainly noradrenergic fibers from the brain stem and LC, which are activated in response to immediate stressors (Daniel and Rainnie, 2015).

407

POSTERIOR BNST CIRCUITRY The principal nucleus (prBNST) is the largest region of the posterior BNST. It receives input from the accessory olfactory system via the posterodorsal medial amygdala and receives GABAergic input from the lateral septum and central amygdala (Gu et al., 2003; Dong and Swanson, 2004a,b; Myers et al., 2014). The prBNST is primarily GABAergic (Bota et al., 2012) and projects to the lateral septum, medial amygdala, posterior amygdala, medial lateral hypothalamus, medial preoptic area, periventricular hypothalamus, ventromedial hypothalamus, ventral premamillary nucleus of the hypothalamus, periaqueductal gray, and the pontine micronutrition center (Barrington’s nucleus) (Gu et al., 2003). The prBNST also projects to the ventral, magnocellular, dorsomedial, anterodorsal, and anteroventral nuclei of the BNST (Dong and Swanson, 2004a,b; Myers et al., 2014). It has a large population of GABAergic CRFR2-expressing neurons that project to the medial amygdala, paraventricular hypothalamus, periaqueductal gray, and locus coeruleus (Bota et al., 2012; Henckens et al., 2017). Approximately onefifth of prBNST neurons express enkephalin and are glutamatergic, although the importance of this cell population is unknown (Poulin et al., 2009). The prBNST further has a high expression of androgen receptors and distinct alpha and beta estrogen receptors and may be the most sexually dimorphic nucleus (Gu et al., 2003). The intrafascicular subnucleus (ifBNST) receives input from the accessory olfactory system via the posteromedial medial amygdala. GABAergic inputs also come from the lateral septum, central amygdala, and medial amygdala. The ifBNST has high expression of kainate receptors and is also strongly regulated by glutamatergic input from the CA1 of the hippocampus, ventral subiculum, paraventricular thalamus, and basomedial amygdala (Bota et al., 2012; Myers et al., 2014). The ifBNST sends projections to the lateral septum, substantia innominata, ventral tegmental area, medial amygdala, basomedial amygdala, posterior amygdala, medial preoptic nucleus of the hypothalamus, anterior hypothalamus, lateral medial hypothalamus, ventromedial nucleus (VMH), ventral premamillary nucleus of the hypothalamus, nucleus reuniens of the thalamus, and periaqueductal gray. It also projects to the ventral, magnocellular, anteroventral, and anterodorsal nuclei of the BNST (Dong and Swanson, 2004a,b). Similar to the prBNST, the ifBNST has high expression of enkephalin, but the majority of these cells are GABAergic (Poulin et al., 2009). The ifBNST also has high expression of androgen receptors, but not of estrogen receptors (Bota et al., 2012), and there is less data available about possible sexual dimorphism of this region compared to the prBNST. Like the ifBNST, the transverse subnucleus (trBNST) receives input from the accessory olfactory system via the

408

I. MAITA ET AL.

posteromedial cortical nucleus and the medial amygdala. It has similar GABAergic and glutamatergic inputs and has high expression of kainite receptors (Bota et al., 2012; Myers et al., 2014). The trBNST sends primarily GABAergic projections to the lateral septum, nucleus accumbens, ventral tegmental area, substantia innominata, central amygdala, medial amygdala, basomedial amygdala, lateral medial preoptic nucleus of the hypothalamus, anterior hypothalamus, dorsomedial lateral hypothalamus, dorsal premamillary nucleus, ventral posterior hypothalamus, thalamus, and pontine central gray. It also projects within the BNST to the anterolateral, anteroventral, anterodorsal, ventral, magnocellular, rhomboid, and fusiform nuclei (Dong and Swanson, 2004a,b).

BNST–HYPOTHALAMUS CIRCUITRY The BNST projects to several regions of the hypothalamus, including the paraventricular nucleus, ventromedial nucleus, lateral hypothalamus, supraoptic region, and medial preoptic area (MPOA) (Fig. 27.4). The densest BNST projections to the hypothalamus are via inhibitory projections from the BNST to the PVN and VMH, which regulate the stress-induced endocrine response (Kash et al., 2015). The greatest number of hypothalamic projections comes from the dmBNST, which inhibits the magnocellular and parvocellular neurons in the PVN and also

inhibits the supraoptic region and MPOA (Dong and Swanson, 2006). Other inputs to the PVN include GABAergic inputs from the fusiform BNST, rhomboid BNST, alBNST, amBNST, and CRF+ ovBNST neurons. Posterior subnuclei, including the principal and transverse BNST, also inhibit the parvocellular neurons of the PVN (Dong and Swanson, 2004a,b). The hypothalamus connects reciprocally to the BNST: the PVN projects to BNST neurons expressing oxytocin and CRF receptor 2 (CRF2R) (Dong and Swanson, 2006). In addition to the dmBNST, the MPOA also receives GABAergic inputs from the rhomboid BNST and principal BNST (Dong and Swanson, 2004a,b). In turn, the amBNST receives inputs from the MPOA (Dong and Swanson, 2006). The VMH is inhibited by the principal BNST. GABAergic amBNST neurons project to the VMH shell and contribute to an opposing circuit described later (Yamamoto et al., 2018). Projections to the LH are mainly GABAergic, originating from CRF+ ovBNST (Dong et al., 2001; Dong and Swanson, 2004a,b), fuBNST (Dong et al., 2001), and adBNST neurons (Kim et al., 2013; Giardino et al., 2018), though the avBNST sends some glutamatergic projections (Gungor et al., 2018). GABAergic adBNST neurons express either CRF or cholecystokinin (CCK), and these distinct neuronal subtypes have opposing effects on target hypocretin/ orexin-expressing LH neurons, as described later

Fig. 27.4. The BNST is highly interconnected with the hypothalamus. Red arrows represent afferent, GABAergic projections. Green arrows represent afferent, glutamatergic projections. AH, anterior hypothalamus; alBNST, anterolateral BNST; amBNST, anteromedial BNST; ARC, arcuate nucleus; dmBNST, dorsomedial BNST; DMH, dorsomedial nucleus of hypothalamus fuBNST, fusiform BNST; HYP, hypothalamus; ifBNST, intrafascicular BNST; jxBNST, juxtacapsular BNST; LH, lateral hypothalamus; mgBNST, magnocellular BNST; MN, mamillary nucleus; MPOA, medial preoptic area; ovBNST, oval BNST; PH, posterior hypothalamus; PMN, premamillary nucleus; POA, preoptic nucleus; prBNST, principal BNST; PVN, paraventricular nucleus; rhBNST, rhomboid BNST; SON, supraoptic nucleus; tBNST, transverse BNST; TMH, tuberomamillary nucleus; vBNST, ventral BNST; VMH, ventromedial nucleus; MN, mamillary nucleus.

FUNCTIONAL ANATOMY OF BNST–HYPOTHALAMUS CIRCUITRY (Giardino et al., 2018). Furthermore, the supraoptic nuclei receive inputs from the alBNST while fusiform nuclei neurons project to the dorsomedial nucleus of the hypothalamus. The BNSTcircuitry appears to be well conserved across species. Imaging in nonhuman primates repeats findings from rodent studies, with strong structural and functional connectivity between the BNST and the amygdala (Oler et al., 2017). BNST connectivity is evolutionarily conserved in humans, with functional connections including amygdalar, basal ganglia, and prefrontal areas, as well as the thalamus, anterior hippocampus, periaqueductal gray, and midcingulate cortex (Avery et al., 2014; Tillman et al., 2018). Imaging studies also demonstrate that the BNST has functional connectivity with the hypothalamus (Torrisi et al., 2015), but not at the level of specificity in rodent tracing studies. In rodents and humans, activation of the BNST–hypothalamus circuitry is thought to mainly disinhibit the hypothalamus, promoting activation of the HPA axis. BNST–hypothalamic projections are implicated in several behaviors, including valence surveillance, addiction, feeding, and social behaviors.

BNST–HYPOTHALAMUS CIRCUITRY IN VALENCE SURVEILLANCE The majority of BNST research has focused on how it regulates the HPA axis to mediate sustained fear states, fear learning, and the stress response. HPA axis activation results in what is commonly known as the stress response: a fast rise in adrenaline production is followed by glucocorticoid release, increased heart rate, and increased blood pressure. Until the early 2000s, BNST–hypothalamus connections were thought to mainly promote negative affective responses mediated by glucocorticoid release (Dong et al., 2001). However, the BNST also regulates avoidance behaviors (Kim et al., 2013; Yamamoto et al., 2018) and visceral sensitivity (Song et al., 2020). While the amygdala regulates the short-term, acute stress response, the BNST is implicated in long-term, sustained threat responses in rodents (Walker et al., 2009; Avery et al., 2016; Lebow and Chen, 2016; Goode et al., 2019; Goode and Maren, 2019). A “valence surveillance role” has now emerged, which suggests that opposing circuitry in the BNST regulate both positive and negative affective responses (Kim et al., 2013; Jennings et al., 2013b; Gungor and Pare, 2016). Most behavioral studies focus on the BNST and avoidance, but the BNST also regulates approach, reward, and motivation. Rodent experiments demonstrate that opposing neural connections to the hypothalamus can activate and terminate stress responsivity. More specifically, the basomedial amygdala regulates the stress response via direct and indirect connections to the hypothalamus. Direct

409

glutamatergic neurons connect the BMA and VMH core, activating an endocrine response to stress. The BMA also projects indirectly to the VMH: glutamatergic BMA inputs excite amBNST interneurons that project to and inhibit the VMH shell, which in turn inhibits the VMH core. Excitation of BMA-VMH or BMA-amBNST results in amplification or inhibition, respectively, of a stress response (Yamamoto et al., 2018). The dual projections from the BMA allow a tight regulation of the HPA axis response, allowing either precise amplification of HPA axis activation or a return to homeostasis. Differential projections, including those to the hypothalamus, define the functions of several BNST subnuclei. The ovBNST drives avoidance via differential circuits and inhibition of the surrounding anterodorsal region of the BNST, which in turn decreases avoidance. Neural firing in the anterodorsal BNST is correlated with safety. Activation of anterodorsal BNST circuitry increases avoidance behaviors, decreases respiratory rate, and decreases motivated behaviors via connections to the LH, PBN, and VTA, respectively (Kim et al., 2013). Neuromodulators add another layer of complexity to BNST circuits. Parallel GABAergic circuits to the orexin-expressing lateral hypothalamus play opposing roles. Stimulation of projections to the LH from the CRF-expressing anterodorsolateral BNST neurons is aversive, while those from the CCKexpressing posterior principal nucleus are rewarding (Giardino et al., 2018). Neurons expressing CRF are located near those expressing enkephalin, which blocks the action of CRF, cytoarchitecture that may permit the BNST to control approach and avoidance response (Lebow and Chen, 2016). The BNST itself, especially the CRF circuitry, undergoes neuroplasticity in response to chronic or early life stress, including increased dendritic arborization, postsynaptic receptor expression (Hu et al., 2020a,b), CRF signaling (Forray and Gysling, 2004; Hu et al., 2020a,b), and firing rates (Conrad et al., 2011; Glangetas and Georges, 2016; Hu et al., 2020a,b). The BNST is a major source of stress-induced neural plasticity and changes to BNST sensitivity have repercussions in the hypothalamus. There is some evidence that stress alters sensitivity of the BNST–hypothalamus circuitry: adrenalectomized rats exhibit decreased total and CRF+ projection density from the BNST to the PVN (Mulders et al., 1997). However, the close, direct, and reciprocal relationship between the BNST and PVN suggests that changes to the BNST are sufficient to alter the hypothalamus. For example, glutamate-dependent high-frequency stimulation of the BNST invokes long-lasting suppression of evoked field potentials in the PVN (Tartar et al., 2006). The functional role of the BNST in sustained threat response is conserved in humans. Unpredictable longterm (40 s) (Alvare et al., 2011) and short-term

410

I. MAITA ET AL.

(1.75–5.75 s) threats (Choi et al., 2012; Klumpers et al., 2015) elicit increased human BNST activity, though middle (10–20 s) phases of cued anticipation elicit the greatest BNST response (McMenamin et al., 2014). BNST activity correlates with increasing risk of shock, skin conductance (anxiety), and temporal (Somerville et al., 2010) and spatial (Mobbs et al., 2010) proximity to physical threat, imagined threat (Coaster et al., 2011), and aversive images (Grupe et al., 2013). The valence surveillance role of the BNST implicates it in maladaptive mood, including anxiety, posttraumatic stress disorder (PTSD), and both negative and positive valence aspects of depression. Recent studies provide emerging evidence for heightened BNST responses to unpredictable threat in individuals with anxiety disorders (Straube et al., 2007; Brinkmann et al., 2017; Figel et al., 2019) or with higher levels of anxiety symptoms (Somerville et al., 2010, 2013; Brinkman et al., 2018). Furthermore, studies of BNST connectivity show correlations with anxiety symptoms (Andreescu et al., 2015; Brinkmann et al., 2018; Clauss et al., 2019) and altered in anxiety disorders (Torrisi et al., 2019). Recently, the BNST has emerged as a therapeutic target against mood disorders. Long-term deep brain stimulation in the BNST improves severe obsessive–compulsive disorder symptoms (Winter et al., 2018), treatment resistant anorexia nervosa, and major depressive disorder (Blomstedt et al., 2017).

THE BNST IN ADDICTION Circuitry between the BNST and reward and moodrelated areas contribute to drug-associated behaviors, including drug-seeking and the affective symptoms of withdrawal and intoxication (Vranjkovic et al., 2017). For an in-depth analysis of the role of BNST circuitry in addiction-related behaviors, see reviews by Koob and Volkow (2010), Stamatakis et al. (2014), and Vranjkovic et al. (2017). Importantly, rodent experiments indicate that the valence surveillance capabilities of the BNST play an integratory role for stress and addiction. Inactivation of the BNST reduces drug-seeking behavior (Buffalari and See, 2011). Several distinct neurotransmitters and neuromodulators are implicated in this role of the BNST in drug-related behavior (Vranjkovic et al., 2017). First, CRF is important for addiction and drug-related behavior (Sarnyai et al., 2001). CRF inputs to the BNST, like those from BNST interneurons, the PVN, and the CeA, promote drug-seeking behaviors in rodents, especially after exposure to stress (Silberman et al., 2013). CRF signaling in the BNST is necessary (Vranjkovic et al., 2014) and sufficient for stressinduced reinstatement of cocaine-seeking behavior (Erb and Stewart, 1999). The CRF upstream regulator pituitary adenylate cyclase activating polypeptide (PACAP), and its receptor protease activating compound

(PAC1), also drive drug reinstatement (Miles et al., 2019). Similarly, alcohol binging is both driven by and induces neuroplasticity in CRF signaling in the BNST (Francesconi et al., 2009; Silberman et al., 2013; Pleil et al., 2015). The effects of CRF on drug-motivated behavior can be attributed to BNST-VTA signaling. CRF1R binding in VTA-projecting BNST neurons increases BNST-VTA signaling, which may drive drug-seeking behavior. Alcohol withdrawal reduces CRF-induced excitation in the BNST, suggesting that withdrawal induces neuroplasticity in BNST-VTA circuitry (Silberman et al., 2013). The link between CRF and drug use has potent translational implications, suggesting that neuroplasticity induced by both stress and drugs promote a vicious cycle of mood disorder and substance abuse. Alcohol exposure also induces neuroplasticity in glutamatergic BNST signaling. Chronic and acute alcohol consumption in rodents increase glutamate sensitivity and LTP in the BNST (Weitlauf et al., 2004), an effect that is dependent on NMDA receptor subunit GluN2B (Kash et al., 2008, 2009; Wills et al., 2012). Kash et al. (2009) hypothesized that alcohol-induced plasticity in the BNST promotes alcoholism, when they observed that intermittent, but not chronic, exposure to ethanol vapor led to upregulation of NMDA receptors with GluN2B, driving alcohol consumption. Interestingly, in a study of the adult human brain, gene expression of GRIN2B, which encodes for GluN2B, was upregulated in the hippocampus of alcoholics and cocaine addicts (Enoch et al., 2014). The BNST regulates dopaminergic neurons in the VTA (Jalabert et al., 2009), which plays a major role in reward behaviors. However, the role of dopamine within the BNST is not fully understood. Exposure to addictive drugs including morphine, nicotine, cocaine, and ethanol increases dopamine in the BNST (Carboni et al., 2000). Alcohol self-administration in rats is dependent on D1 dopamine receptors (Eiler et al., 2003; Krawczyk et al., 2011). Dopaminergic inputs to the BNST may arise from the VTA and periaqueductal gray. Noradrenergic signaling drives affective symptoms of cocaine intoxication. Adrenergic receptor antagonism in the rodent vBNST decreases the anxiogenic effects of cocaine (Wenzel et al., 2014) and stress-induced drug reinstatement (Leri et al., 2002; Vranjkovic et al., 2014). In humans, pharmacologic intervention of noradrenaline signaling has therapeutic effects against cocaine addiction. Propranolol, a nonspecific beta-adrenergic receptor antagonist, reduces withdrawal and guanfacine, an alpha 2a agonist, decreases craving (Kampman et al., 2001; Fox et al., 2012; Fox and Sinha, 2014). The BNST, especially in the CRF and noradrenaline signaling pathways, is a potential therapeutic target for addiction.

FUNCTIONAL ANATOMY OF BNST–HYPOTHALAMUS CIRCUITRY

BNST–HYPOTHALAMUS CIRCUITRY IN FEEDING The ovBNST (and BNSTon the whole) is heavily involved in regulation of feeding. CRF+ ovBNST neurons send dense projections to the lateral hypothalamus, which functions as an interface between motivation and the control of feeding behavior (Dong and Swanson, 2004a,b; Jennings et al., 2013a; Nieh et al., 2015; O’Connor et al., 2015). CRF activates GABAergic LH neurons (Petrovich, 2018) and can induce anorexic feeding behavior (Ciccocioppo et al., 2003). PKCd+ ovBNST neurons also inhibit LH-projecting neurons from the ventrolateral region of the BNST, attenuating inflammation-induced anorexia (Wang et al., 2019). The intersection of CRF, PKCd, and feeding behavior suggests that stress-induced changes to feeding may arise from BNST–hypothalamus connectivity. BNST–hypothalamic connections also affect the digestive tract in ways other than feeding. Visceral hypersensitivity, like that seen in Irritable Bowel Syndrome, has been linked to stress (Holschneider et al., 2016; Fuentes and Christianson, 2018) and excitation of CRF+ neurons in the PVN. Colorectal distension (CRD)-induced visceral hypersensitivity and early life stress increased neural sensitivity and firing of CRF+ parvalbumin neurons in the PVN. Inhibition of PVN neurons via GABAergic avBNST inputs in turn reduces CRD-induced visceral hypersensitivity (Song et al., 2020). Though further research is needed, the BNST and hypothalamus are potential therapeutic targets for maladaptive feeding behaviors, from anorexia (Blomstedt et al., 2017) to obesity (Baldermann et al., 2019).

