International Review of Neurobiology Volume 26

INTERNATIONAL RNlDN OF

Neurobiology VOLUME 26

Editorial Board

W. Koss ADEY JULIUS AXELROD

SEYMOUR KETY

Ross BALDESSAKINI

CONAN KORNETSKY

SIR ROGERBANNISTER

ABELLA-JTHA

FLOYDBLOOM

BORISLEREDEV

DANIELBOVET

PAULMANDELL

PHILLIPBRADLEY

HUMPHRY OSMOND RODOLFOPAOLETTI SOLOMON SNYDER STEPHENSZARA SIRJOHN VANE

JosB DELCADO SIRJOHN ECCLES JOEL

ELKES

H.J. EYSENCK KJELLFuxe Bo HOLMSTEDT PAULJANSSEN

KEITH KILLAM

MARATVARTANIAN RICHARDWYATT OLIVER ZANGWILL

INTERNATIONAL REVIEW OF

Neurobiology Edited by JOHN R. SMYTHIES Deparfment of Psychiatry and The Neurosciences Program University of Alabama Medical Center Birmingham, Alabama

RONALD J. BRADLEY The Neurosciences Program University of Alabama Medical Center Birmingham, Alabama

VOLUME 26

1985

ACADEMIC PRESS, INC. (Horcourt Brace Jovanovich, Publishers)

Orlando Toronto

San Diego Montreal

New York Sydney

London

Tokyo

COPYRIGHT 8 1985, BY ACADEMIC PRESS, INC. ALL RIGHTS RESERVED. NO PART OF THIS PUBLICATION MAY BE REPRODUCED OR TRANSMITTED IN ANY FORM OR BY ANY MEANS. ELECTRONIC OR MECHANICAL, INCLUDING PHOTOCOPY, RECORDING, OR ANY INFORMATION STORAGE AND RETRIEVAL SYSTEM, WITHOUT PERMISSION IN WRITING FROM THE PUBLISHER.

ACADEMIC PRESS, INC. Orlando, Florida 32887

United Kin dom Edition ublished by

ACADEM~CPRESS I&. (LONDON) LTD. 24-28 Oval Road, London NWI 1DX

LIBRARY OF CONGRESS CATALOG

ISBN 0-12-366826-3 PRINTBD IN THE UNIT@D S " E S OP AMERICA

85868788

9 8 7 6 5 4 3 2 1

CARD

NUMBER: 59-13822

CONTENTS

............

CONTRIBUTORS .........................

vii

The Endocrinology of the Opioids

MARKJ . MILLANAND ALBERTHERZ 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . I1. Characteristics. Modulation. and Possible Koles of Endocrine-Like Opioid Peptidesystems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111. Opioid Mechanisms in the Control of Endocrine Secretion . . . . . . . . . . . . . . . 1V. ConcludingComments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Keferences . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

1

2 32 58 59

Multiple Synaptic Receptors for Neuroactive Amino Acid TransmittersNew Vistas NAJAM

A . SHARIF

I . Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Identification . . . . . . . . . . . . . . . . . . 111. Receptors in General . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . IV . Radioreceptor Assays (RRAs) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . V. Inhibitory Amino Acid Receptors ................................... VI . Excitatory Amino Acid Receptors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . VII . Summary and Concluding Remarks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . VIII . An Additional Note . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Note.Added in Proof . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

I1 . Neuroactive Amino Acids-Transmitter

85 86 89 92 96 108 136

139 140

149

Muscarinic Receptor Subtypes in the Central Nervous System

WAYNEHoss

AND JOHN

ELLIS

I . Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

151

I1 . Characterization of Muscarinic Receptors . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

153 168 177 182 190

I11. IV . V. VI . VII .

Responses Elicited by Muscarinic Activation . . . . . . . . . . . . . . . . . . . . . . . . . . . Relationships among Subpopulations and Responses . . . . . . . . . . . . . . . . . . . . Regulation of Muscarinic Receptors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Solubilization of Muscarinic Receptors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Concluding Remarks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . V

191

192

CONTENTS

vi

Neural Plasticity and Recovery of Function after Brain Injury JOHN F

. MARSHALL

.................................................... . ................................... ......................... IV. Neural Events Mediating Recovery: Morpholo@cdl Adaptations........... V . Neural Events Mediating Recovery: Neurocheniical Adaptations .......... V1. Conclusions and Future Directions .................................. References ...................................................... 1. Introduction

201 202 209 210 223 239 241

11 TheoriesofKecovcryofFunction 111. Neural Events Mediating Kecovery: Overview

From lmmunoneurology to Immunopsychiatty: Neuromodulating Activity of Anti-Brain Antibodies

BRANISLAV D.JANKOVIC 1. Introduction .................................................... I1. Brain Antigens .................................................. 111. Neuroimmunologkal Diseases and Anti-Brain Antildies ................ 1V. Biological Activity of Anti-Brain Antibodies ........................... V. Immunopsychiatric Diseases........................................ VI . Concluding Remarks.............................................. References ......................................................

249 256 262 269

293 300 302

Effect of Tremorigenic Agents on the Cerebellum: A Review of Biochemical and Electrophysiologicol Data

V . G. LONGOAND M . MASSOTTI I . Introduction .................................................... I1 ElearophysioIogicalData .......................................... 111. Neurochemkd Uata .............................................. IV. Condunions ..................................................... V . Summary ....................................................... References ......................................................

.

INUEX................................................. CONTENTS OF RECENTVOLUMES ............................

315 316 323 324 327 328

331 337

CONTRIBUTORS Numbers in parentheses indicate the pages on which the authors’ contributions begin. JOHN ELLIS,

Department of Psychiatry Neuroscience Research Unit, University

qf Vermont College of Medicine, Burlington, Verwiont 05405 ( 15 1 )

ALBERTH ERZ, Departvnent of Neuropharrricicolo~~, Max-Planck-Imtitut f u r Psychiatrie, 0-8033 Martinsried, Federal Republic of Germany ( 1 ) WAYNEHoss, Center for Brain Research, University of Rochester School of Medicine and Dentistry, Rochester, New York 14642 (15 1 ) BRANISLAV D. J A N K O V I ~ ,Immunology Research Center, 11221 Belgrade, Yugoslavia (249)

V . G. LONCO, Department of Pharmacology, Istituto Superiore di Sanita, Rome, Italy (315)

F. MARSHALL, Department of Psychobiology, University of Calqornia, Irvine, Irvine, Calzfornia 9271 7 (201)

JOHN

M . MASSOTTI, Department of Pharmacology, Istituto Superiore di Sanita, Rome, Italy ( 315)

MARKJ . MILLAN, Department of Neuropharmacology, Max-Planck-Institut f u r Psychiatrie, 0-8033 Martinsried, Federal Republic of Germany ( 1) NAJAM A . SHARIF, Department of Biochemistry, Queen’s Medical Centre, Nottingham NG7 2 U H , England, and Parke-Davis Research Unit, Addenbrookes Hospital, Cambridge CB2 2QB, England (85)

vii

This Page Intentionally Left Blank

THE ENDOCRINOLOGY OF THE OPlOlDS By Mark J. Millan and Albert Herz Department of Neuropharmacalogy Max-Planck-lnrtitut fur Psychiatrie Martinsried, Federal Republic of Germany

I. Introduction. . . . . . . . . . . . . .

.................

11. Characteristics, Modulation, a

I

ocri ne- Li ke

Opioid Peptide Systems. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A. General Organization o f Mechanisms for Pituitary Control. . . . . . . . . . . B. Characteristics of Opioid Peptide Systems. ........................ C . Relationship of Opioid Peptides to the Neurohypophyseal Tract . . . . . . D. Comparative Control of Anterior and Intermediate Lobe Secretion of

.... E. Modulatior

2 2 5 10

13

and

............................... 011s . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111. Opioid Mechanisms in the Control of Endocrine Secretion. . . . . . . . . . . . . . A. Opioid Control of Ariterior Lobe Secretion. . . . . . . . . . . . . . . . . . . . . . . .

B. Opioid Control of Secretion of P-Endorphin, Adrenocorticotropin, and a-Melanocyte Stiniiilating Hormone C . Opioid Control of and Oxytocin . . . . . . . . . . D. Opioid Control of Periph retion . . . . . . . . . . . . IV. Conclucling Comments ................. Refererices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

‘LO

28 32 32 45 48

54 58 59

1. Introduction

It has been recognized for a generation that the administration of synthetic opioid alkaloids such as morphine is associated with pronounced alterations in the endocrine secretion of the pituitary in both animals and man. T h e discovery of the existence of specific receptors for opioids in mammalian tissue and the identification of their naturally occurrent ligands, the endogenous opioid peptides (endorphins), naturally encouraged questions as to their physiological role in endocrine mechanisms. Indeed, the influence of endogenous opioid peptides upon hypophyseal secretion has proven very comparable to that of synthetic opioids. It is now apparent that there is a multiplicity of both opioid 1 IN 1 F.KNATIONAL REVIEW 01. NkLIKOBIOLOGY, V O L 26

Copyright 0 1Y85 by Academic Press, Inc. A11 rights of reproduc~ionin any form reserved.

ISBN 0-12-366826.3

2

MARK J . MILLAN A N D ALBERT H E K Z

receptor types and of their opioid peptide ligands. Although our knowledge of the role of opioid mechanisms in the control of endocrine secretion in general has been greatly amplified in recent years, the significance of the particular receptor types and individual pools of specific opioids remains largely elusive. Those opioid peptides as yet systemically examined have revealed a widespread but differential distribution. They are by no means restricted to the central nervous system (CNS) but occur in significant quantities in discrete networks in a variety of other tissues including the pituitary and peripheral organs. Such a localization is suggestive of a far broader spectrum of functions than in, for example, the familiar control of nociception and mood. Indeed, a role of extra-CNS, in addition to CNS, pools in regulation of endocrine secretion is possible. However, an especially attractive proposition would be a liberation of opioids into the systemic circulation in order to operate upon remote tissues in a hormone-like fashion themselves. Alternatively, they might act locally o r indirectly attain CNS population of opioid receptors. Section I1 of the present article offers a description of the organization, characteristics, release, and modulation of hypophyseal and other endocrinologically relevant pools of opioid peptides. I n addition, the question of their possible target sites and functions is addressed. In Section I11 the nature and role of opioid mechanisms in the control of the endocrine secretion of various tissues is considered.

II. Characteristics, Modulation, and Possible Roles of Endocrine-Like Opioid Peptide Systems

A. GENERAL ORGANIZATION OF MECHANISMS FOR PITUITARY CONTROL There are important differences between the particular pituitary lobes, anterior (AL), intermediate (IL), and neural (NL), as concerns their relationships with the hypothalamus and other tissues and, correspondingly, the mechanisms of their control (Fig. 1 ) . The AL is considered to receive no (or a comparatively minimal) direct neuronal input, although a minor serotoninergic innervation of uncertain origin has been discovered (Friedman et al., 1983; Westlund and Childs, 1982). It is, thus, subjected to control by releasing and inhibiting factors which attain the AL via the portal vessels subsequent to secretion into the capillary network of the external median eminence. Central neuronal pathways control AL activity via convergent actions

3

T H E ENDOCRINOLOGY OF T H E OPIOIDS

’‘

CATECHOLAMINES

PEPTIDES - - - + - ? h O N i N OPIOIDS

ACTHIR-E/R-LPH LH FSH TSH OH

a-MSH/R-E

VP/DYN OT/ME LE

]

\

SYMPATHETIC NORADRENERGIC INNERVATION

PRL

FIG. 1. General organization of mechanisms for the control of the endocrine secretion of the pituitary. Abbreviations: Ach (acetylcholine), ACTH (adrenocorticotropin), DA (dopamine), DYN (dynorphin), P-EP @-endorphin), FSH (follicle-stimulating hormone), GABA (y-aminobutyric acid), GH (growth hormone), IF (inhibiting factors), L (lobe), LE (leucine-enkephalin), LH (luteinizing hormone), P-LPH (P-lipotropin), magnocell. (magnocellular), ME (methionine-enkephalin), a-MSH (a-melanocyte-stimulating hormone), OT (oxytocin), PRL (prolactin), RF (releasing factors), TSH (thyroid stimulating hormone), VP (vasopressin). It must be noted that the fact that divergently projecting tracts emanate from a single “cell body” is not intended to imply that such a condition exists in viuo. For example, largely separate populations of DA or VP somata project to either the external median eminence or the neural lobe.

upon the release of these factors. There is a massive tubero-infundibular dopaminergic projection from the arcuate region of the hypothalamus to the external (vascularized) median eminence, possibly supplemented by a nigral input (Bjorklund et al., 1973; Hokfelt et al., 1978; Kizer et al.,

4

M A K K J . MILLAN AND A I B E R T HEKZ

1976; Moore and Johnston, 1982). In addition, a GABAergic and cholinergic innervation of the external eminence, largely from the arcuate region, is known (Carson et al., 1977; Walaas and Fonnum, 1978). The serotoninergic input (primarily from the midbrain raphe) travels to both internal and external regions of the median eminence (Hokfelt et al., 1978; Moore and Johnston, 1982; Steinbusch and Nieuwenhuys, 1981). The brainstem-derived adrenergic and noradrenergic contribution, largely in the ventral bundle, runs mainly to the znternal zone (Hokfelt et al., 1978; Jonsonn et al., 1972; Moore and Johnston, 1982). T h e IL, in contrast to the AL, has a linked vascular supply and receives a neuronal input most prominently from dopaminergic neurons in the arcuate area, a population largely independent of its counterpart running to the median eminence (B-j8rklund et al., 1973). These may represent major sites for central integration of IL control. Also of note is the GABAergic innervation of the IL, of a CNS derivation, probably the arcuate or posterior hypothalamus (Oertel et al., 1982; Kacagni el ul., 1979; Vincent et ul., 1982). ‘I’he NL is distinctive in that it does not contain cells but is comprised of a heavy population of fibers, blood vessels, and specialized glial cells, pituicytes. Neurons are derived primarily from the hypothalamic niagnocellular paraventricular nucleus (PVN) and supraoptic nucleus (SON), although other accessory nuclei contribute; these synthesize and release vasopressin (VP) or oxytocin (OT) (Sofroniew and Weindl, 198 1) in addition to certain opioids (Section II,C,l) and a variety of other peptides (e.g., Martin et al., 1983a; Rossier et al., 1979a). Evidently, central rieuronal networks rnay indirectly influence the activity of the N L via an interaction with these hypothalamic perikarya. Actions of substances at the terminal level within the NL are also, however, of importance. Further, the NL (as with the IL) receives a direct dopaminergic and GABAergic input (Bjorklund et al., 1973; Oertel et al., 1982; Vincent et al., 1982). Serotoriinergic fibers are present in the NL and IL in a position suggestive of a modulatory role; those in the IL, at least, appear to originate in the brain, possibly the midbrain raphe and dorsomedial nucleus of the hypothalamus (Baumgarten et al., 1972; Friedman P t al., 1983; Leranth et ul., 1983; Mezey et al., 19846; Saavedra et al., 1975, 1983; Steinbusch and Nieuwenhuys, 1981). T h e acetylcholine in the IL and AL, probably partly in nerve fibers, is of uncertain significance (Bridges et ul., 1973; Fischer arid Moriarty, 1977; Conte-Devolx et al., 1981; Saavedra et al., 1975). Pituicytes rnay play a role in secretory control of the N L (Lightman et al., 1983b; Van Leeuwen et al., 1983). Each lobe is potentially subject to modulation by humoral factors occurrent in the systemic circulation, such as corticosteroids or catecho-

THE ENDOCRINOLOGY OF THE OPIOIDS

5

lamines; these may operate either directly on the pituitary o r indirectly via the median eminence or brain loci (see below). Further, there exists a network of vascular interconnections between the various lobes, the portal circulation, and the median eminence (Bergland and Page, 1979; Page, 1982). Thus, agents delivered to the portal vessel may influence the IL and NL; further, there may be interlobe interactions, and a retrograde flow from the pituitary to the median eminence and brain may occur (Section II,F,2). Finally, the NL and IL receive a peripheral sympathetic noradrenergic input, largely to their vascular zones (Baumgarten et al., 1972; Bjorklund et al., 1973).

B. CHARACTERISTICS OF OPIOID PEPTIDESYSTEMS Inspection of Fig. 2 reveals that three basic cell types are distinguished: those containing P-endorphin (P-EP), methionine-enkephalin (ME), or dynorphin (DYN). This differentiation is in accordance with their independent neuronal localization and possession of separate biosynthetic precursors (see Hollt, 1983). These have been designated for @-EP, ME, and DYN, respectively, as proopiomelanocortin (POMC), proenkephalin A, and proenkephalin B or prodynorphin [see below for leucine-enkephalin (LE)]. 1. @-Endorphin

POMC is the common precursor for P-EP, adrenocorticotropin (ACTH), a-melanocyte-stimulating hormone (a-MSH), and related bioactive peptides. It is cleaved to yield ACTH, from which a-MSH is subsequently split, and P-lipotropin (P-LPH), from which P-EP, the 3 1residue C-terminal component, is generated (see Hollt, 1983; Hope and Lowry, 1981; Lis et al., 1982; Mains and Eipper, 1981). Certain C-terminal abbreviated fragments of P-EP such as C’-fragment (/3-EPI-27)and des-hist-C‘-fragment (P-EP1-2fi)may also be produced (Hope and Lowry, 1981; Zakarian and Smyth, 1982). These are of retained but diminished opiate activity. a- and y-endorphin ( P - E P I Land ~ ~ P-EPI-17), found in brain and pituitary, are of uncertain significance since they do not seem to be “physiologically” generated from POMC (Vaudry et al., 1980; Verhoef‘ et al., 1980; Hiillt, 1983; Hope and Lowry, 1981). The C-terminal dipeptide glycyl-glutamine, generated upon cleavage to P-EP1-27 may in addition be of significance (Parish et al., 1983).There are major intertissue differences in the processing, modification, and storage of P-EP- and ACTH-related species. In the AL, ACTH greatly predominates over a-MSH; p-LPH (which is opiate inactive) and P-EP are

ti

MARK J. MILLAN AND A L B E K I

CNS CENTERS OF ENDOCRINE CONTROL

A1

I1

NL

I

I

I?

CIRCULATION

.,

FIG. 2. Organization of endocrinologicdlly relevant opioid peptide systems of the hypothalamo-pituitary axis. Abbreviations: AL (anterior lobe), IL (intermediate lobe), N L (neural lobe), PVN (paraventricular nucleus), SON (supraoptic nucleus). Symbols: 0 , pendorphin cell bodies; V, dynorphin cell bodies; enkephalin cell bodies. The question mark addressed to dynorphin cell bodies in the AL indicates that these have not, as yet, actually been visualized therein. That adjacent to the NL signifies that a NL secretion of dynorphin or enkephalin into the circulation is inferred but not proven.

present in approximately equimolar amounts. In the IL, however, AC'I'H and p-LPH are almost entirely processed to yield a-MSH and pEP. p-EP is, subsequently, partially processed to C'-fragment and des1981; Chang and Loh, 1983; Hope and hist-C'-lragrnent (Akil et d., Lowry, 1981; Mains and Eipper, 1981; Zakarian and Sniyth, 1982). aMSH is N-acetylated, tesulting in a potentiation of its biological activity. However, the great majority of' p-EP-like species also undergo acetylation resulting in a total loss of opioid activity (Akil ut ul., 1981, 1983; Cahill el al., 1983; Evaris rt al., 1982; Seidinger ijtld Hiillt, 1980; Weber et

THE ENDOCRINOLOGY OF T H E OPIOIDS

7

al., 1981, 1982d; Zakarian and Sniyth, 1982). P-EP is, further, absent from the NL. In the brain, P-EP-synthesizing somata are largely confined to the hypothalamic arcuate region (Finlay et al., 1981). These perikarya innervate other hypothalaniic and extrahypothalamic nuclei (such as the SON, PVN, suprachiasmatic nucleus, midbrain, and amygdala) involved in the control of pituitary secretion and also the median eminence (Bugnon et al., 1979; Finlay et ul., 1981). In the brain, preliminary processing resembles the IL, with a heavy preponderance of P-EP and a-MSH over P-LYH and ACTH. In such major endocrinologically relevant tissues as the hypothalamus or midbrain, virtually none is acetylated; whether a portion is so deactivated elsewhere requires clarification (Akil et al., 1983; Weber et al., 1981; Zakarian and Sniyth, 1982). 2. Dynorphin and Related Peptides Proenkephalin B or prodynorphin encodes a family of opioids, each of which bears an N-terminal-located LE, comprising DYN (the full 17residue molecule), DYN1-8, riniorphin (= dynorphin B), a-neoendorphin (a-NE), and P-neoendorphin (a-NE1-9). These appear to be colocalized and comodulated and to exhibit a similar preference for the K type of opioid receptor (Hollt, 1983; Maysinger et al., 1982; M. J. Millan et al., 1983b, 1984a,b; Schulz et al., 1982b; Watson et al., 1982b, 1983b; Weber et al., 1982e; Weber and Barchas, 1983). Such communalities allow us to collectively refer to them, for brevity, as DYN-related peptides. The relative proportions of particular species may vary from tissue to tissue (Maysinger et al., 1982; Millan et al., 198313; Seizinger et al., 1984; Weber et al., 1982a,c). In the AL, in fact, in analogy to POMC, processing is terminated prematurely such that only high-molecular-size peptides are present (Seizinger et al., 1981). Further, DYN-related species may be lacking in the IL. In the NL, however, substantial quantities are neuronally localized consisting primarily of the five above-mentioned DYN-related peptides (Section II,C, 1). In the brain, such authentic DYN-related species similarly predominate and have been visualized in perikarya in, for example, the suprachiasmatic nucleus, PVN, SON, arcuate nucleus, limbic system, and midbrain, i.e., regions of importance in endocrine control (Khachaturian et al., 1982b; Watson et al., 1983a,b; Weber et al., 1982b,e,f; Weber and Barchas, 1983). Further, it is possible that a pathway containing DYN-related peptides projects from the PVN to the external median eminence (Roth et al., 1983). 3. Enkephalins Although ME is derived from proenkephalin A, the origin of LE is less clear since it is contained not only within this precursor molecule hut

8

MARK .J. MILLAN A N D A L B E K I I I E R Z

also in proenkephalin B , the DYN precursor (see Hollt, 1983).However, since there is a general parallelism in the distribution of ME and LE in the CNS, with an excess of ME over LE which is quite similar to the ratio of ME: LE sequences in proenkephalin A (6: I ) , this is probably a source o f a large proportion of LE measured (Hollt, 1983; Wesche et al., 1977; Yang et al., 1977). In fact, it is possible that the precursor actually yields only four copies of ME plus one heptapeptide and one octapeptide (Cterminal ME extensions) which, together, account for the six ME copies (see Hijllt, 1983). These extensions are opioid active and present in the brain of the rat but have not, as yet, been recorded in major aniounts.in pituitary. Further longer sequences incorporating ME or ME/LE (BAM peptides, etc.) and other proenkephalin A intermediaries are also occurrent in the brain but not pituitary (Bloch et al., 1983a; Hijllt et al., 1982a; Ikeda et al., 1982; Khachaturian el nl., 1982a; Liston et a1.,1983; Sanders et d., 1984; Watson et al., 1983a; Weber et al., 1982f). Recent studies in the gut suggest that ME and LE can occur both together and in independent neurons (Larsson and Stangaared-Pederson, 1982). Indeed, it is possible that LE is also formed from the DY N-precursor proenkephalin B, as may be the case in VP-containing magnocellular neurons innervating the NL which, as discussed in Section lI,C, 1, contain LE, DYN, and a-NE but not ME (Martin et al., 1983a; Watson et al., 198%; Weber et ul., 1982b). I n contrast, M E (and LE) appear to coexist in 0.r neurons of the NL in the absence of DYN (Martin et ul., 1983a; Watson et al., l982a) (see Section I I , C , l ) . In the AL and II,, ME and LE also appear to be present, although the nature of these pools is comparatively unclear; it has been suggested that, in the AL, they might be contained in established endocrine cells, such as somatotrophs (Duka et nl., 1978; Tramu and Leonardelli, 1979; Weber et d., 1978). The enkephalinergic innervation of the external median eminence is especially rich and may possess multiple origins such iis the PVN or arcuate nucleus, and ME has also been claimed t.o exist in the somatostatin input which conies from the anterior periventricular region (Beauvillain et ul., 1984; Hiikfelt et al., 1980; Traniu and Leonardelli, 1979; Tramu et al., 1981). Enkephalinergic neurons are widely dispersed in the brain, including hypothalamic and extrahypothalamic sites of endocrine regulation (Khachaturian et ul., 1982a; Watson et al., 1983a). 4. Perif herd Tissues: Or&+is of Ofioid Peptides in Systemic Circulation P-EP, DYN, and enkephaliris are not restricted to the CNS and pituitary but are variously distributed in a spectrum of peripheral tissues. Of special note is the occurrence of P-EP (and DYN) in the placenta (Liotta et al., 1982; but see Weindl et al., 1983). Further, certain endocrine-like

THE ENDOCRINOLOGY OF THE OPIOIDS

9

cells have been suggested to synthesize POMC and its products, such as /3-El'; for example, somatostatin &cells and glucagon a-cells of the pancreas, testosterone-irianufacturing Leydig's cells of the testis, gastrinproducing cells of the gut, and follicular cells of the ovary (Grube et al., 1978; Larsson, 1981; Lini et al., l983b; Margioris et al., 1983; Pintar et al., 1984; Shu-Dong P t ~ l . 1982; , Watkins et d., 1980). Interestingly, in contrast to other manimalian species, P-EP has been detected in the human adrenal medulla (Evans et al., 1983). Whether the above cell populations operate in an endocrine-like fashion is unclear, and as measured by radioimmunoassay, hypophysectomy eliminates immunoreactive (ir)-P-EY from systemic plasma in which its levels appear to reflect hypophyseal activity (Przewlocki et al., 1982). In trunk plasma, levels of ir-DYN cannot be determined for an evaluation of the DYN-releasing activity of the pituitary since plasma ir-DYN is not derived therefrom and does not disappear with pituitary ablation (Hollt, unpublished; Spampinato and Goldstein, 1983). A partial derivation from the gut is conceivable in view of its ability, zn iiitrv, to discharge large quantities of ir-DYN (Kromer et al., 1981). Further, the adrenal medulla may contribute since DYN is found therein (primarily in noradrenergic cells) and an in uztro release of DYN therefrom has been demonstrated (Dumont et al., 1983). Enkephalins appear to coexist with acetylcholine in splanchnic nerve terminals innervating the adrenal medulla chromaffin cells (Schultzberg et al., 1978). They are also, in fact, synthesized in chromaffin cells (predominantly, in contrast to DYN, in those containing adrenaline) and coexist with adrenaline in individual granules from which their release is concomitantly regulated (Chaminade et al., 1983; Hanbauer et al., 1982; Lang el al., 1982, 1983; Roisoin et al., 1983; Rossier et al., 198 1; Viveros et al., 1980; Yang et al., 1980). Since the higher molecular weight proenkephalins are also stored and released (particularly in response to powerful stimuli), we shall refer collectively to ME-related peptides. Splanchnic stimulation, nicotinic agonists, or insulin promote the release of ME-related peptides and greatly elevate their levels in plasma (Chaminade et al., 1983; Govoni et al., 1981; Hanbauer et al., 1982; Ryder and Eng, 1981). Further, under basal and stimulated conditions, plasma effluent in the adrenal vein contains much higher levels of these than elsewhere (Clement-Jones, 1982; Lang et al., 1982; Yang et al., 1980). Moreover, hypophysectomy does not deplete these levels in systemic plasma (Hanbauer et al., 1982). Thus, the adrenal medulla may be the predominant source of ir-ME and ir-LE in the systemic circulation, although this has not, as yet, been conclusively proven (see Smith et al., 1981). In addition, a contribution of gut endocrine cells is possible (Alu-

10

M A U K J . MILLAN AND ALBERT HEUZ

met et al., 1978; Laasberg et al., 1980). Sympathetic ganglionic nerve endings in which enkephalins are costored with noradrenaline may also deliver to the circulation (Klein et al., 1982).

5. Opioid Keceptors There is a paucity of opioid receptors in both the AL and IL, whereas in the NL these are present in significant quantities (Simantov and Snyder, 1977; Wamsley et ul., 1982). A recent report (Lightman et al., 1983b) suggests that at least a component of the total population of opioid receptors in the NL may be localized not on nerve terminals but on pituicytes which may be involved in mediating their modulatory roles therein (see also Van Leeuwen et ul., 1983). T h e limbic system, in particular the amygdala, and in primates the hypothalamus, are rich in opioid receptors (Atweh and Kuhar, 1977; Wamsley et al., 1982). Opioid receptors are also present in numerous peripheral tissues (Hughes, 1981). C. RELATIONSHIP OF OPIOID PEPTIDESTO NEUKOHYPOPHYSEAL TRACT

THE

1. Organization The finding of LE-containing fibers in the NL and that levels of irLE therein are depleted by stalk transection (Rossier et al., 1979a) initiated a number of studies of the relationship of opioids to VP and O T which culminated in the immunohistocherriical visualization of particular opioids within VP- or OT-containing neurons. Martin et al. (1983a) reported an association of LE with VP neurons in the NL. In addition, an occurrence of ME plus, possibly, LE with O T neurons therein was seen (but see also Coulter et al., 1981; Lotstra et al., 1982; Micevych and Elde, 1980; Reaves and Hayward, 1980; Martin et al., 1983b; Rossier et al., 1979a). The existence of both ME and LE within O T neurons would be consistent with their joint occurrence in proenkephalin A. The observed absence of ME from VP neurons supports the contention that ME, rather than cross-reacting LE, is the species seen in OT fibers. In the rat, the ME-extension octapeptide was recently visualized in OT neurons by Martin et al., (1983b). However, others have not found such proenkephalins in the NL or magnocellular nuclei in significant amounts either radioimmunologically or immunohistochemically (Weber et al., 1982f; Pittius, unpublished). Additional work is required, thus, to determine the exact identity of ME and its relatives in this axis, since imrnunohistochemically there is a particular danger of inadvertent cross-reaction leading to erroneous conclusions. In the case of LE it is

T H E ENDOCRINOLOGY OF THE OFIOIDS

11

uncertain whether this exists entirely as the free pentapeptide in vim, since it might be generated from DYN species postmortem (Martin and Voigt, 1982). Indeed, a coexistence of DYN, DYNI-8, and a-NE with VP but not OT has also been indicated in NL terminals and somata of the SON and PVN (It0 et al., 1981; Martin et al., 1983b; Watson et al., 1982a; Weber et al., 1982b; Whitnel et al., 1983). Further, immunohistochemically rimorphin (DYN B) is similarly detectable in this neurohypophyseal tract (Weber et al., 1982f; Weber and Barchas, 1983). Complementary biochemical studies have confirmed the presence of each of these (plus P-neoendorphin) in the NL (Seizinger et al., 1984; Weber et al., 1982a,c). Thus, ME and LE appear to exist in OT neurons and LE and DYN species in VP neurons of the magnocellular nuclei-NL axis. However, no definitive conclusion can, as yet, be made as to whether these opioids (the situation is particularly unclear as concerns LE) might also occur independently of VP and OT therein (van Leeuwen et al., 1983). Additional evidence for a cooccurrence of LE and DYN-related species with VP rather than OT has been provided in studies of Battleboro rats which are congenitally unable to synthesize VP but possess an intact mechanism for OT. These rats display depressed levels of ir-LE in the NL and manifest variable, sex-dependent reductions in their NL content of ir-DYN-related peptides (Cox et al., 1980; Hollt et al., 1981b; Martin et al., 1983b; Rossier et al., 1979a; Weber et al., 1983). These deficiencies support a coexistence of LE and DYN-related species with VP and presumably reflect defects in systems for transport of peptides or elaboration of granules, etc. However, although decreased in their levels, LEand DY N-related species ;we Jtill present. ‘This finding, together with the characterization of a distinctive precursor for VP in contrast to those for opioids (Land el al., 1982), demonstrates that, although costored, VP and LE/DYN-related species are not cojointly synthesized from a coninion precursor. In agreement with the above data, selective elimination of either the PVN or SON results in a parallel diminution in levels of ir-DYN, irDYNI-8, and ir-a-NE in the neurointermediate lobe (NIL)-or NL plus adhering IL-and hypothalamus of rats (Millan et al., 1983b; M. H. Millan et al., 1984). ‘The magnitude of these decreases in ir-DYN, irDYNI-8, and ir-a-NE is strongly and positively intercorrelated and also correlates with the magnitude of the fall in levels of ir-VP produced. 2. Modulation The possibility of a parallel release of these costored peptides in the NL is supported by preliminary evidence for a common subcellular localization of OT with ME-LE and of VP with LE- and DYN-related

12

M A K K ,I. MILLAN A N D ALBERT 1IEKZ

species in granules or organelles assumed to be related to secretion (Martin et al., 1983a; Molineaux and Cox, 1982; Whitnel et al., 1983). Further, electrical stimulation of the isolated NIL in uilro induces a concomitant outflow of ir-VP and ir-DYN (Maysinger et al., unpublished). A common physiological control of the release of VP-, LE-, and DYNrelated peptides is indicated in the following observations. 1 . Endocrinological maneuvers resulting in an elevation (dexamethasone) o r suppression (adrenalectomy) of corticosteroid levels in systemic plasma result in, respectively, parallel rises and falls in NIL levels of ir-VP, ir-LE, and ir-DYN-related peptides (Hollt et al., 1981b). 2 . Dehydration concomitantly depletes the NIL content of ir-VP, irLE, and ir-DYN-related peptides (Hollt et al., 1981b). Stress affects neither ir-VP nor ir-DYN in the NIL but elevates the levels of each in the hypothalamus (Millan ef al., 1983~). 3. Discrete interruption of the ascending ventral noradrenergic bundle which innervates the magnocellular PVN and SON results in a decrease in NIL ir-VP, reflecting a disinhibition of its secretion; the NIL content of ir-DYN-related peptides is depressed in parallel (Millan et al., 19844. 4. Chronic treatment with haloperidol induces a rise in the content of ir-DYN and ir-VP in the NIL, suggestive of a common dopaminergic (or catecholaminergic) modulation of outflow (Hollt, 1981). 5. Under unstimulated conditions, there is a strong positive correlation between levels of ir-VP- and ir-DYN-related peptides in the NIL and hypothalamus, indicative that, in the steady state, these peptides are also maintained in parallel (Millan et al., 1983b). A steady state intercorrelation, parallel distribution, and comodulation of various DYN-related species in the AL and discrete brain and spinal cord structures has also been demonstrated. However, in each case, no correlation is manifested with VP, which is also modulated independently (M. H. Millan et al., 1984; Millan rl al., 1983a-c, 1984a,b). Further, the PVN was identified as a major source of extrahypothalamic ir-VP, in contrast to its lack of contribution to DYN-related species (M. 14. Millan et al., 1982b, 1984). The hypothalamic-neurohypophyseal axis in which VP- arid DYN-related peptides are colocalized and comodulated is evidently, thus, unique. Likewise, no evidence for a cooccurrence or coregulation of ME with O'r extrinsic to the NIL was acquired. The lack of influence of lesions of the PVN or SON upon AL pools of ir-DYN-related species is demonstrative that these are not, in distinction to their NL counterparts, derived from magnocellular nuclei (Millan et al., 1983b; M. H. Millan el al., 1984). l ' h e higher molecular

THE ENDOCRINOLOGY OF THE OPIOIDS

13

weight of AL as compared to hypothalamic species of DYN and a-NE similarly mitigates against a hypothalamic origin and may explain why DYN in the AL is refractory to visualization by iinmunohistochemistry (Seizinger et al., 1981). Figure 3 offers a diagrammatic illustration of the comparative organization of CNS and pituitary networks of DYN and VP.

D. COMPARATIVE CONTROL OF ANTERIOR AND INTERMEDIATE LOBE SECRETION OF /I-ENDORPHIN Evidence has been acquired for the common localization of ACTH/ P-LPHIP-EP and a-MSHIP-EP in secretory granules of cells in, respectively, the AL and IL; these appear to release these species in parallel in proportions similar to those in which they are stored (e.g., Evans et al., 1982; Guilleman et al., 1977; Lis et al., 1982; Pelletier et al., 1977; Tilders et al., 1981; Weber et al., 1979). In analogy to their distinctive processing of POMC, however, the AL and IL are subject to quite different mechanisms of control. For a consideration of these, in vitro studies of isolated lobes are most appropriate since indirect actions confuse the situation in vivo. Further, in vivo, since most antisera against P-EP recognize the opiate-inactive P-LPH and acetylated forms, it is impossible, in the absence of a comprehensive chromatographic separation, to specify origins

AL

IL

NL

FIG. 3. Comparative organization of CNS and pituitary systems of vasopressin and dynorphin. Abbreviations:AL, IL, NL, PVN, and SON as in Fig. 2; ARC (arcuate nucleus), SCN (suprachiasmatic nucleus). Symbols: 0 , V P cell bodies; V, DYN cell bodies.

14

MAKK J . MILLAN A N D ALBERT HEKZ

(or functional activity) of ir-P-EP determined. (Comparable problems are also confronted in ACTH and a - M S H assays.) A diagrammatic illustration of the major modes of control of AL as compared to IL secretion of ir-P-EP is presented in Fig. 4. Opioid mechanisms of regulation are discussed in Section 1II.B. 1, Anterior Lobe Corticotrophs u. Direct Actions. A possibly multifactorial corticotropin-releasing factor (CRF) is regarded as the primary stimulant of corticotrophic outflow of ACTHIP-LPHIP-EP into the systemic circulation. Recently, a 4l-residue polypeptide, for which specific receptors exist in the AL (Wynn et ul., 1983), was isolated and which fulfills a number of criteria for recognition as a CRF. It is present in high amounts in the external median eminence and portal plasma, is “stimulated” by adrenalectomy, and promotes a secretion of ACTHIP-LPHIP-EP from the AL (Bruhn et d.,

ANTERIOR LOBE

INTERMEDIATE LOBE

Ach OPIOIDS

OPIOIDS

OPlOl D S

O-EP

SECRETION

CIRCULATION

R-A0

CIRCULATION

FIG.4. Humoral and neurorial control of anterior as compared to intermediate lobe secretion of p-endorphin and coreleased peptides. Abbreviations: Ach, ACTH, DA, P-EP, GABA, P-LPH, a-MSH, and VP as in Fig. 1; P-AD (p-adrenoceptor agonists), WAD (aadrenoceptor agonists), CS (corticosteroids), CRF (corticotropin-releasing factor). Note that for anterior lobe, data refer primarily to studies on CRF: (VP) is included since CRF may exist in a subpopulation of VP-containing neurons. Question marks signify tentatively ascribed actions requiring consolidation. Please see text for details and specification of literature in support or in contradiction of particular actions.

THE ENDOCRINOLOGY OF THE OPIOIDS

15

1984b; Burlet et al., 1983; Gibbs and Vale, 1982; Kolodziejczyk et al., 1983; Merchenthaler et al., 1983; Moldow and Fischman, 1982; Rivier et al., 1982a; Swanson et al., 1983; Vale et al., 1981). Further, antibodies against this block the release of ACTH in vivo, e.g., in response to stress (Conte-Devolx et ul., 1983; Rivier et al., 1982b). However, VP has also been implicated as a participant in control of corticotrophic activity. VP from the PVN is similarly heavily represented in the external median eminence; these pools are modified by corticosteroid manipulations (Vandesande et al,, 1977; Zimnierman et al., 1977). It is also found in portal plasma in substantial quantities and is a potent stimulant of the release of ACTHIP-LPHIP-EP via a direct action on the AL wherein receptors for VP are found (Buckingham, 1980; Gaillard et al., 1984; Gillies and Lowry, 1982; Knepel et al., 1984; Przewlocki et al., 1979b; Rivier et al., 1984; Vale et al., 1979; Zimmerman et al., 1977). Certain studies have found antisera against VP to quench the CRF activity of median eminence extracts and introduction of antibodies against VP in the brain to partially block the release of ACTH engendered by stimulation of the PVN (Carlson et al., 1982; Gillies and Lowry, 1982; Linton et al., 1983). This is understandable since the PVN is the predominant origin of the median eminence population of VP (and CRF) fibers and its destruction and stimulation, respectively, suppresses and enhances corticotrophic secretion of ACTHIP-LPHIP-EP (Bruhn et al., l984a,b; Carlson et al., 1982; Dornhorst et al., 1980; Ixart et al., 1982; Makara et al., 1980; M. H. Millan et al., 1984; Tilders et al., 1982). Indeed, a colocalization of at least a component of the pool of this novel CRF with VP in a subpopulation of VP neurons projecting to the external median eminence was recently reported, although this awaits further description (Burlet et al., 1983; Kolodziejzyk et al., 1983; Roth et al., 1982; Sawchenko et al., 1984; Tramu et al., 1983). Further, both in vitro and in vivo, VP potentiates the stirnulatory influence of CRF upon corticotrophic secretion (Giguere and Labrie, 1982; Gillies et al., 1982; Lamberts et al., 1984). Thus, PVN-produced and -coreleased CRF and VP may act synergistically on the AL to facilitate the outflow of ACTHIP-LPHIP-EP, although its physiological significance remains to be more fully clarified (Carlson and Gann, 1984; Mormede, 1983; Rivier and Vale, 1983). In addition, there is anatomical and physiological evidence for a possible contribution of N L stores of VP (and/or OT) to the modulation of AL corticotrophic activity, although the significance of this is somewhat uncertain (see Antoni et al., 1983; Baertschi, 1980; Baertschi et al., 1980a; Beny and Baertschi, 1980; Karteszi et al., 1982; Legros et al., 1984). The activation of adrenoceptors, characterized as the a-postsynaptic type, within cultured AL cells provokes a secretion of ACTHIP-LPHIPEP (Giguere et al., 1981; Pettibone and Mueller, 1981a,b; Raymond et al.,

16

MARK J . MILLAN A N D ALBEH’I‘ H E K Z

1981).The external median eminence contains very little adrenaline and noradrenaline, the concentrations of which in portal plasma are not higher than in systemic plasma; thus, pools of these derived from the medulla or sympathetic nerve terminals may, it is suggested, represent the “physiological” ligdnds of these receptors (Hokfelt et al., 1978; Reymond et al., 1983). An interesting analogy to VP is that agonists at these a-adrenoceptors similarly potentiate the action of CRF in promoting corticotrophic generation of CAMPand release of ACTH (Bruhn et al., 1984; Gigukre and LabriC, 1983). Very recently, a direct action of serotonin in enhancing corticotrophic secretion and facilitating the action of VP thereon was reported (Spinedi and Negro-Vilar, 1983). Glucocorticoids, via rapid and delayed direct negative feedback actions, comprise major inhibitors of corticotrophic activity. They are effective under stimulated and, generally, basal conditions in moderating outflow of ACTHIP-LPHIP-EP (Giguere et al., 1982; Jones and Gillham, 1980; Imura et al., 1982; Koch et al., 1980; Przewlocki et al., 197913; Sakakura et al., 1981; Vale et al., 1979). In addition, they powerfully depress synthesis of mRNA encoding POMC but not the processing of this precursor (Birnberg el al., 1983; Dokas, 1983; Eberwine and Roberts, 1984; Imura et al., 1982; Roberts et al., 1979; Schachter et al., 1982; Siinantov et d., 1980). T h e slight effects of mineralocorticoids or gonadal hormones, if seen, presumably merely reflect their affinities for the AL glucocorticoid receptor (Giguere et al., 1982; Simantov et al., 1980; Roberts et al., 1979; Vale et al., 1979). [Zn vivo effects of gonadal hormones upon AL pools of ACTHIP-LPHIP-EP (Section I I , E , l ) are probably exerted via the CNS, although peripheral tissues could also be involved.] Somatostatin diminishes corticotrophic outflow of AC‘I’HIP-LPHIPEP in cultured AL cells; this effect constitutes a further putative example of multiple actions of releasing factors upon various AL hormones (Richardson and Schonbrunn, 1981). Angiotensin I1 directly stimulates AL outflow of ACTHIP-LPHIP-EP, which partially accounts for its ability to release these in uivo (Anhut et al., 1982; Beuers et al., 1980; Sobel, 1983). Even if of physiological relevance, whether the pool is derived from the periphery or median eminence is unclear. Finally, there are reports that cholecystokinin, a peptide present in the median eminence, can stimulate AL release of ir-P-EP both i n vivo and i n vztro (Matsumura et al., 1983; Meyer et al., 1982). It should be emphasized that the functional significance of these effects of neuropeptides awaits clarification. h. Central Nercrous System Integration. The CRF (VP) pathway secreting into the portal vessels is the major axis for the coordination and summation of CNS neuronal control of AL secretion of ACTHIP-LPHI P-EP. CRF neurons may be modulated either by actions on their somata

THE ENDOCRINOLOGY OF THE OPIOIDS

17

internal to the blood-brain barrier, as generally appears to be the case, or on their terminals in the median eminence, external to this. (The information and references pertaining to the in vitro studies mentioned below may be found in the following reviews: Buckingham, 1980; Jones and Gillham, 1980; Jones et al., 1976, 1981.) Acetylcholine, via a predominently nicotinic action, is excitatory to the release of CRF from hypothalamic tissue but inactive on the isolated median eminence. Corresponding in vivo evidence for a central cholinergic promotion of ACTH release has been acquired, and anticholinesterases o r cholinomimetics have been reported to elevate levels of ir-pEP in the circulation (Conte-Devolx et al., 1981;Jones et al., 1981; Risch et al., 1981, 1982). There is a divergence of views as to the nature of serotoninergic control of ACTH secretion but a clear balance of evidence supports an excitatory role consonant with its ability to provoke CRF release, in vitro, from hypothalamic tissue, but not from the median eminence. Whether it acts directly on CRF or via acetylcholine is, at present, unclear. Both acetylcholine and serotonin have been implicated in the regulation of circadian rhythms of corticotrophic secretion (Fuller, 1981; Jones et al., 1981). Recently, direct pharmacological evidence for a facilitatory role of central serotoninergic neurons upon release of AL ir-P-EP both tonically and under stress was obtained (Bruni et al., 1982; Sapun et al., 1981; Sapun-Malcolm et al., 1983). A substantial body of evidence favors a suppressive impact of noradrenaline (or adrenaline), via a central a-adrenoceptor and CRF, upon AL secretion of ACTHIP-LPHIP-EP (Roth et al., 1981; see Ganong, 1980; Mezey et al., 1984a; Millan et al., 1982). In vitro, a noradrenalineeffected inhibition of, in certain cases, basal and stimulated CRF release has been seen. The ventral bundle, which heavily innervates the PVN (Sawchenko and Swanson, 1982) and which is activated by stress, has been identified as the probable physiological substrate for ACTH control (Gann et al., 1978; Rose et al., 1976; see Ganong, 1980; Millan Pt al., 1982). It has recently been revealed that it is the ventral bundle, in contrast to the locus ceruleus, which is the noradrenergic substrate inhibitory to the AL secretion of ir-P-EP into the circulation both tonically and under stress (M. H. Millan et al., 1982a; Millan et al., 1982, 1984a). ‘I’hiscentral action represents a striking contrast to the role of a-adrenoceptors within the AL which promote AL outflow of ACTHIP-LPHIPEP. A further interesting comparison is that an activation of p-adrenoceptors in the brain, possibly via an angiotensin II-VP link, might enhance AL secretion of ir-P-EP (Knepel et al., l980a, 1981, 1982b). [Indeed, the possibility has been raised that a central action of angiotensin I1 in promoting VP release may partially mediate its enhancement of AL corticotrophic secretion in vivo (Anhut et al., 1982; Beuers Pt al.,

18

MAKK.J. MILI.AN A N D A L B E R T HEKZ

1982; Gariong Pt al., 1982; Knepel et al., 1980a, 1982a,b; Summy-Long et al., 198la,b; but see Spinedi and Negro-vilar, 1984).] I n contrast to the AL, via an action in the brain, VP may suppress CRF (and thereby P-EP) release (Plotsky et al., 1984). GABA comprises, with noradrenaline, the other major neurotransmitter inhibitory to hypothalamic CRF outflow. I n vivo studies have suggested a GABAergic inhibition of corticotrophic release both basally and under stress, but its relationship to AL P-EP has not, as yet, been directly examined. The actions of GABA and noradrenaline in suppressing CRF release appear to be exerted independently (Jones et al., 1976). Corticosteroids not only act via the AL to suppress corticotrophic secretion but also possess central sites of action. In addition to possible actions in the hypothalamus, limbic system structures such as the amygdala and hippocampus appear to play a major mediating role (Buckingham, 1980; Jones et al., 1981; Sakakura et al., 1981). I n vitro studies have demonstrated the potent suppressive effects of corticosteroids on both the release and generation of CRF. An additional short-loop ACTH feedback upon CRF is also considered to occur Uones et al., 1976).

2. Intermediate Lobe Melanotrophs A conspicuous feature of the isolated NIL (or cultures therefrom) is the high rate of unstimulated outflow of a-MSHIP-EP; this release is, parodoxically, suppressed or unaffected rather than enhanced by elevated concentrations of potassium (Przewlowki et al., 197913; see Tomiko et al., 1981). Further, the cells manifest spontaneous action potentials (Douglas and Taraskevich, 1978). These peculiarities are accounted for by the interruption of a tonic inhibitory network upon NIL removal. [The cases of reduced release probably reflect a mobilization by potassium of dopamine from severed fibers (e.g., Randle et al., 1983b).] Indeed, it is now established that dopamine is the primary substrate neuronally (and, possibly, humorally via portal vessels) exerting a tonic inhibition upon IL release of a a-MSHIP-EP. Dopamine and congeners, in vitro, dose-dependently block spontaneous action potentials and the release of a-MSHIP-EP via an action at a D2 subtype of receptor (Cote et ad., 1982; Douglas and Taraskevich, 1978; Munemura et al., 1980; Vale et al., 1979; Vermes et al,, 1980). I n vivo studies have corroborated the picture of a dopaminergic brake upon IL a-MSHIP-EP secretion. Levels of these peptides in the systemic circulation are, thus, increased by dopaminergic antagonists upon chronic exposure to which a rise in the IL content of ir-P-EP and the amount of mRNA coding for POMC is seen (Chen et al., 1983; Hdlt, 1981; Hollt et al., 1982b). A direct IL site of action has been indicated and, in the dog, chromatographic analysis of

THE ENDOCRINOLOGY OF THE OPIOIDS

19

the ir-P-EP occurrent in plasma and the lack of effect of dexamethasone has been forwarded as evidence that the IL is the exclusive origin of this ir-P-EP (Sharp et al., 1982a,b). However, in rats in uiuo, although haloperidol clearly disinhibits IL release, it also promotes a concomitant dexamethasone-preventable rise in ACTHIP-LPHIP-EP from the AL, probably due to a blockade of hypothalamic a-adrenoreceptors (Giraud et al., 1980; Hollt, 1981; Oliver et al., 1976). A population of P-adrenoceptors exists in the IL, the activation of which by adrenaline or analogs dose dependently augments a-MSHIPEP release in a propranolol-reversible fashion (Cote et al., 1982; Tilders et al., 1980, 1981; Vermes et al., 1980).This is an interesting counterpart to the a-adrenoceptor stimulation of AL ACTHIP-EP release, and the adrenal medulla may be the source of these ligands. In response to certain stressors, an activation of these P-adrenoceptors may mediate the mobilization of IL a-MSHIP-EP (Berkenbosch et al., 1983).I n viuo, via a direct IL action, isoprenaline or adrenaline are also potent stimulators of the dexamethasone-resistant outflow of a-MSHIP-EP (Berkenbosch et al., 1981). However, a dexamethasone-blockable rise in ACTH levels is also seen originating from the AL. This influence on the AL is apparently underlain by an action in the brain (possibly an activation of the angiotensin II-VP axis rather than an occupation of AL a-adrenoceptors) (Knepel et al., 1980a, 198213). In contrast to most previous studies of hypothalamic CRF extracts, the 4 1 -residue CRF was found to stimulate IL a-MSH release via a direct action on this lobe wherein fibers of brain origin containing CRF appear to exist (Al-Noaemi et al., 1982; Meunier et al., 1982; Meunier and LabriC, 1982; Proulx-Ferland et al., 1982; Saavedra et al., 1984). However, the physiological significance of such an action remains to be clarified (Conte-Devolx et al., 1983). IL release of irP-EP is, in any case, refractory to VP (Przewlocki et al., 1979b). Interestingly, both P-adrenoceptor occupation and CRF lead, via an induction of adenylcyclase, to an accumulation of CAMP,the transducer of their actions (Cote et al., 1982; Meunier and LabriC, 1982). Indeed, addition of CAMP analogs alone stimulates IL release of P-EP (Vale et al., 1979). Dopamine, in contrast, depresses basal cAMP and antagonizes the increases in cAMP induced by P-agonists and CRF (Meunier and Labrie, 1982). Recent studies are, in line with older data, evidential of a primarily inhibitory impact of GABA upon IL melanotroph secretion, although the nature of this is rather complex and a dual population of GABAA and GABAB receptors may either, respectively, promote or suppress release (Demeneix et al., 1984; Hadley et al., 1977; Tomiko et al., 1983; see Taraskevich and Douglas, 1982; Vincent et al., 1982; J. P. Loeffler, personal communication). Although it has been postulated that nor-

20

MARK J . MILLAN A N D ALHEKT HEKZ

adrenaline from sympathetic neurons might activate DP-or P-receptors in the IL, such a possibility has not received experimental substantiation (Briaud et al., 1979; Kraicer et al., 1980; Munemura el al., 1980; Voigt et al., 1978). Since the great majority of studies have not confirmed initial reports of an inhibition by Pro-Leu-Gly (an oxytocin fragment) and have failed to detect this in rat brain, the status of this and related putative melanocyte-inhibiting factors is in great question (Manberg et al., 1982; Thody et al., 1980). The role of acetylcholine and serotonin in the control of IL melanotrophic secretion has not, as yet, been satisfactorily evaluated. Depending on the species examined, experimental conditions employed, and concentrations applied, either a promotion of or a lack of effect upon the release of a-MSH and related peptides has been observed (see Briaud et al., 1979; Conte-Devolx et al., 1981; Douglas and Taraskevich, 1978; Fischer and Moriarty, 1977; Hadley et al., 1977 Kraicer et al., 1980; Thornton and Geschwind, 1975; Voigt et al., 1978). Further, serotonin was documented not to modify the spontaneous action potentials manifested by IL cells, and most pertinently, a failure of serotonin or cholinomimetics to influence IL release of ir-P-EP has been documented (Douglas and Tarakevich, 1978; Vale et al., 1979; Vermes et al., 1980; Jackson and Lowry, 1983; Randle et al., 1983a). Very recently, Laniberts el al. ( 1 983b) did, in fact, provide evidence for a serotoninergic control of IL P-EP outflow but, on balance, acetylcholine and serotonin (in contrast to dopamine, P-adrenoceptor agonists, and, probably, GABA) appear to be of minor significance in control of IL secretion. An important distinction to the AL is the relative insensitivity of IL release of a-MSHIP-EP to glucocorticoids. However, there may be minor or secondary actions on the IL, and exogenous corticosteroids may slightly, but reproducibly, inhibit IL synthesis of POMC (F. Berkenbosch, personal communication; Eberwine and Roberts, 1984; Fischer and Moriarty, 1977; Imura et ul., 1982; Schachter et al., 1982; Vale et al., 1979). Finally, the possibility of a mineralocorticoid-specific in uivo action upon IL pools of ir-P-EP has been raised (Lim et al., 1983a).

E. MODULATIONOF OPIOID PEPTIDESYSTEMS UNDER PHYSIOLOGICAL A N D OTHER C~NDITIONS 1. P-Endorphin a. Circadian Rhythmicity. Basal levels of ACTHIP-LPHIP-EP in the plasma of rats are not constant but manifest an apparently parallel, circadian rhythm (Takahishi et al., 1981; Vuolteenaho et al., 1982). Con-

THE ENDOCKINOLOCY O F T H E OPIOIDS

21

sistant with this rhythm, circadian changes in AL ir-P-EP have been seen in rats (Kerdelhue et al., 1983); these variations in the AL content of ACTHIP-LPHIP-EP and in corticotrophic secretion may be related more to the diurnal rhythmicity of CRF in the hypothalamus arid synthesis of POMC in the AL, and are ultimately coordinated by the suprachiasrnatic nucleus (see Imura et al., 1982; Suda et al., 1979; Hiillt et al., unpublished). An IL rhythmic outflow of P-EP is suggested by changes in its levels in this lobe seen during the day and the circadian rhythmicity displayed by a-MSH (Gibson et al., 1983; Monnet et al., 1981; Kerdelhue et al., 1983; 'I'ilders and Smelik, 1975; Vuolteenaho et al., 1982). A parallel diurnal rhythmicity in plasma of AC'I'H, P-LPH, and P-EP (which was not clearly coupled to sleep patterns) has also been detected in man (Baser, 1981; Dent et al., 1981; Panerai et nl., 1982; Petraglia et al., 1983; Shanks et al., 1981). b. Pain, Stress, and Shock. There is evidence, in fact, that both the AL and IL contribute to basal circulating levels of ir-P-EP in rats (Berkenbosch et al., 1983; Cahill et al., 1983; Evans et al., 1982; Mueller, 1980; see Millan et al., 1981a; Przewlocki et al., 1982). Original studies indicated that systemic stressors (such as ether) and neurogenic stressors (such as leg break) preferentially activate, respectively, corticotrophic and melanotrophic secretion. Indeed, dependent upon the model utilized and the parameters selected, a participation of both AL and IL pools of P-EP and related species in the response to acute stressors has been indicated (Berkenbosch et al., 1983; Millan et al., 1981a, 1984d; Przewlocki et al., 1982; Vernies et al., 1981). Indeed, in animals, the imposition of a broad range of stressors or related stimuli including pain, laparotomy, ether, and insulin hypoglycemia leads to a liberation of ir-P-EP into the systemic circulation; this may be regarded, as with ACTH, as a characteristic stress reaction (e.g., Farrell et al., 1982; Mueller, 1981; see Millan, 1981, and Olson et al., 1982). In the AL of rats under acute stress, in addition to a promotion of release, evidence for an acceleration in the biosynthesis and processing of POMC to P-EP has been acquired (Shiomi and Akil, 1982). Similarly, in man, a diversity of acute stressors such as surgery, extreme exercise, or fear elicits a rise in circulating levels of ir-P-El' (see Millan, 1981; Olson et al., 1982). Interestingly, in man, plasma levels of ir-P-El' measured in the course of surgery significantly predicted postoperative morphine requirements (Cohen et al., 1982). I t may be mentioned briefly that a contribution of an enhanced secretion ofP-EP under stress to such associated physiological changes as antinociception and hyperthermia seen in animals and man is rather questionable, a problem discussed elsewhere (Millan, 1981).

22

MARK.]. MILLAN AND ALBERT HEKZ

The consequences of chronic, recurrent rather than acute, individual encounters with stressors is, of course, of more than academic interest. It is probable that the discharge of ir-P-EP into the circulation undergoes adaptation upon repetitive exposure to a particular stressor (Millan et al., unpublished). However, recent studies have revealed longterm changes in various tissue pools of ir-P-EP, the nature of which appears to be dependent upon the type of stressor imposed. Thus, Shiomi and Akil(l982) observed an enhancement in the production of POMC in the IL of rats subjected to recurrent foot shock. In contrast, rats suffering from chronic arthritic pain manifested no alteration in IL levels of ir-P-EP but a pronounced increase in these in the AL (Millan et al., l984b). Since this was paralleled by an elevation in plasma levels of ir-P-EP, it appears to reflect a facilitation in the synthesis and secretion of AL P-EP in response to chronic pain. However, ir-P-EP has not invariably been found to be depressed in the plasma of patients experiencing chronic pain; thus patients afflicted with, e.g., persistent headaches o r arthritic or rheumatoid conditions have exhibited variable decreases, increases, or no change in the plasma content of ir-P-EP (Atkinson et al., 1983; Baldi et al., 1982; Denko et al., 1982; Facchinetti et d., 1981; Nappi et ul., 1982a,b; Panerai et al., 1982). It is indubitably of importance to determine the reasons (e.g., differences in therapy, duration, and precise nature of pain, etc.) underlying these contrasting changes. In view of the current attention to the influence of acupuncture (low frequency, high intensity, peripheral stimulation) upon nociceptive processing and its therapeutic efficacy in the alleviation of pain, its influence upon circulating P-EP is of interest. In the horse, human volunteers, and patients undergoing surgery, an increase in circulating levels of ir-P-EP was elicited by acupuncture (Abbate et al., 1983; Bossut et al., 1983; Malizaia et al., 1979; Masala el al., 1983; Nappi et al., l982b) [incidentally, the surgical incision was noted not to produce a further rise (Masala et al., 1983)]. However, notwithstanding the significant improvement in pain symptomatology effected in chronic pain patients by acupuncture, there was no accompanying elevation in plasma ir-P-EP in the study of Kiser rt d., (l983), and such a rise is not invariably seen (Szczudlik and Lypkca, 1983). Further, in rats, a naloxone-sensitive ac~ipunrture-induced antinociception in the apparent absence of a rise in circulating P-EP levels has been seen (Pert rt ul., 1981). Thus, as with acute stress, a causal interrelationship between the inlluence of acupuncture upon nociception and secretion of ir-pEP is very questionable. The rises in circulating ir-P-EP accompanying electroconvulsive shock and spinal, hypovolemic, septic, or endotoxin shock are presum-

THE ENDOCRINOLOGY OF THE OPIOIDS

23

ably related to their “stressful” properties (Alexopoulos et al., 1983; Carr et al., 1982; Holaday et al., 1981; Holaday, 1983; Rees et al., 1983). c. P r e p a n c y . Although in the course of pregnancy there are divergent reports as to whether women exhibit elevated levels of plasma ir-pEP, it is accepted that these reliably rise coincident with the onset of labor and delivery, which are, needless to say, exceedingly painful and stressful events. Whether the placenta contributes to maternal plasma irP-EP is unresolved (e.g., Akil et al., 1979; Csontos et al., 1979; Goland et al., 1981; Kimball et al., 1981; Stark and Frantz, 1983; Steinbrook et al., 1982). Elevated levels are also seen in man and animals in fetal plasma which are not of a maternal origin and may fluctuate independently thereof (Goland et al., 1981; Wardlaw et at., 1981; Stark and Frantz, 1983; Steinbrook et al., 1982). Indeed, P-EP may be synthesized by both the fetus and the P-EP-impermeable placenta, and ir-P-EP in the fetal circulation may respond to stress in that its levels are, in analogy to adults, positively correlated with the degree of hypoxia experienced (Chernick and Craig, 1982; Csontos et al., 1979; Goland et al., 1981; Kimball et al., 1981; Liotta et al., 1982; Stark et al., 1982; Stark and Frantz, 1983; Wardlaw et al., 1981; Yanagida and Corssen, 1981). There are, however, conflicting data as to whether ir-P-EP levels rise in amniotic fluid in fetal distress (Gautray et al., 1977; Petrucha et al., 1983; Riss and Bieglmayer, 1983; Stark and Frantz, 1983). d . Reproductive Status. Many authors have attempted to relate changes in pituitary P-EP levels to patterns of reproductive activity (Forman et d., 1983b). For example, a rise in IL and plasma ir-P-EP was apparent on the afternoon of proestrus in the studies of Ishizuka et al. (1982) around the time of the surges in circulating prolactin and luteinizing hormone which are known to be subject to an opioid inhibition (Section III,A,l and 3). Consonant with a possible modulation by sex steroids, gonadectomy and/or gonadal hormonal therapy result in a complex pattern of sex-dependent modifications in IL, AL, and plasma levels of ir-P-EP (Forman et al., 1983a; Lee et al., 1980; Lim and Funder, 1984; Mueller, 1980; Petraglia et al., 1982; Tejwani et al., 1983). Further, estradiol attenuates the stress-induced rise in ir-P-EP in the plasma of male rats (Mueller, 1980). Changing patterns of reproductive function may be related to certain age and sex differences in the pituitary content of ir-P-EP. For example, a prepubertal rise in pituitary ir-P-EP has been seen in rats, and menopause was paralleled by a decrease in circulating ir-P-EP in women (Forman et al., 1983a; Genazzanni et al., 1981; Lee et al., 1980). In addition, aged rats have been reported to possess elevated contents of ir-P-EP in both the AL and IL in addition to plasma (Forman et al., 1981; Missale et al., 1983).

24

M A R K ,J. MILLAN A N D A L H K R T HERZ

Interestingly, 0-EP in the ovaries may also be regulated by gonadotropins (Shaha et al., 1984). e. Nutritional Status. The fact that in normal and foodshifted circumstances the circadian peaks of plasma ir-P-EP occur just prior to feeding concurs with behavioral data pointing to a relationship of opioids to control of ingestive behavior (Davis et al., 1983; Takahishi et al., 1981; see Morley et al., 1983b). Glucoprivation induced by 2-deoxy-glucose results in a rise in plasma ir-P-EP in the rat (Davis et al., 1983; Yim et al., 1981). Insulin-elicited hypoglycemia is, in contrast, only marginally effective, but has been observed to initiate a pronounced elevation in plasma ir-P-EP in man (Davis et al., 1983; Nakao et al., 1978). Food deprivation does not, in fact, result in significant alterations in the pituitary content of ir-P-EP, and a role of circulating P-EP in control of ingestive behavior has been questioned (Gambert et al., 1980; Konecka et al., unpublished; Wallace et at., 1981). Incidentally, an increase in levels of ir-P-EP extracted from the gut has been detected in fasted rats, and these gut pools proposed to be activated in response to ingestion of food (Matsumura et al., 1982a,b; Orwoll and Kendall, 1980). A number of authors have observed that pituitary (probably IL) and (in certain cases) plasma pools of' ir-P-EP of congenitally obese mice and rats show partially sex- and age-dependent elevations in levels of ir-/3-EP (Garthwaite et al., 1980; Gibson et al., 1981; Govoni and Yang, 1982; Gunion and Peters, 1981; Margules et al., 1978; Rossier et al., 1979b; Wallace et al., 1981). Whether this is causal of or even related to hyperphagia is controversial; nongenetic experimentally induced hyperphagia is not, in general, associated with comparable changes (Gunion and Peters, 1981; Millan et al., 1982, 1983a). Interestingly, in a study of obese hyperandrogenic women, elevated plasma levels of ir-P-EP positively correlating with body weight were detected (Givens et al., 1980). f . Blood Pressure. In a different congenital deficiency, that of spontaneous hypertension, Hutchinson el al. (198 1) observed higher and lower levels of ir-P-EP in, respectively, the NIL and plasma. However, hypertensive rats also reveal alterations in CNS opioid peptide levels, and the relationship of any pituitary shifts to the naloxone-susceptible hyperteiision is unclear (see also Kouchich et al., 1983; Zamir et al., 1980). In man, hypertensive patients did not differ in basal plasma levels ir-P-EP, but in contrast to normotensive controls manifested a marked rise in this in response to clonidine (Farsang et al., 1983). g. Drugs. In view of the similarities of certain of the behavioral and physiological effects of alcohol and opioids, which may even show cross tolerance, and that naloxone blocks certain of the actions of alcohol, the influence of ethanol treatment upon hypophyseal pools of P-EP is of special interest (see, e.g., Hluni, 1980; Naber et al., 198111). Acutely ap-

THE ENDOCRINOLOGY OF THE OPIOIDS

25

plied to rats it does not, in fact, influence levels of ir-B-EP in either the AL or IL of rats, although a naloxone-preventable rise in plasma ir-P-EP has been seen (Allen et al., 1981; Schulz et al., 1980a; Seizinger et al., 1983). Chronic exposure of rats to nontoxic (in contrast to toxic) doses similarly fails to modify AL or 1L ir-P-EP in rats, although the plasma content under such conditions has not been reported and a change in this was seen in mice (Crabbe et al., 1981; Schulz et al., 1980a; Seizinger et al., 1983). Indeed, in man, neither in volunteers upon acute ingestion nor in alcohol addicts has an alteration in plasma ir-P-EP been observed (Genazzanni et al., 1982; Kimball et al., 1980; Naber et al., 1981a), despite the fact these addicts manifest a dysfunction in corticotrophic secretion of ACTH (see Knych and Prohaska, 1981; Schulz et al., 1980a). Studies of the biosynthesis and maturation of POMC in rats have, nevertheless, revealed important consequences of chronic consumption of ethanol, the mode of intake of which may be a critical variable. Thus, in both the AL and IL, an enhancement of the de novo synthesis of POMC is detected. However, as in the AL, the ratio of P-LPH to P-EP and, in the IL, of acetylated P-EP to P-EP is also in each case accentuated, and the functionally opioid-active pool of P-EP is reduced (Seizinger et al., 1983). Further, a parallel rise in the in vivo IL spontaneous outflow of ir-P-EP which might counterbalance an elevated synthesis of total P-EP and explain the lack of changes in steady state levels of ir-P-EP therein has been reported by Gianoulakis et al., (1981). Certain of the effects of amphetamine are, in analogy to alcohol, also sensitive to blockade by naloxone (see Schulz et al., 1980b). Chronic (but not acute) injection of amphetamine evoked a rise in AL, in contrast to NIL, ir-P-EP in guinea pigs while a dextroamphetamine-induced parallel rise in plasma ir-P-EP and CS was observed upon a single application in man (Cohen et al., 1981; Schulz et al., 1980b). These observations are indicative of an interaction with AL pools of P-EP. A further finding of note is that the intraarterial infusion of caffeine evoked an elevation in circulating ir-pEP in rats (Arnold et al., 1982). This was shown to be a CNS- rather than pituitary-mediated action and proved, interestingly, to be antagonizable by naloxone. Finally, presumably via a facilitation of GABA transmission, benzodiazapines attenuate stimulated P-EP release (Britton et al., 1983). h. Affective Disorders. In the clinical domain, patients with disorders of ACTH secretion such as Cushing’s disease exhibit elevated levels of irP-EP in plasma (Akil et al., 1979; Besser, 1981; Krieger et al., 1980). It would presently be premature to recognize any “abnormalities” in ir-pEY levels in the circulation of patients suffering from affective disorders (cf. other hormones) as a useful marker for particular syndromes. However, it has been indicated that a subgroup of schizophrenics may possess

26

MARK J. MILLAN AND ALBERT H E R 2

elevated plasma ir-P-EP and/or a rise in these under haloperidol therapy. while heniodialysis appears not to clear P-EP (or a putative Leu+ EP) from the circulation (Emrich et al., 1980; Naber et al., 1982, 1983; Ross et al., 1979; Van Ree and de Wied, 1981). Original investigations, in line with ACTH data, found no major changes in plasma ir-P-EP in endogenous depressive patients, although diurnal rhythmicity may be of relevance and a rise upon switching to mania was reported (see Alexopoulos et al., 1983; Brambilla et al., 1981; Cohen et al., 1984; Fang et al., 1981; Pickar et al., 1980). A recent study, however, observed elevated levels and an augmented physostigmine-induced rise in ir-P-EP; on this basis, a cholinergic (muscarinic) supersensitive control of P-EP release in depression was postulated (Risch, 1982; Risch et al., 1983). In analogy to cortisol (but possibly not ACTH), a lower likelihood of escape from dexamethasone suppression of ir-P-EP levels was seen in depressed subjects (Fang et al., 1981; Matthews et al., 1982). Although the results which have hitherto been forthcoming are to a large extent disappointing, it should be superfluous to emphasize the need for considerable additional study in this field. 2. Dynorphin NIL ir-DYN undergoes a circadian rhythmicity in its levels, these being high during the day and low at night; the hypothalamus, interestingly, shows a reciprocal pattern (Przewlocki at al., 1983). Daytime NIL levels of ir-DYN have been found to be elevated by food deprivation and the NIL of obese mice to contain an abnormally high content of ir-DYN; thus, a relationship of these pools to ingestive behavior may be anticipated (Fergurson-Segall et al., 1982; Przewlocki et al., 1983). T h e depletion of NIL ir-DYN (plus ir-VP) induced by water deprivation reflects the classic response of neurohypophyseal neurons to dehydration (Przewlocki el al., 1983). Acute foot-shock stress fails to modify NIL irDYN or ir-VP (Millan et al., 1981b, 1984b), in contrast to NIL ir-P-EP. An examination of the probable parallelism of NIL-localized DYN with VP in the response to hemorrhage, hypoxia or other stimuli operating via noriosmotic control mechanisms should be undertaken. l’he relatively little information on AL ir-DYN available indicates a modulation independent of this NIL pool; thus, a diminution in response to foot shock and a lack of effect of dehydration has been seen (Hiillt el al., 1981b; Millan et al., l98lb). Further, and interestingly, chronic arthritic pain is associated with a pronounced rise in AL ir-DYN, i.e., a shift in the opposite direction to that produced by acute footshock (Millan et al., 198413). Where measured, levels of ir-a-NE and ir-DYNI-8 have been

THE ENDOCRINOLOGY OF THE OPIOIDS

27

modulated in parallel with those of ir-DYN in both the AL and NIL (Millan et al., 1984a,b). 3 . Enkephuliris In the NIL, levels of ir-LE, in line with its costorage with ir-DYN, behave comparably; for example, a reduction in dehydrated rats and an elevation in obese mice have been seen (Fergurson-Segall et al., 1982; Hijllt et al., 1981b). AL (in contrast to IL) pools of ir-ME and ir-LE reveal intersex differences. These AL pools are modulated by, or in relation to, circulating gonadal hormones as indicated in the alterations seen during the estrus cycle and produced by castration or administration of exogenous gonadal steroids (Hong et al., 1982; Kumar et al., 1979). I n analogy to P-EP, elevations in pituitary levels of enkephalins have been seen in aged rats (Kumar et al., 1980; Missale et al., 1983). Consonant with the independent origin of enkephalins as compared to P-EP in the circulation, the content of these displays differential fluctuations under certain conditions. For example, in contrast to ir-P-EP, plasma ir-ME is not suppressed to dexamethasone but is influenced by acute ethanol and manifests no circadian rhythmicity in its levels (Clement-Jones, 1982; Medbak et al., 1983; Panerai et al., 1982; Shanks et al., 1981; Yanase et al., 1984). Insulin, a typical stimulant of chromaffin catecholamine outflow, precipitates a pronounced decrease in the adrenal medulla content and a rise in plasma levels of ir-LE and ir-MErelated peptides (Ryder and Eng, 1981; Viveros et al., 1980). Stress is likewise a promoter of adrenal medullary mobilization of both catecholamines and ME-related peptides into the circulation (Alessi et al., 1982; Howlett et al., 1984; Lewis et al., 1982; Panerai et al., 1982). Further, hypovolemic shock produces a liberation of ir-ME and ir-LE from these cells, the levels of which are 10-fold higher in adrenal as compared to femoral vein plasma (Lang et al., 1982). Of related interest is the preliminary evidence that there is no reduction in levels of circulating ir-ME in patients suffering from chronic pain (cf. ,f3-EP) (Dodds et al., 1983; M. ‘r.Jones, personal communication; Kiser et al., 1983). Acupuncture was observed to elicit a rise, however, in plasma ir-ME in chronic pain patients which significantly and positively correlated with the degree of pain relief thereby achieved (Kiser et al., 1983). In this context, evidence for a relationship between adrenal medullary secretion and antinociceptive processes in rats exposed to noxious foot shock is of relevance (Lewis et al., 1982). Nevertheless, it would presently be unjustified to draw any firm conclusions concerning a putative role of circulating ME in the generation of antinociception. Interestingly, in the light of a possible role in modulation of blood

28

MAKK.]. MILLAN A N D ALBERT HEKZ

pressure (Section II,F,3), adrenal medullary ir-ME was shown to be lower in hyper- as compared to normotensive rats (De Wald and Lewis, 1983; Di Guilio et al., 1979).

TISSUES AND POSSIBLE FUNCTIONS F. TARGET 1 . Local Actiom It is important to note that opioids might subserve functions at o r adjacent to their site of release. These are of great potential significance in opioid mechanisms of control of endocrine secretion and are discussed in Section 111, which may be consulted for references. For example, the possibility of a direct feedback of ME/LE release from terminals in the NL upon their own outflow and that of costored O T has been raised. An intmlobe role of AL P-EP in promotion of prolactin secretion has also been considered. Further, the vascular interconnections between the lobes of the pituitary allow for mutual exchange of substances and offer an anatomical basis for interlobe interactions. There are,' for example, indications for an inhibitory impact of AL-derived P-EP upon VP neurons in the NL, and contrariwise, for a facilitatory impact of NLoriginating VP upon AL corticotrophs secreting ACTH and P-EP. 2 . Actions on the Brain or Median Eminence Perhaps the simplest evidence in support of an influence of opioids upon the operation of the brain is the capacity of systemically administered opioid peptides, in both animal and clinical studies, to modify not only endocrine and autonomic parameters but also complex behaviors and moods integrated in higher centers of the brain (see Olson et al., 1980). The structural organization of the blood-brain barrier (BBB) renders it essentially impermeable to hydrophilic molecules, including neurotransmitters and peptides. Further, a failure to find specific uptake mechanisms for enkephalins in the BBB has been documented (Partridge and Mietus, 1981). Rather, enkephalinase activity was detected in brain capillaries which, via an acceleration of degradation, would counter enkephalin penetration. Indeed, although still a matter of controversy, most investigations have indicated that parenterally applied opioid peptides will only cross the BBB to a minor degree under resting conditions. This penetration is, nevertheless, often significant with an accumulation in CSF and, although less clearly, in nervous tissue internal to the BBB (see Gerner et al., 1982; Merin et al., 1980; Ohlsson el ul., 1982; Partridge and Mietus, 1981; Rapoport et al., 1980). It is, further, necessary to make the following qualifications.

THE ENDOCRINOLOGY OF THE OPIOIDS

29

1. T h e BBB is a dynamic structure, the permeability of which may display fluctuations arid is capable of modulation. Indeed, many of the various experimental conditions under which an enhanced penetrability has been seen are characterized as associated with a rnobilization of particular opioids; these include seizures, immobilization stress, tissue injury, and hyperosmolarity (Belova and Jonsson, 1982; Pechura et al., 1982). Thus, under physiological conditions in which opioids are activated, an enhanced penetration could occur (see below), a possibility requiring experimental verification or refutation. 2, Opioids, via the systemic circulation, may interact with brain opioid receptors located externally of the BBB. Of special interest are, for example, the subfornical organ and area postrema which lie in close proximity to autonomic regulatory centers in, respectively, the hypothalamus and the brainstem. Further, the median eminence and mediobasal arcuate hypothalamus possess opiate receptors, the activation of which by central or circulating opioids may be involved in endocrine regulation (see Section 111). Indeed, an uptake of systemically infused opioid peptides by this region has consistently been seen (see above references). 3. I n agreement with the above possibility, anatomical studies have provided evidence that portal plasma may be able to flow in a retrograde direction from the pituitary toward the brain (see Bergland and Page, 1979; Page, 1982; Palkovits and Mezey, 1981). In fact, a considerable portion of the drainage from the AL is via the NL and a majority of the possible exit routes from the NL might allow for the transport of plasma to the brain. Moreover, in rats the portal vessels have been found to contain very high quantities of peptides, including ACTH, a-MSH, and VP, which are greatly reduced by hypophysectomy (Lissitzky et al., 1980; Porter et al., 1981), indicative of a possible primarily pituitary origin. Consistent with these findings, intrapituitarily injected ACTH, P-LPH, or neurotensin has been reported to attain the CNS via a vascular stalk pathway (Mezey et al., 1981; Palkovits and Mezey, 1981). Subsequent to reaching the median eminence via a vascular route (which may also be attained by systemic blood via the portal vessels), the arcuate nucleus is accessible by connecting arterioles. Further, access to the CSF and hence nervous tissue might occur by either retrograde transport in axons projecting to the median eminence or diffusion through the fenestrated median eminence capillaries to the CSF via the pericapillary space o r tanycyte cells (see Bergland and Page, 1979; Palkovits and Mezey, 1981). Supplementary direct pituitary-to-CSF transfer may also occur. Nevertheless, further studies of the putative physiological occurrence and role of such mechanisms of access are definitely required. Of special note is the possibility that under certain conditions, e.g., stress,

30

MARK J. MILLAN A N D ALBERT IIERZ

this access might be potentiated. Indeed, in contrast to the basal state, electrical seizures resulted in a substantial entrance of pituitary ACTH (and, presumably, P-EP) into the brain in sheep, possibly via a transiently open BBB (Bergland et al., 1980).

3. Artions in the Puriphery It is possible that opioids discharged into the systemic circulation may act in a conventional hormone-like fashion to activate receptors resident in remote tissues. Indeed, opioid mechanisms are participants in the control of the motility and secretion of the gut (see Konturek, 1980) and may modify steroidogenesis in the adrenal cortex and stimulate the release of insulin and glucagon from the pancreas (see Section 111,D). In addition, a diversity of direct actions on, for example, the spleen, kidney, liver, and reproductive organs has been recorded. I n such studies, however, there is virtually no concrete evidence for a physiological involvement of circulating opioids. Thus, does an experimental effect of opioids, in vivo or in vitro, represent an action on opioid receptors activated in the physiological state by opioids arriving in the circulation, present in extrinsic innervation (e.g., the vagus) or released from intrinsic neurons? In certain cases, such as the gut or pancreas, direct evidence for an involvement of intrinsic systems has been obtained. Equivalent specific data for circulating opioids appear to be lacking. Further, concentrations of P-EP in plasma, even under stimulation, are substantially less than those documented as experimentally efficacious, although there may exist transport systems which release opioids only at their loci of action. CNS-occurrent opioids are established to be regulators of respiratory and cardiovascular function and, in addition to opioids in peripheral neurons, actions of circulating opioids may fulfill such a role (see Holaday, 1983). T h e intraatrial injection of opioids produces a bradycardia, apnea, and hypotension (or biphasic response) within 1 sec, reflecting an excitation of J-receptors in the alveoli of the lung (Willette and Sapru, 1982). In addition, ME and LE (but not P-EP) interfere with the chronotropic actions of noradrenaline on the isolated heart (Eiden and Ruth, 1982). Adrenal medullary-originating ME-related peptides are good candidates for such roles since adrenal venous effluent travels almost directly to the right heart and, thereby, to the pulmonary circulation. Further, stimulation of the splanchnic nerve in reserpinized dogs results in only a marginal outflow of catecholamines but a pronounced release of ME-related peptides accompanied by a rapid naloxone-reversible hypotension (Hanbauer et al., 1982). Moreover, the intravenous application of opioids may result in bradycardia and hypotension (Lemaire et

THE ENDOCRINOLOGY OF THE OPIOIDS

31

al., 1978; Wei et al., 1980). Holaday (1983) recently raised the possibility of a hormone-like action of adrenal ME-related peptides in the deleterious, naloxone-sensitive depressor effects of spinal shock. In addition, blood vessels may be optimally available to circulating opioids. Indeed, a peripheral vasodilatory effect of opioids may be related to their hypotensive and hypothermic properties. Both in vivo and in vitro, direct naloxone-attenuated vasodilatory actions of opioids upon arterioles are apparent (Hanko and Hardebo, 1978; Ronai et al., 1982; Wong et al., 1981). It is also possible that circulating opioids may interact with peripheral nerve endings (Bartho and Szolcsany, 1981; see Millan, 1982). Finally, of special note are data acquired with P-EP evidential of binding to andlor a modification of the activity of adipocytes, blood platelets, human complement, leucocytes, lymphocytes, and monocytes, targets particularly accessible to circulating ligands but not necessarily displaying opioid-like interactions (see below) (Berk et al., 1983; Gilman et al., 1982; Hazum et al., 1979; Kay and Morley, 1983; Lopker et al., 1981; Mehrishi and Mills, 1983; Rolf and Brune, 1981; Schwandt, 1981; Schwandt et al., 1981; Schweigerer, 1983; Schweigerer et al., 1982; Shavit et al., 1984; Van Epp et al., 1983; Wybran et al., 1979). Those opioids as yet established to be mobilized into the systemic circulation are released concomitantly with a costored nonopioid counterpart. It is highly improbable that such a parallelism is fortuitous, and a coordinated (synergistic or antagonistic) function of these in response to stimuli provoking their cosecretion may be anticipated. ACTH displays, in fact, a weak affinity for the opioid receptor, but the significance of this is very unclear (Akil et al., 1980). ACTH and opioids exhibit discordant actions and ACTH can antagonize the antinociception elicited by opioids, though not necessarily via an opioid receptor action (Smock and Fields, 1981; see Holaday and Loh, 1982). In fact, ACTH modifies many opioid-sensitive parameters and there are a number of' instances where opioids or ACTH (in certain cases, via corticosteroids) may act oppositely or modify each other's actions (see also Section III,D,2) (Baldwin et al., 1974; Desphand and Mitchell, 1983; Ferri et al., 1982b; Fratta et al., 1981; Gilman et al., 1982;Jacquet, 1977; Kasson and George, 1983; Racz et al., 1982; see Holaday and Loh, 1982).Further, aMSH also affects certain opioid-sensitive measures and may, in some cases, also act differently to opioids in, for example, its influence upon hypophyseal secretion of hormones such as prolactin and luteinizing hormone (Alde and Celis, 1980; Gluckman et al., 1980; Khorram et d., 1982; Lichtensteiger and Monnet, 1979; Miyake and Yen, 1981). The ability of enkephalins to antagonize the chronotropic actions of catecholamines on the heart and the opposite pressor and depressor actions

32

MARK.J. MILLAN

A N D ALRERL' MERZ

of, respectively, adrenaline and opioids similarly points to functional opioid-catecholamine interactions (Eiden and Ruth, 1982). Evidently, the possibility of functionally antagonistic or other interactions between opioids and their coreleased counterparts, between which there may be a critical regulatable balance, demands a fuller exploration, in particular as concerns extra-CNS loci. It is interesting to compare these relationships to the opposite metabolic roles of the independently localized and differentially modulated pancreatic hormones, insulin (anabolic) and glucagon (catabolic). It is arguable that specific nonopioid actions of P-EP may be of significance. For example, P-EP exhibits high-affinity specific binding to transformed lymphocytes (the behavior of which may be modified by P-EP) and human complement via a C-terminal, nonopioid action (Gilman et al., 1982; Hazum et al., 1979; Kay and Morley, 1983; Mehrishi and Mills, 1983; Schweigerer, 1983; Schweigerer et al., 1982). The lipolytic, glycerol-liberating action of P-EP might similarly be a nonopioid C-terminal action, though there are discrepancies in the literature as concerns this phenomenon (Richter et al., 1983; Schwandt, 1981; Schwandt et al., 1981). Further, actions of opioids upon steroidogenesis in the adrenal cortex and upon chromaffin secretion might also be nonopioid-like (see Section 111,D). A related observation is that DYN exerts major nonopioid-like actions in the CNS (Walker et al., 1982). Thus, naloxoneresistant effects should not be dismissed as artifactual or irrelevant but may be indicative of genuinely important nonopioid roles of circulating ligands. Further, it is worth recalling that the majority of P-EP in the IL is opiate inactive and this N-acetylated P-EP may, in analogy to AL pLPH, possess a quite distinctive spectrum of functional activities. Moreover, it is conceivable that, via its N-terminus, P-EP subserves an opioidlike neuromodulator neurotransmitter function in the CNS whereas, via its C-terminus, it may exert nonopioid hormone-like actions upon peripheral tissues attained through the systemic circulation.

111. Opioid Mechanisms in the Control of Endocrine Secretion

A. OPIOID CONTROL OF ANTERIOR LOBESECRETION A major strategy in the elucidation of the role of opioids in the regulation of adenohypophyseal secretion has been to evaluate the influence of systemic application of morphine o r other agonists upon levels of particular hormones in the plasma of rats under resting conditions

33

THE ENDOCRINOLOGY OF THE OPIOIDS

(see Meites et al., 1979, and below for individual references). The pattern of effects acquired (see Table I) has been reproduced by the intracerebroventricular (icv) administration of particular opioid peptides such as P-EP or enkephalin derivatives stabilized against enzymatic degradation. These actions are preventable by selective opiate antagonists, exhibit dose dependency, and develop tolerance. Opiate antagonists, contrariwise, in moderate doses, produce an opposite pattern of alterations in circulating levels of these hormone (Table I). These effects of antagonists are suggestive of a tonic control of AL secretion by opioid mechanisms under basal conditions. A participation of opioid systems in control of AL secretion raises more detailed questions as concerns 1. Sites and mechanisms of action 2. T h e particular opioids/opioid receptor types involved 3. T h e physiological conditions under which opioid mechanisms are active 1. Luteinizing Hormone ( L H ) a. Sites and Mechanisms of Action. The median eminence contains substantial amounts of LH-releasing hormone (LHRH), a decapeptide; this originates predominantly from the medial preoptic nucleus and, possibly, partially from the arcuate region (Elde and Hokfelt, 1979; Hokfelt et al., 1978; Palkovits, 1982; Popolov et al., 1980). Although alterations in opioid activity may modify binding of LHRH in the AL (Barkan et al., 1983), the hypothesis that opioids suppress LH secretion via actions in the AL is countered by the following arguments. First, opioid agonists or antagonists are unable, i n vitro, to modify AL secretion of LH either in the presence or in the absence of LHRH and d o not disrupt the secretory response to LHRH in vivo (Cicero et al., 1977, 1979; Ferin et al., 1982; Grossman et al., 1981a; Marko, 1982). Second, application of opioid agonists or naloxone to discrete brain regions (specified below) TABLE I THEEFFECTSOF O P I O I D A c o N r s r s (SUCH AS MORPHINE)O R A N I A C O N I S T S (SUCH AS NALOXONE) UPON 1-EVELS OF PARTICULAR ADENOHYPOPHYSEAL HORMONES IN T H E ~ONDITIONS" SYSTEMIC CIRCULATION OF ADULTRxrs U N D E R RESTINC;

Op~oidagonists Opioid antagonists

Luteinizing hormone

Folliclestimulating hormone

Prolactin

Growth hormone

-

rt_

+

+

+

+

+, Increase; -, decrease; +, no clear effect.

-

-

Thyroidstimulating hormone -

?

34

M A R K J . MILLAN AND ALBERT HER2

can depress serum LH (Johnson et al., 1982; Kalra, 1981).Third, LHKH antagonists blunt naloxone-elicited rises in plasma LH (Blank and Roberts, 1982). Fourth, morphine suppresses secretion of LHRH into pituitary portal plasma (Ching, 1983). Attention is thus focused on an opioid modulation of LHRH which appears to be expressed at multiple sites (see Fig. 5A) including the medial preoptic and arcuate-median eminence regions of the hypothalamus (Grandison et al., 1980; Kalra, 1981; Panerai et al., 1983; Schulz et al., 1981). Dopamine is a potent promoter of the in vztro release of LHRH from mediobasal hypothalamic tissue. In the work of Kordon and colleagues (Drouva et al., 1981; Kordon et al., 1979; Rotsztejn et al.,

A

CIRCULATION

CIRCUL ATlON

FIG.5. (A) Opioid mechanisms in the control of the secretion of luteinizing hormone (LH). Abbreviations: DA (dopamine), 5-HT (serotonin), LHRH (luteinizing hormone releasing hormone), N A (noradrenaline). The question mark above NA indicates that this may partially mediate opioid actions upon LH but that the precise interrelationship between N A and opioid actions is unclear (see text). (B) Opioid mechanisms in the control of the secretion of prolactin (PRL). In this figure and in Fig. 6, the schema presented are of necessity simplified models rather than definitive characterizations of established mechanisms of control. Question marks indicate tentatively drawn interactions demanding confirmation. Arrows do not necessarily imply direct actions. In instances represented by multiple arrows, there may be multiple opioid pools active; e.g., the opioid network acting via 5-HT in Fig. 5 may differ from that interacting directly with LHRH.

THE ENDOCRINOLOGY OF THE OPIOIDS

35

1978a,b), opioids proved inactive alone but powerfully attenuated the dopamine- o r potassium-elicited stimulation of LHRH release. Naloxone reversed this action and was inactive alone. In independent studies, further, naloxone alone promoted the release of LHRH in this tissue while the K-agonist bremazocine reduced bioassayable LHRH in hypothalamic fragments (Kalra and Kalra, 1984; Marko, 1982; Wilkes and Yen, 1981). With respect to dopamine, the action of opioids was postulated to be upon LHRH neurons, resulting in a disruption of its ability to instigate LHRH release. Consistent with this supposition, tentative evidence for the existence of opioid-binding sites upon LHRH neurons has been presented (Rotsztejn et al., 1982). In addition, an opioid reduction in the release of dopamine may be of importance in LHRH control (see Section III,A,3,a). However, microinjection and lesion studies of the dorsal raphe and amygdala have indicated that an engagement of opioid mechanisms in these structures can also inhibit the outflow of LH into the circulation (Johnson, 1982; Lakoski and Gebhart, 1981, 1982; Parvisi and Ellendorf, 1980; but see Wiesner et al., 1984). At least for the raphe, the serotoninergic pathway ascending therefrom to the mediobasal hypothalamus and median eminence may represent the intervening pathway. Thus, dorsal raphe stimulation suppresses serum LH and, in vitro, serotonin moderates hypothalamic outflow of LHRH (Arendash and Gallo, 1978; Kordon et al., 1979). Indeed, opioids may enhance the activity of ascending serotoninergic neurons and hypothalamic serotonin turnover (Van Loon and de Souza, 1978; see Moore and Johnston, 1982; WeilFugazza et al., 1979). Further, pharmacological evidence for a serotoninergic mediation of opioid actions upon LH is available, at least for immature female rats (Blank and Bohnet, 1983; Ieiri et al., 1980b; Tach6 et al., 1979). However, this serotoninergic link requires further evaluation. As an additional complication, noradrenaline has been identified as a primarily excitatory regulator of, e.g., episodic LH secretion, and a disturbance of hypothalamic noradrenergic or adrenergic transmission interferes with the capacity of opioids to depress LH release via an action probably related to a-adrenoceptors (Adler and Crowley, 1984; Blank and Bohnet, 1983; Kalra, 1981; Kalra and Crowley, 1982; Kalra and Gallo, 1983; Kalra and Kalra, 1984; Kalra and Sinipkins, 1981; Koh et al., 1983; Van Vugt et al., 1981). Evidently, thus, at least three major neurotransmitter networks are involved in the actions of opioids upon LH in the rat (i.e., DA, serotonin, and noradrenaline), although their precise interrelationships and physiological conditions of activity remain to be resolved (see Fig. 5A).

36

MARK J. MILLAN A N D ALBEKT HEKZ

b. Opioids and Receptor Types. In general, a comparison of the effects of particular opioid peptides is not very informative since contrasting potencies may, for example, reflect differential diffusion or susceptibility to degradation. Indeed the opioids applied might not, in the natural state, physiologically attain and activate the relevant receptor population. Techniques for the selective manipulation of discrete opioid systems are evidently desirable. A recently developed approach of the use of highly specific and purified antisera for the neutralization of individual opioids has been exploited for an examination of opioid control of LH release in prepubertal female rats. T h e direct introduction of antisera against P-EP into the mediobasal hypothalamus produced, in analogy to naloxone, an elevation in levels of LH in plasma. Antisera to DYN were also effective, but less so, whereas those against ME were inactive. These studies (Schulz et al., 1981) implicate P-EP and DYN as regulators of LH release in immature female rats and, since injections in other regions proved ineffective, again identify the mediobasal hypothalamus as a major site of tonically active opioid control. Recently, Forman et al. (1983a,b) similarly demonstrated antiserum against P-EP to elevate LH in adult male rats. Recently, it was also attempted to evaluate which receptor type may be involved by a comparison of the effects of a relatively selective pagonist (morphine) to a K-agonist (ethylketocyclazocine, EKC) in immature female rats (Schulz et al., 1982a). Since these were more or less equipotent, it appears that both p- and K-receptors may be of significance. A contribution of K-types is substantiated by the reduction in serum LH produced by the K-agonist bremazocine and in the LH-releasing effects of WIN 44:441-3, a preferential K-antagonist (Marko and Romer, 1983; Pechnik et al., 1981), and is consonant with the above findings concerning DY N, a relatively selective K-ligand. Nevertheless, recent work has, contrariwise, suggested that (at least in ovariectomized females) it is p- rather than K-receptors which are of major importance (Pfeiffer et al., 1983). Thus, further studies are required to clarify this discrepancy. c. Physiologzcal Conditions of Opioid Control. In contending that opioids are powerful tonic inhibitors of LH secretion, it is essential to be aware of, for adult females, the cyclicity of release. Thus, naloxone is, in humans and monkeys, most effective during the luteal phase but inactive in the early follicular phase near menstruation (Blankstein et al., 1981; Browning et al., 1981; Quigley and Yen, 1980; Ropert et al., 1981). Indeed, in monkeys, these times correspond to, respectively, minima and maxima in hypothalamic P-EP secretion, as evaluated in portal

THE ENDOCRINOLOGY OF THE OPIOIDS

37

plasma (Wardlaw et al., 1980a, 1982b; Wehrenberg et al., 1982). Further, discrete hypothalamic tissues in rats may reveal shifts in their P-EP content at proestrus (around the time of the LH surge) including, for example, a rise in that of the median eminence (Barden et al., 1981; Knuth et al., 1983; Wardlaw et al., 1982a). Two factors of importance in an understanding of opioid modulation of LH release are, first, sex and, second, age. T h e LH-stimulating effect of naloxone is particularly striking in immature female rats, and changes in the magnitude of its actions between birth and maturity might be related to shifts in hypothalamic LHRH levels (Blank et al., 1979; Gabriel et al., 1983; Ieiri et al., 1979). Subsequent to puberty, naloxone is considerably less potent. Indeed, a progressive disengagement of opioid control might be related to the initiation of pubertal events, a possibility supported by the fact that gonadectomized immature females show age-related decreases in the LH-suppressive effects of opioids parallel in time course to the alterations seen with the feedback suppression of LH induced by gonadal steroids (Barkan et al., 1983; Bhanot and Wilkinson, 1983a,b; Wilkinson and Bhanot, 1982). Moreover the opioid-suppressible oscillations in LH levels (see below) shown by mature females emerge coincidentally at maturity with the moderation of opioid inhibition of tonic LH secretion. In immature males, however, naloxone is virtually ineffective in promoting LH release (Blank et d., 1979; Ieiri et d.,1979; Schulz et d., 1982a). This intersex distinction points to testosterone (plasma levels of which are higher in males) as a possible explanatory factor. Testosterone exerts a strong negative feedback brake on LH release, and its elimination by castration, in fact, reveals a potent LH-releasing action of naloxone in immature males (K. Schulz, personal communication). A “masking” effect of testosterone might, thus, account for this intersex difference. Clonidine, as with naloxone, promotes LH release in both male and female adult rats. However, in distinction to immature males which exhibit a pronounced rise in LH levels, immature females are comparatively irresponsive to this (Schulz et al., 1982a). Thus, in immature females there is a demonstratable opioid, but in males an adrenergic, control mechanism for LH in immature rats. It is well established that the proestrus surge of LH underlies the initiation of ovulation in rats. Opioids applied during the critical period of proestrus prevent the occurrence of ovulation via an interference with the requisite LH surge (Ieiri et al., 1980a; Koves et al., 1981; Packman and Rothchild, 1976; Pang et al., 1977). Further, naloxone potentiates the duration andlor magnitude of the proestrus-related elevation of

38

MARK J. MILI.AN A N D ALBERT HEKZ

LH levels which may result in an increase in the number of ova shed (Gabriel et al., 1983; Ieiri et al., 1980a; Koves et al., 1981; Marton et al., 1981). Recently, the possibility was raised that alterations in the activity of opioid networks (e.g., those involved in control of ovulation) might be related to, for example, the pathophysiology of amenorrhea, the lowered testosterone levels seen in senescent males, and the fact that aged females become anovulatory due to loss of cyclicity of LH levels (Blankstein et al., 1981; Forman et al., 1981; Grossman et al., 1982b; Meites, 1982; Melis et al., 1984; Reid et al., 1983; Steger et al., 1981). Further, a familiar clinical experience is the abnormally high incidence of unsuccessful fertilizations, etc., in human heroin addicts, which may partially reflect a disruption of LH control (see Meites et al., 1979). As mentioned above, naloxone facilitates pulsatile LH release during certain phases of the menstrual cycle in primates, including humans (Quigley and Yen, 1980; Ropert et al., 1981; Van Vugt et al., 1983; Veldhuis et al., 1983). Further, P-EP inhibited pulsatile LH outflow in male castrates, and in ovariectomized female rats, morphine and naloxone, respectively, depressed and augmented certain parameters of this pulsatile release (Gabriel et al., 1983; Kinoshita et al., 1980; Sylvester et al., 1982). Microinjection of morphine into the raphe, moreover, inhibited this pulsatile secretion (Johnson et al., 1982). Opioids may, thus, be participants in the coordination and control of pulsatile LH secretion. This is also known to be regulated by gonadal hormones, and in fact, naloxone counteracted the estrogen- and progesterone-induced suppression of this pulsatile release, leading to the conclusion that opioids may mediate sex steroid hormone negative feedback on LH (Sylvester et al., 1982). In support of the conjecture of an opioid involvement in sex steroid feedback on LH are observations of interactions of rialoxone and morphine with the effects of sex steroids upon LH; for example, testosterone and estrogen failed to inhibit LH release in naloxone-pretreated rats (Bhanot and Wilkinson, 1983a; Cicero et al., 1980; Foresta et al., 1983; Kalra and Simpkins, 1981; Van Vugt et al., 1982). Since naloxone alone (or gonadectomy) raise LH, these data must be interpreted with caution. Nevertheless, sex steroids have been indicated to stimulate the hypothalamic release of P-EP in addition to elevating ME levels in the medical preoptic hypothalamus (Dupont et al., 1980; Wardlaw et al., 1982a,b), and a further evaluation of a possible opioid mediation of sex steroid feedback is justified (e.g., Gabriel et al., 1983). Whether opioids play a role in the modulation of circulating levels of LH in man under stress requires further study. However, of interest is the ability of naloxone to antagonize the blockade of LH release and

T H E ENDOCRINOLOGY OF THE OPIOIDS

39

consequent antiovulatory effects of stress in rats indicative of opioidergic mediation of these stress effects (Briski et al., 1984; Hulse et al., 1982; Hulse and Coleman, 1983; Ixart et al., 1980; Pontiroli et al., 1982). Finally, P-EP was suggested to regulate lordosis behavior in the rat via an interaction with LHRH in the central gray (Siranath-Singhe et al., 1983). 2. Follicle-Stimulating Hormone ( F S H ) In addition to LH, the release of FSH is promoted by LHRH, an action not susceptible to blockade by opioids (Grossman et al., 1981a; Pang et al., 1977). However, under a variety of conditions the secretion of LH and FSH may be dissociated, evidential of an at least partially independent control of FSH and the existence of a putative FSH-controlling factor (see McCann et al., 1983). Our current lack of such a structurally characterized factor underlies our relative ignorance of the nature of opioid control of FSH, but it is pertinent that LH and FSH may also be distinguishable in this respect. Thus, in resting rats, opioids do not, in contrast to LH, clearly depress FSH (Cicero et al., 1976; Grossman et al., 1981a; Meites et al., 1979). However, its circulating levels are elevated by naloxone, evidential of a tonic endorphinergic brake upon its release (Grossman et al., 1981a; Morley et al., 1980; Meites et al., 1979; Ropert et al., 1981). If this were “maximally” active, it might explain the lack of additional effect of exogenously applied opioids. Alternatively, the apparent lack of action may reflect the much longer half-life of FSH in contrast t o LH. Further, opioid agonists interfere with, and antagonists promote, the preovulatory surge of FSH in the rat, although less consistently than with LH (Ieiri et al., 1980a; Koves et al., 1981; Marton et al., 1981; Meites et al., 1979; Pang et al., 1977). 3 . Prolactin (PRL)

a. Sites and Mechanism of Action. Dopamine is discharged from tubero-infundibular neurons into the median eminence, from which it attains the AL to suppress PRL secretion. Although a specific peptide for PRL control may be awaiting discovery, an interaction of opioids with dopamine constitutes a major mechanism of opioid PRL regulation. In distinction to other AL hormones, a possible opioid modulation of PRL release via an action in the AL must be seriously taken into consideration. Thus, in the presence of hypothalamic extracts, morphine was observed to increase PRL outflow (Hall et al., 1976). Further, although there are contradictory reports, opioids have generally been reported to prevent the suppressive impact of dopamine upon PRL release in a dose-dependent and naloxone-reversible fashion (Cheung, 1982; Enjalbert et al., 1979; Login and McCleod, 1979; Muraki and Tokunaga,

40

MARK J . MILLAN AND ALBERT HERZ

1978; Voigt et al., 1983). Opioids are, however, almost invariably inactive alone and naloxone is ineffective in the presence or absence of dopamine (Enjalbert et al., 1979; Grandison and Guidotti, 1977; Grandison et al., 1980; Matsushita et al., 1982; Rivier et al., 1977; Shaar et al., 1977). This pattern of effects is comparable to the relationships of opioids to dopamine in the control of hypothalamic LHRH release (see above) but, of course, functionally opposite. Nevertheless, the following points comprise the major lines of evidence that the principal action of opioids underlying the rise in PRL is a centrally effected reduction in release of dopamine. First, a naloxone analog unable to cross the blood-brain barrier only blocked the opioidinduced rise in PRL if administered centrally (Panerai et al., 1981). This rise was also blocked by stalk transection (Wardlaw et al., 1980b). Second, application of opioids systemically, icv, or into discrete brain regions suppresses plasma PRL and elevates hypophyseal portal levels of dopamine (Grandison and Guidotti, 1977; Gudelsky and Porter, 1979; Haskins et al., 1981; Johnson, 1982; Reymond et al., 1983). Third, opioids inhibit the turnover, synthesis, and release of dopamine in the median eminence region (Alper et al., 1980; Deyo et al., 1979; Ferland et al., 1980; Van Loon et al., 1980b; Wilkes and Yen, 1980) (naloxone facilitates this in vitro outflow of dopamine but fails, surprisingly, to modify dopamine turnover or portal vessel levels of dopamine). Fourth, subeffective doses of dopamine receptor antagonists and morphine act synergistically to elevate PRL secretion (Van Loon et al., 1980a; Van Vugt et al., 1979). Whether, in addition to these effects upon DA release, opioids act via other mechanisms (e.g., via a putative PRL-releasing factor) to control PRL release remains to be clarified (Aritd and Porter, 1984). There is evidence that opioids can act (but not exclusively) at the mediobasal hypothalamus to suppress PRL release, possibly via a direct interaction with dopamine neurons (Grandison and Guidotti, 1977; Grandison et al., 1980; Halasz et al., 1981; Haskins et al., 1981; Haskins and Moss, 1983; Spies et al., 1980; Wilkes and Yen, 1980). However, there are indications for serotonin-opioid interactions in PRL control, and a majority of studies have indicated that a serotoninergic network may mediate the actions of opioids upon PRL in rats (though not monkeys) (Ferland et al., 1980; Halasz et al., 1981; Koenig et al., 1979, 1980; Preziosi et al., 1981; Spampinato et al., 1979; Tache et al., 1979; Van Loon et al., 1980a; Wehrenberg et al., 1981). In line with this proposition, a disruption of serotoninergic transmission attenuated the decrease in median eminence turnover of dopamine produced by morphine (Demarest and Moore, 1981). An involvement of serotonin may partially

T H E ENDOCRINOLOGY OF T H E OPIOIDS

41

relate to a nonhypothalamic site of action (Halasz et al., 1981) since, in analogy to LH, introduction of morphine into the dorsal raphe elevates serum levels of PRL (Johnson, 1982). A summary of possible modes of PRL control is presented in Fig. 5B. b. Opioids and Receptor Types. I n analogy to LH, the icv (but not intravenous) administration of antisera against P-EP depressed basal levels of PRL in the circulation and blunted the response of these to stress, suggestive of a role of P-EP in control of PRL secretion (Ragavan and Frantz, 1981). As discussed below, AL P-EP may be involved under stress. In a recent study in rats, it was maintained that the receptor type (of unknown identity) mediating opioid-induced rises in PRL differed from that for elevation of ir-growth hormone (Spiegel et al., 1982). Other studies have found the K-like opioids DYN and a-NE, and a putative synthetic K-agonist, to elevate levels of PRL (Matsushita et al., 1982; Pechnick et al., 1981) or have suggested a preferential contribution of p- and 6- o r p- and K-receptors (Koenig et al., 1982, 1983; Pfeiffer and Pfeiffer, 1983). Evidently, much remains to be resolved. c. Physiologacal Conditions of Control. Two types of stimuli which are clearly characterized as stimulants of PRL secretion are acute stress and suckling, for both of which a functionally intact serotoninergic network is necessary (see Barofsky et al., 1983; Ferland et al., 1980). Naloxone powerfully inhibits the elevation in circulating PRL provoked by stressors such as heat exposure, ether inhalation (which reveals cross tolerance to morphine), immobilization, and foot shock in rats (Deyo and Miller, 1982; Ferland et al., 1980; Grandison and Guidotti, 1977; Meites et al., 1979; Shin, 1978; Van Vugt et al., 1977). Stress also mobilizes hypothalamic, midbrain, and AL pools of P-EP (Millan, 1981; Millan et al., 1981a), and a depression by P-EP of dopamine release could relate to these by action upon stress-elicited secretion of PRL. Dexamethasone, which blocks the response of AL but not hypothalamic pools of P-EP to stress and does not directly affect lactotrophs, eliminates the PRL reaction to foot shock stress (Rossier et al., 1980). Thus, an interaction of AL /3-EP with dopamine in the AL or median eminencelarcuate region may be the active mechanism in these circumstances (Rossier et al., 1980; but see Deeter and Muller, 1981; Holaday and Loh, 1982). Further, CRF was recently observed to naloxone-reversibly stimulate the release of PRL in viuo perhaps via a mobilization of AL P-EP (Schulte et al., 1983). In man, the situation is rather equivocal since naloxone moderated the rise in PRL accompanying surgery but not gastroscopy and only in certain instances of insulin hypoglycemia or exercise (Grossman et al., 1981b; Mayer et al., 1980; Moretti et al., 1983; Morley et al., 1980; Pontiroli et al., 1982; Saltiel et al., 1982; Serri et al.,1981; Spiler and Molitch,

42

MAKK J . MILLAN A N D ALREKT HERZ

1980; Wakabayashi et al., 1980). An opioid involvement in sucklinginduced increases in plasma PRL in rats was indicated by the marked attenuation in these rises produced by naloxone in rats (Ferland et al., 1980). I n both animals and man, PRL is secreted episodically, and naloxone was documented by Ferland et al. (1980) to block nocturnal rises in PRL in postmenopausal women. However, in neither rats nor man did Martin et al. (1979) find naloxone to affect sleep-related patterns of PRL levels. Further, naloxone is rather less effective in depressing circulating PRL in man as compared to rats. In analogy to LH and FSH, naloxone is able to block the proestrus surge of PRL (Ieiri et al., 1980a). Since estrogen is involved in the initiation of this rise, it is pertinent that naloxone also attenuates the estrogen-instigated rise in serum PRL, although the opioid nature of this mechanism has been questioned (Ahmed et al., 1981; Deyo and Miller, 1982; Grandison and Guidotti, 1977). The ability of naloxone to block the release of PRL evoked by Arg-vasotocin (a putative pineal hormone) is of related interest in that an action of Argvasotocin is dependent on the presence of gonodal hormones (Blask and Vaughan, 1980; Blask et al., 1984). Both female and male immature rats (up to 20 days) respond to morphine with a rise in circulating PRL, whereas in neither sex does naloxone suppress PRL levels in plasma at this age (Ieiri et al., 1979). This intersex similam’ty with PRL in immature rats may be compared to the intersex differences as concerns LH. Of clinical interest is the inability of naloxone to moderate the elevated levels of PRL manifested by hyperprolactinemic patients (Blankstein et al., 1979). 4. Growth Hormone (GH) Somatostatin (SS), or GH-inhibiting factor, was originally sequenced as a tetradecapeptide, but a 28-residue peptide of which SS1-14 is the Cterminus or its SSI-Ip counterpart may also be of significance (Bakhit et al., 1983; Millar et al., 1983). Cell bodies synthesizing SS in the anterior periventricular hypothalamus heavily innervate the median eminence (Elde and Hokfelt, 1979; Palkovits, 1982). The increases and decreases produced by, respectively, opioid agonists and antagonists in plasma GH in vivo reflect actions in the brain (possibly partly in the mediobasal hypothalamus) rather than on the AL (Casanueva et al., 1980, 1981; Chihara et al., 1978; Grandison et al., 1980; Halasz et al., 1981; Kokka and George, 1974; Martin et al., 1975; Panerai et al., 1981; Rivier et al., 1977; Shaar et al., 1977). An attractive proposition would, evidently, be an opioid-induced depression in hypothalamic SS activity. Indeed, opioids were reported by Drouva et al. (1981) to naloxone-reversibly

THE ENDOCRINOLOGY OF THE OPIOIDS

43

inhibit stimulated (but not spontaneous) SS outflow from mediobasal hypothalamic neurons (Drouva el al., 1981; Negro-War, 1982; Sheppard et al., 1979). However, icvp-EP, a potent stimulant of GH secretion, failed to alter levels of SS in portal plasma, and in rats in which SS was neutralized with an antiserum against it, a GH response to morphine was still exhibited (Abe et al., 1981; Chihara et al., 1978; Dupont et al., 1977). These observations are in contradiction of an opioid regulation of GH predominantly via SS, and the recently isolated GH-releasing factor (Bloch et al., 1983b; Guillemin et al., 1982; Spiess et al., 1983) may be involved in the mediation of the influence of opioids upon GH (Miki et al., 1984). Although the situation is far from clear, in the rat an a- (but not p-) adrenergic mechanism has been implicated as involved in the effects of opioids upon GH; serotonin and DA are apparently of rather minor significance (in contrast to PRL) while GABA may also be involved (Eriksson et al., 1981; Halasz et al., 1981; Katakami et al., 1981a,b; Koenig et al., 1980; Tach6 et al., 1979; Terry et al., 1982). I n the dog, however, a major role of histaminergic and cholinergic rather than adrenergic networks in the expression of opioid influences upon GH has been indicated (Casanueva et al., 1980, 1981; Cocchi et al., 1980; Farris and Richards, 1980). A comparable situation to the dog of an involvement of cholinergic mechanisms appears to prevail in man (Delitala et al., 1983; Penalva et al., 1983). In rats, the receptor types underlying opioid actions upon GH are likewise distinguishable from those controlling PRL, and neither p - nor K-opioid receptors appear to be of major importance in GH control, although &receptors may be (Koenig et al., 1982, 1983; Pechnick et al., 1981; Spiegel et al., 1982). However, such work remains at a preliminary phase. Most authors favor a tonic role of opioids in GH control, at least in rats (see Meites, 1979),and a modulation of pulsatile outflow in hamsters has been indicated (Nicoski and Borer, 1983). Naloxone did not, however, modify ultradian or nocturnal pulsatile patters of GH levels in rats or man (Martin et al., 1979; Tannenbaum et al., 1979). Further, opioids do not appear to be responsible for the maintenance of elevated GH levels by acromegalic patients (Blankstein et al., 1979). Interestingly, in analogy to PRL, naloxone did attenuate the increase in plasma GH produced by suckling in the rat (Miki et d., 1981). In man, as with PRL, opioids were generally suggested to be partial mediators of the GH rise accompanying exercise, but not gastroscopy, whereas there are discrepant data as concerns insulin hypoglycemia (Grossman et al., 1981b; Mayer et al., 1980; Moretti et al., 1983; Morley et al., 1980; Serri et al., 1981; Spiler and Molitch, 1980; Wakabayashi et al., 1980). Nevertheless,

44

MAKK.1. MILLAN AND ALBERT HEKZ

the GH response to arginine infusion has been found to be blunted by naloxone (Morley et al., 1980). Finally, naloxone was recently observed to temper the elevation in serum GH evoked by antinociceptive acupuncture-like stimulation in chronic pain patients (Pullan et al., 1983).

