Introduction to Basic Aspects of the Autonomic Nervous System: Volume 1

By Otto Appenzeller, Guillaume J. Lamotte and Elizabeth A. Coon

Introduction to Basic Aspects of the Autonomic Nervous System, Sixth Edition, Volume One is an all-encompassing reference on the autonomic nervous system's basic function, dysfunction and pathology. This volume describes the anatomy of the autonomic nervous system and its role in the regulation of blood pressure, body temperature, respiration, micturition, digestion and renal function. Additional chapters focus on the autonomic modulation of the neuroendocrine system, sexual function, and immunity. There is also a chapter on mummies and the autonomic nervous system.

With these chapters, readers will gain extensive knowledge on the autonomic nervous system's anatomy, functional organization and neurochemistry, which is critical to care for patients with autonomic disorders and guide patient-oriented research.

• Provides an extensive reference on the autonomic nervous system and its crucial functions and dysfunction
• Discusses all aspects of autonomic physiology and pathology
• Outlines several physiological processes regulated by the autonomic nervous system, including thermoregulation, blood pressure, micturition, respiration, digestion and renal function
• Features chapters on the modulation of the neuroendocrine system, sexual function, immunity, and a new chapter on mummies and the autonomic nervous system

• Language: English
• Publisher: Academic Press
• Release date: Jul 15, 2022
• ISBN: 9780323955850

Otto Appenzeller

Dr. Appenzeller MD, PhD is Professor Emeritus at the University of New Mexico in the Departments of Neurology and Medicine. He is also President of the New Mexico Health Enhancement and Marathon Clinics.

Book preview

Front Cover for Introduction to Basic Aspects of the Autonomic Nervous System - Volume 1 - 6th Edition - by Otto Appenzeller, Guillaume J. Lamotte, Elizabeth A. Coon

Introduction to Basic Aspects of the Autonomic Nervous System

Volume 1

Sixth Edition

Otto Appenzeller

University of New Mexico, Albuquerque, NM, United States

New Mexico Health Enhancement and Marathon Clinics Research Foundation, Albuquerque, NM, United States

Guillaume J. Lamotte

Department of Neurology, Movement Disorders, and Autonomic Disorders, The University of Utah, Salt Lake City, UT, United States

Elizabeth A. Coon

Department of Neurology, Autonomic Disorders, Mayo Clinic, Rochester, MN, United States

Table of Contents

Cover image

Title page

Copyright

Foreword

Acknowledgments

Introduction

Chapter 1. Autonomic anatomy, histology, and neurotransmission

Abstract

1.1 Historical perspective

1.2 Anatomy of the autonomic nervous system

1.3 The intrinsic cardiac nervous system; its role in cardiac pacemaking and conduction

1.4 Human sympathetic preganglionic vasomotor and sudomotor neurons

1.5 The neurogenic control of the cutaneous circulation

1.6 Neurotransmitters

1.7 Peptides

1.8 The neurogenic control of the circulation in skeletal muscles

1.9 Shear stress and arterial lumen

1.10 Catecholamines and neurologic disease

References

Chapter 2. Neurogenic control of the circulation, syncope, and hypertension

Abstract

2.1 Neurogenic control of the circulation

2.2 Syncope

2.3 Hypertension

References

Chapter 3. Thermoregulation

Abstract

3.1 Introduction

3.2 Thermoregulatory pathway and peripheral thermoreceptors

3.3 Vasomotor responses

3.4 Local blood vessel changes and environmental temperature

3.5 The hypothalamus

3.6 The measurement of temperature

3.7 Abnormalities of body temperature

3.8 Temperature regulation during fever

3.9 Temperature regulation and exercise

3.10 Periodic fevers

3.11 Peptides

3.12 Sweating and thermoregulation

References

Further reading

Chapter 4. Autonomic modulation of the neuroendocrine system

Abstract

4.1 The hypothalamus and neuroendocrine modulation

4.2 The autonomic nervous system and the pineal gland

References

Chapter 5. Neurogenic control of respiration

Abstract

5.1 Introduction

5.2 The control of respiratory movements

5.3 Pulmonary respiratory reflexes

5.4 Breathlessness

5.5 The nasal cycle

5.6 The cerebrospinal fluid and the control of pulmonary ventilation

5.7 Abnormalities in respiration related to disease of the nervous system

References

Chapter 6. The enteric nervous system

Abstract

6.1 Introduction—anatomy and physiology

6.2 Appetite, satiety, hunger, and the hypothalamus

6.3 Normal swallowing and its disorders

6.4 The stomach

6.5 The intestine

6.6 The rectum and defecation

6.7 Vomiting

6.8 Other reflexes

6.9 Rectal biopsy in the diagnosis of neurological disease

6.10 The irritable bowel syndrome

6.11 Diarrhea and noninvasive secretagogues

6.12 Idiopathic intestinal pseudoobstruction

6.13 Enteric nervous system stem cell therapy for enteric neuropathies

6.14 The gut–brain axis

References

Chapter 7. Autonomic modulation of immunity

Abstract

7.1 Autonomic nervous system and inflammation

7.2 Autonomic innervation of the main immune organs

7.3 Effects of ablation experiments on the immune system

7.4 Neural control of immunity in hypertension

7.5 Multiple sclerosis and autonomic function

7.6 Paraneoplastic syndromes and autonomic failure

7.7 Experimental autonomic neuropathy

7.8 Physical activity and nutritional influence on immune function

References

Chapter 8. Neurogenic control of sexual function

Abstract

8.1 Innervation of sexual organs

8.2 Central nervous system

8.3 Sexual activity in females

8.4 Sexual responses in males

8.5 Neuroanatomical sex differences

8.6 Sexual function in patients with nervous system lesions

References

Chapter 9. Neurogenic control of renal function

Abstract

9.1 Introduction

9.2 Sympathetic innervation of the kidneys

9.3 Sensory renal nerves

9.4 Regulation of renal sympathetic nerve activity

9.5 Renal nerves in diseases

9.6 Neural regulation of thirst and salt appetite

9.7 The renin-angiotensin-aldosterone system

9.8 Special considerations

References

Chapter 10. Neurogenic control of micturition

Abstract

10.1 Introduction

10.2 The anatomy of the bladder and urethra

10.3 Muscles of the pelvic diaphragm and the perineum

10.4 Peripheral innervation of the bladder and urethra

10.5 Influences of the central nervous system on micturition

10.6 Reflex micturition

10.7 Voluntary micturition

10.8 Pathology in humans

References

Chapter 11. The archeology of autonomic function

Abstract

11.1 Diseases with autonomic nervous system involvement in mummy studies

11.2 Paleoneurobiology

11.3 Statistical methods in bioarchaeology

References

Index

Copyright

Academic Press is an imprint of Elsevier

125 London Wall, London EC2Y 5AS, United Kingdom

525 B Street, Suite 1650, San Diego, CA 92101, United States

50 Hampshire Street, 5th Floor, Cambridge, MA 02139, United States

The Boulevard, Langford Lane, Kidlington, Oxford OX5 1GB, United Kingdom

Copyright © 2022 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.