BNST–HYPOTHALAMUS CIRCUITRY IN SOCIAL BEHAVIOR The posterior BNST, particularly the prBNST, is part of a neural circuit implicated in reproductive behavior. Reproductive behavior is often described in two stages: appetitive behaviors (the approach and investigation of a social stimulus) and consummatory behaviors (the elicited response typically referring to mounting, intromission, and ejaculation). The prBNST is implicated in appetitive rather than consummatory behaviors. Exposure to opposite-sex odors results in activation of the rostral accessory bulb, while exposure to same sex odors results in activation of the posterior olfactory bulb. However, only the rostral olfactory bulb projects to the prBNST, suggesting that it is activated preferentially during the investigation of potential mates (Hashikawa et al., 2016). Both the medial preoptic area and the ventrolateral part of the ventromedial hypothalamus express high levels of ERa and progesterone receptors and integrate information from both the main and accessory olfactory systems to play important roles in mediating

411

copulatory behaviors. Both nuclei receive dense projections from the prBNST (Gu et al., 2003; Hashikawa et al., 2016). Interestingly, while prBNST provides dense input to the medial preoptic area, these nuclei are implicated in different aspects of reproductive behavior. Lesions to the medial preoptic area terminate consummatory behaviors but not appetitive behaviors, whereas lesions to the posterior BNST decrease chemoinvestigation of opposite-sex odors but only delay, or do not effect, copulatory behaviors (Powers et al., 1987; Been and Petrulis, 2010). Within the prBNST, there is a significant increase in cell activity after investigation of an opposite-sex odor (Kim et al., 2015). Specifically, inhibition of aromataseexpressing neurons in the prBNST of males eliminates preference for female odors, while activation promotes male–male mating. These neurons do not appear to control sex recognition or mating behaviors in females. While there are fewer studies of reproductive behavior in females, the posterior BNST is necessary for solicitation behaviors in response to male odor stimuli (Martinez and Petrulis, 2011; Bayless et al., 2019). While the medial amygdala plays a prominent role carrying sex-specific odor information, more research is necessary to determine the exact role of the posterior BNST in reproductive behaviors. It is possible that the BNST plays an integrative role in rodent chemoinvestigation (genital sniffing), by receiving olfactory and pheromonal signals, and then relaying that information to the periaqueductal gray, and eventually, motor neurons in the spinal cord (Been and Petrulis, 2011; Hashikawa et al., 2016). Reproductive and social defense behaviors activate similar but different neural pathways in the brain. In rodents, both behaviors depend upon chemosensory cues and gonadal steroids that activate both the main olfactory system and the accessory olfactory system. The posterior BNST receives dense projections from both of these systems via the medial and central amygdala and send projections to nuclei in the hypothalamus that regulate aggression and social defense behaviors (KollackWalker and Newman, 1995; Nelson and Trainor, 2007; Coria-Avila et al., 2014). Specifically, the anterior hypothalamic nucleus, the ventromedial nucleus, and the dorsal premamillary nucleus comprise a circuit implicated in innate defensive behaviors. In particular, the ventromedial hypothalamus expresses ERa receptors and integrates information from the main and accessory olfactory systems. The ifBNST sends dense projections to all three of the hypothalamic nuclei implicated in this aggression circuit (Canteras et al., 1997; Canteras, 2002; Dong and Swanson, 2004a,b; Hashikawa et al., 2016). The ventromedial hypothalamus is implicated in male aggression, and while some argue it is not essential for female aggression, inactivation of populations of estrogen receptor alpha-expressing cells in the

412

I. MAITA ET AL.

ventromedial hypothalamus reduces female aggression, while activation of these cells elicits attack (Hashikawa et al., 2017). The posterior BNST sends dense projections to many subnuclei in the anterior BNST, but the purpose of these neurons remains poorly understood. The posterior BNST may indirectly regulate the HPA axis via inhibitory microcircuitry. Posterior BNST lesions elevate plasma ACTH, CRH mRNA, and corticosterone, while anterior BNST lesions weaken the plasma corticosterone response and cause a decrease of CRH mRNA (Herman et al., 1994; Choi et al., 2007). This suggests that the posterior BNST inhibits the HPA axis, while the anterior BNST activates it. The posterior BNST, specifically the prBNST and trBNST, sends moderate to low projections to the paraventricular hypothalamus, with the densest projections sent to the anterior parvocellular part of the paraventricular hypothalamus (Gu et al., 2003; Dong and Swanson, 2004a,b). However, CRH and AVP are released from the medial parvocellular part of the paraventricular hypothalamus to trigger ACTH secretion from the anterior pituitary (Herman et al., 1994; Choi et al., 2007, 2008b). The dorsomedial and fusiform subnuclei of the anterior BNST, which receive input from the prBNST and trBNST, respectively, express CRF and send dense projections to the medial parvocellular part of the paraventricular hypothalamus (Dong et al., 2001; Dong and Swanson, 2004a,b, 2006). While it is possible that posterior BNST inhibition of the HPA axis is due to direct projections to the paraventricular hypothalamus, different BNST subregions differentially regulate the HPA axis and the posterior BNST sends dense projections to the anterior BNST, but only moderate to weak projections to the region of the paraventricular hypothalamus that mediates CRH release. Taken together, this suggests the possibility of an inhibitory microcircuit within the BNST that differentially mediates the HPA axis. While there is substantial research into the role of the posterior BNST in reproductive behavior, the specific projections from the posterior BNST to hypothalamic regions implicated in these behaviors, such as the medial preoptic area, ventromedial, and ventral premamillary nucleus, are not well studied. Lesions disconnecting the ventral posterior BNST from the medial preoptic area of the hypothalamus eliminates mating in sexually experienced male gerbils (Sayag et al., 1994). It is clear from tracing studies that the posterior BNST, particularly the prBNST, sends dense projections to hypothalamic regions involved in reproductive behavior (Gu et al., 2003; Dong and Swanson, 2004a,b; Hashikawa et al., 2016), but more research is necessary in order to determine the exact role these projections play in the mediation of these behaviors.

SEX DIFFERENCES IN THE BNST Both the BNST and hypothalamus differ in mammalian males and females, by functionality and morphology. The human hypothalamus (Madeira and Lieberman, 1995) and BNST (Hines et al., 1985, 1992; Segovia and Guillamón, 1993; Chung et al., 2002; Swaab, 2007; Forger, 2009) are larger in males. Specifically, the central nucleus of the BNST is larger in males and contains twice the number of neurons as is found in females. Most interestingly, transgender women (assigned male at birth) have a female-typical, smaller central nucleus with less somatostatin neurons, and transgender men (assigned female at birth) have male-typical BNSTs (Zhou et al., 1995; Kruijver et al., 2000; Swaab, 2007). These sex differences occur early in development and are at least in part due to differences in hormone exposure in early developmental stages (Chung et al., 2002). This notion is supported by rodent studies where sex differences in the BNST were suppressed via alteration of gonadal hormones during development, through orchiectomy or androgynization (Segovia and Guillamón, 1993). Increased BNST size might suggest that males are more prone to a stress response, but HPA axis activation is more intense and longer lasting in female rats after stress (Critchlow et al., 1963; Kant et al., 1983; Williams et al., 1985; Brett et al., 1986; Heinsbroek et al., 1988). In rodents, corticosteroid release during acute stress is greater in male rats relative to females. However, corticosteroids may have a greater impact in females due to a less sensitive negative feedback system. In females, CRF1R couples to a greater extent with the Gsa subunit of G proteins than in males. Therefore, CRF1R binding in females induces a greater neuronal response (Bangasser et al., 2010). Binding of CRF2R, thought to have opposite effects relative to CRF1R, is greater in males in some subregions of the BNST (Weathington et al., 2014). There are also sex differences in expression of a variety of receptors in BNST. Males have more muscarinic receptors in the hypothalamus (Avissar et al., 1981; Egozi et al., 1982). High-affinity binding sites on muscarinic receptors decrease in aging male rates at a greater number than female rats (Gurwitz et al., 1987). Gonadal steroid-dependent sex differences in the vasopressin and vasotocin pathways are highly conserved between species, especially those in the BNST (De Vries and Panzica, 2006). Gonadal hormone pathways themselves are sexually divergent within BNST and hypothalamus structures. Estrogen receptor ERa and aromatase are more densely expressed in the BNST of male rats (Wu et al., 2009; Tabatadze et al., 2014). However, ERa immunoreactivity is higher in the MPOA and BNST of females (Brock et al., 2015). Androgen receptors are

FUNCTIONAL ANATOMY OF BNST–HYPOTHALAMUS CIRCUITRY found in somatostatin PVN, VMH, arcuate nucleus, and posterior BNST neurons in males but are sparsely expressed in these regions in females (Herbison, 1995; Brock et al., 2015). Testosterone action is limited in the BNST, VMH, and preoptic area in females, not only due to differences in receptor density but due to decreased velocity of cytochrome P-450 aromatase conversion of testosterone to estrogen (Roselli et al., 1996). Morphologic sexual dimorphism of ERa and aromatase in BNST–hypothalamus circuitry have been well established (Handa and Weiser, 2014), but the behavioral implications of these differences on stress and addiction is a neglected area of research. The BNST–hypothalamus neural circuitry is sexually dimorphic, as well. Perhaps the most sexually dimorphic nucleus, the prBNST sends dense projections to areas such as the anteroventral periventricular nucleus of the hypothalamus, and the ventral premamillary nucleus is significantly denser in male rats than in females (Gu et al., 2003). The periventricular nucleus appears to play a critical role in controlling the estrous cycle in females, though sex differences in prBNST connectivity likely affects more than hormonal cycling. Conserved circuitry between the BNST and anterior regions of the insula, which may be implicated in anxiogenic effects of alcohol withdrawal, are also sexually dimorphic, wherein human male BNST–insula projections are denser in the left hemisphere, but females lack a hemisphere difference (Flook et al., 2020). Females have a stronger anatomic connectivity between the BNST and several areas implicated in mood disorder, reward, and addiction, including the nucleus accumbens, amygdala, hippocampus, pallidum, caudate, putamen, and subgenual cingulate cortex (Avery et al., 2014). Sexual dimorphism in the BNST may have functional ramifications, as well. Females are twice as likely to develop stress-induced mood disorders (Kessler et al., 1994), including posttraumatic stress disorder (Breslau, 2001). Female PTSD patients show a stronger correlation between the risk and severity of PTSD and two key single-nucleotide polypeptides (SNPs): the SNPs for CRF2R and protease activating compound receptor. Binding at PAC1, the specific G-protein coupled receptor for pituitary adenylate cyclase activating polypeptide, increases avoidance in rodents (Hammack et al., 2009). Chronic stress increases both PACAP and PAC1 expression in the anterolateral BNST (Hammack et al., 2009; Hu et al., 2020a). This sex-specific link between PTSD and CRF signaling may be associated with the makeup of the PAC1 gene, which hosts a single SNP in a putative estrogen response element that predicts PTSD diagnosis and symptoms in females (Ressler et al., 2011). Rodent approaches to studying psychiatric diseases, including depression, anxiety, PTSD, and addiction,

413

usually exhibit sexually dimorphic behavioral responses. However, the directionality of these differences is not consistent. Some paradigms, behavioral tests, and genetic models suggest that males are more susceptible to stress and mood disorders, while others suggest the opposite. Even similar stressors can result in behavioral sex differences (Kokras and Dalla, 2014). This variability indicates a need for validating stress paradigms in both males and female rodents (Yohn et al., 2019) and in postmortem tissue of human patients.

CONCLUDING COMMENTS The BNST is a hub for positive and negative valence information, and its tight regulation of areas like the hypothalamus and VTA drives behavioral responses to everchanging environmental conditions. The integratory role of the BNST makes it a key player in a wide variety of typical functions in mammals, including mood, feeding, and social behavior. It also makes it an investigatory target for maladaptation in these functions, including mood disorder, eating disorder, and addiction. Though rodent studies have promoted it as a prospective therapeutic target, the BNST has proven difficult to study in vivo in humans due to its minute size and indiscernibility from bordering structures. Further molecular analysis of human postmortem material, as well as recent developments in 3T and 7T MRI, will help improve the scope of human BNST research (Avery et al., 2014; Torrisi et al., 2015; Theiss et al., 2017). In recent years, the BNST has been used as a therapeutic target, including deep brain stimulation and manipulation of key BNST signaling pathways. These recent developments foster optimism for the future of the BNST as a potential therapeutic target, and as a window into the sexual dimorphism of mood disorder.

REFERENCES Allen Mouse Brain Atlas (2004).Allen institute for brain science. Available from: mouse.brain-map.org. Alvare RP, Chen G, Bodurka J et al. (2011). Phasic and sustained fear in humans elicits distinct patterns of brain activity. Neuroimage 55: 389–400. Andreescu C, Sheu LK, Tudorascu D et al. (2015). Emotion reactivity and regulation in late-life generalized anxiety disorder: functional connectivity at baseline and posttreatment. Am J Geriatr Psychiatry 23: 200–214. Avery SN, Clauss JA, Winder DG et al. (2014). BNST neurocircuitry in humans. Neuroimage 91: 311–323. Avery SN, Clauss JA, Blackford JU (2016). The human BNST: functional role in anxiety and addiction. Neuropsychopharmacology 41: 126–141. Avissar S, Egozi Y, Sokolovsky M (1981). Studies on muscarinic receptors in mouse and rat hypothalamus: a comparison of sex and cyclical differences. Neuroendocrinology 32: 295–302.

414

I. MAITA ET AL.

Baldermann JC, Hahn L, Dembek TA et al. (2019). Weight change after striatal/capsule deep brain stimulation relates to connectivity to the bed nucleus of the stria terminalis and hypothalamus. Brain Sci 9: 264. Bangasser DA, Curtis A, Reyes BA et al. (2010). Sex differences in corticotropin-releasing factor receptor signaling and trafficking: potential role in female vulnerability to stress-related psychopathology. Mol Psychiatry 15 (877): 896–904. Bayless DW, Yang T, Mason MM et al. (2019). Limbic neurons shape sex recognition and social behavior in sexually naive males. Cell 176: 1190–1205. Been LE, Petrulis A (2010). Lesions of the posterior bed nucleus of the stria terminalis eliminate opposite-sex odor preference and delay copulation in male Syrian hamsters: role of odor volatility and sexual experience. Eur J Neurosci 32: 483–493. Been LE, Petrulis A (2011). Chemosensory and hormone information are relayed directly between the medial amygdala, posterior bed nucleus of the stria terminalis, and medial preoptic area in male Syrian hamsters. Horm Behav 59: 536–548. Blomstedt P, Naesstrom M, Bodlund O (2017). Deep brain stimulation in the bed nucleus of the stria terminalis and medial forebrain bundle in a patient with major depressive disorder and anorexia nervosa. Clinical Case Reports 5: 679–684. Bota M, Sporns O, Swanson LW (2012). Neuroinformatics analysis of molecular expression patterns and neuron populations in gray matter regions: the rat BST as a rich exemplar. Brain Res 1450: 174–193. Breslau N (2001). The epidemiology of posttraumatic stress disorder: what is the extent of the problem? J Clin Psychiatry 62: 16–22. Brett LP, Levine R, Levine S (1986). Bidirectional responsiveness of the pituitary–adrenal system in old and young male and female rats. Neurobiol Aging 7: 153–159. Brinkmann L, Buff C, Feldker K et al. (2017). Distinct phasic and sustained brain responses and connectivity of amygdala and bed nucleus of the stria terminalis during threat anticipation in panic disorder. Psych Med 47: 2675–2688. Brinkmann L, Buff C, Feldker K et al. (2018). Inter-individual differences in trait anxiety shape the functional connectivity between the bed nucleus of the stria terminalis and the amygdala during brief threat processing. Neuroimage 166: 110–116. Brock O, De Mees C, Bakker J (2015). Hypothalamic expression of oestrogen a and androgen receptor is sex-, age-, and region-dependent in mice. J Neuroendocrinol 27: 264–276. Buffalari DM, See RE (2011). Inactivation of the bed nucleus of the stria terminalis in an animal model of relapse: effects on conditioned cue-induced reinstatement and its enhancement by yohimbine. Psychopharmacology (Berl) 213: 19–27. Canteras NS (2002). The medial hypothalamic defensive system: hodological organization and functional implications. Pharmacology Biochemistry and Behavior 71: 481–491. Canteras NS, Chiavegatto S, Do Valle LR et al. (1997). Severe reduction of rat defensive behavior to a predator by discrete hypothalamic chemical lesions. Brain Res Bull 44: 297–305.

Carboni E, Silvagni A, Rolando MTP et al. (2000). Stimulation of in vivo dopamine transmission in the bed nucleus of the stria terminalis by reinforcing drugs. J Neurosci 20: RC102. Choi DC, Furay AR, Evanson NK et al. (2007). Bed nucleus of the stria terminalis subregions differentially regulate hypothalamic-pituitary-adrenal axis activity: implications for the integration of limbic inputs. J Neurosci 27: 2025–2034. Choi DC, Evanson NK, Furay AR et al. (2008a). The anteroventral bed nucleus of the stria terminalis differentially regulates hypothalamic-pituitary-adrenocortical axis responses to acute and chronic stress. Endocrinology 149: 818–826. Choi DC, Furay AR, Evanson NK et al. (2008b). The role of the posterior medial bed nucleus of the stria terminalis in modulating hypothalamic–pituitary–adrenocortical axis responsiveness to acute and chronic stress. Psychoneuroendocrinology 33: 659–669. Choi JM, Padmala S, Pessoa L (2012). Impact of state anxiety on the interaction between threat monitoring and cognition. Neuroimage 59: 1912–1923. Chung WCJ, 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. Ciccocioppo R, Fedeli A, Economidou D et al. (2003). The bed nucleus is a neuroanatomical substrate for the anorectic effect of corticotropin-releasing factor and for its reversal by nociception/orphanin FQ. J Neurosci 23: 9445–9451. Clauss JA, Avery SN, Benningfield MM et al. (2019). Social anxiety is associated with BNST response to unpredictability. Depress Anxiety 36: 666–675. Coaster M, Rogers BP, Jones OD et al. (2011). Variables influencing the neural correlates of perceived risk of physical harm. Cogn Affect Behav Neurosci 11: 494–507. Conrad KL, Louderback KM, Gessner CP et al. (2011). Stressinduced alteration in anxiety-like behavior and adaptations in plasticity in the bed nucleus of the stria terminalis. Physiol Behav 104: 248–256. Coria-Avila GA, Manzo J, Garcia LI et al. (2014). Neurobiology of social attachments. Neuroscience & Biobehavioral Reviews 43: 173–182. Critchlow V, Liebelt RA, Bar-Sela M et al. (1963). Sex difference in resting pituitary–adrenal function in the rat. Am J Physiol 205: 807–815. Daniel SE, Rainnie DG (2015). Stress modulation of opposing circuits in the bed nucleus of the stria terminalis. Neuropsychopharmacology 41: 103–125. De Vries GJ, Panzica GC (2006). Sexual differentiation of central vasopressin and vasotocin systems in vertebrates: different mechanisms, similar endpoints. Neuroscience 138: 947–955. Dong HW, Swanson LW (2004a). Organization of axonal projections from the anterolateral areas of the bed nuclei of the stria terminalis. J Comp Neurol 468: 277–298. Dong HW, Swanson LW (2004b). Projections from bed nuclei of the stria terminalis, posterior division: implications for cerebral hemisphere regulation of defensive and reproductive behaviors. J Comp Neurol 471: 396–433. Dong HW, Swanson LW (2006). Projections from bed nuclei of the stria terminalis, dorsomedial nucleus: implications for cerebral hemisphere integration of neuroendocrine,