5. Thyroid Stimulating Hormone (TSH) Although there are claims of actions of opioids in suppressing TSH release at the level of the AL in vitro, these appear not to relate to opioidergic mechanisms since the effects are not naloxone sensitive Uudd and Hedge, 1983; May et al., 1979). Indeed, other studies have demonstrated that neither in the presence nor in the absence of thyrotropin-releasing hormone (TRH) do opioids decrease AL outflow of TSH in vitro, and that intrapituitary injections of @-EP do not depress circulating levels of TSH in uiuo (see Judd and Hedge, 1982; Sharp et al., 1981). Further, in the majority of instances, opioid agonists did not decrease, or antagonists increase, the in vim response of TSH to TRH (Delitala et al., 1981a; Mitsuma and Nogimora, 1983a,b; Morley et al., 1980; Zanoboni et al., 1981; see Krulich, 1982). Moreover, in cold-exposed rats, a morphine-evoked rise in hypothalamic TRH (possibly reflecting a decrease in release) was seen. Indeed, both ablation and brain microinjection studies have demonstrated a central site of action of opiates in depressing TSH secretion. Somata synthesizing TRH project to the external median eminence from the PVN of the hypothalamus (Brownstein et al., 1982; Hokfelt et al., 1978, 1980; Palkovits, 1982), and, in vitro, opiates inhibited K-stimulated release of TRH from the mediobasal hypothalamus (Tapia-Arancibia and Astuer, 1982). In microinjection studies, not only sites in this anterior hypothalamic region but also posterior hypothalamic loci have been identified as of importance for opioid action uudd and Hedge, 1982; Lomax and George, 1966; Lomax et al., 1970; but see Mannisto et al., 1984). An involvement of dopamine in the effects of central opioids upon TRH is an attractive possibility. Thus, although its control of TSH is complex, dopamine enhances hypothalamic TRH release in vitro (see Krulich, 1982), and opioids exert inhibitory effects upon tubero-infundibular neurons (see Section III,A,3,a). Pharmacological evidence for an opioid-dopamine interaction in control of TSH release is also available (Delitala et al., 1981a,b; Mitsuma and Nogimori, 1983; Sharp et d., 1981). In contrast to other AL hormones, opioids do not appear to tonically modulate TSH secretion in the rat, although they possibly do so in man (Agnati et al., 1979; Meites, 1979; Grossman et al., 1981b, 1982b,d). An opioid suppression of the TSH activation evoked by thyroidectomy has

THE ENDOCRINOLOGY OF THE OPIOIDS

45

been observed in the rat (Muraki et al., 1980). Further, opioids prevent the cold exposure-elicited rise in TSH and naloxone, conversely, blocks the heat-elicited fall in TSH (Mannisto et al., 1984; Mitsuma and Nogimori, 1983; Sharp et al., 1981). A role of opioids in TSH control under these typical stimuli for TSH modulation may, thus, be inferred. Further, these actions are not confined to temperature stressors, since naloxone likewise blocks the TSH fall associated with restraint stress Uudd and Hedge, 1982). Finally, a role in control of TSH release under suckling has been tentatively attributed to opioids (Riskin et al., 1984).

OF SECRETION OF P-ENDORPHIN, B. OPIOIDCONTROL ADRENOCORTICOTROPIN, AND a-MELANOCYTE STIMULATING HORMONE

As described in Section IJ,B,l, POMC, the common precursor for ACTH, a-MSH, j3-LPH, and j3-EP, is processed differently in AL corticotrophs as compared to IL melanotrophs. In the former, ACTH, PLPH, and P-EP are costored and coreleased, as opposed to a-MSH and j3-EP in the latter. This parallelism renders it convenient to consider ACTH, a-MSH, and P-EP together in this section. Corticosteroids (CS) have frequently been utilized for the indirect monitoring of corticotrophic secretion. This practice is, in general, acceptable, since the major means whereby opioids modify CS release is via AL corticotrophs rather than directly (see Section III,D,2) upon the adrenal cortex, although the possibility of such direct effects or dissociations in ACTH and CS release (e.g., Wilkinson et al., 1983) must be borne in mind. T h e acute administration of opioid agonists to rats generally leads to an activation of corticotrophic secretion as reflected in a rise in CS levels, an action which develops tolerance (de Souza and Van Loon, 1982; Endrooczi, 1980; George et al., 1974; Jezova et al., 1982; Kokka and George, 1974; see Buckingham, 1980). This appears to be a central effect mediated via AL pools of ACTH since it is abolished by hypophysectomy, dexamethasone, or lesions of the median eminence, is reproduced by administration of opioids into the brain, and is accompanied by a rise in serum ACTH and P-EP (de Souza and Van Loon, 1982; Harasz et al., 1981; Jezova et al., 1982; Levin et al., 1981; see Buckingham 1980; and Kokka and George, 1974). Further, in vitro, the AL secretion of ACTH and P-EP is not affected by opioids whereas they stimulate, in vitro, hypothalamic release of CRF (Buckingham, 1982; Buckingham and Cooper, 1984; Endroczi, 1980; Levin et al., 1981). Indeed, this in vivo rise in ACTH is associated with changes in hypothalamic levels of

46

M A R K J . MILLAN A N D ALBERT HEKZ

CRF, and injections of opioids into the mediobasal hypothalamus also increase circulating CS (Buckingham, 1982; Endroczi, 1980; Lotti et al., 1969; Van Ree et al., 1976; see Kokka and George, 1974). Thus a direct hypothalamic site of action is indicated. Nevertheless, it is unclear whether, in vivo, the ir-P-EP in plasma originates entirely from the AL. Indeed, a parallel enhancement of IL melanotroph secretion by opioids is suggested in the occurrence, upon icv application of P-EP, of a degranulation of these cells (Saland et al., 1982b) and an elevation in circulating levels of a-MSH (De Rotte et al., 1981; van Wimersma Greidanus et al., 1979, 1981). There is evidence that this is a central effect related to a disinhibition of the dopaminergic brake upon IL outflow of a-MSHIP-EP (De Rotte et al., 1981). In contrast to other reports (De Rotte et al., 1981; Przewlocki et al., 19’78),it has been claimed that P-EP in uitro, will liberate a-MSHIP-EP via a direct action on the IL (Celis, 1980), a finding demanding corroboration. Interestingly, in analogy to the interaction of opioids and dopamine on AL PRL release, P-EP naloxone-reversibly antagonized the dopamine-effected suppression of IL a-MSHIP-EP release in frog NILS in vztro (Saland et ul., 1982a). This effect could not, however, be obtained in the rat (Voigt et al., 1983). Finally, suggestive of an endogenous opioid stimulation of a-MSH in the frog, naloxone blocked dark-background adaptation via a central effect in this species (Mennin and Saland, 1980). Recently, it was reported that morphine will naloxone-reversibly attenuate the stimulation-evoked outflow of VP from mediobasal hypothalamic tissue in vztro (Knepel and Reimann, 1982). Such an action, especially if applicable to the subpopulation of VP fibers containing CRF, would offer a means whereby opioids might in certain cases depress AL secretion of ACTHIP-EP and CS. Although dysphoric properties have been argued to partially account for the CS-elevating effects of certain narcotics (Lahti and Collins, 1982), the above mechanistic analyses strongly counter the assertion that an opioid alteration of the release of ACTH, CS, a-MSH, and P-EP is merely a secondary repercussion of a nonspecific stresslike effect. However, opioid antagonists, (in rats, paradoxically), also generally elicit a release of CS, particularly persistently in man, in which opioid agonists may suppress ACTH and CS (Rabinowe et al., 1983; Taylor et al., 1983b). This antagonist action is blocked in rats by hypophysectomy or dexamethasone but not by hypothalamic deafferentation and is accompanied by a rise in plasma ACTH, while no effect on AL secretion of ACTH is seen in vitro (Buckingham, 1982; Eisenberg, 1980; Grossman et al., 1982a; Jezova et al., 1982; Morley et al., 1980; Siege1 et al., 1982; see Holaday and Loh, 1982). ‘These observations are indicative of a predom-

T H E ENDOCRINOLOGY OF T H E OPIOIDS

47

inantly central, probably hypothalamic, mediated influence on AL corticotrophs (see also Eisenberg, 1984; Levin et al., 1981). However, the mechanism of the naloxone-evoked rise in corticotrophic secretion is obscure since, in vitro, naloxone does not affect AL or IL outflow of ir-pEP (Przewlocki et al., 1978; Vale et al., 1979), hypothalamic release of CRF, or median eminence outflow of VP (Buckingham, 1982; Knepel and Reimann, 1982). Further, in rats naloxone consistently fails to elicit a rise in ir-@-EPlevels in plasma, although an increase was seen in dogs and man (Allen et al., 198 1; Arnold et al., 1982; H d l t et al., 1978; Levin et al., 1981, 1984; Naber et al., 1981a). Moreover, naloxone fails to affect the secretory activity of IL melanotrophs (de Rotte et al., 1981; Saland et al., 1982a,b). An additive effect of naloxone and stress upon circulating ACTH and CS has been observed by certain authors (Ixart et al., 1980; Siege1 et al., 1982). In contrast, other authors have found opioid agonists and antagonists not to affect or, respectively, to potentiate and blunt the stress-elicited release of ACTH and CS in rats (Arrigo-Reina et al., 1980; Dai and Chan, 1983; Ferri et al., 1982a; Gibson et al., 1979a; Tapp et al., 1981). T h e data suggestive of an opioid promotion of ACTH and CS release under stress would be reinforced by the ability of morphine, in vivo, to augment the response of hypothalamic CRF to stress (Buckingham, 1982). Nevertheless, the various data are, evidently, contradictory. Further, in rats, dogs, o r human infants treated acutely with naloxone or tolerant to morphine, an exaggeration or lack of effect rather than an attenuation in the increase in plasma ir-P-EP provoked by stresslike stimuli has been seen (Carr et al., 1982; Millan et al., 1981a; Nichols et al., 1983; Rees et al., 1983; Rossier et al., 1980). Possibly the contribution of the IL to ir-P-EP but not ACTH relates to this distinction. Further, in man, naloxone has produced variable affects upon stress-induced alterations in CS or ACTH levels (Engquist et al., 1981; Grossman et al., 1982a; Morley et al., 1980; Pontirolli et al., 1982; Serri et al., 1981; Spiler and Molitch, 1980). Interspecific differences or, for example, the exact nature of stressors employed may account for these discrepant data. Evidently, a great deal requires resolution as concerns the nature of opioid control of AL corticotrophs and IL melanotrophs under both basal conditions and stress. Long-term (at least 4 weeks) exposure to morphine, but not necessarily to other opiates, results in a pronounced fall in NIL levels of ir-P-EP (Millan et al., 1981a; Przewlocki et al., 1979a; Wiister et al., 1980). A corresponding decrease in the an vitro spontaneous outflow of ir-P-EP from the NIL is found (Hollt et al., 1980). Further, in plasma, a clear diminution in levels of ir-P-EP is seen (Millan et al., 1981a). In contrast to the NIL, no change is detected in the AL content of ir-P-EP, but a

48

MARK J. MILLAN A N D ALBERT HERZ

disturbance of corticotrophic secretion is suggested by the reduction in plasma levels of CS observed and a possible impairment of the capacity of the hypothalamus to secrete CRF (Borrell ct al., 1975; Millan et al., 198la; Przew4ocki et al., 1979a; see Buckingham, 1980; Meites et d., 1979). In fact, heroin addicts may o r may not manifest depressed plasma levels of CS, but were found to likewise show a reduction in ir-P-EP (Cushman and Kreek, 1974; Ho et al., 1977, 1980). Chronically morphinized rats exhibit a pronounced decrease in the N I L content of mRNA encoding the common ACTH/P-EP precursor, POMC (Gianoulakis et al., 1981; Hiillt et al., 1981a). This is evidential of a selective disruption of the biosynthetic machinery in this lobe, although the processing of the precursor into its end products appears not to be altered. Interestingly, however, the ability of these rats to respond to stress with a discharge of ir-P-EP and ACTH into the circulation is not compromised (Kokka and George, 1974; Millan et nl., 198la). It should be pointed out that effects of' chronic morphine upon pituitary ir-P-El' were not reproduced by other opiates and that it is questionable whether they represent opioid-specific effects (Wiister et al., 1980). C. OPIOID CONTROL OF NEURAL LOBESECRETION OF VASOPRESSIN A N D OXYTOCIN

1. Vasopressin a. Sites and Mechanisms of Action. Until recently, the antidiuretic actions of opiates, i.e., their ability to reduce urine flow in the rat (Bisset et al., 1977; Fujimoto, 1971; Huidobro-Tor0 and Huidobro, 1981), had commonly been attributed to an enhancement in the NL outflow of VP. This antidiuresis is, indeed, exerted dose dependently and naloxone reversibly and shows stereospecificity in addition to developing tolerance. However, the hypothesis that this antidiuresis is a consequence of a hypersecretion of VP reflecting a direct specific action of opioids is in contradiction to more direct studies of opioid interactions with VP considered below. It is also countered by the following observations (see Huidobro and Huidobro-Toro, 1979; Huidobro-Toro, 1980; HuidobroToro and Huidobro, 1981; Hurwitz, 1981; Skowsky et al., 1982; Walker and Murphy, 1984). First, the application of opioids results in profound alterations in cardiovascular parameters including a vasodilation and hypotension, a secondary effect of which could be an induction of VP release. Such actions, related to a promotion of VP outflow, may be effected within the brain (Leander, 1983b). Second, possibly partly d u e to such peripheral hemodynamic perturbations, an interference with

T H E ENDOCRINOLOGY O F T H E OPIOIDS

49

renal blood flow leads to a reduction in the glomerular filtration rate. Third, there are qualitative differences between VP and morphine as concerns their influence upon electrolyte excretion: VP encourages the elimination of monovalent electrolytes whereas morphine leads to their retention. Finally, morphine still behaves as an antidiuretic in hypophysectomized or Brattleboro rats which lack VP. It has in fact been reported that opioids may elevate levels of ir-VP in the plasma of rats in vivo; these effects are, however, generally insensitive to naloxone (Baertschi et al., 1981; Firemark and Weitzman, 1979; Iovino et al., 1983; Weitzman et al., 1977). Thus, although an action of opioids in promoting VP secretion cannot be entirely dismissed, it is very unlikely to represent their major influence in this respect. These reservations are strengthened by the fact that in the majority of in viuo studies opioids diminished or failed to effect (possibly due to problems with detection limits) circulating levels of ir-VP depending on dose, time of sampling, and other such variables (Aziz et al., 1981; Grossman et al., 1980; Iovino et al., 1983; Knepel et al., 1981; Miller, 1975, Reid et al., 1981; Rockhold et al., 1983; Van Wimersma Greidanus et al., 1979, 1981; Zerbe et al., 1982). Pointing to a possible site of action of opioids in suppressing VP release external to the BBB within which VY somata are localized, are observations of the efficacy of intravenous P-EP (Grossman et al., 1980; Knepel et al., 1981). Indeed, in the isolated NIL, in vitro, opioid agonists generally moderate the outflow of ir-VP evoked by electrical stimulation or other manipulations; in certain cases they depressed, and in others failed, to affect basal release. Naloxone is, in contrast, generally ineffective in modifying basal release, but under appropriate conditions may augment stimulated secretion (Bicknell and Leng, 1982a,b; Christensen and Fjalland, 1982; Clarke and Patrick, 1981; Iversen et al., 1980; Knepel et al., 1983a; Lightman et al., 1982; Lutz-Bucher et al., 1982; LutzBucher and Koch, 1980; Maysinger et al., unpublished; Racke et al., 1982a; Weitzman et al., 1977; Zein et al., 1984). I t must be emphasized that the precise mode and parameters of stimulation employed, in particular whether corresponding to the pattern physiologically generated in viuo, appear critical in determining the nature of action of opioid agonists and antagonists upon in vitro outflow of ir-VP (or ir-OT) from the NIL (Maysinger et al., unpublished). Attention to this variable may offer an explanation for certain discrepant data. T h e clear majority of data is, in any case, evidential of the occurrence of an opioid mechanism for a reduction in ir-VP release within the NIL. A direct presynaptic action on terminals is an attractive postulate but one which lacks experimental verification. Indeed, on the basis of morphological analyses of

50

MARK J. MILLAN A N D ALBERT H E R 2

the NL at the light and electron microscopic level, the possibility that the specialized glial cells (pituicytes) might mediate the effects of opioids upon neurohypophyseal neurons upon VP has been convincingly raised (Lightman et al., 1983b; Van Leeuwen et al., 1983; Van Leeuwan and De Vries, 1983). A supplementary locus of action in the brain is favored by the ability of centrally applied opioids to depress circulating ir-VP and to block the excitatory impact of hypertonic saline upon the discharge of VP-like neurohypophyseal neurons (Clarke and Merrick, 1982; Knepel et al., 1981). Morphine, likewise upon central administration, reduces the spontaneous phasic activity of neurons in the SON, the firing of which is restored by naloxone (Clarke et al., 1980). However, in vitro experiments with the hypothalamic slice have suggested that in both the SON and the PVN it is predominantly nonneurohypophyseal (or unidentified) rather than VP-like neurons which are directly inhibited by perfusion with opiates (Arnauld et al.,1983; Clarke et al., 1981; Muehlethaler et al., 1980; Pittman et al., 1980). Further, recent work of Wakerley et al. (1983) pointed to an extrahypothalamic site of opioid action. The central opioid control of VP release probably represents, thus, primarily a control of afferent input to VP somata in contrast to a possible minor, direct action upon these cells. These indirect sites of action remain to be identified. There is a heavy, dehydration-responsive dopaminergic innervation of the NL and IL from the arcuate hypothalamus (Bjorklund et al., 1973; Holzbauer et al., 1980). Dopamine is a potent attenuator of the stimulated, but not basal, outflow of ir-VP from the NIL in uitro (Lightman et al., 1982; Racke et al., 1982a; Vizi and Volbekas, 1980a,b; see Mathison, 1981). Rack6 et al. (1982a) recently evaluated the actions of dopamine in detail and revealed an extremely complex pattern of effects (see also Bicknell and Leng, 1982b). Of pertinence is the possible interrelationship of this dopamine network to opioids in the control of ir-VP secretion. First, it has been demonstrated that in the NIL, in vitro, opioids will inhibit stimulated dopamine release (Vizi and Volbekas, 1980b). In addition, it is possible that opioids may decrease the activity of this dopamine system via a central action. Thus a reduction of this dopaminergic brake upon VP release could offer a mechanism for opioid promotion of VP secretion (Lightman et al., 1983a; Vizi and Volbekas, 1980a,b). This might in theory help reconcile certain discrepancies concerning the direction of action of opioids on release of ir-VP. However, since morphine did not affect turnover of dopamine in the posterior pituitary, at least under basal conditions, a putative inhibitory interaction of opioids with dopamine therein under stimulation requires evaluation (Moore and Demarest, 1982). Second, it was postulated that, in vitro, the parallel dopamine-effected depression in IL release of P-EP might lead to an

T HE ENDOCRINOLOGY OF THE OPIOIDS

51

indirect enhancement of ir-VP secretion via a reduction in the inhibitory impact of IL P-EP upon VP (Racke et al., 1982a). There could thus be dual, contrasting effects of both DA and opioids upon VP. The predominant and direct influence of these is, however, very probably inhibitory. In Fig. 6, possible modes of opioid control of NIL VP secretion are summarized. b. Opioids and Receptor lypes. 'I'he use of particular agonists has not, to date, proven notably instructive in that agonists with preferential p(morphine), 6- (D-Ala*-D-leucine-enkephah), o r K- (DYN) activity in each case have been variably reported to attenuate liberation of ir-VP from the NIL in vitro (Lightman et al., 1982; Lutz-Bucher and Koch, 1980; Maysinger et al., unpublished). The possibility of a DYN/K-receptor link has aroused much interest since it could represent an autoreceptor-like mechanism of negative feedback by DYN cojointly released with VP from common neurons (Watson et al., 1982a). Interestingly, in the rat and mouse K-agonists elicit (in distinction to p-agonists such as morphine) a pronounced diuresis, an action partially reflecting a reduction in the release of VP and not manifested in Brattleboro rats (Leander, 1983a,b; Rathbun et a!., 1983; Slizgi and Ludens, 1982; Slizgi et al., 1984; Vonvoigtlander et al., 1983). Further, the lack of effect of icv-applied EKC, a prototypic K-ligand, upon ir-VP release led to the postulation of a site of action external to the brain, i.e., in the NL (Slizgi and Ludens, 1982). However, opposite data have been acquired with a more selective K-agonist, U-50488H, and a hypothetical role of DYN/K-receptors in the OPlOlDS (indirect)

ARCUATE NUCLEUS

OPlOlDS

OPlOlDSOPlOlDS

CIRCULATION

FIG.6. Opioid mechanisms in the control of the secretion of vasopressin and oxytocin from the neural lobe. For abbreviations, see Fig. 1. See also legend to Fig. 5.

52

M A R K ] . MILLAN A N D ALBERT HERS:

NL is contradicted by the failure of parenterally applied quaternary naloxone, which does not penetrate the BBB, to antagonize the EKCinduced depression in plasma levels of ir-VP (Carter arid Lightman, 1984; Leander, 1983a). Further, if DYN coreleased with VP was the opioid inhibitory to this VP secretion, an opioid inhibition should be apparent under all conditions eliciting VP-release, which is not the case in vivo (e.g., Knepel et al., 198%). An additional counterargument is that, in vitro, naloxone alone does not consistently modulate stimulated ir-VP outflow, and that DYN is only weakly effective in reducing this ir-VP release (Bicknell and Leng, 1982b; Carter and Lightman, 1984; Clarke and Patrick, 1983; Maysinger et al., unpublished). The former observation perhaps tends to compromise the hypothesis of a general role of any NL (or IL) opioid in VP control. An alternative theory of a role of AL pools of P-EP, to which the NL is accessible via an interconnecting vascular network, has been advanced by Knepel and co-workers (1980b, 198l,1982c,d). Naloxone potentiated the in vivo secretion of ir-VP elicited by isoprenaline, angiotensin 11, foot shock stress, or nonhypotensive hypovolemia but not hypertonic saline. Each maneuver, with the exception of saline, mobilized AL P-EP. Experiments involving blockade of AL release of P-EP with dexamethasone which, for example, potentiated the ir-VP response to angiotensin I1 or hypovolemia, led to the conclusion that AL P-EP may inhibit NL outflow of ir-VP in vivo. Comparable findings were acquired by Baertschi et al. (1980b) with the stimulus of hemorrhage. There are, further, preliminary indications that P-EP originating in the IL might also be of significance in this respect (Lutz-Bucher and Koch, 1980; Knepel and Nutto, 1983; Knepel et al., 1983b; Rack6 et al., 1982a). c. Physiological Conditions. Basal circulating levels of ir-VP are irresponsive to opioid antagonists, in line with the majority of in vitro studies indicating that opioids are not major determinants of VP release under basal conditions (e.g., Knepel et al., 1980b; Rosella-Dampman et al., 1983). Alterations in blood pressure and blood volume monitored by peripheral cardiovascular receptors and of blood osmolarity by hypothalamic osmo- or sodium- receptors comprise major factors governing VP secretion. Recent studies have shown opioids to attenuate the ir-VP response of hyperosmotic stimuli such as sodium chloride or dehydration. However, the role of endogenous opioids is currently unclear in view of the general lack of a naloxone potentiation of these rises in ir-VP and the failure of naloxone to block the fall in plasma ir-VP produced by overhydration (Brownell et a1.,1980; Clarke and Merrick, 1982; Crossman et al., 1980; Cuiol et al., 1984; Knepel et al., 1982c, 1984; Ishikawa and Schrier, 1982; Kamoi et al., 1979; Lightman et al., 1980; Rosella et al., 1981; Roselka-Dampman et al., 1983; Summy-Long et al., 1981c, 1984; van

THE ENDOCRINOLOGY OF THE OPIOIDS

53

Wimersma Greidanus et al., 1979; Wade, 1983). Similarly, an inability of naloxone to modify the influence upon ir-VP of nicotine or hypotension induced by ganglionic blockade was reported (Knepel et al., 1982~).In contrast, unequivocal evidence for a physiological inhibition by endogenous opioids of the elevation in circulating ir-VP evoked by stress has been provided, while results concerning hypoxia or hypovolemidhemorrhage are presently contradictory (Baertschi et al., 1980a,b; Knepel et al., 1982c,d, 1984; Forsling and Aziz, 1983; Rockhold et al., 1984; Rosella et al., 1981; Rosella-Damprnan et al., 1983; Summy-Long et al., 1984). An important mechanism of control, related to the osmotic model, is the renin-angiotensin system. A role of opioids in the regulation of the effect of angiotensin I1 upon VP is indicated by the ability of agonists and antagonists to, respectively, blunt and augment the secretion of irVP evoked by angiotensin I1 in vivo. Angiotensin I1 might act either on the NL, on circumventricular organs, or within the BBB, e.g., on magnocellular somata, but the physiologically active pool(s) of angiotensin I1 have not, as yet, been conclusively identified (Dreifuss et al., 1981; Knepel et al., 1982c; Lutz-Bucher et al., 1982; Mitchell et al., 1982; Simpson, 1981; Sladek et al., 1982; Summy-Long et al., 1981a,b,1983). Of related interest are the actions of opioids and naloxone in, respectively, attenuating and enhancing the VP response to P-adrenoceptor agonists which appear to act partially via the angiotensin axis and also directly on the NL (see Knepel et al., 1980a,b, 1981, 1982b; Racke et al., 1982b; Ramsay et al., 1978). 2. Oxytocin (OT) The electrical activity of OT-like SON neurons (classified on electrical criteria) was found not, in contrast to VP-like neurons, to be modified by the icv administration of morphine in the studies of Clarke et al. (1980). Further, they showed morphine suppresses the release of OT evoked by various stimuli in uivo in the absence of alterations in the electrical activity of OT-containing neurons (Clarke et al., 1979). Nevertheless, these authors have recently reopened the question of a possible brain-localized opioid mechanism for the inhibition of OT secretion (Wright et al., 1982). Indeed, recently evidence for an opioid inhibition of OT neurons at the hypothalamic level has been presented (Wakerley et al., 1983). Opioids and morphine have been observed to suppress, in vitro, the stimulated (and, on one occasion, basal) release of ir-OT from the NIL (Bicknell and Leng, 1982a,b; Lutz-Bucher, 1980; Lutz-Bucher et al., 1982; Maysinger et al., unpublished). Of special interest is the ability of naloxone to exaggerate the secretion of ir-OT from the NIL produced by electrical stimulation (Bicknell and Leng, 1982b; Clarke

54

MARK.J. MILLAN A N D ALBEKT HER%

and Patrick, 198 1; Maysinger et al., unpublished). This amplification suggests that an opioid ligand intrinsic to the NIL may underlie the inhibition of ir-OT secretion. Importantly, this action persists in the absence of the intermediate lobe, indicative that the opioid subject to antagonism may originate within the neural lobe itself (Bicknell arid Leng, 1983). Vizi and Volbekas (1980a,b) have, in distinction to other studies, suggested an opioid promotion of OT release at the NL. They found, thus, P-EP to disinhibit the release of ir-O?' from the N I L via an alleviation of a dopamine-effected reduction in ir-0'1 outflow. A further evaluation of the putative significance of this possible functionally opposite influence on O T release is required. The major action of opioids may be regarded as inhibitory (see Fig. 6). There is, at present, no evidence for a role of opioids in the modulation of OT release under resting conditions. However, in analogy to PRL, opioids attenuate the elevation in circulating ir-OT elicited by suckling in the mother hemorrhage or osmotic stress (Haldar and Sawyer, 1978; Summy-Long et al., 1984; Haldar et al., 1982; Wright et al., 1983). Consistent with these findings, opioids block the 01' secretion evoked by acetylcholine, a physiological component of the reflex arc for suckling-induced OT release (Haldar et al., 1982). Whether the above actions reflect physiological functions is in need of clarification. Nevertheless, of note is that naloxone, in low doses, prevented the disturbance of suckling-evoked O T liberation engendered by stress (Haldar and Bude, 1981; Summy-Long et al., 1984).

D. OPIOID CONTROL OF PERIPHERAL ORGAN ENDOCRINE SECRETION Opioid actions upon certain organs may be expressed indirectly via AL hormones; e.g., the reduction in circulating LH and TSH is coupled to, respectively, a fall in plasma testosterone levels and a fall in thyroid function. There are, nevertheless, other opioid mechanisms which may influence peripheral organ secretion: these comprise opioids in the circulation, opioids in neurons intrinsic to or innervating these organs, and opioids in the C N S , the actions of which can be mediated neuronally or hormonally. Only in certain cases have the particular opioid pools involved been identified. 1. Adrenal Medulla

The administration of opioids results in a centrally effected enhancement of sympathetic outflow involving a liberation of adrenaline (plus noradrenaline) into the circulation (Pfeiffer et al., 1983; van Loon and

T I i E ENDOCRINOLOGY OF THE OPIOIDS

55

Appel, 1981). Opioid effects at hypothalamic sites, possibly on glucoresponsive neurons, may relate to these actions (On0 et al., 1980; A. Pfeiffer et al., 1983). Since the rise in adrenaline is particularly pronounced and adrenal denervation abolishes the influence of opioids upon adrenaline, this must predominantly originate in adrenal medulla chromaffin cells. Sympathetic nerve terminals also contribute to the rises seen in noradrenaline (A. Pfeiffer et al., 1983; van Loon et al., 1981a,b). These actions are exerted in the brain, in which a cholinergic link is involved, and in the case of adrenaline are inhibited by somatostatin (van Loon el al., 1981a,b). Angiotensin I1 is also implicated as a potentiator of morphine-induced enhancement of sympathetic outflow (Appel and van Loon, 1983). The effects do not appear to be a secondary consequence of cardiovascular perturbations (van Loon et al., 1981a,b; see Taborsky et al., 1981). Opioids, similarly via the splanchnic nerve, provoke a mobilization into systemic plasma of the enkephalins which are colocalized with adrenaline in these chromaffin cells (see Section 1I,B,4) (Govoni et al., 1981; Laasberg et al., 1980). It should, perhaps, be noted that in shock beneficial effects of naloxone have been proposed to be, in fact, related to an enhancement of sympathetic nervous outflow and a rise in circulating catecholamines (Eddy et al., 1984; Holaday, 1983; Holaday et al., 1983; Schadt et al., 1983). Further, in man, in contrast to rats, there are indications that administration of naloxone results in a stimulation of adrenal medullary secretion (Grossman et al., 1982c; Manelli et al., 1983). Specific opioid receptors, probably on chromaffin cells, d o in fact exist in the medulla (Kumakura et al., 1980; see Dean et al., 1982). Kumakura et al. (1980) documented that in cultured preparations opioids interfere with the nicotine- but not potassium-evoked outflow of catecholamines. This consisted of, in contrast to hexamethonium, a noncompetitive functional reduction in available nicotine receptors. T h e physiological ligands might represent opioids derived from the adrenal cortex, opioids from the systemic circulation, enkephalins costored with acetylcholine in splanchnic nerve terminals, or enkephalins or DYN feeding back from chromaffin cells. A recent publication suggested, in fact, a particularly high affinity of /3-EP for this receptor (Dumont and Lemaire, 1983). However, the “opioid receptor” mediating this action is peculiar in its characteristics, and the nature and significance of this effect are matters of controversy (Castanas et al., 1984; Costa et al., 1983; Dean et al., 1982; Saiani and Guidotti, 1982). Interestingly, denervated medullas develop nonneurogenic mechanisms of opioid secretion in which opioids also enhance, but to a lesser degree, the outflow of catecholamines and result in an induction of dopamine-P-hydroxylase but not (cf. intact medullas) tyrosine hydroxy-

56

M A K K . J . MILLAN AND ALBERT HERZ

lase (Anderson and Slotkin, 1975). A similar situation is encountered in neonatal rats prior to splanchnic innervation. In thee?, naloxone potentiates and opioid agonists retard nonneurogenic-evoked secretion, possibly indicative of a role of intrinsic opioids in secretory control in the immature state (Ghantry el al., 1982). 2. Adr, nal Cortex In addition to opioid actions mediated via ACTH, many authors have observed direct effects of opioid peptides upon the function of the adrenal cortex. It has been documented that ME will depress the production of corticosterone and mineralocorticoids in the adrenal cortex, whereas P-EP was reported to facilitate corticosterone formation and naloxone to, dependent upon dose, elevate and depress steroidogenesis; corresponding interactions with ACTH were also seen (Gibson et al., 1979b; Gullner and Hill, 1983; Heybach and Vernikos, 1981; Kan et al., 1977; Lamberts et al., 1983a; Lymongrover et al., 1981, 1983; Matsuoka et al., 1981; Pham-Huu-Trung et al., 1982; Racz et al., 1982; Shanker and Sharma, 1979; Szalay and Stark, 1981; Zampa et al., 1981). However, it has not been established that these actions are mediated by specific opioid receptors; agonists and antagonists may behave similarly. The results obtained depend on the exact conditions of study, and in each case conflicting data are available (see above references). Thus, although the adrenal cortex may be a significant site of direct opioid agonist or antagonist action, their effects demand considerable clarification. 3. The Pancreas Evidence that alterations in the endocrine secretion of the pancreas contribute to the influence of opioids upon glucose homeostasis has accumulated. I n vitro, low doses of opioids promoted the release of both glucagon and insulin from the pancreas, an action preceded by a decline in the outflow of somatostatin (Hermansen, 1982, Ipp et al., 1978; Sachse et al., 1981; see Schusdziarra et al., 1983b). Since somatostatin is inhibitory to insulin and glucagon secretion, this initial effect was suggested to intervene in the depression in insulin and glucagon release. Subsequent work has, however, questioned this possibility (Green et al., 1983b). In vim, infusion of opioids elevates plasma levels of glucagon and insulin, most prominently in the pancreatic vein, and results in a hyperglycemia in diabetic dogs lacking an insulin response but not in their normal counterparts (Ipp et al., 1982). A fall in circulating somatostatin has also been seen in the dog (Schusdziarra et al., 1983e). In man, an increase in insulin, glucagon, and blood sugar has been seen, though

THE ENDOCRINOLOGY OF THE OPIOIDS

57

not invariably (Morley et al., 1980; Reid and Yen, 1982; Stubbs et al., 1978). Recent studies in both dogs and man in vivo have expanded these data and indicated a physiological role of endogenous opioids in the control of pancreatic secretion in response to ingestion of nutrients. Dependent upon the exact nature of the meal consumed, naloxone resulted in variable modifications of circulating levels of insulin, glucagon, somatostatin, and also pancreatic polypeptide (Feldman et al., 1980; Feldman and Li, 1982; Holland et al., 1983; Konturek et al., 1983; Rewes et al., 1983; Schusdziarra et al., 1983d). P-Casomorphines are opioid peptides (exorphins) derived by the cleavage of the milk constituent casein; their ingestion has, interestingly, been shown to result in an alteration in postprandial circulating levels of pancreatic polypeptide, somatostatin, and insulin (Morley et al., 1983a; Schusdziarra et al., 1983a,b; Schick et al., 1983). The general absence of clear effects of naloxone alone upon insulin secretion, in vitro or in vivo, under unstimulated conditions suggests a lack of major tonic opioid control (Hermansen, 1982, 1983; Ipp et al., 1978; Recant et al., 1980; Zampa et al., 1981). However, in obese mice in which hyperinsulinemia is associated with hyperphagia, naloxone in vitro moderated the exaggerated insulin outflow, indicative of a possible relationship of pancreatic opioid mechanisms to obesity (Recant et al., 1980). This action, together with the fact that in normal rats perfusion of the pancreas with glucose elicited a rise in the secretion of ir-dynorphin and ir-P-EP (Sachse et al., 1981), focuses attention on an involvement of intrinsic opioid mechanisms in the control of pancreatic endocrine secretion. Interestingly, EKC (a K-agonist) has been reported not to affect insulin release, notwithstanding the activity of DYN, and preceptors suggested to be the major mediators of opioid stimulation of insulin release; &receptors might even, under certain conditions, underlie opposite actions (see Schusdziarra et al., 1983~). Opioid control of pancreatic endocrine secretion is complex and not only stimulatory effects but complex patterns of changes may be seen dependent upon the species receptor types activated, opioid concentration, glucose concentration, and physiological conditions (Green et al., 1983a; Hermansen, 1983; Konturek et al., 1983; Rudman et al., 1983; Schusdziarra et al., 1983c-e). Finally, rats suffering from diabetes-which are supersensitive to hypophagic actions of naloxone-possess elevated levels of opioid activity in the NIL; since ir-P-EP therein is unaltered, DYN-related peptides may underlie this change (Taylor et al., 1983a).An interrelationship between NIL opioids, ingestive behavior and pancreatic secretion might, thus, be inferable.

58

MARK J. MILLAN A N D ALBERT HERZ

4 . Other Hormones T h e gut manufactures a variety of hormones which, in view of opioid control of gut exocrine secretion and motility, are promising candidates for opioid regulation. Indeed, there is evidence that opioids (possibly reflecting a role of intrinsic opioidergic systems) may modulate the basal or stimulated secretion of gut pools of secretin, somatostatin, and possibly cholecystokinin and gastrin (Chey et al., 1980; Chiba et al., 1980; Feldman and Li, 1982; Konturek, 1980; Konturek et al., 1983; Mclntosh et al., 1982, 1983; Morley et al., 1983a). However, these studies require considerable further mechanistic evaluation and elucidation of physiological significance. In view of the evidence for opioid-angiotensin interactions at the level of the CNS and pituitary it is of special interest that opioids may control the activity of the renin-angiotensin axis in the periphery; thus, naloxone was found to suppress the release of renin instigated by renal artery constriction and opioids to increase plasma renin in human subjects (Rabinowe et al., 1983; Szilagayi and Taylor, 1983; Uberti et al., 1983). The mechanism of action remains to be clarified. Finally, a recent intriguing study suggested a role of opioids in the enhancement of the secretion from the placenta of a chorionic gonadotrophic hormone related to hypophyseal LH (Valette et al., 1983).

IV. Concluding Comments

At the very least, the impression should have been gained from the present article that the field of the endocrinology of the opioids is both vast and exceedingly complex, in reflection of the multiplicity and heterogeneity of opioid peptide systems. As a consequence, the use of refined techniques for the selective manipulation and evaluation of the activity of discrete pools is essential for ascertaining the roles of individual systems. Concerning endocrine-like opioid systems, it is salutary to reflect that, despite the volume of work devoted to the endocrinology of the opioids and the resultant advances in our understanding in recent years, it is currently difficult to unequivocally ascribe a definitive role to any particular species. Many conditions, for example, stress, nutritional status, degree of hydration, and phase of circadian or estrus cyclicity, have (1) biochemically been shown to be associated with alterations in the activity of endocrine-like opioid systems and (2) related to opioid-medi-

THE ENDOCRINOLOGY OF THE OPIOIDS

59

ated changes in behavioral, physiological, and endocrinological measures. It is the unification of these complementary sources of information and the assignation of particular functions to particular endocrine-like pools of opioids which is now of importance. Indeed, a substantial part of this paper was devoted to a particular aspect of the function of opioid mechanisms, including endocrine-like opioid peptide systems, i.e., that of their modulation of the secretion of hormones into the systemic circulation. The ability of opioids to modify these endocrine parameters is among their most striking properties and, especially in the case of antagonists, no less impressive than their role in antinociceptive processes. There are multifarious sites of attack and mechanisms of action by which opioid mechanisms may modulate endocrine secretion. Similarly, a diversity of opioid ligands and receptor types has been implicated in particular cases. An elucidation of the roles of individual ligands and receptor types at specific loci and their relationships to other control systems under specific conditions has not, as yet, been achieved. In conclusion, an integrated and thorough resolution of the significance of opioid mechanisms in endocrinology will certainly remain a stimulus and challenge for the future. Acknowledgments

T h e helpful comments of Drs. D. Maysinger, R. Schulz, and B. R. Seizinger are most appreciated. M. J. Millan was supported by the Deutsche Forschungsgemeinschaft. References

Abbate, D., Santamaria, A., Bramballa, A., Panerai, A. E., and Di Giulo, G. H. (1983). Lancet 2, 1309. Abe, H., Chihara, K., Chiba, T., Matsukura, S., and Fujita, T. (1981). Endocrinology 108, 1939-1943. Adler, B. A,, and Crowley, W. R. (1984). Neuroendocrinolgy 38, 248-253. Agnati, L. F., Benfenati, F., Cortelli, P., D’Alessandro, R., Ghisoli, E., and Zarnpa, G. A. (1979). Neurosci. Lett. 12, 307-3 1 1. Ahmed, H. H., Van Vugt, D. A., Ruiz de Galaretta, C. M., Fanjul, L. F., and Meites, J. (1981). Life Sci. 29, 2757-2762. Akil, H., Watson, S. J., Barchas, J. D., and Li, C. H. (1979). Life Sci. 24, 1659-1666. Akil, H., Hewlett, W. A., Barchas, J. D., and Li, C. H. (1980). Eur. J . Pharmacol. 64, 1-8. Akil, H., Ueda, Y., Lin, H. L., and Watson, S. J. (1981). Neuropeptides 1, 429-446. Akil, H., Lin, H. L., Ueda, Y., Knobloch, M., Watson, S. J., and Loy, D. H. (1983). Life Sci. 33 (Suppl. l), 9-12. Alde, S., and Celis, M. E. (1980). Neuroendocrinology 31, 116-120. Alessi, N., Taylor, L., and Akil, H. (1982). Life Scz. 31, 1875-1878. Alexopoulos, G. S., Inturrisi, L. E., and Lipman, R. (1983).Arch. Gen. Psychol. 40, 181-183. Allen, J. P., Ho, A. K. S., Kepic, T. A., Biesecker, L., Mizera, A,, and Hofreiter, B. (1981). Neur0.m. Abstr. 7, 797.

60

MARK J. MILLAN A N D ALBERT HERZ

Al-Noaemi, M . C., Edwardson, J. A., and Hughes, D. (1982).J. Physiol. (London) 332,85P86P. Alper, R. H., Demarest, K. T., and Moore, K. E. (1980). I. Neural Tramm. 48, 157-165. Alumet, J., Hakanson, R., Sundler, F., and Chang, K.-J: (1978). Histochembtry 56, 187196. Anderson, T. R., and Slotkin, T. A. (1975). Biochem. Phurmacol. 25, 1071-1074. Anhut, H., Knepel, W., Holland, A., and Meyer, D. K. (1982). Regul. P e p . 4, 83-90. Antoni, F. A., Holmes, M. C., and Jones, M. T. (1983). Peptides 4, 411-415. Appel, N. M., and Van Loon, G. R. (1983). Peptides 4, 59-62. Arita, J.. and Porter, J. C. (1984).Neuroendocrinology 38, 62-67. Arendash, G. W., and Gallo, R. W. (1978). Endocrinology 102, 1199-1206. Arnauld, E., Cirino, M., Layton, B. S., and Renaud, L. P. (1983). Neuroendocrinology 36, 187- 196. Arnold, M. A., Carr, D. B., Togasaki, D. M., Pian, M. C., and Martin, J. B. (1982). Life Sci. 31, 1017-1024. Arrigo-Reina, R., Costa, G., Ferri, S., Scoto, G. M., Spedaro, C., and Spampinato, S. (1980). In “Advances in Pharmacological Research” 0. Knoll, ed.). pp. 12 1-127. Pergamon, Oxford. Atkinson, J. H., Kreilier, E. F., Risch, S. C., Morgan, C. D., Azad, R. F., Ehlers, C. L., and Bloom, F. E. (1983). Psychiatt. Res. 9, 319-327. Atweh, J. F., and Kuhar, M. J. (1977). Bruin Res. 129, 1-12. Aziz, L. A., Forsling, M. L., and Woolf, C. J. (1981).J. Physiol. (London) 311, 401-409. Baertschi, A. J. (1980). Am. J . Physiol. 239, 463-469. Baertschi, A. J.. Friedli, M., Munoz, J., Tsacopoulos, M., and Coles, J. A. (1980a). In “Synthesis and Release of Adenohypophyseal Hormones” (M. Jutisz and W. K. McKerns, eds.), pp. 775-781. Plenum, New York. Baertschi, A. J., Wicki, M. J., and Friedli, M. (1980b). Neurosci. A M . 6, 696. Baertschi, A. J., Zinig, H. H., and Dreifuss, J. J. (1981). BrainRes. 220, 107-119. Bakhit, C., Benoit, R.,and Bloom, F. E. (1983). Nature (London) 306, 524-526. Baldi, E., Salmon, S., Anselmi, B., Spillantini, M. G., Capelli, G., Brocchi, A., and Sicuteri, F. (1 982). Cephulugiu 2, 77-8 1. Baldwin, D. M., Haun, C. K., and Sawyer, C. H. (1974). Brain Res. 80, 291-301. Barden, N., Merand, Y., Rouleau, D., Garon, H., and Dupont, A. (1981). Bruin Res. 204, 44 1-445. Barkan, A., Regiani, S.,Duncan, J., Pepevasiliou, S., and Marshall,J. C. (1983).EndocnnolO# 112,387-389. Barofsky, A.-L., Taylor, J., and Massari, V. J. (1983). Endocrinology 113, 1894-1899. Bartho, L., and Szolcsanyi,J. (1981).Eur. J. Phannacol. 73, 101-104. Baumgarten, H. G., Bjorklund, A., Holstein, A. F., and Nobin, A. (1972).2.Zellforsch. 126, 483-517. Beauvillain, J.-C., Tramu, G., and Garaud, J.-C. (1984).Brazn Res. 301, 389-393. Belova, T. J., and Jonsson, G. (1982). Acta Physiol. Scand. 116, 21-29. Beny, J. L., and Baertschi, A. J. (1980). Neuroendocrinology 31, 261-264. Bergland, R. M., and Page, R. B. (1979). Science 204, 18-24. Bergland, R. M., Blume, H., Hamilton, A., Monica, P., and Paterson, R. (1980). Science 210,541-543. Berk, L. S., Tan, S. A., Carmona, M., and Eby, W. C. (1983). Clin. Res. 31, 23A. Berkenbosch, F., Vermes, I., Binnekade, R., and Tilders, F. J. H. (1981).L f e S c i . 29,22492256. Berkenbosch, F., Tilders, F. J. H., and Vermes, I. (1983). Nature (London) 305, 237-239. Besser, G. M. (1981). Front. Horm. Res. 8, 80-106. Beuers, V., Hertting, G., and Knepel, W. (1980). Br. J . Pharmacol. 76, 579-585.

THE ENDOCRINOLOGY OF THE OPIOIDS

61

Bhanot, R., and Wilkinson, M. (1983a). Endocrinology 112, 399-401. Bhanot, R., and Wilkinson, M. (1983b). Endocrinology 113, 596-603. Bicknell, R. J., and Leng, G. (1982a).J. Physiol. (London) 332, 87P-88P. Bicknell, R. J., and Leng, G. (1982b). Nature (London) 298, 161-162. Bicknell, R. J., and Leng, G. (1983). Prog. Bruin Res. 60, 323-351. Birnberg, N. C., Lissitzky, J. C., Hinman, M., and Herbert, E. (1983). Proc. Nutl. Acud. Sci. U.S.A. 80, 6982-6986. Bisset, G. W., Chowdrey, M. S., and Feldberg, W. (1977). Br. J. Phunnacol. 62, 370-371. Bjorklund, A., Moore, R. Y., Nobin, A., and Stenevi, U. (1973). Brain Res. 51, 171-191. Blank, M. S., and Bohnet, H. G. (1983). Neuroendocrinol. Lett. 5, 3-8. Blank, M. S., and Roberts, D. L. (1982). Neuroendocrinology 35, 309-312. Blank, M. S., Panerai, A. E., and Fiesen, H. G. (1979). Science 203, 1129-1131. Blankstein, J., Reyes, F., Winter, J., and Faiman, C. (1979). Clzn. Endocrinol. 11, 475-479. Blankstein, J., Reyes, F., Winter,J., and Faiman, C. (1981). Clin. Endocrinol. 14, 287-294. Blask, D. E., and Vaughan, M. K. (1980). Neuroscz. Lett. 18, 181-184. Blask, D. E., Vaughan, M. K., Champney, T. H., Johnson, L. Y., Vaughan, G. M., Becker, R. A., and Reiter, R. J. (1984). Neuroendocrinology 38, 56-61. Bloch, B., Baird, A., Ling, N., Benoit, R., and Guilleman, R. (1983a). Bruin Res. 263, 251257. Bloch, B., Brazeau, P., Ling, N., Bohlen, B., Esch, F., Wehrenberg, W. B., Benoit, R., Bloom, F., and Guilleman, R. (1983b). Nature (London) 301,607-608. Blum, K. (1980).In “Advances in Neurotoxicology” (F. Manzo, ed.), pp. 71-89. Pergamon, Oxford. Borrell, J., LlorCns, I., and Borrell, S. (1975). Eur. J . Pharmacol. 31, 237-242. Bossut, D. F. B., Leshin, L. S., Stromberg, M. W., and Malven, P. V. (1983). Peptides 4,501507. Brambilla, F., Genazzani, A. R., Facchinetti, F., Parrini, D., Petraglia, F., Sacchetti, E., Scarone, S., Guastalla, A., and DAntona, N. (1981). Psychoneuroendocrinology6, 321330. Briaud, B., Koch, B., Lutz-Bucher, B., and Miahle, C. (1979). Neuroendocrinology 28, 377385. Bridges, T. E., Fisher, A. W., Goshee, J. L., Lederis, K., and Santolaya, R. C. (1973). Z. Zellforsch. 136, 1-8. Briski, K. P., Quigley, K., and Meites, J. (1984). Life Scz. 34, 2485-2493. Britton, K. T., Stewart, R. D., and Risch, S. C. (1983). Psychophurmacol. Bull: 19, 757-759. Brownell, J., Del Pozo, E., and Donatsch, P. (1980). Actu Endocrinol. 94, 304. Browning, A. J. F., Butt, W. R., and London, D. R. (1981). Actu Endocrinol. 97, 191. Brownstein, M. J., Eskay, R. L., and Palkovits, M. (1982). Neuropepptides 2, 197-201. Bruhn, O., Plotsky, P. M., and Vale, W. N. (1984a). Endocrinology 114, 57-62. Bruhn, T. O., Sutton, R. E., Rivier, C. L., and Vale, W. W. (1984b). Neuroendocrinology 39, 170- 175. Bruni, J. F., Hawkins, R. L., and Yen, S. S. C. (1982). Life Sci. 30, 1247-1254. Buckingham, J. C. (1980). Phurmacol. Rev. 31, 253-273. Buckingham, J. C. (1982). Neuroendocrinology 35, 1 1 1-1 16. Buckingham, J. C., and Cooper, T. A. (1984). Neuroendocrinology, 38,411-417. Bugnon, C., Bloch, B., Lenys, D., and Fellman, D. (1979). Cell. TissueRes. 199, 177-196. Burlet, A., Tonon, M. C., Tankosic, P., Coy, D., and Vaudry, M. (1983). Neuroendocrinology 37,64-72. Cahill, C., Watson, S. J., Knobloch, M., and Akil, H. (1983). Life Sci. 33 (Suppl. l ) , 53-55. Carlson, D. E., and Gann, D. S. (1984). Endocrinology 114, 317-324. Carlson, D. E., Dornhorst, A., Seif, S. M., Robinson, A. G., and Gann, D. S. (1982). Endocrinology 110,680-682.

62

MARK J. MILLAN AND ALBERT HEKZ

Carr, D. B., Bergland, R., Hamilton, A., Blume, H., Kasting, N., Arnold, M., Martin, J. B., and Rosenblatt, M. (1982). Science 217, 845-848. Carson, K. A., Nemerhoff, C. R., Rone, M. S., Young-Blood, W. W., Prange, A. J., Hanker, J. S., and Kizer, J. S. (1977). Brain Res. 129, 169-173. Carter, D. A., and Lightman, S. L. (1984). Neuroundocrinol Lull. 6, 95-100. Casanueva, F., Betti, R., Frigerio, C., Cocchi, B., Mentegazza, P., and Miiller, E. E. (1980). Endocrinology 106, 1239-1245. Casanueva, F., Betti, R., Cocchi, D., Chieli, T., Mentegazza, P., and Miiller, E. E. (198 1). Endocrinology 108, 157-163. Castanas, E., Giraud, P., Audigier, Y., Drissi, R., Bondourespue, F., Conte-Devolx, B., and Oliver, C. (1984). Mol. Pharmacol. 25, 28-45. Celis, M. E. (1980). Can. J . Physiol. Phannacol. 58, 326-329. Chaminade, M., Foutz, A. S., and Rossier, J. (1983). Life Sci. 33 (Suppl. l), 21-24. Chang, T.-L., and Loh, Y. P. (1983). Endocrinology 112, 1832-1838. Chantry, C. J., Seidler. F. J., and Slotkin, T. A. (1982). Neuroscience 7, 673-678. Chen, C. L. C., Dionne, F. T., and Roberts, J. L. (1983). Proc. Natl. Acad. Sci. U.S.A. 80, 2211-2215. Chernick, V., and Craig, R.J. (1982). Science 216, 1252-1253. Cheung, C. Y. (1982). Neurosci. Abstr. 8, 51. Chey, W. Y., Coy, D. H., Konturek, S. J., Schalley, A. V., and Tasler, J. (1980).J. Physiol. (London) 298,429-436. Chiba, T., Taminato, T., Kadowaki, S., Inoue, F., Mori, K., Seino, Y., Abe, H., Chihara, K., Matsukuro, S., Fujita, T., and Goto, Y. (1980). Endocrinology 106, 145-149. Chihara, K., Arimura, A., Coy, D. H., and Schally, A. V. (1978). Endocrinology 102, 281290. Ching, M. (1983). Endocrinology 112, 2209-221 1. Christensen, J. D., and Fjalland, B. (1 982). Acta Phannacol. Toxicol. 50, 113- 116. Cicero, T. J., Meyer, E. R., Bell, R. D., and Koch, G. A. (1976). Endocrinology 98,367-372. Cicero, T . J., Badger, T. M., Wilcox, C. E., Bell, R. D., and Meyer, E. R. (1977). J . Pharmacol. Exp. Ther. 203, 548-555. Cicero, T. J., Schainker, B. A., and Meyer, E. R. (1979). Endocrinology 104, 1286-1291. Cicero, T. J., Meyer, E. R., Gabriel, S. M., Beel, R. D., and Wilcox, C. E. (1980). Brain Res. 202, 151-164. Clarke, G., and Patrick, G. (1981). Neurosci. Lett. 39, 175-180. Clarke, G., and Merrick, L. (1982). Br. J. Pharmacol. 75, 122P. Clarke, G., Wood, P., Merrick, L., and Lincoln, D. W. (1979). Nature (London) 282, 746748. Clarke, G., Lincoln, D. W., and Wood, P. (1980).J. Physiol. (London) 303, 59P-60P. Clarke, G., Noble, R., and Waker1ey.J. B. (1981).J.Physiol. (London) 319, 115P-116P. Clement-Jones, V. V. (1982). I n “Neuropeptides: Basic and Clinical Aspects” (C. Fink and L. J . Whalley, eds.), pp. 182-195. Churchill Livingstone, Edinburgh. Cocchi, D., Casanueva, F., Betti, R., Zanardi, P., Motta, T., and Mentegazza, P. (1980). I n “Neuroactive Drugs in Endocrinology” (E. E. Miiller, ed.), pp. 191-200. Elsevier, Amsterdam. Cohen, M. R., Nurnberger, J. I., Pickar, D., Gershon, E., and Bunney, W. E. (1981). L f e Sci. 29, 1243-1247. Cohen, M. R., Pickar, D., Dubois, M., and Bunney, W. E. (1982). Psychiatr. Res. 6, 7-12. Cohen, M. R., I’ickar, D., Exstein, E., Gold, M. S., and Sweeney, D. R. (1984). A m . J. Psyrhiatry 141, 628-632. Conte-Devolx, B . , Oliver, C., Giraud, P., Gillioz, P., Castanas, E., Lissitzky, J. C:., Boudouresclue, F., and Millet, Y. (1981). L f e Sci. 28, 1067-1073.

T H E ENDOCRINOLOGY OF THE OPIOIDS

63

Conte-Devolx, B. Rey, M., Boudouresque, F., Giraud, P., Castanas, E., Mildet, Y., Codaccioni, J. L., and Oliver, C. (1983). Peptides 4, 301-304. Costa, E., Guidotti, A., Hanbauer, I., and Saiani, L. (1983). Fed. Proc., Fed. Am. Soc. Exp. Biol. 42, 2946-2952. Cote, T. E., Eskay, R. L., Frey, E. A., Grewe, C. W., Munemura, M., Stoof, J. L., Tsurato, K., and Kebabian, J. W. (1982). Neuroendocrinology 35, 217-224. Coulter, H. D., Elde, R. P., and Unverzagt, S. L. (1981). Peptides 2 (Suppl. I ) , 51-55. Cox, B. M., Ghazarossian, V. E., and Goldstein, A. (1980). Neurosci. Lett. 20, 85-88. Crabbe, J. C., Allen, R. G., Gaudette. N. b.,Kosobud, A., and Young, E. R. (1981). Proc. West. Pharmacol. Sac. 24, 29-31. Csontos, K., Rust, M., Hollt, V., Mahr, W., Kromer, W., and Teschemacher, H. J. (1979). Life Sci. 25, 835-844. Cushman, P., and Kreek, M. J. (1974). In “Narcotics and the Hypothalamus” (E. Zimmerman and R. George, eds.) pp. 161-172. Raven, New York. Dai, S., and Chan, M.-Y. (1983). Pharmacology 27, 180-184. Davis, J. M., Lowy, M. T., Yim, G. K. M., Laub, D. R., and Malven, P. V. (1983). Peptides 4, 79-83. Dean, D. M., Lemaire, S., and Livett, B. G. (1982).J.Neurochem. 38, 606. Deeter, W. T., and Mueller, G. P. (1981). Proc. Soc. Exp. B i d . Med. 168, 369-372. Delitala, G., Devilla, L., and Arata, L. (1981a). Acta Endocrinol. 97, 150-156. Delitala, G., Grossman, A., and Besser, G. M. (1981b). L$e Sci. 29, 1537-1544. Delitala, G., Grossman, A., and Besser, B. M. (1983). Clin. Endocrinol. 18, 401-405. Demarest, K. T., and Moore, K. E. (1981). Life Sci. 28, 1345-1351. Demeneix, B. A., Desaulles, E., Feltz, P., and Loeffler, J-P. (1984). Br. 1. Phurmacot. 82, 183- 190. Denko, C. W., Aponte, J., Gabriel, P., and Petricev, M. (1982).J . Rheurnatol. 9, 827-833. Dent, R. R., Guillemault, C., Albert, L. H., Posner, B. I., Cox, B. M., and Goldstein, A. (1981).J . Clin. Endocrinol. Metab. 52, 942-947. De Rotte, A. A., van Wimersma Greidanus, T. B., van Ree, J. M., Andringa-Bakker, E. A. D., and d e Weid, D. (1981). L$e Sci. 29, 8 2 5 4 3 2 . De Souza, E. B., and van Loon, G. R. (1982). Endocrinology 111, 1483-1491. Desphand, N., and Mitchell, I. (1983). Horm. Res. 17, 52-56. De Wald, D., and Lewis, R. V. (1983). Peptides 4, 121-123. Deyo, S. N., and Miller, R. J. (1982). Life Sci. 3 1, 2 17 1-2 175. Deyo, S. N., Swift, R. M., and Moller, R. J. (1979). Proc. Natl. Acad. Sci. U.S.A. 76, 30063009. Di Guilio, A. M., Yang, HI.-Y.T., Fratta, W., and Costa, E. (1979). Nature (London) 278, 646-647. Dodds, W. N., Jones, C. A., Murphy, C., Jayson, M. I. V., and Rees, J. M. H. (1983). Ann. Rheum. DZF. 42, 228. Dokas, L. A. (1983). Bruin Res. Rev. 5, 177-218. Dornhorst, A., Carlson, D. E., Seif, S. M., Robinson, A. G., Zimmerman, E. A., and Gann, D. S. (1980). Endocrinology 108, 1420-1424. Douglas, W. W., and Taraskevich, P. S. (1978).J. Physiol. (London) 285, 171-184. Dreifuss, J. J., Gahwiler, B. H., and Mathison, R. D. (1981). I n “Role of Peptides in Neuronal Communication” 0. L. Barker and T. G. Smith, eds.), pp. 135-158. Dekker, New York. Drouva, S. V., Epelbaum, J., Tapia-Arancibia, L., La Plante, E., and Kordon, C. (1981). Neuroendocrinology 32, 163-167. Duka, Th., Hollt, V., Przewlocki, R., and Wesche, D. (1978).Biochem. Biophys. Res. Cummun. 85, 1119-1127.

64

MARK J. MILLAN A N D ALBERT HEKZ

Dumont, M., and Lemaire, S. (1983). Neurosci. Abstr. 9, 1010. Dumont, M., Day, R., and Lemaire, S. (1983). Life Sci. 32, 287-294. Dupont, A., Cusan, L., Garon, M., Labrik, F., and Li, C. H. (1977). Proc. Natl. Acad. Sci. U.S.A. 74, 358-359. Dupont, A., Barden, N., Cusan, L., Merand, Y., Labrik, F., and Vaudry, H. (1980). Fed. Proc., Fed. Am. SOC.Ex$. Biol. 39, 2544-2550. Eberwine, J. H., and Roberts, J. L. (1984).J. Biol. Chem. 259, 2166-2170. Eddy, L. J., Joiner, D. W., and Glenn, T. W. (1984). Rex. Commun. Clin. Pathol. Phamacol. 43, 113-125. Eiden, L. E., and Ruth, J. A. (1982). Peptides 3,475-482. Eisenberg, R. M. (1980). Life Sci. 26, 935-943. Eisenberg, R. M. (1984). Life Sci. 34, 1185-1191. Elde, R., and Hokfelt, T . (1979). Annu. Rev. Physiol. 41, 587-602. Emrich, H. M., Hollt, V., Bergman, M., Kissling, W., Schmid, W., Zerssen, D. V., and Herz, A. (1980). In “Neural Peptides and Neuronal Communication” (E. Costa and M. ‘Trabucchi,eds.), pp. 489-502. Raven, New York. Endroczi, E. (1980).In “Progress in Psychoneuroendocrinology” (F. Brambilla, G. Racagni, and D. de Wied, eds.), pp. 87-97. Elsevier, Amsterdam. Engquist, A., Hicquet, J., Saurbury, N., and Blichert-Toft, M. (1981). Acta Anaesth. Scand. 25, 17-20. Enjalbert, A., Ruberg, M., Arancibia, S., Priam, M., and Kordon, C. (1979).Nature (London) 280, 595-597. Eriksson, E., Eden, S., and Modigh, K. (1981). Neuroendocrinology 33, 91-96. Evans, C. J., Weber, E., Erdelyi, E., Lorenz, R., and Barchas, J. D. (1982). Neurosci. Abstr. 8, 476. Evans, C. J., Erdelyi, E., Weber, E., and Barchas, J. D. (1983). Science 221, 957-960. Facchinetti, F., Nappi, G., Savoldi, F., and Genozzani, A. Z. (1981). Cejhatagza 1, 195-201. Fang, V. S., Tricou, B. J., Robertson, A., and Meltzer, H. Y. (1981). Life Sci. 29,931-938. Farrell, P. A., Gates, W. K., Maksud, M. G., and Morgan, W. P. (1982).J. Appl. Physiol. 52, 1245-1249. Farris, T. S., and Richards, G. E. (1980). Life Sci. 27, 1345-1349. Farsang, C., Vajda, L., Kapocsi, J., Malisak, Z., Alfoldi, S., and Varga, K. (1983). J . Clin. Endocrinol. Metab. 56, 865-866. Feldman, M., and Li, C. H. (1982). Regul. Pept. 4, 311-315. Feldman, M., Walsh, J. H., and Taylor, F. L. (1980). Gastroenterology 79, 294-298. Fergurson-Segall, M., Flynn, J. J., Walker, J., and Margules, D. L. (1982). Lqe Sci. 31, 2233-2236. Ferin, M., Wehrenberg, W. B., Lam, N. Y., Alston, E. J., and Van de Wiele, R. L. (1982). Endocrinology 111, 1652-1655. Ferland, L., Labrik, F., Cusan, L., Dupont, A,, LtSpine, J., Bedulieu, M., Denizeau, F., and Lemay, A. (1980). I n “The Endocrine Functions of the Brain” (M. Motto, ed.), pp. 271-296. Raven, New York. Ferri, S., Candeletti, S., Cocchi, D., Giagnoni, G., Rossi, T., Scoto, G., and Spaminato, S. (1982a). In “Regulatory Peptides: From Molecular Biology to Function” (E. Costa and M. Trabucchi, eds.), pp. 109-1 15. Raven, New York. Ferri, A., Cocchi, D., Locatelli, V., Spampinato, S., and Muller, E. (198213).Eur. J . PhamaC O ~ .77, 143-145. Finlay, J. C. W., Linstrom, P., and Petrusz, P. (1981). Neuroendocrinology 33, 28-42. Firemark, M. M., and Weitzman, R. E. (1979). Neuroscience 4, 1895-1902. Fischer, J. L., and Moriarty, C. M. (1977). Endocrinology 100, 1047-1054. Foresta, C., Marra, S., Federspil, G., and Scandellari, S. (1983). Horm. Metub. Res. 15,567568.

THE ENDOCRINOLOGY OF THE OPIOIDS

65

Forman, L. J., Sonntag, W. E., Van Vugt, D. A,, and Meites, J. (1981). Neurobiol. Aging, 2, 281-284. Forman, L. J., Sontag, W. E., Hylka, V. W., and Meites, J. (1983a). Life Sci. 33,993-999. Forman, L. J., Sonntag, W. E., and Meites, J. (1983b). Proc. SOC.Exp. Biol. Med. 173, 14-16. Forsling, M. K., and Aziz, L. A. (1983).J. Endocrinol. 99, 77-86. Fratta, W., Rosetti, Z. L., and Poggio, R. (1981). Neurosci. Lett. 24, 71-74. Friedman, E., Krieger, D. T., Mezey, E., Lkranth, C. S., Brownstein, M. J., and Palkovits, M. J. (1983). Endocrinology 112, 1943-1947. Fujimoto, J. M. (1971). I n “Narcotic Drugs: Biochemical Pharmacology” (D. H. Clouet, ed.), pp. 366-393. Plenum, New York. Fuller, R. W. (1981). Neuroendocrimlogy 32, 118-127. Gabriel, S. M., Simpkins, J. W., and Kalra, S. P. (1983). Endocrinology 113, 1806-1811. Gaillard, R. C., Schonenberg, P., Favrod-Come, C. A., Muller, A. F., Marie, J., Bockaert, J., and Jard, S. (1984). Proc. Natl. Acad. Sci. U.S.A. 81, 2907-2911. Gambert, S. R., Garthwaite, T . L., Pontzer, C. H., and Hagen, T. C. (1980). Science 210, 1271-1272. Gann, D. S., Ward, D. G., and Carlson, D. E. (1978). Recent Prog. Horm. Res. 34, 357-401. Ganong, W. F. (1980). Fed. Proc., Fed. Am. SOC.Exp. Biol. 39, 2923-2930. Ganong, W. F., Shinsako, J., and Reid, I. A. (1982). Ann. N.Y. Acad. Sci. 394, 619-624. Garthwaite, T. L., Martinson, D. R., Tseng, L. F., Hagen, T. C., and Menahan, L. A. (1980). Endocrinology 107, 671. Gautray, J. P., Jolivet, A., Vielh, J. P., and Guilleman, R. (1977).A m . J . Obstet. Gynecol. 129, 21 1-212. Genazzani, A. R., Facchinetti, R., Ricci-Denaro, M. G., Parrini, D., Petraglia, F., La Rosa, R., and D’Antona, N. (1981).J. Endocrinol. Invest. 4, 375-381. Genazzani, A. R., Nappi, G., Facchinetti, F., Mazzella, G. L., Parrini, D., Sinforiani, E., Petraglia, F., and Savoldi, F. (1982).J. Clin. Endocrinol. Metab. 55, 583-586. George, J. M., Reier, C. E., Lanese, R. R., and Power, J. M. (1974).J. Clin. Endocrinol. Metab. 38, 736-741. Gerner, R. H., Sharp, B., and Catlin, C. H. ( 1 982).J. Clin. Endocrinol. Metab. 55,358-360. Gianoulakis, C., Woo, N., Drouin, J. N., Seideh, N. G., Kalant, M., and Chretih, M. (1981). Life Sci. 29, 1973-1982. Gibbs, D., and Vale, W. (1982). Endocrinology 111, 1418-1420. Gibson, A., Ginsburg, M., Hall, M., and Hart, S. L. (1979a). Br.J. Pharmacol. 65, 139-146. Gibson, A., Ginsburg, M., Hall, M., and Hart, S. L. (1979b). Br.J. Pharmacol. 65,671-676. Gibson, M. J., Liotta, A. S., and Krieger, D. T. (1981). Neuropeptides 1, 349-362. Gibson, M. J., Colurso, G . J., Applebaum, L., and Krieger, D. T. (1983). Pepttdes 4, 305309. Giguere, G., and Labrik, F. (1982). Endocrinology 111, 1752-1754. Giguere, V., and Labrik, F. (1983). Biochem. Biophys. Res. Commun. 110, 456-462. Gigutre, V., CBtC, J., and LabriC, F. (1981). Endocrinology 109, 757-762. Gigukre, V., CBt6, J., and Labrit!, F. (1982). Endocrinology 110, 1225-1230. Gillies, G., and Lowry, W. (1982). In “Frontiers in Neuroendocrinology” (W. F. Ganong and L. Martini, eds.), pp. 45-74. Kaven, New York. Gillies, G. E., Linton, E. A,, and Lowry, P. J. (1982). Nature (London) 299, 355-357. Gilman, S. D., Schwartz, J, M., and Milner, R. J. (1982). Proc. Natl. Acad. Sci. U.S.A. 79, 4226-4230. Giraud, P., Lissitzky,J. C., Derolx, B. C., Gilloz, P., and Oliver, C. (1980).Eur.J. Pharmacol. 62,215-217. Givens, J. R., Wiedemann, E., Andersen, R. N., and Kitabchi, A. E. (198O).J. Clin. Endocrinol. Metab. 50, 975-976.

66

M A R K J . MILLAN A N D ALBERT HERZ

Gluckman, P. D., Martin-Hennenberg, C., Leisti, S., Kaplan, S. L., and Grurnbach, M. M. ( 1 980). Life Sci. 27, 1429- 1433. Goland, R. S., Wardlaw, S. L., Stark, R. L., and Fratz, A. G. (1981). J . Clin. Endocrinol. Metab. 52, 74-78. Govoni, S., and Yang, H. -Y. T. (1982).J . Neurochern. 36, 1829-1833. Govoni, S., Hanbauer, I., Hexum, T. D., Yang, H. -T. T., Kelly, G. D., and Costa, E. (1981). Neuropharmacology 20, 639-645. Grandison, L., and Guidotti, A. (1977). Nature (London) 270, 357-359. Grandison, L., Frdtta, W., and Guidotti, A. (1980). Life Scz. 26, 1633-1642. Green, I. C., Perrin, J., Penham, E., Yaseen, A., Ray, K., and Howell, S. L. (1983a). Diabetes 32,685-690. Green, I. C., Ray, K., and Perrin, D. (1983b). Horm. Metab. Kes. 15, 124-128. Grossman, A,, Besser, G. M., Milles, J. J., and Baylis, P. H. (1980). Lancet 2, 1108-1 110. Grossman, A., Moult, P. J. A., Gaillard, R. C., Delitala, G., Toff, W. D., Rees, L. H., and Besser, G. M. (1981a). Clin. Endocrinol. 14, 41-47. Grossman, A,, Stubbs, W. A., Gaillard, R. C., Delitala, G., Rees, L. H., and Besser, G. M. (1981b). Clzn. Endocrinol. 14, 381-386. Grossman, A., Gaillard, R. C., McCartney, P., Rees, L. M., and Besser, G. M. (1982a). Clin. Endocrinol. 17, 279-286. Grossman, A., Moult, P. J. A., McFutyre, H., Evans, J., Silverstone, R., Rees, L. H., and Besser, G. M. (1982b). Clin. Endocrind. 17, 379-388. Grossman, A,, Smith, R., Van Loon, G. R., Brown, G. M., Rees, L. H., and Besser, G. M. (1982~). Neuroendocrinol. Lett. 4, 223-232. Grossman, A., West, S., Williams, J., Evans, J., Rees, L. H., and Besser, G. M. (1982d). Clin. Endocrinol. 16, 317-320. Grube, D., Voigt, K. M., and Weber, E. (1978). Histochemistry 59, 75-79. Gudelsky, G. A., and Porter, J. C. (1979). Life Sci. 25, 1697-1702. Guilleman, R., Vargo, T., Rossier, J., Minick, S., Ling, N., Rivier, C., Vale, W., and Bloom, F. (1977). Science 197, 1367-1369. Guilleman, R.,Brazeau, P., Bohlen, P., Esch, F., Ling, N., and Wehrenberg, W. B. (1982). Science 218, 585-587. Guiol, C., Montaslruc, J.-L., and Montastruc, P. (1984). f3r.J. Phnrmacol. 81, 85-90. Gullner, H. G., and Hill, J. R. (1983).J. Clzn. Invest. 71, 124-128. Gunion, M. W., and Peters, R. H. (1981). Arn.J. Physiol. 241, 173-184. Hadley, M. E., Davis, M. D., and Morgan, C. M. (1977). Front. Horm. Res. 4, 94-104. Halasz, B., Nagy, B., Molnir, J., Marton, J., Banky, Z., Lukats, O., and Vizi Szilveszter, E. (1981).I n “Advances in Physiological Science” (E. Stark, G. B. Makara, B. Halasz, and Gy. Rappay, eds.), Vol. 14, pp. 315-326. Pergarnon, Oxford. Haldar, J., and Bude, V. (1981). Proc. SOC.Exp. Biol. Med. 168, 10-14. Haldar, J., and Sawyer, W. H. (1978). Proc. SOC.Exp. Bzol. Med. 157, 476-480. Haldar, J., Hoffman, D. L., and Zimmerman, E. A. (1982). Pefitides 36, 663-668. Hall, T. R., Advis, J. P., Smith, A. F., and Meites, J. (1976). ZRCS J . Med. Sci. 4, 559. Hanbauer, I., Kelly, G. D., Saiani, C., and Yang, H. T. T. (1982). Peptides 3, 469-473. Hanko, .J. H., and Hardebo, J. E. (1978). Eur. J . Pharmacol. 51, 295-297. Haracz, J. L., Bloom, A. S., Wang, R. I. H., and Tseng, L. F. (1981).Neuroendocrinology 33, 170-175. Haskins, J. T., and Moss, R. L. (1983). Brain Res. 268, 185-188. Haskins, J. T., Gudelsky, G. A., Moss, R. L., and Porter, J. C. (1981). Endocrinology 108, 767-77 1 . Hazum, E., Chang, K. J., and Cuatrecasas, P. (1979). Science 205, 1033-1035.

THE ENDOCRINOLOGY OF THE OPIOIDS

67

Hermansen, K. (1982). Diubetologia 23, 174. Hermansen, K. (1983). Endocrimlogy 113, 1149-1 154. Heybach, J. P., and Vernikos, J. (1981). Eu7.J. Pharmacol. 75, 1-6. Ho, W. K. K., Wen, H. L., Fung, K. P., Ng, Y. H., Au, K. K., and Ma, L. (1977). Clin. Chem. A C ~ 75, U 415-419. Ho, N. K. K., Wen, L., and Ling, N. (1980). Neuropharmacology 19, 117-120. Hokfelt, T., Eddo, R., Fuxe, K., Johansson, O., Ljungdahl, A,, Goldstein, M., Luft, R., Efendic, S., Nilsson, C., Terenius, L., Ganten, D., Jeffcoate, J. L., Rehfeld, J., Said, S., Perez de la Mora, M . , Possani, L., Tapia, R., Teran, L., and Palacios, R. (1978). In “The Hypothalamus” (R.J. Baltlessarini and J . B. Martin, eds.), pp. 69-134. Raven, New York. Hokfelt, T., Johansson, O., Elde, R., Goldstein, M., Jeffcoate, S., and White, S. (1980). In “Neuroactive Drugs in Endocrinology” (E. E. Muller, ed.), pp. 19-33. Elsevier, Amsterdam. Holaday, J. W. (1983). Biochem. Phamacol. 32, 573-585. Holaday, J. W., and Loh, H. H. (1982). In “Hormonal Proteins and Peptides”(C. H. Li, ed.), Vol. 10, pp. 203-291. Academic Press, New York. Holaday, J. W., Tortella, F. C., and Belenky, G. C. (1981). In “The Role of Endorphins in Neuropsychiatry” (H. M. Emrich, ed.), pp. 14% 157. Karger, Basel. Holaday, J. W., D’Amato, R. J., Ruvio, B. A., Feuerstein, G., and Faden, A. I. (1983). Circ. Shock 11, 201-210. Holland, A,, Schick, R., and Schusdziarra, V. (1983) Acta Endocrinol. Suppl. 253, 79. Hollt, V. (1981). In “The Role of Endorphins in Neuropsychiatry”(H. M. Emrich, ed.), pp. 1-18. Karger, Basel. Hollt, V. (1983). Trends Neurosci. 6, 24-26. Hollt, V., Przewlocki, R., and Herz, A. (1978). Life Sci. 23, 1057-1066. Hollt, V., Haarmann, I., Przedocki, R., and Jerlicz, M. (1980). In “Neural Peptides and Neuronal Communication” (E. Costa and M. Trabucchi, eds.), pp. 399-407. Raven, New York. Hollt, V., Haarmann, I., and Herz, A. (1981a).J. Neurochem. 37, 619-626. Hollt, V., Haarmann, I., Seizinger, B. R., and Herz, A. (1981b). Neuroendocrinolo~y 33, 333-339. Hollt, V., Haarmann, I., Grirnm, C., Herz, A., Tulunay, C . F., and Loh, H. H. (1982a). Life Sci. 31, 1883-1886. Hollt, V., Haarmann, I., Seizinger, B. R., and Herz, A. (1982b). Endocrinology 110, 18851891. Holzbauer, M., Sharman, D. F., Godden, U., Mann, S. P., and Stephens, D. B. (1980). Neuroscience 5 , 1959-1968. Hong, J. S., Yoshikawa, K., Hudson, P. M., and Uphouse, L. L. (1982). Life Sci. 31,21812184. Hope, J., and Lowry, P. J. (1981). Front. H u m . Res. 8, 44-61. Howlett, T . A,, Tomlin, S., Ngahfoomg, L., Rees, L. H., Bullen, B. A,, Skrinar, J. S., and McArthur, J. W. (1984). Br. M r d . Bull. 288, 1950-1952. Hughes, J. (1981). Trends Phurmacol. Sci. 2, 21-24. Huidobro, F., and Huidobro-Toro, J. P. (1979). Eur. J . Pharmacol. 59, 55-64. Huidobro-Toro, J. P. (1980). Br.J. Pharmacol. 71, 51-56. Huidobro-Toro, J . P., and Huidobro, F. (1981).J. Pharmacol. Exp. Ther. 217, 579-585. Hulse, G. K., and Coleman, G . J. (1983). Phamacol. Biochem. Behav. 19, 795-799. Hulse, G., Coleman, G . , Nicholas, J., and Greenwood, K. (1982).J. Reprod. Fertil. 66,451456.

68

MARK J. MILLAN AND ALBERT HER2

Hurwitz, A. (1981).J. Phannacol. Ex#. Ther. 216, 90-94. Hutchinson, J. S., Lim, A., Di Nocolantino, R., Clements, J. A,, and Funder, J. W. (1981). Clin. Ex#. P h a m c o l . Physiol. 8, 455-457. Ieiri, T., Chen, H. T., and Meites, J. (1979). Neuroendocrinology 29, 288-292. Ieiri, T., Chen, H. T., Campbell, G. A., and Meites, J. (1980a). Endocrinology 106, 15681570. Ieiri, T., Chen, H. T., and Meites, J. (1980h). Life Sci. 26, 1269-1274. Ikeda, Y., Nakao, K., Yoshimasa, T., Yanaihara, N., Numa, S., and Imura, H. (1982). Biochem. Biophys. Res. Commun. 107, 656-662. Imura, H., Nakai, Y., Nakao, K., Oki, S., and Tanaka, I. (1982). In “Neuroendocrine Perspectives” (E. E. Muller and R. M. Macleod, eds.), Vol. 1 , pp. 137-167. Elsevier, Amsterdam. Iovino, M., Rouhani, S., and Voenath, S. (1983). Neuroendocrinol. Lett. 5, 239-241. Ipp, E., Dobhs, R., and Unger, R. H . (1978). Nature (London) 276, 190-191. Ipp, E., Dhorajwala, J., Pugh, W., Moossa, A. R., and Rubenstein, A. H. (1982). Endocrinology 111,2110-21 16. Ishikawa, S.-E., and Schrier, R. W. (1982).J. Clin. Invest. 69, 666-672. Ishizuka, B., Wilkes, M. M., Stewart, R. D., Rammussen, D. D., and Yen, S. S. C. (1982). Life Sci. 31, 2113-2120. Ito, S., Iwanaga, T., Yui, R., Yamaguchi, K., Hama, H., Kamoi, K., and Shibata, A. (1981). Life Sci. 29, 1457-1461. Iversen, L. L., Iversen, S. D., and Bloom, F. E. (1980). Nature (London) 284, 350-351. Ixart, G., Szafarczyk, A., Malaval, F., Nougier-Soulk, J., and Assenmacher, I. (1980). C . R . Acad. Sci. Paris 291, 265-270. Ixart, G., Alonso, G . , Szanfarczyk, A., Malaval, V., Nougvier-Soulk, J., and Assenmacher, I. (1982). Neuroendocrinology 35, 270-276. Jackson, S., and Lowry, P. L. (1983). Neuroendocrinology 37, 248-257. Jacquet, Y. (1977). Science 201, 1032-1034. Jezova, D., Vigas, M., and Jurcovicova, J. (1982). Life Sci. 31, 307-314. Johnson, J. H. (1982). Neuroendocrinology 35, 169-172. Johnson, J. H., Maughan, G. T., and Anderhub, L. (1982). Life Sci. 30, 1473-1478. Jones, M. T., and Gillham, B. (1980). In “Synthesis and Release of Adenohypophyseal Hormones” (M. Jutisz and K. McKerns, eds.), pp. 587-615. Plenum, New York. Jones, M. T., Hillhouse, E., and Burden, J. (1976). I n “Frontiers in Neuroendocrinology” (L. Martini and W. F. Ganong, eds.), Vol. 4, pp. 195-226. Raven, New York. Jones, M. T., Gillham, B., Di Renzo, G., Beckford, V., and Holmes, M. C. (1981). Front. Honn. Res. 8, 12-32. Jonsson, G., Fuxe, K., and Hokfelt, T. (1972). Brain Res. 40, 271-281. Judd, A. M., and Hedge, G. A. (1982). Life Sci. 31, 2529-2436. Judd, A. M., and Hedge, J. M. (1983). Endocrinology 113,706-710. Kalra, S. P. (1981). Endocrinology 109, 1805-1810. Kalra, S. P., and Crowley, W. K. (1982). Endocrinology 111, 1403-1405. Kalra, S. P., and Gallo, R. V. (1983). Endocrinology 113, 23-28. Kalra, S. P., and Kaira, P. S. (1984). Neuroendocrinology 38, 418-416. Kalra, S. P., and Simpkins, J. W. (1981). Endocrinology 109, 776-782. Kamoi, K., White, K., and Robertson, G. L. (1979). Clin. Res. 27, 254a. Kan, K. W., Baird, A., Seidah, N. G., Lis, M., Chrktien, M., Ling, N., Guilleman, R., and Solomon, S. (1977). Biochem. Biophys. Res. Commun. 79, 553-559. Karteszi, M., Makara, G. B., Lasz16, F. A,, Rappay, G., and Stark, E. (1982). Ann. N.Y. Acad. Sci. 394, 630-633. Kasson, B. G . , and George, R. (1983).j.Phannacol. Exp. Ther. 224, 273-281.

THE ENDOCRINOLOGY OF THE OPIOIDS

69

Katakami, M., Kato, Y., Matsushita, N., Hiroto, S., Shimatsu, A., and lmura, H. (1981a). Neuroendocrinology 33, 129- 135. Katakami, H., Kato, Y., Matsushita, N., Shimatsu, A,, and Imura, H. (1981h). Endocrinology 109, 1033-1036. Kay, N. E., and Morley, J. E. (1983). Clin. Res. 31, 346A. Kerdelhue, B., Karteszi, M., Pasqualini, C., Reinherg, A., Mezey, E., and Palkovits, M. (1983). Brain Res. 261, 243-248. Khachaturian, H., Lewis, M. E., and Watson, S. J. (1982a). Life Sci. 31, 1879-1882. Khachaturian, H., Watson, S. J., Lewis, M. E., Coy, D., Goldstein, A., and Akil, H. (1982b). Peptides 3, 94 1-954. Khorrarn, O., Mizunuma, K., and McCann, S. M. (1982). Neuroendocrinology 34,433-437. Kirnball, C. D., Huang, S. M., Torget. C. E., and Huock, J. C. (1980). Lancet 2, 418-419. Kimball, C. D., Chang, C. M., Huang, S. M., and Huock, J. C. (1981).Am.J. Obstet. Gynecal. 140, 157-169. Kinoshita, F., Nakai, Y., Katakami, H., Kato, Y., Yajima, H., and lmura, H. (1980). Life Sci. 27,843-846. Kiser, R. S . , Gatchel, R. J., Bhatia, K., Khatami, M., Huang, X.-Y., and Altshuler, K. 2. (1983). Lancet 2, 1394-1396. Kizer, J. S., Palkovits, M., and Brownstein, M. J. (1976). Brain Res. 108, 363-370. Klein, K. L., Wilson, S. P., Dzielak, D. J., Yang, W.-H., and Viveros, D. H. (1982). Neuroscience 7, 2255-2261. Knepel, W., and Nutto, D. (1983). Naunyn-Schmiedeberg’s Arch. Pharnacol. S 322, R83. Knepel, W., and Reimann, W. (1982). Brain Res. 238,484-488. Knepel, W., Anhut, H., Nutto, D., and Hertting, G. (1980a). Eur. J. Pharmacol. 68, 359363. Knepel, W., Nutto, D., Anhut, H., and Hertting, G. (1980b). Eur. J . Pharmacol. 65, 449450. Knepel, W., Nutto, D., and Hertting, G. (1981). Neuropeplides 2, 67-73. Knepel, W., Anhut, H., Nutto, D., Holland, A., and Mayer, D. K. (1982a). Eur.1. P h a m a C O ~ .80, 247-249. Knepel, W., Benner, K., and Hertting, G. (1982b). Eur. J. Phannacol. 81, 645-654. Knepel, W., Nutto, D., Anhut, H., and Hertting, G. (1982~).Eur. J. Phannacol. 77, 299306. Knepel, W., Nutto, D., and Hertting, G. (1982d). Neuroendocrinology 34, 353-356. Knepel, W., Nutto, D., and Anhut, H. (1983a). Regul. Pept. 7, 9-919. Knepel, W., Nutto, D., and Meyer, D. K. (1983b). Life Scz. 33, (Suppl. l), 499-502. Knepel, W., Homolka, L., Vlaskovska, M., and Nutto, D. (1984). Endocrinology 114, 17971804. Knuth, V. A., Sikand, G. S., Casamera, F. P., Havlicek, V., and Friesen, H. G. (1983). Life Sci. 33, (Suppl. l), 1443-1450. Knych, E. T., and Prohaska, J. R. (1981). LifeSci. 28, 1987-1994. Koch, B., Lutz-Bucher, B., and Mialhe, C. (1980). In “Synthesis and Release of Adenohypophyseal Hormones” (M. Jutisz and W. K. McKerns, eds.), pp. 561-585. Plenum, New York. Koenig, J. L., Mayfield, H. A., McCann, S. M., and Krulich, L. (1979). Life Sci. 25, 853864. Koenig, J. L., Mayfield, M. A,, Coppins, R. J., McCann, S. M., and Krulich, L. (1980). Bruin Res. 197, 453-468. Koenig, J. L., Mayfield, M. A., and Krulich, L. (1982). Neurosci. Abstr. 8, 56. Koenig, J. L., Mayfield, M. A., and Krulich, L. (1983). Fed. Proc., Fed. Am. SOC.Exp. Biol. 42, 301.

70

MARK.]. MILLAN A N D ALBERT HERZ

Koh, T., Nakai, Y., Kinoshita, F., Tsuji, S., Tsukada, T., and Imura, H. (1983). Eur. J. Pharmacol. 89, 275-278. Kokka, N., and George, R. (1974). In “Narcotics and the Hypothalamus” (E. Zirnmerman and R. George, eds.), pp. 137-157. Raven, New York. Kolodziejcyk, E., Baertschi, A. J., and Tramu, G. (1983). Neuroscience 9, 261-270. Konturek, S. J. (1980). Am. J . Gastroenterol. 74, 285-291. Konturek, S. J., Kwiecien, N., Obtotowicz, W., Swierczek, J., Bielanski, W., Olesky, J., and COY,D. H. (1983). Gut 24, 740-745. Kordon, C., Enjalbert, A., Epelbaum, J., and Rotsztejn, W. (1979). In “Brain Peptides: A New Endocrinology” (A. M. Gotto, E. J. Peck, and A. E. Boyd, eds.), pp. 277-293. Elsevier, Amsterdam. Kouchich, F. J., Quock, R. M., and Tseng, L.-F. (1983). Fed. Proc., Fed. Am. Sac. Exp. Biol. 42, 654. Koves, K., Marton, J., Molnar, J., and Halasz, B. (1981). Newoendocrinology 32, 82-86. Kraicer, J., Zimmermann, E. A., and Moor, B. C. (1980). I n “Catecholamines and Stress: Recent Advances” (E. Usdin, R. I. J . Kvetnasky and M. Kopin, eds.), pp. 131-136. Elsevier, Amsterdam. Krieger, D. T., Liotta, D. S., Suda, T., and Yamaguchi, K. (1980). In “Brain and Pituitary Peptides” (W. Wuttke, A. Weindl, K. H. Voigt, and R. R. Dries, eds.), pp. 35-45. Karger, Basel. Kromer, W., Hollt, V., Schmidt, H., and Herz, A. (1981). Neuroendocrinol. Lett. 25,53-56. Krulich, L. (1982). Neuroendocrinology 35, 139-147. Kumakura, K., Karoum, F., Guidotti, A., and Costa, E. (1980). Nature (London) 283,489492. Kumar, M. S. A,, Chen, C. L., and Muther, T. G. (1979). L f e Sci. 25, 1687-1696. Kumar, M. S. A., Chen, C:. L., and Huang, K. H. (1980). Newohiol. A p n g 1, 153-155. Laasberg, L. H., Johnson, E. E., and Hedley-White, J. (1980).J. Pharmacol. Exp. Ther. 212, 496-502. Lahti, R. A., and Collins, R. J. (1982). Pharmacol. Biochem. Behuu. 17, 107-109. Lakoski, J. M., and Gebhart, G. F. (1981). Neuroendocrinology 33, 105-111. Lakoski, J. M., and Gebhart, G. F. (1982). Brain Res. 232, 231-237. Lamberts, S. W. J., Bons, E. G., and del Pozo, E. (1983a). Life Scz. 32, 755-758. Lamberts, S. W. J., Bons, E. G., Uitterli, P., and Hackeng, W. H. (1983b).J. Endocrinol. 96, 395-400. Lamberts, S. W. J., Verleun, T., Oosterlom, R., d e Jong, F., and Hackeng, W. H. L. (1984). J . Clin. Endocrinol. Metab. 58, 298-303. Land, H., Schutz, G., Schmale, H., and Richter, D. (1982). Nature (London) 295, 299-303. Lang, R. E., Briickner, U. B., Hermann, K., Kempf, B., Rascher, W., Sturm, V., Unger, Th., and Ganten, D. (1982). In “Regulatory Peptides: From Molecular Biology to Function” (E. Costa and M. Trabucchi, eds.), pp. 363-368. Raven, New York. Lang, R. E., Taugner, G., Gaida, W., Ganten, D., Kraft, K., Unger, Th., and Wunderlich 1. (1983). Eu7.J. Pharmacol. 86, 117-120. Larsson, L.-I. (1981). Proc. Natl. Acud. Scz. U.S.A. 78, 2990-2994. Larsson, L.-l., and Stangaared-Pedersen, K. (1982).J . Neurosci. 2, 861-878. Leander, D. (1983a).J. Pharmacol. Exp. Ther. 224, 89-94. Leander, J. D. (1983b).J. Pharmacol. Ex$. Ther. 227, 35-41. Lee, S., Panerai, A. E., Bellabarba, D., and Friesen, H. G. (1980). Endocrinology 107,245248. Legros, J. J., Chiodera, P., Geenon, V., Smitz, S., and von Frcnckel, R. (1984). J . Clzn. Endocrinol. Metah. 58, 10.5- 109. Lemaire, I., Tseng, R., and Lemaire, S. (1978). Proc. Natl. Acad. Sci. U.S.A. 75,6240-6242.

THE ENDOCRINOLOGY OF THE OPIOIDS

71

Lkranth, E. S., Plakovits, M., and Krieger, D. T . (1983). Neuroscience 9, 289-296. Levin, E. R., Sharp, B., Meyer, N. V., and Carlson, M. E. (1981). Endocrinology 109, 146151. Levin, E. K., Sharp, B., and Carlson, H. E. (1984). Life Sci. 35, 1535-1545. Lewis, J, W., Tordoff, M. G., Sherman, J. E., and Liebeskind, J. C. (1982). Science 217, 557-559. Lichtensteiger, W., and Monnet, F. (1979). Life Sci. 25, 2079-2087. Lightman, S. L., Langdon, N., and Forsling, M. L. (1980).J. Clin. Endocrinol. Metab. 51, 1447-1449. Lightman, S. L., Iversen, L. L., and Forsling, M. L. (1982). J . Neurosci. 2, 78-81. Lightman, S. L., Forsling, M., and Todd, K. (1983a). Endocrinology 113, 77-80. Lightman, S. L., Ninkovic, M., Hunt, S. P., and Iversen, L. L. (1983b). Nature (London), 305,235-237. Lim, A. T. W., and Funder, J . W. (1984). Endocrinology 115,575-580. Lim, A. T., Khalid, B. A. K., Clements, J., and Funder, J. W. (1983a).J. Clin. Invest. 69, 1191-1 198. Lim, A. T., Lolait, S., Barlow, J., Toh, B. M., and Funder, J. W. (1983b). Nature (London) 306,709-7 1 1. Linton, E. A,, Gilles, G. E., and Lowry, P. J. (1983). Endocrinology 1878-1883. Liotta, A. S., Houghten, R., and Krieger, D. T. (1982). Nature (London) 295, 593-595. Lis, M., Larivikre, N., Maurice, G., Julesz, J., Seidah, N., and Chrktien, M. ( 1 982). Life Sci. 30, 1159-1164. Lissitzky, J. C., Giraud, P., Conte-Devolx, B., Gillioz, P., Boudouresque, F., Eskay, R. L., and Oliver, C. (1980). Neurosci. Lett. 19, 191-195. Liston, D. R., Vanderhaeghen, J. J., and Rossier, J. (1983). Nature (London) 302, 62-65. Login, I. S., and Macleod, R. M. (1979). Eur. J . Pharmacol. 60, 253-255. Lomax, P., and George, R. (1966). Brain Res. 2, 361-367. Lomax, R., Kokka, N., and George, R. (1970). Neuroendoc&zology 6, 146-152. Lopker, A., Abood, L. G., Hers, W., and Lionetti, F. J. (1981). Biocheni. Pharmacol. 29, 1361-1365. Lotstra, F., Vierendeels, G., Deschepper, C., and Vanderhaeghen, J. J. (1982). Neurosci. Lett. Suppl. 10, S299. Lotti, V., Kokka, N., and George, R. (1969). Neuroendocrinology 4, 326-332. Lutz-Bucher, B., and Koch, B. (1980). Eur. J . Pharmacol. 66, 375-378. Lutz-Bucher, B., Koch, B., and Alzein, M. (1982). Neuroendocrinol. Lett. 4, 103-107. Lymangrover, J. R., Dokas, L. A., Kong, A., Martin, R., and Saffran, M. (1981).Endocrinology 109, 1132-1 137. Lymangrover, J. R., Keku, E., and Eldridge, J. C . (1983). L f e Sci. 34, 1605-1612. McIntosh, C. H. S., Mordhorst, T., Nishimura, E., Pederson, R. A., Thompson, B., and Brown, J. C. (1982). Proc. Int. Symp. Gastrointest. Harm., 4th, Stockholm p. 6. McIntosh, C. S. H., Kwok, Y. N., Mordhurst, T., Nishimura, E., Pederson, R. A,, and Brown, J. C. (1983). Can.J. Physiol. Pharmacol. 161, 657-663. McCann, S. M., Mizaniima, €I., Samon, W. K., and Lumpkin, M. P. (1983). Psychonruroend o c r i ~ ~ 8, o l299-308. ~~ Mains, R. E., and Eipper, B. A. (1981). I n “Neurosecretion and Brain Peptides” (J. B. Martin and K. L. Bick, eds.), pp. 35-47. Raven, New York. Makara, G. B., Stark, E., Karteszi, M., Gellinger, E., Rappay, Gy., and Szabo, D. (1980). Neuroendocrinology 3 1, 237-243. Malizaia, E., Andreucci, D., Paolucci, D. F., and Crescenzi, A. (1979). Lancet 2, 535536. Manberg, P. J., Youngblood, W. W., and Kizer, J. S. (1982). Brazn Res. 241, 279-284.

72

MARK J. MILLAN A N D ALBERT HERL

Mannelli, M., Maggi, M., de Feo, M. L., Cuomo, S., Forki, G., Moroni, F., and Giusti, G. (1983). New En& J . Med. 11, 654-655. Mannisto, P. T., Kuahala, P., Tuominen, R., and Mattila,J. (1984). Lqe Sci. 35, 1101-1 107. Margioris, A. N., Liotta, A. S., Vaudry, C., Bardin, W., and Krieger, D. T. (1983).Endocrinology 113,663-67 1. Margules, D. L., Moisset, B., Lewis, M. J., Shibuya, H., and Pert, C. B. (1978). Science 202, 988-991. Marko, M. (1982). Acta Endocrinol. 99 (Suppl.), 85. Marko, M., and Romer, D. (1983). Life Sci. 33,233-240. Martin, J. B., Audet, J., and Saunders, A. (1975). Endocrinology 96, 839-847. Martin, J. B., Tolis, G., Woods, I., and Gayda, H. (1979). Brain Res. 168, 210-215. Martin, R., and Voigt, K. H. (1982). Life Sci. 31, 1729-1732. Martin, R., Geis, R., Holl, R., Schafer, M., and Voigt, K. H. (1983a). Neuroscience 8, 213227. Martin, R., Moll, V., and Voigt, K. H. (1983b). Life Sci. 33 (Suppl. I), 64-72. Marton, J., Molatr, J., Koiies, K., and Halasz, B. (1981). Life Sci. 28, 737-743. Masala, A., Satta, G., Alagra, S., Zolo, T. A., Rovasio, R. P., and Rassus, P. (1983). Acta Endocrinol. 103,469-472. Mathison, R. (1981). i n “Neurosecretion: Molecules, Cells, Systems” (U. S. Farner, and K. Lederis, eds.), pp. 167-175. Plenum, New York. Matsumura, M., Fukuda, N., Saito, S., and Mori, M. (1982a). Regul. P e p . 4, 173-181. Matsumura, M., Saito, S., and Fuijino, M. (1982b). Regul. Pept. 3, 173-181. Matsumura, M., Yamanoi, A., Yamamoto, S., and Saito, S. (1983). Neuroendocrinology 36, 443-448. Matsuoka, M., Mulrow, P. J., Franco-Saenz, R., and Li, C. H. (1981).J. Clin. Invest. 68, 752-759. Matsushita, N., Kato, Y., Shimatsu, A., Katakami, H., Fujino, M., Matsuo, H., and Imura, H. (1982). Biochem. Biophys. Res. Commun. 107, 735-741. Matthews, J., Akil, H., Greden, J., and Watson, S. (1982). Life Sci. 31, 1867-1870. May, P., Mirtler, J., Manougian, A., and Ertel, N. (1979). Horm. Metab. Res. 11, 30-33. Mayer, G., Wessel, J., and Kobberling, J. (1980). Clin. Endocrinol. 13,413-416. Maysinger, D., Hollt, V., Seizinger, B. R., Mehraein, P., Pasi, A., and Herz, A. (1982). Neuropeptides 2, 22 1-225. Medback, S., Mason, D. F. J., and Rees, L. H. (1983). Regul. Pept. 7, 195-206. Mehrishi, J. N., and Mills, I. H. (1983). Clin. Immunol. Immunopathol. 27, 240-249. Meites, J. (1982). Neuroendocrinology 34, 151-156. Meites, J., Bruni, J. F., van Vugt, D. A., and Smith, A. F. (1979). Life Sci. 24, 1325-1336. Melis, G. B., Paoletti, A. M., Gambacciani, M., Mais, V., and Fioretti, P. (1984). Neuroendocrinology 39, 60-63. Mennin, S. P., and Saland, L. C. (1980). Neuroendocrinology 31, 385-389. Merchenthaler, I., Vigh, S., Petrusz, P., and Schally, A. (1983). Regul. Pept. 5, 295-305. Merin, M., Hollt, V., Przewlocki, R., and Herz, A. (1980). Life Sci. 27 281-289. Meunier, H., and Labrik, F. (1982). Prog. Neuro-Psychophannacol. 6, 41 1-415. Meunier, M., Lefkvre, G., Dumont, D., and Labrit, F. (1982). Life Sci. 31 2129-2135. Meyer, D. K., Anhut, H., Nutto, D., Beinfeld, M. C., and Knepel, W. (1982).Neuropeptides 2,371-375. Mezey, E., De Kloet, E. R., Feminger, A., Chretien, M., Kiss, J., and Palkovits, M. (1981). Adv. Physiol. Sci. 14, 295-301. Mezey. E., Kiss, J. Z., Skirboll, L. R., Goldstein, M., and Axelrod, J. (1984a). Nature (London) 310, 140-141. Mezey, E., Leranth, C. S., Brownstein, M. J., Friedman, E., Krieger, D. T., and Plakovits, M . (1984b). Bruin. RPS.294, 231-237.