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.

ISBN: 978-0-323-95584-3

For Information on all Academic Press publications visit our website at https://www.elsevier.com/books-and-journals

Publisher: Nikki P. Levy

Acquisitions Editor: Natalie Farra

Editorial Project Manager: Sam Young

Production Project Manager: Kiruthika Govindaraju

Cover Designer: Vicky Pearson Esser

Cover Image Designer: Negin Badhinian

Typeset by MPS Limited, Chennai, India

Foreword

Phillip A. Low

Otto Appenzeller, Emeritus Professor of Neurology at the University of New Mexico, is a pioneer in the autonomic field, converting this Cinderella of medicine from a research curiosity to medical practice. His initial book enabled some of that thrust. The foreword by Ray Adams to the fifth edition captures well the situation with clinicians of the day when a young investigator from Sydney, Australia, arrived at the Massachusetts General Hospital. He reveled in the academic environment and the influence of neurologic giants like Ray Adams and C. Miller Fisher. In turn, he provides novel insights into autonomic dysfunction in disorders like Guillain–Barré syndrome and acute pandysautonomia. The sustainability of the effort is evident in this sixth edition, now grown into 2 volumes with 25 chapters. The volumes reflect Otto’s interest in autonomic physiology and the historical evolution of some autonomic disorders. The book also provides historical aspects of some autonomic disorders.

The sixth edition greatly benefited from the addition of two new authors, Elizabeth Coon and Guillaume Lamotte, who bring into the field cutting-edge autonomic neurology and autonomic neuroscience. Their involvement has resulted in a revision of each of the chapters. Additionally, they have added their experience derived from the clinical autonomic laboratory, adding the quantitative dimension, describing the severity and distribution of autonomic failure. Their contributions include standardized autonomic function tests, modern biomarkers, novel chapters on postural tachycardia syndrome, and the dysautonomias related to Covid-19 infection. This blend of the unique aspects of Otto’s contributions with evolving new approaches promises to give the book a new life.

Acknowledgments

Nearly half a century since the appearance of the first edition should have increased the number of individuals who importantly contributed to the success of this book. Surprisingly, however, modern technology and easy access to databases have allowed many helpful labors for the sixth edition to remain anonymous. This is, for the most part, regrettable because human interactions added significant dimensions to the readability, relevance, and clinical immediacy of the book. We would like to thank Dr. Negin Badihian for her assistance on the figures and the cover of the book. We must also not forget those who contributed generously to previous editions: Drs. T. K. Von Storch, G. B. Marcus, and D. Scott (first edition), Dr. G. Ogin (second edition), Dr. E. Collins and Prof. Yen Tsai (third edition), Drs. M. Appenzeller, P. Appenzeller, S. Wood, and R. Greene, and my son Tim Appenzeller, now on the editorial staff of science (fourth edition). Previous editions were typed, proofread, and shepherded through the publishing process by many devoted helpers: Mrs. Grace Wilson, Mrs. Polly Gauthier, and Mrs. Vi Farmer (first edition), Mrs. Connie Sokolowski (second edition), Ms. Katherine Miller (third edition), and Ms. Pamela Livingston (fourth edition). The authors themselves used computer technology to produce the fifth and sixth editions.

Throughout the quarter of the century of the development of the book, and more so recently, our families have supported our endeavors with great understanding and sacrifice and we acknowledge with gratitude the only human interactions for the sixth edition.

Many new illustrations were taken from other sources, and references to the generous permissions for reproduction are found in the legends. Previously published materials are cited in the list of references at the end of the book.

Elsevier Science has shown a remarkably sustained interest in the Autonomic Nervous System.

Introduction

Otto Appenzeller, Guillaume J. Lamotte and Elizabeth A. Coon

Since the appearance of the fifth edition, a number of important advances in basic aspects of the autonomic nervous system needed to be made available to clinicians. These include advances in our understanding of the role of central autonomic areas and autonomic nerves thanks to the use of more sophisticated techniques but also through the development of new ideas. Beyond the control of body homeostasis, the autonomic nervous system is regulated both through modality-specific afferents and their reflex circuits and by functionally selective groups of neurons with increasing degrees of complexity at all levels of the central nervous system. The first volume of the sixth edition focuses on basic aspects of the autonomic nervous system. We discuss the neuroanatomy and physiology of the autonomic nervous system and its role in the regulation of important body processes including circulation, thermoregulation, gastrointestinal function, micturition, renal function, respiration, inflammation, immunity, and neuroendocrine function. We added a chapter on the archeology of autonomic function highlighting the role of paleopathology in the exploration of the autonomic nervous system. The reader will find key references to historical studies with foundational work that still influences the field of autonomic medicine today as well as up-to-date references.

Chapter 1

Autonomic anatomy, histology, and neurotransmission

Abstract

Bodily functions that can proceed independently of volitional activity are regulated atleast in part by reflex mechanisms served by afferent, efferent, and central integrating structures, which are included in the autonomic nervous system. Knowledge of the anatomy, functional organization and neurochemistry of the autonomic nervous system is critical for the diagnosis and management of patients with autonomic disorders. This chapter discusses the anatomy, histology, development, and function of the different components of the autonomic nervous system. We also discuss the role of the different neurotransmitters involved in autonomic neurotransmission such as acetylcholine, norepinephrine, and epinephrine among others.

Keywords

Sympathetic; parasympathetic; norepinephrine; acetylcholine; epinephrine; intrinsic cardiac nervous system; anatomy; histology; neurotransmitters

1.1 Historical perspective

Galen (1528) seems to have been the first to refer to a part of the autonomic nervous system when he described a nerve trunk lying along rib heads and recognized its connecting fibers with the spinal cord. He thought that this was a branch of the vagus and believed that through it the viscera received sensitivity from the brain and power to move from the spinal cord. Galen also observed at least three enlargements along the course of the sympathetic chain and described the ganglion at the entrance of the chain into the abdomen, which might have been the semilunar ganglion of the celiac plexus. He suggested a widely accepted belief thereafter that there was sympathy or consent between body parts and regarded the peripheral nerves as tubes through which animal spirits were distributed to bring about this sympathy.

Galen’s belief that the vagi and sympathetic trunks were a single unit, both functionally and anatomically, was shared by subsequent anatomists until Estienne (1545) recognized the sympathetic trunks as a distinct anatomical structure. Willis (1664) gave the name of intercostal nerve to the sympathetic trunk and believed that the cerebellum was responsible for involuntary movements, which he distinguished from voluntary motion.

Although Willis recognized the innervation of the heart by the vagus nerve, the functional significance of this was not discovered until Lower’s (1669) description of the effects of vagus section on heart rate. The suggestion that involuntary movements are initiated by local stimulation due to nerve irritation is attributed to Whytt (1751) and he used this as an explanation for the reaction of the pupil to light. Whytt’s importance to neurophysiology is further enhanced by his suggestion that all sympathy or consent must be referred to the central nervous system since it occurs between body parts whose nerves make no connection with each other so that the transmission of sympathy cannot involve the flux of matter and must, therefore, occur in an area where all nerves have their origin (Whytt, 1765).