FUNCTIONAL ANATOMY OF BNST–HYPOTHALAMUS CIRCUITRY autonomic, and drinking responses. J Comp Neurol 494: 75–107. Dong HW, Petrovich GD, Watts AG et al. (2001). Basic organization of projections from the oval and fusiform nuclei of the bed nuclei of the stria terminalis in adult rat brain. J Comp Neurol 436: 430–455. Egozi Y, Avissar S, Sokolovsky M (1982). Muscarinic mechanisms and sex hormone secretion in rat adenohypophysis and preoptic area. Neuroendocrinology 35: 93–97. Eiler 2nd. WJ, Seyoum R, Foster K et al. (2003). D1 dopamine receptor regulates alcohol-motivated behaviors in the bed nucleus of the stria terminalis in alcohol-preferring (P) rats. Synapse 48: 45–56. Enoch MA, Rosser AA, Zhou Z et al. (2014). Expression of glutamatergic genes in healthy humans across 16 brain regions; altered expression in the hippocampus after chronic exposure to alcohol or cocaine. Genes Brain Behav 13: 758–768. Erb S, Stewart J (1999). A role for the bed nucleus of the stria terminalis, but not the amygdala, in the effects of corticotropinreleasing factor on stress-induced reinstatement of cocaine seeking. J Neurosci 19: RC35. Figel B, Brinkmann L, Buff C et al. (2019). Phasic amygdala and BNST activation during the anticipation of temporally unpredictable social observation in social anxiety disorder patients. NeuroImage Clin 22: 101735. Flook EA, Feola B, Avery SN et al. (2020). BNST-insula structural connectivity in humans. Neuroimage 210: 116555. Forger NG (2009). Control of cell number in the sexually dimorphic brain and spinal cord. J Neuroendocrinol 21: 393–399. Forray MI, Gysling K (2004). Role of noradrenergic projections to the bed nucleus of the stria terminalis in the regulation of the hypothalamic-pituitary-adrenal axis. Brain Res Brain Res Rev 47: 145–160. Fox H, Sinha R (2014). The role of guanfacine as a therapeutic agent to address stress-related pathophysiology in cocainedependent individuals. Adv Pharmacol 69: 218–265. Fox HC, Seo D, Tuit K et al. (2012). Guanfacine effects on stress, drug craving, and prefrontal activation in cocaine- dependent individuals: preliminary findings. J Psychopharmacol 26: 958–972. Francesconi W, Berton F, Koob GF et al. (2009). Intrinsic neuronal plasticitiy in the juxtacapsular nucleus of the bed nuclei of the stria terminalis (jcBNST). Prog Neuropsychopharmacol Biol Psychiatry 33: 1347–1355. Fuentes IM, Christianson JA (2018). The influence of early life experience on visceral pain. Front Syst Neurosci 12: 2. Giardino WJ, Eban-Rothschild A, Christoffel DJ et al. (2018). Parallel circuits from the bed nucleus of the stria terminalis to the lateral hypothalamus drive opposing emotional states. Nat Neurosci 21: 1084–1095. Glangetas C, Georges F (2016). Pharmacology of the bed nucleus of the stria terminalis. Curr Pharmacol Rep 2: 262–270. Goode TD, Maren S (2019). Role of the bed nucleus of the stria terminalis in aversive learning and memory. Cold Spring Harbor Lab Press 24: 480–491. Goode TD, Ressler RL, Acca GM et al. (2019). Bed nucleus of the stria terminalis regulates fear to unpredictable threat signals. eLife 8. https://doi.org/10.7554/eLife.46525.

415

Grupe DW, Oathes DJ, Nitschke JB (2013). Dissecting the anticipation of aversion reveals dissociable neural networks. Cereb Cortex 23: 1874–1883. Gu G, Cornea A, Simerly RB (2003). Sexual differentiation of projections from the principal nucleus of the bed nuclei of the stria terminalis. J Comp Neurol 460: 542–562. Gungor NZ, Pare D (2016). Functional heterogeneity in the bed nucleus of the stria terminalis. J Neurosci 36: 80388038-49. Gungor NZ, Yamamoto R, Pare D (2018). Glutamatergic and gabaergic ventral BNST neurons differ in their physiological properties and responsibilities. Neuropsychopharmocol 43: 2126–2133. Gurwitz D, Egozi Y, Henis YI et al. (1987). Agonist and antagonist binding to rat brain muscarinic receptors: influence of aging. Neurobiol Aging 8: 115–122. Hammack SE, Cheung J, Rhodes KM et al. (2009). Chronic stress increases pituitary adenylate cyclase-activating peptide (PACAP) and brain derived neurotrophic factor (BDNF) mRNA expression in the bed nucleus of the stria terminalis (BNST): roles for PACAP in anxiety-like behavior. Psychoneuroendocrinology 34: 833–843. Handa RJ, Weiser MJ (2014). Gonadal steroid hormones and the hypothalamo-pituitary-adrenal axis. Front Neuroendocrinol 35: 197–220. Hashikawa K, Hashikawa Y, Falkner A et al. (2016). The neural circuits of mating and fighting in male mice. Curr Opin Neurobiol 38: 27–37. Hashikawa K, Hashikawa Y, Tremblay R et al. (2017). Esr1 + cells in the ventromedial hypothalamus control female aggression. Nat Neurosci 20: 1580. Haufler D, Nagy FZ, Pare D (2013). Neuronal correlates of fear conditioning in the bed nucleus of the stria terminalis. Learn Mem 20: 633–641. Heinsbroek RP, Van Haaren F, Van de Poll NE (1988). Sex differences in passive avoidance behaviour of rats: sexdependent susceptibility to shock-induced behavioral depression. Physiol Behav 43: 201–206. Henckens MJAG, Printz Y, Shamgar U et al. (2017). CRF receptor type 2 neurons in the posterior bed nucleus of the stria terminalis critically contribute to stress recovery. Mol Psychiatry 22: 1691–1700. Herbison AE (1995). Sexually dimorphic expression of androgen receptor immunoreactivity by somatostatin neurons in rat hypothalamus paraventricular nucleus and bed nucleus of the stria terminalis. J Neuroendocrinol 7: 543–553. Herman JP, Cullinan WE, Watson SJ (1994). Involvement of the bed nucleus of the stria terminalis in tonic regulation of paraventricular hypothalamic CRH and AVP mRNA expression. J Neuroendocrinol 6: 433–442. Hines M, Davis FC, Coquelin A et al. (1985). Sexually dimorphic regions in the medial preoptic area and the bed nucleus of the stria terminalis of the Guinea pig brain: a description and an investigation of their relationship to gonadal steroids in adulthood. J Neurosci 5: 40–47. Hines M, Allen LS, Gorski RA (1992). Sex differences in subregions of the medial nucleus of the amygdala and the bed nucleus of the stria terminalis of the rat. Brain Res 579: 321–326.

416

I. MAITA ET AL.

Holschneider DP, Guo Y, Mayer EA et al. (2016). Early life stress elicits visceral hyperalgesia and functional reorganization of pain circuits in adult. Neurobiol Stress 3: 8–22. Hu P, Maita I, Kwok C et al. (2020a). Chronic stress induces anxiety-associated negative valence behavior by activating a Corticotropin-releasing hormone-related molecular signature in the oval bed nucleus of the Stria terminalis. J Neurosci 40: 2519–2537. Hu P, Maita I, Phan ML et al. (2020b). Early-life stress alters affective Behaviors in adult mice through persistent activation of CRH-BDNF Signaling in the oval bed nucleus of the Stria terminalis. Transl Psychiatry. https://doi.org/10.1038/ s41398-020-01070-3. Jalabert M, Aston-Jones G, Herzog E et al. (2009). Role of the bed nucleus of the stria terminalis in the control of ventral tegmental area dopamine neurons. Prog Neuropsychopharmacol Biol Psychiatry 33: 1336–1346. Jennings JH, Rizzi G, Stamatakis AM et al. (2013a). The inhibitory circuit architecture of the lateral hypothalamus orchestrates feeding. Science 341: 1517–1521. Jennings JH, Sparta DR, Stamatakis AM et al. (2013b). Distinct extended amygdala circuits for divergent motivational states. Nature 496: 224–228. Kampman KM, Volpicelli JR, Mulvaney F et al. (2001). Effectiveness of propranolol for cocaine dependence treatment may depend on cocaine withdrawal symptom severity. Drug Alcohol Depend 63: 69–78. Kant GJ, Lenox RH, Bunnell BN (1983). Comparison of stress response in male and female rats: pituitary cyclic AMP and plasma prolactin, growth hormone and corticosterone. Psychoneuroendocrinology 8: 421–428. Kash TL, Matthews RT, Winder DG (2008). Alcohol inhibits NR2B-containing NMDA receptors in the ventral bed nucleus of the stria terminalis. Neuropsychopharmacology 33: 1379–1390. Kash TL, Baucum 2nd AJ, Conrad KL et al. (2009). Alcohol exposure alters NMDAR function in the bed nucleus of the stria terminalis. Neuropsychopharmacology 34: 2420–2429. Kash TL, Pleil KE, Marcinkiewcz CA et al. (2015). Neuropeptide regulation of signaling and behavior in the BNST. Mol Cells 38: 1–13. Kessler RC, McGonagle KA, Nelson CB et al. (1994). Sex and depression in the National Comorbidity Survey II: cohort effects. J Affect Disord 30: 15–26. Kim S, Adhikari A, Lee SY et al. (2013). Diverging neural pathways assemble a behavioral state from separable features in anxiety. Nature 496: 219–223. Kim Y, Venkataraju KU, Pradhan K et al. (2015). Mapping social behavior-induced brain activation at cellular resolution in the mouse. Cell Rep 10 (2): 292–305. Klumpers F, Kroes MC, Heiland I et al. (2015). Dorsomedial prefrontal cortex mediates the impact of serotonin transporter linked polymorphic region genotype on anticipatory threat reactions. Biol Psychiatry 78: 582–589. Kokras N, Dalla C (2014). Sex differences in animal models of psychiatric disorders. Br J Pharmacol 171: 4595–4619. Kollack-Walker S, Newman SW (1995). Mating and agonistic behavior produce different patterns of Fos immunolabeling in the male Syrian hamster brain. Neuroscience 66: 721–736.

Koob GF, Volkow ND (2010). Neurocircuitry of addiction. Neuropsychopharmacology 35: 217–238. Krawczyk M, Sharma R, Mason X et al. (2011). A switch in the neuromodulatory effects of dopamine in the oval bed nucleus of the stria terminalis associated with cocaine self-administration in rats. J Neurosci 31: 8928–8935. Kruijver F, Zhou J, Pool C et al. (2000). Male-to-female transsexuals have female neuron numbers in a limbic nucleus. J Clin Endocrinol Metab 85: 2034–2041. Lebow MA, Chen A (2016). Overshadowed by the amygdala: the bed nucleus of the stria terminalis emerges as key to psychiatric disorders. Mol Psychiatry 21: 450–463. Leri F, Flores J, Rodaros D et al. (2002). Blockade of stressinduced but not cocaine-induced reinstatement by infusion of noradrenergic antagonists into the bed nucleus of the stria terminalis or the central nucleus of the amygdala. J Neurosci 22: 5713–5718. Madeira MD, Lieberman AR (1995). Sexual dimorphism in the mammalian limbic system. Prog Neurobiol 45: 275–333. Martinez LA, Petrulis A (2011). The bed nucleus of the stria terminalis is critical for sexual solicitation, but not for opposite-sex odor preference, in female Syrian hamsters. Horm Behav 60: 651–659. McMenamin BW, Langeslag SJE, Sirbu M et al. (2014). Network organization unfolds over time during periods of anxious anticipation. J Neurosci 34: 11261–11273. Miles OW, May V, Hammack SE (2019). Pituitary adenylate cyclase-activating peptide (PACAP) and the dark side of addiction. J Mol Neurosci 68: 453–464. Mobbs D, Yu R, Rowe JB et al. (2010). Neural activity associated with monitoring the oscillating threat value of a tarantula. PNAS 107: 20582–20586. Mulders WHAM, Meek J, Hafmans TGM et al. (1997). Plasticity in the stress-regulating circuit: decreased input from the bed nucleus of the stria terminalis to the hypothalamic paraventricular nucleus in rats following adrenalectomy. Eur J Neurosci 9: 2462–2471. Myers B, Dolgas CM, Kasckow J et al. (2014). Central stressintegrative circuits: forebrain glutamatergic and GABAergic projections to the dorsomedial hypothalamus, medial preoptic area, and bed nucleus of the stria terminalis. Brain Struct Funct 219: 1287–1303. Nelson RJ, Trainor BC (2007). Neural mechanisms of aggression. Nat Rev Neurosci 8: 536–546. Nieh EH, Matthews GA, Allsop SA et al. (2015). Decoding neural circuits that control compulsive sucrose seeking. Cell 160: 528–541. O’Connor E, Kremer Y, Lefort S et al. (2015). Accumbal D1R neurons projecting to lateral hypothalamus authorize feeding. Neuron 88: 553–564. Oler JA, Birn RM, Patriat R et al. (2012). Evidence for coordinated functional activity within the extended amygdala of non-human and human primates. Neuroimage 61: 1059–1066. Oler JA, Tromp DPM, Fox AS et al. (2017). Connectivity between the central nucleus of the amygdala and the bed nucleus of the stria terminalis in the non-human primate: neuronal tract tracing and developmental neuroimaging studies. Brain Struct Funct 222: 21–39.

FUNCTIONAL ANATOMY OF BNST–HYPOTHALAMUS CIRCUITRY Petrovich GD (2018). Lateral hypothalamus as a motivationcognition interface in the control of feeding behavior. Front Sys Neurosci 12: 14. https://doi.org/10.3389/fnsys. 2018.00014. Pleil KE, Rinker JA, Lowery-Gionta EG et al. (2015). NPY signaling inhibits extended amygdala CRF neurons to suppress binge alcohol drinking. Nat Neuro 18: 545–552. Poulin JF, Arbour D, Laforest S et al. (2009). Neuroanatomical characterization of endogenous opioids in the bed nucleus of the stria terminalis. Prog Neuropsychopharmacol Biol Psychiatry 33: 1356–1365. Powers JB, Newman SW, Bergondy ML (1987). MPOA and BNST lesions in male Syrian hamsters: differential effects on copulatory and chemoinvestigatory behaviors. Behav Brain Res 23: 181–195. Ressler KJ, Mercer KB, Bradley B et al. (2011). Post-traumatic stress disorder is associated with PACAP and the PAC1 receptor. Nature 470: 492–497. Roselli CE, Klosterman SA, Fasasi TA (1996). Sex differences in androgen responsiveness in the rat brain: regional differences in the induction of aromatase activity. Neuroendocrinology 64: 139–145. Russell JS, Trouche S, Reijmers LG (2020). Functional characterization of the basal amygdala-dorsal BNST pathway during contextual fear conditioning. eNeuro 7. https://doi. org/10.1523/ENEURO.0163-20.2020. Sarnyai Z, Shaham Y, Heinrichs SC (2001). The role of corticotropin-releasing hormone in addiction. Pharmacol Rev 53: 209–244. Sayag N, Hoffman NW, Yahr P (1994). Telencephalic connections of the sexually dimorphic area of the gerbil hypothalamus that influence male sexual behavior. Behav Neurosci 108: 743. Segovia S, Guillamo´n A (1993). Sexual dimorphism in the vomeronasal pathway and sex differences in reproductive behaviors. Brain Res Rev 18: 51–74. Silberman Y, Matthews RT, Winder DG (2013). A corticotropin releasing factor pathway for ethanol regulation of the ventral tegmental area in the bed nucleus of the stria terminalis. J Neurosci 33: 950–960. Somerville LH, Whalen PJ, Kelley WM (2010). Human bed nucleus of the stria terminalis indexes hypervigilant threat monitoring. Biol Psychiatry 68: 416–424. Somerville LH, Wagner DD, Wig GS et al. (2013). Interactions between transient and sustained neural signals support the generation and regulation of anxious emotion. Cereb Cortex 23: 49–60. Song Y, Meng Q, Wu K et al. (2020). Disinhibition of PVN-projecting GABAergic neurons in AV region in BNST participates in visceral hypersensitivity in rats. Psychoneuroendocrinology 117: 104690. Stamatakis AM, Sparta DR, Jennings JH et al. (2014). Amygdala and bed nucleus of the stria terminalis circuitry: implications for addiction-related behaviors. Neuropharmacology 76: 320–328. Straube T, Mentzel H, Miltner WHR (2007). Waiting for spiders: brain activation during anticipatory anxiety in spider phobics. Neuroimage 37: 1427–1436. Swaab DF (2003). Bed nucleus of the stria terminalis (BST) and the septum. In: Handbook of clinical neurology,

417

Vol. 79 (3rd series Vol. 1). The human hypothalamus: basic and clinical aspects, part I. Chapter, 7: 149–162. Swaab DF (2007). Sexual differentiation of the brain and behavior. Best Pract Res Clin Endocrinol Metab 21: 431–444. Tabatadze N, Sato SM, Wooley CS (2014). Quantitative analysis of long-form aromatase in the male and female rat brain. PLoS One 9. https://doi.org/10.1371/journal.pone. 0100628. Taha SA, Fields HL (2005). Encoding of palatability and appetitive behaviors by distinct neuronal populations in the nucleus accumbens. J Neurosci 25: 1193–1202. Tartar JL, King MA, Devine DP (2006). Glutamate- mediated neuroplasticity in a limbic input to the hypothalamus. Stress 9: 13–19. Theiss JD, Ridgewell C, McHugo M et al. (2017). Manual segmentation of the human bed nucleus of the stria terminalis using 3 T MRI. Neuroimage 146: 288–292. Tillman RM, Stockbridge MD, Nacewicz BM et al. (2018). Intrinsic functional connectivity of the central extended amygdala. Hum Brain Mapp 39: 1291–1312. Torrisi S, O’Connell K, Davis A et al. (2015). Resting state connectivity of the bed nucleus of the stria terminalis at ultra-high field. Hum Brain Mapp 36: 4076–4088. Torrisi S, Alvarez GM, Gorka AX et al. (2019). Resting-state connectivity of the bed nucleus of the stria terminalis and the central nucleus of the amygdala in clinical anxiety. J Psychiatry Neurosci 44: 313–323. Vranjkovic O, Gasser PJ, Gerndt CH et al. (2014). Stressinduced cocaine seeking requires a beta-2 adrenergic receptor-regulated pathway from the ventral bed nucleus of the stria terminalis that regulates CRF actions in the ventral tegmental area. J Neurosci 34: 1204–1214. Vranjkovic O, Pina M, Kash TL et al. (2017). The bed nucleus of the stria terminalis in drug-associated behavior and affect: a circuit-based perspective. Neuropharmacology 122: 100–106. Walker DL, Miles LA, Davis M (2009). Selective participation of the bed nucleus of the stria terminalis and CRF in sustained anxiety-like versus phasic fear-like responses. Neuropsychopharmol Biol Psychiatry 33: 1291–1308. Walter A, Mai JK, Lanta L et al. (1991). Differential distribution of immunohistochemical markers in the bed nucleus of the stria terminalis in the human brain. J Chem Neuroanat 4: 281–298. Wang Y, Kim J, Schmit MB et al. (2019). A bed nucleus of stria terminalis microcircuit regulating inflammationassociated modulation of feeding. Nat Commun 10: 1–3. Weathington JM, Hamki A, Cooke BM (2014). Sex- and region-specific pubertal maturation of the corticotropinreleasing factor receptor system in the rat. J Comp Neurol 522: 1284–1298. Weitlauf C, Egli RE, Grueter BA et al. (2004). High-frequency stimulation induces ethanol-sensitive long-term potentiation at glutamatergic synapses in the dorsolateral bed nucleus of the stria terminalis. J Neurosci 24: 5741–5747. Wenzel JM, Cotton SW, Dominguez HM et al. (2014). Noradrenergic B-receptor-antagonism within the central nucleus of the amygdala or bed nucleus of the stria

418

I. MAITA ET AL.

terminalis attenuates the negative/anxiogenic effects of cocaine. J Neurosci 34: 3467–3474. Williams TDM, Carter DA, Lightman SL (1985). Sexual dimorphism in the posterior pituitary response to stress in the rat. Endocrinology 116: 738–740. Wills TA, Klug JR, Silberman Y et al. (2012). GluN2B subunit deletion reveals key role in acute and chronic ethanol sensitivity of glutamate synapses in bed nucleus of the stria terminalis. PNAS 109: E278–E287. Winter L, Heitland I, Saryyeva A et al. (2018). Acute Effects of electrical stimulation of the bed nucleus of the stria terminalis/internal capsule in obsessive-compulsive disorder. World Neurosurg 111: e471–e477. Wu MV, Manoli DS, Fraser EJ et al. (2009). Estrogen masculinizes neural pathways and sex-specific behaviors. Cell 139: 61–72.