T H E ENDOCRINOLOGY OF THE OPIOIDS

73

Micevych, A., and Elde, R. (1980).J. Comp. Neurol. 190, 135-146. Miki, N., Sonntag, W. E., Forman, J., and Meites, J. (1981). Proc. Soc. Exp. B i d . Med. 168, 330-333. Miki, N., Ono, M., and Shikume. K. (1984). Endocrinoiogy 114, 1950-1952. Millan, M. H., Millan, M. J., and Herz, A. (1Y82a).Neurosci. Lett. 29, 269-273. Millan, M. H., Millan, M. J., and Herz, A. (1982b). Neurosci. Lett. 34, 165-170. Millan, M. H., Millan, M. J., and Herz, A. (1984). Neurocndocrinology 38, 108-1 16. Millan, M. J. (1981). In “The Role of Endorphins in Neuropsychiatry” (H. M. Emrich, ed.), pp. 49-67. Karger, Basel. Millan, M. J. (1982). Methods Find. Exp. Clin. P h a m o l . 4, 445-462. Millan, M. J., Przewlocki, R., Jerlicz, M., Gramsch, Ch., and Herz, A. (1981a). Bruin Res. 208, 325-338. Millan, M. J., Tsang, Y. F., Przewlocki, R., Hollt, V., and Herz, A. (1981b). Neurosci. Lett. 24,75-79. Millan, M. J., Millan, M. H., and Herz, A. (1982). Neuroendocrinology 35, 449-456. Millan, M. J., Millan, M. H., and Herz, A. (1983a). Brain Res. 263, 283-294. Millan, M. J., Millan, M. H., and Herz, A. (198313). Neuroendocrinology 36, 310-319. Life Sci. 35 (Suppl. l), 29Millan, M. J., Millan, M. H., Tsang, Y. F., and Herz, A. (1983~). 32. Millan, M. J., Millan, M. H., Clonkowski, A., and Herz, A. (1984a). Brain Res. 298, 243252. Millan, M. J., Millan, M. H., Colpaert, C., and Herz, A. (1984b). Pain (Suppl. 12), 5350. Millar, R. P., Sheward, W. J., Wegener, I., and Fink, G. (1983). Bruin Res. 260, 334-337. Miller, M. (1975). Neuroendocrinology 19, 241-251. Missale, C., Govoni, S., Croce, L., Bosio, A., Spano, P. F., and Trabucchi, M. (1983). J . Neurochem. 40, 20-24. Mitchell, L. D., Barron, K., Brody, M. J., and Johnson, A. K. (1982). Peptides 3,503-507. Mitsuma, T., and Nogimori, T. (1983a). Life Sci. 32, 241-248. Mitsuma. T., and Nogimori, T. (1983b). Acta Endocrinol. 104,437-442. Miyake, A., and Yen, S. S. C. (1981). Life Sci. 29, 2637-2640. Moldow, R. A., and Fischman, A. J. (1982). Peptides 3, 143-148. Molineaux, C. J., and Cox, B. M. (1982). Life Sci. 31, 1765-1768. Monnet, F., Reubi, J. C., Eberle, A., and Lichtensteiger, W. (1981). Neuroendocrinology 33, 284-287. Moore, K. E., and Demarest, K. T. (1982). In “Frontiers in Neuroendocrinology 7” (W. F. Ganong and L. Martini, eds.), pp. 161-190. Raven, New York. Moore, K. E., and Johnston, C. A. (1982). I n “Neuroendocrine Perspectives” (E. E. Muller and R. M. McLeod, eds.), Vol. 1, pp. 23-68. Elsevier, Amsterdam. Moretti, C., Fabbri, A., Gnessi, L., Cappa, M., Calzolar, A., Fraoli, F., Grossman, A., and Besser, G. M. (1983). Clin. Endocrinol. 18, 135-138. Morley,J. E., Baranetsky, N. G., Wingert, T. D., Carlson, M. E., Hershman, I. M., Melmed, S., Levin, S. R., Jamison, K. R., Weitzman, R., Chang, R. J., and Varner, A. A. (1980). J . Clin. Endocrinol. Metub. 50, 251-257. Morley, J., Levine, A. S., Yamada, T., Gebhard, R. L., Prigge, W. F., Shafer, R. B., Goetz, F. C., and Silvis, S. E. (1983a). Gastroenterology 84, 1517-1523. Morley, J. E., Levine, A. S., Yim, G. K., and Lowy, M. T. (1983b). Neurosci. Biobehau. Rev. 7, 281-305. Mormede, P. (1983). Nature (London) 302, 345-346. Muehlethaler, M., Gaehwiler, B. H., and Dreifuss, J. J. (1980). Bruin Res. 197, 264268. Mueller, G. P. (1980). Proc. Soc. Exp. Biol. Med. 165, 75-81. Mueller, G. P. (1981). Life Sci. 29, 1669-1674.

74

MARK J . MILLAN A N D ALBERT HERZ

Muneniura, M., Cote, ‘1‘. E., Tsuruta, K., Eskay, R. L., and Kebabian, J. W. (1980). Endocrinology 107, 1676-1683. Muraki, T., and Tokunaga, Y. (1978).Jpn.J. Pharmacol. 28, 803-806. Muraki, T., Nakadate, T., Tokunaga, Y., and Kato, R. (lY80).J. Endocrinol. 86, 357-362. Naber, D., Pickar, D., Davis, N. C., Cohen, R. M., Jimerson, J. C., Elchisak, M. A,, Defraites, E. G., Kelin, N. H., Risch, S. C., and Buchsbaum, M. S. (1981a).PsychophamaC O ~ O ~74, Y 125-128. Naber, D., Soble, M. G., and Pickar, D. (1981b). Pharmacopsychiatry 14, 160-161. Naber, D., Albus, M., Miiller, F., Burke, M., Welter, D., Ackenheil, M., and Hippius, H. (1982). Psychopharmacol. Bull. 18, 224-226. Naber, D., Munch, U., Wissman, J., Grosse, R., Ritt, R., and Welter, D. (1983). Acta Psychiatr. Neurol. Scand. 67, 265-27 1. Nakao, K., Nakai, Y., Ori, S., Horii, K., and Imura, H. (1978).J. Clin. Invest. 62, 13951401. Nappi, G., Fdcchinetti, F., Bono, G., Micielli, G., Parrini, D., Martignoni, E., Petraglia, F., and Genezzani, A. R. (1982a). Headache 22, 276-279. Nappi, G., Facchinetti, F., Legnante, G., Parrini, D., Petraglia, F., Savoldi, F., and Genazzani, A. R. (1982b). Acupunct. Electro-Ther. Res. 7 , 93-103. Negro-Vilar, A. (1982). Peptides 3, 305-310. Nichols, W. D., Accurso, F. J., Morris, H. G., and Grunstein, M. M. (1983). Clin. Res. 31, 142A. Nicoski, D. K., and Borer, K. T. (1983). Neurosci. Abstr. 9, 289. Oertel, W. H., Magnaini, E., Tappaz, M. L., Wiese, V. K., Dahl, A. L., Schmechel, D. E., and Kopin, I. J. (1982). Proc. Natl. Acad. Sci. U.S.A. 79,675-679. Ohlsson, A. E., Tsu-Ching, F., Jones, D., Martin, B. R., and Dewley, W. L. (1982). J . Pharmacol. Exp. Ther. 221, 362-367. Oliver, C., Usategui, R., Vaudry, H., Rozenberg, I., and Mourre, A. M. (1976).J. Endocrinol. 68, 351-352. Olson, R. D., Kastin, A. J., Olson, G. A,, and Coy, D. H. (1980). Psychoneuroendocrinology 5, 47-52. Olson, G. A,, Olson, R. D., Kastin, A. J., and Coy, D. M. (1982). Peptides 3, 1039-1072. Ono, T., Oomura, Y., Nishino, H., Sasaki, K., Muramoto, K., and Yano, I. (1980). Bruin Res. 185, 208-212. Orwoll, E. S., and Kendall, J. W. (1980). Endocrinology 107,438-441. Packman, P. M., and Rothchild, J. A. (1976). Endocrinology 99, 7-10. Page, R. B. (1982). Am. J . Physiol. 243, E427-E442. Palkovits, M. (1982). Peptides 3, 299-303. Palkovits, M., and Mezey, E. (1981). Front. Horn. Res. 8, 122-138. Panerai, A. E., Casaneuva, F., Martini, A., Mentagazza, P., and Di Giulo, A. M. (1981). Endocrinology 108, 2400-2402. Panerai, A. E., Martini, A,, De Rosa, A., Salerno, F., Di Guilo, A. M., and Mentegazza, P. (1982). In “Regulatory Peptides: From Molecular Biology to Function” (E. Costa and M. Trabucchi, eds.), pp. 139-149. Raven, New York. Panerai, A. E., Martini, A., Casanueva, F., Petraglia, F., Di Giulo, A. M., and Mantegazza, P. (1983). Life Sci. 32, 1751-1756. Pang, C. N., Zimmerman, E., and Sawyer, E. H. (1977). Endocrinology 101, 1726-1732. Parish, D. C., Srnyth, D. G., Normanton, J. K., and Wostencroft, J. E l . (1983). Nature (1
THE ENDOCRINOLOGY OF THE OPIOIDS

75

Pechura, C. M., Watkins, L. R., Povlishock, J. T., Becker, D. P., and Hayes, R. L. (1982). Neurosci. Abstr. 8, 849. Pelletier, G., Leclerc, R., Labrie. F., C M , J., Chretien, M., and Lis, M. (1977). Endocrinology 110, 770-776. Penalva, A,, Villanueva, L., Casanueva, F., Cavagnini, F., Gomez-Pan, A., and Muller, E. E. ( 1983). Psychofiharmacology 80, 120- 124. *Pert,A., Dionno, R., Ng, L., Bragin, E., Moody, T. W., and Pert, C. B. (1981). Brain lies. 224, 83-93. Petraglia, F., Penalua, A,, Locatelli, V., Cocchi, D., Panerai, A. E., Genazzani, A. R., and Muller, E. E. (1982). Endocrinology 111, 1224-1229. Petraglia, F., Facchinetti, F., Parrini, D., Micieli, G., De Luca, S., and Genazzani, A. R. (1983). Horm. Res. 17, 147-152. Petrucha, R. A,, Goebelsman, V., Tung, T. T.,Haase, H. R., and Lobo, K. A. (1983).Am.J. Obstet. Gynecol. 146, 646-65 1. Pettibone, D. J., and Mueller, G. P. (1981a). Brain Res. 221, 409-414. Pettibone, D. J., and Mueller, G. P. (1981b). Endocrinology 109, 798-802. Pfeiffer, A,, and Pfeiffer, D. G. (1983). Endocrinology 112A, 128. Pfeiffer, A., Feuerstein, G., Zerbe, R. L., Faden, A. I., and Kopin, I. J. (1983). Endocrinology 113,929-938. Pfeiffer, D. G., Pfeiffer, A., Shimohigashi, Y., Merriam, G. R., and Lorriaux, D. L. (1983). Peptides 4, 647-649. Pham-Huu-Trung, M. T., de Smitter, N., Bogio, A., Bertagna, X., and Giraud, F. (1982). Endocrinology 110, 1819-1821. Pickar, D., Cutler, N. R., Naber, D., Post, R. M., Pert, C. B., and Bunley, W. E. (1980). Lancet 1, 937. Pintar, J . E., Schachter, B. S., Herman, A. B., Durgerain, S., and Krieger, D. T . (1984). Science 225, 632-634. Pittman, Q., Hatton, B. D., and Bloom, F. E. (1980). Proc. Natl. Acud. Sci. U.S.A. 77,5527553 1. Plotsky, P. M., Bruhn, T. O., and Vale, W. (1984). Endocrinology 115, 1639-1641. Pontiroli, A. E., Ginseppina, B., Stella, L., Cresenti, A., and Girdrdi, A. M. (1982).J . Clin. Endocrinol. Metab. 55, 378-380. Popolov, H. B., King, J. C., and Gerall, A. A. (1980). Physiol. Behav. 27, 855-861. Porter, J. C., Gudelsky, G. A., Nansel, D. D., Pilotte, N. S., Burrows, G. H., and Tilders, F. J. H. (1981). Front. H o r n . Kes. 8, 139-148. Preziosi, P., Cerrito, F., and Vacca, M. (1981). Life Sci. 32, 2423-2430. Proulx-Ferland, L., LabriP, F., Dumont, D., Cbtk, J., Coy, D. H., and Sveiraf, J. (1982). Science 217, 62-63. Przewlocki, R., Hollt, V., and Herz, A. (1978). I n “Characteristics and Functions of Opioids” (T. van Ree and L. Terenius, eds.), pp. 285-286. Elsevier, Amsterdam. Przewlocki, R., Hollt, V., Duka, Th., Kleber, G., Haarmann, I., and Herz, A. (1979a).Brain Res. 174, 357-361. Przewlocki, R., Hollt, V., Voigt, K. K., and Herz, A. (1979b). Life Sci. 24, 1601-1608. Przewiocki, R., Millan, M. J., Gramsch, Ch., Millan, M. H., and Herz, A. (1982). Brain Res. 242, 107-117. Przewlocki, R., Lason, W., Konecka, A. M., Gramsch, Ch., Herz, A., and Reid, L. D. (1983). Science 219, 71-73. Pullan, P. T., Finch, P. M., Yuen, R. W. M., and Watson, F. E. (1983). Life Sci. 32, 17051710. Quigley, M. E., and Yen, S. S. C. (1980).J. Clin. Endocrinol. Metab. 51, 179-181. Rabinowe, S. L., Taylor, T., Dheky, R. G., and Williams, G. H. (1983). Clin. Res. 31,680A.

76

M A R K J . MILLAN AND ALBERT HERZ

Racagni, G., Apud, J. A., Locatelli, V., Cocchi, D., Nistoco, G., de Giorgio, R. M., and Miller, E. E. (1979). Nature (London) 281, 575-578. Racke, K., Ritzel, H., Trapp, B., and Muscholl, E. (1982a). Naunyn-Schmkdeberg's Arch. Pharmol. 319956-65. RackO, K., Rothlander, M., and Muscholl, E. (1982b). E u r . 1 . Pharmacol. 82, 97-100. Racz, K., Varga, I., Glaz, E., Kiss, R., Vida, S., Lada, G., di Gloria, K., Medzihradszky, K., Lichtwald, K., and Vecsei, P. (1982). J. Clin. Endocrinol. Metab. 54, 656-660. Ragavan, V. V., and Frantz, A. G. (1981). Endocrinology 109, 1769-1771. Ramsay, D. J., Reid, I. A., Keil, L. C., and Ganong, W. F. (1978).Endocrinology 103,54-59. Randle, J. C. R., Moor, B. C., and Kraicer, J. (1983a). Neuroendocrinology 37, 141-149. Randle, J. C. R., Moor, B. C., and Kraicer, J. (1983b). Neuroendocrinology 37, 131-140. Rapport, S. I., Klee, W. A., Pettigrew, K. D., and Ohno, K. (1980). Science 207, 84-86. Rathbun, R. C., Kattau, R. W., and Leander, J. D. (1983). Pharmucol. Biochem. Behav. 19, 863-866. Raymond, V., Lkpine, J., Gigukre, V., Lissitzky,J. C., C M , J., and LabriC, F. (1981). Mol. Cell. Endocrinol. 22,295-303. Reaves, T. A., and Hayward, J. N. (1980). J. Comp. Neurol. 193, 777-788. Recant, L., Voyles, N. R., Luciano, M.,and Pert, C. B. (1980). Peptides 1, 309-313. Rees, M., Bowen, J. C., Payne, J, G., and MacPhee, A. A. (1983). Surgery 93, 386-390. Reid, R. L., and Yen, S. S. C. (1982). Endocrinology 52, 592-594. Reid, R. L., Yen, S. S. C., Artrnen, H., and Fisher, D. A. (1981). Lancet 1, 1169-1170. Reid, R. L., Quigley, M. E., and Yen, S. S. C. (1983).J . Clin. Invest. 57, 1107-1110. Rewes, B., Lenz, N., and Brantl, V. (1983). Acta Endocrinol. Suppl. 253, 79-80. Reymond, M. J., Kaur, C., and Porter, J. C. (1983).'Brain Rex 262, 253-258. Richardson, 1. I., and Schonbrunn, A. (1981). Endocrinology 108,281-289. Kerscher, P., and Schwandt, P. (1983). Life Sci. 33, '143-746. Richter, W. 0.. Risch, S. C. (1982). Biol. Psychiatty 17, 1071-1079. Risch, S. C., Kalin, N. H., Cohen, R. M., Weker, J. L., Insel, T. R., Cohen, M. L., and Murphy, D. L. (1981). Peptides 2 (Suppl. I), 95-97. Risch, S. C., Janowsky, D. S., Siever, L. J., Judd, L. J., Rausch, J. L., Huey, L. Y., Beckman, K. A., Cohen, R. M., and Murphy, D. L. (1982). Peptides 3, 319-322. Risch, S. C., Janowsky, D. S., and Gillin, J. C. (1983). Peptides 4, 789-792. Riskind, R. N., Millard, W. J., and Martin, J. B. (1984). Endocrinology 114, 1232-1237. Riss, P. A., and Bieglmayer, C. (1983). Br. J . Obstet. Gynaecol. 90, 49-50. Rivier, C., and Vale, W. (1983). Nature (London) 305, 325-327. Rivier, C., Vale, W., Ling, N., Brown, M., and Guilleman, R. (1977). Endocrinology 100, 238-241. Rivier, C., Brownstein, M., Spiess, J., Rivier, J., and Vale, W. (1982a). Endocrinology 110, 272-278. Rivier, C., Rivier, C., and Vale, W. (1982b). Science 218, 377-378. Rivier, C., Rivier, J., Mormede, P.,and Vale, W. (1984). Endocrinology 115, 882-886. Roberts, J. L., Budarf, M. L., Baxter, J. D., and Herbert, E. (1979). Biochemitty 18,49074915. Rockhold, R. W., Crofton, J. T., Wang, B. C., and Share, L. (1983).J . Pharmacol. Ex#. Ther. 224, 386-390. Rockhold, R. W., Corfron, J. W., Brooks, D. P., and Share, L. (1984). Fed. Proc., Fed. Am. Soc. Exp. Biol. 43, 806. Roisin, M. P., Artold, A., Henry, J. P., and Roissier, J. (1983). Nturoscitnce 10, 83-88. Rolf, L. H., and Brune, G. G. (1981). Psychofihannacology 73, 48-50. Ronai, A. Z., Harsang, L. G., Berzetei, A. P., Bajusz, S., and Vizi, E. S. (1982). Eur. J . Phamacol. 79, 337-338.

THE ENDOCRINOLOGY OF THE OPIOIDS

77

Ropert, J. F., Quigley, M. E., and Yen, S. S. C. (1981).J. Clin. Endocrinol. Meta’b. 52, 583585. Rose, J. C., Goldsmith, P. C., Holland, F. J., Kaplan, S. L., and Ganong, W. F. (1976). Neuroendocrinology 22, 352-363. Rosella, L. M., Keil, L. C., and Summy-Long, J . Y. (1981).Fed. Proc., Fed. Am. SOC.Exp. B i d . 39, 255. Rosella-Dampman, D. M., Keel, L. C., Chee, O., and Sammy-Long,J. Y. (1983).J. Pharmacol Exp. Ther. 226, 373-381. Ross, M., Berger, P. A,, and Goldstein, A. (1979). Science 205, 1163-1 164. Rossier,J., Battenberg, E., Pittman, Q., Bayon, A., Koda, L., Miller, R., Guilleman, R., and Bloom, F. (1979a). Nature (London) 277, 651-653. Rossier, J., Rogers, J., Shibasaki, T., Guilleman, R., and Bloom, F. E. (1979b). Proc. Natl. Acad. Sci. U.S.A. 76, 2077-2080. Rossier, J., French, E., Rivier, C., Shibasaki, T., Guillemin, R., and Bloom, F. E. (1980). Proc. Natl. Acad. Sci. U.S.A. 77, 666-669. Rossier, J., Dean, D. M., Livett, B. G., and Udenfriend, S. (1981). Life Scz. 28, 781-789. Roth, K. A., Katz, R. J., Sibel, M., Mefford, I. N., Barchas, J. D., and Carroll, B. J. (1981). Llfe Sci. 28, 2389-2394. Roth, K. A,, Weber, E., and Barchas, J. D. (1982). L f e Sci. 31, 1857-1860. Roth, K. A., Weber, E., Barchas, J. D., Chang, D., and Chank, J. K. (1983). Science 219, 189- 191. Rotsztejn, W., Drouva, S. V., Patton, E., and Kordon, C. (1978a). Eur. J . Pharmacol. 50, 285-286. Rotsztejn, W., Drouva, S. V., Pattou, E., and Kordon, C. (1978b). Nature (London) 274, 281-283. Rotsztejn, W., Drouva, S. V., Pollard, K., Sokoloff, P., Pattou, E., and Kordon, C. (1982). Eur. J . Pharmacol. 80, 139-141. Rudman, D., Berry, C. J., Riedeburg, C. H., Hollins, B. M., Kutner, M. H., Lynn, M. J., and Chawla, R. K. (1983). Endocrinology 112, 1702-1709. Ryder, S. W., and Eng, J. (1981).J. Clin. Enducrinol. Metab. 52, 367-369. Saavedra, J. M., Palkovits, M., Kizer, J. S., Brownstein, M., and Zivin, J. A. (1975). J . Neurochem. 25, 257-260. Saavedra, J. M., Rougeot, C., and Drey, F. (1983). Neuroendocrinology 37, 164-165. Saavedra, J. M., Rougeot, C., Culman, J., Israel, A,, Niwa, M., Tonon, M. C., Vaudry, H., and Day, F. (1984). Neuroendocrinology 39, 93-95, Sachse, G., Laube, H., Svedberg, J., Breidenbach, Th., and Teschemacher, H. J. (1981). Diabetologia 21, 322. Saiani, L., and Guidotti, A. (1982).J. Neurochem. 39, 1669-1676. Sakakura, M., Yoshioka, M., Kobayashi, M., and Takebe, K. (1981). Neuroendocrinology 32, 174-178. Saland, L. C., Mennin, S. P., Selinfreund, R., and Rasmussen, P. (1982a). Regul. Pept. 3, 37 1-381. Saland, L. C., Ortiz, E., and Munger, A. T. (1982b). Cell Tissue Rex 225, 217-222. Saltiel, H., Passa, P. H., Kuhn, J. M., Fiet, J., and Canivet, J. (1982). Nouv. Presse Med. 11, 847-850. Sapun, A. I., Farah, J. M., and Mueller, G. P. (1981). Endocrinology 109,421-426. Sapun-Malcolm, D., Farah, J. M., and Mueller, G. P. (1983). Life Sci. 33, 95-102. Sawchenko, P. E., and Swanson, L. W. (1982). Brain Res. Rev, 4, 275-325. Sawchenko, P. E., Swanson, L. L., and Vale, W. W. (1984).J. Neuroscience 4, 1 1 18-1 129. Schachter, B. S., Johnston, L. K., Baxter, J. D., and Roberts, J. L. (1982). Endocrinology 110, 1 142- 1 144.

78

M A K K , J . MILLAN A N D ALBERT HEKZ

Schadt, J. C., Caddis, R. R., McKown, M. D., and Franklin, D. (1983). Fed. Proc., Fed. Am. SOC.Ex$. Biol. 42, 483. Schick, R., Holland, A., d e la Fuente, A., Spacht, J., Maier, V., and Brantl, V. (1983).Acta Endocrinol. Suppl. 253, 80-81. Schulte, H. M., Chrousos, G. P., Gold, P. W., Oldfield, E. H., Hoban, M. C., Cutler, G. B., and Loriaux, D. L. (1983). Acta Endocrinol. Suppl. 283, 32. Schultzberg, M., Lundberg, J. M., Hokfelt, T., Terenius, L., Brandt, J., Elde, R. P., and Goldstein, M. (1978). Neuroscience 3, 1169-1 186. Schulz, R.,Wiister, M., Duka, Th., and Herz, A. (1980a).Psychopharmacology 68,221-227. Schulz, R., Wiister, M., and Herz, A. (1980b). Eur.J. Phannacol. 63, 313-319. Schulz, R., Wilhelrn, A., Pirke, K. M., Gramsch, C., and Herz, A. (1981). Nature (London) 294, 757-759. Schulz, R.,Wilhelm, A., Pirke, K. M., and Herz, A. (1982a). Life Sci. 31, 2167-2170. Schulz, R., Wiister, M., and Herz, A. (1982b). Peptides 3, 973-976. Schusdziarra, V., Holland, A., Schick, R., de la Fuente, A., Klier, M., Brantl, V., and Pfeiffer, E. F. (1983a). Diabctologia 24, 113-1 16. Schusdziarra, V., Schick, R., d e la Fuente, A., Holland, A., and Brantl, V. (1983b). Endocrinology 112, 1948-1951. Schusdziarra, V., Schick, R., d e la Fuente, A., Specht, J., Klier, M., Brantl, V., and Pfeiffer, E. F. (1983~). Endocrinology 112, 885-889. Schusdziarra, V., Schick, R., Holland, A., d e la Fuente, A., Speck, J., Maier, V., Brantl, V., and Pfeiffer, E. F. (1983d). Peptides 4, 205-210. Schusdziarra, V., Specht, J., Schick, A., de la Fuente, A., Holland, A,, and Pfeiffer, E. F. (1983e). Horm. Metabol. Res. 15, 407-408. Schwandt, R. (1981). Front. Horn. Res. 8, 107-121. Schwandt, R., Richter, W., and Morley, J. S. (1981). Neuropeptides 1, 211-216. Schweigerer, L. (1983). Biochem. Biophys. Res. Commun. 113, 839-844. Schweigerer, L., Bhakdi, S., and Teschemacher, H. (1982). Nature (London) 296,572-574. Seizinger, B. R., and Hollt, V. (1980). Biochem. Biophys. Res. Commun. 96, 535-543. Seizinger, B. R., Hollt, V., and Herz, A. (1981). Biochem. Biophys. Res. Commun. 103, 256263. Seizinger, B. R., Bovermann, K., Maysinger, D., Hollt, V., and Herz, A. (1983).Phannacol. Biochem. Behav., 18, (Suppl. l ) , 367-369. Seizinger, B. R., Iliillt, V., and Herz, A. (1984). Endocrinology 115, 662-671. Serri, O., Rasio, E., and Somma, M. (1981).,]. Clin. Endocrinol. Metab. 53, 206-208. Shaar, C. J., Frederickson, R. C. A., Dininger, N. B., and Jackson, L. (1977). Life Scz. 21, 853-860. Shaha, C., Margioris, A,, Liotta, A. S., Krieger, D. T., and Bardin, C. W. (1984). Endocrinol00115,378-384. Shariker, G., and Sharma, R. K. (1979). Biochern. Lliophys. Xes. Commun. 86, 1-5. Shanks, M. F.,Clement-Jones, V., Linsell, C. J., Mullen, P. E., Kees, L. H., and Besser, G. M. (1981). Brain Res. 212, 403-409. Sharp, B., Morley, J. E., Carlson, H. E., Gordon, J., Briggs, J., Melmed S., and Hershrnan, J. M. (1981). Brain Res. 219, 335-344. Sharp, B., Kasson, B., Marshak, D., Ross, R., George, R., Sowett, J., and Yarnado, T. (1982a). Life Sci. 31, 98 1-985. Sharp, B., Ross, P., Levin, E., and Sowers, J. (1982b). Endocrinology 110, 1828-1830. Shavit, Y., Lewis, J. W., Terman, G. W., Gale, R. P., and Liebeskind, J. C. (1984). Science 223, 188-191. Sheppard, M. E., Kronheim, S., and Primstone, P. L. (1979).J.Neurochem. 32, 647-649. Shin, S. H. (1978).J. Endocrinol. 79, 397-398. Shiomi, H., and Akil, H. (1982). Life Sci. 31, 2185-2188.

THE ENDOCRINOLOGY OF THE OPIOIDS

79

Shu-Dong, T., Phillips. D. M., Halmi. N., Krieger, D. T., and Bardin, C. W. (1982). B i d . Reprod. 27, 755-764. Siegel, R. A,, Chowers, I., Conforti, N., Feldman, S., and Wedenteld, J. (1982).Brain Res. 249, 103-109. Simantov, R., and Snyder, S. (1977). Brain Res. 124, 178-184. Simantov, R., Zang, D., and Yelmeda, S. (1980). I n “Neural Peptides and Neuronal Communication” (E. Costa and M. Trabucchi, eds.), pp. 247-256. Raven, New York. Simpson, J. B. (1981). Neuroendocrinology 32, 248-256. Siranath-Singhi, D. J., Whitting, R. E., Audsley, A., and Fraser, H. M. (1983). Nature (London) 301, 62-64. Skowsky, R., Smith, P., and Swan, L. (1982). Clin. Res. 26, 108A. Sladek, C. D., Blair, M. L., and Ramsay, D. L. (1982). Endocrinology 111, 599-607. Slizgi, G. R., and Ludens, J. H. (1982).J. Pharmacol. Exp. Ther. 220, 585-591. Slizgi, G. R., Taylor, C .J . , and Ludens, J. H. ( 1984).5. Phannacol. Exp. Ther. 230,64 1-645. Smith, R., Grossman, A., Gaillard, R., Clement-Jones, V. J., Ratter, S., Mallinson, J., Lowry, P. J., Besser, G. M., and Rees, L. H. (1981). Clin. Endocrinol. 15, 291-300. Smock, T., and Fields, H. L. (1981). Brain Res. 212, 202-206. Sobel, D. 0. (1983). Neuroendocrinology 36, 249-253. Sofroniew, M. W., and Weindl, A. (1981). I n “Endogenous Peptides and Learning and Memory Processes” (J. L. Martinez, R. A. Jensen, K. B. Messing, H. Rigter, and J. L. McGaugh, eds.), pp. 327-369. Academic Press, New York. Sonders, M., Barchas, J., and Weber, E. (1984). Biochem. Bio$hys. Res. Comm. 122,892-898. Spampinato, S., and Goldstein, A. (1983). Neuropeptides 3, 193-212. Spampinato, S., Locatelli, V., Cocchi, D., Vicentini, L., Bajusz, S., Ferri, S., and Mhller, E. E. (1979). Endocnnology 105, 163-170. Spiegel, K., Kourides, 1. A., and Pasternak, G. W. (1982). Science 217,745-747. Spies, H. G., Quadri, S. H., Chappel, S. C., and Norman, R. L. (1980). Neuroendocrinology 30,249-256. Spiess, J . , Rivier, C., and Vale, W. (1983). Nature (London) 303, 532-535. Spiler, I. J., and Molitch, M. E. (1980).J. Clin. Endocrinol. Metab. 50, 516-520. Spinedi, E., and Negro-Vilar, A. (1983). Endocrinology 112, 1217-1223. Spinedi, E., and Negro-War, A. (1984). Life Sci. 34, 721-729. Stark, R. I., and Frantz, A. G. (1983). Clin. Perinatol. 10, 653-667. Stark, R. I., Wardlaw, S. L., Daniel, S. S., Husain, M. K., Sanocka, U. M., Stanley-James, L., and Van d e Wiele, R. L. (1982). A m . J . Obstet. Gynecol. 143, 204-215. Steger, R. W., Sonntag, W. E., Van Vugt, D. A., and Meites, J. (1981). Fed. Proc., Fed. Am. Soc. Exp. Biol. 39, 585. Steinbrook, R. A,, Carr, D. B., Datta, S., Naulty, J. S., Lee, C., and Fisher, J. (1982).Anesth. Analg. 61, 893-897. Steinbusch, H. W. M., and Nieuwenhuys, R. (1981). I n “Serotonin: Current Aspects of Neurochemistry and Function” (B. Haber, S. Gabay, M. Issidorides, and S. Alivasator, eds.), pp. 7-35. Plenum, New York. Stubbs, W. A., Delitala, G., and Jones, A. (1978). Lancet 2, 1225-1227. Suda, M., Hayaishi, O., and Nakagawa, H., eds. (1979). “Biological Rhythms and their Central Mechanism.” Elsevier, Amsterdam. Summy-Long, J. Y., Keil, L. C., Deen, K., Rosella, L. M., and Severs, W. B. (1981a).J . Pharmacol. Exp. Ther. 217, 619-629. Summy-Long, J. Y., Keil, L. C., Deen, K., and Severs, W. B. (1981b).J . Phannacol. Exp. Ther. 217, 630-637. Summy-Long, J. Y., Rosella, L. M., and Keil, L. C. (1981~). Brain Res. 221, 343-357. Summy-Long, J. Y., Keil, L. C., Sells, G., Kirby, A,, Chee, O., and Severs, W. B. (1983).Am. J . Physiol. 244, R522-R529.

80

MARK J . MILLAN A N D ALBERT HERZ

Summy-Long, J. Y., Miller, D., Rosella-Dampman, L., Hartnian, R., and Emmert, S. (1984). Brain Hes. 309, 362-366. Swanson, L. W., Sawchenko, P. E., Rivier, J., and Vale, W. (1983). Neuroendocrinology 36, 165- 186. Sylvester, P. W., Van Vugt, D. A., Aylsworth, C. A., Hanson, E. A., and Meites, J. (1982). Neuroendocrinology 34, 269-273. Szalay, K. S., and Stark, E. (1981). Lge Sci. 29, 1355-1361. Szczudlik, A., and Lypka, A. (1983). Acupunct. Electro-Ther. 8, 127-130. Szilagayi, J. E., and Taylor, A. A. (1983). Fed. Proc., Fed. Am. Soc. Exp. Biol. 42, 895. Taborsky, G. J., Halter, J. B., and Porte, D. (1981). Endocrinology 109, 319-321. Tache, Y., Charpenet, C., Chrktien, M., and Collu, R. (1979). In “Central Nervous System Effects of Hypothalamic Hormones and other Peptides” (R. Collu, A. Barbeau, 0. R. Ducharmeand, and G. G. Rochefort, eds.), pp. 301-303. Raven, New York. Takahishi, M., Motumatsu, T., Okajima, T., Oma, H., Kuto, K., and Ibayashi, M. (1981). In “Advances in Endogenous and Exogenous Opioids” (K. Takagi and E. J. Simon, eds.), pp. 241-243. Elsevier, Amsterdam. Tannenbaum, G. S., Panerai, A. E., and Friesen, H. G. (1979). Life Sci. 95, 1983-1990. Tapia-Arancibia, L., and Astuer, H. (1982). Neuroendocrinology 34, 166-168. Tapp, W. N., Mittler, J . C., and Natelson, B. H. (1981). Pharmacol. Biochem. Behau. 14,749751. Taraskevich, P. S., and Douglas, W. W. (1982). Nature (London) 299, 733-734. Taylor, P. K., Dawson, G., and Sedikario, S. (1983a). Life Sci. 32, 1935-1942. Taylor, T., Dlughy, R. G., and Williams, G. H. (1983b).J. Clin. Endocrinol. Metab. 57,592596. Tejwani, G. A,, Vaswani, K. K., Barbacci, J. C., Richard, C. W., and Bianchine, J. R. (1983). Life Sci. 33 (Suppl. I), 519-522. Terry, L. C., Growley, W. R., and Johnson, M. D. (1982).J. Clin. Invest. 69, 104-112. Thody, A. J., Lever d e Vries, L. H., and Tilders, F. J. H. (1980). Acta Endocrinol. 93, 300305. Thornton, V. F., and Geschwind, I. I. (1975). Gen. Corn$. Endocrinol. 26, 346-353. Tilders, F. J . H., and Smelik, P. G. (1975). Neuroendocrinology 17, 296-308. Tilders, F. J. H., Berkenbosch, F., and Smelik, P. G. (1980). In “Cathecholamines and Stress: Recent Advances” (E. Usdin, R. Kvetnasky, and I. J. Kopin, eds.), pp. 125-130. Elsevier, Amsterdam. Tilders, F. J. H., Post, M., Jackson, S., Lowry, P. J.. and Smelik, P. G. (1981). Acta Endocrinol. 97,343-351. Tilders, F. J. H., Schipper, J., Lowry, P. J., and Vermes, F. (1982). Regul. Pept. 5, 77-84. Tomiko, S. A., Taraskevich, P. S., and Douglas, W. W. (1981). Neuroscience 6,2259-2267. Tomiko, S. A., Taraskevich, P. S., and Douglas, W. W. (1983) Nature (London) 301, 706707. Trdmu, G., and Leonardelli, J. (1979). Brain Res. 168, 457-471. Tramu, G., Beauvillain, J. C., Croix, D., and Leonardelli, J. (1981). Brain Res. 215, 235255. Tramu, G., Croix, C., and Pillez, A. (1983). Neuroendocrinology 37, 467-469. Uberti, E. C. D., Trasforini, G., Salvadori, S., Margutti, A., Tomatis, R., Bianoconi, M., Mantoni, C., and Pansini, R. (1983).J. Clin. Endocrinol. 57, 1179-1185. Vale, W., Rivier, J., Guilleman, R., and Rivier, C. (1979). In “Central Nervous System Effects of Hypothalamic Hormone and other Peptides” (E. Collu, A. Barbeau, J. R. Ducharme, and S. G. Rochefort, eds.), pp. 163-176. Raven, New York. Vale, W., Spiess, J., Rivier, C., and Rivier, J. (1981). Science 213, 1394-1396. Valette, A., Tafani, M., Porthe, G., Pontonnier, G., and Cros, J. (1983).Llfe Sci. 33 (Suppl. I), 523-526.

THE ENDOCRINOLOGY OF THE OPIOIDS

81

Vandesande, F., Dierickx, K., and de May, I. (1977). Cell Tissue Res. 180, 443-452. Van Epp, D. E., Saland, L., Taylor, C., and Williams, R. C. (1983). Prog. Bruin Res. 59,361374. Van Leeuwen, F. W., and De Vries, G. J. (1983). Prog. Bruin Res. 60, 343-351. Van Leeuwen, F. W., Pool, C. W., and Sluiter, A. A. (1983). Neuroscience 8, 229-241. Van Loon, G. R., and Appel, N. M. (1981). Bruin Res. 204, 236-241. Van Loon, G. R., and De Souza, D. P. (1978). Life Sci. 23,971-978. Van Loon, G. R., De Souza, E. B., and Shin, S. H. (1980a). Neuroendocrinology 31,292-296. Van Loon, G. R., Ho, D., and Kim, C. (1980b). Endocrinology 1 0 6 , 7 6 4 0 . Van Loon, G. R., Appel, N. M., and Ho, D. (1981a). Bruin Res. 212, 207-214. Van Loon, G. R., Appel, N. M.. and Ho, D. ( 198 1 b). Endocrinology 109,46-52. Van Ree, J. M., and de Wied, D. (1981). Neuruphnrmucology 20, 1271-1277. Van Ree, J. M., Spaapen-Kok, W. B., and de Weid, D. (1976). Neuroendocrinology 22,318324. Van Vugt, D. A., Bruni, F., and Meites, J. (1977). Life Scz. 22, 85-90. Van Vugt, D. A,, Bruni, J. F., Sylvester, P. W., Chen, H. T., Ieiri, T., and Meites, J . (1979). Life Sci. 24, 2361-2368. Van Vugt, D. A,, Aylsworth, C. F., Sylvester, P. W., Leung, F. C., and Meites, J. (1981). Neuroendocrinology 13, 26 1-264. Van Vugt, D. A,, Sylvester, P. W., Aylsworth, C. F., and Meites, J. (1982). Neuroendocrinology 34,274-278. Van Vugt, D. A,, Bakst, G., Dyrenfurth, I., and Ferin, M. (1983). Endocrinology 113, 18581864. Van Wimersma Greidanus, T. B., Thody, T. J., Verspaget, H., de Rotte, G. A,, Goedemans, H. J. H., Croiset, G., and Van Ree, J. M. (1979). Life Sci. 24, 579-586. Van Wimersma Greidanus, T. B., Van Ree, J. M., Goedemans, H. J. H., Van nam, A. F., Andringa-Bakkar, E. A. D., and de Weid, D. (1981). Life Sci. 29,783-788. Vaudry, H., Pelletier, G., Guy, J., Leclerc, R., and Jegou, S. (1980). Endocrinology 106, 15 12- 1520. Veldhuis, J. D., Rogol, A. D., and Dufau, M. L. (1983). Clin. Res. 31, 278A. Vermes, I., Mulder, G. H., Smelik, R. G., and Tilders, F. J. H. (1980). Life Sci. 27, 17611768. Verrnes, I., Berkenbosch, F., Tilders, F. J. H., and Smelik, P. G. (1981). Neurosci. Lett. 27, 89-93. Verhoef, J,, Loeber, J. G., Burbach, J. P. H., Gispen, W. H., Witter, A., and de Weid, D. (1980). Life Sci. 26, 851-859. Vincent, S. R., Hokfelt, T., and Wu, J. Y. (1982). Neuroendocrinology 34, 117-125. Viveros, 0. M., Diliberto, E. J., Hazurn, E., and Chang, K. J. (1980). Adv. Bzochem. Psychuphal+ntacol. 18, 191-204. Vizi, E. S., and Volbekas, V. (1980a). Neuroendocrinology 31,46-52. Vizi, E . S., and Volbekas, V. (1980b). I n “Neuropeptides and Neural Transmission” (C. A. Marsan and W. 2. Traczyk, eds.), pp. 257-270. Raven, New York. Voigt, K. H., Weber, E., Breitig, D., and Martin, R. (1978). Actu Endocrinol. 215, 17-25. Voigt, K. K., Frank, D., Diiker, E., Martin, R., and Wuttke, W. (1983). Life Sci. 33 (Suppl. I), 507-510. Vonvoigtlander, P. F., Lahti, R. A., and Ludens, J. H. (1983).J. Phurmacol. Exp. Ther. 224, 7-12. Voulteenaho, O., Leppalouoto, J., and Mannisto, P. (1982).Actu Physzol. Scund. 115, 515516. Wade, C. E. (1983). Fed. Proc., Fed. Am. Soc. Exp. B i d . 42, 1 1 16. Wakabayashi, I., Demura, R., Miki, N., Ohmura, E., Miyoshi, H., and Shizume, K. (1980). J . Clin. Endocrinol. Metub. 50, 597-599.

82

MARK J . MILLAN A N D ALBEKT IIERZ

Wakerley, J. B., Noble, R., and Clarke, G. (1983). Neuroscience 10, 73-81. Walaas, I., and Fonnurn, F. (1978). Brain Res. 153, 549-562. Walker, J. M., Moises, H. C., Coy, D. H., Baldrighi, G., and Akil, H. (1982). Science 218, 1136- 1138. Walker, L. A., and Murphy, J . C. (1984).J. Pharmacol. Esp. T l w . 229, 404-408. Wallace, M., Fraser, C. D., Clements, J. A., and Funder, J. W. (1981). Endocrinology 108, 189- 192. Wamsley, J. K., Zarbin, M. A,, Young, W. S., and Kuhar, M. J. (1982). Neuroscience 7,595613. Wardlaw, S. L., Wehrenberg, W. B., Ferin, M., Carmel, P. W., and Frantz, A. G. (1980a). Endocrinology 106, 1323-1326. Wardlaw, S. L., Wehrenberg, W. B., Ferin, M., and Frantz, A. G. (1980b). Endocrinology 107, 1663-1666. Wardlaw, S. L., Stark, R. I., Daniel, S., and Frantz, A. G. (1981). Endocrinology 108, 17101715. Wardlaw, S. L., Thoron, L., and Frantz, A. G. (1982a). Brain Res. 245, 327-331. Wardlaw, S. L., Wehrenberg, W. B., Ferin, M., Antunes, J. L., and Frdntz, A. G. (1982b).J. Clin. Endocrinol. Metab. 55, 877-88 1. Watkins, W. B., Bruni, J. F., and Yen, S. S. C. (1980).J. Hktochem. Cytochem. 28, 11701174. Watson, S. J., Akil, H., Fischli, W., Goldstein, A., Zimmerman, E., Nilaver, G., and Greidanus, T. B. (1982a). Science 216, 85-87, Watson, S. J., Khachaturian, H., Coy, D. H., Taylor, L., and Akil, H. (1982b). L f e Sci. 31, 1773-1776. Watson, S. J., Khachaturian, H., Akii, H., Coy, D. H., and Goldstein, A. (1983a). Science 218, 1134-1 136. Watson, S. J., Khachaturian, H., Taylor, L., Fischli, W., Goldstein, A., and Akil, H. (1983b). Proc. Natl. Acud. Sci. U.S.A. 80, 891-894. Weber, E., and Barchas, J. D. (1983). Proc. Natl. Acad. Sci. U.S.A. 80, 1125-1129. Weber, E., Voigt, K. H., and Martin, R. (1978). Proc. Natl. Acad. Sci. U.S.A.75,6134-6138. Weber, E., Martin, R., and Voigt, K. H. (1979). Life Sci. 25, 1111-1118. Weber, E., Evans, C. J., and Barchas, J. D. (1981). Biochem. Biophys. Res. Commun. 103,982989. Weber, E., Evans, C. J., and Barchas, J. D. (1982a). Nature (London) 299, 77-79. Weber, E., Evans, C. J., and Barchas, J. D. (1982b).I n “Regulatory Peptides: From Molecular Biology to Function” (E. Costa and M. Trabucchi, eds.), pp. 519-526. Raven, New York. Weber, E., Evans, C. J., Chang, J. K., and Barchas, J. D. (1982~).Biochem. Biophys. Res. Commun. 108, 81-88. Weber, E., Evans, C. J., Chang, J . K., and Barchas, J. D. (1982d).J. Neurochem. 38, 436447. Weber, E., Roth, K. A., and Barchas, J. D. (1982e). Proc. Natl. Acad. Sci. U.S.A. 79, 30623066. Weber, E.. Roth, K. A., Evans, C. J., Chang, J. K., and Barchas, J . D. (1982f). LfeScz. 31, 1761- 1764. Weber, E., Geis, R., Voigt, K. H., and Barchas, J . D. (1983). Bruin Rex. 260, 166-171. Wehrenberg, W. B., McNicol, D., Wardlaw, S. L., Frantz, A. G., and Ferin, M. (1981). Endocrinology 109, 544-547. Wehrenberg, W. B., Wardlaw, S. L., Frantz, A. G., and Ferin, M. (1982). ErLdocrinology 111, 879-88 1.

THE ENDOCRINOLOGY OF THE OYIOIDS

83

Wei, E. T., Lee, A., and Chang, J. K. (1980). Life Sci. 26, 1517-1522. Weil-Fugazza, J., Godefroy, C., and Besson, J. M. (1979). Brain Res. 175, 291-301. Weindl, A,, Rust, M., and Graeff, H. (1983). L f e Sci. 33 (Suppl. l ) , 777-780. Weitzman, R. E., Fisher, D. A,, Minick, S., Ling, N., and Guillemin, R. (1977). Endocrinology 101, 1643-1646. Wesche, D., Hdlt, V., and Herz, A. (1977). Naunyn-Schmiedeberg'sArch. Pharmacol. 301,7982. Westlund, K. N., and Childs, G. V. (1982). Endocrinology 111, 1761-1763. Whitnel, M. W., Gainer, H., Cox, B. M., and Mohnedux, C. 1 . (1983). Science 222, 11371139. Wiesner, J. B., Koenig, J. I., Krulich, L., and Moss, R. L. (1984). L f e Sci. 34, 1462-1473. Wilkes, H. M., and Yen, S. S. C. (1980). Life Sci. 27, 1387-1391. Wilkes, H. M., and Yen, S. S. C. (1981). Life Sci. 28, 2355-2359. Wilkinson, W., and Bhanot, R. (1982). Endocrinology 110, 1046-1048, Wilkinson, K. W., Shinsako, J., and Dallman, M. F. (1983). Endocrinology 110, 1599-1606. Willette, R. N., and Sapru, H. N. (1982). Neuroscience 21, 1019-1026. Wong, T. M., Koo, A., and Li, C. H. (1981). Int. J. Pep. Protein Res. 18, 420-422. Wright, D. M., Clarke, G., and Pill, C. E. J. (1982). Regul. P e p . 4, 384. Wright, D. M., Pill, C. E. J., and Clarke, G. (1983). Life Sci. 33 (Suppl. I ) , 495-498. Wuster, M., Schulz, R., and Herz, A. (1980). Brain Res. 189, 403-411. Wybran, J., Appelboom, T., Famaey, J. P., and Govaerts, A. (1979).J. Immunol. 123,10681070. Wynn, P. C., Aguilera, G., Morell, J., and Catt, K. J. (1983). Biochen. Siophys. Rex Commun. 110, 602-608. Yanagida, H., and Corssen, G. (1981).Anaesthesiology 55, 515-519. Yanase, -1'., Nawatd, H., Higuchi, K., Kata, K.-I., and Ibayashi, H. (1984). L f e Sci. 35, 1869-1875. Yang, H.-Y. T., Hong, J. S., and Costa, E. (1977). Neuropliurmacology 16, 303-307. Yang, H.-Y. T., Hexum, T., and Costa, E. (1980). Lije Sci. 27, 1 119-1125. Yim, G. K. W., Davis, M., Lowy, M. T., Lamb, B., and Malven, P. (1981). Fed. Proc., Fed. Am. Sac. Exp. Biol. 40, 286. Zakarian, S., and Smyth, D. G. (1982). Bi0chem.J. 202, 561-565. Zamir, N., Simantov, R., and Segal, M. (1980). Brain. Res. 184, 299-310. Zampa, G. A., Benfenati, F., Ghilosi, E., Corbucci, G., Vecchi, P., Zini, I., Battastini, N., and Agnati, L. F. (1981).J. Endocrinol. Invest. 4, 423-429. Zanoboni, A., Zecca, L., Zanussi, C., and Zanoboni-Muciaccia, W. (1981). Neuroendocrinology 33, 140-143. Zein, M. A., Lutz-Bucher, B., and Koch, B. (1984). Neuroendocrinology 39, 392-396. Zerbe, R. L., Henry, D. P., and Robertson, G. L. (1982). Peptides 1, 199-201. Zimmerman, E. A., Stillman, N. A., Recht, L. D., Antunes, J. L., and Carrnel, P. W. (1977). Ann. N.Y. Acad. Sci. 297, 405-418.

This Page Intentionally Left Blank

MULTIPLE SYNAPTIC RECEPTORS FOR NEUROACTIVE AMINO ACID TRANSMITTERS-NEW VISTAS By Najam A. Sharif Department of Biochemistry Queen's Medical Centre Noeingham, England and Parke-Davis Research Unit Addenbrwker Hospital Cambridge, England

..................................

11. Neuroactive Amino Acids-Transmitter ldentification . . . . . . . . . . . . . . . . 111. Receptors in General . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

A. Definitions and Properties . , . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . , B. Criteria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..................... IV. Radioreceptor Assays (RRAs). . . . A. Usesof RRAs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B. Methodology of RRAs . . . . . . . . . . . . . , . . . . . . . . . . . . . . . . . . . . . . . . . .. .. ... . . C. Interpretations of Binding Measurements . . . , D. Multiplicity of Amino Acid Receptors.. . . . . . . . . , . . . . . . . . . . . . . . . . ....................... V. Inhibitory Amino Acid Receptors. . . A. GABA Receptor Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ........................... B. Glycine Receptors. . . . . .

.

A. Introduction

.............................

Amino Acid Receptors . . . C. Other Agonist Binding Studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . D. Antagonist Binding Studies . . . . . . . . . . . . . . . . . . . . . E. Pharmacology of Response Mechanisms. . . . . . . . . . . . . . . . . . . . . . . . . F. Regulation of Excitant Amino Acid Receptors.. . . . . . . . . . . . . . . . . . . G. Neuropathology of Acidic Amino Acids.. . . . . . . . . . . . . . . . . . . . . . . . VII. Summary and Concluding Remarks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . VIII. An Additional Note . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References .......................................

85 b6 89 89 90 92 92 93 95 96 96 96 105 108 108 112 123 124 126 129 135 136 139 140

1. Introduction

The main objective of this article is to review the recent advances in our limited understanding of receptor biochemistry and pharmacology of amino acid (AA) neurotransmitters. The author wishes to stress the 85 INTERNATIONAL REVIEW OF NEUROBIOLOGY, VOL. 26

Copyright Q 1985 by Academic Press, Inc. All rights of reproduction in any form reserved. ISBN 0-12-366826-3

86

NAJAM A . SHAKIF

vertebrate, in particular, mammalian, central nervous system (CNS). However, reference will be made to invertebrates, and limited results from electrophysiological, behavioral, and anatomical studies will be discussed for contrast and comparison. Receptor identification by in vitro ligand binding and measurements of biochemical response mechanisms will be considered in depth. The level of CNS activity is governed by the balance between neuronal excitation and inhibition. T h e known neuroactive AAs can be classified broadly either as excitatory or inhibitory. By far the major excitatory AA transmitters are glutamic (Glu) and aspartic (Asp) acids, while y-aminobutyric acid (GABA) and glycine (Gly) subserve neurotransmission at most inhibitory synapses. Although several other endogenous AAs (e.g., homocysteate, taurine) exhibit activity upon application to exposed neuronal populations (Watkins and Evans, 1981), rigorous evidence in favor of their transmitter function is largely lacking. However, it must be stressed that both Glu and Asp do not satisfy all the criteria normally ascribed to neurotransmitter candidates (Werman, 1966), but the case for these AAs and GABA and Gly is much stronger. Detailed discussion of this subject and appraisal of AA receptor properties relevant to AA physiology, pharmacology, pathology, and behavior have been previously documented (Curtis and Johnston, 1974; Watkins, 1978; Johnson, 1978; Usherwood, 1981; DeFeudis, 1979; Johnston, 1979; Enna and Gallager, 1983; Nistri and Constantini, 1979; Watkins and Evans, 1981; Roberts et al., 1981; Snodgrass, 1983; Sharif, 1984). Due to space considerations, the need to focus on most recent developments, and inadvertent bias, the author regrets not being able to quote all the chronologically relevant references. Therefore, readers are strongly urged to consult also other reviews and the original reports cited in their bibliographies as the source for more detailed and exact information.

II.

Neuroactive Amino Acids-Transmitter

Identification

An endogenous substance may be considered a neurotransmitter candidate and ascribed to a neuronal pathway providing a number of criteria are met and identified (Werman, 1966). In essence, the putative neurotransmitter, enzymes, and substrates for its synthesis must be present in presynaptic terminals from which the transmitter can be released upon stimulation in vivo and in vitro. Following evoked release and postsynaptic action at receptors, the transmitter must be removed from the synaptic cleft by a specific mechanism. Presynaptic stimulation

RECEPTORS FOR NEUKOACTIVE AMINO ACID TRANSMITTERS

87

must result in a response that can be mimicked by exogenous administration of the suspected transmitter and suppressed by its antagonists. Furthermore, it is desirable to be able to identify the functional correlates associated with the pharmacological manipulation of postsynaptically localized binding sites for that transmitter. However, the equation of these entities, whether monitored in vivo o r in vitro, with specific receptors for that transmitter requires satisfaction of additional prerequisites (Burt, 1978; Hollenberg and Cuatrecasas, 1979). Although evidence for excitant and depressant AAs as neurotransmitters will not be presentejl exhaustively, it is pertinent to note that Glu, Asp, GABA, and Gly exhibit a heterogeneous distribution within the mammalian CNS, being enriched, together with their metabolizing enzymes, in synaptosomal fractions (De Belleroche and Bradford, 1973; E. Roberts, 1979; Johnson, 1978; Fagg and Foster, 1983). The existence of'transmitter pool-linked metabolic enzymes for AAs has been verified (Fonnum et al., 1970; Bradford and Ward, 1976; Hamberger et al., 1979; Wenthold, 1980) and GABA and Glu visualized immunohistochemically (Storm-Mathisen et al., 1983). Ca2+-dependentrelease of endogenous and exogenous (radioactive) AAs from in vivo and in vitro preparations by depolarizing stimuli has been demonstrated following de novo synthesis and/or accumulation of tracer AAs by high-affinity uptake (Fagg and Lane, 1979). The latter processes have been identified as mechanisms for terminating synaptic AA transmitter action (Neal, 197 1; Iversen and Bloom, 1972; Bennett et al., 1972; Fagg and Lane, 1979). Iontophoresis of AAs on neurons results in depolarization- or hyperpolarization-induced excitation or inhibition consistent with observations of in vivo stimulation of nerve pathways (Curtis, 1963; Curtis and Watkins, 1963; Curtis and Johnston, 1974; Watkins and Evans, 1981). Similarly, blockade of AA- and neuronal-induced responses by known antagonists has been shown (Spencer, 1976; Watkins and Evans, 1981; Evans and Watkins, 198 1). More recently, multiple sodium-independent, postsynaptic binding sites for numerous radiolabeled AAs and their analogs have been identified in vitro on cell plasma membranes (Roberts, 1974; Michaelis et d.,1974; Roberts and Sharif, 1981; Enna and Gallager, 1983; DeFeudis, 1978; Snodgrass, 1983; Sharif, 1984) and visualized by autoradiography (ARG) (Foster et al., 1981b; Halpain et al., 1983). T h e latter technique has provided additional evidence in support of the neurotransmitter function of AAs in different parts of the CNS. Measurement and evaluation of putative AA receptor-mediated processes (e.g., 22Na+fluxes, cyclic nucleotide generation, and efflux of labeled transmitters) (Teichberg et al., 1981; Foster and Roberts, 1981; Scatton and Leyman, 1982; Enna and Gallager, 1983) have supported the presence of pharmacologically responsive AA receptors. The gen-

88

NAJAM A. SHAKIF

era1 distribution of AA receptors determined by ARG and radioligand binding correlates well with the distribution of AA neurotransmitter innervation in the CNS determined enzymatically following specific lesion experiments (McGeer et al., 1978; Coyle, 1983).These observations, together with other neurochemical and behavioral data, have implicated AA-mediated neurotransmission in the retina, cerebral cortex, hippocampus, cerebellum, striatum, spinal cord, etc. (Fagg and Foster, 1983). Their contribution to the control and coordination of such functions as vision, cognition, memory, posture, locomotion, and many vegetative functions cannot be overemphasized. The irpportance of AAs as major neurotransmitters is underlined further by the manifestation of such neuropathological disorders as Huntington’s disease (HD), epilepsy, dementia, Parkinson’s disease, spinal paralysis, etc. (Lloyd et al., 197’7; McGeer et al., 1978; Coyle, 1983a,b; Davidoff, 1983), where AA neurotransmitter function is thought to be grossly affected due to hypoactivity of hyperactivity of major excitatory and/or inhibitory neuronal pathways. Although, taken together, all the outlined evidence favors a transmitter function for GABA, Gly, Glu, and Asp, we must be aware of a few limitations which apply chiefly to the latter two AAs. For instance, it is known that both Glu and Asp (and other excitant AAs) are able to excite almost all neurons in the CNS, an observation which led to the early (incorrect) hypothesis of a single communal receptor for all excitatory AAs (Curtis et al., 1960, 1967, 1968, 1974). Secondly, the ubiquitous distribution of Glu and Asp and their well-known involvement in metabolic functions (Johnson, 1978; Hertz, 1979; Watkins and Evans, 1981; Cotman et al., 1981) impose constraints on the relevance and interpretations of lesion-induced changes (in glutamatergic and aspartergic systems), a technique that has been heavily employed to elucidate AAutilizing neuronal projections (Fagg and Foster, 1983). In the final analysis, however, these discrepancies d o not detract from, o r profoundly alter, the accumulating multidisciplinary evidence in favor of a neurotransmitter role for Glu and Asp (and GABA and Gly). However, a historical point to bear in mind is that unlike the early establishment of acetylcholine, and perhaps some biogenic amines, as important transmitter substances, serious and concerted studies to delineate similar roles for amino acids date back less than two decades. In time these AAs and related substances will undoubtedly gain recognition as the major and most important neurotransmitters in the CNS. Since most recent reviews on the current subject have neglected to include important fundamental background information, a brief account will now be presented.

RECEPTORS FOR NEUROACTIVE A M I N O ACID TRANSMITTERS

89

111. Receptors in General

A. DEFINITIONS AND PROPERTIES Historically, the receptor concept can be attributed to Langley (1905, 1907), Dale (1906), and Ehrlich (1909). The results of their classic pharmacological approach to studying the actions of biologically active substances invoked the existence of a “receptive substance” (Langley, 1907) with which drugs could react to evoke a given response. The term “receptor” originated from Clark (1926) in his attempt to explain the action of acetylcholine on the frog heart. He subsequently proposed that receptor-ligand interaction was reversible and that the induced response was proportional to receptor occupation/activation (Clark, 1936). Using current terminology, a receptor can be considered a macromolecular element of cells with which drugs, hormones, neurotransmitters, and antibodies interact to produce diverse biological responses. Thus, a true receptor comprises a ligand recognition site and one or more transducer/effector components. Cell-surface receptors are all proteins embedded in plasma membranes. Such glyco-, lipoproteins form integral components of membranes, and many are coupled to effector components (e.g., adenylate cyclase, ion channels) which transduce ligand-receptor interaction into biochemical o r physical signals which, in turn, cause the expression of the relevant biological response. Physiologically active receptors, under normal conditions, are present in minute quantities and exhibit a fast turnover rate, partly because of their proteinaceous nature. Related to their low numbers is a high affinity for their specific ligand(s). Moreover, the relative concentration of available endogenous ligand can cause hypotrophy or hypertrophy of receptors, leading to diminution or augmentation of the response mechanism. Sometimes, however, only a change in receptor affinity compensates in the absence o r presence of alterations in receptor number. These types of changes are often encountered in neurodegenerative disease states and during chronic drug therapy (Creese and Sibley, 1981). As noted earlier, classical study of receptors relied on bioassays (e.g., ileal contractions in organ bath). Since the 1960s, the advent and application of the iontophoretic technique (Krnjevik and Phillis, 1963; Curtis, 1964) has allowed the monitoring of extracellular and intracellular activities of innumerable neuroactive substances. With advances in biochemistry came techniques to quantify intracellular products of receptor activation, namely, secondary messengers (e.g., cyclic AMP and GMP

90

NAJAM A. SIIAKIF

and the resultant protein phosphorylation; 22Na+and 45Ca2+fluxes) (Sutherland and Rall, 1960; Greengard, 1979; Rodbell, 1980). However, all these former methods, although extremely important, are rather indirect means of studying recognition sites of synaptic receptors. More direct methods have relied on labeling receptors with radioactive ligands (Lin and Goodfriend, 1970; Yamamura et al., 1978; Snyder and Goodman, 1980). These radioreceptor binding techniques yield information about the properties and lead to possible localization (Young and Kuhar, 197913) of the ligand recognition site of the physiological receptors, but often the receptor-mediated response can be studied concurrently for receptor identification (Snyder, 1983; Bareis et al., 1983). Despite some limitations of receptor binding studies, biochemical technology coupled with this technique has a lot to contribute to the study, isolation, purification, and eventually the reconstitution of receptors (Snyder, 1983) with concomitant development of new drugs able to interact with these proteins. A current controversy in neuropharmacology concerns the use and abuse of the terms “receptors” and “binding sites” in descriptions of radiolabeling studies. From above, it is clear that the term “receptor” should be reserved for cases where a response can be related to ligand binding, preferably in the same experiment and tissue. However, since “spare” (Furchgott, 1964) and “silent” (Haga et al., 1977) receptors are known to exist, the strict adherence to such a definition may also be incorrect. In the view of this author, and perhaps many others (Kahn, 1976; Iversen et al., 1983), it may be possible to dissociate the functional concept from the radioligand binding sites and equate them with the classic pharmacological receptors, providing some important criteria are satisfied (see Section 111,B).Thus, “receptor binding” will be used in the broadest sense throughout this article. Needless to say, this semantic aphasia will attract at least a few opponents and the debate will rage on! B. CRITERIA Certain documented and verifiable characteristics of ligand binding are generally employed to assess the identification and measurement of receptors for informational molecules such as hormones, neurotransmitters, and drugs (Kahn, 1976; Burt, 1978; Hollenberg and Cuatrecasas, 1979). Binding of ligands to their specific recognition sites, under physiological conditions, occurs rapidly and reversibly, consistent with observations of the onset and cessation of nontrophic responses in vitro and in

RECEPTORS FOR NEUROACTIVE AMINO ACID TRANSMITTERS

91

vivo. I n reality, the time course of binding is generally dependent on, and is proportional to, the temperature and concentrations of ligands and receptors. Thus, time to equilibrium with association rates of lo41071Mlsecis shorter at high, nondenaturing temperatures, and steady states are achieved rapidly in a few minutes, Quenching of the equilibrated receptor-ligand mixture with unlabeled ligand (or dilution) results in dissociation of a reversible radioligand. Off rates in the range of 10-5- 1O-3/sec normally indicate reversibility of action consonant with cessation of response. A similar (minimal) criterion for receptor identification is saturability. Specific binding to receptor recognition sites is of a finite magnitude (saturable), while the nonspecific interactions of the radioligand increase and are proportional to the concentration of the latter. The high capacity (low affinity) of nonspecific binding has presently no defined correlates and is often problematical. However, the relatively low density of specific sites is compensated by their high affinity for their respective ligands. Receptors for informational molecules must possess high affinity (1/ dissociation constant; Kd = 1O-l0M) since under physiological conditions the circulating (central or peripheral) andlor neuronally released levels of their ligands are correspondingly low. The high affinity contributes to an acceleration of ligand-receptor interaction and in some cases partial receptor occupancy results in response mediation. Thus, the high affinity may be an explanation for the “spare” receptor concept (Furchgott, 1964). In general, the distribution of receptor binding for transmitters in various brain regions should parallel the density of innervation of that structure by nerve terminals which have the capacity to synthesize and release that substance. Similarly, the presence of a neuroactive substance in a CNS area does not always correlate with its receptor levels since that region may just be involved in synthesis and export of that compound (e.g., some peptide hormonesltransmitters) (Burt and Sharif, 1984). Similarly, more metabolically active CNS regions may exhibit higher content of AAs which will not necessarily be linked to a transmitter pool. Receptor binding should be mostly associated with the synaptosomal fraction of the target tissue if a synaptic action of the drug or transmitter is to be inferred. However, receptor denaturation can often result during subcellular fractionation, and therefore an erroneous conclusion may be drawn from binding results. Providing appropriate precautions are taken to retard receptor lability during preparations, this criterion can often be easily demonstrated. In another sense, the destruction of binding properties of membranes during prolonged handling, as during

92

NAJAM A. SHARIF

fractionation, o r following exposure to proteolytic enzymes, proteinmodifying reagents, extremes of temperature, pH, and/or ionic strength can help identify the receptor as a protein, which is also an important property to demonstrate. Receptor identification is not complete unless the pharmacological specificity of radioligand binding correlates well with the rank order of potency of competing agonists to elicit, and antagonists to block, one o r more series of quantitative biochemical and/or physiological responses. Although a reduced affinity for the nonbiological isomer of the receptor-ligand can often strengthen receptor identification (Goldstein et al., 197l), the significance of such stereoselectivity is lost, for example, in light of stereospecific opiate binding to filters (Snyder et al., 1975). Although an absolute correlation between the pharmacology of the response mechanism and binding is the best key to receptor identification, discrepancies due to dissimilar preparations and assay conditions and ligandlreceptor degradation, as for neuropeptides (Burt and Sharif, 1984),can result; for these explanations should be sought and the situation remedied. In conclusion, proteinaceous neurotransmitter receptors should be localized on cell membranes, and should exhibit saturability, high affinity, and a high degree of pharmacological specificity for their homologous ligands.

IV. Radioreceptor Assays (RRAs)

A. USESOF RRAs Although the first demonstration of receptor labeling involved muscarinic (Paton and Rang, 1965) and nicotinic cholinergic receptors (Changeux et ad., 1970), RRAs for AAs were instigated in the mid-1970s (Roberts, 1974; Michaelis et al., 1974; Zuking et al., 1974, 1975; Enna and Snyder, 1975). Today, the worldwide adoption of RRAs has been made possible by the commercial availability of radioligands of high specific activity, purity, and affinity, coupled with the relative simplicity and low running costs of the technique. Neurobiologists have applied RRAs to answer many varied questions. For example: (1) Do binding sites for putative transmitters exist in the CNS (and periphery)? If so, are they receptors with distinct pharmacological properties? (2) How do these neurotransmitter receptors relate to neuronal pathways, and do they adapt to presynaptic andlor postsynaptic alterations in vivo? (3)

RECEPTORS FOR NEUROACTIVE AMINO ACID TRANSMITTERS

93

Where are the receptors localized? Are they neuronallglial, presynaptic/ postsynaptic, junctional/extrajunctional? (4)Are there ontogenetic patterns of development of receptors, and what factors determine them? Do functional correlates also change with ontogeny? ( 5 ) How d o receptors respond during drug therapy and how are they regulatedkhanged in disease states? Other uses of binding assays include evaluation of the therapeutic value of new drugs and their possible side effects due to their interaction with more than one receptor system (Peroutka et al., 1977; Snyder and Yamamura, 1977; Snyder et al., 1974; Sharif and Burt, 1984). RRAs have provided evidence for and allowed quantification of biologically active substances and endogenous receptor ligands (Enna and Snyder, 1976; Sharif and Roberts, 1980; Toffano et al., 1978; Zackzek et al., 1983), thus affording development of new drugs. Aside from the versatility of RRAs as screening tools (Creese and Snyder, 1977), their application and importance in receptor visualization (Kuhar, 1983), solubilization, and reconstitution (Hollenberg and Cuatrecasas, 1979) in determination of receptor coupling to transducer mechanisms and in structure-activity studies (Snyder, 1983) has proven and will undoubtedly continue to prove fruitful. One future goal for AA receptor research will rely on all these techniques to develop monoclonal receptor antibodies for immunohistochemical localization of these moieties.

B. METHODOLOGY OF RRAs Since the synaptic properties of AA neurotransmitters differ vastly from other neuroactive agents, conditions for demonstration of receptor binding for AAs will be outlined briefly. Receptor labeling is usually best accomplished by employment of radiolabeled antagonists because of their greater affinity for receptors and reduced metabolic activity. I n the absence of potent and specific organic antagonists for most AA receptors, the detection and characterization of these entities has relied almost exclusively on agonist binding until very recently (Butcher el al., 1983). This has presented numerous problems. The putative AA neurotransmitters are substrates for high-affinity uptake systems (Logan and Snyder, 1972), and Gly, Asp, and Glu can be modified enzymatically. In addition, agonist binding is confounded by its rapid dissociation kinetics, which pose problems during assay termination. Further, the RRA may suffer in sensitivity and specificity due to the ubiquity of endogenous AAs which can compete with the radioprobe during the assay incubation (Sharif and Roberts, 1980).

94

NAJAM A. SHARIF

Effective measures to obviate these drawbacks have been necessitated in order to label AA receptors. Although in vitro preparations for study of the latter have included slide-mounted tissue sections (Kuhar, 1983), synaptic plasma membranes (Snyder and Bennett, 1976; Snyder and Goodman, 1980; Sharif, 1984), and synaptic junctional complexes (Foster et al., 198la), broken cell membranes of varying purity have been most widely utilized. To eliminate possible interference from endogenous competing ligands and to affect destruction of uptake sites, membranes have normally been subjected to vigorous washings. Thus, specific [3H]GABA binding is maximized by mild detergent treatment (Enna and Snyder, 1977) and by repeated freeze-thawing (Enna and Snyder, 1977; Napias et al., 1980). T o achieve similar goals for excitatory AA receptor binding, mild sonication and heat treatment (37"C/30 min) plus washings appeared optimal and necessary (Sharif and Roberts, 1980; Sharif, 1984). However, depletion of uptake components here by freezing was accompanied by drastically reduced receptor binding activity (Sharif and Roberts, 1980; Sharif, 1984). The difficulties in studying AA receptors in vitro by RRAs are obviously reduced by those procedures. T h e investigator is, however, plagued by a serious dilemma. On the one hand, removal of endogenous competing ligands from the tissue preparation by physical means is essential and warranted. On the other hand, will not these procedures perturb receptodmembrane structure and/or remove important cofactors, and therefore produce nonphysiological conditions? Similarly, receptor identification (Burt, 1978) requires enrichment of binding sites in fractions of synaptic elements, but lengthy subcellular fractionations coupled with detergent solubilization and extensive washings (Matus et al., 1981; Foster et al., 1981a) are surely deviations from the natural state too? Perhaps these questions are best left unanswered since biochemical RRAs are best performed in test tubes, and available evidence dictates use of RRAs in the search for new useful drugs, which is perhaps the ultimate goal of current research. Membrane preparation is followed by their dispersion in a sodiumfree medium (buffered to physiological pH) in order that Na+-dependent uptake processes (Logan and Snyder, 1972) should not be expressed during the binding assay. Attempts to further block these and catabolic enzyme sites with antagonists have not been successful without a concomitant reduction in receptor binding activity (Sharif and Roberts, unpublished observations). The rapid association and dissociation kinetics of AA receptor binding, due in part to their relatively low (micromolar) affinity (Sharif, 1984) compared to nanoniolar affinity of peptide receptors (Burt and Sharif, 1984), have dictated the microcentrifuge method of separation of free and receptor-bound radioactivity

KECEPTOKS FOR NEUROACTIVE AM I NO ACID TRANSMITTERS

95

(Roberts, 1974; Enna and Snyder, 1975). However, binding assays have also been successfully terminated by rapid filtration in conjunction with suitable precautions (Baudry and Lynch, 1981; Iwata et al., 1982). Rapid superficial rinsing of the pelleted and filtered membranes allows elimination of adsorbed label and thus reduces artifacts and apparent nonspecific binding. Thus, at a very simplistic level, the technique of radioreceptor labeling involves incubation of radioisotopic derivatives of the putative neurotransmitter or its agonist/antagonist analog with cellular fractions of suitable nervous tissue under optimized conditions. The “nonspecific” component, which often represents hydrophobic interactions and/or adsorption of the radioligand, can be resolved from the specific receptor binding by addition of a 1000-fold excess of the unlabeled ligand to a second batch of assay tubes (“blanks”), and the whole RRA taken to equilibrium and terminated. Subtraction of the blank values from the total binding (in absence of unlabeled ligand) yields the “specific” receptor binding. Further details regarding RRA methodologies can be found in several recent reviews (Yamamura et al., 1978; Marangos et al., 1983). C. INTERPRETATIONS OF BINDING MEASUREMENTS This apparent simplicity of RRAs is, however, fraught with inherent problems such as artifactual binding to nonprotein materials (Loh et al., 1974; Cuatrecasas and Hollenberg, 1975) and a host of interpretational difficulties (Yamamura et al., 1978). Secondly, it is not often possible to distinguish between binding occurring to presynaptic and/or postsynaptic (let alone glial) elements unless specific deafferentation or neuronal depletion studies are conducted concurrently. Moreover, binding assays cannot determine whether a ligand-site interaction involves agonist or antagonist conformation of the receptor or when a particular conformation is favored, unless labeled analogs of both types are employed. Receptor binding studies can yield information about the rates of association and dissociation of the radioligand from its recognition sites, and this can indicate the affinity parameters ( K d = Koff/Kon)and can help differentiate multiple binding components. The dissociation constant ( K d , Uaffinity) can also be obtained from manipulation of saturation andlor competition data by Scatchard analyses. A close agreement between the Kd values derived from kinetic and the latter method should be sought. An index of the apparent density of binding sites (BmaX) is also yielded by the Scatchard plots. Further manipulation of these data allows detection of dynamic aspects of ligand-protein interaction such as cooperativity and heterogeneity of binding sites by Hill and/or Scatchard

96

NA-JAM A. SHAKIF

plots. In addition, quantification of analog affinity of receptors determined by competition experiments allows calculation of inhibition constants (Ki, of competing compounds, and this knowledge is used to describe the pharmacological specificity and structural requirements of the receptor under study. Design and development of more specific and potent analogs can thus be initiated (Barlow, 1980; Creese and Snyder, 1977). In-depth appraisals of the quantitative aspects of RRAs have recently been published (Barlow, 1980; Weiland and Molinoff, 1981; Molinoff et al., 1981).

D. MULTIPLICITY OF AMINOACIDRECEPTORS The concept of neurotransmitter receptor heterogeneity is now well documented and is classically exemplified by nicotinic and muscarinic cholinergic receptors and a and /3 adrenergic receptors (Snyder and Goodman, 1980). Activation of subclasses of receptors by homologous compounds of a common drug category obviously confers many subtle variations in the mechanism of eliciting the net biological response(s). By analogy, then, receptors for inhibitory and excitatory AAs would also be expected to exhibit heterogeneity, whether in the recognition sites or in coupled proteins o r both. In any case, the existence of distinct subpopulations of any receptor is ultimately proven by isolation of these moieties, a situation that appears to be a distant goal for AA receptors on the whole.

V.

Inhibitory Amino Acid Receptors

Although available evidence favors a neurotransmitter role for GABA and glycine, other neutral AAs such as p-alanine and tauririe exhibit neuroactivity and may also fulfill such a function at inhibitory synapses. A. GABA RECEPTOR SYSTEMS 1, GABAergzc Pharmacology GABA is almost exclusively synthesized by decarboxylation of L - G ~ in the vertebrate CNS, and neuronal levels may exceed millimolar concentrations (Fonnum et al., 1970; Johnson, 1978), a factor that contrib-

RECEPTORS FOR NEUROACTIVE AMINO ACID TRANSMITTERS

97

uted to its early discovery (Awapara et al., 1950; Roberts and Frankel, 1950). Subsequently, Curtis and co-workers were able to demonstrate its inhibitory properties almost throughout the CNS (Curtis, 1963). Neuropharmacologic studies of the GABA system followed soon after with a view to determining its mechanism of action, defining neuronal elements mediating its responses, and resulting in design, development, and discovery of GABA mimetics and GABA antagonists. Interest has pivoted on combating such neurological diseases as epilepsy and other seizure states which may be consequential to GABAergic underactivity. The rationale of drug action here has been that activation of GABAergic transmission (uptake blockade and/or receptor stimulation) would effectively reduce neuronal firing and help alleviate the disorders (Meldrum, 1978, 1982). However, since evidence has accrued linking GABAergic hyperfunction (in nigrostriatal pathway) to Parkinsonism (Reisine et al., 1977), the search for region-directed GABA antagonists (Marsden, 1979) and specific glutamic acid decarboxylase (GAD) inhibitors (Wood and Peesker, 1975) also seems warranted. Early vertebrate electrophysiological studies revealed that GABA and its congeners activate receptor-channel complexes to gate chloride ions and cause cellular hyperpolarization with the resultant cessation of neuronal firing (Curtis and Watkins, 1965). Limited demonstration of the existence of these postsynaptic markers, viz. ligand recognition-site and ionophore, has been accomplished by in vitro biochemistry. Like most neuroactive AAs, GABA is a nonrigid molecule and can therefore assume numerous configurations. However, interaction with the receptor requires distinct ligand geometry. Systematic structureactivity studies have defined the optimal GABA state to be “partially folded” and planar for the vertebrate receptors (Johnston, 1978; Krogsgaard-Larsen and Falch, 1981), while mixed conformations are effective at mollusc neurons (Azanza and Walker, 1975), where both depolarizing and hyperpolarizing receptors prevail. GABA agonists include the straight-chain 3-aminopropanesulfinate (3-APS) and trans-4aminocrotonic acid, the cyclic kojic amine, muscimol (MUS) and its derivatives o r analogs, and isoguvacine (ISO) (Johnston, 1978). However, bicuculline (BIC)-like compounds appear to be selective recognition-site antagonists (Johnston, 1978) in the vertebrate CNS, while picrotoxin (PIC) appears specifically to block the activity of the coupled Cl- channel (Curtis and Johnston, 1974) (see Table I below and Fig. 1).

2. Postsynaptic GABA Receptor Binding While electrophysiological studies indicated a single GABA recognition site, in vitro radioligand-binding techniques have detected a sodium-

98

NAJAM A. SIlARIF

TABLE I GABA RECEPTOR CLASSIFICATION” Vertebrates Conipounds

GABAA

GABAB

Invertebrates

Agoriists Muscirnol 3-APS Baclofen

Potent Potent Inactive

Very weak Inactive Potent

Potent Very weak

Antagonists Bicuculline Picrotoxin Penicillin

Potent (competitive) Potent (not competitive) Weak

Inactive Inactive ?

Weak Potent (not competitive) Weak

Modulators Benzodiazepines Phenobarbitane Phenytoin

Potent (potentiators) Potent (potentiator) Very weak

? ? ?

Inactive Inactive Inactive (muscle)

a

?

This classification is adapted from Sinlmonds (1983). ?, Unclear or not known.

dependent site, corresponding to the presynaptic uptake mechanism, and a Na+-independent site representing the postsynaptic receptor (Enna and Maggi, 1979). In fact, initial attempts to label GABA receptors were confounded by the more numerous and avid uptake sites (Sano and Roberts, 1963). “Fortuitous” employment of Tris-containing buffer allowed manifestation of the receptor component (Elliott et al., 1965; DeFeudis, 1973). Physically disruptive paradigms such as freezethawing, extensive washings (Zuking et al., 1974; Enna and Snyder, 19’75),and membrane exposure to detergents (Enna and Snyder, 1977) were routinely utilized in subsequent studies to destroy uptake mechanisms and thus maximize (and augment) [3H]GABA receptor binding. While these procedures appear unnecessarily nonphysiological, the low ratio of receptorshptake sites renders receptor detection a major problem. Despite the relative ubiquity of GABA in the vertebrates, the brain regional distribution of its Na+-independent receptors is uneven (Zuking et al., 1974; Enna et al., 1975; Snodgrass, 1983). In the rat and cat CNS, GABA receptors are enriched in cerebellum > caudate > substantia nigra, while in the monkey and human brain the profile is generally cerebral cortex > substantia nigra = cerebellum (Snodgrass, 1983). However, while the pharmacological specificity appears uniform, there may also be species differences in the relative affinity of GABA receptors

99

RECEPTORS FOR NEUROACTIVE AMINO ACID TRANSMITTERS

0

HO

6 H

@

(2; 0

---.

Y

(8)

0 (9)

FIG. 1. Inhibitory amino acid agonists and antagonists. (1) Glycine (Gly);(2) p-alanine; (3) y-aminobutyrate (GABA); (4)y-aminovalerate; (5) rnuscimol (MUS); (6) thiomuscimol; (7) isoguvacine (ISO); (8) strychnine; (9) bicuculline (BIC) methiodide.

for this ligand, indicating exercise in caution in interspecies extrapolations. Crude synaptic membrane preparation from rat brain dispersed in Tris-citrate and assayed at 4°C for 5 min for [3H] GABA binding displayed dissociation constants (Kd) of 0.1-0.4 ,uM (Zuking et al., 1974; Enna and Snyder, 1975). This single population of GABA receptors exhibited high affinity for GABA (IC50= 0.4 p M ) and BIC (1Cs0 = 5 p M ) , and specific [“IGABA binding was also displaced by imidazoleacetic acid and 3-APS (IC50s= 0.25 /.LM)consistent with their neuronal depressant properties via CAB A receptors, while PIC, p-alanine, and Gly were inactive. This selective GABA site seems to fulfill the pharma-

100

NAJAM A . SIIARIF

cological requirements for a physiologically relevant ("classical") CNS GABA receptor. In addition, similar concentrations (14 p M ) of GABA and BIC are known to cause half-maximal stimulation and blockade, respectively, of rat superior cervical ganglion activity (Bowery and Brown, 1974), indicating a peripheral existence of a similar system. Following Triton X-100 or deoxycholate treatment, followed by excessive washings of membranes, multiple [3H]GABA binding components emerge. A population of high-affinity sites (& range: 4-28 nM, B,, range: 0.38-2 pmol/mg protein) and a second class of low-affinity ones (& range: 27-470 nM, B,,, range: 5-17 pmoVmg protein) have been demonstrated in rat brain (Enna and Snyder, 1977; Greenlee and Olsen, 1979), rat cerebellum (Wong and Horng, 1977), mouse brain (Chude, 1979), and cultured brain cells (Ticku et al., 1980). Purified postsynaptic densities exhibit predominantly the high-affinity component for [3H]muscimol (& = 4-50 nkf) (Matus et al., 1981). The possible mechanisms contributing to the elevation of binding affinity in these latter studies probably include destruction of uptake sites and removal of limiting lipids (Fiszer de Plazas and De Robertis, 1967; Johnston and Kennedy, 1978), endogenous Naf and GABA (Napias et al., 1980), and a thermostable inhibitor protein (GABA-modulin) (Toffano et al., 1978). The physiological relevance of these super highaffinity GABA binding sites remains unclear. However, they may represent relatively specific sites of action of the modulin peptide. Postsynaptic GABA receptors have also been studied employing [3H]muscimol ([3H]MUS) (Snodgrass, 1978; Williams and Risley, 1979). This alkaloid is a conformationally restricted analog of GABA which exhibits about 10 times the potency of the latter in iontophoretic tests and at competing for [3H]GABA binding (Krogsgaard-Larsen and Falch, 1981). 3H-labeled MUS bound to a single site of high-affinity sites (& = 3 nM), and unlabeled MUS, 3-APS, GABA (IC& range: 6-24 a), and BIC (IC50 = 1.8 p M ) were very potent displacers. T h e possibility that presynaptic GABA receptors may be labeled with [3H]MUS was partially supported by the demonstration of a MUS- and BIC-sensitive [?H]GABA release response (Snodgrass, 1978). White and Snodgrass ( 1983) have recently described [3H]isoguvacine ([3H]ISO) binding to rat brain membranes. This GABA agonist exhibited features similar to [3H]GABA binding, but [3H]IS0interaction was plagued by very high nonspecific binding. In concurrent experiments Kd values were 8,30, and 76 nM for [3H]MUS, [3H]GABA,and ["HIISO, respectively, while the density of [3H]MUS binding sites (0.96 pmoVmg protein) appeared to be twice those for other radioligands. The authors suggest that [3H]IS0 is probably not a useful probe for GABA receptors.

RECEPTORS FOR NEUROACTIVE AMINO ACID TRANSMITTERS

10 1

especially since the GABA uptake mechanism also has high affinity for this compound. However, further studies with this radioligand may be more rewarding. The use of another agonist to label receptors for GABA had its own problems. [P-(Chloropheny1)GABAl (L-baclofen) is an active neuronal depressant able to hyperpolarize frog motoneurons and binds to specific sites (in the presence of 2.5 mh4 Ca2+)on rat brain membranes ( K d = 132 nM) (Hill and Bowery, 1981). However, [3H]baclofen could not be readily displaced by most GABA analogs (IC50s= 12, 1 1 pM for MUS and APS), and both BIC and PIC were also inactive. The physiological significance of this BIC-insensitive site remains to be evaluated. However, since a similar site has been found on rat atrium and vas deferens (Bowery et al., 1981), on rat anococnogeous muscle (Muhyaddin et al., 1983), and in rat brain on noradrenergic terminals (Karbon et al., 1983), this GABAB site could be a peripheral-type GABA receptor akin to those described for certain benzodiazepine (BZD) (Sholnick and Paul, 1982) and neuropeptide receptors (Burt and Sharif, 1984). In the same context, these GABAB receptors, being BIC-insensitive, may have a strict requirement for interacting with GABA-like compounds in a “folded” conformation. Evidently, baclofen shows promise as an anticonvulsant and antispasticity drug (Snodgrass, 1983), and therefore its mechanism of action should be the subject of further research. The instability of labeled BIC blunted early efforts to use antagonist binding to study GABA receptors (Olsen et al., 1976). More recently, Mohler and Okada (1978) have demonstrated [3H]BIC methiodide interaction with rat brain cerebellar membranes consonant with labeling of GABA receptors. However, the binding parameters ( K d = 380 nM) and pharmacology (BIC, IC5,, 2 0.24 pM; GABA, 1 0 5 0 = 0.46 p M ) appear at variance with those of [3H]GABA binding. It is conceivable that BIC is able to bind selectively with an antagonist conformation of the GABA receptor, but more compelling evidence for this notion is not presently at hand, although differential enhancement of agonist binding by Triton, with no effects on antagonist binding (Snodgrass, 1983), may indicate such a phenomenon. Olsen and collaborators have been instrumental in studying the in vitro neuropharmacology of the GABA ionophore. a-[3H]Dihydropicrotoxinin (C~!-[~H]DHP) binds to rat brain preparations, but the blanks are rather high (Ticku and Olsen, 1978; Olsen et al., 1980), which makes the estimated K d ( 1-2 pM) questionable. Pharmacologically, GABA and MUS were weak inhibitors, while numerous barbiturates and BZDs were relatively potent competitors of 13H]DHP binding. A number of bicyclophosphate esters, noncompetitive GABA antagonists (Bowery et al.,

102

NA-JAM A. S H A K I F

1976), also exhibited marked potency against [3H]DHP binding (Olsen et al., 1980). T h e described properties of DHP binding are inconsistent with labeling of postsynaptic GABA receptors, but definitive evidence for its interaction with the C1- channel still needs to be presented. An adjunct to in vitro receptor binding is the development of autoradiographic (ARG) means of visualizing recognition sites for GABA using [?HIMUS (Palacios et al., 1980). The information provided by this technique has complemented the test-tube results in terms of receptor distribution in the CNS (Enna and Snyder, 1975). Furthermore, the differential localization of GABAA and GABAB receptors by tritium-film ARG has also been achieved (Wilken et at., 1981; Crossman et al., 1983), thus providing further evidence for the existence of GABA receptor subtypes. Use of ARC in mutant mice cerebellums has helped define the morphological localization of GABAB sites on granule cell dendrites (Bowery et al., 1983a,b).

3. Other GABA Receptors? The preceding discussion has centered around the biochemical properties of postsynaptic GABA receptors and coupled mechanisms. Physicochemical evidence indicates that presynaptic and nonneuronal GABA receptors also occur in the CNS. GABA-mediated inhibition of monosynaptic reflexes (by reduced Glu release) was reported in the early 1950s (Frank and Fuortes, 1957). Activation of presynaptic GABA receptors appeared to be responsible for this phenomenon. Bicuculline-insensitive,(-)baclofen-activated presynaptic GABA receptors on sympathetic nerve terminals have been demonstrated (Bowery et al., 1983a-c); and similar GABA receptors may be involved in ['Hldopamine release from brain slices [see Chesselet (1984) for review]. Release of [3H]GABA from CNS preparations is under negative feedback control through pharmacologically responsive GABA autoreceptors (Mitchell and Martin, 1978; Snodgrass, 1978; Arbilla et al., 1979; Starke, 1981).In a study of GABA receptor binding to different cellular preparations, glial cells bound the label to a minimal extent (DeFeudis et al., 1980), indicating a minor contribution in measurements of GABA receptor-binding density.

4. GABA Receptors, Disease States, and Behavior In general, the correlation between human disorders and anticipated alterations in GABAergic function has not been good. Exceptions to this include Huntington's and Parkinson's disease. In the latter state a marked deficiency of GABA receptors in substantia nigra and hippo-

RECEPrORS FOR NEUROACTIVE AMINO ACID TRANSMITTERS

103

campus was reported (Lloyd and Dreskler, 1978). The degeneration of the nigrostriatal dopaminergic pathway appears to be the key factor involved. The same authors showed that Huntington’s disease may be caused by massive reduction in human putamen and caudatal GABA receptors (Lloyd and Dreskler, 1978), with a concomitant increase in cerebellar sites. Kainate (KA)-induced striatal lesions (in rats) of‘ GABAergic interneurons produced symptoms resembling Huntington’s disease (Coyle and Schwarcz, 1976), and a marked supersensitivity of GABA receptors was seen which could be blocked by degenerating the nigrostriatal pathway. Consistent with the GABAergic neuron loss in Huntington’s disease (Lloyd and Dreskler, 1978) are observations of elevated GABA receptor binding to nigral membranes in humans (Enna et al., 1976; Cross and Waddington, 1981) and rats (Coyle and Schwarcz, 1976). Behavioral and electrophysiological (Waszezak et al., 198 1) supersensitivity to intranigral muscimol injections in similarly lesioned rats confirmed these biochemical observations. While GABAergic dysfunction in extrapyramidal systems has captured much attention in the past, the reports of reduced [3H]GABA binding in Alzheimer’s diseased frontal cortex and caudatal membranes (Reisine et al., 1977), increased [3H]GABA binding in ataxic patient’s cerebellum (Kish et al., 1983), and altered GABA-receptor affinity in patients with a history of epilepsy (Lloyd et al., 1980) deserve further study. With the disclosure of GABAergic influence on the release of hypothalamic/hypophyseal hormones, detection of [SHIMUSbinding sites on cerebral vessels and ovarian membranes, and the reports of analgesic effects of GABA mimetics (reviewed in DeFeudis, 1983), more emphasis should now be placed on GABA function in modulating behavior and peripheral systems.

5 . GABA-Benzodiazepine Interactions An interesting facet of GABAergic pharmacology involves the electrophysiological (Choi et al., 1977; MacDonald and Barker, 1978) and biochemical (Guidotti et al., 1978; Skerrit et al., 1982) observations of the facilitation of GABAergic function and enhancement of GABA receptor binding capacity by BZDs. The displacement of GABA-modulin (Guidotti et al., 1978) from GABA receptors by BZDs was advanced as an allosteric mechanism of BZD-induced increase in GABA binding. The early hypothesis of a close molecular association between GABA and BZD receptors based on these findings (cf. Skolnick and Paul, 1982) has repeatedly been supported by the reciprocal interactions between these drugs at the receptor level; thus most GABA agonists potentiate BZD

104

NAJAM A . SHARIF

binding affinity (Tallman et al., 1978; Karobath et al., 1979; Young and Kuhar, 1979b) in a BIC- and C1--sensitive manner. Differential stimulation of BZD binding by GABA during development (Mallorga et al., 1980) and in adult brain regions (Karobath and Sperk, 1979) indicates that only a distinct subtype of GABA receptors is coupled to BZD recognition sites and vice versa. T h e existence of GABAA and GABAB receptors (Hill and Bowery, 1981; Wilkin et al., 1981; Bowery el al., 1983a-c) and high-affinity (nanomolar Kd) (peripheral and central) (Skolnick and Paul, 1982) and iow-affinity (micromolar Kn) (Bowling and DeLorenzo, 1983) BZD receptors has complicated the distinction between these possibilities. T h e varied evidence for heterogeneity of receptors for GABA (Simmonds, 1983) and BZDs (Chiu and Rosenberg, 1983) has recently been reviewed. Reports of the marked anion dependency of GABA-stimulated and -unstimulated RZD binding (cf. Skolnick and Paul, 1982) and the inhibition of [3H]DHP binding by physiologically relevant levels of barbiturates (Olsen, 1981) followed the observations of enhancement of BZD binding affinity by the latter drugs (Leeb-Lundberg et al., 1980) which was BIC and PIC sensitive. Numerous pyrazolopyridine BZDs (e.g., etazolate) exhibit properties similar to barbiturates in their enhancement of BZD receptor affinity and GABA binding capacity (cf. Skolnick and Paul, 1982). Some limited correlation between the anesthetic properties of barbiturates (Olsen, 198l ) , anxiolytichonconvulsant actions of pyrazolpyridines, (Skolnick and Paul, 1982), and the latter effects is apparent. In the light of all these complex interactions between GABA receptors and psychotropic drugs, a multicomponent model involving the former and various regulatory sites has been proposed (Olsen, 1981; Skolnick and Paul, 1982). Although the complete isolation of this receptor-ionophore complex (GABA-BZD-barbiturate-C1channel) has not been accomplished yet, the solubilization and copurification of a MUS and BZD binding protein has been reported in partial support of this proposal. I n any event, this model helps explain that barbiturates and pyrazolopyridines may enhance BZD binding by interacting with the Clk ionophore, and that GABA mimetics bind to GABA recognition sites to increase availability of BZI) receptors; whereas BIC interacts with the GABA receptor, PIC would bind with the chloride channel to inhibit the GABA- and barbiturate-stimulated BZD binding. A more detailed appraisal of these concepts is dealt with elsewhere (Skolnick and Paul, 1982; Olsen, 1981). Although much is now known about the mammalian GABA and BZD receptors, early attempts to study similar entities in preparations of

RECEPTORS FOR NEUROACTIVE AMINO ACID TRANSMITTERS

105

invertebrates (cf. Usherwood, 1981) were unsuccessful; the first demonstration of C3H]FLU binding sites, which are stimulated by GABA mimetics, in fly thoracic muscle has, however, been accomplished (Abalis et al., 1983). Two binding sites were identified ( K ~ s= 24 and 994 nM),and the high affinity component displayed a drug specificity somewhat different from central and peripheral mammalian BZD receptors. The investigation of phylogenetic distribution of BZD receptors (Nielsen et al., 1978) should now be extended and reexamined.

6. GABA Receptor Regulation Aspects of this have been mentioned above without direct reference to regulation per se. Thus, availability of GABA receptors in vitro is modulated by detergent treatment and the ionic milieu (including chaotropic effects of ions), by endogenous peptide(s), by barbiturates, and by BZDs. Changes in GABA receptor population produced by chronic GABA-replacement therapy and lithium treatment (“down-regulation”) and by hormonal manipulations in young and old animals with a view to alterations in behavior have been reviewed (DeFeudis, 1983; Enna and Gallager, 1983). Early determinations of GABAergic stimulation of cyclic adenosine monophosphate (CAMP)(Ferendelli et al., 1980) have not been foIlowed up in detail since. In view of the invoivement of phosphorylation/dephosphorylation of phosphoproteins and the importance of secondary messengers in neurotransmission (Greengard, 1979), delineation of mechanisms of GABA-receptor coupling to guanine binding proteins (Rodbell, 1980) may afford further insights into regulation of GABAergic function.

B. GLYCINE RECEPTORS Biochemical Pharmacology

The simple zwitterionic AA glycine (Gly) exhibits several properties consistent with a putative neurotransmitter function. Unlike other AA transmitters, Gly levels predominate only in the caudal mammalian CNS, where they are known to mediate postsynaptic inhibition. Highaffinity uptake mechanisms able to sequester labeled Gly exist in the brainstem and spinal cord, while the cerebral cortex appears deficient in such transport sites. Similarly, the IS+-stimulated release of C3H]Gly has the same distribution as the glycinergic markers. The electrophysiological evidence for a spinal role of Gly and coupling of its receptors to a C1- ionophore is discussed elsewhere (Curtis and Johnston, 1974).

106

NAJAM A. SHARIF

However, suffice it to say that iontophoresed strychnine appeared to be a selective glycinergic antagonist, devoid of activity against GABAinduced inhibitions (Curtis et al., 1967) (Fig. 1). These early demonstrations of hyperpolarizing actions of Gly prompted a search for specific receptors for this AA by biochemical means. Here again, radioligand binding techniques have been most useful to date (for reviews, see Snyder and Bennett, 1976; DeFeudis, 1978; Snodgrass, 1983). Early studies of [3H]Glybinding (Valdes and Orrego, 1975) (to transport sites) were superceded by more elegant glycine-sensitive strychnine binding in order to demonstrate labeling of postsynaptic Gly “receptors.” Saturable and reversible binding of [3H]strychnine to a single class of sites on rat spinal cord membranes (& = 30 nM) exhibited halfmaximal inhibitions by unlabeled strychnine and Gly of 0.035 and 10 /AM, respectively, at 25°C (Young and Snyder, 1973). However, this drug selectivity, plus evidence of cooperative interactions between Gly and strychnine, and differential inactivation of Gly-inhibition of [3H]strychnine binding by sulfhydryl reagents (Young and Snyder, 1974a), indicated that separate binding sites existed for these glycinergic ligands and that [3H]strychnine was not a specific probe for Gly recognition sites. Concurrent data from the same laboratory (Young and Snyder, 1974b) on the parallel nature of noncompetitive, cooperative anion inhibition of [3H]strychnine binding and anion reversal of spinal hyperpolarizations, with C1- being the most potent, suggested that strychnine interacted with the presumed C1- ionophore coupled to Gly receptors. T h e alternate explanation invoked the possibility that strychnine was labeling an antagonist conformation of Gly receptors. However, the problem of labeling the postsynaptic Gly (agonist) binding sites remained until recently. [3H]Gly binding to synaptosome-enriched fractions of rat CNS tissues was of high affinity ( K d = 80-160 nM) and was specifically displaced by Gly (> strychnine > other inhibitory AAs) (DeFeudis et al., 1977). These results further support the notion of distinct Gly and strychnine binding sites. Although it is not known with certainty whether all Gly receptors are coupled to C1- channels, the studies of [3H]strychnine binding provide a very useful index of the density of Gly receptors. With this constraint in mind, the regional distribution of Gly receptors exhibits a rostra1 to caudal enrichment in rat (and monkey) CNS (thalamus < hypothalamus < midbrain < pons < spinal cord) as determined by [3H]strychnine binding to membranes (Young and Snyder, 1973) and by in nitro ARG (Zarbin et al., 1981). Consistent with this profile (including Gly level and CNS sensitivity data) is the recent demonstration of markedly reduced

RECEPTORS FOR NEUROACTIVE AMINO ACID TRANSMITTERS

107

[3H]strychnine binding sites in spinal membranes of spastic mice (White and Heller, 1982). These data appear important in light of recent proposals of hyperactivity of a specific glycinergic autoimmune disorder in the debilitating motoneuron disease (MND) (Bowery, 1983). However, a deficiency of thyrotropin-releasing hormone (TRH) in MND and the therapeutic benefits of TRH in this state (Engel et al., 1983) and other spinal disorders should not be ignored. Similarly, though BZDs displaced specific [3H]strychnine binding from spinal membranes (Young and Snyder, 1974c), the tentative link between Gly receptors and BZDs has weakened over the years (Snodgrass, 1983; Skolnick and Paul, 1982) and is especially eroded in view of the potent effects of certain BZDs on TRH receptor binding (Sharif and Burt, 1984). Novel biochemical studies involving solubilization (Pfeiffer and Betz, 1981) and affinity-chromatographic purification (Pfeiffer et al., 1982) of [3H]strychnine binding sites of rat spinal membranes have emerged recently. This solubilized protein of 48,000 molecular weight bound the label with high affinity (Kd = 11.3 nM),being analogous to properties of membrane strychnine sites (Young and Snyder, 1973). Further elegant experiments by these authors have demonstrated photoaffinity labeling of [3H]strychnine into rat cord membranes which was glycine sensitive, being reduced in a concentration-related manner in the presence of nonradioactive strychnine, and had an affinity constant of 9.7 nM (Graham et al., 1983). In addition, the sodium dodecyl sulfate/polyacrylamide gel electrophoresis of photoaffinity-labeled protein yielded a single polypeptide of M , 48,000, comparing well with previous solubilization results (Pfeiffer and Betz, 1981). With the appreciation of a putative retinal neurotransmitter function of Gly, apparent homogeneous [3H]strychnine binding sites have been detected in ovine retina (Borbe et al., 1981). In conclusion, though GABA has dominated the inhibitory AA field for many decades, the biochemical pharmacology of Gly-mediated events in the CNS is slowly being recognized and will ultimately gain prominence and importance on par with GABA. The recent demonstration of [3H]strychnine sites in human spinal cord (ca. Kd = 8 and 120 &) and their selective reduction (17-47%) in substantia nigra ( K d = 4 and 58 nM) of Parkinson’s diseased patients (Lloyd et al., 1983) should provide a further impetus to research on Gly function. Other inhibitory AAs include p-alanine and taurine, which are decarboxylate products of Asp and cysteate, respectively. Electrophysiological and neurochemical evidence in favor of their neurotransmitter function is presently limited to in uivo inhibition of cell firing (Curtis et aL., 1968) and enrichment of taurine in synaptosomal fractions of rat CNS tissues.

108

NA-JAM A . SHAKIP

However, the greatest impediment in the unequivocal demonstration of specific effects and sites of actions of both taurine and p-alanine, distinct from GABA and Gly, has been the lack of enthusiasm in tackling these tasks and in consequence the status of these AAs as putative transmitter agents occupies the void recently left by Glu and Asp.

VI. Excitatory Amino Acid Receptors

A. INTRODUCTION Much has been written about the possible role of excitatory AAs in mediating neurotransmission in the CNS, and extensive data support this notion (Johnson, 1978; Watkins, 1978; Watkins and Evans, 1981; Roberts et al., 198 1, for reviews). Evidence from diverse neurochemical and electrophysiological experiments employing various techniques and compounds has recently culminated in the elucidation of some key AAusing pathways (Fagg and Foster, 1983) and in the postulation of three types of excitatory AA receptors (Watkins and Evans, 1981). These receptors are specifically activated by quisqualate (QA),N-methyl-D-aspartate (NMDA) and KA, and relatively specific organic antagonists block the actions of these agonists. Historically, Hayashi ( 1954) first demonsuated the potent excitant properties of extracellularly applied L-G~uon central neurons. T h e inability of intracellularly injected L-GIu to evoke motoneuronal excitations (Coombs et al., 1955) indicated an extracellular localization of mechanisms responsible for Glu’s actions. Subsequent studies with other acidic AAs revealed a qualitatively similar depolarization response to L - A s ~L-, Glu, and L-cysteate (Curtis et al., 1960), leading the authors to hypothesize the existence of a common receptor for these AAs. Although some structural modifications of these AAs resulted in enhanced potencies (Curtis and Watkins, 1963), a degree of specificity was still absent, thus reinforcing the belief in a single receptor mechanism. T o a certain extent, since the permeability changes ensuing after various acidic AA applications were similar (Hosli et al., 1976), the same inference was evoked. The first hint of receptor multiplicity came from the suprapotent excitatory effects of heterocyclic AAs like KA, ibotenate (Ibo), QA, and domoate, which all possess the Glu skeleton (Johnston et al., 1968). Reports of differential excitation of feline spinal neurons by L - A s and ~ LGlu (Duggan, 1974) and by KA and NMDA (McCulloch et al., 1974), the

RECEPTORS FOR NEUROACTIVE AMINO ACID TRANSMITTERS

109

different potencies of Glu and Asp on thalamic cells (Hall et al., 1979), and the ability of Glu-desensitized neurons to respond to L - A s ~(Dostrovsky and Pomeranz, 1977) invoked the prevalence of subtypes of acidic AA receptors in the mammalian CNS. The ability of Glu and its linear analogs to adopt many conformations (Ham, 1974) and the above observations resulted in the postulate of “Glu-preferring” and “Asp-preferring” receptors (Buu et al., 1976; Johnston, 1979) capable of recognizing Glu in an extended state and in a folded state, respectively. Thus, while Glu could be assumed to bind to both receptor types, Asp and its congeners can interact only with the latter type. However, because of the greater efficacy of cyclic analogs of Glu (containing different rigid conformations of the latter) and the similarity of depolarizations of NMDA and KA to L-ASPand L-G~u(McCulloch et al., 1974), a better definition of these receptors was necessitated. Watkins (1978) proposed the “Kainate receptors” and “NMDA receptors” based on the distinct actions of these agonists. While the latter type could be envisaged to accommodate L - A s ~NMDA, , and folded L-G~u, the former could be anticipated to bind extended L-Glu and KA (see Fig. 2). A better resolution of these issues has been achieved through attempted antagonism of agonist-induced and synaptically evoked excitations of spinal neurons. The relatively nonspecific antagonists of AAinduced depolarizations, L-glutamic acid diethyl ester (GDEE) (Haldeman et al., 1972) and 3-amino- 1-hydroxy-2-pyrrolidone (HA-966) (Davies and Watkins, 1977),were surprisingly superceded by Mg2+.Low micromolar levels of MgCI:! potently depressed NMDA-induced responses of spinal neurons of frogs in vitro (Evans et al., 1977) and of cats in vivo (Davies and Watkins, 1977), but were devoid of activity against LGlu, KA, and QA. Some transition metal cations (Coy+,Ni‘+) could substitute for Mg2+.These observations were interpreted in terms of divalent cations acting at the ion channels believed to be coupled to the excitatory AA receptor(s) (Watkins and Evans, 1981). Detailed structure-activity studies of synthetic organic compounds on frog and mammalian spinal neurons revealed that certain analogs of Glu, specifically long-chain mono- and diamino dicarboxylic acids, have antagonistic properties. Again, blockade of NMDA- and dorsal rootinduced excitations were more effectively provided by these agents than against other agonists (Watkins, 1981). The profile of activity appeared to be D-a-aminoadipate (D-cx-AA)> D-a-aminosuberate (D-a-AS)= ~ - a diaminopimelate > D-a-diaminosuberate (D-a-DAS) > D-a-diaminoadipate (Watkins, 1978). Further modifications of these carboxylate antagonists to produce phosphonate analogs yielded compounds with

110

OH 0 (10)

(11)

FIG. 2. Excitatory amino acid agonists. ( I ) aspartate (Asp); (2) N-methyl-D-aspartate (NMDA); (3) glutamate (Glu); (4) 4-F-glutamate; (5) cysteate; (6) homacysteate; (7) cis-2.3piperidine dicarboxylate (PDA); (8) iboteriate (Ibo); (9) Kainate (KA); (10) quisqualate (QA); ( 1 1 ) dornoate.

enhanced potency. Thus, 2-amino-5-phosphonovalerate (APV) (D-CY-AA analog) and 2-amino-7-phosphonoheptanoate(APH) are the most potent NMDA antagonists known today (Evans and Watkins, 1981), having considerably less affinity for KA, QA, o r L-Glu receptors. Other relatively potent NMDA blockers include y-D-glutamylglycine (Y-D-GG)and cis-2,3-piperidine dicarboxylate (PDA), but these dipeptide and cyclic aspartergic analogs also depress QA and KA responses (Watkins, 1981; Davies et al., 1983). In the same context, GDEE and 2-amino-4-phos-

111

RECEPTORS FOR NEUROACTIVE AMINO ACID TRANSMITTERS

HO&OH H

O

U

o0 H

0

0

0 OH2

d

O

H

0

(5)

(8)

(9)

(10)

FIG.3. Excitatory amino acid antagonists. (1) a-aminoadipate (a-AA);( 2 ) a-aminosuberate (&-AS); (3) 2-amino-3-phosphonopropionate(APP); (4) 2-amino-4-phosphonobutyrate (APB); (5) 2-amino-5-phosphonovalerate(APV); (6) y-D-glutamylglycine (Y-D-GG); (7) P-D-aspartyl-P-alanine;(8) glutamate diethyl ester (GDEE); (9) L-methionine-DL-sdf(HA-966). oximine; (10) 3-amino-l-hydroxy-2-pyrrolidone

phonobutyrate (APB) appear to be relatively specific antagonists of QAtype Clu receptors (Evans and Watkins, 1981) (see Fig. 3). Based on the differential susceptibility of agonist-induced excitations of mammalian and frog spinal neurons to the inorganic and organic antagonists, at least three subtypes of excitatory AA receptors have been identified (Watkins and Evans, 1981; Watkins, 1981). The agonist and antagonist selectivity of these defined NMDA, quisqualate (L-Clu), and kainate receptors are shown in Table 11. Whether these receptor types exist in other parts of the CNS with similar pharmacological properties remains to be established by detailed, parallel studies. A point to note is that while electrophysiological evidence for these receptors is substantial, biochemical identification and characterization of these entities (to be discussed ahead) has lagged behind somewhat.

112

NAJAM A. SHARIF

TABLE I1 A

CLASSIFICArION OF

EXCITAIORY AA KECEPTORS~

Receptor types

Putative agonists

NMDA

Ibo NMDA AMPA u-Hioniocysteate L-Homocysteate NMLA D-As~ L-As~

2-APV D-CZ-AS D-Q-AA M#+ PDA

Quisqualate (L-Glu)

Quisqualate L-Cysteate

GDEE PDA DL-APB(?)

L-Glu

Putative antagonists

L-As~

L-Homocysteate 1bo Kainate

Donioate Kainate L-Clu Quisqualate NMDA

PDA Carboxyphenylgl ycine Y-D-GG Y-D-GAMS

a This table is primarily based on, and has been modified from the data and concepts outlined by Watkins and Evans (1981), Evans and Watkins (198I), and Koberts et al. (198I). For abbreviations see the niain text. Others are as follows: AMPA, a-amino3-hydroxy-5-methyl-4-isoxazolepropionate; NMLA, N-methyl-L-asparlate; Y-D-GAMS,yu-glutamylaniiriomethylsulfonate. Compounds are listed in decreasing potency.