Du Petit (1727) pointed out that the sympathetic trunk was not directly connected with the brain and was a separate structure from the vagus, but it was Winslow (1732) who first designated the paravertebral chain as the great sympathetic nerve.

Johnstone (1764) said that the movements of the heart and intestine are involuntary because the sympathetic ganglia blocked the actions of the will and prevented them from reaching these structures and that this blockade also accounted for the relative insensitivity of the viscera.

Studies by Bichat (1802) led to the concept of animal life and organic life. He pointed to the continuous action, which was the hallmark of organic life distinct from the intermittent activity of animal life, a concept, which is still widely expressed by the terms visceral and somatic, respectively. Bichat was aware of the different appearances of gray and white rami communicantes but did not recognize their significance.

The term vegetative nervous system was used by Reil (1857). He thought that the rami communicantes served as connectors between the animal and vegetative nervous systems. An early description of nerve cell bodies in sympathetic ganglia appeared in Ehrenbergs (1833) writing, together with comments on the microscopic structure of nerve fibers. In the 19th century, the ciliary, sphenopalatine, otic, and submandibular ganglia were thought of as part of the autonomic nervous system but their functional significance was not appreciated.

Meissner’s (1857) mention of the submucous plexus and the description of the myenteric plexus by Auerbach (1864) in the second half of the 19th century concluded the anatomic studies, which paved the way for the physiologic work on vasomotor function by Bernard (Bernard, 1878). His studies led to the concept of sympathetic vasoconstrictor action but not until sometime later was he able to demonstrate vasodilator nerves in arteries supplying the submandibular gland after stimulation of the chorda tympani. Bernard thought that the sympathetic reflexes were mediated by the spinal cord and that on stimulation of some areas of the brain impulses were discharged through the sympathetic fibers.

Gaskell (1886) gave a detailed description of the anatomy of the rami communicantes and recognized that the efferent fibers within these nerves arise in the spinal cord and that corresponding fibers can be found in some cranial nerves. He also recognized the connection with the central nervous system of the peripherally located ganglion masses through medullated fibers and divided these fibers into bulbar, thoracolumbar, and sacral groups. Langley and Dickinson (1889) used the action of nicotine on ganglia to study the relation of the nerve fibers to the peripheral ganglion cells and proposed the name autonomic nervous system. When this term was coined, the different distribution and functional effects of the thoracolumbar and craniosacral outflows were known and Langley separated the former from the rest of the autonomic nervous system. After the discovery of substances that either produced actions similar to those obtained by stimulation of the thoracolumbar or of the craniosacral outflows, he coined the term parasympathetic for the latter (Langley, 1901).

1.2 Anatomy of the autonomic nervous system

Body functions, which can proceed independently of volitional activity, are regulated at least in part by reflex mechanisms served by afferents, efferents, and central integrating structures, which are included in the autonomic or vegetative nervous system. Although the activity of this system is essentially autonomous, it is not entirely free from voluntary control. The vegetative nervous system is made up of all neurons, which lie outside the central nervous system and is concerned with visceral innervation. The only exceptions are those that are part of the afferent system of the cerebrospinal nerves and are located in the posterior root ganglia or some sensory cranial nerve ganglia. The neurons in the brain, brainstem, and spinal cord, through which these autonomic neurons are functionally connected, are also included in the vegetative nervous system.

There are two main divisions of this system, the sympathetic and parasympathetic, each usually made up of preganglionic and postganglionic neurons. The cell bodies of the preganglionic neurons lie in the brain or spinal cord and those of the postganglionic neurons in the autonomic ganglia. The preganglionic sympathetic neurons are in the thoracic and upper lumbar cord and this part of the vegetative nervous system is also called the thoracolumbar division (Fig. 1.1). Preganglionic neurons of the parasympathetic nervous system are in the brainstem and sacral cord and this is also termed the craniosacral division (Figs. 1.2 and 1.3). The viscera are mostly innervated by both sympathetic and parasympathetic fibers. Many structures supplied by the vegetative nervous system have, however, a single innervation only, such as some blood vessels and sweat glands.

Figure 1.1 Diagram to show the pre- and postganglionic fibers of the autonomic innervation of the thoracic and abdominal viscera. Dotted lines: postganglionic fibers of the thoracolumbar division. Short solid lines on viscera: postganglionic fibers of craniosacral division. Drawn by M. Norviel.

Figure 1.2 Diagram to show the brainstem nuclear masses which form part of the craniosacral division of the autonomic nervous system. Left: dorsal aspect; right: lateral aspect. Drawn by M. Norviel.

Figure 1.3 Diagram to show the pre- and postganglionic fibers of the autonomic innervation of the abdominal and pelvic viscera. Dotted lines: postganglionic fibers of the thoracolumbar division. Short solid lines on viscera: postganglionic fibers of craniosacral division. Drawn by M. Norviel.

The preganglionic fibers of the sympathetic nervous system arise in the intermediolateral and intermediomedial cell columns of the spinal cord and join the ventral roots of T1 to L2. Variations of this occur and preganglionic fibers from C7 (Harman, 1900) and as low as L4 cord segments have been demonstrated (Monro, 1959; Randall, Cox, Alexander, & Coldwater, 1955). Some preganglionic pathways remain intraspinally for up to 12 segments before exiting through ventral roots to reach the sympathetic chain. The axons to thoracic ganglia arise from ipsilateral sympathetic preganglionic neurons but those reaching lumbar ganglia are of bilateral origin. Therefore, crossed and uncrossed intraspinal preganglionic pathways exist (Faden & Petras, 1978). The preganglionic fibers synapse in the sympathetic chain or traverse several of the ganglia up or down the chain before synapsing, or may pass through the ganglia to synapse in collateral ganglia near viscera. These latter fibers form the splanchnic nerves (Figs. 1.1, 1.4, and 1.5). They also contain some postganglionic fibers, particularly near their termination (Kuntz, 1956; Kuntz, Hoffman, & Schaeffer, 1957). The long postganglionic sympathetic fibers join peripheral nerves to be distributed to blood vessels, skin, and other structures or form visceral nerves such as the cardiac nerves. Some fibers form plexuses like those distributed to the head. Other postganglionic sympathetic fibers come from the accessory or aberrant ganglia, which may be found in neve trunks or communicating rami (Alexander, Kuntz, Henderson, & Ehrlich, 1949; Kuntz, 1953) (Figs. 1.4 and 1.6).

Figure 1.4 Diagram to show the relation of pre- and postganglionic fibers in chain and collateral ganglia. Drawn by M. Norviel.

Figure 1.5 Diagram to show the paravertebral sympathetic ganglia and splanchnic nerves in man. Drawn by M. Norviel.

Figure 1.6 Diagram of cervical sympathetic ganglia to show the formation of visceral nerves. Drawn by M. Norviel.

