Mechanisms of Drug Action on the Nervous System (Cambridge Texts in Physiological Sciences)

Mechanisms of drug action on the nervous system

Mechanisms of drug action on the nervous system SECOND EDITION Ronald W. Ryall Lecturer in Pharmacology, University of Cambridge and Fellow of Churchill College, Cambridge

The right of the University of Cambridge to print and sell all manner of books was grunted by Henry VIII in 1534. The University has printed and published continuously since 1584.

CAMBRIDGE UNIVERSITY PRESS Cambridge New York New Rochelle Melbourne Sydney

Published by the Press Syndicate of the University of Cambridge The Pitt Building, Trumpington Street, Cambridge CB2 1RP 32 East 57th Street New York NY 10022, USA 10 Stamford Road, Oakleigh, Melbourne 3166, Australia © Cambridge University Press 1979, 1989 First published 1979 Second edition 1989 British Library cataloguing in publication data Ryall, Ronald W. Mechanism of drug action on the nervous system.—2nd ed. 1. Drugs affecting nervous system. Action. Mechanisms I. Title 615\78 Library of Congress cataloguing in publication data Ryall, Ronald W. Mechanisms of drug action on the nervous system / Ronald W. Ryall.—2nd ed. p. cm.—(Cambridge texts in the physiological sciences : 1) Bibliography: p. Includes index. ISBN 0-521-25424-8. ISBN 0-521-27437-0 (pbk.) 1. Neuropharmacology. I. Title. II. Series. [DNLM: 1. Nervous System—drug effects. QV 76.5 R988m] RM315.R9 1989 615'.78—dcl9 DNLM/DLC 88-20353 ISBN 0 521 25424 8 hard covers ISBN 0 521 27437 0 paperback (first edition ISBN 0 521 22125 0 hard covers ISBN 0 521 29364 2 paperback) Transferred to digital printing 2004

BO

To Audrey

CONTENTS

Preface to the second edition Preface to thefirstedition List of abbreviations

xiii xvii xix

1 Introduction

1

2 Techniques Routes of drug administration Systemic administration Local administration Electrophysiological methods Biochemical and histochemical techniques

7

PERIPHERAL NERVOUS SYSTEM 3 Neuromuscularjunction Techniques Synaptic transmission The acetylcholine receptor Activation of the receptor Sites of drug action at the neuromuscularjunction Prejunctional drug action Postjunctional drug action Pharmacological characterisation of neuromuscular blocking agents Myaesthenia gravis Denervation supersensitivity 4 Autonomic nervous system Neurotransmitters Drug action in the autonomic nervous system Ganglionic sites of action

14

43

viii

Contents

The structure and function of sympathetic nerves The metabolism of catecholamines The uptake and storage of catecholamines Receptors for noradrenaline Membrane and intracellular consequences of adrenoceptor activation Directly and indirectly acting sympathomimetic amines Inhibition of uptake mechanisms Miscellaneous drug actions The importance of uptake mechanisms in the actions of some adrenergic neurone blocking drugs Other antihypertensive drugs Denervation supersensitivity Cholinergic transmission at autonomic postganglionic nerve endings Muscarinic receptors Cholinesterase inhibitors

CENTRAL NERVOUS SYSTEM

5 Central neurotransmitters and neuromodulators Acetylcholine Amino acids Catecholamines and 5-hydroxytryptamine Polypeptides

80

6 The blood-brain barrier The nature of the blood-brain barrier Factors affecting rate of transfer of substances to and from the brain Developmental aspects Neurotoxicity Summary

93

7 General anaesthetics Types of general anaesthetic Gaseous anaesthetics

101

Contents

ix

Volatile anaesthetics Soluble (intravenous) anaesthetics

Mechanisms of anaesthesia Physico-chemical theories Difficulties with physico-chemical theories

Localisation of the effects of anaesthetics on neurones Pre- and postsynaptic effects Differential effects on excitatory neurotransmitters Effects on presynaptic inhibition Selective effects upon different areas of the brain and on spinal reflexes

Conclusions Tolerance to anaesthetics 8 Pain and analgesia Peripheral pain mechanisms Peripheral nerve fibres Activation of pain receptors and mediators The action of aspirin The action of capsiacin Central pain pathways Processing in the spinal cord Morphine-like analgesics Structure of morphine-like drugs Actions of morphine-like drugs The opiate receptor Localisation of the receptor Endogenous ligands for opiate receptors Analgesia and opioid peptides Multiple receptors for opioid peptides Involvement of opioid peptides in pain Sites of opiate action Descending control and analgesia Cellular actions of opiates Tolerance to opiates

118

9 Drug interactions with inhibitory amino acids Convulsants Anxiety-reduction and sedative-hypnotics

144

Contents

Benzodiazepines Pharmacokinetics Pharmacological actions

Benzodiazepine receptors Other anxiety-reducing, sedative-hypnotic drugs Anti-epileptic drugs Characterisation of epileptic seizures The use of drugs in epilepsy Pharmacological mechanisms General conclusions 10

Drugs used in schizophrenia

171

Theories of schizophrenia Drugs used in schizophrenia The dopamine receptor Multiple receptors for dopamine

Extrapyramidal side-effects of antischizophrenic drugs Mechanisms in drug-induced dyskinesias

Summary 11 Affective and manic depression Endogenous depression

193

Monoamine oxidase inhibitors Tricyclic antidepressants Other classes of antidepressants

Mechanisms of antidepressant action Long-term effects of antidepressants

Conclusions Manic depression 12 Disorders associated with defined brain lesions Spasticity Wilson's disease Parkinson's disease Drug treatment

Huntington's disease Biochemical and structural changes

203

Contents

xi

Treatment Alzheimer's disease Selected reading Index

217 225

PREFACE TO THE SECOND EDITION

Since the first edition of this book was published in 1979 there have been some major advances in all areas of knowledge concerning the physiology and pharmacology of the nervous system, although some advances are much greater than others. Each of these advances can on the whole be attributed to advances in technology. New technology for the measurement of receptor binding and immunological techniques, combined with the production of monoclonal antibodies, have been responsible for the greatly increased understanding and apparent complexity of receptors and their ligands. Only recently have some notes of caution been raised concerning the interpretation of ligand binding data as necessarily reflecting the properties of receptors. Advances in electrophysiological techniques now permit not only noise analysis of single channel events but also direct recording of the electrical currents flowing through those channels: I refer of course to single channel current recording with 'patch-clamp' techniques and the use of modern, high frequency, voltage-clamp amplifiers. Considerations of space in a book of this size and scope does not allow more than a brief mention of the technology. Finally, the age of the computer has brought with it considerable benefits, together with some consequential difficulties. Among the benefits is the ability to analyse complex events or to build up complex pictures of three dimensional objects which was not possible in an earlier generation. Complex experiments are now easier to perform than ever before. Among the potential hazards is the proliferation of trivial data which are impediments rather than aids to understanding. There has been an explosion of information concerning the presence of polypeptides in the peripheral and central nervous

xiv

Preface to the second edition

systems, but there is still a deficit in our understanding of their functions at specific locations. Despite immense efforts, a therapeutic advance from such studies has yet to appear. All too often excessive enthusiasm can lead to false conclusions: the mere presence of a particular substance in a nerve or that beautiful coloured picture of its distribution in nerve networks that stirs one's artistic imagination or the fact that binding sites exist, even when they are high affinity sites, does not preclude the possibility that the substance may normally function neither as a neurotransmitter nor as a neuromodulator. The next few decades will hopefully tell us whether the present has been guilty of the scientific crime of stretching conclusions beyond those justified by the data available. Two areas in particular have 'blossomed' since the first edition: these are the discovery of the wealth of endogenous opioid and other peptides and their receptors and the benzodiazepines and their receptors. In the case of the opioids both receptors and endogenous peptide ligands for them have been identified. As I write, I am conscious that fresh interest is currently being evoked in the possibility that morphine itself might be an endogenous ligand for the receptor! In contrast, the benzodiazepine receptors are probably still looking' for their endogenous ligands, although plenty of the synthetic variety are available and enthusiastic suggestions for candidates abound. In both of these areas there are drugs available which have a therapeutic use. I am intrigued that, despite the enormous amount of basic science that has been carried out, the increasing sophistication of technology and the tremendous increase in understanding of the actions of known drugs, there have been few major new conceptual advances in new drugs for the treatment of diseases of the peripheral or central nervous system over the last decade. One begins to wonder whether the 'sight of the wood is getting lost by the obscurity of the trees'. Put in another way, is there now too much emphasis on molecular mechanisms and too little on function and its disturbance in disease states? If the answer is yes then we shall continue to see an expansion in knowledge of the action of currently popular drugs but will see few new drugs developed.

Preface to the second edition

xv

However, there are some grounds for optimism. In the field of transmitters the excitatory amino acids are currently in vogue and are beginning to eclipse the 'conventional1 neurotransmitters and even the peptides. This renewed interest (attention was first drawn to them in the 1950s) is largely attributable to the production of interesting new compounds, rather than the study of old ones. There is a good involvement in the function of endogenous excitatory amino acids and their receptors and the current speculation on the possible role of NMDA receptors in memory and even Alzheimer's disease and stroke gives grounds for enthusiasm and hope for the possible emergence of radically new therapies. It is disappointing that no radically new therapies have been developed for the treatment of psychotic mental illness although in this case it is perhaps attributable to the poor understanding of the underlying neuropathology. There is hope, but little evidence on which to base it, for the development of new anti-schizophrenic compounds which interact with 5HT3 receptors. It is surprising that we can say little more than was possible ten years ago about the action of general anaesthetics. The difficulty may still be our lack of knowledge about the nature of the synaptic transmitters at specific synaptic locations, despite the extensive knowledge about how they work on isolated bits of membrane. The approach to this book and its objectives have changed little, if at all. Essentially, the objective has been to present a 'story' about each class of drugs which will give undergraduate students in science and medicine some broad insight into basic disorders of the nervous system and the way in which the drugs work to alleviate symptoms. If others find the book of value then this will be an added bonus since this was not the objective. A narrative style has been adopted and there is a deliberate avoidance of references in the text to help attain this objective. No attempt has been made to mention all drugs which are used: only those which have attributes making them worth special mention are presented. Nevertheless, the average student will not find that there are too many drug names to remember. He should take hope in the thought that before too long, especially if he is a medical student, each of these drugs, or perhaps a 'better' version, will soon become a 'household word'. The student is not spared con-

xvi

Preface to the second edition

troversy, where this is appropriate, although I would hope that I have not included too much. Most of the book has been rewritten and reorganised, although a few chapters are little changed. The section on the autonomic nervous system has been expanded considerably since this was not given adequate treatment in the first edition. The book now gives a fairly balanced perspective of the action of most classes of drugs which act on the nervous system. Stimulant drugs, such as amphetamine, ritalin, lysergic acid diethylamide, cannabis (is it a stimulant?) and cocaine are not particularly useful therapeutically, even though many of them are of importance as drugs of abuse. Too little is known about their mechanisms to give a coherent picture and too little space was available to do justice to the problem of abuse. However, references to some of them will be found sporadically through the book. Perhaps I can be excused for this deliberate omission.

PREFACE TO THE FIRST EDITION

In recent years there have been many important advances in knowledge concerning the mechanisms of chemical synaptic transmission, the identification of the neurotransmitters and the mechanisms by which drugs act on the nervous system. These advances have necessitated a change in approach to the teaching of the pharmacology of the nervous system to undergraduate science and preclinical medical students from a basically therapeutic orientation to one which is more mechanistically minded. In giving such courses to students in Cambridge, the author has become painfully aware of the need for an undergraduate text which could fulfil the needs of students in this respect. There are of course many excellent textbooks of therapeutics available but few of them attempt to cope in detail with mechanisms of drug action, especially on the central nervous system, except from rather specialised viewpoints. It was therefore considered to be unnecessary to discuss therapeutic applications in detail in this book, although an attempt has been made to give a fairly balanced account of the physiological basis, applications and mechanisms of action of each class of drugs, within the limitations imposed by the objective of producing a concise account of drug actions. Advances are occurring at such a rate that some of the concepts which are current today may be superseded tomorrow: this is probably true for any subject that is 'alive' and progressing. However, this does create problems in deciding what to omit and what to include. As far as possible, the basic approach adopted in this volume is to present a coherent 'story' which will enable the student to develop concepts and, perhaps, ideas of his own. Only in this way is it likely that a continuation of progress can be

xviii

Preface to the first edition

assured and that future medical graduates will not see drugs simply as liquids in bottles to be administered in an empirical manner without understanding to patients with diseases of the nervous system: a reasonable concept, compatible with contemporary information, even if subsequently found to be incorrect in detail, is surely better than no concept at all. Nevertheless, where concepts are relatively insecure, or mechanisms completely unknown, no attempt has been made to disguise this fact in order to present a 'story': such an approach could lead to unjustified complacency.

ABBREVIATIONS

ACh ACTH Adr AMP ANS ATP Bmax CAT (scan) CIO P-CCE CCK CNS COMT CSF DA DOPA d-Tc EC50/ED50 ECF ECT EEG EMG epp EPSP GABA GAD GDP GMP GTP

Acetylcholine Adrenocorticotropic hormone Adrenaline Adenosine monophosphate Autonomic nervous system Adenosine triphosphate Concentration at which all receptors are saturated Computer aided axial tomography Decamethonium P-carboline-3-carboxylic acid ethyl ester Cholecystokinin Central nervous system Catechol-O-methyl transferase Cerebrospinal fluid Dopamine Dihydroxyphenylalanine d-Tubocurarine Concentration/dose to cause 50% effect Extracellular fluid Electroconvulsive therapy Electroencephalogram Electromyogram End plate potential Excitatory postsynaptic potential y-Amino-butyric acid Glutamic acid decarboxylase Guanosine diphosphate Guanosine monophosphate Guanosine triphosphate

xx

Abbreviations

HC-3 5-HT 5-HTP IC50/ID50

KA KD K, LHRH LSD MAO (-A or -B) mepp MPTP NA Ni

Ns NMDA NMR 6-OHDA PET PG PTMA PTP SIF SPECT TEA TOH TRF VIP

Hemicholinium 5-Hydroxytryptamine 5-Hydroxytryptophan Concentration/dose to cause 50% inhibition Affinity constant = \/KD Dissociation constant Dissociation constant for inhibition Luteinising hormone releasing hormone Lysergic acid diethylamide Mono amine oxidase (A & B forms) Miniature end plate potential N-methyl-4-phenyl-1,2,3,6-tetrahydropyridine Noradrenaline Inhibitory regulatory protein Stimulatory regulatory protein N-methyl-D-aspartic acid Nuclear magnetic resonance 6-Hydroxydopamine Positron emission tomography Prostaglandin Phenyltrimethyl ammonium Post-tetanic potentiation Small intensely fluorescent neurone Single photon emission tomography Tetraethyl ammonium Tyrosine hydroxylase Thyrotrophin releasing factor Vasoactive intestinal polypeptide

Introduction

The junctions between nerve cells and either other nerve cells or the organs which they innervate are called synapses and form major sites for the targetting of new drugs with therapeutic effects and for the actions of toxic substances, including those produced by living systems. The synapses act as transducers, transforming the fairly constant energy of the action potential, which conducts nerve impulses from one end of the nerve to the other, into a variety of events upon the next cell in the chain. It is the variety of transmitters and modulators and the effects which they produce which enables drugs to act in relatively selective ways, either directly on the transmission process or the events which follow it. Side-effects are often due to unwanted actions upon some other system. However, it is perhaps less obvious that even if the drugs are highly specific in their actions then they may still have serious side-effects arising from the fact that individual processes are common to many nervous pathways, not all of which are equally affected by disease. Ideally, the development and design of new drugs for specific applications would arise from a thorough understanding of the basic physiology of the system and the pathology of the process which has led to dysfunction, together with a thorough understanding of the action of drugs and transmitters which affect the system. One would also know how modification to the structures of existing drugs would bring about changes in their pharmacological actions: the modern techniques of structure-activity relationships, as resolved by sophisticated computer simulations and calculations, can be considered to be in their infancy and have yet to lead to therapeutically useful drugs. Only rarely has a rational approach actually yielded important therapeutic

2

Introduction

advances, and it is significant that many of the drugs in current use, or their prototypes, are of plant origin and their major effects were known and exploited long before the advent of contemporary physiology, pharmacology or pathology. Serendipity has contributed a great deal to the contents of the modern pharmacopoeia and many of the drugs now used for one purpose were originally introduced for quite another. Despite these observations, the search for new drugs has undoubtedly yielded great improvements upon the activity of the prototypes and has occasionally, e.g. with the benzodiazepines, led to a completely different pharmacological approach to a therapeutic problem. Most of the drugs used to treat disorders of the nervous system produce an alleviation of symptoms, without curing the underlying disease. However, the alleviation of distressing symptoms is a worthwhile therapeutic objective and the curing of the disease and the reversal of the pathological changes which have already occurred is a tantalising goal to strive for. In some instances, the underlying pathology is fairly well established, as in the case of Parkinson's disease, and drug therapy rests upon a rational basis. In other illnesses, such as schizophrenia, little or nothing is known of the underlying pathological process although the pharmacological action of therapeutically effective drugs has been soundly evaluated. In such cases there is a tendency to develop a plethora of theories, most of them resting upon very insecure foundations, and the temptation to accept those theories without question must be strongly resisted. Uncritical acceptance can only slow the discovery of the real causes and produce complacency and a false sense of security. Although the sites of drug action will be evaluated in more detail in subsequent chapters, it is appropriate first to consider the question more generally. Most junctions between nerves and postjunctional structures, particularly in mammals, operate by the release of a chemical transmitter substance from the nerve termination acting upon specific sites or receptors upon the postjunctional membrane. Most of this book will be concerned with drug action at such sites. In addition, local anaesthetics and some neurotoxins impair impulse conduction in nerves by altering specific membrane conductances to particular ions and thereby

Introduction

3

will have effects on synaptic transmission, but they will not be considered in detail. Chemically transmitting junctions between a nerve and its postjunctional element are of two types. There are those in which the postjunctional element is a muscle or a gland and activation of the synapse results in either secretion or its inhibition, or contraction or its inhibition. There are of course also those synapses in which both the pre- and postjunctional elements are neuronal. In this instance the effect of the transmitter will be to modify the electrical excitability of the postsynaptic neurone so as either to excite it or to make it either more or less susceptible to excitation. Generally, the term inhibition is applied to a reduction in excitability of the postsynaptic neurone. However, the advent of the peptides as possible neurotranmitters has created some additional terminology in which the endogenous substance is said to modulate the excitability of the postsynaptic membrane. Such modulation may either increase or decrease the excitability of the cell. There are many potential targets upon which a substance could act at synapses. These targets are illustrated in Fig. 1.1, which forms the basis of much that will be described in this text. The sites of action may be broadly divided into prejunctional and postjunctional sites. At prejunctional sites there are susceptible processes involved in the conduction of the nerve impulses into the nerve terminal or with the entry of calcium ions into the terminal, which are both essential for the release of the neurotransmitter. Axonal transport may be involved not only in transporting transmitter-related substances, such as precursors, synthesising or degrading enzymes or even polypeptides manufactured in the cell body, but may also be involved in the transport of trophic factors responsible for maintaining the structural integrity of the junction. Such trophic factors may also control the efficacy of synaptic transmission in a dynamic fashion, perhaps by controlling, among other possibilities, the density of receptors upon the post- or prejunctional membranes. Other prejunctional processes include: uptake of precursors and the re-uptake of released transmitters; synthesis or degradation of transmitters or their precursors in cell membranes,

4

Introduction

organelles or cytoplasm; uptake into organelles in which there are storage mechanisms for the transmitter and subsequently the storage mechanism itself; the release mechanism, whether this be calcium-dependent, vesicular release or, as some believe, non-vesicular release; prejunctional receptors. When drugs act prejunctionally they must do so so efficiently that transmitter release is significantly decreased or increased. If the process upon which the drug acts is not rate limiting then the process of transmission will not be significantly altered unless Fig. 1.1. Cellular processes susceptible to drug action at chemically transmitting junctions. Axonal transport

Na + -

Synthesis and degradation Uptake into storage organelles Prejunctional receptors

Precursors Uptake"

Neuronal uptake Prejunctional re-uptake

oo Exocytosis and release 0 of transmitter Diffusion Postjunctional receptors

Postjunctional uptake

u

/ Intracellular 'messengers'

Postjunctional effect

Extrajunctional receptors

Introduction

5

the process is affected to such an extent that it does become rate limiting. Examples of such situations may be found in Chapter 4, which considers the biosynthetic pathway for catecholamines, in which the only rate limiting step is tyrosine hydroxylase. Postjunctionally, there are also many sites at which drugs could act. Again the process affected by the substance should either be rate limiting or should be so severely affected that it becomes so after drug treatment. The action of d-tubocurarine, acting upon skeletal muscle (Chapter 3), falls into this category. Here, the reduction of the end plate potential does not become rate limiting for the production of muscle action potentials until about 90% of the postjunctional acetylcholine receptors are occupied by the antagonist. The most common and best understood mechanism for postsynaptic action is upon specific subjunctional receptor sites. These sites may be considered to be chemically selective binding sites which modify the properties of the postjunctional cell. This may be effected either by opening or closing adjacent or integral ion channels, thereby affecting ion currents across the cell membrane, or by indirectly affecting function via a second messenger such as one of the cyclic nucleotides. Other postjunctional sites are also available. These include the enzymes which degrade the transmitter into inactive products; the enzymes may be located in the cell membrane or free within the cytoplasm or contained within or bound to the surface of cytoplasmic organelles. There may also be receptor sites at a distance from the junctional regions where the transmitter is liberated and the characteristics of the extrajunctional receptors are not necessarily identical to those located subjunctionally. Drug action, for obvious reasons, will usually involve reversible effects. Where the effects are not reversible the action of the drug is likely to be prolonged, limited by the de novo reconstitution of the inhibited process, or even toxic and possibly fatal if the process is essential, severely affected or perhaps completely eliminated. One can find examples of many such irreversible actions among the toxins and venoms produced by living organisms to their own peculiar advantage. Irreversible toxic effects could ensue from selective effects of

6

Introduction

the toxic substance upon particular junctions, e.g. 6hydroxydopamine selectively destroys adrenergic nerve terminals because it is selectively taken up by the adrenergic nerve terminal, or the toxicity may be more or less non-selective. Tetrodotoxin, by blocking sodium channels, will reduce the excitability of all electrically excitable tissues, whereas botulinus toxin, which prevents the release of acetylcholine at all cholinergic junctions, will selectively block the process of neurotransmission at all such junctions, both peripherally and in the central nervous system, but will leave unaffected adrenergic transmission or transmission mediated by other substances. Formation of free radicals could be the basis of some irreversible and toxic actions. The kinetics of drug action depends upon many processes. At the periphery, it may be limited by absorption from the site of administration, by binding to plasma proteins, by blood flow to the target organ, by rate of elimination from the plasma, either by excretion, metabolism or uptake by tissues, and by the kinetics of the drug-receptor interactions. For actions on the central nervous system, there is an additional barrier to access. This is the blood-brain barrier which for the most part behaves like a lipid membrane. The blood-brain barrier will be considered in detail in Chapter 6. If the substance is highly lipid soluble it will rapidly pass through the barrier at a rate dependent upon the concentration difference across it. If the lipid solubility is low then the rate of access will be determined not only by the lipid solubility but also by the blood flow. In summary, there are many possible sites of drug action at synapses. In practice, the most useful therapeutic substances will be those which affect processes peculiar to a limited range of junctions. Often the target will be the interaction between the endogenous transmitter or neuromodulator and its receptor sites, but other selective processes such as transmitter release or degradation may also yield the required degree of selectivity in drug action. Unfortunately, few drugs are absolutely specific in their effects. Examples of specific and less specific drug actions will be found throughout this book.

Techniques

In this chapter only some of the techniques which can be employed to investigate the action of drugs upon the nervous system and the process of synaptic transmission will be examined. There is no space in a book of this size to describe the many techniques in detail and no attempt is made to discuss behavioural techniques which are frequently employed in the evaluation of drugs acting on the central nervous system. The emphasis will be on the more recent developments in technology. Many of the techniques are common to studies of the peripheral or central nervous system. However, special techniques are required to overcome the problems associated with the access of drugs to the brain and the determination of the site of action. Since the first edition of this book was written, there have been a number of advances in the development of in vitro techniques for examining the properties of slices of brain tissue (brain slice techniques) and for maintaining cultures of neuronal tissue. These have enabled us to bypass the problems associated with the access of drugs to the brain although they have introduced some additional problems of their own. There have also been advances in electrophysiological techniques for examining the properties of single ion channels in isolated segments of cell membranes in vitro. It is likely that future major progress will accrue from the development of the techniques to the point when they can be applied to intact, functioning tissues. In other disciplines there have been advances in techniques for visualising receptors and enzymes and in synthetic methods for determining the structure of complex molecules.

8

Techniques

Routes of drug administration The route of drug administration is particularly important in studies of the actions of drugs on the brain since access of drugs to this structure may be limited. The access via the bloodbrain barrier is discussed in more detail in Chapter 6. Systemic administration The easiest way to give a drug is into the general circulation. An intravenous injection will give a rapidly equilibrating concentration in the blood. The blood levels attained during experiments can be compared with the therapeutic levels attained in man and in this way the relevance of experimental data to the therapeutic situation can be assessed. Oral administration, or other routes such as subcutaneous or intra-peritoneal injections may also be employed when appropriate but absorption via these routes is likely to be slower and more variable. Many of these routes may be adequate for study of peripheral mechanisms but may not be appropriate for the study of the actions of some drugs upon the central nervous system to which they may not have easy access via the blood stream. Clinically useful drugs will usually have access since clearly there is no other way in which they can be administered except occasionally under very supervised conditions, e.g. by direct injection into the cerebro-spinal fluid. Other substances, such as most neurotransmitters, have a very limited ability to cross the blood-brain barrier and so other means need to be found to administer them in order to study their effects. There are disadvantages with systemic administration when attempting to define the locus and mechanism of drug effects with any precision, because the drug may be exerting its effect indirectly on that part of the system under study. It may even be producing an effect on CNS function by changing the inflow of sensory information. Local administration A more restricted locus of drug action may be obtained by a variety of techniques. These include injection into or perfusion of the cerebro-spinal fluid (CSF), local irrigation of superficial parts of the brain with solutions of known composition, local

Routes of drug administration

9

perfusion of deeper areas by perfusion through 'push-puir cannulae, microinjection of small amounts into defined areas and microelectrophoresis, microelectro-osmosis or pressure microejection from micropipettes. The most restricted site of drug action is probably attained by those techniques in which the substance is ejected from micropipettes. All of these techniques suffer the disadvantage that the concentrations attained in the tissue are unknown, although they can often be estimated, but the rate of administration or the amount injected can be regulated and monitored. The more localised the administration the more uncertain is the concentration which is effective. This is because a concentration gradient is set up with the highest concentration at the focus of the injection and concentrations falling off exponentially with distance at a rate dependent upon the diffusion of the substance in the medium and factors affecting its rate of removal. It has commonly been assumed that the microinjection of small quantities, when small is usually defined as a few micrograms, produces low concentrations. It can readily be calculated that in fact within millimetres of the injection site the concentration can be in the millimolar range. In pharmacological terms such millimolar concentrations are usually considered to be high. This is the penalty that one perhaps has to pay for localised administration. In contrast, low, known concentrations can be established by in vitro techniques in which parts of tissues are bathed in solutions containing known drug concentrations. However, such techniques have their own but quite distinct limitations. Even when slices are superfused with solutions of known concentrations, the concentration at the receptor sites within the tissue will be lower than that in solution. It is still uncertain how far one may extrapolate data obtained from the highly artificial environments of slices and cultures of tissue, particularly the latter in which the cells may have changed their properties during the culturing process, to the functioning nervous system in vivo. Furthermore, there is the major drawback that much but not all of the functional integrity of the nervous system, upon which the nature of the system depends, is lost in these isolated systems. Nevertheless, the techniques have permitted answers to be obtained to

10

Techniques

some very fundamental questions. Provided that the assumptions made in extrapolating from data obtained with in vitro techniques are always kept in mind, then the in vitro techniques have undoubtedly led to some important advances and will continue to do so for some time yet. It may be concluded that no one technique for drug administration is ideal or will provide all of the answers that are needed. Advances will continue to be made by laboratories working with a variety of techniques. Electrophysiological methods The electrical activity of the nervous system can be observed in many ways. It is a relatively simple matter to record action potentials in peripheral nerves: in recent years it has been possible to record from or stimulate single nerve fibres in man. It is also relatively easy, even in man, to record spinal reflexes with electrodes placed upon the peripheral nerves. However, such recordings will generally tell us little about how a drug can produce a change in the record. The recording of the electrical activity of large areas of the brain, expressed as electroencephalograms (EEGs) or as evoked potentials has also been widely used and is of some value in the diagnosis of disorders or in the predictions of possible clinical applications of new drugs. However, EEGs and evoked potentials tell us little about the precise localisation or mechanism of drug actions. Computer technology has been applied to EEG recording, and computer-generated maps of activity in the brain may now be obtained. However, it must be remembered that an EEG represents the summed activity of many neurones in a relatively large part of the brain: asynchronous neuronal activity is therefore largely unobserved. At the next level of resolution, electrolyte-filled microelectrodes have been used for several decades to record the activity of single nerve cells in situ. These recordings may be obtained from the just extracellular environment and consist of recordings of action potentials. These recordings cause the least change to the neurones under observation, but are limited to recordings of the changes in the frequency or latency of firing of the cells. Since the action potential is the only line of fast com-

Electrophysiological methods

11

munication in the nervous system, this limitation may not be so important. Nevertheless, it does not allow a deeper understanding of the intracellular and membrane processes involved in drug action. Major advances in this area have been made with studies in which the pipette is stationed inside the cell membrane (intracellular recording). The technique can be employed in vivo as well as in vitro. It is possible to record either the voltage changes across the membrane or the current fluxes when the membrane voltage is 'clamped' to known potentials. Voltage clamp techniques have considerably advanced in recent years to the state at which the electronic circuitry will now stabilise voltage changes occurring at frequencies up to about 30 kHz, a frequency which is at the upper end of the biological frequency range. In principle, the voltage clamp technique is quite simple. The potential is monitored with a micropipette stationed inside the cell. Any tendency to change is counterbalanced by ejecting a current through either the same micropipette, or an independent pipette, of such a magnitude and polarity as to just counterbalance the change: this is the technique of negative feedback. The amplitude of the feedback current is equal and opposite to the change in membrane current induced by changes in membrane properties. The monitoring of membrane currents and voltage enables us to estimate the changes in electrical resistance and conductance of the cell membrane to ions. The great advantage of monitoring changes in membrane currents, rather than membrane voltage, is that it is now known that certain membrane ion conductances are sensitive to changes in membrane voltage. Voltage clamping enables studies to be made without needing to consider this complicating factor. Such techniques have given insight into the manner in which neurotransmitter substances may control neuronal activity. The final stage of resolution has been achieved in recent years by attaching a very small piece of cell membrane to a micropipette orifice with a very tight seal. This is the technique of fc patch-clamping\ The small area of membrane ensures that only one, or at most a few, ion channels in the membrane will be conducting (open) at any one instant. The tightness of the seal between membrane and micropipette ensures that the small current

12

Techniques

flowing through the single ion channel can be measured. Although most studies have been carried out on pieces of membrane which have been detached from the cell, it is also possible to make whole cell patch clamps. It has not so far been possible to make useful studies in vivo. It is possible to change the composition of the medium on either side of the membrane. The technique has given much insight into the changes in single ion channel conductances which accompany neurotransmitter action. Biochemical and histochemical techniques One of the most widely used techniques developed in the last decades is the ligand binding technique. Specific, saturable binding of specially prepared ligands to binding sites in nervous tissue has yielded a great deal of information on drug receptor interactions. However, there is still a tendency to equate ligand binding with receptor binding and this has led to a certain confusion and in some cases to an unjustified multiplicity of postulated receptor sites. Awareness of the problem should ultimately lead to a solution. The important thing to remember is that a binding site cannot be assumed to be the same as a receptor site until it has been unequivocally shown that the occupation of that binding site correlates with changes in tissue response which are identical to those caused by receptor occupation. Histologists have been most active in devising methods for visualising binding sites in nervous tissue. In this they have been greatly aided by the development of immunohistochemical methods for labelling binding sites with immunoreactive substances. The specificity of the antibodies used has in turn been dependent upon advances in cloning techniques for preparing monoclonal antibodies. In general, the results have corroborated those obtained with more conventional techniques but they have allowed extensive studies to be made of the distribution of binding sites for polypeptides for which no other techniques were available. Biochemical techniques now being used to study the nervous system include the countless studies on the role of cyclic nucleotides and the more recent studies on the phosphoinositol system,

Biochemical and histochemical techniques

13

which may be involved in the action of various neurotransmitters, especially in the mediation of the muscarinic actions of acetylcholine. Among the most recent developments are the techniques of CAT (computer aided tomography) and PET (positron emission tomography) scanning. These techniques are improving our knowledge of brain structure and changes in disease. Both techniques rely heavily upon sophisticated computer technology to build up complex three-dimensional images of some aspects of brain structure or function. PET scanning has the potential to tell us about changes in function in selected neuronal systems in living subjects. Although in their infancy, high hopes are held by some that these techniques will provide important answers to questions which are at present incapable of being answered by other techniques. Nevertheless, some caution must be expressed since the technology is exceedingly expensive and its use requires careful justification. Of course, there are many other techniques being used by biochemists to study the function of the nervous system and the action of drugs on it, but these are too numerous to mention here: reference to them will be found at appropriate places in the text.

Neuromuscular junction

The neuromuscular junction between motor nerve fibres and skeletal muscles is a good place at which to begin a study of the ways in which drugs can influence the processes of synaptic transmission. Other synapses may be considerably more complex. At the neuromuscular junction the system is both anatomically and functionally relatively simple and a functional system can be isolated from the whole organism. The majority of mammalian skeletal (striated) muscle fibres are focally innervated by motor axon terminals, each one forming a single junction on any one muscle fibre at the motor end plate. A motoneurone in the ventral horn of the spinal cord gives rise to a single axon which branches within the muscle to innervate from one to six muscle fibres, these groups of muscle fibres together comprising a single motor unit. This motor unit therefore behaves as a single functional entity (Fig. 3.1). The large, extrafusal muscle fibres which generate the major component of the force developed in muscle contraction are innervated by large a-motoneurones through myelinated A-fibres. The small intrafusal fibres of the muscle spindles are innervated by the small y-fibres of motor nerves. Some of the muscle fibres in amphibia and in birds, e.g. the rectus abdominis muscle in the frog and the biventer cervicis muscle in the chick respectively, have multiple endings upon them and this results in important differences in their physiological responses to nerve stimulation and in their pharmacological responses to drugs.

15

Neuromuscularjunction

Motor end plate Axon

Synaptic vesicles containing ACh Mitochondrion

Myelin sheath

Myofibrils

.. Spinal motoneurone Peripheral motor axon Single end plates Motor unit (focal innervation) Terminal branches Individual muscle fibres

Multiple innervation Separate motor axons

Multiple end plates

- Muscle fibre

Fig. 3.1. Innervation of skeletal muscle. After Couteaux, R. (1958). Exp. Cell Res., Supply 5, 294.

16

Neurom uscularjunction

Techniques There are many ways in which the actions of drugs may be studied upon neuromuscular transmission. These range from the simplest of techniques, in which recordings are made of the contractions of the muscle to stimulation of the peripheral end of the sectioned motor nerve and of the responses to administered drugs, either in vitro or in vivo, to the most sophisticated modern techniques of intracellular or patch clamp recording. The electromyogram (EMG) can be recorded in vivo via a metal electrode inserted into the muscle. Depending upon the size of the exposed electrode tip, recordings of either gross muscle action potentials or of the action potentials generated by a single motor unit can be obtained. When the muscle is maintained in isolation, perfused with physiological salt solution, then intracellular recording of end plate, miniature end plate and muscle action potentials becomes possible with fine capillary microelectrodes filled with potassium chloride solution. In order to record the miniature end plate potentials (mepps) or end plate potentials (epps), the micropipette must be located close to the end plate region. Modern voltage clamp amplifiers are able to clamp the fast voltage transients of these intracellular potentials (see Chapter 2) thus enabling the currents flowing through the membrane to be determined. The techniques of noise analysis and of patch clamp recording have been described in more detail in the previous chapter but were first applied in studies of transmission at the neuromuscular junction, and have greatly increased our understanding of the way in which transmission is effected and of the ways in which it is modified by drugs. Synaptic transmission The transmitter at the neuromuscular junction is acetylcholine (ACh) which is packaged in vesicles in the presynaptic terminals. The arrival of an action potential in the terminal causes a transient increase in the permeability of the terminal membrane to Ca2+, which in turn precipitates the simultaneous liberation of ACh from a number of vesicles. The ACh content of each vesicle produces a single miniature end plate po-

Synaptic transmission

17

tential (mepp), after diffusing across a narrow synaptic cleft of about 150 A in width and combining with specific receptors located at the end plate region on the postsynaptic membrane. The simultaneous release of many vesicles, of quanta, of ACh results in the production of the much larger end plate potential (epp). In normal muscles the receptors for ACh are only found at the junctional regions of the muscle fibres, but after denervation or before innervation during development, the receptors are far more diffusely located over the muscle fibre membrane. The activation of the receptors by ACh gives rise to local changes in membrane permeability and to local changes in membrane potential which do not propagate along the muscle fibres but only decay electrotonically, i.e. the miniature and end plate potentials decrease exponentially with distance along the fibres, the rate of decrease in amplitude being dependent upon the properties of the muscle fibres. In muscle fibres which are focally innervated, the epp gives rise to a propagated muscle action potential which in turn invades the transverse tubular system, to produce a translocation of Ca from the sarcoplasmic reticulum, so activating the contractile mechanism. At low frequencies of stimulation there is evoked a single, all or nothing twitch, but at higher frequencies of 30 to 100 Hz the twitches fuse to produce a tetanic contraction, many times the amplitude of the all or nothing twitch. The tetanic contraction at high frequencies of stimulation is due to the incomplete relaxation of the sliding filaments between each arriving nerve impulse. The transmitter at multiple innervated junctions is released from many sites on each fibre and at each site there is a local non-propagated epp which does not give rise to a propagated action potential. Instead the epps summate to produce slow and graded depolarisations, evoking in turn slow and graded contractures of the muscle fibres. After dissociation of the ACh-receptor complex, the ACh is hydrolysed by the enzyme acetylcholinesterase which is located predominantly in the postjunctional folds of the muscle fibre membrane. The by-product, choline, is either removed in the circulation or else taken up again by the nerve terminals by an active transport process: choline is the precursor of ACh synthesised by

18

Neuromuscularjunction

acetylcholinesterase. In the absence of electrical activity in the motoneuronal axon, individual vesicles release their content of ACh in a random fashion, and each unit amount of ACh produces a quantal change in the membrane potential. The amplitudes of the spontaneous miniature potentials range in size from about 0.1 to 0.7 mV, due to random liberation of single or multiple quanta of transmitter, and there are slight variations between different species. Each quantum probably represents the ACh released from a single vesicle. When a nerve impulse arrives in the terminal there is a sudden release of ACh from about 100 to 200 vesicles, causing a large epp to occur after a delay of about 0.5 ms. The synaptic delay is made up of a short period for the ACh molecules to diffuse across the cleft (about 0.15 ms) and a somewhat longer period required for the release process to be completed. The acetylcholine receptor Peripheral receptors for ACh fall into two types, characterised by their biochemical and pharmacological properties. Muscarinic receptors are activated by ACh itself, by some other choline esters such as acetyl-p-methylcholine and by the cholinomimetic substance muscarine. These muscarinic receptors are blocked by the antagonist atropine. Muscarinic receptors are found in effector organs innervated by postganglionic parasympathetic autonomic fibres and at a few Fig. 3.2. The acetylcholine receptor complex. Glycoprotein subunits a

65 000 /

NA_/

/

\ 40 000

50000 Hydrophilic pore

The acetylcholine receptor

19

other locations, including the central nervous system, as discussed elsewhere in this book. Nicotinic receptors are found at the neuromuscular junction and in autonomic ganglia and at a few sites in the central nervous system. At the periphery, the nicotinic receptors are activated by ACh and by nicotine, but not by acetyl-p-methylcholine, and are blocked by curare-like substances at the neuromuscular junction or by hexamethoniumlike substances in autonomic ganglia. It is therefore evident that there are differences between those nicotinic receptors found at the neuromuscular junction and those located in ganglia. This is further exemplified by the fact that the receptors at the neuromuscular junction are selectively activated by phenyltrimethylammonium, whereas those in ganglia are selectively activated by dimethylphenylpiperazinium. The nicotinic ACh receptor has been biochemically isolated in relatively abundant quantities from the electroplaque organ of the Torpedo; this organ is composed of stacks of motor end plates, arranged as a series of batteries, and capable of generating upon command from the central nervous system rather unpleasantly high voltages to shock unwary intruders. Immunological studies indicate that the receptor isolated from Torpedo is virtually identical to those found elsewhere. The nicotinic ACh receptor is a complex of four different glycoprotein subunits (a,p,y,5), arranged in a rosette of five subunits in which the a-unit is represented twice. The overall molecular weight of the complex is about 250 000 daltons and the molecular weight of each subunit is about 40 000 to 65 000 daltons. The subunits are each translated from a separate messenger RNA on ribosomes of the rough endoplasmic reticulum. The five subunits in the receptor are arranged around a central hydrophilic ion channel in the muscle membrane (Fig. 3.2). The toxin abungarotoxin, about which we shall hear more later, and other cholinergic receptor ligands bind with high affinity to the asubunit but the role of the other subunits is unknown. There is a very high density of receptors at the end plate region. The subsynaptic density is about 104 receptor complexes per square micrometre and the receptors extend, at reduced densities, for about 400 juim from the end plate.

20

Neurom uscularjunction

Activation of the receptor The interaction of ACh, or of any other suitable agonist, with the nicotinic receptor causes an increase in the permeability of the postjunctional membrane to cations, especially to Na + and K\ with a size limit of not more than twice the size of the hydrated K+ ion. The influx of Na+ ions causes a depolarisation of the postsynaptic membrane. The amplitude of the depolarisation depends upon a number of factors, including the concentration of ACh in the region of the receptors, the receptor density and the duration of the lifetime of the ionic channels opened by the transmitter-receptor interaction. The smallest mepp observed at the end plate has an amplitude of about 0.3 mV and corresponds to the net effect resulting from the opening of many channels activated by the release of the transmitter content of one vesicle, i.e. about 10-50 000 molecules of ACh. There have been two major advances in the last decade which have greatly increased our understanding of the molecular events accompanying receptor activation at the neuromuscular junction. These advances have utilised the two new techniques of noise analysis and patch clamping described in Chapter 2. Katz & Miledi were the first to approach this problem with noise analysis, which is essentially a statistical analysis of the small variations in membrane potential or current observed with the now relatively gross technique of intracellular recording at the end plate region. If each interaction between ACh molecules and receptors causes a constant and equal channel opening, so permitting a fixed charge to cross the membrane, the depolarisation produced by this unitary event (shot effect) is given by the expression: a = 2EVV mV where 'a' is the elementary depolarisation, E2 is the statistical variance of the membrane potential about the mean value of the depolarisation (V) caused by the external application of ACh. It will be noticed that V is independent of the concentration of ACh employed and this was confirmed experimentally. The amplitude was about 0.3 jwV in frog muscle and 0.7 jaV in rat muscle. The current flowing through an open channel is of the order of 2

Activation of the receptor

21

Table 3.1. Membrane noise and depolarisers of the neuromuscular junction (frog sartorius) a ACh Carbachol Suberyldicholine Acetylthiocholine Decamethonium

JJV

0.3 0.1 0.4 0.08 0.05

rms 1.0 0.3-0.4 1.65 0.12 0.1

After Katz & Miledi (1973). a is the elemental shot-effect; r is the duration of opening of the ionic gate.

pA and is independent of the cholinomimetic employed. However, different agonists evoke different amplitudes of elementary depolarisations (see Table 3.1). This is due to the fact that the duration of the channel opening produced by different agonists is not the same, as is also shown in the table: the duration of the channel opening was measured from the spectral density of the current noise recorded with a focal electrode . For ACh the value of Y, the channel open time, was 1 ms. The passive electrical properties of the cell membrane slow down the processes of depolarisation and of repolarisation so that the time constant for the decay of the depolarisation produced by ACh is about 10 ms, i.e. it is much longer than the channel open time (Fig. 3.3). The high potency of carbachol in depolarising the membrane, despite the small magnitude and brief time course of the elemental depolarisation, is due to its resistance to acetylcholinesterase, possibly allowing repeated interactions of a single agonist molecule with the receptor. Recording of membrane current noise under voltage clamp conditions, rather than of membrane depolarisation, gave direct measurements of the open channel lifetimes which agreed with the earlier results of Katz & Miledi in their pioneering studies. Investigations of the molecular events at the neuromuscular junction have now reached a new level of sophistication with the

22

Neurom uscular junction

recent introduction by Neher & Sakmann of the elegant technique of patch clamping. With this technique the amplitude and duration of the molecular events initiated by a single ACh-receptor interaction can be observed and measured directly, rather than by statistical inference. In general, the main conclusions reached by Katz & Miledi have been confirmed. However, the technique has also revealed other interesting phenomena for which there is at present no certain explanation. Among the more unexpected findings there are some which seem to show that there are multiple conductance states of a single ion channel and that there are 'flickering' transitions between open and closed states. Clearly this new technique is bringing a wealth of new information which may in the future bring new insights into the molecular events accompanying drug receptor interactions. Already these studies at the neuromuscular junction have spurred on efforts to study similar phenomena at other synaptic junctions, particularly those in the central nervous system. Sites of drug action at the neuromuscular junction Drugs affecting neuromuscular transmission may act at one or more of three possible sites and these are summarised,

Fig. 3.3. Diagram showing single channel currents flowing when recordings are made from one (upper record) or two (lower record) channels simultaneously. The large current steps, labelled c, are the algebraic sums of the two individual steps, a and b.