Iontophoretic studies coupled with intracellular recordings have demonstrated that AA excitants cause cellular depolarizations by enhancing the membrane permeability to Na+ ions (Zieglgansberger and Puil, 1973; Hosli and Hosli, 1978). Because of this ionophore-receptor association it should be remembered that drug actions in viuo and in vitro may involve interactions with one or the other (or both) components of this complex and that allosteric effects are likely. B. BIOCHEMICAL CIIARACTERIZATION OF EXCITATORY AMINO ACID RECEPTORS

1. Glutamate Receptors The dicarboxylic AA glutamate is likely to be an excitatory transmitter of the hippocampal perforant and hippocampal septa1 pathways,

RECEPTORS FOR NEUROACTIVE AMINO ACID TRANSMITTERS

113

cerebellar granule cells, corticostriatal pathway, lateral olfactory tract, primary sensory afferents (and spinal interneurons), and auditory and retinal pathways (see Watkins and Evans, 1981; Roberts et al., 1981; Fagg and Foster, 1983, for reviews). As noted before, extracellular Glu must produce its effects by interacting with proteins located on neuronal plasma membranes (Coombs et al., 1955). The study of these recognition proteins has attracted much attention, with many attempts to demonstrate their existence and define their biochemical binding properties. In the absence of potent and specific pharmacological antagonists for excitatory AAs, the early studies relied exclusively on agonist radiolabels. The first reports of Glu binding described interaction of L-[ 14C]Glu of low specific activity with brain membrane fragments. While Roberts (1974) found a single class of [14C]Glu-bindingsites (& = 8 p M ) on rat cortical membranes, Michaelis et al., (1974) detected two components in whole brain membranes suspended in a Na+-free medium (Kdl = 0.2 /& K dI 2 , 4 p M ) . In contrast, triphasic binding isotherms were exhibited by cortically derived proteolipids (Kds = 0.3-55 p M ) (DeRobertis and Fiszer d e Plazas, 1976). These descriptions of Na+-independent Glu recognition sites, although limited in many respects, helped set the scene for more refined and detailed studies. The presence of high-affinity (& = 700 nM), stereoselective, Na+independent (postsynaptic) binding sites for L - [ ~ H ] G on ~ u fresh cerebellar synaptic membranes was first described by Foster and Roberts (1978) using a centrifugation assay. Following optimization of membrane preparation, which involved mild sonication, incubation (37"C/30 min), and extensive washings, a higher density of [3H]Glu receptor binding sites of high affinity (& = 360 nM, B,,,,, = 1 1 7 pmollmg protein) was demonstrated on rat cerebellar membranes (Sharif and Roberts, 1980, 198la,b, 1984). Typical characteristics of these sites included optimal binding affinity at pH 7.1 and 37"C, attaining equilibrium within 10 min in buffered-Tris medium. Subcellular fractionation of rat cerebellum revealed much enrichment of Na+-independent ["H]Glu binding activity in the synaptosomal preparation (Foster and Roberts, 1978; Sharif and Roberts, 1984). Under these defined, optimal conditions the homogeneous population of noninteracting binding sites exhibited an uneven distribution in the rat CNS but was consistent with known density of glutamatergic innervation of these regions. On the basis of the differential pharmacological specificity of the [3H]Glu binding to the cerebellar membranes (Table 111), which closely resembles the profile of compound actions on CNS neurons (Watkins and Evans, 1981), these recognition sites can be identified as glutamate receptors. Note the marked potency of sulfur-containing AAs homocysteate, cysteate, and cysteine

114

NAJAM A. SI-IARIF

TABLE I11 PHARMACOLOGY OF Glu A N D Asp RECEPTOR BINDINC" Inhibition of Competing analog (drugs) m-Homocysteate 1.-Cysteate L-Cysteine sulfinate cis-Cyclopentyl Glu L-mi

Ibo 4-F-Glu y-Methyl Glu *DL-2-APBa L-As~ *DL-~-AP Hept. D-Glu * D L - ~ - AHex. P "DL-P-APV *D-2-APV *L-~-APV *HA-Y66 "DL-a- Aminoadipate *oL-a-Aminosu berate *uL-a-Diaminopimelate o~-Threo-3-hydroxyAsp D-As~

[3H] Glu binding [ K , (+)I

Inhibition of ['HI Asp binding [K, ( 4 0 1

0.3 1 0.40 3.12 4.60 6.24 6.32 7.04 8.0 10.0 20.0 52.0 57.6 ND 80.0 ND ND 100.0 ND 80.6 80.0 80.0 100.00

100.0 N Db 82.8 ND 1.32 6.60 ND ND >100.0 3.30 49.6 ND 19.86 23.34 25.16 39.73 6.95 238.41 6.62 ND 100.0 302.64

a Compounds with K,s > 1 mM against radioligand binding included: NMDA, kainate, giycine, baclofen, diazepam, m-C-allylglycine, Dr.-Asp-p-liydroxamate, L-methionine sulfoxamine, and (RS)-c~-amino-3-hydroxy-5-methyl-4-isoxazolep1-opionate. Data are modified from Sharif and Roberts (1984) and refer to binding of radioligands to sonicated, preincubated, and washed rat cerebellar membranes. DL-2-AP Hex., uL-2-amino~~-2-amino-7-phosphonoheptanwate. For other 6-phosphonohexanoate; D L - ~ - AHept., P abbreviations see the main text. *, Putative antagonists. ND, Not determined.

sulfinate and that of cyclic analogs of Glu, e.g., cis-cyclopentyl Glu and Ibo. Of the phosphonate esters, DL-APBand DL-APHwere more potent competitors than DL-APVand other NMDA antagonists against [3H]Glu binding. Characterization of similar receptors in the rat hippocampus has also been accomplished. Two distinct C3H]Glu binding sites were apparent in the hippocampus. T h e Na+-independent component, representing the postsynaptic site, possessed a slower rate of association and dissociation than the Na+-dependent sites (uptake mechanism) (Baudry and Lynch,

KECEPTORS FOR NEUROACTIVE AMINO ACID TRANSMITTERS

115

1981). These sites were differentiated further on pharmacological grounds; thus potent inhibitors of [“H]Glu binding (e.g., QA, Ibo, and DL-homocysteate) were almost devoid of activity against [3H]Glu uptake and vice versa. Similarly, the Na+-independent binding sites ( K d = 0.75 pM, B,,, = 6.5 pmol/mg protein) had a greater affinity but a lower B,,, = 75 pmol/mg protein), density than the uptake sites (Kd = 2.4 and a differential brain regional distribution was also evident. Most recently, evidence for two Na+-independent [3H]Glu binding sites on hippocampal membranes has been presented (Werling and Nadler, 1982). The corresponding half-maximal equilibrium binding occurred at 0.0 11 and 0.57 pM, and Glu receptor binding appeared to be potentiated by nanomolar levels of excitant AA agonists and antagonists (Werling and Nadler, 1982). Further, computer analysis of inhibition curves revealed that [3H]Glu binding to the low-affinity (Kd = 0.57 f l )component was selectively reduced by QA, D- and L-homocysteate and L-wAA, while Land D-GIu and Asp exhibited a preference for the 0.011 & ./ affinity sites. Using QA- and Ibo-defined nonspecific binding, Werling et al. (1983) have differentiated these two [3H]Glu binding sites. T h e 0.02 ,uM K d site (GluA) equilibrated more quickly and denatured during freezing as compared to the G ~ Usite B (1 pM Kd). Furthermore, lesion studies have shown that the QA-sensitive (GluB) sites are presynaptically located in perforant path fibers, while the GluA sites predominate on dentate granule cells (Werling and Nadler, 1983). Although the low-affinity hippocampal site resembles a pharmacologically and kinetically defined postsynaptic Glu receptor, the high-affinity recognition site has presently no apparent physiological relevance. However, since neuropeptide receptors generally possess nanomolar affinity constants (Burt and Sharif, 1984), this binding component may be a recognition site for a small endogenous peptide capable of interacting with receptors for excitant AAs. Indeed, a dipeptide [N-acetylaspartylglutamate (NAAG)] isolated from rat brain has recently been shown to have convulsant effects upon intrahippocampal injection and inhibits [3H]Glu binding relatively potently but partially (Zaczek et d., 1983). Previous reports of such very high-affinity [3H]Glu binding sites include one in striatum of I 1 nM Kd (Biziere et al., 1980) and two similar components in rat cortical membranes (Fig. 4) when 0.05-25 nM concentrations of the radioligand were employed in saturation experiments. In addition to the extensive radioreceptor studies conducted in rat CNS tissues, apparent high-affinity (& values of 0.01 and 0.8 p M ) Na+independent [3H]Glu binding sites responsive to antagonism by APB, a purported depressant of excitant AA-induced depolarizations (Watkins and Evans, 1981), have been detected in bovine retina (Mitchell and

m,

116

NAJAM A. SHARIF

Redburn, 1982). The pharmacological specificity of the 0.8 pM K d site here matched well with the electrophysiologically characterized putative Glu receptors located on bipolar cells. Specific [3H]Glu binding to feline CNS tissues displayed a heterogeneous distribution (amygdala > hippocampus > hypothalamus = visual cortex % spinal cord) (Head et al., 1980). Cat cerebellar membranes possessed low-affinity ( K d = 1.8 pM) and high-affinity (Kd = 0.33 pM) [3H]Glu binding sites that displayed marked stereoselectivity and to which L-Glu, DL-homocysteate, and Lcysteate bound avidly. In order to correlate the C3H]Glu binding activity measured in crude synaptosomal membranes with specific interactions with discrete synaptic proteins, Cotman et al. (1981) have studied radiolabeling of detergent-solubilized and gradient-purified synaptic junctional complexes (SJCs). Substantial enrichment of Na+-independent [3H]Glu (and [5H]Asp)binding (& = 0.45 pM, B,,, = 91 pmoVmg protein) was demonstrated in SJCs by double-label saturation techniques (Foster et al., 1981 a,b). T h e prevalence of this homogeneous class of [3H]Glubinding sites in forebrain SJCs resembles those found on synaptic membranes (Foster and Roberts, 1978; Sharif and Roberts, 1980, 1981a,b, 1984; Werling and Nadler, 1982; Table IV below) in terms of equilibrium data and thus represents compelling evidence that the Glu receptor binding studied in both preparations may be associated with some synaptic function, presumably mediation of the postsynaptic actions of Glu. In the same context an acidic, hydrophobic glycolipoprotein has been isolated from rat brain by Triton extraction and subsequent concanavalin A affinity chromatography (Michaelis et al., 1981, 1983). Using Millipore filtration, these investigators have found this M , 14,300 protein to bind [3H]Glu with an affinity constant of 0.13 p M at low protein concentrations, while at higher levels two components ( K d l = 0.17 pM, K d 2 = 0.8 pM) emerged, indicating cooperative interactions, probably due to protein aggregation. Although preliminary pharmacological data suggested that a postsynaptic Glu receptor protein was being studied here, more detailed analyses employing some new putative antagonists may help substantiate these findings. Similarly, attempts to reconstitute these [3H]Glu binding proteins into liposomes in order to bestow some functional properties (Chang and Michaelis, 1981, 1982) may help clarify the issues discussed above. A preliminary report of the development of monoclonal antibodies to this purified Glu binding protein (Roy and Michaelis, 1983) represents another exciting advance in this sphere of research. Presently, we can conclude that though numerous reports describing binding characteristics of radioactive Glu to in vitro mammalian CNS

RECEPTORS FOR NEUROACTIVE AMINO ACID TRANSMITTERS

117

preparations have been published since the mid- 1970s, the relative progress has been slight in terms of' defining receptor identification. T h e major constraints contributing to this deficit have obviously included the nonavailability of suitable pharmacological antagonists [a situation partially remedied recently (Evans and Watkins, 198l)l and the adoption of dissimilar preparative (fresh vs frozen; homogenates vs synaptic membranes vs SJCs) and assay procedures (centrifugation vs filtration). Thus, the resultant interpretative problems have been enormous. As discussed previously (Sharif and Roberts, 1980, 1984; Roberts and Sharif, 1981), future studies of receptor binding of excitatory AAs would benefit immensely from application of common, optimal tissue preparations and assay methodologies. Table IV shows a few selected examples of Glu receptor binding studies for the reader's reference. It is apparent from this list that Na+-independent Glu binding sites described to date have dissociation constants in the low-high micromolar range. This low affinity profile is, however, consistent with the concentration range of L - G ~ likely to be encountered by synaptic receptors asjudged by the threshold of excitations (in vitro and in vivo) induced by this AA and its agonist analogs (Watkins and Evans, 1981). Localization of Na+-independent [3H]Glu binding sites by autoradiography on cryostat-cut rat brain sections has indicated a high density of receptors in hippocampus > striatum > cortex > cerebellum (Greenamyre et al., 1983; Halpain et al., 1983). The relatively high-affinity binding sites (& = 0.5-1.0 ,& measured I) in these studies exhibited a drug selectivity similar to that determined in membrane preparations indicating localization of Glu receptors. Other progress in this area is the detection of specific [3H]Glu binding sites (& = 0.6 pA4, B,,, = 12 pmol/mg protein) on neuroblastoma cell line (N18-RE-105) (Malouf et al., 1982). QA and Ibo were potent displacers of binding but KA and NMDA were inactive, thus indicating labeling of a Glu receptor. T h e synergistic effects of Ca2+and C1- ions in stimulation of [3H]Glu binding have helped differentiate between Glu receptor subtypes (Fagg et al., 1982). While these APB-sensitive Ca2+lC--dependent sites predominate in the brainstem, thalamus, and hippocampus, the ion-independent variety (APB-insensitive) had the highest density in striatum and thalamus (Whittmore et al., 1983). Further differences between these populations included lability to freezing and detergents. Much evidence outlined by the latter authors suggests that the APB-sensitive Glu receptors may be presynaptically localized. However, APB may also have direct actions via a C1--channel coupled to a hyperpolarizing Glu receptor. This aspect will be discussed further ahead in Section VI,E,F.

TABLE IV SOMEGLUTAMATE BINDING STUDIES~ Rat brain region

Tissue preparation

Assay conditions

Radioligand employed

Whole brain

Crude membranes

Na+-free filtration

L-[ ‘4C]Glu (0.1-30 pA4)

Cerebral cortex

Crude membranes

Na+-free filtration

L-[

Cerebral cortex

Proteolipid fractions

Sephadex chromatography

Cerebral cortex

Crude membranes

* Na+ centrifugation

Kd (pM)

Bmax

(pmoYmg protein)

Reference

0.18 2.10

44,401 -

Michaelis et al. ( 1974)

’“C]Glu (0.4-8.7 pM)

4.0 8.3

200

Roberts ( 1974)

L-[ ‘4CJGlu (0.06-250 p M )

0.3 5.0 55.0

530 32,000 166,000

L-[~H]GIu (0.005-1.6 pM)

1.34 0.37

28

210 (+Na+) 8.4 (-Na’)

DeRobertis et al. (1976) Sanderson and Murphy (1982)

Cerebellum

Synaptic membranes

Na+-free centrifugation

L-[~H]GIu (0.001-1.8 pM)

0.74

73

Foster and Roberts (1978)

Cerebellum

Synaptic membranes

Na+-free centrifugation (sonicated + preincubated)

L-[~H]GIu (0.001-1.8 p M )

0.36

117

Sharif and Roberts (1980)

Striatum

Crude membranes

Na+-free centrifugation

L-[’H]GIu (0.001-1 pM)

0.68

70

Roberts et al. ( 1982b)

Hippocampus

Synaptic membranes

2

Na+ filtration

L-[’H]GIu (0.05- 10 pM)

2.40 0.77

75 (+ Na+) 6.5 (-Na+)

Baudry and Lynch (1981)

Hippocampus

Synaptic membranes

Na+-free centrifugation

L-[’H]GIu (0.001-1 p M )

0.01 1 0.57

2.5 47.0

Werling and Nadler ( 1982)

Forebrain

SJCS

Na+-free centrifugation

L-[’H]GIu (0.01-1 p M )

0.45

91

Foster et al. ( 1981a)

0

T h e s e data have been adapted from Sharif (1984).

RECEPTORS FOR NEUROACTIVE AMINO ACID TRANSMITTERS

119

2. Aspartate Receptor Binding Studies The central actions of L-G~uand L - A s ~could be predicted to be mediated via a common receptor site because of their structural similarities. L-ASPis a dicarboxylic AA, being a shorter homolog of L-G~u. However, the multiplicity of excitatory AA receptors based on numerous pieces of neurophysiological evidence (Watkins and Evans, 1981) invokes the existence of relatively distinct recognition sites for L-ASPand/ or its more potent analog NMDA (Table I1 above). Up until recently a few concerted attempts to demonstrate the latter receptors biochemically have been undertaken. Fiszer de Plazas and DeRobertis (1976) first described the Na+-independent binding of ~ - [ l ~ C ] A sto p hydrophobic rat cortical proteins. Three kinetically distinct sites of variable affinity (& = 0.2-50 pA4) and capacity (B,,, = 3-617 nmolimg protein) were found. While the highaffinity component showed selective interaction with L - A s and ~ NMDA, L-G~uand KA did not compete for this site. The unexpectedly high B,,, for [ 14C]Aspbinding may have resulted from sequestering of the label by the lipids in the proteolipid fraction used, and this has tended to detract from a possible physiological relevance of these early observations. A similar paradoxical demonstration of high affinity (Kd = 10 nM) [3H]NMDA binding to neural membranes (Snodgrass, 1979) has been difficult to reproduce in several laboratories, perhaps indicating an artifactual observation. Alternatively, subtle differences in tissue preparation and assay conditions between the investigators may have contributed to the lack of success. In any event, these studies need following u p since NMDA receptors are the best defined excitatory AA receptors (Watkins and Evans, 1981). Another approach to labeling Asp receptors has been to use L[3H]Asp. A single class of specific, saturable (& = 0.87 pM, B,,, = 44 pmol/mg protein), and reversible (dissociation half-life = 32 sec) binding sites for [3H]Aspwere detected on fresh rat cerebellar membranes using saturation analysis (Sharif and Roberts, 1981a, 1984). As with r3H]Glu binding to similarly prepared membranes, [3H]Aspinteraction was optimum at physiological p H and temperature and was highest in synaptosoma1 fractions (Sharif and Roberts, 1981a, 1984).T h e apparent success in differentiating between [3H]Glu and [3H]Asp binding to cerebellar membranes rested mainly on their differential pharmacology of inhibition by putative analogs. T h e potent inhibitors of [3H]Asp binding included L-G~u,L - A s ~ ,Ibo, DL-~-AS,HA-966, and DL-APV, while compounds that showed much affinity for [3H]Glu binding (DL-homocysteate, L-cysteine sulfinate, L-G~u,DL-APB)were almost de-

120

NAJAM A. SIIARIF

void of activity in displacing [3H]Asp from membranes (Table 111above; Sharif and Roberts, 1981a, 1984). Other distinguishing features of [YH]Aspbinding include half the affinity and a third of receptor capacity compared to [3H]Glu binding, and an apparently different mechanism of stabilization of Asp receptors by lyophilization than for Glu receptors (Sharif and Roberts, 1981a,b). Dissimilar equilibrium binding parameters of Glu and Asp receptors in SJCs and a differential sensitivity to detergents (Foster et al., 1981a) provide further support for a biochemical distinction between receptors for these AAs. p described above have Many characteristics of ~ - [ ~ H l A sbinding been verified and extended recently to show that Asp receptors are heterogeneously distributed in the rat CNS (cerebellum = cortex > pons > spinal cord) and localized postsynaptically when frozen-thawed membranes in Na+-free Tris-C1- buffer were used (Di Lauro et al., 1982). Although spinal NMDA receptors are the best electrophysiologically characterized excitatory AA receptors, little attention has been directed at studying these moieties biochemically. Notably a somewhat discouraging feature of recent [3H]Asp binding determinations may be the relative inactivity of NMDA as a displacer (Sharif and Roberts, 1981a; Di Lauro et al., 1982). Similarly, though ARG of ~-[’HlAsp(a marker for Glu uptake sites) binding has been described (Parsons and Rainbow, 1983), L - A s ~ receptors have not been localized in this way. Interest in this field will no doubt increase following the reports of neurotoxic properties of Ibo (Schwarcz et al., 1983a), NMDA (Olney et al., 1979), and a rigid NMDA-analog, quinolinic acid (Schwarcz et al., 1983b), all of which are believed to interact selectively with NMDA receptors (Watkins and Evans, 1981).

3. KA Receptor Banding Studies KA is a rigid analog of L-G~uof natural origin which has powerful neuroexcitatory and neurotoxic properties (McGeer et al., 1978). It appears to activate receptors distinct from those of other acidic AAs (Watkins and Evans, 1981) even though KA was considered a prototypic ligand for Glu receptors for a long time. Evidence that has suggested autonomous sites of action of KA and Glu (and Asp) includes the GDEE sensitivity of Glu-induced but not KA-induced depolarizations of feline spinal neurons with the opposite specificity being exhibited by Y-D-GG (Watkins and Evans, 1981), the potentiatory effects of KA on Glu responses in cortex (Shinozaki and Konishi, 1970) and at invertebrate neuromuscular junction (Shinozaki and Shibuya, l974), and greater po-

RECEPTORS FOR NEUROACTIVE AMINO ACID TRANSMITTERS

121

tency of KA than other excitant AAs at evoking striatal dopamine release (Roberts and Sharif, 1978) and stimulating cerebellar cyclic guanyl nucleotide (G. Foster and Roberts, 1980). Moreover, KA elicits endoge~ other agonist nous Glu release (Ferkany et al., 1982), while L - G and acidic AAs inhibit such efflux via presumptive autoreceptors (McBean and Roberts, 1981). Additional support has come from binding studies. The initial studies by Simon et al. (1976) demonstrated specific binding of [3H]KA to rat brain membranes at 4°C in a Tris-citrate buffer. A single site of 60 nM Kd of relatively low density (1 pmol/mg protein) appeared concentrated in synaptosomal fractions, displayed a nonuniform (CNS) localization, and avidly interacted with KA and QA. Here LGlu inhibited [3H]KA binding with 1/48 the potency of KA. Since most of these features of [3H]KA binding differed from E3H]Glu binding, it was proposed that KA interacts with a subpopulation of Glu receptors. Recent studies have, however, revealed ["JKA binding sites of two affinities ( K d , = 4-16 nM, Kd2 = 27-66 nM) in rat brain membranes (London and Coyle, 1979a), and such multiplicity was further supported by biphasic dissociation kinetics of the label and a differential pharmacology of displacement by AA analogs. Thus, while KA > QA > L - G ~ inhibited [3H]KA binding to the high affinity site, dihydrokainate > Lglutamine > D - G preferred ~ binding to the other site. In addition, the high-affinity component exhibited a discrete brain regional distribution (striatum > cortex > hippocampus > cerebellum) different from the low-affinity site which predominated in the cerebellum and pons (see Table V). An intact glutamatergic innervation is required for KA to exert its neurotoxic effects (McCeer et al., 1978). Much evidence now indicates that KA activates prolonged release of excitatory AAs by interacting with receptors on presynaptic elements (Ferkany et al., 1982; Collins et al., 1983). Receptor ARG has provided compelling correlation between tissue sensitivity to KA and the density of [3H]KA binding sites there. High levels of [3H]KA sites were autoradiographically visualized in pigeon and human cerebellum (Henke et al., 1981) and in rat olfactory bulb, striatum, hippocampus, amygdala, and cortex (Unnerstall and Wamsley, 1983; Monaghan and Cotman, 1982). The importance of such localization of [3H]KA binding sites is underlined by the good correspondence between symptomology of status epilepticus, HD, and KA-induced neuronal damage in the rat hippocampus and striatum, respectively (Coyle, 1983a,b). Beaumont et al. (1979) have found 53-55% lower [3H]KA binding in HD patient putamen and caudate, supporting the findings of Schwarcz and Coyle (1977) from the animal model of HD. One enigma

122

NAJAM A. SHAKIF

TABLE V PHARMACOLOGY 01;K A I N A T E AND NMDA BINDING" Competing drug

Kainate binding K , (WW

NMDA binding K , (PM)

0.001 0.023 3.3 6.6 7.1 19.4 155.0 > 100 > 100 > 100 >lo0

0,600

~~

Kainate Quisqualate L-Glutamate Ibotenate Dihydrokaindte m-Homocystedte L-Glutamate D-Glutamate L-GlUtdniate diethylester N-Methyl-waspartate 2-Amino-4-phosphonobutyrate N-Methyl-L-aspartate D- Aspartate L-Aspartate

0.009 0.053 2.300 0.440 0.017 0.155 0.320 0.900

Kainate binding data were modified from London and Coyle (l979a) and refer to the high-affinity component in rat forebrain membranes. The NMDA data are taken from Snodgrass (1983) involving mouse cerebellar membranes.

of KA toxicity that remains is whether the high- and/or the low-affinity receptors are involved and whether KA has a direct as well as an indirect mechanism of action. Other studies of ["JKA binding have addressed the questions of phylogenetic distribution (London et al., 1980) and the ontogeny of KA receptors (Slevin and Coyle, 1981) with the findings that nonchordates and lower vertebrates (e.g., hydra and frog, respectively) have about 2.7 and 420 times the density of KA sites of human cerebellum, and that [3H]Gluand [SH]KAbinding sites develop at different rates in rat brain. A further distinction between the latter sites has been niade on the basis of differential effects of Na+-cholate on their binding properties (Michaelis et al., 1981). The foregoing discussion has outlined the receptor binding characteristics for the three most studied radioligands, viz. [3H]Glu,[3H]Asp, and [3H]KA. In the absence of many overlapping properties, it may be just to conclude that the receptors labeled by these compounds represent separate recognition sites. The super high-affinity components of [3H]Glu and [3H]KA binding are difficult to interpret-perhaps they represent sites for endogenous peptides able to modulate the activity of the coupled AA receptors. The existence of an unidentified non-AA inhibitor of cerebellar [3H]Glu binding (Sharif and Roberts, 1980) and

RECEPTORS FOR N E U R O A C T I V E AMINO ACID TRANSMITTERS

123

the recent discovery of a brain dipeptide (NAAG) that competes for [3H]Glu binding (Zaczek et al., 1983) add weight to this postulate.

C. OTHERAGONIST BINDING STUDIES 1. Cysteine S u F n a t e Binding

The powerful excitatory actions of sulfur-containing acidic AAs have suggested to some (Watkins and Evans, 1981) a synaptic role for such compounds. Among the latter are sulfonic and sulfinic acids such as cysteate, homocysteate, cysteine sulfinate, and homocysteine sulfinic acid. L-Cysteine sulfinate (CSA) is structurally related to Asp and Glu and exhibits some putative neurotransmitter properties (for reviews, see Recasens et al., 1982; Iwata et al., 1982). Notably most of these sulfinic/ sulfonic AAs are potent competitors of postsynaptic [3H]Glu binding (Sharif and Roberts, 1981a, 1984; Mewett et al., 1983). The binding activity of [3H]CSA has been evaluated recently using frozen-thawed brain membranes in a centrifugation (Kecasens et al., 1982) and a filtration (Iwata et al., 1982)assay. Na+-independent binding (15 min/5"C) was of high affinity and low capacity (Kd = 0.1 p M , B,,,,, = 2.4 pmol/mg protein), while binding to uptake sites in the presence of high Na+ had the converse profile (Recasens et al., 1982). Similarly, whereas CSA was the most powerful inhibitor of Na+-independent bind~ ing ( K , = 0.09 p M , 3-fold > ~.-Gluand 80-fold > L-AsP), L - A s best antagonized [3H]CSA transport. The profile of postsynaptic binding density was cerebellum > olfactory bulb > cortex 9 retina, while striatal, cortical, and hypothalamic uptake activities were the highest. The dissimilarities of [YH]CSAbinding to receptor and uptake sites has therefore allowed easy distinction between presynaptic and postsynaptic interactions of this ligand. l'hese features, although requiring confirmation in other laboratories, appear to resemble those reported for [3H]Glu binding more than for ["]Asp binding (Sharif and Roberts, l981a,b, 1984), and in common with previous suggestions (see Watkins and Evans, 1981) it seems reasonable to propose that [SH]CSAmay be binding to a subpopulation ofr.-Glu receptors. This is borne out by the observation of potent inhibition of [3HJC;lubinding by honiologs of CSA t o cerebellar (Table I11 above; Slevin et al., 1983), hippocampal (Werling and Nadler, 1982), retinal (Mitchell and Kedburn, 1982), and brain membranes (Mewett et al., 1983). Whether these sulfur-containing AAs function as neurotransmitters at some excitatory synapses in the CNS needs to be determined.

124

N A J A M A . SHAKIF

2. A M P A Binding (RS)-a-amino-3-hydroxy-5-methyl-4-isoxa~olepropionate(AMPA) is a conformationally restricted Glu analog having a structure similar to Ibo and QA (Krogsgaard-Larsen et al., 1980). AMPA caused GDEEsensitive (D-a-AA-insensitive) excitations of feline spinal neurons, but was a poor inhibitor of [3H]KA and E3H]Glu binding (Krogsgaard-Larsen et al., 1980; Honore et al., 1981). In binding studies, [3H]AMPA labeled two sites of 9 and 2440 nM Kds in frozen brain membranes, which upon repeated freezing-thawing exhibited only one component of intermediate affinity (Kd = 278 nM) (Honor6 et al., 1982). The inhibition profile of [3H]AMPA binding (AMPA > QA > L-G~u > KA > homo-Ibo Ibo > D-LX-AA) seems fairly different from pharmacological activities of these compounds against other acidic AA/analog binding (Tables I1 and IV above), although some similarity to [3H]Glu binding does prevail, but where AMPA was devoid of activity. These disparities are very difficult to interpret, and direct comparisons are impossible because of the dissimilarities in tissue preparations, etc.

D. ANTAGONIST HINDINC. STUDIES 1. DL-APB Binding The phosphonate analog of Glu, APB, has documented antagonist properties against excitatory AA-induced responses in the CNS (Evans and Watkins, 1981; Watkins and Evans, 1981). Although in many cases APB appears relatively nonselective, the L-isomer carries the antagonist activity. Tested in receptor binding paradigms, DL-APBwas found t o be 10 times more active at displacing [SH]Glu than ["]Asp from cerebellar membranes (Sharif and Roberts, 1984; Table I11 above). Results from other studies in different tissues (Mitchell and Redburn, 1982; Slevin et al., 1983; Fagg et nl., 1983a,b) were similar, indicating that DL-APBprobably interacts with a QA-type, r,-Glu receptor. In the study by Fagg P t al. (1982) it became apparent that the presence of Ca2+/C1-in assay buffer potentiated the inhibition of ["H]Glu binding by the niajority of phosphonates. This issue will be discussed in detail ahead. Binding of DL-['HH]APB to brain synaptic membranes in HEPESKOH buffer containing 2.5 mM CaC12 was saturable and reversible (tb dissociation 1.5 niin) (Butcher et al., 1983). Saturation isotherms indicated a single binding component (& = 1.3 pM, B,,,,3,= 12-39 pmol/mg L-

RECEPTORS FOR NEUROACTIVE AMINO ACID TRANSMITTERS

125

protein), but Hill plots yielded coefficients of 1.26-1.35, which can be interpreted as site-site interactions or the presence of multiple binding components. Binding activity appeared concentrated in synaptosomal fractions and exhibited the following brain regional distribution: hippocampus 2 striatum > cortex > cerebellum > pons > spinal cord. Various analogs of Glu competed for DL-[~H]APB binding (QA > Lhomocysteate > L-Glu > L-cysteate > L-ASP> L-cysteine sulfinate for agonists; DL-WAS> L-APB > L-a-AA > DL-APB> DL-CY-AA > DL-APX > DL-APH> DL-APVfor putative antagonists). It is noticeable again that sulfur-containing agonists were the most potent inhibitors of binding (along with QA and the phosphonates). Binding assays conducted in the presence of 2.5 mM salts revealed that halide anions (Cl- > Br- > F-) and divalent cations (Ca" > Mn2+= Mg2+ = Sr2+> Ba2+)enhanced DL-[~HH]APB binding by increasing B,,, with no change in Kd. Many of the properties of the described DL[3H]APB binding suggest the labeling of a QA-type Glu receptor. However, some of the anomalous findings include the potent displacer activities of specific NMDA antagonists (although NMDA itself was inactive) and the relatively low binding affinity (Kd = 1.3 /AM),especially since APB is an antagonist. However, the racemicity of the radioligand is probably responsible for this paradox. Another question that needs attention is the localization of the [3H]APB binding site-the authors suspect that a presynaptic autoreceptor is perhaps being labeled. 2. D-APV Binding The radiolabeling of NMDA receptors on rat brain membranes has been achieved. Specific D-[~H]APVbinding, though 40% of the total, exhibited high affinity (Kd = 0.47 p M ) , a heterogeneous brain distribution (hippocampus > cortex > striatum > cerebellum = pons), and an avid selectivity for D-APV(> L - G > ~ APH > L - A s = ~ NMDA > D-GG% APB = QA S- KA) with a Ki for D-APVof 0.62 pM (Olverman at al., 1984). T h e high potency of both L-Glu and L - A s here ~ is another example of how their flexibility enables them to interact with most of the subclasses of excitatory AA receptors. 3. DL-APHBinding Detailed structure-activity studies (Evans and Watkins, 198 1) have shown that APV and the heptanoate homolog (APH) are very specific NMDA (L-Asp) antagonists. Many other neurobiological investigations have confirmed this premise, including antagonism of NMDA-induced cyclic GMP synthesis (Roberts et al., 1982) and [3H] acetylcholine (ACh) efflux (Scatton and Leyman, 1982) and blockade of NMDA-induced

126

NA-JAM A. SHAKIF

seizures (Croucher et al., 1982) and NMDA-induced neurotoxicity (Schwarcz et al., 1983a). Preliminary studies of Ferkany and Coyle ( 1983) have demonstrated the specific binding of DL-[~H]APH to frozen, crude mitochondria1 f'ractions of rat brain. Only low-affinity ( K d = 3.6 p M ) binding was detected using Tris-citrate buffer in the absence of any added ions. The hippocampus (> cortex > striatum > cerebellum) was enriched in binding sites for DL-[:~H]APH, and binding in brain membranes was inhibited by QA > Ibo > L-Glu > D-c~-AA > DL-homocysteate > DL-APB> DL-APV + quinolinate. Many disparities of this work with electrophysiological and other neurochemical studies argue against the site labeled with DL[3H]APH being an NMDA receptor despite some evidence outlined above. Previous data (Roberts et al., 1982a; Table 111 above) suggest that the inhibitory activity of APH against [:'H]<;lu binding is probably mediated via a Glu receptor. However, it is obvious that the radiolabeled phosphonates will greatly enhance our knowledge of the excitatory AA receptor heterogeneity and thus help define the properties of these entities. One remaining question about the radioligand binding studies described above is why the spinal cord has been ignored as the tissue of choice even though the definition of heterogeneous excitatory AA receptor has been determined mainly in this region (Watkins and Evans, 1981). The relatively low density of these receptors here may be one factor, but this should not be a deterrent.

E. PHARMACOLOGY OF RESPONSE MECHANISMS I have alluded to this subject from time to time during the previous discussions without much emphasis. Comparison of the drug selectivity of acidic AA binding sites is made in Table I1 above, and the differences generally support the concept of distinct receptors for these ligands. Further evidence has come from qualitative and quantitative analysis of evoked responses in vitro. T h e pharmacology of receptors mediating neuronal depolarization has already been detailed in Section VI,A. One of the earliest response mechanisms studied was the AA-induced release of [3H]dopamine from striatal slices (Roberts and Sharif, 1978; Roberts and Anderson, 1979). This Ca2+-dependent efflux was dose related to L-G~uconcentration (EGO= 63 pM) and was antagonized by GDEE. T h e rank order of potency of agonists was KA > L-glu > L - A s> ~ D-G~u > DL-homocysteate > glutamine % D - A s (Roberts ~ and Sharif, 1978). Of the phosphonate analogs, DL-APBwas tested, and it

RECEPTORS FOR NEUROACTIVE AMINO ACID TRANSMITTERS

127

was about 75% as active an inhibitor as GDEE, whereas HA-966 and D-(YAA were inactive (Roberts and Anderson, 1979). Since exogenously administered L-G~ustimulates the production of cGMP (Biggio and Guidotti, 1976; Garthwaite and BalBzs, 1978), a detailed examination of the receptor mediating this response was undertaken (Foster and Roberts, 1981; Roberts etal., 1982a). Rat pup cerebellar slices exposed to excitatory AAs generated cGMP in a Ca2+-dependentmanner, and the profile of AA activity was NMDA > 4F-Glu > Ibo > DL-homocysteate > cysteine sulfinate > L-G~u.All the excitants’ actions were similarly antagonized by DL-APBand D-a- AA. However, while GDEE blocked KA and L-G~u effects, DL-APVwas effective against KA > NMDA > Ibo > L - A s > ~ L - G responses ~ (Roberts et al., 1982a). Similarly, while DL-APHwas a potent antagonist of L - A s ~> NMDA > Ibo > L - G responses, ~ Y-D-GGsuppressed effects of KA > LAsp > NMDA > Ibo > L-G~u. Some of these data correspond well with electrophysiological findings, and an NMDA-type receptor may be primarily involved. However, the absolute identification of a particular receptor type for the cGMP response mechanism must await further studies because the agonist specificity is not particularly clear-cut. Perhaps more than one receptor type is coupled to a guanylate cyclase. Although these studies indicate that all excitatory AAs and their agonist analogs stimulate cerebellar cGMP synthesis, albeit slightly differentially, protoveratrine depolarization of slices and the resultant cGMP response were selectively antagonized by GDEE, indicating that a QA-type Glu receptor may be involved in the mediation of responses to an endogenous AA transmitter like L-G~u(Foster and Roberts, 1981). The ionic mechanism transducing the actions of extracellularly applied excitant AAs, in particular Glu, is now thought to be enhanced Na+ permeability which causes depolarization of the neuron (Zieglgansberger and Puil, 1973; Hosli and Hosli, 1978). Luini et al. (1981) have studied the pharmacological specificity of the efflux of 22Na+from preloaded striatal slices. The relative efficacy of excitatory AAs to release 22Na+appeared to be as follows: NMDA > Dbhomocysteate > KA > QA B L - G > ~ L - A s ~The . most potent antagonists for the various ago>DL-~nists were as follows: for NMDA, DL-APV> Y-D-GG> DL-(Y-AA AS; for KA, KA-phenylthiolactone > KA-iodolactone > Y-D-GG. N o particular organic compound blocked the effects of either QA o r L-Glu. These results indicate that an NMDA-type receptor capable of activating the Na+ channels on striatal neurons is likely to be the most physiologically relevant excitatory AA receptor in this tissue. Furthermore, w e can deduce that NMDA o r a compound of similar structural constraints may be the endogenous ligand for these receptors. However, even though

128

NA-JAM A. SHARIF

Scatton and Leymann (1982) have also identified NMDA-type receptors that mediate [:'H]ACh release from striatal slices, a few words of caution in interpretation of these results should be voiced. High micromolar-millimolar concentrations of agonist excitant AAs were necessary to induce the changes in cGMP levels (Foster and Roberts, 1981) and to evoke release of 22Na+ (Luini et al., 1981) and [3H]ACh (Scatton and Leymann, 1982), and similar levels of antagonists were effective in blocking these responses. The agonists' actions are complicated by avid uptake mechanisms that obviously necessitate the addition of high doses to see an effect, with the consequence that those excitants that are not substrates for the transport sites (e.g., NMDA, KA) look more efficacious. The data outlined above, however, seem valid even with these constraints since specific blockade of NMDA receptors was associated with the diminution of the respective responses. It is also pertinent to note that iontophoretic/electrophysiologicalinvestigations are prone to the above-mentioned problems, but despite this the pharmacology of these in vitro biochemical responses provides the best correlation with the neurophysiological studies, indicating that the same receptors are being studied by both techniques. The disparities between electrophysiological and binding studies are even more difficult to explain. On the basis of cross-desensitization of the 22Na+efflux response in hippocampal slices, Lynch and co-workers have proposed the reclassification of Watkins' three-receptor model into a synaptic receptor activated by DL-homocysteate (Gl), an extrasynaptic Glu receptor (G2), an NMDA, and a KA receptor (Baudry et al., 1983a,b). Their detailed analyses indicate that the G1 is stimulated by DL-homocysteate, does not desensitize, and is blocked by D-cY-AA and DL-APB;G2 is stimulated by LGlu, desensitizes, and is not blocked by DL-APBor D-(Y-AA;while the NMDA receptor is sensitive to D-~Y-AA and desensitizes, the KA receptor exhibits the opposite sensitivity. Baudry et al. (1983a,b) propose that the G1 receptor represents the postsynaptic entity with which the endogenous excitatory AA transmitter(s) interacts, and since G2 and NMDA receptors cross desensitize they may be extrasynaptic and coupled to the Na+ channel (Fagni et al., 1983). Since similar response mechanisms were also detected in striatal slices (see above), this classification may be quite useful for defining receptors for excitatory AAs in the mammalian CNS as a whole. Obviously, different tissues may contain some and not all the different receptor types, and this limits gross generalizations. Another aspect of excitant AA function concerns their convulsant and neurotoxic properties (Olney, 1981). KA, Ibo and NMDA injected into rat brain in nanomolar quantities produce axon-sparing neuronal

RECEPTOKS FOR NEUKOACTIVE AMINO ACID TRANSMITTEKS

129

death (Coyle, 198Ya,b), probably by activating their specific receptors. While KA is now believed to exert its neurotoxic actions by releasing high concentrations of endogenous excitatory AAs (Ferkany et al., 1982), enhancing Ca2+and Na+ influx (Berdichevsky et al., 1983) and probably overstimulating postsynaptic KA receptors, both Ibo and NMDA probably have direct lesioning effects (Coyle, 1983a,b). In support of this hypothesis and the existence of receptors for these latter ligands, several NMDA (and Ibo) antagonists (APV, APH, D-a-AA) are known to block specifically the neurotoxicity due to NMDA (Olney et al., 1971, 1979) and Ibo (Schwarcz et al., 1983a). T h e same antagonists potently suppress NMDA- and/or sound-induced seizures (Meldrum et al., 1983) and kindling (Peterson et d., 1983). The above discourse provides important information about the biochemical and physiological response mechanisms of excitatory AAs, and the relatively specific blockade of these functions by organic antagonists is further evidence for the mu1tiplicity”of receptors for these AAs. Koerner et al. ( 1983) have recently assessed electrophysiologically the activity of a number of y-substituted Glu analogs on hippocampal dentate cells and found L-serine-O-sulfateand L-Glu-tetrazole to be agonists, while L-APB and L-O-phosphoserine possessed antagonist activity. The latter two compounds also inhibited [3H]Glu binding with relatively high potency (Foster et al., 1982). Other new nonhomologous organic agents able to alter excitatory AA activity include ‘2-chloroadenosine (Dolphin, 1983), caroverine (Ishida and Shinozaki, 1983), streptomycin (Stone and Perkins, 1983), kynurenate (KYA) (Ganong et al., 1983), and quinolinate (Quin) (Stone and Perkin, 1981). The latter two compounds are endogenous tryptophan metabolites, but while Quin is an agonist with powerful neurotoxic properties similar to KA (Schwarcz et aZ., 1983b), KYA antagonizes synaptic and acidic AA-induced excitations (Ganong et al., 1983) and apparently blocks the lesioning effects of Quin (Foster and Schwarcz, 1983). Further multidisciplinary studies with these drugs may provide new insights into the properties and regulation of synaptic functions of acidic AAs and may lead to the development of more potent compounds.

F. REGULATION OF EXCITANT AMINOACID RECEPTORS Proteinaceous receptors adapt rapidly to changes in their environment. In vivo receptor properties depend on the relative availability of their specific ligand(s), other cofactors, and the status of their neighboring proteins and the surrrounding membrane lipids. The employment

130

NAJAM A. SHARIF

of radioligand binding studies in the determination of receptor changes during or following chronic drug therapy and neuronal lesions can help answer questions related to tolerance, addiction, and withdrawal syndromes. In vitro experiments that simulate these phenomena can aid in our understanding of similar processes that may operate in uivo under normal and pathological conditions. Electrophysiological (Watkins and Evans, 1981 ) and biochemical (Luini et al., 1981) studies have provided evidence for the coupling of a Na+ channel to excitatory AA receptor(s). Binding experiments have shown that monovalent cations (Na+ > K+) at low millimolar concentrations reduce receptor interaction of [3H]Glu (Baudry and Lynch, 1979; Sharif and Roberts, 1981a,b) and [3H]Asp (Sharif and Roberts, 1981a), probably by acting at the ionophore and/or by causing protein aggregation. This exemplifies a negative feedback mechanism that probably occurs in vivo at these excitatory synapses. On the contrary, divalent cations like Ca2+ enhance [3H]Glu and [3H]Aspbinding to brain membranes at low millimolar levels. A mechanism involving exposure of new Glu sites due to actions of a thiol protease-induced proteolysis appears to be involved since micromolar inhibitors of this enzyme and temperatures < 20°C antagonize this effect (Baudry and Lynch, 1980; Vargas et al., 1980; Baudry et al., 1981a,b). Furthermore, the stimulatory action of Ca2+is absent in neonates and certain brain regions of adult rats and is lost following detergent treatment of membranes. Similarly, the increased binding sites detected following membrane incubation, their pronounced cold lability (Sharif and Roberts, 1980, 1984), and the sharp pH and temperature optimums of binding all implicate a possible enzyme-mediated stimulation of [3H]Glu binding, a n d o r the inactivation of an endogenous inhibitor (Sharif and Roberts, 1980). The “heat’’-induced [3H]Glu binding (Sharif and Roberts, 1980) has now been confirmed in the cerebellum (Honor6 et al., 1981) and hippocampus (Werling and Nadler, 1982; Baudry and Lynch, 1981) and may involve the removal of certain inhibitory substances including endogenous phospholipids and excitatory peptides (Sharif, 1984). Although Baudry and co-workers have been quite influential in arguing in favor of the Ca2+-dependentproteolysis mechanism of enhancing hippocampal [SH]Glubinding, reports of similar phenomena following in viuo electrical kindling (Savage et al., 1983) and in uitro long-term potentiation (LTP) induced by electrical stimulation (Baudry et al., 1980), sonication, and lyophilization of cerebellar membranes (Sharif and Roberts, 1980) weaken this proposal slightly. It is likely, however, that the latter biophysical treatments produce the same net result as the

RECEPTORS FOR NEUKOACTIVE AMINO ACID TRANSMITTERS

131

Ca2+-activatedevents by displacing limiting membranes and inhibitory substances from the receptors and/or partially extracting receptor proteins from their lipid environment, thus exposing new binding sites. Thus, in many ways, controlled in vitro ultrasonication of membranes may be a useful model of in vivo kindling and other seizure states, but only insofar as it concerns what happens to receptor properties and the lipid content of membranes during these dysfunctions. Another facet of Ca2+ and C1- ion regulation of [3H]Glu binding concerns their ability to enhance the potency and alter the rank order of activity of homologous phosphonate antagonists in their competition for forebrain [3H]Glu binding (Fagg et al., 1982, 1983a,b) without affecting the displacement by L-G~u or L-ASP,an observation contradicted in cerebellar (Sharif and Roberts, 1984) and hippocampal (Larder and McLennan, 1983) membranes. These ions failed to modify the actions of these compounds on E3H]Aspbinding. T h e properties of the Ca2+-dependent, Na+-independent [3H]Glu binding apparently correlate well with the DLAPB sensitive sites at the perforant path-granule cell synapses of rat hippocampus (Koerner and Cotman, 1981). An apparent anomaly is that while the actions of QA and Ibo are increased by CaC12, L - G and ~ LAsp do not exhibit the same sensitivity (Fagg et al., 1982). Interestingly, the binding of DL-[~H]APB is similarly enhanced in the presence of C1and Ca2+ ions (Butcher et al., 1983), but this effect is lost on Triton solubilization of membranes to make SJCs (Foster et al., 1981a). Fagg et al. (1983a,b) have interpreted these results in terms of C1-/Ca2+-dependent, APB-sensitive sites being associated with a C1- channel in support of the inhibitory actions of Glu at some cerebellar (Yamamoto et al., 1976) and other central synapses (Watkins and Evans, 1981). However, the localization and the mechanism of the C1-/Ca2+-induced augmentation of compound potency remain to be elucidated. A possible explanation for the stimulatory actions of CaC12 may be as follows: partial cyclization of acidic AAs, accomplished by chelation of Ca2+ by anionic groups of the compounds, would prime the ligands for receptor binding (McLennan et al., 1982). The possible depletion of Ca2+from the ligandCa2+complex by high-affinity calcium-binding protein(s) (CBP) (Grab et al., 1979; Cheung, 1982) would activate the latter and help concentrate the ligands at the synaptic sites. Further, the Ca2+-inducedactivation of CBPs, including Ca2+-dependentpeptidases, may lead to favorable conformational changes, even proteolysis, and result in exposure of new binding sites (Baudry et al., 1981a,b; Vargas et al., 1980). Thus, ligand chelation of Ca2+ ions with the resultant concentration of the former, and the stimulation of CBPs by the Ca2+,may help explain the above observations. Alternatively, Ca2+ and other divalent cations may com-

I32

NAJAM A. SHARIF

plex with the putative AA receptor inhibitors (see Sharif, 1984, for review), disinhibit the receptor, and consequently allow a greater access of the ligands to the recognition sites. However, these remain speculations. Possible endogenous inhibitors of excitatory AA receptor binding include guanyl nucleotides (GN) (Sharif and Roberts, 1980, 1981b), phospholipids (Michaelis et al., 1981; Foster et al., 1982), small peptides (Kanazawa et al., 1965; Sano et al., 1966; Francis et al., 1980; Zaczek et al., 1983), and an unidentified substance(s) (Sharif and Roberts, 1980). T h e cCMP response and GN effects confirm cyclase coupling of some Glu receptors. An exciting new concept involves the possible transmitter role of these small peptides at acidic AA receptors. NAAG has recently been isolated from rat brain and shown to be a potent (APB-sensitive) neuroexcitant and displacer (partial) of [3H]Glu binding (Zaczek et al., 1983). T h e brain regional distribution (spinal cord > medulla > tegmentum > hippocampus > cortex > cerebellum) of NAAG is modified by specific lesions (Koller et al., 1983) indicating its neuronal localization. Another peptide (phenylalanyl glutamate), which resembles the structure of domoate (a KA analog), selectively enhances [3H]APH binding at low-high micromolar levels but inhibits [3H]KA' binding (Ki = 5 FM) (Ferkany et al., 1983). These studies indicate that similar peptides containing acidic AAs may operate in whole o r in part at some excitatory AA synapses, since they may coexist with the transmitters in glutamatergic/aspartergic terminals. The super high-affinity ( K d s = 5 and 70 nM) [3H]Glu binding sites found in rat cortical membranes (Sharif, unpublished; Fig. 4) and in other tissues previously (Bizere et al., 1980; Mitchell and Redburn, 1982; Werling and Nadler, 1982) may represent templates associated with Glu receptor proteins for these modulatory peptides. The heat-induced elevation of [3H]Glu binding (Sharif and Roberts, 1980; Honore et al., 1981; Werling and Nadler, 1982) may involve induction of new sites due to the presence of endogenous Glu in the membrane preparation (Sharif and Roberts, 1984), as has been demonstrated for peptide transmitters (see Hirsch and Margolis, 1979). Similar ligand-induced enhancement of radiolabeling has been observed with 5-15 mM Glu in locust neuromuscular preparations (M. T. Filbin, personal communication), with 10 mM Glu and/or Ca2+in neuroblastoma cells (Malouf et al., 1982), and with nanomolar Glu analog agonistsl antagonists in hippocampal membranes (Werling and Nadler, 1982). Since high concentrations of endogenous excitatory AAs are thought to be released during tetanic electrical stimulation (Peterson et al., 1983), the increased [3H]Glu binding detected in kindled (Savage et al., 1983)

RECEPTORS FOR NEUROACTIVE AMINO ACID TRANSMITTERS

BOUND/FREE

133

LOG (3H)GLU

FIG.4. (A) Eadie-Hofstee plots of specific [3H]Glu binding to rat cortical membranes. Biphasic plots indicate the presence of two apparent subclasses of super high-affinity binding sites of different capacities as shown. (B) Hill plots of the same data, indicating positive cooperativity between these sites. (Sharif, unpublished data.)

and LTP (Baudry et al., 1980) tissues may be related to these phenomena. The molecular mechanism behind the latter are currently not known, but probably involves both presynaptic and postsynaptic elements (Cotman et al., 1981), and perhaps desensitized receptors are being measured in these studies. Another form of receptor up-regulation is exemplified by the effects of denervation. Thus, while decortication led to a 34% increase of striatal Clu receptor density (Roberts et al., 1982b; see Fig. 5), increased receptor affinity was detected in hippocampal preparations of decommissured rats with the development of a concomitant supersensitive response to applied L - G in ~ the 22Na+efflux assay (Baudry et al., 1983a,b). Similar studies in locust (Gration et al., 1979) and rat hippocampus (Segal, 1977) are also associated with supersensitivity of Glu receptors determined from depolarization criteria. Other interesting findings related to these changes include the hypersensitivity of neonatal cerebellar guanylate-cyclase-coupled Glu receptors (Garthwaite and Balazs, 1978) and of Glu-activated hippocampal 22Na+-releasemechanisms of rat pups (Baudry et al., 1983a,b). The neonatal cerebellar Glu supersensitivity appears to correlate well with absence of presynaptic fibers and the developmental peak of Glu receptor binding and Glu uptake (Sharif and Roberts, 1984) and coincides with the first demonstration of endogenous

134

NAJAM A. SHAKIF

0

0.2

0.4

05

0.8

u)

[3H]Glutamate ( p M )

FIG.5. Apparent up-regulation of [3H]Glu binding sites. Specific [SH]Glubinding to striatal membranes after deafferentation of the corticostriatal glutamatergic pathway. (Redrawn from Roberts et al., 1982b.)

Glu release and appearance of KA toxicity (Foster et al., 1981; Sharif and Roberts, 1984). Furthermore, these ontogenetic changes correlate well with the peak of protein deposition (De Barry et al., 1980) and synaptogenesis (Altman, 1972). Since NMDA and KA become less potent in releasing 22Na+from decommissured hippocampal slices (Baudry et al., 1983a,b), while L-G~u exhibited a greater activity, these results taken in conjunction with endogenous Glu/Asp release studies (Ferkany et al., 1983) and neurotoxicity studies (Nadler and Cuthbertson, 1980) suggest that both NMDA and KA receptors are presynaptically localized, as opposed to Glu receptors which appear to be postsynaptic in this tissue. Previous [SH]DA release studies (Giorguiff et al., 1977; Roberts and Sharif, 19’78; Roberts and Anderson, 1979) suggest that Glu receptors on nigrostriatal terminals may become supersensitive following cortical aspiration (Roberts et al., 1982b), but this requires confirmation. Membrane-bound receptors for excitatory AAs exhibit pronounced lability at low temperatures (see Sharif, 1984, for review). This coldinduced denaturation of [3H]Glu binding sites on cerebellar membranes (Sharif and Roberts, 1980, 1984) extends to similar sites on membranes of striaturn (Vincent and McGeer, I980), hippocampus (Werling and Nadler, 1982), neuroblastoma cells (Prasad et al., 1980), invertebrate neuromuscular tissue (Filbin et al., 1980), and forebrain SJCs (Foster et

RECEPTORS FOR NEUROACTIVE AMINO ACID TRANSMITTERS

135

al., 1981a). Recent evidence suggests the specific denaturation of the Ca2+/C1--dependent (APB-sensitive) ["]Glu binding sites during membrane freezing (Fagg et al., 1983a). Binding sites for [3H]KA (Viveros and Orrego, 1982), ~ - [ ~ H l A(Sharif sp and Roberts, 1981a), and [3H]CSA (Iwata et al., 1982) respond similarly to freezing. The initial phase of apparent receptor loss may involve protein folding since [3H]Glu binding activity could be partially recovered by short periods of warming at physidogical temperatures (Sharif and Roberts, 1980, 1984). However, irreversible receptor denaturing may ensue during membrane freezing by cleavage of important disulfide bonds andlor oxidation of functionally important thiol residues (Sharif and Roberts, 1984). The protection afforded by micromolar levels of exogenous L-G~uagainst sulfhydryl reagents (Sharif and Roberts, 1984) has provided partial evidence in support of this mechanism. Following an extensive evaluation of suitable conditions for preserving binding activity during membrane storage, lyophilization proved most effective not just in affording protection from cold lability of Glu and Asp receptors (Sharif and Roberts, 1980, 1981a), but also in enhancing the membrane binding activity. Cerebellar membranes could be stored for up to 6 weeks under dessication folIowing freeze-drying without loss of viability. This approach has recently been employed to stabilize membranes for binding of [3H]GABA (Chang et al., 1981) and [3H]KA (Viveros and Orrego, 1982). Studies with membranes prepared from previously frozen-intact cerebellums (Sharif and Roberts, 1984) have revealed l3H]Glu binding activity analogous to that from fresh tissues. Taken together, it would appear that postmortem CNS tissues (frozen whole) may prove viable for analyses of excitatory AA receptor properties in disease states. G. NEUROPATHOLOGY OF ACIDICAMINOACIDS

This topic is relevant from the point of completeness of this article, but more so because hypoactivity and hyperactivity of central excitatory AA systems may be directly related to the etiology of a number of pathological conditions. Thus, a chronic overstimulation of Glu receptors may be one of the key factors responsible for HD and epileptic conditions (Coyle, 1983a,b). I n contrast, low activation of similar receptors in the mesolimbic dopaminergic system has been proposed as an alternative precipitating factor for schizophrenia (Kim et al., 1980). Elevated endogenous release of acidic AAs, in conjunction with high dietary intake, has been linked to the so-called "Chinese-restaurant syn-

136

NAJAM A. SIIARIF

drome,” epilepsy, senile dementia, hippocampal stroke (Olney et al., 1979), and also amyotrophic lateral sclerosis, a form of irreversible paralysis otherwise known as MND (Barrow et al., 1974). Further, an inborn error of Glu metabolism that is manifested late in life has been implicated in olivopontocerebellar atrophy (Plaitakis et al., 1982), which is characterized by neuronal death in the cerebellum, pons, spinal cord, and substantia nigra with resultant ataxia, spasticity, and Parkinsonian symptoms. These examples are important in light of the close correlation between excitatory and neurotoxic potency of many of the acidic AAs and their analogs (Olney et al., 1971; McGeer et al., 1978; Coyle, 1983a,b). Although the precise events leading up to neuronal death in response to a neurotoxic challenge remain ill defined, some limited progress has been made in elucidating the mechanism of KA toxicity. KA appears to interact with presynaptic receptors to release massive amounts of endogenous acidic AAs. Some molecules of KA may also interact postsynaptically, in concert with the released AAs, to cause cellular excitation. Following excess deplorization, the neuronal energy supplies may become depleted and the cellular water and electrolyte balance become disrupted. The ensuing osmotic lysis of intercellular organelles may culminate in the overall death of the respective cells. This description may be oversimplified, and therefore readers are referred to recent “state-ofthe-art” publications on this subject for more definitive information (McGeer et al., 1978; Coyle, 1983a,b; Fuxe et al., 1983). The neurotoxicity of NMDA and Ibo can be blocked by coinjection of NMDA antagonists D-(Y-AA,APV, and APH (Olney et al., 1979; Schwarcz et al., 1983a). Similar antagonism of quinolinate-induced lesions (Schwarcz et al., 1983b) by kynurenate (Foster and Schwarcz, 1983) has been demonstrated. To date, however, the neurotoxicity of KA has proved resistant to organic antagonists. Therefore, possible tools to prevent development of HD-like symptoms remain elusive, although since quinolinate mimics the KA effects kynurenate could prove useful.

VII. Summary and Concluding Remarks

The contention of AA-mediated neurotransmission in the mammalian CNS via multiple postsynaptic receptors has received overwhelming support from electrophysiological, anatomical, and neurochemical research over the last few years. The major contributions to the elucidation of such neurotransmitter roles for these AAs include the development

RECEPTORS FOR NEUROACTIVE AMINO ACID TRANSMITTERS

137

of relative potent and specific organic antagonists, the mapping of neuronal pathways utilizing AAs, and the identification, characterization, and visualization of receptors by radioreceptor assays and autoradiographic techniques. Many aspects of this research have been described and discussed in this article. GABA receptors exist in high- and low-affinity forms as determined by binding of agonist drugs. T h e low-affinity sites can apparently be converted to the high-affinity entities by freeze-thawing and detergent treatment. These sites may represent recognition templates for modulatory peptides like GABA-modulin, but this needs to be confirmed. The properties of GABA receptors are further regulated by the activation of coupled BZDIbarbituratetionophore binding sites, and vice versa. The current consensus on the pharmacological effects of BZDs is the facilitation of GABA actions, and this has functional implications in the use of these agents as anticonvulsant and anxiolytic drugs. Novel BIC-insensitive GABABreceptors activated by baclofen have been found to have pharmacology and CNS distribution different from classical GABAA receptors. These GABAB receptors appear to be predominantly localized presynaptically, and many appear to be associated with noradrenergic terminals. Glycine is the caudal inhibitory neurotransmitter, and its receptors in the brainstem and spinal cord membranes have been mainly studied indirectly by labeling the coupled C1- ionophore with [SH]strychnine. Deficiency of these entities seems to be symptomatic of motoneuron disease and of spasticity in mice. In general, research on Gly is increasing as its importance in normal and pathological states becomes more clear. There has been a veritable revolution in the excitatory AA field since the discovery of selective neuronal lesioning (“excitoxic”)agents and the synthesis of some potent antagonists. Although some doubt still exists as to the nature of the endogenous excitatory AA(s) mediating neurotransmission, Glu and Asp are now considered serious candidates at certain synapses. Furthermore, the relative lack of cross-desensitization between the actions of numerous analogs of Glu and Asp and the relatively specific antagonism of their responses by homologous straight-chain phosphonate analogs of these AAs have indicated the existence of multiple receptors for acidic AAs. The current receptor classification involves those proteinaceous templates activated by NMDA, KA, and QA (L-Gh). The respective potent antagonists for these excitants are D-APV, 7-DGG, and GDEE. T o date the best defined are NMDA receptors, and they appear to be associated with the neurotoxic effects of NMDA and Ibo and also with some seizure states. I n addition, NMDA receptors are known to mediate release of other classical transmitters and activate Na+

138

NAJAM A . SHARIF

channels and receptor-coupled cyclases. Yet the receptors for ["INMDA have not been successfully radiolabeled. Although L[3H]Aspbinding sites have been characterized, NMDA is devoid of activity in competing for the radioligand. However, the recent development of radioreceptor assays for [3H]APV and [SH]APH (NMDA antagonists) will undoubtedly prove useful in this goal. KA and other heterocyclic analogs of Glu are powerful excitants and have considerable neurotoxic potencies. Many facets of the differential actions of KA (and their limited blockade) and the development of its toxic properties from those of both Glu and Asp are additional strong indications of the autonomous nature of its receptors. KA has proved a useful tool for selectively lesioning certain neuronal pools for delineating neuronal pathways. It has also provided a good animal model of HD since the pattern of degeneration and consequential behavior alterations resemble the changes seen in this disease. Although removal of proximal acidic AA-utilizing terminals has proved a sound protective mechanism against KA neurotoxicity in the rat, no organic compound has been successfully applied to the same end. In contrast, the neurotoxic effects of NMDA and Ibo are adequately attenuated by low doses of APV, APH, and D-cx-AA.Similarly, while quinolinate produces lesions similar to KA, another tryptophan metabolite, kynurenate, blocks these actions of quinolinate but not of KA. Authors of these reports suggest that quinolinate, a cyclic analog of Asp, may be responsible for initiating and/or propagating neuronal damage in a manner similar to that found in HD. These hypotheses obviously need following up. A start has been made in studying the binding characteristics of radiolabeled phosphonate antagonists, but it's too early to draw adequate conclusions from the early reports published thus far. However, although ["]APV and DL-[~H]APBapparently label NMDA and Glu (QA-type) receptors, these antagonists exhibit surprisingly high dissociation constants. In the case of ["H]APB a low-affinity presynaptic receptor template may be involved. Another important observation concerns the differentiation of APBsensitive, Ca2+/C1--dependent and -independent Glu receptors in brain membranes. T h e former have been confirmed in many laboratories, and interestingly [3H]APB binding shows a strong Ca2+ and C1- dependency. The importance of Ca2+has also been realized with respect to LTP and induction of Glu binding sites in viuo and in vitro by kindling and LTP. However, ultrasonication and lyophilization of membrane suspensions have been able to duplicate these changes, indicating some caution

RECEPTORS FOR NEUROACTIVE AMINO ACID TRANSMITTERS

139

in interpretation of such binding measurements in models of neuronal plasticity. An adjunct of receptor binding, namely autoradiography, has allowed the visualization of Glu and KA binding sites in the rat CNS. T h e results from such studies have provided further support for the differential localization of such receptors, and these can be assumed to be the postsynaptic markers of acidic AA innervations in the CNS, thus providing complementary information to classical lesion techniques. Other exciting advances in this field include (1) the detection and purification of an endogenous dipeptide (NAAG) containing acidic AAs, which has excitatory actions in vivo and a high potency at competing for [3H]Glu binding in vitro; (2) development of biochemical bioassays for acidic AAs such as 22Na+fluxes and stimulation of cyclic nucleotide levels; and (3) isolation and purification of a brain Glu-binding protein (GBP) and the recent production of antibodies to the latter. Similarly, new biochemical technology may aid in the ultimate determination of the identity of the natural, endogenous ligands operative at the so-called “acidic-AA-utilizing” nervous pathways. Future studies should address these questions and those of the cellular need for AA receptor heterogeneity, and whether this multiplicity exists at the molecular level or not. Selective deletion and isolation of the various receptor proteins may provide some answers. Since a putative Glu receptor protein has apparently been isolated, its reconstitution in an artificial lipid environment should be attempted next in order to study its functional properties. Moreover, the chemical composition and structure of the GBP can now be elucidated. And some day in the near future perhaps we will be able to unravel the genetic mechanism involved in its production. We have now entered yet another era of molecular neurobiology which will certainly help answer some of the outstanding questions raised above. Contemporary neurochemistry awaits these novel developments with much hope and anticipation.

VIII. An Additional Note

I wish to draw attention to the following papers which have been published recently. Thus, while the concept of BZD receptor heterogeneity has gained further credence from physical separation and autoradiographic studies (Lo et al., 1983), it has become apparent that perhaps

140

NA.JAM A. SHARIF

the low-affinity GABA receptors are the major species responsible for stimulating BZD binding (Burch et al., 1983).In addition, phosphorylation-induced inactivation of GABA-modulin (Wise et al., 1983) supports the possible physiological relevance of this endogenous polypeptide at GABA synapses where Ca2+ ions are now known to modulate GABA binding (Corda and Guidotti, 1983).A useful review of GABA/ BZD pharmacology has also appeared (Squires, 1984). In the excitatory AA field the progress has been more rapid. Thus, NMDA-insensitive ~ - [ ~ H l A binding sp to rat cord (Butcher and Roberts, 1984)exhibited many properties of similar sites previously characterized in rat cerebellum (Sharif and Roberts, 1981).T h e Asp recognition sites in the cord appeared enriched in synaptic preparations, exhibited a Kd of 1.4 f l ,and were very cryolabile. Denervation supersensitivity has been demonstrated for specific L-[~H]GIubinding (40-85% increase) in experimentally induced paraplegia in dogs (McBride et al., 1984)where a concomitant down-regulation of GABA receptors (in the same lumbosacral cord membranes) was also evident. The former results corroborate enhanced [3H]Glu binding seen in striatal membranes after decortication (Roberts et al., 1982b; Fig. 5). T h e other important findings concern the identification of the quinolinate excitotoxin in most mammals, including man (Wolfensberger et al., 1983),and the significant increase in its levels seen in cortical regions of aged rats (Moroni et al., 1984)-these data could be relevant to the possible etiology/consequence of the development of senile dementia as well as the previously proposed role of this endogenous compound in causing Huntington’s disease. Other relevent information on this subject may be found in the comprehensive review by Foster and Fagg (1984). Acknowledgments

I extend my sincere thanks to Dr. David R. Burt for his support and encouragement, and I gratefully acknowledge the skillful typing of Ms. Evelyn Elizabeth. References

Abalis, I. M., Eldefrawi, M. E., and Eldefrawi, A. T. (1983). Pest. Biochem. Physiol. 20, 3948. Altman, J. (1972).J . Comp. Neurol. 145, 328-339. Arbilla, S., Kamal, L., and Langer, S. 2. (1979).J. Physiol. (London) 159, 410-435. Awapara J., Landina, A. J., Fuerat, R., and Seale, B. (1950).j.Biol. Chem. 187, 35-39. Azanza, M. J., and Walker, R. J. (1975). Comp. Biochem. Physiol. C 50, 15-161. Bareis, D. L., Manganiello, V. C., Hirata, F., Vaughan, M., and Axelrod, J. (1983). Proc. Natl. Acad. Sci. U.S.A. 80, 2514-2518. Barlow, R. B. (1980). “QuantitativeAspects of Chemical Pharmacology.” Univ. Park Press, Baltimore, Maryland.

RECEPTORS FOR NEUROACTIVE AMINO ACID TRANSMITTERS

14 1

Barrow, M. V., Simpson, C. F., and Miller, E. J. (1974). 4. Rev. Biol. 49, 101-128. Baudry, M., and Lynch, G. (1979). Nature (London) 282, 748-4750. Baudry, M., and Lynch, G. (1980). Proc. Natl. Acad. Sci. V.S.A. 77, 2298-2302. Baudry, M., and Lynch, G . (198l).J. Neurochem. 36, 811-820. Baudry, M., Oliver, M., Creager, R., Wieraszko, A,, and Lynch, G. (1980).Life Scz. 27,325330. Baudry, M., Arst, D., Oliver, M., and Lynch, G. (1981a). Dev. Brain Res. 1, 37-48. Baudry, M., Smith, E., and Lynch, G. (1981b). Mol. Pharmacol. 20, 280-286. Baudry, M., Kramer, K., Fagni, L., Recasens, M., and Lynch, G. (1983a). Mol. Pharmacol. 24,222-228. Baudry, M., Kramer, K., and Lynch, G. (1983b). Mol. Pharmacol. 24, 229-234. Beaumont, K., Maurin, U., Reisine, T. D., Fields, J. Z., Spokes, E., Bird, E. D., and Yamamura, H. I. (1979). Life Sci. 24, 809-816. Bennett, J. P., Logan, W. J., and Snyder, S. H. (1972). Science 178, 997-999. Berdichevsky, E., Riveros, N., Sanchez-Amass, S., and Orrego, F. (1983).Neurosci. Lett. 33, 85-90. Biggio, G., and Guidotti, A. (1976). Brain Res. 107, 365-375. Biziere, K., Thompson, H., and Coyle, J. T. (1980). Brain Res. 183,421-433. Borbe, H. O., Muller, W. E., and Wollert, U. (1981). Brain Res. 205, 131-139. Bowery, N. G. (1983). TIPS 4, 329-330. Bowery, N. G., and Brown, D. A. (1974). Brt. J. P h a m c o l . 50, 205-218. Bowery, N. G., Collins, J. F., and Hill, R. D. (1976). Nature (London) 261, 601-604. Bowery, N. G., Doble, A,, Hill, D. R., Hudson, A. L., Turnbull, M.J., and Warrington, R. (1981). In “Amino Acid Transmitters” (F. V. DeFeudis and P. Mandel, eds.), pp. 333341. Raven, New York. Bowery, N. G., Hill, D. R.,and Hudson, A. L. (1983a). Br. J . Pharmacol. 78, 191-206. Bowery, N. G., Price, G. W., Turnbull, M. J., and Wilkin, G. P. (1983b). Br. J. Pharmacol. 79, 189P. Bowery, N. G.,Price, G. W., Turnbull, M. J., and Wilkin, G. P. (1983~). Br.J. Pharmacol. 79, 277P. Bowling, A. C., and DeLorenzo, P. J. (1982). Science 216, 1247-1251. Bradford, H. F., and Ward, H. K. (1976). Bruin Res. 110, 115-125. Burch, T. P., Thyagarajan, R., and Ticku, M. K. (1983). Mol. Pharmacol., 23, 52-59. Burt, D. R. (1978). In “Neurotransmitter Receptor Binding” (H. I. Yamamura, S. J., Enna, and M. J. Kuhar, eds.), pp. 41-55. Raven, New York. Burt, D. R., and Sharif, N. A. (1983). Handb. Neurochem. 6, 353-378. Butcher, S. P., and Roberts, P. J. (1984). Biochem. Pharmacol. 33, 2439-2446. Butcher, S. P., Collins, J. F., and Roberts, P. J. (1983). Br. J . Pharmacol. 80, 355-364. Buu, N. T., Puil, E., and Van Gelder, N. M. (1976). Gen. Pharmacol. 7, 5-14. Chang, H. H., and Michaelis, E. K. (1981).J. Biol. Chcm. 256, 10084-10087. Chang, H. H., and Michaelis, E. K. (1982). Biochim. Bi@hys. Acta 688, 285-294. Chang, L. R., Barnard, E. A,, Lo, M. M. S., and Dolly, J. 0. (1981). FEES Lett. 126, 309312. Changeux, J.-P., Kasai, M., Huchet, M., and Meunier, J. C. (1970). C.R. Acad. Sct. (Paris) 270,2864-2867. Cheung, W. Y. (1982). Sci. Am. 246,62-71. Chiu, T. H., and Rosenberg, H. C. (1983). TIPS 4, 348-350. Choi, D. W., Farb, D. H., and Fischbach, G. D. (1977). Nature (London) 269, 342-344. Chude, 0. (1979).J . Neurochem. 33, 621-629. Clark, A. J. (1926).J. Physiol. (London) 61, 531.

142

NAJAM A. SHARIF

Clark, A. J., ed. (1936).“Heffter’s Handbuch det experimentellen Pharmakologie,” Vol. 4. Springer-Verlag, Berlin, and New York. Collins, G. G. S., Anson, J., and Surtees, L. (1983).BrainRes. 265, 157-159. Coombs, J. S.,Eccles, J. C., and Fatt, P. (1955).J.Physiol. (London) 130,326-373. Corda, M. G.,and Guidotti, A. (1983).J.Neurochem. 41, 277-280. Cotman, C. W., Foster, A. C., and Lanthorn, T. (1981).Adu. Biochem. Psychopharmacol. 27, 1-28. Chyle, J. T. (lY83a). Handb. Psyrhopliuriiiacol. 17, 237-269. Coyle, J. T.(1983b).TINS 6, 1-2. Coyle, J. T.,and Schwarcz, R. (1976).Nature (London) 263, 244-246. Creese, I., and Sibley, D. R. (1981).Annu. Rev. Phannacol. Toxicol. 21, 357-391. Creese, I., and Snyder, S. H. (1977).Nature (London) 270, 180-182. Cross, A. J., and Waddington, J. L. (1981).J.Neurochem. 37, 321-324. Crossman, A. R., Latham, A., Patel, S., and Slater, P. (1983).Br. J . Pharmacol. 80, 39P. Croucher, M.J., Collins, J. F., and Meldrum, B. S. (1982).Science 216, 899-901. Cuatrecasas, P.,and Hollenberg, M. D. (1975).Biochem. Eiophys. Res. Commun. 62, 31-41. Curtis, D. R. (1963).Pharmacol. Rev. 15, 333-364. Curtis, D. R. (1964).In “Physical Techniques in Biological Research” (W. L. Nastuk, ed.), pp. 144-190. Academic Press, New York. Curtis, D. R., and Johnston, G. A. R. (1974).Ergebn. Physiol. Biol. Chem. Exp. Pharmakol. 69, 97- 188. Curtis, D. R., and Watkins, J. C. (1963).J.Physiol.(London) 166, 1-14. Curtis, D.R., and Watkins, J. C. (1964).J.Physiol. (London) 166,1-14. Curtis, D.R., and Watkins, J. C. (1965).Pharmacol. Rev. 17, 347-391. Curtis, D. R.,Phillis, J. W., and Watkins, J. C. (1960).J. Physiol. (London) 150, 656-682. Curtis, D.R.,Hosli, L., and Johnston, G. A. R. (1967).Nature (London)215, 1502-1503. Curtis, D.R.,Hosli, L., Johnston, G. A. R., and Johnston, I. H. (1968).Exp. Bruin Res. 5, 235-258. Dale, H. H. (1906).J. Physiol. (London) 34, 163. Davidoff, R. A., ed. (1983).“Handbook of Spinal Cord,” Vol. 1. Dekker, New York. Davies, J., and Watkins, J. C. (1977).Brain Res. 130, 364-368. Davies, J., Evans, R. H., Francis, A. A,, Jones, A. W., Smith, D. A. S., and Watkins, J. C. (1983).Neurochem. Res. 7, 1119-1133. De Barry, J., Vincendon, G., and Gdmbos, G . (1980).FEES Lett. 109, 175-179. De Belleroche, J. S., and Bradfoord, H. F. (1973).J.Neurochem. 21, 441-451. DeFeudis, F. V. (1973).Can. J . Pharmacol. 51, 873-878. DeFeudis, F. V. (1978).Cen. Phannacol. 9, 139-144. DeFeudis, F. V. (1979).Int. Rev. Neurobiol. 21, 129-216. DeFeudis, F. V. (1983).Gen. Pharmacol. 14, 313-319. DeFeudis, F. V., Fando, J. L., and Munoz, 0. L. M. (1977).Expenentia 33, 1068-1070. DeFeudis, F. V., Ossola, L., Schmitt, G., and Mandel, P. (1980).J.Neurochem. 34,845-849. DeRobertis, E., and Fiszer De Plazas, S. (1976).J.Neurochem. 26, 1237-1243. Di Lauro, A., Meek, J. L., and Costa, E. (1982).J.Neurochem. 38, 1261-1267. Dolphin, A. C. (1983).Neurosci. Lett. 39, 83-89. Dostrovksky, J. O., and Pomeranz, B. (1977).Neurosci. Lett. 4, 315-319. Duggan, A. W. (1974).Exp. Erain Res. 19, 522-525. Ehrlich, P.(1909).Ber. Dtsch. Chem. Ges. 42, 17. Elliott, K. A., Khan, R. T., Bilodeau, F., and Lovell, R. A. (1965).C a n . J . Biochem. 43,407416. Engel, W. K., Siddique, T., and Nicoloff, J. T. (1983).Lancet 2, 73-75.

RECEPTORS FOR NEUROACTIVE AMINO ACID TRANSMITTERS

143

Enna, S. J., and Gallager J. P. (1983). Int. Rev. Neurobiol. 24, 181-212. Enna, S. J., and Maggi, A. (1979). Life Sci. 24, 1727-1738. Enna, S. J., and Snyder, S. H. (1975). Bruin Res. 100, 81-97. Enna, S. J., and Snyder, S. H. (1976).J. Neurochem. 26, 221-224. Enna, S J., and Snyder, S. H. (1977). Mol. Pharmacol. 13, 442-453. Enna, S. J.. Kuhar, M. J., and Snyder, S. H. (1975). Brain Res. 93, 168-174. Enna, S. J., Bennett, J. P., Jr., Bylund, D. B., Snyder, S. H., Bird, E. D., and Iversen, L. L. (1976). Brain Res. 116, 531-537. Evans, R. H., and Watkins, J. C. (1981). Life Sci. 28, 1303-1308. Evans, R. H., Francis, A. A., and Watkins, J. C. (1977). Experientia 33,489-491. Fagg, G. E., and Foster, A. C. (1983). Neuroscience 9, 701-709. Fagg, G. E., and Lane, J. D. (1979). Neuroscience 4, 1015-1036. Fagg, G. E., Foster, A. C., Mena, E. E., and Cotman, C. W. (1982).J. Neurosci. 2,958-965. Fagg, G . E., Foster, A. C., Mena, E. E., and Cotman, C. W. (1983a). Eur. J . Pharmacol. 88, 105-1 10. Fagg, J. E., Mena, E. E., Monaghan, D. T., and Cotman, C. W. (1983b). Neurosci. Lett. 38, 157-162. Fagni, L., Baudry, M., and Lynch, G. (1983).J. Neurosci. 3, 1538-1546. Ferkany, J. W., and Coyle, J. T. (1983). Life Sci. 33, 1295-1305. Ferkany, J. W., Zaczek, R., and Coyle, J. T. (1982). Nature (London) 298, 757-759. Ferkany, J. W., Zaczek, R., Markl, A., and Coyle, J. T. (1983). SOC.Neurosci. Abstr. 9, 262. Ferrendelli, J. A., Blank, A. C., and Grooss, R. A. (1980). Brain Re$. 200, 93-103. Filbin, M. T., Lunt, G. G., and Donnellan, J. F. (1980). In “Receptors for Neurotransmitters, Hormones and Pheromones” (D. 8 . Sattelle, ed.), pp. 153-160. Elsevier, Amsterdam. Fiszer de Plazas, S., and De Robertis, E. (1967). Brain Res. 5, 31-44. Fiszer de Plazas, S., and De Robertis, E. (1976).J . Neurochem. 27, 889-894. Fonnum, F., Storm-Mathisen, J., and Walerg, F. (1970). Brain Res. 20, 259-275. Foster, A. C., and Fagg, G. E. (1984). Brain Res. Rev. 7, 103-164. Foster, A. C., and Roberts, P.J. (1978).J. Neurochem. 31, 1467-1477. Foster, A. C., and Schwarcz, R. (1983). SOC.Neurosci. Abstr. 9, 1185. Foster, A. C., Mena, E. E., Fagg, G. E., and Cotman, C. W. (1981a).J. Neurosci. 1,620-625. Foster, A. C., Mena, E. E., Monaghan, D. T., and Cotman, C. W. (1981b). Nature (London) 289,73-75. Foster, A. C., Fagg, G. E., Harris, E. W., and Cotman, C. W. (1982). Brain Res. 242,374377. Foster, G. A., and Roberts, P. J. (1980). Lfe Sci. 27, 1467-1477. Foster, G. A., and Roberts, P. J. (1981). Br. J . Pharmucol. 74, 723-729. Foster, G. A., Roberts, P. J., Rowlands, G. J., and Sharif, N. A. (1981). Br.J. Pharmacol. 70, 235P. Francis, A. A., Jones, A. W., and Watkins, J. C. (1980).J . Neurochem. 35, 1458-1460. Frank, K., and Fuortes, M. G. F. (1957).J . Physiol. (London) 16, 39-40. Furchgott, R. F. (1964). Annu. Rev. Phanacol. 4, 21-50. Fuxe, K., Roberts, P. J., and Schwarcz, R., eds. (1983). WennerCrenInt. Symp. Excitotoxins. 39. Garthwaite, J., and Balks, R. (1978). Nature (London) 275, 328-329. Ganong, A., Lanthorn, T. H., and Cotman, C. W. (1983). Brain Res. 273, 170-174. Gioguieff, M. F., Kemel, M. L., and Glowinski, J. (1977). Neurosci. Lett. 6, 73-78. Goldstein, A., Lowney, L. I., and Pal, B. K. (1971). Proc. Natl. Acad. Sci. U.S.A. 68, 17421747.

144

NAJAM A . SHARIF

Grab, D. J., Berzina, K., Cohen, R. S., and Seikevitz, P. (1979).J . B i d . Chem. 254, 86908696. Graham, D., Pfeiffer, F., and Betz, H. (1983). Eur. J. Biochem. 131, 519-525. Gration, K. A. F., Clark, R. B., and Usherwood, P. N. R. (1979). Neuropharmacology 18, 201-208. Greenamyre, J. T., Young, A. B., and Penney, J. B. (1983). Neurosci. Lett. 37, 155-160. Greengard, P. (1979). Fed. Proc., Fed. Am. Soc. Exp. Baal. 38, 2208-2217. Greenlee, D., and Olsen, R. W. (1979). Biochem. Biophys. Res. Commun. 88, 380-387. Guidotti, A., Toffano, G., and Costa, E. (1978). Nature (London) 275, 553-555. Haga, T., Ross, E. M., Anderson, H. J., and Gilrnan, A. G. (1977). Proc. Nutl. Acud. Sci. U.S.A. 74, 2016-2020. Haldernann, S., Huffman, R.D., Marshall, K. C., and McLennan, H. (1972). Bruin Res. 39, 4 19-425. Hall, J. G., Hicks, T. P., McLennan, H., Richardson, T. L., and Wheal, H. V. (1979).J. Physiol. (London) 286, 29-39. Halpain, S., Parsons, B., and Rainbow, T. C. (1983). Eur. J . Phannacol. 86, 313-314. Ham, N. S. (1974). In “Molecular and Quantum Pharmacology” (E. D. Bergmann and B. Pullman, eds.), pp. 261-268. Reidel, Dordrecht. Harnberger, A. C., Chiang, G. H., Sandoval, E., and Cotman, C. W. (1979). Brain Res. 168, 53 1-54 1. Hayashi, T. (1954). K e i J . Med. 3, 183-192. Head, K. A,, Tunnicliff, G., and Matheson, G. K. (1980). Can. J. Biochem. 58, 534-538. Henke, H., Beaudet, A., and Cuknod, M. (1981). Brain Res. 219, 95-101. Hertz, L. (1979). Prog. Neurobiol. 13, 277-323. Hill, D. R.,and Bowery, N. G. (1981). Nature (London) 290, 149-152. Hirach, J. D., and Margolis, F. L. (1979). Brain Res. 174, 81-94. Hollenberg, M. D., and Cuatrecasas, P. (1979). I n “The Receptors, a Comprehensive Treatise” (R.D. OBrien, ed.), pp. 193-214. Plenum, New York. Honork, T., Lauridsen, J., and Krogsgaard-Larsen, P. (1981). J. Neurochem. 36, 13021304. Honore, T., Lauridsen, J., and Krogsgaard-Larsen, P. (1982). J . Neurochem. 38, 173178. Hosli, L., and Hosli, E. (1978). Rev. Physiol. Biochem. Pharmacol. 81, 136-188. Hosli, J. S., Andres, P. F., and Hosli, E. (1976). Pflugers Arch. Physiol. 363, 43-48. Ishida, M., and Shinozaki, H. (1983). Brain Res. 283, 174-177. Iversen L. L., and Bloom, F. E. (1972). Bruin Res. 41, 131-143. Iversen, L. L., Iversen, S. P., and Snyder, S. H., eds. (1983). Handb. Psychopharmacol. 17. Iwata, H., Yarnagami, S., and Baba, A. (1982).J. Neurochem. 38, 135-141. Johnson, J. L. (1978). Prog. Neurobiol. 10, 155-202. Johnston, G. A. R . (1978). Annu. Rev. Phurmacol. Toxicol. 18, 269-289. Johnston, G. A. R. (1979). Adv. Biochem. Physiol. 27, 177-185. Johnston, G . A. R., and Kennedy, S. M. E. (1978). I n “Amino Acids as Chemical TransmitLen” (F. Fonnum, ed.), pp. 50-65. Plenum, New York. Johnston, G. A. R.,Curtis, D. R.,Davies,J., and McCulloch, R. M. (1968). Biochem. PharmaC O ~ .17, 2488-2489. Kahn, C . R. (1976).J. Cell Biol. 70, 261-286. Kanazawa, A., Kakimoto, Y., Nakajima, T., Shimizu, H., Takesada, M., and Sano, I. (1965). Biochim. Biophys. Acta 97, 460-464. Karbon, E. W., Duman, R., and Enna, S. J. (1983). Brain Res. 274, 393-396. Karobath. M., and Sperk, G. (1979). Proc. Nutl. Acud. Sci. U.S.A. 76, 1004-1006.

RECEPTORS FOR NEUROACTIVE AMINO ACID TRANSMITTERS

145

Karobath, M., Plecheta, P., Lippitsch, M., and Krogsgaard-Larsen, P. (1979). Nature (London) 278, 748-749. Kim, J. S., Kornhuber, H. H., Schmid-Burg, K., and Holzmuller, B. (1980). Neuroscience 20,239-244. Koerner, J. F., and Cotman, C. W. (1981). Bruin Res. 216, 192-198. Koerner, J. F., Johnson, R. L., Freund, R. K., Robinson, M. B., and Crooks, S. L. (1983). Brain Res. 121, 299-309. Koller, K. J., Zaczek, R.,and Coyle, J. T . (1983). Soc. Neurosci. Abstr. 9, 1184. Krnjevik, K., and Phillis, J. W. (1963).J. Physiol. (London) 165, 274-304. Krogsgaard-Larsen, P., Honore, T., Hansen, J. J., Curtis, D. R., and Lodge, D. (1980). Nature (London) 284, 64-66. Krogsgaard-Larsen, P., and Falch, E. (1981). Mol. Cell. Biochem. 38, 129-146. Kuhar, M. J. (1983). Hundb. Psychopharmacol. 17, 299-320. Langley, J. N. (1905).J. Physiol. (London) 33, 34. Langley, J. N. (1907).J. Physiol. (London) 36, 347. Larder, A. P., and McLennan, H. (1983). Neurosci. Lett. 32, 59-63. Leeb-Lundberg, F., Snowman, A., and Olsen. R. W. (1980). Proc. Nutl. Acad. Sci. U.S.A. 77, 7468-7472. Lin, W. Y., and Goodfriend, T. L. (1970). A m . J . Physiol. 218, 1319-1328. Lloyd, K. G., Shemen, L., and Hornykiewicz, 0. (1977). Bruin Res. 127, 269-278. Lloyd, K. G., Worms, P., Bossi, L., and Morselli, P. L. (1980).Proc. Epilepsy Int. Sympo., 12th pp. 301-326. Lloyd, K. G., De Montis, G., Javoy-Agid, F., Beaumont, K., Lowenthal, A., Constantinidis, J., and Agid, Y. (1983). I n “Molecular Pharmacology of Neurotransmitter Receptors” (T. Segawa et ul., eds.), pp. 233-238. Raven, New York. Lo, M. M., Strittmatter, S. M., and Snyder, S. H. (1983). Proc. Nutl. Acud. Sci. U.S.A. 79, 680-684. Logan, J. P., and Snyder, S. H. (1972). Bruin Res. 11, 199-2 12. Loh, H. H., Cho, T. M., Wu, Y. C., and Way, E. L. (1974). Life Scz. 14, 2231-2245. London, E. D., and Coyle, J. T. (1979a).Mol. Phurmacol. 15, 492-505. London, E. D., and Coyle, J. T. (1979b). Eur. J . Pharmacol. 56, 287-290. London, E. D., Klernm, N., and Coyle, J. T. (1980). Bruin Res. 192,463-476. Losj, S. J., Perry, T. L., and Hornykiewicz, 0. (1983). Brain Rex. 370-373. Luini, A,, Goldberg, O., and Teichberg, V. I. (1981). Proc. Nutl. Acud. Sci. U.S.A. 78,32503254. McBean, G. J., and Roberts, P. J. (1981). Nature (London) 291,593-594. McBride, W. J., Hall, P. V., Chernet, J. T., Patrick, J. T., and Shapiro, S. (1984). J . Neurochem. 42, 1625- 163 1 . McCulloch, R. M., Johnston, G. A. R.,Gaine, C;.J., and Curtis, D. R. (1974).Exp. Bruin Res. 21, 515-518. MacDonald, R. L., and Barker, J. L. (1978). Nature (London) 271, 563-564. McGeer, E. G., Olney, J. W., and McGeer, P. L., eds. (1978). “Kainic Acid as a Tool in Neurobiology.” Raven, New York. McLennan, H., Hicks, T. P., and Liu, J. R. (1982). Neuropharmacology 21, 549-554. Mallorga, P., Hamburg, M., Tdhnan,J. F., and Gallager, D. W. (1980). Neurophurmacology 19,405-408. Malouf, A. T., Schnaar, R. L., and Coyle, J . T. (1982). SOC.Neurosci. Abstr. 8, 560. Marangos, P. J., Campbell, I., and Cohen, R. M., eds. (1983). Methods Neurobiol. 1. Marsden, C. D. (1979). In “GABA-Neurotransmitters” (P. Krogsgaard-Larsen, J. ScheelKruger, and H. Kofod, eds.), pp. 207-295. Munksgdard, Copenhagen.

146

NAJAM A . SHAKIF

Matus, A., Pehling, G., and Wilkinson, D. (1981).J. Neurobiol. 12, 67-73. Meldrum, B. S. (1978). Lancet 2, 304-306. Meldrum, B. S. (1982). Clin. Neuropharmucol. 5, 293-316. Meldrum, B. S., Croucher, M. J., Czuczwar, S. J., Collins, J. F., Curry, K., Josheph, M., and Stone, T. W. (1983). Neuroscience 9, 925-930. Mewett, K. N., Oaks, D. J., Olverman, H. J., Smith, D. A. S., and Watkins, J. C. (1983). In “CNS Receptors-From Molecular Pharmacology to Behavior” (P. Mandel and F. V. DeFeudis, eds.), pp. 163-209. Raven, New York. Michaelis, E. K., Michaelis, M. L., and Boyarsky, L. L. (1974). Bzochim. Bzophys. Actu 367, 338-348. Michaelis, E. K., Michaelis, M. L., Chang, H. H., Grubbs, R. D., and Kuonen, D. R. (1981). Mol. Cell. Biochem. 38, 163-179. Michaelis, E. K., Michaelis, M. L., Stormann, T. M., Chittenden, W. L., and Grubbs, R. D. (1983).J. Neurochem. 40, 1742-1753. Mitchell, P. R., and Martin, I. L. (1978). Nature (London) 274, 904-905. Mitchell, C. K., and Redburn, D. A. (1982). Neurosci. Lett. 28, 241-246. Mohler, H., and Okada, T. (1978). Mol. Pharmacol. 14, 256-265. Molinoff, P. B., Wolfe, B. B., and Weiland, G. A. (1981). Life Sci. 29, 427-443. Monaghan, D. T., and Cotman, C. W. (1982). Brain Res. 281, 91-100. Moroni, F., Lombard;, G., Monetti, G., and Aldinio, C. (1984) Neurosci. Lett. 47, 51-55. Muhyaddin, M. S., Roberts, P. J., and Woodruff, G. N. (1983).Eur.J. Pharmacol. 92,Y-14. Nadler, J. V., and Cuthbertson, G. J. (1980). Brain Res. 195,45-56. Napias, C., Bergmann, M. O., Van Ness, P. C., Greenlee, D. V., and Olsen, R. W. (1980). Life Sci. 27, 1001-1011. Neal, M. J. (1971).J. Physiol. (London) 215, 103-117. Nielsen, M., Braestrup, C., and Squires, R. F. (1978). Brain Res. 141, 342-346. Nistri, A., and Constantini, A. (1979). Prog. Neurobiol. 21, 219-258. Olney, J. W. (1981). Adv. Biochem. Psychopharmacol. 27, 375-384. Olney, J. W., Ho, 0. L., and Rhee, V. (1971). Exp. Brain Res. 14, 61-76. Olney, J. W., Degubareff, T., and Labruyere, J. (1979). Life Sci. 25,537-540. Olsen, R. W. (1981).J . Neurochem. 37, 1-13. Olsen, R. W., Ban, M., and Miller, T. (1976). Brain Res. 102, 283-299. Olsen, R. W., Leeb-Lundberg, F., and Napias, C. (1980). Brain Res. Bull. SuHl. 25, 217221. Olverman, H. J., Jones, A. W., and Watkins, J-C. (1984). Nature (Londun) 307, 460462. Palacios, J. M., Young, W. S., and Kuhar, M. J. (1980). Proc. Nutl. Acad. Scz. U.S.A. 77,670674. Parsons, B., and Rainbow, T. C. (1983). Neurosci. Lett. 36, 9-12. Paton, W. D. M., and Rang, H. P. (1965). Proc. R. Sac. Ser. B 163, 1-44. Peroutka, S. J., U Prichard, D. C., Greenberg, D. A., and Snyder, S. H. (1977). NeuropharWW,CO~O~Y 16, 549-556. Peterson, D. W., Collins, J. F., and Bradford, H. F. (1983). Brain Res. 275, 169-172. Pfeiffer, F., and Betz, H. (1981). Bruin Res. 226, 273-279. Pfeiffer, F., Graham, D., and Betz, H. (1982).J. Bzol. Chem. 257, 9389-9393. Plaitakis, A,, Bed, S., and Yahr, M. (1982). Science 216, 193-196. Prasad, K. N., Nayak, M., Prasad, E. J., Cummings, S., and Pattisapu, K. (1980).Life Sci. 27, 2251-2259. Recasens, M., Varga, V., Nanopulos, D., Saadoun, F., Vincendon, G., and Benavides, J. (1982). Bruin Res. 239, 153-173.

RECEPTORS FOR NEUROACTIVE AMINO ACID TRANSMITTERS

147

Reisine, T. D., Fields, J. Z., Yarnamura, H. I., Bird, E., Spokes, E., Schreier, P. S., and Enna, S. J. (1977).Life Sci. 21, 335-344. Roberts, E., and Frankel, S. (1950).J. Biol. Chem. 187, 55-63. Roberts, P. J. (1974). Nature (London) 252, 399-401. Roberts, P. J., and Anderson, S. D. (1979).J. Neurochem. 32, 1539-1545. Roberts, P. J., and Sharif, N. A. (1978).Brain Res. 157, 391-395. Roberts, P. J., and Sharif, N. A. (1981).Adv. Biochem. Psychophamacol. 27, 295-305. Roberts, P. J., Storm-Mathisen,J., and Johnston, G. A. R., eds. (1981).“Glutamate: Transmitter in the Central Nervous System,” pp. 1-226. Wiley, New York. Roberts, P. J., Foster, G. A., Sharif, N. A., and Collins, J. F. (1982a). Brain Res. 238, 475479. Roberts, P.J., McBean, G. J., Sharif, N. A., and Thomas, E. M. (1982b).Brain Res. 235,8391. Rodbell, M. (1980).Nature (London) 284, 17-22. Roy, S., and Michaelis, E. K. (1983).SOC.Neurosci. Abstr. 9, 573. Sanderson, C., and Murphy, S. (1982).Dev. Brain Res. 2, 329-339. Sano, K., and Roberts, E. (1963).Biochem. Pharmacol. 12, 489-502. Sano, I., Kakirnoto, Y., Kanazawa, A., Nakajima, T., and Shimizu, H. (1966).J. Neurochem. 13, 711-719. Savage, D. D., Werling, L. L., Nadler, J. V., and McNamara, J. 0.(1983).Eur. J . Phamacol. 85,225-226. Scatton, B., and Leyrnan, J. (1982).Nature (London) 297, 422-424. Schwarcz, R., and Coyle, J. T. (1977).Life Sci. 20, 431-436. Schwarcz, R., Collins, J. F., and Parks, D. A. (1983a).Neurosci. Lett. 33, 85-90. Schwarcz, R., Whetsell, W. O., and Mangano, R. M. (1983b).Science 219, 316-318. Segal, M. (1977). Brain Res. 119,476-479. Sharif, N. A. (1984).Handb. Neurochem. 6, 239-259. Sharif, N. A., and Burt, D. R . (1984).J. Neurochem. 43, 742-746. Sharif, N. A., and Roberts, P. J. (1980).J . Neurochem. 34, 779-784. Sharif, N. A., and Roberts, P. J. (1981a).Brain Res. 211, 293-303. Sharif, N. A., and Roberts, P. J. (1981b).Biochem. Pharmacol. 30, 3019-3022. Sharif, N. A., and Roberts, P. J. (1984).Neurochem. Res. 9, 81-101. Shinozaki, H., and Konishi, S. (1970).Brain Res. 24, 368-371. Shinozaki, H., and Shibuya, I. (1974).Neuropharmacology 13, 1057-1065. Simrnonds, M. A. (1983). TINS 6,279-281. Simon, J. R., Contrera, J. F., and Kuhar, M. J. (1976).J. Neurochem. 26, 141-147. Skerrit, J. H., Willow, M., and Johnston, G. A. R. (1982).Neurosci. Lett. 29, 63-66. Skolnick, P., and Paul, S. M. (1982).Int. Rev. Neurobaol. 23, 103-138. Slevin, J. T., and Coyle, J. T. (1981).J. Newochem. 37, 531-533. Slevin, J. T., Collins, J. F., and Coyle, J. T. (1983). Brain Res. 265, 169- 172. Snodgrass, S. R. (1978). Nature (London) 273, 392-394. Snodgrass, S. R. (1979). SOC.Neurosci. Abstr. 5 , 572. Snodgrass, S. R. (1983). Handb. Psychopharmacol. 17, 167-239. Snyder, S. H. (1983). Handb. Psychopharmacol. 17, 1-12. Snyder, S. H., and Bennett, J. P. Jr. (1976).Annu. Rev. Physiol. 38, 153-175. Snyder, S. H., and Goodman, R. R. (1980).J. Neurochem. 35, 5-15. Snyder, S. H., and Yarnamura, H. 1. (1977).Arch. Gen. Psychiatry 34, 236-239. Snyder, S. H., Greenberg, D., and Yamamura, H. I. (1974).Arch. Gen. Psychiatry 31,58-61. Snyder, J. H., Pasternak, G. W., and Pert, C. B. (1975).Handb. Psychopharmacol. 5, 329360.

148

NAJAM A. SHAKIF

Spencer, J. (1976). Brain Res. 102, 91-101. Squires, R. F. (1984). In “Handbook ojNeurochemistly” (A. Lajtha, ed.) Vol. 6, pp. 261-306. Plenum, New York. Starke, K. (1981). Annu. Rev. Pharmacol. Toxicol. 21, 7-30. Stone, T. W., and Perkins, M. N. (1981). Eur. J. Phermacol. 72, 41 1-412. Stone, T. W., and Perkins, M. N. (1983). Brain Res. 331, 347-349. Storm-Mathisen, J., Leknes, A. K., Bore, A. T., Vaaland, J. L., Edminson, P., Haug, F. M. S., and Ottersen, 0. P. (1983). Nature (London) 301, 517-520. Sutherland, E. W., and Rall, T. W. (1960). Pharmacol. Rev. 12, 265-299. Tallman, J., Thomas, J. W., and Gallager, D. (1978). Nature (London) 274, 383-385. Teichberg, V. I., Goldberg, D., and Luini, A. (1981). Mol. Cell. Biochem. 393, 281-295. Ticku, M. K., and Olsen, R. W. (1978). LifeSci. 22, 1643-1651. Ticku, M. K., Huang, A., and Barker, J. L. (1980). Mol. Pharmacol. 17, 285-289. Toffano, G., Guidotti, A., and Costa, E. (1978). Proc. Natl. Acad. Sci. U.S.A. 75,4024-4028. Unnerstall, J. R., and Wamsley, J. K. (1983). Eur.J. Pharmacol. 86, 361-371. Usherwood, P. N. R. (1981). Adv. Biochem. Psychopharmacol. 27, 183-193. Valdes, F., and Orrego, F. (1975). Brain Res. 97, 277-290. Vargas, F., Greenbaum, L., and Costa, E. (1980). Neuropharmacology 19, 791-794. Vincent, S. R., and McGeer, E. G. (1980). Brain Res. 184, 99-108. Viveros, N., and Orrego, F. (1982). Brain Res. 236, 492-496. Waszczak, B. L., Hume C., and Walters, J. R. (1981). L f e Sci. 28, 2411-2420. Watkins, J. C. (1978). In “Kainic Acid as a Tool in Neurobiology” (E. G. McGeer, J. W. Olney, and P. L. McGeer, eds.), pp. 37-69. Raven, New York. Watkins, J. C. (1981). In “Glutamate: Transmitter in the Central Nervous System’’ (P. J . Roberts, J. Storm-Mathisen, and G . A. R. Johnston, eds.), pp. 1-24. Wiley, New York. Watkins, J. C., and Evans, R. H. (1981). Annu. Rev. Pharmacol. Toxicol. 21, 165-204. Weiland, G. A., and Molinoff, P. B. (1981). L f e Sci. 29, 313-330. Wenthold, R. J. ( 1 980). Brain Res. 190,293-297. Werling, L. L., and Nadler, J. V. (1982).J. Neurochem. 38, 1050-1062. Werling, L. L., and Nadler, J. V. (1983). Brain Res. 276, 344-347. Werling, L. L., Dorman, K. A., and Nadler, J. V. (1983).J. Neurochem. 41, 586-593. Werman, R. (1966). Comp. Biochem. Physiol. 18, 745-766. White, W. F., and Heller, A. H. (1982). Nature (London) 298, 655-657. White, W. F., and Snodgrass, S. R. (1983).J. Neurochem. 40, 1701-1708. Whittmore, S. R., Mena, E. E., Monaghan, D. T., and Cotman, C. W. (1983). Brain Res. 277,99-107. Williams, M., and Risley, E. A. (1979).J. Neurochem. 32, 713-718. Wilkin, G. P., Hudson, A. P., Hill, D. R., and Bowery,W. G. (1981). Nature (London) 294, 584-587. Wise, B. C., Guidotti, A., and Costa, E. (1983). Proc. Natl. Acad. Sci. U.S.A. 80, 886-890. Wolfensberger, M., Amsler, U., Cuenod, M., Foster, A. C., Whetsel, W. O., and Schwarcz, R. (1983). Neurosci. Lett. 41, 247-252. Wong, D. T., and Horng, J. S. (1977). Life Sci. 20, 445-452. Wood, J. D., and Peesker, S. J. (1975). J . Neurochem. 25, 277-282. Yamamoto, C., Yamashita, H., and Chujo, T. (1976). Nature (London) 262, 786-787. Yamamura, H. I., Enna, S. J., and Kuhar, M. J., eds. (1978). “Neurotransmitter Receptor Binding.” Raven, New York. Young, W. S., and Kuhar, M. J. (1979a). Nature (London) 280, 393-395. Young, W. S., and Kuhar, M. J. (1979b). Brain Res. 179, 255-270. Young, A. B., and Snyder, S. H. (1973). Proc. Natl. Acad. Sci. U.S.A. 70, 2832-2836.

RECEPTORS FOR NEUKOACTIVE AMINO ACID TRANSMITTERS

I49

Young, A. B., and Snyder, S. H. (1974a). Mol. Pharmacol. 10, 790-809. Young, A. B., and Snyder, S. H. (1974b). Proc. Natl. Acad. Sci. U.S.A. 71, 4002-4005. Young, A. B., and Snyder, S. H. (1974~ ). Proc. Natl. Acad. Sci. U.S.A. 71, 2246-2250. Zaczek, R., and Coyle, J. T . (1982). Neuropharmacology 21, 15-26. Zaczek, R., Koller, K., Cotter, R., Heller, D., and Coyle, J. T. (1983). Proc. Natl. Acad. Sci. U.S.A. 80, 1 1 16-1 119. Zarbin, M. A., Wamsley, J. K., and Kuhar, M. J. (1981).J. Neurosci. 1, 532-547. Zieglgansberger, W., and Puil, E. A. (1973). Exp. Brain Res. 17, 35-49. Zuking, S. R., Young, A. B., and Snyder, S. H. (1974). Proc. Natl. Acad. Sci. U.S.A. 71, 4802-4807. Zuking, S. R., Young, A. B., and Snyder, S. H. (1975). Proc. Natl. Acad. Sci. U.S.A. 83,525530.

NOTE ADDEDI N PROOF This addendum hopes to highlight some new and exciting developments in the AA transmitter field. First, a putative peptide precursor for the natural ligand of BZD receptors has been isolated (Guidotti et al., 1983) and found to have proconvulsant actions similar to the P-carbolines (Squires, 1984), some of which have inverse agonist effects and are anxiogenic. The close coupling of GABA-BZD sites has been reinforced by the downregulation of GABAergic activity upon chronic BZD treatment (Gallager et al., 1984), and this taken further in the purification ofthe GABA-BZD-CI- ionophore complex (Sigel el al., 1984). Second, novel molecular biologic studies have demonstrated the expression of functional GABA, Gly, Glu, and KA receptors in Xenopus oocyte membranes following injections of messenger ribonucleic acid isolated from rat (Houamed et al., 1984) and human (Gunderssen et al., 1984) brain. Monaghan et al. (1984) have apparenily identified four pharmacologically distinct [SHjGlubinding sites in rat hippocampus, and siniilar sites have been localized in postsynaptic densities (Fagg and Matus, 1983), indicating the association of Glu recognition sites with the terminal fields of niany acidic AA-releasing projections. A relatively new observation involves the ability of Ca2+ions to augment Glu binding only in the presence of CI(Mena et al., 1982), and that both I.-GIu and Q A have a similar CI--dependence for inhibiting high-affinity (Kd = 383 NM) ['HJNAAG binding to brain synaptic membranes (Koller and Coyle, 1984). In a similar vein a number of peptide spider venoms potently inhibit postsynaptic Glu-receptor interactions (Michaelis el al., 1984). While Chesselet (1984) has reviewed the literature on the nature of the presynaptic receptors mediating transmitter release from brain preparations, Snyder ( 1 983) has presented an overview for the postsynaptic receptors in the mammalian brain. APPENDEDREFERENCES

Chesselet, M.-F. (1984). Neuroscience 12, 347-375. Fagg, G . E., and Matus, A. I. (1983). SOC.Neurosci. A6str. 9, 344.9. Gallager, D. W., Lakoski, J., Gonsalves, S. F., and Rauch, S. L. (1984) Nature (London) 308, 74-76. Suidotti, A , , Forchetti, C. M., Corda, M. G., Korikel, D., Bennett, C. D., and Costa, E. (1983). Proc. Null. Acad. Scz. USA 80, 3531-3533. hnderssen, C. B., Miledi, R., and Parker, I. (1984). Nuture (London) 308, 421-424. 3ouamed. K. M., Bilbe, G., Smart, T. G., Constanti, A,, Brown, D. A., Barnard, E. A,, and Richards, B. M. (1984). Nature (London) 310, 318-321.

150

NAJAM A. SHARIF

Koller, K. J., and Coyle, J. ‘r.(1984). Eur. J . Pharmacol. 104, 193-194. Mena, E. E., Fagg, G. E., and Cotman, C. W. (1982). Bruzn R a . 243, 378-381. Michaelis, E. K., Galton, N., and Early, S. L. (1984). Pro. Nutl. Acud. Sci. USA 81, 55715574. Monaghan, D. T., Holets, V. K., Toy, D. W., and Cotinan, C. W. (1984). Nature (Loiidorr) 306, 176-178. Sigel, E., Stephenson, F. A., Mamalaki, C., and Barnard, E. A. (1984).J . B i d . Chem. 258, 6965-ti97 1 .