Yamamoto R, Ahmed N, Ito T et al. (2018). Optogenetic study of anterior BNST and basomedial amygdala projections to the ventromedial hypothalamus. eNeuro 5. https://doi.org/ 10.1523/ENEURO.0204-18.2018. Ye J, Veinante P (2019). Cell-type specific parallel circuits in the bed nucleus of the stria terminalis and the central nucleus of the amygdala of the mouse. Brain Struct Funct 224: 1067–1095. Yohn CN, Dieterich A, Bazer AS et al. (2019). Chronic nondiscriminatory social defeat is an effective chronic stress paradigm for both male and female mice. Neuropsychopharmacology 44: 2220–2229. 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.

Handbook of Clinical Neurology, Vol. 179 (3rd series) The Human Hypothalamus: Anterior Region D.F. Swaab, F. Kreier, P.J. Lucassen, A. Salehi, and R.M. Buijs, Editors https://doi.org/10.1016/B978-0-12-819975-6.00027-3 Copyright © 2021 Elsevier B.V. All rights reserved

Chapter 28

Roles of the bed nucleus of the stria terminalis and amygdala in fear reactions ANNELOES M. HULSMAN1,2, DAVID TERBURG3,4, KARIN ROELOFS1,2, AND FLORIS KLUMPERS1,2* 1

Experimental Psychopathology & Treatment, Behavioural Science Institute, Radboud University, Nijmegen, The Netherlands 2

Affective Neuroscience, Donders Institute for Brain, Cognition and Behaviour, Centre for Cognitive Neuroimaging, Radboud University, Nijmegen, The Netherlands 3

Department of Experimental Psychology, Utrecht University, Utrecht, The Netherlands

4

Department of Psychiatry and Mental Health, University of Cape Town, Cape Town, South Africa

Abstract The bed nucleus of the stria terminalis (BNST) plays a critical modulatory role in driving fear responses. Part of the so-called extended amygdala, this region shares many functions and connections with the substantially more investigated amygdala proper. In this chapter, we review contributions of the BNST and amygdala to subjective, behavioral, and physiological aspects of fear. Despite the fact that both regions are together involved in each of these aspects of fear, they appear complimentary in their contributions. Specifically, the basolateral amygdala (BLA), through its connections to sensory and orbitofrontal regions, is ideally poised for fast learning and controlling fear reactions in a variety of situations. The central amygdala (CeA) relies on BLA input and is particularly important for adjusting physiological and behavioral responses under acute threat. In contrast, the BNST may profit from more extensive striatal and dorsomedial prefrontal connections to drive anticipatory responses under more ambiguous conditions that allow more time for planning. Thus current evidence suggests that the BNST is ideally suited to play a critical role responding to distant or ambiguous threats and could thereby facilitate goal-directed defensive action.

OVERVIEW Based on its anatomical integration with the hypothalamic–pituitary–adrenal axis and the autonomic nervous system, the bed nucleus of the stria terminalis (BNST) is ideally situated to play a modulatory role in emotion and motivation. It is considered a hub that on the one hand receives descending cortical and striatal information on higher level goals and associated motor plans and on the other hand receives ascending information from intero- and exteroceptive systems on changes in homeostasis (Dumont, 2009). The BNST also shares intimate

bidirectional connections with the amygdala. The strongly overlapping functions ascribed to these regions have even raised the question whether the regions should be considered separate or as integral parts of a larger “extended amygdala” (Megerman and Murphy, 1975; Shackman and Fox, 2016). Nevertheless, the amygdala proper is studied considerably more than the BNST over the last decades (Lebow and Chen, 2016). To further understand the role of these regions, in this chapter, we will focus on the realm of research where both have received most attention: fear processing. This is not to deny that both regions also play an important role in

*Correspondence to: Floris Klumpers PhD, Experimental Psychopathology & Treatment Section, Behavioural Science Institute, Radboud University, Postbus 9104, Nijmegen 6500 HE, The Netherlands. Tel: +31-64-399-1031, E-mail: [email protected]

420

A.M. HULSMAN ET AL.

other emotional and motivational processes. Indeed, there is substantial evidence for a role in positively valenced affective processes (Jennings et al., 2013; Avery et al., 2016; Lebow and Chen, 2016; Ch’ng et al., 2018). The current chapter will first provide a comprehensive discussion of the state of the art on the consecutive roles of the BNST and amygdala in fear processing. Subsequently the conclusions will be integrated within a neural model of how these regions together contribute to fear processing. Before we start, however, we will briefly provide the reader with our operational definition of fear.

DEFINING FEAR A recent survey of eminent scientists in the field uncovered widely diverging definitions for fear and anxiety (Mobbs et al., 2019). This necessitates a brief introduction to the terminology we will use in this chapter when speaking of fear responses. In the emotion literature, fear is often defined as a response to specific, directly perceivable threat. This can be contrasted by anxiety which is defined as a diffuse, future-oriented response to ambiguous threat (Barlow et al., 1996). Making a distinction between various types of threat-related responses is highly valuable as the dimensions may correspond also to differences in the subjective feelings, behavior, and physiology depending on the concreteness of a threat. However, a strict fear/ anxiety dichotomy in terminology is not likely to be reflected in the neural circuits. Therefore, for a more precise understanding of how the BNST and amygdala contribute to the diverse aspects of threat processing, we like to define fear in an inclusive manner as: a state triggered by a perceived threat (intern/extern origin) that, depending on urgency (imminence, intensity) and context (resources, coping potential), involves a variety of behavioral, physiological (autonomic nervous system), and subjective (thinking and feeling) reactions with the goal to avoid harm. In our definition, not all elements (behavioral, cognitive, physiological, subjective) need to be present in order to speak of a fear response. It is noteworthy that our definition includes several elements such as defensive responses in the context of ambiguous threats (often referred to as anxiety) as well as elements beyond subjective feelings. This terminology can be debated (LeDoux and Pine, 2016; Fanselow and Pennington, 2017). However, regardless of the semantics, when discussing the roles of the BNST and amygdala it is of critical importance to dissect findings both with respect to threat urgency and the three levels at which emotional processes can be viewed: the subjective level (feelings), the physiological level (bodily responses), and the behavioral level (Bradley and Lang, 2000).

THE BED NUCLEUS OF THE STRIA TERMINALIS Anatomy The BNST is considered part of the extended amygdala and is located at the end of the stria terminalis (see Fig. 28.1). Rodent research suggests that the BNST consists of 12–18 subnuclei, and there is clear evidence from postmortem studies for at least three subnuclei in the human BNST; namely the central, lateral, and medial BNST (Walter et al., 1991). However, given the small size of the BNST relative to the resolution offered by functional neuroimaging, the current literature on human BNST function typically treats the BNST as one functional unit. The BNST has extensive connections with limbic regions (amygdala, hypothalamus, hippocampus, periaqueductal gray (PAG), infralimbic cortex), striatal regions (nucleus accumbens (NAcc), ventral tegmental area), and prefrontal regions (Avery et al., 2014; Shackman and Fox, 2016). The BNST is therefore ideally situated to evaluate risks and then through connections to prefrontal, striatal, hypothalamic, and brain stem regions instantiate appropriate cognitive, motor, endocrine, and autonomic responses.

Current theories of the BNST in behavioral, subjective, and physiological fear responses OVERVIEW Classic theories of the BNST in fear responding are predominantly based on rodent research (Walker et al., 2003; Davis et al., 2010; Miles et al., 2011). These

Fig. 28.1. Coronal view of the bed nucleus stria terminalis (BNST; in blue) and the amygdala (in red) shown on a T1-weighted MRI image of the human brain.

ROLES OF THE BNST AND AMYGDALA IN FEAR REACTIONS studies provided evidence that the BNST is indispensable in the production of more sustained fear states that arise in the context of ambiguous threats. For example, lesions of the BNST blocked fear responses when rats were exposed to light environments (Walker et al., 2003). For rodents, the light in itself poses no direct harm, yet light environments are generally anxiogenic due to the increased risk of detection and predation. When exposed to the indirect threat created by the light, BNST lesions blocked potentiation of the startle response—a robust cross-species indicator of fear. Critically, however, these same lesions did not block startle potentiation when the animals were exposed to cues that predicted a highly aversive electric shock would follow soon. These findings together with others were the foundation for the theory that the BNST is important for generating fear responses during ambiguous or contextual threat such as light exposure but not for specific, more imminent threats such as shock exposure (Walker et al., 2009; Davis et al., 2010). In humans the BNST is substantially larger and more developed than in rodents (Avery et al., 2014) (Fig. 28.1), suggesting the relevance of a full translational approach to investigate this structure’s function (Fox and Shackman, 2019). In this chapter we will therefore, whenever possible, mainly focus on the existing human literature. Recently, advances in neuroimaging have led to an increase in human studies probing the role of the BNST in fear responding. These studies suggest that, in accordance with the rodent literature, the human BNST is involved in future threat anticipation (Straube et al., 2007; Klumpers et al., 2015; Hashemi et al., 2019), particularly when threat is uncertain, unpredictable, and/or distant in time or space (Mobbs et al., 2010; Alvarez et al., 2011; Grupe et al., 2013; Goode and Maren, 2017; Klumpers et al., 2017; Clauss et al., 2019). Neuroimaging investigations of the BNST are complicated due to the BNST’s small size for neuroimaging purposes, which makes it hard to ascribe activation with certainty. Nevertheless, such findings have also been supported by high-resolution imaging studies. For example, Naaz et al. (2019) presented subjects with cues that either signaled that an aversive loud scream would certainly be delivered shortly or that signaled that the scream would perhaps be delivered after a longer and unknown delay. The BNST showed significantly stronger responses in the latter, more ambiguous condition. Several studies have pointed toward an important role for the BNST in driving individual differences in emotional responding. Indeed, the BNSTshows differentiation in size, connectivity, and function with, for example, larger volumes in males compared to females (Zhou et al., 1995; Chung et al., 2002; Flook et al., 2020). It has been hypothesized that this sexual dimorphism may explain the consistently higher prevalence of anxiety

421

disorders in females (McLean et al., 2011). However, the influence of this sexual dimorphism on fear responses is still unclear due to a lack of adequately powered studies on this topic. Involvement of the BNST in pathological anxiety has been suggested by its contribution to key symptoms of anxiety disorders, namely intolerance of uncertainty and unpredictability (Sierra et al., 2015). Indeed, the BNST has been linked to individual differences in normative anxiety. In a study by Somerville et al. (2010), participants viewed a continuously fluctuating line that they thought reflected their own or another’s internal state. If this line exceeded a threshold, electric shocks were accumulated and delivered at the end of the experiment. Participants were instructed to remain calm to avoid accumulating shocks. The results showed stronger responses in the BNST to increasing line height (i.e., increased risk of receiving shocks), with exaggerated responses as a function of trait anxiety. Other studies have provided further support for this finding and suggested that high trait anxiety is associated with increased and/or more sustained BNST activation (Somerville et al., 2010, 2013) and altered BNST connectivity (McMenamin et al., 2014; Brinkmann et al., 2018). Clinical models of stress and anxiety have generally focused on the amygdala, yet there is a growing body of research that also links the BNST to anxiety disorders, such as generalized anxiety disorder (Yassa et al., 2012; Buff et al., 2017), posttraumatic stress disorder (Brinkmann et al., 2017b), social anxiety disorder (Clauss et al., 2019; Figel et al., 2019), panic disorder (Brinkmann et al., 2017a), and spider phobia (Straube et al., 2007; M€unsterk€otter et al., 2015). Most of these studies showed that, relative to healthy controls, anxiety patients show enhanced BNST activation during uncertain or unpredictable threat anticipation, in contrast to enhanced amygdala activation during certain or predictable threat anticipation. This suggests not only a transdiagnostic neural model of fear but also provides evidence for a potential functional dissociation between the BNST and amygdala. Behavioral responses In addition to contributing to passive threat anticipation, the BNST is also involved with active fear-like behavior. For example, BNST stimulation in male mice has been reported to stimulate anxiety-like behavior in an openfield test (Giardino et al., 2018) and freezing responses in conditions of contextual threat (rather than specifically cued threat) (Sullivan et al., 2004). Recent work has begun to examine the role of the BNST in avoidance. A pharmacological rodent study showed that, only in females, oxytocin receptors in the BNST play a role in increasing avoidance behavior after social defeat stress

422

A.M. HULSMAN ET AL.

(Duque-Wilckens et al., 2018). Comparatively less human research has focused on the role of the BNST in behavioral responses. In one notable exception, subjects performed a free-operant avoidance task in which they were exposed to a threat-avoidance context and a safety context. They learned that in the threat-avoidance context a button press avoided monetary loss, whereas in the safety context no response was necessary. Results showed BNST involvement in anticipation of the decision to avoid (Schlund et al., 2013). Together these findings are consistent with the idea that the BNST plays a role in driving anxiety-like behavioral responses, particularly during situations when threat is more distant. In these more ambiguous situations, the BNST has been suggested to facilitate goal-directed behavioral responses by contributing to the valence assignment that ultimately drives approach/avoid decisions (Lebow and Chen, 2016; Ch’ng et al., 2018; Klumpers and Kroes, 2019). Subjective responses Evidence for the involvement of the BNST in subjective responses of fear primarily comes from deep brain stimulation (DBS) research. DBS is a treatment method for various brain disorders in which electrodes are surgically implanted at specific locations of the brain. In patients with treatment resistant obsessive–compulsive disorder (OCD) or major depressive disorder (MDD), these electrodes in many cases target the nucleus accumbens (Baas et al., 2014) or the BNST. Due to the small distance and extensive connections between the NAcc and BNST, it is conceivable that DBS to the NAcc also affects the BNST and thereby decreases anxiety. Indeed, these studies provided converging evidence that DBS targeting the BNST and/or NAcc strongly reduces reported OCD symptoms and anxiety (Baas et al., 2014; Luyten et al., 2016; Raymaekers et al., 2017). Furthermore, DBS in the BNST of a patient with comorbid MDD and anorexia nervosa has been proven effective in reducing foodrelated anxiety (Blomstedt et al., 2017). Besides evidence from DBS research, the role of the BNST in subjective responses of fear is further supported by MRI research. One interesting study showed that in contrast to the amygdala, the threat response in the BNST is increased for individuals with high relative to low selfreported anxiety (McMenamin et al., 2014). Together there is relatively strong causal evidence that activation of the BNST directly contributes to subjective feelings of fear. Physiological responses A large body of evidence has linked BNST function to autonomic and endocrine responses to stress in rodents (reviewed by Crestani et al., 2013). Unfortunately, there

is a relative lack of studies linking BNST activity to physiological responses of fear in humans. In the aforementioned DBS study (Baas et al., 2014) the authors found that, in contrast to subjective fear responses, DBS in the NAcc/BNST did not affect fear-potentiated startle. Furthermore, MRI research showed moderate to weak correlations between threat responses in the BNST and skin conductance during a contextual threat conditioning task (Alvarez et al., 2015) and with heart rate responses during stress (Banihashemi et al., 2015). In conclusion, despite a relative paucity of studies that have investigated the role of the human BNST in physiological threat responses, the evidence so far is consistent with the idea that the BNST may facilitate autonomic and endocrine responses during more diffuse states of threat and stress. Rodent studies have specifically shown evidence that the BNST instantiates these changes by connecting limbic forebrain regions such as the amygdala and medial prefrontal cortex to the hypothalamus and brain stem (Crestani et al., 2013).

THE AMYGDALA Overview The interplay of amygdala and fear has been extensively studied for several decades. Core feature in this research is the amygdala’s crucial role in threat conditioning, both for acquiring new stimulus-threat associations and expressing the resulting conditioned physiological responses and behavior. Following this evidence, the amygdala is often seen as a fear-module responsible for rapid threat detection and subsequent activation of automated mechanisms underlying physiological reactions and freeze/fight/flight behaviors (Isaac and Thiemer, 1975). This view was further extended based on attention research providing evidence that the amygdala can bias automated attention toward threat (Ohman, 2005), thereby facilitating more rapid cognitive processing of threat (Phelps and LeDoux, 2005). According to these views, the amygdala thus not only promotes threat reactivity via the autonomous nervous system and associated automatic behaviors but also influences cognitive processing in such a way that potential threat has priority. Recent accounts take this notion even further by showing that the amygdala has context-dependent control over threat physiology and responsivity in favor of either automated freeze/fight/flight or instrumental goal-directed behaviors (Campese et al., 2016; Cain, 2019). The latter view is driven by an increased interest in active escape experiments as compared to the traditional passive threat paradigms that have been used to study threat reactivity and conditioning (LeDoux et al., 2017; Cain, 2019). Moreover, the recent availability of more

ROLES OF THE BNST AND AMYGDALA IN FEAR REACTIONS precise human lesion models and neuroimaging techniques has strongly facilitated the translation of animal neuroscience data to the human case (Terburg et al., 2018). Together these studies provide for robust evidence that the amygdala has bidirectional control over threat reactivity, thus providing for the behavioral flexibility that is necessary to deal with acute threat.

Anatomy The amygdala is not one singular structure but a collection of nuclei sometimes referred to as the amygdaloid nucleus, which acts as a hub between cortex, striatum, and subcortex (Janak and Tye, 2015). In human neuroimaging, three subdivisions of the amygdala can be distinguished: dorsal, ventrolateral, and medial (Bickart et al., 2014), of which the first two are of particular relevance for threat behaviors. These subdivisions strongly correspond to histological mapping of amygdala microstructure in humans (Bzdok et al., 2013), where the dorsal region consists of the central (CeA) and medial (MeA) nuclei, and the ventrolateral region represents the basolateral complex (BLA), including the lateral, basolateral, basomedial, and basoventral nuclei (Amunts et al., 2005). The dynamic interaction of CeA and BLA coordinates threat reactivity. The BLA receives multimodal information from the thalamus and orbitofrontal cortex and is therefore often considered to be the “sensory” amygdala (LeDoux and Daw, 2018). The BLA can excite (directly) and inhibit (via the intercalated cells, IC) the CeA, which is the main output region of the amygdala, particularly in relation to threat reactivity. Important in this organization is that the BLA also communicates with a corticostriatal network consisting of orbitofrontal and ventromedial prefrontal cortices and nucleus accumbens (Barbas et al., 2003; LeDoux and Brown, 2017; LeDoux and Daw, 2018). Via this route the BLA can influence cognition, subjective experience, and goal-directed motivation. The CeA on the other hand projects to the BNST (Shackman and Fox, 2016) and to the reactive threat system, most particularly brainstem regions, i.e., periaqueductal gray (PAG), locus coeruleus (LC) and the reticular formation (RF) (Isaac and Thiemer, 1975; LeDoux and Daw, 2018). Via this route the CeA has substantial control over subjective, physiological, as well as behavioral threat reactivity. Finally, the BLA as well as CeA project to the hypothalamus resulting in another indirect line of influence on downstream threat reactivity as well as influence on hormonal stress responses (Reppucci and Petrovich, 2016). Functional connectivity work in humans has confirmed this BLA/CeA organization as the BLA has been identified as a main hub of the perception network and the CeA as a main hub of the aversion network (Bickart et al., 2014). Fig. 28.4 shows

423

a visualization of this amygdala connection model and a summary of the behavioral outcomes.