Sites of drug action at the neuromuscularjunction

23

together with examples in Fig. 3.4: i) On axonal conduction of the impulse into the nerve terminal, ii) On presynaptic terminals, affecting transmitter storage, release or synthesis. iii) On the postjunctional cell, either at the end plate or on the contractile mechanism at a step beyond the end plate. Therapeutically useful drugs which cause muscle relaxation are to be found among those acting on the motor end plates, producing either non-depolarising blockade of transmission or depolarising, desensitising block. A therapeutically useful facilitation of transmission is produced by neostigmine or germine, acting either as inhibitors of acetylcholinesterase or by facilitating transmitter release (see below). Actions at other sites are produced by a variety of toxins and venoms and sometimes as a side-effect during therapy with certain antibiotics. Prejunctional drug action Conduction. Although local anaesthetics block nerve conduction, they non-selectively block fast, voltage-dependent, sodium channels in all excitable tissues and so will not be considered here. Anticholinesterase agents and, in particular, neostigmine, are used in the treatment of myaesthenia gravis. The beneficial facilitation of transmission produce is most likely due to a number of factors. The inhibition of acetylcholinesterase reduces the breakdown of released ACh and so facilitates transmission when the concentration of the transmitter is rate limiting. However, neostigmine also causes a repetitive firing of action potentials both in the nerve terminals and in the muscle fibre in response to a single motor nerve volley. The depolarisation of the terminal is sufficient to cause impulses to propagate antidromically up the axon. Although the details of the mechanism are not certain, it does not seem to be related to inhibition of cholinesterase. Similar repetitive firing of action potentials is known to be caused by germine, which is not a cholinesterase inhibitor but is also useful in myaesthenia. Thus both inhibition of acetylcholinesterase and repetitive firing of action po-

24

Neuromuscularjunction

tentials may contribute to the mechanisms by which these drugs are effective in myaesthenia gravis. Germine, used as the monoacetate, is a veratrum alkaloid. It does not change the sensitivity of the muscle to ACh nor does it change the frequency of mepps or the amplitude of epps in doses which cause repetitive firing in frog muscle, but there may be a slight increase in mepp frequency in the diaphragm muscle of the rat. The repetitive firing is probably due to a delayed closure of Na+ channels, although other mechanisms have not been excluded. Tetrodotoxin blocks neuromuscular transmission by blocking the voltage-dependent fast Na+ channels in all excitable membranes, including the presynaptic terminals and muscle fibres, so preventing the initiation of action potentials. Fig. 3.4. Sites and structures of drugs acting at the neuromuscular junction. Presynaptic Repetitive activation:

neostigmine

germine CH 3

OCN(CH 3)2

S

JL.CH3

(CH3)3

^J^^^ O H

CH 3 OH

X

OH

OH Synthesis: diphenylbutyl acetate, hexamethylene-l,4-(l-naphthylvinyl)pyridinium-6-trimethylammonium hemicholinium (HC-3)

Uptake:

CH 3 CH 3 Storage: black widow spider venom, l-bungarotoxin Mobilisation: neomycin, streptomycin Release: botulinus toxin

CH3 CH 3

Sites of drug action at the neuromuscular junction Postsynaptic Cholinestcrase inhibitors: edrophonium

neostigmine O II OCN(CH 3 ) 2

^V/^N + (CH 3 ) 3

OH

^X^N-C2H CH3

Competitive blocking agents (reversible): d-tubocurarine (d-TC) OH

CH 3 O

X

Q

OCH3

CH3 CH3

pancuronium

O-CH 2 CH 2 N + (C2H5)3 -O-CH 2 CH 2 N + (C 2 H 5 ) 3 ^O-CH 2 CH 2 N + (C 2 H 5 ) 3

Competitive blocking agents (non-reversible): a-bungarotoxin Desensitising blocking agents: decamethonium (CH3)3N+(CH2)1ON+(CH3)3

succinylcholine

(CH3)3N+CH2CH2OCCH2

I

(CH3)3N+CH2CH2OCCH2 O Metaphilic antagonists (reversible): dinaphthyl decamethonium (DNC-10). Metaphilic antagonists (non-reversible): DNC-10 mustard

25

26

Neuromuscularjunction

Synthesis of transmitter. Choline acetyltransferase is blocked nonselectively by a number of substances such as diphenylbutyl acetate and some styrylpyridine analogues. The most potent of the latter substances is hexamethylene-1, 4-(l-naphthylvinyl)pyridinium-6-trimethylammonium, but this substance blocks acetylcholinesterase in vitro to about the same degree. The block of neuromuscular transmission which occurs in vivo is probably related to a direct effect on the contractile mechanism rather than to an inhibition of choline acetyltransferase. Inhibition of choline uptake. Hemicholinium (HC-3) is a quaternary ammonium substance which prevents the synthesis of ACh by blocking the uptake of the precursor, choline, into the terminal. In early experiments with HC-3 it was noted that intravenous injection in rabbits caused muscle relaxation and respiratory paralysis which was aggravated by exercise. It was erroneously thought that the effect was due to an action on the central nervous system, even though such a quaternary compound, being fully ionised, would be unlikely to cross the blood-brain barrier. Since synthesis of ACh is only rate limiting for transmission in normal muscles at high rates of stimulation, HC-3 has little effect on contractions of a muscle excited through its nerve at low frequencies of stimulation but there is a marked decline in the twitches if the rate of stimulation is increased to about 1 Hz and the reduction can be effectively counteracted by the intravenous administration of choline. This indicates that HC-3 blocks transmission by competing with choline at uptake sites on the terminal membrane. The conclusion is supported by the observation that the sizes of mepps as well as epps are reduced by HC-3, that the measured output of ACh on stimulation is diminished, and by biochemical determinations of choline uptake in the presence and absence of HC-3 in vitro. At high doses, HC-3 may also have a postjunctional, curare-like effect. Triethylcholine is transported almost as readily as choline and blocks transmission in a manner similar to that of HC-3. An earlier suggestion that triethylcholine might be acetylated, stored and then subsequently released as a false transmitter is only of historical interest as the first reported suggestion that substances

Sites of drug action at the neuromuscularjunction

27

taken into nerve terminals could act in this way. In fact, attempts to isolate the acetylated derivative have been unsuccessful. Tetraethylammonium (TEA) also has an action like that of triethylcholine but is better known for its ability to block some potassium channels. It may also have some postsynaptic blocking action, especially pronounced in autonomic ganglia. Storage of ACh. A number of venoms are known to prevent the storage of ACh in the presynaptic vesicles. In frog muscle, black widow spider venom causes at first a great increase in the number of mepps recorded with an intracellular electrode and this is followed by a block of transmitter output and a block of neuromuscular transmission. Electron microscopy has shown that at this stage the vesicles have disappeared. In cats, similar effects are followed by a complete disintegration of the terminals. (3-Bungarotoxin, obtained from the venom of a Formosan snake, and not to be confused with a-bungarotoxin which has a postsynaptic site of action, has a presynaptic action akin to that of the venom of the black widow spider. Mobilisation of transmitter. Not all of the ACh in the terminal is immediately available for release, much of it needing first to be 'mobilised'. Mobilisation, which may involve the movement of vesicles up to the release site on the presynaptic membrane, is dependent on the frequency of stimulation, is increased by an increase in extracellular Ca2+ or by depolarisation of the terminals by potassium ions. Some antibiotics, notably neomycin, streptomycin and kanamycin, occasionally cause muscle weakness during antibiotic therapy; there is some evidence to suggest that the weakness is caused by a reduction in transmitter mobilisation leading to a decrease in the quantal content of the epp. Release of ACh. A highly toxic substance with a molecular weight of about 60 000 is produced by the anaerobic bacterium Clostridium botulinum. Botulinus toxin produces neuromuscular paralysis by virtue of the fact that it very effectively prevents the liberation of ACh from the terminals without destroying the

28

Neuromuscular junction

vesicles. It therefore acts upon the release process. The fatal dose for a mouse is about 2 X 10"10 g kg"1. Making the assumption that a mouse has about 106 muscle fibres, Burgen has estimated that the fatal dose of the toxin would correspond to an availability of about 40 molecules of the toxin at each end plate! Even though the toxin is poorly absorbed from the gastrointestinal tract, the high potency is sufficient to ensure that is is easy to ingest toxic quantities. The toxic effects of botulinus toxin are attributable to a paralysis of cholinergic systems throughout the body, including not only those at the neuromuscular junction but also those of the autonomic nervous system. The symptoms include paralysis of eye muscles, causing diplopia, ptosis, dilated pupils and an inability to accommodate. Later there is muscle weakness, respiratory distress, difficulty in swallowing or speaking, constipation and urinary retention. The toxin does not affect the sensitivity of the membrane to ACh. There is no effect on the arrival of the impulse in the terminal nor is there any effect on the quantal size of the mepps. However, the frequency of the spontaneous mepps is greatly reduced and stimulation of the motor nerve elicits only a small or no epp and the assayed output of ACh declines. This is all consistent with the proposed mechanism of action. Simpson has suggested that the action of the toxin is dependent on ACh release. The block of transmission is prevented by maintaining a low concentration of Ca2+ in the medium to block ACh release, or by a maintained depolarisation of the terminals. However, it is more likely that the toxin blocks Ca2+ entry to the terminal, upon which the ACh release is dependent, or else changes the coupling between calcium entry and release from the vesicles: the block may be alleviated by the use of calcium ionophores, which promote the entry of Ca2+ or by TEA which prolongs action potentials by blocking repolarising potassium currents, so increasing the voltage-dependent Ca2+ entry. Guanidine sometimes alleviates muscle weakness in myaesthenia. It increases epp amplitude but there is no change in the size or frequency of the mepps. There is no effect on the

Sites of drug action at the neuromuscularjunction

29

postjunctional sensitivity of the membrane to ACh or on resting membrane potential or resistance. As the effect is calcium dependent it seems likely that guanidine increases the ability of nerve impulses to release ACh from the nerve terminals. Postjunctional drug action Drugs causing a block of neuromuscular transmission are used as adjuvants in surgery to achieve muscle relaxation, especially in abdominal surgery, or to prevent severe convulsions in tetanus or those caused by electroconvulsive therapy (ECT) in the treatment of psychiatric disorders. The useful drugs (Fig. 3.4) are either non-depolarising (competitive) or depolarising (desensitising) blocking agents: the metaphilic antagonists are only mentioned here as an example of a possibly different type of action but these drugs are not clinically useful. d-Tubocurarine (d-Tc) is a synthetic drug which is the active principle of the South American Indian arrow poison. In addition to blocking the nicotinic receptors at the neuromuscular junction, it also blocks those in autonomic ganglia, resulting in a fall in blood pressure, decreased tone and motility of the gastointestinal tract and a block of vagal actions. Histamine release can be an important problem which can lead to hypotension, bronchospasm and an increase in bronchial and salivary secretion in man. These effects may be reduced by the administration of antihistamines. The search for alternative drugs has concentrated on a reduction of these undesirable side-effects, together with different durations of action. Gallamine is a short-acting, purely synthetic drug with a minimal effect on sympathetic ganglia but it may nevertheless cause tachycardia by reducing vagal effects by a mechanism which is not certain. Pancuronium, a synthetic steroidal neuromuscular blocking drug, is more potent than d-tubocurarine and has a similar duration of action but does not block ganglia or release histamine. Fazadinium, (AH 8165), belonging to the azo-bisaryl-imidazopyridinium series has an action which is brief in du-

30

Neuromuscularjunction

ration and rapid in onset with minimal effects on the cardiovascular system: the brief duration is due to rapid metabolism in the liver. The differential ability of some of these drugs to block nicotinic receptors at the neuromuscular junction but not those in autonomic ganglia re-emphasise the conclusion that the nicotinic receptors at these two sites are different. All of the drugs so far mentioned act as competitive antagonists to ACh at the receptor sites. Another group of drugs, typified by decamethonium (CIO) and succinylcholine, act differently and are classed as depolarising, desensitising agents. A characteristic of the depolarising agents is that they produce an initial muscle fasciculation before the onset of neuromuscular blockade. This uncoordinated contraction of many muscle fibres may cause physical damage and result in muscle pain postoperatively. CIO has minimal effects on ganglia or on the release of histamine and therefore has minimal effects upon the circulation. Succinylcholine has a very brief duration of action because it is an excellent substrate for plasma cholinesterase and so is rapidly degraded. However, in some patients with very low levels of the esterase, there may be a very prolonged action if the drug is used in the normal dosages. These agents depolarise not only skeletal muscle but also autonomic ganglia. On vagal ganglia the depolarisation may be sufficient to cause bradycardia and hypotension. Hexafluorenium has a mild neuromuscular blocking action and in addition inactivates plasma cholinesterase. Benzoquinonium has characterisitics of both the competitive and the depolarising blocking drugs and additionally inhibits cholinesterase. It is therefore used to prolong the action of succinylcholine. a-Bungarotoxin is a toxic rather than therapeutic drug but it has been of great importance experimentally for the isolation of the nicotinic receptor with which it combines irreversibly to form a very stable drug-receptor complex. Mechanisms of neuromuscular blockade by competitive receptor an-

tagonists. The theory of drug-receptor interaction based upon the

Sites of drug action at the neuromuscularjunction

31

law of mass action predicts that in the presence of a reversible, competitive antagonist the log dose-response curve for an agonist will be shifted to the right, the curve will remain parallel to the control curve obtained in the absence of the antagonist and the maximum response of the tissue will be unchanged. Curarelike drugs should therefore be expected to conform to these criteria since they are believed to be competitive, reversible antagonists of ACh at the nicotinic receptor at the motor end plate. Dose-response curves for ACh are easily obtained upon the frog rectus muscle preparation. At low concentrations of d-Tc producing dose ratios for the agonist (ACh) of about 100 the curves remain reasonably parallel and the maximum response does not change appreciably. Thus, at these concentrations of antagonist there seems to be a competitive interaction with ACh at the receptor. At higher concentrations there is a progressive departure from parallelism and the maximum response declines. On the frog sartorius muscle in which Jenkinson compared the depolarisation of the end plate produced by carbachol in the presence and absence of d-Tc, the departure from parallelism and the reduction of the maximum response to ACh was even more marked at relatively low concentrations of the antagonist. In the tibialis muscle of the cat, Paton & Waud showed that gallamine caused very marked discrepancies from the anticipated curves. The situation in such experiments is complex and difficult to explain simply in terms of competitive antagonism but it was suggested by Paton & Waud that the response to agonists may not be equilibrium responses and that the antagonists in some way alter the equilibria. Thus, shifts in dose response curves in this instance does not give unequivocal evidence for or against the competitive nature of the anticholinergic action of d-Tc or gallamine. Evidence showing that the interaction of d-Tc with ACh at the motor end plate is competitive in nature was first obtained by Katz & Miledi in their experiments on membrane noise. They predicted that if the antagonism was competitive then the amplitude of the elemental shot effect produced by the interaction of one molecule of ACh with the receptor should be

32

Neuromuscularjunction

unaffected by the presence of the antagonist. They clearly showed that neither d-Tc, which is reversible, nor abungarotoxin, which is not, changed the amplitude of the ACh shot effect, fca\ The reduction of the depolarisation produced by a given concentration of ACh is therefore due to a lower frequency of activation of the receptors. At low frequencies of activation of the neuromuscular junction it is certain that competitive antagonism of ACh at the receptor site is the major mechanism involved and this is supported by the failure to show any reduction in ACh output by direct assay. However, the results obtained from direct assays of the amount of ACh released in the presence of a cholinesterase inhibitor, required to prevent hydrolysis by acetlcholinesterase, must be subject to a note of caution. Furthermore, there is a large component of the assayed ACh which is released in a non-quantal fashion, perhaps from non-vesicular sites, and this may mask any small change in quantal release produced by the antagonist. There is electrophysiological evidence that d-Tc causes a reduction in quantal release when the motor nerve is stimulated at higher frequencies. This prejunctional effect may therefore also contribute to neuromuscular paralysis under physiological conditions. If so, then it seems likely that there are prejunctional nicotinic receptors which facilitates the release of ACh at relatively high frequencies of stimulation. The interaction of ACh with its receptor and the reduction of the effect by a competitive antagonist can be represented as the molecular model shown below, which may later be compared with other mechanisms. ACh + R ^ AChR ^ ACI12R ^ ACI12R* dTc + R ^ dTcR In this model the interaction of ACh with the receptor is shown as a bimolecular interaction, in accord with the experimental evidence. R' represents the closed and R*' represents the open channel configuration of the receptor. Mechanism of neuromuscular block by depolarising agents. It will

be evident from some of the previous discussion that

33

Sites of drug action at the neuromuscular junction

decamethonium (CIO) and succinylcholine may act as agonists at the nicotinic receptor and that this fact must be considered in relation to the mechanism by which they block neuromuscular transmission. The discussion of the mechanism is complicated by the fact that these substances cause depolarisation, desensitisation and neuromuscular blockade and by the fact that there are differences in the effects obtained in different species or even between different muscles in the same animal. The systemic administration of CIO or succinylcholine in mammals causes a muscle fasciculation which is relatively shortlived and may be seen in the experiment illustrated in Fig. 3.5 as a slight elevation of the baseline tension immediately following the intravenous injection of 2 mg of succinylcholine. During this phase of the effect, the muscle twitches elicited by nerve Fig. 3.5. Contractions of the tibialis muscle of the cat to electrical stimulation of the peripheral stump of the sciatic nerve every 5 s. An intravenous injection of 1 mg of succinylcholine caused a facilitation of the contractions; a larger dose caused only a transient facilitation followed by a block of transmission during which tetanic stimulation at 30 Hz caused a small but maintained contraction. D-tubocurarine (0.5 and 1 mg) caused only a block of transmission during which a tetanus (100 Hz) was not maintained but gave rise to post-tetanic potentiation. Neostigmine (0.2 mg) reversed the blockade of neuromuscular transmission.

A

AT

ti 1 mg

2 mg 30 Hz

0.5 mg 1 mg 30 Hz 100 Hz

0.2 mg

34

Neuromuscularjunction

stimulation are at first enhanced but are later reduced and ultimately completely blocked. The fasciculation is a reasonably In 1951, Burns & Paton carried out experiments on the gracilis muscle of the cat with extracellular recording which suggested that the time course of the depolarisation by CIO in this muscle ran parallel to the time course of the neuromuscular blockade. This supported the idea that blockade of neuromuscular transmission was the result of excessive depolarisation of the end plate leading to an inactivation of the voltage dependent Na + channels in the surrounding, extrajunctional membrane, so preventing the initiation of propagated action potentials upon which the contraction of focally innervated muscle depends. However, the close correspondence between depolarisation and block is not seen on frog muscle (Fig. 3.6). Even in the early work on mammalian muscles it was evident that different muscles in the cat showed different pharmacological responses to decamethonium and that there were species differences. In all muscles of the monkey, dog, rabbit and rat and in the soleus muscle of the cat it was thought that decamethonium and succinylcholine produced a 'dual block', changing with time from a depolarising to a non-depolarising type. However, in the tibialis muscle and gracilis muscles of the cat and probably in some muscles in man the block was thought to be solely due to the prolonged depolarisation of the end plate. In 'Phase IV block in man, which occurs late during prolonged administration of depolarising agents, the block of neuromuscular transmission is relieved by anticholinesterase agents. Neostigmine does not relieve depolarisation or desensitisation block and the mechanism of Phase II block is unclear. One suggestion is that it is the result of the entry of the blocking agent into the interior of the muscle through open ion channels. In those muscles, for example frog muscle, in which depolarisation is clearly not the explanation of neuromuscular blockade Katz & Thesleff introduced the concept of receptor desensitisation. Desensitising block of the responses of frog muscle to ACh by a small, sustained application of ACh is illustrated in Fig. 3.6a. Here, the depolarisation produced by brief iontophoretic pulses of ACh are depressed by the sustained

Sites of drug action at the neuromuscularjunction

35

iontophoretic administration of a small quantity of ACh which produced only a transient depolarisation of the end plate. A possible molecular mechanism for such desensitisation could be: ACh + R - ACI12R* - AChRd Fig. 3.6. Desensitisation, depolarisation and neuromuscular block, (a) Iontophoresis of ACh on frog muscle. A prolonged (15 s) administration of ACh caused a depolarisation which was not sustained and the effect of repeated small administrations of ACh was reduced by the prolonged administration, even when the depolarisation produced initially was minute. After Katz, B. & Thesleff, S. (1967). J. PhysioL, 138, 63. (b) The intra-arterial injection of CIO in a cat (gracilis muscle) caused a block of neuromuscular transmission and a depolarisation of the muscle with only a slightly different time-course. After Burns, B.D. & Paton, W.D. (1951). J. PhysioL, 115, 41. (c) Bath application of CIO to frog muscle. The addition of decamethonium to the bath caused a sustained block of neuromuscular transmission but only a transient depolarisation. After Thesleff, S. (1955). Acta PhysioL Scand, 34, 218. In (a) and (c) the neuromuscular block is due to desensitisation but in (b) it is due to depolarisation. 5 m

ACh

ACh

ACh

ACh (c) Depolarisation Block

(b)

Depolarisation

Intra-arterial CIO

Block

Time

CIO in bath

3 Time

36

Neuromuscularjunction

in which Rd represents a desensitised configuration of the receptor associated with closed ionic channels, open ionic channels again being represented by R*. The second, desensitising phase of the reaction is considered to be a slow process compared with the first activation of the channel. Similar reactions could occur with the blocking agents. Changeux has characterised three affinity states of the ACh receptor isolated from Torpedo electroplaques. The highest affinity state, with a dissociation constant for ACh of 3 X 10~9 M, represented about 20% of the receptor population and corresponds to Rd. The lowest affinity state with a constant of about 10'4 M corresponds to R, the resting state of the receptor. Binding to the high affinity state, Rd, is facilitated by raising the Ca2+ concentration, which also increases desensitisation. It seems likely that binding to Rd blocks the sodium channel, rather than maintaining it in the resting state, but this type of channel blocking may be different from that produced by other channel blockers which bind to a different site in the channel. Although CIO enters the muscle fibre at the end plate region, the uptake, in contrast to the depolarisation, is not antagonised by d-Tc. It is unlikely that CIO acts intracellularly to block transmission, even though it is taken up. Using modern techniques, Wray (1981) has produced convincing evidence that, at least in the tenuissimus muscle of the cat, the blockade of transmission by CIO, succinylcholine or by high concentrations of ACh itself is due to depolarisation and not to desensitisation in this tissue. In rat muscles, which are less sensitive to the blocking agents, the mechanism is not due to depolarisation but to desensitisation. In man the mechanism seems to be depolarisation as in the tenuissimus muscles of the cat. A diagrammatic representation of the data obtained by Wray is shown in Fig. 3.7. At low concentrations of ACh (1 to 2 X 10"6 M) there was a maintained depolarisation of the end plate but a gradual decline in twitch tension attributable to a blockade of transmission. From his noise analysis records it was evident that there was a maintained increase in the frequency of opening of the ion channels by the agonist. At higher concentrations (5 to 10 X 10"6 M)

Sites of drug action at the neuromuscular junction

37

the frequency of channel opening at first increased and then rapidly decreased. This was accompanied, as might be expected, by a decrease in the mean level of depolarisation. Thus desensitisation was revealed in these experiments by a decreased frequency of opening of ion channels in the presence of high but not low concentrations of ACh, although block of transmission occurred at all concentrations in this range (2 to 10 X 10"6 M). The desensitisation had a relatively slow time course. Fig. 3.7. Diagram of depolarisation, channel opening and neuromuscular block in the tenuissimus muscle of the cat and rat in response to bath application of acetylcholine or succinylcholine. After Wray, 1981. /. Physiol, 310, 37. CAT Tension to nerve stimulation

Depolarisation of muscle fibres

Frequency of channel opening

RAT Tension to nerve stimulation

Depolarisation of muscle fibres

Frequency of channel opening

Low concentration

High concentration

38

Neurom uscularjunction

At concentrations which block transmission, CIO and succinylcholine depolarise by about 20 mV. Although the channel open time is shorter than for ACh (Table 3.1), the frequency of channel opening is greater so that the overall level of depolarisation is similar. Wray found that at concentrations which just blocked transmission there was no evidence of desensitisation, although it did occur, as with ACh, at higher concentrations. In similar experiments on rat muscles, which are less sensitive to the blocking agents, he has shown that blockade of transmission only occurred at concentrations which caused a decline in the frequency of channel opening: the mechanism in rat muscle is therefore desensitisation and not depolarisation. In conclusion, it will be appreciated that the mechanism of blockade of neuromuscular transmission by CIO and succinylcholine is complex. Both desensitising and depolarising block may occur, but the relative importance of these two effects depends both upon the particular muscle and the animal species. Metaphilic antagonism. Rang & Ritter found that the dinaphthyl derivative of decamethonium (DNC-10) only blocked the action of cholinergic agonists on frog and chick muscle in the presence of an agonist (Fig. 3.8). The rate of onset of antagonism was faster Fig. 3.8. (a) Lack of apparent effect of metaphilic antagonists after washout when desensitising agent is not administered during exposure to the antagonist, contrasted with the marked effect when both are present together (b). (a)

(b)

I Desensitising agonist Time -

Characterisation of blocking agents

39

if the agonist was given at frequent intervals and absent if no agonist was present during exposure to the antagonist. The effect of the metaphilic antagonist was greater with some agonists than with others. For example, the blocking action of DNC10 was greater when decamethylene-bis-trimethylammonium (C10TMA) was used as the agonist than it was with either CIO or carbachol. C10TMA itself caused more desensitisation than either carbachol or CIO. This type of antagonism was called metaphilic antagonism. It probably represents s special case of desensitisation in which the ligand binds only to the desensitised configuration of the receptor (Rd). Metaphilic antagonists have not yet provided any clinically useful compounds and have not been as extensively studied as other neuromuscular blocking agents. Pharmacological characterisation of neuromuscular blocking agents Neuromuscular blocking agents which act competitively, e.g. d-tubocurarine, can be distinguished from those which cause depolarisation, e.g. decamethonium, by a number of tests in whole animals or on isolated tissues. (Some of these have already been referred to in the preceding discussion and are illustrated in Fig. 3.5). i) A muscle depolarisation is not caused by agents which act competitively. They therefore do not cause an initial fasciculation of the muscle when injected intravenously in mammals nor do they cause a contracture of multiply innervated muscle fibres in frogs or birds. In birds, which have both focally innervated and multiply innervated fibres, even in the same muscle, the competitive blockers cause a flaccid paralysis in contrast to the spastic paralysis, characterised by opisthotonus and rigid limb extension, produced by depolarising blocking agents in this species. ii) Cholinesterase inhibitors, e.g. neostigmine, reverse block by d-Tc but not by CIO or succinylcholine, due to an increase in the safety factor for transmission.

40

Neuromuscular junction

iii)

The tension developed during tetanic stimulation of motor nerves at frequencies of 30 to 50 Hz is not maintained during partial blockade by a competitive agent but it is maintained during partial block by depolarising agents. The decline in tetanic tension after the administration of drugs such as d-Tc is attributed to two factors, a reduction in safety factor for transmission, caused by the reduced number of receptors available to ACh, and to the progressive reduction in ACh release with each stimulus in the tetanic train. This combined effect reduces the amplitude of the epps until they are no longer able to elicit action potentials. In untreated muscles, even though transmitter depletion still occurs during the tetanus, the end plate potentials remain above threshold for the initiation of muscle action potentials. Post-tetanic potentiation (PTP) is readily observed after partial neuromuscular block by the competitive but not the depolarising agents. PTP is attributed to an increase in the intracellular concentration of free Ca2+ in the terminals during the tetanus which only slowly declines over a period of minutes after cessation of the stimulus. This causes an increased liberation of ACh with each stimulus at low frequencies of stimulation and thereby an augmentation of the amplitude of the epps, partially reduced to near threshold levels by d-Tc. The effect is observed to only a small degree in untreated muscles where the safety factor for transmission is high. Potentiation is not evident during the period of high frequency of stimulation because under these conditions there is transmitter depletion from the terminals.

iv)

Myaesthenia gravis About one in every 6000 persons is afflicted with a crippling disease of the neuromuscular junction called myaesthenia gravis. The disease strikes either suddenly or slowly and may appear at any age between 10 and 40 years. The incidence of the disease is greater in women than in men but the

Myaesthen ia gravis

41

distribution of the disease is unaffected by geography, race or climate. There are marked changes in the structure of the end plate region including a reduction in the number and size of the nerve terminals and an increase in the diameter of the synaptic cleft. There is a reduced amplitude of the mepps recorded electrophysiologically. The symptoms are similar to those produced by neuromuscular blocking agents like d-Tc and there is a marked increase in the sensitivity to this agent. The muscle weakness increases during repetitive muscle activity and decreases with rest. It tends to increase throughout the day. In about 10% of patients there is a thymoma and in about 80% there are other abnormalities in the thymus, including mild thymitis. Possibly a viral infection is the causative factor. Thymectomy may give marked improvement but there is complete remission in only about one third of the patients. About 90% of patients have circulating antibodies which react with the ACh receptors on skeletal muscle and immunosuppressive agents produce some improvement. There is a correlation between the severity of the symptoms and the antibody titre. Symptoms may improve dramatically with plasmaphoresis to remove the circulating antibodies. Similar, although not necessarily identical, antibodies have been induced experimentally by the administration of purified ACh receptor proteins. It is therefore almost certain that the disease is of the autoimmune type. There is a correlation between the incidence of myaesthenia gravis and other autoimmune diseases and with the occurrence of neoplasms. Myaesthenic symptoms are reduced by the administration of the anticholinesterase, neostigmine; this is used in preference to physostigmine because it does not pass the blood-brain barrier. Germine monoacetate has also been found to produce beneficial results. A very short acting anticholinesterase, edrophonium, is employed as a diagnostic agent for the disease; if the symptoms improve, the test is diagnostic for myaesthenia. Related conditions, such a myaesthenic syndrome, are not improved by cholinesterase inhibitors and may in fact be made worse. In myaesthenic syndrome there may be a different defect

42

Neuromuscularjunction

in transmission in which there is a deficiency in the release mechanism. Denervation supersensitivity Muscles which have been denervated by cutting the motor nerve supply and allowing sufficient time for the fibres distal to the cut to degenerate become hypersensitive to ACh. The hypersensitivity is due to a spread of receptors along the muscle fibres, in contrast to the restricted distribution about the end plate in normal muscle. Similar effects may be produced by irreversibly blocking the receptors with a-bungarotoxin. Several factors may account for the spread of the receptors, or perhaps more strictly their restriction to the end plate region in normal fibres, including muscle activity and 'trophic' factors liberated from the nerve. The characteristics of the new receptors formed after degeneration of the nerve are similar, but not identical, to those of the normal end plate receptors. Although the new receptors remain nicotinic in type, the channel open times are longer than those at the normal end plate and the action potentials are no longer blocked by tetrodotoxin. In addition the supersensitivity to d-Tc is not as great as that to ACh indicating some change in the receptor affinity. There is some increase in the levels of cyclic adenosine monophosphate, possibly indicating that cyclic nucleotides play a role in the development of the supersensitivity.

Autonomic nervous system

The autonomic nervous system (ANS) is an efferent system conveying impulses from the central nervous system to the periphery. It controls the activity of most bodily functions except those of the skeletal muscles which are controlled by the somatic motor system discussed in the preceding chapter. It therefore affects such diverse physiological activities as salivary secretion, sweating, movement and secretions of the gastrointestinal tract, heart rate, calibre of the blood vessels, secretions of the pineal gland, contraction of the urinary bladder and its internal sphincter, penile erection, adrenaline secretion by the adrenal glands, accommodation in the eye and control of pupil diameter. Pharmacological interference with the ANS will therefore lead to widespread effects. A major application of drugs which selectively reduce vasomotor tone is in the treatment of cardiovascular disease. There are two major divisions of the ANS. The sympathetic division originates from preganglionic nerve cell bodies lying predominantly in the intermediolateral part of the thoracic and upper lumbar regions of the spinal cord. The parasympathetic division originates from nerve cells located in the brain stem and in the sacral spinal cord. Both systems may be distinguished from the skeleto-motor system by the interposition of a peripheral ganglionic synapse between the preganglionic neurone in the CNS and the peripheral, innervated effector organ (Fig. 4.1). The ganglia of the sympathetic division are situated either in the paravertebral chain ganglia or in one of the more distal sympathetic ganglia, such as the cervical sympathetic, stellate, coeliac, inferior mesenteric, or hypogastric ganglia. The parasympathetic ganglia are usually located within

44

Autonomic nervous system

or very close to the effector organ and so, unlike the sympathetic nerves, give rise to only very short postganglionic fibres, for example, those in the heart or on the wall of the urinary bladder. The parasympathetic nerve to the orbit, which synapses in the ciliary ganglion, is an exception to this generalisation. The postganglionic nerve fibres in both systems are mostly nonmyelinated C-fibres in contrast to the myelinated B-fibres of the preganglionic nerves. Neurotransmitters The most important transmitters in the autonomic nervous system are ACh and noradrenaline (NA). However, recent studies have shown that, particularly in the intestine, there is an abundance of neuronally located bioactive peptides such as substance P, enkephalins, vasoactive intestinal polypeptide (VIP), cholecystokinin and a number of other peptides. Sometimes the peptides co-exist with the more conventional transmitters. For example, ACh and VIP have been shown to be Fig. 4.1. Comparison of structure and transmitters in skeleto-motor and autonomic nervous system. Skeleto-motor

Central nervous system

Sympathetic

Parasympathetic t

A fibre

B fibre

B fibre Myelin

Myelin ^ ^ ^ Peripheral nerve ACh-^

^ ^ Sympathetic * ganglion C fibre

Effector organ

ACh^^ ACh^, HUH TTTff

Skeletal muscle

NA-^,

^ \)

Parasympathetic ganglion

t ^ C fibre

Smooth muscle or glands

Neurotransmitters

45

present in the same nerve fibres, but more than one peptide is usually not present simultaneously in the same neurone. This may not always be the case since there is some evidence that somatostatin and substance P may co-exist in some primary afferent fibres. The major part of the substance P at the periphery is associated with non-myelinated, sensory afferent fibres. There is good evidence for a role of substance P and of luteinising hormone-releasing factor in autonomic ganglia but the evidence for a physiological role of other peptides is as yet lacking. In addition to the peptides, there is growing evidence that purines may be involved in transmission at some sites in the viscera and on blood vessels. ACh is the transmitter liberated from preganglionic terminals in both the sympathetic and parasympathetic systems and excites the postganglionic neurones. It also functions as the transmitter at the postganglionic parasympathetic junctions, where it may either be excitatory, e.g. in the alimentary tract and the urinary bladder, or inhibitory as in the heart, or it may evoke secretions as in the salivary glands. The first conclusive evidence that ACh was the transmitter from the vagus nerve to the heart was obtained by Loewi in the 1920s, who demonstrated that the substance released on stimulation of the vagus in isolated frog hearts, slowed a second perfused heart in the same way as ACh: using the same technique he also showed that an adrenaline-like substance was liberated on stimulation of the sympathetic supply. It was not until the 1930s that it was shown that ACh was also the transmitter in sympathetic ganglia and at the neuromuscular junction. NA is liberated from postganglionic sympathetic nerve fibres and may be either excitatory as in the heart or inhibitory as in the intestine. The chromaffin cells of the adrenal medulla may be considered to be modified sympathetic neurones which are able to synthesise adrenaline from noradrenaline by N-methylation. In this case the amine is liberated into the circulation, where it exerts effects similar to those of NA. Dopamine may also be a neurotransmitter at some sites, e.g. in some ganglia and in the kidney, although its precise function at these sites remains uncertain. Some sympathetic nerves contain ACh, as in the sympathet-

46

Autonomic nervous system

ic supply to the sweat glands and to some blood vessels in sketetal muscle, but these fibres are probably of minor significance. Drug action in the autonomic nervous system Drugs may act at peripheral sites such as ganglia, terminals, postjunctional receptors or on postreceptor mechanisms of the autonomically innervated cell or they may act on central mechanisms in autonomic reflex pathways. Occasionally, they may also exert effects upon the afferent limb of the pathway. An example of this last action is the marked fall in blood pressure and heart rate which accompanies an intravenous injection of a large quantity of 5-hydroxytryptamine and other agents which stimulate vagal afferent fibres in the right atrium: this is the Bezold-Jarische reflex for which there is little evidence of a physiological function. Ganglionic sites of action At the turn of the century Langley noted that the local application of nicotine to autonomic nerves only caused effects when the agent was painted on at discrete spots, which we now call ganglia. He used this simple technique to map out most of what we know of the distribution of autonomic nerves to their effector organs. He observed that at low concentrations applied to sympathetic ganglia there was an excitant effect, similar to that produced by electrical stimulation of the nerve. At higher concentrations, there was also at first an excitant effect but this waned and at this time stimulation of the nerve was ineffective. This dual action of nicotine is due to an initial excitation of the ACh receptors followed by a blocking action due to desensitisation, probably similar in mechanism to that at the neuromuscular junction. In addition to the cholinergic synapses in ganglia, which are probably the most important to consider, there are also monoaminergic synapses which utilise either NA or dopamine (DA) as the neurotransmitter and, at least in some species, peptides may also be involved.

47

Ganglionic sites of action

Nicotinic receptors. The most obvious effects in ganglia resulting from the electrical stimulation of preganglionic nerves are due to the activation of the cholinergic receptors. The largest postsynaptic potential with the shortest latency is caused by the Fig. 4.2. Ganglionic mechanisms, (a) Diagrammatic representation of sequence of ganglionic potentials evoked by a stimulus (S) to the preganglionic nerve. N, surface-negative wave of depolarisation; P, surface-positive wave of hyperpolaristion; LN, late surface-negative wave of depolarisation. (b) postulated ganglionic synapses causing the recorded wave form. N, Nicotinic receptors (blocked by hexamethonium or D-tubocurarine); M, muscarinic receptors (blocked by atropine); A, adrenoceptor (blocked by dibenamine), SIF, small intensely fluorescent neurone. Note: there is evidence that in some situations a cholinergic step directly mediates the inhibition without the intervention of a SIF neurone.

DAorNA SIF neurone

48

Autonomic nervous system

interaction of ACh with nicotinic receptors on the postganglionic neurones. This is the equivalent of the N-wave, first recorded extracellularly from isolated ganglia of the rabbit by Eccles & Libet and shown diagrammatically in Fig. 4.2. The nicotinic receptors in ganglia are similar to but not identical to the nicotinic receptors at the neuromuscular junction. They can be distinguished by the actions of agonists and antagonists at the receptors. At both sites ACh, nicotine and carbachol are stimulants and d-tubocurarine is a competitive antagonist. However, the nicotinic receptors in ganglia but not at the neuromuscular junction are activated by dimethylphenylpiperazinium (DMPP) or tetramethylammonium and blocked competitively by tetraethylammonium or hexamethonium (C6). Phenyltrimethylammonium, which activates the nicotinic receptors at the neuromuscular junction, and decamethonium, which causes block, are not similarly effective in ganglia (Table 4.1). Studies with the bisquaternary ammonium series showed that peak ganglion blocking activity was present with a chain length of 5-6 carbon atoms (pentamethonium and hexamethonium) whereas 10 carbon atoms (decamethonium) are required at the neuromuscular junction. Alpha bungarotoxin binds to specific binding sites in ganglia as it does at the neuromuscular junction. However, it does not block transmission in ganglia, possibly because it has two opposite effects. Besides blocking the receptors, it also increases the mean channel lifetime by a factor of two, and this effect may counteract the block of transmission. Muscarinic receptors. Electrical stimulation of the preganglionic fibres in the cervical sympathetic nerve of the cat causes a contraction of the nictitating membrane which is blocked by hexamethonium and is due to the activation of nicotinic receptors in the superior cervical sympathetic ganglion. When the stimulation is terminated, the nictitating membrane relaxes but there follows a small second contraction which is resistant to hexamethonium but blocked by atropine, a known selective antagonist at muscarinic receptors. Electrical stimulation of the hypothalamus causes a contraction of the membrane which is only

Ganglionic sites of action

49

Table 4.1. Comparison of effects of agonists and antagonists at the neuromuscular junction and in autonomic ganglia

Muscarinic receptors a) Agonists: ACh Carbachol Muscarine Acetyl-beta-methylcholine McNeil A-343 (Mi) b) Antagonists (competitive): Atropine (Mi) and (M2) Pirenzapine (Mi)

Neuromuscular junction

Autonomic ganglia

+ + 0 0 0

+ + + 4+

0 0

44-

Nicotinic receptors c) Agonists: ACh 4Carbachol + Nicotine + Dimethylphenylpiperazinium (DMPP) 0 Phenyltrimethylammonium (PTMA) + d) Antagonists (competitive): D-tubocurarine 4Hexamethonium 0 Block release of ACh Botulinus toxin

4-

+ 44+ 0 + + +

0 = no effect 4- = effective; Mi muscarinic receptors are found in ganglia and in the brain and are activated by McNeil A-343 and blocked by pirenzapine; ACh and carbachol are effective on all muscarinic and nicotinic receptors. partially blocked by either hexamethonium or atropine administered separately but is completely blocked when both agents are administered simultaneously under conditions in which there is no effect on the frequency of impulses in the preganglionic nerve. It is therefore clear that this response involves both nicotinic and muscarinic receptors. The intra-arterial administration to the ganglion of acetyl-(3methylcholine, muscarine or McNeil A-343 (Table 4.1), all of which activate muscarinic receptors but not nicotinic receptors,

50

Autonomic nervous system

causes contractions of the membrane which are blocked by atropine but not hexamethonium: McNeil A-343 actually causes a rise in blood pressure, unlike other agonists at muscarinic receptors, because it selectively activates ganglionic receptors and not those on the blood vessels which cause vasodilatation and a fall in blood pressure. There is now some evidence to support the notion that there may be different subtypes of muscarinic receptors. The electrophysiological observations of Eccles & Libet in 1961 on transmission in the isolated rabbit superior cervical ganglion demonstrated the importance of both nicotinic receptors and muscarinic receptors in ganglionic transmission (Fig. 4.2). There was an early depolarising N-wave which is blocked by hexamethonium or d-tubocurarine and is attributable to activation of the nicotinic receptors by neuronally released ACh. There follows a hyperpolarising P-wave, which is blocked either by atropine acting on muscarinic sites or by dibenamine, an alphaadrenergic receptor antagonist. The P-wave was followed by another late depolarising negative wave called the LN-wave which is completely blocked by atropine. It was postulated that the P-wave was caused by muscarinic activation of a monoaminergic interneurone which synapsed with the postganglionic neurone: this explanation accounted for the block by dibenamine (see Fig. 4.2). Since all waves were blocked by botulinus toxin which prevents the release of ACh in ganglia, just as it does at the neuromuscular junction, all effects are clearly mediated via cholinergic preganglionic nerve fibres. The postulate of Eccles & Libet has received ample support from histochemical studies demonstrating the presence in the ganglia of small, intensely fluorescent (SIF) neurones containing either dopamine or noradrenaline according to the species. However, more recently the hypothesis has been questioned on several grounds. The most compelling is that muscarinic receptor agonists induce a hyperpolarisation of the ganglion cell membrane, by an increase in the potassium ion conductance, which occurs at low Ca2+ concentrations in the medium. This indicates that activation of the muscarinic receptor can directly hyperpolarise the membrane, without the intervention of

Ganglion ic sites of action

51

synaptic transmission. The slow inhibitory synaptic potential produced by stimulation of the preganglionic nerve has also been shown to be due to an increase in potassium ion conductance, in which no second messenger is involved, and which is monosynaptically mediated via a cholinergic synapse in the bullfrog, mudpuppy, rabbit, rat and cat. Nevertheless, there is evidence that dopamine receptor activation does facilitate the responses due to activation of muscarinic receptors and these effects of dopamine are blocked by the dopamine Dl receptor antagonist, butaclamol, but not by antagonists which are selective for the D2 receptor. Potentiation of these effects by theophylline, an inhibitor of phosphodiesterase, indicates that cylcic AMP could be involved as a second messenger. In this context it is of interest that dopamine and nerve stimulation both increase the level of cyclic AMP in the ganglion. Other evidence indicates that the receptors in ganglia are not dopamine but a-2 adrenoceptors and that the SIF cells have no processes and are too sparse to carry their postulated role. Clearly, there are still some controversial issues to be resolved in the mechanism of the inhibition in ganglia. Peptide involvement in ganglionic transmission. In addition to the

late negative wave demonstrated in mammalian ganglia, in the frog there is also a late slow excitatory postsynaptic potential (EPSP) which is probably due to luteinising hormone releasing hormone (LHRH), co-released with ACh from preganglionic fibres. In this species, stimulation of cholinergic, myelinated B-fibres gives rise only to cholinergically mediated responses. However, stimulation of the preganglionic, non-myelinated C-fibres (Fig. 4.3) causes activation of cells via the release of both ACh and LHRH. There is also an indirect activation of the cells due to diffusion of the LHRH from the C-fibres. The effects of exogenous LHRH are similar to those of the endogenously released substance and both are blocked by LHRH analogues which are antagonists of LHRH in the pituitary. There is also evidence for the involvement of the polypeptide substance P in ganglionic transmission (see Otsuka & Konishi,

52

Autonomic nervous system Fig. 4.3. Diagrammatic representation of cholinergic and peptide mediated effects in the inferior mesenteric ganglion of the guinea pig and in frog sympathetic chain ganglia. A. Inferior mesenteric ganglion in guinea pig

SP

Post

Inf. mes. g. Diagram of synaptic potentials

seconds minutes Stimulate preganglionic nerve B. Frog sympathetic ganglion

B-cell

B-fibres (myelinated)
ACh ACh + LHRH

Muscarinic

LHRH

mV

250 ms

1 min

On stimulus to third, fourth and fifth nerve

Stimulate nerves 3, 4 and 5 repetitively

4 min Stimulate seventh and eighth nerves repetitively

Ganglionic sites of action

53

1983). This peptide is associated with sensory afferent fibres which make synaptic contact with postganglionic neurones in the inferior mesenteric ganglion in the guinea pig and elsewhere, and appears to function as an excitatory modulator of transmission. Ionic mechanisms. The fast cholinergic EPSP in ganglia is due mainly to an increase in the conductance of the cell membrane to sodium ions, as at the neuromuscular junction. The slow muscarinic cholinergic EPSP and the peptide mediated EPSPs are due, not to increased conductance to Na\ but to a decreased conductance of specific potassium channels. This latter effect has been most studied for the cholinergic receptor-mediated effect in mammalian ganglia, where the voltage-dependent potassium current has been referred to as the M-current (M for muscarinic). This potassium current differs from the potassium current which flows during the after hyperpolarisation of the action potential and is not Ca2+ dependent. The M-current increases with depolarisation of the neurone and tends to stabilise the cell to make it less excitable. Inactivation of the current via the muscarinic receptors causes the membrane to depolarise and to become more excitable. Many of the excitatory effects of ACh in the central nervous system which involve the participation of muscarinic receptors also probably act via the same mechanism. The slow muscarinic inhibition in ganglia involves an increase the membrane permeability to potassium ions. This action may be calcium activated. Ganglionic blocking agents. Although now of mainly historical interest, the ganglionic blocking agents were the first really effective drugs for the control of blood pressure in severe hypertension. The first substances introduced were the polymethylene compounds, hexamethonium and related analogues. Since these are quaternary ammonium compounds which are fully charged at body pH they are poorly absorbed on oral administration and needed to be given systemically, not an ideal route for the chronic administration of a drug. Since they do not discriminate between sympathetic and parasympathetic ganglia, they produced a vari-

54

Autonomic nervous system

ety of side-effects due to block of the parasympathetic system. Particularly unpleasant was the mydriasis, paralytic ileus, dryness of the mouth and eyes and difficulties with micturition, to name but a few. Unsuccessful programmes were undertaken to find agents which could selectively block sympathetic but not parasympathetic ganglia. In the mid-1950s two non-quaternary amines, mecamylamine and pempidine, were introduced. These were no more selective for sympathetic ganglia than were the quaternary drugs but were better absorbed from the gut. Since they also penetrate the bloodbrain barrier to some degree there were also some unpleasant side effects upon the central nervous system. The reign of the ganglionic blocking agents was finally terminated by the introduction of drugs with a selective effect upon sympathetic nervous system in 1959. These were the adrenergic neurone blocking agents which have now been supplemented by a range of other agents acting in a relatively specific fashion upon the sympathetic nervous system, either at the postganglionic nerve terminals, on adrenergic peripheral receptors and mechanisms or upon the central nervous system (Fig. 4.4). It is therefore appropriate that we should now turn our attention to other aspects of drug action in the autonomic nervous system. The structure and function of sympathetic nerves Unlike the skeleto-motor nerve fibres, postganglionic sympathetic nerves do not terminate as discrete nerve terminals, although they are often referred to as such. Instead, there is a sympathetic ground plexus which consists of a network of fine fibres with swellings at intervals along them. These swellings are called varicosities. The varicosities contain many small, granular vesicles with a diameter of 30-60 nM and which have electron dense cores. There are also larger electron dense granules with a diameter of 90-120 nM. The granules are the storage sites for NA. The release process involves the exocytosis of vesicles, probably the large vesicles which then become small vesicles. The concentration of NA in the terminal part of the axon is about 100

The structure andfunction of sympathetic nerves

55

Fig. 4.4. Ganglionic blocking drugs and drugs used in the treatment of essential hypertension. Ganglionic blocking agents

CH,

NHCH 3

(CH3)3N+—(CH 2)6~N + (

I

CH3 CH 3

Hexamethonium

Mecamylamine

Pempidine

Drugs affecting peripheral noradrenergic terminals ^NH S

NH 2

Guanethidine OCH 3

\-°CH3

-OC—/

OCH 3 Reserpine

H3CO,

Other drugs Cl O O

H

Cl

I

H Hydrochlorothiazide

OH HO

H

or v-

NH 2 SO 2

N N

c

. n

Clonidine OH I OCH 2 CHCH ? NHCH(CH 3 ) 2

CH 3 CH^CGOOH NH 2

a-methyl DOPA

Propranolol

56

Autonomic nervous system

times that in the cell body and conducting part of the axon. Each varicosity may contain 6 X 10~15 g and each vesicle may contain 4 X 10~18 g of NA, corresponding to about 15 000 molecules. Not all smooth muscle fibres are directly innervated by the sympathetic nerves and there are electrically transmitting junctions, called 'tighf junctions, between the muscle fibres. The proportion of directly innervated fibres to indirectly activated fibres varies from one tissue to another. As many as six varicosities, perhaps from more than one fibre, or as few as one, may lie close to a smooth muscle cell. The amplitude of the smooth muscle response to stimulation may depend upon the number of junctions near to it and the frequency of stimulation. In contrast to skeletal muscle, there is little if any specialisation of the postjunctional cell and the distance of the cell from a varicosity may be quite large (up to 1 JLIM) SO that diffusion distances for the transmitter may be quite long. Stimulation of sympathetic fibres to smooth muscle cells give rise to inhibitory or excitatory junctional potentials and excitatory potentials may give rise to propagated action potentials. Here again there is a difference from skeletal muscle in that the action potentials have a large calcium component rather than the sodium component with which we are by now familiar. The effects of the autonomic neurotransmitters on ionic mechanisms in the heart are exceedingly complex (see Noble), involving multiple potassium ion conductance channels, calcium channels and interaction with the sodium conductance. The metabolism of catecholamines Noradrenaline is synthesised from its precursors tyrosine, phenylalanine or phenylethylamine via a series of enzymatic conversions by phenylalanine hydroxylase, tyrosine hydroxylase, DOPA decarboxylase and dopamine-p-hydroxylase (Fig. 4.5). Tyrosine hydroxylase (TOH) is the rate limiting step in the reaction. The enzyme is selective for the L-isomer of tyrosine but it will also hydroxlate a number of analogues. It is loosely associated with the endoplasmic reticulum. TOH is synthesised in the cell body and is transported down the axon to the terminals. The

The metabolism ofcatecholamines

57

rate of synthesis is augmented by an increase in activity of the sympathetic nerve and is reduced by substances which interfere with either protein or RNA synthesis. Examples of substances in this class are puromycin, cycloheximide and actinomycin-D. The enzyme has a high affinity for tyrosine, with a KD of 1010 M. The concentration of circulating tyrosine is in the same range so that the enzyme is, perhaps not surprisingly, well equipped to deal with its substrate. TOH is inhibited by a-methyltyrosine and by amethylphenylalanine. The latter substance has been employed in Fig. 4.5. Metabolic pathways for catecholamines.