MUSCARINK RECEPTOR SUBTYPES IN THE CENTRAL NERVOUS SYSTEM By Wayne Hoss

Center for Brain Research University of Rochester School of Medicine and Dentistry Rochester, New York

and John Ellis

Department of Psychiatry Neuroscience Research Unit University of Vermont College of Medicine Burlington, Vermont

.............................................. Muscarinic Receptors ..................

III.

IV.

V.

VI. VII.

A, Methodology and B. Classical Antagonist Binding ..................... C. Agonist Binding.. . . . . . . . . D. Nonuniform Antagonist Binding E. Sulfhydryl and Disulfide Reagents . . . . . . . . . . . . . . . . . F. Metals and GTP . . . ............................ G. Affinity Labeling.. . . . . . . . . . . . . . . . . . . . . . . . . . Responses Elicited by A. Turnover of Phosphatidylinositol . . . . . . . B. Cyclic Nucleotides.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C. Ion Fluxes.. . . . . . . . . . . . . ....... D. Presynaptic Recep Relationships among Subpopulations and Responses . . . . . . . . . . . . . . . . . . A. Relationships between Subpopulations ........................ B. Relationships between S Regulation of Muscarinic Receptors . . . . . . . . . . . . . . . . . . . . . . . . A. Sensitization and Desensitization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B. Hormones . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C. Copper-Deficient Animals .............. D. Protein Phosphorylation. . . . . . . . . . . . . . Solubilization of Muscarinic Concluding Remarks . . . . . . ................ References . . . . . . . . . . . . . . . . . . . .................

151 153 156

165

172 175 177

182

191 192

1. Introduction

Acetylcholine is a transmitter in both the central and peripheral nervous systems. In a manner similar to many other hormones, the 151 INTERNATIONAL REVIEW OF NEUROBIOLOGY, VOL. 26

Copyright 6 1985 by Academic Press, Inc. All rights of reproduction in any form reserved. ISBN 0-12-366826-3

152

WAYNE IIOSS AND JOHN ELLIS

effects of acetylcholine are mediated by cell surface receptors that recognize and bind acetylcholine, subsequently initiating biochemical pathways that lead eventually to the cellular responses. These responses include both excitatory and inhibitory synaptic actions in the CNS (Crawford and Curtis, 1966; Stone, 1972), as well as secretion (Babkin, 1950; Lundquist et al., 1980) and muscle contraction (Parker, 1972; Rang, 1966) in the periphery. Receptors for acetylcholine are of two major types-nicotinic and muscarinic-differing in location, pharmacology, and associated responses (Dale, 1914). The muscarinic receptor, which is found primarily in the CNS and nonskeletal muscle, is the subject of this article. The focus is narrowed further in that this article will concentrate on muscarinic receptors in the CNS, mentioning peripheral systems only for the purpose of comparison. In addition to interneurons in, for example, the caudate nucleus and the cerebral cortex, there are a number of known cholinergic pathways in the brain. These include the septohippocampal pathway (Raisman, 1966; Mosko et al., 1973) as well as those from the nucleus basalis of Meynert to the frontal cortex (Johnston et al., 1979, 1981) and the pathway coursing through the habenula en route to the interpeduncular nucleus-ventral tegmental area (Fibiger, 1982). The profound psychotomimetic effects of the potent antimuscarinic agents in man (Abood and Biel, 1962; Abood, 1968), together with the involvement of cholinergic neurons (Davies and Maloney, 1976; Perry et al., 1977; Bartus et al., 1982), especially those originating in the nucleus basalis of Meynert (Whitehouse et al., 1982) in senile dementia of the Alzheimer type attest to the importance of the central muscarinic system for cognitive function and memory. A decline in the number of muscarinic receptors with aging has also been noted (Perry, 1980; Nordberg and Winblad, 1981). Lesioning studies have demonstrated the involvement of the septohippocampal pathway in representational memory employing a task in which rats are trained to alternate between two sources of food in a T-maze (Thomas and Brito, 1980). That muscarinic receptors are involved in the behavioral deficits produced by antimuscarinic agents is bolstered by the correlation between the affinity constants of the anticholinergics for muscarinic receptors and the behavioral disturbances elicited by these agents in animals (Baumgold et al., 1977). Comparable relationships exist for the peripheral systems comparing muscle contraction with affinity constants (Hulme et al., 1978; Yamamura and Snyder, 1974a). Recent experimental findings in a number of laboratories suggest that muscarinic receptors are heterogeneous. Both agonists and antago-

MUSCARINIC RECEPTOR SUBTYPES I N THE

CNS

153

nists distinguish between different subclasses of the receptor on the basis of affinity. Whether these subtypes are distinct entities, each linked to separate effector systems, or represent different functional or conformational states of the same receptor is a central issue. T h e evidence for multiple muscarinic receptors and their regulation is discussed in the following sections.

II. Characterization of Muscarinic Receptors

A. METHODOLOGY AND CRITERIA FOR RECEPTOR BINDING These issues have been discussed in detail elsewhere (Burt, 1978b; Cuatrecasas and Hollenberg, 1975), but will be briefly reviewed here because of the major role that binding studies have played in the development of the current view of the muscarinic receptor. It is important to stress that detail must be paid to possible artifacts that may be introduced by the various methods employed. Additionally, the criteria outlined below constitute necessary but not sufficient conditions to conclude that the binding site is a bona fide receptor. Furthermore, each criterion should be reconsidered when features of the assay, such as method, tissue preparation, or labeled ligand, are altered. The computer-assisted methods of analysis that are discussed below will be largely unproductive if insufficient attention is paid to these criteria, o r to possible artifacts of methodology. 1. Methods

Of the many methods that may be used to measure binding to particulate fractions (see Bennett, 1978), by far the most common are filtration and centrifugation. Filtration is, in many ways, the simpler of the two and places less stringent restrictions on the assay volume. On the other hand, the possibility that the labeled ligand may bind nonspecifically (or, worse yet, “specifically”; cf. Cuatrecasas and Hollenberg, 1976) to the filter itself must be rigorously excluded. Kinetic studies must be carried out to ensure that the ligand does not dissociate appreciably during the washing phase of the assay. Centrifugation assays may yield higher values of nonspecific binding, but seem to represent the better method overall due to the ease with which the concentration and stability of the unbound ligand can be determined (in the supernatant). Neither labeled nor unlabeled ligands should be degraded or taken up to any significant extent, or bound by soluble proteins or other factors.

154

WAYNE HOSS A N D JOIIN ELLIS

I n all binding assays, it is necessary to distinguish between specific binding and nonspecific binding. By definition, nonspecific binding is nonsaturable and of low affinity and should, therefore, be proportional to ligand concentration. In practice, nonspecific binding is determined as that which remains in the presence of a concentration of unlabeled ligand which occupies essentially all of the specific sites. In an extensive study of opiate receptor binding, Fischel and Medzihradsky (1981) pointed out possible pitfalls associated with using active and inactive isomers to define specific binding.

2. Criteria T h e total binding and nonspecific binding having been determined, the difference between these values is the specific binding. If this difference represents a finite population of receptors, the specific binding should saturate. The affinity of the binding can be estimated initially according to the concentration of ligand that occupies half of the receptors; for successful assays, this value is usually in or below the nanomolar range. More precisely, the affinities of labeled and unlabeled antagonists, as determined by binding assays, should agree with affinities determined by Schild analysis against responses that are pharmacologically well characterized. Ideally, comparisons of this type (between binding and response) should be made between assays conducted in identical preparations and under identical conditions, but this is not always possible. Extensive comparisons of the binding and response properties of classical muscarinic antagonists have revealed excellent agreement between the two measures of antagonist affinity (Snyder et al., 1975; Birdsall et al., 1977). Even when identical preparations can be employed, the binding and response properties of agonists may differ, due to possible nonlinear coupling between occupancy and response. This feature of agonists can be both disturbing and useful, as discussed in Section IV,B. The last criterion in this abbreviated list is that receptors should be found in tissues that exhibit the appropriate pharmacological response(s), but not in tissues that lack such responses. Many studies have demonstrated that the tissues which possess muscarinic receptors according to binding studies are those that respond to muscarinic stimulation.

3. Methods of Analysis Many methods are now available for the analysis of data collected in binding studies, some of which are cited here and below (Weiland and Molinoff, 1981; Molinoff et al., 1981). However, sophisticated computer software cannot in itself assign the model for any set of observations. What it can be used for is to choose between well-defined models, assum-

MUSCARINIC RECEPTOR SUBTYPES IN THE

CNS

155

ing that the data are free of artifact o r systematic error. The possibility of such systematic error must often be carefully examined in separate experiments, the design of which must vary from case to case. T h e effects of some common artifacts have been described recently on the basis of computer simulations (Munson, 1983). One possible source of systematic error relates to the degree of equilibrium that is achieved during the incubation. For this reason, as well as others, kinetic studies are invaluable. Once on- and off-rates have been determined, the time required to obtain a given degree of equilibrium can be readily estimated (McPherson and Zettner, 1975). It is important to note that the time to equilibrium is markedly dependent on ligand concentration. Further, the presence of competitive inhibitors can alter the time course of the approach to equilibrium (Aranyi, 1980; Ehlert et al., 1981a; Moltusky and Mahan, 1984). Thus, these factors must be taken into account and attainment of equilibrium should be verified empirically. Most paradigms assume that all ligands (labeled and unlabeled) are exposed to the receptor simultaneously; this is especially important when ligands of very high affinity are used (Rodbard et al., 1971). The most common assumption regarding the binding of ligands to the muscarinic receptor is that of competitive inhibition, and in general, this assumption has been validated. Nonetheless, competitivity cannot be assumed but should be tested directly. The discussion regarding gallamine (Section II,D) illustrates some of the complexities that may arise. A variety of user-friendly programs are now available for the analysis of complex competitive interactions (cf. Feldman, 1972; Munson and Rodbard, 1980; DeLean et al., 1982). These programs are commonly used to discern receptor heterogeneity based on direct and indirect binding assays. Subtypes may be defined by the use of selective labeled ligands or by the competition for the binding of a nonselective labeled ligand by an unlabeled selective ligand. In the muscarinic system, there are no ligands of sufficient selectivity to reliably label a single subpopulation of receptors. It is usually difficult to demonstrate saturability of radiolabeled ligands of moderate selectivity due to the presence of a low-affinity binding component; therefore, the fulfillment of the criteria for receptor binding (Section II,A,2) depends in part on comparison with indirect binding assays in which the unlabeled form of the selective ligand competes for the binding of a labeled, nonselective ligand (Birdsall et al., 1978). Furthermore, the binding of nonselective ligands (antagonists) can be correlated more definitively with affinities in response assays, and receptor subtypes are best defined in indirect binding assays in which the labeled ligand is nonselective (Molinoff et al., 1981; Minneman and Molinoff, 1980).

156

WAYNE HOSS AND JOHN ELLIS

The use of excessively high receptor concentrations may lead to problems of interpretation in both direct and indirect binding assays. When the receptor concentration exceeds one-tenth the dissociation constant of the labeled ligand, the resulting depletion of the free ligand concentration will affect estimates of the affinity of the ligand-receptor interaction (Chang et al., 1975). If the free ligand concentration can be accurately determined, however, true values of affinity will be obtained from direct binding assays (Chang et al., 1975). When indirect binding studies are performed, the observed affinity of the unlabeled ligand is a function of the concentration of both the labeled ligand and the receptor. The effect of the concentration of labeled ligand is readily accounted for (Cheng and Prusoff, 1973), while the effect of receptor concentration is not (Jacobs et al., 1975). Recently, however, a correction for the effect of receptor concentration has been published (Linden, 1982); computer methods may also correct for the depletion of radioligand (DeLean et al., 1982). Investigators should beware of the contribution of nonspecific binding to radioligand concentration; binding to filters, for example, occurs after the assay. In extreme conditions, errors in the specific activity of the radioligand may lead to erroneous estimates of depletion. Estimates of the depletion of high-affinity unlabeled ligands are even more difficult. Finally, similar artifacts may occur when response assays are conducted in small volumes; in some cases, the magnitude of the error involved can be assessed by performing binding assays under identical conditions. B. CLASSICAL ANTAGONIST BINDING The binding of classical muscarinic antagonists, in the CNS and in the periphery, satisfies the criteria outlined above (Section I1,A) for specific receptor binding. Extensive studies by many laboratories had indicated for many years that these antagonists bind to a single, homogeneous population of receptors with the same affinities that they exhibit in response assays. Thus, the numbers of receptor sites @Imax) determined by different labeled antagonists are in good agreement and different unlabeled competitors displace the same amount of binding of a given labeled ligand (Hulme et al., 1978). Antagonists have been found to bind to different regions of the CNS with the same affinities (Ellis and HOSS,1980; Birdsall et al., 1980) and also to bind to central and peripheral receptors with the same affinities, which agree with affinities obtained from response measurements (Beld et al., 1975; Snyder et al., 1975). Within the CNS, muscarinic receptor densities in a variety of

MUSCARINIC RECEPTOR SUBTYPES IN THE

CNS

157

species follow the following pattern: striatum > hippocampus = cerebral cortex > diencephalon = tectum > medulla-pons > cerebellar hemispheres. In the last few years, it has become apparent that the interactions of antagonists with muscarinic receptors are not as simple as the earlier studies indicated. It is now known that many of the classical antagonists express different affinities in different tissues and that, at least under some conditions, they can be shown to bind heterogeneously within a given tissue. Other (truly selective) antagonists bind heterogeneously both within and among tissues. These deviations from ideal mass-action behavior are discussed below (Section 11,D). However, it is important to emphasize the fact that under many normal conditions the classical antagonists define a single homogeneous population of receptors, within which agonists and the truly selective antagonists define their respective subpopulations of muscarinic receptors. As discussed in Section II,A, the use of a nonselective labeled ligand improves the feasibility of complicated binding assays involving selective unlabeled ligands.

C. AGONISTBINDING Under the same conditions that give rise to homogeneous binding curves for antagonists, the binding of agonists deviates from the form of a single site. Hill coefficients for the inhibition of the binding of labeled antagonists by agonists such as carbachol are considerably less than unity throughout the CNS (Birdsall et al., 1978, 1980; Ellis and Hoss, 1980, 1982). Alternative plots of such data show that log dose-occupancy curves are flattened relative to the mass-action isotherm or that Scatchard plots are concave-up. The binding of two labeled agonists, oxotremorine-M (Birdsall et aZ., 1978) and cis-methyldioxolane (Ehlert et al., 1980a), has been investigated. T h e binding observed in these studies at the highest ligand concentrations that are technically feasible is, in general, considerably less than that obtained from antagonist binding studies. Also, unlabeled agonists compete with greater affinity for the binding of labeled agonists than for the binding of labeled antagonists. Explanations for the binding characteristics of agonists have included (1) that the presence of agonists or antagonists shifts the proportion of receptors that are in the agonist or antagonist state (Snyder, 1975; Snyder and Bennett, 1976); (2) that increasing concentrations of agonist progressively desensitize the receptor (Birdsall et al., 1978); ( 3 ) that the interaction of agonists with the receptor is negatively cooperative (Birdsall et al., 1978); and (4)that there are multiple subpopulations of mus-

158

WAYNE HOSS A N D JOHN ELLIS

carinic receptors that are discerned by agonists but not by antagonists (Birdsall et al., 1978). Since the binding curves of agonists are stable over time, model 2 seems unlikely (Birdsall et al., 1978). T h e strongest support for model 4 over the others has come from studies in which a given proportion of receptors is blocked by irreversible or very slowly reversible antagonists in the presence or absence of agonist. The results of such experiments show that there are preformed subpopulations of receptors that do not interconvert under the conditions of the binding assays. Qualitatively similar results have been obtained for inuscarinic receptors derived from forebrain and brainstem regions (Birdsall Pt al., 1978; Ellis and Hoss, 1980). These studies illustrate that the choice of experimental design, rather than extensive curve fitting, can be the key to discrimination between complex models of receptor action. Thus, the shapes of the binding curves of agonists seem to be due to the presence of subpopulations of receptors which possess different affinities for agonists but homogeneous affinities for antagonists. The observation that agonist IC50 values (concentration that inhibits 50% of the binding of labeled antagonist) varied across brain regions (Aronstam et al., 1977, 1978a) suggested that these differences might be due to differing proportions of the subpopulations in the different regions. Analyses of the occupancy curves for agonists found that the binding in different regions of the brain could not be explained as representing different proportions of two subpopulations, and it was concluded that the brain must possess at least three different subpopulations (Ellis and Hoss, 1980). More detailed analysis of the binding of agonists demonstrated that the postulation of a third site significantly improved the agreement between the model and the data (compared to two sites) in the medullapons (Birdsall et al., 1980). Further, the same three sites were sufficient to describe the binding of agonists in the other regions. Sokolovsky et al. (1983) have pointed out that it is probably not possible at the present time to discriminate between models of great complexity, such as a model of three sites in comparison with a model of two sites plus cooperativity or isomerization. These authors suggested that most studies neither require nor justify the effort involved in substantiating a third site. T h e development of selective antagonists (Section I1,D) may generate fresh approaches to these questions. D. NONUNIFORM ANTAGONIST BINDING Initial reports of heterogeneity in the binding of muscarinic agonists, but not antagonists (Birdsall et al., 1978; Hulme et al., 1978), were cause for both excitement and disappointment. While these data suggested

MUSCARINIC RECEPTOR SUBTYPES IN T H E

CNS

1.59

that the many physiological and behavioral effects that are mediated by the muscarinic system might be related to distinct subpopulations, they did not hold much promise for the exploitation of this information. First, receptors are most usefully classified according to the affinities of antagonists, because of the difficulties that can be encountered in relating the binding data and response data of agonists (Burgen, 1979; Swillens and Dumont, 1980; Kenakin, 1983). Second, all of the agonists tested seemed to have the same order of affinities for the subpopulations (Birdsall et al., 1980). This feature led Birdsall et al. (1977) to propose that the receptor heterogeneity that is perceived by agonists is due to constraints that the coupling of effectors places on the properties of the activated state of a single receptor. If these constraints did not affect the ground state of the receptor, then the binding of antagonists would not be affected. Within the general guidelines of this theory, however, there are two possibilities for the existence of antagonist selectivities. First, there may be minor variations in the receptive site of the different subpopulations that are overshadowed by coupling constraints in the case of agonists and that are not discerned by classical antagonists, but may be discerned by novel antagonists. Second, the coupling constraints may induce minor variations in a common receptor site, such that a similar situation prevails. In either event, selective pharmacological manipulation of muscarinic subpopulations might be feasible and it is of considerable importance that various forms of selectivity can now be demonstrated for antagonists. 1. Selectivity between Tissues and Responses It has been noted above that years of study with classical muscarinic antagonists had indicated that the muscarinic receptors in different tissues were indistinguishable. In their study of heterogeneous antagonists, Fisher et al. (1976) pointed out that the earlier conclusion was the “correct and logical conclusion” from the data that these earlier studies had collected, and they suggested that studies with molecules of greater rigidity and lower affinity would be necessary to detect heterogeneity in the effects of antagonists. The rigid antagonists that this group has synthesized do indeed display widely differing affinities in antagonizing different muscarinic responses in vivo. Molinoff et al. (198 1) have pointed out that the determination of receptor subtypes by in vivo studies is complicated by problems of access barriers and metabolic effects that may vary from tissue to tissue. However, these reservations do not apply as strongly to comparisons within a tissue. Based on the potencies of these relatively low-affinity, rigid antagonists, Fisher et al. (1980a) have concluded that there are subtypes of central muscarinic receptors. Support for the existence of separate muscarinic receptors has also

160

WAYNE HOSS A N D JOHN ELLIS

come from studies of the response properties of antagonists in vitro. Although a previous report had found little evidence for differences between the muscarinic receptors of ileum, bronchial muscle, and iris (Barlow et al., 1972), studies which compared antagonist affinity constants of atrium to those of ileum did find such differences (Barlow et al., 1976, 1980). The authors noted that quaternary salts were more selective than their tertiary analogs. Other examples of tissue and response selectivity will be presented in succeeding sections. 2. Heterogeneous Binding Properties of Classical Antagonists As stated above, under many experimental conditions, especially those in which physiological buffers and salts are included, the equilibrium binding curves of classical antagonists follow the mass-action isotherm. However, recent studies have found the binding of even the classical antagonists to be more complex in several ways. These deviations from the mass-action formulation can be grouped under three separate headings: ( 1 ) conditions of low ionic strength, in which equilibrium binding curves suggest heterogeneity of receptors; (2) kinetic studies that indicate that there are multiple conformational states of the receptor-ligand complex (isomerization); and (3) studies that indicate complex cooperative interactions in the binding of antagonists. Thus, in the striatuni, Ehlert et al. (1981b) have found the binding of ["H]quinuclidinyl benzilate (["I-IJQNB) to be biphasic when assays were performed in 50 mM sodium HEPES, 10 mM MgC14, and 10 /.LM 5'guanylyl-[/3-y]-imidodiphosphate[Gpp(NH)p]. T h e additional presence of apomorphine ( 1 p M ) appeared to convert the low-affinity sites to high affinity, resulting in a homogeneous binding curve for [:'H]QNB. A somewhat similar phenomenon is observed in the heart, where the binding of N-["H]methylscopolamine (NMS) suggests the presence of two populations of receptors when assays are conducted in 20 nlM HEPES, in the absence of inorganic ions (Hulme et al., 1981a). T h e inclusion of 100 p M Gpp(NH)p converts the binding curve to that expected of a single, homogeneous, high-affinity site. The fact that these results are obtained in the absence of added Mg2+may be related to the observation that added Mg'+ is also not necessary for the modulation of agonist binding by guanyl nucleotides in heart tissue (Hulme et al., 1981b). A difference in the level of endogeneous membrane-bound Mg" may explain these differences and also be responsible for the opposite effects of low ionic strength on antagonist binding in the heart and cerebral cortex. In contrast to the effects described above for cardiac receptors, the a f h i t y of the cortical receptor for antagonists is increased in low ionic strength buffers, and homogeneity of binding is preserved (Birdsall et nl., 1979b).

MUSCARINIC RECEPTOR SUBTYPES IN THE

CNS

161

There are now a large number of reports to indicate that the binding of muscarinic antagonists involves an isomerization step, although binding at equilibrium is entirely in accord with mass-action kinetics. These findings are not paradoxical, as the properties of binding under equilibrium conditions are independent of the number of conformational states (Prinz, 1983). Support for the concept of isomerization then comes necessarily from detailed kinetic studies of on- and off-rates. Studies in many different tissues and with several different labeled ligands have found that dissociation of labeled antagonists from the receptor is not a monoexponential process, and that the dissociation curves change qualitatively, depending on the preceding time of association (Galper et al., 1977; Galper and Smith, 1978; Klein, 1980; Schimerlik and Searles, 1980; Kloog and Sokolovsky, 1978a,b; Jarv et al., 1979). Other studies have found that the association kinetics of antagonists are not compatible with a simple bimolecular process or that equilibrium dissociation constants do not agree with the ratio k J k 1 (see Sokolovsky et al., 1983). Similar findings concerning the binding of the /3-adrenergic antagonist [1251]iodohydroxybenzylpindolol (IHYP) have led to a similar model of receptor isomerization (Ross et al., 1977). Burgisser et al. (1981) noted that many of these studies involved the use of racemic mixtures of labeled ligands. They went on to show that the presence of the inactive form of the radiolabeled ligand can give rise to artifactually complex binding phenomena. Tolkovsky (1982) has examined a similar kinetics problem concerning the binding of [3H]etorphine to membranes from sheep caudate nucleus. He suggested that two independent sites were responsible, but also discussed the additional complexities introduced by the use of a racemic labeled ligand. Thus, while the bulk of evidence supports the possibility of isomerization in the binding of muscarinic antagonists, some past experiments might bear reinvestigation with the purified isomers [e.g., (-)["H]QNB] that are now available. Several studies have reported that, even under equilibrium conditions, the binding of classical muscarinic antagonists differs from the simple mass-action isotherm in buffers that approximate physiological ionic composition. On the basis of binding studies with several antagonists, Henis and Sokolovsky (1983) concluded that there are negatively cooperative interactions between muscarinic receptors in the adenohypophysis of the rat, but not in the medulla-pons or the cerebral cortex. On the other hand, Kloog and Sokolovsky (1978a) found curvilinear Scatchard plots for the binding of [3H]scopolamine and ["HIN-methyl4-piperidyl benzilate (4NMPB) to homogenates of whole mouse brain, which indicate either negative cooperativity or the presence of heterogeneous sites (Sokolovsky et al., 1983). Hedlund et al. (1980, 1982) have recently reported very complex patterns of binding for the classical mus-

162

WAYNE HOSS A N D JOHN ELLIS

carinic antagonist QNB and for 4NMPB in rat cortex. Plots of binding versus concentration of labeled ligand show two plateaus, a finding which requires the assumption of either positive cooperativity or complex interactions of negative and positive cooperativity (see Hedlund et al., 1982). It has been noted that the optimal buffer for demonstrating this phenomenon is Krebs buffer containing 5 mM HEPES (Hedlund et al., 1982). Markedly different results were obtained with 50 mM phate buffer, although the ionic strengths of the two buffers were equal (Hedlund, 1981).

3 . Selective Antagonists In 1980, Hammer et al. reported that the tricyclic compound pirenzepine displayed heterogeneous binding profiles within and among tissues, under conditions in which classical antagonists such as [3H]NMS bind homogeneously. Pirenzepine had previously been shown to antagonize muscarinic responses with widely varying affinities, depending on the location of the receptor (see Hammer, 1982), but the additional finding of heterogeneous binding underlined its importance as a prototypical selective antagonist. Pirenzepine binds with highest affinity in the forebrain, and with lesser affinities in the brainstem, heart, and ileum (Hammer et al., 1980). The recent availability of [3H]pirenzepine (Watson et al., 1982) adds another useful tool to the armament of receptorologists, but its relatively low affinity and selectivity may limit its utility (Section 11,A). Gallamine is a neuromuscular blocking agent that has been found to express antimuscarinic effects (Clark and Mitchelson, 1976; Bird and Aghajanian, 1976; Rathbun and Hamilton, 1970). In an attempt to screen selective muscarinic antagonists, we chose to investigate such unusual antimuscarinics. We agreed with the reasoning of Fisher et al. ( 1976) that drugs of greatest selectivity would have relatively low affinities, and felt that investigation of well-known antagonists would prove less fruitful. Gallamine competed more potently for the binding of [3H]QNB in the brainstem than in the forebrain, in a manner that suggested heterogeneity of binding sites in both regions (Ellis and Hoss, 1982). When sites that had low affinity for carbachol were selectively eliminated, the affinities of both gallamine and carbachol were increased, suggesting that the eliminated sites also had low affinity for gallamine. Furthermore, in the brainstem, the sites that were left after this treatment appeared to possess homogeneous affinity for gallamine, although not for carbachol (Ellis and Hoss, 1982). These findings, and the observation that [3H]QNBcould completely overcome the inhibition

MUSCARINIC RECEPTOR SUBTYPES IN THE

163

CNS

by gallamine, led us to conclude that the interaction between gallamine and [3H]QNB was competitive. A subsequent report by Stockton et al. (1983) presented convincing evidence, especially kinetic evidence, that gallamine's interaction with [3H]NMS is not competitive. The contrast between these two reports led us to compare the effects of gallamine on the binding properties of [3H]NMS and [3H]QNBin a single system (Ellis and Lenox, 1984). It can be seen in Table I that we were able to replicate the results of Stockton et al. (1983) when [3H]NMS was the labeled ligand. That is, the presence of 100 /.&I gallamine dramatically slows the rate of dissociation of [3H]NMS from the receptor; also, concentrations as low as 3 slow the rate of association of [3H]NMS with the receptor approximately 10-fold. At the same time, similar kinetic studies in the same membrane preparations did not find evidence of noncompetitive inhibition of the binding of [3H]QNB.At concentrations of 15 and 100 /.&I, gallamine did not significantly alter the kinetics of association or dissocation of [3H]QNB. Therefore, w e do not feel that noncompetitive interactions obscured the previously reported studies in which [3H]QNB was employed as the labeled TABLE I ["HINMS A N D ["HIQNB '1'0 MEMBRANES

EYFECIS OF G A L L A M t N E O N T H E B I N D I N G K i N F r l C s OF BRAIN

RAI

Percentage inhibitionb

Experimental conditions Association

["HINMS ( 1 IN) + 3 pM gallamine [3H]QNB (0.5 nM) + 15 p M gallamine (SH]QNB (5 IN) + 100 pA4 gallamine

0.52 ? 0.10 min 4.6 ? 0.35 niin 3.9 ? 0.4 min 4.5 2 0.7 min 0.28 2 0.04 min 0.17 .t 0.08 niin

36% -

58% 47%

Dissociation",'

["INMS + Q N B + QNB + gallaniine (100 p M ) [JH]QNB + Q N B + QNB + gallamine ( 1 5 p M ) + QNB + gallamine (100 p M )

12.2 2 0.5 87 ? 4 10.8 ? 0.4 11.3 It 0.5 11.7 2 0.4

min min hr hr hr

Half times (tin) were determined by rnonoexponeiitial fits. Percentage inhibition was calculated at 10 min for 5 nM [:'H]QNB and at 30 rnin for the other cases. Dissociation was initiated by the addition o f 1 p M unlabeled Q N U , with or without gallamine, as indicated. "

164

WAYNE HOSS A N D JOHN ELLIS

ligand (Ellis and HOSS,1982). One possible explanation for the data presented in Table I is that gallamine does bind allosterically, but that the binding of gallamine or QNB reduces the affinity of the other to such an extent that ternary complexes cannot be demonstrated. Such an interaction would be indistinguishable from true competivity. Another possible explanation is that gallamine may bind to an allosteric site as well as to the site to which classical antagonists bind. According to this scheme, QNB must be insensitive to the same allosteric interaction that affects the binding of NMS. It is not unlikely that the positively charged NMS might interact with the receptor in a different manner than does the very lipophilic QNB. Dunlap and Brown (1983) have suggested that gallamine interacts both competitively and allosterically with cardiac muscarinic receptors. Evaluation of this and other possibilities must await future studies. The atypical agonist 3-(M-chlorophenyl-carbamoyloxy)-2-butynyltrimethyl ammonium (McN-A-343) raises blood pressure and heart rate when applied systemically, apparently by stimulating ganglionic receptors with greater affinity than it expresses in heart or smooth muscle (Roszkowski, 1961; Hammer and Giachetti, 1982). It has been suggested that McN-A-343 may bind to the same allosteric site to which gallamine appears to bind in the heart, although it displaces the binding of [3H]NMS to cortical receptors by a competitive mechanism (Birdsall et al., 1983~).It is somewhat surprising that guanyl nucleotides regulate the binding of McN-A-343 in the heart, as they do that of typical, presumably competitive agonists (Birdsall et al., 1983~). It is also interesting that McN-A-343 has higher affinity for L sites than for H sites in the cortex (Birdsall and Hulme, 1983), while gallamine seems to possess an order of affinities similar to that of typical agonists (Ellis and HOSS, 1982). Despite the controversy over the nature of gallamine's interaction with the muscarinic receptor, there is agreement that it distinguishes receptor subtypes (Ellis and Hoss, 1982; Birdsall and Hulme, 1983; Birdsall et al., 1983d; Dunlap and Brown, 1983). In view of the paucity of antagonists that distinguish subpopulations of muscarinic receptors, gallamine will undoubtedly prove to be a useful tool in future characterizations of the muscarinic receptor.

A N D DISULFIDE REAGENTS E. SULFHYDRYL

Sulfhydryl alkylating reagents such as N-ethylmaleimide (NEM) and membrane oxidizing reagents such as 5,5'-dithiobis (2-nitrobenzoic acid)

MUSCARINIC RECEPTOR SUBTYPES I N THE

CNS

165

(DTNB) and potassium ferricyanide increase the affinity of agonists for the muscarinic receptor (Aronstam et al., 1978a; Aronstam and Eldefrawi, 1979a). The effect of NEM is seen as an increase in the number of high-affinity receptors (Aronstam et al., 1977). T h e presence of agonists but not antagonists increases the ability of NEM to enhance agonist binding, indicating that agonist binding results in the exposure of sequestered sulfhydryl groups. On the other hand, reducing agents, such as dithiothreitol (DTT) or 2-mercaptoethanol, decrease agonist affinity (Aronstam and Eldefrawi, 1979a). T h e effects of oxidizing and reducing agents appear, moreover, to be reversible (Aronstam and Eldefrawi, 1979a; Hedlund and Bartfai, 1979). The effects on agonist binding can be observed at concentrations of the alkylating, oxidizing, o r reducing reagents that have no effects on antagonist binding. In contrast, NEM decreased the affinity of oxotremorine in cardiac membranes (Harden et al., 1982). The sulfhydryl reagent p-chloromercuribenzoate(PCMB) inhibits both antagonist and agonist binding (Aronstam et al., 1978a). Pretreatment with NEM protects against the inhibition of antagonist but not agonist binding, suggesting the interaction of at least two different sulfhydryl groups with the muscarinic receptor. Recently, these findings have been corroborated and extended by a detailed study of the effects of PCMB on muscarinic receptors in the rat cerebral cortex (Birdsall et al., 1983a,b). In conclusion, these studies suggest that the state of membrane sulfhydryl and disulfide moieties may regulate muscarinic receptors. As discussed in detail below (Section V,C), Cu may regulate muscarinic receptors in uiuo by binding to membrane sulfhydryl groups.

F. METALSAND GTP Transition and heavy metals affect muscarinic receptors by inhibiting the binding of antagonists at higher concentrations (Aronstam et al., 1978a; Aronstam and Eldefrawi, 1979b) and by increasing the binding of agonists at lower concentrations that do not affect antagonist binding (Aronstam et al., 1978a). The effect on antagonist binding is reversible and competitive, with Hg2+having the greatest inhibitory potency (I& = lO-’M). Other metals, including Cu2+,Fe2+,and Pb2+,were much less effective, having IC50 values between lop5 and M. Increasing the availability of sulfhydryl groups did not affect the inhibition of antagonist binding by heavy metals. Thus, the inhibitory effects of metals at high concentrations seem to be due to a direct interaction with the recep-

166

WAYNE IIOSS AND JOHN ELLIS

tor that does not involve sulfhydryl groups. It is conceivable that inhibition of muscarinic cholinergic receptors may be one of the toxic effects of metals such as Hg2+ and Pb2+. At lower concentrations, metals such as Cu2+,Cd2+,Pb2+,and Zn2+ increased agonist binding without affecting antagonist binding (Aronstam et al., 1978a). Since there was no further increase in agonist binding after pretreatment with NEM, sulfhydryl groups are probably involved in the ability of lower concentrations of metals to increase agonist binding. There is a striking difference between the effect of Cu2+ on forebrain and brainstem receptors (Farrar and HOSS,1984). For example, 5 pM Cu2+ significantly inhibits QNB binding in the forebrain, but has almost no effect on the brainstem. Further, as shown in Fig. 1, 3 p M Cu greatly increases Carbachol (CCh) binding in the forebrain but not in the brainstem. Thus, Cu distinguishes between forebrain and brainstem receptors in vitro. The inclusion of 1 pM Cu2+,which had no effect on Q N B binding, increased the fraction of sites with high affinity for carbachol from 42 to 70% according to a two-site fit in forebrain (Farrar and Hoss, 1984). Thus, low concentrations of Cu2+can affect the distribution between high- and low-affinity agonist receptors in vitro. The effects 6C

40

-s v

m

20

0

1

I

I

5

10

15

Z

Cu ( u M ) FIG. 1. T h e effect of' increasing medium Cu on the displacement of 50 pM [SH]QNB by 5 and I pM carbamylcholine in forebrain ( 0 )and brainstem (A)preparations, respectively. The reduction in percentage Q N B bound (%B) indicates increased carbamylcholine binding. The data represent the means of three experiments each performed in triplicate.

MUSCARINIC RECEPTOR SUBTYPES IN THE

CNS

167

of Cu2+ on agonist binding were reversible with triethylenetetramine (Farrar and HOSS,1984), suggesting that the effects were not owing to some Cu-related oxidative or catalytic process. In contrast to the effects of Cu and other transition metals, the effects of alkali and alkaline earth metals appear to be absent or weak (Birdsall et al., 197913). Guanine nucleotide in the presence of Mg2+decreases the binding of agonists but not antagonists to muscarinic receptors (Sokolovsky et al., 1980))in brainstem but not in forebrain regions of mouse. In that study, guanosine triphosphate (GTP) induced an apparent interconversion between high- and low-affinity receptors. In a subsequent report, Gurwitz and Sokolovsky (1980) demonstrated that micromolar concentrations of GTP could reverse the increase in agonist affinity induced by 1 mM Mn2+,Ni2+,or Co2+. T h e GTP effect on agonist binding, which is weaker in brain than in heart (Berrie et al., 1979; Ehlert et al., 1980b), is presumably a reflection of the activity of the regulatory protein complex Ni, which binds GTP and mediates the coupling between receptors and the inhibition of adenylate cyclase Uakobs, 1979). The hydrolysis of GTP terminates the coupling between receptor occupancy and adenylate cyclase. Muscarinic receptor-induced inhibition of adenylate cyclase has been demonstrated in several tissues including NG108-15 cells (Nathanson et al., 1978) and rabbit heart Uakobs et al., 1979) in addition to rat brain (Olianas et al., 1983). In brain, the effect requires GTP. The muscarinic receptor-induced inhibition of adenylate cyclase is discussed in detail in Section II1,B. G. AFFINITY LABELING Affinity labeling of muscarinic receptors in brain has been accomplished by using nanomolar concentrations of tritiated propylbenzilylcholine mustard. Specificity is demonstrated by the ability of 1 p M atropine to inhibit the labeling. Subsequent solubilization and electrophoresis have given molecular weights of approximately 80,000 for guinea pig, rat, and frog brains (Birdsall et al., 1979a) or 75,000 for bovine caudate membranes (Ruess and Lieflander, 1979). More recently, Amitai and collaborators (1982) have synthesized two photoaffinity labels for muscarinic receptors based on the incorporation of an azido group into one of the phenyl rings of 3-quinuclidinyl benzilate and N-methyl-4-piperidyl benzilate (NMPB). T h e tritiated probes bound in a potent, specific, and reversible manner in the dark. Photolysis in the presence of nanomolar concentrations of a~ido-N-[~H]methyl-

168

WAYNE HOSS A N D

.JOHN ELLIS

4-piperidyl benzilate (a~ido[~H]NMpB) produced an irreversible labeling of receptors that was inhibited by atropine. Gel electrophoresis showed that the label was associated with a single protein of molecular weight 86,000 in the rat cerebral cortex. In a subsequent report (Avissar et al., 1983), the use of the azido[3H]NMPB probe showed that, in brainstem areas, specific binding occurs to a 180,000-Da protein in addition to the 86,000-Da species. Further, both proteins can be dissociated to 40,000-Da peptides by treatments that cleave ester bonds. Based on these findings, they propose that muscarinic receptors can exist as an 80,000-Da dimer of two 40,000-Da subunits joined by covalent bonds or a tetramer of 160,000-Da comprised of two dimers, and further, that the dimer corresponds to the low and the tetramer to the high agonist affinity state of the receptor. It is also possible that the higher molecular weight form involves the receptor associated with some other protein. It is of interest in this regard that agonists can induce the association between receptor and guanine nucleotide regulatory protein (see, for example, Smith and Linibird, 1981; Kilpatrick and Caron, 1983).

111. Responses Elicited by Muscarinic Activation

As mentioned in the Introduction, muscarinic systems mediate a wide variety of physiological and behavioral responses. This section, however, will be restricted to biochemical responses, with minor exceptions. By this, we mean those responses that can be demonstrated in vitro in tissue slices, cultured cells, or subcellular fractions. These responses are more compatible with the biochemical portions of this review that deal with receptor binding. More to the point, these responses are more readily related to binding parameters because they can be carried out under conditions that can also be used in binding studies. This advantage will be discussed below and in Section IV,B. A. TURNOVER OF PHOSPHATIDYLINOSITOL Stimulation of the turnover of phosphatidylinositol (PhI) can be elicited by a wide range of hormones in an equally large variety of physiological systems, including many regions of the brain (Michell and Kirk, 1981). On the basis of the close association of PhI turnover and calcium mobilization, coupled with the lack of calcium dependence in the PhI

MUSCARINIC RECEPTOR SUBTYPES IN THE

CNS

169

response (but see below), Michell(l975) has suggested that the turnover of PhI may be the stimulus that leads to a rise in intracellular Ca2+levels. Several mechanisms have been proposed to explain the apparent link between PhI metabolism and calcium mobilization. Most generally, the hydrolysis of PhI may result in local changes in membrane structure that lead to the opening of a calcium channel (Jones et al., 1982). A more specific version of this theory suggests that PhI 4,5-biphosphate (PhIPz), which has high affinity for Ca2+,may project into the calcium channel to bind and hold Ca2+.Cleavage of the inositol triphosphate would then allow an unimpeded flow of Ca2+ into the cell (Jones et al., 1982). Another suggestion is that phosphatidic acid (PhA), regenerated from the 1,2-diacylglycerol that is produced by the hydrolysis of PhI, may function as a Ca2+ ionophore (Putney et al., 1980). T h e potency of PhA in partitioning calcium into chloroform from aqueous solutions is roughly equal to that of the calcium ionophore A23187 (Putney et al., 1980). Addition of exogeneous PhA stimulates several responses that are associated with PhI turnover/calcium flux (Putney et al., 1980; Salmon and Honeyman, 1980). T h e recently discovered phospholipid-dependent protein kinase (C-kinase; Nishizuka, 1983) represents another possible transduction mechanism, since its activity is enhanced by 1,2-diacylglycerol. T h e diacylglycerol which results from cleavage of the head group of PhI may be the endogenous activator, since the presence of the fatty acids common to PhI confer the greatest activity. Activation of this CAMP-independent kinase could lead to the phosphorylation of membrane proteins and, consequently, to alterations in membrane permeability. Whether the association between PhI turnover and Ca2+ mobilization is as general as originally suggested (Michell, 1975) is a matter of some controversy (see Hawthorne, 1982; Michell, 1982; Cockcroft, 1981). Thus, in some systems the turnover of PhI is dependent to some extent on extracellular Ca2+(see Cockcroft et al., 1980; Cockcroft, 1981). For example, the acetylcholine-induced Phl effect in synaptosomes requires micromolar levels of extracellular Ca2+(Griffin et al., 1979), while the muscarinic stimulation of the formation of cGMP in brain slices requires Ca2+ in the millimolar range (Hanley and Iversen, 1978). I n pineal cells (Smith and Hauser, 198l), the adrenergic stimulation of PhI effect requires Ca2+only in the absence of 1 mM inositol. This suggests that a requirement for calcium may indicate the metabolic needs of the cell rather than a direct interaction of calcium in the initiation of PhI turnover. T h e concentrations of PhI, PhIP, and PhIP:, are determined by complex metabolic pathways (Jones et al., 1982) which may be sensitive to drastic fluctuations in Ca2+ levels. Therefore, while the finding

170

WAYNE HOSS A N D J O H N ELLIS

that PhI turnover is independent of external Ca2+ represents strong evidence against a role for Ca2+in the response, the converse finding is not so conclusive. Vasopressin has recently been reported to stimulate degradation of PhI in purified liver plasma membranes (Wallace et al., 1982). The use of such simplified systems may provide a more definitive answer to the question of precedence in PhI turnover/Ca2+mobilization. Whatever the mechanism, muscarinic agonists stimulate the labeling and/or breakdown of PhI in many systems, including sympathetic ganglia, parotid gland, heart, smooth muscle, and brain (Michell and Kirk, 1981). Table I1 demonstrates the muscarinic nature of the cholinergic stimulation of the turnover of PhI in slices of rat forebrain. The maximal stimulatory effect is approximately 100% and is blocked by scopolamine. Other studies have indicated that a stimulation of about 50% can be elicited in slices derived from the tectum (not shown). The doseresponse curve for the stimulation elicited by carbachol is shown in Fig. 2,along with the occupancy curve for carbachol, inferred from competition versus [SH]NMS. Several features of Fig. 2 deserve comment. T h e occupancy curve for carbachol is markedly flattened, as in membranes, while the PhI response can be adequately fitted by a one-site curve. The curve for the PhI response does not agree with either KH or K L (derived from a two-site fit to the carbachol binding curve). The affinity of carbachol is much lower than the value commonly found in membrane studies, as previously reported by Gilbert et al. (1979). Comparisons of binding and response curves of agonists are fraught with difficulties (see Section IV,A). However, we have detected neither desensitization nor a receptor reserve in studies to date and tentatively suggest that the PhI response of brain slices may be associated with a single population of muscarinic receptors, having an affinity between the values of K H and K L

Conditions

CPM/nmol inorganic phosphate in PhSlPhI spota

Control + 10 nlM carbachol + I @ scopolanririe + 10 nlM carhadiol + 1 pM scopolamine + 1 mM eallaniine

12.8 28.5 13.5 12.3 13.4

The labeling of phosphatidylserine (PIIS) is not affected by carbachol, so that the increase seen is due to Phl alone. The values shown are the averages of quadruplicate determinations. The SEM values were within 10%of the means. Experimental details are given in the legend to Fig. 2.

MUSCARINJC RECEPTOR SUBTYPES IN T H E

CNS

171

FIG.2. Comparison of carbachol-stimulated Phl turnover to the occupancy curve for CCh. f, the fraction of maximal response or occupancy. Tissue slices (150 x 150 pm) were prepared from the forebrains of male Sprague-Dawley rats and incubated in KrebsRinger bicarbonate buffer (KKB). T h e incorporation of ["PIP; into PhI in a period o f 4 5 niin (37%) was measured by extracting the phospholipids (Folch rt nl., 1957), separating them by thin layer chromatography (Skipsky et ul., 1964), and scraping the Phl region of the plate. Radioactivity was determined by Cerenkov counting and P, was determined later, as described by Plesums and Bunch (1971). Binding assays were performed with identical slices under identical conditions, by competition for the specific binding of 0.5 nM [SH]NMS. l ' h e affinity of ["HINMS was 0.83 nM (single-site kinetics). T h e dose-response curve for the PhI effect (A) is shown superimposed on the occupancy curve for carbachol ( 0 ) and mass-action binding curves froni a two-site fit of the binding data (41% highaffinity sites). All curves were drawn by computer and the response curve is based on a one-site model, ECSo = 37 pM. Analysis of the curves was performed by previously described methods (Ellis and Hoss, 1982).

obtained by a two-site fit. This site may be related to the H site of the three-site fit of Birdsall et al. (1980). Fisher et al. (1983) have recently reported similar comparisons of binding and response in synaptosomal preparations derived from cerebral cortex. In this synaptosomal study, the binding of carbachol occurred with higher affinity and the response with lower affinity, com-

172

WAYNE HOSS AND-JOHN ELLIS

pared to our data in slices (Fig. 2). The response curve agreed well with the law of mass action, with an affinity very near to that of the low affinity site (approximately 100 @) derived from a two-site fit of the binding data. Bethanecol, pilocarpine, arecoline, and oxotremorine were found to be partial agonists of low efficacy in the response assays and did not distinguish heterogeneity in binding assays (Fisher et al., 1983). This finding contrasts with the ability of oxotremorine to distinguish subpopulations in membrane studies (Ellis and HOSS,1980; Birdsall et al., 1980). The above studies (Fig. 2; Fisher et al., 1983) differ qualitatively from those of Michell et aE. (1976) in smooth muscle and pancreas, where the response curves for PhI turnover were found to be flattened and to follow the binding curves of agonists. However, since the binding and response curves in the latter report were not generated under identical conditions, the results are difficult to interpret. Furthermore, smooth muscle may exhibit densensitization in the binding of agonists (Young, 1974), while brain does not (Birdsall et al., 1978). In parotid acinar cells, Weiss and Putney (1981) found the dose-response curve for the PhI effect to be shifted to the left relative to the methacholine occupancy curve. This might suggest that a high-affinity subpopulation of receptors is involved o r that there is a receptor reserve associated with the response. Cohen et al. (1983) have reported that neurally derived cell lines (NlE-115 and NG108-15) exhibit a robust but somewhat variable Phl response to muscarinic agonists. The response was found to desensitize within 20 min in the presence of 1 mM carbachol, a result that has not generally been reported for other systems (Weiss and Putney, 1981; Miller, 1977; Kirk and Michell, 1981; Fisher and Agranoff, 1980).

B. CYCLICNUCLEOTIDES Agonists acting at muscarinic receptors stimulate guanylate cyclase activity in intact cells and inhibit adenylate cyclase in intact cells and cellfree systems, under appropriate conditions. Presumably, these cyclic nucleotides then serve as second messengers to modulate intracellular metabolism, especially the activity of protein kinases, leading eventually to alterations in the ionic conductance of the cell membrane. 1. Guanylate Cyclase Muscarinic agonists stimulate the activity of a guanylate cyclase in brain (Hanley and Iversen, 1978; Ferendelli et al., 1970) and in cultured

MUSCARINIC RECEPTOR SUBTYPES IN THE

CNS

173

cells derived from neural tissue (Richelson et al., 1978). Stimulation of the cultured cell line N1E-115 can lead to a 200-fold increase in the intracellular concentration of cGMP (Matsuzawa and Nirenberg, 1975). T h e response is dependent on the presence of physiological levels of extracellular Ca2+ (Schultz et al., 1973; Hanley and Iversen, 1978). It is likely that the guanylate cyclase involved is soluble, as the membranebound form is inhibited by calcium (Ferendelli et al., 1976). Also, sodium azide stimulates only the membrane-bound enzyme and this stimulation is linearly additive to that induced by oxotremorine (Hanley and Iversen, 1978). Thus, it might be assumed that muscarinic receptor activation leads to an influx of Ca2+which, in turn, stimulates the activity of guanylate cyclase. However, it is doubtful that intracellular levels of Ca2+ ever rise to a level sufficient to stimulate the cytosolic enzyme (Deguchi et al., 1983). Indeed, Snider et al. (1981) used aequorin to show that stimulation of N1E-115 cells by carbachol did not result in noticeable Ca2+ uptake. These findings suggest that there may be another endogeneous activator of cytosolic guanylate cyclase, and two such activators have been proposed. Snider and Richelson (1983) found that the guanylate cyclase activity of N 1E-115 cells was dramatically enhanced by thrombin. Based on a combination of direct and indirect evidence, these authors (Snider and Richelson, 1983; McKinney and Richelson, 1984) have suggested that an arachidonate hydroperoxide formed by hormone-stimulated, calcium-dependent pathways may be the endogeneous activator. Deguchi et al. (1983) isolated from rat brain an activator of soluble guanylate cyclase which was identified as L-arginine. D-Arginine was not effective, and the activation by (micromolar) L-arginine required the presence of low levels of Ca'+. The authors postulated that Ca2+ is mobilized by hormone action, possibly by the ionophore action of' phosphatidic acid, and that the combination of the mobilized Ca'+ and Larginine would be sufficient to activate the cyclase, although it was allowed that other factors may be involved (Deguchi et al., 1983). The kinetics of the guanylate cyclase response are complex. In brain slices and in cultured neuroblastoma (N 1E-115) cells, intracellular cGMP levels peak at 1-2 min and then decline rapidly (Hanley and Iversen, 1978; El-Fakahany and Richelson, 1980). The declining phase appears to be the result of a turning off of the stimulatory phase (McKinney and Richelson, 1984), and the action of phosphodiesterase is at least partly responsible for the reduction in cGMP levels (Matsuzawa and Nirenberg, 1975; Hanley and Iversen, 1978). T h e time course is very similar when phosphatidic acid (Deguchi et d.,1983) or Mn2+(El-Fakahany and Richelson, 1980) is the stimulant, which suggests that the turning-off process is not mediated by the receptor itself (McKinney and Richelson,

174

WAYNE HOSS A N D .JOHN ELLIS

1984). The time course of the decline of cGMP levels is very similar to that of desensitization (El-Fakahany and Richelson, 1980), which might indicate that the two processes are different manifestations of the same event.

2 . Adenylate Cyclase Muscarinic agonists inhibit the activity of the enzyme adenylate cyclase in intact cells and cell-free preparations derived from many sources, including heart Uakobs et al., 1979), neurally derived cell lines (Nathanson et al., 1978), and brain (Olianas et al., 1983). Hormone systems which inhibit adenylate cyclase share many of the properties of the P-adrenergic receptor-mediated activation of adenylate cyclase. T h e inhibitory system possesses three main components: a receptor, a GTPbinding regulatory protein, and the catalytic moiety (Cooper, 1982). The presence of GTP o r its stable analogues reduces the affinities of the inhibitory agonists, and GTP is required for the inhibition of adenylate cyclase activity. Stimulatory and inhibitory hormones have both been found to stimulate low-K, GTPase activities in a manner that suggests that GTP hydrolysis terminates the coupling between hormone and adenylate cyclase Uakobs et al., 1983b). A great deal of evidence now indicates that the GTP-regulatory protein involved in the inhibitory process is different from that which leads to stimulation, and the designations N, and N; have been given to the stimulatory and inhibitory sites, respectively. Much of this evidence has been summarized recently Uakobs et al., 1983b). Additionally, several very recent reports have shown that pertussis toxin selectively disrupts the coupling of inhibitory hormones to adenylate cyclase by an ADPribosylation, leaving the stimulatory system unaffected (Murayama and Ui, 1983; Kurose et al., 1983; Bokoch et al., 1983). Hydrolysis-resistant GTP analogs do not have completely parallel effects on stimulatory and inhibitory systems. These stable analogs lead to a persistent activation of adenylate cyclase in many cell types through interaction with N, (Ross and Gilman, 1980). Recent evidence suggests that in the cyc- variant of the S49 lymphoma cell line, which lacks a functional N,, the stable analogs inhibit adenylate cyclase activated, for example, by forskolin Uakobs, et al., 1983a). In light of the unique aspects of inhibitory systems, care must be taken in assuming that principles determined for stimulatory systems will necessarily apply to inhibitory systems. The muscarinic, opiate, and a-adrenergic receptors found on NG 108-15 cells inhibit adenylate cyclase in similar manners (Sabol and Nirenberg, 1979). The a-adrenergic response has been studied in greater detail, due in large part to the fact that there is a selective phar-

MUSCAKINIC RECEPTOR SUBTYPES IN THE

CNS

175

macology of a-receptors and that the inhibition of adenylate cyclase is mediated by the a2 subtype (Sabol and Nirenberg, 1979; Lenox et al., 1983). Indeed, there are interesting parallels between the adrenergic system (a1 plus a2) and the muscarinic system. That is, a1 receptors are associated with the turnover of the PhI and Cap+mobilization, while ap receptors mediate the inhibition of adenylate cyclase (Berridge, 1980). The binding of agonists to a2 receptors is sensitive to guanyl nucleotides, while the binding to a1 receptors is not (Lefkowitz et al., 1981). In the presence of sufficient GTP or Gpp(NH)p, Hill coefficients for the binding of agonists to a2 receptors are raised from about 0.6 to 1.O (Lenox et al., 1983; Tsai and Lefkowitz, 1979). The fact that guanyl nucleotides raise the Hill coefficients for muscarinic agonist binding in the brainstem and heart, but not quite to unity (Hulme et al., 1981b), may indicate the similar presence of a muscarinic subtype not regulated by GTP. C. IONFLUXES Muscarinic agonists modulate the firing rate of neurons via alterations in membrane potassium conductance. Sympathetic neurons can display both inhibitory and excitatory responses (Kuba and Koketsu, 1978). T h e decreases in firing rate in the heart (Vaughn-Williams, 1957; Giles and Noble, 1976) and in neurons (Ben-Ari et al., 1976; Horn and Dodd, 198 1) result from increases in potassium conductance. Increases in the firing rates of central and sympathetic neurons have been attributed to decreases in potassium conductance. At least two potassium channels may be involved in excitation. One is a voltage-sensitive channel, the M-channel, which opens as the cell depolarizes (Brown, 1984). The other channel is opened by increases in intracellular calcium levels (North and Tokimasa, 1984). Both channels serve to restore the resting membrane potential and inhibit repetitive firing. The action of muscarinic agonists is to close these channels and to increase membrane resistance, thereby rendering the neuron more sensitive to other excitatory inputs. This feature suggests that muscarinic synapses may serve a modulatory role in synaptic transmission. T h e slowness of muscarinic responses (Purves, 1976; Bolton, 1976) is also consistent with a modulatory role and suggests that second-messenger systems may be interposed between receptor activation and effects on conductance. Cyclic GMP has been suggested as a candidate for such a second messenger role, but there is disagreement over whether cGMP mimics the effects of muscarinic agonists on membrane conductance (Stone et al., 1975; Phillis et al., 1974; Hartzell, 1982).

176

WAYNE HOSS A N D JOHN ELLIS

D. PRESYNAPTIC RECEPTORS In the classical view of the neuron, receptors reside in the dendrites or soma of the postsynaptic cell. However, receptors are also found on nerve terminals, where they presumably modulate the release of transmitter; when these presynaptic receptors are sensitive to the transmitter whose release is modulated, they are considered autoreceptors (Coyle and Snyder, 1981; Starke, 1977). There is considerable evidence for the existence of muscarinic autoreceptors and presynaptic receptors. For example, muscarinic agonists inhibit the release of norepinephrine from peripheral (Fuder et al., 1982; Muscholl, 1980) and central (Westfall, 1974) fibers. Cholinergic agonists have also been shown to modulate the release of dopamine (DA) in striatal slices and synaptosomes (Westfall, 1974; De Belleroche and Bradford, 1978). Szerb (1977) has shown that the potentiation by muscarinic antagonists of the release of acetylcholine from cortical and hippocampal terminals is due to the blockade of muscarinic autoreceptors. The potentiation is greatest when the concentration of acetylcholine is artificially raised by the use of acetylcholinesterase inhibitors in the assay; however, a significant effect can also be demonstrated in the absence of cholinesterase inhibition, indicating that the autoreceptors limit the release of acetylcholine under physiological conditions. The hippocampus is an ideal region in which to study muscarinic autoreceptors because its extrinsic cholinergic innervation derives entirely from the septum via the fornix. Lesions of the septum o r fornix reduce levels of choline acetyltransferase to 5-10% of control levels and abolish muscarinic regulation of acetylcholine release (Kamiya et al., 1981; Szerb et al., 1977). However, it has been demonstrated repeatedly that such lesions do not significantly reduce the number of muscarinic receptors in the hippocampus, as would be expected if the cholinergic terminals that are lost possess muscarinic receptors (Yamamura and Snyder, 1974b; Dudai and Segal, 1978; Overstreet et al., 1980; Fisher et al., 1980b; Kamiya et al., 1981). There are several possible reasons for the failure to detect autoreceptors by binding assays. First, it has been suggested that the autoreceptors may have low affinity for the ligands that are used in the binding assays (Szerb et al., 1977); however, the discrepancy in affinities may be at least partly related to the comparison of slices to homogenates (Gilbert et al., 1979). Second, the number of autoreceptors may be very small by comparison with the number of postsynaptic receptors. Third, the reduction in autoreceptors may be obscured by an approximately equal increase in postsynaptic receptors (denervation supersensitivity). A recent study by McKinney and Coyle (1982) suggests a combination of the latter two reasons. Ablation of the

MUSCARINIC RECEPTOR SUBTYPES IN T H E

CNS

177

nucleus basalis of Meynert led to an acute decrease (14%)in the density of cortical muscarinic receptors, with a return to normal levels by 5 weeks postlesion. Binding studies have been more successful in confirming the existence of presynaptic muscarinic receptors on the terminals of catecholamine neurons. T h e administration of 6-hydroxydopamine, centrally and peripherally, has been shown to lead to a reduction in the binding of muscarinic ligands which is attributable to the destruction of catecholaniine terniinals (Sharma and Banerjee, 1978; Gurwitz et al., 1980; De Belleroche et ol., 1982). However, there is not unanimous agreement, as some studies have reported transient reductions (Kato et al., 1978) or no reduction (Story et al., 1979; cf. Muscholl, 1980) under similar conditions.

IV. Relationships among Subpopulations and Responses

Previous sections of this article have indicated that subpopulations of muscarinic receptors may be defined in several ways. These include the subpopulations that account for the binding of most agonists; the subpopulations that have differing affinities for nonclassical antagonists (especially pirenzepine) and atypical agonist (McN-A-343), both within and among tissues; and subpopulations that may be responsible for discrete responses. A pressing question is whether the subpopulations defined by these different means are related in simple ways. Some approaches that have been applied to this problem are described below. A. RELATIONSHIPS BETWEEN SUBPOPULATIONS

T h e natural distribution of subpopulations between tissues may provide circumstantial evidence for the equivalence or nonequivalence of subpopulations. Studies of the binding of agonists have suggested that the postulation of three sites (SH, H, L) can explain the agonist binding properties of all brain regions (Birdsall et al., 1980). However, the binding of the antagonist pirenzepine suggests the presence of heterogeneity in the cortex, but not in the brainstem (Hammer et al., 1980). Since there is a fairly even distribution of SH, H, and L sites in the brainstem, it is clear that there cannot be a one-to-one relationship between the subpopulations defined according to affinities for pirenzepine and those defined by agonists. However, the availability of [3H]pirenzepine and the

178

WAYNE HOSS A N D J O H N ELLIS

agonist [3H]oxotremorine-M has allowed further comparisons to be drawn within tissues. When binding assays are conducted with low concentrations of these ligands, those sites that have highest affinities are selectively labeled (Birdsall et al., 1978). Comparisons of such studies indicate that, in the cortex, the L sites possess high affinity for pirenzepine, while the SH and H sites have lower affinity for pirenzepine (Birdsall and Hulme, 1983). In view of the data described above, this relationship would not be expected to hold for the brainstem and may indicate a greater complexity than is allowed by the SH, H, L classification scheme. It may be that there are subsets of receptors, each with a different agonist affinity, within the subpopulations defined by pirenzepine (Birdsall et al., 1984). When muscarinic receptors are blocked by unlabeled, irreversible antagonists in the presence of low concentrations of the agonist carbachol, the subpopulation(s) with highest affinity for carbachol are selectively protected from blockade (Birdsall et al., 1978; Ellis and Hoss, 1980). Membrane preparations which have been pretreated in this manner, to enrich the proportion of sites with high affinity toward carbachol, also exhibit a greater affinity for gallamine, compared to untreated tissue (Ellis and HOSS,1982). This implies that gallamine has a different pattern of selectivity than pirenzepine, which would be expected to display lower affinity toward such a preparation (see above). The paradigm of selective protection is particularly useful, because it does not require the preparation of new labeled ligands. Robson and Kosterlitz (19’79) used a similar approach to characterize subpopulations of opiate receptors. B. RELATIONSHIPS BETWEEN SUBPOPULATIONS AND RESPONSES

Muscarinic receptors are linked to many different responses in many tissues and the ability to activate or selectively antagonize these responses would be of great therapeutic import. There is an obvious analogy to the benefits that have derived from the subdivision of adrenergic receptors (Weiner, 1980). T h e characterization of receptors may be facilitated by the study of in vitro systems, where problems of distribution and metabolism of drugs can most easily be monitored and controlled. An ideal preparation would allow the study of multiple responses, under conditions that are amenable to the conduct of binding assays. In this way, interassay differences in buffer composition, tissue source, and tissue preparation would be avoided altogether and the information obtained in binding assays could be used to best advantage. There are many instances where dose-response curves generated in intact tissue from

MUSCARINIC RECEPTOR SUBTYPES IN THE

CNS

179

one source have been compared to binding curves generated in subcellular preparations from another tissue, often performed in separate laboratories. It is our opinion that such comparisons introduce unnecessary complexities into an already difficult task. It is known that the binding curves of agonists and antagonists can be markedly influenced by buffer conditions (Hulme et al., 1981b; Birdsall et al., 1983d), tissue source (Ellis and Hoss, 1982; Birdsall and Hulme, 1983), and even the state of the tissue (i.e., intact versus cell-free systems; Gilbert et al., 1979). Even when such an ideal system can be utilized and metabolism or uptake of ligands minimized, a major problem remains in relating binding and response. This is that “there is no reason to suppose that there is a linear relationship between the proportion of receptors occupied by a drug and the size of response that is observed” (Stephenson, 1975). For this reason, receptors are best characterized in terms of the affinities of competitive antagonists, which, in response assays, are obtained from Schild plots (Arunlakshana and Schild, 1959; Furchgott, 1978; Tallarida, 1981). Because a competitive antagonist acts by preventing the access of agonists to the receptor, the affinity of such an antagonist should be independent of the choice of agonist used to stimulate the system and should be identical to that observed in binding assays, providing that binding and response are studied under identical conditions. However, when an agonist mediates similar (or interacting) responses at different receptors, complications may arise; such complexities cannot yet be ruled out for the muscarinic subpopulations, and the use of Schild analysis to evaluate the possibility should be kept in mind (Furchgott, 1978). Finally, it should be pointed out that despite the difficulties associated with defining receptors according to agonist affinities (Kenakin, 1983), the relationship between the binding and response curves of agonists can be particularly useful in investigating the nature of receptoreffector coupling (Swillens and Dumont, 1980). A large body of indirect evidence has been suggested to indicate that many muscarinic responses may be associated with the L-agonist subpopulation (Birdsall et al., 1977; Strange et al., 1977; Fisher et al., 1983; Hanley and Iversen, 1978), but many of these studies are based on intertissue comparisons and all rely on agonist affinities. McKinney et al. (1983) have recently reported that several agonists are much more potent in inhibiting adenylate cyclase than in stimulating guanylate cyclase in N1E-115 cells. Gil and Wolfe (1983) found the muscarinic stimulation of the turnover of PhI in parotid slices to be antagonized by lower concentrations of pirenzepine ( 16-fold) than were required to antagonize the muscarinic inhibition of adenylate cyclase in ventricular homogenates. T h e finding by the same authors that the classical antagonist

180

WAYNE HOSS AND JOHN ELLIS

atropine exhibited a fivefold selectivity in the same direction illustrates the difficulty of such intertissue comparisons. Nonetheless, the conclusions of McKinney et al. (1983, above) and Gil and Wolfe (1983, above) are consistent in that if the adenylate cyclase response is attributed to the high-affinity agonist subpopulation, the finding of Birdsall and Hulme (1983) that the high-affinity subpopulation has low affinity for pirenzepine leads to the expectation that pirenzepine should antagonize the adenylate cyclase response with low affinity. These results are in accord with the suggestion that the muscarinic system may be similar to the aadrenergic system, in which a1 receptors mediate calcium-related responses and ap receptors inhibit adenylate cyclase (Section 111,B).Physiological and biochemical studies (Hammer and Giachetti, 1982; Potter et al., 1984) have suggested that pirenzepine binds to M1 sites with high affinity in the forebrain and in ganglia and to My sites with low affinity in the brainstem, smooth muscle, and heart. The above tissue distributions are relative rather than absolute, but agree with the greater potency displayed by pirenzepine in antagonizing the stimulatory effects of muscarinic agonists in sympathetic ganglia, compared to its ability to block inhibitory muscarinic effects on the heart and smooth muscle (Hammer and Giachetti, 1982; Brown et al., 1981; Barlow d al., 1981). A receptor (subfclassification based on a single antagonist (i.e.,pirenzepine) faces a high risk that extrareceptor interactions may influence the classification. This problem emphasizes the importance of developing or discovering additional selective antagonists. As part of our interest in this area, we have exanlined studies in the literature concerning the antimuscarinic potencies of antidepressants and antipsychotics (Table 111). Potency in inhibiting the binding of ["H]QNB (Q.50)was determined by Snyder et al. (1974) in rat brain membranes, while potency in inhibiting the muscarinic stimulation of guanylate cyclase (cCs0) was determined by Richelson and Divinetz-Roniero (1977) in neuroblastoma cells. Because of the marked interassay differences, the values are useful only by comparison between the classes of drugs or relative to the classical antimuscarinic atropine. The two classes of drugs do not differ significantly by either of the individual measures (QS", cG50), but the ratio of potencies in the two assays does separate the two groups. The antidepressants are relatively more potent in the binding assay, while the antipsychotics are more potent in the guanylate cyclase assay ( p < 0.001, Table 111). One possible explanation for this finding is that the antipsychotics are selective for the subpopulation of receptors that is associated with the stimulation of guanylate cyclase, while the antidepressants are more potent at other sites. An alternative possibility is that the greater potency of the antipsychotics in the guanylate cyclase assay may be re-

MUSCARINIC RECEPTOR SUBTYPES IN THE

TABLE 111 ANTIMUSCARINIC POTENCIES OF VARIOUSAGENTS, EXPRESSED AS W H I C H INHIRIT T H E

Antimuscarinics QNB Atropine Scopolamine Antidepressants Amitriptyline Doxepine Imipramine Nortript ylirie Desipramine Antipsychotics Clozapine Thioridazine Promazine Chlorproniazine Fluphenazine Perphenazine Acetophenazine Haloperidol Trifluoperazine

CNS

THE

181

CONCENTRATIONS

MUSCARINICALLY s I I M U L A T E D FORMATION OF CcMP (CG50) OR T H E BINDING OF [yH]QNB(Qs,,) BY 50%

0.3

2

0.4

100 300 400 1,000 2,000

10 44 78 57 170

3 60 2,000 2,000 2,000 4,000 4,000 7,000 20,000

26 150 650 1,000 12,000 1 1,000 10,000 48,000 13,000

p > 0.1h

p > 0.Ib

5

10 6.8 5.1 18 12 0.115 0.4 3.1 2 0.17 0.36 0.4 0.15

1.5

p < 0.001b

“ T h e ratio cGS(JQ50 may reflect the selectivity of these drugs toward muscarinic subpopulations. To account for differences between the assays, atropine may be taken for comparison as an antagonist which does not distinguish subpopulations. The data are from the literature: Richelson and Divinetz-Romero (1977), neuroblastoma cells; Snyder et al. (1974), rat brain membranes. * Two-tailed t tests for differences between antipsychotic and antidepressant values.

lated to the inhibitory effects of phenothiazines and related compounds on calcium uptake in some systems (Hoss and Formaniak, 1984). T o investigate the first possibility, we have examined the binding characteristics of some of the drugs in Table 111. The niost interesting result to date is the similarity between the binding characteristics of clozapine and pirenzepine in the brainstem and forebrain (Table IV). There is a consistency between Tables I11 and IV in that the pirenzepine-like selectivity of clozapine (Table IV) suggests that it should antagonize calciumrelated responses with high affinity (see above), as it does (Table 111). However, it is also possible that clozapine and pirenzipine may have direct effects on calcium fluxes, due to their structural similarities to the

182

WAYNE HOSS AND JOHN ELLlS

TABLE IV BINDING PROPERTIES OF TYPICAL A N D ATYPICAL MUSCARINIC I.ICANDS".~ Forebrain Conipourid

nH

Atropine N-Methylscopolamine Carbachol Pirenzepine Clozapirie

0.94 1.w2 0.46 0.68 0.70

Brainstem

PI'& 8.6 9.4 5.5 7.0 7.6

PIC50

11H

8.9 9.4 7.0 6.2 6.8

0.96 0.97 0.44 0.98 0.92

a The binding of the unlabeled ligands to crude synaptic membrane preparations was determined by competition for 0.2 nM [3H]NMS by previously described methods (Ellis ) -log 1C50 and Hoss, 1980). Results are expressed in terms of Hill coefficients ( 7 ~ ) ~and

(pIG0). Kesults are averages of two to six experiments.

phenothiazines (see above). Such effects could complicate the evaluation of the relationship between subpopulation and response, because of the likelihood that calciuni-related and calcium-independent responses may be mediated by different muscarinic subpopulations (above). Experiments are underway to test the extrareceptor influences of niuscarinic hgdnds on calcium flux. In summary, some consensus is developing in favor of the M1, M2 subclassification scheme, analogous to the a1,a2 adrenergic system, in which M1 sites possess high affinity for pirenzepine and may be linked to calcium mobilization. As in the a adrenergic system, there may be multiple agonist states for one or more of the antagonist sites (Birdsall et al., 1984). However, it must be remembered that neither this scheme nor several others that have been proposed are yet supported by well-developed pharmacologies, so that all must be considered tentative (Birdsall and Hulme, 1983). V. Regulation of Muscarinic Receptors

A. SENSITIZATION AND DESENSITIZATION Although some controversy remains, especially regarding denervation supersensitivity, there is now general agreement that the muscarinic receptor responds-at least under some conditions-in a manner similar to other CNS receptors, becoming sensitized after denervation of afferent pathways or chronic exposure to antagonists and becoming

MUSCARINIC RECEPTOR SUBTYPES IN T H E

CNS

183

desensitized subsequent to stimulation or chronic agonist treatment. First considered here are results obtained with neuroblastoma-glioma hybrid cell lines. In the neuron-derived cell lines, extended activation of the receptor with muscarinic agonists produces a severe loss (up to 88%) of muscarinic receptors (Klein et al., 1979), resulting in a concomitant decrease in the ability of muscarinic agonists to activate PhI turnover (Siman and Klein, 1981) and inhibit adenylate cyclase (Klein et al., 1979). T h e agonist-induced decrease in the steady state number of receptors was owing to an increased rate of receptor degradation (Klein et al., 1979). Whereas antagonists inhibited the agonist-induced loss of receptors, they did not themselves directly affect receptor number. Inactivation of the muscarinic stimulation of cyclic GMP formation also occurs, albeit on a shorter time scale, that is, minutes instead of hours (Taylor and Richelson, 1979). There is evidence that this short-term effect on guanylate cyclase involves the activation-inactivation of calcium channels in the NIE-115 cell line (El-Fakahany and Richelson, 1980). In the CNS, denervation of cholinergic pathways results in a modest increase, if any, in the number of muscarinic receptors in the target tissue. Thus, Westlind et al. (1981) find a 20% increase in the dorsal hippocampus while Yamarnura and Snyder (1974b) find no change in the whole hippocampus after septa1 lesions in rats. Lesion of the nucleus basalis produced a 28% increase in the high agonist affinity site without a significant change in B,,,, in rat cortex (McKinney and Coyle, 1982). These findings can be compared with data from peripheral tissues showing a lack of denervation supersensitivity in rat sympathetic ganglia (Burt, 1978a) and cat iris (Sachs et al., 1979),but a robust response in rat salivary gland (Pimoule el al., 1980). In contrast to the denervation studies, a more consistent picture emerges from chronic treatments with cholinergic agonists and antagonists. Raising acetylcholine levels by chronic inhibition of acetylcholinesterase activity with di-isopropyl fluorophosphate caused a 32 % reduction in total muscarinic receptor level, owing to a 47% decrease in the low agonist affinity sites (McKinney and Coyle, 1982). These results corroborated the earlier findings of Gazit et al. (1979) using Tetram, which is an organophosphate inhibitor of acetylcholinesterase. Multiple infusions of carbamyl choline into the spinal cord of rats yielded a desensitization to the antinociceptive effect of the cholinergic agonist within 12 hr (Taylor et al., 1982). There was a concomitant loss of muscarinic receptors, reaching a value of 57% after the 12-hr period. .Chronic intravenous infusion of oxotremorine in mice produced not only a tolerance to a variety of physiological and behavioral parameters,

184

WAYNE HOSS AND JOHN ELLIS

but also a loss of muscarinic receptors in several brain regions (Marks et al., 1981). It is also of interest that tolerance was observed at doses that did not cause a loss of receptors, suggesting that mechanisms other than receptor regulation are also operative in viuo. T h e observed loss of muscarinic receptors with chronic oxotremorine treatment was extended to developing animals by Ben-Barak et al. (1981). Chronic atropine treatment resulted in a doubling of the number of high agonist affinity sites, together with a smaller increase in low affinity sites (McKinney and Coyle, 1982). Likewise, chronic scopolamine treatment (10 mg/kg once daily for 10 days) leads to modest increases in total number of muscarinic receptors in several brain regions, both during development and at maturity (Ben-Barak et al., 1981). We have recently demonstrated a behavioral tolerance to low doses of scopolamine using a working memory task (Messer et al., 1983). Lesioning studies have demonstrated the importance of the septohippocampal pathway for this behavior (Thomas and Brito, 1980). Similar results are obtained whether scopolamine is administered systemically or directly into the hippocampus. As shown in Fig. 3, animals drop to chance after the first injection, but develop tolerance to scopolamine within 2 days. In contrast to the other examples cited above for tolerance to muscarinic antagonists, there are no striking or apparent changes in muscarinic receptors when

sol.

1

2

3

4

5

6

7

Injections (ip)

FIG. 3. Effect of scopolamine on representational memory in a T-maze. Male hooded rats of the Long-Evans strain were trained to perform a nonmatching to sample task in a T-maze. After rats demonstrated proficiency (100%correct for three consecutive sessions), they were divided into two groups, one receiving 2 mg/kg daily ip injections of scopolamine ( 0 )and the other receiving saline vehicle (0) 15 niin before testing.

MUSCARINIC RECEPTOR SUBTYPES IN THE

CNS

185

u

O7 bfprnol)

6

5 4 -lOg(CCh)

3

FIG. 4. Binding of QNB and carbachol to niuscarinic receptors from the hippocampi of trained rats. (A) Scatchard plot used to estimate the values of K,i and B,,,,,, for the specific binding of [3H]I-QNB where b and f refer to bound and free QNB, respectively. Leastsquares analysis gave B,,,, = 2.43 pmol/mg protein and K d = 3.53 x lo-” M ; (B) T h e inhibition (i) of QNB binding by various concentrations of CCh. T h e synibols represent the experimental values and the line is calculated from the best fit to a two-site model. Data analysis as described elsewhere (Ellis and Hoss, 1980) gave ICan= 4.96 X M , K,, = 7.45 X lo-@M , K , = ~ 5.80 X M , and OL (fraction of receptors with high affinity) = 0.32. T h e binding assays were performed on a washed membrane fraction prepared from the hippocampi of trained rats sacrificed 24 hr after the last injection (see legend to Fig. 3).

they are examined in hippocampal homogenates after intraperitoneal injections (Fig. 4,Table V). Preliminary autoradiographic localization after intrahippocampal administration suggests that there may be some increase in receptor density, especially in the dentate gyrus (Fig. 5 ) . Whether there are more dramatic localized changes in subtype distribution must await further experimentation and quantitation of the autoradiograms.

B. HORMONES Although the primary mode of interaction between cholinergic and other transmitter systems appears to be at the cellular level, for example, cholinergic input from the septum onto glutamate-containing granule cells in the dentate and dopaminergic synapses onto cholinergic interneurons in the caudate nucleus, there is also evidence for a direct interaction between muscarinic receptors and other transmitters or neuro-

I86

WAYNE HOSS A N D .JOHN ELLIS

EFYECI. OF

Para meter QNB:

n,,, Kl ChrbdchOl IG,,, a

KH KL

TABLE V DAILYSCOPOLAMINE INJE(:.IlONS O N MUSCARINIC HIPPOCAMPUS OF TRAINED RATS''

h!XEYI.OKS IN 1'HE

Expet-imentals

*

2.36 0.27 p m o l / I i l g (5.31 -t 0.96) x I 0 V i 1 M

*

(4.33 0.37) X M 0.280 t 0.048 (8.05 t 3.5) x lo-" M (4.19 2 1 . 1 ) X 10F5M

Controls 2.00 t 0.27 p n d m g protein (5.03 _t 1.6) x 10-'1 M (3.10 2 0.301 (7.27 (4.21 -t

Values are i t l e a n s 2 SEM of five t o six exper-irnents, each Expcriiiierital details are giveii in the legend t o Fig. 4.

1.0) x 10-5M

* 0.01!) * 2.6) x 0.90)

X

lo-* M M

perforined

in tiiplicate.

modulators occurring at the level of the plasma membrane. These effects include steroid hormones in the pituitary and brain, dopamine in the brain, and vasoactive intestinal peptide in the periphery. Interestingly, the binding of tritiated classical antagonists in the adenohypophysis and preoptic area of the hypothalamus in rats and mice is heterogeneous, yielding curved Scatchard plots best fit by a two-site model (Avissar et al., 1981). Further, there is a sexual dimorphism in the binding characteristics, showing differences not only between males and females, but also among females at various stages of the estrous cycle. A potent and specific effect of P-estradiol and progesterone was observed only in the hypothalamic and pituitary areas mentioned above, resulting in a decrease in both the proportion of high agonist affinity receptors and the value of the high-affinity constant for oxotremorine (Sokolovsky et al., 1981). Antagonist binding was, however, unaffected by these hormones. These studies suggest an involvement of muscarinic receptors in the feedback mechanisms by which estrogens affect the release of gonadotropin. Under certain ionic conditions, [50 mM Na/HEPES buffer containing 10 RUW MgC12 and 10 mM Gpp(NH)p], the binding of ["H]QNB to muscarinic receptors in the striatum of rats was heterogeneous (Ehlert et al., 1981b). Apomorphine, and to a lesser extent dopamine and isoproterenol, enhanced the binding of [3H]QNB by increasing the fraction of receptors in the high-affinity antagonist form. The dopaminergic effect was inhibited by fluphenazine but not by haloperidol, suggesting that the effect was mediated by dopamine receptors of the D-1 subtype. There is now considerable support for the notion that vasoactive intestinal peptide (VIP), which is also found in the CNS (Giachetti et al.,

MUSCARINIC RECEPTOR SUBTYPES IN 'THE

CNS

FIG. 5. Autoradiographic localization of rnuscarinic receptors in trained rats. Rats were training as described in the legend to Fig. 3. (A) Scopolamine, 30 pg in O.$ pl, or (B) saline was injected bilaterally through guide cannulas aimed at the CA3 region of each hippocampus four times over a period of 12 days. Brains were lightly fixed by perfusion with 0.1% formaldehyde in phosphate buffer 24 hr following the last injection and 24-pm coronal sections cut on a cryostat microtome. The sections, which were mounted on microscope slides, were incubated with 4 nM ["Il-QNB for 1 hr and rinsed twice with buffer. The slides were subsequently exposed to X-ray film for 5 days at 4°C. The prints were made using the autoradiograms as negatives. Nonspecific binding, which was evaluated by including excess atropine with adjacent sections, was virtually absent. Scatchard analysis indicated that the total specific binding per section (80 fmol) was the same for both animals.

1977), potentiates the action of acetylcholine in the periphery (see for example, Ahren and Lundquist, 1982). Evidence that VIP increases the affinity of acethylcholine, as well as other cholinergic ligands, for muscarinic receptors in the cat salivary gland has been obtained by Lundberg et al. (1982).

I88

WAYNE IIOSS A N D J O H N ELLIS

C. COPPER-DEFICIENT ANIMALS Among the metals that were shown to affect muscarinic receptors in vitro, Cu2+was highly potent, showing effects well within the physiological range of Cu concentration in brain (see Section 11,F). In order to assess a possible role for Cu in vivo, the receptor was characterized in rats made Cu deficient by a dietary regimen (Farrar and Hoss, unpublished). As shown in Table VI, in forebrain regions there was a decrease in both the affinity of the receptors for [3H]1-QNB and the density of receptors in the Cu-deficient animals compared with control animals. Cu treatment in vitro of homogenates from deficient animals did not reverse the in vivo effects but rather decreased receptor occupancy and ligand affinity in a manner similar to Cu treatment of control homogenates. Interestingly, minimally deficient rats displayed very similar changes in receptor properties compared with more severely deficient animals. Minimal Cu deficiency produced robust effects on the binding of agonists, increasing ICSOand derived dissociation constants values (Table VII). The addition of Cu to the assay medium caused an apparent reversal of the in vivo Cu deficiency, decreasing ICSOand derived dissociation constants to values near those observed with homogenates from normal animals in the presence of Cu. In summary, Cu deficiency has dramatic effects on both receptor number and the binding of agonists to muscarinic receptors in the CNS, suggesting that Cu may have an endogeneous role in the regulation of the receptor. Since Cu2+is a sulfhydryl ligand with a propensity for forming stable square planar complexes, perhaps Cu participates in the dimer-tetramer equilibrium corresponding to low and high agonist affinity data, respectively, as suggested by Avissar and her associates (Avissar et al., 1983). TABLE VI ECFECTOF DIETARY C U DEFICIENCY ON RECEPTORNUMBER A N I ) K c , FOR <2NB I N R A T FOREBRAIN"

Animal Control Deficient Minimally deficient

Receptor number (pmol/mg protein)

(X

Kd lo-"' M )

CUh

(wp/g dry wt)

2.26 5 0.02 1.50 2 0.05

0.62 t 0.04 1.07 2 0.12

17.0 5.6

1.60 2 0.00

0.91 2 0.14

12.8 5 0.95

5 5

0.94 0.42

* Experimental details are similar to those given in the legend to Fig. 4. Values are the means SEM of three to four experiments, each performed in triplicate. * Data shown for cortex.

*

MUSCARINIC RECEPTOR SUBTYPES IN T H E

EFFECTOF

Control Minimally deficient

189

CNS

TABLE VII DEPICIENCY O N THE B I N D I N G OF (;ARBAMYLCHOI.INE M U S C A R I N I C R E C E P T o H S I N R A T FOREBRAIN"

DIETARY <;U

5.44

* 0.20

0.22

k

0.05

29.8

2

3.0

0.48

It

0.02

TO

41.0 2 6.9

0.41

* 0.01

* 8.2

0.41

* 0.02

118

Experimental details are similar to those given in the legend to Fig. 4. Values are the means 2 SEM of three to four experiments, each performed in triplicate. (I

D. PROTEINPHOSPHORYLATION Burgoyne (1980, 1981, 1983; Burgoyne and Pearce, 1981) has presented a variety of evidence that phosphorylation of proteins inhibits the specific binding of the muscarinic antagonist [3H]QNB by a reduction in receptor number. Further, the reduction in muscarinic receptors that occurs when primary cerebellar cultures are incubated with cholinergic agonists is accompanied by an increase in phosphorylation of several membrane proteins, one of which has a molecular weight similar to that of the component that binds the irreversible muscarinic antagonist PrBCM (Burgoyne and Pearce, 1981). In studies of membranes from rat brain, Burgoyne also concluded that subpopulations of muscarinic receptors, defiued by affinities for agonists, were differentially regulated by phosphorylation (Burgoyne, 1983). It is quite likely that muscarinic receptors are involved in the regulation of the phospholipid-dependent protein kinase (C-kinase) described by Nishizuka and his associates (Nishizuka, 1983; see Section 111,A). However, the possibility that Ckinase may be involved in the phosphorylation-induced decrease in muscarinic receptors has not been investigated to date. As described in Section V,A, down-regulation of muscarinic receptors, due to continuous agonist occupancy, appears to occur in two phases. In the first phase, a decrease in the ability to elicit a response (Richelson and El-Fakahanay, 1981; Burgoyne and Pearce, 1981) occurs over a period of minutes, while receptor binding sites are unaltered. Then, over a period of hours, the number of binding sites decreases (Shifrin and Klein, 1980; Gazit et al., 1979; Burgoyne and Pearce, 1981). This loss in binding capacity is accompanied by a decrease in the binding affinity of the agonist carbachol, with a concomitant loss in the ability of guanyl nucleotides to shift the agonist binding curve toward lower affinity (Chin et al., 1980). It is possible to accommodate much of the data regarding phosphorylation and down-regulation by postulating, speculatively, that phosphorylation

190

WAYNE HOSS A N D . J O I l N ELLIS

“marks” the receptors for degradation o r for some other irreversible alteration (possibly internalization). If the action of chronic agonist occupancy is envisioned as favoring kinase activity over an ongoing phosphatase activity, such an action would enhance the probability of degradation. This postulate would explain the findings that ( 1 ) the effects of phosphorylation on receptor loss are not reversed by phosphatase (Burgoyne, 1983); (2) peptides that are phosphorylated with a time course similar to that of receptor loss are smaller than the “intact” receptor (Burgoyne and Pearce, 1981); and (3) the return of receptor density to normal levels in culture, following removal of agonist, requires protein synthesis (Shifrin and Klein, 1980). Additionally, it would be possible that the phosphorylation per se might be responsible for short-term desensitization, as reported for the luteinizing hormone (Hunzicker-Dunn et al., 1979) and /3-adrenergic (Stadel et al., 1983) receptors. Sokolovsky (1984) has recently suggested that protein phosphorylation may play a role in the dimerization of muscarinic receptors (Section 11,G).

VI. Solubilization of Muscarinic Receptors

Successful solubilization of the receptor from brain has been achieved using digitonin (see for example, Aronstam et al., 1978b), sodium cholatehodium chloride (Carson, 1982), and the zwitterionic detergent CHAPS (Gavish and Sokolovsky, 1982). Increased stability of the solubilized receptor was realized using glycerol with digitonin (Baron, 1982) and lecithin with sodium cholate/sodium chloride (Hulme et al., 1983). In general, the solubilized receptors are homogeneous with respect to the binding of classical antagonists. Whereas digitonin-solubilized receptors from brain are also homogeneous with respect to agonist binding (Hurko, 1978; Baron, 1982), CHAPS-solubilized receptors may be heterogeneous as evidenced by flattened binding curves (Gavish and Sokolovsky, 1982), although these authors do not mention this point. Agonist binding to receptors solubilized with CHAPS is, moreover, decreased by GTP in a Na+-dependent manner (Kuno et al., 1983). It is also of interest that digitonin-solubilized receptors retain their response to low concentrations of Cu2+ (Baron, 1982), suggesting perhaps that the low-affinity agonist, form of the receptor has been preferentially solubilized.

MUSCARINIC RECEPTOR SUBTYPES IN THE

CNS

191

VII. Concluding Remarks

There is now compelling evidence that binding sites defined using tritiated classical muscarinic antagonists such as Q N B represent physiological receptors that mediate many of the actions of acetylcholine in brain as well as other tissues. That agonists and nonclassical antagonists distinguish different populations of the receptor on the basis of affinity is equally apparent. Studies of the binding of radioligands to homogenized tissue, together with autoradiographic analyses of brain sections, are providing detailed maps of muscarinic receptor distribution throughout the brain. The recent achievement of producing the first image of the receptor in a living person using computed tomography after injection of 4-[1231]iodoQNB(Eckelman et al., 1984) raises hope for the clinical assessment of the state of CNS muscarinic receptors in various neuropathological conditions. Muscarinic receptors are ubiquitous in the brain, in terms of receptor density and of their widespread involvement in behavioral phenomena (Karczmar, 1977). However, it is presently difficult to analyze specifically the cholinergic components of these processes, due to the diffuse nature of the cholinergic system. That is, cholinergic agents have such diverse effects that it is difficult to examine specific functions in isolation. This feature underscores the importance of the cholinergic system, but severely limits its experimental and therapeutic utility. For example, muscarinic antagonists continue to have a place in the treatment of Parkinson’s disease, owing to their lack of the adverse effects associated with L-DOPA therapy (Weiner, 1982). The nonselective antimuscarinics that are currently available exacerbate the dementia that is often associated with Parkinson’s disease (Barbeau, 1980; Ruberg P t al., 1982). It is possible that the therapeutic and adverse effects may be controlled by independent subpopulations of muscarinic receptors, and that the development of drugs with greater selectivity toward the muscarinic subpopulations will prove beneficial in this and other neurological disorders. However, the number and functions of the various subpopulations distinguished on the basis of differential affinities of some muscarinic ligands remains unclear. T w o current approaches-development of specific ligands, especially antagonists, and identification of the effector systems linked to each receptor subtype-hold promise for unraveling this complex system. T h e importance of carrying out binding studies under conditions identical to those used for biochemical assays requires

192

WAYNE HOSS A N D JOHN ELLIS

attention if correlations between receptor occupancy and function are to be meaningful. There may be an advantage in studying responses that are closely linked to agonist occupancy. In this regard, muscarinic stimulation of the breakdown of inositol phospholipids can be readily investigated in brain (Downes, 1982); recently, muscarinic and opiate-specific GTPase activities, presumably associated with the regulation of adenylate cyclase, have also been demonstrated in brain (Franklin and Hoss, 1984; Onali et ad., 1983). The studies of Fisher et al. (1980a) have indicated that the use of rigid antagonists of low affinity may be the key to detecting subtle differences in receptor structure or conformation. Although receptor subtypes are best defined according to the affinities of selective, competitive antagonists, events occurring beyond receptor occupation may contribute to the empirical selectivity of agonists. Thus, an agonist of low efficacy may initiate a full-blown response at a synapse that exhibits a large receptor reserve, but mainly antagonist properties where a receptor reserve is lacking. Such coupling phenomena probably explain muscarinic agonist-antagonist properties that have been reported recently in brain (Fisher et al., 1983; Nordstrom et al., 1983). The allosteric site that seems to bind gallanline and the agonist McN-A-343 may offer another means to adjust the gain between muscarinic binding and response. The synaptic environment continually subjects muscarinic receptors to the dynamic processes of sensitizationfdensensitization, upldown regulation, and couplingluncoupling. It seems likely that these regulatory mechanisms contribute to the development of tolerance and may complicate long-term drug regimens. The regulation of muscarinic receptors by transition metals, sulfhydryl reagents, and protein phosphorylation may provide insight into the molecular mechanisms involved in these processes. Subpopulations of muscarinic receptors may be regulated differentially (Smit et al., 1980; Chin et al., 1980; Burgoyne, 1983; Farrar and Hoss, 1984). T h e development of ligands with greater subpopulation selectivities, coupled with recent advances in affinity labeling and the solubilization of active receptors, can be expected to clarify further the interrelationships among muscarinic subpopulations.

References

Abood, L. G. (1968). In “Drugs Affecting the Central Nervous System” (A. Burger, ed.), pp. 127-165. Dekker, New York. Abood, L. G., and Biel, J. H. (1962). Int. Rev. Neurobiol. 4, 218-271. Ahren, B., and Lundquist, I . (1982). Experientza 38, 405-406.

MUSCARINIC RECEPTOR SUBTYPES IN THE

CNS

193

Amitai, G., Avissar, S., Balderman, D., and Sokolovsky, M. (1982). Proc. Natl. Acad. Sci. U.S.A. 79, 243-247. Aranyi, P. (1980). Biochem. Biophys. Acta 628, 220-227. Aronstam, R. S., and Eldefrawi, M. E. (1979a). Biochem. Pharmacol. 28, 701-703. Aronstam, R. S., and Eldefrawi, M. E. (197913). Toxicol. Appl. Pharmacol. 48, 489-496. Aronstam, R. S., Hoss, W., and Abood, L. G. (1977). Eur. J . Phurmacol. 46, 279-282. Aronstam, R. S., Abood, L. G., and Hoss, W. (1978a). Mol. Phanacol. 14,575-586. Aronstam, R. S., Schnessler, D. C., and Eldefrawi, M. E. (197813). Life Sci. 23, 1377-1382. Arunlakshana, O., and Schild, H. 0. (1959). Br. J . Pharmacol. Chemother. 14, 45-58. Avissar, S., Egozi, Y., and Sokolovsky, M. (1981). Neuroendocrinology 32, 303-309. Avissar, S., Amitai, G., and Sokolovsky, M. (1983). Proc. Natl. Acad. Sci. U.S.A. 80, 156-159. Babkin, B. P. (1950). “Secretory Mechanism of the Digestive Glands,” 2nd Ed., Hoeber, New York. Barbeau, A. (1980). J. Neural Trans. Suppl. 16, 187-193. Barlow, R. B., Franks, F. M., and Pearson, J. D. M. (1972). Br. J . Pharmacol. 46, 300-314. Barlow, R. B., Berry, K. J., Glenton, P. A. M., Nikolau, N. M., and Soh, K. S. (1976). Br. J . Pharmacol. 58, 613-620. Barlow, R. B., Burston, K. N., and Vis, A. (1980). B r . J . Pharmacol. 68, 141P-142P. Barlow, R. B., Caulfield, M. P., Kitchen, R., Roberts, P. M., and Stubley J. K. (1981). Br. J . Pharmacol. 73, 1838. Baron, B. (1982). Ph.D. thesis, Center for Brain Research, University of Rochester School of Medicine and Dentistry. Bartus, R. T., Dean, R. L., Beer, B., and Lippa, A. S. (1982). Science 217,408-417. Baumgold, J., Abood, L. G., and Aronstam, R. (1977). Brain Res. 124, 331-340. Beld, A. J., Van den Hoven, S., Wouterse, A. C., and Zegers, M. A. P. (1975). Eur. J. Phanacol. 30, 360-363. Ben-Ari, Y.,Dingledine, R., Kanazawa I., and Kelley, J. S. (1976).J . Physio2. (London) 261, 647-661. Ben-Barak, J., Gazit, H., Silman, I., and Dudai, Y. (1981). Eur. J. Phanacol. 74, 73-81. Bennett, J. P. (1978). In “Neurotransmitter Receptor Binding” (H. I. Yamamura, S. J. Enna, and J. Kuhar, eds.), pp. 57-90. Raven, New York. Berridge, M. J. (1980). Trendc Pharmacol. Sci. 1, 419-424. Berrie, C. P., Birdsall, N. J. M., Burgen, A. S. V., and Hulme, E. C. (1979). Biochem. Biophys. Res. Commun. 87, 1000-1005. Bird, S. J., and Aghajanian, G. K. (1976). Neuropharmacology 15, 273-282. Birdsall, N. J. M., and Hulme, E. C. (1983). Trends Pharmacol. Sci. 4, 459-463. Birdsall, N. J. M., Burgen, A. S. V., and Hulme, E. C. (1977). Adu. Behav. Biol. 24,25-33. Birdsall, N . J. M., Burgen, A. S. V., and Hulme, E. C. (1978). M o l . Phanacol. 14,723-736. Birdsall, N. J. M., Burgen, A. S. V., and Hulme, E. C. (1979a). Br. J . Pharmacol. 66, 337342. Birdsall, N. J. M., Burgen, A. S. V., Hulme, E. C., and Wells, J. W. (1979b). B r . J . PhunnaC O ~ .67, 371-377. Birdsall, N. J. M., Hulme, E. C., and Burgen, A. S. V. (1980). Proc. R. SOC.London Biol. 207, 1-12. Birdsall, N. J. M., Burgen, A. S. V., Hulme, E. C., and Wong, E. H. F. (1983a). Br. J . Pharmacol. 80, 187-196. Birdsall, N. J. M., Burgen, A. S. V., Hulme, E. C., and Wong, E. H. F. (1983b). Br. J . Pharmacol. 80, 197-204. Birdsall, N. J. M., Burgen, A. S. V., Hulme, E. C., Stockton, J. M., and Zigmond, M. J. (1983~). Br. J. Pharmacol. 78, 257-259.

194

WAYNE IlOSS AND JOHN EI.LIS

Birdwll, N J . M., I-lulme, E. C., and Stockton,J. M. (I983d). C,’o/l Sj)rrngHarbo? Sym/~48, 53-56. birdsall, N. ,I. M., Hulme, E. C., and Stockton, J. M. (1!)84). 111 “Subtypes of Muscarink Receptors” (B. 1. Hirschowitz, R. Hammer, A. (hchetti, ,I..]. Keirns, and R. R. Levine, eds.), pp. 4-8. Elsevier, Amsterdani. Bokoch, G. M., Katada, T., Northrup, J. K., Hewlett, E. R., and Gilman, A. J. (1983).J. B i d . Chem. 258, 2072-2075. Bolton, T. B. (1976). Proc. R . Sue. 194, 99-1 19. Brown, 1). A. (1984). In “Subtypes of Muscarink Receptors” (B. I. llirschowitz, K. Hanimer, A. (;i;ichctti, ,I. ]. Keirns, ant1 R. K. Levinc, eds.), pp. :32-34. Elscvier, Ainsterdam. Brown, D. A , , Forward, A., and Marsh, S. (1981). Br. J. Phurmucol. 71, 362-364. Burgen, A. S . V. (1979). Br. Med. Bull. 35, 269-273. Burgisser, E., Lefkowitz, R. J., and DeLean, A. (1981). Mol. Pharmacol. 19, 509-512. Burgoyne, R. D. (1980). FEBS Lett. 122, 288-292. Burgoyne, R. D. (1981). FEBS Lett. 127, 144-148. Burgoyne, R. D. (1983).J . Neurochem. 40, 324-331. Burgoyne, R. D., and Pearce, B. (1981). Dev. Bruin Res. 2, 55-63. Burt, D. R. (1978a). BruinRex 143, 573-579. Burt, D. R. (1978b). In “Neurotransmitter Receptor Binding” (H. I. Yamamura, S. J. Enna, and M. J. Kuhar, eds.), pp. 41-56. Raven, New York. Carson, S. (1 982). Biochem. Phurmucol. 31, 1806- 1809. (:hang, K.-J., Jacobs, S., and Cuatrecasas, P. (1975). Birichmi. Biophys. Acta 406, 294-303. Cheng, Y.-C., and Prusoff, W. H. (1973). Biochem. Pharmacol. 22, 3099-3108. Chin, H. R., Shifrin, G., and Klein, W. L. (1980). Fed. Proc., Fed. Am. Sac. Exp. Biol. 39, 1618. Clark, A. L., and Mitchelson, F. (1976). B r . 1 . Pharmacol. 58, 323-331. Cockcroft, S. (1981). Trends Pharmacol. Sci. 2, 340-342. Crockcroft, S., Bennett, J. P., and Gomperts, (1980). FEBS Lett. 110, 115-118. Cohen, N. M., Schmidt, D. M., McGlennen, R. C., and Klein, W. L. (1983).J. Neurochem. 40,547-554. Cooper, D. M. F. (1982). FEBS Lett. 138, 157-163. Coyle, J. T., and Snyder, S. H. (3981). In “Basir Neurocheniistry” ((;. J . Siegel, R. W. Alhers, B. W. Agranoff, and R. K. Katznian, eds.), 3rd Ed., pp. 205-217. Little, Brown, Boston, Massachusetts. Cuatrecasas, P., and Hollenberg, M. D. (1975). Biochem. Biophys. Res. Commun. 62, 3 1-4 1. Cuatrecasas, P., and Hollenberg, M. D. (1976). Adv. Protein Chem. 30, 251-451. Crawford, J. M., and Curtis, D. R. (1966).J. Physiol. (London) 186, 121-138. Dale, H. H. (1914).J . Pharmacol. Exp. Ther. 6 , 147-190. Davies, P., and Maloney, A. J. F. (1976). Lancet 2, 1403. De Belleroche, J., and Bradford, H. F. (1978). Brain Res. 142, 53-68. De Belleroche,J., Kilpatrick, I. C., Birdsall, N . J. M., and Hulme, E. C. (1982). Brain Res. 234,327-337. Deguchi, T., Ohsako, S., Nakane, M., Ichihawa, M., and Yoskioka, M. (1983). J . Neurotranrm. 518, 369-378. DeLean, A., Hancock, A. A., and Lefkowitz, R. J. (1982). Mol. Pharmucol. 21, 5-16. Downes, C. P. (1982). Cell Calcium 3, 413-428. Dudai, Y., and Segal, M. (1978). Bruin Res. 154, 167-171. Dunlap, J., and Brown, J. H. (1983). Mol. Phuimucol. 24, 15-22. Eckelman, W. C., Reba, R. C., Rzeszotarski, W. J., Gibson, R. E., Hill, T., Holman, B. L.,

MUSCARINIC RECEPTOR SUBTYPES IN T H E

CNS

195

Budinger, T., Conklin, J. T., Eng, R., and Grissom, M. P. (1984). Science 223, 291292. Ehlert, F. J., Dumont, Y., Roeske, W. R., and Yamamura, H. I. (1980a).Life Sci. 26, 961967. Ehlert, F. J. Roeske, W. R., Rosenberger, L. B., and Yamamura, H. I. (1980b). Life Sci. 26, 245-252. Ehlert, F. J., Roeske, W. R., and Yamamura, H. I. (1981a). Mol. Phai-macol. 19, 367-371. Ehlert, F. J., Roeske, W. R., and Yamamura, H. I. (1981b). Life Sci. 28, 2441-2448. El-Fakahany, E., and Richelson, E. (1980). Proc. Natl. Acad. Sci. U.S.A. 77, 6897-6901. Ellis, J., and Hoss, W. (1980). Brain Res. 193, 189-198. Ellis, J., and Hoss, W. (1982). Biochem. Pharmacol. 31, 873-876. Ellis, .J., and Lenox, K. H. (1985). Biochem. Pharmacol., in press. Farrar, J. R., and Hoss, W. (1984). Biochem. Pharmarof.,33, 2849-2856. Feldman, H . A. (1972). Anal. Biochem. 48, 317-338. Ferrendelli, J. A., Steiner, A. L., McDougal, J. R., and Kipnis, D. M. (1970). Biochem. Biophys. Res. Commun. 41, 1061-1067. Ferrendelli, J. A., Rubin, E. H., and Kinscerf, D. A. (1976).J. Neurochem. 26, 741-748. Fibiger, H. C. (1982). Brain Res. Rev. 4, 327-388. Fischel, G. V., and Medzihradsky, F. (1981). Mol. Pharmacol. 20, 269-279. Fisher, A,, Grunfeld, Y., Weinstock, M., Gitter, S., and Cohen, S. (1976).Eur.1. Pharmacol. 38, 131-139. Fisher, S. K., and Agranoff, B. W. (1980).J. Neurochem. 34, 1231-1240. Fisher, A,, Abraham, S., Lachman, C., Lass, Y., Akselrod, S., Ackerman, E., and Cohen, S. (1980a). Mongr. Neural Sci. 7, 41-54. Fisher, S. K., Boast, C. A,, and Agranoff, B. W. (1980b). Brain Res. 189, 284-288. Fisher, S. K., Klinger, P. D., and Agranoff, B. W. (1983).J. Biol. Chem. 258, 7358-7363. Folch, J., Lees, M., and Stanley, G. H. S. (1957).J . Biol. Chem. 226, 497-509. Franklin, P. H., and Hoss, W. (1984).J. NeurochPm., 43,~1132-1 135. Fuder, H., Rink, D., and Muscholl, E. (1982). Naunyn-Schmiedeberg’s Arch. Pharmacol. 318, 210-219. Furchgott, R. F. (1978). Fed. Proc., Fed. Am. SOC.Exp. Biol. 37, 115-120. Galper, J. B., and Smith, T. W. (1978). Proc. Natl. Acad. Sci. U.S.A. 75, 5831-5835. Galper, J. B., Klein, W., and Catteral, W. A. (1977).J.Biol. Chem. 252, 8692-8699. Gavish, M., and Sokolovsky, M. (1982). Biochern. Biophys. Res. Commun. 109, 434-445. Gazit, H., Silman, I., and Dudai, Y. (1979). Brain Res. 174, 351-356. Giachetti, A., Said, S. I., Reynolds, R. C., and Koniges, F. C. (1977). PTOC.Natl. Acad. Sci. U.S.A. 74, 3424-3428. Gil, D. W., and Wolfe, B. B. (1983). Abstr. SOC.Neurosci. 9, 581. Gilbert, R. F., Hanley, M. R., and Iversen, L. L. (1979). Br. J . Pharmacol. 65, 451-456. Giles, W., and Noble, S. J. (1976).J. Physiol. (London) 261, 103-123. Griffin, H. D., Hawthorne, J. N., Sykes, M., and Orlacchio, A. (1979). Biochem. Pharmacol. 28, 1143-1 147. Gurwitz, D., and Sokolovsky, M. (1980). Biochem. Biophys. Res. Commun. 96, 1296-1304. Gurwitz, D., Kloog, Y., and Sokolovsky, M. (1980). Life Sci. 26, 79-84. Hammer, R. (1982). Scand. /. Gmtroenterol. 17, 59-67. Hammer, R., and Giachetti, A. (1982).Life Sci. 31, 2991-2998. Hammer, R., Berrie, C. P., Birdsall, N. J. M., Burgen, A. S. V., and Hulme, E. C. (1980). Nature (London) 283, 90-92. Hanley, M. R., and Iversen, L. L. (1978). Mol. Pharmacol 14, 246-25.5. Harden, T. K., Scheer, A. G., and Smith, McH., M. (1982). Mol. Pharmacol. 21, 570-580.

196

WAYNE HOSS AND ]OHN ELLIS

Hartzell, H. C. (1982). Trends Pharmacol. Sci. 3, 213-214. Hawthorne, J. N. (1982). Nature (London) 295, 281-282. Hedlund, B. (1981). I n “Cholinergic Mechanisms: Phylogenetic Aspects, Central and Peripheral Synapses, and Clinical Significance” { C . Pepeu, ed.), pp. 647-652. Plenum, New York. Hedlund, B., and Bartfai, T. (1979). Mol. Pharmacol. 15, 531-544. Hedlund, B., Unden, A., Grynfarb, M., and Bartfai, T. (1980). Acta Chem. Scand. Ser. B 34, 767. Hedlund, B., Grynfarb, M., and Bartfai, T. (1982). Naunyn-Schmiedeberg’s Arch. Pharmacol. 320,3-13. Henis, Y. I., and Sokolovsky, M. (1983). Mol. Pharmacol. 24, 357-365. Horn, J., and Dodd, J. (1981). Nature (London) 292, 625-627. Hoss, W. P., and Formaniak, M. (1984). Neurochpm. Res., 9, 109-120. Hulme, E. C., Birdsall, N. J. M., Burgen, A. S. V., and Mehta, P. (1978).Mol. Pharmacol. 14, 737-750. Hulme, E. C., Berrie, C. P., Birdsall, N. J. M., and Burgen, A. S. V. (1981a). Eur. J. Pharmacol. 73, 137-142. Hulrne, E. C., Berrie, C. P., Birdsall, N. J. M., and Burgen, A. S. V. (1981b). In “Drug Receptors and Their Effectors” (N. J. M. Birdsall, ed.), pp. 23-34. Macrnillan, New York. Hulme, E. C., Berrie, C. P., Haga, T., Birdsall, N. J. M., Burgen, A. S. V., and Stockton, J. (1983).J. Recept. Res. 3, 301-311. Hunzicker-Dunn, M., Derda, D., Jungmann, R. A., and Birnbauner, L. (1979). EndocrinolO ~ Y104, 1785-1793. Hurko, 0. (1978). Arch. Biochem. Biophys. 190, 434-445. Jakobs, K. H. (1979). Mol. Cell. Endocn’nol. 16, 147-156. Jakobs, K. H., Aktories, K., and Schultz, G. X. (1979). Naunyn-Schmiedeberg’s Arch. P h a m a C O ~ .310, 113-119. Jakobs, K. H., Aktories, K., and Schultz, G. X. (1983a). Nature (London) 303, 177178.