Preganglionic fibers of the parasympathetic chain arise in the visceral brainstem nuclei (e.g., the dorsal motor nucleus of the vagus) and the second to fourth sacral segments. They are distributed by the third, seventh, ninth, and tenth cranial nerves and the bulbar accessory to the head and neck, thorax, and abdominal viscera. The distal ganglia of the descending colon and pelvic organs receive preganglionic fibers from the sacral segments (Figs. 1.2 and 1.3).

In 2016, Espinosa-Medina et al. proposed to reclassify the sacral portion of the autonomic nervous system as being sympathetic, thereby revising Langley’s classical distinction between the thoracolumbar sympathetic and sacral parasympathetic divisions (Espinosa-Medina et al., 2016). The authors studied transcription factors and molecular markers in mouse embryos at multiple embryonic stages. This study revealed that neuronal nitric oxide synthase and the transcription factor Foxp1 were detected in spinal but not vagal preganglionic neurons, while transcription factors Phoxb2, Tbx2, 3, and 20 were expressed in vagal preganglionic but not spinal neurons. In Olig2–/– mice depleted from spinal motor neurons, the pelvic ganglia, just like the sympathetic ganglia, developed independently of outgrowing preganglionic axons. Based on these findings, the authors concluded that the sacral autonomic outflow was sympathetic and not parasympathetic. While this study shed light on the ontogenesis of the autonomic nervous system, it only provides evidence about the spinal (not sympathetic) nature of both the thoracolumbar and sacral autonomic pathways and it does not take into account the complexity of the autonomic innervation of pelvic organs.

The enteric nervous system has been named the largest and most complex division of the peripheral nervous system (Jessen, Mirsky, & Hills, 1987). The neurons are found grouped together in large numbers and have complex synaptic interactions. These enteric neuronal circuits are surprisingly independent of central nervous system input for basic control of gut activity. Nevertheless, there is a close analogy between nonneural supporting cells in the enteric ganglia, the enteric glial cells, and most central nervous system glial cells, including astrocytes. The antiquated view that the enteric plexuses are cholinergic parasympathetic relays had to be reappraised; enteric ganglia are highly complex integrative neural structures, and they can be compared with the physiologic biochemical and histologic features of the central nervous system (Gabella, 1976; Jessen & Mirsky, 1983).

1.2.1 Autonomic ganglia

Autonomic ganglia are divided into paravertebral (chain) ganglia and prevertebral (collateral) ganglia, which are the synaptic sites of sympathetic fibers, and peripheral (terminal) ganglia in which the synapses of parasympathetic fibers are found. Embryologically, they arise either entirely from the neural crest (Hammond, 1946; Yntema & Hammond, 1947, 1955) or the basal plate of the neural tube (Brizzee, 1949; Brizzee & Kuntz, 1950), or from both sites (Triplett, 1958). It may be that their origin varies with the species studied. Neurons and glial cells of the enteric nervous system are derived from the neural crest. There is evidence for a vagal level neural crest source of the enteric nervous system, whereas the contribution of sacral neural crest neurons that eventually lie in the lower intestine remains controversial (Yntema & Hammond, 1953; Young, Hearn, & Newgreen, 2000 Gut). The neurons of the parasympathetic ganglia of the head and neck also arise from the neural crest (Hammond & Yntema, 1958).

The paravertebral sympathetic ganglia lie on both sides of the vertebral bodies. The ganglia are attached to the ventral roots of the thoracic and lumbar segments by myelinated axons of the preganglionic neurons, which reach them through the ventral roots. There is a whitish appearance to the fiber bundles, which also contain myelinated visceral afferents and they are called white communicating rami. The postganglionic fibers from these ganglia also connect the chain to the ventral nerve roots, since they are distributed through the nerves to the periphery. These fibers are not myelinated, or only thinly myelinated, and look grayish in the fresh specimen. They are called gray communicating rami. There may be more than one gray communicating ramus to a spinal nerve (Mitchell, 1953).

The superior cervical ganglion is usually related to the upper four cervical levels. The middle cervical ganglion is inconstant but if present is related to the fifth and sixth cervical segments (Becket & Grunt, 1957; Jamieson, Smith, & Anson, 1952; Kuntz, 1953). The inferior cervical ganglion is related to the seventh and eighth cervical segments and in eighty-two percent of cases examined it is fused with the first thoracic ganglion into a large mass of neurons, the so-called stellate ganglion. When the first thoracic ganglion remains separate, it is also called the stellate ganglion (Hoffman, 1957). There are up to 11, but often fewer, ganglia on each side in the thoracic region. The number in the lumbar and in the sacral region is four each but varies considerably (Kuntz, 1953). The chain ganglia in the coccygeal region are fused into the coccygeal ganglion or ganglion impar.

Preganglionic sympathetic fibers destined for the abdominal and pelvic organs pass through the chain ganglia without synapsing and form the splanchnic nerves, which end in collateral ganglia usually situated around branches of the abdominal aorta. The postganglionic fibers originating in these collateral ganglia pass along branches of the aorta to supply the viscera.

Above the diaphragm, all sympathetic preganglionic fibers synapse in the chain ganglia, and postganglionic fibers are distributed to the viscera. Those supplying structures in the head take their origin in the superior cervical ganglia and are distributed along blood vessels.

Terminal or peripheral ganglia are small collections of autonomic neurons on or within the walls of various organs. They are mainly synaptic sites for preganglionic parasympathetic fibers and are often called parasympathetic ganglia. They comprise cranial ganglia, such as the ciliary, sphenopalatine, otic, submandibular, and Langley’s ganglion, and the cervical ganglia of the uterus. In the gastrointestinal tract, these neurons form plexuses known as myenteric (Auerbach) and submucosal (Meissner) plexuses.

1.2.2 The normal histology of ganglia

Descriptions of the normal histology of human sympathetic ganglion cells have been given by many authors (De Castro, 1932; Stöhr, 1928, 1943a, 1943b, 1948). The ganglion cells are different in various sites. Thus, the superior cervical ganglion contains a large number of polymorphic cells, whereas the celiac ganglion is composed predominantly of large stellate cells. All ganglia are surrounded by a connective tissue capsule that extends into the depth of the tissue and separates the cells by septa into small compartments (Martin, 1937). Numerous attempts to establish a pathologically significant increase in the amount of connective tissue and to correlate this with certain diseases have failed. The amount and density of the connective tissue in sections vary with the plane of the cut. Samples from near the surface or the end of the ganglion usually contain large amounts of dense connective tissue, whereas those from near the center show thick septa, often containing nerve bundles. These septa also contain numerous arterioles and venules. In children and fetuses, there is little connective tissue (Spiegel & Adolf, 1920). The thin trabeculae in infantile sympathetic ganglia hardly separate the densely packed ganglion cells. With advancing years, however, there is an increase in the connective tissue, which penetrates between the cells and probably accounts for the growth of the sympathetic ganglia. Numerous mast cells are often found within the capsule and in the trabeculae. Herzog and Sepúlveda (1940) stated that these cells are rarely if ever, found among the neurons. Their significance and function within the sympathetic chain are not known.