/

Y-CH2CHNH2 \=/

•

\-CH2CHNH2

HO—/

I

\ — /

COOH

COOH

Tyrosine

Phenylalanine

OH HO'-f

N

DOPA

DOPA )-CH 2 CHNH 2 | COOH

• decarboxylase

CH2CH2NH2

HO Dopamine

Phenylethanolamine CHCH2NH2 — ^ H I N-methyl transferase OH Noradrenaline MAO and catechol O-methyltransferase

-CHCOOH I OH 3-methoxy-4-hydroxy mandelic acid

CHCH 2 NHCH 3

- / Adrenaline

AH

58

Autonomic nervous system

the treatment of phaeochromocytoma, in which there is an overproduction of adrenaline by the tumour. DOPA decarboxylase is free in the cytoplasm and is not rate limiting. It is therefore difficult to control the rate of NA synthesis by blocking this enzyme. Inhibitors of dopa decarboxylase include a-methyldopa, an agent used in the control of high blood pressure where it is probably acting by quite a different mechanism, a-methyl-meta-tyrosine and carbidopa, which helps to alleviate some of the side-effects associated with the use of L-dopa in Parkinson's disease, as discussed later in this book. Dopamine-(3-hydroxylase is located in the storage granules and converts dopamine to noradrenaline. Finally, at some sites, notably in the adrenal medulla, the noradrenaline is N-methylated to form adrenaline by phenylethanolamine N-methyl transferase. Catecholamines are deaminated by the mitochondrial enzyme monoamine oxidase (MAO) or O-methylated by catechol-O-methyl transferase, which is cytoplasmically localised. There are very large amounts of MAO in tissues such as the liver and the heart which will be responsible for the removal of most of the circulating amines, including some of those taken in in the diet. Tyramine is contained in quite large amounts in products such as cheese. Normally, this is deaminated in the liver. If the MAO is inhibited then the tyramine may be converted to octopamine which may indirectly cause the release of NA from terminals (see section on uptake and release) to cause hypertensive crises. The enzyme, which is relatively non-specific, is therefore of importance as a major mechanism for the detoxification of pharmacologically active amines ingested in the diet. Inhibitors of MAO include a number of substances such as iproniazid, nialamide and phenelzine which were formerly used in the treatment of mental depression. Use has been discontinued because of the high toxicity of these substances. Nevertheless, inhibitors and substrates have provided useful experimental tools to investigate the nature of the enzyme and two forms have now been demonstrated, with differing relative distributions in different tissues. 5-Hydroxytryptamine is selectively metabolised by MAO-A whereas benzylamine or (3-phenylethylamine are se-

The uptake and storage of catecholamines

59

lectively metabolised by MAO-B in the human heart for example. Dopamine and tyramine are substrates for both enzymes. The antagonist, chlorgyline, has a much higher affinity for MAO-A than it has for MAO-B. Deprenyl is said to have a higher affinity for MAO-B than for MAO-A but it is metabolised to amphetamine which may be responsible for some of its effects. The uptake and storage of catecholamines Neuronal uptake. Unlike the cholinergic system where the action of ACh is terminated by hydrolysis by cholinesterase, the action of the adrenergic transmitter is brought to an end by reuptake, either into the terminals or into postsynaptic structures. These two events are commonly referred to as neuronal uptake (uptake-1), and non-neuronal uptake (uptake-2). The interrelationship between re-uptake, storage, release and breakdown is shown in Fig. 4.6. Fig. 4.6. Relation between uptake, storage, release and breakdown of catecholamines. TOH

, Axonal transport

Tyrosine DOPA decarboxylase Dopamine-j3-hydroxylase Deaminated metabolites

60

Autonomic nervous system

Iversen, in the 1960s, measured the initial rate, i.e. the rate of uptake in the first two minutes, of tritiated DL-NA into the tissues of the perfused rat heart (Fig. 4.7). He showed that there is a high affinity uptake with a KD of about 0.7 X 10~6 M, which saturates at about 1.4 nM min 1 g"1 (Table 4.2). The uptake is stereospecific for the L-isomer of NA. Adrenaline has a lower affinity for the process than NA and isoprenaline, a beta-agonist, is not a substrate but is an inhibitor. The uptake of NA is highly efficient and in heart slices the tissue concentration is approximately five times higher than the concentration in the medium, whereas in perfused intact hearts the ratio is even higher at 30-40 times that in the medium. Degeneration of the sympathetic nerves or the administration of cocaine, which is an inhibitor of uptake-1, both prevent the uptake. Other inhibitors of uptake-1 include imipramine, guanethidine and bretylium (Table 4.2). Uptake and storage by granules. Analysis of the rate of uptake with time shows that there are two exponential components to the uptake process. The first component falls rapidly with a half life of about 5 minutes. The second component has a half-life of about 20 minutes. These two components represent uptake-1 and vesicle storage respectively. The slow component is abolished by the drug reserpine which depletes the terminals of their catecholamine stores by preventing the storage of NA in the granules. Reserpine does not affect uptake-1 as reflected by the fast component of the curve in Fig. 4.7, whereas cocaine abolishes both components because it blocks uptake-1. The storage granules can be isolated from the adrenal medulla or from the splenic nerve. Such isolated granules take up NA from the medium by an ATP-dependent process with kinetics similar to those of the slow component. Granular uptake is blocked by reserpine, is unaffected by cocaine, is not selective for adrenaline or noradrenaline and shows little evidence of stereoselectivity. The KD for granule uptake is relatively high (8 X 10"4 M). Extraneuronal uptake. At a concentration above about 0.6

The uptake and storage ofcatecholamines

61

Table 4.2. Substrates and inhibitors of uptake-1, uptake-2, COMT and MAO Km (uM)

Vmax (nM g -1 min"1)

A: Nor adrenaline

Iversen: Uptake-1 Uptake-2 Trendelenburg: Uptake-2 O-methylation Deamination

B: Iversen: Uptake-1 (neuronal)

Dopamine DL-noradren L-noradren D-noradren DL-adren Isoprenaline

Uptake-2 (non-neuronal)

Dopamine DL-noradren DL-adren DL-isoprenaline

0.27 252

1.18 100

60 1.7 138

50 1.2 25

0.69 0.67 0.27 1.39 1.4 Not a substrate 590 252 52 23

140 100 64 16

C: Relative affinities for uptake-1 and uptake-2 U1/U2 2350 Dopamine 374 DL-noradrenaline 239 DL-Amphetamine 233 Tyramine 68 Phenylethylamine 37 DL-adrenaline 5 Isoprenaline D: Inhibitors of uptake-1 Desmethylimipramine Imipramine Cocaine Guanethidine Bretylium Phenoxybenzamine

1.45 1.36 1.45 1.72 1.04

(Not a substrate)

ID50 for Ui (M) 1 9 4 3 1

X X X X X

10"8 10"8 lO"7 10"6 10"5


See Trendelenburg, 1979. Trends in Pharmac. ScL, 1, 4. for references

62

Autonomic nervous system

jig ml"1 in the perfusion medium (Fig. 4.7) there is a second component in the uptake process with both a high KD and a high VMAX. This second phase is termed uptake-2, and represents uptake into non-neuronal tissue. The large dissociation constant suggested that the mechanism was not very important physiologically. Uptake-2 shows little stereospecificity but isoprenaline, which is not a substrate for uptake-1, is a substrate for uptake-2. Trendelenburg has more recently re-evaluated the significance of uptake-2, and has concluded that, contrary to previously held Fig. 4.7. Rates of uptake of noradrenaline in rat heart, showing neuronal uptake component (uptake 1), non-neuronal uptake (uptake 2) and granular uptake. U2 - low affinity uptake Initial rate of uptake of 3 H-DL-NA U1 - high affinity uptake

Perfusion concentration (/zg/ml) KD 0.7

5 min (neuronal uptake, Ul) Log rate of uptake of 3 H-DL-NA

Ti = 20 min (granular uptake)

Perfusion time (min)

I 60

Cocaine blocks component a. Component b is reduced consequently. Reserpine blocks component b. Component a is not affected.

The uptake and storage ofcatecholamines

63

views, uptake-2 is almost as efficient at removing NA as uptake-1 but that uptake-2 only operates effectively at low concentrations of the neurotransmitter. Trendelenburg demonstrated that the tissue concentration of tritiated NA taken up by uptake-2 does not rise at low concentrations because the amine is rapidly O-methylated by COMT operating with a low KD of 1.7 |utM. For this reason, NA removed by uptake-2 is rapidly converted into O-methylated metabolites which were measured by Trendelenburg. However, O-methylation by COMT saturates at a relatively low level (Vmax = 1.2 nM g~l min"1). Only when COMT is saturated will levels of NA begin to rise, so accounting for the apparent threshold for uptake-2. Since there is no storage mechanism for NA in the nonneuronal tissue, uptake-2 becomes less efficient at high concentrations and the diffusional efflux of NA will tend to rise in parallel with the intracellular concentration. Thus at high concentrations uptake-2 has little net effect. At a medium concentration of 1 JULM, O-methylation accounts for about 50% of the metabolised NA. Since COMT is located extraneuronally, uptake-2 is clearly important only at low concentrations because at higher concentrations O-methylation becomes the rate limiting factor. Uptake-2 is blocked by phenoxybenzamine, which also blocks uptake-1 and by steroids such as hydrocortisone and oestradiol. On the isolated nictitating membrane of the cat the concentration of NA required to cause a contraction is about 10 juM and this is unaffected by hydrocortisone at a concentration of 28 juiM. After block of uptake-1 by cocaine, NA acts at lower concentrations in the region of 1 juiM. At this lower concentration of NA, hydrocortisone now potentiates the effect of NA. On other preparations the beta-stimulant or inhibitory effects of isoprenaline, acting at low concentrations since this agent is not a substrate for uptake-1, are potentiated by hydrocortisone, whereas the effect of NA, acting at higher concentrations is not. Such sensitisation of the tissue by steroids is absent if COMT is inhibited. It is therefore apparent that COMT must be operative for uptake-2 to be effective at terminating transmitter action. Monoamine oxidase is not important in this context, even though it is located both in the neurones and in non-neuronal tissue, because its dissociation constant is much larger (KD 138 |uiM).

64

Autonomic nervous system

Receptors for noradrenaline It is now generally accepted that there are four types of receptors for noradrenaline (NA). There are two classes of alphareceptors, a-1 and a-2 and two beta-receptors, (3-1 and (3-2. A classification of these receptors, based upon the actions of agonists and antagonists, is given in Table 4.3. In general, activation of a-1 receptors causes a contraction of smooth muscle, including that of the blood vessels, pilomotor muscles, dilator pupillae, vas deferens, nictitating membrane, splenic capsule, sphincters of the intestine and urinary bladder and of the bile duct. An exception is the relaxation of the smooth muscle of the intestine. Receptors of the a-2 variety are located presynaptically on the terminals of sympathetic nerves, where they mediate inhibition of transmitter release, but they are also found in the central nervous system and there are some located postsynaptically at the periphery. Beta-receptors of the (3-1 type mediate an increase in the heart rate and an increased force of contraction: they are also found in the central nervous system. (3-2 receptors are well-known for their involvement in relaxing the bronchioles where the selective (3-2 agonist, salbutamol, is of great value in the treatment of asthmatic conditions. Isoprenaline was formerly used for this purpose but its (3-1 stimulant effect on the heart was undesirable and in some cases fatal. The endogenous catecholamines, adrenaline from the adrenal medulla and noradrenaline released from sympathetic nerves, of course act upon all receptors, although their affinities for each may vary. A simple example of this multiple action may be seen when noradrenaline is injected into the circulation before and after the administration of phenoxybenzamine, which has a relatively greater antagonist effect at a-1 receptors than it has on beta receptors or a-2 receptors. Before the administration of the antagonist, NA causes a rise in blood pressure due to activation of a-1 receptors on the blood vessels. After phenoxybenzamine NA now elicits a fall in blood pressure due to activation of (3-2 receptors, previously masked by the a-1 effect (Fig 4.8). The effects of three different agonists, noradrenaline, adrenaline and isoprenaline

Receptorsfor noradrenaline

65

Table 4.3. Catecholamine receptors: classification and characterisation Receptor

A: Agonists: Methoxamine Phenylephrine Noradrenaline Adrenaline Clonidine Isoprenaline Dobutamine Methoxyphenamine Salbutamol B: Antagonists: Prazocin Phenoxybenzamine Phentolamine Tolazoline Yohimbine Dichloroisoprenaline (partial agonist) Propranolol (non-selective) Practolol (cardio-selective) Metoprolol (cardio-selective) Butoxamine (3-2-selective) Labetolol

a-1

a-2

3-1

3-2

+ + + +

+ + + + +

0 0

0 0

+ +

+ +

0 0 0 0 0

'a-blockers'

++++ +++ ++ ++ +

'fi-blockers' 0

0

0

0 0 0 0

+ +

0 0

0

+ + + + ++

0 0 0 0 0

0 0 0 0 0

0

+

+

+ +++ +++ +

+ + + ++

0 0 0 0 0 0 0 0 'mixed, selective a-1 and non-selective 3 0

+ +

Note: a '+' indicates that the agent is effective as an agonist or antagonist. The number of+"s indicate the relative selectivity for a given agent for the different receptors and are not quantitative. Most of these substances are not absolutely selective and sufficiently high concentrations are likely to affect all receptors.

66

Autonomic nervous system

on the systolic and diastolic blood pressure, the heart rate and the peripheral resistance are shown diagrammatically in Fig. 4.9. The complexity is produced by the fact that the different agonists have different relative effects upon the various adrenoceptors, causing a variety of direct and reflex effects upon the measured parameters. Presynaptic receptors. There are presynaptic receptors for catecholamines and ACh located on presynaptic terminals (Fig. 4.10). They are involved in the Ca2+-evoked release of Fig. 4.8. 'Adrenaline reversal' by a-blocking drugs in the cat. Diagram of increases in arterial blood pressure caused by adrenaline or noradrenaline before and after the intravenous injection of phenoxybenzamine. Phenoxybenzamine Arterial blood pressure

NA

Adr

NA

Adr

«i

Fig. 4.9. Receptor participation in the cardiovascular effects of catecholamines in man.

Peripheral resistance (PR) Pulse r a t e Systolic BP Diastolic BP Mean BP Primary Mechanisms Receptors

Noradrenaline

Adrenaline

Increased

Decreased

Decreased

Decreased (reflex) Increased Increased Increased

Increased

Increased

Increased Decreased Increased

Increased Decreased Decreased

Increased PR

Cardiac stimn Decreased PR 01 02

Isoprenaline

Cardiac stimn Decreased PR 01 02

Receptorsfor noradrenaline

67

neurotransmitters. To a large degree, presynaptic receptors have been revealed by techniques measuring directly the transmitter output and from measurements of radioligand binding. Such techniques avoid the complications inherent in deductions made from recordings of responses such as contractile responses or electrophysiological responses from innervated organs. NA can act back on presynaptic alpha receptors to modulate its own release by nerve impulses. These receptors do not modulate release by other mechanisms such as the indirect action of tyramine which will be considered anon. An example of this effect is the enhancement of the tachycardia produced by sympathetic nerve stimulation (a beta-effect) by alpha-adrenolytics such as phenoxybenzamine. In all tissues studied, the alphablockers enhance NA release. These observations show that the effect of the presynaptic alpha-receptors is one of feedback-inhibition. Clonidine, a selective a-agonist is usually effective at such sites. The feedback-inhibition is most apparent at low frequencies of nerve stimulation.

Fig. 4.10. Presynaptic receptors and transmitter release. Sympathetic

Parasympathetic

68

Autonomic nervous system

Feedback-facilitation of transmitter release may also occur. Thus beta-agonists, such as salbutamol enhance the release of NA in some, e.g. human blood vessels, but not in all tissues. The effect is blocked by p-2 antagonists but is not mimicked by (3-1 agonists. In contrast, in the vessels of the hind limb of the cat, the presynaptic beta receptors appear to be of the (3-1 variety. There are also presynaptic receptors for catecholamines upon non-adrenergic neurones. Thus, ACh release from cholinergic nerves is decreased by NA acting via a-2 receptors. Conversely, vagal stimulation of cholinergic fibres to the heart in vivo or in vitro decreases the NA released by stimulation of the sympathetic nerve to the heart. Not only are there autoreceptors for catecholamines upon sympathetic noradrenergic nerves, but there are also NA receptors upon parasympathetic nerve terminals and muscarinic receptors for ACh upon noradrenergic nerves. Regulation of adrenoceptors. The number, and occasionally the affinity, of adrenoceptors is regulated by a variety of agents such as agonists, antagonists, steroids and guanine nucleotides. These are summarised in Table 4.4. These actions may be important in relation to some of the effects of long term treatment with antagonists. Membrane and intracellular consequences of adrenoceptor activation There are two major mechanisms involved in the interaction of catecholamines with their receptors, the first involving changes in calcium conductances and the phospho-inositol (TF) response and the second related to activation or inhibition of adenylate cyclase. These are illustrated in Fig. 4.11 Excitation of smooth muscle via a-1 receptors, e.g. in the uterus, nictitating membrane and vascular smooth muscle, is accompanied by an increase in intracelllular free calcium. This is due in part to an increase in the calcium ion conductance of the cell membrane and in part to the release of calcium bound to the inner surface of the cell membrane.

69

Consequences ofnoradrenaline activation Table 4.4. Regulation of catecholamine receptors Agent

Effect

Noradrenaline p-antagonists GTP

Decrease number of p-receptors Increase number of P-receptors Decrease KA for P-agonists but not antagonists: decrease a-2 not a-1 receptors Decrease a receptors

a-agonists Denervation or Oestrogen

Increase a-receptors

Fig. 4.11. Consequences of adrenoceptor activation. A. a^-receptor activation OLx activation

breakdown of phosphoinositides ('PI' response)

Inositol triphosphate

Diacyl glycerol

I

Directly opens Ca2+ channel

Activation of protein kinase C

Release of sequestered Ca2+ -•Increased [Ca2+]

B. a2- and ^-receptor activation

p

Ns

Adenylate cyclase

ATP

Ni

oc2

c-AMP Activation of protein kinase

/32 relaxation Inhibit Ca-calmodulin complex Facilitate Ca2+ efflux Inactivate myosin light chain kinase

j3x excitation (heart) Increased Ca2+ ion conductance

70

Autonomic nervous system

The increased calcium level accelerates the breakdown of polyphosphoinositides. The products of the breakdown, inositol triphosphate and diacylglycerol, may lead to the activation of Ca2+-dependent kinases and protein kinase-C, respectively. Effects mediated through beta- or a-2-adrenoceptors are related to inhibition or activation of adenylate cyclase through the coupling of the receptors to the inhibitory (N) or stimulatory (Ns) regulatory proteins which influence adenylate cyclase through guanosine triphosphate (GTP). Beta receptor activation leads to activation of adenylate cyclase whereas a-2 activation leads to inhibition. In the heart, where (3-1 receptors mediate an increase in heart rate and force of contraction, the consequence of activation of protein kinase is an increase in the conductance of the muscle fibre membranes to calcium ions. In situations in which the effects of beta stimulation are to cause relaxation, as in the relaxation of bronchiolar smooth muscle by activation of (3-2 receptors, a similar activation of protein kinase may lead to several actions responsible for the relaxation. These may include: (i) Inhibited formation of the Ca2+-calmodulin complex, (ii) Facilitated efflux of Ca2+ by Ca-ATP-ase. (iii) Inactivation of the light chain protein kinase involved in the contractile process. Before leaving the subject of presynaptic receptors it is as well to remember that autoregulation of transmitter release in the cholinergic system also occurs. This conclusion is based upon experiments in which it was shown that 'cold' ACh or tremorine, an agonist at muscarinic receptor sites, decreased the measured release of 3H-ACh from the ileum of the guinea pig and that the effect was reduced by atropine acting as an antagonist at muscarinic sites. Directly and indirectly acting sympathomimetic amines Sympathomimetic agents such as noradrenaline, adrenaline, isoprenaline, phenylethylamine, salbutamol and clonidine (Table 4.2) act directly upon their respective re-

Inhibition of uptake mechanisms

71

ceptors. Other sympathomimetics act by an indirect mechanism. Tyramine and amphetamine may be considered to be examples of this type. Both substances are taken up by uptake-1 into the nerve terminals and are then taken into the vesicles where they displace the stored noradrenaline. Noradrenaline is released to act upon the postjunctional receptors. This action may explain the peripheral actions of amphetamine but other actions are also important for its effects upon the central nervous system. The amines taken up into the granules may themselves be stored and subsequently released as 'false transmitters' by nerve impulses. If the affinity of these amines for the receptors is low then they will be ineffective when released. However, a-methyldopa is taken up, metabolised to amethylnoradrenaline, stored and released. At peripheral a-1 receptors a-methylnoradrenaline is not greatly less potent than noradrenaline itself and so transmission at these sites is unlikely to be impaired as a consequence of this action. The indirect action of tyramine is easily demonstrated (Fig. 4.12). For example, the rise in blood pressure or the contraction of the nictitating membrane of the cat following intravenous injection is prevented by treatment either with cocaine, which blocks uptake-1, or by reserpine, which depletes the stores of their catecholamine content. It is also absent when the nerve is sectioned and allowed to degenerate. None of these procedures reduce the action of directly acting amines. Repeated administration of indirectly acting amines at short intervals leads to a progressive reduction in the responses as the terminals are depleted of their noradrenaline content. Finally, tyramine produces no response from the nictitating membrane if the nerve supply is sectioned and allowed to degenerate. Some substances such as ephedrine and meta-tyramine have mixed direct and indirect actions upon the receptors. Inhibition of uptake mechanisms Inhibition of uptake-1 readily leads to a potentiation of the responses to amines which have a high affinity for uptake-1,

72

Autonomic nervous system Fig. 4.12. False transmitters and uptake. Diagram of responses of any suitable test object such as the nictitating membrane of the cat to noradrenaline (NA), adrenaline (A) and tyramine (T), before and after the administration of cocaine or reserpine. NA is a good substrate for uptake 1 and acts directly on the receptor and so the effect is potentiated by cocaine. A also acts directly but is a poor substrate for uptake 1 and so is not potentiated. Tyramine is a good substrate for uptake 1 but can only act via the displacement of NA from the granules. The action of T is therefore blocked by cocaine. Reserpine depletes the terminals of their catecholamine contents: pretreatment therefore only blocks the action of indirectly acting agents such as T. Control

After cocaine

After reserpine

After reserpine + NA infusion

Miscellaneous drug actions

73

e.g. noradrenaline, but has less effect on directly acting amines with a lower affinity, such as adrenaline, and abolishes the action of indirectly acting amines such as tyramine. The action of cocaine is illustrated in Fig. 4.12 and similar effects are obtained with imipramine. Inhibition of uptake-2 by agents such as the steroids only leads to potentiation of the action of directly acting amines when they are present at concentrations which do not saturate catechol-<9methyltransferase, as described previously. Miscellaneous drug actions Inhibition of monoamine oxidase or catechol-o-methyltransferase.

Inhibition of these enzymes slightly potentiates the action of catecholamines but the effects are not marked. This is due to the fact that the major determinants of the action of the neurotransmitter are the uptake mechanisms. Reserpine. Reserpine is avidly taken up by the storage vesicles and is stored at up to 10 000 times the concentration of catecholamines. At low concentrations, sufficient to block sympathetic neurotransmission, the impairment of uptake and binding of noradrenaline in the vesicles is reversible. At higher concentrations the effect is irreversible. An injection of reserpine causes a rapid depletion of amines from the nerve terminals both at the periphery and in the central nervous system. The effect lasts for several days. The action is accompanied by marked sedation, hypothermia and hypotension. If intraneuronal deamination of the released amine is prevented by the administration of the inhibitor of monoamine oxidase then the injection of reserpine causes stimulation and an elevation of the blood pressure. This is due to the displaced noradrenaline leaking from the terminals to interact with the postjunctional receptors. In order to block sympathetic nerve transmission there needs to be a very large depletion of catecholamine from the terminals. In the heart, this needs to exceed 90%. The rate and extent of the depletion varies from tissue to tissue and is greatest in those tissues in which there is a high rate of turnover of noradrenaline, as in the heart.

74

Autonomic nervous system

The importance of uptake mechanisms in the actions of some adrenergic neurone blocking drugs Guanethidine. Guanethidine causes a fall in blood pressure by blocking sympathetic nerve transmission. This property has found application in the management of essential hypertension. If uptake-1 is blocked by cocaine or imipramine, the fall in blood pressure is delayed and may be temporarily replaced by a rise in pressure. Guanethidine is a substrate for uptake-1, which is necessary for its action, and is stored in the granules, displacing noradrenaline and so depleting it in the process. This action may contribute to the fall in pressure after a time but the fall antecedes the depletion. The initial fall in pressure is due to the membrane stabilising action of the compound. Tolerance to repeated administration rarely occurs. Bretylium. Bretylium is no longer in use as a antihypertensive agent because a marked tolerance rapidly develops. It blocks sympathetic neurotransmission by preventing the release of noradrenaline. Like guanethidine, it enters the terminal by uptake-1, thus competing with NA for uptake. This action may partly explain the tolerance if potentiating the action of NA more than compensated for the reduction in release. Bretylium is accumulated in sympathetic nerves at concentrations which would be local anaesthetic if applied outside those nerves. It seems therefore that the block of transmission is due to an electrical stabilisation of the terminal membrane, and there is little depletion of the NA store. Bethanidine. Bethanidine has an action similar to that of guanethidine but its action is more rapid in onset and shorter in duration. The hypotensive action may be completely reversed by block of uptake-1, showing again that entry in to the terminal is essential for its action. Other antihypertensive drugs One major disadvantage of the antihypertensive drugs so far discussed is that they tend to cause a degree of postural

Other antihypertensive drugs

75

hypotension, related to the fact that the sympathetic vasomotor control is rather crudely blocked either in the terminals or at ganglia. A number of other agents seem to lower the blood pressure in a more circumspect fashion. One of the more interesting of these drugs is clonidine. It will be recalled that clonidine is an agonist at mainly presynaptic a- receptors and inhibits the release of noradrenaline. It will therefore come as a surprise that when first injected intravenously there is a rise in blood pressure preceding the fall. The rise is due to an activation of a-1 receptors associated with vasoconstriction, which overrides the action on the a-2 receptors. Direct injection into the brain however causes only the expected fall in pressure, associated with an a-2 effect. It is particularly effective when injected into the medullary centres associated with vasomotor control, especially the nucleus tractus solitarius. It is unknown whether the effect is pre- or postsynaptic, and one might say that it matters little. The net outcome of the central action is a somewhat more coordinated decrease in blood pressure than is achieved with agents such as guanethidine. Clonidine reduces those reflexes with an efferent pathway in the sympathetic vasomotor nerves but increases, synergistically, vagally mediated reflexes. Alpha-methylDOPA is metabolised to a-methylnoradrenaline. We have already seen that this false transmitter is too potent for this mechanism to be the explanation for the useful antihypertensive action of this substance. Again, the action is probably on the central nervous system, although the mechanism is less well established than it is for clonidine. It seems likely that amethylnoradrenaline is a more potent activator of central a-2 receptors than noradrenaline. Beta-blocking agents such as propranolol have useful antihypertensive effects. Part of the action may be attributed to a block of p-1 effects on the heart but this is not the only, or even the major action. There are a number of other explanations which have been proposed. The secretion of renin by the kidney is under sympathetic control: renin is part of the reninangiotensin system in which angiotensin II is an extremely pow-

76

Autonomic nervous system

erful vasoconstrictor. A reduction in the sympathetically evoked release of renin may contribute to the antihypertensive effect. Although less certain, an action of the beta-blockers on the central nervous system has also been suggested. The converting enzyme which converts angiotensin I to angiotensin II is inhibited by drugs such as captopril which have some antihypertensive effect. The same, or a very similar enzyme also metabolises bradykinin, a potent vasodilator, to inactive products. Inhibition of the converting enzyme may therefore lower blood pressure by this additional action. Alpha-blocking agents, such as phentolamine, are generally of little value in the treatment of hypertension, and tolerance may soon develop. Nevertheless, they have been of value in lowering blood pressure in phaeochromocytoma, a chromaffin cell tumour causing the massive overproducton of adrenaline which enters the circulation. An alternative approach to this problem would be to reduce the synthesis of the amine. Mixed alpha- and beta-blockers, such as labetalol, have been more successful. Diuretics, in particular the thiazides, are also useful in the management of essential hypertension but these probably act mainly by reducing the circulating blood volume. Denervation supersensitivity Cannon and his colleagues showed in the 1930s that section and subsequent degeneration of sympathetic nerves led to a marked supersensitivity of the denervated structure to exogenous catecholamines. The increase in sensitivity ranged from 10- to 100-fold in different tissues, such as the spleen on section of the splenic nerves, or the nictitating membrane on section of the postganglionic cervical sympathetic nerve or the heart on removal of the stellate ganglion. Unlike skeletal muscle, where denervation supersensitivity is almost entirely due to a proliferation of ACh receptors, in smooth muscle the increase in sensitivity is mostly explained by the associated loss of neuronal re-uptake of noradrenaline, with only a lesser role for the increase in receptor numbers.

Muscarinic receptors

11

Cholinergic transmission at autonomic postganglionic nerve endings Cholinergic transmission in the autonomic nervous system occurs in ganglia, as already discussed, and at all postganglionic endings of parasympathetic nerves and at the endings of a few sympathetic nerves. Muscarinic receptors The endogenous neurotransmitter at postganglionic cholinergic endings is acetylcholine, which acts upon receptors which are sensitive to the agonist muscarine: they are therefore called muscarinic receptors. Muscarinic agonists which are relatively selective for the muscarinic receptors compared with their actions on nicotinic receptors include muscarine, acetyl-pmethylcholine (methacholine), bethanecol, pilocarpine and oxotremorine: the last of these agents has stimulant effects on the central nervous system. Acetylcholine and carbachol do not distinguish between muscarinic and nicotinic receptors. Agonists. Muscarinic agonists are used in conditions in which a selective activation of smooth muscle of the gastrointestinal tract or urinary bladder is required. Another use is in glaucoma. Generally, an action on the nicotinic receptors in autonomic ganglia would be undesirable and so the selective agonists are used more than those which do not distinguish between nicotinic and muscarinic receptors. Both acetylcholine and methacholine are good substrates for acetylcholinesterase, but the other agonists are not. Antagonists. Muscarinic receptors are blocked by atropine and by a number of related antagonists such as hyoscine, and atropine methyl nitrate, a quaternised derivative which, being fully ionised, does not cross the blood-brain barrier and so avoids the problem of actions on the central nervous system. The muscarinic receptors on smooth muscle differ from those in ganglia in being relatively insensitive to the antagonist pirenzepine or the agonist McNeil A-343.

78

Autonomic nervous system

Cholinesterase inhibitors The use of inhibitors of cholinesterase in myaesthenia gravis is discussed in the preceding chapter. Agents used for this purpose are fully reversible and either very short acting or have a medium duration of effect. Edrophonium is very short acting and forms an ionic bond with the enzyme and is therefore a competitive inhibitor. Neostigmine is longer acting and is weakly covalent in its binding to the enzyme: it is a substrate which carbamylates the esteratic site on the enzyme but is only slowly hydrolysed. Pyridostigmine acts similarly to neostigmine but can be taken by mouth since its absorption is better, even though it also has a quaternary nitrogen atom in its structure. Physostigmine (eserine) has a similar action but is non-quaternary and so crosses the blood-brain barrier. Organophosphate anticholinesterase drugs were developed as pesticides and as nerve gases in the First World War. Some also have a use in the treatment of glaucoma. The original nerve gas developed was DFP (diisopropyl phosphofluoridate). Orgnophosphate insecticides include tetraethyl pyrophosphate, which was one of the earliest, and others which are more recent such as parathion and malathion. Ecothiophate is used in the treatment of glaucoma. Their toxicity is high due to the fact that they form covalent links to the enzyme. The phosphorylated enzyme is extremely stable after it has 'aged' for a few hours but the enzyme may be re-activated by treatment with pralidoxime or newer reactivators such as obidoxime, provided that treatment is instituted at an early stage. The toxic actions of all of these substances are due entirely to the inactivation of cholinesterase causing vastly potentiated activity in the parasympathetic nervous system, at all autonomic ganglia and at the skeletal neuromuscular junction. The symptoms include secretions in the respiratory tract which together with bronchoconstriction may well be fatal, excessive stimulation of the alimentary canal and urinary bladder, occular effects and marked cardiovascular actions. If the inhibitor penetrates to the central nervous system then there will also be effects of CNS origin. The relative degree of each effect may well depend upon the

Cholinesterase inhibitors

79

route of administration. Many of the substances are well absorbed through the skin and through the airways, making them exceedingly hazardous to use.

Central neurotransmitters and neuromodulators This chapter is devoted to a consideration of some basic aspects of transmitter action which are not treated elsewhere in this book. A more detailed treatment of opiate, GABA, benzodiazepine, 5-HT and dopamine receptors will be found in Chapters 8-11. Of the large number of drugs which act upon the central nervous system, many appear to do so by an effect upon synaptic transmission. In the evolution of the nervous system a variety of chemical substances have assumed the role of chemical mediators and some of these have become associated with particular functions to a variable extent, although there is considerable overlap. Perhaps as a consequence of this specialisation, derangement of neuronal functions may sometimes be associated with defects in the operation of one transmitter system whilst others appear to operate in a more or less normal fashion. Drugs which act in the most selective fashion are therefore likely to be those with the most specific site of action on a process which in turn is relatively specific for a particular transmitter. Such drugs could act either by modifying selective binding of the transmitter to its receptor site or by effects on the synthesis, degradation or release of the transmitter. They could also operate on processes subsequent to the binding of the transmitter to its receptor but uniquely associated with that transmitter. The overlapping functions of neuronal pathways utilising the same transmitter leads to the generalisation that drugs selectively operating upon a transmitter system may affect a number of functions not all of which are necessarily deranged in disease. The topography of the connections of neurones employing different transmitters, a knowledge of changes which occur in dis-

Central neurotransmitters and neuromodulators ease states and an understanding of the mechanism of action of useful drugs may therefore lead to a rational approach to therapy. Over the last few decades there have been many advances in the development and application of techniques for the identification of central neurotransmitters and for mapping their presence in the CNS, coupled with advances in neurophysiological and micropharmacological methods which have led to a greater understanding of central neurotransmitters. Table 5.1 and Fig. 5.1 lists a number of endogenous substances which are currently under consideration as neurotransmitters in the CNS. In order to establish the identity of a neurotransmitter it is necessary to fully substantiate only two criteria. The first is that the putative transmitter is released when the presynaptic nerve is stimulated and the second is that the putative transmitter, when administered as an exogenous substance to the postsynaptic neurone, reproduces exactly the effect of the endogenously released substance. Close examination of this statement reveals that it embodies a number of accessory assumptions which will provide a partial solution to the question,fcissubstance X a transmitter?'. If a substance can be released then it must be assumed to be present and probably synthesised. Thus much of the supporting evidence for the identity of a particular transmitter is based upon demonstrations of the presence and distribution of the substance and its associated synthetic or degradative enzymes within the CNS. Release of the substance when particular pathways within the brain are activated is technically more difficult to demonstrate, but this has been done in many instances. Identity of action of exogenously administered and endogenously released neurotransmitter embodies a number of subsidiary considerations, all of which are capable of experimental verification. There are two which are probably the most informative. The first is the demonstration that the pharmacology of the exogenous substance is identical to that of the endogenous transmitter. Pharmacological observations within this category are likely to include structure-activity studies with agonists, studies with specific receptor antagonists, enzyme activators or inactivators, uptake inhibitors, etc. The second is that the precise mem-

81

82

Central neurotransmitters and neuromodulators

Table 5.1. Neurotransmitters, their receptors and their significance in the central nervous system Transmitter

Receptors

ACh

Muscarinic

Significance

Parkinson's disease Alzheimer's disease Huntingtorfs chorea (Nicotinic) Nociception a Noradrenaline a-2 Antihypertension H-1 Motion sickness Histamine D1,D2 Schizophrenia Dopamine Tardive dyskinesia? Parkinson's Disease Depression 5HT-1, -2 and -3 5-hydroxytryptamine Nociception GLY Convulsants Glycine Non-ketotic hyperglycinaemia GABA Convulsants, GABA-A GABA-B antispastic anti-epileptics benzodiazepine actions BDZ-1 and BDZ-2 Benzodiazepines Enkephalins Analgesia m 8, y, K opiate Neurokinins Nociception Substance P, NK-A & NK-B Capsiacin Excitatory transmission Acidic amino acids NMDA, kainate, quisqualate Epilepsy? Stroke? Prostaglandins ?? Central analgesic PG Antipyresis ??? Other neuropeptides e.g. somatostatin, cholecystokinin, neurotensirI, bombesin, angiotensin, vasoactive: intestinal polypeptide (VIP), etc. brane mechanism by which the transmitter causes excitation or inhibition is identified in both instances. Almost inevitably, most investigations of this type will be electrophysiological in nature. Such evidence is most complete for the inhibitory transmitter glycine but modern techniques and technology are rapidly in-

83

Acetylcholine

creasing our understanding of the actions of neurotransmitters. Subsidiary evidence relating to the mechanisms of action can be derived from binding or biochemical studies such as enzyme activation, as in the example of dopamine-activated adenylate cyclase, or inactivation. Acetylcholine Acetylcholine is found in many parts of the brain, especially high concentrations being found in the caudate nucleus and Fig. 5.1. Putative central neurotransmitters or modulators. O

CH,

CH.CH.NH,

H 3 CCOCH 2 CH 2 N—CH 3 ^CH3

>X

^N' H

5-hydroxytryptaminc HO HO

CHCH2NH2

HO

OH

HO

CHCH2NHCH3 OH Adrenaline

HOOCCH2CH2CH2NH2 y-Aminobutyric acid (GABA) HOOCCH,CHCOOH

I

NH :

HOOCCH2NH2 Glycine HOOCCH,CH,CHCOOH II NH, Glutamic acid Tyr-Gly-Gly-Phe-Leu Leucine enkephalin

Aspartic acid Arg-Pro-Lyc-Pro-Gln-Gln-Phe-Phc-Gly-Leu-Met-NH: Substance P

84

Central neurotransmitters and neuromodulators

cerebral cortex. In general, the distribution of the transmitter corresponds to the distribution of choline acetyltransferase and of the muscarinic cholinergic receptor, but there is increasing interest in nicotinic receptors in the CNS. In order to study the topography of cholinergic neurones a sensitive technique for visualising either acetylcholine or choline acetyltransferase is needed. A histochemical technique for visualising nerve tracts containing acetylcholinesterase was the first useful technique to be developed and extensive maps were obtained. However, there are a number of esterases which may not function in neurotransmission and such maps cannot be interpreted without supporting evidence. In particular, acetylcholinesterase occurs in a high concentration in cerebellum, and nerve fibres containing the enzyme have been demonstrated with the copperthiocholine technique. However, there is little acetylcholine, choline acetyltransferase or cholinergic receptor material in cerebellum and neurophysiological evidence of cholinergic transmission is lacking. Acetylcholine usually excites central neurones when administered by microelectrophoresis, but inhibitory effects have also been shown, especially in nucleus reticularis of the thalamus and in the brain stem. There is some neurophysiological evidence for cholinergic pathways to the cerebral cortex and hippocampus but, in general, precise definition of the pathways is not yet possible. The synapse formed by the termination of collaterals of motor axons on Renshaw cells in the spinal cord is the best characterised site of cholinergic transmission in the CNS (Fig 5.2). Here, the action of acetylcholine released from presynaptic terminals of the axon collaterals exerts a short latency monosynaptic excitatory action on the Renshaw cell followed by a long-latency, prolonged period of excitation with a different mechanism. The short-latency excitation causes cell firing lasting for about 50 ms and is blocked by pharmacological agents which block nicotinic receptors for acetylcholine, such a dihydro-(3erythroidine and d-tubocurarine, and is prolonged by inhibitors of cholinesterase. The major action of acetylcholine administered micro-

Amino acids

85

electrophoretically is also due to excitation of nicotinic receptors. The release of acetylcholine is blocked by the administration of botulinus toxin and direct assays from perfusates of spinal cords have demonstrated an increased release after antidromic stimulatiion of ventral roots. The nicotinic actions of acetylcholine released by stimulation of motor axons or applied from micropipettes is potentiated by the administration of inhibitors of acetylcholinesterase. The long-latency excitation after antidromic stimulation of ventral roots lasts for up to 5 s and is due to the interaction of released acetylcholine with muscarinic receptors because it is antagonised by atropine, well known for its ability to block peripheral muscarinic receptors. After blockade of nicotinic receptors with dihydro-p-erythroidine, a muscarinic action can be demonstrated by the iontophoresis of acetyl-p-methylcholine or muscarine. The membrane mechanisms of the action of acetylcholine on Renshaw cells have not been clearly elucidated although nicotinic effects appear to be associated with an increase in membrane conductance. Muscarinic actions of acetylcholine are the most prominent elsewhere in the nervous system. Krnjevic has shown that the muscarinic excitatory effect on cerebral cortical neurones is associated with a depolarisation and a decreased membrane conductance, reflecting a selective decrease in permeability to potassium ions. Experiments performed on hippocampal slices have demonstrated that the muscarinic actions of acetylcholine resemble those at the periphery and are produced by reducing a potassium ion conductance which is voltage dependent, i.e. it resembles the 4 M-currenf in ganglia.

Amino acids There are three amino acids commonly considered to be transmitters in the CNS. These are glycine, y-amino-butyric acid (GABA) and glutamic acid. Some other amino acids, such as taurine and aspartic acid, have also occasionally been proposed as transmitters but the evidence for this is weak.