Jakobs, K. H., Aktories, K., and Srhultz, G . X. (I983b).J. ltecep. Rer. 3, 137-149. Jacobs, S., Chang, K.-]., and Cuatrecasas, P. (1975). HiochPni. H i o p h y . ~Res. Cornmurs. 66, 687-692. Jaw, J., Hedlund, B., and Bartfai, T. J. (1979).]. Biol. Chem. 254, 5595-5598. Johnston, M. V., McKinney, M., and Coyle, J. T. (1979). Proc. Natl. Acad. Sci. U.S.A. 76, 5392-5396. Johnston, M. V., McKinney M., and Coyle, J . T. (1981). Ex$. Brain Res. 43, 159-172. Jones, L. M., Kirk, C. J., and Michell, R. H. (1982). Scand.J. Gmtroenterol. 17(Suppl.), 3341. Kamiya, H., Rotter, A,, and Jacobowitz, D. M. (1981). Brain Res. 209,432-439. Karczmar, A. G. (1977). Adv. Behav. Biol. 24, 697-708. Kato, G., Carson, S., Kernel, M. L., Glowinski,J., and Gioruieff, M. F. (1978). Life Sci. 22, 1607- 1614. Kenakin, T. P. (1983). Trendr Pharmacol. Sci. 4, 291-295. Kilpatrick, B. F., and Caron, M. G. (1983).J. Bzol. Chem. 258, 13528-13534. Kirk, C. J., and Michell, R. H. (1981). Biochem.]. 194, 155-165. Klein, W. L. (1980). Biochem. Bzophys. Res. Commun.93, 1058-1066. Klein, W. L., Nathanson, N., and Nirenberg, M. (1979). Biochem. Biophys. Res. Commun. 90, 506-512. Kloog, Y., and Sokolovsky, M. (1978a). Brain Res. 144, 31-48. Kloog, Y., and Sokolovsky, M. (1978b). Biochem. Biophys. Res. Commun. 81, 710-717. Kuba, K., and Koketsu, K. (1978). Prog. Neurobiol. 11, 77-169.

MUSCARINIC RECEPTOR SUBTYPES I N T H E

CNS

197

Kuno, T., Shirakawa, O., and Tanaka, C. (1983). Biochem. Biophys. Res. Commun. 112, 948953. Kurose, H., Katada, T., Amano, T., and Ui, M. (1983).J. Biol. Chem. 258, 4870-4875. Lefkowitz, R. J., DeLean, A., Hoffman, B. B., Stadel, J. M., Kent, R., Michel, T., and Limbird, L. (1981). Adu. Cyclic Nucleotide Res. 14, 145-161. Lenox, R. H., Ellis, J., Van Riper, D. A,, Ehrlich, Y. H., Peyser, J. M., Shipley, J. E., and Weaver, L. A. (1983). In “Frontiers in Neuropsychiatric Research’ (E. Usdin, M. Goldstein, A. J. Friedhoff, and A. Georgota, eds.), pp. 331-356. Macmillan, New York. Linden, J. (1982).J. Cyclic Nucleotide Res. 8, 163-172. Lundberg, J. M., Hedlund, B., and Bartfai, T. (1982). Nature (London) 295, 147-149. Lundquist, I., Ahren, B., Hakanson, R., and Sundler, F. (1980). In “Diabetes 1979” (W. K. Waldhausl, ed.), pp. 57-63. Excerpta Medica, Amsterdam. McKinney, M., and Coyle, J. T. (1982).J. Neurosci. 2, 97-105. McKinney, M., and Richelson, E. (1984). Annu. Rev. Pharmacol. Toxic01 24, 121-146. McKinney, M., Stanstrom, S., and Richelson E. (1983). SOC.Neurosci. Abstr. 9, 581. McPherson, R. A., and Zettner, A. (1975). Anal. Biochem. 64, 501-508. Marks, M. J., Artman, L. D., Patinkin, D. M., and Collins, A. C. (1981).J. Pharmacol. Exp. Ther. 218, 337-343. Matsuzawa, H., and Nirenberg, M. (1975). Proc. Natl. Acad. Sci. U.S.A. 72, 3472-3476. Messer, W. S., Thomas, G. J., and Hoss, W. P. (1983). SOC.Neurosci. Abstr. 9, 826. Michell, R. H. (1975). Biochem. Biophys. Acta 415, 81-147. Michell, R. H. (1982). Nature (London) 296,492-493. Michell, R. H., and Kirk, C. J. (1981). Trends Pharmacol. Sci. 2, 86-89. Michell, R. H., Jafferji, S. S., and Jones, L. M. (1976). FEBS L e a 69, 1-5. Miller, J. C. (1977). Bi0chem.J. 168, 549-555. Minneman, K. P., and Molinoff, P. B. (1980). Biochem. Pharnacol. 29, 1317-1323. Molinoff, P. B., Wolfe, B. B., and Weiland, G. A. (1981). Life Sci. 29, 427-443. Mosko, S., Lynch, G., and Cotman, C. W. (1973).J. Camp. Neurol. 152, 163-174. Motulsky, H. J., and Mahan, L. C. (1984). Mol. Pharmacol. 25, 1-9. Munson, P. J. (1983).J. Receptor Res. 3, 249-259. Munson, P. J., and Rodbard, D. (1980). Anal. Biochem. 107, 220-239. Murayama, T., and Ui, M. (1983).J. Biol. Chem. 258, 3319-3326. Muscholl, E. (1980). Am. J . Physiol. 239, H713-720. Nathanson, N. M., Klein, W. L., and Nirenberg, M. (1978). Proc. Natl. Acad. Sci. U.S.A. 75, 1788-1791. Nishizuka, Y. (1983). Philos. Trans. R . Sac. London Ser. B 302, 101-112. Nordberg, A., and Winblad, B. (1981). Life Sci. 29, 1937-1944. Nordstroni, O., Alberts, P., Westlind, A., Unden, A., and Bartfai, T. (1983). Mol. Pharmacot. 24, 1-5. North, R. A., and Tokimasa T. (1984). I n “Subtypes of Muscarinic Receptors” (B. I. Hirschowitz, R. Hammer, A. Giachetti, J. J. Keirns, and K. R. Levine, eds.), pp. 35-38. Elsevier, Amsterdam. Olianas, M. C., Onali, P., Neff, N. H., and Costa, E. (1983). Mol. Pharmacol. 23, 393-398. Onali, P., Olianas, M. C., Schwartz, J. P., and Costa, E. (1983). Mol. Pharmacol. 24, 380386. Overstreet, D. H., Speth, R. C., Hruska, R. E., Ehlert, F., Durnont, Y., and Yanianiura, H . I . (1980). Brain Res. 195, 203-207. Parker, R. B. (1972).J . Pharmacol. Exp. Ther. 180, 62-70. Perry, E. K. (1980). Age Ageing 9, 1-8. Perry, E. K., Perry, R. H., Blesser, G., and Tomlison, B. E. (1977).J. Neurol. Sci. 34, 247265.

198

WAYNE HOSS AND JOHN ELLIS

Phillis, J. W., Koustopoulos, C. K., and Limacher, J. J. (1974). C a n . J . Phys. Pharmacol. 52, 1226-1229. Pimoule, C . , Briley, M., Arbilla, S., and Langer, S. Z. (1980). Naunyn-Schmiedeberg’s Arch. Phannacol. 312, 15-18. Plesums, J., and Bunch, W. H. (1971). Anal. Biochem. 42, 360-362. Potter, I,. T., Flynn, D. D., Hanchett, H. E., Kalinoski, D. L., Luber-Narod, J., and Mash, D. C. (1984). Trends Phamacol. Sci. Suppl. 22-31. Prinz, H. (1983).J. Recept. Res. 3, 239-248. Purves, R. D. (1976). Nature (London) 261, 149-151. Putney, J. W., Weiss, S. J., Van Den Walle, C. M., and Haddas, R. A. (1980). Nature (London) 284, 345-347. Raisman, G. (1966). Brain 89, 317-348. Rang, H. P. (1966). Proc. R. Soc. London Bid. Sci. 164, 488-510. Rathbun, F. J.. and Hamilton, J. T. (1970). Can. Anne& SOC. J. 17, 574-590. Richelson, E., and Divinetz-Romero, F. G. (1977). Bid. Psychiatry 12, 771-785. Richelson, E., and El-Fakahany, E. (1981). Biochern. Pharmacol. 30, 2887-2891. Richelson, E., Prendergast, F. G., and Diventz-Romero, F. G. (1978). Biochem. Pharmacol. 27,2039-2048. Robson, L. E., and Kosterlitz, H. W. (1979). Proc. R. Soc. London Ser. B 205, 425-432. Rodbard, D., Ruder, H. J., Vaitukaitis, J., and Jacobs, H. S. (1971).J. Clin. Endocrinol. 33, 343-355. Ross, E. M., and Gilrnan, A. G . (1980). Annu. Rev. Biochem. 49, 533-564. Ross, E. M., Maguire, M. E., Sturgill, T. W., Biltonen, R. L., and Gilman, A. G. (1977). J . Bi d . Chem. 252, 5761-5775. Roszkowski, A. P. (1961).J. Pharmacol. Exp. Ther. 132, 156-170. Ruberg, M., Ploska, A., Javoy-Agid, F., and Agid, Y. (1982). Bruin Res. 232, 129-139. Ruess, K.-P., anti LieHander, M. (1979). Biochum. Hiophys. He.,. Commun. 88, 627-633. Sabol, S. L., and Nirenberg, M. (1979).J. B i d . Chem. 254, 1913-1920. Sachs, D. I., Kloog, Y . , Korczyn, A. D., Heron, D. S., and Sokolvsky, M. (1979). Biochem. Pharmacol. 28, 1513-1518. Salmon, D. M., and Honeyman, T. W. (1980). Nature (London) 284, 344-345. Schimerlik, M. I., and Searles, R. (1980). Biochemistry 19, 3413-3434. Schultz, G., Hardman, J. G., Schultz, K., Baird, C. E., and Sutherland, E. W. (1973). Proc. Natl. Acad. Sci. U.S.A. 70, 3889-3893. Sharnia, V. K., and Bannerjee, G. P. (1978). Nature (London) 272, 276-278. Shifrin, G. S., and Klein, W. L. (1980).J. Neurochem. 34, 993-999. Siman, R. G., and Klein, W. L. (1979). Proc. Natl. Acad. Sci. U.S.A. 76, 4141-4145. Siman, R. G., and Klein, W. L. (1981).J. Neurochem. 37, 1099-1108. Skipsky, V. P., Peterson, R. F., and Barclay, M. (1964). Biochem. J . 90, 374-378. Smit, M. H., Ehlert, F. J., Roeske, W. R., and Yamamura, H. I. (1980). Fed. Proc., Fed. Am. Soc. Exp. B i d . 39, 388. Smith, S. K., and Limbird, L. E. (1981). Proc. Natl. Acad. Sci. U.S.A. 78,4026-4030. Smith, T. L., and Hauser, G. (1981).J. Neurochem. 37, 427-435. Snider, R. M., and Richelson, E. (1983). Science 217, 566-567. Snider, R. M., Richelson, E., and Blinks, J. R. (1981). Soc. Neurobcz. Abstr. 8, 124. Snyder, S. H. (1975). Biochem. Pharmacol. 24, 1371-1374. Snyder, S. H., and Bennett, J. P. (1976). Annu. Rev. Physiol. 38, 153-175. Snyder, S. H., Greenberg, D., and Yamamura, H. I. (1974). Arch. Gen. Psychiatry 3 1 , 5 8 4 9 . Snyder, S. H., Chang, K. J., Kuhar, M. J., and Yamamura, H. I. (1975). Fed. Proc., Fed. Am. Soc. Exp. B i d . 34, 1915-1921.

MUSCAKINIC RECEPTOR SUBTYPES IN T H E

CNS

199

Sokolovsky, M. (1984). In “Subtypes of Muscarinic Receptors” (G. I. Hirschowitz et al., eds.), pp. 17-21. Elsevier, Amsterdam. Sokolovsky, M., Gurwitz, D., and Galron, R. (1980). Biochern. Biophys. Res. Commun. 94, 487-492. Sokolovsky, M., Egozi, Y., and Avissar, S. (1981). Proc. Natl. Acad. Sci. U.S.A. 76, 55545558. Sokolovsky, M., Gurwitz, D., and Kloog, J. (1983). Adv. Enrymol. 55, 137-196. Stadel, J. M., Nambi, P., Shorr, R. G. L., Sawyer, D. F., Caron, M. G., and Lefkowitz, R. J. (1983). Proc. Natl. Acad. Sci. U.S.A. 80, 3173-3177. Starke, K. (1977). Rev. Physiol. Biochem. Pharmacol. 77, 1-124. Stephenson, R. P. (1975). INSERM 50, 15-28. Stockton, J. M., Birdsall, N. J. M., Burgen, A. S. V., and Hulme, E. C. (1983). Mol. Pharmacol. 23, 551-557. Stone, T . W. (1972).J. Physiol. (London) 225, 485-499. Stone, T . W., Taylor, D. A,, and Bloom, F. E. (1975). Science 187, 845-847. Story, D. D., Briley, M. S., and Langer, S. Z. (1979). Eur. J . Pharmacol. 57, 423-426. Strange, P. G., Birdsall, N. J. M., and Burgen, A. S. V. (i977). Biochem. Sac. Tram. 5, 189191. Swillens, S., and Dumont, J. E. (1980). Mol. Cell. EndocrZnol. 20, 233-242. Szerb, J. C. (1977). Adv. Behav. Biol. 24, 49-60. Szerb, J. C., Hadhazy, P., and Dudar, J. D. (1977). Brain Res. 128, 285-291. Tallarida, R. J. (1981). Trends Pharmacol. Sci. 2, 231-234. Taylor, J. E., and Richelson, E. (1979). Mol. Pharmacol. 15, 462-471. Taylor, J. E., Yaksh, T. L., and Richelson, E. (1982).J. Neurochem. 39, 521-524. Thomas, G. J., and Brit,, G. N. 0. (1980). J . C o r p . Physiol. Psychol. 94, 808-818. Tolkovsky, A. M. (1982). Mol. Pharmacol. 22, 648-656. Tsai, B. S., and Lefkowitz, R. J. (1979). Mol. Pharmacol. 16, 16-68. Vaughn-Williams, E. M. (1957).,J. Physiol. (London) 140, 327-341. Wallace, M. A., Randazzo, P., Li, S., and Fain, J. N. (1982). Endocrinology 111, 341-343. Watson, M., Roeske, W. R., and Yamamura, H. I. (1982). Life Sci. 31, 2019-2032. Weiland, G. A., and Molinoff, P. B. (1981). Life Sci. 29, 313-330. Weiner, N. (1980). In “The Pharmacological Basis of Therapeutics” (L. S. (;oodman and A. G. Gilman et al., eds.), pp. 135-175. Macniillaii, Ncw York. Weiner, M. (1982). Geriatrics 37, 81-91. Weiss, S. J., and Putney, J. W. (1981). Biochem. J. 194, 463-468. Westfall, T. C. (1974). Life Sci. 14, 1641-1652. Westlind, A,, Grynfarb, M., Hedlund, B., Bartfai, T., and Fuxe, K. (1981). Brazn Res. 225, 131-141. Whitehouse, P. J., Price, D. L., Struble, K.G., Clark, A. W., Coyle, J. T., and DeLong, M. K. (1982). Science 215, 1237-1239. Yamamura, H. I., and Snyder, S. H. (1974a). Mol. Pharmacol. 10, 861-876. Yamamura, H. I., and Snyder, S. H. (1974b). Brain Res. 78, 320-326. Young, J. M. (1974). FEBS Lett. 46, 354-356.

This Page Intentionally Left Blank

NEURAL PLASTICITY AND RECOVERY OF FUNCTION AFTER BRAIN INJURY By John

F. Marshall

Department of Psychobiology University of California, lrvine Irvine, Californio

.............................. .................... A. Alternate Strategies. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B. Redundancy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C. Vicarious Functioning. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . D. Diaschisis . . . . . . . . . . . . . . . . . . . . . . . . . . . ............ 111. Neural Events Mediating Recovery: Overview . . . . . . . . . . . . . . . . . . . . . . . . I. Introduction. . . . . . . . . . . . . . .

11. Theories of Recovery of Functlon

IV. Neural Events Mediating Recovery: Morphological Adaptations.. ........ A. Axonal Growth and Its Determinants . . . . . . . . . . . . . . . . . . . . . . . . . . . . B. Axonal Growth and Recovery of Function . . . ................. V. Neural Events Mediating Recovery: Neurochemical Adaptations . . . . . . . . . A. Neurochemical Adaptations and Their Determinants. . . . . . . . . . . . . . . . B. Neurochemical Adaptations and Recovery of Function . . . . . . . . . . . . . . VI. Conclusions and Future Directions . . . . . . . . . .. References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..

20 1 202 203 203 204 206 209 2 10 210 217 223 224 233 239 24 1

1. Introduction

After humans or animals suffer central nervous system (CNS) damage they often show impairments in behavioral o r physiological functions that abate as the time since injury increases. This phenomenon, termed recovery of function, has long intrigued clinicians and scientists both for its medical significance and because of the varieties of neural plasticity that its study may reveal. The question that is central to this field of research is, what adaptations occur in surviving nerve cells that are responsible for the behavioral improvement? Embodied in this question are two separate issues: (1) the identification of neural changes that occur in response to CNS injury, and (2)an analysis of their contribution to the recovery process. One thesis of this article is that this question can only be answered by employing particular model systems of brain injury, 20 I INTERNATIONAL REVIEW O F NEUROBIOLOGY, VOI.. 26

Copyright 0 1985 by Academic Press. Inc. All rights of reproduction i n any form reserved. ISBN 0-12-366826-3

202

JOHN F. MAKSHALL

in which the neural and behavioral events occurring after the damage may be studied concurrently. Interest in the neural mechanisms of recovery of function has increased significantly in the past decade because of advances in understanding how neurons can change their structure or function after CNS damage. Nowhere have these advances been more dramatic than in the research on axonal sprouting. Although axonal growth after peripheral nerve injury had long been recognized (reviewed by Edds, 1953), its occurrence in the mammalian CNS was not discovered until the pioneering work of Liu and Chambers (1958). I n the 25 years since then the view of the brain and spinal cord as static organs has yielded to the overwhelming evidence that many CNS neurons undergo both structural and biochemical changes according to use or specific experiences, during normal synapse turnover, or in response to injury in adulthood (Diamond et al., 1975; Uylings et al., 1978; Juraska et al., 1980; Cotman et al., 1981; Greenough and Green, 1981; Tsukahara, 1981). The purpose of this article is to review the progress that has been made in understanding the neural events that contribute to recovery from brain injury and to suggest approaches that appear promising. I start by considering the theories that have influenced this field in the past. I then describe the forms of neural plasticity that are thought to mediate recovery of function, reviewing those instances of nervous system injury in which neurobiological explanations of a recovery sequence have been suggested.

II. Theories of Recovery of Function

The recovery that humans and animals frequently show from the initial motor, sensory, or cognitive consequences of CNS injury represented a disturbing challenge to those neurologists and neuroscientists of the nineteenth and early twentieth centuries who adhered to the tenet that functions are strictly localized to particular brain structures. Flourens, who in 1824 described recovery from brain injury in experimental animals, concluded that behaviors could not be localized to specific cortical regions. However, during the latter half of the nineteenth and the first part of the twentieth centuries, the overwhelming evidence for localization of function in the brains of animals and humans silenced the critics (see Rosner, 1974). Several theories were advanced that attempted to account for the existence of behavioral recovery, while still acknowledging the validity of the localizationist doctrine.

RECOVERY OF FUNCTION AFTER BRAIN INJURY

203

A. ALTERNATE STRATEGIES After damage to the CNS, humans and animals may develop strategies, or tricks, that allow them to circumvent their lost capacities. These alternative strategies permit the subject to achieve the same behavioral end using means that are very different from neurologically intact individuals. For example, many humans that suffer damage to the visual cortex and have a permanent scotoma demonstrable during perinietry testing show few obvious impairments of visual functions in their day-today lives. This compensation is achieved in part through the patient’s eye movements that scan the visual field, thereby circumventing the scotomatous retinal regions (Teuber et al., 1960). Also, monkeys with lesions of the dentate and interpositus cerebellar nuclei display a syndrome of limb ataxia followed by tremor from which they subsequently recover. These dyskinesias result from an inability to fixate the affected limb at the precise moment that a phasic movement is initiated (Goldberger and Growden, 1973). During recovery the monkey avoids limb positions that are likely to result in these oscillatory dyskinesias, a compensation that depends upon somatosensory information from the affected limb. Some neurologists argue that most, if not all, instances of behavioral recovery can be attributed to behavioral tricks rather than the intrinsic capacity of the injured CNS for neuronal compensation. However, this conclusion is difficult to accept in light of the growing number of model systems in which correlates between neuronal plasticity and behavioral recovery have been found (see Sections 1V and V).

B. REDUNDANCY After damage to the CNS, the recovery from a particular behavioral impairment may depend upon uninjured neurons that normally contribute to that behavior. J. Hughlings Jackson (reprinted in 1958) proposed one form of redundancy, a hierarchical representation of functions in the CNS. After damage to higher centers, lower ones remain intact to subserve the function. Alternatively, neurons located at the same level of the neural axis as the injured cells may contribute to recovery. For example, Lashley (1929) concluded that all cortical regions contributed equally to maze learning in the rat (“equipotentiality”) and that the extent of the learning impairment depended only on the amount of cortical tissue excised (“mass action”). More recent work supports the view that considerable redundancy

204

JOIIN F. MARSHALL,

exists, even within topographically organized sensory and motor systems (Frommer, 1978). T h e severe and persistent motor deficits resulting from unilateral pyramidotomy in monkeys depend upon the extent of pyramidal tract injury, with some recovery occurring unless virtually all fibers have been severed (Beck and Chambers, 1970; Lawrence and Kuypers, 1968). Also, cats with damage to 98-99% of optic tract axons can perform a visual pattern discrimination (Galambos et al., 1967). T h e number of spared fibers, rather than their position within the damaged optic or pyramidal tracts, appears critical to the maintenance of sensory or motor functions. However, the survival of redundant neurons does not offer a sufficient explanation for behavioral recovery because it does not specify what events occur within the surviving neurons during the postoperative period to mediate the recovery. For example, after damage to the optic nerve fibers, an impairment in acuity is evident with as little as 33% damage to the tract, and the rate of recovery from the acuity deficit is inversely related to the extent of fiber loss (Jacobson et al., 1979). Also, although an irreversible impairment in orienting toward somatosensory stimuli does not occur unless 95% or more of the nigrostriatal dopaminergic afferent fibers are damaged (Marshall, 1979), significant but brief impairments in this behavior result from destruction of only 2530% of these terminals (Marshall et al., 1980). In this system as well the duration of the impairment is inversely proportional to the extent of damage to this population of dopaminergic cells (Marshall, 1979). That significant behavioral deficits do result from even small lesions within redundantly organized systems suggests that compensatory neural events within the surviving elements underlie the behavioral improvement. Thus, the spared neurons provide a substrate in which the dynamic events leading to behavioral improvement may occur.

C. VICARIOUS FUNCTIONING According to this theory, brain regions that survive the injury have a latent ability to carry out the functions of the damaged system. This hypothesis, in contrast to the redundancy theory, states that neurons not previously involved in a particular function may alter their properties to assume that function after injury. T h e evidence typically presented to support vicarious functioning takes the following form. After damage to brain region A, a behavior is initially impaired and subsequently recovers. When region B is damaged concurrently with or subsequent to area A injury, this behavior is irreversibly lost; however, damage to region B

RECOVERY OF FUNCTION AFTER BRAIN INJURY

205

alone fails to produce the deficit. For example, Lashley (1938) showed that rats with lesions of the striate cortex lose the ability to perform a brightness discrimination that they had learned prior to surgery, but that they can relearn this discrimination postoperatively. However, when the pretectal regions are removed along with the striate cortex no relearning of the brightness discrimination occurs. Because injury to the pretectum and tectum by themselves does not affect brightness discrimination, Lashley concluded that the pretectum is induced to assume a role in brightness discrimination learning after the striate cortex is injured and that the brightness relearning of the destriate rat depends upon this modification of pretectal capacities. However, sequential lesion experiments can be misleading. As noted previously, monkeys with combined dentate-interpositus lesions appear to recover from their ataxia and tremor by employing an alternative strategy for limb fixation that depends upon somatosensory information from the affected limb (Goldberger and Growden, 1973). After spontaneous recovery from deep cerebellar nuclear lesions has occurred, these dyskinesias can be reinstated by damage anywhere along primary somatosensory projections (Goldberger, 1974). Rather than arguing that somatosensory systems vicariously take over the function of the damaged cerebellar nuclei in movement, Goldberger ( 1 974) proposed that the reduction in somatosensory afference resulting from the second lesion compromises the behavioral strategy that the animal has developed. Whereas sequential lesion experiments cannot provide unequivocal evidence for vicarious functioning, electrophysiological experiments can. T h e hypothesis predicts that the properties of neurons in regions presumed to take over the impaired behaviors should be altered by the lesion with a time course corresponding to the behavioral recovery. Experiments that test these predictions have been conducted using visual structures of the cat in which receptive field properties have been especially well characterized, and the results do not support vicarious functioning. After ablation of the visual cortex, cats show an initial impairment in discriminating patterns, but extensive postoperative training promotes relearning. The lateral suprasylvian gyrus is especially important to the residual pattern discrimination abilities, since its removal after recovery from the initial injury results in a pattern discrimination loss that is not reversed by retraining (Baumann and Spear, 19’77). Removal of the suprasylvian gyrus alone, however, has a negligible effect on pattern vision (Wood et al., 1974). Spear and Baumann (1979) investigated whether after visual cortex damage the receptive field characteristics of lateral suprasylvian gyrus

206

JOHN F. MARSHALL

cells change to become like those of visual cortical neurons. In the intact cat, lateral suprasylvian gyrus cells have the following properties: most respond to a particular direction of moving stimuli, few respond to stationary light, many are driven by either eye, and (unlike visual cortex neurons) they are insensitive to the spatial orientation of bars or slits in the receptive field. After visual cortex removal, few cells are responsive to the direction of moving stimuli, many respond to stationary lights, they are driven almost exclusively by the contralateral eye, but they remain insensitive to orientation. These changes in receptive field properties occur within a few hours of the visual cortex removal and persist for months afterward, even in cats that have been successfully retrained to perform a pattern discrimination (Fig. 1). The alterations in suprasylvian gyrus receptive field properties observed immediately after visual cortex removal suggest that certain properties of these cells (directional selectivity, ipsilateral ocular driving) depend upon information carried by the direct projections that these areas receive from the visual cortex. The persistence of these electrophysiological changes despite extensive retraining indicates that these extrastriate visual regions are not induced to assume the properties of visual cortex neurons as a result of the injury. The failure of lateral suprasylvian gyrus cells to adopt an orientation sensitivity typical of visual cortex cells further supports this conclusion. Thus, these experiments provide no evidence for vicarious functioning of spared visual structures. Presumably, the lateral suprasylvian gyrus provides the destriate cat with redundant information concerning visual form that allows discriminations to be achieved after extensive retraining.

D. DIASCHISIS von Monakow (1914) proposed that behavioral functions are lost after brain injury for two reasons. First, the injury destroys neurons. Because these neurons are not replaced, he argued that the impairments resulting from their loss must be irreversible. Second, the destruction of nerve cells alters the excitability of neurons that normally receive axonal inputs from the damaged cells, a phenomenon termed diaschisis, or neural shock. This trans-synaptic effect of the injury, von Monakow argued, is reversible. As the excitability of the denervated neurons returns to normal, behavioral functions that are impaired as a result of the diaschisis should be restored. Both the precise definition of the excitability changes and the mechanism by which injury induces diaschisis were not specified.

RECOVERY OF F UNC T I ON AFTER BRAIN INJURY

207

A. NORMAL

20

20

O I S M D

' 1 2 3 4 5 6 7

B. ACUTE 3R SHORT-TERM LESION

u

'1"" "9 1

N.55

60

40

2

2oL ' 1 2 3 4 5 6 7

C LONG-TERM LESION AND BEHAVIORAL RECOVERY 80

-

60 40 -

20

~

ib ':\b

' $ S M O

2 3 4 5 6 7

RECEPTIVE FIELD OCULAR DOMINANCE

FIG. 1. Response properties of cells in lateral suprasylvian cortex of three groups of cats. (A) 24 normal cats; (B) 7 cats with acute or short-term bilateral removal of areas 17, 18, and 19; (C) 5 cats which had demonstrated behavioral recovery from long-term bilateral removal of areas 17, 18, and 19. Left shows the percentage of cells ( N , number of cells in sample) in each of the following four receptive field classes: I, indefinite; S, stationary; M, movement sensitive; D, direction selective. Right shows the percentage of cells in each of seven ocular dominance groups, ranging from 1 (driven exclusively by the contralateral age) to 7 (driven exclusively by the ipsilateral eye). [Reprinted from Spear, P., and Baumann, T. (1979). Erp. Bruin Res. 35, 177-192, with permission of Springer-Verlag.]

T h e clearest behavioral example of neural shock is the depression of spinal reflex activity that occurs after transection of the spinal cord (Sherrington, 1906). Depending upon species, the period of areflexia lasts from minutes to days (Mountcastle, 1974). More encephalized animals display a more prolonged areflexia, suggesting that the duration of the shock to spinal centers depends upon the volume of the descending supraspinal projections that are severed.

208

JOHN F. MAKSHAI.1.

Electrophysiological changes suggestive of neural shock have been identified in the spinal cord and brainstem. Cord potentials that reflect spinal interneuronal response to dorsal root stimulation are acutely depressed after transection in dog and cat, paralleling the depression of spinal reflexes (Stewart et al., 1940). Barnes et al., (1962) recorded the membrane potentials of lumbar motoneurons after cooling the thoracic spinal cord in cats (which eliminated the crossed extensor reflex). Spinal cooling induces a prompt 2-to 8-mV hyperpolarization of motoneurons, and the resting potential returns to normal within 30 sec of rewarming. Unfortunately, the effects of cooling on the response of motoneurons to afferent driving was not determined. Somatosensory deafferentation depresses the spontaneous activity of neurons in the lateral cuneate nucleus for 24 hr (Kjerulf and Loeser, 1973; Kjerulf et a/., 1973). However, the discharge rate of cells in the dorsal horn or nucleus gracilis is unchanged even immediately after dorsal rhizotomy (Millar et al., 1976; Devor and Wall, 1978; Brenowitz and Pubols, 1981). Furthermore, West et al. (1976) obtained no evidence for diaschisis in the hippocampus. The hypothesis predicts that after denervating the granule cells of the dentate gyrus of their entorhinal afferents, their response to remaining inputs (i.e., commissural-associational) should diminish. However, the extracellular field potentials elicited by commissural stimulation were unchanged from control levels in rats 30 min to 1 1 days following removal of the ipsilateral entorhinal cortex. Also, the experiments of Wickelgren and Sterling ( 1969), Rosenquist and Palmer (1971), and Spear and Baumann (1979) fail to provide evidence for diaschisis in the visual system. The alterations in receptive field characteristics in the superior colliculus and lateral suprasylvian gyrus after visual cortex ablation appear due to the interruption of information conveyed to these structures via the visual cortex, not to transsynaptic excitability changes. Even immediately after visual cortex injury, the tectal and suprasylvian neurons display receptive field properties indicative of their continued responsiveness to retinal and thalamic visual inputs. Therefore, it is clear that neural shock is not a general consequence of injury. Its occurrence in any neuron may depend upon the quantity of the synaptic input that the cell loses as a consequence of the injury. In those instances where diaschisis does occur, its dissipation could well contribute to functional restoration. Because the acute loss of spinal reflex activity after transection is such a powerful behavioral demonstration of neural shock, further electrophysiological investigations of the

RECOVERY OF FUNCTION AFTER BRAIN INJURY

209

spinal interneurons and motoneurons of such animals should provide insight into the cellular events underlying this phenomenon.

111. Neural Events Mediating Recovery: Overview

Although the preceding theories have provided a major organizational influence on research in this field, for three reasons they have not yielded answers as to how recovery from CNS injury occurs. First, these theories have often proven resistant to experimental test. Second, in cases where they have been tested (i.e., vicarious functioning and diaschisis), the results have been either strongly disconfirmatory o r equivocal. Third, each theory seeks to provide a general account of recovery rather than explaining recovery from particular instances of CNS injury. If, for example, diaschisis occurs only in structures that are nearly completely denervated as a result of injury, then the dissipation of neural shock can contribute to behavioral recovery only under these special circumstances. In contrast to the theories of recovery discussed in the previous section, more recent accounts of behavioral restoration emphasize the capacity of the CNS to undergo specific synaptic changes in response to injury. These forms of neural adaptation may lead to behavioral recovery by normalizing synaptic transmission within the injured nervous system. I n the following two sections I consider these types of neuronal plasticity and review the evidence suggesting that they may contribute to specific instances of behavioral recovery after CNS injury. T h e neurobiological explanations for recovery sequences fall into two general categories: morphological and neurochemical. T h e morphological accounts emphasize the ability of the injured CNS to form new synapses by, e.g., axon sprouting. I n contrast, neurochemical explanations stress the contribution of existing synapses to modify their activity by, e g . , increasing transmitter synthesis and release or elevating the postsynaptic response to transmitter. This classification provides a conceptual framework for the organization of the remainder of this review. However, before considering these explanations in detail, it is important to recognize that the successful analysis of the neuronal events mediating a recovery sequence depends critically upon the choice of an appropriate model system. Two criteria must be fulfilled. 1. The relationship between the CNS region under investigation and the behavior of interest must be thoroughly characterized. Damage to

210

JOHN I;. M A R S H A L L

the brain region should result in a well-defined behavioral impairment followed by postoperative improvement. The time course and behavioral characteristics of the recovery should be thoroughly documented. Ideally, it should be possible to associate the behavioral abnormality with the injury of one or more identified populations of nerve cells within the damaged region. The relationship of the extent of injury to this population of neurons to the severity and duration of the ensuing behavioral impairments should be known. Fulfilling this criteria is a major hurdle because the behavioral consequences of selective injury to only a few cell populations in the CNS can be stated with confidence. 2. The varieties of neuronal plasticity that occur within these identified cells, their targets, or their afferents must either be known or be amenable to study using available techniques. The extent of knowledge concerning the neurobiology of the system under investigation limits the questions that can be posed.

IV. Neural Events Mediating Recovery: Morphological Adaptations

After its injury the nervous system may undergo structural alterations that permit lost functions to be restored, such as regenerative and collateral axon sprouting, expansion of dendritic surface area (Cotman et al., 198 l), and a replacement of lost neurons by cell division similar to that occurring in the olfactory epithelium (Graziadei and Graziadei, 1978). The present article limits itself to the axonal growth responses, which have been investigated extensively. A. AXONAL GROWTH AND ITS DETERMINANTS 1. Regenerative Growth

Peripheral nerves have a striking capacity to grow after injury (Edds, 1953; Guth, 1956). The sprouting can take two forms. In regenerative sprouting, the distal end of the severed axon degenerates and the proximal stump emits a growth cone that can enter the Schwann cell sheaths. Under optimal circumstances, these growing processes follow the denervated sheaths to the vacated tissue, at which point they form synaptic contacts. Collateral sprouting occurs when only part of the innervation to a tissue is removed. Uninjured axons extend collateral branches (or extend the surface area of their terminals) to increase the innervation of the partly denervated tissue.

RECOVERY OF FUNCTION AFTER BRAIN IN-JURY

21 1

Although CNS neurons do not ordinarily display the extensive regenerative growth typical of peripheral axons ( R a m h y Cajal, 1928; Bernstein and Bernstein, 1973), several conditions favor central growth. One such condition is the positioning of a target tissue appropriate for reinnervation in the vicinity of the transected fibers. When smooth muscle grafts that normally receive an adrenergic sympathetic innervation (e.g., iris or portal vein) are positioned along the course of brain catecholaminergic axon bundles, regenerating sprouts grow across the scar tissue at the graft-CNS interface and innervate the transplant (Bjorklund and Stenevi, 1971 ; Bjorklund et al., 1975a). Physiological experiments demonstrate that these axons establish functional contacts with the graft (Bjorklund el al., 1975a). Similarly, severed CNS acetylcholine-containing axons grow into an iris graft positioned near the axotomy (Svengaard et al., 1976; Emson et al., 1977). Central transplants of embryonic brain tissue can also induce regenerative growth of severed axons. When transplants of embryonic hippocampus are placed into a cavity along the severed fimbria of adult rats, fimbrial cholinergic axons from the host animal innervate the transplant during the subsequent 1-2 months (Kromer et al., 1981a). After innervating the transplant, the growing cholinergic axons often continue to grow into the host hippocampus, their normal target tissue (Kromer et al., 1981b; Fig. 2). The pattern of reinnervation by these axons is similar to that of the normal cholinergic lamination. Thus, the host hippocampus can attract the ingrowth of regenerating axons provided that those axons have been stimulated to grow into its vicinity by the adjoining embryonic transplant. A second condition that influences the regenerative growth of CNS axons is the extent of glial scarring at the site of injury (Guth and Windle, 1970). Whereas mechanical transection or electrolysis of brain tissue leads to the local formation of a dense scar, the monoamine neurotoxins 6-hydroxydopamine (6-OHDA), 5,6-dihydroxytryptamine (5,6-DHT), and 5,7-dihydroxytryptamine (5,7-DHT) do not. When CNS monoamine-containing axons are damaged by the injection of these neurotoxins into the parenchyma or CSF, regenerative growth from injured axons can occur. Thus, after intraspinal injection of 6-OHDA, the number of fluorescent catecholamine axons in the spinal cord decreases dramatically, but the brainstem catecholamine cell bodies that give rise to the spinal projection survive (Nygren and Olson, 1977). Fluorescent profiles reappear in the spinal cord within 3 days and extensively reinnervate the spinal cord during the next 1-2 months (Nygren et al., 1971). Regenerating noradrenergic axons reinnervate the spinal cord in a craniocaudal direction. When the noradrenergic innervation of the entire cord is

212

JOHN F. MARSHALL

FIG. 2. Schematic representation of acetylcholinesterase-positive fibers in an animal containing a hippocampal bridge implant (I in a-d), 5.5 months survival time. The symbols P 0.0 mm to P 6.3 mm represent coronal sections taken at indicated distances from site of fusion of implant with the lateral septum (LS). Acetylcholinesterase-positive fibers from the host septum reach the host hippocampus (HPC) and dentate gyrus (DG) via the bridge implant. Other abbreviations: AD, anterodorsal thalamic nucleus; C, caudate-putamen; CC, cingulate cortex; DB, diagonal band; DHPC, dorsal hippocampus; LH, lateral habenula; MG, medial geniculate; MH, medial habenula; RC, retrosplenial cortex; S, subiculum; SC, superior colliculus; SM, medial septum; VHPC, ventral hippocampus. [Reprinted from Kromer, L., Bjorklund, A,, and Stenevi, U. (1981b). Brain Res. 210, 173-200, with permission of Elsevier Biomedical Press.]

RECOVERY OF FUNCTION AFTER BRAIN INJURY

213

destroyed, the reinnervation of cervical levels requires approximately 4 weeks while that of thoracic segments takes 3 months. By 6.5 months postoperatively, the reinnervation of lumbar and sacral segments is only partial (Nygren and Olson, 1977). The regrowth of the spinal noradrenergic axons can also be followed by measuring the high-affinity [3H]norepinephrine(NE) uptake into different spinal segments, and the reappearance of uptake sites occurs in a craniocaudal direction (Bjorklund and Lindvall, 1979). When 5,6-DHT is injected into the cerebral ventricles in doses that destroy serotonin (5-HT)-containing axons but not their cell bodies, spinal serotonergic axons also regrow (Nygren et al., 1974; Bjorklund and Wiklund, 1980; Wiklund and Bjorklund, 1980). Again, the ingrowth of fluorescent fibers and the return of [“H]-5-HT uptake occur in a craniocaudal direction. After intraventricular injections of these neurotoxins, brain 5-HT and NE axons can also grow (Bjorklund et al., 1973, 1975b; Wuttke et al., 1977; Bjorklund and Lindvall, 1979). Monoamine-containing axons that regenerate in the brain, like those in the spinal cord, preferentially reinnervate structures near to the origin of axonal sprouting (Bjorklund et al., 1975b). A third determinant of CNS axonal regeneration is the neuroglial environment through which the sprouting fibers grow (Guth and Windle, 1970). Because Schwann cells provide a substrate for the growth of regenerating peripheral axons, their role in CNS regrowth has received recent attention. When a length of sciatic nerve forms a bridge between the cranial and caudal ends of the severed dog or rat spinal cord, axons of CNS origin grow through the peripheral nerve segment during the subsequent 1-3 months (Kao et al., 1977; Richardson et al., 1980; David and Aguayo, 1981; see Fig. 3). Many of the growing fibers become ensheathed by Schwann cells, and the sciatic nerve bridges can sustain growth for distances as great as 35 mm (David and Aguayo, 1981). Although the growing fibers may be derived from cells axotomized during the spinal transection, this point has not been established. I n other experiments (Horvat, 1980; Weinberg and Raine, 1980), segments of peripheral nerve have been transplanted into the midbrain o r spinal cord of rodents. These grafts become invaded by axons and ensheathed by Schwann cells.

2 . Collateral Growth Although collateral sprouting in the mammalian CNS was first demonstrated by Liu and Chambers in 1958, only by the mid-1970s was this form of synaptic plasticity recognized as a common neural adaption to CNS injury (reviewed by Kerr, 1975).

2 14

Bu

JOIIN F. MAKSHA1.L

A

4 mm

c c

35 mm

65mm

[

FIG. 3. (A) Diagram of the dorsal surface of’the rat CNS, showing a peripheral nerve “bridge” linking the medulla and thoracic spinal cord. Cross sections depict the regions where the ends of the nerve graft were inserted; (B) approximate rostrocaudal position of 1472 labeled CNS neurons (dots) demonstrated in 7 grafted rats. In the brdinstem the territory occupied by 450 of these cells extended along 4 mm, whereas 1022 labeled neurons were scattered along a 6.5-mm segment ofthe spinal cord. [Reprinted from David S., and Aguayo, A. J. (1981). Scieriu 214, 931-933, with permission of the American Assoriation for the Advancement of Scicnce.]

The dentate gyrus of the rat hippocampus has proven to be an extremely useful system in which to study collateral sprouting. The apical dendrites of dentate granule cells extend through the width of the molecular layer and receive afferent projections from several sources in a precise laminar organization (reviewed by Cotman and Nadler, 1978). In brief, the extensive projection from the ipsilateral entorhinal cortex (perforant path) terminates in the outer two-thirds of the molecular layer, as does a sparse projection from the contralateral entorhinal cortex. Axons from the ipsilateral and contralateral hippocampal pyramidal cells (i.e., associational and commissural fibers) terminate in the inner one-third of the molecular layer. T h e septohippocampal fibers project to narrow zones of the molecular layer, one in the inner molecular layer and a second in the outer layer. Removal of one entorhinal cortex induces an extensive rearrangement of the remaining afferent fibers in the molecular layer. T h e chohnesterdse-staining septa1 axon terminals to the outer molecular layer increase their distribution (Lynch et al., 1972), the commis-

RECOVERY OF FUNCTION AFTER BRAIN IhVURY

215

sural-association afferent fibers move into the denervated outer molecular layer (Lynch et al., 1973), and the projection of the contralateral entorhinal cortex to the denervated molecular layer increases (Steward et al., 1974). Electrophysiological experiments indicate that the redistributed afferents form functional contacts (Lynch et al., 1973; Steward et al., 1974). In contrast, damage to the commissural and/or associational systems does not induce a major synaptic rearrangement (Lynch et al., 1974). No changes in the distribution of entorhinal or cholinesterase-staining afferent fibers to the molecular layer occur. After damage to the commissural projection alone, however, the synaptic density of the inner molecular layer does increase postoperatively, perhaps reflecting a proliferation of the associational fibers to this zone (McWilliams and Lynch, 1978). A sprouting of GABA-containing interneurons in response to damage of extrinsic afferent fibers may also occur (Nadler et al., 1974). Although collateral sprouting in the CNS is now a well-documented phenomenon, the stimulus responsible for its appearance is poorly understood. In the neuromuscular junction several stimuli have been considered: (1) degeneration products of the injured nerve might induce a growth of neighboring axons; (2) trophic factors transported down the axon and released into the junction may normally limit the growth of neighbors; and (3) muscle inactivity may change some properties of the muscle fibers so that they attract new growth. Several findings point to the importance of muscle inactivity in initiating sprouting. Whereas denervated’ muscles can attract the ingrowth of axon terminals from a foreign nerve implanted on their surface, chronic electrical stimulation of the denervated muscle prevents this ingrowth (L@moand Slater, 1978). Furthermore, a terminal sprouting from intact axons innervating the rodent soleus muscle occurs if the muscle activity is decreased by conduction block of its axonal input, prevention of acetylcholine (ACh) release at the neuromuscular junction, or acetylcholine receptor blockade (Brown and Ironton, 1977; Brown et al., 197’7; Holland and Brown, 1980). Trophic factors transported by the axon may also contribute to sprouting. When a salamander sensory nerve is exposed to a concentration of colchicine that interrupts fast axoplasmic transport, the peripheral fields of the neighboring nerves enlarge, just as they do after section of the experimental nerve (Aguilar et al., 1973). In these experiments the colchicine treatment does not interfere with the exposed nerve’s axon conduction o r structural integrity (Diamond et al., 1976). The sprouting that occurs after colchicine application or treatments that pre-

216

JOHN F. MARSHALL

vent neuromuscular transmission is clearly not dependent upon nerve degeneration. Thus, products of neuronal degeneration are not necessary for sprouting. Apparently only one experiment has investigated the signal leading to growth responses in the mammalian CNS. Goldowitz and Cotman (1980) found that they could induce an increased density of synapses in the inner molecular layer of the dentate gyrus by applying colchicine to the commissural afferent fibers that innervate this zone. The source of this hyperinnervation was the commissural projection. Because the colchicine treatment led to little o r no degeneration of the exposed axons, they suggest a role for axonally transported trophic substances in the regulation of synaptogenesis. Of course, these findings do not rule out a role for neuronal activity in CNS synaptic remodeling. Collateral sprouts may compete with regenerating axons for vacated synaptic space. I n partially denervated skeletal muscles, remaining afferent fibers sprout within several days to occupy denervated end plates. But if the damaged nerve is allowed to regenerate, these regenerating axons reclaim synaptic space from the collateral sprouts (Brown and Ironton, 1978; Thompson, 1978; but see Guth, 1962). A similar competitive advantage of regenerating axons over collateral sprouts has been seen in the sympathetic trunk innervation of the pupil (Guth and Bernstein, 1961), sensory nerve innervation of cutaneous fields (Slack, 1978; Devor et al., 1979; Jackson and Diamond, 1981), and the vagal inputs to the cardiac ganglia (Roper, 1976). Sometimes the regenerating fibers fail to reclaim all of their original synaptic territory and some collateral innervation persists (Guth and Bernstein, 1961; Roper, 1976; Brown and Ironton, 1978). When this occurrs, the time lag between the collateral innervation and arrival of regenerating axon terminals appears critical. When the regeneration is delayed, allowing the collateral growth to fully establish its innervation, the regenerating axons are less successful in reestablishing their territory (Slack, 1978; Thompson, 1978). The interaction of sprouting and regenerating axons in the CNS is poorly understood, but the possibility that an early collateral sprouting response may retard the synaptic connections of subsequent regenerating axons must be considered. After the central serotonergic innervation of the subcommissural organ is damaged, this organ becomes reinnervated, but by non-5-HT-containing terminals (Wiklund and M@llgArd, 1979). These investigators hypothesized that the collateral growth may have prevented the serotonergic regeneration. Similarly, Kromer et al., (198 1 b) considered this possibility in their studies of septal regeneration into hippocampus. The septal cholinergic axons begin to reinnervate the host hippocampus by 6- 12 weeks postoperatively. Based on published

RECOVERY OF FUNCI'ION AFTER BRAIN INJURY

217

work (McWilliams and Lynch, 1979), the nonfimbrial afferent fibers to the host hippocampus should have undergone significant collateral sprouting well before this time. The finding that the septa1 regeneration into the host hippocampus is at best 30% complete may reflect competition from an established collateral innervation. AND RECOVERY OF FUNCTION B. AXONAL GROWTH

1 . Peripheral Nervous System Explanations of functional recovery based upon axonal growth after nervous system injury are not new, especially for the peripheral nervous system. T h e discovery of collateral sprouting of somatosensory nerves (Weddell et al., 1941) resulted from the observation that skin sensitivity could return to normal after cutaneous nerve injury. Also, a collateral sprouting of motor nerves contributes to the recovery of muscle tension after damage to spinal ventral roots (reviewed by Edds, 1953). A similar relationship between synaptic remodeling and recovery has been observed in the sympathetic nervous system. The CNS control of the cat nictitating membrane is mediated by preganglionic axons exiting in the thoracic ventral roots and synapsing in the superior cervical ganglion. Several months after severing 90% of the preganglionic axons controlling one nictitating membrane, electrical stimulation of the remaining 10% of the fibers on that side evokes a membrane response equal to that resulting from stimulation of all the preganglionic axons on the intact side. This compensation is achieved in several stages. By 4-5 days postoperatively the decentralized membrane becomes supersensitive to NE, resulting in a partial compensation. Within 2 weeks the intact preganglionic axons sprout collaterals in the partially denervated superior cervical ganglion, leading to a further normalization of the response to stimulation and a return of normal membrane sensitivity to NE (Murray and Thompson, 1957). Functional studies suggest that the collateral sprouts are eventually replaced by regenerating terminals of the severed preganglionic axons (Guth and Bernstein, 196 1). These experiments demonstrate elegantly how postsynaptic supersensitivity, collateral sprouting, and regenerative growth can interact sequentially to contribute to the recovery of function at peripheral synapses.

2 . Central Nervous System Determining the neural events mediating recovery from CNS injury is more difficult than for peripheral nerves. T h e function of a peripheral nerve is typically apparent from its anatomy. Also, peripheral

218

JOHN F. MARSHALL

nerves and their target tissues can be isolated relatively simply so that any plastic changes may be studied directly. Nevertheless, during the last decade important advances have been made in understanding the neuronal events mediating recovery from CNS injury. Much of the success in this field has occurred in neonatal animals, in which the capacity for injury-induced growth is far greater than in adults. a. Retanotectal Projection. T h e integrity of the retinal projection to the superior colliculus of the Syrian hamster is essential for its head-turning toward visual stimuli (Schneider, 1969). In neonates in which the superficial layers of the superior colliculus have been removed, the ingrowing retinal fibers terminate exclusively in the deeper layers of this structure, rather than synapsing in their normal superficial position. In those hamsters in which the terminations are distributed throughout the mediolatera1 extent o f the deeper layers, adult visual orientation is remarkably normal, indicating that the terminations in the inappropriate tectal layers are functional. In other hamsters in which the terminations are restricted to the lateral aspect of the deep layers, however, orientation occurs only toward stimuli in the lower visual field (corresponding to the retinotopic organization of this structure) (Schneider, 1973). In hamsters with unilateral removal of the superficial collicular layers soon after birth, axons from the contralateral retina not only synapse in the deeper layers of the injured tectum but also decussate across the tectal midline where they terminate in the medial aspect of the superficial layers of the intact superior colliculus. This anomalous crossed projection can be induced to grow throughout the entire superficial gray layer of the intact colliculus if the eye projecting to that collicdus (i.e., ipsilateral to the injured tectum) is also removed at birth. As adults, such hamsters frequently show misdirected head orientation: stimuli presented to the intact eye may elicit head-turning toward the opposite visual field (Schneider, 1973). When this anomalous crossed projection is transected in the adult hamster, the misdirected turning is nearly totally abolished (Schneider, 19’79). b. Pyramidal Tract. Damage to the neonatal hamster brain can induce redirected axonal growth in pathways other than the retinotectal projection. After one pyramidal tract is interrupted between the fourth and eighth days of life, the severed axons grow out dorsolaterally in the brainstem to occupy an anomalous position. During the subsequent weeks, however, these redirected axons grow and form synaptic connections in the dorsal column nuclei of the brainstem and in the dorsal horn of the spinal cord. The positioning and ultrastructural appearance of these terminals appears remarkably normal, despite the anomalous course that the regenerating axons take to reach these sites. The ability

RECOVERY OF FUNCTION AFTER B R A I N INJURY

219

of severed pyramidal tract fibers to regenerate into the spinal cord diminishes when they are injured after the eighth day of life and is absent in hamsters operated on as adults (Kalil and Reh, 1979, 1982). This regeneration has important behavioral consequences for the manual dexterity of the hamsters. Normal adult hamsters shell sunflower seeds with their teeth, using the digits of the forepaws to grasp and rotate the seeds. Hamsters with interruption of one pyramidal tract in adulthood shell seeds clumsily. They use the affected paw to support the seed but lose the use of digits. As a result, the time required for them to she11 seeds increases significantly, relative to neurologically intact adults. In contrast, the hamsters given unilateral pyramidotomy at 4-8 days of age show good use of the affected forepaw in this behavior as adults, and the times required to shell seeds are nearly normal (Fig. 4). When the infant-operate animals undergo a second surgery as adults to sever the regenerated pyramidal axons, the impairments of manual usage are evident. These results indicate that integrity of the regenerated fibers is responsible for the sparing of manual function in the infant operates (Reh and Kalil, 1982). 240 220 -

NORMAL

INFANT LESION (4-8 days)

FIG.4. Histogram comparing the average time required for animals with pyramidal tract lesions at three postnatal ages to shell and eat sunflower seeds. Open bars represent scores of hamsters with unilateral lesions; stippled bars represent scores of animals with bilateral lesions; cross-hatched bars represent the scores of sham-operated animals. [Reprinted from Reh, T., and Kalil, K. (1982).J. Covnp. Neurol. 211,276-283, with permission of Alan R. Liss, Inc.]

220

JOHN F. MARSHAL.L

c. Hippocampus. In spite of the extensive work done on synaptic remodeling in the partially denervated hippocampus, the behavioral consequences of sprouting in this structure are poorly understood. The hippocampus appears to contribute to spatial tasks (O’Keefe and Nadel, 1978). One measure of this ability is the learned alternation task, in which an animal must alternate between running down the left and right arms of a maze on successive trials to obtain food reward. Bilateral lesions of the entorhinal cortex, which denervates the dentate gyrus of each hemisphere of its major source of extrinsic afferents, lead to a longlasting impairment in alternation behavior. Rats with such lesions are equally likely to enter the left and right arms of a maze regardless of their choice on the previous trial. Each entorhinal cortex projects heavily to the ipsilateral dentate gyrus and sparsely to the contralateral dentate. After unilateral entorhinal cortex lesion, the projection of the contralateral entorhinal cortex to the ipsilateral dentate gyrus increases markedly between days 8 and 12 postoperatively (Steward et al., 1974, 1976). T h e possible functional consequences of this sprouting of a homologous source of afferents to the dentate have been examined by Loesche and Steward (1977). Unilateral lesions of the entorhinal cortex produce a decrement in alteration performance that returns to preoperative levels during the first 12 days postoperatively. Furthermore, after recovery has occurred, a secondary lesion of the remaining entorhinal cortex or of its crossed projection to the dentate (via dorsal psalterium) reinstates the deficit (however, see Ramirez, 1980, cited in Finger and Stein, 1982, pp. 98-99). Based on these observations, Loesche and Steward (1977) suggest that the sprouting of crossed entorhinal-dentate projections may underlie the recovery of the spatial task. However, approximately two-thirds of the recovery of alternation performance that these investigators observe occurs by postoperative day 7, prior to the reported onset of sprouting. One possibility not discussed by these authors is that the early improvement in alternation performance depends upon a supersensitivity (see Section V) of the partially denervated granule cells to the neurotransmitter released by the intact crossed projection. This would account both for the early onset of the recovery and for its dependence upon the integrity of the crossed projection. After bilateral lesions of the fimbria, which eliminate the cholinergic innervation of the hippocampus, rats developed impairments in another spatial task, performance of an eight-arm radial maze. Rats with fimbrial lesions that received bilateral transplants of fetal septal cholinergic cells at the site of the fimbrial damage showed a gradual improvement in maze performance, whereas lesioned rats not receiving septal trans-

RECOVERY OF FUNCTION AFTER BRAIN INJURY

22 1

plants exhibited no improvement with repeated trials. Administration of the acetylcholinesterase inhibitor, physostigmine, to the rats receiving transplants further enhanced their maze performance. Lesioned animals not given the transplants exhibited no improvement with physostigmine (Low et al., 1982). d. Spinal Cord. Goldberger and Murray (see review, 1978) have undertaken an extensive research program to investigate the role of neural remodeling in the recovery of limb movements after limb deafferentation or spinal cord injury. Using several different surgical preparations they have identified morphological changes that appear to contribute to the recovery of motor function. i. Spinal hemisection. Cats that have spinal hemisections (sparing the dorsal columns) at lower thoracic or upper lumbar levels show a considerable initial depression of the motor functions of the ipsilateral hindlimb. This motor depression affects reflexes intrinsic to that limb (tendon, cutaneous flexor), contralateral segmental reflexes (crossed extensor), descending reflexes (vestibular, ear scratch), limb tone, and the use of the limb in spontaneous movements such as locomotion. During the 2-3 weeks postoperatively certain reflexes show a considerable recovery. T h e intrinsic reflexes recover and become hyperactive. The contralateral segmental reflexes recover but do not become hyperactive. The descending reflexes never recover, their pathways having been transected. During this time the use of the ipsilateral hindlimb in locomotion recovers to such an extent that the cat appears normal. Murray and Goldberger (1974) suggest that the marked increase in the intrinsic segmental reflexes that follows hemisection contributes to the recovery of that limb’s ability to function during locomotion. In support of this view, deafferentation of the affected hindlimb of a cat with spinal hemitransection completely prevents recovery. The reflex and locomotor recovery may be due to an expansion of the dorsal root terminal fields in the lumbar spinal cord. Murray and Goldberger (1974) found an increased termination of dorsal roots within the intermediate layers and the base of the dorsal horn at spinal levels caudal to the hemisection producing the motor impairments described above. Electrophysiological receptive field correlates of this expanded projection to the dorsal horn (Brenowitz and Pubols, 1981) suggest that terminal expansion occurs within the first 3 weeks after the hemisection, a time course consistent with the behavioral recovery. ii. Complete limb deafferentation. In an otherwise intact cat, sectioning all of the lumbar dorsal roots supplying one hindlimb results in an immediate flaccid paralysis of the limb. The limb is dragged passively during spontaneous locomotion, lacks tone, and cannot be driven to

222

JOHN F. MARSHALL

activity by the residual pathways (postural reflex systems arising in the trunk, contralateral segmental reflex pathways, or the descending projections). By 2 days postoperatively the reflex control of the limb by these residual pathways begins to recover, and during the second week the descending reflexes (vestibular placing, ear scratch) become hyperactive. Shortly thereafter the cat appears to use cues from its trunk to place the limb accurately during locomotion. The hyperactivity of the supraspinal reflexes suggested to Goldberger and Murray (1974) that the descending projections to the lumbar spinal cord might sprout following hindlimb deafferentation. In confirmation, cats that received upper lumbar transections 12 months after unilateral deafferentation showed a greater distribution of degeneration products in laminae IV through VIII of the ipsilateral lumbar cord. iii. Spared root preparation. I n this preparation, all of the lumbar and sacral dorsal roots except one (typically L6) are cut unilaterally. T h e behavioral effects of this surgery are much less severe than are those of deafferentation or hemisection. If the spared root innervates the foot, then within 3-4 days the cat shows ipsilateral segmental reflexes (placing, hopping) and the hindlimb is accurate in locomotion. By the second week some segmental reflexes (placing, tendon reflexes) become hyperactive. In contrast, at no time are the reflexes mediated by the descending pathways depressed or hyperactive. In cats with long-standing section of all but a single root, the distribution of that spared root is significantly increased (Liu and Chambers, 1958; Goldberger and Murray, 1978), particularly its projection to the intermediate gray zone, Clarke’s nucleus, the dorsal horn, and parts of the nucleus gracilis (Fig. 5 ) . Receptive field electrophysiological correlates of the sprouting in the lateral portion of the dorsal horn have been observed (Pubols and Goldberger, 1980). iv. Generalizations. In each preparation, the recovery of spontaneous movements of the affected hindlimb occurs concurrently with a postoperative increase in the reflex functions mediated by certain of the residual pathways to the lumbar spinal cord. I n the case of the hemisected or spared root preparations, intrinsic reflexes return; in the case of complete deafferentation, the descending reflexes are restored. The reflex hyperactivity occurs during the second and third weeks and is accompanied by a dramatic improvement of the locomotor function of hemisected or completely deafferented cats. T h e view that the hyperactivity of residual reflex pathways contributes importantly to the recovery of spontaneous locomotion is supported by the finding that the concurrent destruction of these pathways prevents recovery. The axonal projections that mediate the hyperactive reflexes in each

RECOVERY OF FUNCTION AFTER BRAIN INJURY

223

FIG. 5 . Tracing of L6 spinal segment of a cat (spared root preparation) in which the dorsal roots of one side, sparing L6, were cut. One year later L6 was cut on both sides. Four days later the animal was killed and the spinal cord stained by Nauta and Fink-Heimer techniques. The figure shows the L6 projections on the experimental (left) and control (right) sides. The degeneration around the motoneurons of the horn appears similar on the two sides. The experimental side shows an increase in the projection within the lateral part of the dorsal horn and in the intermediate gray zone. The increased degeneration presumably reflects sprouting of terminals and/or axon collaterals into some of the denervated regions at some time during the postoperative year. [Reprinted from Goldberger, M., and Murray, M. (1978). In “Neuronal Plasticity” (C. Cotman, ed.), pp. 73-96, with permission of Raven Press.]

preparation show a heavier than normal distribution to the lumbar segments of the spinal cord, apparently having sprouted in response to the denervation. However, except for the study of Brenowitz and Pubols (1981), the relationship between the time course of the terminal expansion and that of the functional improvement is unknown. Also, it is not known which of the sprouted terminal fields (dorsal horn, intermediate layer, Clarke’s nucleus) contribute to the recovery of motor function.

V. Neural Events Mediating Recovery: Neurochemical Adaptations

In addition to the morphological forms of plasticity described above, central synapses can respond to injury by modifying their rate of chemical transmission. Neurons spared from injury may contribute to a recovery of function by increasing their rate of transmitter synthesis and

224

JOHN F. MARSHALL

release or by decreasing the rate of inactivation of released transmitter. Also, neurons postsynaptic to the injured pathway may show an enhanced response to the transmitter released by surviving terminals (postsynaptic supersensitivity). Within the CNS these adaptations are best documented for the central monoamine-containing neurons, where the regulation of transmitter synthesis, release, inactivation, and postsynaptic sensitivity have been studied most extensively.

A. NEUROCHEMICAI, ADAPTATIONS AND

THEIR

DETERMINANTS

1. Synthesis and Release

After incomplete damage to a central catecholamine (CA)-containing projection, the surviving nerve terminals of that pathway typically show an enhanced rate of transmitter synthesis and release. After extensive but subtotal damage to nigrostriatal dopaminergic projection, for instance, the rate of incorporation of [3H]tyrosineinto [3H]dopamine (DA) in the neostriatum is increased, when expressed as a proportion of the endogenous DA remaining in this structure (Agid et al., 1973). Similarly, the ratios of the dopamine metabolites dihydroxyphenylacetic acid and homovanillic acid to DA are enhanced, suggesting an increased rate of transmitter metabolism by the surviving terminals (Sharman et al., 1967; Hefti et al., 1980). Similar increases in the ratio of homovanillic acid to DA occur in patients with Parkinson’s disease, in which many of the nigral DA-containing neurons degenerate (Bernheimer and Hornykiewicz, 1962). Also, after partial injury to the central noradrenergic axons that innervate the hippocampus or cerebellum, the activity of tyrosine hydroxylase (TH), the rate-limiting enzyme in CA biosynthesis, is increased within the remaining NE-containing terminals of these structures. At 1.5-5 days after the injury T H activity is enhanced due to an allosteric modification of the T H molecule which increases its affinity for the reduced pteridine cofactor. By 3 weeks postoperatively this allosteric modification is no longer evident, but the surviving cerebellar and hippocampal noradrenergic nerve terminals contain a greater than normal number of T H molecules (Acheson and Zigmond, 1981; Fig. 6). Similarly, an increased rate of 5-HT metabolism occurs within the remaining terminals after incomplete damage to the bulbospinal 5-HTcontaining axons (Bjorklund and Wiklund, 1980). After incomplete injury to a monoamine-containing projection, the elevated rate of transmitter synthesis and release within the residual

RECOVERY OF FUNCTION AFTER BRAIN INJURY

-*

1.6

12

1.2

9

-.

. P6 c E

s kE

225

0.8

:

5 a E

Y

0.4

. i

0

0

Control

5 21 Post 60HDA (days)

Control

5 21 Post W H D A (days)

FIG.6. Kinetics of effect of pterin cofactor (GMPH,) on tyrosine hydroxylase activity of hippocampus 5 or 21 days after intraventricular injection of 6-OHDA (250 pg) in rats. Tyrosine hydroxylase activity was measured in the presence of varying concentrations of GMPH, in Tridacetate buffer, pH 6.2. Values represent the mean Tt SEM for five separate animals. Note that this analysis was carried out at the pH optimum for the enzyme in untreated control rats, thus minimizing the change in the K , at 5 days postlesion. [Adapted from Acheson, A., and Zigmond, M. J. (1981).J. Neurosci. 1, 493-504.1

neurons is at least partly the result of an increased rate of action potentials (“impulse flow”) in the surviving axons. Electrical stimulation that activates peripheral (Zigmond, 1980) or central (Roth et al., 1974, 1975, 1976) CA-containing neurons or central serotonergic cells (Shields and Eccleston, 1972) increases the release of transmitter from their nerve terminals, and compensatory increases in monoamine synthesis result. Similarly, drugs that act as antagonists of brain DA receptors increase the firing rate of substantia nigra dopaminergic cells (Bunney et al., 1973) and lead to increased DA synthesis and release (Carlsson and Lindqvist, 1963; Zivkovic et al., 1974). Three events appear to contribute to the enhanced CA synthesis during increased impulse flow. First, the accelerated release of CA from the nerve terminal initially reduces the intraterminal CA concentration, which may increase synthesis by reducing end-product inhibition (Spector et al., 1967; Carlsson et al., 1976). Second, electrical stimulation of central DA or NE neurons leads to an allosteric modification of the T H molecule similar to that seen after incomplete injury (Roth et al., 1975;

226

JOHN F. MARSHALL

Murrin and Roth, 1976). This activation occurs within minutes of the onset of electrical stimulation (Salzman and Roth, 1980), requires Ca2+ (Harris and Roth, 1971; Simon and Roth, 1979), and may result from a Ca2+-dependent phosphorylation of the T H molecule (George et al., 1982). Third, a later induction of additional T H molecules results from more prolonged periods of electrical stimulation (Zigmond, 1980) or from drug treatments that lead to sustained demands for increased synthesis in CA-containing cells (Reis et al., 1975; Thoenen et al., 1969; Fluharty et al., 1982). 2. Inactivation After monoamines are released into the synaptic cleft, their principal route of inactivation is by reuptake into the nerve terminals (Iversen, 197 1). After monoamine-containing neurons have been axotomized, the nerve terminals distal to the injury cease releasing transmitter immediately. However, the high-affinity uptake sites that concentrate monoamines in the nerve terminals continue to function until the anterograde degeneration affects the integrity of the nerve terminal membrane (typically, at least 24 hr postoperatively). This lag between the time at which release ceases and uptake sites are lost can limit the rate of recovery from partial injury to monoaminergic pathways, as shown elegantly by Zigmond et al. (1981). The pineal gland is innervated by noradrenergic axons whose cell bodies reside in the left and right superior cervical ganglia (SCG). These noradrenergic neurons regulate the large circadian rhythm of pineal serotonin N-acetyltransferase (NATase) activity, since this rhythm is largely abolished by bilateral SCG removal (denervation) or by removal of the cervical sympathetic trunk innervating each SCG (decentralization). After unilateral pineal denervation the NATase rhythm is depressed during the initial 24-hr cycle but its magnitude is fully restored by the second day postoperatively. In contrast, unilateral decentralization leads to a major decrease in NATase rhythmicity that is not reversed by day 3 . Zigmond et al., (1981) hypothesized that after unilateral decentralization and during the first cycle following unilateral denervation the uptake capacity of the inactive neurons persists, thereby limiting the effectiveness of NE released from the active terminals. During the second day after denervation, however, the degeneration of the terminals and attendant loss of uptake sites enhances the postsynaptic action of the NE released by the active terminals. In support of this view, administration of the NE uptake blocker desmethylimipramine (DMI) to unilaterally

RECOVERY OF FUNCTION AFTER BRAIN IN.JURY

227

decentralized animals led to a restoration of NATase activity (Zigmond et al., 1981; Fig. 7 ) . In addition to providing evidence for the role of monoamine uptake sites in recovery of function after partial lesions, these experiments demonstrate that a monoamine molecule can be taken up not only by the neuron that releases it but also by neighboring neurons.

3. Postsynaptic Supersensitivity Cannon and Rosenblueth (1949) suggested that a loss of nervous input renders tissues supersensitive to naturally occurring substances and drugs. This “law of denervation” includes smooth, cardiac, and skeletal muscle, neurons of the central and peripheral nervous systems, and the pineal gland. Because denervated tissues typically require a

I

T

Sham

Saline

1

T

Sham DMI

Unil.

Unil.

Dec.

Dec.

Bil. Dec.

Surgery

None

DMI

DMI

Drug

FIG. 7. Effect of desmethylimipramine (DMI) on pineal N-acetyltransferase activity in normal and decentralized animals. Rats were sham-operated (Sham), unilaterally decentralized (Unil. Dec.), or bilaterally decentralized (Bil. Dec.). The next day they were injected with saline or DMI or were untreated. All rats were killed 5-8 hr after the onset of darkness. Rats given DMI were injected ip with this drug 30 min before the onset of darkness (20 mg/kg) and again 5 hr after the onset of darkness (10 mgkg). Two hours later they were killed and enzyme activity was measured. [Adapted from Zigmond, R., Baldwin, C., and Bowers, C. W. (1981). Proc. Nutl. Acud. Sci. U.S.A. 78, 3959-3963.1

228

JOHN F. MARSHALL

smaller quantity of agonist to elicit a threshold response (without necessarily changing the maximal response), Trendelenburg (1963) indicated that supersensitivity is best quantified by the horizontal shift of the doseresponse curve. Trendelenburg (1963) identified both pre- and postsynaptic components of denervation supersensitivity. Those mechanisms that change the concentration of the agonist at the postsynaptic membrane are presynaptic; those that alter the response of the tissue to a particular concentration of agonist are postsynaptic. For example, removal of the sympathetic input to the cat nictitating membrane causes it to become supersensitive to the application of NE for two reasons. First, adrenergic nerve terminals, which would otherwise take up and remove NE from the synaptic cleft, degenerate. Second, the membrane becomes more sensitive to NE. An example of how presynaptic mechanisms may contribute to the recovery of pineal function after partial adrenergic denervation has been given. T h e remainder of this section considers only postsynaptic supersensitivity. Denervation-induced postsynaptic supersensitivity may well be a special case of a more general principle: the regulation of tissue sensitivity by the rate of neurotransmission. Thus, supersensitivity occurs in smooth and skeletal muscles whose neural inputs have been silenced, even when those afferent fibers are structurally intact. Also, skeletal muscles or CNS tissue can become subsensitive after chronic exposure to drugs that enhance transmission at neuromuscular or central synapses (Chang et al., 1973; Martres et al., 1975). Research on supersensitivity has focused on three tissues: skeletal muscle, smooth muscle, and CNS structures. For smooth and skeletal muscles several methods may be used to quantify the extent of pharmacological supersensitivity. These include measuring (1) the extent of contractile force in response to systemic or bath application of an agonist, or (2) the electrophysiological response of the muscle cell to iontophoretic application of the agonist. Iontophoresis has also proven useful within the CNS, as have biochemical measures of cellular sensitivity (e.g., catecholamine-stimulated adenylate cyclase). When the CNS target of a drug is well established, behavioral measures of sensitivity are also valuable. Denervation of skeletal muscle or CNS regions can also lead to marked changes in the density of membrane binding sites for ligands of the endogenous transmitter. In skeletal muscle there is a quantitative relationship between sensitivity of the muscle fiber to ACh iontophoresis and its binding of a-[1251]bungarotoxin (Hartzell and Fambrough,

RECOVERY OF FUNCTION AFTER BRAIN INJURY

229

1980). However, the postsynaptic supersensitivity of smooth muscle appears unrelated to membrane receptor changes (Fleming, 1976). Therefore, changes in membrane binding sites may underlie some instances of supersensitivity but cannot be used uncritically as an index of tissue sensitivity. Two issues have dominated research concerning the regulation of postsynaptic sensitivity. First, what is the stimulus (or stimuli) that acts on the postsynaptic cell to regulate the extent of its sensitivity? Second, what postsynaptic cellular events mediate the change in sensitivity? a. Stimuli Regulating Sensitivity. Although the sensitivity of denervated skeletal muscle may be controlled by several stimuli, recent research has focused on distinguishing between the roles of (1) muscle contractile activity, and (2) trophic factors transported by the nerve. Many investigations have supported the importance of decreased muscle activity. Whereas denervation of the rat diaphragm o r soleus muscles leads to an enhanced ACh sensitivity in the extrajunctional regions of the muscle within a few days, this spread of sensitivity can be limited if electrical stimulation is intermittently applied to the distal segment of the cat nerve or to the denervated muscle Uones and VrbovA, 1971; Drachman and Witzke, 1972; Lqmo and Rosenthal, 1972). Just 10 sec of 10-Hz stimuli applied once every 2.75 hr is sufficient to reduce the sensitivity of the denervated soleus to that of innervated fibers (L@mo and Westgaard, 1975). Also, procedures that chronically silence the muscle while leaving the nerve structurally intact (e.g., nerve anesthesia or tetrodotoxin injections, or intramuscular injections of d-tubocurarine, botulinum toxin, hemicholinium-3, a- o r P-bungarotoxin) can result in an enhanced extrajunctional response to ACh iontophoresis and a proliferation of extrajunctional ACh receptors (Thesleff, 1960; Lqmo and Rosenthal, 1971; Berg and Hall, 1975; Chang et al., 1975; Lavoie et al., 1976; Pestronk et al., 1976a,b). Evidence for axonally transported factors comes from experiments (e.g., Tiedt et al., 1977; Warnick et al., 1977) in which the application of substances that block axon transport (colchicine, vinblastine) to the motor nerve induces an increased ACh sensitivity of the muscle. Because of the possible spread of colchicine from the site of application on the nerve to the muscle and because this drug may affect the nerve in ways other than by blocking axon transport, the interpretation of those experiments has been controversial (Fambrough, 1979). Even under favorable conditions, colchicine o r vinblastine application to the sciatic nerve produces an ACh supersensitivity in the soleus or extensor digitorum

230

JOHN F. MARSI-IAI.1.

longus muscles only 10-20% of that observed following denervation (Warnick et al., 1977), suggesting that the contribution of axonally transported trophic factors to the supersensitivity is likely to be small. T h e level of noradrenergic receptor activation appears to be very important in controlling the sensitivity of smooth muscle. For example, the time course of development of postsynaptic supersensitivity in the nictitating membrane is similar when it is induced by denervation, decentralization, or chronic administration of the catecholamine-depleting drug reserpine (Fleming et al., 1973). Furthermore, postsynaptic supersensitivity has been produced by adrenergic blocking drugs, botulinum toxin, or sympathetic ganglionic blockers (Emmelin, 1961; Trendelenburg and Weiner, 1962). The sensitivity of central neurons also appears to be importantly modulated by the level of receptor activation. For instance, destroying the dopaminergic innervation of the neostriatum by an intracerebral 6OHDA injection results in behavioral supersensitivity to dopamine receptor stimulants, increased neostriatal DA-stimulated adenylate cyclase activity, enhanced electrophysiological response of neostriatal neurons to DA iontophoresis, and an increased binding of ligands for the DA receptor to neostriatal membranes (Ungerstedt, 1971; Feltz and deChamplain, 1972; Schoenfeld and Uretsky, 1972; Mishra et al., 1974, 1978; Yarbrough and Kostopoulos, 1975; Creese et al., 1977; Fig. 8). All of these effects can be produced without neuronal degeneration by chronic administration of DA receptor blocking agents (Tarsy and Baldessarini, 1973; Gianutsos et al., 1974; Iwatsubo and Clouet, 1975; Yarbrough, 1975; Burt et al., 1977). Furthermore, chronic treatment with the DA precursor, L-dopa, prevents the supersensitivity otherwise resulting from DA insufficiency (Gudelsky et al., 1975; Friedhoff et al., 1977; Ezrin-Waters and Seeman, 1978). Also, chronic administration of drugs that directly stimulate DA receptors or induce the release of DA into the synaptic cleft decreases the number of neostriatal binding sites for ligands of the DA receptor (Quik and Iversen, 1978; Kaneno and Shimazono, 1981). Destroying the noradrenergic innervation of the neocortex by intracerebral 6-OHDA injections results in increased cortical NE-stimulated adenylate cyclase activity and elevated binding of ligands for the padrenoceptor to cortical membranes (Huang et al., 1973; Kalisker et al., 1973; Harden et al., 1977; Skolnick et al., 1978; Minneman et al., 1979; U’Prichard et al., 1980). Similar changes, without neuronal degeneration, are produced by chronic reserpine treatment (French et al., 1974; Baudry et al., 1976; Palmer et al., 1976; Sharma et al., 1981). Opposite

FIG.8. Autoradiograph, using tritium-sensitive film, depicting the proliferation of ipsilateral neostriatal [3H]spiroperidol binding sites in vitro 28 days after left ventral tegmental6-OHDA injection. The brain section was incubated in 37°C buffer containing 1.1 nMISH]spiroperidol, washed and exposed to film for 4 weeks. ["H]Spiroperidol binding to dopamine D-2 receptors is elevated in striatum denervated of its DA input.

232

JOHN F. MARSHALL

changes result from chronic treatment with drugs that increase the release of NE into the synaptic cleft or decrease its uptake into the presynaptic terminal (Martres et al., 1975; Banerjee et al., 1977; Sarai et al., 1978; Minneman et al., 1979; Hal Manier et al., 1980). b. Cellular Events Mediating Supersensitivity Innervated skeletal muscle shows a high degree of ACh sensitivity at the end plate and little or no ACh sensitivity of the surrounding membrane. In denervated muscle the sensitivity of the extrajunctional membrane to ACh can increase more than 1000-fold (Axelsson and Thesleff, 1959; Miledi, 1960). This extrajunctional ACh supersensitivity is explained by the dramatic increase in the number of nicotinic ACh receptors in the extrajunctional region of the muscle fiber. In the rat diaphragm, for example, denervation increases the density of extrajunctional receptors labeled by a['251]bungarotoxinfrom fewer than 5 per pm2 to 1695 per pm2 (Hartzell and Fambrough, 1980). The added receptors are synthesized de novo and inserted into the membrane (Fambrough, 1979). The cellular events underlying supersensitivity in smooth muscle appear to be quite different. The postsynaptic supersensitivity is relatively nonspecific; i.e., the supersensitivity to NE, ACh, histamine, K+, and other transmitters and ions does not vary by more than one order of magnitude (Hudgins and Fleming, 1966; Westfall, 1970; Westfall et al., 1972; Fleming et al., 1973). No difference is found between the adrenergic receptor binding of normal and denervated smooth muscle (Page and Neufeld, 1978; Seidel et al., 1982). Thus, the cellular mechanism of the supersensitivity appears to occur after the receptor occupancy (Fleming et al., 1973). Two candidates have been suggested: ( 1 ) a 7-10 mV depolarizing shift in the resting membrane potential resulting from a decreased activity of the membrane Na+, K+ electrogenic pump (Urquilla et al., 1978; Gerthoffer et al., 1979; Wong et al., 1981), and (2) a redistribution of intracellular calcium (Garrett and Carrier, 197 1 ; Carrier, 1975; Westfall, 1977). Within the CNS, the finding that denervation or pharmacological blockade of DA, NE, 5-HT OF muscarinic ACh receptors induces them to increase in number suggests that a receptor proliferation may contribute to the postsynaptic supersensitivity. However, events subsequent to receptor occupancy may also contribute to the supersensitivity of denervated CNS neurons. For example, after injury to nigrostriatal dopaminergic neurons, the denervated neostriatum shows pharmacological supersensitivity not only to DA and DA receptor stimulants but also to NE, 5-HT, 5'-AMP, and 5'-GMP (Costall et al., 1976; Satoh et al., 1976; Setler et ad., 1978; Neve, Grisham, Hansen, and Marshall, unpublished findings).

RECOVERY OF FUNCTION AFTER BRAIN INJURY

233

B. NEUROCHEMICAL ADAPTATIONS AND RECOVERY OF FUNCTION Cannon and Rosenblueth (1949) and Stavraky (1961) suggested that a denervation-induced supersensitivity to transmitter may be one mechanism by which animals with nervous system injury recover from the functional consequences of that damage. This explanation requires that there be a source of transmitter remaining after the injury that is able to reach the denervated cells. I n those model systems in which postsynaptic supersensitivity does appear to contribute to recovery of function, a small proportion of the original innervation of the target tissue remains intact, providing a source of neurotransmitter to drive the postsynaptic cells. The other forms of neurochemical plasticity described previously (enhanced synthesis and release of transmitter or decreased rate of inactivation) have been considered as possible mechanisms of recovery by relatively few investigators (Stricker and Zigmond, 1976; Zigmond et al., 1981). 1 . Spinal Serotonergw Innervation The activity at spinal serotonergic synapses importantly modulates certain nociceptive reflexes (e.g., heat-induced tail flick). After intrathecal injections of 5,6-DHT that produce extensive but incomplete damage to the spinal serotonergic innervation, rats show an initial decline in the latency of the tail flick, followed by complete recovery within 2 weeks. This recovery depends upon spinal 5-HT receptor activity because it is reversed by intrathecal administration of the serotonin receptor antagonist metergoline (Berge et al., 1983). Although the descending 5-HT-containing projections do reinnervate the lumbar spinal cord between the second and sixth months after their injury by 5,6-DHT (Nygren et al., 1974), the spontaneous normalization of the nociceptive reflex latency begins within 3 days of the injury (Berge et aE., 1983). Within several days of interrupting the spinal 5-HT innvervation, the spinal cord becomes supersensitive to 5-HT precursors or receptor stimulants (Nygren et al., 1974; Barbeau and Bedard, 1981). The temporal course of this supersensitivity closely matches that of the normalization of the nociceptive reflex. These findings suggested to Berge et al., (1983) that the reflex normalization depends upon neurochemical adaptations occurring at the surviving serotonergic synapses rather than on the growth of 5-HT-containing axons.

2. Mesostriatal Dopaminergzc Projection a. The Model System. When the ascending catecholamine-containing projections are interrupted by electrolytic lesions, knife cuts, o r 6-hy-

234

JOHN F. MARSHALL

droxydopamine injections, animals develop impairments in several behaviors, including feeding, drinking, sexual and aggressive behaviors, exploratory locomotion, and sensorimotor capacities (see Ungerstedt, 1974; Marshall and Teitelbaum, 1977; Fink and Smith, 1979; Stricker and Zigmond, 1976). Many animals with such brain injury show a remarkable recovery from the initial behavioral deficits. One behavioral impairment resulting from this brain damage that is particularly amenable to study is somatosensory inattention, or the failure to orient toward tactile stimuli applied to the body surface (Marshall et al., 1971, 1974; Marshall, 1979). In rats with unilateral injury this impairment is restricted to the contralateral body surface. The somatosensory inattention results specifically from the interruption of mesostriatal dopaminergic projections. First, the impairment occurs after injections of 6-OHDA into the DA-containing cell bodies of the substantia nigra or into any of several loci along their axons (Marshall et al., 1974; Marshall, 1979; Kozlowski and Marshall, 1980). Second, after such injections the extent of the impairment in somatosensory localization correlates highly with the degree of neostriatal DA depletion (Marshall, 1979; Kozlowski and Marshall, 1981). Third, injections o f 6-OHDA into the neostriatum that significantly decrease the dopaminergic innervation of this structure without damaging the DA-containing terminals of limbic or cortical regions result in impairments in somatosensory localization, whereas injections of this neurotoxin directly into limbic or cortical sites do not (Marshall et al., 1980; Dunnett and Iversen, 1982). Fourth, in rats made inattentive to tactile stimuli by prior mesostriatal injury, somatosensory localization is reinstated either by transplants of fetal nigral dopaminergic cells that reinnervate the lateral neostriatum (Dunnett et al., 1981) o r by intrastriatal injections of the DA receptor stimulant apomorphine (Marshall et al., 1980). b. Recovery of Somatosensory Localization. The recovery of somatosensory localization in rats with mesostriatal 6-OHDA injections assumes a characteristic temporal and somatotopic organization. As the rats recover, orientation to touch of the snout occurs first (Fig. 9). Only later does touch of progressively more caudal regions of the affected side elicit head orientation (Marshall et al., 1971; Marshall, 1979). The recovery begins by the fourth day postoperatively and is complete by the twenty-eighth day (Marshall, 1979; Kozlowski and Marshall, 1981). The recovery of somatosensory localization depends upon the survival and continued functioning of a small population (5-10%) of the dopaminergic terminals in the neostriaturn. When the mesostriatal damage is so extensive as to deplete the DA content of this structure by more than 95%, recovery never occurs. Rats that do recover have less exten-

RECOVERY OF FUNCTION AFTER BRAIN INJURY

3 dllys

7days

5 days

42 days

235

Group M e a n

0 0-0.9 1.0-1.9

=

2.0 - 2.9

3.0-3.9

FIG. 9. Somatotopic organization of the recovery of orientation toward tactile stimuli on the body surface contralateral to extensive, but incomplete, injury to the mesostriatal dopaminergic projection. The extent of orientation to each body region was rated from 0 to 4, where 0 = no turning toward the side contacted, and 4 = precise localization and biting of the tactile probe. Shadings depict the average scores for a group of 15 rats.

sive striatal DA depletions (Marshall, 1979; Kozlowski and Marshall, 1981). In rats that recover spontaneously from their impaired orientation to touch, the somatosensory impairments are reinstated by administering low doses of a DA receptor blocking agent, spiroperidol, or a catecholamine synthesis inhibitor, a-methyl-para-tyrosine, suggesting that a continued synthesis, release, and receptor action of DA at this small population of residual dopaminergic synapses is necessary for recovery (Marshall, 1979). The findings that intrastriatal apomorphine injections o r nigral transplants restore orientation (Marshall et al., 1980; Dunnett et al., 1981) indicate that an activation of denervated neostriatal DA receptors is also sufficient for recovery. c. Metabolic Cohelates of Recovery. The [ 14C]2-deoxyglucose(2DG) autoradiographic procedure (Sokoloff et al., 1977) has been used to visualize the metabolic activity of the neostriatum during recovery from the somatosensory impairments. Rats given unilateral mesostriatal6-OHDA injections that have a marked impairment in contralateral somatosensory localization at the third day postoperatively show significant hemispheric asymmetries of [14C]2DGuptake at this time. The uptake of this glucose analog is depressed ipsilaterally to the lesion in forebrain structures that normally receive a dense dopaminergic innervation (i.e., the neostriatum and limbic forebrain structures-Schwartz, 1978; Kozlowski and Marshall, 1980), while its uptake is enhanced ipsilaterally

236

JOHN F. MARSHALL

in basal gangliar structures that receive neostriatal efferent fibers (Le., globus pallidus, entopenduncular nucleus, and substantia nigra pars reticulata-Schwartz, 1978; Kozlowski and Marshall, 1980; Wooten and Collins, 1981). Most of the lesion-induced asymmetries of [I4C]2DGincorporation are reversed by the instrastriatal injection of apomorphine (a procedure that also restores somatosensory localization), indicating that they result from decreased neostriatal dopaminergic receptor stimulation (Kozlowski and Marshall, 1980). This pattern of altered [14C]2DGuptake is also seen at G weeks postoperatively in rats that have shown no recovery of somatosensory orientation postoperatively. However, in animals that recover spontaneously, the hemispheric asymmetries in the anterior neostriatum, globus pallidus, and substantia nigra pars reticulata are no longer evident at G weeks postoperatively. Moreover, the time course with which these basal gangliar asymmetries are reversed is quite similar to that of the behavioral recovery (Kozlowski and Marshall, 1983, Fig. 10). d. Compensatory Changes at Residual Neostriatal Dopaminerpc Synapses. The behavioral improvement and normalization of neostriatal [14C]2DG uptake are not likely to be due to a regrowth or collateral sprouting of dopaminergic axons within this structure. Electron microscopic, fluorescence microscopic, and biochemical experiments have failed to find evidence for a postsynaptic normalization of DA terminal number after 6OHDA injections along these axons (Hokfelt and Ungerstedt, 1973; Neve et al., 1982; Reis et al., 1978). The recovery of somatosensory localization is likely to depend upon an increased functioning within the small population (5- 10%)of surviving dopaminergic synapses. This recovery could be mediated by an elevated synthesis and release of DA by the residual dopaminergic terminals, an elevated response of the postsynaptic neurons to the available DA, or both (Stricker and Zigmond, 1976). From this perspective, rats with greater than 95% destruction of this neuronal population fail to recover from their somatosensory impairments because these pre- and postsynaptic compensations are insufficient to restore the dopaminergic receptor activity of the neostriatum to a level compatible with this behavior. To determine the contribution of pre- and postsynaptic events to the recovery sequence, my colleagues and I have adopted the strategy of determining their postoperative time courses. The well-defined time course of the recovery of somatosensory localization leads to strong predictions regarding the temporal limits of the neural events hypothesized to mediate them. The time course of dp*--.lopmentof postsynaptic super-

RECOVERY OF FUNCTION AFTER BRAIN 1N.JURY

j

5

I4

237

4.2

Days Postoperatively FIG. 10. Time course for recovery of contralateral somatosensory localization ( 0 )compared to the time courses for the normalization of [I4C]2DG uptake in the anterior nrostriatum (A),globus pallidus (A), ventral internal capsule (m), and substantia nigra pars reticulata (0)of recovering rats. Values represent the asymmetries in each of these measures at 3, 5, 14, and 42 days postoperatively as a percent of the asymmetry at 3 days. Negative values represent a reversal of the direction of the asymmetry. T h e asymmetry in somatosensory localization is not fully reversed by 42 days because of residual impairments in orienting to touch of the most caudal points. [Reprinted from Kozlowski, M. R., and Marshall, J. F. (1983).Brain Res. 259,249-260, with permission from Elsevier Biomedical Press.]

sensitivity to DA agonists corresponds closely to that of the behavioral recovery (Neve et al., 1982; Fig. 11). This correspondence suggests that the compensatory events critical for recovery are located on neostriatal elements postsynaptic to the residual dopaminergic innervation. In contrast, the time course of elevated DA synthesis matches that of recovery poorly (Altar and Marshall, unpublished findings, Fig. 1I), because it is maximal at the onset of the recovery (3 days postoperatively). Thus, an increased DA synthesis appears not to limit the rate at which recovery occurs.

238

JOHN F. MARSHALL

I:* /.if: 000

a

*om'

NC

-h 0

r O

000

7 x

. f"

(DZN.

FIG. 11. Time course of recovery of contralateral somatosensory localization (0)compared to the time course of the supersensitivity to the dopamine receptor stimulant apomorphine (x). Also depicted is the time course of elevated dopamine synthesis (A), as indicated by the accumulation of DOPA after administration of the DOPA decarboxylase inhibitor NSD-1015 (100 mg/kg, ip) 30 min before sacrifice.

What postsynaptic changes mediate this recovery? Neve et al. (1982) used the in vivv binding of [3H]spiroperidol to the neostriatum to label one class of dopamine receptors in this structure (D-2) that does not stimulate adenylate cyclase activity (Kebabian and Calne, 1979; Seeman, 1980). The binding of this ligand to the denervated neostriatum was significantly elevated by 4 days after extensive mesostriatal injury, and the denervation-induced elevation in binding continued to increase during the first month postoperatively. Also, using autoradiography to visualize the binding of ["Hlspiroperidol to rat forebrain sections in vztrv, Neve et al. (1983)observed a denervation-induced increase in [4H]spiroperidol binding to D-2 sites in the neostriatum within the first week postoperatively. These results are consistent with the hypothesis that an increased density of D-2 receptors on neostriatal cells mediates the observed postsynaptic supersensitivity and the recovery of somatosensory localization. The findings do not eliminate the possibility that changes in neostriatal neurons subsequent to the binding of DA to its receptors may also contribute. Possible mechanisms include alterations in the resting membrane potential of neostriatal cells (Yarbrough, 1976; Kamata et al., 1980; van der Krogt and Belfroid, 1980) or a redistribution of intracellular Ca'' or Ca2+-dependentregulatory proteins (Gnegy et al., 1977a,h; Gnegy and Lau, 1980).

RECOVERY OF FUNCTION AFTER BRAIN INJURY

239

VI. Conclusions and Future Directions

During the past 15 years many investigators have been interested in providing neurobiological explanations for the recovery of function that follows specific types of CNS injury. Much of this research has focused on the behavioral consequences of injury-induced axon growth, and two approaches have yielded significant advances. First, the extensive axonal growth that can occur following brain injury in the neonatal animal may help determine the motor function of that animal as an adult. After neonatal pyramidotomy this growth leads to a later sparing of forepaw dexterity (Reh and Kalil, 1982), whereas the anomalous retinotectal connections that result from neonatal collicular injury can result in misdirected head orientation toward visual cues as an adult (Schneider, 1973). Second, the axonal outgrowth that occurs from fetal neurons transplanted into the brains of adult rats can result in the replacement of a severed pathway and concomitant improvements in motor functions (Dunnett et al., 1981) o r the performance of a spatial task (Low et al., 1982). These experiments emphasize the capacity of immature nerve cells for axonal growth and the formation of functional circuits. What can be said concerning the behavioral consequences of axonal growth in adult CNS neurons? T h e only system in which strong arguments can be made is the cat spinal cord. Sectioning of the dorsal roots innervating one hindlimb leads to an increased distribution of the axons descending to the ipsilateral lumbar spinal cord, a hyperactivity of descending reflex controls over the denervated limb, and an improvement in the use of that limb during locomotion (Goldberger and Murray, 1978). Of course, full recovery never occurs because the intrinsic limb reflexes are never restored. However, the locomotor recovery of the denervated hindlimb indicates that hindlimb afferents and descending axons share control over the interneuron-motoneuron pool responsible for this coordinated behavior such that one source of inputs can substitute for the other. A major goal for future research in this field is to identify other systems in the adult CNS in which a collateral sprouting after injury has behavioral consequences (facilitative or deleterious) so that the interaction between afferent sources can be studied further. Compared to these morphological approaches, less attention has been paid to the possibility that neurochemical adaptations at surviving synapses may mediate behavioral recovery. Although the idea that neurochemical changes such as denervation supersensitivity may mediate

240

JOHN F. MARSHALL

recovery is not new (Cannon and Rosenblueth, 1949), direct experimental support for this hypothesis is. Alterations in transmitter synthesis, release, and postsynaptic sensitivity mediate the recovery of ingestive behaviors and sensorimotor functions after injury to brain catecholaminergic neurons (Stricker and Zigmond, 1976; Marshall, 1980), and the normalization of spinal nociceptive reflexes after injury to bulbospinal serotonergic projections suggests a similar conclusion (Berger et d.,1983). For the recovery of somatosensory localization after nigrostriatal injury, a supersensitivity of the neostriatal neurons to dopamine appears to determine the rate at which this behavior is restored (Neve et al., 1982), and a proliferation of dopamine receptors of the D-2 class may mediate this enhanced cellular sensitivity. Neurochemical adaptations may underlie the recovery that occurs after damage to many CNS systems. For instance, the recovery of somatosensory localization after nigrostriatal injury has a characteristic time course (beginning by 3-4 days, complete by 2-3 weeks) that is similar to the time course of recovery of other behaviors after other forms of brain injury (e.g., Loesche and Steward, 1977; Berge et al., 1983). This similar time course may reflect the operation of a common underlying process (i.e., the development of a postsynaptic supersensitivity), and a contribution of postsynaptic supersensitivity to the behavioral improvement should be suspected whenever this time course of behavioral recovery is encountered. Additional evidence which suggests that neurochemical adaptations may contribute to other instances of recovery has been reviewed in Section II,B (Redundancy). After damage to several fiber systems, recovery occurs only if some small population of the axons escapes injury. I have argued that this relationship between the extent of damage to a population of neurons and recovery may suggest that neurochemical adaptations within the surviving axons or their postsynaptic cells subserve recovery. The dependence of recovery on the survival of a small population of axons within the injured projection is not expected if the recovery is mediated by the collateral sprouting of axons from another source into the denervated region. Some investigators have argued that supersensitivity may alter only the central excitatory state of the denervated neurons, thereby mediating generalized aspects of recovery, but that the recovery of patterned motor responses must be mediated by the more selective process of axon sprouting (Goldberger, 1974). This view of supersensitivity implies a lack of specificity of the partially denervated neurons to the neurohumoral agents reaching their surface. However, recent evidence suggests that the supersensitivity of denervated CNS neurons is mediated, at least in part, by a proliferation of receptors for the depleted neurotransmit-

RECOVERY OF FUNCTION AFTEK B R A I N IN-JURY

24 1

ter. This proliferation can be quite specific. Thus, after nigrostriatal injury the density of D-2 receptors in the neostriatum increases significantly, but the binding of ligands for the muscarinic or nicotinic cholinergic receptors, p-adrenoceptors, opiate, and GABA receptors does not increase in this structure (Kato et al., 1978; McGeer et al., 1979; Reisine et al., 1979; Schallert et al., 1980; Suga, 1980). One important future direction is to determine the cellular processes responsible for the development of denervation supersensitivity in other populations of CNS neurons in order to determine whether this degree of specificity is commonplace. At present, however, there is no reason to suppose that the supersensitivity of denervated CNS neurons cannot be a transmitterselective process responsible for the recovery of patterned responses in many systems. Acknowledgments

This article was written while the author was a fellow in Neuroscience of the Alfred P. Sloan Foundation. Some of the author’s research described herein was supported by National Science Foudation. References

Acheson, A. L., and Zigmond, M. J. (1981).J. Neurosci. 1, 493-504. Agid, Y., Javoy, F., and Glowinksi,J. (1973). Nature (London) New Biol. 245, 150-151. Aguilar, C. E., Bisby, M. A., Cooper, E., and Diamond, J. (1973).J. Physiol. (London) 234, 449-464. Axelsson, J., and Thesleff, S. (1959).J. Physiol. (London) 147, 178-193. Banerjee, S. P., Kung, L. S., Riggi, S. J., and Chanda, S. K. (1977). Nature (London) 268, 455-456. Barbeau, H., and Bedard, P. (1981). Neurophannacology 20, 61 1-616. Barnes, C. D., Joynt, R. F., and Schottelius, B. A. (1962). Am. J . Physiol. 203, 1 13-1 1 16. Baudry, M., Martres, M.-P., and Schwartz, J.4. (1976). Brain Res. 116, 1 1 1 - 24. Baumann, T. P., and Spear, P. D. (1977). Brain Res. 138,445-468. Beck, C. H., and Chambers, W. W. (1970).J . Comp. Physiol. Psychol. 70, 1-22 Berg, D. K., and Hall, Z. W. (1975)..]. Physiol. (London) 244, 659-676. Berge, 0 . - G . , Fasmer, 0. B., Flatmark, T., and Hole, K. (1983).Pharmacol. Biochem. Behau. 18, 637-643. Bernheimer, H., and Hornykiewicz, 0. (1962). Naunyn-Schmiedeberg’s Arch. Exp. Pathol. 243,295-296. Bernstein, M. E., and Bernstein, J. J. (1973). Znl. J . Neurosci. 5 , 15-26. Bjorklund, A., and Lindvall, 0. (1979). Brain Res. 171, 271-293. Bjorklund, A., and Stenevi, U. (1971). Brain Res. 31, 1-20. Bjorklund, A., and Wiklund, L. (1980). Bruin Res. 191, 109-127. Bjorklund, A., Nobin, A., and Stenevi, U. (1973). Brain Res. 50, 214-220. Rjorklund, A., Johansson, B., Stenevi, U., and Svendgaard, N.-A. (1975a).Nature (London) 253,446-448. Bjorklund, A., Baumgarten, H. G., Lachenmayer, L., and Rosengren, E. (197513). Cell Tissue Res. 161, 145-155.

242

JOHN P. MARSHALL

Brenowitz, G. L., and Pubols, L. M. (1981). Brain Res. 216, 45-59. Brown, M. C., and Ironton, R. (1977). Nature (London) 265,459-461. Brown, M. C., and Ironton, R. (1978).J . Physiol. (London) 278, 325-348. Brown, M. C., Goodwin, G. M., and Ironton, R. (1977).J. Physiol. (London) 267,42P-43P. Bunney, B. S., Walters, J. R., Roth, R. H., and Aghajanian, G. K. (1973).J. Pharmacol. Ex$. Ther. 185, 560-57 1 . Burt, D. R., Creese, I., and Snyder, S. H. (1977). Science 196, 326-328. Cannon, W. B., and Rosenblueth, A. (1949). “The Supersensitivity of Denervated Structures: A Law of Denervation.” Macmillan, New York. Carlsson, A., and Lindqvist, M. (1963). Acta Pharmacol. Toxicol. 20, 140-144. Carlsson, A., Kehr, W., and Lindqvist, M. (1976).J. Neural Transmission 39, 1-19. Carrier, 0. (1975). Fed. Proc., Fed. Am. Soc. Exp. Biol. 34, 1975-1980. Chang, C. C., Chen, T. F., and Chuang, S.-T. (1973).J. Physzol. (London) 230, 613-618. Chang, C. C., Chuang, S.-T., and Huang, M. C. (1975).J. Physiol. (London) 250, 161-173. Costall, B., Naylor, R. J., and Pycock, C. (1976). Eur. J . Pharmacol. 35, 275-284. Cotman, C. W., and Nadler, J. V. (1978). In “Neuronal Plasticity” (C. W. Cotman, ed.), pp. 227-271. Raven, New York. Cotman, C. W., Nieto-Sampedro, M., and Harris, E. W. (1981). Physiol. Rev. 61,684-784. Creese, I., Burt, D. R., and Snyder, S. H. (1977). Science 197, 596-598. David, S., and Aguayo, A. J. (1981). Science 214, 931-933. Devor, M., and Wall, P. D. (1978). Nature (London) 276, 75-76. Devor, M., Schonfeld, D., Seltzer, Z., and Wall, P. D. (1979).J. Comp. Neurol. 185, 211220. Diamond, M. C., Lindner, B., Johnson, R., Bennett, E. L., and Rosenzweig, M. R. (1975).J. Neurosci. Res. 1, 109-119. Diamond, J., Cooper, E., Turner, C., and Macintyre, L. (1976). Science 193, 371-377. Drachman, D. B., and Witzke, F. (1972). Science 176, 514-516. Dunnett, S. B., and Iversen, S. D. (1982). Brain Res. 248, 121-127. Dunnett, S. B., Bjorklund, A., Stenevi, U., and Iversen, S. D. (1981).Brain Res. 229, 209217. Edds, M. V. (1953). Q. Rev. Bzol. 28, 260-276. Emmelin, N. (1961). Pharmacol. Rev. 13, 17-37. Emson, P. C., Bjorklund, A., and Stenevi, U. (1977). Brain Res. 135, 87-105. Ezrin-Waters, C., and Seeman, P. (1978). Life Sci. 22, 1027-1032. Fambrough, D. M. (1979). Physiol. Rev. 59, 165-227. Feltz, P., and dechamplain, J. (1972). Brain Res. 43, 601-605. Finger, S., and Stein, D. G. (1982). “Brain Damage and Recovery: Research and Clinical Perspectives.” Academic Press, New York. Fink, J. S., and Smith, G. P. (1979).J. Comp. Physiol. Psychol. 93, 34-65. Fleming, W. W. (1976). Rev. Neurosci. 2, 43-90. Fleming, W. W., McPhillips, J. J., and Westfall, D. P. (1973).Ergeb. Physiol. Biol. Chem. Exp. Pharmakol. 68, 56- 119. Flourens, P. (1824). “Recherches expkrimentales sur les proprietks et les fonctions d u systhme nerveux dans les animaux verthbres.” Crevot, Paris. Fluharty, S. J., Snyder, G. L., Stricker, E. M., and Zigmond, M. J. (1982). Neurosci. Abstr. 8, 190. French, S. W., Palmer, D. S., and Narod, M. E. (1974).Fed. Proc., Fed. Am. SOC.Exp. Biol. 34, 297. Friedhoff, A. J., Bonnet, K., and Rosengarten, H. (1977). Res. Commun. Chem. Pathol. Pharmacol. 16, 41 1-423.

RECOVERY OF FUNCTION AFTER BRAIN INJURY

243

Frommer, G. P. (1978). I n “Recovery from Brain Damage” (S. Finger, ed.), pp. 217-280. Plenum, New York. Galambos, R., Norton, T. T., and Frommer, G. P. (1967). Exp. Neural. 18, 8-25. Garrett, R. L., and Carrier, O., Jr. (1971). Eur.1. Pharmacol. 13, 306-311. George, R. J., Haycock, J. W., and Waymire, J. C. (1982). Neurosci. Abstr. 8, 817. Gerthoffer, W. T., Fedan, J. S., Westfall, D. P., Goto, K., and Fleming, W. W. (1979).J . Pharmacol. Exp. Ther. 210, 27-36. Gianutsos, G., Drawbaugh, R. B., Hynes, M. D., and Lal, H. (1974). Life Sci. 14,887-898. Gnegy, M. E., and Lau, Y. S. (1980). Neuropharmacology 19, 319-323. Gnegy, M. E., Lucchelli, A,, and Costa, E. (1977a). Naunyn-Schmiedeberg’sArch. Pharmacol. 301, 121-127. Gnegy, M., Uzonov, P., and Costa, E. (1977b).J. Phannacol. Exp. Ther. 202, 558-564. Goldberger, M. E. (1974). I n “Plasticity and Recovery of Function in the Central Nervous System” (D. G. Stein,J.J. Rosen, and N. Butters, eds.), pp. 265-337. Academic Press, New York. Goldberger, M. E., and Growden, J. H. (1973). Exp. Neurol. 39, 307-322. Goldberger, M. E., and Murray, M. (1974).J . Camp. Neural. 158, 37-54. Goldberger, M. E., and Murray, M. (1978). I n “Neuronal Plasticity” (C. W. Cotman, ed.), pp. 73-96. Raven, New York. Goldowitz, D., and Cotman, C. W. (1980). Brain Res. 181, 325-344. Graziadei, P. P. C., and Graziadei, G. A. M. (1978). In “Neuronal Plasticity” (C. W. Cotman, ed.), pp. 131-153. Raven, New York. Greenough, W. T., and Green, E. J. (1981). In “Aging: Biology and Behavior” (J. McGaugh, ed.), pp. 159-200. Academic Press, New York. Gudelsky, G. A., Thornburg, J. E., and Moore, K. E. (1975). Life Sci. 16, 1331-1338. Guth, L. (1956). Physiol. Rev. 36, 441-478. Guth, L. (1962). Exp. Neurol. 6, 129-141. Guth, L., and Bernstein, J. J. (1961). Exp. Neurol. 4, 59-69. Guth, L., and Windle, W. F. (1970). Exp. Neural. Suppl. 5, 1-43. Hal Manier, D., Gillespie, D. D., and Sulser, F. (1980). Nautzyn-Schmiedeberg’s Arch. PharnaC O ~ .313, 113-1 18. Harden, T. K., Wolfe, B. B., Sporn, J. R., Poulos, B. K., and Molinoff, P. B. (1977).J . Pharmacol. Exp. Ther. 203, 132-143. Harris, J. E., and Roth, R. H. (1971). Mol. Pharmacol. 7, 593-604. Hartzell, H. C., and Fambrough, D. M. (1972).J . Gen. Physiol. 60, 248-262. Hefti, F., Melamed, E., and Wurtman, R. J. (1980). Brain Res. 195, 123-137. Hokfelt, T., and Ungerstedt, U. (1973). Brain Res. 60, 269-297. Holland, R. L., and Brown, M. C. (1980). Science 207, 649-651. Horvat, J.-C. (1980). C . R . Acad. Sac. Paris 290, 127-130. Huang, M.,,Ho., A. K. S., and Daly, J. W. (1973). Mol. Pharmacol. 9, 711-717. Hudgins, P. M., and Fleming, W. W. (1966)./. Pharmacol. Exper. Ther. 153, 70-80. Iversen, L. L. (1971). B r . J . Pharmacol. 41, 571-591. Iwatsubo, K., and Clouet, D. H. (1975). Biochern. Pharmacol. 24, 1499-1503. Jackson, J. H. (1958). In “Selected Writings of.John Hughlings Jackson” (J. Taylor, ed.), Vol. 11, pp. 45-75. Basic Books, New York. Jackson, P. C., and Diamond, J. (1981). Science 214, 926-928. Jacobson, S. G., Eames, R. A,, and McDonald, W. I. (1979). Exp. Brain Res. 36, 491-508. Jones, R., and Vrbovi, G. (197 1). J . Physiol. (London) 216, 67P-68P. Juraska, J. M., Greenough, W. T., Elliott, C., Mack, K. J., and Berkowitz, R. (1980).Behav. Neural Biol. 29, 157-167.