Current theories of the amygdala in behavioral, subjective, and physiological fear responses Three major lines of evidence underlie the present view of human amygdala functioning. Animal research, mostly from rodents and to a lesser extent nonhuman primates, human neuroimaging research mainly using functional magnetic-resonance imaging and stimulation and lesion research in humans. Human lesion research is predominantly driven by research into Urbach–Wiethe disease (UWD), which is a recessive function mutation within the extracellular matrix protein 1 gene (ECM1) (Hamada et al., 2002) that often results in focal bilateral amygdala calcification rendering it dysfunctional. These focal calcifications turn UWD therefore into a unique amygdala-lesion model for human behavior. The first UWD study with respect to affective processing, a case study in patient SM-046, showed that full amygdala calcification can result in a selective deficit in the ability to recognize fearful facial expressions (Adolphs et al., 1994). This was one of the first translations of animal amygdala functioning with respect to fear processing to the human case, which sparked many human neuroimaging studies into amygdala functioning. Eventually a view emerged that the human amygdala’s role in reacting to threat stimuli is most apparent on preconscious levels (Whalen et al., 1998, 2004), which was argued to serve an automatic alarm function (Liddell et al., 2005). Follow-up research in SM-046 therefore focused on attentional mechanisms and showed that she was not necessarily impaired in fear recognition per se, but rather lacked the automated mechanism that directs attention to fearful eyes (Adolphs et al., 2005). More recent UWD research has provided a human model for amygdala subregion functioning after research started in a South African population with UWD (Thornton et al., 2008). This population dates back to the 17th century when Dutch/German settlers introduced the disorder which spread locally due to the founder effect (Van Hougenhouck-Tulleken et al., 2004). Interestingly, it was found that this population has focal and bilateral calcifications limited to the BLA, while particularly the CeA remained intact and functional (Terburg et al., 2012; van Honk et al., 2016). Research in this group has not only translated evidence from animal research (Isaac and Thiemer, 1975) showing the BLA is crucial for threat conditioning (Klumpers et al., 2015) but also showed that the attentional mechanism that is lacking after full amygdala damage in SM-046 (Adolphs et al., 2005) is strengthened after BLA damage

424

A.M. HULSMAN ET AL.

(Terburg et al., 2012). This dissociation eventually culminated in a cross-species examination of amygdala subregion functioning with regard to the interplay of autonomic threat reactivity and goal-directed threat behaviors (Terburg et al., 2018). In the next paragraph, we will discuss this intriguing interaction of threat physiology and behavior.

THREAT PHYSIOLOGY AND BEHAVIOR The CeA has direct control over autonomous reactions through projections to the hypothalamus and several midbrain nuclei like the PAG, LC, and RF (Isaac and Thiemer, 1975; Phelps and LeDoux, 2005; Reppucci and Petrovich, 2016; Schipper et al., 2019). As such, it has substantial influence over physiological mechanisms such as heart rate and blood pressure, endocrine reactions like the cortisol stress response and the expression of autonomous behaviors such as noradrenalin release, freezing, startle reflexes, and pupil dilation (see Fig. 28.4). This provides the CeA with an excellent position to influence all aspects of automatic threat reactivity. The BLA can both excite (directly) and inhibit (via the intercalated cells) the CeA (Isaac and Thiemer, 1975; de la Mora et al., 2010; Gregoriou et al., 2019), and across species it has been confirmed that these BLA projections are particularly important for acquisition (Klumpers et al., 2015) and extinction (Busti et al., 2011; Giustino et al., 2020) of stimulus–response associations in relation to threat. The CeA is subsequently necessary for the expression of these conditioned responses (Isaac and Thiemer, 1975), but recent cross-species research has extended this function by describing the CeA as a neural switch between active and passive threat behaviors. According to this view, projections within the CeA, from the lateral (CeL) to medial (CeM) nucleus, serve either the expression of downstream autonomous threat reactions like freezing and startle reflexes or their inhibition which provides for the possibility to actively respond to the threat at hand (Ciocchi et al., 2010; Gozzi et al., 2010; Haubensak et al., 2010; Fadok et al., 2017). Intriguingly it seems to be the case that the BLA can activate this switch as glutamatergic projections from BLA to CeL in mice reduce passive anxiety in favor of active exploration (Tye et al., 2011). In humans it is impossible to study CeA functioning at this microlevel, but neuroimaging findings are in line with this theory. By manipulating the imminence of a threat during an active avoidance task, it was shown that when a threat comes so close that active escape becomes impossible, brain activity shifts from a network of BLA and vmPFC to CeA and PAG, corresponding to a behavioral switch from strategic avoidance to panic (Mobbs et al., 2007, 2009). This evidence does not yet prove that

the BLA can control the active-to-passive switch within the CeA, but a recent cross-species examination of autonomous threat responses in relation to threat imminence bridged these lines of human and rodent evidence (Terburg et al., 2018). In this UWD and rat study, humans with BLA damage were unable to regulate their startle response in a situation where they anticipated to escape a highly imminent threat. This was reflected in increased brainstem reactivity in a location consistent with the RF, driver of motor reflexes (Davis et al., 1982), and a lack of CeA-RF functional connectivity. Together this suggests that, when goal-directed escape is necessary, input from the BLA is crucial for regulation of autonomous threat reactivity by the CeA (see Fig. 28.2). In a parallel experiment in rats, this idea was confirmed and extended. BLA silencing strongly impaired the ability of rats to snap out of their freezing behavior in favor of goal-directed escape during imminent threat. Moreover, as could be expected from the earlier evidence in mice (Tye et al., 2011) a glutamatergic pathway from BLA to CeL was crucial for this switch from passive to active threat reactivity, as well as the ability to regulate startle reflexes. Combined, this thus confirms that the BLA can directly control the neural passive-to-active switch within the CeA (Ciocchi et al., 2010; Gozzi et al., 2010; Haubensak et al., 2010; Fadok et al., 2017). The CeA passive-to-active switch can facilitate active behavior in two ways. The first is nonspecific fight/flight behavior, related to panic, generated by switching from freezing to flight/fight in the downstream PAG (Mobbs et al., 2009; Benarroch, 2012; Hermans et al., 2013; Tovote et al., 2016; Fadok et al., 2017; Hashemi et al., 2019). This behavior is the final stage of Michael Fanselow’s threat imminence model, where distant threat evokes active avoidance strategies, imminent threat evokes autonomous threat anticipation and freezing both serving to rapidly react defensively, e.g., startle reflexes and nonspecific flight, when a threat is delivered (Fanselow et al., 2019). The second CeA passive-toactive switch function can provide for the possibility for goal-directed escape behavior. This is the switch function that is activated by glutamatergic projection from the BLA, which effectively downregulates the PAG’s autonomous reactions thereby providing for a window for goal-directed escape action (Gladwin et al., 2016). The actual escape behavior can subsequently be initiated in the NAcc, possibly in response to input from the BLA (LeDoux and Daw, 2018). Indeed, in both the human and animal component (Terburg et al., 2018), the escape behavior involved a specific action (either a button press or a shuttle from one cage to the other), which are instrumental goal-directed actions that are subserved by this BLA-NAcc pathway (Namburi et al., 2015; LeDoux et al., 2017).

ROLES OF THE BNST AND AMYGDALA IN FEAR REACTIONS

425

Fig. 28.2. Human BLA damage is associated with increased potentiation of the startle response during imminent yet escapable threat. In this study, patients with BLA damage due to Urbach–Wiethe disease were compared to matched healthy controls. The results show that healthy subjects suppress fear measured by startle potentiation when imminent threat (imm) required a quick avoidance response. This pattern is significantly reduced in patients, suggesting a critical role for the BLA in the instrumental downregulation of fear responses. There were no differences when threat was easily escapable (distant; dis) or inescapable (ines). Strikingly, the same patients show reduced startle responses in a passive fear conditioning context where healthy subjects acquired fear responses (Klumpers et al., 2015). Together with other research this suggests that the human BLA provides bidirectional instrumental control over fear. Adapted from Terburg D, Scheggia D, Triana Del Rio R et al. (2018). The basolateral amygdala is essential for rapid escape: a human and rodent study. Cell 175: 723–735.e716 with permission.

In sum, the amygdala orchestrates acute threat reactivity. The BLA promotes threat conditioning and the CeA activates the associated responses through its projections to the downstream threat system. Hormonal reactions are subsequently activated in the hypothalamus, freezing and nonspecific flight in the PAG, noradrenalin release in the LC, and motor reflexes in the RF. Furthermore, the BLA uses its sensory input to adjust the CeA to the situation at hand. Most notably, the BLA can prevent the CeA from activating threat reactivity in favor of goal-directed escape behavior when this possibility arises. Impairment of this modulatory BLA-CeA pathway can thus result in reduced escape efficiency but can also underlie the hypervigilance for threat observed in UWD patients with BLA-specific calcifications resulting in increased unconscious processing and attention for facial threat (Terburg et al., 2012) and increased threat biases in their emotional judgments (de Gelder et al., 2014). Given that many anxiety disorders are associated with such hyperresponsivity to threat cues (Bishop,

2008; Klumpers et al., 2017), this mechanism might contribute to the often reported involvement of the amygdala in anxiety-related disorders (Phelps and LeDoux, 2005).

SUBJECTIVE EXPERIENCE Many studies have linked amygdala (re)activity to the subjective experience of fear in healthy populations and in relation to anxiety disorders (Phelps and LeDoux, 2005). Although full amygdala damage indeed can result in severely reduced subjective experience of fear (Feinstein et al., 2011), a direct subjective amygdala function seems unlikely as other brain pathways can take over part of this function in some cases (Becker et al., 2012). Moreover, anxiety-related disorders are even quite prominent in humans with BLA lesions due to UWD (Thornton et al., 2008), suggesting that subjective fear can also increase after damage to the amygdala. Taken together this suggests that the amygdala affects subjective experience of fear via indirect mechanisms.

426

A.M. HULSMAN ET AL.

Subjective experience of fear and anxiety can be argued to reflect uncertainty about if, or when, a threat will strike. The amygdala is however responsive to acute threat or to visual imagery associated with direct threat. Particularly of interest in this regard is that the amygdala responds to acute danger and actual threat exposure, but not to uncertain threat anticipation (as discussed later). Moreover, amygdala reactivity to threat imagery in itself is not related to subjective experience of anxiety, but this association only becomes apparent in task-interference settings (Bishop, 2008). It is therefore unlikely that amygdala reactivity to threat directly evokes the subjective experience of fear and anxiety. The alternative interpretation would thus be that the amygdala contributes to subjective feelings of fear and anxiety indirectly (LeDoux and Brown, 2017). As noted earlier, the amygdala has a profound function in the bidirectional control of threat physiology and behavioral tendencies (Terburg et al., 2018) and boosts memory consolidation under stress (de Voogd et al., 2017). These can be considered precursors of subjective affect that can strongly interfere with cognitive processing (Bishop, 2008; Terburg et al., 2012), and without the effects on physiology and cognition, the subjective fear might even be absent (Feinstein et al., 2011). An important conclusion that should therefore be taken from this indirect role is that one should not speak of subjective fear, but of threat (or fear) reactivity, when describing the functional role of the amygdala. An intriguing illustration of this distinction comes from a recent amygdala stimulation study, showing that direct electrical stimulation of the amygdala can result in strong physiological threat reactivity with hardly any experience of fear or anxiety (Inman et al., 2018). Notably, the scarce subjective experiences that were activated by this stimulation were not limited to the experience of fear and anxiety and could even involve pleasant feelings, showing that the amygdala can indirectly influence subjective fear and also emotional experience in general (LeDoux and Brown, 2017).

DIFFERENCES BETWEEN THE AMYGDALA AND BNST IN CONTRIBUTIONS TO FEAR Despite the substantially larger quantity of evidence for the amygdala, the previous sections indicate that both BNST and amygdala play a role in various aspects of fear processing. As mentioned at the start of this chapter, classic rodent work has outlined dissociable roles for the amygdala and BNST in fear. Whereas the central nucleus of the amygdala has been implicated in both phasic and sustained responses to threat, the BNST has mainly been implicated in sustained responses to uncertain, unpredictable, and/or distant

threats (Davis et al., 2010). These findings are the basis of one of the most influential models of the extended amygdala (Davis et al., 2010). However, the majority of human studies fail to statistically test the interaction between region and condition, meaning that they do not directly compare BNST and amygdala activity as a function of how acute a threat is. Moreover, rodent research suggests that bidirectional projections between the central amygdala and BNST facilitate many fear-like responses, such as approach–avoidance behavior (Miller et al., 2019), fear memory indicated by freezing (Asok et al., 2018), and anxiogenic behavior in the elevated plus maze test (Yamauchi et al., 2018). The strong bidirectional projections indeed suggest that there is a great degree of coordination between these regions that could explain the many common and complementary functions of the amygdala and BNST. Therefore, there is ongoing debate about whether the BNST and amygdala have distinct functions or form one functional unit together (Shackman and Fox, 2016). Several recent neuroimaging findings in humans have however started to shed light on the potentially complimentary roles of these regions.

Threat imminence Klumpers et al. (2017) were one of the first to make a direct comparison between the BNST and amygdala in humans. In two relatively large and independent samples (n ¼ 108 and n ¼ 70), they investigated potential differential contributions of the BNST and amygdala to defensive responding as a function of threat imminence in time. In both experiments, two cues were presented indicating either threat or safety from electrical shocks (shock anticipation phase). While safe cues were never paired with a shock, a subset of the threat cues were paired with shock delivery (shock confrontation phase). The results showed a shift in neural activity from the BNST toward the amygdala when moving from a state of threat anticipation to confrontation (see Fig. 28.3).

Temporal predictability and response duration Other recent studies focused on the dissociative roles of BNSTand amygdala as a function of temporal predictability and response duration. In two independent studies, participants viewed neutral and negatively valenced affective pictures that appeared with a predictable or unpredictable timing (Somerville et al., 2013; Pedersen et al., 2019). Both studies used a mixed block event-related design that enables comparing responses over a block of trials (i.e., sustained responses) with responses to a single trial (i.e., phasic responses). Findings by Somerville et al. (2013) suggested a double dissociation in which the BNST exhibits sustained activation, but not phasic activation to

ROLES OF THE BNST AND AMYGDALA IN FEAR REACTIONS

427

Fig. 28.3. Amygdala and BNST show dissociable patterns of activity along the threat imminence continuum. In this functional MRI study, it was shown that in two independent samples of participants the BNST showed relatively stronger activity than the amygdala during uncertain anticipation of aversive electrical stimulation. On the other hand, the amygdala showed relatively stronger activation during the direct confrontation with this stimulation. These findings are consistent with a relatively stronger involvement of the BNST in situations when threat is uncertain, whereas the amygdala may be more strongly involved in acute situations. Adapted with permission from Klumpers F, Kroes MCW, Baas J et al. (2017). How human amygdala and bed nucleus of the stria terminalis may drive distinct defensive responses. J Neurosci 37: 3830–3816.

negative vs neutral pictures, whereas the amygdala shows opposite patterns. Additionally, the BNST showed the previously mentioned patterns for unpredictable vs predictable pictures. However, as the interaction between region (BNST vs amygdala) and type of activity (phasic vs sustained) was not statistically tested, no strong conclusions could be drawn on the dissociative roles of the BNST and amygdala. This was resolved by Pedersen et al. (2019) who showed that the BNST exhibits both phasic and sustained responses to negative vs neutral pictures. Yet, these sustained responses were significantly stronger than the phasic responses. However, there was no difference in the strength of phasic and sustained responses in the amygdala. This lack of differential responding in the amygdala could be explained by the use of relatively low threatening stimuli (i.e., affective pictures) rather than high threatening stimuli (i.e., electric shocks) that are commonly used in animal work (Delgado et al., 2011; Klumpers and Kroes, 2019). There was no evidence that predictability affected phasic or sustained responses in either the BNST or amygdala. In conclusion, there is no hard evidence for a

strict double dissociation in terms of temporal predictability or response duration. However, these studies did provide evidence for partial functional segregation, in which the BNST is more strongly involved in sustained responding to threat compared to the amygdala.

Outcome predictability Another potential factor of dissociation between the BNST and amygdala as suggested by Davis et al. (2010) is outcome predictability. This was elegantly studied by Clauss et al. (2019) in a mixed sample of healthy controls (n ¼ 35) and participants who met criteria for social anxiety disorder or another anxiety disorder (n ¼ 9). In the predictable condition, participants were trained to associate one cue with a fearful face and another cue with a neutral face. In the unpredictable condition, participants encountered a novel cue that was randomly followed by either a fearful or neutral face. Relative to the amygdala, the BNST showed increased responding to unpredictable cues. There was no significant

428

A.M. HULSMAN ET AL.

MPFC Goal directed behaviour (avoidance + escape) Subjective experience Striatum NAcc

BNST

Hormonal stress-response

H OFC

CeA M L

IC BLA

PAG

Freezing and non-specific flight

Brainstem LC

Noradrenalin release RF

Motor responses (startle)

Threat learning Flexible threat response

Fig. 28.4. Behavioral, physiological, and subjective levels of fear responding and their relation to the neural networks of the amygdala (in blue) and BNST (in green). Color coded are the downstream threat reactivity network (in red) and the upstream corticostriatal network connected to the BNST (in green). This model suggests stronger involvement of the amygdala in direct physiological modulation and acute threat responding through the network downstream threat reactivity network. Conversely, stronger BNST connections to corticostriatal regions may facilitate goal-directed behavior and subjective experience. Connections within the amygdala are either excitatory (triangles) or inhibitory (dashes). BLA, basolateral amygdala complex; BNST, bed nucleus of the stria terminalis; CeA, central amygdala with its lateral (L) and medial (M) subdivisions; H, hypothalamus; IC, intercalated cells; LC, locus coeruleus; mPFC, medial prefrontal cortex; NAcc, nucleus accumbens; OFC, orbitofrontal cortex; PAG, periaqueductal gray; RF, reticular formation.

dissociation between these areas in their response to fearful and neutral images following predictable or unpredictable cues. However, there was a dissociation in their response to images following unpredictable cues that was moderated by social anxiety: the threat effect shifted from the amygdala to the BNST with increased social anxiety. Together these studies did not provide evidence for a full double dissociation in threat anticipation. However, the model by Davis et al. (2010) is largely supported. Indeed, the BNST is often more involved in uncertain, unpredictable, and/or distant threats than the amygdala.

At the same time, opposite patterns are not always observed for the amygdala. It is also not completely clear to what extent this dissociation applies to behavioral, subjective, and physiological measures of fear. Yet, an indication for dissociable functions is provided by differences in BNST and amygdala functional connectivity patterns. Research has consistently shown robust interconnectivity between the BNST and amygdala (Avery et al., 2014; Klumpers et al., 2017; Gorka et al., 2018). However, studies directly comparing amygdala and BNST connectivity patterns show distinct functional

ROLES OF THE BNST AND AMYGDALA IN FEAR REACTIONS coupling to other brain regions. These studies suggested that the BNST connected more with a striatal/prefrontal network that also includes the hypothalamus, whereas the amygdala showed greater coupling to a ventrocaudal network including occipital cortex and brainstem (Klumpers et al., 2017; Gorka et al., 2018). These relative differences in how the regions are functionally connected could very well contribute to each region having a slightly different functional role.