86

Central neurotransmitters and neuromodulators Early discharge

(b) 40-| N

(c)

Nicotinic excitation

Frequ enc

X

20 ms

PST

-

Late discharge

ill. i

20-

o-

J

Time (s)

(d)

0J

Time (s)

Fig. 5.2. Pharmacology of cholinergic synapses on Renshaw cells. (Curtis, D.R. & Ryall, R.W. 1966. Exptl Br. Res., 2, 81-96 and Ryall, R.W. 1975. In Handbook of Psychopharmacology, ed. Iversen, L.L., Iversen, S.D. & Snyder, S.H. 4, pp. 83-128. New York, Plenum Press.) Stimulation (arrows) of the ventral root (VR) with single electric shocks gives rise to excitatory-inhibitory-excitatory sequences in Renshaw cells (RC). The motoneurones (M)

Amino acids

87

Glutamic acid is ubiquitous in the central nervous system and serves metabolic as well as transmitter functions. It is likely to be a major excitatory transmitter in the CNS. When applied to nerve cells it causes a depolarisation, due to an increased permeability to sodium ions. Depolarisation is also sometimes due to changes in the conductance of voltage-dependent potassium channels. Depolarisation is usually sufficient to increase the firing rate of neurones. Antagonists for glutamic acid have revealed that there are multiple receptors for excitatory amino acids. Based on agonist and antagonist actions, the receptors have been designated as N-methyl-D-aspartic acid (NMDA), quisqualate and kainate receptors. The best antagonists are those which act selectively on the

Fig. 5.2 (cont.) have axons which branch within the grey matter to form collaterals which synapse on Renshaw cells. ACh is liberated to interact first with nicotinic receptors, the activation of which evokes repetitive action potentials (early discharge) lasting for about 50 ms (a). The action potentials occur at about 1 ms intervals (frequency 1 kHz) at the beginning of the discharge. The early discharge is attenuated by dihydro-j8-erythroidine which competitively antagonises the effect of ACh on nicotinic receptors. The full sequence of events is shown schematically in (e) and as a peristimulus historgram on the same time scale in (c). The record in (b) is from an actual experiment. The histograms are computed from several repetitions of the stimulus on the same cell and represent the average firing frequency at various times before and after the stimulus. The early discharge is followed by a 'pause1, during which the excitatory background activity is suppressed. The pause is due to a combination of mutual inhibitory connections between the Renshaw cells, where the transmitter is probably glycine, and to desensitisation. The late discharge follows the pause and has a prolonged time course lasting for several seconds. It is not blocked by dihydro-j3-erythroidine but is abolished by atropine, a competitive antagonist of ACh at muscarinic sites.

88

Central neurotransmitters and neuromodulators

NMDA receptor and it has been shown that some polysynaptic processes are inhibited by such antagonists. The voltage-dependency of the NMDA-activated channels is due to the rather unusual phenomenon of voltage-dependent block by Mg2+ ions. There is interest in the possible therapeutic action of antagonists to excitatory amino acid receptors to reduce brain cell damage in stroke victims and it seems likely that some advances may be made within the next few years. Selective antagonists are available for glycine and GABA (Chapter 9) and the evidence that glycine or GABA is the transmitter at certain sites is strong. There is strong evidence that there are two receptors for GABA, GABA-A and GABA-B. Glycine is distributed mainly in the spinal cord and medullary parts of the brain stem. It invariably inhibits neurones by increasing membrane permeability to chloride and potassium ions, causing a hyperpolarisation. It is the mediator of postsynaptic inhibition at a number of identified synapses, including the recurrent inhibitory synapses of Renshaw cells upon motoneurones, the inhibitory synapses of interneurones mediating reciprocal IA inhibition on motoneurones and on Renshaw cells and the inhibitory synapses on motoneurones of interneurones excited by IB fibres from Golgi tendon organs. Strychnine and some related drugs are specific receptor antagonists and has been shown to interact with glycine at binding sites in in vitro systems. The inhibitory effect of glycine is generally more marked on spinal neurones than on supraspinal neurones and it probably acts as a postsynaptic inhibitory transmitter mainly at spinal cord synapses. In contrast, GABA and its associated enzymes are found in higher concentrations in supraspinal than in spinal regions of the CNS. At a number of supraspinal sites it has been identified as the postsynaptic inhibitory transmitter, acting by a membrane mechanism indistinguishable from that of glycine. In the spinal cord it is the most likely mediator of presynaptic inhibition in which afferent terminals become depolarised and transmission reduced by a mechanism which is the subject of considerable controversy. The antagonist bicuculline blocks GABA-A receptors and inhibits some actions of GABA on spinal and

Catecholamines and 5-hydroxytryptamine

89

supraspinal neurones. Presynaptic depolarisation is also blocked and seems attributable to activation of GABA-A receptors. However, there is evidence that GABA-B receptors may be involved in the presynaptic inhibition of transmitter release by GABA. Baclofen is a selective agonist for the GABA-B receptor. Picrotoxin inhibits presynaptic inhibition and some of the postsynaptic effects of GABA by blocking the chloride channels in the membrane. Visualisation of GABA-containing neurones has been more successful than visualisation of glutamate or glycine-containing neurones. Autoradiography, in which labelled amino acid is taken up by a tissue followed by photographic determination of the distribution in a tissue slice, has been extensively used but suffers from the disadvantage that the amino acids may be taken up not only by neurones by also by glial cells. Sensitive immunohistochemical methods for glutamic acid decarboxylase (GAD), which forms GABA from its precursor glutamic acid, have confirmed the presence of specific fcGABA-utilising' pathways in the CNS. Catecholamines and 5-hydroxytryptamine The first highly successful demonstrations of pathways in the CNS containing noradrenaline (NA), dopamine (DA) and 5-hydroxytryptamine (5-HT) were made with a fluorescence technique in which tissue sections are treated with formaldehyde vapour under carefully controlled conditions to form fluorescent isoquinolines. Later investigations with an immunohistochemical technique for dopamine-(3-hydroxylase have confirmed the original observations for NA made with the fluorescence method. In more recent studies condensation with glyoxylic acid is used to form fluorescent products. Major dopaminergic pathways (the nigrostriatal and mesolimbic systems) originate from the region of the substantia nigra and ventral tegmental area and project principally to the basal ganglia, limbic system and cerebral cortex (Fig. 5.3). Another pathway (the tuberoinfundibular system) originates from cells of the arcuate nucleus and projects mainly to the median eminence of the hypothalamus.

90

Central neurotransmitters and neuromodulators

Noradrenergic pathways are more diffuse with cell bodies in the pons and medulla, including a localised group of neurones in the locus coeruleus. There are rostral projections to many parts of the brain and caudal projections to the spinal cord. Microelectrophoretic administration of NA to CNS Fig. 5.3. The origin and distribution of monoaminergic fibres in the brain. Noradrenaline and dopamine pathways redrawn after Livett, B.G. (1973) Brit Med. Bull, 29, 93. Cerebral cortex Noradrenaline

\S:

Hippocampus /

~~. I A T

Locus coeruleus Y-N

^Cerebellum

Olfactory bulb Cingulate gyrus Hypothalamus

Amygdala

Medulla Descending fibres Dopamine Nucleus accumbens

Olfactory tubercle Striatum Arcuate nucleus 5-hydroxytryptamine

Raphe

Catecholamines and 5-hydroxytryptamine

91

neurones usually causes inhibition but excitatory effects have also been demonstrated. There have been rather few studies of the membrane mechanism affected by NA but there is some indication that the inhibitory effect is associated with a hyperpolarisation and an increase in membrane resistance. There is also some evidence that, at least in cerebellum, the effect may be mediated via cyclic AMP. It might be expected that agents which antagonise the effects of NA at peripheral a- and P-receptors might also act in the CNS. Although antagonism has been demonstrated in some studies, it has not been possible to demonstrate specific effects in others. In part this was due to the unrecognised presence of a-2 receptors in earlier studies. Although the amount of adrenaline in the CNS is small relative to that of NA, the presence of phenylethanolamine-Nmethyltransferase has been demonstrated and specific immunohistochemical methods have led to preliminary indications of adrenergic transmission with a rather restricted distribution. 5-HT is located in the cell bodies of neurones predominantly situated in the raphe nuclei in the brain stem. The tryptaminergic neurones project both rostrally and caudally into the spinal cord. Although an immunohistochemical technique is available for DOPA-decarboxylase, the enzyme responsible for the decarboxylation of DOPA is immunologically indistinguishable from that which decarboxylates 5-hydroxytryptophan and so is of limited usefulness. 5-HT may either excite or inhibit neurones when it is administered microelectrophoretically. Lysergic acid diethylamide and related substances have been found to antagonise these effects in some investigations but not in others. It has been suggested that 5-HT systems may be important as a 'trigger' system for slowwave sleep whereas noradrenergic systems originating from the locus coeruleus are involved in 'deep' or 'paradoxical' sleep. In the last few years there has been a growing awareness of three different binding sites for 5-HT, but it was not until 1987 that a binding site of the 5HT3 variety was demonstrated. It is not yet known how many of these sites represent functional receptors.

92

Central neurotransmitters and neuromodulators

Polypeptides It has long been known that the polypeptide, substance P occurs in the CNS and is localised in the synaptosomal fractions of brain homogenates with a subcellular distribution similar to that of acetylcholine. More recently, immunological techniques have been employed to demonstrate the presence of particularly high concentrations especially in the dorsal grey matter of the spinal cord and in the caudate nucleus. Again, there may be several binding sites although the nomenclature is somewhat unsettled. Besides substance P, two other endogenous neurokinins, A and B have been identified. The elucidation of the nature of central receptors will be delayed until more potent specific antagonists are available, which act effectively in the CNS. Another group of polypeptides, namely the enkephalins, have also been shown to have rather specific distributions. There is also evidence for nerve terminals containing other polypeptides such as neurotensin, somatostatin and thyrotrophin releasing factor (TRF) in the central nervous system. Microelectrophoretic studies with synthetic polypeptides have shown that they are potent excitants or depressants of neurones in the CNS but further elucidation of their function at specific synapses awaits future developments.

The blood-brain barrier

Ehrlich showed in the mid-nineteenth century that when acidic dyes such as trypan blue were injected into the bloodstream most of the tissues of the body were stained, except for the brain. The blood-brain barrier is essentially the restraint imposed upon the passage of substances from the blood into the brain or vice versa. The permeability of the blood-brain barrier and the rate at which substances pass across it is determined by the nature of the barrier and by the concentration gradients across it, as well as by the physico-chemical properties of the substances. The barrier determines whether and how fast an administered drug enters the brain and how fast it will leave it as the blood level decreases. It also determines the access and egress of other substances such as metabolites and precursors. The nature of the blood-brain barrier The properties of the barrier arise as a consequence of the special properties of the capillary endothelium which separates the brain substance from the blood capillaries or the choroidal epithelium which separates the cerebro-spinal fluid (CSF) from the blood in the choroidal plexus. The cavities of the brain (cerebral ventricles) are filled with cerebro-spinal fluid (CSF), which differs only slightly in composition from the extracellular fluid surrounding the brain cells. The relationship between, blood, brain, extracellular fluid and the CSF is shown in Fig. 6.1. The CSF is secreted by the choroidal plexus lining the cerebral ventricles. In man the rate of secretion is about 0.3 to 0.5 ml min"1. The rate of turnover is about 10% per hour. The CSF is vir-

94

The blood-brain barrier

tually in equilibrium with the extracellular fluid of the brain (ECF). The choroid plexus is a secretory epithelium with tight junctions interposed between the epithelial, but not the endothelial cells. The ECF comprises about 15% of the brain volume: early estimates of about 2-3% were based upon an incomplete understanding of the kinetics of the equilibrium between blood, ECF and CSF. The CSF and ECF are not simply plasma filtrates, as shown by the fact that the sodium ion concentration is about 8% higher and chloride 17% higher in CSF than plasma and that potassium, calcium, glucose, phosphate and urea are lower in CSF than plasma: CSF is also virtually devoid of protein in the adult and is slightly more acid than plasma. The capillary endothelium differs in structure from that in most other places by the presence of tight junctions between the endothelial cells, also found in some other special tissues such as the kidney tubules. These junctions prevent the simple passive Fig. 6.1. The blood-brain barrier. Blood capillary Active Tight junction T . .. transport / Lipid I A diffusion

Astrocytic processes Basement membrane Extracellular space ICSF Lipid diffusion

Active

transport

Tight

junction Epithelial cells Endotheliurr.

Open junction Blood capillary in choroid plexus

Transfer across the blood-brain barrier distribution of solutes between blood and brain. In order to enter or leave they must first pass through the endothelial cell membranes. This may be achieved either by diffusion, in which case the lipid solubility and concentration of the substance will determine the rate of flux, or by active transport for some molecules, e.g. sugars and amino acids. The feet of large astrocytes cover about 85% of the capillary endothelium and it is possible that the properties of these neuroglial cells may contribute to the properties of the bloodbrain barrier. Factors affecting rate of transfer of substances to and from the brain A number of factors determine how fast a substance enters or leaves the brain and whether it will distribute equally to all regions. Blood flow. The brain represents about 2% of the body weight but receives about 16% of the cardiac output. The blood flow is about 0.5 ml g"1 min"1 compared with one-tenth of this value in skeletal muscle. There are also regional differences within the brain, with a ten-fold difference between different regions. With some drugs in which blood flow might be rate-limiting, e.g. those with a low lipid solubility and which are not actively transported across the barrier, the regional differences would determine which parts of the brain would be first exposed to high drug concentrations. This fact is often overlooked when experimenters try to assess the locus of drug action by determining which parts of the brain show changes in their activity before others after systemic administration of a drug. Thiopentone is highly lipid soluble and therefore rapidly enters the brain. Nevertheless, one minute after an injection of radiolabelled thiopentone the label first appears in the cortex, geniculate bodies and inferior colliculus, precisely those areas with the highest blood flow. After 30 min the entire brain is uniformly labelled. The volatile anaesthetic halothane is found largely in the grey

95

96

The blood-brain barrier

matter 2 min after administration commences but after 5 min has largely redistributed to the white matter, but the reticular formation seems to retain the drug, even after it has disappeared from other regions. Secretion by the choroid plexus. The diuretic acetazolamide is first found in the ependymal lining of the ventricles and in the CSF before it is found in the brain tissue: it is secreted by the choroid plexus into the CSF. Some hours after administration there are regional differences in concentration which are probably achieved by differences in drug binding to the tissues. Drug binding. Some drugs, such as the anti-epileptic drug phenytoin or the anti-schizophrenic drug chlorpromazine reach a concentration in the brain which is many times higher (ten-fold and twenty-fold higher respectively) than in the blood and may be maintained for very long periods of time. These high concentrations are due to the removal of free drug by binding: the equilibria are determined by the concentrations of free rather than bound substance. Conversely strong binding to plasma proteins will delay entry to the brain, providing that the binding is strong enough. This is partly why Ehrlich found that trypan blue did not pass easily into the brain. Lipid solubility. A high lipid solubility is associated with a high rate of penetration into the brain. This has been extensively studied with the barbiturate anaesthetics, where the kinetics of action are exceedingly important in clinical use. Thus thiopentone, a short, fast-acting barbiturate with an oil/water partition coefficient of about 2000 penetrates brain with a half time of about 1.4 min. This contrasts with barbitone which has a coefficient of about 1.0 and a half time for penetration of 27 min. In reality, a substance does not have to be highly lipid soluble in order to gain useful acces to the CNS. Morphine, with a partition coefficient of only 0.02, nevertheless is a useful analgesic: with such substances, the blood flow is likely to be rate limiting but is sufficient for most of the morphine to be removed from arterial blood in a single passage through the brain.

Transfer across the blood-brain barrier

97

Lipid solubility of a weak electrolyte is dependent on the degree of ionisation: the less ionised (more undissociated) the molecule, the greater the lipid solubility and ease of access to the brain. Weak bases are less dissociated at alkaline pH whereas weak acids are less dissociated at acid pH. Quaternary compounds are completely ionised over a very large range of pHs, covering any that occur either physiologically or pathologically, and therefore do not gain access to the brain. This property has been employed to restrict the action of drugs which can be quaternised to the periphery. Methylatropine has a pK of 4.35 and is fully ionised at pH 7.4. It therefore has all of the peripheral effects of atropine but lacks the central side-effects. Hexamethonium is a quaternary ganglionic blocking agent. It has numerous undesirable side effects, but these are all attributable to its effects at the periphery. Mecamylamine is a secondary amine which is 1.6% non-ionised at pH 7.4, sufficient for therapeutic doses to gain entry to the brain and exert unwanted effects on the CNS. Neostigmine is fully ionised at body pH and so can be successfully used as an anticholinesterase agent in myaesthenia gravis without effects on the CNS. In contrast, physostigmine and DFP are not fully ionised and so are not used. Tremorine is used to produce experimental Parkinsonism but it does not cross the blood-brain barrier. It is effective on the CNS by virtue of the fact that it is metabolised to oxo-tremorine which does cross. Penicillin is a water soluble cyclic peptide with a carboxyl group which is completely ionised at physiological pH. It therefore enters the brain only very slowly, but as a weak acid it is effectively secreted by the choroid plexus. It is therefore not very useful in treating infections of the CNS such as meningitis. Chloramphenicol and tetracycline penetrate far more readily. Breakdown of the blood-brain barrier. In meningitis, or as the result of trauma, oedema, seizure activity, arterial hypertension, marked or prolonged hypoxia or ischaemia, ionising radiation, plaque formation, lead or mercury encephalopathy or other

98

The blood-brain barrier

causes, the blood-brain barrier may not be very effective. Thus sufficient penicilHn may cross to combat infection. However, the direct application of penicillin to the cerebral cortex causes seizures and convulsions have been observed when penicillin has been given in meningitis. Active transport across the blood-brain barrier. Amino acids and

glucose are actively transported across the blood-brain barrier, even against a concentration gradient. Many transmitters are primary amines which do not easily gain access to the brain. However, amino acid precursors are readily transported. This is taken advantage of with treatment with L-DOPA in Parkinson's disease, alpha-methyl-DOPA for hypertension, and with 5-hydroxytryptophan (precursor of 5hydroxytryptamine) which has had some success in treating Down's syndrome. Developmental aspects In the foetus the blood-brain barrier is more permeable to water soluble substances than in the adult. This is due to the fact that the capillary endothelium changes from one in which there are few tight junctions between the cells to one in which tight junctions are characteristic of the endothelium, as in the adult. The rate of development of the blood-brain barrier varies with the species, and can be monitored in sheep by measuring the ability of the brain to take up sucrose. The permeability to sucrose decreases by about 90% between 50 days' gestation and birth (at about 145 days), and there is a further small decrease after birth. The protein level in lumbar CSF in man falls from 1 mg ml"1 at 1 year. In the human foetus, spontaneously or therapeutically aborted at 14 to 40 weeks, the level of protein is at its maximum level at about 20 weeks, at which time it reaches a mean value of 7.7 mg ml"1. Another consequence of this slow development of the barrier is that the foetus and the infant is more susceptible to the toxic actions of various substances which would not normally pene-

Neurotoxicity

99

trate to the central nervous system. Thus kernicterus in the newborn is due to bilirubin crossing the blood-brain barrier, causing a yellow staining of the basal ganglia and signs of motor disorder. Neurotoxicity The permeability of the blood-brain barrier is diminished in disease and malnutrition. This can sometimes lead to very severe conditions of neurotoxicity, even in the adult. An example of this is lathyrism. Lathyrism. Lathyrus species, e.g. Lathyrus sativus, Vetch1 are not usually an important part of the diet, but it may be planted in India and North Africa within a cereal planting and the pods used as food under famine conditions. If the rainfall is good then the cereal thrives and the Lathyrus does not. In drought, the cereal fails but the Lathyrus gives a reasonable crop which can assume a large importance in the diet. In small amounts there are no ill-effects, but if the intake of Lathyrus reaches 30 to 50% of the diet over a period of 3 to 6 months the neurotoxic signs of lathyrism appear. These consist of exaggeration of reflexes, leg stiffness and, in severe cases, paralysis, all of which are due to the irreversible destruction of spinal neurones. Two toxic substances have been isolated; these are i) beta-N-oxalyl-L-alpha-beta-diaminoproprionic acid and ii) beta-N-oxalyl-amino-L-alanine. These substances are neurotoxic in neonatal rats before the blood-brain barrier has matured but not in the adult rat unless they are introduced directly into the CNS, e.g. by lumbar puncture. It is therefore clear that they do not cross the blood-brain barrier when it is mature and intact. In severe malnutrition, the integrity of the barrier is impaired, perhaps by acidosis and vitamin deficiency and the neurotoxins may enter the CNS to produce their effects. Supporting this suggestion is the observation that if animals are first made acidotic by giving ammonium chloride or acetazolamide, a carbonic anhydrase inhibitor used as a

100

The blood-brain barrier

diuretic, then the amino acids produce neurotoxic symptoms in the adult. It is a simple matter to destroy the neurotoxic amino acids in Lathyrus sp. by boiling, however, this also destroys essential vitamins and is not such a rational solution. Summary The blood-brain barrier explains differences in the rates of penetration of drugs into the CNS. Normally, substances are either actively transported or must have a high lipid solubility in order to diffuse passively into the brain. In the latter case, the substance will be relatively non-ionised at physiological pH. A knowledge of the blood-brain barrier, and its susceptiblity to injury can explain some unexpected neurotoxic effects, especially in the neonate and in the foetus, and can predict the time course of drug action. It may be possible in the future to induce a temporary increase in the permeability of the barrier in order to allow certain agents to enter which otherwise could not do so. The most exciting example of this would be to allow entry for antibodies to treat specific disorders. However, clearly there needs to be extreme caution with such an approach.

General anaesthetics

The general anaesthetics comprise a large group of compounds of diverse chemical structure which cause a loss of sensation and perception leading in adequate doses to a complete, but reversible loss of consciousness. Although they may cause an attenuation of pain sensation in subanaesthetic doses, this is not invariably so, and the anaesthetics as a class are distinct from the analgesic drugs which selectively reduce the sensation of pain but do not lead to a loss of consciousness at therapeutic concentrations. The general anaesthetics depress function at all levels of the central nervous system and so may depress respiration, circulation, temperature control, voluntary reflex movements and pain, but the degree of depression of each of these functions is not the same with all agents, even at a similar depth of anaesthesia. It therefore follows that the characteristics of anaesthesia with different agents may vary. Types of general anaesthetic The chemical structures of a variety of different general anaesthetics are given in Fig. 7.1. The diversity of structures is readily apparent and must be taken into consideration in evaluating the mechanisms of action. The physical form of anaesthetics also varies from gaseous substances such as the inert gases, nitrous oxide and cyclopropane, or volatile substances such as ether, chloroform and halothane, to soluble substances typified by the barbiturates, eugenols, steroid anaesthetics and cyclohexylamine. To a large degree, the physical form and properties of the anaesthetic determine the manner in which the

102

General anaesthetics

anaesthetic is used, and some of the advantages and disadvantages of each class. Gaseous anaesthetics Nitrous oxide was the first anaesthetic to be used in clinical practice in the middle of the nineteenth century, although its Fig. 7.1. Structures of general anaesthetics. Gaseous:

Nitrous oxide

N:O H, C

Cyclopropane

CH :

Volatile :

Diethyl ether

CH3CH,OCH2CH3

Chloroform

CHCIj

Halothane

CFjCHCIBr

Methoxyfluorane

CH3OCF:CHCU

Soluble (intravenous): Barbiturates (fig. 7.2)

O—CH,CON(OHs .OCH,

Propanidid (a eugenol) CH:COOCH,CH:CH,

Alphaxalone

Ketamine O

NHCH3

Types ofgeneral anaesthetic

103

use was closely followed by that of the volatile anaesthetics, chloroform and ether. At atmospheric pressure, nitrous oxide has a rather low anaesthetic potency. It is administered by inhalation and to achieve adequate depths of anaesthesia the concentration in inspired air must be increased to such a level that anoxia and cyanosis ensues. Used alone, for example in dentistry, it can therefore only be administered for very brief periods. However, at subanaesthetic concentrations it has good analgesic properties and so is often used to supplement anaesthesia induced by other agents. At the higher atmospheric pressures which can be achieved in hyperbaric chambers, nitrous oxide may be used to achieve anaesthesia at concentrations mixed with oxygen which do not cause anoxia. Cyclopropane is another gaseous anaesthetic administered by inhalation. It has a higher anaesthetic potency than nitrous oxide and so does not cause anoxia. However, it is an explosive gas and its use has declined. Both nitrous oxide and cyclopropane have a low solubility in blood such that a rapid equilibrium is achieved between the concentration in the inspired gas mixture and the blood. Consequently the concentration in the alveolus is not greatly reduced during each inspiration and the rate of equilibration of blood and lung concentrations is chiefly determined by the cardiac output, rather than the rate and depth of respiration. The induction of anaesthesia is therefore rapid because the rate of equilibration between blood and brain is not a limiting factor. Volatile anaesthetics Ether, chloroform, halothane and related drugs may be vaporised at room temperature by passing a stream of gas over the surface of the liquid. The degree of vaporisation and the concentration in the administered gas mixture is determined mainly by the temperature, and adequate means of controlling temperature or compensating for changes are necessary if reproducible concentrations in the gas mixture are to be obtained. The volatile anaesthetics are all more soluble in blood than nitrous oxide or cyclopropane. The solubility is such that during

104

General anaesthetics

each inspiration the concentration of gas in the alveoli is greatly diminished. The rate of attainment of equilibrium between blood concentration, and therefore tension in blood, and the concentration in the inspired gas mixture is therefore slow and determined by the rate and depth of respiration, rather than by the cardiac output. It has been estimated that 90% equilibration of the blood with diethylether administered at a constant concentration requires about 20 h of administration. The equivalent time for nitrous oxide would be about 1 h. If highly soluble volatile anaesthetics were to be administered at concentrations just sufficient for anaesthesia at equilibrium then the induction of anaesthesia would be an extremely slow process. This disadvantage can be overcome by one of two methods. First, the concentration of inspired gas during the first stages of induction is greatly in excess of that which would cause anaesthesia at equilibrium. An anaesthetic concentration in blood is therefore rapidly attained. The concentration in the inspired air may then be gradually reduced towards the equilibrium concentration as anaesthesia progresses. The rate of onset of anaesthesia may also be accelerated by increasing the rate of respiration, e.g. by the administration of increased concentrations of carbon dioxide in the gas mixture. The plateau principle states that at equilibrium the rate of administration of a substance is equal to the rate of elimination. It therefore follows that anaesthetics which only slowly attain equilibrium concentrations in blood will also be eliminated at a slow rate. This is clearly a major disadvantage with anaesthetics from which rapid and complete recovery is usually desirable to avoid post-operative complications. In addition such drugs are highly lipid soluble. During prolonged periods of administration they will therefore be accumulated by body fat. The rate of attainment of equilibrium between blood and fat is a slow process due to the fact that blood flow in the body fat is a rate-limiting process. There will therefore be a rather prolonged 'hangover' effect if the duration of administration is sufficiently prolonged to cause significant increases in the concentration of anaesthetic in

Types ofgeneral anaesthetic

105

the body fat, from which it will be even more slowly eliminated, with consequent maintained low levels of anaesthetic in the blood. Ether is an inflammable, and therefore dangerous, anaesthetic. It is very irritant to the respiratory mucosa and causes copious secretions and may cause laryngospasm. It causes less respiratory depression than chloroform and may cause intense excitement during the early stages of induction. Its major advantage is that it is inexpensive. Chloroform is not inflammable but is particularly liable to cause liver damage. Halothane is a relatively modern drug which was synthesised in a deliberate effort to achieve a non-inflammable and safe volatile anaesthetic. It is less prone to cause liver damage than chloroform in normal use but repeated exposure, as may be experienced by the anaesthetist, may have toxic effects on the liver. It has analgesic properties and produces a smooth induction of anaesthesia with minimal excitement. However, it causes a marked circulatory depression at anaesthetic concentrations. It also causes some respiratory depression. Circulatory depression is less marked with the related anaesthetic, methoxyfluorane, but respiratory depression is more evident with this agent. Many anaesthetics, including cyclopropane, chloroform, halothane and methoxyfluorane appear to sensitise the myocardium to circulating catecholamine and may cause ventricular fibrillation. Soluble (intravenous) anaesthetics Von Bayer synthesised barbituric acid by condensing together urea and malonic acid in 1864. Barbituric acid is not itself an anaesthetic but is the parent of a long line of derivatives which are. The first of these was sodium barbitone, synthesised by Fischer & Von Mering in 1903, but its action is too prolonged for modern use. Phenobarbitone was synthesised in 1912. It is also quite long acting and used more as a sedative and antipileptic in subanaesthetic doses, rather than as an anaesthetic. The first really useful, short acting barbiturate was

106

General anaesthetics

thiopentone, introduced during the Second World War. The introduction of this substance led to relatively safe and short-lasting anaesthesia. The structures of some of the more important barbiturates are shown in Fig. 7.2. In clinical anaesthesia only those with the shortest duration of action, e.g. thiopentone, methohexitone and thiohexitone are widely used. The duration of the anaesthetic effect is governed by two factors which determine the plasma level. Redistribution of the drug is probably the most important with thiopentone, whereas metabolism becomes rate limiting for methohexitone and thiohexitone, although redistribution still occurs. Redistribution in body tissues which are well perfused by blood, e.g. muscle, are more important than redistribution into body fat, which is poorly perfused by blood, even though the drugs have a high fat : plasma partition coefficient. The high fat solubility of the barbiturates leads to a good absorption from the gastrointestinal tract. However, they are not administered by this route for anaesthesia because the ratio between toxic and therapeutic plasma concentrations is too small and absorption too unreliable for this route of administration to be safe. Nevertheless, when used as sedatives or hypnotics in subanaesthetic concentrations, the oral route is relatively safe and certainly more convenient than intravenous injection. Eugenols such as propanidid are esters and rapidly hydrolysed by plasma and liver esterases. They are therefore extremely short acting and have been used for out-patients. Unlike the barbiturates, they rarely cause laryngospasm or bronchospasm. The soluble steroidal anaesthetics such as alphaxalone are also short acting anaesthetics which cause good muscle relaxation. Cyclohexylamine ('Ketamine') produces a rather unusual type of anaesthesia, sometimes called 'dissociative' anaesthesia. The drug was developed from a series of psychotomimetic substances and the use of the first of these with anaesthetic properties, phencyclidine, had to be abandoned because of the severity and prolonged nature of its psychological side-effects which included hallucinations and mania lasting for days. These effects are less marked and prolonged with ketamine but during induction the

Types ofgeneral anaesthetic

107

patient may experience odd sensations or even frank hallucinations. The muscles are not relaxed during anaesthesia, and the limbs may move involuntarily, but reflexes to painful stimuli are attentuated. The major advantage of ketamine is that it causes minimal effects on the cardiovascular system and may therefore be particularly indicated in those patients with cardiac disease or in the elderly. In general, the short-acting intravenous anaesthetics are used for induction of anaesthesia because they cause a smooth and rapid induction, with minimal stress or excitement. However, the anaesthesia is usually maintained during more prolonged surgery by the administration of a volatile or gaseous anaesthetic, or combination of these, often with additional analgesic drugs and other agents. Mechanisms of anaesthesia Physico-chemical theories The great diversity in the chemical structures of substances causing anaesthesia, ranging from the inert gases to such complex molecules as steroids, renders it improbable that anaesthesia is explicable in terms of interaction with a single, specific receptor molecule. In this respect the anaesthetics comprise a class of compounds differing from all of the others discussed in this book. A certain natural tendency to search for a single explanation for the phenomenon of anaesthesia has led to several physico-chemical theories of anaesthetic action based upon some non-specific action of the substances upon cell membranes. While such theories have the merit of simplicity and rather precise supporting data, they often fail to account for certain discrepancies in the data and pay no heed to the qualitatively different aspects of anaesthesia with different agents. Purely physico-chemical effects upon cell membranes may indeed be an adequate explanation for some of the chemically simpler anaesthetics, but the apparent correlations between anaesthetic potency and physico-chemical parameters for more complex molecules may simply reflect the physical environment in which

108

General anaesthetics Fig. 7.2. Structures of barbiturates. Anaesthetics are listed in order of decreasing duration of effect. NH2

/ )= C

Urea

\ CH2

+ NH,

O H II N — C

HOOC

2 5CH 2 6/ \l N — C II H

HOOC

o

Malonic acid

Barbituric acid

Anaesthetics derived from barbituric acid by substitution in positions 2, 3, 5 :

H ? N

C CH, \ / ' " Na—O — C C \ / \ N— C C,H5 H M O /

Sodium barbitone

H ?} N— C

/ Phenobarbitone

C2H, /

\

O=C

C H

II O

H ? N— C /

Amylobarbitone

O=C \

C2H, \ /

C / \ N H

C II O

CH 2 CH 2 CH (CH 3 ) 7

the anaesthetic acts, rather than the precise mechanism of action. In this case, it may be more useful to consider the anatomical locus of anaesthetic action in the central nervous system and be satisfied with a less fundamental explanation of the mechanism which nevertheless may be more intelligible in terms of its effect upon the physiological function of the brain.

109

Median isms of anaesthesia

H N

Pentobarbitone

? C

/ O=C

\

C2HS /

\ C NH

-C II O

\

CHCH 2 CH,CH 3 I CH 3

CH 3 O I II

N Hexobarbitone

o=c

\

s=c

\

CH 3

/

c N— C

H

Thiopentone

C \

£ O II

H N-

-c

N H

C II O

C\H5

\ \

CH3 O I II N— C Methohexitone

CH 2 CH=CH, \

O=C \

/

N—C H n O

C \ C H O sCCH2CH3 i CH3

H ?

N —C Thiohexitone

\

CHCH,CH2CH3 I CH3

CH2CH = CH2

\

N— C H U O

/ \

CHC = CCH2CH3

Physico-chemical theories of anaesthetic action will therefore only be considered rather briefly here and the demanding reader is referred to other reviews of the subject. Overton & Meyer were among the first to notice a correlation between lipid solubility and anaesthetic potency. However, lipid solubility is an important factor in gaining access to the brain or

110

General anaesthetics

even to cell membranes. Ferguson correlated anaesthesia with thermodynamic activity but such molecular properties are related to other intrinsic physical properties of substances such as solubility, boiling point and molar refractivity (colligative properties), all of which show a high degree of correlation with anaesthesia. Mullins proposed a critical volume theory from which has derived the idea that when anaesthetics dissolve in cell membranes they cause those membranes to expand. When the expansion reaches a critical point then normal membrane properties are so disrupted that anaesthesia ensues. That some anaesthetics do indeed alter the physico-chemical properties of membranes is shown by the fact that not only do they cause membrane expansion (at least in red blood cells) but they cause a disorganisation of membrane structure, as elegantly shown by the increased membrane fluidity in studies with nuclear magnetic resonance (NMR) techniques. Other theories, such as those of Pauling, ascribe anaesthesia to interaction of anaesthetics with the aqueous phase of cell membranes, forming clathrates which alter membrane function. These clathrates may be stabilised by protein side chains in the physiological milieu. (i)

(ii)

Difficulties with physico-chemical theories Detailed studies of the effect of changes in pressure or temperature upon anaesthesia show that the effects are not always quite as predicted from the known effects of these factors on the clathrate formation. It is also difficult to account satisfactorily, in physico-chemical terms, for the fact that in a homologous series of anaesthetic compounds, the anaesthetic potency may increase in parallel with some physico-chemical characteristic up to a point beyond which the anaesthetic potency may decrease, whereas properties such as lipid solubility may continue to increase. It is difficult to understand why some substances should be anaesthetics, whereas very closely related substances, with rather similar physico-chemical characteristics

Mechanisms of anaesthesia

111

Fig. 7.3. Closely related anaesthetics and convulsants.

O=C

Anaesthetic

Convulsant

CF,CHFBr

CF,CH : Br

/ \

O H || N— C

H

N —C H II O

\ / C / \

/

C\H S \ / '

\

c / \

o=c CHCH.CH.CH, 1 CH.,

N—C CHCHXHCH, H || I I O CH, CH, 5-ethyl-5-(l, 3-dimethylbutyl) barbiturate

Pentobarbitone

O

o

H || N—C

\ / "

\

/

N— C H II 0

C

H II N— C

CH, \ / O=C C \ / \ N—C CH=CHCH(CHo H II

C,H S

/

o= C

°n

N—C

CM,

/

\

CH,CH,CH(CH,)^

O

5-ethyl-5-(8, 8-dimethylallyl) barbiturate

Amy lobarbi tone

Fig. 7.4. Steroid anaesthetics. Alphaxalone (anaesthetic)

A-16-Alphaxalone (antagonist)

HO

112

General anaesthetics

should not only be lacking in anaesthetic action but in fact have quite the opposite effect on the central nervous system in causing convulsions. Examples of such opposing effects are shown in Fig. 7.3. (iii) Alphaxalone (Fig. 7.4) is a potent steroid anaesthetic, whereas the closely related A-16-alphaxalone not only lacks anaesthetic action but antagonises the effect of alphaxalone in isolated guinea pig cortex. Alphaxalone apparently does not change membrane fluidity, as measured by NMR but it is not known whether A-16alphaxalone antagonises the action of other anaesthetics. It is of course possible that the steroidal anaesthetics do indeed interact with specific membrane receptors, but this has yet to be demonstrated. (iv) Anaesthetics may depress other central nervous system functions in addition to consciousness. Thus they may have analgesic actions, depress respiration or circulatory control. If they had only one action on the central nervous system then one would expect all agents to have similar relative potencies in all of these effects. Yet one finds that halothane has good analgesic properties but the barbiturates may have anti-analgesic actions; halothane depresses the blood pressure to a greater extent than does methoxyfluorane but the relative potencies are reversed for depression of respiration. There are no adequate explanations for these discrepancies but they should make one wary of accepting in toto any purely physico-chemical theory of anaesthetic action. Localisation of the effects of anaesthetics on neurones Pre- and postsynaptic effects In sufficient concentrations, anaesthetics depress excitable cells in all parts of the nervous system, but effects are noted in some areas or on some parts of neurones at lower concentrations or earlier than effects at other sites. While a differential access of the anaesthetic to different parts of the nervous system may provide a partial explanation for such different sensitivities, it is very

Localisation of the effects ofanaesthetics on neurones

113

unlikely that this factor predominates. It is more probable that some physiological functions or some neurones are more affected by a quantitatively similar effect of the anaesthetic than are others. For example, Weakly has shown pentobarbitone reduces the presynaptic release of transmitter responsible for the monosynaptic excitatory postsynaptic potential in motoneurones, even at subanaesthetic concentrations. However, he found, in agreement with others, that higher concentrations also caused a decrease in postsynaptic excitability. Richards has examined the effects of a range of anaesthetics on pre- and postsynaptic phenomena in isolated cerebral cortex. Although all of the agents tested reduced the amplitude of evoked postsynaptic potentials they did not appear to do this by decreasing the electrical excitability of the postsynaptic membrane and there were differences between the agents, indicating that some anaesthetics (e.g. halothane) were acting to reduce the release of excitatory transmitter from the nerve terminals, whereas other (e.g. trichloroethylene) reduced the action of the transmitter on the postsynaptic membrane, as shown by the ability of trichlorethylene but not halothane to reduce the excitatory effect of glutamic acid. Thus it may be concluded that general anaesthetics depress central neurones by effects on both pre- and postsynaptic mechanisms but the former are likely to be more susceptible with many agents. In very high concentrations the general anaesthetics may also depress nerve conduction, but this is not found at therapeutic concentrations. Differential effects on excitatory neurotransmitters A substance with a non-specific depressant action on cell membranes might be presumed to reduce the excitation of a particular neurone by any excitatory agent which is administered to it. It is therefore rather surprising to find that general anaesthetics selectively depress the excitation of cerebral cortical neurones by iontophoretically administered acetylcholine without reducing, but even enhancing, excitation by glutamic acid. The selective depression of excitation by acetylcholine is

114

General anaesthetics

particularly pertinent to the question of anaesthesia because it is highly probable that cholinergic projections to the cerebral cortex are important in maintaining consciousness. The effects of pentobarbitone on the membrane potential and resistance are similar to those of intracellularly injected calcium ions, which hyperpolarise the membrane and decrease resistance probably by increasing the membrane permeability to potassium ions. The probable mechanism of excitation of cortical neurones by acetylcholine is a reduction in the membrane permeability to potassium ions. It is therefore likely that both anaesthetics and acetylcholine act on the same membrane permeability, but they have opposite effects. A plausible explanation of the action of general anaesthetics which selectively reduce excitation by acetylcholine is that they reduce the sequestration of free internal calcium ions by brain mitochondria, so increasing the internal free calcium concentration and increasing permeability to potassium. Observations on the effects of anaesthetics on Ca2+ uptake by brain mitochondria are unfortunately few, but it is known that halothane at anaesthetic concentrations markedly reduces Ca2+ uptake by rat brain mitochondria. The mechanism by which the sequestration of Ca2+ by mitochondria is suppressed by anaesthetics is unknown; it could be by a non-specific action on the mitochondrial membrane but it could also be due to a more specific effect on metabolic processes. It is perhaps significant that the release of acetylcholine from the cerebral cortex, as measured by assay into a Perspex cup containing eserinised Ringer solution placed on the exposed surface of the cortex, is reduced by general anaesthetics and is higher in alert than in quiescent behavioural states. There may be several explanations for the reduction in release and the relative importance of each of these is not established. There is likely to be a direct depressant action of the anaesthetic on the release of the transmitter from presynaptic terminals but releases would also be suppressed by any process which decreases the activity of the cholinergic neurones. The cholinergic projection from the midbrain reticular forma-

Conclusions

115

tion is probably reduced in activity during anaesthesia, both by depressant effects on the projecting neurones themselves and by the reduction of afferent input to the cells during anaesthesia. Effects on presynaptic inhibition Presynaptic inhibition serves to reduce the release of transmitter from the central terminals of primary afferent nerve fibres and thereby reduces the flow of afferent information from peripheral receptors to the brain. Many general anaesthetics increase presynaptic inhibition, although the mechanism is obscure. They also increase a number of prolonged postsynaptic inhibitory processes in the brain and it may be significant that GABA is thought to be a common mediator of these effects. Modern studies with voltage clamp techniques have demonstrated that barbiturates have several actions on neuronal membranes. It has been confirmed that the action of GABA is potentiated by increasing the duration of channel opening. This effect on the GABA-A receptor correlates with many of the actions of the barbiturates. However, there are also direct gabamimetic actions at higher concentrations and there are also effects on Ca2+ ion conductances. The barbiturates enhance the binding of GABA to its binding sites, perhaps by decreasing the rate of dissociation from the binding site, by a chloride-dependent mechanism. A potentiation of inhibitory processes in the brain may therefore be supplementary to the other actions of general anaesthetics in causing a loss of consciousness. Selective effects upon different areas of the brain and on spinal reflexes It has long been known that general anaesthesia has a greater depressant effect upon polysynaptic spinal reflexes than upon monosynaptic reflexes. This may merely reflect the greater susceptibility of complex pathways to drug-induced depression. The same explanation may apply to the often-repeated demonstration that general anaesthetics tend to decrease activity in the reticular formation of the brain before effects can be detected elsewhere, although the selective effects upon cholinergic mechanisms may also be involved.

116

General anaesthetics

It was noted earlier that some inhibitory processes in the brain may be prolonged by general anaesthetics. However, the inhibition of spinal reflexes by stimulation of inhibitory regions of the reticular formation of the brain stem is reduced by concentrations of anaesthetics lower than those which block facilitation by stimulation of excitatory regions. Conclusions Even if general anaesthetics do act at the molecular level by a common non-specific action, and this seems by no means certain, then it is apparent that different neuronal membranes at different sites, including intracellular membranes, may be more susceptible than others. This may result from a particular membrane process being more important for the function of some cells than for others. Particular attention is drawn to the actions of some anaesthetics upon GABA-mediated inhibitory processes. In addition, different regions of the brain or even different pathways within the same area show differential susceptibilities to a given anaesthetic for reasons which are not entirely clear. Anaesthesia, then, results from a depression of a number of functions of the brain and it is indeed fortunate that the selectivity of action results in a relatively selective depression of consciousness with minimal effects on some vital functions such as respiration and circulatory control. Such selectivity is clearly only marginal and the anaesthetics will depress all neuronal function in concentrations only slightly higher than those required for anaesthesia. Tolerance to anaesthetics Tolerance and physical dependence upon barbiturates may develop with continued use, although it may be evident even after the administration of a single dose: the plasma level of a barbiturate may be higher at the point of awakening from anaesthesia than it was on induction. The tolerance is not so severe as with opiates. This will clearly not be of great importance in the use of these agents as anaesthetics but will be more significant

Tolerance to anaesthetics

117

when they are used chronically as sedatives or hypnotics. Pharmacokinetic factors, including the induction of liver microsomal enzymes, e.g cytochrome-P450, which metabolises the barbiturate, are also bought into play in long-term tolerance. The tolerance depends upon the continuous presence of the agent and so is more evident with barbiturates which have a long half-life, such as phenobarbitone than with shorter-acting agents such a pentobarbitone. The tolerance seems to depend mainly upon neural adaptation to the continued presence of the drug, rather than to increased metabolism, because it has been clearly demonstrated that during the development of tolerance, sedative effects at the same blood level become progressively less. Tolerance may be produced by intravenous administration and so is not dependent upon changed absorption. There is a cross-tolerance between the barbiturates and other classes of non-specific depressants of the nervous system (e.g. alcohol). Cross-tolerance to the opiates is less, indicating that in the latter case quite different mechanisms are involved. Neural tolerance to the action of barbiturates has some significance in anaesthesia because it has been shown in rats that the brain level of barbital determined at the time of awakening is significantly increased after only five daily injections. In man, it has been shown that the plasma level of thiopentone at the time of awakening is about three times higher when sufficient thiopentone was given to produce anaesthesia lasting for about 5 h as compared with the level after a dose causing anaesthesia for only 1 h. The practical significance of such observations is that in order to obtain a prolonged anaesthesia with thiopentone, progressively higher plasma levels will need to be maintained which will increase the amount redistributing into fatty tissues and may lead to post-operative complications. The mechanism of neural tolerance is completely unknown, but it may reasonably be assumed to represent homeostatic adaptation of the nervous system to the continued presence of the anaesthetic.