The size of the ganglion cells varies. Nevertheless, three subdivisions have been described. Large cells with a diameter of 35–155 mm, medium-sized cells with a diameter of 25–32 mm, and small cells with a diameter of 15–22 mm (De Castro, 1932). A total of 50 to 70% of all ganglion cells are of medium size. All cells have a large, somewhat transparent nucleus that contains a clearly delineated nucleolus. Cells with two or more nuclei can be found particularly in fetuses and in children (De Castro, 1932; Herzog, 1931) but are not common in adults. The neurons are filled with a fine neurofibrillary network, which leaves only the nuclear zone free. Occasionally the neural processes appear empty but this may also be the result of pathological swelling of the cells (Von Doring, Herzog, Krucke, & Orthner, 1955). The sympathetic ganglion cells can be further subdivided according to the length of their cell processes (De Castro, 1932). It has been emphasized, however, that the distinction between dendrites and axons can be difficult, although the existence of the latter seems to have been proven by Cajal and Stöhr. Cells with long neural processes are particularly common in the alimentary canal. Ganglion cells with short and accessory processes are also found. These often branches and twist within the capsule of gliocytes that surround the cell. Sometimes they enclose small spaces that could easily be mistaken for vacuoles. Numerous morphologic types of synapses within the sympathetic chain are described. These include club-like or pear-shaped endings and dendritic skeins (Cajal, 1911). Their distribution is quite haphazard but the large ganglion cells have numerous synaptic endings (Herzog, 1931).

1.2.3 Blood supply to autonomic ganglia

There is little information about the blood supply of the sympathetic ganglia. Occasionally sizable venous channels and arterioles are described in the center of the larger ones. In animals, a rich capillary and venular bed can be found in the superior cervical ganglia (De Castro, 1932). Based on observation with India ink injections, Ranvier (1888) described venous sinuses within the sympathetic ganglia. There are lymphatics in the sympathetic chain and they are occasionally invaded by carcinoma, which then makes these channels visible (Von Doring et al., 1955). A rich lymphatic network surrounds the cervical sympathetic ganglia and it is claimed that this is proof of their high metabolic activity. These lymph vessels drain into the cervical lymph nodes (Rouvière, 1929). Each sympathetic ganglion cell is embedded in a fine fibrillary meshwork, which forms various sized holes into which the cells fit snugly. The nuclei of the supporting cells can be seen among this fine network. The cytoplasm of these supporting cells cannot be recognized with ordinary histological methods and requires silver carbonate stains for visualization. These cells have been called capsular cells, satellite cells (Cajal, 1911), or amphicytes (Stöhr, 1928). They surround the ganglion cells and their processes and appear analogous to the oligodendrocytes of the central nervous system (Von Doring et al., 1955). They were first clearly defined by Del Hortega Rio and Prado (1942) who named them gliocytes. The function of gliocytes is obscure. It is believed by some that they have endocrine functions (Nageotte, 1910). De Castro (1932) suggested that they are the site of acetylcholine formation and Sulkin and Kuntz (1948) showed that ascorbic acid is present in the gliocytes and is markedly reduced in hypertensive patients. They also observed hyperplasia of the gliocytes in animals given diphtheria toxin and in patients with tuberculosis and pneumonia, and ascribed this to toxic irritation. Conflicting statements and claims are made about specific functions and activities of gliocytes but they likely play an important role in the control of the microenvironment of sympathetic ganglia (Hanani, 2010). The formation of lactate in satellite gliocytes is induced by nicotinic cholinergic synapses directly involved in neuron-glial interactions and in controlling the activity of the Lactose dehydrogenase enzyme system in sympathetic neurons (Gorelikov & Savel’ev, 2008).

1.2.4 Development of the autonomic nervous system

During development, a complex, as yet poorly understood process takes place, which must address two problems:

1. How are synapses established between appropriate neurons?

2. How are the number of cells and the number of synaptic contacts between them regulated?

The problem of qualitative accuracy, that is, the way in which synaptic contacts between appropriate partners are achieved, remains largely unsolved. However, there is somewhat more agreement about how quantitative regulation, that is, the relationship between pre and postsynaptic neuronal populations and synaptic contacts between them, is established. It has, for example, been shown that the innervating population of neurons is well matched to the capacity of target structures. Neurons are overproduced during development and compete for survival in early embryonic life. Presynaptic neurons are dependent on some activity of their targets. If, for example, an increase in target size is achieved experimentally, the survival and size of the innervating neurons are increased, and conversely, artificially decreasing the target size increases neuronal death. Evidence has also been found that a trophic factor is produced by the target for which innervating neurons compete during development. Thus, it is clear that there is no preordained neuronal pool that innervates a target population, but that the number of neurons is adjusted and depends upon feedback mechanisms. In addition, there are also quantitative adjustments of neuronal contacts by a normally occurring elimination of synaptic cells. These features have experimentally been demonstrated during the development of muscle, but synapse elimination occurs also in autonomic ganglia in which synaptic contact is between pre and postganglionic neurons. Thus, in the mature rat submandibular parasympathetic ganglion, about 80% of neurons are innervated by single preganglionic axons. But at birth, these neurons receive an average of five different axons. Only after the first few weeks of life is the adult one-to-one pattern established. In developing sympathetic ganglia, which are innervated by a larger number of axons during adulthood, the findings are comparable. In the superior cervical ganglia, normally innervated by about six axons in the adult, up to a dozen different axons are found during development. Thus, initial connections are eliminated during the first postnatal weeks even in those neurons, which continue to receive more than one presynaptic axon (Purves & Lichtman, 1980).

At three months gestation, the normal human sympathetic chain consists of aggregates of darkly staining, somewhat oval nuclei measuring 6–9 mm in diameter. The cytoplasm of these cells cannot be made out. They contain from three to six chromophilic nucleoli. These nuclei are often arranged in rosettes and accumulate in groups but also extend along the course of the nerve fibers from the ganglion to the ganglion. No pigment can be seen. Neural processes cannot be identified with silver stains. The rami communicantes and connecting interganglionic fibers do not take myelin stains. The cells described resemble neuroblasts. Numerous lymphocytes are diffusely scattered throughout the ganglia but mast cells are not found. No gliocytes can be seen but elongated cells arranged in rows, presumably fibroblasts, are related to small capillaries. Large vessels are not visible (Figs. 1.7 and 1.8).

Figure 1.7 Sympathetic chain from a fetus of 3-month gestation. The arrangement of nuclei in rosettes is seen (cresyl violet). From Appenzeller (1966).

Figure 1.8 Sympathetic chain from a fetus of 3-month gestation. No neurofibrils or neural processes are seen (Bodian protargol). From Appenzeller (1966).