Integrative model of the role of the amygdala and BNST in fear Based on the literature discussed previously, it seems the BNST and amygdala have clearly overlapping yet also potentially somewhat complimentary roles functions in fear responding. Both regions are strongly interconnected and together serve as a hub between the cortex and subcortical regions. This way, these regions together coordinate emotional reactions that are instantiated downstream through parallel projections to the striatum (e.g., for instrumental motor responses), the hypothalamus (e.g., blood pressure change and corticosteroid release), and the brainstem (e.g., fast defensive freezing and startle reflex) (Davis and Whalen, 2001; Fox et al., 2018). Yet studies have suggested functional distinctions that seem to match the differences in connectivity to other brain regions, both upstream and downstream. We summarized the existing evidence of the current review in Fig. 28.4. Notably, the model extends current theories about the potentially dissociative roles of the BNST and amygdala to behavioral tendencies in response to threat. Indeed, most theories on defensive action converge on the amygdala outputs to the nucleus accumbens as a central path for avoidance behavior (LeDoux and Daw, 2018; Cain, 2019; Klumpers and Kroes, 2019), whereas amygdala outputs to the PAG/brainstem support fast reflexes, nonspecific flight, and freezing. The amygdala CeA outputs may be controlled by the BLA, which facilitates goaldirected behavior. Many theories however have not yet included a role for the BNST, which is—as reviewed previously—likely to play an important role in driving defensive behavior particularly under conditions of more distant threat. Given the dissociations in temporal dynamics and functional connectivity of the BNST and amygdala, we conjecture that the BNST is more involved in organizing and planning instrumental behavior, whereas the amygdala plays a role in controlling more acute responses. On a subjective level, the BNST might therefore be more associated with subjective phenomena such as stress (Hu et al., 2020), worry, and intolerance of uncertainty whereas the amygdala is primarily associated with acute physiological and behavioral responses through stronger downstream connections with the brain

429

stem (primarily from the CeA). In addition, the strong connections to sensory regions of the BLA allow for fast learning about appropriate responses in a range of potentially threatening conditions. In this way, coordinated activity in a range of situations can be associated with a relatively stronger activity of the BNST network when threats are uncertain, while with increasing threat imminence, the amygdala network becomes more dominant. Together the studies reviewed in this chapter provide strong support for a critical role of the BNST in fear responding. While many of the proposed functions are dependent on the strong interactions with the amygdala, there is also evidence for distinct contributions from each region. The particular contribution of the BNST to situations of more ambiguous threat provides an impetus for further investigating the role of this region as it could be argued that much of the stress and fear we experience in modern times does not result from acute life-threatening situations.

REFERENCES Adolphs R, Tranel D, Damasio H et al. (1994). Impaired recognition of emotion in facial expressions following bilateral damage to the human amygdala. Nature 372: 669–672. Adolphs R, Gosselin F, Buchanan TW et al. (2005). A mechanism for impaired fear recognition after amygdala damage. Nature 433: 68–72. Alvarez RP, Chen G, Bodurka J et al. (2011). Phasic and sustained fear in humans elicits distinct patterns of brain activity. Neuroimage 55: 389–400. Alvarez RP, Kirlic N, Misaki M et al. (2015). Increased anterior insula activity in anxious individuals is linked to diminished perceived control. Transl Psychiatry 5: e591. Amunts K, Kedo O, Kindler M et al. (2005). Cytoarchitectonic mapping of the human amygdala, hippocampal region and entorhinal cortex: intersubject variability and probability maps. Anat Embryol 210: 343–352. Asok A, Draper A, Hoffman AF et al. (2018). Optogenetic silencing of a corticotropin-releasing factor pathway from the central amygdala to the bed nucleus of the stria terminalis disrupts sustained fear. Mol Psychiatry 23: 914–922. Avery SN, Clauss JA, Winder DG et al. (2014). BNST neurocircuitry in humans. Neuroimage 91: 311–323. Avery SN, Clauss JA, Blackford JU (2016). The human BNST: functional role in anxiety and addiction. Neuropsychopharmacology 41: 126–141. Baas JM, Klumpers F, Mantione MH et al. (2014). No impact of deep brain stimulation on fear-potentiated startle in obsessive-compulsive disorder. Front Behav Neurosci 8: 305. Banihashemi L, Sheu LK, Midei AJ et al. (2015). Childhood physical abuse predicts stressor-evoked activity within central visceral control regions. Soc Cogn Affect Neurosci 10: 474–485. Barbas H, Saha S, Rempel-Clower N et al. (2003). Serial pathways from primate prefrontal cortex to autonomic areas may influence emotional expression. BMC Neurosci 4: 25.

430

A.M. HULSMAN ET AL.

Barlow DH, Chorpita BF, Turovsky J (1996). Fear, panic, anxiety, and disorders of emotion. Nebr Symp Motiv 43: 251–328. Becker B, Mihov Y, Scheele D et al. (2012). Fear processing and social networking in the absence of a functional amygdala. Biol Psychiatry 72: 70–77. Benarroch EE (2012). Periaqueductal gray: an interface for behavioral control. Neurology 78: 210–217. Bickart KC, Dickerson BC, Barrett LF (2014). The amygdala as a hub in brain networks that support social life. Neuropsychologia 63: 235–248. Bishop SJ (2008). Neural mechanisms underlying selective attention to threat. Ann N Y Acad Sci 1129: 141–152. Blomstedt P, Naesstr€om M, Bodlund O (2017). Deep brain stimulation in the bed nucleus of the stria terminalis and medial forebrain bundle in a patient with major depressive disorder and anorexia nervosa. Clin Case Rep 5: 679. Bradley MM, Lang PJ (2000). Measuring emotion: behavior, feeling, and physiology. Cognitive neuroscience of emotion, Oxford University Press, New York, NY. Brinkmann L, Buff C, Feldker K et al. (2017a). Distinct phasic and sustained brain responses and connectivity of amygdala and bed nucleus of the stria terminalis during threat anticipation in panic disorder. Psychol Med 47: 2675–2688. Brinkmann L, Buff C, Neumeister P et al. (2017b). Dissociation between amygdala and bed nucleus of the stria terminalis during threat anticipation in female post-traumatic stress disorder patients. Hum Brain Mapp 38: 2190–2205. Brinkmann L, Buff C, Feldker K et al. (2018). Inter-individual differences in trait anxiety shape the functional connectivity between the bed nucleus of the stria terminalis and the amygdala during brief threat processing. Neuroimage 166: 110–116. Buff C, Brinkmann L, Bruchmann M et al. (2017). Activity alterations in the bed nucleus of the stria terminalis and amygdala during threat anticipation in generalized anxiety disorder. Soc Cogn Affect Neurosci 12: 1766–1774. Busti D, Geracitano R, Whittle N et al. (2011). Different fear states engage distinct networks within the intercalated cell clusters of the amygdala. J Neurosci 31: 5131–5144. Bzdok D, Laird AR, Zilles K et al. (2013). An investigation of the structural, connectional, and functional subspecialization in the human amygdala. Hum Brain Mapp 34: 3247–3266. Cain CK (2019). Avoidance problems reconsidered. Curr Opin Behav Sci 26: 9–17. Campese VD, Sears RM, Moscarello JM et al. (2016). The neural foundations of reaction and action in aversive motivation. Curr Top Behav Neurosci 27: 171–195. Ch’ng S, Fu J, Brown RM et al. (2018). The intersection of stress and reward: BNST modulation of aversive and appetitive states. Prog Neuropsychopharmacol Biol Psychiatry 87: 108–125. Chung WCJ, 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. Ciocchi S, Herry C, Grenier F et al. (2010). Encoding of conditioned fear in central amygdala inhibitory circuits. Nature 468: 277–282.

Clauss JA, Avery SN, Benningfield MM et al. (2019). Social anxiety is associated with BNST response to unpredictability. Depress Anxiety 36: 666–675. Crestani CC, Alves FH, Gomes FV et al. (2013). Mechanisms in the bed nucleus of the stria terminalis involved in control of autonomic and neuroendocrine functions: a review. Curr Neuropharmacol 11: 141–159. Davis M, Whalen PJ (2001). The amygdala: vigilance and emotion. Mol Psychiatry 6: 13–34. Davis M, Gendelman DS, Tischler MD et al. (1982). A primary acoustic startle circuit: lesion and stimulation studies. J Neurosci 2: 791–805. Davis M, Walker DL, Miles L et al. (2010). Phasic vs sustained fear in rats and humans: role of the extended amygdala in fear vs anxiety. Neuropsychopharmacology 35: 105–135. de Gelder B, Terburg D, Morgan B et al. (2014). The role of human basolateral amygdala in ambiguous social threat perception. Cortex 52: 28–34. de la Mora MP, Gallegos-Cari A, Arizmendi-Garcia Y et al. (2010). Role of dopamine receptor mechanisms in the amygdaloid modulation of fear and anxiety: structural and functional analysis. Prog Neurobiol 90: 198–216. de Voogd LD, Klumpers F, Fernandez G et al. (2017). Intrinsic functional connectivity between amygdala and hippocampus during rest predicts enhanced memory under stress. Psychoneuroendocrinology 75: 192–202. Delgado MR, Jou RL, Phelps EA (2011). Neural systems underlying aversive conditioning in humans with primary and secondary reinforcers. Front Neurosci 5: 71. Dumont EC (2009). What is the bed nucleus of the stria terminalis? Prog Neuropsychopharmacol Biol Psychiatry 33: 1289–1290. Duque-Wilckens N, Steinman MQ, Busnelli M et al. (2018). Oxytocin receptors in the anteromedial bed nucleus of the Stria terminalis promote stress-induced social avoidance in female California mice. Biol Psychiatry 83: 203–213. Fadok JP, Krabbe S, Markovic M et al. (2017). A competitive inhibitory circuit for selection of active and passive fear responses. Nature 542: 96–100. Fanselow MS, Pennington ZT (2017). The danger of LeDoux and Pine’s two-system framework for fear. Am J Psychiatry 174: 1120–1121. Fanselow MS, Hoffman AN, Zhuravka I (2019). Timing and the transition between modes in the defensive behavior system. Behav Processes 166: 103890. Feinstein JS, Adolphs R, Damasio A et al. (2011). The human amygdala and the induction and experience of fear. Curr Biol 21: 34–38. Figel B, Brinkmann L, Buff C et al. (2019). Phasic amygdala and BNST activation during the anticipation of temporally unpredictable social observation in social anxiety disorder patients. Neuroimage Clin 22: 101735. Flook EA, Feola B, Avery SN et al. (2020). BNST-insula structural connectivity in humans. Neuroimage 210: 116555. Fox AS, Shackman AJ (2019). The central extended amygdala in fear and anxiety: closing the gap between mechanistic and neuroimaging research. Neurosci Lett 693: 58–67. Fox AS, Oler JA, Birn RM et al. (2018). Functional connectivity within the primate extended amygdala is heritable and

ROLES OF THE BNST AND AMYGDALA IN FEAR REACTIONS associated with early-life anxious temperament. J Neurosci 38: 7611–7621. Giardino WJ, Eban-Rothschild A, Christoffel DJ et al. (2018). Parallel circuits from the bed nuclei of stria terminalis to the lateral hypothalamus drive opposing emotional states. Nat Neurosci 21: 1084–1095. Giustino TF, Ramanathan KR, Totty MS et al. (2020). Locus coeruleus norepinephrine drives stress-induced increases in basolateral amygdala firing and impairs extinction learning. J Neurosci 40: 907–916. Gladwin TE, Hashemi MM, van Ast V et al. (2016). Ready and waiting: freezing as active action preparation under threat. Neurosci Lett 619: 182–188. Goode TD, Maren S (2017). Role of the bed nucleus of the stria terminalis in aversive learning and memory. Learn Mem 24: 480–491. Gorka AX, Torrisi S, Shackman AJ et al. (2018). Intrinsic functional connectivity of the central nucleus of the amygdala and bed nucleus of the stria terminalis. Neuroimage 168: 392–402. Gozzi A, Jain A, Giovannelli A et al. (2010). A neural switch for active and passive fear. Neuron 67: 656–666. Gregoriou GC, Kissiwaa SA, Patel SD et al. (2019). Dopamine and opioids inhibit synaptic outputs of the main island of the intercalated neurons of the amygdala. Eur J Neurosci 50: 2065–2074. Grupe DW, Oathes DJ, Nitschke JB (2013). Dissecting the anticipation of aversion reveals dissociable neural networks. Cereb Cortex 23: 1874–1883. Hamada T, McLean WH, Ramsay M et al. (2002). Lipoid proteinosis maps to 1q21 and is caused by mutations in the extracellular matrix protein 1 gene (ECM1). Hum Mol Genet 11: 833–840. Hashemi MM, Gladwin TE, de Valk NM et al. (2019). Neural dynamics of shooting decisions and the switch from freeze to fight. Sci Rep 9: 4240. Haubensak W, Kunwar PS, Cai H et al. (2010). Genetic dissection of an amygdala microcircuit that gates conditioned fear. Nature 468: 270–276. Hermans EJ, Henckens MJ, Roelofs K et al. (2013). Fear bradycardia and activation of the human periaqueductal grey. Neuroimage 66: 278–287. Hu P, Maita I, Kwok C et al. (2020). Chronic stress induces anxiety-associated negative valence behavior by activating a corticotropin-releasing hormone-related molecular signature in the oval bed nucleus of the stria terminalis. J Neurosci 40: 2519–2537. Inman CS, Bijanki KR, Bass DI et al. (2018). Human amygdala stimulation effects on emotion physiology and emotional experience. Neuropsychologia 145: 106722. Isaac O, Thiemer K (1975). Biochemical studies on camomile components/III. In vitro studies about the antipeptic activity of (–)-alpha-bisabolol (author’s transl). Arzneimittelforschung 25: 1352–1354. Janak PH, Tye KM (2015). From circuits to behaviour in the amygdala. Nature 517: 284–292. Jennings JH, Sparta DR, Stamatakis AM et al. (2013). Distinct extended amygdala circuits for divergent motivational states. Nature 496: 224–228.

431

Klumpers F, Kroes MCW (2019). Roles of the amygdala and basal forebrain in defense: a reply to Luyck et al. and implications for defensive action. Neuropsychol Rev 29: 186–189. Klumpers F, Morgan B, Terburg D et al. (2015). Impaired acquisition of classically conditioned fear-potentiated startle reflexes in humans with focal bilateral basolateral amygdala damage. Soc Cogn Affect Neurosci 10: 1161–1168. Klumpers F, Kroes MCW, Baas J et al. (2017). How human amygdala and bed nucleus of the stria terminalis may drive distinct defensive responses. J Neurosci 37: 38303816. Lebow MA, Chen A (2016). Overshadowed by the amygdala: the bed nucleus of the stria terminalis emerges as key to psychiatric disorders. Mol Psychiatry 21: 450–463. LeDoux JE, Brown R (2017). A higher-order theory of emotional consciousness. Proc Natl Acad Sci U S A 114: E2016–E2025. LeDoux J, Daw ND (2018). Surviving threats: neural circuit and computational implications of a new taxonomy of defensive behaviour. Nat Rev Neurosci 19: 269–282. LeDoux JE, Pine DS (2016). Using neuroscience to help understand fear and anxiety: a two-system framework. Am J Psychiatry 173: 1083–1093. LeDoux JE, Moscarello J, Sears R et al. (2017). The birth, death and resurrection of avoidance: a reconceptualization of a troubled paradigm. Mol Psychiatry 22: 24–36. Liddell BJ, Brown KJ, Kemp AH et al. (2005). A direct brainstem-amygdala-cortical ’alarm’ system for subliminal signals of fear. Neuroimage 24: 235–243. Luyten L, Hendrickx S, Raymaekers S et al. (2016). Electrical stimulation in the bed nucleus of the stria terminalis alleviates severe obsessive-compulsive disorder. Mol Psychiatry 21: 1272–1280. McLean CP, Asnaani A, Litz BT et al. (2011). Gender differences in anxiety disorders: prevalence, course of illness, comorbidity and burden of illness. J Psychiatr Res 45: 1027–1035. McMenamin BW, Langeslag SJ, Sirbu M et al. (2014). Network organization unfolds over time during periods of anxious anticipation. J Neurosci 34: 11261–11273. Megerman J, Murphy RA (1975). Myosin from arterial smooth muscle: isolation following actin depolymerization. Biochim Biophys Acta 412: 241–255. Miles L, Davis M, Walker D (2011). Phasic and sustained fear are pharmacologically dissociable in rats. Neuropsychopharmacology 36: 1563–1574. Miller SM, Marcotulli D, Shen A et al. (2019). Divergent medial amygdala projections regulate approach-avoidance conflict behavior. Nat Neurosci 22: 565–575. Mobbs D, Petrovic P, Marchant JL et al. (2007). When fear is near: threat imminence elicits prefrontal-periaqueductal gray shifts in humans. Science 317: 1079–1083. Mobbs D, Marchant JL, Hassabis D et al. (2009). From threat to fear: the neural organization of defensive fear systems in humans. J Neurosci 29: 12236–12243. Mobbs D, Adolphs R, Fanselow MS et al. (2019). Viewpoints: approaches to defining and investigating fear. Nat Neurosci 22: 1205–1216.

432

A.M. HULSMAN ET AL.

Mobbs D, Yu R, Rowe JB et al. (2010). Neural activity associated with monitoring the oscillating threat value of a tarantula. Proc Natl Acad Sci U S A 107: 20582–20586. M€ unsterk€otter AL, Notzon S, Redlich R et al. (2015). Spider or no spider? neural correlates of sustained and phasic fear in spider phobia. Depress Anxiety 32: 656–663. Naaz F, Knight LK, Depue BE (2019). Explicit and ambiguous threat processing: functionally dissociable roles of the amygdala and bed nucleus of the stria terminalis. J Cogn Neurosci 31: 543–559. Namburi P, Beyeler A, Yorozu S et al. (2015). A circuit mechanism for differentiating positive and negative associations. Nature 520: 675–678. Ohman A (2005). The role of the amygdala in human fear: automatic detection of threat. Psychoneuroendocrinology 30: 953–958. Pedersen WS, Muftuler LT, Larson CL (2019). A highresolution fMRI investigation of BNST and centromedial amygdala activity as a function of affective stimulus predictability, anticipation, and duration. Soc Cogn Affect Neurosci 14: 1167–1177. Phelps EA, LeDoux JE (2005). Contributions of the amygdala to emotion processing: from animal models to human behavior. Neuron 48: 175–187. Raymaekers S, Luyten L, Bervoets C et al. (2017). Deep brain stimulation for treatment-resistant major depressive disorder: a comparison of two targets and long-term follow-up. Transl Psychiatry 7: e1251. Reppucci CJ, Petrovich GD (2016). Organization of connections between the amygdala, medial prefrontal cortex, and lateral hypothalamus: a single and double retrograde tracing study in rats. Brain Struct Funct 221: 2937–2962. Schipper P, Hiemstra M, Bosch K et al. (2019). The association between serotonin transporter availability and the neural correlates of fear bradycardia. Proc Natl Acad Sci U S A 116: 25941–25947. Schlund MW, Hudgins CD, Magee S et al. (2013). Neuroimaging the temporal dynamics of human avoidance to sustained threat. Behav Brain Res 257: 148–155. Shackman AJ, Fox AS (2016). Contributions of the central extended amygdala to fear and anxiety. J Neurosci 36: 8050–8063. Sierra H, Cordova M, Chen CJ et al. (2015). Confocal imagingguided laser ablation of basal cell carcinomas: an ex vivo study. J Invest Dermatol 135: 612–615. Somerville LH, Whalen PJ, Kelley WM (2010). Human bed nucleus of the stria terminalis indexes hypervigilant threat monitoring. Biol Psychiatry 68: 416–424. Somerville LH, Wagner DD, Wig GS et al. (2013). Interactions between transient and sustained neural signals support the generation and regulation of anxious emotion. Cereb Cortex 23: 49–60. Straube T, Mentzel HJ, Miltner WHR (2007). Waiting for spiders: brain activation during anticipatory anxiety in spider phobics. Neuroimage 37: 1427–1436.