8 Pain and analgesia

Pain is essentially a subjective sensation. It can therefore only be described in terms of an individual's expression of it and is subject to that individual's previous experience. For the experimenter this poses a number of problems. A subjective experience can be quantified in man but not in experimental animals, because the animal is not able to express its experience in terms that are intelligible to us. However, we can measure the various effects of and reactions to a stimulus which can itself be described as painful on the basis of the effect of that stimulus administered to man. If the observation made is itself an objective observation such as a reaction time or the frequency of firing of a neurone to the noxious stimulus there is no problem until attempts are made to extrapolate the objective observation to the subjective phenomenon of pain. Particularly difficult to evaluate are experiments on animal behaviour in which the behaviour is expressed in anthropomorphic terms. Furthermore, the objective tests, even in man, do not usually assess the quality (degree of unpleasantness), as distinct from the perception or intensity of the pain. In contrast, nociception is the perception of a painful stimulus. The ability of a neurone or an animal to detect the stimulus, or the threshold stimulus intensity to produce a response can be accurately measured. Provided that we bear in mind the difficulties in relating nociception to the quality of pain sensation, then experimental studies in man and animals can and have greatly advanced our understanding of the neuronal control mechanisms involved in the control of nociception by physiological neuronal systems and by drugs. The major advance in the understanding of the mode of action of the major, opiate analge-

Peripheral pain mechanisms

119

sics came with the discovery of opiate receptors in the 1970s and the tremendous expansion of our knowledge of these receptors and their ligands which is still increasing at the present time. Peripheral pain mechanisms Peripheral nerve fibres Pain sensations are evoked by intense stimuli in any sensory modality, e.g. pressure, heat, cold, light or sound. In certain pathological conditions such as trigeminal neuralgia, pain is evoked by normally innocuous stimuli. Clearly in this state the normal interpretation or control of the pain pathways in the central nervous system is lacking. Usually, pain is only produced when the intensity of the stimulus is sufficient to cause some minimal tissue damage: this can be expected to release endogenous agents which excite the nerve terminals. However, it is reasonably certain that the sensation of pain is only produced by activation of a specific set of peripheral receptors which respond only to stimuli in the painful range of intensities. Most of these are polymodal nociceptors. In the central nervous system there is some maintenance of selective pain pathways as information flows from periphery to the brain, although this is clearly not absolute. Pain may be evoked in superficial structures such as the skin or it may arise from deep structures such as joints or viscera. Deep pain is usually of a dull, aching variety and, in pain of visceral origin, may be referred to a point on the skin surface. Superficial pain may be relatively well localised. Pain can be either sharp and pricking or it can be a dull, aching or burning sensation. Such differences may be related to either the receptor types activated or the fibres which conduct the impulses into the central nervous system. Little is known about the structure of the peripheral nociceptors except that they are probably bare nerve terminals. The nerve fibres involved are either myelinated, A-8 fibres, conducting at speeds of 2.5 to 11 or 30 m s"1, depending on the nerve, or unmyelinated C-fibres conducting at speeds of less than 0.5 m s"1. The A-8 fibres may be in part responsible for the sharp, prick-

120

Pain and analgesia

ing pain sensation and the C-fibres for the dull, aching qualities of pain. In man, the quality of pain changes with peripheral nerve injury or block. Local anaesthetics first block the C-fibres, although at higher concentrations the A-fibres are also blocked. In the abnormality of carpal tunnel syndrome, in which there is entrapment of the median nerve with consequential compression and block of conduction in the A-fibres, there are paresthesias and a radiating, aching pain which is attributed to the inbalance between the C-fibre and A-fibre input. If the A-fibres are not conducting but the C-fibres are intact, as in various peripheral neuropathies, e.g. demyelinating diseases, then the second, burning pain becomes more intense. Not all C-fibres mediate the sensation of pain, e.g. there are thermoceptors which detect temperature changes at non-noxious levels. Of interest is the observation that many C-fibres contain the polypeptide substance P. This peptide can coexist with other peptides such as somatostatin or dynorphin in the primary afferents. Consequently there is much speculation about the role of these peptides as transmitters or modulators at the central terminals of the pain fibres. In experiments in man, it has been shown that the A-8 fibres have a relatively high threshold to excitation by noxious heat compared with C-fibres. However, after very mild injury, as in sunburn, the hyperalgesia is accompanied by a marked increase in the sensitivity of the A-8 fibres, compared with minimal effects on C-fibres. Furthermore, there are different C-fibres mediating the sensations of pain and itch. Activation of pain receptors and mediators Tissue injury releases histamine which increases capillary permeability and causes a wheal. It also releases bradykinin. Histamine also releases bradykinin in human skin. Bradykinin is a very powerful excitant of peripheral nociceptors and is relatively selective, having little effect on other sensory receptors: although histamine and a number of other agents excite nociceptors, they are relatively non-selective. Distant from the

121

Peripheral pain mechanisms

site of the original injury the tissue becomes reddened due to vasodilatation. This is the 'flare' of the classical Lewis's triple response. It is due to an axon-reflex in which the bradykinin activates a peripheral terminal of a C-fibre, setting up impulses which travel orthodromically into the CNS and antidromically down nearby branches, liberating substance P at the distant terminals. Substance P is a powerful vasodilator (Fig. 8.1). The stimulant action of bradykinin on the terminals is greatly potentiated by the presence of prostaglandins, which are also released in tissue injury.

Fig. 8.1. Peripheral components in pain perception. Spinal cord Dorsal root ganglion

Substance P - containing non-myelinated C-fibres (A-5 fibres are also nociceptive) Pain receptor

Axon reflexes Substance P s>v

-- Blood vessels

Noxious stimulus

- • Bradykinin/histamine release -

-Pain receptor activation

Sensitisation Release of PGs

Synthesis of PGs -

- Aspirin

122

Pain and analgesia

The action of aspirin Aspirin inhibits cyclo-oxygenase which is involved in the synthesis of prostaglandins. The production of prostaglandins is thereby inhibited. In conditions in which tissue damage, and the consequent liberation of prostaglandins is implicated, aspirin acts as a pain-killer by virtue of the fact that it reduces the release of prostaglandin, so that the action of bradykinin on nociceptors is not potentiated. The analgesic effect seems not to be due to the anti-inflammatory action since there is no correlation between the analgesic and anti-inflammatory effects (Table 8.1). The peripheral site of action of aspirin as a pain killer was established by Lim in crossed circulation experiments in dogs. Aspirin inhibits the synthesis of prostaglandins both centrally and peripherally. Paracetamol is also a good analgesic but only inhibits prostaglandin synthesis centrally. It must be supposed that there are also central components to the analgesic action of the prostaglandin synthesis inhibitors, although the peripheral action of aspirin and a central action of paracetamol might predominate. Aspirin is usually taken by mouth for some relatively mild types of pain, such as headache, neuralgia and pain arising from some inflammatory conditions. It is not usually effective in very severe pain such as the enduring pain of terminal cancer. Nevertheless, it has some value as an adjunct since it is relatively safe and does not lead to tolerance. The side-effects of aspirin include a prolongation of bleeding time, gastric ulceration and haemorrhage, headache, dizziness, confusion, hyperventilation, disturbance of acid-base balance, and various autonomic effects ranging from sweating and nausea to vomiting and diarrhoea. In more severe intoxication, hallucinations, convulsions and kidney damage may develop. Other aspirin-like agents (Fig. 8.2) include paracetamol, phenacetin, indomethacin and phenylbutazone. Paracetamol is widely used and is fairly well tolerated except for occasional skin rashes. However, high doses may cause methaemoglobinaemia and haemolytic anaemia or may lead to potentially fatal hepatic and renal necrosis. Indomethacin and phenylbutazone are poorly tolerated in a high number of patients and may cause serious

Central pain pathways

123

Table 8.1. Lack of correlation between analgesic and anti-inflammatory effects of aspirin-like analgesics

Aspirin Phenacetin Paracetamol Indomethacin Phenylbutazone

Analgesic action

Anti-inflammatory action

4-44-44-44-4-a 4-

4-40 0 + 44- +

0, no action; +, +4- or + + + refer to approximate order of potency. a The analgesic effect of indomethacin may be the indirect consequence of anti-inflammatory action.

blood dyscrasias such as thrombocytopenia, leucopenia, agranulocytosis and aplastic anaemia. The action of capsiacin Capsiacin is the khot-stuff from peppers. It causes the release of substance P from C-fibres and ultimately the fibres degenerate. A first injection of capsiacin gives pain and inflammation, due to the release of substance P and other mediators. A second injection of capsiacin has no effect since the substance P has already been depleted. Central pain pathways Processing in the spinal cord A-8 and C-fibres enter the spinal cord by the lateral division of the dorsal roots and run a short distance in Lissauer's tract before entering the dorsal horn. Most C-fibres terminate in the superficial laminae (lamina I of Rexed and the substantia gelatinosa). A-8 fibres penetrate more deeply. There is a great deal of processing of the noxious input before the resultant information is relayed to the higher levels of the nervous system by projection neurones belonging to various ascending pathways.

124

Pain and analgesia

Processing occurs both in the substantia gelatinosa and in deeper laminae of the dorsal horn. The substantia gelatinosa has attracted a great deal of attention because it contains the highest levels of most of the conventional and peptide neurotransmitters and modulators (see Chapter 5). This should not lead one to ignore the importance of processing at other sites. Descending control from the brainstem. In addition to purely

segmental processing of noxious input in the spinal cord there is a descending control of the spinal nociceptive neurones by pathways originating in the brainstem. In general, these pathways inhibit transmission of nociceptive information to the spinal Fig. 8.2. Structures of some aspirin-like analgesics. COOH ,OCOCH3

Aspirin

OH Paracetamol

OC 2 H 5 Phenacetin

Phenylbutazone

CH 3 CH 2 CH 2 CH 2

125

Central pain pathways

nociceptive neurones (Fig. 8.3). Areas in the brainstem which are thought to be particularly important in reducing the activation of nociceptive neurones by a noxious input include the periaqueductal grey, nucleus raphe magnus and lateral regions of the reticular formation. The periaqueductal grey probably relays through the raphe nuclei. It will be recalled that the nucleus raphe magnus is the major site of origin of 5hydroxytryptamine-containing fibres in the spinal cord. Some of these descending control systems travel down the cord in the dorso-lateral funiculus and are involved in tonically inhibiting the neurones which respond to noxious stimuli. We shall see that various hypotheses attribute the analgesic action of morphinelike drugs to effects in the spinal cord, in the brainstem and at even higher levels of the neuraxis. Fig. 8.3. Spinal and supraspinal mechanisms in nociception. Cortex

Thalamus Reticular formation

Periqueductal grey

Raphe nuclei

^ Descending NA-containing fibres - Descending 5-HT-containing fibres *— Other descending fibres Ascending tracts Spinal cord Sensory input Projection neurones

Substantia gelatinosa

126

Pain and analgesia

Ascending pathways. The tracts conveying noxious information from the spinal cord to the brain include, to a variable extent according to the species, spinothalmic (crossed antero-lateral), spino-cervical (dorso-lateral), spino-reticular and postsynaptic dorsal column pathways. Some of the spinothalmic fibres may give off collaterals to the reticular formation and so there may be some overlap in these pathways. The multiplicity of pathways explains why antero-lateral cordotomy does not always completely remove pain and that pain may return some time after surgery, presumably as the alternative pathways increase their functional participation. The ascending pathways may relay in the ventro-basal complex of the thalamus (specific thalamic nuclei) on their way to the cortex. This neo-spino-thalamic system is thought to be activated mainly by A-8 fibres and to be responsible for the somatotopic localisation of the pain. Alternatively, paleospinothalamic fibres relay in intralaminar thalamic nuclei such as nucleus centralis lateralis and nucleus centromedianus, giving rise to diffuse projections to cortex and hypothalamus. The paleospinothalamic system is strongly activated by C-fibre inputs and is thought to be involved in the perception of dull, aching or burning pain sensations. Morphine-like analgesics The medical use of opium as a pain-relieving drug dates back at least to the twelfth century and the use of soporific sponges. The absorption of morphine after ingestion is unreliable and the systemic administration was not possible until the active principle from opium was identified as morphine in the nineteenth century. Opium may contain as much as 10% by weight of morphine and 0.5% of codeine, together with other alkaloids which are not analgesic. Structure of morphine-like drugs Many substances have been synthesised in a largely vain attempt to overcome the problems of abuse and tolerance present in all of the commonly used drugs in this class. Morphine, codeine and heroin are phenanthrene derivatives (Fig. 8.4) but

127

Morphine-like analgesics

HO

O'

H3CO

OH

O

OH

Codeine

Morphine N — CH 3

C H J C H J C —C—CH

H

I

OCCH3 II

2 CHN

I CH3

Methadone N-CH2CH-C \

COCH2CH3

OH Pentazocine

Mependine

HO

O

OCH3

Buprenorphine

Antagonists N—CH 2 CH=CH 2

HO

O Nalorphine

OH

N—CH 2CH—CH 2

HO

O

O

Naloxone

Fig. 8.4. Structures of some morphine-like drugs and antagonists.

128

Pain and analgesia

this structure is not essential for analgesic activity as is shown by the structures of meperidine and methadone. Although pentazocine can cause addiction, it is said to be less addictive than most opiates in equivalent analgesic doses. Many of the opiates are conformationally restricted and there are some similarities in the shape of the molecules in three dimensions to the D(-) isomer of morphine, which is the active isomer. Substitution of an allyl group on the nitrogen atom produces compounds which are 'narcotic antagonists1, i.e. they antagonise under suitable conditions the analgesic, euphoric and respiratory depressant effects of opiates and may precipitate withdrawal symptoms in addicts. Naloxone is an example of a pure antagonist. Some drugs are partial agonists, e.g. nalorphine, which when given alone acts as an agonist like morphine, but will counteract the action of morphine when given concurrently or subsequently. Pentazocine, although having the antagonist structure, is a useful analgesic, lacking the side-effects associated with the partial agonist nalorphine. The latest drug said to be less addictive and producing less tolerance and withdrawal is buprenorphine, a substituted morphine-like compound, with a higher potency than morphine. Actions of morphine-like drugs The analgesic action of morphine-like drugs is their most important medical use, followed closely by the euphoric, sedative and anxiety-relieving effect when used pre-operatively or in the context of a hospice. The abuse potential and tolerance are the chief drawbacks and the major toxic hazard is respiratory depression. Morphine-like drugs are antitussive. However, the cough-suppressant action is not correlated with the other actions of morphine and it is advisable to use other agents, such as codeine or dextrorphan, with a lower abuse potential. A spasmogenic action on the bowel is the basis of the early use of morphine in treating dysentery and is still the basis of certain remedies for the diarrhoea which often manifests itself to the unwary traveller in tropic climes. Morphine may cause a spasm

The opiate receptor

129

of the sphincter of Oddi, a rise in biliary pressure and severe epigastric pain. Morphine stimulates the synthesis of prostaglandins and is emetic. Both effects are abolished by chlorpromazine, a dopamine receptor antagonist, and apomorphine, a dopamine receptor agonist, is itself emetic. The probable mechanism of emesis with morphine is that activation of opiate receptors indirectly affects both dopaminergic and prostaglandin systems and leads to emesis by a stimulant action on the chemoreceptive trigger zone in the medulla oblongata. Opiates generally have minimal cardiovascular effects. A pinpoint pupil diagonostic for the morphine addict. The miosis shows less tolerance than the other effect of morphine. There are some species differences in the actions of morphine. For example, the pupils are dilated in cats and monkeys. Some animals are excited by morphine (cats, pigs, cows, sheep, goats, lions, tigers, bears and horses) whereas in man, monkey and the rat there is sedation. Withdrawal symptoms are characteristic of all members of this class of analgesics, although they may vary in extent. The variation is, at least in part, due to kinetics. Thus, methadone is often used to replace morphine or herion in addicts but is itself addictive. However, methadone is eliminated more slowly, and the withdrawal symptoms are thereby less severe. Antagonists precipitate withdrawal symptons in addicts. The opiate receptor Morphine-like agents show high affinity, saturable, stereospecific binding to sites in brain homogenates. The binding correlates well with the analgesic potency, although there are some exceptions. Table 8.2 shows some relative potencies as analgesics, for binding affinity and for lipid solubilities for morphine, etorphine and codeine. There is clearly no correlation between binding and potency. However, etorphine is far more lipid soluble than morphine and so enters the brain more readily. Codeine on the other hand is more potent as an analgesic than

130

Pain and analgesia

Table 8.2. Relative potency for analgesia, binding affinity to a brain homogenate and lipid solubility for morphine, etorphine and codeine

Morphine Etorphine Codeine

Analgesia

Binding

Lipid solubility

1 1000 0.2

1 20 0.001

300

would be predicted from its binding or lipid solubility. Codeine is demethylated to morphine in the liver and the product is at least in part responsible for the analgesia. Clearly, problems of access to the CNS and metabolic changes to the molecule must be considered when attempting to correlate a parameter such as ligand binding with activity in vivo. Localisation of the receptor Ligand binding, subcellular fractionation and histochemical techniques have been employed to study the distribution of the opiate receptors. The synaptosomal fraction, containing pinched-off nerve endings, of brain homogenates contains most of the opiate binding site. There is a wide variation in the amount of binding in different brain regions, with the highest concentrations in sensory, limbic and neuroendocrine systems. There are high concentrations in the amygdala and the periaqueductal grey. The local microinjection of morphine produces analgesia at the latter but not at the former site. However, microinjection in the amygdala causes a retrograde amnesia which is blocked by naloxone. There is also a fairly high concentration in the basal ganglia, which are involved in the control of movement, but not in the cerebellum. It is of interest that morphine produces stereotypy, in which behaviour, including movements, is repeated endlessly. There is a very high concentration of receptors in the intralaminar nuclei of the thalamus. Receptors in neuroendocrine regions are found in the posterior pituitary and the hypothalamus, which may be related to the actions of opiates on pituitary and hypothalamic function.

The opiate receptor

131

In the spinal cord, there is a very high density of receptors in the marginal cell layer and the dorsal layer of the substantia gelatinosa, regions which receive a high density of innervation from peripheral C-fibres. Some of these receptors may be on the terminals of the primary afferent fibres since the binding decreases after section of the dorsal roots. There is an excellent correlation between the binding of opiates to guinea pig intestine and to rat brain homogenates. There is also a correlation with the ability to inhibit the contractions of the intestine and all of these effects show an excellent correlation with analgesic potency. It is now well established that there are at least three clearly differentiated opiate receptors, jut, 8 and k: these are described in more detail in the following sections. Endogenous ligandsfor opiate receptors The first endogenous ligands for the opiate receptor to be isolated were the pentapeptide enkephalins, isolated by Kosterlitz and his co-workers in the mid 1970s. This was a significant conceptual advance in which for the first time an endogenous ligand was sought and found for a receptor which was already known to exist, rather than the reverse. This conceptual innovation has now been repeated for other classes of compounds including the benzodiazepines and some anticonvulsants. The endogenous ligands found in those early studies were methionine enkephalin (met-enkephalin) and leucine enkephalin (leu-enkephalin). Beta-endorphin with 31 amino acids contains the met-enkephalin sequence and is found in the pituitary. Still more recently, a new series of endogenous peptides, the dynorphins, have attracted a great deal of interest. All of these polypeptides are derived from one of three large precursors, pro-enkephalin-A, pro-opiomelanocortin or prodynorphin. Pro-enkephalin-A contains six copies of metenkephalin and one of leu-enkephalin. Pro-opiomelanocortin contains just one copy of metenkephalin and pro-dynorphin contains three copies of leu-enkephalin and the dynorphin sequences, which are absent from the first two precursors. The

132

Pain and analgesia

precursors also contain some larger amino acid sequences, e.g. gamma-melanocyte stimulating hormone, ACTH, (3-lipotropin. The peptides frequently coexist with other peptides or small molecule neurotransmitters: for example, 5-hydroxytryptamine may coexist with either enkephalins or dynorphins in different brain regions and substance P may coexist with dynorphin in some primary afferent C-fibres. The highest levels of enkephalins occur in laminae I and II of the dorsal horn of the spinal cord and in the periventricular and periaqueductal grey matter of the brainstem, dorso-medial parts of the thalamus, hypothalamus, globus pallidus and central nucleus of the amygdala. Immuno-reactive morphine was shown to be present in the brain in small quantities for the first time in 1985, but it was not known whether this was of endogenous or exogenous origin: its concentration is less than that of the enkephalins. It was demonstrated in 1987 that the mammalian liver is able to synthesise morphine from the same precursors that are used by Papaver somniferum. Thus the wheel may be turning full circle in the sense that it may yet prove to be the case that the oldest known opiate, morphine, which is more stable than the opioid peptides and has a greater affinity for the jut-receptor, is a major endogenous ligand for the opiate receptor. Analgesia and opioid peptides Leu-and met-enkephalin are analgesic when injected directly into the cerebral ventricles but not when given systemically. Beta-endorphin is also analgesic and is more potent than the enkephalins. The resistance of enkephalins to enkephalinases is increased by substitution of an amide group, e.g. met-enkephalin amide, or by replacing one glycine residue by D-alanine: such compounds have a higher analgesic potency but are still not effective on systemic administration. A carbinol derivative (FK33-824) of met-enkephalin has been synthesised and has about ten times the affinity of metenkephalin and five times the affinity of morphine for binding to the receptor. It is more effective than morphine as an analgesic

The opiate receptor

133

when administered into the cerebral ventricles and or when administered intravenously or subcutaneously. Such compounds are of interest in the search for new classes of chemical agents which are analgesic without the abuse potential of existing agents. They clearly demonstrate that large molecular weight peptides are able to enter the CNS to produce useful effects. Unfortunately, although analgesic in man, FK33-824 produced a number of undesirable side-effects, including heaviness of the muscles and feelings of oppression of the throat and chest, leading to anxiety. Like morphine, it increased secretion of prolactin and produced EEG changes but some of the other effects of morphine were lacking. It is of considerable interest that none of the side-effects were antagonised by naloxone, indicating a completely different mechanism of action. It remains to be established that new compounds related to the endogenous ligands can produce analgesia without tolerance or addiction. The signs to date are not hopeful since repeated intracerebral administration of enkephalins clearly produces tolerance, as with morphine. Multiple receptors for opioid peptides There is now good evidence for the existence of at least three, and possibly more, receptors for opioid peptides. Kosterlitz

Fig. 8.5. Structures of opioid peptides. Endogenous Metenkephalin Leu-enkephalin

Tyr_)_Gly Tyr I Gly

Gly ! Phe GlyTphe

Non-specific aminopeptidase /3-endorphin dynorphin Synthetic (more stable) FK33-824 Protein precursors Pro-enkephalin-A Pro-opiomelanocortin Pro-dynorphin

Met Leu

Carboxydipeptidase =enkephalinase

31 residues 17 residues MeTyr Gly Gly Phe Met CONH2 Tyr D Ala Gly Phe Met CONH2 Tyr_D Ala Gly MePhe Met (O)ol 263 residues 265 residues 256 residues

134

Pain and analgesia

and his co-workers have differentiated two receptors, [x and 8, on the basis of parallel biological assays and binding experiments. Table 8.3 shows the relative potencies of three peptides and morphine in inhibiting contractions of the guinea pig ileum and mouse vas deferens. If only one receptor was involved then the relative potencies should be the same in both assay systems. There is clearly a wide divergence. Similarly, the relative ability of the same peptides to inhibit the binding of labelled naloxone or leu-enkephalin differed. It will be noted that the relationship between the two biological assays and the two binding experiments is similar (compare columns a/b with c/d). This indicates that two receptors are involved. The jut receptor is the one present to a major extent on the guinea pig ileum and correlates with the inhibition of the binding of naloxone, whereas the 8 receptor is the one present in the mouse vas deferens and correlates with the displacement of the binding of leu-enkephalin. In other experiments they showed that the effects of the peptides on the guinea pig ileum are more easily antagonised by naloxone than are the effects on the mouse vas deferens by a factor of about ten: i.e. LJA receptors are more easily blocked by naloxone than 8 receptors. Morphine has a much greater affinity for LJA than for 8 receptors and leu-enkephalin has a greater affinity for jui than for 8 receptors. Furthermore, the analgesic action of the peptides correlates far better with their effects on the guinea pig ileum than it does with the effect on the mouse vas deferens. It is therefore apparent that, with this series of peptides at least, analgesia is due to an interaction with jut rather than with 8 receptors. The kappa receptors are revealed by an examination of the binding characteristics of ethylketazocine, which is itself an analgesic compound which is chemically unrelated to morphine or the opioid peptides. The binding of this substance is only partially displaced by ligands for the LJX and 8 receptors at low concentrations. This is due to about 40% cross-reactivity for ethylketazocine for the jui and 8 sites, the remainder being due to interaction with the K-site. Of particular interest is that the dynorphins interact fairly selectively with the K-site. To summarise, there are three well-defined binding sites for

135

The opiate receptor

Table 8.3. Evidence for two opiate receptors from biological assays on guinea pig ileum and mouse vas deferens and from ligand displacement studies with naloxone or leu-enkephalin as the displaced ligand. Figures in columns are the relative potencies (metenkephalin = 1) or ratios of potencies (a/b or c/d) Binding assay (Inhibition of binding)

Biological assay

Ligand G.Pig Ileum

Mouse Vas deferens

a

b

Ratio a/b

Nalox

Leu-enk

c

d

Ratio c/d

Met-enkephalin N-methylmetenkeph -amide D-ala-metenkeph Morphine

1

1

1

1

1

1

3.7 6 1.4

0.33 5.6 0.024

1.2 0.54 0.39

RECEPTOR

M

8

11 1.1 58 u/8

0.08 1.7 0.005 8

15 0.32 78 u/8

Data from Kosterlitz & Paterson (1980). Proc. R. Soc. 13, 210, 113 enkeph, enkephalin; nalox, naloxone; leu-enk, leu-enkephalin; G. Pig, guinea pig

opioid peptides and non-peptide analgesics, ji, 8 and K. Morphine and FK33-824 have a relatively selective affinity for the |u site, whereas the ketazocines and dynorphins are more selective for the K-sites. The naturally occurring enkephalins have a higher affinity for the 5 sites than for jm- or K-sites. The potent analgesic etorphine has only a five-fold differential between its affinities for the three binding sites: the affinity for the K-site is largest and the affinity for the jui site is lowest. Analgesia may be due to interaction with either ja- or K-sites. Existing non-peptide agonists at the K-site are relatively non-selective and cross react to a large degree with the other sites. Newer ligands derived from the dynorphins are more promising as selective K-agonists and, with some optimism, might give rise to a new series of analgesic compounds with lower abuse potential than existing agents.

136

Pain and analgesia

An alternative view that the opiate receptor may be a single entity with distinct |d-, 8- and K-binding sites, all allosterically interacting with each other has also been proposed on the basis of binding studies. However, if this model does represent the true situation then the preponderance of the different sites must vary in different tissues, perhaps dependent on environmental factors including mono- and divalent ion concentrations and GTP regulatory proteins. Involvement of opioid peptides in pain There is evidence that under some conditions an endogenous opioid system is brought into play in when nociceptors are activated. (i) There is an increase in the release of enkephalins from the perfused spinal cord when unmyelinated C-fibres are stimulated. (ii) The analgesia produced by a variety of acupuncture techniques is attenuated by the administration of the opiate antagonist, naloxone and is accompanied by an elevated level of endorphins in the CSF. This indicates that acupuncture activates the endogenous opiate system, (iii) In experimental animals continuous electrical stimulation of the skin over a period of about 20 min, analogous to electro-acupuncture, decreases the excitation of nociceptive dorsal horn neurones by a noxious stimulus. The attenuation is reversed by the administration of naloxone, again indicating that the electrical stimulation over a long period has activated the endogenous enkephalinergic system. (iv) There is evidence that naloxone can increase the perceived unpleasantness of pain following dental operations in some patients but not in others. Interestingly, those patients in which naloxone had an effect also responded to placebo treatment, indicating that part of the placebo response could have been the activation of the endogenous opiate system, (v) Naloxone increases some spinal reflexes, possibly by

Sites of opiate action

(vi)

(vii)

137

reducing background inhibition by enkephalinergic neurones. Stress has also been shown to increase the levels of endogenous opioid peptides in the brain in rats and to cause a stress-induced analgesia. Furthermore, the increased levels of peptides correlates with an increase in the threshold level of a stimulus used to asses the level of analgesia. In some chronic pain conditions the levels of endorphins in the cerebro-spinal fluid are decreased, indicating that a decreased activity of the endogenous peptide system may contribute to the chronic pain.

Sites of opiate action Studies with the iontophoretic administration of morphine and opioid peptides have shown that analgesic substances may either excite or depress the activity of neurones at various levels of the neuraxis, with depression being more frequently observed than excitation. Other studies have been interpreted as showing activation of descending systems which inhibit the activation of projection neurones by the noxious stimulus. However, the evidence for the latter view as a mechanism for the analgesia produced by systemically administered morphine-like substances is controversial. The spinal cord and trigeminal nucleus represent sites at which sensory information first enters the CNS. The depressant actions of opiates have been most studied in the spinal cord. The excitation of neurones in lamina IV and V in the dorsal horn by noxious stimuli is selectively inhibited by morphine or stable enkephalins, without inhibiting the responses of these neurones to a non-noxious e.g. mechanoreceptor input. The inhibition is observed whether the opiate is administered into the substantia gelatinosa or into the immediate vicinity of the deeper neurones and is antagonised by naloxone. The selective effect indicates that the action is not due to postsynaptic but to presynaptic inhibition. In keeping with this conclusion it has been demonstrated both in vivo and in vitro that morphine diminishes the stimulation

138

Pain and analgesia

or potassium evoked release of substance P from the spinal cord or trigeminal nucleus: substance P is localised in primary afferent C-fibres but is not confined to these fibres. It is interesting that activation of non-myelinated C-fibres causes the release of endogenous enkephalins in the spinal cord, suggesting that pain is associated with the activation of an endogenous opioid system, mimicked by the effects of analgesic opioids. Excitatory effects on dorsal horn neurones have also been described and are most evident on those neurones which do not receive an excitatory input from nociceptors are inhibitory to the projection neurones in the dorsal horn: if so, then the excitation would be synergistic with the inhibitory effect on the excitation of the projection neurones by noxious inputs. Such an explanation indicates that there are several mechanisms by which the input of nociceptor information to the projection neurones is modified by analgesic substances. An excitatory action on Renshaw cells has also been reported. This excitation is blocked by the nicotinic receptor antagonist, dihydro-p-erythroidine and was therefore ascribed to the release of ACh from the terminals of motor axon collaterals. In quadriplegic man the ability of morphine to inhibit polysynaptic flexor reflexes is similar to its effect in normal man. Taken with other evidence this indicates that a major component in the analgesic action of opiates is their ability to inhibit transmission of noxious information to projection systems at the level of input to the CNS, i.e. at the spinal level or its equivalent supraspinally, e.g. the trigeminal system. The spinal action of morphine is that employed clinically in pain relief by intrathecal administration of morphine. The use of this technique in surgery avoids to a large degree the side effects associated with actions in other parts of the nervous system, e.g. respiratory depression. In the brain, the usual effect of morphine is inhibition but there is a marked excitation of neurones in the hippocampus which has been ascribed by various authors to disinhibition, i.e. the inhibition of inhibitory interneurones, or to facilitation: these differing interpretations have not been satisfactorily resolved.

Cellular actions of opiates

139

Descending control and analgesia It is quite clear that stimulation of certain supraspinal sites both causes analgesia and inhibits the activation of spinal neurones by noxious stimuli. It is also clear that under some conditions there are tonically acting descending systems with similar effects. Since the microinjection of morphine into some brain sites similarly causes analgesia and inhibition of spinal neurones it has been proposed that one of the important effects in opiate analgesia is the activation of descending inhibitory control systems. However, the concentrations achieved at and around the injection sites are most probably in the millimolar range and would seem to have no relevance to the analgesia achieved by systemic administration, after which the concentration in the brain does not exceed micromolar levels. The descending projection most frequently invoked in descending control of nociception is one from the raphe nuclei to the spinal cord: this projection contains 5-hydroxytryptamine. Iontophoretically administered 5-HT has effects similar to those of morphine in selectively inhibiting the noxious input to spinal neurones. However, other descending systems such as the noradrenergic and dopaminergic systems may also have a role to play. The extent to which they are involved in opiate analgesia is uncertain. Although there is uncertainty about the activation of descending systems by opiates, there can be little doubt that the clear supraspinal effects of morphine contribute to the overall analgesic action, which includes changes in the affective reaction to a painful stimulus as well as changes in perception of it. Equally, the ubiquitous although not even distribution of opiate receptors throughout the CNS may explain some of the side effects of opiates. Fig. 8.6 shows in diagrammatic form some of the proposed sites of action of morphine-like analgesics.

Cellular actions of opiates A reduction in transmitter release is a notable feature of opiate actions. This is established for the release of acetylcholine,

140

Pain and analgesia

noradrenaline and substance P. The selectivity for nociceptive pathways must therefore be due to a selective localisation of the opiate receptors on the terminals of the fibres of neurones in these pathways. In some situations a hyperpolarisation of the post-synaptic membrane has also been demonstrated. Selective interactions with excitatory or inhibitory amino acids, proposed by some investigators have not received much support from other studies. The problem then is to find the mechanism or mechanisms which might cause the reduced release and hyperpolarisation. Adenylate cyclase inhibition is one action of opiates which has been proposed as a primary action. However, this is probably a secondary, rather than a primary, action. In any case, the action has been best established on cultured neuroblastoma and glioma cells which have predominantly 8 type binding sites. It is not clear whether the effect on cyclic AMP occurs with activation of the other opiate receptors. Certainly, the electrophysiological effects of opiates are unchanged by cyclic-AMP, its derivatives or phosphodiesterase inhibitors. However, inhibition of cyclicAMP production has also been observed in the striatum, probably again mediated through a 8-type receptor. An increased conductance of the membrane to potassium ions is a currently favoured mechanism. This has been demonstrated at various sites and certainly explains the hyperpolarisation seen, for instance, on neurones of the locus coeruleus, a major source of noradrenaline containing neurones in the brain. Presynaptic depolarisation accompanies presynaptic inhibition. If the presynaptic inhibition is due to the depolarisation, Fig. 8.6. Proposed sites of action of morphine-like analgesics. Trigeminal neurones

Motoneurones

Limbic forebrain \

_ . T Projection

neurones

Periaqueductal ^ey

" fc . Primary afferent

£^™§

Brainstem reticular formation

Neurones in substantia

gelatinosa

Tolerance to opiates

141

then presynaptic hyperpolarisation should increase, not decrease transmitter release. Thus an increased potassium conductance causing a hyperpolarisation of the terminals is not an adequate explanation for the reduced transmitter release seen with opiates. However, the increased conductance of the terminal membranes might prevent action potentials from invading the terminals: this would clearly reduce transmitter release. Another possibility is that the influx of calcium through a depolarisation dependent calcium channel is directly reduced by activation of opiate receptors. While this is probably the mechanism in some situations, it is unlikely to be the mechanism at all sites and under all circumstances. Finally, if the increased potassium conductance is the primary mechanism, then the resulting hyperpolarisation would tend to reduce the voltage-dependent entry of calcium into the terminals. Since transmitter release is calcium dependent, this would explain why both hyperpolarisation and reduced transmitter release has been observed with opiates. However, this does not explain the excitatory effects of opiates unless they are all due to disinhibition, which is most unlikely. Tolerance to opiates In isolated systems, morphine administration decreases the level of cyclic-AMP but the effect does not persist: i.e. there is tolerance. There is a 'ebound' increase in the level after withdrawal of morphine. Although a decrease in opiate receptor numbers is produced in similar preparations after peptide administration, it is not seen with morphine. It is therefore unlikely that this mechanism accounts for the tolerance to morphine. To date, a convincing explanation of tolerance based on changes in cyclic nucleotide formation has not been forthcoming. More useful hypotheses explain tolerance in terms of changes in specific functions of the nervous system. One interesting observation is that prolonged electrical stimulation of the nucleus raphe magnus at first inhibits the jawopening reflex in response to noxious stimuli but later becomes ineffective. At this time, an injection of 5-hydroxytryptophan, the

142

Pain and analgesia

precursor of 5-hydroxytryptamine, restores the reflex. This indicates that there can be an adaptive reaction of control systems in the pain pathway with prolonged activation of the pathway. Similar adaptations might occur when pain pathways are tonically inhibited by opiates. Another interesting observation (Fig. 8.7) is that there is an increased number of (3-noradrenergic receptors, with no change in opiate receptors, in the cerebral cortex after chronic treatment with opiates. This could be a compensatory response to the maintained decreased release of noradrenaline caused by opiates and is revealed by an increased stimulation of adenyl cyclase by the amine. The result of the increased number of receptors would be Fig. 8.7. A possible mechanism for tolerance to opiates, (after Schwartz, J.-C. 1979, Trends in NeuroscL, 2, 137. Terminal

/^-receptors ooo ooo ooo

Normal

Transmitter release

Response

Normal

Normal

Normal

Normal

Decreased

Decreased

Increased

Decreased

Normal (i.e. tolerance)

Increased

Normal

Increased

Decreased

/i-receptors

Acute morphine Morphine

Chronic morphine

Abstinence or Naloxone"

Abstinence Clonidine

m

Increased (i.e. withdrawal symptoms)

Normal (Clonidine prevents withdrawal symptoms)

Tolerance to opiates

143

to restore the system to its original, 'pre-opiate 1 state. When the opiate is withdrawn, the increased release of catecholamine would lead to an increased response. If similar changes occurred in other systems in which transmitter release is reduced by opiates, then the changes in receptor numbers could be the basis of tolerance- and abstinence-evoked withdrawal symptoms. The a-2 agonist, clonidine, ameliorates withdrawal symptoms in the addict and is itself analgesic. Clonidine is known to prevent transmitter release, like morphine, and itself has some analgesic effect. The cellular action of clonidine appears to be the same as morphine in increasing potassium conductances and reducing calcium fluxes. However, this identical ionic effect is not achieved through interaction with opiate receptors.

Drug interactions with inhibitory amino acids A number of drugs interact with the inhibitory neurotransmitters glycine and y-aminobutyric acid (GABA) either as competitive receptor antagonists (strychnine and bicuculline), as non-specific channel blockers (picrotoxin), to reduce their release from terminals (tetanus toxin) or by more complex interactions with the receptors leading to a potentiation of the effect of the transmitter (benzodiazepines, some barbiturates and some anti-epileptic drugs). The functions of glycine and GABA as inhibitory transmitters and the actions of antagonists are discussed in Chapter 5. Fig. 9.1 summarises drug interactions with inhibitory amino acids and their receptors. The pharmacological and clinical implications of drugs which modulate the effect of the inhibitory amino acids, either to reduce or to enhance their effects, are numerous and are subdivided in this chapter into three groups (convulsants, anxiety-reducing drugs and anti-epileptic drugs): some drugs, e.g. benzodiazepines and barbiturates, are both anxiety-reducing and anti-epileptic. The boundary between these is therefore indistinct. The barbiturates are considered in more detail in Chapter 7, principally as anaesthetic agents. It should be remembered that not all of the actions of anxiety-reducing and anti-epileptic drugs can be explained simply in terms of their interactions with the inhibitory transmitters, and there may be additional factors involved in determining which drug or group of drugs is more effective in particular clinical states. Convulsants Strychnine, bicuculline, picrotoxin and tetanus toxin all reduce the effectiveness of amino acid-mediated inhibitory

Convulsants

145

neurotransmission but they do so in different ways. Strychnine and bicuculline are competitive antagonists at the glycine and GABA-A receptors, respectively (see Chapter 5). Picrotoxin blocks the chloride channels which are opened by both amino acids. Tetanus toxin selectively blocks the release of inhibitory amino acids from the nerve terminals, probably by a mechanism similar to that of botulinus toxin at the neuromuscular junction (Chapter 3). Historically, whilst convulsant agents have been used as general stimulants (e.g. nux vomica extracts were used as a general fc pep' preparation and contained strychnine; picrotoxin was used for a time in the treatment of psychotic depression), such uses are clearly dangerous and are no longer in vogue. More recently, strychnine has been tried in the treatment of the fairly rare condition known as non-ketotic hyperglycinaemia, an inherited disorder of glycine metabolism in which the oxidation of glycine and the formation of serine is deficient. In this condition, there is a three-fold increase in the plasma concentration of glycine and a 20-fold rise in CSF levels. The disease appears in the first few days or weeks of life and may lead to a premature death. The majority of children survive to develop severe Fig. 9.1. Summary of drug interactions with inhibitory amino acids and their receptors. Convulsants

Anti-epileptics

GABA/GLYCINE

Antispastic

Anti-anxiety

Sedative Hypnotic General anaesthetic

146

Drug interactions with inhibitory amino acids

hypotonia, myoclonic jerks, generalised seizures and severe mental retardation. The condition is resistant to all forms of therapy, including restriction of glycine in the diet. Treatment with strychnine daily for several months to a few patients gave modest improvement in muscle tone with some increase in alertness, indicating that perhaps the condition was due to excessive inhibition caused by the high levels of glycine in the CNS. Strychnine is still used as a rodent poison and the most likely encounter with the drug will be in accidental poisoning. It is widely used experimentally to elucidate the involvement of glycine as a transmitter in inhibitory pathways. The only known use for bicuculline is an experimental one to elucidate the involvement of GABA in inhibitory transmission. However, since the antagonist selectively blocks only GABA-A receptors, negative results will not necessarily eliminate an action of GABA interacting with GABA-B receptors. Baclofen ((3chlorophenyl-GABA) is an agonist for the GABA-B receptors and is an anti-spastic agent. Tetanus toxin is produced by Clostridium tetani and is the only agent known to block the release of both glycine and GABA. The bacillus is a normal inhabitant of the gut of horses and sheep. It is an anaerobic organism and on infection of deep penetrating wounds it liberates the extremely powerful exotoxin. The toxin is transported along both sensory and motor nerves into the CNS, after which it spreads along the neuraxis. One of the first symptoms may be fclock-jaw' but it later causes generalised convulsions. The only effective treatment is with the antitoxin and prophyllactic therapy is advisable. Substances which block glutamic acid decarboxylase (GAD), the enzyme which synthesises GABA from glutamic acid, may also cause convulsions at high doses. Examples of such drugs include carbazide, thiosemicarbazide, allyglycine and 3mercaptoproprionic acid. In contrast, GABA levels in the brain and the sensitivity to GABA are both increased when GABA transaminase is inhibited by amino-oxyacetic acid, but there is not a good correlation with the anticonvulsant action.

Anxiety-reduction and sedative-hypnotics

147

Anxiety-reduction and sedative-hypnotics It is possible that the reduction of anxiety, the production of sedation and the induction of sleep (hypnotic action) may be but a continuum of progressive dulling of function in the CNS. Even anti-epileptic action and anaesthesia (Chapter 7) could be viewed as an extension of this dulling. This postulate would be compatible with the observation that drugs may be found in each category which potentiate the inhibitory effect of GABA on CNS function. However, it ignores the fact that not all drugs in these classes have this property. Furthermore, different drugs may have the same, or similar, pharmacological actions but to different relative degrees: this is difficult to reconcile with a common pharmacological mechanism for all therapeutic effects. In addition, some drugs may possess one or several, but not all, of the effects. Thus, depression of the nervous system can be achieved by many different mechanisms. The drugs in this class do not directly bind to GABA receptors, but there are a number of specific binding sites which have been postulated. It is likely that some of these may interact with the binding of GABA to its receptor binding site, and that the action of GABA may be potentiated by several different mechanisms. This is best documented for the interaction of benzodiazepines with specific benzodiazepine receptors. Some difficulties arise in the interpretation of many pharmacological tests which supposedly differentiate between the various states of anxiety reduction, sedation, hypnosis and motor incoordination. Drugs which appear to differ in the various experimental tests do not come up to expectations in man. In the following account, particular attention will be given to those classes of drugs for which reasonable evidence for a pharmacological mechanism exists. Others will receive only a brief mention for the sake of completeness. It so happens that currently the most important therapeutic agents in the category of anti-anxiety/sedative/hypnotic drugs are the benzodiazepines and that more is known about these substances than any other in this class. Anxiety is a normal part of ordinary human existence and should only require treatment when it becomes excessive, result-

148

Drug interactions with inhibitory amino acids

ing in an inability to cope with the normal stresses of everyday life. Excessive anxiety is a frequent component of neurotic conditions. There is a maintained contact with reality, which readily differentiates the condition from more severe psychotic states in which this contact is lost or impaired. There is a high rate of spontaneous recovery, which makes evaluation of therapeutic measures difficult. Often, there is a marked improvement with placebos. Consequently, it is necessary to conduct carefully controlled trials to evaluate any treatment. Benzodiazepines In the past, barbiturates were the substances most commonly used to produce a reduction in anxiety, daytime sedation or sleep. Whilst still employed as anaesthetic agents (Chapter 7), the low therapeutic index, reflecting a low safety margin, is a contra-indication for their use in out-patients. Their use has been almost entirely supplanted by the benzodiazepines which are generally considered to be relatively safe, non-toxic, nonaddictive, non-habituating drugs with a minimum of disturbing side-effects. They are now the most widely used of all psychotropic drugs, representing a high proportion of all drug prescriptions. In the United States in the year 1977, 8000 tons of benzodiazepines were dispensed. There were a total of 80 million prescriptions for Librium, Valium and flurazepam alone. Their use has if anything increased since then but there is a growing awareness of the problems of overprescription and in the United Kingdom the benzodiazepines have become a major problem as drugs of abuse. Withdrawal symptoms occur when they are given over long periods of time and are then abruptly withdrawn. The withdrawal symptoms can be severe, and may not occur immediately. Habituation occurs frequently, due in part to the long half-lives and the conversion to active metabolites. There may be a potentiation of the depressant effects of alcohol. In the elderly or in those with liver dysfunction there may be excessive reactions to 'normal' doses. Increasingly, there are problems associ-

149

Benzodiazepines

ated with overdosage, as one might expect in a very large susceptible population exposed to the drugs. Tranquillizer withdrawal' groups have been set up in numerous places to provide self-help techniques for overcoming the problem. Although these drugs were introduced into clinical practice in 1960 and there is no certain evidence of cumulative and nonreversible damage or teratogenesis resulting from prolonged treatment, this should not lead to complacency. In general terms, prolonged drug treatment of any description is potentially a hazard and should only be undertaken with due regard to the balance of possible benefits and long-term toxic effects, which are notoriously difficult to evaluate. Perhaps a lesson should be learned from the recent studies on tardive dyskinesia with antiFig. 9.2. Structures of benzodiazepine drugs. Barbiturates : see table 6.2 NHCH, O

CH,

II

I

O

II

H,NCOCH,CH,CCH,OCNH, I CH, Meprobamate

Chlordiazepoxide (Librium)

Diazepam (Valium)

Nitrazepam (Mogadon)

150

Drug interactions with inhibitory amino acids

schizophrenic drugs, indicating previously unsuspected possibilities of structural damage to the CNS with prolonged therapy in susceptible patients (Chapter 11). Paradoxical effects, including an increase in hostility, disturbing dreams and an increase in anxiety may be occasionally observed, and are a potential risk in patients with additional psychotic symptoms. Increased aggression has also been noted in volunteers when an element of frustration is introduced into the evaluation procedure. In mice it has been demonstrated that librium may have the usual calming effect when the mice are kept in small groups but in large groups there is an increased aggression, leading to an increase in the mortality rate. Thus environmental factors may influence the behavioural effect. Pharmacokinetics The benzodiazepines are highly lipid soluble. They are therefore well absorbed when taken orally. However, there are marked differences between different members of the group. They are strongly bound to plasma proteins and the concentration of drug in the CSF is approximately that of the free drug in plasma. They are broken down in the liver, but the breakdown products are frequently themselves active and may have a longer half-life than the parent compound. It is likely that clinical usage is more dependent on the different pharmacokinetic properties of different benzodiazepines than on differences in potency or to intrinsic differences in pharmacological mechanisms. Thus, in general, a night-time hypnotic such as Mogadon would need a rapid onset of action but a relatively short duration of effect, whereas a daytime sedative such as Librium or Valium would require a more prolonged action. Occasionally, for example in dental practice, benzodiazepines are given intravenously as anaesthetics, or as supplements to anaesthesia. Although relatively safe and patients can become rapidly ambulatory, there are considerable risks in a too rapid release from the surgery because of persistent impairment of motor coordination and reaction time. There may also be some retrograde and anterograde amnesia.