244

J O H N F. MARSHALL

Kalil, K., and Reh, T. (1979). Science 205, 1158-1161. Kalil, K., and Reh, T. (1982).J. Comp. Neurol. 211, 265-275. Kalisker, A., Rutledge, O., and Perkins, J. P. (1973). Mol. Pharmucol. 9, 619-629. Kamata, K., Kamata, M., Ino, H., and Kameyama, T. (1980). Jpn. J. Phurmucol. 30, 401404. Kaneno, S., and Shimazono, Y. (1981). E u r . 1 . Pharmucol. 72, 101-105. Kao, C. C., Chang, L. W., and Bloodworth, J. M. B., Jr. (1977). Exp. Neurol. 54, 591-615. Kato, G., Carson, S., Kernel, M. L., Glowinski, J., and Giorguieff, M. F. (1978). Life Sci. 22, 1607-16 14. Kebabian, J. W., and Calne, D. B. (1979). Nature (London) 277, 93-96. Kerr, F. W. L. (1975). Exp. Neurol. 48, 16-31. Kjerulf, T. D., and Loeser, J. D. (1973). Exp. Neurol. 39, 70-85. Kjerulf, T. D., O’Neal, J. T., Calvin, W. H., Loeser, J. D., and Westrum, L. E. (1973). Exp. Neurol. 39, 86-102. Kozlowski, M. R., and Marshall, J. F. (1980). Brain Res. 197, 167-183. Kozlowski, M. R., and Marshall, J. F. (1981). Exp. Neurol. 74, 318-323. Kozlowski, M. R., and Marshall, J. F. (1983). Brain Res. 259, 249-260. Kromer, L. F., Bjorklund, A,, and Stenevi, U. (1981a). Bruin Res. 210, 153-171. Kromer, L. F., Bjorklund, A,, and Stenevi, U. (1981b). Bruin Res. 210, 173-200. Lashley, K. S. (1929). “Brain Mechanisms and Intelligence.” Univ. of Chicago Press, Chicago, Illinois. Lashley, K. S. (1938). J. New. Ment. Dis. 88, 733-755. Lavoie, P.-A., Collier, B., and Tenenhouse, A. (1976). Nature (London) 260, 349-350. Lawrence, D. G., and Kuypers, H. G. J. M. (1968). Brain 91, 1-14. Liu, C.-N., and Chambers, W. W. (1958). Arch. Neurol. Psychol. 79, 46-61. Loesche, J., and Steward, 0. (1977). Bruin Res. Bull. 2, 31-39. L@mo,T., and Rosenthal, J. (1971).J. Physiol. (London) 216, 52P-53P. L@mo,T., and Rosenthal, J. (1972).J. Physiol. (London) 221, 493-513. L@mo,T., and Slater, C. R. (1978).J. Physiol. (London) 275, 391-402. LGmo, T., and Westgaard, R. H. (1975).J. Physiol. (London) 252, 603-626. Low, W. C., Lewis, P. R., Bunch, S. T., Dunnett, S. B., Thomas, S. R., Iversen, S. D., Bjorklund, A., and Stenevi, U. (1982). Nature (London) 300, 260-262. Lynch, G., Matthews, D. A., Mosko, S., Parks, T., and Cotman, C. (1972). Bruin Res. 42, 3 1 1-318. Lynch, G., Deadwyler, S., and Cotman, C. (1973). Science 180, 1364-1366. Lynch, G., Stanfield, B., Parks, T., and Cotman, C. W. (1974). Brain Res. 69, 1-11. McGeer, P. L., McGeer, E. G., and Innanen, V. T. (1979). Bruin Res. 169, 433-441. McWilliams, J. R., and Lynch, G. (1978).J. Comp. Neurol. 180, 581-615. McWilliams, J. R., and Lynch, G. (1979).J. Comp. Neurol. 187, 191-198. Marshall, J. F. (1979). Brain Res. 177, 311-324. Marshall, J. F. (1980). I n “Neural Mechanisms of Goal-Directed Behavior” (R. F. Thompson, L. H. Hicks, and V. B. Shvyrkov, eds.), pp. 167-176. Academic Press, New York. Marshall, J. F., and Teitelbaum, P. (1977). Hundb. Psychophurmucol. 7, 20 1-229. Marshall, J. F., Turner, B. H., and Teitelbaum, P. (1971). Science 174, 523-525. Marshall, J. F., Richardson, J. S., and Teitelbaum, P. (1974).J. Comp. Physiol. Psychol. 87, 808-830. Marshall, J. F., Berrios, N., and Sawyer, S. (1980). J . Comp. Physiol. Psychol. 94, 833-846. Martres, M.-P., Baudry, M., and Schwartz, J.-C. (1975). Nature (London) 255, 731-733. Miledi, R. (1960).J. Physiol. (London) 151, 1-23, Millar, J., Basbaum, A. I., and Wall, P. D. (1976). Exp. Neurol. 50, 658-672.

RECOVERY OF FUNCTION AFTER BRAIN IN-JURY

245

Minneman, K. P., Dibner, M. D., Wolfe, B. B., and Molinoff, P. B. (1979). Science 204, 866-868. Mishra, R. K., Gardner, E. L., Katzman, R., and Makman, M. H. (1974). Proc. Natl. Acad. Sci. U.S.A. 71, 3883-3887. Mishra, R. K., Wong, Y.-W., Varmuza, S. L., and Tuff, L. (1978). Life Sci. 23, 443-446. Monakow, C., von (1914). “Die Lokalisation inder Crosshirnrinde und d e r Abhau der Funktion durch Korticale Herde.” Bermann, Wieshaden. Mountcastle, V. B. (1974). In “Medical Physiology” (V. B. Mountcastle, ed.), Vol. 1, pp. 662-667. Mosby, St. Louis, Missouri. Murray, J. G., and Thompson, J. W. (1957).J. Physiol. (London) 135, 133-162. Murray, M., and Goldberger, M. E. (1974).J. Comp. Neurol. 158, 19-36. Murrin, L. C., and Roth, R. H. (1976).Mol. Pharmacol. 12, 463-475. Nadler, J. V., Cotman, C. W., and Lynch, G. S. (1974).Exp. Neurol. 45,403-412. Neve, K. A., Kozlowski, M. R., and Marshall, J. F. (1982). Brain Res. 244, 33-44. Neve, K. A,, Altar, C. A,, Wong, C. A,, and Marshall, J. F. (1983). Neurosci. Abstr., in press. Nygren, L.-C., and Olson, L. (1977). Hictochenzie 52, 281-306. Nygren, L.-C., Olson, L., and Seiger, A. (1971). Histochemze 28, 1-15. Nygren, L.-G., Fuxe, K., Jonsson, G., and Olson, L. (1974). Brain Res. 78, 377-394. OKeefe, J., and Nadel, L. (1978). In “The Hippocampus as a Cognitive Map.” Oxford Univ. Press, London and New York. Page, E. D., and Neufeld, A. H. (1978). Biochem. Pharmacol. 27, 953-958. Palmer, D. S., French, S. W., and Narod, M. E. (1976).J. Pharmacol. Exp. Ther. 196, 167171. Pestronk, A., Drachman, D. B., and Griffin, J. W. (1976a).Nature (London) 260,352-353. Pestronk, A., Drachman, D. B., and Griffin, J. W. (1976b). Nature (London) 264, 787-789. Puhols, L. M., and Goldberger, M. E. (1980).J.Neurophysiol. 43, 102-117. Quik, M., and Iversen, L. L. (1978). Naunyn-Schmiedeberg’s Arch, Phannacol. 304, 141-145. Ramon y Cajal, S. (1928). I n “Degeneration and Regeneration of the Nervous System” (R. M. May, trans.). Oxford Univ. Press, London and New York. Reh, T., and Kalil, K. (1982).J. Comp. Neurol. 211, 276-283. Reis, D. J., Joh, T. H., and Ross, R. A. (1975).J. Pharmacol. Exp. Ther. 193, 775-784. Reis, D. J., Gilad, G., Pickel, V. M., and Joh, T. H. (1978). Brain Res. 144, 325-342. Reisine, T. D., Nagy, J. I., Beaumont, K., Fibiger, H. C., and Yamamura, H. I. (1979). Brain Res. 177, 241-252. Richardson, P. M., McGuinness, U. M., and Aguayo, A. J. (1980). Nature (London) 284, 264-265. Roper, S. (1976). Nature (London) 261, 148-149. Rosenquist, A. C., and Palmer, L. A. (1971). Exp. Neurol. 33, 629-652. Rosner, B. S. (1974). In “Plasticity and Recovery of Function in the Central Nervous System’’ (D. G. Stein, J. J. Rasen, and N. Butters, eds.), pp. 1-29. Academic Press, New York. Roth, R. H., Salzman, P. M., and Morgenroth, V. H., I11 (1974). Bzochem. Pharmacol. 23, 2779-2784. Roth, R. H., Morgenroth, V. H., 111, and Salzman, P. M. (1975). Naunyn-Schmiedebergk Arch. Pharmucol. 289, 327-343. Roth, R. H., Murrin, L. C., and Walters, J. R. (1976). E u r . ] . Pharmacol. 36, 163-171. Salzman, P. M., and Roth, R. H. (1980).J. Pharmacol. Exp. Ther. 212, 64-73. Sarai, K., Frazier, A., Brunswick, D., and Mendels, J. (1978).Biochem. P h a m c o l . 27,21792181. Satoh, H., Satoh, Y., Notou, Y., and Honda, F. (1976). Eur. J . Pharmacol. 39, 365-377.

246

JOHN F. MARSHALL

Schallert, T., Overstreet, D. H., and Yamamura, H. I. (1980). Pharmacol. Biochem. Behav. 13, 187-192. Schneider, G. E. (1969). Science 163, 895-902. Schneider, G. E. (1973). Bruin Behav. Evol. 8, 73-109. Schneider, G. E. (1979). Neuropsychologia 17, 557-583. Schoenfeld, R., and Uretsky, N. (1972). Eur. J. Pharmacol. 19, 115-1 18. Schwartz, W. J. (1978). Brain Res. 158, 129-147. Seeman, P. (1980). Pharmacol. Rev. 32, 229-313. Seidel, E. R., Miller, T . A., and Johnson, L. R. (1982). Life Sci. 31, 749-756. Setler, P. E., Malesky, M., McDevitt, J., and Turner, K. (1978). Life Sci. 23, 1277-1284. Sharma, V. K., Harik, S. I., Busto, R., and Banerjee, S. P. (1981).Exp. Neural. 72, 179-194. Sharman, D. F., Pokier, L. J., Murphy, G. F., and Sourkes, T . L. (1967). Can.J. Physiol. Pharmacol. 45, 57-62. Sherrington, C. S. (1906). In “The Integrative Action of the Nervous System.” Yale Univ. Press, New Haven, Connecticut. Shields, P. J., and Eccleston, D. (1972).J. Neurochem. 19,265-272. Simon, J. R., and Roth, R. H. (1979). Mol. Pharmacol. 16, 224-233. Skolnick, P., Stalvey, L. P., Daly, J. W., Hoyler, E., and Davis, J. N. (1978). Eur. J . Pharmacol. 47,201-210. Slack, J. R. (1978). Brain Res. 146, 172-176. Sokoloff, L., Reivich, M., Kennedy, C., Des Rosiers, M. H., Patlak, C. S., Pettigrew, K. D., Sakurada, O., and Shinohara, M. (1977).J. Neurochem. 28, 897-916. Spear, P. D., and Baumann, T. P. (1979). Exp. Brain Res. 35, 177-192. Spector, S., Gordon, R., Sjoerdsma, A., and Udenfriend, U. (1967).Mol. Pharmacol3,549555. Stavraky, G. W. (1961). In “Supersensitivity Following Lesions of the Nervous System.” University of Toronto, Toronto. Steward, O., Cotman, C. W., and Lynch, G. S. (1974). Exp. Brain Res. 20, 45-66. Steward, O., Cotman, C. W., and Lynch, G. S. (1976). Brain Res. 114, 181-200. Stewart, W. B., Hughes, J., and McCouch, G. P. (1940).J. Neurophysiol. 3, 139-145. Stricker, E. M., and Zigmond, M. J. (1976). I n “Progress in Physiological Psychology and Psychobiology” (J. M. Sprague and A. N. Epstein, eds.), Vol. 6, pp. 121-188. Academic Press, New York. Suga, M. (1980). Life Scz. 27, 877-882. Svengaard, N.-A., BjBrklund, A., and Stenevi, U. (1976). Brain Res. 102, 1-22. Tarsy, D., and Baldessarini, R. J. (1973). Nature (London) New Biol. 245, 262-263. Teuber, H.-L., Battersby, W. S., and Render, M. B. (1960). In “Visual Field Defects after Penetrating Missle Wounds of the Brain.” Harvard Univ. Press, Cambridge, Massachusetts. Thesleff, S. (1960). Physiol. Rev. 40, 734-752. Thoenen, H. J., Mueller, R. A., and Axelrod, J. (1969).J. Pharmacol. Exp. Ther. 169, 249254. Thompson, W. (1978). Acta Physiol. Scand. 103, 81-91. Tiedt, T. N., Wisler, P. L., and Younkin, S. G. (1977). Exp. Neurol. 57, 766-791. Trendelenburg, U. (1963). Pharmacol. Rev. 15, 225-276. Trendelenburg, U., and Weiner, N. (1962).J. Pharm. Ex#. Ther. 136, 152-161. Tsukahara, N. (1981). Annu. Rev. Neurosci. 4, 351-379. Ungerstedt, U. (1971). Acta Physiol. Scand. Suppl. 367, 69-93. Ungerstedt, U. (1974).I n “The Neurosciences: Third Study Program” (F. 0. Schmitt and F. G. Worden, eds.), pp. 695-703. MIT Press, Cambridge, Massachusetts.

RECOVERY OF FUNCTION AFTER BRAIN INJURY

247

U'Prichard, D. C., Reisine, T. D., Yamamura, S., Mason, S. T., Fibiger, H. C., Ehlert, F., and Yamamura, H. (1980). Life Sci. 26, 355-364. Urquilla, P. R., Westfall, D. P., Goto, K., and Fleming, W. W. (1978).J . Pharmacol. ESP. Ther. 207, 247-355. Uylings, H. B. M., Kuypers, K., Diamond, M. C., and Veltman, W. A. M. (1978). Exp. Neurol. 62, 658-677. Van der Krogt, J. A,, and Belfroid, R. D. M. (1980). Biochem. P h a m c o l . 29, 857-868. Warnick, J. E., Albuquerque, E. X., and Guth, L. (1977). Exp. Neurol. 57, 622-636. Weddell, G., Guttmann, L., and Gutmann, E. (1941).J.Neurol. Psychiatr. 4, 206-225. Weinberg, E. L., and Raine, C. S. (1980). Brain Res. 198, 1-1 1. West, J. R., Deadwyler, S. A , , Cotman, C. W., and Lynch, G. S. (1976).BehavioralBiol. 18, 419-425. Westfall, D. P. (1970). B r . 1 . Pharmacol. 39, 110-120. Westfall, D. P. (1977).J . Pharmacol. ESP. Ther. 201, 267-275. Westfall, D. P., McClure, D. C., and Fleming, W. W. (1972).5. Pharmacol. Ex@. Ther. 181, 328-338. Wickelgren, B. G., and Sterling, P. (1969).J. Neurophysiol. 32, 16-23. Wiklund, L., and Bjorklund, A. (1980). Brain Res. 191, 129-160. Wiklund, L., and M&$rd, K. (19791.J. Neurocylol. 8, 469-480. Wong, S. K., Westfall, D. P., Fedan, J. S., and Fleming, W. W. (1981).J. Pharmacol. Exp. Ther. 219, 163-169. Wood, C. C., Spear, P. D., and Braun, J. J. (1974). Brain Res. 66, 443-466. Wooten, G. F., and Collins, R. C. (1981).J . Neurosci. 1, 285-291. Wuttke, W., BjBrklund, A., Baumgarten, H. G., Lachenmayer, L., Fenske, M., and Klemm, H . P. (1977). Brain RPS.134, 317-331. Yarbrough, G. G. (1975). Eur. J . Pharmacol. 31, 367-369. Yarbrough, G. G. (1976). Neuropharmacol. 15, 335-338. Yarbrough, G. G., and Kostopoulos, G. K. (1975). Fed. Proc., Fed. Am. SOC.Exp. Biol. 34, 331. Zigmond, R. E. (1980). Fed. Proc., Fed. Am. Sac. ESP. Biol. 39, 3003-3008. Zigmond, R. E., Baldwin, R. C., and Bowers, C. W. (1981). Proc. Natl. Acad. Sci. U.S.A. 78, 3959-3963. Zivkovic, B., Guidotti, A., and Costa, E. (1974). Mol. Pharamacol. 10, 727-735.

This Page Intentionally Left Blank

FROM IMMUNONEUROLOGY TO IMMUNOPSYCHIATRY: NEUROMODULATINGACTIVITY OF ANTI-BRAIN ANTIBODIES Branislav D. Jankovit Immunology Research Center Belgrade, Yugoslavia

I. Introduction. . . . . . .'. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A. Immune Microenviro B. Neuroimmune Relationship . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C. Neuroimmunology.. . . . 11. Brain Antigens.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A. Lipids, Organelles, Tubulin, Receptors, and Proteins B. Neuroectodermal S- 100 Protein. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111. Neuroirnmunological Dis A. Experimental Allergic Encephalomyelitis. . . . . . . . . . . . . . . . . . . . . . . . . . B. Multiple Sclerosis. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

1V. Biological Activity of Anti-Brain Antibodies A. In Vzuo Activity.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B. In Vitro Activity . . . .

D. Higher Nervous Activity.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

C. Schizophrenia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . D. Psychiatric Diseases and Cell-Mediated Immunity

249 250 252 255 256 256 259 262 263 266 267 268 269 270 274 285 288 293 293 294 295 297 299 300 302

1. Introduction

Although the brain is the most fascinating organ to which a scientist can address himself in biomedical research, and despite the fact that the brain is associated with all structures and functions in the organism, for many decades of this century immunologists tended to ignore the interconnections between the immune system and the nervous system (Spector, 1980). The great majority of explorers in immunosciences, and the 249 INTERNATIONAL REVIEW OF NEUROBIOLOGY, VOI.. 2fi

Copyright 0 1985 by Academic Press, Inr. All rights of reproduction in any form reserved. ISBN 1~i2-3fim2fi-3

250

RRANISLAV D. ~ J A N K O V I ~ :

authors of leading immunological theories in particular, stuck firmly to their basic training, experience, opinions, scepticism, adopted dogmas, and acquired prejudices, thus deliberately or through disinterest thwarting and delaying meaningful communication among different sciences. Fortunately, a couple of decades ago interdisciplinary developments forced scientists in different fields to listen and learn from one another. That led to the recognition of neuroimmunology (immunoneurology) as an important discipline encompassing the great problems of intercommunication between the nervous system and the immune system (Katz, 1966). In tracing the pathway from immunoneurology to immunopsychiatry, the common denominator throughout this article is the antibrain antibody, which reacts with its antigenic counterpart in the central nervous system (CNS). Several lines of research are brought together to portray the cooperative efforts of immunoscientists, neuroscientists, and other scientists to explore the biological galaxies organized and functioning in the form of the neuron and the brain.

.

A. IMMUNEMICROENVIRONMENT Contemporary immunological investigations of the lymphocyte receptors (Cinader, 1977) originated from the experiments of Paul Ehrlich on “cellular receptors” and the ideas of Sir Henry Dale, and from vast pharmacological and neurophysiological experience. These studies at molecular and subcellular levels demonstrated the outstanding plasticity of the cell membrane in reacting and recognizing very many endogenous and exogenous ligands, and the importance of receptors in intercellular communications. They also pointed to the great need for more sophisticated analyses of anatomical and physiological interconnections between lymphoid and nonlymphoid, structural and functional correlates of the immune microenvironment (JankoviC, 1973, 1979). ?’he imniune niicroenvironment (milieu) reflects the homeostatic properties of the organism and the dynamic harmony of “le milieu interieur” of Claude Bernard (1878). In simple terms, the following structural and functional components constitute this microenvironment: Lymphoid cells, both T- and B-lymphocytes, at different stages of maturation, activated specifically by antigens o r nonspecifically by mitogens (Resch and Kirchner, 1981). Those cells respond to an immense number of stimuli from various sources and of different physicochemical composition.

NEUROMODULATING ACTIVITY OF ANTI-BRAIN ANTIBODIES

25 1

Nonlymphoid c~lls,such as epithelial cells, macrophages, reticulocytes, etc., which interact with lymphoid cells and cooperate with other cells in normal and inflammatory reactions (Weissmann et al., 1979). Neurons, both cholinergic arid adrenergic, which take part in microenvironmental events by means of their neurohumoral activity (Pinsker and Willis, 1980). Visting cells, both lymphoid and nonlymphoid, which influence the structural and functional hierarchy of the immune milieu. Hormones, released by endocrine glands, which modulate the immune recognition mechanisms in the immune microenvironment (Pierpaoli and Sorkin, 1972). NeuropeptideJ, which are produced by “regulatory cells” (Grossman, 1979) of the brain and nonnervous tissues (Iversen et al., 1978). Available evidence indicates that peptides may reach via the bloodstream the remote target cells and alter their activity (endocrine function), influence neighboring cells by diffusion through interstitial space (paracrine function), and stimulate or inhibit postsynaptic cells (neurocrine function). The ubiquitous distribution of biologically active peptides (Guillernin, 1978), their presence in the blood, and the multiple functions they may exert (Lord et al., 1977) would imply that neuropeptides may easily reach the immune microenvironment. Receptors, the functional units of the cell membrane which are coupled to response signals (Schulster and Levitzki, 1980) and make possible the immune microenvironment circuits. Biologically active substances, such as lymphokines and monokines (Pick, 198 l), interleukins (Mdller, 1982), and pharmacologically active agents (Werner and Floc’h, 1978) complement, immunoglobulins, etc., which are produced by lymphoid and nonlymphoid cells in situ or elsewhere. Zons, such as sodium, potassium, calcium, and magnesium, which influence the biophysical properties of the cell membrane and affect the transmission of information. Despite the importance of the ion flux between extracellular and intracellular space, there is a general lack of evidence of ionic dissipation and migmtion in the immune microenvironment. T h e investigation in that direction may bring quite new information about the function of lymphocytes. As an example, the lithium cation, known as an effective agent in the treatment of manic-depressive diseases, has been shown to affect in uivo a large variety of biological systems (Bunney and Murphy, 1976) including the immune system UankoviC et al., 1978, 1979a,b, 1980a, 1982a). T h e study of this cation and other physiological and nonphysiological ions may have a great impact on notions relevant t o immune stimulation and inhibition.

252

SKANISLAV

I). . J A N K O V I ~

Higher nervous activzty, which accomplishes its influence on the immune microenvironment through emotional, aggressive, stress-related, and various behavioral performances. Studies on the relationship between higher nervous activity and immunity were begun long ago (Metalnikov, 1934; Lukianenko, 1961) but are still in an initial stage. It has been shown that neurological and psychiatric diseases can be accompanied by abnormal immune responses. Impaired in vitro correlates of cellmediated immunity were found in patients with Guamanian Parkinson’s dementia (Hoffman et al., 1978), Huntington’s disease (Morrel, 1979), and cerebral tumors (Brooks et al., 1972). Immunological abnormalities were also observed in subjects with psychoses (Kovaleva et al., 1977), bereavement and depression (Bartrop et al., 1977), and emotional stress (Solomon, 1969). Experiments in animals on avoidance learning and stressful stimuli demonstrated diminished delayed hypersensitivity (Rasmussen et al., 1959), prolonged survival of skin grafts (Wistar and Hildemann, 1960), and impaired lymphocyte reactivity (Marsh and Rasmussen, 1960). Immunological deviations in psychiatric diseases with special reference to schizophrenia have been recently reviewed (Solomon, 1981). Investigations in the opposite direction, i.e., the influence of immune factors and mechanisms on structural and functional correlates of the nervous system, will be partly covered in the present article. ‘The outlined concept of the immune microenvironment implies that the transmission of immune information within the lymphoid tissue is not exclusively confined to or conducted by immune recognition, but rather that neuronal and endocrine components are also encoded in the modulation and transduction processes which take place in the immune microenvironment under in uiuo conditions. Therefore, the immune microenvironment possesses the potential to act as a locus of communications between lymphoid, nervous, and endocrine systems. In other words, the immune microenvironment is a multisystem composed of components derived from the lymphoid system, the nervous system, and the endocrine system UankoviC, 1973, 1979). Consequently, the structure and function of one system cannot be studied in isolation from the two other systems. As for the immune system, all current theories claiming to explain the immune functions have paid little or no attention to the neuroendocrine correlates of immune phenomena.

B. NEUROIMMUNE RELATIONSHIP Several developments have suggested some analogies between the overall structures and functions of the nervous system and the immune

NEUROMODULATING ACTIVITY OF ANTI-BRAIN ANTIBODIES

TABLE I THENEUROIMMtJNO RELAPIONSHIP: SOME ANALOGIES BETWEEN STRUCTURES A N D FIJNCTIONS OF T H E NERVOUS SYSTEM A N D

THE

253

THE 0 V E R A I . I .

IMMUNE SYSTEM

Category

Nervous system

Immune system

Structural unit Central organ Origin Cell-cell continuity Cell-cell interactions

Neuron Brain Neuroepitheliuni Synapses Neuron-neuron Neuron-glial cell Neuron-other cells For hormones, acetylcholine, Fc fragment, lectins, etc. Neurostenin For immune system Neurotransmitters, etc. Neuropeptides With lymphocytes Electric activity of the neuron From stimulation to inhibition Primary information; memory Immunoneurological and immunopsychiatric diseases

Lymphocyte Thymus Epithelial rudiment Synapselike contacts Lymphocyte-lymphocyte Lymphocyte-macrophage Lymphocyte-other cells For hormones, acetylcholine, Fc fragment, lectins, etc. Neurostenin For nervous system Lymphokines, etc. Thymopeptides With neurons Electric activity of the lymphocyte From stimulation to inhibition Primary response; memory response Lymphoid tissue (cell) disorders

Membrane receptors

Contractile protein Modulators” Soluble mediators Endocrine function Common antigens Biophysics Sensitivity range Recognition events Autoimmunity

a T h e term “modulator” (Florey, 1967) is in wide use. Little effort has been made to distinguish a modulator from a transmitter (Dimuskes, 1980). Therefore, this term is employed in a variety of contexts to describe actions at different levels of organization (Kandel et al., 1979).

system. Some of them are listed in Table I. It is quite obvious that the immune system provides strong stimulus for analysis of comparable events in the nervous system, and vice versa. Recent studies at the stem cell level performed independently in Australia and Yugoslavia furnished fresh evidence of the neuron-lymphocyte structural and functional relationship, other than the known antigenic similarity between the brain and thymus (Reif and Allen, 1964). In examining the mouse brain for the presence of pluripotential stem cells, and using the colony-forming unit in the spleen as an assay system, Bartlett (1982) found that a large number of hematopoietic stem cells in the brain of the adult mouse were comparable to those contained in adult bone marrow. Concerning himself primarily with B-cells, the author speculated that elevated levels of immunoglobulin G in the cerebro-

254

BRANISIAV

D. J A N K O V I ~

spinal fluid (CSF) of patients with multiple sclerosis (MS) may be associated with secretion of B-lymphocytes within the CNS. The experiment of Jankovi6 and JankoviC (1983), however, was concerned with T-lymphocytes. Viable cells from the brains of l0-day-old chick embryos were placed either directly on the chorioallantoic membranes (CAM) of 12day-old host chick embryos, or first cultured for 4 days and then transferred onto the recipient's chorioallantoic membrane (Fig. 1A). Light and electron microscope analyses revealed that many of the chorioallantoic pocks produced by implanted embryonic brain cells were composed of cells which exhibited the morphology typical of the embryonic thymocytes (Fig. IB,C). It was concluded that the brain of the chick embryo

FIG. 1. (A) Scheme of an experiment designed to study the relationship between the chick embryo brain and thymus at the stem cell level. Brain cells from IO-day-old donor chick embryos were transferred directly on the chorioallantoic nienlbrane (CAM) of 12day-old recipient embryos, or brain cells were first cultured for 4 days and then placed on the CAM of the host chick embryo. In both instances, implanted brain cells induced the formation of pocks within 3 days. (Pock: a colony of cells derived from the implanted cell.) Light and electron microscope analyses revealed that pocks were composed of cells which exhibited the morphology typical of embryonic thymocytes (B and C). (B) An electron micrograph of the CAM pock (X6000), 15-day-old host embryo. (C) An electron micrograph of the embryonic thymus (x6000), 15-day-old normal chick embryo.

NEUROMODULATING ACTIVITY OF A N T I - B R A I N ANTIBODIES

255

contains stem cells which may give rise to thymocytes. A relevant finding from neuroembryology suggests that the neural crest contains pluripotential cells from which originate aggregates of cells of different phenotypes (Bronner-Fraser and Cohen, 1980). Thus, the described studies constitute further evidence of the interconnections between the nervous system and the immune system. C. NEUROIMMUNOLOGY

A hundred years ago, Louis Pasteur (1885) described an acute and frequently fatal neurological disease in some patients vaccinated with rabbit brain infected with attenuated rabies virus. This observation pointed to the link between the neurological disease and immunity. Probably the first indication of the functional importance of the CNS in immunity came from Besredka (1919) who, impressed by dramatic neurological signs which occurred in acute anaphylactic shock, used the intracerebral route in studies of the nervous origin of anaphylactic symptoms. The ingenious experiments of Metalnikov (1934) on the brain-immunity relationship opened new avenues for research. An important impetus to immunological investigations of the nervous system followed the production of demyelinating disease in experimental animals (Morgan, 1946; Kabat et al., 1946). The term “immunoneurology” was first introduced by Schmitt (1964), who suggested that in neurosciences the unknown correlates between the structure and function of the neuron and the brain might be found with the aid of immunological tools and concepts, “especially in view of the remarkable discovery of Mihailovid and JankoviC” (Levine, 1965). These investigators (Mihailovid and JankoviC, 1961) demonstrated that the electroencephalographic activity and behavior of the cat can be profoundly affected by antibrain antibodies injected into the lateral ventricle of the cat brain. In general, immunoneurological studies are mainly focused on seven problems: ( 1 ) cytochemical mapping and immunochemical definition of neural antigens and markers; (2) immunoneurobiology of receptors, transmitters, mediators, and modulators; (3) neuroimmune connections during embryogenesis; (4) immunoneurological diseases such as experimental allergic encephalomyelitis in animals and multiple sclerosis in humans; ( 5 ) the role of the nervous system in immunity; (6) structural and functional dissection of the neuron, brain, and mind by means of immunological agents, and antibrain antibodies in particular; and (7) immunopsychiatric diseases. Immunoneurological diseases are an integral part of any textbook of

256

BKANISLAV D.

JANKOVIC:

immunopathology, while other immunoneurological topics have been the subject of numerous comprehensive reviews. The principal aim of this article is to provide concise information about the biological activity of antibrain antibodies and some new insights into the pathogenesis of certain psychiatric diseases. For the sake of convenience, the term “antibrain antibody” (JankoviC, 1972) used in this article refers to different immunoglobulin molecules capable of reacting with antigenic determinants situated in the neuron, glial cell, and subcellular and macromolecular constituents of these cells. At the beginning, it seems appropriate to make some general remarks on brain antigens.

II. Brain Antigens

The entire range of antigenic potentialities of any brain structure is immense. The antigenic composition of the brain is made even more complex by the widespread network of communications between brain structures, pleomorphic variability of the neuron, changes in the neuron-glia interplay, high protein synthesis and turnover in the neuron, continuous flow of information, kaleidoscope of memory, neurotransmitter and neuroendocrine activities, mechanisms underlying the vital functions, wide scale of chemical composition of macromolecules, variety of membraneous and intracellular receptors, progressive cellular transformations related to aging, removal of cell debris, dynamic state of the blood-brain-blood barrier, and diversity of interactions between the brain and other tissues of the organism. In accordance with our present immunological knowledge, each of the brain antigens can activate immunocompetent cells and trigger humoral and cell-mediated immune processes. However, the purpose of this article is not to provide a complete list of nervous tissue antigens but rather to emphasize a few points which are relevant to the biological activity of antibrain antibodies. For further information the reader is referred to several comprehensive reviews (Thompson, 1976; Zomzely-Neurath and Walker, 1980; Bock, 1982). A. LIPIDS,ORGANELLES, TUBULIN, RECEPTORS, AND PROTEINS The idea of brain antigens first emerged in Pasteur’s time from neuroparalytic accidents caused by injection of the rabies vaccine grown

NEUROMODULAI’ING A C l I V I T Y OF ANTI-BRAIN ANTIBODIES

257

in rabbit spinal cord. At the beginning of this century, in a series of investigations Witebsky and co-workers established the organ specificity of lipids extracted by alcohol from whole brain tissue (Witebsky and Steinfeld, 1928), and demonstrated by means of anti-caudate nucleus and anti-hippocampus serum that gray matter could be distinguished from white matter (Reichner and Witebsky, 1934). Other investigators extended studies and provided new information about the antigenicity of the brain tissue (Lewis, 1933; Bogoch et al., 1964; MacPherson and Liakopoulou, 1966). Great impetus to the study of brain antigens was given by the production of allergic encephalomyelitis in experimental animals (Morgan, 1946; Kabat et al., 1946). In this domain, studies focused on isolation and chemical characterization (Kies et al., 1958; Bergstrand, 1977), and pathophysiology (Paterson, 1976) of the encephalitogenic myelin basic protein. At the subcellular level, many efforts have been made to separate synaptosomes, synaptic vesicles, synaptic membranes, mitochondria, nerve endings, and other neuronal constituents (Lajtha, 1969). Several distinctive and cross-reacting classes of antigenic determinants are revealed in synaptosomes, synaptic membranes, and mitochondria (Mickey et al., 1971). I n synaptic plasma membranes and synaptic vesicles there is a variety of proteins and glycoproteins (Morgan et al., 1973a) which may have common antigenic determinants. It appears, however, that the plasma membrane of cortical synaptosomes differs serologically from the plasma membranes of other organs, and from other fractions of the brain (Henchman et al., 1972). Experiments with subcellular particles of the brain are confronted by two major problems: the purity of preparations and the specificity of antisera produced with brain organelles. As regards the purity, a preparation of synaptosomal plasma membranes of 80% purity (Morgan et al., 1971) was found to be contaminated with mitochondria1 membranes and glial membranes (Morgan et al., 1972). Moreover, synaptic vesicle preparations of 90% purity contained synaptosomal plasma membranes (Morgan et al., 1973b). T h e purification of nervous tissue antigens still remains one of the major immunoneurological problems. The specificity of anti-brain organelle sera depends largely on sequential absorption procedures which may yield monospecific antisera (McMillan et al., 1971). Since microtubules are cytoskeleton structures (Shelanski, 1973), and have functions in axoplasmatic transport (Fernandez et al., 1970, 1971) and release of neurotransmitters of synapses (Rasmussen, 1970), a number of immunological undertakings were devoted to the tubulin. Antisera against this colchicine-binding and vinblastine-precipitated poly-

258

BKANISLAV D. J A N K O V l C

morphic protein reacted with mitochondrial, myelin, and synaptic vesicle fractions, thus suggesting that tubulin is a component of some membranes, and that it exists as a soluble pool of tubulin subunits (Twomey and Samson, 1970). Of the known membrane receptors, the acetylcholine receptor (AChR) is biochemically and electrophysiologically probably the best characterized neurotransmitter receptor. Its importance in immunology derives from the fact that anti-AChR antibody impairs the AChR-mediated neurotransmission from motor neurons to striated muscle fibers (Katz, 1966). Such an inhibition of AChR by antibodies is the characteristic pathophysiological feature of myasthenia gravis (Lennon, 1976; Fuchs, 1979; Lindstrom, 1979; Seybold and Lindstrom, 1982). From the immunological point of view, the muscle fiber component is more important than the neuronal component in the pathogenesis of myasthenia gravis. More relevant to the neuronal membrane receptors is the finding that anti-dopamine receptor serum inhibits depolarization of' the TCXl 1 line of mouse neural cells, and dopamine binding and dopamine stimulation in nervous tissue (Myers et al., 1976). Dopamine receptors may serve as nonspecific immunomodulators since phenothiazine drugs have been shown to affect cell-mediated immune responses (Mitrova and Mayer, 1976; Ferguson et al., 1978). Haloperidol, a blocker of brain dopamine receptors (Niemegeers and Janssen, 1979), is capable of modulating immune processes (Shaskan and Lovett, 1980, 1981), although it is not clear whether its action is primarily confined to the nervous system (Hall and Goldstein, 1981) or to the immune system or whether it affects both (Shaskan and Lovett, 1981). Recently, dopamine receptors were detected on lymphocytes (LeFur et al., 1980). T h e acidic protein isolated from fibrous astrocytes of the human brain (Eng et al., 197 l), designated as glial fibrillary acidic (GFA) protein, was found to have some antigenic properties in common with the brain and spinal cord of various mammalian and submammalian (shark, goldfish, turtle, and chicken) species (Dahl and Bighami, 1973). In immunofluorescent assays, human astrocytes were selectively stained with specific antibodies against GFA protein (Eng et al., 1971; Uyeda et al., 1972; Bighami et al., 1972). Glia-specific antigens were demonstrated in cells cultured from human and rabbit brain (Wahlstrom et al., 1973). Besides GFA protein, nervous tissue contains various soluble acidic proteins that are not detectable in other tissues (Moore, 1965; Moore and Perez, 1965; Liakopoulou and MacPherson, 1970; Hatcher and MacPherson, 1970; Bennett, 1974). T h e acidic protein separated from the rat brain by Bennett and Edelman (1968) is probably the protein 14-3-2 already characterized by Moore and Perez (1965). The data on nervous system-specific

NEUKOMODULATING ACTIVITY OF ANTI-BRAIN ANTIBODIES

259

proteins 14-3-2, antigen alpha, and neuron-specific enolase are presented by Zomzely-Neurath (1982). B. NEUROECTODERMAL S- 100 PROTEIN Of all acidic proteins isolated from the brain, the S-100 protein (Moore, 1965) has become the most important topic in immunoneurology. This protein appears to satisfy the immunochemical requirements for purity. There is now an increasing body of evidence that S-100 protein is a heterogeneous class of molecules (Uyemura et al., 1971; Dannies and Levine, 1971; Stewart, 1972; Abe et al., 1973; Isobe et al., 1977). An improved prodedure for isolating S- 100 protein from bovine and rat brain enabled the separation of two populations of S-100 proteins: III-IVa-I and III-IVb-1. These two fractions exhibited all the properties of S-100 protein and reacted with anti& 100 serum, but differed in the distribution of protein among the four units obtained in sodium dodecyl sulfate gels (Mahadik et al., 1979a). Complement-fixation and rocket immunoelectrophoresis assays revealed that the immunological reactivity of the III-IVa-1 protein was significantly lower than that of the III-IVb-1 protein. These two subpopulations of S-100 proteins showed quantitative differences in reactivity with anti-S- 100 serum (Mahadik et al., 1979b). Thus, the S-100 proteins expressed different physicochemical and immunological properties due to the heterogeneity of S-100 proteins in the brain. Although S-100 protein occurs in the brain soon ifter birth (Svidorov et al., 1972) and makes up 0.5% of soluble proteins of the C N S , it is not known whether this protein is synthesized in the neuron, in the glia, or in both. The localization of S-100 protein in the brain is still a matter of dispute (Moore et al., 1968). S-100 protein is predominantly localized in the glial cells (Hyden and McEwen, 1966; Benda et al., 1968; Cicero et al., 1969, 1970; Lightbody et al., 1970; Perez et al., 1970; Martus and Mughal, 1975; Hansson et al., 1976; Ludwin et al., 1976). On the other hand, S-100 protein was found in the neuron (Hyd6n and McEwen, 1966; Packman et al., 1971; Miani et al., 1972; Sviridov et al., 1972; Rusca et al., 1972; Mahadik et al., 1974; Schubert et al., 1974; Donato and Michetti, 1974; Haglid et al., 1974; Donato et al., 1975). As for intracellular distribution, S-100 protein is variously referred to as a membranebound protein (Haglid and Stavrou, 1973; Haglid et at., 1973b; Rusca et al., 1972; HydCn and Ronnback, 1975), a nucleus-associated protein (Hyden and McEwen, 1966; Sviridov et al., 1972; Mahadik et al., 1974; Michetti et al., 1974), o r a synaptic membrane-bound protein (Haglid et

260

RKANISLAV D. . J A N K O V I ~

al., 1974; Donato, 1978). It has been reported that cerebral cortex synaptosomes contain two forms of S-100 protein: soluble and membrane bound. The latter form is distributed in the synaptosomal membranes, intraterminal mitochondria, and synaptic vesicles (Donato et al., 1975). In any case, S-100 protein is present in astrocytes and oligodendrocytes in higher concentrations than in neurons (Adams, 1977; Bock, 1978; Zomzely-Neurath and Keller, 1977). S-100 protein was detected in human glioblastoma, astrocytoma, spongioblastoma, and acoustic neuroma cells, but not in ependymoma, plexus papilloma, chromophobe adenoma of the pituitary, and polymorphans oligodendroglioma cells (Haglid et al., 1973a). It has been claimed that S-100 protein shows strict tissue specificity (Moore and McCregor, 1965), and that S-100 protein is similar if not identical in the brain of a variety of vertebrate species (Moore and McCregor, 1965; Levine and Moore, 1966; Kessler et al., 1968). However, further investigations revealed that S- 100 protein is present in nonnervous tissues, too. S- 100 protein was detected in the interstitial cells of the rat pineal gland (MGller et al., 1978). Although ultrastructural studies indicated that those cells have the morphology of fibrillary astrocytes (Sheridan and Reiter, 1973; Povlishock et al., 1975), it has been stated (Mglller et al., 1978) that some of the interstitial cells might be oligodendrocytes, thus implying that S- 1OO-containing interstitial cells are macroglial and neuroectodermal in origin. The GFA protein was also detected in the pineal gland (M@lleret al., 1978). Immunohistochemical studies showed that the folliculostellate (follicular, chromophobe) cells of the rat pituitary anterior lobe have S-100 protein (Nakajima et al., 1980). Those cells of unknown function in the adenohypophysis are supposed to belong to the neuroectodermal cell category. In the adrenal medulla there are S-1OO-positivecells (Cocchia and Michetti, 1981). Cocchia et al. (1981) described the presence of S-100 protein in the normal human skin; the protein was specifically located in melanocytes and cells with the morphological features of Langerhans' cells. Melanocytes are of neuroectodermal origin but Langerhans' cells of the skin are generally believed to be of mesodermal origin (Breathnach, 1980; Sting1 et al., 1980). I t has been reported that S-100 protein can be detected in human chondrocytes (Stefansson et al., 1982). Although chondrocytes are considered to originate from the neural crest '(LeLievre, 1974) whereas chondrocytes found elsewhere are developmentally closer to the mesoderm (Holzer and Detwiler, 1953; Hall, 1978), S-100 protein seems to occur both in neuroectodermal and mesodermal chondrocytes (Stefansson et al., 1982).

NEUROMODULATING ACTIVITY OF

ANTI-BRAIN ANTIBODIES

26 1

Of particular interest is the detection of S-100 protein in cells of continuous lines of human malignant melanomas (Gaynor et al., 1980). Gaynor and associates (198 1) suggested that S-100 protein may perhaps serve as a marker for human malignant melanomas. However, S-100 protein was also found in compound naevi, and this may indicate that S100 is a constituent of melanocyte-derived cells which are not associated with malignant transformation. The relative absence of S- 100 protein from normal human skin may be due to the low density of melanocytes (Cochran, 1970). The content of S-100 protein in normal melanocytes remains to be elucidated. On the basis of the foregoing evidence, it appears that S-100 protein is shared by nervous and non nervous cells of neuroectodermal origin. Consequently, cells associated with malignant transformation of cells of neuroectodermal origin are expected to contain significant amounts of S-100 protein (JankoviC et al., 1982a). T h e fact remains, however, that under normal conditions the highest concentration of S-100 protein is in the brain. S-100 protein is believed not to be present in detectable amounts in the CSF and serum. T h e appearance of S-100 in CSF in humans might serve as an indicator of active injury of the nervous parenchyma (Michetti et d., 1980). The S-100 protein is absent from the CSF of patients suffering from psychic disorders o r neurological diseases with no apparent signs of nervous tissue damage, but present in patients with extensive neurological lesions (Michetti et d., 1979). The circulation of S-100 in the CSF and blood probably reflects alterations in the brain-blood barrier. T h e localization of S-100 protein in cytoplasmic processes of astrocytes and oligodendrocytes in the immediate vicinity of brain capillaries (Tabuchi et al., 1983), taken together with the finding that S-100 is able to interact with artificial membranes and change their permeability (Rusca et al., 1972), would indicate the indirect involvement of S-100 protein in the permeability of the brain-blood barrier. In this connection, S- 100 protein is reported to increase the transport of cations through the membrane in the presence of calcium ions (Calissano and Bangham, 1971). The use of S-100 protein from the human brain in the evaluation of certain mental disorders will be discussed in Section V, Immunopsychiatric Diseases. The physiological function of S-100 protein is unclear. Changes of the bioelectrical pattern of the neuron from the snail ganglion induced by anti-S-100 antibodies (SaviC et al., 1979) may be ascribed to the capacity of antibodies to combine with membrane-associated S- 100 molecules of the neuron (Rusca et al., 1972). On the other hand, anti-S-100 serum injected into the cortex of rat brain failed to change the cortical EEG

262

BRANISLAV D.

JANKOVIC

pattern (Karpiak et al., 1976a). The biological activity of anti-S-100 antibodies will be considered later on, together with other anti-brain antibodies. Since anti-S- 100 protein antibodies injected intraventricularly into adult rats during the course of training prevented animals from further learning, it was concluded that S-100 is associated with processes underlying learning and behavior (Hyden and Lange, 1970, 1971). Antisera to S-100 protein (Karpiak et al., 1976a) induced changes in certain behavioral paradigms. To conclude, only a small number of brain antigens have been isolated in a sufficiently pure state. Antigenic correlates of the brain neuroendocrine function and of neuronal and glial receptor activity are still poorly explored. S-100 protein constitutes a class of brain macromolecules which helps, to some extent, the immunoneurological studies of structural and functional events that occur during embryonic development, postembryonic maturation, and aging of the brain. It is to be expected, however, that a better immunochemical characterization of known neuronal and glial antigens, their wider use in experimental and clinical research, and the isolation of new antigens from the nervous tissue will contribute a great deal to the comprehension of immunoneuroendocrine interconnections.

111. Neuroirnmunologicol Diseases and Anti-Brain Antibodies

Studies of autoimmune responses to nervous tissue antigens are concerned with neurological diseases in animals and humans. In 1946, experimental allergic encephalomyelitis (EAE) was described by Morgan, and Kabat et al. Since then, much attention has been devoted to this demyelinating disease, and EAE is generally regarded as an animal model of immunoneurological disorders such as MS, rabies vaccineassociated acute disseminated encephalitis, and postinfectious and postvaccinal encephalitis. In spite of certain dissimilarities between EAE and MS, there is a sufficient number of common pathological and immunological features to justify the assumption that EAE is an experimental counterpart of MS in humans (Paterson, 1979; Weigle, 1980). EAE is characterized by inflammatory lesions distributed in various parts of the brain and spinal cord. Histopathologically, these lesions are perivenous collections of lymphocytes, histiocytes, plasma cells, and polymorphonuclears (Waksman, 1959). Another important feature of EAE is the break-

NEUROMODULATING ACTIVITY OF ANTI-BRAIN ANTIBODIES

263

down of the myelin sheath (Kies and Alvord, 1959). While EAE can be readily produced with encephalitogenic myelin basic protein, the antigen which induces autoimmune response in MS is not known. EAE can be passively transferred to nonsensitized hosts with the regional lymph node cells from EAE donors (Paterson, 1960). In contrast, circulating antinervous tissue antibodies from EAE-sensitized animals failed to produce lesions typical of EAE when injected intravenously into naive recipients (Morgan, 194’7; Kabat et al., 1948; Hill, 1949; Koprowski, 1962). It was deduced, therefore, that the effector mechanism of disease is related to cell-mediated immunity and not to humoral immunity. Central issues relevant to the understanding of EAE and the relationship between EAE and MS have been comprehensively and critically covered by Paterson in his series of reviews (1966, 1972, 1976, 1979, 1982). However, in spite of the belief that sensitized lymphocytes rather than anti-brain antibodies are of importance in the EAE and MS, the fact still remains that circulating anti-nervous tissue antibodies are regularly present in both diseases. This portion of the article will consider, therefore, what role anti-nervous tissue antibodies may have in the pathogenesis of EAE and MS. A. EXPERIMENTAL ALLERGIC ENCEPHALOMYELITIS In order to test the EAE-inducing capacity of anti-brain antibodies, guinea pigs were repeatedly injected through a cannula inserted permanently into the lateral ventricle of the brain with pooled sera from guinea pigs which developed EAE following injection of bovine spinal cord in complete Freund’s adjuvant UankoviC et al., 1965). In another experiment, normal rabbits received several injections of “EAE 7-globulin” (i.e., the serum fraction containing anti-brain antibodies from rabbits with EAE) directly into the cerebral cavity UankoviC et al., 1966a). Thus, in both experiments the blood-brain barrier was bypassed. Histological analysis of brain and spinal cord sections revealed that animals given several intraventricular injections of anti-brain antibodies developed lesions typical of EAE (Table 11). It was suggested, therefore, that EAE can be passively transferred to normal animals by means of antibrain antibodies and that, in addition to sensitized lymphocytes, circulating antibodies can also be involved in the pathogenesis of EAE. These results were substantiated with findings that normal rabbits developed neurological symptomatology following intraventricular and suboccipital inoculation of serum from donor rabbits immunized with brain emulsified in complete Freund’s adjuvant (Simon and Simon, 1975). Both

264

BKANISLAV D. . J A N K O V I ~

~ N C I D E N C EOF

EAE

IN

TABLE I1 GUINEA PIGSA N D RABBITSINJECTED INro THE LATERALV E N I K I C L E O F T H E BRAIN W I T H ANTI-BRAIN ANI.IBODIES Number of animals with EAE lesions (material injected intraventIicularly)

Animals

Anti-brain antibodies

Normal serii 1x1

Guinea pigs Rabbits

10115“ 1712 1

Oil0 011 1

Saline

N Dh 018

Numerator, number of’ animals with lesions; denominator, number of’ animals in

group. ND, Not determined.

groups of investigators reported on striking electroencephalographic changes in the recipient animals caused by intracerebral administration of anti-brain antibodies. In tissue culture (Bornstein and Appel, 1961), the serum from animals with EAE exerted a demyelinating effect on neonatal rat cerebrum (Bornstein and Appel, 1965). A similar phenomenon was produced with y-2-globulin fraction in the presence of complement (Appel and Bornstein, 1964). It appears, however, that heat-labile complement components are not necessary for myelination inhibition by antisera against whole spinal cord and galactocerebroside (Dorfman et al., 1979). Equivocal results were obtained with antimyelin activity of anti-cerebroside antibodies (Appel and Bornstein, 1964; Bornstein and Raine, 1970; Dubois-Dalcq et al., 1970; Fry et al., 1974; Hruby et al., 1977; Dorfman et al., 1978). Some data indicated that sera from guinea pigs with EAE and animals immunized with myelin basic protein (MBP), although containing antibodies against MBP, did not possess demyelinating properties, thus implying that there are anti-neuronal antibodies which do not exhibit demyelinating activity, and that some nervous tissue antigens are not capable of inducing the formation of demyelinating antibody (Seil et al., 1968). On the other hand, a serum-demyelinating factor (a complement-fixing autoantibody) responsible for demyelination in vitro was found to react with a constituent of myelin of both central and peripheral nervous tissue which is neither MBP nor cerebroside (Lebar et al., 1976). Of interest is that normal rabbit and human sera exerted depressant (Seil et al., 1975) and nondepressant (Bornstein and Crain, 1965; Lumsden, 1972) affects. This dual activity was ascribed to the different levels of “nonspecific depressants” in normal sera (Crain and Bornstein, 1975). Anti-myelin protein serum impaired the incorporation of precur-

NEUROMODULATING ACTIVITY OF ANTI-BRAIN ANTIBODIES

265

sors into MBP, proteolipid, and sulfatides of incubated slices of the rabbit spinal cord (Pellkofer and Jatzkewitz, 1976). Progressive paralysis was induced in normal guinea pigs by means of intraperitoneal injections of plasma from EAE-challenged donors (Pabst and Dupuy, 1970). Experiments with cell-free preparations other than serum revealed that perivascular mononuclear cell infiltrates, but not clinical signs of EAE, were produced in the brain of rats injected intravenously with supernates derived from cultured lymph node cells of rats immunized with guinea pig spinal cord (Whitacre and Paterson, 1977). T h e transfer activity of the supernates was inhibited by the addition of brain antigens. This kind of passive transfer of EAE failed when donor animals were sensitized with MBP of heterologous and homologous origin (Whitacre and Paterson, 1980). T h e cell-free lymph from EAE sheep sensitized with homologous spinal cord induced the formation of EAE lesions when transferred to normal recipient sheep (Willenborg, 1982). Although the cell-free lymph from EAE animals contained antibodies to myelin basic protein, it was not clear whether these antibodies or antibodies of some other specificities or not yet defined lymph factors were responsible for the passive induction of EAE in the recipient animals. The presence of plasma cells at the site of EAE lesions (Campbell, 1949; Campbell and Good, 1950) may be taken as evidence that humoral immune processes are active in EAE. It has been reported that mononuclear cells of the EAE lesions in the brain release anti-brain antibodies (Lennon et al., 1972). Specific immunoglobulins surrounding nervous tissue blood vessels were detectable days before the appearance of perivascular aggregates of mononuclear cells (Oldstone and Dixon, 1968). Transfer of EAE in guinea pigs with lymph node fragments from EAE animals in Millipore chamber could be attributed to the activity of antibody-secreting cells (Lamoureaux et al., 1967). However, direct evidence that immune response can be maintained in the brain, that antibodies can be produced locally in the brain (Heremans, 1973), and that memory B-cells are present in the brain emerged from the study of anti-virusproducing plasma cells in the brain of mice infected intracerebrally with 6/94 virus (Gerhard and Koprowski, 1977). Pertinent to the function of anti-brain antibodies in EAE are experiments on protective action of antibodies. Thus, complement-fixing antibody to nonencephalitogenic cerebroside (Niedieck, 1964) has been related to resistance to EAE (Paterson et al., 1965). This experiment is in accordance with an earlier observation (Paterson and Harwin, 1963) that prolonged treatment with convalescent serum protected rats against EAE. Although there was no correlation between antibody titer and degree of protection, the authors assumed that complement-fixing anti-

266

BRANISLAV D.

.JANKOVIC

bodies to the rat brain exerted the protective activity. Some investigators believe that the protective action of anti-brain antibodies has not been proven conclusively (Alvord et al., 1970). However, recent findings showed that the prolonged treatment of rats with antiserum to guinea pig spinal cord strikingly reduced the severity of clinical EAE. This protective effect was attributed to an “enhancing” antibody and not to the presence of hemagglutinating antibodies to galactocerebroside (Hughes, 1974). The “inhibitory” antibodies were effective only when given in the early stage of EAE development. Since the reduction of clinical EAE was significant, whereas the diminution of histological changes was only slightly affected by protective antibodies, it was suggested that those antibodies act chiefly against antigens responsible for the neurological signs (Hughes, 1974). This assumption was based on the view that the neurological symptoms and the histological lesions in EAE are caused by different antigens (Lennon et al., 1971; Paterson, 1972). An interesting idea has been recently launched by Day (1981) in a critical review of the role of myelin basic protein in the pathogenesis of EAE. The author hypothesized that the induction of EAE does not require specific antigen but rather a “bystander” antigen situated in the local environment of the myelin. This bystander model encompasses several biochemical processes, including the removal of MBP fragments from the vicinity of the myelin sheath by means of antibody. In conclusion, although it is generally believed that cell-mediated immunity plays the major role in EAE, anti-nervous tissue antibodies may also be involved in processes underlying the development of disease (Weigle, 1980; Paterson, 1982). Further studies should elucidate the participation of anti-brain antibodies and B-cells in the pathogenesis of EAE in general, and the inflammatory demyelinating mechanism in particular. An interesting possibility would be that EAE and intracerebral Arthus’ reaction represent a continuous spectrum of the pathological events (Roizin and Kolb, 1959). The role of Arthus’ reaction and immune complexes in EAE is by no means excluded. B. MULTIPLESCLEROSIS The presence of anti-nervous tissue antibodies in sera of patients with MS was reviewed many years ago (McAlpine et al., 1965) as well as recently (Trotter and Brooks, 1980). Current trends in MS research (Waksman, 1981) also include the problem of anti-brain antibodies in MS. It is well established that sera from MS patients cause demyelination of cultured nervous tissue (Bornstein and Appel, 1961; Bornstein, 1973;

NEUKOMODULATING ACTIVITY

OF ANTI-BRAIN ANTIBODIES

267

Grundke-Iqbal and Bornstein, 1979) similar to demyelinating activity of sera from EAE animals. While in EAE antibody binds to the myelin basic protein, in MS it is still not established whether antibody present in the CSF combines with myelin basic protein or with its peptide fragments (Gutstein and Cohen, 1978). Thus, both the specificity of antigen for MS and the nature of demyelinating factor(s) in MS sera are unknown. Evidence has been presented that IgG is most likely the active factor from MS serum (Grundke-Iqbal and Bornstein, 1979). Immunoelectrophoretic analysis showed an increase in immunoglobulins in CSF of MS patients (Tourtellotte and Parker, 1967; Lumsden, 1972) but the abnormal amount of immunoglobulins in CSF appears not to be pathognomonic of any particular disease. The enigmatic oligoclonal IgG in CSF of MS patients may be a product of B-cells which are activated by endogenous mitogens released at the injured sites of the CNS during the MS process (Paterson and Whitacre, 1981). Another possibility would be that the elevation of oligoclonal IgG of unknown specificity is an epiphenomenon reflecting “a defect in immunoregulation” (Trotter and Brooks, 1980). Complement-fixing anti-brain antibodies have been found in patients with MS but not in patients with most other neurological diseases (Ryberg, 1978, 1982). These antibodies exerted specificities for different components of the brain homogenate, and only galactosylceramide and sulfatide antigens have been immunochemically identified (Ryberg, 1978). Various aspects of humoral immunity in MS were recently reviewed (Karcher et al., 1979; Ryberg, 1982). In conclusion, an important and largely unanswered question concerns the identification of neuroantigens and anti-brain antibodies in serum and CSF of MS patients, and their involvement in the initiation and development of MS. C. CIRCULATING NERVOUSTISSUE FRAGMENTS Proteins of central nervous origin were demonstrated in the CSF of patients with neurological and psychiatric disorders (Dencker and Swahn, 1962). Brain injuries may be followed by release of large amounts of CNS antigens into the circulation (Thomas, 1975; Thomas et al., 1978). In addition, there is a physiological loss of neurons and clearance of neuronal components from the brain associated with aging (Ordy and Brizzee, 1975). These phenomena are integral parts of the neuromodulatory mechanisms operating in the organism. An interesting finding by Paterson and colleagues (Day et al., 1978a,b; Fujinami et al., 1978; Peskovitz et al., 1978; Paterson, 1979) is

268

BRANISLAV D. J A N K O V l k

that serum and CSF of normal Lewis rats and rats with EAE contain fragments of MBP, designated as MBP-SFs (myelin basic protein serum factors). These circulating fragments of MBP, which react with antiMBP antibodies, are supposed to act as neuroautotolerogens; i.e., MBPSFs are capable of performing an immunoregulatory function and affecting the development of EAE (Paterson, 1979). When considering the proposed neuroautotolerogenic function of MBP-SFs, it should be borne in mind that some immunodeterminants of MBP molecules may exert EAE-inhibiting activity, while others display encephalitogenic activity (Paterson, 1980). The clinical status of EAE may reflect different affinities of MBP-SFs (Day et al., 1978b). Whether MBP-SFs are similar to a serum factor from MS patients which has been shown to suppress EAE in animals (Tsaregorodtseva and Chernigovskaya, 1969) remains to be elucidated. T h e myelin breakdown process in MS leads to the presence of MBP and its fragments in CSF (Cohen et al., 1976; Whittacker, 1977; Paterson et al., 1981). It has been claimed that the levels of basic proteins parallel the clinical states of disease (Cohen et al., 1980). T o conclude, circulating MBP-SF molecules and other macromolecules of the neuron-glia complexes are derivatives of a physiological process of aging of the CNS and destructive processes related to EAE and MS. If so, the anti-brain antibodies which regularly occur in EAE and MS may take part in the removal of MBP-SFs and other circulating neuron-glia components from the blood and CSF. This mechanism for the clearance of CNS particles includes the formation of immune complexes which may then be excreted from the organism o r may combine with tissues and induce immunopathological changes (Lawley and Frank, 1980).

D. EPILEPSY Since antibodies against different brain structures and isolated neurons may evoke typical epileptic discharges (see below), one is tempted to assume that autoimmune processes, in this case anti-brain autoantibodies, may be involved in the pathogenesis of epilepsy. Kopeloff (1942) and Kopeloff et al. (1942) described recurrent convulsive seizures in animals produced by immunological means. In fact, the earliest reports of epileptic symptoms following intracerebral injection of anti-brain antibodies are those of Delezenne (1900) and Armand-Delille (1906). Epileptiform spikings were recorded in animals whose cerebral cavities o r tissue were injected with anti-brain antibodies (MihailoviC and JankoviC,

NEUROMODULATING ACTIVITY OF

ANTI-BRAIN ANTIBODIES

269

1961, 1965; Bowen, 1968; Walker, 1970; MihailoviC and CupiC, 1971; Karpiak et al., 1973, 1974; Ettlinger and Lowrie, 1976). Recurrent epileptiform episodes in the cortical electroencephalogram were induced when antiserum to total brain ganglioside was applied onto and into the sensorimotor cortex (Karpiak et al., 1976a). T h e epileptiform activity appears to be a common bioelectrical reaction of a neuron exposed to antibodies against different constituents of the nervous tissue. Antiserum against an actomyosin-like (neurostenin) protein from brain-induced focal epileptigenic activity (Bowen et al., 1976). Although the facts cited above are relevant to the development of epilepsy, the immunological interpretation of this disease cannot conclusively be accomplished until more information is available.

IV. Biological Activity of Anti-Brain Antibodies

The brain has a well-developed organ specificity; i.e., an anti-brain serum reacts with brains of unrelated species but distinguishes the brain from other organs within a species (Landsteiner, 1946). Owing to this organ specificity, the brains of phylogenically very distant species have some antigens in common so that, for example, an anti-brain antibody produced against mammalian brain may react with brain and neurons from lower animals. Twenty-five years ago, in the search for anti-brain antibodies which would allow structural and functional dissection of the brain, a series of experiments was set up to define antigenic differences between the frontal cortex, occipital cortex, temporal cortex, cerebellar cortex, caudate nucleus, thalamus, cerebral white matter, medulla, and spinal cord of the cat brain (JankoviC et al., 1960). Complement-fixation reaction and absorption analysis of antisera against different brain structures revealed pronounced cross reactivity. In other words, serological tests failed to discriminate clearly among various brain regions. T h e in vivo experiments, however, demonstrated that anti-brain antibodies were capable of distinguishing one brain structure from another (MihailoviC and JankoviC, 1961). It was deemed, therefore, that anti-brain antibodies could provide a powerful instrument for analysis of brain structures and functions. Indeed, one of the most significant developments in probing the immunoneuro relationship has been the successful application of anti-brain antibodies in analyses of the neuron and brain. Certain aspects of anti-brain antibody activity have been discussed previously (MihailoviC and JankoviC, 1965; JankoviC and RakiC, 1969;

270

BKANISLAV D. . J A N K O V I ~ :

JankoviC, 1972; Stein et al., 1976; Shtark, 1978; Rapport and Karpiak, 1978; Rapport et al., 1979; Barondes, 1982).

A. In Vzvo ACTIVITY 1. Intracerebral Application The earliest experiments with “neurotoxic” antisera were performed on dogs injected into the frontal lobes with duck antisera against dog forebrain, cerebellum, brain stem, and spinal cord. Dogs injected into the brain tissue quickly became paralyzed, and some animals exhibited epileptic symptoms, such as salivation and clonic and tonic convulsions. On the other hand, injections of 0.5-0.6 ml of normal duck serum into the cerebral tissue did not produce abnormalities (Delezenne, 1900). Histopathological changes were observed in the brain of dogs which were killed by intracerebral injections of anti-dog brain serum prepared in guinea pigs (Armand-Delille, 1906). These experiments are of value for the history of neuroimmunology.

2. Intraventricular Application The first intraventricular application of anti-brain antibodies to the study of brain functional region specificity was conducted by MihailoviC and JankoviC (1961). The rationale of this experiment was that an antibrain antibody, because of its extraordinary specificity,can react with the neuron and brain structure in a more physiological way than other foreign chemical and physical agents. For this purpose, y-globulin fraction from rabbit anti-cat caudate nucleus serum was injected into the lateral ventricle of the cat brain. Thus administered, anti-caudate nucleus antibodies induced pronounced modifications in electrical activity characterized by transient irritative phenomena, general accentuation of background activity, and progressive decrease in the amplitude of action potentials. These electrographic disturbances were strictly confined to the caudate nucleus and could not be observed in other brain structures. The possibility of cross reactivity due to residual components of circulating blood was excluded since anti-cat brain serum absorbed with cat blood produced the same effects on bioelectrical pattern as antiserum which was not absorbed with cat blood constituents. In another experiment with anti-hippocampus antibodies, long-lasting epileptiform discharges appeared in the hippocampus following intraventricular administration of anti-hippocampus antibodies. Normal rabbit sera or antisera

NEUROMODULAlING ACTIVITY OF ANTI-BRAIN ANTIBODIES

27 1

from which anti-brain antibodies were removed by absorption procedures did not affect bioelectrical activity of the caudate nucleus and hippocampus (MihailoviC and JankoviC, 1965). Since an antibody-antigen reaction at the cell level may induce the release of mediators such as histamine, serotonin, and acetylcholine, the content of histamine-like substance in the frontal cortex, caudate nucleus, hippocampus, and CSF was measured in cats after intraventricular injection of anti-caudate nucleus and anti-hippocampus antibodies. A predominant increase of histamine-like substance was found in the caudate nucleus and hippocampus after injection of anti-caudate and antihippocampus antibody, respectively (MihailoviC et al., 1964). Impedance changes were studied in cats intraventricularly injected with antibodies against cat thalamus and caudate nucleus (MatejiC et al., 1969). Recordings obtained showed that the interaction between antibrain antibodies and corresponding brain structures resulted in decreased resistance and increased capacitance. Anti-cat thalamus antibodies applied through the cannula into the lateral ventricle of the cat brain produced long-lasting desynchronization, single spike activity, and generalized large discharges in all cortical and subcortical structures (Radulovai‘ki and JankoviC, 1966). The experiment, originally designed to study the role of anti-brain antibodies in the pathogenesis of experimental allergic encephalomyelitis (see Section III,A), in which normal rabbits were treated intraventricularly with y-globulin fraction from the sera of rabbits which developed EAE, revealed that anti-brain antibodies provoked electroencephalographic abnormalities in the form of high-voltage slow activity in the occipital cortex and caudate nucleus and midbrain reticular formation, and irregularity in hippocampal theta rhythm (Fig. 2). The EEG pattern after multiple injections of anti-brain antibodies, particularly in rabbits with severe paralysis, was characterized by a phasic state with flattened background activity in all cortical and subcortical structures, but was most pronounced in the caudate nucleus and in the frontal and occipital cortex UankoviC et al., 1966a,b). The injection of antibodies into the cerebral cavity deserves an additional comment: namely, intraventricular, subarachnoid, and intracisternal routes of injections may be considered “extreme conditions” (Willenborg, 1982) of testing the biological activity of antibodies in,viva In fact, there is no serious reason why an injection into the cerebral cavity, and injection through a cannula (Feldberg and Sherwood, 1953) permanently inserted into the lateral ventricle in particular, should be more artificial than an intravenous injection. The injection through a chronic

272

RRANISLAV D.

L-RFr

JANKOVI~:

I

Cx

L-ROCC Cx

L

Cd

L Fr-Occ

CX

lsec FIG. 2. Changes in EEG activity of rabbit No. 77 after the ninth intraventricular injection (20 min) of anti-brain antibodies from a rabbit which developed experimental allergic encephalomyelitis. Note the high-voltage slow waves in caudate nucleus and occipital cortex, and irregularities in the hippocampal theta rhythm. Abbreviations: I., left; R, right; Fr Cx, frontal cortex; Occ Cx, occipital cortex; Cd, caudate nucleus; Hippo, dorsal hippocampus; Sept, septum pellucidum; and RF, midbrain reticular formation.

cannula avoids the trauma caused with the introduction of a needle directly into the cerebral tissue. In the latter case, the needle injures the cerebral tissue, breaks the brain-blood barrier by damaging blood vessels, and causes leakage from the site of injection. In contrast, the injection of antibodies via a chronic cannula has some advantages: the brainblood barrier is bypassed, there is small chance for anti-brain antibodies to cross react with antigens from other tissues, concentration of antibodies is higher (intravenous in-jection is accompanied by dilution of antibodies in the circulation), accessibility of antibodies to brain antigen is much easier, and clearance of antibodies from the site of injection is prolonged, which allows a better contact between antibodies and tissue antigens. Thus, contrary to an opinion expressed by Karpiak and Rapport (1975) that intraventricular cannula causes some ambiguities, the cannula inserted permanently into the cerebral cavity enables well-controlled conditions of experimentation. Incidentally, subarachnoid (Sherwin et al., 1963) and intraventricular (JankoviC et al., 1961; MitroviC et al., 1964) injections of antigen proved to be effective routes of stimulation of the body’s immune machinery. Besides, although the cat is peculiarly resistant to sensitization with antigens (Wilson and Miles, 1964), anaphylactic shock can be elicited in cats

NEUROMODULATINC ACTIVITY OF ANTI-BRAIN

273

ANTIBODIES

sensitive to bovine y-globulin by means of antigen injected into the lateral ventricle of the brain (Jankovit et al., 1969a).

3. Topical Application T h e brains of the lobster (Palanurus vulgaris) and cockroach (Blatta orientalis L.) were used in an experiment with local application of antibrain antibodies (Jankovit et al., 1969b). Antibodies were raised in rabbits with homogenized lobster brain and y-globulin fraction containing antibrain antibodies was applied directly to the surface of cockroach brain; electrical recordings were made. In 27 out of 32 tested insects, anti-brain antibodies produced bioelectrical abnormalities characterized by an increase in the amplitude of continuous spikes (Fig. 3) and the appearance of bursting activity. This experiment also provided evidence of the immunophylogenic relationship between lobster, rabbit, and cockroach, based on the known organ specificity of the brain. Namely, a number of rabbits inoculated with lobster brain in complete Freund's adjuvant developed experimental allergic encephalomyelitis in spite of the wide phylogenic distance between the lobster and the rabbit. T h e reaction between anti-lobster brain antibodies and cockroach brain indicated that the lobster and cockroach brain share some antigenic determinants. In mammals, antisera against nerve-ending membranes from the rabbit and cat cortex induced epileptiform discharges when applied topically upon the cerebral cortex (De Robertis et al., 1966). Anti-rat brain synaptosome membrane antibodies applied to the brain surface consistently produced a decrease in the amplitude of the slow potential wave, thus suggesting that anti-membrane antibodies exerted a suppressive effect on the transsynaptic component of the parallel fiber-evoked field potential (Jarosch and Precht, 1972). This effect was complement independent and reversible since there was no permanent injury of the neural membrane at the site of anti-brain antibody application. before

30 IIV

I+

''"

AFTER APPLICATION O F ANTI -LOBSTER BRAIN ANTIBODY + 1 min

I8.C

FIG. 3. Increased amplitudes of spikes of cockroach brain induced by application of anti-lobster brain antibodies on the surface of the cockroach brain.

274

HKANISLAV D. . J A N K O V ~ ~

B. In Vitro ACTIVITY 1 . Axon Studies of low (Schmitt and Davison, 1961) and high (Huneeus-Cox, 1964a) molecular constituents of the squid axon did not provide conclusive evidence on their function in the neuron. Intraaxonal perfusion with proteases destroyed both the action potential and the membrane potential (Rojas and Luxoro, 1965; Rojas, 1965; Huneeus-Cox, 1967). Perfusion of the giant axon of Dosidicus pgas with anti-axoplasm protein antibodies may inhibit the action potential without affecting the membrane resting potential (Huneeus-Cox, 196413). Similar results were obtained in experiments with lobster giant axons (Mihailovii. et al., 1965): antibodies against the ventral nerve cords of lobster (Palinurus vulgaris) affected the bioelectrical properties of the axon. The conduction of action potentials was blocked when the axon was immersed in artificial seawater containing antibodies. It has been speculated that after damage of a nerve in the brain, axonal growth is inhibited by antibodies produced to brain autoantigens released from the injured neuron (Berry and Riches, 1974).

2 . Subesophageal Ganglion of the Snail Increasing attention has been paid to the bioelectrical effects of antibodies against different subcellular and protein components of the brain. Mollusc ganglia were introduced by De Robertis and colleagues to the study of biological activity of anti-brain antibodies under in vitro conditions. Antisera to nerve ending membranes from the rabbit and cat cortex produced dramatic bioelectrical effects of mollusc (Cryptomphallus mpersa) neurons (De Robertis et al., 1966, 1968; Wald et al., 1968). Antibodies against nerve ending membrane exerted a pronounced cytotoxic activity when incubated with mollusc neurons (De Robertis et al., 1968). Ultrastructural analysis of those nervous preparations following incubation with antibodies revealed extensive vacuolization of cytoplasm and breakdown of the surface layer of cytoplasm. The incubation of isolated nerve endings of the cat cerebral cortex with anti-nerve ending serum resulted in various degrees of lysis and disintegration of nerve endings (De Robertis et a!.,1971). Relevant to this is the observation that antibodies prepared in rabbits against nerve endings from guinea pig cerebellum and cerebral cortex stained the cerebellar and cerebral cortex in immunoflourescent assays (Kornguth et al., 1969). Bioelectr i d effects of anti-brain antibodies on the neurons of isolated subesophageal ganglion of the snail (Helix pomatia) have been ex-

NEUKOMODULATING

ACTIVITY OF ANTI-BRAIN ANTIBODIES

275

tensively studied by Jankovik and colleagues. Using different experimental conditions, these investigators evaluated the neuromodulating capacity of antibodies to rat brain S- 100 protein, microsomes, synaptic vesicles, and synaptic membranes UankoviC et al., 1982a). Purified antibodies to S-100 protein altered the spontaneous activity of all tested neurons of the snail ganglion (Fig. 4).It should be emphasized that this effect was reversible since the removal of antibodies from the medium in which the ganglion was dipped was accompanied by reappearance of spontaneous spike activity. However, bioelectrical changes of the same

A

B #stl

1 sec

C 4Ab

D

FIG. 4. Pacemaker potentials of snail (Helix pornatia) ganglion neuron and changes produced by a n t i 6 100 brain protein antibodies. (A) Hyperpolarization induced by a train of stimuli (st) applied through the left pallial nerve o f the snail ganglion. (B) Generation of one spike by means of direct stimulation of the neuron (stl). ( C ) Addition of antibodies (Ah). Progressive inhibition of the spontaneous beating activity with slight depolarization of the membrane. (D) T h e same neuron 10 min after antibody addition. Train of stimuli and single stimulus d o not generate spontaneous firing.

276

B R A N I S L A V D. J A N K O V I k

A

B 4Ab

C LI-

4St

I IOsec

4 91 FIG.5. Different effects of anti-ratbrain microsome antibodies on spontaneous activity of the snail ganglion neuron. (A) Brief excitation after electrical stimulation (st) of the left pallial nerve. (B) Prompt excitation and striking depolarization after the application of antibodies (Ab), and disappearance of spontaneous activity. (C) The same antibody-treated

NEUROMODULATING ACTIVITY OF ANTI-BRAIN ANTIBODIES

277

ganglion preparation can be elicited again with a new addition of antibodies (SaviC et al., 1979a). Similar effects of the removal of antibodies were obtained in vitro with inhibition of myelin formation by means of antisera; myelinated axons appeared in nervous tissue explants within 5 days after the antiserum was discarded (Dorfman et al., 1976). Anti-rat brain microsome antibodies when brought into contact with beating neurons of the snail ganglion induced either membrane depolarization and disappearance of spontaneous spike activity or membrane hyperpolarization and inhibition of spontaneous discharges (Soltes et al., 1979). Representative recordings of those two types of bioelectrical changes generated by anti-microsome antibodies are shown in Fig. 5. In the case of a neuron with spontaneous bursting activity, anti-microsome antibodies caused depolarization followed by a short-lasting electrical silence and subsequent occurrence of modified low-amplitude pacemaker activity (Fig. 6). T h e effect produced by antibodies to rat brain synaptic vesicles on the bioelectrical activity of a pacemaker neuron of the snail ganglion was characterized by a slightly higher frequency of action potentials which increased with time (Fig. 7 ) .Of considerable interest was the response of a nonpacemaker (silent) neuron to anti-synaptic vesicle antibodies. The addition of these antibodies to the silent neuron induced the generation of spontaneous beating activity with apparent excitatory reactions to electric stimuli (Fig. 8). However, nonpacemaker properties of the same neuron reappeared after washing the neuron with special Ringer’s solution for molluscs. A striking finding was that the stimulation of the nerve trunk of a silent neuron washed in this way induced a short-lasting discharge. Such a response would imply that the contact of the nonpacemaker neuron with anti-brain antibodies was of sufficient “modulating” strength to reveal imperceptible functional changes which can be evoked by electrical stimulation of the neuron. This review of studies of the spontaneous bioelectrical activity of snail neurons exposed to different anti-brain antibodies will be concluded with an experiment in which antibodies against the rat brain synaptic membrane were employed. Figure 9 illustrates the modulation of spontaneous activity by means of anti-synaptic membrane antibodies. The characteristic feature was delayed occurrence of an increased frequency of action potentials. neuron does not respond to electrical stimuli 15 min after the addition of antibodies. (D) Short-lasting hyperpolarization effect following electrical stimulation (st) o f the nerve trunk. (E) Hyperpolarization and inhibition of bioelectrical activity produced by addition of antibodies (Ab). (F) The same antibody-treated neuron. Note the absence of responses to train of electrical stimuli. Recordings made 15 min after the addition of antibodies.

278

BRANISLAV D. J A N K O V I ~ :

A

B +Ab

4 0

C 4 st

Y

I IOsec

FIG.6. Bursting pacemaker activity of the F1 neuron o f t h e right parietal ganglion of the snail, and changes produced by anti-rat brain microsome antibodies. (A) Spontaneous bursting activity and single electrical stimulations (st). (B) Addition o f antibodies (Ab) induces short-lasting inhibition o f spontaneous activity and hyperpolarization. (C) T h e same neuron 5 min after the addition of antibodies. Note the disappearance of bursting activity and occurrence of modified low-aniplitude spontaneous activity. Electrical stirnulation does not affect the altered bioelectrical pattern of the neuron.

Immunohistochemical analysis revealed that sections of the snail subesophageal ganglion stain differently when exposed to anti-brain antibodies. For example, while anti-microsome antibodies reacted exclusively with cytoplasmic structures of the neuron, anti-synaptic membrane antibodies reacted with membranes and nerve fibers (Fig. 10). A similar immunoflourescent pattern was obtained with rat superior cervical ganglion treated with flourescent anti-ganglion serum and antisynaptic vesicle serum (JankoviC et al., 197513). In the above-mentioned experiments different anti-brain antibodies were used to study the spontaneous bioelectrical function of the neuron. However, anti-brain antibodies were also employed in studies of passive properties and evoked action potentials (AP) of the neuronal membrane. Anti-rat brain microsome and anti-rat brain synaptic vesicle antibodies were found to affect the evoked spike activity of the ganglion neuron of the land snail. Antibodies exerted different effects depending on the ionic concentration of Ringer’s solution. In normal Ringer’s solution, a concentration of 200 p g of antibody protein/ml was ineffective, whereas a higher concentration of 750 pg of antibody protein/ml inhibited the generation of evoked potentials (SaviC et al., 1982, 1983; Soltes et

NEUROMODULATINC ACTIVITY OF ANTI-BRAIN ANTIBODIES

279

I 10 sec FIG. 7. Delayed effects induced by antibodies against the rat brain synaptic vesicles. (A) Spontaneous bioelectrical activity of the snail ganglion neuron and excitatory effect following stimulation (st) of the left pallial nerve with a train of stimuli. (B) Slightly increased frequency of action potentials of the neuron immediately after addition of antibodies (Ab). (C) Stimulation of the same nerve trunk 5 min later produced a short-lasting excitation. (D) Increased frequency of spontaneous spike activity 20 min after the addition of antibodies.

al., 1983). In sodium-free Ringer's solution, anti-microsome antibodies increased the calcium-dependent part of the evoked AP. In calcium-free Ringer's solution, however, these antibodies decreased the amplitude of the evoked AP. The most interesting effects of anti-brain antibodies were observed in sodium-free Ringer's solution with blocked calcium channels of the neuronal membrane. After the blocking of calcium channels with CdClp, electrical stimuli failed to evoke APs. However,

280

RRANISLAV D. J A N K O V I ~ :

C

5 sec

D

40

r-

10 sec

FIG.8. A silent (nonpacemaker) neuron of the snail subesophageal ganglionic complex and effects produced by anti-rat brain synaptic vesicle antibodies. (A) Stimulation (st) of the nerve trunk provokes a short-lasting beating activity. Direct stimulation (stl) of the neuron elicits a single spike. (B and C) The same neuron. Application of antibodies (Ab) induces the appearance of long-lasting spontaneous beating activity with an excitatory reaction after stimulation (st) of the nerve trunk. (D) Fifteen minutes later, reappearance of silent properties of the neuron after washing of the ganglion with Ringer's solution. Stimulation (st) of the nerve trunk induces a short-lasting discharge, whereas direct stimulation (stl) of the neuron elicits only one action potential.

application of anti-brain antibodies induced a prompt generation of evoked spikes (SaviC et al., 1983). These results indicate that the biological activity of anti-brain antibodies depends on the ionic composition of the extracellular medium, and calcium in particular. Anti-brain antibodies altered the passive electrical properties (resistance, time constant, and capacitance) of the neuronal membrane. Differ-

NEUKOMODULATING ACTIVITY OF ANTI-BRAIN ANTIBODIES

28 1

A

4stl

4stl

C

4 51

4st

F 4St

4 10 sec

FIG.9. A neuron with spontaneous beating activity of the snail subesophageal ganglion and effects produced by anti-rat brain synaptic membrane antibodies. Spontaneous spike activity and effects of electrical stimulation (st) of the nerve trunk (A) and direct stimulation (stl) of the neuron (B). (C) There are no changes in the pacemaker activity immediately after the addition of antibodies (Ab). (D) Increased frequency of spikes which occurs 5 min after antibody addition. The neuron does not react to single stimulus and train of electrical stimuli. (E and F) The same neuron 20 min after antibody addition. Reappearance of normal bioelectrical activity. Direct stimulation (stl) of the neuron elicits a single spike (E). Stimulation of the nerve trunk (st) produces an effect similar to that before exposure of the neuron to antibodies (F). These recordings illustrate the slightest changes which can be produced by antibodies.

282

BRANISLAV D. J A N K O V X ~

FIG. 10. Immunofluorescence microphotographs (x600) of the subesophageal ganglion of the snail (Helix pomatzu) showing different localizing properties of anti-nervous tissue antibodies. (A) A section of the ganglion treated with anti-rat brain microsome antibodies. Note the concentration of fluorescing material in the neuronal cytoplasm, and unstained neuronal membranes and nuclei. (B) A section of the ganglion exposed to antirat brain synaptic membrane antibodies. Note the strong fluorescence of membraneous structures and unstained cell bodies.

ent anti-brain antibodies produced different effects. Anti-synaptic vesicle antibodies caused an increase of resistance and time constant. On the other hand, anti-microsome antibodies induced a decrease of resistance and time constant. However, both antibrain antibodies exerted similar effects on capacitance (Soltes et al., 1983). 3. Superior Cervical Ganglion of the Rat Yet another model for studying anti-brain antibodies is the sympathetic superior cervical ganglion of the rat (JankoviC et al., 1975a; SaviC et al., 1976, 197913). The isolated superior cervical ganglion was treated with rabbit anti-rat brain synaptic membrane antibodies purified on chromatography columns. T h e surface action potential, the slow inhibitory postsynaptic potential (s-IPSP) and the slow excitatory postsynaptic potential (s-EPSP) were recorded before and after treatment of the ganglion with anti-synaptic membrane antibodies (50 p g of antibody nitrogen). This treatment of the superior cervical ganglion resulted in abolition of the s-IPSP and the consequent increase of the action potential.

NEUROMODULATINC ACTIVITY OF ANTI-BRAIN ANTIBODIES

283

4st

4st

4stl

,

, 1 sec

D 4Ab

Cst

4St

Cstl

I

10 sec

FIG. 11. Spontaneous activity of snail ganglion neuron and effects produced by antibodies against rat superior cervical ganglion, Normal bioelectrical activity of the neuron and effects of stimulation (st) of the left pallial nerve with a single stimulus (A) and train of stimuli (B). (C) Direct application of stimulus (stl) on the neuron elicits one spike. (D) Addition of antibodies (Ab) evokes a strong polyspike discharge and striking depolarization followed by complete disappearance of spontaneous activity, (E) The same neuron 10 min after addition of antibodies. Electrical stimulations do not elicit spontaneous bioelectrical activity.

Of interest from the immunological point of view is an experiment in which the snail ganglion was exposed to antibodies against the rat superior cervical ganglion. In spite of the great biological distance between the snail and the rat, antibodies produced against rat ganglion had striking effects on the bioelectrical activity of the snail ganglion neurons: initial excitatory activity was followed by complete disappearance of the spike activity (Fig. 11).

2 84

RRANISLAV D. J A N K O V I ~

An attempt was made to approach the recognition mechanisms related to the collision between antigen and antibody at cell membrane level. For this purpose, the cervical ganglion was used as the indicator of the antigen-antibody reaction (JankoviC et al., 1975~).The procedure consisted of two steps: the ganglion was first exposed to a given concentration of anti-ganglion antibodies prepared in rabbits, and thus “sensitized” ganglion (coated with rabbit Ig molecules) was exposed to chicken anti-rabbit Ig antibodies. It was expected that the reaction between rabbit Ig and chicken anti-Ig at the neuronal membrane level would affect the bioelectrical activity of the ganglion; i.e., the ganglion would record this antigen-antibody reaction. Indeed, amplitudes of surface action potentials of isolated ganglion treated in this way were 25-41% above the initial values. This experiment demonstrated that the collision between an Ig molecule (bound by specific forces to the neuronal membrane) and corresponding anti-Ig antibody (a molecule which does not normally bind to the neuronal membrane) exerted a stimulatory effect on the neuronal membrane. In this case, the neuronal membrane served as a recorder of an irrelevant antigen-antibody reaction. Most probably, energy produced by antigen-antibody reaction was of sufficient strength to induce changes in the neuronal membrane and consequent increase of the bioelectrical activity. On these grounds, one may assume that under in vivo conditions an idiotypic antibody in the blood and CSF may react with anti-brain antibody molecules which are attached to the neuron and cause functional abnormalities. This possibility should be taken into account when considering the pathogenesis of autoimmune neurological and psychiatric disorders.

4. Nervous Tissue Explants Kimura (1928) was the first to apply anti-brain serum to cultures of neuronal cells. Since several aspects of the activity of anti-brain antibodies on cultured nervous tissue have been considered earlier (see Section III,A), only a few remarks relating to bioelectrical activity will be made here. Propagation of nervous impulses was affected by exposure of cultured brain explants to serum from animals with EAE and patients with MS (Bornstein and Crain, 1965).This effect was reversible, since normal bioelectrical activity may be reestablished after removing the tissue from the serum. Reversible depression of polysynaptic reflex responses was produced by means of serum from patients during acute exacerbations of MS applied to isolated spinal cord of the frog. This factor was complement dependent and its effect was attributed to the action on synaptic processes in internuncial neurons, or to functional alteration of the subsynaptic membrane of metaneuron dendrites (Cerf and Carles, 1966).

NEUROMODULATING ACTIVITY OF ANTI-BRAIN ANTIBODIES

285

Addition of antiserum with complement to human meningioma cells in tissue culture produced depolarization and a decrease in resistance (Prieto et al., 1967). 5. Neuron and Anti-Thymocyte Antibody The antigenic relationship between the brain and lymphoid cells is well established (Reif and Allen, 1964; Golub and Day, 1975). Lymphocytes share some antigenic determinants with brain microsomes UankoviC et al., 1975a), tubulin (JankoviC et al., 1977a), cervical ganglion UankoviC et al., 1977b), synaptic membranes UankoviC et al., 1977c), synaptic vesicles UankoviC et al., 1979c), membrane and fiber components of neurons (Barclay and HydCn, 1979); and astrocytes in longterm cultures (Pruss, 1979). If those antigens are associated with the neuronal membrane, then they may be involved in membrane functions. To test this hypothesis, anti-rat thymocyte serum, which in immunofluorescent assays has been shown to react with membranes, axons, and nerve endings of the snail ganglion neuron, was used for the treatment of isolated snail ganglion UankoviC et al., 1980b). Both pacemaker and nonpacemaker neurons exhibited significant bioelectrical changes when exposed to anti-thymocyte serum (Fig. 12). These results may have an important bearing upon the pathogenesis of immunoneurological and immunopsychiatric diseases, since patients with MS contain both antibrain and lymphocytotoxic antibodies (Schocket et al., 1977), and the latter antibodies may join anti-brain antibodies in the pathological mechanisms related to MS.

C. OTHERANTI-BRAIN ANTIBODIES

1. Anti-Ganglioside Antibodies Flourescent antibody technique revealed that antibodies against purified gangliosides localize in the body of nerve cell of the bovine brain (Bogoch, 1960). In an attempt to ensure the reproducibility of tests and to rely on less complex antigenic mixture, Rapport, Karpiak, and colleagues used anti-ganglioside sera in neuroimmunological studies (Karpiak et al., 1976a, 1978, 1981; Karpiak and Rapport, 1979; Rapport, 1981). It was expected that anti-ganglioside antibodies would predominantly interfere with synaptic function since synaptic membranes have a large amount of gangliosides. In spite of a wide distribution of gangliosides in the brain (Rapport and Mahadik, 1977), the interference of antiganglioside antibodies with CNS functions was selective (Rapport, 198l),

kst1

4stl

IIII

IIIIIII

A Il J

w

7

1 sec

1st FIG. 12. The influence of anti-rat thymocyte membrane antibodies on beating activity of the snail ganglion neuron. Hyperpolarizing effects of stimulation (st) of the left pallial nerve with a single stimulus (A) and train of stimuli (B). (C) Direct stimulation (stl) of the neuron elicits one spike. (D) Application of antibodies (Ab) against thymocyte membranes evokes prompt disappearance of the spontaneous activity. (E) Stimulation of the nerve trunk (st) and direct stimulus of the neuron (stl) do not provoke spike discharges.

thus confirming an editorial commentary in Nature (London) (1968) that the immunological technique used by MihailoviC and JankoviC (1961) “promises to be a particularly powerful method of investigation of the brain.” Anti-ganglioside antibodies applied locally produced recurrent epileptiform spiking (Karpiak et al., 19?6b), inhibited a learned avoidance response (Karpiak et al., 1978), and suppressed synapse formation

NEUROMODULATING ACTIVITY OF ANTI-BRAIN ANTIBODIES

287

and processes related to dendritogenesis and myelinogenesis (Rapport, 1981). Antibodies specific to bovine gangliosides blocked the neuritic outgrowth from regenerating goldfish retinal explants (Spirman et al., 1982). Mice with sciatic nerve injury produced antibodies to gangliosides and myelin protein (Schwartz et al., 1982), and the lymphocytes of those mice were sensitized to the nervous tissue autoantigens. Since surface gangliosides of the myelin structure play a role in the process of regeneration (Ceccarelli et al., 1975), and administration of gangliosides accelerates the rate of sciatic nerve regeneration (Gorio et al., 1980) and increases the number of neurites (Hauw et al., 1981), it can be deduced that anti-myelin ganglioside antibodies are involved in mechanisms underlying certain immunoneurological diseases (Nagai et al., 1980). It has been speculated that the function of anti-ganglioside antibodies concerns the release and uptake of neurotransmitters. Namely, anti-ganglioside antibodies increase the release of GABA induced by depolarization with K+ (Frieder and Rapport, 1980), without affecting the spontaneous release (Frieder and Rapport, 1981), enhance the output of norepinephrine while that of serotonin remains unaffected, and inhibit the GABA binding of synaptosome-enriched fractions of the cerebral cortex (De Feudis et al., 1980). 2. Monoclonal Anti-Brain Antibodies The technology of monoclonal antibody production (Kohler and Milstein, 1975) is widely expected to provide a new way of obtaining homogeneous antibodies that will be useful in investigating many problems related to the structure and function of the cell. The list of monoclonal antibodies employed in neurosciences is only just started. It is possible to discriminate between neurons of neural tube origin and neurons of neural crest origin by means of monoclonal antibodies (Cohen and Selvendran, 1981 ; Vulliamy et al., 1981). Monoclonal antibody assay detected one antigen in the chick embryo retina that may be involved in directing retinal projections within the nervous system (Trisler et al., 198 1). Monoclonal antibodies to nicotinic acetylcholine receptors have been obtained (Gomez et al., 1979; Moshly-Rosen et al., 1979; ContiTronconi et al., 1981), and when injected into animals produced pathological effects similar to myasthenia gravis (Tzartos and Lindstrom, 1980; Tzartos et al., 1981). Attempts have been made to localize glioma (Phillips et al., 1982) and other neural tumors in humans (Kennett et al., 1980) by monoclonal antibodies. In studies of the leech nervous system, of 300 monoclonal antibodies that reacted with nervous tissue components,

288

BRANISLAV D.

JANKOVIC

about 40 were found to be specific for restricted sets of neurons and only a few of them could be involved in biosynthesis and response to ligands (Zisper and McKay, 1981). Immunohistological evaluation of monoclonal antibodies against brain proteins S- 100 and 14-3-2 revealed that anti-S-100 localized in the epithelial cells of the cerebellar cortex and in astrocytes of the hippocampus, dental gyrus, and neocortex, but not in neurons and other glial cells (Haan et al., 1982). It should be mentioned, however, that S-100 has been shown to bind to synaptosomes and glial nuclei in nitro (Donato, 1981; Donato and Michetti, 1981). Thus the issue of heterogeneity of S-100 protein and the localization of its individual components remains to be solved (see Section 11,B).So far, in spite of the very promising monoclonal antibody technology, none of the anti-neuron monoclonal antibodies is sufficiently specific to identify only a welldefined population of neurons, but rather labels more than one kind of neuron (McKay et al., 1981). Nevertheless, the bioelectrical activity of the neuron can be analyzed by means of monoclonal reagents. D. HIGHERNERVOUSACTIVITY 1. Sleep Sleep is an extraordinarily difficult challenge for neurochemistry (Karnovsky and Reich, 1977) and has been scarcely touched by immunologists. On the assumption that the processes of sleep and wakefulness are controlled from different centers Uouvet, 1965), it was anticipated that the injection of different anti-brain antibodies into the lateral ventricle of the brain would provide more information about the location and function of the sleep-waking structures. For this purpose, antibodies were produced in rabbits against the frontal, occipital, pontine, nucleus reticularis pontis caudalis and caudal third of nucleus reticularis pontis oralis, and midbrain reticular formation tissue from the cat brain. Purified anti-brain antibodies were used for intraventricular injection of cats bearing monopolar and bipolar deep electrodes. Thirty-one cats were first deprived of paradoxical sleep Uouvet et al., 1964) for 72 hr, then injected intraventricularly with rabbit normal y-globulin lacking antibrain activity and transferred to a sound-proof chamber, and EEG was recorded for 24 hr. After that, the animals were allowed to rest for 72 hr, then again were paradoxical sleep deprived for 72 hr, and injected with different anti-cat brain antibodies. Further EEG recordings lasted for 24 hr. T h e effect of anti-brain antibodies on sleep and wakefulness are presented in Table 111 (RadulovaCki and JankoviC, unpub-

289

NEUROMODULATING ACTIVITY OF AN'II-BRAIN ANTIBODIES

TABLE I11 PERCENTAGE OF WAKING STAI'E,SLOWSLEW(SS), A N D PARADOXICAL SLEEP(PS) FOLLOWING INJECrION OF ANTI-BRAIN ANTIBODIES I N T O T H E LATERAL VENTRICLE OF T H E BRAIN OF CAIS DEPRIVE11 OF PARAI1OXICAL SLEEP" ~

Kind of sleep Group FC Normal globulin Anti-FC antibody

Cat number

Waking state

ss

PS

23.2 26.1 (1.12)

61.7 65.9 (1.07)

15.1 8.0 (0.53)

32.5 33.5 (1.03)

53.8 58.8 (1.09)

13.7 7.7 (0.56)

24.8 25.4 (1.02)

66.1 66.0 (0.99)

9.1 8.6 (0.94)

14.6 36.2 (2.48)

63.3 55.5 (0.88)

22.1 10.3 (0.37)

17.6 39.2 (2.23)

62.4 57.2 (0.92)

28.0 3.5 (0.12)

3

OCC Normal globulin Anti-OCC antibody

11

MBRF Normal globulin Anti-MBRF antibody

13

NRPC Normal globulin Anti-NRPC antibody

23

Pons Normal globulin Anti-pons antibody

28

Numbers in parentheses indicate the ratios of the experimental values (after intraventricular injection of anti-brain antibody) to the control values (after administration of normal globulin). FC, frontal cortex; OCC, occipital cortex; MBKF, midbrain reticular formation; NRPC, nucleus reticularis pontis caudalis and the caudal third of nucleus reticularis pontis oralis.

lished results). Analyses based on individual cats showed that the application of anti-pons antibodies caused the most pronounced reduction of paradoxical sleep and a striking increase of waking state. A moderate decrease of paradoxical sleep and increase of waking state was observed in cats treated with antibodies against nucleus reticularis pontis caudalis and the caudal third of nucleus reticularis pontis oralis. T h e results summarized here suggest that anti-brain antibodies may be used in the analysis of the sleep-waking state. T h e mechanism of antibody action is not clear, but the possibility remains that antibodies may interfere with the release of neurotransmitters Uouvet, 1976) which are involved in the sleep mechanisms. Anti-reticular formation antibodies were found to produce a specific decrease in rapid-eye-movement sleep without altering slow-wave sleep, thus indicating that some specific protein molecules participate in the

290

BRANISLAV I). . J A N K O V I ~

regulation of rapid-eye-movement sleep (Drucker-Colin et al., 1980). Those antibodies also affected some phasic elements of rapid-eye-movement sleep such as multiple unit and eye movement bursts. The described effects of anti-brain antibodies were almost identical to those produced by protein synthesis inhibitors on the phasic periods of rapideye-movement sleep (Drucker-Colin et al., 1979). 2. Learning, Memory, and Behavior Investigations of behavioral effects of anti-brain antibodies have been initiated by JankoviC and co-workers UankoviC et al., 1968; Mihailovid et al., 1969). In the first experiment UankoviC et al., 1968), cats with electrodes implanted in cortical and subcortical sites and a silver disk electrode attached to the dorsum of the right hind leg were trained to discriminate between a “positive” tone of 800 Hz coupled with an aversive stimulus and a “negative” tone of 700 Hz. Fully trained cats received a single injection of anti-brain antibodies through intraventricular cannula, and defensive conditioned reflexes were tested after the injection. The results (Table IV) showed that the administration of anticat brain ribonucleoprotein antibodies was followed by significant and long-lasting changes in conditioned responses. In the second experiment (MihailoviC et al., 1969), monkeys were trained to perform delayed alteration and visual discrimination tasks. After intraventricular injection of anti-monkey caudate nucleus and anti-monkey hippocampus antibodies, animals were tested for retention. Monkeys injected with both kinds of anti-brain antibodies were significantly impaired in delayed performance, whereas animals inoculated with control normal y-globulin retained criterion performance on both tasks. Recently, quantitative analysis of the locomotor activity of rats injected via a cannula into the lateral ventricle of the brain with either anti-rat brain S- 100 protein o r anti-rat brain synaptic membrane antibodies revealed that anti-brain antibodies caused pronounced changes in the open field behavior of rats (MarkoviC and JankoviC, 1983). These experiments established reasonable grounds on which the study of mechanisms of some immunopsychopathological phenomena might be attempted. A variety of experiments have shown that anti-brain antibodies interfered with processes associated with consolidation of memory, learning rates in training, and development. Thus, repeated injections of antibrain antibodies affected the learning and produced irreversible damage to the brain (MacPherson and Chinerman, 1971). Anti-synaptosoma1 plasma membrane antibodies applied through a cannula in the rat brain induced impairment in memory retrieval (Kobiler et al., 1976). In an independent study, Karpiak et al. (1977) found that anti-ganglioside

TABLE IV

Loss OF DIFFERENTIATION I N TRAINED CATSFOLLOWING INTRAVENTRICULAR INJECTION OF ANTI-CATBRAINRIBONUCLEOPROTEIN (RNAP) ANTIBODY Response to “positive” tone (800 Hz)

Response to “negative” tone (700 Hz)

Number of cats

Total stimuli

Number correct (%)

Number incorrect (%)

Total stimuli

Correct

Material injected

(96)

Incorrect (%)

Anti-cat brain RNAP antibody“ Anti-cat liver RNAP antibody” Rabbit normal y-globulin Saline

14 5 7 9

1916 342 472 448

1050 (54.8) 329 (96.2) 446 (94.5) 424 (94.6)

866 (45.2) 13 (3.8) 26 (5.5) 24 (5.3)

1454 366 494 450

849 (58.4) 348 (95.1) 466 (94.3) 422 (93.7)

605 (41.6) 18 (4.9) 28 (5.7) 28 (6.2)

Anti-cat brain RNAP and anti-cat liver RNAP antibodies were produced in rabbits.

292

RKANISLAV D. J A N K O V I ~

serum inhibited the consolidation phase of memory but not the acquisition phase. Antisera to a brain synaptic membrane fraction, pentanol extract antigen, S- 100 protein, and ganglioside GMI injected intraventricularly into rats immediately after training impaired the learning process (Kapport et al., 1979). Learned avoidance response is altered in rats by antibodies to ganglioside GMI (Karpiak et al., 1978). Anti-S-100 antibodies, when applied into the cortex of the rat brain, inhibited the behavioral performance (maze learning) of the rat but did not alter cortical EEG patterns (Karpiak et al., 1976a). These results suggested that an anti-brain antibody may affect behavior without influencing bioelectrical phenomena. S- 100 protein was correlated with behavior (Hyden and Lange, 1970) and anti-S-100 antibodies injected intraventricularly into adult rats during the course of training prevented animals from further learning (Hyden and Lange, 1971). It has been reported that actively immunized rats with anti-brain microsome and anti-liver microsome antibodies circulating for a month required more time and trials to learn a visual discrimination problem than control rats immunized with egg albumin (MacPherson and Shek, 1970). These results on the effect on learning processes of circulating antibodies against brain and liver microsome wait for supporting evidence from other laboratories. A number of experiments were concerned with development. Twomonth-old rats from mothers injected with antibodies against rat brain synaptic membrane fraction demonstrated marked behavioral deficits (Karpiak and Rapport, 1975). Rat pups injected subcutaneously for 14 consecutive days with anti-rat hippocampus synaptosome membrane antibodies showed impairments in T-maze learning, passive avoidance task and one-way avoidance learning (Hofstein et al., 1980). It was assumed that antibodies injected into neonatal rats might reach the brain by passing through the immature blood-brain barrier. The treatment of pregnant and newborn rats with anti-ganglioside and anti-galactocerebroside antibodies induced long-lasting behavioral abnormalities (Adinolfi et al., 1982). The same investigators reported that rats born to mothers injected with anti-myelin antibodies showed motor deficiency in the open field test. I n spite of this short list as yet available in the literature on the interference of anti-brain antibodies with learning, memory, and behavior, differences in materials and procedures used for the production of antibodies, variety of antibody specificity, distinct localization in vivo of antibodies, and diversity of conceptual and methodological approaches, the fact remains that all tested anti-brain antibodies have induced behavioral changes.

NEUKOMODULATING ACTIVITY OF

ANTI-BRAIN ANTIBODIES

293

V. lmmunopsychiatric Diseases

The term “immunopsychiatry” has been introduced by JankoviC et al. (1980a) to denote that both antibody-mediated and cell-mediated immune mechanisms are implicated in the pathogenesis of certain mental diseases. The following main developments have opened promising avenues for immunopsychiatric investigations: (1) demonstration that the electrical activity of the cat brain is influenced region specifically by intraventricular injection of antibodies to homologous regional antigen (MilhailoviC and JankoviC, 1961); (2) clinical observation that a serum globulin fraction called “taraxeip” from schizophrenic patients given to human volunteer subjects produced symptoms resembling the schizophrenic psychotic state (Heath, 1966); (3) changes of memory induced by anti-brain antibodies injected into the cerebral cavity of the cat brain (JankoviC et al., 1968); and (4)relationship between cerebral atrophy of unknown origin and delayed hypersensitivity to neurotissue antigens in humans (JankoviC et al., 1977a). While the first three experimental and clinical studies dealt with the biological activity of anti-brain antibodies, the fourth one concerned the role of cellular (cell-mediated) immunity in the pathogenesis of certain mental disorders. Reports from different laboratories provided a reasonable body of evidence in favor of the capacity of anti-brain antibodies to affect the neuron and behavior (see Section V,B, Anti-Brain Antibodies in Different Diseases). The immunological basis of schizophrenia is still under consideration, whereas the participation of cell-mediated immunity in the development of some psychiatric diseases waits to be confirmed or assailed. Therefore, brief statements regarding the immunological dimensions of schizophrenia and the problem of neurotissue hypersensitivity in psychiatric diseases are appropriate here. But first a few words about anti-brain antibodies in aging and in patients suffering from different diseases.

A. AGINGAND DEMENTIA Naturally occurring antibodies reactive to brain constituents were detected in animals (Martin and Martin, 1975a,b; Harbeck et al., 1976). The amount of these antibodies augmented with aging (Nandy, 1973; Martin and Martin, 1975a). Testing of animal and human sera against neurons and myelin basic protein yielded an increased number of positive results with advancing age (Eddington and Dalessio, 1970). In humans, antibodies with binding capacity to brain antigens have

294

BRANISLAV D. .JANKOVIk

been detected in normal sera (Allerand and Yahr, 1964). Anti-brain antibodies were found in 20430% of apparently normal subjects (Raskin, 1955; Hughes and Field, 1967; Eddington and Dalessio, 1970), but the amount of autoantibodies was higher in aged people (Strickland and Hooper, 1972). Using a cytotoxic assay system in which mouse undifferentiated neural tumor cells served as target cells for normal human sera, Chaffee et al. (1978) found, however, that many persons of age 20-25 have higher levels of anti-brain antibody than do persons either older or younger. This discrepancy in results certainly requires further investigation. Immune reactions in aging humans have been reviewed by Nandy (1977). T h e relationship between autoimmunity and aging, with special reference to amyloidosis as a possible immunological entity associated with the aging process, has been commented on by Land and Carlo (1975). The current immunoneurological concept of aging is based on the belief that the immune system, which at a younger age exerts defensive functions, provides destructive mechanisms in senescence (Makinodan, 1976). T h e autoimmune process has been implicated in the pathogenesis of neuronal alterations in dementia (Tower, 1978). ‘The presence of amyloid in the core of plaques of the dementia brain has been associated with disorders of immune function (Nielsen ~t al., 1977; Perry et al., 1978). Several aspects of the dementia syndrome are presented elsewhere (Terry, 1978; Small and Jarvik, 1982). U. ANTI-BRAIN ANTIBODIES I N DIFFERENT DISEASES Antibodies capable of reacting with neural cell nuclei of the guinea pig brain were demonstrated in the sera of patients suffering from sensory carcinomatous neuropathy (Wilkinson and Zeromski, 1965). The 7-S y-globulin from serum and CSF of patients with neurological diseases exhibited affinity to the glia and myelin sheath of normal human nervous tissue (Allerand and Yahr, 1964). Approximately half of the patients with Huntington’s disease contained antibodies to neuronal antigens (Husby et al., 1978). These antigens are present in the normal brain and do not appear to be unique to the brain of patients with Huntington’s disease (Husby et al., 1979). Lymphocytes from patients with Huntington’s disease produced in vitro positive delayed sensitivity reactions to human nervous system antigens (Barkley et al., 1977a,b). Antibodies against brain structures were observed in patients with cerebrovascular accidents (Motycka and Jezkova, 19’75).Recent data bearing on anti-brain antibodies in different neurological diseases were presented

NEUROMODULATING A C T I V I T Y OF ANTI-BRAIN ANTIBODIES

295

in a dispersed fashion in three monographs (Rose, 1979; Karcher et al., 1979; Ader, 1981). Patients with systemic lupus erythematosus (SLE) complicated with neurological symptoms produced antibodies against neuronal cytoplasm (Diederichsen and Pyndt, 1970; Quismorio and Friou, 1972). It was supposed that anti-neuronal antibodies in the serum and CSF of these patients were related to the pathogenesis of the brain dysfunction in SLE (Wilson et al., 1979; Williams et al., 1981). This assumption was supported by findings that anti-neuronal antibodies of the IgG and IgA classes were present in the brain of an SLE patient who had grand ma1 seizures and psychoses (Inoue et al., 1982). Pertinent to this are reports on the immune complex deposits in the choroid plexus in SLE patients with neurological signs (Atkins et al., 1972; Lampert and Oldstone, 1973; Sher and Pertschuk, 1974; Oldstone and Lampert, 1974; Gershwin et al., 1975). It is probable that deposits of immune complexes in the choroid plexus may induce breakdown of the blood-brain barrier. An interesting suggestion was that lymphocytotoxic antibodies from SLE patients may be responsible for the neurological symptoms (Bluenstein and Zvaifler, 1976). Sera from children with rheumatic chorea and active rheumatic carditis were found to contain IgG antibody reacting with cytoplasmic antigens from neurons of human caudate and subthalamic nuclei (Husby et al., 1976). This antibody cross reacted with Group A streptococcal membranes, and to a lesser extent with Group A wall preparations, but not with Group A carbohydrate and Group D streptococcal membranes. It appears that the presence of anti-neuronal antibody correlated with the severity and duration of the disease (Husby et al., 1976). Similar results were obtained by other investigators (Kingston and Glynn, 1976), who showed that antibodies from patients with Sydenham’s chorea reacted with human brain structures such as fibers running into the brain from the external pial-glial membranes, fibrous astrocytes in the corpus callosum, and ependyma and choroid plexus.