Sullivan GM, Apergis J, Bush DEA et al. (2004). Lesions in the bed nucleus of the stria terminalis disrupt corticosterone and freezing responses elicited by a contextual but not by a specific cue-conditioned fear stimulus. Neuroscience 128: 7–14. Terburg D, Morgan BE, Montoya ER et al. (2012). Hypervigilance for fear after basolateral amygdala damage in humans. Transl Psychiatry 2: e115. Terburg D, Scheggia D, Triana Del Rio R et al. (2018). The basolateral amygdala is essential for rapid escape: a human and rodent study. Cell 175: 723–735.e716. Thornton HB, Nel D, Thornton D et al. (2008). The neuropsychiatry and neuropsychology of lipoid proteinosis. J Neuropsychiatry Clin Neurosci 20: 86–92. Tovote P, Esposito MS, Botta P et al. (2016). Midbrain circuits for defensive behaviour. Nature 534: 206–212. Tye KM, Prakash R, Kim SY et al. (2011). Amygdala circuitry mediating reversible and bidirectional control of anxiety. Nature 471: 358–362. van Honk J, Terburg D, Thornton H et al. (2016). Consequences of selective bilateral lesions to the basolateral amygdala in humans. In: Living without an amygdala, The Guilford Press, New York, NY. Van Hougenhouck-Tulleken W, Chan I, Hamada T et al. (2004). Clinical and molecular characterization of lipoid proteinosis in Namaqualand, South Africa. Br J Dermatol 151: 413–423. Walker DL, Toufexis DJ, Davis M (2003). Role of the bed nucleus of the stria terminalis versus the amygdala in fear, stress, and anxiety. Eur J Pharmacol 463: 199–216. Walker DL, Miles LA, Davis M (2009). Selective participation of the bed nucleus of the stria terminalis and CRF in sustained anxiety-like versus phasic fear-like responses. Prog Neuropsychopharmacol Biol Psychiatry 33: 1291–1308. Walter A, Mai JK, Lanta L et al. (1991). Differential distribution of immunohistochemical markers in the bed nucleus of the stria terminalis in the human brain. J Chem Neuroanat 4: 281–298. Whalen PJ, Rauch SL, Etcoff NL et al. (1998). Masked presentations of emotional facial expressions modulate amygdala activity without explicit knowledge. J Neurosci 18: 411–418. Whalen PJ, Kagan J, Cook RG et al. (2004). Human amygdala responsivity to masked fearful eye whites. Science 306: 2061. Yamauchi N, Takahashi D, Sugimura YK et al. (2018). Activation of the neural pathway from the dorsolateral bed nucleus of the stria terminalis to the central amygdala induces anxiety-like behaviors. Eur J Neurosci 48: 3052–3061. Yassa MA, Hazlett RL, Stark CE et al. (2012). Functional MRI of the amygdala and bed nucleus of the stria terminalis during conditions of uncertainty in generalized anxiety 2disorder. J Psychiatr Res 46: 1045–1052. 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.

Section 5 Preoptic area

This page intentionally left blank

Handbook of Clinical Neurology, Vol. 179 (3rd series) The Human Hypothalamus: Anterior Region D.F. Swaab, F. Kreier, P.J. Lucassen, A. Salehi, and R.M. Buijs, Editors https://doi.org/10.1016/B978-0-12-819975-6.00028-5 Copyright © 2021 Elsevier B.V. All rights reserved

Chapter 29

The median preoptic nucleus: A major regulator of fluid, temperature, sleep, and cardiovascular homeostasis MICHAEL J. MCKINLEY1,2*, GLENN L. PENNINGTON1, AND PHILIP J. RYAN1 1

Florey Institute of Neuroscience and Mental Health, University of Melbourne, Parkville, VIC, Australia 2

Department of Anatomy and Physiology, University of Melbourne, Parkville, VIC, Australia

Abstract Located in the midline lamina terminalis of the anterior wall of the third ventricle, the median preoptic nucleus is a thin elongated nucleus stretching around the rostral border of the anterior commissure. Its neuronal elements, composed of various types of excitatory glutamatergic and inhibitory GABAergic neurons, receive afferent neural signals from (1) neighboring subfornical organ and organum vasculosum of the lamina terminalis related to plasma osmolality and hormone concentrations, e.g., angiotensin II; (2) from peripheral sensors such as arterial baroreceptors and cutaneous thermosensors. Different sets of these MnPO glutamatergic and GABAergic neurons relay output signals to hypothalamic, midbrain, and medullary regions that drive homeostatic effector responses. Included in the effector responses are (1) thirst, antidiuretic hormone secretion and renal sodium excretion that subserve osmoregulation and body fluid homeostasis; (2) vasoconstriction or dilatation of skin blood vessels, and shivering and brown adipose tissue thermogenesis for core temperature homeostasis; (3) inhibition of hypothalamic and midbrain nuclei that stimulate wakefulness and arousal, thereby promoting both REM and non-REM sleep; and (4) activation of sympathetic pathways that drive vasoconstriction and heart rate to maintain arterial pressure and the perfusion of vital organs. The small size of MnPO belies its massive homeostatic significance.

INTRODUCTION The preoptic region of the brain is located rostral to the hypothalamus and is properly classified as part of the telencephalon medium (Le Gros-Clark et al., 1938). However, its close relationship, both functionally and morphologically, to the anterior hypothalamus has provided justification for inclusion of the preoptic region in discussions of hypothalamic function. In this regard, the median preoptic nucleus (MnPO), while relatively small in size, has several essential homeostatic roles. In particular, it has a crucial role integrating neural signals from osmoreceptors and hormonal actions on the brain to regulate thirst and antidiuretic hormone

secretion, thereby coordinating body fluid homeostasis. It is also a relay station for incoming signals from peripheral thermosensors to be relayed to downstream thermoregulatory effector regions in the hypothalamus and brainstem. Furthermore, the MnPO influences cardiovascular function by integrating hindbrain signals from arterial baroreceptors with signals from forebrain circumventricular organs (CVOs) responding to bloodborne hormones, then relaying this neural information to the hypothalamus and brainstem to regulate sympathetic outflow. Last, but not least, the MnPO participates in regulating sleep. Furthermore, the aforementioned homeostatic mechanisms also come under the influence

*Correspondence to: Michael J. McKinley, Florey Institute of Neuroscience and Mental Health, University of Melbourne, Parkville, VIC 3010, Australia. Tel: +61-3-8344-7332, E-mail: [email protected]

436

M.J. MCKINLEY ET AL.

of signals (integrated at the level of the MnPO) related to circadian rhythm, reproductive cycle and energy balance coming from hypothalamic sites such as the suprachiasmatic and arcuate nuclei (Guzmán-Ruiz et al., 2015; Krajewski-Hall et al., 2019). The MnPO is the most anterior part of the preoptichypothalamic region, developing from the rostral pole of the unevaginated neural tube. In mammals, including primates, rodents, lagomorphs, ruminants, and marsupials that have been studied (Loo, 1931; Humphrey, 1936; Allen et al., 1999), the MnPO is a thin elongated midline sliver of gray matter. Except for the portion coursing over the rostral boundary of the anterior commissure, it occupies the midline anterior wall of the third ventricle (the lamina terminalis) immediately dorsal to the organum vasculosum of the lamina terminalis (OVLT) and ventral to the subfornical organ. Most atlases of the human brain do not identify the MnPO; however, the human MnPO is approximately 10 mm from top to bottom, spreading 1 mm on either side of the midline. In horizontal section its width is approximately 1 mm (Fig. 29.1). The MnPO is comprised of many small to mediumsized neurons dispersed along its length as well as unmyelinated fibers running along its longitudinal axis. There has been little study of the human MnPO; therefore we

are reliant on studies mainly but not exclusively, in rats and mice for details and properties of cells, neural connections, fiber tracts, and functional correlates found within the MnPO.

NEUROCHEMICAL PROPERTIES OF MnPO NEURONS The MnPO contains many small to medium-sized densely packed neurons throughout its extent. Based on in situ hybridization methods and immunohistochemistry in rats (Grob et al., 2003) or GFP-expressing reporter mice crossed with either VGluT2-ires-Cre or VGAT-iresCre mice (Abbott et al., 2016), the distribution within the MnPO of excitatory glutamatergic (MnPOGlut neurons) or inhibitory GABAergic neurons (MnPOGABA neurons) was investigated. Those studies show in both rats and mice that there are two major populations of neurons, one glutamatergic and the other GABAergic, dispersed throughout the MnPO, which may be categorized further into subgroups. In rats and mice, many glutamatergic neurons also express the Agtr1a gene that encodes the angiotensin AT1a receptor (Grob et al., 2003; Leib et al., 2017). Some other genes that may be differentially expressed on specific subsets of MnPOGlut neurons are Nxph4

Fig. 29.1. The human median preoptic nucleus (MnPO). (A) Mid-sagittal section at the level of the lamina terminalis. (B) Coronal section through the lamina terminalis. MnPO is lightly stained and indicated by the arrows in both panels. Abbreviations. Ac, anterior commissure; cp, choroid plexus; f, fornix; ic, internal capsule; inf, infundibulum; oc, optic chiasma; pc, prechiasmatic cisterna; 3V, third ventricle. Stain: luxol fast blue/cresyl violet. Magnification bar ¼ 0.5 cm.

THE MEDIAN PREOPTIC NUCLEUS (encoding neurexophilin-4), Adcyap1, (encoding adenylate cyclase activating peptide 1), Etv1 (ETS translocation variant 1, a member of the E-twenty six family of transcription factors), CaMKIIa, and nNOS (neuronal nitric oxide synthase), as well as Slc17a6 (vesicular glutamate transporter 2). Different subsets of these MnPOGlut neurons regulate body fluid homeostasis, thermoregulation, sleep or arterial pressure (Abbott et al., 2017; Allen et al., 2017; Leib et al., 2017; Augustine et al., 2018; Harding et al., 2018). In regard to angiotensin receptors, angiotensin binding sites, i.e., AT1 receptors in mammals other than rats and mice (which express a subgroup termed AT1a) are concentrated on neurons throughout the MnPO of all species that have been investigated, including humans (Song et al., 1992; Lenkei et al., 1997; Allen et al., 1999). The distribution of angiotensin binding sites in the human MnPO is shown in Fig. 29.2. The neurokinin 3 receptor is expressed on a subset of MnPOGlut neurons, which may play a crucial role in thermoregulatory menopausal hot flushes (Padilla et al., 2018; Krajewski-Hall et al., 2019). Another group of MnPOGlut neurons express leptin receptors (Yu et al., 2016). Gonadotropin releasing hormone is expressed in MnPO mainly by MnPOGlut neurons. A group of glutamatergic neurons also express both nNOS and the estrogen receptor ERa (Chachlaki et al., 2017). Other peptides that have been detected by immunohistochemistry within some MnPO neurons are neurotensin, met-enkephalin, cholecystokinin and luteinizing hormone releasing hormone, and to a lesser extent substance P and somatostatin (Kawano et al., 1989; Kawano and Masuko 1992). Glp1r (glucagon-like peptide-1-receptor) is coexpressed by inhibitory MnPOGABA neurons as is Vgat1 (vesicular GABA transporter), but not by excitatory MnPOGlut2 neurons (Augustine et al., 2018). Prostaglandin receptors, EP1, EP3, and EP4 are expressed

437

in MnPO neurons (Nakamura et al., 2000; Oka et al., 2000). The EP3 receptor is likely to be expressed in inhibitory MnPOGABA neurons (Nakamura et al., 2002). Prostaglandin E2 may act at the EP3 receptor to inhibit neuronal activity in the MnPO, thereby disinhibiting febrile mechanisms (Oka et al., 2000). Evidence from microinjection studies shows that there are adenosine AT2A receptors in the MnPO that may mediate sleep influences on pain (Hambrecht-Wiedbusch et al., 2017). Using some of the previously genetic markers for excitatory glutamatergic and inhibitory GABAergic neurons, several Cre mouse lines with the Cre recombinase enzyme gene under the control of promoters for these markers have been developed. To selectively express genes within Cre-expressing neurons, such mice can be stereotactically injected with Cre-dependent viruses, which express a variety of genes (Lowell 2019). They include channel rhodopsin 2, an excitatory optogenetic receptor that responds to blue light (Betley and Sternson 2011), and GCaMP6, a genetically encoded calcium indicator, which fluoresces with increased neuronal calcium concentration (Chen et al., 2013). Several of the neuroanatomical and functional studies described later in this article have utilized injections of Cre-dependent viruses into the MnPO of mice to enable selective stimulation or inhibition of excitatory or inhibitory MnPO neurons, to trace neural pathways, and to monitor neuronal activation by calcium imaging (Lowell, 2019).

NEURAL CONNECTIVITY Afferent neural connections The MnPO receives neural input from several fore-, mid-, and hind-brain sites that send information to the MnPO from peripheral and central sensors that detect changes in body fluid and electrolyte levels, circulating

Fig. 29.2. Autoradiographs of binding of a radiolabeled angiotensin II analogue, Sar1-Ile8-angiotensin II in the lamina terminalis. This radioligand binding represents human AT1 receptors (Allen et al., 1999). White areas indicate high levels of radioligand binding in different parts of the lamina terminalis. (A) OVLT. (B) OVLT and Rostral MnPO. (C) Caudal MnPO. Abbreviations as in Fig. 29.1. PVH, hypothalamic paraventricular nucleus. Magnification bar ¼ 1 cm.

438

M.J. MCKINLEY ET AL. Table 29.1 Afferent neural inputs of the median preoptic nucleus traced in rats and mice Brain region Preoptic Anterior and periventricular preoptic n. OVLT Subfornical organ Ventrolateral preoptic a. Ventromedial preoptic n. Hypothalamus Arcuate n. Dorsomedial n. Lateral hypothalamic a. Paraventricular n (parvo and magno) Suprachiasmatic n. Thalamus Paraventricular n of thalamus Midbrain Lateral parabrachial n. Locus coeruleus Raph
e pontis Medulla oblongata Nucleus of the solitary tract (medial, commissural) Caudal ventrolateral medulla (A1 group) Raph
e pallidus Rostral ventrolateral medulla (C1 group)

Functions

References

VP, thirst, cardiovascular VP, thirst, cardiovascular Sleep

10 4, 12 4, 5, 8, 10 10 4

Thermoregulation, flushes Thermoregulation Thirst, sleep VP, thirst, cardiovascular Thermoregulation

1, 2, 3, 9 4, 10 4 1, 10 2

Thirst

1

Thermoregulation, thirst Arousal

4, 6, 7, 10 10 10

Cardiovascular VP secretion

4, 10, 11 10, 11 4 10, 11

Cardiovascular

References: (1) Allen et al. (2017); (2) Guzma´n-Ruiz et al. (2015); (3) Kawano and Masuko (2000); (4) Leib et al. (2017); (5) Miselis et al. (1979); (6) Nakamura and Morrison (2008); (7) Nakamura and Morrison (2010); (8) Oka et al. (2015); (9) Rance et al. (2013); (10) Saper and Levisohn (1983); (11) Saper and Levisohn (1983); (12) Zimmerman et al. (2016). Abbreviations: a., area; n., nucleus; VP, vasopressin secretion; OVLT, organum vasculosum of the lamina terminalis.

hormones, blood pressure, skin or core temperature, or sleep status. These afferent neural connections are summarized in Table 29.1 and are briefly outlined here. In the forebrain lamina terminalis, two circumventricular organs, the subfornical organ and OVLT both provide excitatory glutamatergic neural input to the entire MnPO from neurons that detect changes in plasma tonicity or the circulating concentrations of hormones such as angiotensin II. As well, there is inhibitory innervation of MnPO neurons by GABAergic neurons originating in the subfornical organ and OVLT (Oka et al., 2015; Augustine et al., 2018). Other nearby preoptic sites that provide afferent neural connections to MnPO are the ventromedial preoptic nucleus (Leib et al., 2017) that may influence thermoregulation and ventrolateral and periventricular preoptic areas (Saper and Levisohn, 1983) that are related to sleep homeostasis. Within the hypothalamus, most parvocellular divisions of the hypothalamic paraventricular nucleus provide afferent connections to MnPO neurons (Saper and Levisohn, 1983). Vasopressinergic neurons within the suprachiasmatic nucleus and a-melanocyte-stimulating

hormone (a-MSH) containing neurons of the arcuate nucleus provide afferent input to GABAergic neurons within the MnPO that influence thermoregulation (Guzmán-Ruiz et al., 2015). Neurons in the dorsomedial nucleus and lateral hypothalamic area also send projections to the MnPO (Saper and Levisohn, 1983; Rance et al., 2013; Leib et al., 2017). In the midbrain, several subdivisions of the lateral parabrachial nucleus that include, dorsal, external lateral, and Kolliker-Fuse subnuclei send efferent projections into the MnPO (Saper and Levisohn, 1983; Nakamura and Morrison, 2008; Nakamura and Morrison, 2010). The dorsal subnucleus relays signals from warmth sensors in the skin and the external lateral subnucleus relays signals from cold sensors in the skin to the MnPO (Morrison 2016). Other signals related to fluid balance and blood pressure are relayed from medullary regions to MnPO. Direct neural inputs from the hindbrain nucleus of the solitary tract (NTS) and from A1 cells in the rostroventrolateral medulla and caudal ventrolateral medulla have been identified (Saper and Levisohn, 1983).

THE MEDIAN PREOPTIC NUCLEUS

Efferent neural connections The efferent neural pathways that emanate from neurons in the MnPO have been mapped in rats (Miselis et al., 1979; Saper and Levisohn, 1983; Gu and Simerly, 1997; Uschakov et al., 2007) and mice (Abbott et al., 2016; Allen et al., 2017; Leib et al., 2017). They terminate mainly in adjacent lamina terminalis and nearby

439

preoptic and hypothalamic sites. Terminal fields are also observed in a few thalamic, midbrain, and medullary sites. Efferent connections are summarized in Table 29.2. Within the lamina terminalis, MnPO neurons send axonal connections that form synapses with neurons in the subfornical organ and OVLT (Miselis et al., 1979; Saper and Levisohn, 1983; Kawano, 2017), thereby

Table 29.2 Efferent neural connections of the median preoptic nucleus in rats and mice Brain region Preoptic region Anteroventral preoptic n. Dorsolateral preoptic a. Medial preoptic area OVLTa Parastrial n. Subfornical organa Suprachiasmatic n. Ventrolateral preoptic n.a Hypothalamus Arcuate n. Dorsomedial n.a Lateral hypothalamic a.a Median eminence Paraventricular n (parvo and magno)a Perifornical areaa Periventricular n. Posterior hypothalamic n. Premamillary n. Supraoptic n.a Tuberomamillary n. Thalamus Paraventricular n. of thalamusa Pedunculopontine tegmental n. Reuniens n. Xiphoid n. Midbrain Barrington’s n. Dorsal raphea Lateral parabrachial n. Locus coeruleusa Median raph
e Periaqueductal graya Medulla oblongata Nucleus of the solitary tract Parapyramidal region Raph
e obscurus Raph
e pallidusa a

Excitatory inhibitory

Functions

References

Thermoregulation

5, 7 1, 7 8 5, 7,11 7 1, 5, 7, 11 7 8, 10

E and I E and I

Thermoregulation, sleep VP, thirst, cardiovasc.