Benzodiazepines

151

The high lipid solubility of benzodiazepines leads them to be effectively transferred across the placenta and to be secreted in milk. They may therefore cause depression of the foetus or neonate. Pharmacological actions Benzodiazepines were originally introduced because of their easily demonstrated 'taming' effects in animals. Perhaps their effects upon such wild animals as lions and tigers have produced some of the most eye-catching advertising in the pharmaceutical industry. Nevertheless, caution must be exercised in extrapolating these changes in behaviour in experimental and wild animals to man. The effects of benzodiazepines in the therapeutic dose range on peripheral functions are slight and attention has been focussed on their central effects. Muscle relaxation. One of the earliest properties to be demonstrated was the muscle-relaxant effect. This has led to their use in the treatment of spasticity, but their effectiveness is dubious or at most slight. They reduce polysynaptic spinal reflexes more than monosynaptic reflexes, but this is true of many depressants of the CNS. Effects on endogenous monoamines. Benzodiazepines have no effect on the endogenous levels of 5-hydroxytryptamine, dopamine or noradrenaline in the brain, but they do affect turnover of these amines, particularly in stress-inducing situations. In behavioural tests, coupled with biochemical measurements of amine turnover, the reduction in anxiety with oxazepam correlated with a reduction in 5-HT turnover, but not with noradrenaline turnover. However, it is unlikely that any of these effects represent a primary target for the action of benzodiazepines. Since they clearly affect function in gabergic systems, the changes in mono-amine turnover is probably the indirect consequence of changes in gabergic transmission. Effect on excitatory action of acetylcholine.

Benzodiazepines

152

Drug interactions with inhibitory amino acids

depress the excitatory action of acetylcholine on cerebro-cortical neurones and this action may also be indirect: it is also found with a number of other general depressants such as the barbiturates. Localisation of action. Attempts to localise the actions of benzodiazepines to particular 'centres' in the CNS have not met with notable success. Systemic administration of Librium or Valium to experimental animals decreases the spontaneous firing of neurones in the hippocampus more than in the pre-optic area of the hypothalamus or reticular formation of the brainstem. In contrast, pentobarbitone exerts its greatest effect under similar conditions in the reticular formation. Similar differences have been observed in EEG recordings. However, it is possible that these small differences reflect differences in the initial localisation of the drugs, rather than any basic differences in their sites of action. Depressant effects are also observed on neurones in other parts of the nervous system. Potentiation of the effects ofGABA. Particularly relevant actions are the ability of benzodiazepines to increase GABA-mediated dorsal root potentials, to enhance presynaptic inhibition in cats and to depolarise primary afferent nerve terminals. The effects are blocked by bicuculline and are greatly reduced after GABA levels are lowered by the administration of thiosemicarbazide. Other effects which indicate a potentiation of the action of endogenous GABA include potentiation of basket cell-evoked inhibition of cerebellar Purkinje cells and potentiation of inhibition in the cuneate nucleus: both types of inhibition are due to postsynaptic, GABA mediated transmission and are blocked by bicuculline. Benzodiazepines can also to some extent reverse the actions of GABA antagonists, presumably by increasing the effect of the available transmitter and thereby competing with the antagonist. GABA causes inhibition mainly by increasing the chloride ion conductance of the neuronal cell membrane (Chapter 5). Benzodiazepines increase the action of GABA on the chloride conductance by increasing the frequency with which chloride

Benzodiazapine receptors

153

channels open in response to a given concentration of GABA. Barbiturates also increase the action of GABA but they do this by increasing the time for which the chloride channel stays open, rather than by increasing the frequency of opening. Decrease in cyclic GMP levels. Benzodiazepines decrease cyclic GMP in the cerebellar cortex and deep cerebellar nuclei, an effect which is similar to that of GABA, and prevent the rise due to isoniazid, picrotoxin and pentylenetetrazol. These effects are almost certainly secondary to changes in the firing rate of the neurones since the cyclic GMP in the cerebellum is preferentially localised in neuroglia, rather than in neurones. Benzodiazepine receptors One of the most significant advances since writing the first edition of this book is the discovery and elucidation of a group of membrane receptors which specifically bind benzodiazepine drugs and which interact with the GABA-A receptor. There have been suggestions for the presence of endogenous ligands which bind to the receptor and modulate interaction between the benzodiazepine recognition sites and the GABA-A receptor. Although it was initially shown that benzodiazepines displaced strychnine from its binding sites, indicating that there was an interaction with glycine receptors, a high affinity binding was not confirmed in later studies and there is no evidence from in vivo neuropharmacological experiments for modification of the actions of glycine. There is a high affinity, saturable, reversible, and stereospecific binding of benzodiazepines to membranes prepared from brain homogenates. The binding site seems to be quite specific for benzodiazepines, and does not bind a wide variety of other putative transmitters, including acetylcholine, noradrenaline, dopamine, 5-hydroxytryptamine, GABA, glycine or glutamic acid. There is no displacement of benzodiazepine binding by a variety of substances such as strychnine, pimozide, clozapine, bicuculline, picrotoxin, etorphine, naloxone, meprobamate, barbiturates and ethanol.

154

Drug interactions with inhibitory amino acids

The affinity for benzodiazepine binding to these sites is some 10 000 times higher than for the displacement of strychnine binding to the glycine receptor. Although radio-labelled diazepam binding to the receptors can be measured at 0 °C it cannot be measured at 37 °C. This is because the dissociation of the labelled compound from the receptor is so fast at the higher temperature that the bound label is lost during the washing procedures. This problem does not arise with flunitrazepam, which dissociates more slowly at both low and high temperatures. Irreversible binding of flunitrazepam, or other compounds with a nitro group at position 7, is achieved by exposing to ultraviolet light after contact with flunitrazepam at 0 °C. This is the process of photoaffinity labelling. The irreversible binding is rather unusual, and not fully explained. However, even though up to 90% of the receptor sites are irreversibly occupied by the label, it does not reduce the binding of other ligands to the extent that one would expect. Furthermore, GABA still facilitates the binding of benzodiazepines (see below). Correlation between binding and pharmacological action. There is a

high correlation between the ability of various benzodiazepines to bind to the specific sites in human or rat brain and such pharmacological actions as antagonism of leptazol-induced convulsions (mice), inhibition of fighting behaviour (mice), tests of motor coordination (mice), muscle relaxant effects in cats, their anti-anxiety effects in man and, perhaps surprisingly, the therapeutic doses recommended by the manufacturers. Regional variation. There is a regional variation in the amount of receptor material: the highest levels are found in the cerebral cortex and the lowest in the spinal cord, with intermediate amounts in the cerebellum, basal ganglia, hypothalamus and thalamus. There is also a species variation. For example, in the cerebellum of rat and man there is a similarly high concentration in the molecular layer. However, the concentration in the granular layer is much greater in man than in the rat. There have been some interesting studies in mutant mice

Benzodiazapine receptors

155

which have differential loss of various cell types in the cerebellum. It is apparent that the Purkinje cells contain the receptor but that the granule cells, which give rise to the parallel fibres, do not. However, if all Purkinje cells are absent then not all receptor binding is lost, perhaps indicating that the presence or absence of particular cell types may determine the expression of receptors on other cells. Correlation with GABA receptors. In general terms, the benzo-

diazepine receptor binding follows GABA receptor binding, but the correlation is not absolute. This suggests that GABA receptors may occur independently of benzodiazepine receptors but that benzodiazepine receptors only occur with GABA receptors. Detailed studies of benzodiazepine receptors in Barnard's laboratory indicate that they may be integral components of the GABA receptor. Since the pharmacological action of benzodiazepines which seems most relevant is the potentiation of GABA-mediated inhibition, it is useful to consider the experimental evidence which supports the hypothesis that benzodiazepines modulate the binding of GABA to GABA-A sites indirectly via a receptorreceptor interaction. The evidence from the binding studies is shown in Fig.9.3, together with a more speculative diagram showing how the various drug interactions might occur. In membranes prepared from a fresh homogenate of rat brain GABA has a low affinity with a KD of about 200 nM. If the fresh membranes are exposed to benzodiazepines at effective concentrations then the KD for GABA decreases about ten-fold, indicating an increase in binding affinity, exactly what one would predict from the pharmacological studies showing that effects of GABA are potentiated. Repeated washing of the membranes in the detergent Triton X100 also converts the low affinity to high affinity sites. If the high affinity sites are now treated with the supernatant from the original centrifugation, or with a thermostable protein separated by chromatography from the supernatant, then the high affinity sites revert to the low affinity state. On this basis is was postulated

156

Drug interactions with inhibitory amino acids

that brain contains a substance (GABA modulin) which can alter by its presence or absence the affinity of the receptor sites for GABA. However, it should be emphasised that the true identity of this endogenous modulatory substance has so far escaped positive identification. Fig. 9.3 illustrates how benzodiazepines, GABA, the endogenous ligand, a competitive antagonist such as bicuculline and a channel blocker such as picrotoxin might all interact at GABA sites. Fig. 9.3. The interactions between benzodiazepines, amino acids and antagonists and their respective receptors in brain homogenates and speculative explanation. GABA binding in brain homogenates (Benzodiazepine binding has similar characteristics) Fresh membranes 0 °C Homogenise Fresh membranes Low affinity for GABA KD approx 200 nM

Wash

Supernatant (S) Chromatography

Add sbenzodiazepine

Triton-X-100

Thermostable protein (p)

High affinity KD approx 20 nM

High affinity KD approx 20 nM Add S or P Low affinity KD approx 200 nM

Bicuculline

Endogenous ligand for receptors \——^— Benzodiazepine Membrane

Chloride channel

k

GABA receptor s Picrotoxin binds to chloride channel

Benzodiazapine receptors

157

Cellular and subcellular localisation of the benzodiazepine receptor.

Kainic acid, which causes the death of neurones, reduces the number of benzodiazepine binding sites by only about 50% after injection into various brain sites. This indicates that at least some of the receptor sites exist on the postsynaptic membranes of neurones which are destroyed by kainate: the remaining 50% may be presynaptic or may be in glial cells. Subcellular distribution studies indicate that cell membranes contain much of the receptor material, with little present in nuclei or microsomal fractions. There is also some in the presynaptic membranes. Endogenous ligands for the receptor. The discovery of the opiate

receptors and their endogenous enkephalin ligands led to a hunt for endogenous ligands for the benzodiazepine receptor. A number of suggestions have been made, but none substantiated. These have included high molecular weight substances with molecular weights in excess of 1000. However, such studies are complicated by the uncertainty regarding the degree of purity of the samples isolated. A number of smaller molecules such as inosine, hypoxanthine and nicotinamide have also been suggested but all have a low affinity (KDin the millimolar range). In 1980 there was great excitement when 1.78 mg of an active substance was isolated from 1800 litres of human urine. This was shown by mass spectrometry and nuclear magnetic resonance spectroscopy to be beta-carboline-3-carboxylic acid ethyl ester (P-CCE). The structure was confirmed by synthesis and it had a very high affinity in the nanomolar range for displacing diazepam binding. Unfortunately, the substance has never been isolated from brain, and almost certainly is an artifact formed from tryptophan by the acid hydrolysis extraction procedure. Nevertheless, P-CCE had pharmacological actions as a kproconvulsanf or 'inverse agonist' (it lowered seizure thresholds without itself causing seizures) and antagonised the anticonvulsant effects of benzodiazepines. More importantly it started a new era in the benzodiazepine receptor story. In 1985, the use of monoclonal antibodies to help identify endogenous ligands for benzodiazepine receptors has led to the suggestion

158

Drug interactions with inhibitory amino acids

that the ligand is a non-protein compound with a specific localisation in neurons. In 1986, a new endogenous ligand for the benzodiazepine receptor was characterised. The substance seemed to be an agonist with a molecular size of about 1000, was localised in synaptosomes and was probably not a peptide. Multiple receptors for benzodiazepines. Early work on the binding

of benzodiazepines to the receptor indicated a single, homogeneous receptor population. The displacement of flunitrazepam binding by P-CCE and the fact that the affinity constant varied from one brain region to another did not fit with the concept of a single receptor population. For example, the Ki for (3-CCE displacement of flunitrazepam binding is about 1 nM in the cerebellum and cerebral cortex but it is 4-7 nM in the hippocampus, corpus striatum and nucleus accumbens. Furthermore, a new class of compounds, the triazolopyridazines (e.g. CL218872) distinct chemically from the benzodiazepines have been developed which also bind to the benzodiazepine receptors and share some of their pharmacological properties. However, CL218872 inhibits diazepam binding with a Hill slope of less than one (0.6 to 0.7): this indicates that there is more than one receptor. Presently, it is considered that there are at least two benzodiazepine receptors which we may conveniently call BZ1 and BZ2. Some differences between them are shown in Table 9.1. The relative proportions of BZ1 to BZ2 receptors varies in different brain regions. In the cerebellum, BZ1 receptors predominate. In the hippocampus there are similar amounts of both and in the basal ganglia and the superficial layers of the superior colliculus BZ2 receptors predominate. The affinity of diazepam and flunitrazepam for BZ1 and BZ2 is about the same. However, p-CCE and the triazolopyridazines (CL218872) have about a ten-fold higher affinity for the BZ1 receptor than for BZ2. The BZ1 receptor is slightly smaller than the BZ2 receptor (51 000 and 55 000 Daltons respectively). Benzodiazepine receptors are also found in the peripheral tissues of the liver, kidney and lung, but they differ in some respects from those in the central nervous system in the binding affinities

Benzodiazapine receptors

159

Table 9.1. Differences betwen the two types of benzodiazepine receptor (BZ1 and BZ2)

A: Regional distribution:

Cerebellum Hippocampus Cerebral cortex Basal ganglia

B: Relative affinities for benzodiazepine receptors:

Diazepam or flunitrazepam Beta-carbolines (p-CCE) Triazolopyridazines (CL218872) C: Size of receptor (Daltons)

BZ1

BZ2

92% 50-60% 75% low

<8% 40-50% 25% high

1 10 10 51000

1 1 1 55000

of some ligands. However, in tissue culture the binding of benzodiazepines to glial cells seems to be similar to the peripheral type of receptor whereas the binding to the neurones has identical characteristics to binding in fresh brain. Further evidence for multiple receptors come from studying the effects of GABA and chloride ion concentration on binding of ligands to the receptors (Table 9.2). It can be seen quite clearly that the presence of GABA increases binding to the 'diazepam' and kCL218872' type receptors but not binding to the k(3-carboline' or 'inactive' type. Increasing the chloride ion concentration affects binding of the various ligands in quite different ways. Such studies have led some to suggest as many as three different binding sites. The pharmacological spectrum of agents which are benzodiazepines or which bind to benzodiazepine receptors. Table 9.3 shows the

range of pharmacological profiles of a number of selected agents. All of these substance are either 1,4-benzodiazepines or interact with the benzodiazepine receptor. It is important to note that a number of quite different chemical structures can bind to the receptor and produce pharmacological agonist effects (e.g. 1,4-benzodiazepines, triazolopyridazines, |3-carbolines, quinolines).

160

Drug interactions with inhibitory amino acids

Table 9.2. Effects of GABA and chloride ion concentration on binding of ligands to benzodiazepine receptors BINDING TYPE

Diazepam

CL218872

p-CCE

Inactive

diazepam nitrazepam Increase Increase a,b,c,

CL218872 oxazepam Increase 0 a,b

(3-CCE harmaline 0 Increase a,c

nicotinamid leptazol 0 0 a

LIGANDS

GABA Chloride Binding sites on BDZ receptor

Binding sites (c) affected by raising the chloride concentration may be associated with the chloride ionophore, the sites (b) affected by raising the GABA concentration may be associated with sites interacting with the GABA receptor and sites unaffected by these procedures are functionally inactive sites.

One 1,4-benzodiazepine, RO-15-1788, is devoid of agonist action and is a relatively good antagonist. This compound is undergoing clinical trial as an agent which may be of value in problems of benzodiazepine overdose, an increasing problem, and as an agent to terminate the actions of benzodiazepines when they are used in anaesthesia. The quinoline, PK9084, is interesting in that it seems to have a good anti-anxiety effect without causing much sedation, clearly of immense value clinically if the indications are substantiated. Finally, trifluadom is a benzodiazepine which does not bind to benzodiazepine receptors. However, it is not without effect. It causes analgesia and it does this by binding preferentially to the kappa opiate receptor. Conclusions. The benzodiazepine receptors are probably integral parts of the GABA-A receptor which provide targets for a range

Other anxiety-reducing, sedative-hypnotic drugs

161

Table 9.3. Pharmacological profiles of substances binding to benzodiazepine receptors

Benzodiazepines (diazepam) Triazolopyridazines (CL218872) Quinolines (PK9084) Ro-15-1788

Anticonflict

Sedative

Anticonvulsant

+++ 0

+ 0

0 0

Antagonist

Types: (i) Broad spectrum Benzodiazepines (ii) Anti-anxiety/anti-convulsant Triazolopyridazines (iii) Anti-anxiety Quinolines (iv) Antagonist only RO-15-1788 (a benzodiazepine derivative)

of substances, not all of which are chemically benzodiazepines. There are at least two receptor subtypes and the interaction of various agents with each of these to different extents may determine the apparently complex range of effects seen with the active compounds. This provides hope for finding newer substances with a better spectrum of pharmacological action than is found in those currently available for clinical use. In particular, a drug which reduced anxiety without undue sedation would seem to be a useful advance. Antagonists may also be of value, particularly in treating overdose. However, the incidence of overdosing might be more effectively combated if the prescription of these drugs were to be more circumspect and less use was made of them as universal panaceas for all the ills of mankind. Other anxiety-reducing, sedative-hypnotic drugs In addition to the barbiturates and benzodiazepines, a large number of other drugs have been introduced for the alleviation of anxiety or as sedative-hypnotics. Some have been in use

162

Drug interactions with inhibitory amino acids

for decades, e.g. chloral hydrate and some agents which break down in the body to chloral hydrate. Others have no major advantages. Most of these have all but disappeared from the clinicians' armamentarium. Meprobamate was the first of the knew' era of anxiety reducing drugs and was first introduced in 1950. It is still used, principally for geriatric patients although proof of a selective anxiety-reducing effect is lacking. It selectively reduces polysynaptic spinal reflexes but is not a useful anti-spastic agent. It does not potentiate the inhibitory action of GABA and nothing is known about its mechanism of action. Anti-epileptic drugs Although included in this section on the basis that some agents, notably anticonvulsant benzodiazepines, phenytoin, barbiturates and valproate, used in the therapy of epilepsy modify the inhibitory action of GABA in the CNS, this mechanism cannot be considered to be a common mode of action for all antiepileptic drugs. It should also be emphasised that the potentiation of the action of GABA by barbiturates and valproate is produced by mechanisms which differ from those of the benzodiazepines. Characterisation of epileptic seizures Epileptic seizures are characterised as being spontaneous, recurrent, episodic and paroxysmal (Hughlings Jackson). Muscle spasms are typical of grand mal or focal (see below) types of epilepsy but are not found in petit mal, characterised by 'absence7 seizures with a transient loss of consciousness. Epilepsy may be produced in many ways. It may be caused by infection, trauma, tumours, etc., but often the causative factor is not apparent. Convulsions may also be produced by a variety of transient, non-recurrent causes and should not in these instances be called or treated as epileptic convulsions (fits). Examples of such transient causes include the convulsions associated with renal failure or evoked by drugs or the fits consequent upon cerebro-vascular accidents in the elderly.

Anti-epileptic drugs

163

Table 9.4. Clinical applications of anti-epileptic drugs Type of epilepsy Grand mal

Petit mal

Status epilepticus

Drugs used

Phenobarbitone Phenytoin Primidone Carbamazepine Valproate (Benzodiazepines) (Bromides obsolete) Ethosuximide Benzodiazepines, especially clonazepam Valproate (Trimethadione) intravenous diazepam or clonazepam intravenous short-acting barbiturate intravenous phenytoin

There have been many attempts at classification of the epilepsies but none are absolutely satisfactory. Since treatment is largely based upon symptoms, the most useful classification is one which gives an indication of the therapeutic regimen to follow. Table 9.4 and Fig. 9.4 show the major drugs used in the treatment of the epilepsies, there are a number of overlaps apparent and it is not always clear which is the drug of choice. Seizures may be broadly classified into two types. Focal seizures are initiated at a specific locus in the brain, often in a cerebral hemisphere. If the initial focal attack occurs in an 'eloquent' area, the seizure itself may be preceded by an 'aura' which can give an indication of the site of the epileptogenic primary site in the brain. If the aura is associated with a site in motor areas of the cortex then it may be reflected by motor phenomena. If it is located in sensory areas then it may be reflected by sensory mani-

164

Drug interactions with inhibitory amino acids

festations, which could be somatic, visual, auditory, olfactory, gustatory or visceral. Autonomic manifestations could be due to a primary locus in central autonomic sites. The aura may be experienced hours or even days before the onset of the motor seizures and a focal discharge of neurones may be detected in the Fig. 9.4. Structures of anti-epileptic agents. H

o=c

O

ii

N - C C22H5S / \ /

c

H «

II

H2C

\

II

W

O

/ \=/

Primidone

Phenobarbitone

Phenytoin C2H5

CH3 C—CH2

oc/

O=C

Vc H

Trimethadione

S

Ethosuximide

COONa

I

CH3CH2CH2CH CH2CH2CH3 Sodium valproate

Carbamazepine

Anti-epileptic drugs

165

electroencephalographic patterns hours or days before the patient is aware of an impending fit. If the primary locus is sufficiently delineated and the epilepsy is both severe and resistant to other forms of treatment, then surgical removal of the locus may lead to improvement, particularly if the problem is clearly associated with an identifiable cause, such as a tumour. As the discharge of neurones in the primary locus increases, so the area of brain tissue affected spreads first across the cortex and finally it may involve many brain areas, including subcortical regions. Ultimately, the explosively developing discharge of brain neurones may lead to a major fit with loss of consciousness. Recurrent fits without recovery of consciousness represent the medical emergency of status epilepticus. Loss of consciousness does not invariably accompany focal seizures, e.g. in seizures of the frontal lobes. The second type of seizure may be called petit mal and is characterised by absence attacks, in which there is only a transient loss of consciousness, sometimes lasting only seconds. The seizure does not originate from a specific locus in the cerebral cortex and most likely arises from a mid-line structure of the brain stem. A bilateral discharge of neurones in the cortex, which can be detected by waves in the electroencephalogram, is typical of petit mal. The frequency of these discharges is about three per second and inter-ictal discharges are rare. Although there is a loss of consciousness, the patient does not necessarily fall and the only outward sign may be an apparent inattention or lack of concentration. There may be a brief period of amnesia. In some cases the attacks may occur very frequently with incidences as high as one hundred attacks being reported. Petit mal usually first occurs before the age of 15 and disappears by age 20 in 80% of the cases. However, some 50% of petit mal patients subsequently progress to grand mal. The use of drugs in epilepsy The use of drugs in epileptic seizures will control the symptoms either completely or at least significantly in a large proportion of the patients. However, some are resistant to all drug therapy. Furthermore, some of the most useful drugs, e.g.

166

Drug interactions with inhibitory amino acids

phenytoin in grand mal, are also highly dangerous and may cause aplastic anaemia. Others, such as the anticonvulsant barbiturates have marked sedative effects in effective dosage and this will clearly impair the quality of life for the sufferer. Bromides were introduced in the last century for quite spurious reasons but were effective in controlling the fits. However, they are extremely toxic and are no longer employed. Long-acting anticonvulsant barbiturates were introduced at the beginning of the century and continue to be employed in grand mal epilepsy, together with phenytoin, which causes less sedation, primidone and the more modern drugs valproate and carbamazepine. With several of these drugs haematological sideeffects must be carefully considered and monitored. In the absence seizures of petit mal, ethosuximide and to a lesser extent trimethadione, the prototype drug for petit mal, are still used. More modern drugs include some benzodiazepines, notably clonazepam, and valproate. Phenobarbitone and phenytoin are both contra-indicated. In status epilepticus, intravenously administered benzodiazepines (diazepam or clonazepam) have been considered to be the drugs of choice. Short-acting barbiturates may also be used but may cause some respiratory depression. Intravenous phenytoin may be effective but may cause toxic reactions if the drug is already being used. Pharmacological mechanisms The effectiveness of many anticonvulsant drugs is similar to their relative effectiveness in reducing post-tetanic potentiation in the spinal cord. This may be due to the potentiation of the action of GABA by some but this cannot be the action of them all. Although benzodiazepines, barbiturates, phenytoin and valproate potentiate the inhibitory action of GABA, different mechanisms are involved. The actions of the first two classes have already been discussed. Possible mechanisms for valproate and phenytoin are described below. Valproate. Valproate is a relatively simple branched-chain fatty acid and was originally used as a vehicle for the administration

Anti-epileptic drugs

167

of other anticonvulsant agents: it was only later discovered that the vehicle itself had considerable anti-epileptic actions, especially in petit mal but also in grand mal seizures. It is said to be relatively free from side-effects. Valproate, in common with phenobarbitone and phenytoin decreases the levels of cyclic G M P in brain but the significance of this action is unknown. It potentiates the action of GABA in iontophoresis experiments but fails to do so in vivo at therapeutic concentrations. There is an increase in GABA levels in brain associated with an inhibition of GABA transaminase but the inhibition is very weak and there is very little correlation between GABA levels and anticonvulsant action. Succinic semialdehyde dehydrogenase is more sensitive but complete inhibition of this enzyme has little effect on GABA levels. Valproate is a potent inhibitor of aldehyde reductase at therapeutic concentrations, but the function of this enzyme, which exists in two forms, is uncertain. A decrease in aspartic acid levels, in parallel with the increase in GABA levels is also produced. Valproate also has direct postsynaptic effects on the membranes of central neurones, possibly caused by a modification of the receptor/ ionophore complex or the lipid environment of the receptor. In conclusion, valproate possesses several interesting effects but it is uncertain which of them is important for the anti-epileptic action. Phenytoin. Phenytoin is one of the oldest drugs used in grand mal epilepsy. Like benzodiazepines, phenytoin potentiates the action of GABA but, unlike the former drugs, it does not bind to the benzodiazepine receptors and does not potentiate GABA actions at therapeutic concentrations. Nevertheless, the chronic administration of phenytoin increases the number of benzodiazepine binding sites and increases the anticonvulsant action of the benzodiazepines. Numerous effects on membrane properties have been demonstrated but usually only at concentrations exceeding the therapeutic level. These effects include decrease in sodium, calcium and potassium conductances and fluxes. Of more interest is the observation that there are specific, satu-

168

Drug interactions with inhibitory amino acids

rable phenytoin binding sites in brain: there are both affinity (KD 6 nM) and low affinity (KD 4.8 JJ,) sites. The values of the dissociation constant for the low affinity site are similar to the concentrations achieved in the CSF after therapeutic doses. The complex binding behaviour of phenytoin is shown in Fig. 9.5. As the concentration of unlabelled phenytoin is increased the amount of labelled phenytoin bound falls in a biphasic pattern: the two components are clearly shown by the Scatchard plot. Binding of phenytoin is inhibited by barbiturates but not by other anticonvulsants (carbamazepine, ethosuximide, trimethadione or valproate), convulsants (pentylenetetrazole, picrotoxin or strychnine), neurotransmitters (ACh, aspartate, dopamine, Fig. 9.5. Binding of phenytoin sodium to brain homogenates (based on data from Burnham et a/., 1981, Can. J. PhysioL, 59, 402. Amount of labelled phenytoin bound [3H-phenytoin]

. KD 6 nM Amount of labelled phenytoin bound

X log [unlabelled phenytoin]

Anti-epileptic drugs

169

adrenaline, noradrenaline, GABA, glycine, glutamate, histamine agonists, 5-hydroxytryptamine), or by ouabain, apomorphine, cyclic AMP or cyclic GMP, morphine, phenylbutazone or tetrodotoxin. In contrast, diazepam and bicuculline enhance phenytoin binding. There is also the suggestion that there may be an endogenous ligand for the binding site since a substance with a molecular weight of 450 has been extracted from calf brain which competes for the phenytoin binding site and which protects from electroconvulsive shocks and from metrazole convulsions. Thus a binding site has been demonstrated but its relevance to a receptor with a role in nervous function remains to be established. In addition, the enhancing effect of diazepam on binding indicates that there could be some relationship to the benzodiazepine receptor. At low concentrations in the therapeutic range (KD approximately 30 |uM), both phenytoin and carbamazepine are now known to block tetrodotoxin-sensitive sodium channels. The block is both voltage and frequency dependent and in this respect resembles the actions of local anaesthetics. It seems likely that the binding sites for the anticonvulsants are the closed or inactivated state of the sodium channels. Whether these are the sites demonstrated in the binding studies is not clear, but the affinities appear to be not too dissimilar. Phenytoin has been shown to produce a number of other effects, but there is no convincing evidence that they are particularly relevant to the therapeutic action. Phenytoin decreases the total sodium and the calculated intracellular sodium ion concentrations in rat brain. It also decreases the elevation of intracellular sodium and the potassium depletion in the brain of rats subjected to subliminal electric shocks. In vitro, phenytoin activates sodium/potassium ATP-ase in synaptosomes, but only at high sodium/potassium ratios in the medium. Phenytoin is bound to messenger RNA and may affect protein synthesis by the microsomes and, at high concentrations, there is a decrease in the activity of oxidative enzymes.

170

Drug interactions with inhibitory amino acids General

conclusions

Anti-epileptic drugs have diverse chemical structures and pharmacological actions. In general they fall into two classes, those which are effective in grand mal epilepsy and those which are effective in petit mal. Many but not all of them potentiate the action of GABA and among those that do so the mechanisms involved in the potentiation also seem to differ. There is no single mechanism to explain the action of all anti-epileptic drugs and for some there is no evidence to give even an indication of their likely mode of action.

10 Drugs used in schizophrenia

There are two major type of mental illness, psychoneurotic and psychotic. Anxiety states are examples of the former and are considered in Chapter 9. Although they may be at times incapacitating, they are generally less severe and less enduring than the psychotic states. Psychotic mental illness is generally thought to be due to an organic brain lesion or disorder. However, the presumed defects have proved to be remarkably difficult to characterise, although some headway has been made in the last few years. There are difficulties in the diagnoses of the different psychotic conditions, although this has improved in recent times with the introduction of antipsychotic drugs. The drugs themselves are not disease specific and treat the symptoms of the illness rather than the illness itself. This has led to some confusion in nomenclature and diagnosis. Three major categories comprising personality disorders, schizophrenia and the affective disorders can be distinguished. There may be overlap of symptoms between these states. Personality disorders tend to have a long history dating back to childhood. The development of maladaptive behaviour, seen by society as behaviour that is inappropriate to its own sense of values, frequently brings the sufferer into conflict with society. Since treatment tends to be social, behavioural and psychotherapeutic, rather than pharmacological, except to treat specific temporary psychotic symptoms, these disorders will not be considered further. It will be convenient to consider schizophrenia and the affective disorders separately, since the drugs used are quite different. However, the overlap of symptoms between these states

172

Drugs used in sch izophren ia

may make a differential diagnosis quite tenuous in a particular patient, and it should be emphasised again that treatment for an individual is based upon the symptoms and not the sometimes necessarily somewhat arbitrary diagnosis. Schizophrenia Schizophrenia means a splitting of the personality from reality. The more literal translation of 'split or dual-personality' has led to the inappropriate, inaccurate and often counter-productive Jekyll and Hyde concept of schizophrenia which is characteristic of some rare forms of hysteria. Schizophrenia tends most often to strike its first blow between the ages of 18 and 25. From this fact the old name of'dementia praecox' originated. However, a first attack can sometimes occur up to age 60. The breakdown is often rapid and catastrophic with marked disorientation, loss of contact with reality, disorders of thought processes and abnormal introspection. Hallucinations, particularly auditory hallucinations (Voices'), and delusions, for example of grandeur or persecution (paranoia), are extremely common and if they persist in the absence of other attributable causes are quite diagnostic. The illness tends to be episodic and the frequency of relapse varies considerably from one individual to another. However, schizophrenic symptoms may be evoked by other causes which must first be eliminated before a clear diagnosis can be made. The amphetamines in particular tend to cause behavioural changes which are difficult to distinguish from schizophrenia. A diagnosis of schizophrenia is usually delayed for a period of at least six months after the first apparent onset of symptoms. The incidence of schizophrenia is about 1 in 100 population and represents the most frequent single cause of severe mental illness. About 25% of those diagnosed as schizophrenic make a fairly complete recovery. Another 25% remain very severely disabled, often for the remainder of their lives. About half make some recovery but retain some degree of impairment, with a variable severity and frequency of relapse. The majority require medication for very long periods. The florid symptoms tend to be

Theories of schizophrenia

173

most marked in the early stages of the illness but in the absence of recovery may progress to the more disabling negative signs of poverty of speech and behavioural withdrawal. Drug treatment is more effective against the positive, more florid symptoms. Theories of schizophrenia The cause of schizophrenia is not known. This has not prevented the development of a number of postulates, ranging from purely behavioural interpretations to theories based upon postulated organic changes of a biochemical, functional or structural nature. Over the last decade or so, knowledge of the mechanisms involved in the actions of anti-schizophrenic drugs has expanded rapidly and a number of new techniques have been developed for visualising small structural and chemical changes in the brain. Correspondingly, there has been a greater readiness to accept the idea that schizophrenia is caused by an identifiable abnormality in brain function or structure. At the present time the illness must now be accepted as organic in origin. It is no longer reasonable to consider it as a 'functional psychosis' (Tyrer & Mackay, 1986). Many of the arguments used to justify ideas rampant in the 1960's that the disorder was purely of psychological origin and attributable to 'schizophrenogenic' parents can now be adequately explained by the clearly defined genetic link in the pattern of occurrence (Hirsch, 1979). The dopamine hypothesis The dopamine theory takes its roots from the discovery that anti-schizophrenic drugs are all antagonists at dopamine receptors in brain. This is not of itself sufficient evidence that there is a defect in the dopamine system in the disorder. Attempts to show changes in dopamine levels or in the levels of metabolites in brain or cerebrospinal fluid have not yielded consistent results. A more promising observation was that there was an increase in the density of dopamine receptors in the brains of patients who died with schizophrenia. Most patients are also treated with

174

Drugs used in schizophrenia

effective antischizophrenic drugs and these drugs themselves are well known to elevate the receptor density. The left amygdala contains a selectively high dopamine but not noradrenaline concentration whereas the right amygdala and other parts of the brain are normal in postmortem brains from schizophrenics. On the reasonable assumption that any effect of the neuroleptic drugs is bilateral, it seems likely that there is indeed an asymmetric disturbance in the function of the dopamine system in schizophrenia. An increase in dopamine receptor density seemed to have been substantiated (Wong et ah, 1986) with the new technique of positron emission tomography (PET) scanning for visualising dopamine in the living brain. The observations were made on the caudate nucleus, after first blocking a proportion of the receptors with haloperidol. There was a two and a half-fold increase in the calculated receptor density in the brains of two groups of schizophrenic patients compared with the density in normal volunteers. Since a small group of the schizophrenic patients were claimed to have had no history of medication with psychotropic drugs, it seems unlikely that the increase was due to the action of the medication itself. Two other groups (Crawley et a/., 1986 and Farde et aL, 1987) have independently failed to confirm these results. The findings of one group showed only a slight increase in density whereas in the other study there was no significant increase either in the receptor density (BMAX) or the affinity (KA). There are three disquieting factors about the first study: i. a number of assumptions were needed in order to reach a conclusion, ii. the data show an unexplained difference in the serum levels of haloperidol in the schizophrenic patients compared with the controls and iii. experiments in which receptors were not blocked with haloperidol failed to reveal a difference between the normal and the schizophrenic brain because, it is said, that blood flow was more rate limiting when the number of unblocked receptors is very large. Perhaps an explanation will be forthcoming. Certainly, the dopamine hypothesis has been the most enduring, and the most resurrected, of all theories of schizophrenia. If this is a reflection of truth then the dopamine theory must be correct. If it

Theories of schizophrenia

175

is not then we are left with no plausible theory to explain schizophrenia. Peptide hypothesis Beta-endorphin is an endogenous ligand for opiate receptors. A possible metabolite of beta-endorphin is des-tyrosine-gamma-endorphin, which has been claimed to produce behavioural changes in experimental animals and to alleviate schizophrenic symptoms in some patients. Unfortunately, these findings have not been substantiated. There are reports that the concentrations of met-enkephalin, cholecystokinin (CCK), vasoactive intestinal polypeptide (VIP), somatostatin and substance P are abnormal in various parts of the brain of patients dying with schizophrenia: some are raised and others lowered. However, these reports clearly need to be substantiated and it cannot really be decided whether the changes are secondary to structural or other changes which have nowr been clearly demonstrated. Structural hypothesis Computer aided tomographic (CAT) scans have revealed clear evidence of small but highly significant structural changes in the cerebral ventricular size of the brains of schizophrenic patients. More detailed measurements with other neuroanatomical techniques reveal a large change in size of the ventricles localised to the left temporal lobe coupled with a significantly thinner cortex in the parahippocampus. The genetic hypothesis It is now clearly established by studies in monozygotic twins that there is an inherited predisposition to schizophrenia. The fact that the concordance rate is not 100%, but is between 38 and 58%, has been taken to indicate that there may be an environmental element. Crow has hypothesised that an alternative explanation of the lack of concordance may be an infection by a retro-virus particle at a very early stage in life or even transmitted in the genes which may influence laterality in the adult in

176

Drugs used in schizophrenia

such a way as to lead to schizophrenia. Although attractive, this hypothesis has received little direct experimental support as yet. Nevertheless, the viral hypothesis could explain the relationship between the season in which individuals are born and the incidence of schizophrenia, in which the highest incidence occurs in those born in the winter months in either global hemisphere. Miscellaneous hypotheses Since there is so much uncertainty concerning the causes of schizophrenia it is not surprising that many theories have been advanced on the basis of very little or even no evidence. Thus those interested in the subject will come across references to inorganic ions such as zinc being associated with schizophrenia. There are also many who put their faith in such causes as 'poor dief or even specific dietary factors such as vitamins or gluten. Placebo effects are evident even in serious mental illness and it is all too easy to find anecdotal or other insubstantial evidence to 'support' the flimsiest theories. The critical evaluator will be ultra cautious without necessarily closing his mind to interesting possibilities. It should be remembered that belief in the curative properties of any treatment for mental illness is likely to be almost as important as the pharmacological effect itself. Finally, there has been some recent interest in ligands for the 5-HT3 receptor as possible treatments for schizophrenia. At the time of writing, only abstracts of papers have been published and these purport to show behavioural effects of the drugs in experimental animals which differ from those of the dopamine receptor antagonists: it has only just been demonstrated that the binding sites occur in brain and it has yet to be shown that these are in fact receptors. It remains to be demonstrated whether these substances have any therapeutic value.

Drugs used in schizophrenia The first drug with a really useful and relatively selective therapeutic action in schizophrenia was chlorpromazine, introduced in the mid-1950s. Previously, the only treatments available were restraint in the form of the 'straight-jackef or the padded

Drugs used in schizophrenia

177

cell, sedation, convulsant or insulin-shock therapy, electroconvulsive therapy (ECT), or surgical methods. The last of these involved surgical removal, or partial removal of the frontal lobes on the dubious premise that these structures could be dispensed with without severe deleterious effects. In fact, there were severe changes in personality and the practice has virtually disappeared and can be considered to be quite unethical. The introduction of chlorpromazine started a new era in the treatment and rehabilitation of patients with schizophrenic mental illness. Many patients who previously could only be treated in hospital can now often live reasonably independent lives in the community, despite their residual disabilities and low resistance to stress. The new emphasis on sheltered housing for the mentally ill provides an alternative to the 'institutionalisation' of long stays in hospital but cannot be considered to be the solution for all severely disabled sufferers, many of whom are discharged into an ill-prepared community as institutions are closed down. Chlorpromazine, a phenothiazine derivative, is still frequently used, although other phenothiazines and drugs from other chemical classes have now been introduced (Fig. 10.1). A recent advance has been the introduction of 'depot' preparations of agents such as flupenthixol which have a prolonged action for several weeks after a single injection. This is of enormous benefit to those who otherwise require oral administration several times a day indefinitely, and who frequently tend to discontinue medication as they undergo a relapse. Chlorpromazine and the other drugs used in schizophrenia all produce side-effects of varying incidence and severity (Fig. 10.2). With some agents the sedative action is more pronounced than it is with others. For example, in equi-antipsychotic doses, pimozide tends to have less sedative action than chlorpromazine. However, the sedative action tends to be most noticeable early in treatment although it may well remain incapacitating at a dose adequate to control schizophrenic symptoms. There may also be hypotensive effects with accompanying orthostatic hypotension. Particularly noticeable is the frequent occurrence of druginduced motor disorders (extra-pyramidal symptoms) which are discussed later. Liver dysfunction occurs in a few patients and

178

Drugs used in schizophrenia

Phenothiazines

Chlorpromazine

I

|j I CH 2 CH 2 CH 2 N(CH 3 ) 2

Thioridazine

> A

N^

\ ^ SCH3

CH 2 CH 2 NT I CH 3

Butyrophenones

Haloperidol

O II CCH 2 CH 2 CH,-N

OH

Diphenylbutylpiperidines

Pimozide

CHCH^CH.CH^-N

O

Fig. 10.1. Examples of antipsychotic drugs.

179

The dopamine receptor Thioxanthenes

Flupenthixol CHCH2CH2 — N

N—CH2CH2OH

Dibenzazepines

o

Clozapine

Cl

CH,

Benzamides Sulpiride

NH2SO2 /

\ - CONHCH2 OCH3

—L NNJ I

C2H5

dermatological problems of contact dermatitis and photosensitivity are sometimes seen. Blood dyscrasias, including agranulocytosis and leucopenia, are encountered rarely. Chlorpromazine causes a marked hypothermia and the drug has been used to produce 'artificial hibernation' which reduces metabolic demands for oxygen in major heart surgery. This action was known and used clinically before its use as an antipsychotic drug. Chlorpromazine markedly potentiates the action of other depressants of the central nervous system. The simultaneous intake of alcohol, for example, could have unexpectedly dramatic results. This action is mainly due to an action on the CNS itself, but is also partly the consequence of inhibition of the cytochrome P-450 system in the liver, reducing the enzymic degradation of drugs.

180

Drugs used in sch izophrenia

Table 10.1. Severity of major side-effects of antischizophrenic drugs at effective antipsychotic doses Extrapyramidal symptoms

Drug

Sedative effect

Hypotensive effect

Chlorpromazine Thioridazine Triflupromazine Prochlorperazine Haloperidol Clozapine Pimozide

Fig. 10.2. Metabolism of dopamine. OH

OCH3 \

XV

COMT -CH,CH : NH : — • -

HO

CH2CH2NH2

Dopamine

OCH3 CHXTOOH Homovanillic acid

The dopamine receptor Our present knowledge of the dopamine receptor has derived almost entirely from the interest in the mode of action of the antischizophrenic drugs. Effects of antischizophrenic drugs on terminals One of the earliest pieces of evidence is that the administration of drugs which are effective in schizophrenia increase the turnover of dopamine in the CNS, as revealed by an increased production of the metabolite, homovanillic acid (Fig. 10.2). Sub-

The dopamine receptor

181

stances with a similar chemical structure but which are not effective in schizophrenia lack this effect. Although some of the neuroleptics also increase the turnover of noradenaline, not all do so and this effect may be more related to the incidence or severity of sedation. Most of the studies have been carried out in the caudate nucleus which derives its dopaminergic innervation from the substantia nigra. Two pieces of evidence indicate that the increase in turnover is due to an effect directly on the terminals and is not due to a change in a feedback loop as originally proposed (Fig. 10.4). First, the injection of kainic acid into the caudate nucleus, which destroys nerve cell bodies but not terminals, does not reduce the ability of antischizophrenic drugs to increase dopamine turnover. Secondly, the increased turnover is not abolished by acute transection of the nigro-striatal pathway. Antagonism of behavioural action of dopamine or dopamine agonists injected directly into the brain There have been many studies showing that apomorphine or amphetamine, both of which probably act upon the dopamine system, or dopamine itself causes turning to one side or stereotypy when injected into the cerebral ventricles of rats with unilateral nigro-striatal lesions. These effects are all readily blocked by systemic administration of antischizophrenic drugs. Antagonism of dopamine agonist effects upon single neurons While the acute effect of antischizophrenic drugs is clearly to reduce the inhibition of dopamine-sensitive neurones by the agonist, Bunney (1984) has emphasised his view that the chronic administration of antischizophrenic drugs causes a depolarisation blockade of dopamine sensitive cells. He claims that it is this effect which is correlated with the slow onset of the therapeutic action. Increased receptor density Prolonged administration of antischizophrenic agents causes an increase in dopamine receptor density (Bmax) in brain.

182

Drugs used in schizophrenia Negative feedback

o A

-
Antipsychotics block DA receptor here

Fig. 10.3. Original proposal that antischizophrenic drugs increased dopamine turnover by decreasing a negative feedback. This proposal is no longer tenable. Fig. 10.4. Proposals for locations of dopamine receptors. See also Tables 10.3 and 10.4. 1982

o

Cortex

Nigro-striatal pathway Substantia nigra D3

D2A

i

i

D2,D3 Dl

Dl,]

Striatum Striato-nigral pathway

1985

O D2A

Cortex

Substantia nigra D2,D1

Multiple receptors for dopamine

183

Table 10.2. Lack of correlation between ability of antischizophrenic drugs to block dopamine-sensitive adenylate cyclase and therapeutic efficacy.

a-flupenthixol chlorpromazine spiroperidol (a butyrophenone)

adenylate cyclase inhibition

therapeutic efficacy

+++ + -f4-

++ + + ++

This could be an alternative explanation for the delayed onset of therapeutic action. Blockade of dopamine-activated adenylate cyclase Adenylate cyclase from various brain regions to which dopaminergic fibres project, such as caudate nucleus, olfactory tubercle and nucleus accumbens, is specifically activated by dopamine or apomorphine. This activation is blocked by all potent antischizophrenic drugs and there is some degree of correlation between the therapeutic and biochemical action. Thus, of the two isomers of flupenthixol, the alpha isomer is most potent clinically and blocks adenylate cyclase, whereas the beta isomer has low clinical effectiveness and has little effect on the enzyme. The correlation is not absolute. This can now be attributed to the presence of multiple receptors for dopamine. The block of dopamine activation of adenylate cyclase does correlate rather well with the overall extent of specific binding of different ligands to brain homogenates and thus also correlates rather poorly with clinical efficacy. Multiple receptors for dopamine Perhaps the first indication that there was not just a single dopamine receptor was the lack of correlation between clinical efficacy and block of dopamine activated adenylate cyclase referred to in the preceding section. This is emphasised by the

184

Drugs used in schizophrenia

data shown in Table 10.2. It is evident that the butyrophenone, spiroperidol, is more efficacious than would be expected from its ability to block the cyclase. In keeping with this observation, there is a very good correlation between the clinical efficacy and the ability of a particular dopamine agonist to displace radiolabelled haloperidol from its binding site in brain. In 1978 it was suggested that there were two receptors for dopamine and that this explained the discrepancies which had been noted up to that time. By 1982 (Table 10.3) dopamine receptors were multiplying at a furious pace, with as many as four postulated and no end in sight. The classification was based upon the binding affinities of a variety of agonists and antagonists and their modification by GTP, together with a variety of other tests including, surgical lesions to interrupt selected pathways, kainic acid injections to destroy nerve cells bodies near the site of injection, 6hydroxydopamine-induced lesions to destroy dopamine nerve terminals. Two features were most noticeable. The Dl receptor seemed to be associated with adenylate cyclase activation by dopamine working at relatively low affinity and a low affinity of butyrophenones for the receptor. The D2 receptor seemed to be characterised by dopamine inhibition of adenylate cyclase at nM concentrations and a high affinity of butyrophenones for the receptor. The evidence for D3 and especially D4 seemed weak. The deduced location of receptors Dl to D3 is shown in Fig. 10.4. A few years later, some new experiments rationalised the position. At the present time there seem to be just two receptors, Dl and D2. However, each of these has two different affinity configurations corresponding to the old D3 and D4 receptors. It will be seen that there are selective agonists and antagonists for the two receptor types and that there is still no clear function for the Dl type in relation either to normal function in brain or to the action of the antischizophrenic drugs (but see Kebabian etal, 1986). Part of the original problem was the interpretation of experiments with 6-hydroxydopamine (6-OHDA). It was shown subsequently that reserpine caused a receptor loss similar to that

Multiple receptors for dopamine

185

Table 10.3. Original classification of dopamine receptors - 1982 Receptor

Dl

Adenylate cyclase activation Adenylate cyclase inhibition Guanine nucleotide sensitivity Agonists: Dopamine Apomorphine Bromocryptine Antagonists: Phenothiazines Butyrophenones Thioxanthenes Sulpiride Location:

Yes No Yes

No Yes Yes

No No Yes 7 No 7

uM

nM

uM nM

a) Presynaptic

D2

D3

Partial Full (nM) Full (nM) Full (nM) Weak No Strong Strong Strong uM

Strong 7 nM Weak Strong 7 Weak (D4?)

nM

Strong Strong No Weak

Cortico-striatal Nigro-striatal Striato-nigral

No No Yes

Yes No No

No Yes No

Striatal Substantia nigra

Yes No

Yes Yes

Yes Yes

b) Postsynaptic

Tests used: 1. Surgical lesions 2. Kainic acid lesions 3. 6-Hydroxydopamine lesions 4. Affinity of receptors for butyrophenones and agonists 5. Effect on adenylate cyclase 6. Regulation of agonist affinity by GTP note 1: Despite weak antagonism sulpiride binds with high affinity to brain and is anti-psychotic. note 2: This classification has been superceded by that shown in Table 10.4.

note 3: After Creese 1982. Trends in NeuroscL 5, 40.