At birth, definite autonomic neurons are found. The cytoplasm is clear and contains a Nissl substance. No neural fibrils can be identified with silver stains, although delicate neural processes are seen to emerge from many cells. The nucleus is large and pale, and the nucleolus chromophilic. Cells congregate in ganglia but many neurons are found between ganglia, within the nerves connecting them, and some extend out for a short distance in the rami communicantes. Neurons with two nuclei are frequently found (Figs. 1.9 and 1.10). The pigment is occasionally seen in some neurons with silver staining. Gliocytes can be seen closely applied to the neurons. Numerous fibroblasts are visible in the vicinity of capillaries. Myelinated fibers are present and are found in small numbers mainly in the rami communicantes. This appearance remains unaltered except that by the age of two years, the axons within the nerves connecting the ganglia show irregular elongated thickenings, which become more prominent with advancing years (Fig. 1.11). The normal sympathetic chain remains then essentially unchanged throughout life, except for the accumulation of increasing amounts of pigment in the ganglion cells and prominent hyaline thickening of small vessel walls in old subjects (Figs. 1.12 and 1.13). Mast cells are abundant, particularly in the perineurium of old patients. A striking increase in silver impregnation of neural processes and fibrils occurs in neurons of old subjects. The increased uptake of silver stains does not appear to be a pathological change as it is found in a variety of unrelated diseases and in subjects dying after accidents (Fig. 1.14). Fibroblasts are numerous and generally accompany vessels but it is impossible to grade the amount of fibrous tissue in the human sympathetic paravertebral chain, particularly if samples and not serial sections are examined.

Figure 1.9 Sympathetic ganglion at birth. The clear nucleus almost fills the cells (cresyl violet). From Appenzeller (1966).

Figure 1.10 Sympathetic ganglion at birth. Delicate neural processes and a few argyrophilic pigment granules are seen (Bodian protargol). From Appenzeller (1966).

Figure 1.11 Sympathetic ganglion of normal adult showing irregularly thickened nerve fibers (Bodian protargol). From Appenzeller (1966).

Figure 1.12 Sympathetic ganglion from a 70-year-old man showing an occasional heavily pigmented neuron (Bodian protargol). From Appenzeller (1966).

Figure 1.13 Hyaline thickening of small vessels in sympathetic ganglion from a 72-year-old patient with diabetes (periodic acid-Schiff). From (Appenzeller (1966)).

Figure 1.14 Heavy silver impregnation of neural processes in sympathetic chain ganglion of an 80-year-old man dying after an accident (Bodian protargol). From Appenzeller (1966).

The staining characteristics of sympathetic neurons are subject to wide variations determined by a variety of uncontrollable influences not related to pathological states. In surgically removed specimens, it can be seen that the normal neuron has a somewhat pale cytoplasm and nucleus (Fig. 1.15). In the same specimen, dark cells with indistinct cytoplasmic detail can be seen and this is most probably due to handling by instruments at the time of removal. In autopsy specimens, the neurons are frequently retracted away from the gliocytic capsules, leaving a clear space. This retraction is often incomplete so that thin tissue strands continue to remain attached to the capsule. This appearance suggests the presence of large vacuoles (Fig. 1.16). The retracted neurons sometimes stain darkly in hematoxylin and eosin as well as cresyl violet stains, and the nucleus is indistinct. Silver staining of these same specimens often shows remarkably well-preserved neural processes. These large extracellular vacuoles are to be distinguished from small foam-like intracellular vacuolization, which is pathological. In silver stains, it is difficult to distinguish the axons from connective tissue, which also takes up the stain. It is clear, however, that in normal specimens thickenings along the course of the axons are seen and that these may sometimes have elongated holes or may be divided into a fine network of fibrils. These appearances cannot be related to any pathological state and must be regarded as artifactual (Fig. 1.17). In electron micrographs, axonal irregularities have also been noted (Pick, Delemos Carmen, & Gerdin, 1964).

Figure 1.15 Artifacts of handling. Normal surgically removed sympathetic ganglion showing shrinkage and hyperchromatism of some neurons (cresyl violet). From Appenzeller (1966).

Figure 1.16 Sympathetic ganglion removed at autopsy showing large pseudovacuoles. Thin strands between the neuron and the gliocytic capsule remain. Artifact of handling and fixation (cresyl violet). From Appenzeller (1966).

Figure 1.17 Marked artifactual thickening of neural processes in sympathetic ganglion (Bodian protargol). From Appenzeller (1966).

1.2.5 Pathologic changes with aging

Age-related changes in the morphologic appearance of human cervical and lumbar ganglia have been examined (Nagashima & Oota, 1974). Twelve cases without disease of the central or peripheral nervous systems were selected for examination of the cervical and lumbar paravertebral sympathetic chain ganglia to delineate the normal changes in the histology of these ganglia with advancing age. Serial sections were obtained and showed a 36% decrease in the number of autonomic neurons in the seventh decade. The average size of neurons, however, remained constant after the second decade ranging from 25 mm to 135 mm, with the largest number being found at 60 mm in diameter. The first onset of neuronal degeneration was noted in the latter part of the second decade, and an increase in connective tissue was noted after the fourth. Age-related histologic changes in the paravertebral sympathetic ganglia were noted in the third decade and were manifested by cellular atrophy and satellitosis. A gradual replacement of neural tissue by fibrous connective tissue was noted, and this eventually formed sclerotic changes and residual nodules. It must, therefore, be stressed that normal aging may be associated with definite pathologic changes in paravertebral sympathetic ganglia, and these histologic changes must not be misinterpreted.

The fine structure of mammalian autonomic neurons (Palay & Palade, 1955) and of the frog (Taxi, 1967) have been extensively studied and both show essentially similar features. Numerous mitochondria unevenly distributed Nissl substance, and well-developed Golgi apparatus is found. In addition, in the cytoplasm of the neurons glycogen granules and numerous dense bodies are seen. It has been suggested that these bodies correspond to granules identified in the light microscope as lipofuscin. In addition, lipid droplets can also be recognized (Taxi, 1967). Gliocyte cytoplasm surrounds the neurons and all processes except in synaptic zones, and it is identified by the great number of filaments and a large amount of vesicles and canaliculi, which form part of the endoplasmic reticulum. Synapses are found in the perikaryon and neuronal processes. The presynaptic ending contains numerous clear vesicles and fewer dense cored vesicles. Mitochondria and glycogen granules are also seen. An active zone of increased density, particularly of the postsynaptic membrane is found at synaptic contacts (Figs. 1.18–1.21).

Figure 1.18 Ultrastructure of reptile sympathetic ganglion. CV, clear vesicles; DB, dense body; DCV, dense cored vesicles; G, Golgi apparatus; Gly, glycogen; GM, gliocyte membranes; L, lipid droplets; M, mitochondrion; N, nucleus; NP, nuclear pore; S, soma; SB, synaptic bulbs. Original ® 15,700. Courtesy of G.L. Colborn, Ph.D.

Figure 1.19 Ultrastructure of reptile sympathetic ganglion. AX, axon; AZ, active zone; CV, clear vesicles; DCV, dense cored vesicles; M, mitochondrion; NF, neurofilament; NT, neurotubule; SB, synaptic bulbs. Original® 30,750. Courtesy of G.L. Colborn, Ph.D.