E and I E and I E and I

VP, thirst, cardiovasc. Circadian rhythm Sleep

E and I E and I E and I E and i E and I E and I E and I

Hot flushes Thermoregulation Thirst, sleep, cardiovasc. VP, thirst, cardiovasc. Arousal, sleep Sleep

E and I E and I

Vasopressin Sleep

1, 5 1, 5, 7 1, 2, 5, 7 1 1, 2, 4, 5, 11 1, 8, 11 1, 7 1 1 1, 2, 5 1, 7

E and I E and I E and I E and I

Thirst

1, 2, 5, 7 1 1, 7 1

E and I I and e E and I E and I I and e E and I

Micturition Fluid, arousal, sleep Thirst, Thermoregulation Arousal and sleep Arousal and sleep Cardiovascular

1 1, 7, 8 1, 7 1, 8 1 1, 7, 8, 9

E and i E and i E and i E and i

Cardiovascular, fluid

1 1 1 1

Thermoregulation

The projection was confirmed by retrograde tracing techniques. References: (1) Abbott et al. (2016); (2) Allen et al. (2017); (3) Gu and Simerly (1997); (4) Kawano and Masuko (1992); (5) Leib et al. (2017); (6) Miselis et al. (1979); (7) Thompson and Swanson (2003) (8) Uschakov et al. (2007); (9) Uschakov et al. (2007); (10) Walter et al. (2019); (11) Zardetto-Smith et al. (1992). Abbreviation: a., area; n., nucleus; Cardiovasc., cardiovascular; VP, arginine vasopressin. Upper case E and I denote substantial excitatory or inhibitory projections respectively, while lower case e and i denote minor glutamatergic and GABAergic projections respectively.

440

M.J. MCKINLEY ET AL.

providing reciprocal connectivity between these closely linked sites. Such efferent links may be excitatory or inhibitory and influence thirst and arterial blood pressure. There are also intranuclear neural connections within the MnPO (Thompson and Swanson, 2003; McKinley et al., 2015). Elsewhere in the preoptic region, the parastrial, suprachiasmatic, and ventrolateral preoptic nuclei receive inputs from the MnPO; the latter influencing sleep mechanisms (Walter et al., 2019). Several major efferent projections of MnPO neurons are observed in the hypothalamus (Table 29.2). These include links to magnocellular neurons in the supraoptic and hypothalamic paraventricular (PVH) nuclei that regulate the secretion of antidiuretic hormone from the posterior pituitary (McKinley et al., 2004) to the perifornical region of the lateral hypothalamus that may influence sleep–wake status and thirst (Uschakov et al., 2007; Leib et al., 2017), and to parvocellular parts of the PVH that influence the sympathetic nervous system and secretion of hormones from the anterior pituitary gland. As well, efferent connections to the tuberomamillary nucleus (Abbott et al., 2016) may have a role in regulating sleep (Chung et al., 2017). The major thalamic target of MnPO efferent projections is the paraventricular nucleus of the thalamus (PVT), which receives both excitatory glutamatergic and inhibitory GABAergic projections. Optogenetic stimulation of the glutamatergic projection to the PVT has been shown to drive water intake in mice (Allen et al., 2017; Leib et al., 2017). Within the midbrain, MnPO efferent projections to the dorsal and median raph
e may influence thirst and fluid intake (Zardetto-Smith et al., 1992). There are also projections to several subdivisions of the periaqueductal gray (Thompson et al., 2003; Uschakov et al., 2009; Abbott et al., 2016) that may affect sleep, thermoregulation, and cardiovascular regulation. In the hindbrain, both direct and relayed efferent projections from MnPO to rostral medial medulla including raph
e pallidus nucleus control thermoregulatory changes in skin blood flow, shivering, and brown adipose tissue (Tanaka et al., 2011; Morrison, 2016). The raph
e pallidus nucleus together with PVH and lateral hypothalamic area are sites of sympathetic premotor neurons. MnPO neurons are labeled following injection of trans-synaptic retrogradely transported neurotropic virus injected into various peripheral target organs such as the adrenal medulla (Westerhaus and Loewy, 1999), kidney (Sly et al., 1999), and brown adipose tissue (Oldfield et al., 2002). It is likely that MnPO efferent terminals in raph
e pallidus, PVH, and lateral hypothalamic area are connecting to sympathetic pathways regulating cardiovascular, body fluid, and temperature regulation. However, transneuronal viral tracing back to MnPO

following injection of pseudorabies virus into the submandibular and submaxillary gland (Hubschle et al., 1998) has been shown to use parasympathetic pathway (Hettogoda et al., 2015).

Collaterals Several efferent connections of the MnPO collateralize to more than one target. Weiss and Hatton (1990) demonstrated collateralization of MnPO axons supplying PVH and supraoptic nuclei. They suggested that such collateral fibers may coordinate hormone release from magnocellular neurons within these two hypothalamic nuclei. Using microinjections of two different retrogradely transported tracer molecules, one injected into the subfornical organ, the other into the supraoptic nucleus, evidence was obtained of branching efferent axons from some neurons in the MnPO to the subfornical organ and supraoptic nucleus in rats (Oldfield et al., 1992). Such an arrangement suggests common signals from MnPO neurons could coordinate fluid balance by influencing both neuroendocrine secretion from the supraoptic nucleus and thirst via the subfornical organ. Previously mentioned intranuclear connections within MnPO may also have a coordinating role. Using a similar strategy of microinjections of different tracers into three different diencephalic sites in mice, Allen et al. (2017) showed that MnPO neurons could have collateral inputs to two or three of the following sites: the lateral hypothalamic area, PVH, and PVT.

BODY FLUID HOMEOSTASIS The MnPO plays a pivotal role in the maintenance of body fluid volume and composition by regulating both the intake and excretion of body water. The initial view of this nucleus was that it integrated signals coming from sensors and receptors located in the subfornical organ, OVLT, and brain stem, then relayed this information via excitatory pathways to various effector sites involved in the generation of thirst and secretion of vasopressin. This is now considered to be an oversimplification, and current opinion suggests that the MnPO provides both excitatory and inhibitory output pathways that regulate both the stimulation and quenching of thirst and water drinking as well as driving or inhibiting ADH secretion to regulate urine output.

Thirst Mangiapane et al. (1983) reported that electrolytic ablation of the MnPO caused disruption of water intake in response to dipsogenic stimuli in rats. Since then, more evidence has accrued showing that this nucleus has a major role in the regulation of thirst. Ablation of neuronal

THE MEDIAN PREOPTIC NUCLEUS cell bodies within the MnPO by means of injection of neurotoxic agents kainic acid or ibotenic acid into it also reduces drinking in rats, showing that the initial results obtained with electrolytic lesions studies were not due to destruction of fibers of coursing through the MnPO (Jones, 1988; Cunningham et al., 1991; Cunningham et al., 1992); rather, neurons had been damaged within the MnPO that were part of a neural pathway driving thirst that had been damaged. More recently, chemogenetic inhibition of MnPOGlut neurons in rats was shown to severely inhibit both osmotic and angiotensin-induced drinking (Marciante et al., 2019). Consistent with the idea that there were neurons within the MnPO that could be labeled “thirst neurons,” was electrophysiologic evidence that some MnPO neurons increased their activity in response to thirst-inducing stimuli. It was shown in anesthetized sheep that systemic hypertonicity produced by intracarotid infusion of hypertonic NaCl or mannitol solution increased the electrical firing of a population of MnPO neurons (McAllen et al., 1990). Relevant to this observation, ablation of the MnPO in sheep severely disrupted osmotically stimulated water drinking in this mammal (McKinley et al., 1999). Fos expression in neurons is an indicator of increased neuronal activity (Morgan and Curran, 1989). When conscious rats were made thirsty by either water deprivation (McKinley et al., 1994), systemic administration of hypertonic NaCl (Oldfield et al., 1991a,b,c), intravenous or intracerebroventricular (icv) angiotensin II (McKinley et al., 1995), or icv infusion of relaxin (McKinley et al., 1997), they exhibited intense Fos expression in the median preoptic nucleus (Fig. 29.3A), consistent with the idea that MnPO neurons participate in generating

441

thirst. If the subfornical organ and OVLT, the sensory circumventricular organs that provide afferent neural input to the MnPO, are ablated in combination in rats, then Fos expression in the MnPO in response to systemic hypertonicity is greatly reduced (Hochstenbach and Ciriello, 1996). In mice, specific ablation of MnPO neurons expressing nNOS caused a large suppression of water drinking that occurred in response to optogenetic (light) stimulation of either subfornical organ or OVLT neurons; acute inhibition of MnPOnNOS neurons also greatly attenuated water intake in dehydrated mice (Augustine et al., 2018). These results suggest that dipsogenic stimuli arising in the systemic circulation do not act directly on the MnPO; rather, osmotic and hormonal dipsogenic signals in the circulation act on the circumventricular organs because they lack a blood–brain barrier and possess the relevant sensors and hormonal receptors (McKinley et al., 2003); these then relay neural signals via the MnPO to other effector sites for the generation of thirst and subsequent ingestion of water. By contrast, sodium ions may act directly on neurons within the MnPO to influence thirst; Na sensors are located within the MnPO (Grob et al., 2004a,b) and Na sensors within the wall of the third ventricle had previously been proposed to regulate water drinking behavior (Andersson and Olsson, 1973; McKinley et al., 1974). For dipsogenic stimuli (e.g., angiotensin II, relaxin) delivered directly into the ventricular cerebrospinal fluid, a direct action on the MnPO is likely, because receptors for these two peptides are expressed by MnPO neurons, and there is direct access from cerebrospinal fluid to the interstitium of the MnPO. By contrast, when these peptides circulate in the bloodstream, they do not have access to the MnPO because of its blood–brain barrier;

Fig. 29.3. Coronal sections showing the distribution of activated neurons, indicated by Fos immunoreactivity (Fos-ir) in the MnPO of rats in response to (A) systemic hypertonicity (due to intraperitoneal administration of 0.6 M NaCl) or (B) a hot environment (exposure to 37°C for 1 h) to stimulate cutaneous thermoreceptors receptors. Note that Fos-ir (seen as brown dots) is more medially placed with a hypertonic stimulus compared with cutaneous heating, where the Fos-ir can be seen in more lateral parts of the rat MnPO. Arrows indicate the MnPO boundary. Counter stained with NADPH-diaphorase. Abbreviations as in Fig. 29.1. Magnification bar ¼ 100 mm.

442

M.J. MCKINLEY ET AL.

therefore their dipsogenic action is dependent on stimulating subfornical organ neurons which relay signals onwards in the brain to generate thirst (Simpson et al., 1978; Sunn et al., 2002). Finally, direct stimulation of neurons within the MnPO initiates water drinking within seconds of stimulation. Recent optogenetic approaches in which channel rhodopsin was expressed in glutamatergic neurons within the MnPO of mice showed that light stimulus to these neurons immediately initiated copious drinking (Abbott et al., 2016; Allen et al., 2017; Leib et al., 2017). Furthermore, such experiments also demonstrated that MnPOGlut neurons are involved in the unpleasant or aversive sensation (negative valence) that is associated with being thirsty, because mice would learn to press a lever or move to a place in their cage that would stop optogenetic stimulation of these thirstassociated neurons in the MnPO (Allen et al., 2017; Leib et al., 2017). The neural pathways downstream from the MnPO that drive thirst or water drinking are beginning to be unraveled. In particular, efferent pathways to the PVT, the PVH, and the lateral hypothalamic area appear to be involved. Optogenetic stimulation of nerve terminals within these three sites that originated from excitatory MnPO neurons resulted in drinking behavior in mice (Allen et al., 2017; Leib et al., 2017). Interestingly, the negative valence associated with photostimulation of MnPO neurons was only observed when their terminals were stimulated in the two hypothalamic sites, not those in the thalamic PVT (Leib et al., 2017). Evidence has also been obtained by using a chemogenetic approach that the efferent pathway from MnPO for osmotically stimulated drinking may differ from that for angiotensin II-induced drinking; the former response being relayed via the lateral hypothalamic area while the latter via the PVH (Marciante et al., 2019). In regard to cortical regions of the brain that may participate in the generation of the conscious sensation of thirst, pseudorabies virus injected into the either insular or cingulate cortex of rats is transported retrogradely back to the MnPO through several synaptic connections that include the paraventricular nucleus of the thalamus (Hollis et al., 2008). In regard to regulation of thirst in humans, the only species where a subjective report of actual thirst is possible, functional imaging studies utilizing positron and functional magnetic resonance imaging have identified several cortical area that were activated in thirsty human subjects. These areas included parts of the cingulate and insular cortices (Denton et al., 1999; Egan et al., 2003). As well, the lamina terminalis containing the MnPO was also activated in some subjects (Egan et al., 2003). Further analysis in thirsty human subjects revealed a functional connectivity between the ventral lamina terminalis,

which included MnPO, and insula and cingulate regions that were not present in the absence of thirst (Farrell et al., 2011).

Inhibition of thirst Optogenetic stimulation of inhibitory GABAergic neurons in the MnPO suppresses water drinking by mice that had been previously denied access to water (Abbott et al., 2016; Augustine et al., 2018; Zimmerman et al., 2019), suggesting that they play a key modulatory role in the satiation of thirst following ingestion of water. Calcium imaging studies reveal that these neurons are activated while water is being ingested and deactivated when drinking stops. They are also activated by ingestion of other fluids (oil, sucrose solution) but not solids such as peanut butter; the distinction between eating and drinking apparently depends on the speed of ingestion, i.e., they are highly activated during periods of concentrated, rapid ingestion compared to sparse, slower ingestion (Augustine et al., 2018). Inhibitory MnPOGABA neurons express Glp1r gene that can be used as a label for them. MnPOGlp1r neurons send efferent projections to excitatory SFOnNOS neurons, transmitting an inhibitory signal which is time-locked to drinking (Augustine et al., 2018). Calcium imaging of MnPOGlp1r neurons demonstrate variable responses for individual, spatially intermingled MnPOGlp1r neurons with drinking activating 28%, but inhibiting 36% of these neurons. Smaller subsets of MnPOGlp1r neurons show activity correlated with access to water alone (even before drinking) or by intragastric infusion of water or hypertonic saline (Zimmerman et al., 2019). How these different groups of inhibitory MnPOGlp1r neurons fit into the homeostatic regulation of thirst remains to be determined.

Antidiuretic hormone secretion The antidiuretic hormone, vasopressin, is synthesized within magnocellular neurons of the hypothalamic supraoptic and paraventricular nuclei. Axons of these magnocellular neurons project, via the median eminence, to the posterior pituitary gland from where vasopressin is secreted into the systemic blood stream. In the kidney, circulating vasopressin acts on distal tubules and collecting duct epithelium to open water channels, causing reabsorption of the glomerular filtrate, thereby concentrating urine and reducing its output. Transynaptic neural tracing from injections of pseudorabies virus into the posterior pituitary gland shows that pseudorabies is retrogradely transported from the posterior pituitary gland back to the MnPO as well as to the adjacent OVLT and subfornical organ in the lamina terminalis (McKinley et al., 2004). Consistent with this

THE MEDIAN PREOPTIC NUCLEUS observation are neuroanatomical tract tracing studies in rodents, outlined earlier in this chapter, showing that neurons within the MnPO send direct efferent projections to the hypothalamic supraoptic and paraventricular nuclei (Miselis et al., 1979; Saper and Levisohn, 1983). Importantly, it was also shown, both in rats and sheep, that efferent projections of MnPO neurons make direct synaptic contacts with vasopressin secreting magnocellular neurons in the supraoptic nucleus (Oldfield et al., 1991b; Zardetto-Smith et al., 1992). As well, stimuli to vasopressin secretion such as systemic infusion of hypertonic saline, angiotensin II or dehydration due to water deprivation activate MnPO neurons (as shown by Fos expression) that project to the supraoptic nucleus (McKinley et al., 1994; Oldfield et al., 1994). However, within the MnPO, both excitatory glutamatergic and inhibitory GABAergic neurons project to the hypothalamic magnocellular supraoptic and paraventricular nuclei (Nissen and Renaud, 1994; Grob et al., 2003; Abbott et al., 2016). How are these efferent neural connections of MnPO to vasopressin-containing neurons significant for vasopressin secretion? In regard to excitatory glutamatergic pathways from MnPO to supraoptic and hypothalamic paraventricular nuclei, evidence from lesion studies in rats and sheep show that ablation of the MnPO severely disrupts osmotically stimulated vasopressin secretion (Mangiapane et al., 1983; Gardiner et al., 1985; McKinley et al., 2004). In mice, chemogenetic inhibition of MnPOGlut neurons blocked osmotically stimulated vasopressin secretion and attenuated that to peripheral angiotensin II (Marciante et al., 2019). However, ablation of the MnPO does not disrupt vasopressin secretion associated with hemorrhage, fever, hypovolemia, or hypotension (Gardiner et al., 1985; McKinley et al., 2004), suggesting that the MnPO neurons specifically relay signals from osmoreceptors and AT1 receptors in circumventricular organs that drive vasopressin release. In this regard, although there are direct projections of osmoresponsive neurons within the subfornical organ and OVLT to the supraoptic nucleus (Oldfield et al., 1994), osmoregulated neural signals may also be relayed via a synapse in the MnPO to the supraoptic (Honda et al., 1990; Oldfield et al., 1991a,b,c; Tanaka et al., 1995) and hypothalamic paraventricular nuclei (Stocker and Toney, 2005). Microinjections of glutamate-receptor antagonists into the MnPO block incoming signals to the MnPO from osmoreceptors in the subfornical organ and OVLT and inhibit the release of vasopressin in response to systemic hypertonicity (Yamaguchi and Watanabe, 2002; Yamaguchi and Yamada, 2006). These data indicate the important role of the MnPO in relaying osmoregulatory signals from osmoreceptors elsewhere in the lamina terminalis to the neurosecretory magnocellular

443

neurons in the supraoptic and hypothalamic paraventricular nuclei to cause vasopressin secretion. It is also likely that Na+ sensors within MnPO (Grob et al., 2004b) have a role in vasopressin release. In regard to the inhibitory GABAergic input from the MnPO to vasopressin-secreting neurons in the hypothalamic supraoptic and paraventricular nuclei (Nissen and Renaud, 1994; Abbott et al., 2016), studies in goats and rats in which the lamina terminalis was ablated have shown that the excretion of water loads is impaired if the MnPO is included in the lesion (Andersson et al., 1975; Johnson et al., 1978; Rundgren and Fyhrquist, 1978). These observations suggest that an impairment in inhibition of vasopressin secretion was caused by ablating the MnPO, despite the fact that excitatory glutamatergic input from MnPO to vasopressin-secreting neurons in the PVH and supraoptic nucleus would have been destroyed. More recently, investigations of this phenomenon in conscious sheep in which the MnPO had been ablated several weeks earlier revealed that intragastric water loading (2.5 L by gavage) failed to initiate a water diuresis within 4 h of its administration, despite plasma osmolality falling to extremely low levels (262 mosmol/kg). Plasma vasopression remained at 1–2 pg/mL, not falling to levels needed for a water diuresis (