186

Drugs used in schizophrenia

Table 10.4. Classification of dopamine receptors - 1985 Receptor Old Class

R

H

"

high D3

(Dl+Ns)

Dl

D2

Dl and D3 Increased c-AMP GTP GDP

D2 and D4 Decreased c-AMP GTP GDP R

2

2

Ca 7Mg -

L

R

H

agonist affinity low high

R

Ca27Mg2"

L

low

D4

(D2+Ni)

Note 1: Antagonists have equal affinities for low and high affinity states Note 2: Ni & Ns are guanine nucleotide binding regulatory proteins. Dl receptors

Location and function:

1. Postsynaptic to DA terminals. Apart from increased c-AMP production, function unknown.

D2 receptors 1. Decreased c-AMP in pituitary, decreased release of prolactin, a-MSH and possible regulation of Ca channels. 2. Decreased ACh release in striatum, leading to increased levels. 3. Decreased DA release and modifies turnover in striatum. (presynaptic autoreceptors). 4. Excites striatal neurones. 5. Decreased firing of DA neurones in substantia nigra. Agonists:

Relative selectivities

Selective Dl to selective D2: SK&F-38393 > > apomorphine = dopamine > bromocryptine > pergolide >> quinpirole

Antagonists:

Selective Dl to selective D2: SCH-23390 > > a-flupenthixol = fluphenazine = chlorpromazine > haloperidol > pimozide > spiperone > domperidone > (-)sulpiride

Adapted from Trends in Neurosci., 6, centrefold.

Extrapyramidal side-effects of antischizophrenic drugs

187

caused by 6-OHDA, even though the former does not destroy the nerve terminals. Furthermore, after destruction of the dopamine terminals with 6-OHDA, the administration of dopamine restored the 'missing' receptors. It became clear that the loss of receptors after 6-OHDA could not be used as evidence that the receptors were presynaptic, as assumed before. These receptors were thought to be high affinity Dl receptors located on the striatal neurones. It seems that the expression of dopamine receptors on the postsynaptic membrane is controlled by the amount of dopamine released in transmission. If this is reduced by lesions, produced surgically or by 6-OHDA, or by depletion of transmitter with reserpine then the postsynaptic receptor density is seen to decline. Elevating the level of dopamine by administration restores the prelesioned condition. Extrapyramidal side-effects of antischizophrenic drugs Disorders of movement known as dyskinesias are frequently observed in patients treated with antischizophrenic drugs. The dyskinesias are of several types. Least troublesome are akathisia, a need for continual movement as if agitated and dystonia, characterised by grimacing and torticollis. More frequent are Parkinsonian symptoms of akinesia and rigidity which tend to be most noticeable during the early stages of treatment and can be well controlled by drugs, especially the anticholinergic drugs, which are used in Parkinson's disease. The tremors are more troublesome and not so readily controlled. All Parkinsonian symptoms diminish rapidly if the antischizophrenic drug is either withdrawn or the dosage is reduced. More disturbing for the patient is tardive dyskinesia, observed in about 20% of patients being treated with drugs over long periods. Another, more descriptive, name for this side-effect is orofacial dyskinesia in which there are excessive and highly distressing movements of the lips, tongue and jaw. The symptoms seem to appear only after prolonged drug treatment and do not always disappear as soon as the drug is withdrawn. The occurrence of dyskinesias in schizophrenic patients was observed before the

188

Drugs used in schizophrenia

introduction of the modern drugs and it is possible that there is a particular tendency to tardive dyskinesia in such patients. Unfortunately, these symptoms do not respond favourably to the anticholinergic drugs and may be made worse by them. Thus the effective treatment of the Parkinsonian side-effects may worsen the tardive dyskinesia. The symptoms may subside when the drug is withdrawn. Clearly, this will not always be possible. Mechanisms in drug-induced dyskinesias If tardive dyskinesia is exacerbated by the antischizophrenic drugs, as seems probable, then the mechanism is obscure. Since the symptoms only occur after prolonged treatment the mechanism is presumed to be something which also follows a long time course. Although not quite on the right time scale, the proliferation of dopamine receptors after chronic treatment with neuropleptic drugs offers a possible but by no means certain explanation. There is a convincing explanation for the Parkinsonian-like dyskinesias. The cause of Parkinson's disease is well known to be the loss of dopamine neurones and of their terminals in the basal ganglia. The dopamine pathway involved is the nigro-striatal pathway originating in the substantia nigra. It is the loss of function in the basal ganglia which gives rise to the term extrapyramidal effects, meaning utilising those nervous pathways outside the pyramidal system. Agents which block dopamine receptors in the basal ganglia have the same ultimate effect as the destruction of the dopamine neurones in basal ganglia disease. Although all useful antischizophrenic drugs bind to the Dl receptor to some degree, we have noted that there is not a good correlation with the incidence of extrapyramidal effects. Nevertheless, a reasonably satisfying explanation has been adduced from the relative ability of different antischizophrenic drugs to block both dopamine receptors (stimulation of adenylate cyclase) and muscarinic receptors for acetylcholine, so effectively combining both cause of the problem and its cure within the same molecule. Table 10.5 summarises some data on the affinity of antischizo-

Extrapyramidal side-effects of antischizophrenic drugs

189

Table 10.5. Dissociation constants and relative affinities of some antipsychotic drugs for binding to muscarinic and dopamine receptors in brain homogenates. (After Miller & Hiley 1974) Muscarinic (M) Dopaminergic (D) Relative affinity K D nM KD nM 25 Thioridazine 55 Clozapine 160 Pimozide 350 Chlorpromazine 12000 Spiroperidol 4000 Trifluoperazine 2200 a-flupenthixol

130 170 140 48 95 19 1

5.2 3.1

0.87 0.14 0.008 0.005 0.0005

Note: Binding to dopamine receptors was measured by activation of adenylate cyclase. Relative affinity is the ratio of the affinity (l/KD) for muscarinic receptors compared with the affinity for dopamine receptors.

phrenic drugs to the Dl receptor, measured by inhibition of dopamine-activated adenylate cyclase and for muscarinic receptors. The relative affinity is the ratio of the affinity (l/^ o ) for muscarinic receptors compared with the affinity for dopamine receptors. The drugs in the table are arranged in descending order of relative affinity for muscarinic/dopamine receptors. It was observed that this order correlates well with the relative incidence of extrapyramidal effects with this series of drugs. Those drugs near the top of the table, such as thioridazine, clozapine and pimozide have a much lower incidence of extrapyramidal effects than have those at the bottom of the table. Dopamine turnover, measured by the production of homovanillic acid has confirmed that the extrapyramidal effects correlate with the anticholinergic properties of the molecule. However, the significance of the anticholinergic action depends upon the region of the brain under study. Table 10.6 shows in summary some data obtained in the caudate nucleus and nucleus accumbens, a part of the limbic forebrain. The number of arrows indicate the relative effects of each of the three compounds studied. Thioridazine, which causes

190

Drugs used in schizophrenia

Table 10.6. Effect of three antipsychotic drugs on dopamine turnover in two regions of the CNS in relation to their abilities to cause extrapyramidal symptoms. Motor side effects

Dopamine turnover (HVA production)

Thioridazine Chlorpromazine Fluphenazine

No anticholinergic present

Anticholinergic present

Caudate Ace

Caudate Ace

0 + +•

0 0 0

+++ +++ +++

Slight Moderate Severe

-I- = increase in DA turnover, Caudate = caudate N., Ace = N. Accumbens Based on data from Crow et aU 1976. The Lancet, 11 Sept, 563.

the least extrapyramidal action and has the highest ratio of anticholinergic to antidopaminergic effect, had the least effect on dopamine turnover in the caudate nucleus. By contrast, fluphenazine had the greatest effect on dopamine turnover in caudate, had the smallest ratio of affinities and had the highest incidence of motor effects. Chlorpromazine came in the middle. In nucleus accumbens all three agents had about the same effect on dopamine turnover. Of particular interest is the finding that an anticholinergic drug abolished all effects on dopamine turnover in the caudate but had no similar effect in the nucleus accumbens. It may be concluded from this study that the cholinergic system is less important in counteracting the dopamine system in the nucleus accumbens than in the basal ganglia. A hypothesis collating these various studies is presented in Fig. 10.5. The cholinergic system may act as a physiological balance to the dopamine system originating from the substantia nigra in the basal ganglia but not in the limbic system. When the control of the basal ganglia is disturbed by blocking the dopamine recep-

Summary

191

Fig. 10.5. Hypothesis relating the action of drugs in the basal ganglia and limbic system to antischizophrenic action and the production of motor disorders.

Block of DA receptors and reduction of DA release is responsible for antipsychotic effect

Block of DA receptors causes extrapyramidal effects

Cholinergic pathway

Block of ACh receptors balances block of DA receptors and reduces extrapyramidal effects

tors, then motor dyskinesias become evident. If there is a counterbalancing block of the cholinergic system, then the dyskinesia are less evident. In the limbic system it is hypothesised that the unopposed block of dopamine receptors is necessary for the antipsychotic action to be in evidence. The penalty for requiring substances to be antagonists at muscarinic receptors and at dopamine receptors is that such compounds may be prone to cause peripheral symptoms attributable to block of muscarinic receptors. Such common symptoms include dryness of the mouth, blurring of vision and constipation. (i)

(ii)

Summary There is increasing evidence that schizophrenia is associated with observable organic brain changes, including structural changes and an increase in the density of D2 receptors for dopamine. Drugs which are effective in treating the symptoms of schizophrenia are most effective on the positive signs

192

(iii)

(iv)

(v)

(vi)

(vii)

Drugs used in schizophrenia

and less effective on negative signs such as withdrawal. There is a good correlation between clinical efficacy and the affinity of the drugs for dopamine D2 receptors where they are antagonists to the transmitter, dopamine: the role of the Dl receptors is uncertain. The presumed target of action is in the limbic system but their is little direct evidence in support of this concept. There is a slowly developing increase in dopamine receptor density. This may explain the discrepancy between the time course of the onset of the therapeutic effect, which is slow, and the pharmacological action which is immediate. Alternatively, it is also possible that in part the slow onset of therapeutic action is due to slowly developing psychological readjustement, although this seems less likely. The Parkinsonian-like side-effects of antischizophrenic drugs are also related to the ability to block dopamine receptors in the basal ganglia, mitigated by their ability to also block muscarinic receptors at the same site. Anticholinergic drugs are therefore an effective countermeasure. Tardive dyskinesia is of slow onset. The mechanism can only be speculated upon and their is no effective countermeasure, except a reduction in dose, which is not always 100% effective. New compounds may be developed in which the primary action is not on dopamine receptors but on some other type of receptor (e.g. the 5-HT receptor) but the clinical efficacy of such drugs has yet to be established.

11 Affective and manic depression

The affective disorders include a variety of conditions characterised by mood changes unrelated to life events, i.e. they are not reactive. There is major depression, sometimes referred to as psychotic or endogenous depression, mania and bipolar or manic depression. Any extreme of mood may be associated with psychosis in which thinking may become irrational and delusional. Since drug therapy should be based upon symptomatology and not on diagnosis this should not cause problems. However, it has sometimes been rather difficult to differentiate clearly manic depression from schizophrenia and the boundaries may merge. In typical cases the distinction is clear. There may also be a tendency for the nature of the illness to change over the years and the diagnosis may change accordingly. Until the major mental illnesses can be characterised completely in terms of specific disorders in structure or function, diagnosis will need to remain linked to symptoms and treatment. A recent promising beginning is the association of a genetic abnormality with the illness.

Endogenous depression Drugs used (Fig. 11.1) in treating major depression include tricyclic compounds like imipramine, tetracyclics like mianserin and monoamine oxidase inhibitors, such as nialamide, which are no longer used to any great degree. Prior to the use of these substances the only available procedures included leptazol or insulin shock therapy and electroconvulsive therapy (ECT). Of these only ECT survives today, although the use of ECT varies greatly from one centre to another. All major theories assert that effective procedures modulate

194

Affective and manic depression

aminergic mechanisms in some way. However, the diversity of action of therapeutic measures defeats most attempts at creating a universal theory to explain the actions of all antidepressant drugs. If it is difficult to explain satisfactorily the mechanism in antidepressant drug action, it is even more difficult to define the aberrant mechanisms in the illness itself. In many studies there have been changes reported in levels of noradrenaline, dopamine or 5-hydroxytryptamine, or their metabolites, in urine, blood or cerebro-spinal fluid. Such changes have been seen in the manic and depressive stages of manic depression, for example. While such alterations may indeed reflect a changes in a monoaminergic system in depression, they do not tell us whether the changes cause the illness or are simply a reflection of its presence. Monoamine oxidase inhibitors The mood-elevating effect of iproniazid was first noted in 1951 when it was introduced for the treatment of tuberculosis. In 1952 it was shown to inhibit monoamine oxidase (MAO) but it was not until five years later that it was tried as an antidepressant. Many drugs with similar actions have now been investigated but severe toxicity, principally on the liver, has caused them to be withdrawn from regular use, although they may still be occasionally employed when all other measures fail. The inhibition of MAO-A prevents amines such as tyramine from being deaminated and thereby detoxified. An agent such as Deprenyl, which affects mainly MAO-B may be more free from such sideeffects but its value in depression remains to be established. The elevation of mood in depressed patients may take some weeks to become apparent. It is tempting to attribute the therapeutic action to inhibition of MAO, which controls the levels of monoamines in the cytoplasm. However, the correlation between ability to inhibit MAO and clinical efficacy is not good and the inhibition of MAO is immediate whereas the clinical improvement may take weeks.

Endogenous depression

195

Fig. 11.1. Structures of antidepressant drugs. Monoamine oxidase inhibitors O C-NHNHCH(CH3)2 CH — C H N H 2

"cH2 Tranylcypromine

Iproniazid (prototype)

CH 2 CH 2 NHNH 2

II O

II O

Nialamide

Phenelzine

Tricyclic

CH2NHCCH2NHNHC

compounds

CH 2 CH 2 CH 2 N(CH 3 ) 2 Imipramine

CH 2 CH 2 CH 2 NHCH 3 Desipramine

CHCH 2 CH 2 N(CH 3 ) 2 Amitriptyline Tetracyclic compounds

CH3 CH 2 CH 2 CH 2 N;

CH3 Mianserin

Iprindole

196

Affective and manic depression

Tricyclic antidepressants The prototype of these drugs is imipramine, which was originally developed as a possible histamine antagonist where its sedative action was noted. In 1958 it was first tested clinically as a sedative in agitated psychotic patients but was not very impressive. Quite coincidentally, it was found to be an antidepressant. The major metabolite of imipramine is desipramine, which was thought to be more active than the parent compound: this seems not to be the case but both are similarly active. Although the MAO inhibitors cause an elevation of mood in normal subjects, the tricyclic antidepressants tend to cause sedation which is accompanied by unpleasant subjective sensations. In depressed patients, imipramine causes less outright euphoria than MAO inhibitors but causes a greater attenuation of the depressive ideas. It is also interesting that the sedative action in normal individuals is rapid in onset with imipramine but the antidepressant action is slow to develop. Other classes of antidepressants There are a variety of other agents which seem to have some use in treating depression. This miscellany includes tetracyclic drugs like mianserin (Fig. 11.1) and other substances such as iprindole and trazodone. Mechanisms of antidepressant action All effective agents seem to interfere in some way or other with monoaminergic systems in the central nervous system. However, the passage of time and more numerous investigations has shown just how complex and varied these actions are. Instead of simplifying the picture it is perhaps becoming more obscure. The difficulties are compounded by the fact that most actions which are observed by pharmacologists are rapid in onset whereas the clinical improvement is very slow over weeks from the commencement of treatment. It would seem rational to concentrate upon only the long-term effects of the drugs if it were not for the fact that long-term effects may only be the long term, and

Mechanisms of antidepressant action

197

Table 11.1. Effects of some antidepressants on uptake ofnoradrenaline and 5-hydroxytryptamine into nerve terminals. Block of uptake (Relative potencies, imipramine = 1)

Desmethylimipramine Amitriptyline Iprindole Mianserin Imipramine KD (M)

NAdr

5-HT

20 18 0 9 1 X 10"6

0.2 0.6 0.3 0.0 2.8 X 10"7

secondary, changes produced by the immediate but maintained short-term action. Inevitably, this means that all actions must at this stage be considered as possibly important mechanisms. In the early days, when the only compounds available were MAO inhibitors and tricyclic antidepressants it seemed reasonable to propose that there was too little monoamine transmitter in depression. This was based upon the fact that either blocking MAO or decreasing amine uptake with tricyclic compounds was beneficial. At that time, although the tricyclic compounds blocked the uptake of both noradrenaline and 5-hydroxytryptamine, the alleviation of clinical depression correlated better with the ability to block the uptake of 5-hydroxytryptamine than with the block of noradrenaline uptake. Other antidepressant compounds with different chemical structures did not fit into this picture at all well. Thus mianserin and iprindole have little effect on uptake mechanisms for 5hydroxytryptamine or noradrenaline respectively (Table 11.2) and yet are effective antidepressants. The picture became even more blurred when it was seen that some antidepressants had a strong affinity for the histamine H-2 receptor. For example, amitriptyline is about seven times as potent as imipramine in this

198

Affective and manic depression

Table 11.2. Effects of some antidepressants as antagonists at various receptors. Receptor block

Desmethylimipramine Amitriptyline Imipramine Iprindole Mianserin Imipramine KD (M)

H2

ACh (Muse)

7 1.0 0.07

0.33 6 1.0 0.04

2.4 X 10"7

2 X 10"7

Pre-a-2

Active Inactive Inactive Active Inactive

Note: blank entries signify unknown result

respect and has a KD\n the nM range. Amitriptyline is also a more effective antagonist than imipramine at muscarinic receptors for acetylcholine. Again these actions may be purely coincidental because substances like iprindole have low activity. The ability of some antidepressants, this time including mianserin, to block presynaptic a-2 receptors for catecholamines at first sight seems promising. On closer investigation this seems less likely because iprindole and imipramine are not effective. Furthermore, the block of the alpha receptor was deduced from experiments in which cocaine was used to block the confusing effect of uptake 1. There is now evidence that cocaine blocks a new 5-HT receptor (5-HT-3) which is located on neurones and mianserin blocks yet another variety of 5-HT receptor (5-HT-2), also found on neurones. Nevertheless, several antidepressants displace the binding of clonidine, an a-2 agonist (Table 11.2), indicating that there is an affinity for these receptors. Suffice it to say that this complex picture is difficult to resolve. Long-term effects of antidepressants There are several long term-effects noted with antidepressant or ECT treatment, including an up-regulation of a-2

Mechanisms

of antidepressant

T a b l e 11.3. Effects of antidepressants cerebral cortex (3-1 Imipramine Clomipramine Desipramine Nortriptyline Mianserin Nialamide Nisoxetine

— — 0 - 0

action

199

on P-7 and a-2 receptor density in

a-2 +++ + ++ + +++ ++ 0

Notes: 1. + = Increase, - = Decrease, 0 = No effect on receptor density 2. p-receptors assessed by dihydroalprenolol binding 3. a-receptors assessed by clonidine binding 4. in limbic forebrain desipramine increased a- but did not decrease Preceptor density 5. nisoxetine decreased P-activated adenylate cyclase without an effect on receptor density

receptors and a down-regulation of P-l-receptors (Table 11.3). At the present time such changes are generally considered to be of vital importance in the generation of the antidepressant action (Tyrer & Marsden, 1985). A variety of effects have been observed with chronic antidepressant treatment. (i) The clinical efficacy of the tricyclic antidepressants correlates well with their ability to displace radio-labelled imipramine from a high affinity (KD = 4 nM) binding sites (Langer & Briley, 1981). Such sites have been identified in postmortem human brain, as well as in the brains of experimental animals. The high affinity binding of imipramine is depressed by about 50% in the platelets of depressed patients. The amount of binding does not correlate with the severity of the symptoms, nor does it change when the patients are treated with antidepressants. The binding site is probably some part of the 5-hydroxytryptamine uptake site on nerve terminals. ECT also decreases imipramine binding but the

200

(ii)

(iii) (iv)

(v)

(vi)

Affective and man ic depression

atypical antidepressants such as iprindole have no effect. It is possible that there is a defect in these sites in the brain of depressed patients, although this has yet to be established. Repeated injections of tricyclic antidepressants over two to three weeks decreases the density of a-1 noradrenergic receptors in the brain, with no change in their affinity, i.e. there is a reduction in Bmax with no change in KD. ECT has a similar effect. This effect is common to many but not all of the newer antidepressants (Table 11.3). One exception is mianserin, which has no significant action on the beta receptors. Repeated administration of antidepressants, including mianserin and nialamide, an MAO inhibitor, causes an up-regulation of a-2 receptors in brain. Although most of the antidepressants decrease the beta receptors and increase the number of a-2 receptors in cerebral cortex, in limbic forebrain desmethylimipramine decreases the beta receptors without affecting alpha receptors. This indicates that the regulation of the two receptor populations is not identical. There is also evidence that chronic treatment with tricyclic compounds, iprindole or pargyline, an MAO inhibitor, decreases the density of receptors for 5hydroxytryptamine in the brain. Chronic treatment with tricyclic antidepressants, iprindole or ECT causes a subsensitivity of a (3-1 activated adenylate cyclase in rat limbic forebrain. While this is often associated with a decreased density of adrenergic receptors, this is not invariably so: nixoxetine decreases the sensitivity without a change in receptor density.

Conclusions It is indeed difficult to draw firm conclusions from these observations. Antidepressant activity seems to be associated with many different types of acute interaction with monoaminergic

Mechanisms of antidepressant action

201

systems coupled with a down regulation of beta receptors and an up regulation of alpha receptors with chronic treatment or ECT. It is possible that the long-term effects on receptor density are regulated by the short-term effects on a variety of amine transmitter systems. As for the biochemical disorder in depression, we seem to be little further down the road than we were 15 years ago. The emerging re-emphasis on multiple receptor types for 5hydroxytryptamine perhaps will shed some new lights on this picture of confusion. However, with about four different types proposed, there may first need to be some rationalisation, as with other systems such as the dopamine system. In the meantime there is no single pharmacological action which will reliably predict the success of a new compound as an antidepressant and reliance still needs to be placed upon the clinical trial. Manic depression The pathology of manic depression is no more established than it is for depression. It is an interesting fact that one of the commonest elements in the earth's crust, lithium, has therapeutic value in the treatment of manic depression. Its first use in modern psychiatry was in 1950 but serious interest did not develop until about 1965. Lithium did not come into general use until 1971, possibly because its low cost did not really make it a very attractive commercial proposition. Lithium has no indication in the treatment of depression but does come into its own in the prophylactic therapy of manic depression. There is very little effect in the first 7-10 days of treatment. The plasma level must be carefully monitored to ensure that it does not rise above about 1.0 mEq/1, or 1.5 mEq/1 in extreme cases. If the plasma level rises higher then toxic effects are observed. Lithium is a simple monovalent cation, slightly larger than sodium ions in the hydrated state. It is not handled effectively by the sodium pump but can diffuse slowly across membranes. The resulting slow accumulation of sodium in cells leads to a progressive depolarisation. Because lithium is so slow to act, and does not cause sedation, it is not effective in initial management, especially in the manic

202

Affective and manic depression

phase of the illness. It is therefore frequently combined with relative large doses of phenothiazines to quieten the patient. The reported actions of lithium include: (i) stimulation of noradrenaline turnover (ii) inhibition of 5-hydroxytryptamine turnover (iii) stimulation of 5-hydroxytryptamine synthesis (iv) inhibition of stimulus induced amine release (v) inhibition or stimulation of amine uptake (vi) changes in the concentration of monoamine precursors and metabolites in CSF (vii) block of choline uptake in erythrocytes (viii) excitation of spinal Renshaw neurones by release of acetylcholine (ix) inhibition of the sodium pump (x) facilitation of synaptic transmission in the hippocampus (xi) increase in evoked potentials in animals and man There seems to be general consensus that the primary action of lithium may be the inhibition of the sodium pump and that all other effects are the secondary effect of this, with the apparent selectivity on different cells being the consequence of differing levels of dependence on the pump. Interesting though these effects may be, they do not tell us a great deal about the way in which lithium exerts its undoubted effect in manic depression.

12 Disorders associated with brain lesions If the cause of an illness is known then it should be possible to devise a therapy to combat that illness. However, if the cause is an anatomical lesion of the brain then it may be possible to devise a rational therapy to treat the symptoms but at first sight it would seem to be impossible to devise a cure because the neuronal loss is irreversible. There are two ways in which the effects of losing neurones might be overcome. One would be to train parallel neuronal circuits to become more efficient: this is the basis of some types of physiotherapy. Alternatively, it may in the future become possible to transplant nerve cells: already this seems to be possible, at least experimentally, with dopaminesecreting cells. Although the major neuronal deficit in Parkinson's disease and effective therapy have been established for many years, attention has more recently shifted first to another basal ganglia disorder: Huntington's chorea, and to the debilitating illness of Alzheimer's disease (senile dementia). In each of these illnesses the major impact has come from studies of changes in neurotransmitter systems, carried out in the hope that specific changes which might be amenable to drug therapy could be devised on a rational basis. Very recently there has also been a new interest in the cell loss which results from the hypoxia and ischaemia of stroke. It will be convenient to group together five disorders of the basal ganglia since these are all characterised by dyskinesias (disorders of movement) of various types.

204

Disorders associated with defined brain lesions

Spasticity Spasticity is muscle rigidity of central origin, as compared with various muscle spasms of peripheral origin such as those occurring in arthritis, migraine or trauma where treatment would be quite different. Spasticity of CNS origin, unlike spasms of peripheral origin, is rarely accompanied by pain. The brain lesions accompanying spasticity are varied and generally not discreetly localised. The extrapyramidal symptoms displayed are correspondingly varied. Typically, stretch reflexes are exaggerated, e.g. in the Babinski sign. There seems in general to be a loss of descending control mechanisms mediated by extrapyramidal pathways with some involvement of pyramidal tracts. Spasticity is often aggravated by anxiety and some of the drugs used with moderate success are anxiety-reducing agents such as diazepam. However, it is emphasised that there is as yet no really effective treatment. Baclofen is another drug for which some success has been claimed, especially in conditions associated with spinal cord lesions such as multiple sclerosis. The chemical name of this substance is p-chlorophenyl-GABA. It is a derivative of the inhibitory amino acid neurotransmitter, GABA. Like GABA, baclofen has inhibitory effects on neurones. However, unlike GABA, baclofen does not increase the conductance of the membrane to chloride ions, nor does it depolarise primary afferent terminals in the spinal cord. Nevertheless, there are specific binding sites to which both baclofen and GABA bind with high affinity. In sympathetic ganglia, in which it is well known that the transmitter is acetylcholine, it has been shown that baclofen and GABA inhibit transmission by decreasing the release of acetylcholine. This effect is not blocked by the competitive GABA-antagonist, bicuculline: it therefore does not interact with what is now termed the GABA-A receptor. The GABA-B receptor for which both GABA and baclofen have affinity is probably associated with a reduction in the permeability of the membrane to calcium ions which are necessary for transmitter release. However, it is also possible that an increase in the conductance of the terminal membrane to potassium ions decreases the probability

Parkinson fs disease

205

of invasion of the nerve terminals by an incoming impulse down the axon. Other drugs which have been used in spasticity include mephanesin, meprobamate and carisoprodal but baclofen and diazepam have made the use of these substances obsolete. Dantrolene is also used but acts peripherally by interfering with the release of calcium from the endoplasmic reticulum of skeletal muscle. Wilson's disease Wilson's disease is also known as hepato-lenticular degeneration because it is accompanied by definable lesions in the lenticular nucleus (globus pallidus and putamen) and in the liver, often presenting as chronic or viral hepatitis, sometimes with cirrhosis. It is a fairly rare familial disorder in which there is deficiency of ceruloplasmin, a copper-carrying protein, in the plasma. Copper is deposited in the basal ganglia and liver, and if untreated leads to permanent brain damage. Restriction of copper in the diet has no benefit but the administration of penicillamine, a copper chelating agent, descended from the original useful drug dimercaprol (British Anti-Lewisite): dimercaprol was developed to counteract arsenical poisoning and was found to be effective but toxic in Wilson's disease. If treatment with penicillamine is instituted before major lesions have been induced then there is hope that further deterioration can be prevented and physiotherapy may produce some recovery. Wilson's disease usually presents between the first and third decades of life. If it occurs early the main symptom is rigidity. Later, tremors, athetosis (swinging of the limbs between flexion and extension) or chorea (sudden, random, coordinated but involuntary flinging movements of the extremities). Parkinson's disease The modern approaches to pharmacological treatment of Parkinson's disease represent one of the few major advances made by a rational development of drugs designed to produce specific effects to correct a known disorder of function based

206

Disorders associated with defined brain lesions

upon an identified pathological lesion. Prior to the advent of Ldihydroxyphenylalanine (L-DOPA), the drugs used were based upon serendipity as in most other areas of therapy. The disease was first described by James Parkinson in 1817, who described patients suffering from what he called the 'shaking palsy' or paralysis agitans, what we now call Parkinson's disease. The treatment which he recommended was letting of blood from the neck followed by the application of vesicants and liniments to cause a purulent discharge. The illness can be considered to be of three types: (i) symptomatic; (ii) postencephalitic; (iii) idiopathic. The symptoms are similar but not necessarily identical in all three types. They consist of tremors, which tend to disappear during volitional movement, unlike the intentional tremors of cerebellar disease. There is also poverty or absence of movement (brady- or akinesia), including difficulties in initiating or arresting any one or a sequence of movements, mask-like facial expression, impairment of handwriting (micrographia), stooped, stiff postures, impaired speech and therefore difficulties in communication and 'cogwheel' rigidity. Mental faculties are generally not impaired. 'Charcot's triad' is the combination of akinesia, rigidity and tremor which are diagnostic. The tremor is relatively fast with a frequency of 3-4 per second with alternating contractions of flexor and extensor muscles, indicating a disturbance of fine motor control. (i) Symptomatic Parkinson's disease may follow injury to the CNS caused by trauma, senile arteriosclerosis, carbon monoxide poisoning, or manganese or other metallic poisoning. The lesions produced by these means may be irreversible. In contrast, Parkinsonian symptoms produced by antipsychotic drugs like the phenothiazines or butyrophenones are reversible upon withdrawal of the drug or reducing the dose. (ii) Postencephalitic Parkinson's disease was first observed as a consequence of an epidemic in 1916-17 of a disease of unknown etiology called encephalitis lethargica (sleeping sickness, Von Economo's disease) and presumed to be a virus. The first case appeared in Vienna in the winter of 1916-17, had spread to England by 1918 and to the USA, the following year. There have been

Parkinson's disease

207

no later resurgences of this illness. The neurological signs of the basal ganglia disorder occurred some years later, although they were seen, atypically, in some young children. (iii) Ideopathic Parkinson's disease is of unknown origin, rarely occurs before the age of 40 and reaches a peak between the ages of 50 and 60 years. The estimated incidence varies from about 1 : 40 to 1 : 500 and is currently at a plateau. Genetic factors are not clearly defined although males are slightly more susceptible than females. There is usually a progressive deterioration over a 10-15 year time span from the onset of symptoms and life expectancy is not greatly increased by drug treatment, even though it is often very effective in relieving the most distressing symptoms. In 1982, classical symptoms of Parkinson's disease suddenly appeared in a group of young drug addicts in California. This was traced to N-methyl-4-phenyl-1,2,3,6,-tetrahydropyridine (MPTP), a toxic substance accidentally produced as a 'designer drug' during the illegal preparation of'synthetic heroin'. It is possible, but not yet proven, that MPTP or a substance like it is present in the environment in small quantities and could be the causal agent of spontaneously occurring Parkinson's disease. The drug is absorbed by inhalation or skin contact. It is probable that MPTP is metabolised by MAO-B in brain to l-methyl-4phenylpyridinium, which may be the final toxic agent since inhibition to MAO-B reduces the toxicity. If such treatment is effective in preventing or reducing degeneration in Parkinsonian patients, then it will be a major step forward in the prevention of progressive deterioration, rather than just reducing symptoms with no effect on the underlying pathology, there is an increased susceptibility to MPTP with age, which may be related to increased activity of MAO in the aging process. MPTP is probably selectively taken up into dopaminergic neurones since these are selectively destroyed, just as in Parkinson's disease. MPTP has a number of pharmacological actions including depletion of dopamine, which may not be the cause of cell death, a block of the uptake of dopamine and of 5-HT and a reduction of dopamine release.

208

Disorders associated with defined brain lesions Drug

treatment

A number of drugs are available for treatment of Parkinson's disease (Fig. 12.1) but L-DOPA reigns supreme with bromocriptine coming a close second. The mechanisms can be separated into several different types: Antagonism at muscarinic receptors for acetylcholine. Belladonna

alkaloids (atropine) were first used by Charcot in 1892. Synthetic anticholinergic compounds include benztropine, ethopropazine, trihexyphenidyl and procyclidine. These are sometimes used in conjunction with L-DOPA but may have some adverse effect on the absorption of the latter. The anticholinergic compounds are the major drugs used in treating the Parkinsonian-like sideeffects of the antischizophrenic drugs. The action of the anticholinergic compounds is usually attributed to restoring the balance of cholinergic and dopaminergic transmission in the basal ganglia. However, these substances also block the uptake of dopamine in synaptosomes and benztropine, at least, causes the release of newly synthesised dopamine and the stimulant action in mice is blocked by a-methyl-/?-tyrosine, which blocks dopamine synthesis, but not by an acute lesion of the nigro-striatal tract (i.e. terminals are still intact). 'Replacement therapy': L-DOPA. The history of the introduction of L-DOPA is one of the few examples of the rational development of a new drug. In 1957 it was known that treatment of experimental animals with reserpine, which depletes nerve terminals of their catecholamine content, replicated some of the symptoms of Parkinson's disease. Especially noticeable in mice or rats treated with reserpine is the tremor, rigidity and akinesia. In 1957 it was shown that this action was reduced by the administration of DOPA. Two years later it was shown that dopamine is highly concentrated in the basal ganglia in normal human postmortem brain. It was in 1961 that dopamine was found to be low in the basal ganglia of patients dying with Parkinson's disease. Over the next five or six years DL-DOPA was tried in moderate doses with only slight improvement and numerous sideeffects. In 1967, large doses of L-DOPA (3-8 g by mouth) were

Parkinson's disease

209

employed. This gave a remarkable improvement in the akinesia and rigidity with lesser effect on the tremor. In 1971 it was shown that the entire dopaminergic projection to the basal ganglia arose from the substantia nigra. It is thought that the effect of administering large quantities of L-DOPA is to increase the amount of dopamine synthesised by those terminals which have not degenerated. Nausea was experienced in about 80% of the patients treated with L-DOPA, with vomiting and anorexia in a smaller proportion. However, these symptoms diminished with continuation of treatment. Involuntary movements of athetosis and chorea were the most disabling of all symptoms and occurred in about 60% of patients. The commonest reason for withdrawal of L-DOPA was the occurrence of psychic side-effects in about 15% of patients. It has been suggested (Paalzow & Paalzow, 1986) that under some circumstances treatment with L-DOPA may exacerbate the Parkinsonian symptoms and that fcon-off effects of L-DOPA during the course of the day may, in part, be related to pharmacokinetic considerations. Prevention of breakdown of L-DOPA. Carbidopa is N-(DL-seryl)-N-

(2,3,4,trihydroxybenzyl)hydrazine. It is an inhibitor of DOPA decarboxylase peripherally, but it is unable to enter the central nervous system. This prolongs the action of L-DOPA by reducing its inactivation. It is ineffective by itself but it enables smaller doses of L-DOPA to be used, with resulting lower blood levels and therefore fewer gastro-intestinal side-effects. However, the unwanted effects on the central nervous system, especially the involuntary movements, are unchanged. Deprenyl is a selective inhibitor of monoamine oxidase-B (MAO-B). MAO-B is present in parts of the CNS and it is thought that the effectiveness of deprenyl in Parkinson's disease is due to the prolongation of the action of endogenous dopamine. It is useful because it does not inhibit the peripheral MAO-A to any great extent so that breakdown of indirectly acting sympathomimetic amines such as tyramine may proceed as normally, with little risk of the hypertensive crises that can ensue when this enzyme is inactivated. However, deprenyl increases the

210

Disorders associated with defined brain lesions HO 3, 4-dihydroxyphenyl-L-alanine (L-DOPA)

HO

"A

V C H 2 'CHNH2

I

CH3

Benztropine

Trihexyphenidyl

Diethazine CH2CH2N(C2H5)2

Ethopropazine CH 2 CHN(CH 5 ) 2 I CH3

severity of hon-off effects with L-DOPA, particularly after prolonged treatment when the effect of L-DOPA is declining. If Parkinson's disease is due to a substance such as MPTP in the environment which needs to be broken down by MAO-B into a more toxic metabolite then treatment with deprenyl may reduce further structural deterioration. The answer to this will not be available until long-term studies are completed.

211

Parkinson's disease

Amantadine

Apomorphine

COOH

I

.CH2-C-CH3 NH-NH2 Carbidopa 1 -methyl-4-pheny 1tetrahydropyridine (MPTP)

Deprenyl

N—CH 3

cv

CH, CH 2 CH—N—CH 2 C=CH CH 3

Fig. 12.1. Drugs and Parkinson's disease.

Release of dopamine. Amantadine was introduced as an antiviral agent in 1964. Amantadine causes the release of dopamine from nerve terminals. The release can be prevented in vivo by a lesion of the nigro-striatal tract. Amantadine also blocks the re-uptake of dopamine into terminals, so increasing the effect of any which is released.

212

Disorders associated with defined brain lesions

Agonist binding to dopamine receptors. Apomorphine binds to

dopamine, Dl type, receptors and activates adenylate cyclase. It is effective in Parkinson's disease but hepato- and nephrotoxicity has prevented its widespread use. The binding of 3H-apomorphine in postmortem brain from Parkinsonian patients is reduced by about 60% compared with normal brain. This indicates that the binding sites are largely located presynaptically on the dopaminergic neurons. In contrast, the binding of 3H-haloperidol, a reasonably selective antagonist acting on D2 receptors for dopamine is increased by about 60%. This indicates that the postsynaptic D2 receptors proliferate. However, treatment with L-DOPA prevents the change in the haloperidol binding sites. Bromocriptine is an ergot derivative which has some selectivity as an agonist at D2 receptors for dopamine. It has found considerable use in Parkinson's disease, particularly when used in combination with L-DOPA. It has a longer duration of effect than LDOPA and causes fewer involuntary movements. It has been suggested that the action of bromocriptine is to potentiate the action of dopamine itself, since it atypically does not convert the high to the low affinity state of the receptor (Goldstein et dl. 1985). It was therefore predicted that it would be most effective if combined with low doses of L-DOPA. A disadvantage of bromocriptine is that hallucinations are more frequent with bromocriptine than with L-DOPA. Lisuride and pergolide act in a similar way to bromocriptine. Peripheral side-effects of nausea and vomiting are reduced by the administration of domperidone, a dopamine receptor antagonist which does not cross the blood-brain barrier. A combination of lisuride with L-DOPA has given promising results in causing less 'on-off effects. Lithium has also been found to reduce these unexplained effects, but the mechanism of action of lithium in this respect is unknown. The mechanisms of the effective anti-Parkinsonian agents is summarised in Fig. 12.2

Fig. 12.2. Sites of drug action in Parkinson's disease and Huntington's chorea. Huntington's disease

Parkinson's disease

Caudate Putamen Globus pallidus

Caudate Putamen Globus pallidus

Globus pallidus 5-HT normal GAB A decreased GAD decreased DA increased Ch. Ac. Tr. decreased ACh receptor decreased

Z. compacta Subs tan tia nigra

T-OH increased GABA receptors increased

5-HT ACh normal GABA DA decreased Apomorphine binding decreased Haloperidol binding increased

DA'

DA neurones degenerated

Z. reticulata Drugs used Tetrabenazine Chlorpromazine Haloperidol Also effective; ? Choline; lecithin; physostigmine

Drugs used L-DOPA (with decarboxylase inhibitor) Anticholinergics Amantadine Bromo crip tine ? Also effective but nephrotoxic, apomorphine

I

214

Disorders associated with defined brain lesions

Huntington's disease Huntington's disease, also known as Huntington's chorea or senile chorea is a hereditary disease, with autosomal dominance, of the basal ganglia and the cerebral cortex. A specific antineuronal antibody (Igg) has been found in 50% of patients, in 23% of close relatives and in 3-6% of the normal population. The disease occurs in both sexes of all races and only rarely does it miss a generation. The ancestry of Huntington's chorea in the USA has been traced to three immigrants from England in 1630. In Venezuela, the illness has been traced to a single German sailor who settled in 1860. Choreiform movements and mental deterioration occur in adult life, usually appearing between the ages of 30 and 60 years, and are accompanied by widespread lesions in the brain leading to death in about 15 years. The movements may be increased by emotional disturbance but tend to disappear during sleep. Chorea is increased by treatment with L-DOPA, which may even precipitate symptoms in siblings in whom there was no other evidence of the disease. Biochemical and structural changes Histologically, there is a considerable atrophy of parts of the brain, especially of the basal ganglia. The levels of tyrosine hydroxylase are increased in the striatum and substantia nigra in the brains of patients dying with Huntington's chorea. Dopamine in the caudate nucleus and nucleus accumbens is increased. It is likely that this apparent increase is only relative, due to the substantial loss of other neurones. There has been a similar increase in 5-hydroxytryptamine levels, which can probably also be attributed to the neuronal loss. In contrast, there is marked loss of substance P in the medial pallidum and substantia nigra, indicating the degeneration of a substance P projection from globus pallidus to substantia nigra. Similarly, there is a substantial loss of angiotensin converting enzyme in the striatum and substantia nigra. There is also some evidence for degeneration of a striato-nigral met-enkephalincontaining pathway.

Alzheimer's disease

215

More interest has been shown in the loss of gabergic neurons projecting from the caudate to the substantia nigra. The greatest loss of glutamic acid decarboxylase, the GABA synthesising enzyme, occurs in the zona reticulata, where the striato-nigral gabergic fibres terminate. There is an increase in the affinity for GABA binding, with no change in flmax in the substantia nigra. In some, but not all, patients dying with Huntington's chorea there is a loss of cholinergic receptors and of choline acetyltransferase from the putamen and caudate nucleus. This enzyme synthesises acetylcholine for choline and is a good marker for cholinergic neurones. Since the reduction of receptor is not always accompanied by a decrease in the synthesising enzyme, it has been postulated that postsynaptic lesions may precede presynaptic changes. In summary (Fig. 12.2) there is good evidence for a loss of gabergic neurones projecting from striatum to substantia nigra in Huntington's chorea. There is also a degeneration of some peptidergic pathways and in some, but not all individuals there is a loss of intra-striatal cholinergic neurones. Treatment It is curious that the most used treatment is with agents such as haloperidol or chlorpromazine: it will be recalled that these agents block receptors for dopamine and that the dopamine system is one which is relatively unchanged in Huntington's chorea. This may be seen as an attempt to restore the balance caused by the loss of other systems. Other agents which have been claimed to have been of some benefit include substances acting to increase the activity of the cholinergic system. These include physostigmine, which blocks acetylcholinesterase, choline itself and lecithin, which is a choline precursor. These agents cannot be considered to be established treatments, but are of interest in that they illustrate the attempts to compensate for selective neuronal loss by either increasing activity in a deficient system, or by blocking the activity in another system, the dopamine system, which may be relatively overactive, due to the loss of systems with which it normally acts in harmony.

216

Disorders associated with defined brain lesions

Alzheimer's disease Alzheimer's disease, or dementia of the Alzheimer type, occurs in about one in six persons over the age of sixty. The personality changes and loss of memory are catastrophic and accompanied by characteristic lesions in the brain in cerebral cortex, amygdala and hippocampus. These changes include the formation of tangles and plaques accompanied by neurochemical changes in a number of neurotransmitters. There is no known cause of the disease but there are many hypotheses. There some observations which indicate that the lesions could be produced by an increased sensitivity to an endogenous excitatory amino acid. First, there is a decreased population of receptors for glutamic acid of the N-methyl-D-aspartate (NMDA) type. A number of toxic actions have been demonstrated with toxic derivatives of the endogenous excitatory amino acids. In particular it has been shown that N-methyl-D-aspartic acid applied to the cerebral cortex produces biochemical changes similar to those of Alzheimer's disease and causes lesions of cholinergic neurones in the nucleus basalis. Uniquely, so far at least, glutamate induces the formation of structures in human spinal cord neurons in tissue culture which are similar to the neurofibrillary tangles characteristic of Alzheimer's disease. These findings suggest that substances which block glutaminergic transmission by acting as antagonists at NMDA receptors may prove to be of value in treating Alzheimer's disease. A number of such substances are now known, including the dissociative anaesthetic phencyclidine, but have yet to be evaluated. In the meantime, there is suggestive evidence that the use of cholinesterase inhibitors may improve memory in patients with Alzheimer's disease, even if the effect is only temporary. There are also pointers to the involvement of NMDA receptors in longterm potentiation and memory which may also lead to therapeutic advances in the future. There is also evidence that antagonists at NMDA receptors may be of prophylactic value in preventing cell death due to hypoxia or ischaemia. Such therapy would be of enormous value in stroke victims.