Figure 1.20 Ultrastructure of reptile sympathetic ganglion. AZ, active zone; CV, clear vesicle; Dr, dense body; DCV, dense cored vesicle; G, Golgi apparatus; Gly, glycogen; M, mitochondrion; N, Nissl substance; R, ribosomes; SB, synaptic bulbs; GM, gliocyte membranes; UM, unit membranes. Original® 37,250. Courtesy of G.L. Colborn, Ph.D.

Figure 1.21 Ultrastructure of reptile sympathetic ganglion to illustrate the relation of the gliocyte cytoplasm and membranes to an axon. AX, axon; DCV, dense cored vesicles; GM, gliocyte membranes; GN, gliocyte nucleus; M, mitochondrion; NT, neurotubules. Original ® 15,700. Courtesy of G.L. Colborn, Ph.D.

1.2.6 Neurotransmitters and the autonomic nervous system

The application of molecular approaches to the study of autonomic function has yielded some advances. The traditional views have been challenged, and it is apparent that single neurons in the autonomic nervous system, and also in the rest of the neuraxis, often use multiple neurotransmitters for signaling. These transmitters are expressed independently, and their production is regulated by the same neurons. Moreover, external influences are being recognized that can change gene expression in the nervous system. Many discoveries have changed long-held concepts of autonomic function.

The peripheral sympathetic nervous system has been used as an easily manipulated model for sometime for the study of nerve function. The neurotransmitter norepinephrine mediates the consequences of sympathetic discharges and is involved in the fight or flight reaction, which has been scrutinized for more than a century. Classically, the actions of norepinephrine have suggested to investigators that sympathetic function is the result of norepinephrine release caused by impulse activity that leads to easily recognizable physiologic effects. Nevertheless, impulse activity has far-reaching effects that are relevant to information storage and occur beyond the momentary electrical discharge. For example, transsynaptic impulse activity is associated with long-term changes in metabolic pathways of norepinephrine in the postsynaptic sympathetic neurons. Stressful stimuli, including environmental stress, which normally increases sympathetic discharge, can induce tyrosine hydroxylase, which is the rate-limiting enzyme in norepinephrine synthesis (Kvetnansky, 1980). The tyrosine hydroxylase molecules increase in response to environmental stimuli, and this increase in the enzyme protein is long-lasting. A more careful definition of induction of this enzyme has shown that environmental stress causes a two- to three-fold increase in the enzyme in sympathetic neurons within days and that the increase remains elevated for at least three days after the increased impulse activity has ceased. Similarly, stimulation of nerves electrically for 0.5–1.5 hours also increases tyrosine hydroxylase for three days (Chalazonitis & Zigmond, 1980). This implies that stimulation of sympathetic nerves directly mimics the effect of environmental stress. A brief stimulus, therefore, causes long-term changes in molecular neuronal mechanisms that lead to amplification in time and that have implications for memory, in particular for long-term effects of environmental stress on sympathetic activity. Such changes in the biochemistry are important functionally also because the increase in tyrosine hydroxylase results in an increase in norepinephrine synthesis, which in turn shows that enzyme activity in the sympathetic nervous system is important in behavioral responses such as the fight or flight reaction (Black, Adler, & Dreyfus, 1987).

Single neurons may release multiple transmitters simultaneously (Hökfelt, Fuxe, & Pernow, 1986). This increases the capacity for information storage in neurons, and the possibility that changes in transmitter species can be manipulated by changing impulse activity has been examined. Sympathetic neurons contain the substance P (SP) and somatostatin. In vivo, denervation of sympathetic ganglia or treatment with pharmacologic substances that block ganglionic transmission causes a marked increase in SP (Kessler & Black, 1982). But sympathetic impulse flow depresses SP, suggesting that transsynaptic stimulation causes a fall in postsynaptic transmitter content. Analysis of denervated cultured sympathetic ganglia showed similar responses, that is, an increase in SP, which mimics surgical or pharmacologic denervation in vivo. Data suggests that a number of molecules are found in neurons to code for environmental events over relatively long periods of time. Some may be differentially regulated by environmental stimuli, but these molecules are not indifferently encoding environmental information, rather they are central to normal neural function. Millisecond activity (neuronal depolarization) remains the cornerstone for immediate responses of the sympathetic and other parts of the nervous system, but this can alter gene expression and lead to the storage of information in neurons. However, the precise intracellular mechanisms that mediate these effects remain unknown.

In the brain, neurons also transduce environmental stimuli into long-lasting transmitter changes and a prime example is the locus coeruleus, which has been extensively studied. This is a bilateral brainstem nucleus that innervates many structures throughout the neuraxis, including the cortex, the cerebellar cortex, multiple segments in the spinal cord, and the blood vessels in the brain. It is thought that the locus coeruleus is important in arousal and attention, and maintaining vigilance. The continuous or tonic release of norepinephrine that occurs during such neuronal activity and during grooming, sleeping, and eating in animals is important in maintaining internal homeostasis. By contrast, the episodic release of norepinephrine induced by environmental cues causes an increase in attention, alertness, and responsiveness to the environment. This important function of the locus coeruleus, which modulates nervous system activity in response to internal and external stimuli, is at the junction of behavioral states where the enzyme tyrosine hydroxylase occupies an important position, analogous to that in the peripheral sympathetic system. Tyrosine hydroxylase controls the biosynthesis of norepinephrine in the central nervous system and the amount of transmitter released in various areas of the brain and spinal cord. Similar effects to those seen in peripheral autonomic ganglia are found in the central nervous system and in explanted cultures the central autonomic neurons from the locus coeruleus.

These experiments indicate that memory function is not confined to specific isolated structures in the nervous system, but that biochemical information storage may be a widespread activity of many central and peripheral neuronal systems and account for such diverse phenomena as the memory for pain in the trigeminal system observed in aviators flying in nonpressurized aircraft. These pilots develop pain in their teeth after dental fillings. Changes in dopamine content in basal ganglia induced by external stimuli such as treadmill running in experimental animals (Freed & Yamamoto, 1985) are also an expression of memory function.

1.2.7 Intercellular junctions

Because a number of cells of the autonomic nervous system act in conjunction, they have relinquished their independence to function as a coherent whole. This orchestration between cells is accomplished through specialized intercellular junctions, which are vital for proper activity of the autonomic nervous system and normal function of all higher forms of life. The two principal preparation methods, which have been most rewarding for the study of the structure and function of specialized intercellular contacts, have been electron microscopy and freeze-fracture.

Several types of junctions between cells are important to the function of tissues as a whole. Tight junctions are areas where plasma membranes of two adjacent cells fuse in a region of intimate contact. These regions encircle many cells. At tight junctions, the plasma membranes are fused at a series of points, which stop the penetration of protein markers between cells. Freeze-fracture studies were particularly helpful in unraveling the three-dimensional aspects of tight junctions. They are characterized by a lace-work of ridges on the cytoplasmic half of the membrane face of the plasma membrane and are complemented by grooves on the external half of the membrane face. The ridges seem to be composed of rows of tightly packed particles (integral membrane proteins), one row is contributed by each of the adjacent plasma membranes. The head-to-head contact of these rows of particles suggests a modified zipper that holds the membranes closely together obliterating the intercellular space. The tightly packed particles form lines of attachment (sealing strands), which provide a physical barrier to the passage of molecules. There seems to be a direct relationship between the number of sealing strands in tight junctions of various tissues and the permeability of junctions. The more permeable ones have fewer numbers of strands compared to tightly sealing junctions, which place great resistance to the movement of ions across membranes. In the urinary bladder, for example, where there are more than six sealing strands per tight junction, the concentration gradients across epithelial cells are considerable, thus allowing for high electrical resistance. The design of tight junctions is an example of biologic engineering at the cellular level, for the network of sealing strands between cells determines the tightness of the seal, and this is varied according to the physiologic needs of tissues. Moreover, the network is so flexible that tight junctions can be stretched, compressed, or twisted without impairment of sealing capacity. If there is disruption of a tight junction at one point due to injury, there will be little effect on the overall tightness of the seals. It seems that the main function of tight junctions is to enable cells to maintain a different internal environment from that found in their surroundings.

Cells adhere to each other due to junctions called desmosomes. Desmosomes are necessary for cells to function as structural units. There are two types of desmosomes: belt-desmosomes and spot-desmosomes. The differences between the two types are due to their associated cytoplasmic filaments. Just below tight junctions, there are belt-desmosomes that form a band linking adjacent cells. This arrangement is conspicuous in epithelial cells. Within the zone of belt-desmosomes, there is an intercellular space filled with filamentous material. There are two sets of filaments: one forms a bundle running along the inside of the plasma membrane and the other originates near the junction and extends in a flat fashion into the cytoplasm. These filaments contain actin, which is the principal protein of muscle cells but is also found in all belted desmosomes. This suggests that filaments are capable of contraction. Contractions of filaments have been demonstrated in the presence of ATP and calcium or magnesium ions for belt-desmosomes of intestinal epithelial cells.

The second type of desmosomes is called spot-desmosome. Spot-desmosomes are different from tight junction and belt-desmosomes because they do not form bands around cells, but are arranged in button-like points of contact between plasma membranes at different levels, the analogy being to rivets or spot welds between plasma membranes. Within spot-desmosomes, the adjacent cell membranes are parallel, separated by a small gap of about 300 Å only. This gap is filled with filamentous material and is bisected by a dense line seen on electron microscopy, which is called the central stratum. A disk-like block is found on the cytoplasmic surface of each plasma membrane, and connected to these blocks are other filaments, which are called tonofilaments. They are not contractile but appear to be responsible for tensile and structural characteristics. The tonofilaments take their origin deep within the cytoplasm of cells, and they cross through the blocks of the spot-desmosomes. There are even thinner filaments that arise within blocks and project from the cell into the intercellular space, there to be connected to the central stratum. They form transmembrane linkers, which provide direct mechanical leverage between tonofilament networks of adjacent cells. There is yet a third type of desmosome that seems to be particularly common in epithelial cells. These cells are thought to be anchors for bundles of tonofilaments and are mainly concerned with the mechanical linkage of epithelial cells and connective tissue.

Direct communication between cells is possible by the transfer of chemical messengers. This occurs through gap junctions. The intercellular space is narrowed to about 30 Å. These junctions do not interfere with the flow between cells of various heavy metal tracers used in their study, and they have been shown on electron microscopy to consist of hexagonal arrays of cylindrical structures, which appear to form pipes or channels bridging the plasma membranes of two adjacent cells and the intercellular space. This arrangement allows the passage of molecules between cells but does not permit entry from the intercellular space into the cytoplasm. It conveniently also allows the free passage of intercellular molecules between the pipes or channels of the gap junction through the intercellular space. Molecules of up to 1,000 Daltons seem to pass readily from one cell to another, through the connecting channels of gap junctions and these include among many other messenger molecules such as steroids and cyclic AMP. In excitable tissue, the gap junctions are important in the transmittal of electrical signals. They are, therefore, also called electrotonic synapses. They account for the capacity of the electrical activity of one cell to be transmitted to an adjacent cell without the intermediary of a neurotransmitter. Such electrical transmission is instantaneous. Electrotonic synapses are, therefore, particularly common in tissues where rapid responses are necessary and synchronization is of importance, for example, in heart muscle and smooth muscle cells of blood vessels. Gap junctions are also of great importance in the function of the electric organs of fish who use electric discharges to stun their prey.

It is not clear yet how the integrity of junctions relates to health or disease, but their structure is correlated with the normal function of tissues. Thus, a tight junction provides an impermeable seal. Desmosomes seem to be important in structural support between adjacent cells, and gap junctions provide communication channels between the cytoplasm of neighboring cells. All types of junctions are important in the integrated activity of tissues, and their disruption might be associated with serious diseases (Staehelin & Hull, 1978) (Figs. 1.22–1.29).

Figure 1.22 Diagram of spot desmosome. Tonofilaments (diameter 100 Å) provide a tensile, strong network throughout the cell and are attached to cytoplasmic plaques. This type of junction couples the networks of tonofilaments of adjacent cells and allows dissipation of shearing stress throughout tissues. Drawn by M. Norviel.

Figure 1.23 Electron micrograph showing multiple forms of desmosomes. Tonofilaments run parallel and perpendicular (arrows) to the plane of apposed cell membranes at contact sites and at junctional specialization (® 49,000). Courtesy of Robert Kelly, Ph.D., Department of Anatomy, University of New Mexico School of Medicine.

Figure 1.24 Diagram of a tight junction. Sealing strands hold adjacent cells together through attachment lines. Closely spaced particles contributed by each cell form sealing strands that act like a modified zipper. The network of sealing strands gives this junction flexibility and allows maintenance of the seal under stress. The cytoplasmic filaments on the cytoplasmic surface reinforce the junction. Drawn by M. Norviel.

Figure 1.25 Freeze-fractured surface of tight junction illustrating sealing strands. An interconnected network of ridges (P face) (arrowhead) and complementary valleys (E face) (arrow) (® 82,000). Courtesy of Robert Kelly, Ph.D., Department of Anatomy, University of New Mexico School of Medicine.

Figure 1.26 Transmission electron micrograph. Numerous tight junctions between apposed cells surrounded by an electron-dense tracer (lanthanum nitrate). Membrane contacts are shown by the exclusion of tracer at sites of membrane fusion (arrows) (® 75,500). Courtesy of Robert Kelly, Ph.D., Department of Anatomy, University of New Mexico School of Medicine.

Figure 1.27 Diagram of gap junction. Nutrients and molecular signals can be exchanged between adjacent cells (double-headed arrow). The communicating channels are composed of six dumbbell-shaped protein units. Channels are about 20 Å in diameter. Fluids and tracers can penetrate gap junction by flowing around the pipes (curved arrow). Drawn by M. Norviel.

Figure 1.28 Transmission electron micrograph of gap junction (communicating junction) between apposed cells. Cell membranes are separated by a 4.0–4.5 nm gap at sites of close membrane apposition (® 50,000). Courtesy of Robert Kelly, Ph.D., Department of Anatomy, University