SELECTED READING

Chapter 1 Introduction Feldman, R.S. & Quenzer, L.F. (1984). Fundamentals of Neuropsychopharmacology. Sinauer. New York. Goodman Gilman, A., Goodman, L.S., Rail, T.W. & Murad, F. (1985). The Pharmacological Basis of Therapeutics. Seventh Edition. MacMillan. New York. Lamble, J.W., Ed. (1980) More about Receptors. Elsevier. Amsterdam. Lamble, J.W., & Abbott, A.C. Eds. (1984). Receptors Again. Elsevier, Amsterdam. Rang, H.P. & Dale, M.M. (1987). Pharmacology. Churchill-Livingstone. London. Snyder, S. (1986). Drugs and the Brain. Scientific American Books. New York. Chapter 2 Techniques Dingledine, R. (1983). Brain Slices. Plenum. New York. McBurney, R.N. (1983). New approaches to the study of rapid events underlying neurotransmitter action. Trends in Neurosci., 6, 297-302. Chapter 3 Neuromuscular junction Adams, P.R. (1978). Molecular aspects of synaptic transmission. Trends in NeuroscL, 1, 141-3. Behan, P.O. (1979). The immunology of myaesthenia gravis. Trends in Neurosci., 1, 31-3. Drachman, D.B. (1983). Myaesthenia gravis: immunobiology of a receptor disorder. Trends in Neurosci., 6, 446-51. Giraudat, J. & Changeux, J-P. (1980). The acetylcholine receptor. Trends in Pharmacol. ScL 1, 198-202. Grob, D. (Ed). (1976). Myasthenia Gravis. Ann. NYAcad ScL 274, 1682. Katz, B. (1966). Nerve Muscle and Synapse. New York: McGraw-Hill. Katz, B. & Miledi, R. (1972). The statistical nature of the acetylcholine potential and its molecular components. /. Physiol, 230, 665-99. Katz, B. & Miledi, R. (1973). The characteristics of'endplate noise' produced by different depolarising drugs. /. Physiol, 230, 707-17. Neher, E. & Sakmann, B. (1976). Single channel currents recorded from

218

Selected reading

membrane of denervated frog muscle fibres. Nature (Lond), 260, 799-802. Neher, E. & Stevens, C.F. (1977). Conductance fluctuations and ionic pores in membranes. Ann. Rev. Biophys. Bioengng., 6, 345. Rang, H.P. & Ritter, J.M. (1970). On the mechanism of desensitization at cholinergic receptors. Mol Pharmacol., 6, 383-90. Rang, H.P. & Ritter, J.M. (1970). The relationship between desensitisation and the metaphilic effect at cholinergic receptors. Mol Pharmacol, 6, 383-90. Thesleff, S. (1955). The mode of neuromuscular block caused by acetylcholine, nicotine, decamethonium and succinylcholine. Acta Physiol Scand, 34, 218-31. Thesleff, S. & Sellin, L.C. (1980). Denervation supersensitivity. Trends in Neurosci., 3, 122-4. Wray, D. (1981). Prolonged exposure to acetylcholine: Noise analysis and channel inactivation in cat tenuissimus muscle./ Physiol., 310, 37-56. Chapter 4 Autonomic nervous system Brown, D. (1982). Peptidergic transmission in ganglia. Trends in NeuroscL, 5, 34-5. Eccles, R.M. & Libet, B. (1961). Origin and blockade of the synaptic responses of curarized sympathetic ganglia. 7. Physiol, 157, 484-503. Jan, Y.N. & Jan, L.Y. (1983). A LHRH-like peptidergic neurotransmitter capable of 'action at a distance' in autonomic ganglia. Trends in Neurosci., 6, 320-5. Langer, S.Z. (1977). Presynaptic receptors and their role in the regulation of transmitter release. Br. J. Pharmac, 60, 481-97. Langer, S.Z. (1980). Presynaptic receptors and modulation of neurotransmission: pharmacological implications and therapeutic relevance. Trends in Neuroscl, 3, 110-12. Laverty, R. (1973). The mechanisms of action of some anti-hypertensive drugs. Br. Med. Bull, 29, 152-7. Libet, B. (1977). The role SIF cells play in ganglionic transmission. Ann. Rev. Pharmacol, 9, 135-47. Otsuka, M. & Konishi, S. (1983). Substance P - the first peptide neurotransmitter. Trends in Neuroscl, 5, 317-20. Trendelenburg, U. (1979). The extraneuronal uptake of catecholamines: is it an experimental oddity or a physiological mechanism? Trends in Pharmac. ScL, 1, 4-6. Chapter 5 Central neurotransmitters and neuromodulators Artola, A. & Singer, W. (1987). Longterm potentiation and NMD A receptors in rat visual cortex. Nature (Lond), 330, 649-52. Beart, P.M. (1982). Multiple dopamine receptors-new vistas. Trends in Pharmacol ScL, 2, 100-2.

Selected reading

219

Bradley, P.B. (1987). Pharmacology: 5-HT3 receptors in the brain? Nature (Lond), 330, 696. Collingridge, G. (1987). Synaptic plasticity: The role of NMDA receptors in learning and memory. Nature (Lond), 330, 604. Creese, I. (1982). Dopamine receptors explained. Trends in Neurosci 5 40-3. Creese, I. (1985). Dopamine receptor subtypes. Trends in Pharmacol. ScL 6, centrefold. Davidoff, R.A. (ed.) (1983). Handbook of the Spinal Cord. Marcel Dekker, New York. Hanley, M.R. & Jackson, T. (1987). Substance K receptor: return of the magnificent seven. Nature (Lond), 329, 766-7. Henry, H.L., Couture, R., Cuello, A.C., Pelletier, G., Querion, R. & Regoli, D. (eds.) (1987). Substance P and Neurokinins. SpringerVerlag, New York. Kilpatrick, G.J., Jones, B.J. & Tyers, M.B. (1987). Identification and distribution of 5-HT3 receptors in rat brain using radioligand binding. Nature (Lond), 330, 746-9. Leff, S.E. & Creese, I. (1983). Dopamine receptors re-explained. Trends in Pharmacol ScL, 4, 463-7. Masu, Y., Nakayama, K., Tamaki, H., Harada, Y., Kuno, M. & Nakanishi, S. (1987). cDNA cloning of bovine substance-K receptor through oocyte expression system. Nature (Lond), 329, 836-8. Porter, R. & O'Connor, M. eds. (1982). Substance P in the nervous system. Ciba Foundation Symposium No. 91. Pitman. London. Richardson, B. & Engel, G. (1986). The pharmacology and function of 5-HT3 receptors. Trends in Neurosci., 9, 424-8. Rogawski, M.A. & Barker, J.L. (1985). Neurotransmitter Actions. Plenum. New York. Chapter 6 The blood-brain barrier Bowman, W.C. & Rand, M. (1980). Textbook of Pharmacology. 2nd Edn. Bradbury, M. (1979). The Concept of a Blood-Brain Barrier, John Wiley, Chichester. Bradbury, M. (1979). Why a blood-brain barrier? Trends in Neurosci, 2, 36-8. Davson, H. (1978). The environment of the neurone. Trends in Neurosci., 2, 39-41. Lund-Anoksen, H. (1979). Transport of glucose from blood to brain. Physiol. Rev., 59, 305-52. Chapter 7 General anaesthetics Dundee, J.W. (1971). Comparative analysis of intravenous anaesthetics. Anesthesiology, 35, 137-48. Nicholl, R.A. (1979). Differential postsynaptic effects of barbiturates on chemical transmission. In, Neurobiology of Chemical Transmission.

220

Selected reading

Otsuka, M. & Hall, Z.W. (eds.). John Wiley & Sons. New York. pp. 267-78. Pender, J.W. (1971). Dissociative anaesthesia,/ Am. Med. Ass., 215, 1126-30. Richards, CD. (1980). The mechanism of anaesthesia. In Topical Reviews in Anaesthesia. Norman, J. & Whitwam, J. (eds.). Vol. 1. Wright. Bristol. Richards, CD. & Hesketh, T.R. (1975). Implications for theories of anaesthesia of antagonism between anaesthetic and non-anaesthetic steroids. Nature (Lond), 256, 179-82. Seeman, P. (1972). The membrane actions of anesthetics and tranquillizers. Pharmacol. Rev., 24, 583-655. Study, R.E. & Barker, J.L. (1981). Diazepam and (-)-pentobarbital: Fluctuation analysis reveals different mechanisms for potentiation of GABA responses in cultured central neurons. Proc. Natl. Acad. Sci. USA., 78, 7180-4. Halsey, et al. (1974). Molecular Mechanisms in General Anaesthesia.

Churchill-Livingstone. Edinburgh. Weakly, J.N. (1969). Effect of barbiturates on 'quanta!' synaptic transmission in spinal motoneurones. J. Physiol, 204, 63-77. Chapter 8 Pain and analgesia Chung, Shin-Ho & Dickenson, A. (1980). Pain, enkephalin and acupuncture. Nature (Lond), 283, 243-4. Duggan, A.W. & North, R.A. (1983). Electrophysiology of opioids. Pharmacol. Rev., 35, 219-82.

Basbaum, A.I. & Fields, H.L. (1984). Endogenous pain control systems: brainstem spinal pathways and endorphin circuitry. Ann. Rev. NeuroscL, 7, 309-38. Hughes, J., Ed. (1983). Opioid peptides. Brit. Med Bull, 39, 1-100. Jesell, T.M. & Iversen, L.L. (1977). Opiate analgesics inhibit substance P release form rat trigeminal slices. Nature (Lond), 268, 549-551. Khatchaturian, H., Lewis, M.E., Schafer, M.K.-H. & Watson, S.J. (1985). Anatomy of the CNS opioid systems. Trends in NeuroscL 8, 111-19. Kosterlitz, H.W. (ed.) (1976). Opiates and Endogenous Opioid Peptides.

North Holland. Amsterdam. Kosterlitz, H.W. (1987). Biosynthesis of morphine in the animal kingdom. Nature (Lond), 330, 606. Kosterlitz, H.W. & Paterson, S.J. (1980). Characterization of opioid receptors in nervous tissue. Proc. R. Soc. B., 210, 113-22. Lord, J.A.H., Waterfield, A.A., Hughes, J. & Kosterlitz, H.W. (1977). Endogenous opioid peptides: multiple agonists and receptors. Nature (Lond), 267, 495-9. Mayer, D.J., Price, D.D. & Rafii, A. (1977). Antagonism of acupuncture analgesia in man by the narcotic antagonist naloxone. Brain Res., 121, 368-72. Miller, R. (1978). Enkephalin: a peptide with morphine-like properties. Trends in NeuroscL, 1, 29-31.

Selected reading

221

Milton, A.S. (1976). Modern views on the pathogenesis of fever and the mode of action of antipyretic drugs. /. Pharm. Pharmacol, 28, 393-9. North, R.A. & Williams, J.T. (1983). How do opiates inhibit transmitter release? Trends in Neurosci., 6, 337-9. Pomeranz, B. (1977). Brain opiates at work in acupuncture. New Scientist, 6th June, 12-13. Roemar, D., Buescher, H.H., Hill, R.C., Pless, J., Bauer, W., Cardinaux, F., Closse, A., Hauser, D. & Huguenin, R. (1977). A synthetic enkephalin analogue with prolonged parenteral and oral analgesic activity. Nature (Lond), 268, 547-9. Schwartz, J-C. (1979). Opiate receptors on catecholaminergic neurones in brain. Trends in Neurosci., 2, 137-9. Wall, P.D. & Melzack, R. (1984). Textbook of Pain. ChurchillLivingstone, Edinburgh. Weber, E., Evans, C.J. & Barchas, J.D. (1983). Multiple endogenous ligands for opioid receptors. Trends in Neurosci., 6, 333-6. Chapter 9 Drug interactions with inhibitory amino acids Barnard, E.A., Darlison, M.G. & Seburg, P. (1987). Molecular biology of the GABA-A receptor/channel superfamily. Trends in Neurosci., 10, 502-9. Barker, J.L. and Matthers, D.A. (1980). GAB A receptors and the depressant action of pentobarbital. Trends in Neurosci., 1, 257. Braestrup, C. and Nielsen, M. (1980). Multiple benzodiazepine receptors. Trends in Neurosci., 3, 301-3. Burnham, W.M., Spero, L., Okazaki, M.M. and Madras, B.K. (1981). Saturable binding of 3H-phenytoin to rat brain membrane fraction. Can. J. Physiol. Pharmacol., 59, 402-7. Chiu, T.H. and Rosenberg, H.C. (1983). Multiple conformational states of benzodiazepine receptors. Trends in Pharmacol. Sci., 4., 34850. Fujimoto, M., Hirai, K. & Okabayashi, T. (1982). Comparison of the effects of GABA and chloride ion on the affinities of ligands for the benzodiazepine receptor. Life Sciences, 30, 51-7. Haefely et al. (1981). General pharmacology and neuropharmacology of the benzodiazepines. Handbook Exptl. Pharmacol, 55 (II), 13-262. Karobath, M. (1979). Molecular basis of benzodiazepine actions. Trends in Neurosci, 2, 166-8. Macdonald, R.L. and McLean, M.J. (1982). Cellular bases of barbiturate and phenytoin anticonvulsant drug action. Epilepsia, 23, Suppl. 1, S7-S18. Martin, I.L. (1980). Endogenous ligands for benzodiazepine receptors. Trends in Neurosci., 3, 299-310. Martin, I.L. (1986). The benzodiazepine receptor. In Neuromethods\ eds. G.B. Baker and A.A. Boulton. 4, 12.1-12.31. Humana Press. Turner, A.J. and Whittle, S.R. (1980). Sodium valproate, GABA and epilepsy. Trends in Pharmacol Sci., 2, 257-60. Willow, M. (1986). Pharmacology of diphenylhydantoin and

222

Selected reading

carbamazepine action on voltage-sensitive sodium channels. Trends in Neuroscl, 9, 147-9. Chapter 10 Drugs used in schizophrenia Bissette, G., Nemeroff, C.B. & MacKay, A.V. (1986). Neuropeptides and schizophrenia. Progr. Brain Res., 66, 161-74. Bunney, B.S. (1984). Antipsychotic effects on the electrical activity of dopaminergic neurons. Trends in Neuroscl, 5, 212-15. Crawley, J.C.W., Crow, T.J., Johnstone, E.C., Oldland, S.R.D., Owen, F., Owens, D.G.C., Smith, T., Veall, N. & Zanelli, G.D. (1986). Uptake of 77Br-spiperone in the striata of schizophrenic patients and controls. Nucl Med. Commun., 7, 599-607. Crow, T.J. (1979). What is wrong with dopaminergic transmission in schizophrenia? Trends in Neuroscl, 2, 52-6. Crow, TJ. (1984). A re-evaluation of the viral hypothesis: is psychosis the result of retroviral integration at a site close to the cerebral dominance gene? Br. J. Psychiat., 145, 243-53. Crow, TJ. & Johnstone, E.C. (1978). ECT - Does it work? Trends in Neuroscl, 1, 51-3. Crow, T.J., Johnston, E.C, Deakin, J.W.F. & Longden, A. (1976). Dopamine and schizophrenia. Lancet, Sept. 11, 563. Farde, L., Weisel, F-A, Hall, H., Halldin, C, Stone-Elander, S. & Sedvall, G. (1987). No D2 receptor increase in PET study of schizophrenia. Arch. Gen. Psych., 44, 671-2. Hirsch, S.R. (1979). Do parents cause schizophrenia? Trends in neuroscl, 2, 49-52. Hornykiewicz, O. (1977)vPsychopharmacological implications of dopamine and dopamine antagonists: a critical evaluation of current evidence. Ann. Rev. Pharmacol, Toxicol, 17, 545-59. Iversen, L.L., Iversen, S.D. & Snyder, S.H. (1975). Handbook of Psychopharmacology. Plenum. New York. Kebabian, J.W., Agui, J.C., van Oene, J.C., Shigematsu, K. & Saavedra, J.M. (1986). The DI dopamine receptor: new perspectives. Trends in Pharmacol Scl, 7, 96-9. Mackay, A.V.P. (1984). High dopamine in the left amygdala. Trends in Neuroscl, 5, 107-8. McCann, S.M., Lumpkin, M.D., Mizunuma, H., Khorram, O., Ottlecz, A. & Samson, W.K. (1984). Peptidergic and dopaminergic control of prolactin release. Trends in Neuroscl, 5, 127-31. Miller, RJ. & Hiley, R. (1974). Antimuscarinic properties of neuroleptics and drug-induced Parkinsonism. Nature (Lond), 248, 596-7. Nemeroff, C.B. & Cain, S.T. (1985). Neurotensin-dopamine interactions in the CNS. Trends in Pharmacol. Scl, 6, 201-5. Phillips, A.G., Lane, R.F. & Blaha, CD. (1986). Inhibition of dopamine release by cholecystokinin: relevance to schizophrenia. Trends in Pharmacol. Scl, 7, 126-8. Reynolds, G.P. (1979). Phenylethylamine - a role in mental illness? Trends in Neuroscl, 2, 265-8.

Selected reading

223

Smythies, J.R. (1984). The transmethylation hypothesis of schizophreniz re-evaluated. Trends in NeuroscL, 5, 45-7. Stevens, J.R. (1979). Schizophrenia and dopamine regulation in the mesolimbic system. Trends in NeuroscL, 2, 102-5. Tyrer, P. & Mackay, A. (1986). Schizophrenia: no longer a functional psychosis. Trends in NeuroscL 9, 537-8. Wang, R.Y., White, F.J. & Voigt, M.M. (1984). Cholecystokinin, dopamine and schizophrenia. Trends in Pharmacol Sci., 5, 436-8. Weissman, M.M. (1986). Psychiatric diagnoses. Science, 235, 522. de Wied, D. (1979). Schizophrenia as an inborn error in the degradation of beta-endorphin - a hypothesis. Trends in Neuwsci., 2, 79-82. Wong, D.F., Wagner, H.N., Tune, L.E., Dannals, R.F., Pearlson, G.D., Links, J.M., Taminga, C.A., Broussolle, E.P., Ravert, H.T., Wilson, A.A., Toung, J.K.T., Malat, J., Williams, J.A., OTauma, L.A., Snyder, S.H., Kuhar, MJ. & Gjedde, A. (1986). Positron emission tomography reveals D2 dopamine receptors in drug-naive schizophrenics. Science, 234, 1558-63. Chapter 11 Affective and manic depression Akasura, M. Tsukamoto, T. & Hasegawa, K. (1982). Modulation of rat brain ai and p-adrenergic receptor sensitivity following long term treatment with antidepressants. Brain Res., 235, 192-7. Emrich, H.M., Aldenhoff, J.B. & Lux, H.D. (eds.) (1982). Basic Mechanisms in the Action of Lithium. Excerpta Med. Int. Congr. Ser, 572, 71-

9. Elsevier. Green, R.A. (1978). ECT-How does it work? Trends in Neuwsci., 1, 534. Green, R.A. & Maayani, M. (1977). Tricyclic antidepressant drugs block histamine H2-receptors in brain. Nature (Lond), 269, 163-5. Harper, B. & Hughes, I.E. (1979). Presynaptic a-adrenoceptor blocking properties of tri- and tetracyclic antidepressant drugs. Br. J. Pharmac, 67,511-17. Hodgkinson, S., Sherrington, R. Gurling, H.M.D. et al. (1987). Molecular genetic evidence for heterogeneity in manic depression. Nature (Lond), 325, 805-6. Kimmelberg, H.K. (1986). New antidepressant drugs: is there anything new they can tell us about depression? Trends in Neuwsci., 9, 314. Langer, S.Z. & Briley, M. (1981). High-affinity 3H-imipramine binding: a new biological tool for studies in depression. Trends in Neuwsci., 4, 28-31. Maas, J.W. (1979). Neurotransmitters and depression. Too much, too little or too unstable? Trends in Neuwsci., 2, 306-8. Maj, J. (1981). Antidepressant drugs: will new findings change the present theories of their action? Trends in Pharmacol Sci., 1, 80-3. Sulzer, F. (1979). Newperspectives on the mode of action of antidepressant drugs. Trends in Pharmacol Sci., 1, 92-4. Tyrer, P. & Marsden, C. (1985). New antidepressant drugs: is there anything new they can tell us about depression? Trends in Neuwsci., 8,427-31.

224

Selected reading

Chapter 12 Disorders associated with defined brain lesions Bartolini, G. (1980). Interactions of striatal dopaminergic, cholinergic and GABA-ergic neurons: relation to extrapyramidal function. Trends in Pharmacol ScL 1, 138-40. Bird, E.D. (1978). Huntington's disease (chorea). Trends in Neurosci., 1, 57-9. Bird, E.D. (1980). Chemical pathology of Huntington's disease. Ann. Rev. Pharmacol. ToxicoL 20, 533-51. Goldstein, M., Liberman, A. & Meller, E. (1985). A possible molecular mechanism for the antiparkinsonian action of bromocryptine in combination with levodopa. Trends in Pharmacol Sci., 6, 436-7. Hefti, F. & Melamed, E. (1980). LDOPA's mechanism of action in Parkinson's disease. Trends in NeuroscL, 3, 229-31. Hornykiewicz, O. (1973). Parkinson's disease from brain homogenate to treatment. Fed. Proc, 32, 183-90. Kindt, M.V., -Nicklas, W.J., Sonsalla, P.K. & Heikkila, R.E. (1986). Mitochondria and the neurotoxicity of MPTP. Trends in Pharmacol Sci., 7, 473-5. Langston, J.W. (1985). MPTP and Parkinson's disease. Trends in NeuroscL 8, 79-83. Langston, J.W. (1987). Closer to Parkinson's disease: MTP and the ageing nervous system. Trends in Pharmacol Sci., 8, 7-8. Larsen, T.A. & Calne, D.B. (1982). Recent advances in the study of Parkinson's disease. Trends in Neurosci., 5, 10-12. Lee, T., Seeman, P., Rajput, A., Farley, I.J. & Hornykiewicz, O. (1978). Receptor basis for dopaminergic supersensitivity in Parkinson's disease. Nature (Lond), 273, 59-61. Lloyd, K.G. & Davidson, L. (1979). 3H-GABA binding in brains from Huntington's chorea patients: altered regulation by phospholipids. Science, 205, 1147-9. Maragos, W.F., Greenamyre, J.T., Penney, J.B. & Young, A.B. (1987). Glutamate dysfunction in Alzheimer's disease: an hypothesis. Trends in Neurosci., 10, 65-8. Paalzow, G.H.M. & Paalzow, L.K. (1986). L-DOPA: how it may exacerbate Parkinsonian symptoms. Trends in Pharmacol ScL 7, 15-19. Von Voigtlander, P.F. & Moore, K.E. (1971). Dopamine: Release from the brain in vivo by amantadine. Science, 174, 408-10.

INDEX

A-fibres, 119-20, 123, 126 AH 8165 (fazadinium), 29-30 Acetazolamide, 96 Acetylcholine (ACh), 53, 77 action, 85 in autonomic nervous system, 44, 45 benzodiazepine effect on, 151 blocking agents: depolarising (desensitising), 25, 30, 39-40; mechanisms, 32-8; metaphilic antagonists, 38; non-depolarising (competitive), 25, 29-30, 30-2, 39-40 mobilisation, 27 at neuromuscular junction, 16-17, 18-22 receptor complex, 17, 18 receptors, 82 fig.; activation, 20-2; and antidepressant action, 198; autoantibodies, 41; muscarinic, 13, 18-19, 77; nicotinic, 19, 30 storage and release of, 27-9 structure, 83 Acetylcholinesterase, 17-18, 23-4, 77 Action potentials, 1, 10-11, 16 Acupuncture, 136 Adenylate cyclase, 140, 183 Adrenaline, 64, 76 cardiovascular effects, 64-5 in CNS, 91 mechanism of action, 70 structure, 83 uptake, 73 Affective disorders, 171 Akathisia, drug-induced, 187 Alpha-methyl DOPA, 55 fig., 75, 98 Alpha-receptors, 64, 68-70 blocking agents, 75-6 Alphaxalone, 106, 111fig.,112 Alzheimer's disease (senile dementia), 203, 216

Amantadine, 211fig.,211 Amino acids in CNS,. 85-9; receptors, 82 fig., 87-8; see also y-Ainino-butyric acid, Glycine, Glutamic acid inhibitory, drug reactions with, 144-70

y-Amino-butyric acid (GABA), 85, 88. 115 and benzodiazepines, 152-3, 155-6, 160 compared with baclofen, 204-5 drug interactions with, 144-70 passim

receptors, 82 fig., 88, 89 structure, 83 Amino-oxyacetic acid, 146 Amitriptyline, 195 fig., 197-8 Amphetamines, 71, 172, 180 Amylobarbitone, 107 fig., 11 fig. Analgesia aspirin-like drugs, 122-3, 124 local, 122-3, 124 morphine-like drugs, 128 opiates, 137-9 stress-induced, 137 Angiotensin, 76, 82 fig. Angiotensin-converting enzyme, 76 Antibiotics, 27 Anticholinesterase agents, 23-4, 97 Antidepressants, 193-4 long-term effects, 198-200 mechanisms of action, 196-8 monoamine oxidase inhibitors, 194, 195 fig. other, 196 tricyclics, 196-7 see also individual drugs

Anti-epileptics, 162, 164 clinical applications, 163, 165-6 mechanisms, 166-9 Antihypertensives, 74-6

225

226

Index

Antineuronal antibody (Igg), 214 Antipsychotic drugs, 178-9 fig. Anxiety, 147-8 Anxiety reducing agents, 147-8, 161-2 Apomorphine, 180, 210fig.,212 Aspartic acid, 83fig.,85 Aspirin, 122-3, 124 fig. Athetosis, 205 Atropine, 77, 208 Autonomic nervous system cholinergic transmission, 77 drug action, 46 ganglionic sites of action, 46-54 neurotransmitters, 44-6 skeletomotor nervous system compared, 44 fig. sympathetic nerves, structure and function, 54-6 Autoradiography, 89 Baclofen ((3-chlorophenyl-GABA), 89, 146, 204-5 Barbitone, 96 Barbiturates, 105-10, 108-9 fig., 110, 112, 116-17, 148 brain penetration, 96, 166 Basal ganglia, disorders of, 203-15 Benzodiazepines dependency, 148-9 long-term effects, 149-50 pharmaco kinetics, 150-1 pharmacological actions, 151-3, 166 receptors: binding affinity, 153-4; BZ1 compared to BZ2, 158-9; cellular localisation, 157; endogenous ligands, 157-8; and GABA receptors, 155-6; regional variation, 154-5, 158, 159; substances binding to, 159-60 structures, 149 Benzoquinonium, 30 Benztropine, 208, 210 fig. Benzylamine, 58 Beta-carbolines (p-CCE), 157 Beta-endorphin, 131, 132, 133 fig., 175 Beta-receptors, 64, 69-70 blocking agents, 75-6 Bethanecol, 77 Bethanidine, 74 Bezold-Jarische reflex, 46 Bicuculline, 88-9, 144-5, 146 Biochemical techniques, 12-13 Black widow spider venom, 24, 27 Blood-brain barrier

breakdown of, 97-8 developmental aspects, 98-9 nature of, 8, 93-5 permeability, factors affecting, 95-8 Bombesin, 82 fig. Botulinus toxin, 24, 27-8, 85 Bradykinin, 76, 120-1 Brain, electrical activity, recording, 10 Brain lesions, associated disorders, 203-16

Brain slice techniques, 7, 9 Brainstem, pain control, 124-5 Bretylium, 55fig.,74 Bromocriptine, 208, 211-12 a-Bungarotoxin, 19, 25, 30, 32, 42 p-Bungarotoxin, 27 Buprenorphine, 127 fig., 128 C-fibres, 119-20, 123, 126 Catechol-o-methyltransferase (COMT), 57 fig., 61, 63, 73 Catecholamines, 89-91, 198 cardiovascular effects, 64-5 metabolism, 56-9 receptors, 68-70 uptake and storage, 59-63; neuronal (uptake-1), 59, 60, 61 fig., 71; non-neuronal (uptake-2), 59, 60-2, 63, 72 CL218872 compound, 158, 159, 160 figCapsiacin, 123 Captopril, 76 Carbachol, 21, 31, 39, 77 Carbamazepine, 166 Carbidopa, 210 Cerebrospinal fluid (CSF), 8, 93-4, 96, 136, 137 Ceruloplasmin, 205 Charcofs triad, 206 Chloral hydrate, 162 Chloramphenicol, 97 Chlordiazepoxide (Librium), 148, 149 Chlorgyline, 59 Chloroform, 102 fig., 103-5 Chlorpromazine, 96, 129, 176-9, 178 figand dopamine turnover, 190 fig. in Huntington's chorea, 213 fig., 214 Cholecystokinin (CCK), 44, 82 fig., 175 Choline, 17,26,202,215 Choline acetyltransferase, 26, 84 Cholinergic pathways, in CNS, 84

227

Index Cholinergic system, autoregulation, 70 Cholinergic transmission, in autonomic nervous system, 77 Cholinesterase inhibitors, 78, 216 Chorea, 205 Choroid plexus, 93-4, 96 Chromaffin cells, 45, 76 Clathrates, 110 Clonazepam, 166 Clonidine, 55fig.,67, 70-1, 75

Clostridium botulinum, 27 Clostridiwn tetanic 146

Clozapine, 179 fig., 188-9 Cocaine, 60, 62 fig., 63, 71 Codeine, 126, 127 fig., 129-30 Computer aided tomography (CAT), 13, 175 Convulsants, 144-6 Cyclohexylamine (Ketamine), 106-7 Cyclopropane, 102 fig., 103, 105 Dl-4 (dopamine) receptors, 182 fig., 183-7, 188 Dantrolene, 205 Decamethonium (C10), 25, 30, 33-8, 39 Decamethylene-bistrimethylammonium (C10TMA), 39 Delta (8) receptors, 133-6 Dementia, senile see Alzheimer's disease Denervation supersensitivity, 43, 76 Depolarising (desensitising) blocking agents, 25, 30 mechanisms, 32-8 pharmacological characteristics, 39-40 Deprenyl, 59, 211fig.,211 Depression bipolar (manic), 193, 201-2 major (endogenous, psychotic), 193, 193-201 Desipramine, 195 fig., 196 Desmethylimipramine, 197 fig., 198 fig. Diazepam (Valium), 148, 149 fig., 154 P-CCE displacement of, 157 and benzodiazepine receptors, 158, 159 pharmacological profile, 161 in spasticity, 204, 205 for status epilepticus, 166 Diethazine, 210 fig.

Dihydro-p-erythroidine, 84, 85, 87, 138 Diisopropyl phosphofluoridate (DFP), 78, 97 Dimercaprol, 205 Dimethylphenylpiperazinium, 19 Dinaphthyl decamethonium (DNC10), 25, 38-9 Diphenylbutyl acetate, 24, 26 Diuretics, 76 Dopa decarboxylase, 56, 57fig.,58 Dopamine, 45 amantadine effect on, 212 anticholinergic drugs effect on, 208 level in depression, 194 level in Huntington's chorea, 214 metabolism of, 180 in Parkinson's disease, 212 receptors, 82 fig., 180-7; classification, 185-6; effect of antischizophrenic drugs, 180-7; location, 182 fig., 185-6 figs role in schizophrenia, 173-5 structure, 83 Dopamine-p-hydroxylase, 57fig.,58 Dopaminergic pathways, 89, 90 fig.

Drugs, 1-3, 8-10 see also individual drugs

Dynorphins, 120, 131, 133 fig., 134-5 Dyskinesias, 203, 204, 205, 206, 212 drug-induced, 187-8; mechanisms, 188-91 Dystonia, drug-induced, 187 Ecothiophate, 78 Edrophonium, 25, 41, 78 Electroconvulsive therapy (ECT), 177, 192, 198 Electroencephalogram (EEG), 10 Electromyogram (EMG), 16 Electrophysiological methods, 7, 10-12 Encephalitis lethargica (sleeping sickness, Von Economo's disease), 206 End plate potentials (epps), 16, 17, 18, 26,28 Enkephalins, 44, 82 fig., 131-2, 134-5 Ephedrine, 71 Epilepsy, 162-5, see also Anti-epileptic drugs Ether, 102 fig., 103-5 Ethopropazine, 208 Ethosuximide, 166 Etorphine, 129-30, 135

228

Index

Eugenols, 106 Evoked potentials, 10 Excitatory postsynaptic potential (EPSP), 51, 53, 56 Extracellular fluid (ECF), 94 FK33-824 compound, 132-3, 135 Fazadinium (AH 8165), 29-30 Flunitrazepam, 154, 158, 159 Flupenazine, 190 fig., 190 Flupenthixol, 177, 179 fig., 189 fig. Flurazepam, 148 GABA see y-Amino-butyric acid GABA-A receptor, 82 fig., 88, 89 GABA-B receptor, 82 fig., 88, 89 GABA modulin, 156 GABA transaminase inhibition, 146 Gallamine, 25, 29, 31 Ganglionic sites of action, 43-4, 46-54 blocking agents, 53-4, 55 ionic mechanisms, 53 muscarinic receptors, 48-51 nicotinic receptors, 46-8 peptide involvement in, 51-3 General anaesthetics, 102-17 convulsants related to, 110, 111 fig., 112 effects, 112-16 gaseous, 102-3 mechanisms, 107-12 soluble (intravenous), 105-7 structures of, 101, 102 tolerance to, 116-17 types of, 101-7 volatile, 103-5 see also individual anaesthetics

Germine, 24 Glaucoma, 77, 78 Glutamic acid, 83, 85 receptors, 87-8 Glutamic acid decarboxylase (GAD), 89, 146 Glycine, 82 fig., 83 fig., 85, 88 drug interactions with, 144-70 passim

Guanethidine, 55 fig., 74, 75 Guanidine, 28 Guanosine monophosphate (GMP), 153 Haloperidol, 178, 180fig.,213fig.,215 Halothane, 95-6, 102 fig., 103-5, 112, 113

Heart muscle activation, 68, 70 Hemicholinium (HC-3), 24, 26 Hexafluorenium, 30 Hexamethonium, 55fig.,97 Hexobarbitone, 109 fig. Histamine, 82 fig., 120 Histochemical techniques, 12 Homovanillic acid, 180 fig. Huntington's disease (Huntington's or senile chorea), 202, 214-15 Hydrochlorothiazide, 55 fig. 6-Hydroxydopamine (6-OHDA), 187 5-Hydroxytryptamine (5-HT), 46, 89 antidepressant effect on, 197-8 in CNS, 90fig.,91 in depression, 194 in Huntington's disease, 214 lithium effect on, 202 in nociception, 139 receptors, 82 fig., 176, 198, 201 in schizophrenia, 176 structure, 83 fig. 5-Hydroxytryptophan (5-HTP), 98 Hyoscine, 77 Hyperglycinaemia, non-ketotic, 145-6 Hypertension, 98, see also Antihypertensives Hypnotics, 147-8, 161-2 Hypoxanthine, 157 Imipramine, 192, 195 fig., 196, 197-8 Immunohistochemical methods, 12 Indomethacin, 122, 123, 124 fig. Inosine, 157 Insulin shock therapy, 177, 192 Intra-peritoneal injections, 8 Intracellular recording, 11, 16 Intravenous injections, 8 Ion channels conductance: measurement, 7, 11-12; states, 21-2 Iprindole, 195 fig., 196, 197-8 Iproniazid, 58, 194, 195 fig. Isoprenaline, 62, 63, 64-5, 70-1 Kainate receptor for glutamic acid, 82 fig., 87 Kainic acid, 157, 181 Kanamycin, 27 Kappa (K) receptors, 133-6 Kernicterus, 99 Ketamine (cyclohexylamine), 106, 107 Lathyrism, 99-100 L-dihydroxyphenylalanine

229

Index (L-DOPA), 98, 206, 213 fig. development of, 208-9 effect in Huntington's disease, 212 for Parkinson's disease, 208-12 Labetalol, 76 Lecithin, 215 Leptazol, 192 Leu-enkephalin, 131, 132, 133 fig., 134-5 Lisuride, 212 Lithium, 201-2 Local anaesthetics, 2-3, 120 Luteinising hormone releasing hormone (LHRH), 45, 51 M-current, 53 Malathion, 78 Mania, 193 MAO-A -B, see Monoamine oxidase McNeil A-343 (muscarine agonist), 77 Mu (u) receptors, 132, 133-6 Mecamylamine, 54, 55 fig., 97 Membrane noise, 7, 11-12, 20-1 Meningitis, 97, 98 Meperidine, 127 fig. 3-Mercaptoproprionic acid, 146 Merprobamate, structure, 149 Met-enkephalin, 131, 132, 133 fig., 134-5, 175 Meta-tyramine, 71 Metaphilic antagonists, 29, 38, 38-9 Methacholine (acetyl-pmethylcholine) 18, 19, 77 Methadone, 127 fig., 129 Methohexitone, 106, 109 fig. Methoxyfluorane, 102 fig., 105, 112 Methylatropine, 97 N-methyl-D-aspartic acid (NMDA), 87-8, 215-16 ct-Methylnoradrenaline, 75 Mianserin, 192, 195 fig., 96, 197-8 Microejection, 9 Microelectro-osmosis, 9 Microelectrophoresis, 9, 84 Microinjection, 9 Miniature end plate potentials (mepps), 16-18, 26, 28 Monoamine oxidase (MAO) in catecholamine metabolism, 58 receptors: MAO-A, 5, 210; MAO-B, 59,207,210-11 substrates, 61 Monoamine oxidase inhibitors (MAOIs), 58-9, 61, 73, 192, 194-5 mechanisms of action, 196-7

Monoclonal antibodies, 12, 157 Morphine, 96, 126, 127 fig. binding affinity, 134-5 characteristics, 129-30 endogenous, 132 Morphine-like drugs, 126-8 action, 128-9 addiction, 126, 128, 129 Motor end plate, 15, 16 Motor unit, 15 MPTP (N-methyl-4-phenyl-l, 2, 3, 6, • tetrahydropyridine), 207. 211 fig., 212 Muscarine, 13, 18, 77 Muscarinic receptors, 18-19, 77 Muscle relaxants, 23, 29, 106, 151 Myasthenia gravis, 23-4, 40-2, 97 Myasthenic syndrome, 41-2 N-wave, 46, 47 fig. Nalorphine, 127 fig., 128 Naloxone, 127 fig., 128, 136-7 Neomycin, 24, 27 Neostigmine, 23, 24-5 figs, 41, 78, 97 Nerve fibres, peripheral, 119-20 Nerve gases, 78 Neural transplants, 203 Neurokinins, 82 fig. Neuromuscular junction acetylcholine, receptors: activation, 20-2; sites, 17; types of, 18-19 synaptic transmission, 16-19 drug action: investigations, 16; sites, 22-39: postjunctional, 29-39; prejunctional, 23-9 synaptic transmission, 16-19 Neurones activity, recording, 10-12 general anaesthetic: effects on, 112-16; tolerance to, 116-17 nociceptive, central control of, 124-5 uptake of catecholamines, 71, 59, 60, 61 fig. Neurotensin, 82 fig. Neurotoxicity, 99-100 Neurotoxins, 2-3 Neurotransmitters central: identification of, 81-3; receptors and their significance, 82 k false\ 71-2 see also individual neurotransmitters

Nialamide, 58, 193, 195 fig. Nicotinamide, 157

230

Index

Nicotine, 19 Nicotinic receptors, 19, 30 at ganglia, 46-7 at neuromuscular junction, 19, 30, 47 Nitrazepam (Mogadon), 149 Nitrous oxide, 102-3 Nociception, 118, 118-19, 139 spinal and supraspinal mechanisms, 124-6 Nociceptors, peripheral, 119, 120-2 Non-depolarising (competitive) blocking agents, 25, 29-32, 39-40 Noradrenaline (NA), 44, 45-6, 76, 89, 180 antidepressant effect on, 197-8 cardiovascular effects, 64-5 in the CNS, 89-91 in depression, 194 feedback facilitation, 68 feedback inhibition, 67-8 lithium effect on, 202 mechanism of action, 70-1 metabolism, 56, 57 receptors, 64-8, 82 fig.; characteristics, 65; classification, 65 structure, 83 uptake and storage, 60, 61-3, 73 Noradrenergic pathways, 89, 90 Octopamine, 58 Opiates, 127-9 action sites, 137-9 cellular effects, 139-41 receptors, 129-36; endogenous ligands, 131-2; evidence for, 133-6; localisation, 130-1 tolerance, 141-3 Opioid peptides, 131-7 analgesic effects, 132-3 endogenous involvement in pain, 136-7 multiple receptors, 133-6 Opium, 126 Organophosphates, 78 Oxotremorine, 77 Pain central pathways, 123-6 measurement of, 118, 118-19 peripheral mechanisms, 119-23 Pancuronium, 25, 29 Parathion, 78 Paracetamol, 122, 123, 124 fig.

Parasympathetic nervous system cholinergic transmission, 77 effect of cholinesterase inhibitors on, 78-9 neurotransmitters, 44, 45 Parkinsonian symptoms, druginduced, 188, 189, 206 Parkinson's disease, 2, 203, 205-12 idiopathic, 207 postencephalitic, 206-7 symptomatic, 206 treatment, 98, 208-12, 213 fig. Patch clamping, 11-12, 16, 20, 22 Pempidine, 54, 55 fig. Penicillamine, 205 Penicillin, 97 Pentazocine, 127 fig., 128 Pentobarbitone, 111 fig., 113, 114, 117 Peptide hypothesis of schizophrenia, 175 Pergolide, 212 Personality disorders. 111, see also Affective disorders; Schizophrenia Phaeochromocytoma, 58, 76 Phenacetin, 122, 123, 124 fig. Phencyclidine, 106, 216 Phenelzine, 58, 195 fig. Phenobarbitone, 105, 108 fig., 117, 166 Phenothiazines, 177, 202 Phenoxybenzamine, 63, 64, 66, 67 Phentobarbitone, 107 fig. Phentolamine, 76 Phenylbutazone, 122, 123, 124, fig. Phenylethylamine, 58, 70-1 Phenyltrimethylammonium, 19 Phenytoin, 96, 166, 167-9 Phosphoinositol system, 12 Photoaffinity labelling, 154 Physostigmine (eserine), 41, 78, 97, 215 Picrotoxin, 89, 144-5 Pilocarpine, 77 Pimozide, 177, 178 fig., 180 fig., 188-9 Pirenzepine, 77 PK9084 compound, 160, 161 Positron emission tomography (PET), 13, 174 Postsynaptic membrane, neuromuscular, depolarisation, 17, 20-1 Pressure microejection, 9 Primidone, 166 Pro-dynorphin, 131, 133 fig. Pro-enkephalin-A, 131, 133 fig.

231

Index Pro-opiomelanocortin, 131, 133 fig. Prochlorperazine, 180 fig. Procyclidine, 208 Propanidid, 106 Propranolol, 55fig.,75 Prostaglandins (PG), aspirin inhibition of, 82 fig., 121, 122, 129 Psychoneurotic illness, 111, see also Anxiety Psychotic illness, 171 Pyridostigmine, 78 Quinolines, 160, 161 Quisqualate receptor for glutamic acid, 82fig.,87 Renin-angiotensin system, 75-6 Renshaw cells, 84, 85, 86-7 fig., 138 Reserpine, 55 fig., 60, 62 fig., 71, 73, 186 RO-15-1788 compound, 160, 161 Salbutamol, 64, 68, 70-1 Schizophrenia, 2, 171, 172-3 dopamine receptors: classification, 185-6; comparison of different, 180 fig., 183-7; effect of antischizophrenic drugs on, 180-7; location, 182 fig., 185-6 figs dopamine theory of, 173-5 drug treatment, 176-92; side-effects, 177, 179, 181, 187-91 non-drug treatment, 176-7 summary, 191-2 theories of, 173-6 Sedatives, 147-8, 161-2 Senile chorea see Huntington's disease Senile dementia see Alzheimer's disease Skeletal muscle contraction of, 17 electrical activity, measurement, 16 innervation of, 14-15 see also Neuromuscular junction Skeletomotor nervous system, 44 Sleep, 91 Smooth muscle activation, 64, 68 innervation, 56 stimulation, 57 Sodium barbitone, 105, 108 fig. Somatostatin, 45, 82 fig., 120, 175

Spasticity, 204-5 Spinal cord lesions, 204 Spiroperidol, 189 fig. Steroidal anaesthetics, 106, 111 fig., 112 Sterotypy, 130, 180 fig. Streptomycin, 24, 27 Stroke, 88, 203, 216 Subcutaneous injections, 8 Substance P, 44, 45, 82 fig. capsiacin effect on, 123 in ganglionic transmission, 51-2 in Huntington's chorea, 214 in pain fibres, 120, 121 in schizophrenics, 175 structure, 83 Succinylcholine, 25, 30, 33-8 Sympathetic nerves, structure and function, 54-6 Sympathetic nervous system cholinergic transmission, 77 neurotransmitters, 44-6 Sympathomimetic amines, 70-1, see also individual agents

Synapses, cholinergic, on Renshaw cells, 84, 85, 86-7 fig. Synaptic delay, 18 Synaptic transmission, neuromuscular, 14, 16-18 Taurine, 85 Tetanus toxin, 144-5, 146 Tetracyclics, 192 Tetracycline, 97 Tetraethyl pyrophosphate, 78 Tetraethylammonium (TEA), 27 Tetrodotoxin, 24 Thiohexitone, 106, 109 fig. Thiopentone, 95, 96, 106, 109 fig., 117 Thioridazine, 178 fig., 180 fig., 188-9 Tissue culture, 7 Torpedo, electroplaque organ of, 19 Tranylcypromine, 195 fig. Trazodone, 196 Tremorine, 97 Triazolopyridazines, 158, 159, 160 fig. Trichloroethylene, 112 Tricyclics, 96, 192, 195 fig., 197 Triethylcholine, 26-7 Trifluoperazine, 189 fig. Triflupromazine, 180 fig. Trihexyphenidyl, 208, 210 fig. Trimethadione, 166 Trizolopyridazines, 161 Trypan blue, 93. 96

232

Index

d-Tubocurarine (d-Tc), 25, 29, 31, 32 Tyramine, 58, 71, 73 Tyrosine hydroxylase (TOH), 56-8 Valproate, 166, 166-7 Vasoactive intestinal polypeptide

(VIP), 44, 82 fig., 175 Voltage clamping, 11, 16, 21 Wilson's disease (hepato-lenticular degeneration), 205