C. SCHIZOPHRENIA The heterogeneous group of psychiatric disorders known collectively as schizophrenia (Bleuler, 1950) encompasses numerous functional changes at the clinical, pharmacological, and biochemical levels. The etiology of schizophrenia is still a matter of dispute. In the light of the wide-ranging antecedent insults known to produce schizophrenic

296

BRANISL.AV D. . J A N K O V I ~

symptoms (Davison and Bagley, 1969), it has been recently suggested that perinatal infections may be implicated in the etiology and pathogenesis of schizophrenia (Murray and Reveley, 1983). However, the hypothesis that an infectious agent may play a dominant role in the pathology of schizophrenia is considered to be unlikely (Crow, 1983a,b) because of the occasional occurrence of schizophrenia in association with numerous infectious diseases. Nevertheless, this assumption is of interest since an infection of the CNS may be accompanied by antibodies that can combine with antigens situated in the brain. Various biological aspects of schizophrenia have recently been discussed (Snyder, 1982; Crow, 1983b). Abnormalities on immunoglobulin level in the serum and CSF of schizophrenics (Fessel, 1962;Jensen et al., 1964; Amkraut et al., 1973) d o not a priori mean that elevated immunoglobulins are pathognomonic for schizophrenia (Solomon et al., 1969). These elevated levels of Ig in schizophrenic patients were interpreted to be a reflection of a “general nonspecific immunohypersensitivity,” whatever this means. Several investigators reported on the presence of anti-brain antibodies in schizophrenic patients (Vartanian, 1966; Semenov, 1968; Kolyaskina and Kushnir, 1969; Gosheva et al., 1969; Vartanian et al., 1978), but some sera from schizophrenics did not precipitate with homogenates prepared from whole brain or brain regions (Fessel, 1963; Jensen et al., 1964; Vulchanov and Hadjieva, 1964; Kety, 1965; Rubin, 1965). The evidence suggesting the immunological nature of schizophrenia was provided by studies of Heath and co-workers. These investigators isolated from the serum of schizophrenic patients a protein designated “taraxein” (denominating confusion and disorder in the mind), a specific subfraction of serum IgG which appeared to be characteristic of schizophrenics but which was not present in normal individuals. Abnormal waves of the caudate nucleus and septa1 area in schizophrenic subjects can be recorded in monkeys injected into the lateral ventricle with taraxein from actively ill schizophrenics (Heath and Krupp, 1967; Heath et al., 1967a,b). At postmortem examination, specific brain regions exposed to taraxein exhibited positive immunoflourescence. It has been concluded that sera from schizophrenic patients contain a specific globulin (taraxein) which may be an antibody that reacts with antigen of the septal-basal caudate region of the brain. Taraxein given to human volunteer subjects induced symptoms resembling the schizophrenic psychotic state (Heath, 1966). On this ground, Heath and colleagues advanced the hypothesis that schizophrenia is an immunological disorder. The original version of this hypothesis was modified to the idea that schizophrenia is characterized by the presence of autoantibodies which

NEUROMODULATING ACTIVITY OF ANTI-BRAIN ANTIBODIES

297

react with receptor sites for transmitters in the neurons of the septal region (Garey et al., 1974). This view was substantiated by changes in EEG and behavior observed in cats injected intraventricularly with antibodies against pre- and postsynaptic membrane of the septal region of the human brain. It has been suggested that in uiuo binding of taraxein with neuronal antigens alters neuronal activity at focal sites and causes psychotic signs and symptoms in schizophrenia (Heath, 1969). Supporting evidence for the immunological concept of schizophrenia came from the work of Baron et al. (1977). However, not all investigators agreed with the immunological nature of schizophrenia (Whittingham et al., 1968; Logan and Deodhar, 1974). Recently, it has been reported that 27% of normal subject sera contained taraxein which induced EEG alterations when injected into monkey, but the incidence of taraxein-positive sera was significantly higher in patients with schizophrenia (Bergen et al., 1980). Under these circumstances, additional supporting evidence would be required to prove the autoimmune theory of schizophrenia.

D. PSYCHIATRIC DISEASES AND CELL-MEDIATED IMMUNITY Although antibody-mediated and cell-mediated processes are integral constituents of the immune machinery of' the body, and both processes are involved in the development of autoimmune diseases, it is quite surprising how little attention has been paid to the in vivo evaluation of structural and functional correlates of cell-mediated immunity in psychiatric diseases. Almost all accounts of cell-mediated immunity in mental patients rely on in uitro tests (Knowles et al., 1970; Bonaster, 1973; Kuritzky et al., 1976). Recent investigations have demonstrated, however, that a simple immunological device such as a skin test with brain antigen can provide important information about sensitization to nervous tissue antigens. In a series of reports, JankoviC and associates (1977d, 1979a, 198Oc, 1981, 1982b,c) described delayed skin hypersensitivity reactions to human brain S-100 protein in patients suffering from cerebral atrophy of unknown origin, senile dementia, depression, nuclear and peripheral schizophrenia, and mental retardation (IQ 0-49). Data collected in double-blind studies during the last 8 years are summarized in Table V. T h e incidence of positive delayed hypersensitivity reactions to S- 100 protein was particularly high in subjects with cerebral atrophy, dementia, and depression, and to a lesser extent in patients with schizophrenia and mental retardation, thus suggesting that cell-mediated immune mechanisms are involved in the pathogenesis of certain mental diseases. It

298

BKANISI.AV D. J A N K O V I C

TABLE V DELAYED SKINHYPERSENSITIVITY REACTIONS TO HUMAN BRAINS-100 PROTEIN IN PSYCHIA~KIC PAITEN~S

Group Cerebral atrophy Dementia Depression Scliizophrenia" Mental retardation Control a

Number of patients tested 124 92 47 260 612 36

Positive reactions (%) S- 100 protein

Liver protein

93.2 92.6 93.8 75.8

20.2 18.8 20.5 7.8 3.0 9.5

70.6 6.1

Nuclear and peripheral forms. Human liver antigen served as control antigen.

should be mentioned that the number of positive reactions to S-100 protein in patients with affective psychoses, neuroses, and reactive psychoses approached that observed in control individuals. (For brain S100 protein see Section I1,B and IV,D.) These investigations also revealed a high frequency of positive local Arthus' reactivity to human brain S- 100 protein in patients with cerebral atrophy, dementia, depression, schizophrenia, and mental retardation (Table VI). Since Arthus' reaction is an accute inflammatory lesion induced by the antigen-antibody complexes, it would follow that antibrain antibodies and circulating brain antigens or their fragments (Paterson, 1979) are also active in diseases mentioned here. Consequently, the positiveness of delayed and Arthus' reactions in certain psychiatric diseases would imply that the operational composition of these autoimmune diseases includes both humoral and cellular immunity. From the practical point of view, skin testing of psychiatric patients with a brain TABLE V1 AIIHUS' SKIN REACTIVITY TO HUMAN BRAIN s-100 P R O T E I N

Group Cerebral atrophy Dementia Depression Schizophrenia" Mental retardation Control

Number of patients tested

IN

PSYCHIAI'RIC PATIENTS

Positive reactions (%) S-100 protein

Liver protein

124 92 47 260 612

83.4 95.6 81.0

11.5 40.4 21.8

36

3.2

71.8

8.3

72.4

4.3 15.6

" Nuclear and peripheral forms. Human liver antigen served as control antigen.

NEUROMODULATING A C T I V I T Y OF ANTI-BRAIN ANTIBODIES

299

antigen which is not encephalitogenic and harmful for patients may help diagnosis and evaluation of disease. It should be clearly stated, however, that positive hypersensitivity to neurotissue antigens suggests that immune mechanisms participate in the pathogenesis of immunopsychiatric diseases but by no means provides evidence that the immune process is the cause of diseases. E. ALCOHOLISM AND HYPERSENSITIVITY TO BRAIN ANTIGENS

It is generally recognized that chronic alcoholism causes severe brain dysfunctions (Jones and Parsons, 1972; Gudeman et al., 1977). Alcohol significantly perturbs the developing nervous system (Borges and Lewis, 1981), diminishes cellular density (Golden et al., 1981), and reduces cortex and white matter (Carlen et al., 1978; Haubek and Lee, 1979). Chronic consumption of ethanol may cause a higher mortality rate of cells and increased output of antigens from dead neurons and glial cells, which enter the circulation, stimulate the immune machinery of the body, and initiate autoimmune processes related to the brain. Investigations in this direction Uankovit et al., 1982d) revealed that almost all heavy chronic alcoholics who were hospitalized because of psychotic symptoms have developed positive delayed and Arthus’ skin hypersensitivity reactions to brain S- 100 protein (Table VII). Approximately onehalf of tested patients also exhibited positive delayed reactions to human liver protein, thus suggesting a correlation between delayed hypersensitivity to liver antigen and alcohol-induced liver disease. On this basis, it can be assumed that immune mechanisms are involved in neurological and psychiatric components of alcoholism. Accordingly, evaluation of TABLE V I I AR.I.HUS’ A N U DELAYED SKIN IIYPERSENSI’I’IVITYREACTIONS TO BRAIN S-100PROI’EIN HOSPITALIZED CHRONIC ALCOHOLICS WITH MENTAL DETERIORATION I iiiinune skin reaction

IN

Positive reactions (%) Subjects‘J

S- I00 protein

Liver protein

Arthus’

Alcoholics Controls

98.3 4.2

28.2 6.8

Delayed

Alcoholics Controls

97.0 6.4

44.1 11.2

Seventy-one alcoholics and 38 control subjects were tested. Human liver antigen served as control antigen. ‘I

300

BRANISLAV D. J A N K O V I ~

hypersensitivity to neurotissue antigens may help early detection of brain impairment in chronic alcoholics.

VI. Concluding Remarks

Over the years, small and isolated groups of investigators have repeatedly drawn attention to the fact that the immune system could play important and hitherto unrecognized roles in nervous system function, and vice versa. T h e immune microenvironment is a “social unit” which unifies cellular and extracellular elements from three different sources: the immune (lymphoreticuloepithelial) system, the nervous system, and the endocrine system. This milieu represents not only a channel of communication and transmission of information but also a locus of integrative and specific activity from which the immune responses arise. This triadic property of the immune milieu forms a basis for an interdisciplinary repertoire of experimental undertakings that may greatly expand our understanding of neuroimmune correlations. The anti-brain antibody serves this goal. Several lines of investigations are brought together to portray the biological activity of anti-brain antibodies. T h e activity of anti-brain antibody implies the high specificity of its action. This belief is based on the well-documented capacity of the antibody to distinguish clearly between distinct antigenic determinants. There are no compelling reasons why an anti-brain antibody should not be capable of recognizing very precisely its antigenic partner situated on or in the neuron, and then inducing structural and biochemical changes in the radius of neuron activity (neuronal membrane, synaptic clefts, surrounding glial cells, neighboring brain-blood compartments, etc.). Evidence from several laboratories showed that the primary functions of the neuron and the brain were affected by anti-brain antibodies. Since the anti-brain antibody originates from cellular and nonsynaptic sites and influences the excitability of the neuron, it can be regarded as fitting the term neuromodulator (Florey, 1967; Bloom, 1980). ‘Ihe satisfaction of that definition rests upon the assurance that the direct effects of anti-brain antibody on neuronal membrane properties are real, as shown in different protocols concerning modulation of the responsiveness of the CNS by means of anti-brain antibodies. This article summarizes some more recent studies on autoimmunity in psychiatric diseases, and emphasizes the importance of cell-mediated

NEUROMODULATING ACTIVITY OF ANTI-BRAIN ANTIBODIES

30 1

immunity in the pathogenesis of certain mental disorders such as cerebral atrophy of unknown origin, dementia, depression, schizophrenia, mental retardation, and brain deterioration syndrome in chronic alcoholism. The high incidence of positive skin hypersensitivity reactions, both Arthus’ and delayed, to human brain antigens in immunopsychiatric patients most probably reflects an autoreactive disease state which is caused by abnormal output and/or structural changes of neurotissue antigens. Accordingly, the operational composition of immunoneurological and immunopsychiatric diseases encompasses both antibody-mediated and cell-mediated immune mechanisms. On this ground one is tempted to assume that immunosuppressive manipulations can be used for the treatment of autoimmune diseases of the CNS. A rather complicated issue arises if one considers the mechanisms underlying the effects produced by anti-brain antibodies. There are several possibilities which are not mutually exclusive: (1) changes at the receptor level (adenylate cyclase receptors, cholinergic receptors, etc.); (2)changes at the membrane level (the physical state of membrane, ionic transport and active transmembrane pumping, enzymatic methylation processes that modulate membrane fluidity, etc.); (3) intracellular changes (ATP and GTP nucleotides, protein phosphorylation, etc.); (4) changes in axonal transport (membrane-confined compartments in anterograde transport and tubulovesicular structures in retrograde transport); ( 5 ) changes in conductibility of the membrane induced by neurotransmitters, neurohormones, and neuromodulators; and (6) other explanations which may be constructed on purely speculative grounds. Each of the above possibilities has appealing features, but none accounts entirely for all aspects of antibody activity on the neuron and the brain. Further research at the cellular level, the circuitry level, and the level of behavior will move experience in the regulation and modulation of the neuron and brain sensitivity by anti-brain antibodies from the plane of speculation to experimental elucidation. Acknowledgments

T h e essential participation of my collaborators, Drs. Lj. MihailoviC, K. Isakovit, V. SaviC, S. Jakulii., J . Horvat, K. MitroviC, Soltes, H.M. Markovit, R. Veskov, Lj. RakiC, and M. Radulovatki, in experiments reported here is gratefully acknowledged. It is a pleasure to thank Drs. V. Savit, S . Soltes, K. Mitrovit, and M. Radulovatki for their generous permission to cite unpublished findings from our laboratory. I am indebted to Drs. B. M. Markovit and S. Soltes for technical assistance in the preparation of this manuscript. T h e experiments from our laboratory were supported by the Republic of Serbia Research Fund, Belgrade, and partly by research grants from the National Institutes of Health, Bethesda, Maryland.

s.

302

BRANISLAV D. . J A N K O V I ~

References

Abe, K., Takaslri, K., and Ando, T. (1973).J. BzochPm. 75, 11-22. Adams, C. E. M. (1977). Br. Mpd. Bull. 33, 15-20. Ader, R., ed. (1981). “Psychoneuroirnmunology.” Academic Press, New York. Adinolfi, M., Rick, I. T., Leibowitz, S., and Gregson, A. N. (1982). In “Padkatrische Fortbildungskurse fur die Praxis” (E. Rosie, ed.), pp. 178-186. Karger, Basel. Allerand, C. D., and Yahr, M. D. (1964). Science 144, 1141-1 142. Alvord, E. C., Jr., Shaw, C.-M., Lisdk, R. P., Falk, G. A,, and Kies, M. W. (1970). Int. Arch. Allergy Af$l. Inimunol. 38, 403-4 12. Amkraut, A., Solomon, G. F., Allansmith, M., McClellan, B., and Rappaport, M. (1973). Arch. Gen. Psychiatry 28, 673-677. Appel, S. H., and Bornstein, M. B. (1964).J. Exp. Med. 119, 303-312. Armand-Delille, P. F. (1906). Ann. Imt. Pasteur 20, 838-858. Atkins, C. J., Kondon, J. J., Quismorio, F. P., and Friou, G. J. (1972). Ann. Intern. Med. 76, 65-72. Barclay, A. N., and Hydkn, H. (1979).J. Neurocheni. 32, 1538-1586. Barkley, I). S., Hardiwidjaja, S., and Menkes, J. €1. (1977a). Science 195, 314. Barkley, D. S., Hardiwidjaja, S. I., Menkes,J. H., Ellison, G. W., and Myers, L. W. (1977b). Cell. Immunol. 32, 385-390. Baron, M., Stern, M., Anavi, R., and Witz, I. P. (1977). Biol. Psychiatry. 12, 199-212. Barondes, S. H. (1982). TINS 5, 280-282. Bartlett, P. (1982). Proc. Natl. Acad. Scz. U.S.A. 79, 2722-2725. Bartrop, R. W., Luckhurst, E., Lazarus, L., Kiloh, L. G., and Penny, R. (1977). Lancet 1, 834-836. Benda, P., Lightbody, J., Sato, G., Levine, L., and Sweet, W. (1968). Sczence 161, 370-371. Bennett, G. S . (1974). Brain Res. 68, 365-369. Bennett, G. S., and Edelrnan, G. M. (1968). J. Biol. Chem. 243, 6234-6241. Bergen, J. R., Grispoon, L., Pyle, H. M., Martinez, J. L., Jr., and Pennell, R. B. (1980). Biol. Psychiatry 15, 369-379. Bergstrand, H. (1977). In “Immunochemistry of Proteins” (M. Z. Atassi, ed.), pp. 315-370. Plenum, New York. Bernard, C. (1878). “Lecons sur les PhCnomknes de la Vie.” Baillieve, Paris. Berry, M., and Riches, A. C . (1974). Br. Med. Bull. 30, 135-140. Besredka, A. (1919). “Anaphylaxis and Anti-Anaphylaxis, and Their Experimental Foundations.” Mosby, St. Louis, Missouri. Bighami, A., Eng, L. F., Dahl, D., and Uyeda, C . T. (1972). Brain Res. 43, 429-435. Bleuler, E. (1950). “Dementia Praecox and the Group of Schizophrenias.” International Univ. Press, New York. Bloom, F. E. (1980). Neurosci. Res. Prog. Bull. 18, 429-435. Bluenstein, H. G., and Zvaifler, N. Y. (1976).J. Clin. Invest. 57, 509-516. Bock, E. (1978).J. Neurochem. 30, 7-14. Bock, E., ed. ( 1 982). “Nervous-System-Specific Proteins.” Blackwell, Oxford. Bogoch, S. (1960). Nature (Lundon) 185, 392-393. Bogoch, S., Rajam, P. C., and Belval, P. C. (1964). Nature (London) 204, 73-75. Bonaster, P. D. (1973). Zh. Nevropatol. Psikhiatr. 73, 878-885. Borges, S., and Lewis, P. D. (1981). Trends Neurosci. 4, 13-15. Bornstein, M. B. (1973). Prog. Neuropathol. 2, 69-90. Bornstein, M. B., and Appel, S. H. (196l).J. Neuropathol. ExP. Neurol. 20, 141-157. Bornstein, M. B., and Appel, S. H. (1965). Ann. N.Y. Acad. Sci. 122, 280-286.

NEUROMODULATING ACTIVITY OF ANTI-BRAIN ANTIBODIES

303

Bornstein, M. B., and Crain, S. M. (1965). Science 148, 1242-1244. Bornstein, M. B., and Raine, C. S. (1970). Lab. Invest. 23, 436-442. Bowen, F. P. (1968). Arch. Neurol. 19, 398-402. Bowen, F. P., Korsakova, J., Casella, D., Nicklas, W. J., and Berl, S. (1976). BTain Res. 102, 363-367. Breathnach, A. S. (1980).J. Invest. Dermatol. 75, 6-11. Bronner-Fraser, M. E., and Cohen, A. M. (1980). Dev. Bid. 15, 1-25. Brooks, W. H., Netsky, M. G., Normansell, D. E., and Horwitz, D. A. (1972).J. Exp. Med. 136, 1631-1647. Bunney, W. E., Jr., and Murphy, D. L. (1976). Neurosci. Res. Prog. Bull. 14, 115-207. Calissano, P., and Bangham, A. D. (1971). Biochim. Biophys. Acta 43, 504-509. Campbell, B. (1949). J . Neuropathol. E x f . Neurol. 8, 347-354. Campbell, B., and Good, R. A. (1950). Arch. Neurol. Psychiatr. 63, 298-310. Carlen, P., Wortzman, G., Holgate, R., Wilkinson, D., and Rankin, J. (1978). Science 200, 1076- 1078. Ceccarelli, B., Aporti, A., and Finesso, M. (1975). Adu. Exp. Med. Biol. 71, 275-293. Cerf, J. A., and Carles, G. (1966). Science 152, 1066-1068. Chaffee, J., Nassef, M., and Robin, S. (1978). I n “Senile Dementia: A Biomedical Approach” (K. Nandy, ed.), pp. 61-72. Elsevier, Amsterdam. Cicero, T. J., Cowan, W. M., Moore, B. W., and Suntzeff, V. (1969). Brain Res. 18, 25-34. Cicero, T . J., Cowan, W. M., and Moore, B. W. (1970). Brain Res. 24, 7-14. Cinader, B., ed. (1977). “Immunology of Receptors.” Dekker, New York. Cocchia, D., and Michetti, F. (1981). Cell TissueRes. 215, 103-112. Cocchia, D., Michetti, F., and Donato, R. (1981). Nature (London) 294, 85-87. Cochran, A. J. (1970). J . Invest. Dermatol. 55, 65-70. Cohen, J., and Selvendran, S. Y. (1981). Nature (London) 291, 421-423. Cohen, S. R., Herndon, R. M., and McKhann, G. M. (1976). New Engl. J . Med. 295, 14551457. Cohen, S. R., Brooks, B. R., Jubelt, B., Herndon, R. M., and McKhann, G. M. (1980). I n “Neurobiology of Cerebrospinal Fluid” H. Wood, ed.), pp. 487-494. Plenum, New York. Conti-Tronconi, B., Tzartos, S., and Lindstrom, J. (1981). Biochemirtly 20, 2181-2191. Crain, S. M., and Bornstein, M. B. (1975). Exp. Neurol. 49, 330-335. Crow, T. J. (1983a). Lancet 1, 173-175. Crow, T. J. (1983b). Lancet 1, 819-820. Experientia 38, 1275-1282. Crow, T. J. (1983~). Dahl, D., and Bighami, A. (1973). Brain Res. 61, 279-293. Dannies, P. S., and Levine, L. (1971).J. Biol. Chem. 246, 6276-6283. Davison, K., and Bagley, C. R. (1969). I n “Current Problems in Neuropsychiatry” (R. N. Herrington, ed.), pp. 113-184. Headley, Ashford (Kent). Day, E. D. (1981). Contemp. Top. Mol. Immunol. 8, 1-39. Day, E. D., Varitek, V. A., Fujinami, R. S., and Paterson, P.Y. (1978a). Immunochemist?y 15, 1-9. Day, E. D., Varitek, V. A., and Paterson, P. Y. (197813). Immunochemistry 15, 437-442. De Feudis, F. V., Yusufi, A. N. K., Ossola, L., Maitre, M., Wolff, P., Rebel, G., and Mandel, P. (1980). Gen. Pharmacol. 11, 251-254. Delezenne, C. (1900). Ann. Inst. Pasteur 14, 686-704. Dencker, S . J., and Swahn, B. (1962). Nature (London) 194, 288-289. De Robertis, E., Lapetina, E. G., Pecci Saavedra,J., and Soto, E. F. (1966). Life Scz. 5, 19791989.

u.

304

BRANISLAV D. . J A N K O V I ~

De Robertis, E., Lapetina, E. G., and Wald, F. (1968). Exp. Neural. 21, 322-335. De Robertis, E., Wald, F., and Lapetina, E. G. (1971). I n “Immunological Disorders of the Nervous System” (L. P. Rowland, ed.), pp. 8-24. Williams & Wilkins, Baltimore, Maryland. Diederichsen, H., and Pyndt, I. C. (1970). Brain 93, 407-412. Dimuskes, R. K. (1980). Neurosci. Res. Prog. Bull. 18, 390-394. Donato, R. (1978). J . Neurochem. 30, 1 105- 1 11 1. Donato, R. (1981).J. Neurochem. 37, 610-618. Donato, R., and Michetti, F. (1974). Expen‘entia 30, 511-512. Donato, K., and Michetti, F. (1981).J. Neurochem. 36, 1698-1705. Donato, R., Michetti, F., and Miani, N. (1975). Brain Res. 98, 561-573. Dorfman, S. H., Holtzer, H., and Silberberg, D. H. (1976). Brain Res. 104, 283-294. Dorfman, S. H., Fry, J. M., Silberberg, D. H., Drose, C., and Manning, M. C . (1978). Brain Res. 147, 410-415. Dorfman, S. H., Fry, J. M., and Silberberg, D. H. (1979). Brain Res. 177, 105-114. Drucker-Colin, R. R., Zamora, J., Bernal-Pedraza, J., and Sosa, B. (1979). Exp. Neural. 63, 458-467. Drucker-Colin, R. R., De Gomez-Puyou, M. T., Del CarmenGutierrez, M., and DreyfusCortes, G. (1980). Exp. Neural. 69, 563-575. Dubois-Dalcq, M., Niedieck, B., and Buyse, M. (1970). Pathol. Eur. 5 , 331-347. Eddington, T. S., and Dalessio, D. .J. (1970).J. Immunol. 105, 248-255. Editorial. (1968). Flure (London) 219, 221-230. Eng, L. F., Vanderhaeghen, J. J., Bighami, A,, and Gerstl, B. (1971). Brain Res. 28, 351354. Ettlinger, G., and Lowrie, M. B. (1976). Lancet 1, 1386. Feldberg, W., and Sherwood, S. L. (1953).J. Physiol. (London) 120, 3P-5P. Ferguson, R. M., Schmidtke, J. R., and Simmons, R. L. (1978). Birth Defects Orig. Artic. Ser. 14,379-405. Fernandez, H. L., Huneeus, F. C., and Davison, P. F. (1970)J. Neurobiol. 1, 395-409. Fernandez, H. L., Burton, P. R., and Samson, F. E., Jr. (1971).J. Cell Biol. 51, 176-192. Fessel, W. J. (1962). Arch. Gen. Psychiatry 6, 132-148. Fessel, W. J. (1963). Arch. Gen. Psychid9 8, 614-622. Florey, E. (1967). Fed. Proc., Fed. Am. SOC.Ex#. Biol. 26, 1164-1176. Frieder, B., and Rapport, M. M. (1980). SOC.Neurosci. Abstr. 6, 6. Frieder, B., and Rapport, M. M. (1981). Trans. Am. Sac. Neurochem. 12, 274. Fry, J. M., Weissbarth, S., Lehrer, G. M., and Bornstein, M. B. (1974). Science 183, 540542. Fuchs, S. (1979). Cur?. Top. Microbial. Immunol. 85, 1-29. Fujinami, R. S., Paterson, P. Y., Day, E. D., and Varitek, V. A. (1978). J. Exp. Med. 148, 17 16-172 1. Garey, R. E., Heath, K. G., and Harper, J. W. (1974). Biol. Psychiatry 8, 75-88. Gaynor, R., Irie, R., Morton, D., and Herschman, H. K. (1980). Nature (London)286,400401. Gaynor, R., Irie, R., Morton, D., Herschman, H. R., Jones, P., and Cochran, A. (1981). Lancet 1, 869-87 1 . Gcrhard, W., and Koprowski, H. (1977). Nnture (London) 266, 360-561. Gershwin, M. E., Hyman, L. R. H., and Steinberg, A. D. (1975).J. Pediatr. 87, 588-590. Golden, C . J., Grabar, B., Blose, I., Berg, R., Coffman,J., and Bloch, S. (1981).Science 211, 508-510. Golub, E. S., and Day, E. D. (1975). Cell. Immunol. 16, 427-431.

NEUROMODULATING ACTIVITY OF ANTI-BRAIN ANTIBODIES

305

Gomez, G. M., Richman, D. P., Berman, P. W., Burres, S. A., Arnason, B. G. W., and Fitch, F. W. (1979). Biochem. Biophys. Res. Commun. 88, 575-582. Gorio, A., Carmignoto, G., Facii, L., and Finesso, M. (1980). Brain Res. 197, 236-241. Gosheva, A. E.. Domasheva, I. V., Kobrinsky, G. D., Kolyaskina, G. I., Kushnir, S. G., and Podozreva, N. P. (1969). Vestn. Akad. Med. Nauk SSSR 4, 70-75. Grossman, M. I. (1979). Fed. Proc., Fed. Am. Soc. Exp. Biol. 38, 2341-2343. Grundke-Iqbal, I . , and Bornstein, M. B. (1979). Brain Res. 160,489-503. Gudeman, H. E., Craine, J. F., Golden, C. J., and McLaughlin, D. (1977). Int. J . Neurosci. 8, 33-40. Guillemin, R. (1978). Neurosci. Re.!. Prog. Bull. (Suppl.) 16, 1-25. Gutstein, H. S., and Cohen, S. R. (1978). Science 199, 301-303. Haan, E. A,, Boss, B. D., and Cowan, W. M. (1982). Proc. Natl. Acad. Sci. U.S.A. 79, 75857589. Haglid, K. G., and Stavrou, D. (1973).J. Neurochem. 20, 1523-1532. Haglid, K. G., Carlsson, C.-A,, and Stavrou, D. (197%).Acta Neuropathol. (Berlin) 24, 187196. Haglid, K. G.,Stavrou, D., Riinnback, I>., Carlsson, C.-A., and Weidenbach, W. (1973b).J. Neurol. Sci. 20, 103-1 1 1 . Haglid, K. G., Hamberger, A., Hansson, H.-A,, Hyden, H., Persson, L., and Riinnback, L. (1974). Nature (London} 251, 532-534. Hall, B. K. (1978). Adv. Anat. Embryol. Cell Biol. 53, 1-49. Hall, N. R., and Goldstein, A. L. (1981). In “Psychoneuroimmunology” (R. Ader, ed.), pp. 521-544. Academic Press, New York. Hansson, H.-A., Persson, L., Ronnback, L., and Haglid, K. G. (1976). Cytofiios 15, 4548. Harbeck, R. J., Hoffman, A. A., Hoffman, S. A., Shucard, D. W., and Carr, R. I. (1976). Fed. Proc., Fed. Am. Sac. Exp. Biol. 35, 437. Hatcher, V. B., and MacPherson, C. F. C. (1970).J . Immunol. 104,633-640. Haubek, A., and Lee, K. (1979). Neuroradiology 18, 77-79. Hauw, J.-J., Fenelon, S., Boutry, J.-M., and Escourolle, R. (1981). C.R. Acad. Sci: (Paris) 292, 569-57 1 . Heath, R. C . (1966). 1nt.J. Neuropsyc/ziutty 2, 597-610. Heath, R. G. (1969). I n “Schizophrenia: Current Concepts and Research” (D. V. Siva Sankar, ed.), pp. 580-585. PJD Publ., Hicksville, New York. Heath, R. G., and Krupp, I. M. (1967). Arch. Gen. Psychiatry 16, 1-9. Heath, R. G., Krupp, I. M., Byers, L. W., and Liljekvist, J. I. (1967a).Arch. Gen. Psychiatv 16, 10-23. Heath, R. G., Krupp, I. M., Byers, L. W., and Liljekvist, J. I. (1967b). Arch. Gen. Psychiatry 16, 24-33. Heremans, J. F. (1973). Curr. Top. Microbiol. Immunol. 45, 132-203. Herschman, H. R., Cotman, C., and Latthews, D. A. (1972).J . Immunol. 108, 1362-1369. Hill, K. R. (1949). Bull. Johns Hopkins Hosp. 84, 302-333. Hoffman, P. M., Robbins, D. S., Nolte, M. T., Gibbs, C. S., Jr., and Gajdusek, D. C. (1978). New Engl. J . Med. 299, 680-685. Hofstein, R., Segal, M., and Samuel, D. (1980). Exp. Neurol. 70, 307-320. Holzer, H., and Detwiler, S. R. (1953).J. Exp. Zool. 123, 335-369. Hruby, S., Alvord, E. C., Jr., and Seil, F. J . (1977). Science 195, 173-175. Hughes, R. A. C. (1974). Immunology 26, 703-711. Hughes, D., and Field, E. J. (1967). Clin. Exp. Immunol. 2, 295-309. Huneeus-Cox, F. (1964a). Science 143, 1036-1037.

306

BKANISLAV D.

JANKOVIC

Huneeus-Cox, F. (1964b). Cungr. Asoc. Latinoam. Cienc. Fisiol. (ALACF). bth, Vinn del Mar, Chile p. 46. Huneeus-Cox, F. ( 1 967). In “Neurosciences Research Symposium Summaries” (F. 0. Schniitt, T. Melnechuk, G . C. Quarton, and G. Adelman, eds.), pp. 360-362. MIT Press, Cambridge, Massachusetts. Husby, G., van d e Rijn, I., Zabriskie, J. B., Abdin, Z. H., and Williams, R. C., Jr. (1976).J. Exp. Med. 144, 1094-1110. Husby, G., Wedege, E., and Williams, R. C., Jr. (1978). Clin. Immunol. Immunopathol. 11, 131-141. Husby, G . , Williams, R. C., Jr.. Bersin, R. M., and Lewis, M. K. (1979). In “Clinical Neuroimmunology” (F. C. Rose, ed.), pp. 90-105. Blackwell, Oxford. Hyden, H., and Lange, P. (1970). Exp. Cell Res. 62, 125-132. Hyden, H., and Lange, P. W. (1971). Proc. Natl. Acad. Sci. U.S.A. 65, 898-904. HydCn, H., and McEwen, B. (1966). Proc. Natl. Acad. Scz. U.S.A. 55, 354-358. Hyden, H., and Ronnbgck, L. (1975). Brain Res. 100, 615-628. Inoue, T., Okamura, M., Amatsu, K., Ohe, A., Kato, N., and Kanayama, Y. (1982). Lancet 1, 852. Isobe, T., Nakajima, T., and Okuyama, T. (1977). Biochim. Biophys. Acta 494, 222-232. Iversen, L. L., NicoI1, R. A,, and Vale, W. W. (1978).Neurosci. Res. Prog. Bull. 16,211-370. JankoviC, B. D. (1972). In “Macromolecules and Behavior” Gaito, ed.), pp. 99-130. Appleton, New York. JankoviC, B. D. (1973).I n “Microenvironmental Aspects of Immunity” (B. D. Jankovit and K. Isakovid, eds.), pp. 1-4. Plenum, New York. Jankovit, B. D. (1979). Immunol. Lett. 1, 145-146. JankoviC, B. D., and Jankovit, D. Lj. (1983). Fed. Proc., Fed. Am. Sac. Exp. Bzol. 42, 1242. .JankoviC, B. D., and RakiC, Lj. (1969). Vestn. Akad. Med. Nnuk SSSR 23,44-51. (In Russian) JankoviC, B. D., IsakoviC, K., and MihailoviC, Lj. (1960). Int. Arch. Allergy Appl. Immunol. 17, 21 1-220. Jankovit, B. D., DraSkoci, M., and IsakoviC, K. (1961). Nature (London) 191, 288-289. Jankovik, B. D., DraSkoci, M., and JanjiC, M. (1965). Nature (London) 207, 428-429. JankoviC, B. D., RakiC, L. M., JanjiC, M., IvanuS, J., and MitroviC, K. (1966a). Experientia 22, 459-460. JankoviC, B. D., RakiC, Lj., Janjit, M., MitroviC, K., and IvanuS, J. (1966b). Pathol. Eur. 2, 87-107. JankoviC, B. D., Rakid, Lj., Veskov, R., and Horvat, J. (1968). Nature (London) 218, 270271. JankoviC, B. D., RakiC, Lj., Veskov, R., and Horvat, J. (1969a). Experientiu 25, 864-865. JankoviC, B. D., RakiC, Lj., and Sestovit, M. (1969b). Experientia 25, 1049-1050. JankoviC, B. D., Horvat, J., MitroviC, K., and Mostarica, M. (1975a). Keio J.Med. 24, 355365. JankoviC, B. D., MitroviC, K., Horvat, J., and SaviC, V. (1975b). Experientia 31, 239-240. JankoviC, B. D., Savit, V., and Horvat, J. (1975~ ). Expen’entza 31, 1093-1095. JankoviC, B. D., Horvat, J., Mitrovit, K., and Mostarica, M. (1977a).Immunochemirtry 14, 75-78. JankoviC, B. D., Horvat, J., MitroviC, K., and Mostarica, M. (1977b). Experientza 33, 15261527. JankoviC, B. D., Horvdt, J . , MitroviC, K., and Mostarica, M. ( 1 9 7 7 ~)Scand.J. . Immunol. 6, 843-847. Jankovit, B. D., Jakulit, S., and Horvat, J. (1977d). Lancet 2, 219-220.

u.

NEUKOMODULATING ACTIVITY OF ANTI-BRAIN ANTIBODIES

307

JankoviC, B. D., Popeskovit, Lj., and IsakoviC, K. (1978). Fed. Proc., Fed. Am. SOC.Exp. B i d . 37, 1651. JankoviC, B. D., JakuliC, S., and Horvat, J. (1979a). Period. Biol. 81, 219-220. JankoviC, B. D., Popeskovit, Lj., and Isakovii., K. (1979b). I n “Function and Structure of the Immune System” (W. Muller-Ruchholtz and H. K. Muller-Hermelink, eds.), pp. 339-344. Plenum, New York. Jankovik, B. D., Horvat, J., and MitroviC, K. (1979~).Experientia 35, 1393-1394. JankoviC, B. D., Isakovid, K., and Popeskovit, Lj. (1980a). Immunol. Lett. 1, 203-208. JankoviC, B. D., Soltes, Horvat, J., and MitroviC, K. (1980b). Immunol. Lett. 1, 275-279. JankoviC, B. D., JakuliC, S., and Horvat, J. (1980~).Clin. Ex#. Immunol. 40, 515-522. JankoviC, B. D., JakuliC, S., and Horvat, J. (1981). Zmmunobiol. Abstr. 160, 47. JankoviC, B. D., Lenert, P., and MitroviC, K. (1982a). Zmmunobiology 161, 488-493. JankoviC, B. D., JakuliC, S., and Horvat, J. (1982b). Bid. Psychiatry 17, 687-697. JankoviC, B. D., Jakulit, S., and Horvat, J. (1982~).Neurosci. Abstr. 7, S103. JankoviC, B. D., JakuliC, S., and Horvat, J. (1982d). Clin. Exp. Zmmunol. 49, 598-602. Jarosch, E., and Precht, W. (1972). Brain Res. 42, 225-229. Jensen, K., Clausen, J., and Osterman, E. (1964). Acta Psychiah. Scund. 40, 280-286. Jones, B., and Parsons, 0. (1972). Arch. Cen. Psychiatry 26, 380-384. Jouvet, M. (1965). Prog. Bruin Res. 18, 20-62. Jouvet, M. (1976). I n “Cholinergic Mechanisms” (P. G. Waser, ed.), pp. 455-476. Raven, New York. Jouvet, M., Vimont, P., Delorme, J. F., andJouvet, M. (1964). C.R. Sac. Biol. (Paris) 158, 756-759. Kabat, E. A,, Wolf, A., and Bezer, A. E. (1946). Science 104, 362-363. Kabat, E. A., Wolf, A., and Bezer, A. E. (1948).J. Ex#. Med. 88,417-426. Kandel, E. R., Krasne, F. B., Strumwasser, F., and Truman, J. W. (1979). Neurosci. Res. Prog. Bull. 17, 521-710. Karcher, D., Lowenthal, A., and Strosberg, A. D., eds. (1979). “Humoral Immunity in Neurological Diseases.” Plenum, New York. Karnovsky, M. L., and Reich, P. (1977). Adv. Neurochem. 2, 213-275. Karpiak, S. E., and Rapport, M. M. (1975). Brain Res. 92, 405-413. Karpiak, S. E., and Rapport, M. M. (1979). Behav. Neural Biol. 27, 146-156. Karpiak, S. E., Jr., Bowen, F. P., and Rapport, M. M. (1973).Brain Res. 59, 303-310. Karpiak, S. E., Jr., Rapport, M. M., and Bowen, F. P. (1974). Neurophysiolopa 12,313-322. Karpiak, S. E., Graf, L., and Rapport, M. M. (1976a). Science 194, 735-737. Karpiak, S. E., Serokosz, M., and Rapport, M. M. (1976b). BrainRes. 102, 313-321. Karpiak, S. E., Graf, L., and Rapport, M. M. (1978). Brain Res. 151, 637-641. Karpiak, S. E., Mohadik, S. P., Graf, L., and Rapport, M. M. (1981).Epilepsia 22, 189-196. Katz, B. (1966). “Nerve, Muscle and Synapse.” McGraw-Hill, New York. Kennett, R. H., Jonak, Z. L., and Bechtol, K. B. (1980). In “Monoclonal Antibodies. Hybridomas: A New Dimension in Biological Analyses” (R. H. Kennett, T. J. McKearn, and K. B. Bechtol, eds.), pp. 155-168. Plenum, New York. Kessler, D., Levine, L. and Fassman, G. D. (1968). Biochemistry 7, 758-764. Kety, S. S. (1965). Znt.J. Psychiatry 1, 409-447. Kies, M. W., Alvord, E. C., Jr., and Roboz, E. (1958). Nature (London) 182, 104-105. Kies, M. W., and Alvord, E. C., Jr., eds. (1959). “‘Allergic’ Encephalomyelitis.” Thomas, Springfield, Illinois. Kimura, R. (1928). Z. Zmmunituetsforsch. 55, 501-502. Kingston, D., and Glynn, L. E. (1976). Br. J . Exp. Pathol. 57, 114-128.

s.,

308

BKANISLAV D. JANKOVIC

Knowles, M., Saunders, M., and McClelland, H. A. (1970).ActaPsychiatr. Scand. 46,64-70. Kobiler, D., Fuchs, S., and Samuel, D. (1976). Brain Res. 115, 129-138. Kohler, G., and Milstein, C. (1975). Nature (London) 256, 195-197. Kolyaskina, G. I., and Kushnir, S. G. (1969). Zh. Neuropatol. Psikiatr. 69, 1679-1682. (In Russian) Kopeloff, L. M. (1942). Am. J . Physiol. 98, 270-271. Kopeloff, L. M., Barrera, S. E., and Kopeloff, N. (1942). Am. J. Psychiatly 98, 881-902. Koprowski, H. (1962). A m . J . DLc. Child. 103, 273-278. Kornguth, S. E., Anderson, J. W., and Scott, G. (1969).J. Neurochem. 16, 1017-1024. Kovaleva, E. S., Bonartsev, P. D., and Prilipko, L. L. (1977).Bul. E h p . Biol. Med. 84, 11361139. Kuritzky, A., Livni, E., Munitz, H., Englander, T., Tyano, S., Wijsenbeek, H., Joshua, H., and Kott, E. (1976).J . Neurol. Scz. 30, 369-373. Lajtha, A. (1969). “Handbook of Neurochemistry. Vol. 2. Structural Neurochemistry.” Plenum, New York. Lamoureaux, G., McPherson, T . A., and Carnegie, P. R. (1967). Clin. Exp. Immunol. 2, 253-255. Lampert, P. W., and Oldstone, M. B. A. (1973). Science 180, 408-410. Land, R. W., and Carlo, D. J. (1975). I n “Neurobiology of Aging” 0. M. Ordy and K. R. Brizzee, eds.), pp. 233-251. Plenum, New York. Landsteiner, K. (1946). “The Specificity of Serological Reactions,” p. 80. Harvard Univ. Press, Cambridge, Massachusetts. Lawley, T. J., and Frank, M. M. (1980). In “Clinical Immunology” (C. W. Parker, ed.), Vol. 1, pp. 143-172. Saunders, Philadelphia, Pennsylvania. Lebar, K., Boutry, J.-M., Vincent, C., Robineaux, K., and Voisin, G. A. (1976).J. Inmunol. 116, 1439-1446. LeFur, G., Phan, T., and Uzan, A. (1980). Life Scz. 26, 1139-1 148. LeLievre, C. (1974).J . Embryol. Exp. Morphol. 31, 453-477. Lennon, V. A. (1976). Irnmunol. Commun. 5, 323-344. Lennon, V. A., Whittingham, S., Carnegie, P. R., McPherson, T. A., and MacKay, I. R. (1971).J . Immunol. 107, 56-62. Lennon, V., Feldmann, M., and Crawford, M. (1972).Int. Arch. Allergy A$$. Immunol. 43, 749-758. Levine, L. (1965). Neurosci. Res. Prog. Bull. 3, 3-7. Levine, L., and Moore, B. W. (1966).I n “Neurosciences Research Symposium Summaries” (F. 0. Schmitt and T. Melnechuk, eds.), pp. 454-458. MIT Press, Cambridge, Massachusetts. Lewis, J. H. (1933).J. Immunol. 24, 193-211. Liakopoulou, A,, and MacPherson, C. F. C. (1970).J. Immunol. 105, 512-520. Lightbody, J., Pfeiffer, S. E., Kornbilth, P. L., and Henchman, H. R. (1970).J. Neurobiol. 4, 41 1-417. Lindstrom, J. (1979). Adv. Immunol. 27, 1-50. Logan, D. G., and Deodhar, S. D. (1974).J. Am. Med. Assoc. 212, 1703-1704. Lord, J. A. H., Waterfield, A. A., Hughes, J., and Kosterlitz, H. W. (1977). Nature (London) 267,495-499. Ludwin, S. K., Kosek, J. V., and Eng, L. (1976).J . Comp. Neural. 165, 197-208. Lukianenko, V. I. (1961). Usp. Sourem. Biol. 51, 170-187. (In Russian) Lunisden, C. E. (1972). In “Multiple Sclerosis: A Reappraisal” (D. McAlpine, C. E. Lumsden, and E. D. Acheson, eds.), pp. 368-432,512-621. Williams & Wilkins, Baltimore, Maryland.

NEUKOMODULATING ACTIVITY OF ANTI-BRAIN ANTIBODIES

309

McAlpine, D., Lumsden, C. E., and Acheson, E. D., eds. (1965). “Multiple Sclerosis. A Reappraisal.” Livingstone, London. McKay, R., Raff, M. C., and Reichardt, L. F., eds. (1981). “Monoclonal Antibodies to Neural Antigens.” Cold Spring Harbor Laboratory, Cold Spring Harbor, New York. McMillan, P. N., Mickey, D. D., Kaufman, B., and Day, E. D. (1971). J . Zmmunol. 107, 1611-1617. MacPherson, C. F. C., and Chinerman, J. (1971). Exp. Neurol. 31, 45-52. MacPherson, C. F. C., and Liakopoulou, A. (1966).J. Immunol. 97, 450-457. MacPherson, C. F. C., and Shek, R. P. N. (1970).Exp. Neural. 29, 1-15. Mahadik, S . P., Laev, H., Graf, L., and Rapport, M. M. (1974). Trans. Am. Sac. Neurochem. 5, 161. Mahadik, S. P., Korenovsky, A,, and Rapport, M. M. (1979a). J. Neurochem. 33, 751-762. Mahadik, S. P., Graf, L., Korenovsky, A., and Rapport, M. M. (1979b). J . Neurochem. 33, 763-77 1 . Makinodan, T. (1976).J. Am. Geriatr. Sac. 24, 249-252. MarkoviC, B. M., and JankoviC, B. D. (1983). Period. Biol. 85(3), 67-68. Marsh, J. T., and Rasmussen, A. F., Jr. (1960). Proc. SOC.Exp. Biol. Med. 104, 180-183. Martin, S. E., and Martin, W. J. (1975a). Proc. Natl. Acad. Scz. U.S.A. 72, 1036-1040. Martin, S. E., and Martin, W. J. (1975b). Cancer Res. 35, 2609-2612. MatejiC, B., RadulovaEki, M., JankoviC, B. D., MitroviC, K., and Rabie, F. (1969). Sud. Med. J . 7, 27-29. Matus, A., and Mughal, S. (1975). Nature (London) 258, 746-748. Metalnikov, S. (1934). “Kole d u Systeme Nerveux et des Facteurs Biologiques et Psychiques dans I’Inimunitt.” Masson, Paris. Miani, N., De Renzis, G., Michetti, F., Correr, S., Olivieri Sangiacomo, C., and Caniglia, A. (1972).J. Neurochem. 19, 1387-1394. Michetti, F., Miani, N., De Renzis, G., Caniglia, A,, and Correr, S . (1974).J. Neurochem. 22, 239-244. Michetti, F., Massaro, A., and Murazio, M. (1979). Neurosci. Lett. 11, 247-251. Michetti, F., Massaro, A., Russo, G., and Rigon, G. (1980).J . Neural. Scz. 44, 259-263. Mickey, D. D., McMillan, P. N., Appel, S. H., and Day, E. D. (1971).J . Immunol. 107, 15991610. MihailoviC, L., and &pic, D. (1971). Brain Res. 32, 97-104. MihailoviC, Lj., and JankoviC, B. D. (1961). Nature (London) 192, 665-666. MihailoviC, Lj., and Jankovic, B. D. (1965). Neurosci. Res. Prog. Bull. 3, 8-17. MihailoviC, Lj., JankoviC, B. D., Beleslin, B., MitroviC, K., and KrialiC, Lj. (1964). Nature (London) 203, 763-765. Mihailovit, Lj., Jankovit, B. D., Beleslin, B., MiloSeviC, D., and cupit, D. (1965). Nature (London) 206, 904-905. MihailoviC, Lj., Divac, I., MitroviC, K., MiloSeviC, D., and JankoviC, B. D. (1969). Exp. Neurol. 24, 325-336. Mitrova, E., and Mayer, V. (1976). Acta Virol. 20, 479-485. MitroviC, K., Draskoci, M., and JankoviC, B. D. (1964). Experientia 20, 7. Moller, G., ed. (1982). Zmmunol. Rev. 63, 1-209. Mdler, M., Ingild, A., and Bock, E. (1978). Brain Res. 140, 1-13. Moore, B. W. (1965). Biochem. Biophys. Res. Commun. 19, 739-744. Moore, B. W., and McGregor, D. J. (1965).J . Biol. Chem. 240, 1647-1653. Moore, B. W., and Perez, V. J. (1965). I n “Physiological and Biochemical Aspects of Nervous Integration” (F. D. Carlson, ed.) pp. 343-359. Prentice-Hall, New York. Moore, B. W., Perez, V. J., and Gehring, M. (1968).J. Neurochem. 15, 265-272.

310

RRANISLAV D. J A N K O V I ~

Morgan, I. M. (1946).J. Bacterial. 51, 614-615. Morgan, I. M. (1947).J. Exp. Med. 85, 131-140. Morgan, I. G., Wolfe, L. S., Mandel, P., and Gombos, G . (1971).Biochim. Biophys. Actu 241, 737-75 1. Morgan, I. G., Reith, M., Marinari, U., Breckenbridge, W. C., and Gombos, G. (1972).I n “Glycolipids, Glycoproteins and Mucopolysaccharides of the Nervous System” (V. Zambotti, G . Tettanianti, and M. Arrigoni, eds.), pp. 209-228. Plenum, New York. Morgan, I. G., Vincendon, G., and Gombos, G. (1973a). Biochim. Biophys. Acta 32,671-680. Morgan, 1. C., Zanetta, J.-P., Breckenbridge, W. C., Vincerrdon, G., and G n h o s , C . (1973b). Bruin Res. 62, 405-411. Morrel, R. M. (1979). Adv. Neural. 23, 443-447. Moshly-Rosen, D., Fuchs, S., and Eshhar, Z. (1979). FEBS Lett. 106, 389-392. Motycka, A., and Jezkova, Z. (1975). Cas. Lek. Cesu. 114, 1455-1457. Murray, R. M., and Reveley, A. M. (1983). Lancet 1, 583. Myers, P. R., Donlon, M., McCarthy, K., Livengood, D., and Shain, W. (1976). Biochem. Bzophy. Res. Commun. 72, 131 1-1316. Nagai, Y., Sakakibara, K., and Ushida, T. (1980). I n “Search for the Cause of Multiple Sclerosis and Other Chronic Diseases of Central Nervous System. First International Symposium of Hertie Foundation” (A. Boese, ed.), pp. 127-138. Verlag Chemie, Weinheim. Nakajima, T., Yamaguchi, H., and Takahashi, K. (1980). Bruin Res. 191, 523-531. Nandy, K. (1973). Prog. Bruin Res. 40, 437-454. Nandy, K. (1977). In “The Aging Brain and Senile Dementia” (K. Nandy and 1. Sherwin, eds.), pp. 181-196. Plenum, New York. Niedieck, B. ( I 964). Z. Inimunztae!s/ior.tch. 126, 40-48. Nielsen, J., Homma, A., and Biorn-Henriksen, T. (1977).J. Gerontol. 32, 554-561. Niemegeers, C. J. E., and Janssen, P. A. .J. (1979). Life Sci. 24, 2201-2216. Oldstone, M. B. A,, and Dixon, F. J. (1968). Am. J . Pathol. 52, 251-264. Oldstone, M. B. A., and Lampert, P. W. (1974). Adv. Biosci. 12, 381-390. Ordy, J. M., and Brizzee, K. B., eds. (1975). “Neurobiology of Aging.” Plenum, New York. Pabst, H., and Dupuy, J. M. (1970). Fed. Proc., Fed. Am. SOC.Exp. E d . 29, 622. Packman, P. M., Blomstrand, C., and Hamberger, A. (1971).J. Neurochem. 18, 479-487. Pasteur, I.. (1885). C . K . S6nnce.s Amd. Sci. 101, 765-774. Paterson, P. Y. (1960).J. Exp. Med. 111, 119-136. Paterson, P. Y. (1966). Adv. Immunol. 5, 131-208. Paterson, P. Y. (1972). In “Multiple Sclerosis: Immunology, Virology and Ultrastructure” (F. Wolfgram, G. W. Ellison, J. G. Stevens, and J. M. Andrews, eds.), pp. 539-567. Academic Press, New York. Paterson, P. Y. (1976). Fed. Proc., Fed. Am. Soc. Exp. Biol. 35, 2428-2434. Paterson, P. Y. (1979). Rev.Infect. Dis. 1, 468-482. Paterson, P. Y. (1980). In “The Suppression of Experimental Allergic Encephalomyelitis and Multiple Sclerosis” (A. N. Davison and M. L. Cuzner, eds.), pp. 11-30. Academic Press, New York. Paterson, P. Y. (1982). FPd. Proc., Fed. Am. Sac. Ex$. Hiol. 41, 2569-2576. Paterson, P. Y., and Harwin, S. M. (1963).J. Exp. Med. 117, 755-774. Paterson, P. Y., and Whitacre, C. C. (1981). Immunol. Tnduy 2, 1 1 1-1 17. Paterson, P. Y., Jacobs, A. F., and Coia, E. M. (1965). Ann. N.Y. Acud. Sci. 122, 292-298. Paterson, P. Y., Day, E. D., and Whitacre, C. C. (1981). Immunol. Rev. 55, 89-120. Pellkofer, R., and Jatzkewitz, H. (1976).J . Neurochem. 27, 351-354.

NEUKOMODULATING ACI‘IVITY OF ANTI-BRAIN ANTIBODIES

3 11

Perez, V. J., Olney, J. W., Cicero, T. J., Moore, B. W., and Bahn, B. A. (1970).]. Neurochem. 17,511-519. Perry, E. K., Tomlinson, B. E., Blesses, G., Bergmann, K., Gibson, P. H., and Parry, R. H. (1978). Br. Med. J. 2, 1457-1459. Peskovitz, M. D., Paterson, P. Y., Kelly, J., and Lorand, L. (1978). Cell. Immunol. 39, 355365. Phillips, J., Sikora, K., and Watson, J. V. (1982). Lancet 2, 1214-1215. Pick, E., ed. (1981). “Lymphokines,” Vol. 2. Academic Press, New York. Pierpaoli, W., and Sorkin, E. (1972). Experientiu 28, 1385-1389. Pinsker, H. M., and Willis, W. D., J r . , eds. (1980). “Information Processing in the Nervous System.” Raven, New York. Povlishock, J. T., Kriebel, R. M., and Seibel, H. R. (1975). Am. J . Anat. 143, 465-484. Prieto, A,, Kornblith, P. L., and Pollen, D. A. (1967). Science 157, 1185-1187. Pruss, R. M. (1981). Nature (London) 280, 688-690. Quismorio, F. P., and Friou, G. J. (1972). Znt. Arch. Allergy Appl. Zmmunol. 43, 740-748. Radulovatki, M., and JankoviC, B. D. (1966). Int. Congr. Psychol., 17th Moscow p. 247. Rapport, M. M. (1981). In “Gangliosides in Neurological and Neuromuscular Function, Development, and Repair” (M. M. Rapport and A. Gorio, eds.), pp. 91-97. Raven, New York. Rapport, M. M., and Karpiak, S. E. (1978).Zn “Senile Dementia: A Biomedical Approach” (K. Nandy, ed.), pp. 73-88. Elsevier, Amsterdam. Rapport, M. M., and Mahadik, S. P. (1977). In “Mechanisms, Regulation and Special Functions of Protein Synthesis in the Brain.” (S. Roberts, A. Lajtha, and W. H. Gispen, eds.), pp. 221-230. Elsevier, Amsterdam. Rapport, M. M., Karpiak, S. E., and Mahadik, S. P. (1979). Fed. Proc., Fed. Am. Soc. Exp. Bid. 38, 2391-2396. Raskin, N. (1955). Arch. Neurol. 73, 645-655. Rasmussen, H. (1970). Science 170, 404-412. Rasmussen, A. F., Jr., Spencer, E. S., and Marsh, J. T. (1959). Proc. SOC.Exfi. B i d . Med. 100, 878-879. Reichner, H., and Witebsky, E. (1934). Z. Zmmunituetsforsch. 81, 410-432. Reif, A. E., and Allen, J. M. V. (1964).J . Exp. Med. 120, 413-433. Resch, K., and Kirchner, H., eds. (1981). “Mechanisms of Lymphocyte Activation.” Elsevier, Amsterdam. Roizin, L., and Kolb, L. C. (1959). I n “Allergic Encephalomyelitis” (M. W. Kies and C. E. Alvord, Jr., eds.), pp. 5-57. C Thomas, Springfield, Illinois. Rojas, E. (1965). Proc. Natl. Acad. Sci. U.S.A. 53, 306-31 I . Rojas, E., and Luxoro, M. (1963). Nature (London) 199, 78-79. Rose, F. C., ed. (1979). “Clinical Neuroimmunology.” Blackwell, Oxford. Rubin, R. T. (1965). Br. J . Psychiatry 111, 1003-1006. Rusca, G., Galissano, P., and Alema, S. (1972). Brain Res. 49, 223-227. Ryberg, B. (1978).J . Neurol. Sci. 38, 357-382. Ryberg, B. (1982). J. Neurol. Sci. 54, 239-261. SaviC, V., Horvat, J., and JankoviC, B. D. (1976). Period. Bid. 78, 42-43. SaviC, V., Solnceva, E., Horvat, J., and JankoviC, B. D. (19’79a).Period Bid. 81, 215-216. SaviC, V. Lj., JankoviC, B. D., and Horvat, J. (1979b). Zmmunol. Lett. 1, 3-5. Savii-, V., Soltes, S., Horvat, J., and Jankovii-, B. D. (1982). Period. B i d . 84, 192-194. SaviC, V., Soltes, S., Horvat, J., and Jankovii-, B. D. (1983). Period. Bid. 85(3), 71-72. Schmitt, F. 0. (1964). Neurosci. Res. Prog. Bull. 2, 43-75.

312

BRANISLAV I). J A N K O V I ~

Schmitt, F. O., and Davison, P. F. (1961). Actual. Neurophysiol. Ser. 3, 355-369. Schocket, A. L., Weiner, H. L., Walker, J., McIntosh, K., and Kohler, P. F. (1977). Clin. Immunol. Immunopathol. 7, 15-23. Sdiubert, D., Heinemann, S., Carlisle, W., Tarikas, 11.. Kimes, B., Patrick, J., Steinhach, 1.H., Culp, W., and Brdndt, B. I,. (1974). Nature (London) 249, 224-227. Schulster, D., and Levitzki, A., eds. (1980). “Cellular Receptors for Hormones and Neurotransmitters.” Willis, Chichester. Schwartz, M., Selam, B. A., and Eshhar, N. (1982)./. Neurochem. 38, 1192-1195. Seil, F. J., Falk, G. A., Kies, M. W., and Alvord, E. C., Jr. (1968). Exp. Neurol. 22,545-555. Seil, F. J., Smith, E., Leiman, A. L., and Kelly, J. M. (1975). Science 187, 951-953. Semenov, S. F. (1963). Int. Rev. NeurobioE. 9, 601-613. Seybold, M. E., and Lindstrom, J. M. (1982). Springer Semin. Immunopathol. 5, 389-412. Shaskan, E. G., and Lovett, E. J. (1980). Res. Commun. Psychol. Psychiatr. Behav. 5, 241-254. Shaskan, E. G., and Lovett, E. J. (1981). In “Biological Psychiatry 1981” (C. Perris, G. Struwe, and B. Sansson, eds.), pp. 73-84. Elsevier, Amsterdam. Shelanski, M. L. (1973). In “Proteins of the Nervous System” (D. J. Schneider, R. H. Angeletti, R. A. Bradshaw, A. Grasso, and B. W. Moore, eds.), pp. 227-241. Raven, New York. Sher, J. H., and Pertschuk, L. P. (1974).J. Pediatr. 85, 385-387. Sheridan, M. N., and Reiter, R. J. (1973). Am. J . Anat. 136, 363-382. Sherwin, A. L., O’Brien, G. J., Richter, M., Casgrove, J. B. R., and Rose, B. (1963). Neurology 13, 703-707. Shtark, M. B. (1978). “Immunoneurophysiology.” Medicina, Leningrad. (In Russian) Simon, J., and Simon, 0..(1975).E x f . Neurol. 47, 523-534. Small, G. W., and Jarvik, t.F. (1982). Lancet 2, 1443-1446. Snyder, S. H. (1982). Lancet 2, 970-973. Solomon, G . F. (1969).Ann. N . Y . Acad. Scz. 164, 335-343. Solomon, G. F. (1981). In “Psychoneuroimmunology” (R. Ader, ed.), pp. 259-278. Academic Press, New York. Solomon, G. F., Allansmith, M., McCleIlan, B., and Amkraut, A. (1969). Arch. Gen. Psychiatry 20,272-277. Soltes, S., Savil., V., Horvat, J . , andJankoviC, B. D. (1979). Period. B i d . 81, 217-218. Soltes, SaviC, V., Horvat, J., andJankoviC, B. D. (1983). Period. Bzol. 85(3), 69-70. Spector, N. H. (1980). In “Physiology of the Hypothalamus” (P. Morgane and J. Panksepp, eds.), pp. 453-517. Dekker, New York. Spirman, N., Sela, B.-A., and Schwartz, M. (1982).J. Neurochem. 39, 874-877. Stefansson, K., Wollmann, R. L., Moore, B. W., and Arnason, B. G. W. (1982). Nature (London) 295, 63-64. Stein, M., Schiavi, R. C., and Camerino, M. S. (1976). Science 191, 435-440. Stewart, J . A. (1972). Biochim. Biophys. Acta 263, 178-192. Stingl, G., Tamaki, K., and Katz, S. I. (1980). Immunol. Rev. 53, 149-174. Strickland, R. G., and Hooper, B. M. (1972). Pathology 4, 259-263. Sviridov, S. M., Korochkin, L. J., Ivanov, V. N., Maletskaya, E. I., and Bakhtina, T . K. (1972).J. Neurochem. 19, 713-718. Tabuchi, K., Ohnishi, R., Furuta, T., and Nishimoto, A. (1983). Experientia 39, 335-337. Terry, R. D. (1978). Fed. Proc., Fed. Am. Sac. Exp. Biol. 37, 2837-2840. Thomas, D. G. T . (1975). IRCS Med. Sci. 3, 26. Thomas, D. G. T., Palfreyman, J. W., and Ratcliffe, J. G. (1978). Lancet 1, 113. Thompson, E. J . (1976). In “Biochemistry and Neurological Diseases’’ (A. N. Davison, ed.), pp. 278-316. Blackwell, Oxford.

s.,

NEUROMODUIATING ACTIVITY OF ANTI-BRAIN ANTIBODIES

3 13

Tourtelotte, W. W., and Parker, J. A. (1967). Nature (London) 214, 683-686. Tower, D. B. (1978). I n “Alzheimer’s Disease: Senile Dementia and Related Disorders (Aging, Vol. 7)” (K. Katzman, R. D. Terry, and K. L. Bick, eds.), pp. 1-4, Raven, New York. Trisler, G. D., Schneider, M. D., and Nirenberg, M. (1981). l’roc. Natl. Acud. Sci. U.S.A. 78, 2 145-2149. Trotter, J. L., and Brooks, B. R. (1980). I n “Neurobiology of Cerebrospinal Fluid” (J. H. Wood, ed.), pp. 465-486. Plenum, New York. Tsaregorodtseva, T. M., and Chernigovskaya, N. V. (1969). Bul. Eks. Bid. Aled. 67, 89-92. (In Russian) Twomey, S. L., and Samson, F. E., Jr. (1973). Brain Res. 37, 101-108. Tzartos, S., and Lindstrom, J. (1980). Proc. Natl. Acad. Sci. U.S.A. 77, 755-759. Tzartos, S., Rand, D. E., Einarson, B. L., and Lindstrom, J. M. (1981).J. Bid. Chem. 256, 8635-8645. Uyeda, C. T., Eng. L. F., and Bighami, A. (1972). Brain Res. 37, 81-89. Uyemura, K., Vincendon, G., Gombos, G., and Mandel P. (1971).J. Neurochem. 18, 429438. Vartanian, M. E. (1966). Proc. World Congr. Psychiatry, 6th, Amsterdam p. 3030. Vartanian, M. E., Kolyaskina, G. I., Lozovsky, D. V., Burbaeva, G. Sh., and lgnatov, S. A. (1978). In “Neurocheniical and Immunologic Components in Schizophrenia” (D. Uergsnia and A. L. Goldstein, eds.), pp. 339-364. Liss, New York. Vulchanov, V. H., and Hadjieva, Y. (1964). Z. Immun. Allergteforsch. 127, 138-144. Vulliamy, T., Rattray, S., and Mirsky, R. (1981). Nature (London) 291, 418-420. Wahlstroni, T., Linder E., and Saksela, E. (1973). Acla Pathol. Microbiol. Scand. Sect. B 81, 768-774. Waksman, B. H. (1959). Int. Arch. Allergy Appl. Immunol. (Suppl.) 14, 1-87. Waksman, B. H. (1981). Immunol. Today 2, 87-93. Wald, F., Mazzuchelli, A,, Lapetina, E. G., and De Robertis, E. (1968). Exp. NPurol. 21, 336-345. Walker, R. A. (1970). I n “Basic Mechanisms in Epilepsy” (R. Jasper and R. Walker, eds.), pp. 510-513. Wiley, New York. Weigle, W. 0. (1980). Adv. Immunol. 30, 159-273. Weissmann, G., Samuelsson, B., and Paoletti, R., eds. (1979). “Advances in Inflammation Research,” Vol. 1. Raven, New York. Werner, G. H., and Floc’h, F., eds. (1978). “The Pharmacology of Immunoregulation.” Academic Press, New York. Whitacre, C. C., and Paterson, P. Y. (1977).J. Exp. Med. 145, 1405-1410. Whitacre, C. C., and Paterson, P. Y. (1980).J. Immunol. 124, 1784-1788. Whittacker, J. N. (1977). Neurology 27, 911-920. Whittingham, S., Mackay, I. R., Jones, I. H., and Davis, H. (1968). Br. Med. J . 1, 347-348. Wilkinson, P. C., and Zeromski, J. (1965). Brain 88, 529-583. Willenborg, D. 0. (1982). Scand. J . Immunol. 16, 437-441. Williams, G. W., Bluenstein, H. G., and Steinberg, A. D. (1981). Clin. Immunol. Immzmopulhol. IS, 126-132. Wilson, G. S., and Miles, M. M. (1964). “Topley and Wilson’s Principles of Bacteriology and Immunology,” Vol. 2, p. 1397. Arnold, London. Wilson, H. A., Winfield, J. B., Lahita, R. G., and Koffler, D. (1979). Arthritis Rheum. 22, 458-462. Wistar, R., and Hildemann, W. H. (1960). Science 131, 159-160. Witebsky, E., and Steinfeld, J . (1928). 2. Immunituetsforsch. 28, 27 1-296.

3 14

BKANISLAV D. . J A N K O V I ~ :

Zisper, B., and McKay, R. (1981). Nature (London) 289, 549-554. Zornzely-Neurath,C. E. (1982).In “Nervous-System-SpecificProteins” (E. Bock, ed.), pp. 1-40. Blackwell, Oxford. Zomzely-Neurath, C., and Keller, A. (1977).Neurochem. Res. 2, 353-377. Zomzely-Neurath, C. E., and Walker, W. A. (1980).In “Proteins of the Nervous System” (R. A. Bradshow and D. M. Schneider, eds.), pp. 1-57. Raven, New York.

EFFECT OF TREMORIGENIC AGENTS ON THE CEREBELLUM: A REVIEW OF BIOCHEMICAL AND ELECTROPHYSIOLOGICAL DATA By V. G. Longo

and M. Massotti Department of Pharmacology lstituto Superiore di Sanitb Rome, holy

I. Introduction. . . . . . . . . . . . . . . . . . . . . 11. Electrophysiological 111. Neurochemical Data

IV. Conclusions . . . . . . .

................................................ References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

327 328

1. Introduction

For some time, the interest of pharmacologists in tremorigenic substances has been mainly related to research on antiparkinsonian drugs. In the late 1940s a paper was published in which nicotine tremors were proposed as a screening method for potential antiparkinsonian agents (Longo and Bovet, 1949). In 1956 Everett described the tremorigenic properties of 1,4-dipyrrolidino-2-butyne(tremorine), which was employed for the same purpose together with its active metabolite oxotremorine. The tremors caused by nicotine and tremorine appear to be primarily of subcortical origin, involving mainly the cholinergic system. Another drug, harmine, produces sustained tremors but no signs of parasympathetic involvment. This compound and its analog harmaline have also been used as models of neurological diseases. Harmine and harmaline, in addition to their antimonoamino-oxidase properties (Pletscher et al., 1959), exhibit some other effects on certain monoaminergic mechanisms, especially on the nigrostriatal dopaminergic pathway. According to Poirier et nl. (1968), harmaline hinders dopamine synthesis from L-Dopa and its conversion to homovanillic acid. In a later investigation, the same group (Larochelle et al., 1971) described a potentiation by harmaline of the postural tremor present in monkeys with 315 I N T E R N A T I O N A L REVIEW OF NECIROHIOLOGY, VOL. 26

Copyrighi 0 1985 hy Academic Press. Inc. All rights of rrprnduction in any form reserved. ISHN o-iz-vx8z6-:4

3 16

V.

C . LONGO AND M . MASSOTTI

lesions of the ventroniedial tegmental area which interrupt the nigrostriatal pathways. Since the early 1970s, harmine has been used in a series of investigations of a more basic nature and has proved an invaluable tool for tracing the pathways connecting some lower brain stem structures to the cerebellum. In parallel with the electrophysiological research, biochemical investigations have been carried out complementary to studies on the tremorigenic effects of the drug.

II. Electrophysiological Data

Investigations (Fuentes and Longo, 1971) carried out in rats and rabbits on the central effects of harmine and harmaline have revealed that tremorigenic doses of these two drugs induce alterations of the electroencephalogram (EEG) consisting of an increase of frequency and voltage of electrical activity recorded from the cerebral cortex. It was also shown that L-Dopa antagonized harmine and harmaline tremors without influencing the effect of these drugs on the EEG. A few years later, quite independently, Llinas and Volkind (1973) and De Montigny and Lamarre (1973) demonstrated by unit recording that, in the cat, harmaline-induced tremors are associated with a rhythmic firing at approximately 10 Hz which is recorded along the olivocerebellar-bulbospinal pathways. I n Fig. 1 there is a schematic representation of these pathways. According to Llinas and Volkind (1973), the drug acts on the inferior olive; then through the olivocerebellar pathway the rhythmic firing is transmitted to the fastigium and through the climbing fibers to the Purkinje cells of the paleocerebellum. These project back to the cerebellar nuclei; the output of the fastigial nucleus activates the spinal motoneurons through the Deiters and reticular nuclei, as demonstrated by the bursting activity recorded from bulbar reticular neurons and from spinal anterior roots. In a series of experiments carried out in the cat, Llinhs and Volkind (1973) demonstrated that the action of harmaline is exerted directly on the cells of the inferior olive. In animals with the inferior peduncles chronically severed, harmaline does not elicit rhythmic firing in Purkinje cells or at the cerebellar nuclei level. These animals also demonstrate a total lack of tremor following drug administration. This finding strongly suggests that harmaline is not acting directly at the level of the cerebellar cortex itself and that the rhythmic climbing fiber firing is of extracerebellar origin. De Montigny and Lamarre (1973) removed the entire cere-

EFFECT OF TREMORIGENIC: AGENTS ON CEREBELLUM

317

to thalamus I

I

FIG. 1. Schematic drawing of the two pathways involved in harmine and harmaline tremors, both including the inferior olive as a pacemaker. The continuous line indicates the neuronal connections of one pathway. The drug acts on the inferior olive; the firing at approximately 10 Hz is transmitted through the climbing fibers to the Purkinje cells. These project back to the fastigium and to the Deiters and reticular nuclei. The second pathway (interrupted lines) is a loop including the inferior olive, the neocerehellar cortex, the dentate nucleus, and the nucleus ruher.

bellum, including all cerebellar nuclei; after administration of harmaline the rhythmic activity was still present in the inferior olive. No rhythmic activity was found in other regions of the medulla such as the lateral reticular, the vestibular, paramedian, and gigantocellularis reticular nuclei. These data indicate that the rhythmic activity is not generated within the loops connecting the cerebellum to the brainstem nuclei and that the rhythm observed in the inferior olive does not derive from other nuclei of the medulla. De Trujillo et al. (1976) carried out a study on the effects of' harmine and harmaline on the rabbit EEG. The authors recorded with gross electrodes and with an ink-writing apparatus 10-Hz spiking activity in the cerebellar vermis and synchronous waves at the level of the lumbar enlargement of the spinal cord (Fig. 2). It should be mentioned at this point that in a paper published in 1970 by Villablanca and Riobo, EEG records of the pontine reticular formation showed bursts of 8- to 12-Hz spiking after harmaline. The authors limited themselves to the description of these pattern, without going into a detailed analysis of its origin.

Control

A

**

FIG. 2. Effects of harmine on electrical activity recorded from various sites of the cerebrospinal axis. Three minutes after administration of 2 mg/kg of harmine the cortical slow waves and spindles are replaced by low-voltage rapid activity, particularly evident in the anterior leads. Spikes at 10 Hz appear in the cerebellar leads and synchronous waves at the same frequency are present in the spinal cord. Unanesthetized rabbit immobilized with gallamine and under artificial respiration. The drawing at upper left indicates recording points. (De Trujillo et al., 1976.)

EFFECT OF TKEMORIGENIC AGENTS ON CEREBELLUM

319

According to De Trujillo et al. (1976), electrode positioning is critical for the recording of the harmine effect: the area of best recording is the lobulus medianus. In this area there is maximal synchronization, giving rise to 10-Hz spiking, as demonstrated also by data obtained with singlecell recording and reported by Llinas and Volkind (1973) (Fig. 3). De Trujillo et al. (1976) also investigated the effect of diazepam on tremors induced by harmine. Doses of 0.5, 1, and 2 mg/kg given iv 5 min before harmine prevented the appearance of tremor. These doses of diazepam also abated tremors when administered after harmine, at the peak time of the effect of the alkaloid. Animals treated with diazepam either before or after harmine exhibited ataxia and sedation. The EEG record indicated that harmine-induced spiking in the cerebellar leads was not modified by diazepam, which instead prevented or blocked spinal paroxysmal activity and induced the appearance of spindles in the cortical cerebral leads (Fig. 4).On the other hand, administration of LDopa in doses (5-10 mg/kg iv) which in previous experiments proved able to prevent or abate harmine tremors did not modify spiking in the cerebellar leads or the corresponding synchronous waves in the spinal cord. T h e adminstration of dopamine (5-10 mg/kg iv) proved able to abate, but not to prevent, the harmine tremors (Huidobro-Tor0 et al., 1975). In the case of dopamine as well, disappearance of the tremors did not correspond to modifications of the record. The effects of harmine on the electrical activity of other subcortical nuclei were investigated by Gogolak et al. (1977) and by Scotti de Carolis

FIG.3. Effects of harmaline on Purkinje cells firing in cats. In the upper tracings (A) are recorded simultaneously the activities of two Purkinje cells near the midline in the cerebellar vermis; the bursts occur in an almost synchronous manner. When simultaneous recordings are obtained from three different units situated at increasing distances from the center of the vermis (as in B), the synchronization is less apparent. (Llinas and Volkind, 1973.)

320

V. G. LONGO AND M . M A S S O T T I A. Harmine ( 2 mg/kg iv) SMC

:

VERM

SPIN

B. Harmine 10 min after diazepam (1 mg/kg iv)

P --

C. Harmine 90 min after diazepam

25pV

1 .w - : / # w / / w 2 5cc

FIG.4. Antagonistic effect of diazepam on spinal EEG seizures due to harmine. T h e upper record (A) was taken 4 min after administration of harmine. Spikes appear in the cerebellar lead, and synchronous waves are present in the spinal lead. T h e middle record (B) illustrates the effect of the same dose of harmine administered after pretreatment with diazepam. Spinal synchronous waves are absent, while cerebellar spikes are still present. The lower record (C) was taken 90 min later, and illustrates the disappearance of the antagonistic effect of diazepam. Rabbit immobilized with gallamine. Leads: SMC, anterior sensorimotor cortex; VERM, cerebellar vermis; SPIN, lumbar enlargement of the spinal cord. (De Trujillo et al., 1976.)

et al. (1978). Both groups of investigators found that after harmine o r harmaline injection a rhythm synchronous with cerebellar spiking can be recorded in the nucleus ruber (Fig. 5). The work carried out by Gogolak et al. (1977) included recording with macro- and microelectrodes and an autocorrelation analysis of the firing pattern. Their results indicate that the level of synchronization induced by the drug is lower in the nucleus

EFFECT OF TKEMORIGENIC AGENTS ON CEREBELLUM

32 1

ruber than in the cerebellar cortex. The conclusion drawn by these authors points to an interconnection between two circuits, both including the inferior olive as a pacemaker. The rubro-olivo-cerebellar-rubral loop includes the inferior olive, the neocerebellar cortex, the dentate nucleus, and the nucleus ruber, while the other pathway includes the inferior olive, the paleocerebellum, the fastigial nucleus, the reticular and the Deiters’ nuclei, and the spinal cord (Fig. 1). Other results relevant to the effects of harmine on the cerebellum should be mentioned at this point. Swedish investigators (Sjolund et al., 1977; Wiklund et al., 1977) demonstrated that olivary neurons projecting to different areas of the cerebellar cortex receive widely different densities of indolamine-containing (probably serotoninergic) terminals. The origin of these neurons is still under debate, but is probably the medial raphe nuclei. Selective destruction of these serotoninergic terminals with 5,6-dihydroxytryptamine causes a significant attenuation of both the tremor and the climbing fibers activity induced by harmaline. Headley et al. (1976) and Headley and Lodge (1976) investigated in the rat the effects of several compounds on the synchronous rhythmical activity which occurs in the inferior olive following the electrophoretic or intravenous administration of harmine and harmaline. Serotonin, as well as a number of tryptamine derivatives, antagonized the established rhythmical activity when applied electrophoretically. These results suggest that the serotoninergic innervation of the inferior olive is of importance for the action of harmaline; however, available data do not permit a straightforward explanation of the mechanisms involved. T h e chemical similarity between harmala alkaloids and serotonin suggests an action upon the same receptor. Is harmaline an agonist or an antagonist of serotonin? According to Headley et al. (1976), harmaline might antagonize the inhibitory serotoninergic input in the inferior olive through a presynaptic action. In favor of this hypothesis is the fact that depletion of serotonin or inhibition of serotonin synthesis potentiates harmaline tremor (Kelly and Naylor, 1974); moreover, Zetler (1957) found serotonin, tryptamine, and LSD among the most active antagonists of harmine tremors in mice. In several investigations the effects of other tremorigenic drugs on the olivocerebellar pathways were studied. Tremorine and oxotremorine (De Montigny and Lamarre, 1974; Knowles and Phillips, 1980; Simantov et al., 1976) and nicotine (personal data) were found to be devoid of any effect. O n the other hand ibogaline (De Montigny and Lamarre, 1974) and ibogaine (personal data) exhibited electrophysiological and behavioral patterns very similar to that of harmine. Ibogaline,

1

EFFECT OF TREMORIGENIC AGENTS O N CEREBELLUM

323

contrary to harmine and harmaline, is devoid of anti-monoamino-oxidase inhibitory properties, which excludes a possible involvement of this effect in the production of the tremor.

111. Neurochemical Data

There has also been a neurochemical approach to the tremors induced by harmaline. T h e cerebellum differs from the other brain areas because the concentrations of cyclic adenosine monophosphate (CAMP) and of cyclic guanosine monophosphate (cGMP) are almost equal, while in other brain regions the concentration of cAMP is about 10-fold that of cGMP. T h e cellular location of cerebellar cGMP is probably the Purkinje cells since levels of cGMP in a mutant strain of mice (nervous mice) with a selective degeneration of these cells is low while cAMP is unchanged. cGMP was therefore considered as the second messenger in the cerebellar cortex, and several studies were carried out on the effects of drugs, among them harmaline, on its level. An increase in the cerebellar cGMP content was reported in a series of papers issued by Costa and co-workers after administration of tremorigenic doses of harmaline (Costa et al., 1975). This has been related to the stimulation of Purkinje cells by the climbing fibers since pretreatment with 3-acetylpyridine (3AP), which destroys these fibers, abolishes the tremors and also the increase in cGMP (Simantov et al., 1976; Guidotti et al., 1975). 3-Acetylpyridine is a nicotinamide antagonist which when administered in rather high doses (LD30-LD50), induces a fairly selective destruction of the inferior olivary nucleus. In animals treated with 3AP, harmaline induces only hyperextension of the forelimbs and ataxia. If these animals receive increasing doses of isoniazid, they exhibit, respectively, an increase in cerebellar cGMP content and convulsions. Since isoniazid decreases cerebellar GABA levels, the increase in cGMP content might reflect a reduction in the availability of GABA at the level of postsynaptic receptors, with eventual prevalence of stimulatory inputs (Ma0 et al., 1975). FIG.5. Effects of harmine on the EEG of the rabbit. The control tracing was recorded under basic conditions in the curarized rabbit (gallamine, 5 mg/kg iv). At the arrow, an acoustic stimulus was applied. The harmine record was taken 3 min after the administration of the drug. The cortical spindles and slow waves were replaced by low-voltage rapid activity; synchronous waves at 10 Hz appear in the leads from the nucleus ruber. Leads: ASM, anterior sensorimotor cortex; PSM, posterior sensorimotor cortex; OPT, optic cortex; HYP, mamillothalamic tract; LNR, left nucleus ruber; RNR, right nucleus ruber.

324

V. G. LONGO AND M. MASSOTTI

Harmaline tremors seem therefore to be connected with an increase of cGMP due to the activation of Purkinje cells through the liberation of an unknown transmitter from the climbing fibers terminals. Some data suggest that this transmitter could be glutamate, since electrophoretic application of glutamate increases the discharge rate of Purkinje cells in normal rats as well as in rats pretreated with 3AP. Another interesting contribution to defining the map of harmalineinduced rhythmic activity is that of Batini et al. (1981), who studied the changes of local cerebral glucose consumption upon administration of tremorigenic doses of the drug in cats. To find the brain structures activated by the drug the autoradiographic method with labeled 2-deoxyglucose was used. In addition to selected parts of the olive and of the vermian and paravermian zones of the cerebellum, other structures displayed increased radioactivity: some reticular nuclei, the nucleus ruber, and the basal ganglia. In animals with unilateral sections of the inferior cerebellar peduncle, the basal ganglia were still marked, suggesting a direct pharmacological action of the drug rather than a transmitted one secondary to the activation of the olive. It seems, therefore, that the striatum plays a role in the tremorigenic effect of harmine. This is confirmed by the data of Kelly and Naylor (1975), who found that destruction of the globus pallidus diminishes harmaline tremors. Although these data also support a role for the dopamine-serotonin mechanism in the modulation of harmaline tremor, we still have no indication as to the precise site of action of the drug. Robertson (1980) suggests that harmaline action may be at least in part the result of its interaction with benzodiazepine-specific recognition sites. This author found that in mice harmaline induces tremors when the whole brain level of the drug is above the molar concentration which is able to displace 13H]diazepam from its low-affinity recognition sites.

IV. Conclusions

On the basis of electrophysiological results and of pharmacological manipulations which have been reviewed, we can forward some hypotheses of the neurophysiological and biochemical bases of the effects of tremorigenic drugs on the cerebellum. T w o types of afferent excitatory neurons impinge on the cerebellar cortex: the climbing fibers, originating in the olive and synapsing directly with Purkinje cells, and the mossy fibers, synapsing with the granular cells. The axons of the granular cells generate the parallel fibers

EFFECT OF TREMORIGENIC AGENTS O N CEREBELLUM

325

which in turn synapse with Purkinje cells; the parallel fibers, however, also impinge on numerous inhibitory neurons: stellate, basket, and Golgi cells. T h e activation of these inhibitory neurons may extinguish the excitation of Purkinje cells exerted through the parallel fibers (Fig. 6). Some harmala and iboga alkaloids such as harmine, harmaline, ibogaline, and ibogaine induce an 8- to 12-Hz tremor. Cerebellar structures play a basic role in the genesis of this tremor. The pacemaker has been localized in the inferior olive neurons, more precisely in those neurons receiving a serotoninergic input. One hypothesis which has been forwarded is that these drugs exert a presynaptic effect antagonizing inhibitory serotoninergic innervation and releasing the tendency of the olivary neurons to fire repetitively. This tendency is due to electrotonic coupling between neurons, as indicated by electrophysiological and anatomical data (gapjunctions) (Llinas et al., 1974). Through the climbing fibers Purkinje cells are activated; the activation of Purkinje cells is accompanied by an increase of the cGMP content of the cerebellar cortex. From the cerebellar cortex through the cerebellar and reticular nuclei rhythmic stimuli reach the spinal motoneurons for the final motor issue. It should be noted at this point that the frequency of this tremor is similar to the so-called physiological tremor and differs from that of Parkinsonism (5-7 Hz) and from cerebellar tremor (3-4 Hz). The basic role of the olivocerebellar connection in originating and maintaining this tremor is demonstrated by the disappearance of the tremors in rats treated with 3AP, which induces degeneration of the olivary neurons. A further demonstration could be gained by checking

I'

MF

FIG. 6. Neuronal connections in the cerebellar cortex. Black cells are inhibitory; white cells are excitatory. For details see text. BC, basket cell; SC, stellate cell; PC, Purkinje cell; GoC, Golgi cell; GrC, granular cell; CF, clirnhing fiber; MF, mossy fiber. (Eccles, 1973.)

326

V. G . L O N G 0 A N D M . MASSWlI'I

the tremorigenic effect of harmaline in mutant mice (nervous mice) devoid of Purkinje cells. We are not aware of experiments that have been performed along this line. On the other hand, we have results which throw further light on some other cerebellar regulatory mechanisms, in particular the role of GABA. GABA is probably the inhibitory neurotransmitter in several cerebellar relays. By raising or lowering its level it is possible to change the firing rate of Purkinje cells and the cerebellar cGMP content. Blocking GABA synthesis with isoniazid causes an increase in cerebellar cGMP and convulsions. This increase is still present in rats pretreated with 3AP and therefore is probably connected with the stimulation of the mossy fibers system. If one analyzes the effect of diazepam on this background, the picture which emerges is not clear. Discordant data exist on the effect of diazepam on Purkinje cells. T h e increase in frequency described by Julien (1972) in the cat has not been confirmed by Haefely et al. (1975) who, on the contrary, found a decrease in the firing rate after injection of the drug in both rats and cats. According to De Trujillo et al. (1976), the EEG of rabbits treated with diazepam shows an increase in voltage in the recording from the surface of the Cerebellum. These data suggest that increased electrical activity is present at the level of the cerebellar cortex, which does not necessarily result from direct stimulation of Purkinje cells. This possibility is in line with the data of Speth et al. (1981), who showed that granule cells possess the majority of'the benzodiazepine receptors, which however are also present in Purkinje cells. Diazepam lowers the cGMP content of the cerebellum and abates or prevents harmaline tremors; it does not block the cerebellar 8- to 12-Hz firing observed after harmaline (De Trujillo et al., 1976) or the cGMP modifications induced by harmaline and 3AP (Biggio et al., 1977a,b). These data indicate that the diazepam-induced lowering of cGMP is probably not linked to direct action of the benzodiazepine on the climbing fiber-Purkinje cells system, whereas it can be ascribed to the increase of the GABAergic tonus in the cerebellum, as confirmed by data which show that diazepam attenuates the isoniazide-induced increase of cGMP (Biggio et al., 1977b). The finding that GABA stimulates the release of glutamate (Levi et al., 1981) would suggest that the increase of electrical activity found in the cerebellar cortex after diazepam might be due to a resultant increase of glutamatergic activity. In line with this hypothesis, there is the inability of diazepam to modify the cerebellar 8- to 12-Hz firing induced by harmaline. T h e possibility that the effect of diazepam at the level of the cerebellum is mediated via a granular cells-parallel fibers-Purkinje cells system, as suggested by the data on the inhibitory action of isoniazide-induced increase of cGMP (Biggio 1977a), appears

EFFECT OF TREMORIGENIC AGENTS ON CEREBELLUM

32’1

to be confirmed by the distribution of diazepam receptors at the level of granule cells and Purkinje cells (Speth et al., 1981). From this hypothesis we can speculate that the regulatory input to Purkinje cells from the mossy fibers is operative under basic conditions. However, this regulation fails to be operative when the climbing fibersPurkinje cells system is activated o r inhibited. Therefore, the antitremorigenic effect of diazepam on harmaline-induced tremors might be due to an action on some relay nuclei (cerebellar or reticular) between the cerebellar cortex and the spinal cord. An effect on more rostra1 modulatory centers such as the basal ganglia or the cortex cannot, however, be excluded. T h e data of Batini et al. (1981), for instance, indicate an increase in metabolism directly exerted by harmaline on the striate. Attention should be called to the fact that tremors can be absent even if all the electrophysiological manifestations are present, as for instance after administration of L-Dopa or even dopamine. The antagonism of harmine tremors by L-Dopa could be explained as an effect of the drug on striatal nuclei which have a modulatory influence on spinal motoneurons. Against this hypothesis there are two sets of data: (1) after L-Dopa administration and cessation of tremors, IO-Hz synchronized activity is still present in the spinal leads. (2) Dopamine is also active in abating harmine tremors. Neuropharmacologists therefore should not be oblivious to peripheral effects which quite often play an important role in the mechanism of action of centrally acting drugs. There are data in the literature which suggest that the peripheral beta-adrenergic receptors are involved in the so-called essential tremor (Young et al., 1975).T h e location of the receptors and their possible relation to muscle spindles is not well understood; however, on the basis of our results, the possibility should be considered of an antitremor effect of L-Dopa and dopamine based on an extracerebra1 mechanism of action.

V. Summary

T h e large group of centrally acting drugs includes only relatively few tremor-producing compounds. On the one hand, there are some harmala and iboga alkaloids; on the other hand, there are the cholinomimetics tremorine and nicotine. Since the cerebellum plays a role in the regulation of tonic and phasic reflexes, several investigations have been devoted to the effects of tremorigenic drugs on cerebellar structures.

328

V. C. LONGO AND M. MASSOTTI

Electrophysiological investigations in cats and rabbits have demonstrated that the tremor induced by harmine or harmaline is associated with a rhythmic activity at about 10 Hz generated in the inferior olive and reaching the spinal motor neurons through the cerebellum and the Deiters’ and reticular nuclei. This activity can be recorded from the appropriate sites with both micro- and macroelectrodes. Nicotine and tremorine are devoid of such an effect. T h e mechanism of action of harmaline is still under debate. Probably it antagonizes the inhibitory serotoninergic innervation of the olive, releasing the tendency of the olivary neurons to fire repetitively. In the rabbit, diazepam antagonizes harmaline tremors. An EEG investigation of this antagonism has indicated that the drug blocks spinal seizures without influencing cerebellar spiking; therefore its site of‘ action is probably at the pontobulbar level. In the rat, the tremor induced by harmaline is paralleled by an increase of cGMP in the cerebellar cortex. This increase has been attributed to activation of Purkinje cells by the climbing fibers, since it is prevented by treatment with 3-acetylpyridine, which destroys these fibers. Acknowledgment

T h e authors are grateful to Patrizia Campagna, who prepared the typed manuscript. References

Batini, C., Buisseret Delmas, C., and Conrath Verrier, M. (1981). Exp. Brain Res. 42, 371382. Biggio, G., Brodie, B. B., Costa, E., and Guidotti, A. (1977a). Proc. Natl. Acad. Sci. U.S.A. 74,3592-3595. Biggio, G., Costa, E., and Guidotti, A. (1977b).J. Pharmacol. Exp. Ther. 200, 207-215. Costa, E., Guidotti, A., and Mao, C. C. (1975). I n “Mechanisms of Action of Benzodiazepines” (E. Costa and P. Greengard, eds.), pp. 113-130. Raven, New York. De Montigny, C., and Lamarre, Y. (1973). Brain Res. 53, 81-95. De Montigny, C., and Lamarre, Y. (1974). Brain Res. 82, 369-373. De Trujillo, G. C., Scotti d e Carolis, A., and Longo, V. G. (1976). Neuropharmacology 16, 31-36. Eccles, J. C. (1973).J. Physiol. (London) 229, 1-32. Everett, G. M. (1956). Nature (London) 177, 1238. Fuentes, J. A., and Longo, V. G. (197 I). Neuropharmacology 10, 15-23. Gogolak, G., Jindra, R., and Stumpf, Ch. (1977). Experientia 33, 1352-1354. Guidotti, A., Biggio, G., and Costa, E. (1975). Brain Rex 96, 101-205. Haefely, W., Kulcshr, A., Mohler, H., Pieri, L., Polc, P., and Schaffner, K. (1975). In “Mechanism of Action of Benzodiazepines” (E. Costa and P. Greengard, eds.), pp. 131-151. Raven, New York. Headley, P. M., and Lodge, D. (1976). Brain Res. 101, 445-459. Headley, P. M., Lodge, D., and Duggan, A. W. (1976). Brain Res. 101, 461-478.

EFFECT OF TREMORIGENIC AGENTS ON CEREBELLUM

329

Huidobro-Toro, J. P., Scotti de Carolis, A., and Longo, V. G. (1975). Phnrmacof. Ezochem. Behav. 3, 235-242. Julien, R. M. (1972). Neuropharmaculogy 11, 683-691. Kelly, M., and Naylor, R. J. (1974).J. Pharmacol. 27, 14-24. Kelly, M., and Naylor, R.J. (1975). Eur. J. Pharmacol. 32, 76-86. Knowles, W. D., and Phillips, M. I. (1980). Neuropharmacology 19, 745-756. Ldrochelle, L., Bedard, P., Poirier, L. J., and Sourkes, T. L. (1971). Neuropharmacology 10, 273-288. Levi, G., Gallo, V., and Raiteri, M. (1981). I n “Glutamate as a Neurotransmitter” (G. Di Chiara and G. L. Gessa, eds.), pp. 127-137. Raven, New York. Llinas, R., and Volkind, R. A. (1973). Exp. Brain Res. 18, 69-87. Llinas, R.,Baker, R., and Sotelo, C. (1974).J. Neurophysiol. 37, 560-571. Longo, V. G., and Bovet, D. (1949). Farmaco 4, 515-525. Mao, C. C., (hidotti, A,, and Costa, E. (1975). Nuu7iyii-SchiiilzedebergsArch. Pharmucol. 289, 369-378. Pletscher, A., Besendorf, H., Bachtold, H . P., and Gey, K. F. (1959). Helv. Physiol. PharmaC O ~ .A C ~ U 17, 202-214. Poirier, L. J., Singh, P., and Boucher, R. (1968). CanJ. Physiol. Pharmacol. 46,585-589. Robertson, H. A. (1980). Eur.1. Pharmacol. 67, 129-132. Scotti de Carolis, A,, Florio, V., and Longo, V. G. (1978). Neuropharmacology 17,295-298. Simantov, R., Snyder, S . H., and Oster-Granite, M. L. (1976). Brain Res. 114, 144-151. Sjolund, B., Bjorklund, A,, and Wiklund, L. (1977). Brain Res. 131, 23-37. Speth, R. C., Bresolin, N., Mimaki, T., Deshmukh, P. P., and Yamamura, H. I. (1981). I n “GABA and Benzodiazepine Receptors” (E. Costa, G . Di Chiara, and G . Gessa, eds.), pp. 27-39. Raven, New York. Villablanca, J., and Riobo, F. (1970). Psycopharmacologza (Berlin) 17, 302-313. Wiklund, L., Bjorklund, A,, and Sjolund, B. (1977). Brain Res. 131, 1-21. Young, R. R.,Growdon, J. H., and Shahami, B. T. (1975). N . Eng1.J. Med. 293,950-953. Zetler, G. (1957). Arch. Expr. Pathol. Pharmacol. 231, 34-54.

This Page Intentionally Left Blank

INDEX A

in CNS, 97-102 regulation, 105 trans-4-Aminocrotonic acid, 9 7 (RS)-a-Amino-3-hydroxy-5-methylisoxazole-4-propionate, binding studies, 124 3-Amino- I-hydroxy-2-pyrrolidone,I09 structural formula, 11 1 2-Amino-4-phosphonobutyrate binding studies, 114, 124-125 structural formula, 1 I I 2-Amino-7-phosphonoheptanoate binding studies, 114, 125-126 structural formula, I10 2-Amino-3-phosphonopropionate, structural formula, 11 I 2-Amino-5-phosphonovalerate, 1 10 binding studies, 114, 125 structural formula, 1 1 1 3-Aminopropanesulfinate, 97, 98 D-a-Aminosuberate, 109 structural formula, 1 1 1 y-Aminovaleric acid, structural formula, 99 Anti-brain antibodies biological activity, 269-292 in vitro activity, 274-285 in vivo activity, 270-273 monoclonal, 287-288 neuroimmunological diseases and, 262269 Antidepressant drugs, antimuscarinic potencies, I8 1 Anti-ganglioside antibodies, 285-287 Antimuscarinic drugs, 180- 182 Antipsychotic drugs, antimuscarinic potencies, 181 Anti-thymocyte antibody, 285 APB, see 2-Amino-4-phosphonobutyrate APH, see 2-Amino-7-phosphonoheptanoate APP, see 2-Amino-3-phosphonopropionate 3-APS, see 3-Aminopropanesulfinate APV, see 2-Amino-5-phosphonovalerate

Acetylcholine receptors, see Muscdrinic receptors ACTH, see Adrenocorticotropin Adenylate cyclase, 174- I75 Adrenal cortex, opioid regulation of, 56 Adrenal medulla, opioid regulation of', 54-56 Adrenocorticotropin, 5 secretion control, 45-48 Affective disorders, opioid peptides and, 25-26 Aging, 293-294 P-Alanine, 107-108 structural formula, 99 Alcohol, opioid peptides and, 24-25 Alcoholism, hypersensitivity to brain antigens and, 299-300 Amino acid neurotransmitters, see also specific transmitters identification, 86-88 Amino acid receptor, see also Receptor excitatory, 108-136 inhibitory, 96-108 multiplicity of, 96 Amino acid receptors, excitatory agonists, 108- 1 I0 antagonists, 109-1 1 1 binding studies, 123-129 classification, 1 12 pharmacology, 126- 129 regulation of, 129-135 D-a-Aniinoadipate, 109 structural formula, 1 1 1 y-Aminobutyric acid identification, 86-88 interaction with henzodiazepines, 103I05 pharmacology, 96-97 postsynaptic binding, 97- 102 structural formula, 99 y-Aminobutyric acid receptors, 96-105 behavior and, 102-103 33 1

332

INDEX

Aspartate receptor, binding studies, 112, 119-120 Aspartic acid identification, 86 neuropathology, 135-136 structural formula, 110 p-u-Aspartyl-p-alanine, structural formula, 1 1 1 Autoimmune diseases, anti-brain antibodies and, 262-269 Axon, autoimmune studies with, 274 Axonal growth determinants of, 210-217 recovery of fi~nctionand, 217-223

B Behavior, anti-brain antibody and, 290292 Benzodiazepines, 98, 10 1, 103- 105 binding studies, 139-140 Bicuculline methiodide, structural formula, 99 Binding site, 90 Blood pressure, opioid peptitles and, 24 Brain, anti-brain antisera studies,270273 brain antigens, 256-262, see also Antibrain antibodies circulating as tissue fragments, 267-268 Brain injury, functional recovery from, 201-247 axonal growth and, 210-223 metabolic correlates of, 235 neurochemical adaptations, 223-238 theories on, 202-209 BZD, see Benzodiazepines

Cerebellum, effect of tremorigenic agents on, 315-329 Cerebral atrophy, 298 fi-Chloromercurihenzoate, 165 3-(M-ChloroplienyI-carbamoyloxy)-2butynyltrimethyl ammonium, 164 Cholecystokinin, 58 Chorionic gonadotrophic hormone, 58 Circadian rhythmicity, of opioid peptide levels, 20-21, 26, 27 Climbing fibers, tremors and, 324-327 Copper deficiency, muscarinic receptors and, 188 Corticotropin-releasing factor, 14-20 Cysteate, structural formula, 110 Cysteine sulfinate, binding studies, 114, 123

D Dementia, 293-294, 298 Depression, 298 u-cu-Diaminopirrielate, 109 D-a-Diaminosuberate, 109 Diaschisis, 206-209 Diazepam, 320, 321 1,4-DipyrroIidino-2-butyne, see Tremorine Disulfide reagents, 164-36.5 5,5'-Dithio (2-nitrobenzoic acid), 164- 165 Dithiothreitol, 165 Domoate, 108 structural tormula, 110 DTNB, see 5,5'-Dithio (2-nitrobenzoic acid) Dynorphin, see also Opioid peptides cellular origins of, 7 secretion modulation, 26-27

E C Carbachol, 170-172, 185 Cell-mediated immunity, psychiatric diseases and, 297-299 Central nervous system, see also Brain; Brain injury, functional recovery from immune system and, 252-255 muscarinic receptor subtypes, 15 1-199 recovery of function in, 217-223

Endocrine secretion opioid controls of, 32-58 opioid peptides and, 1-83 P-Endorphin, .see also Opioid peptides cellular origins of, 5-7 circadian rhythmicity arid, 20-21 secretion control, 13-20, 45-48 secretion modulation, 20-26 Enkephalins, see also Opioid peptides cellular origins of, 7 secretion modulation, 27-28

333

INDEX

Epilepsy, 268-269 N-Ethylmaleimide, 164- 165 Experimental allergic encephalomyelitis, 263-266

F 4-Fluoroglutamic acid, structural formula, 110 Follicle-stimulating hormone, 39

G GABA, .we y-Aminobutyric acid Gallamine, 162-164, 170 Gastrin, 58 GDEE, see Glutainate diethyl ester Glial fibrillary acidic protein, 258 Glutamate diethyl ester, 109, 110 structural formula, 1 1 1 Glutamate receptor binding studies, 116-1 18 biochemical characterization, 112- 118 pharmacology, 113-1 14 Glutamic acid identification, 86 neuropathology, 135- 136 structural formula, 110 y-o-Glutamylglycine, 110 structural formula, 1 1 1 Glycine identification, 86-88 structural formula. 99 Glycine receptor, 105- 108 Growth hormone, 42-44 Guanylate cyclase, 172- 174 Gut, opioid regulation of, 58

H HA-966, see 3-Amino- 1-hydroxy-2-pyrrolidone Harmaline, 315, 316-321, 323-324 Harmine, 315, 322, 324 Hippocampus, recovery of function in, 220-221 Homocysteate, structural formula, 1 10 Hormone, see also spec@ hormones in immune microenvironment, 25 1 muscarinic system and, 18.5-187 opioid regulation of, 32-58

Hypothalamo-pituitary axis, opioid peptide systems of, 6

I Ibogaline, 321-322 Ihotenate, 108 structural formula, 110 Immune microenvironment, 250-252 Immune system, nervous systems and, 252-255 Immunopsychiatric diseases, 293-300 lon fluxes, during muscarinic responses, 175 Ions, role in immune microenvironment, 25 1 Isoguvacine, structural formula, 99

K KA, see Kainate Kainate, 108, 109 neuropathology, 136, 138 structural formula, 110 Kainate receptor, binding studies, 120123

L Learning, anti-brain antibody and, 290292 Limb deafferentation, complete, 22 1-223 Lipids, as antigens in brain, 256-257 Luteinizing hormone opioid control of secretion of, 36-39 sites and mechanisms of action, 33-35 Luteinizing hormone-releasing hormone, 33-35 Lymphoid cells, functions, 250 Lymphokines, in immune environment, 25 1

Magnesium ion, antagonist of excitatory receptors, 109 a-Melanocyte stimulating hormone, 5 secretion control, 45-48 Melanotrophs, in pituitary, 18-20

334

INDEX

Memory, anti-brain antibody and, 290292 Mental retardation, 298 2-Mercaptoethanol, 165 Mesostriatal dopaminergic projection, recovery studies, 233-234 Metal ions, musrarinic receptor antagonists, 165-167 L-Methiorrine-oL-sulfoximirie, structural formula, 111 N-Methyl-maspartate, 108, 109 binding studies, 122 neurotoxicity, 136, 137 structural formula, 110 cic-Methyldioxolane, 157 Mossy fibers, tremors and, 324, 325 Multiple sclerosis, 266-267 Muscarinic receptor affinity labeling, 167-168 agonist binding, 157-158 antagonist binding, 156-157, 158-164 hinding, study methodology, 153- 156 metal ions and, 165-167 presynaptic, 176-177 regulation of, 182-190 sensitization and desensitization, 182185 solubilization of, 190 subpopulations, 177- 182 Muscarinic systems pharmacology, 178- 182 responses of, 168-182 Muscimol, 97, 98 structural formula, 99

N NEM, sec N-Ethylnialeimide Neoendorphin, 7 Neostriatal dopaminergic synapses, recovery studies, 236-238 Nervous tissue explants, autoimmune studies with, 284-285 Neural plasticity, 201-247 axonal growth, 210-223 Neurohypophyseal tract, opioid peptides and, 10-13 Neuroimmune relationship, 252-255 Neuroininiunalogical diseases, anti-hrain antibodies and, 262-269

Neuroimmunology, 255-256 Neuron anti-thymocyte antibody and, 285 axonal growth after injury, 210-223 collateral growth, 2 13-2 17 neurochemical adaptations after injury, 223-238 postsynaptic supersensitivity, 227-232 redundancy in, 203-204 regenerative growth, 210-2 13 role in immune microenvironment, 25 I vicarious functioning, 204-206 Neuropeptides, see ulso Opioid peptides Neuropeptides, in immune environment, 25 1 Neurotoxic antisera, 270-285 Neurotransmitter inactivation mechanisms, 226-227 synthesis and release after injury, 224226 Nicotine, 3 15 NMDA, see N-Methyl-maspartate Nutritional status, P-endorphin and, 24

0 Opioid peptides, see ulso spcczjic peplzde5 cellular origins of, 5-10 functions, 28-32 neurohypophyseal tract and, 10-13 target tissues, 28-32 Opioid peptide receptors, 10, 36, 4 1, 5 1, 55 Opioid peptide system endocrinoIogy of, 1-83 modulation under physiological conditions, 20-28 Organelles, cellular, as antigens in brain, 257 Oxotremorine, 321 Oxytocin, 53-54

P Pain, P-endorphin and, 21-23 Pancreas, opioid regulation of, 56-58 PCMB, sce p-Chloromercuribenzoate PDA, see cis-2,3-Piperidine dicarboxylate Peripheral nervous system immune system and, 252-255 recovery of function in, 2 I 7

335

INDEX

Phenobarbitane, 98 Phosphatidylinositol, turnover of, 168172 Picrotoxin, 97, 98 czs-2,3-Piperidine dicarboxylate, 1 10 structural formula, 110 Pirenzepine, 162 Pituitary anterior lobe corticotrophs, 14- 16 control of secretions by opioids, 2-5, 32-54 intermediate lobe melanotrophs, 18-20 secretion of j3-endorphin, 13-20 Placenta, opioid peptides and, 8, 58 Postsynaptic supersensitivity, 227-232 Pregnancy, @-endorphin and, 23 Prodynorphin, 7 Proenkephalin A, 7 Proenkephalin B, see Prodynorphir. Prolactin, 39-42 Proopiornelanocortin, 5 Protein phosphorylation, 189-190 Purkinje cell, tremors and, 324-327 Pyramidal tract, recovery of function in, 2 18-2 19

QA, see Quisqualate QNB, binding studies, 160-164, 166, 180, 181, 185, 186, 187, 191 Quisqualate, 108 structural formula, 1 I0

Reproductive status, P-endorphin and, 23 Retinotectal projection, recovery of function in, 2 18 Rheumatic carditis, 295 Rheumatic chorea, 295 RRA, see Radioreceptor assays

S Schizophrenia, 295-297 Scopolamine, 170, 184, 186, 187 Secretin, 58 Sensory carcinomatous neuropathy, 294295 Shock, P-endorphin and, 21-23 Sleep, anti-brain antibody and, 288-290 Somatosensory localization, recovery studies, 234-235 Somatostatin, 42, 58 Spinal cord, recovery of function in, 221 Spinal hemisection, 22 1 Spinal serotonergic innervation, recovery studies, 233 S- 100 protein, neuroectodermal, 259-262, 297-300 Stress, opioid peptides and, 21-23, 26, 27 Strychnine, 106- 107 structural formula, 99 Subesophageal ganglion, autoimmune studies with, 274-283 Sulfhydryl reagents, 164-165 Superior cervical ganglion, autoimmune studies on, 282-284 Systemic lupus erythematosus, 295

R T Radioreceptor assays, 92-96 interpretation of binding measurements, 95-96 methodology of, 93-95 uses, 92-93 Receptor assays for, 92-96 as brain antigens, 258 ligand binding, 90-92 properties, 89-90 role in immune mic,roenvironment, 251 Recovery of function, see Brain injury, functional recovery from Redundancy, 203-204 Renin, opioid regulation of, 58

Taurine, 107-108 Thiomuscimol, structural formula, 99 Thyroid stimulating hormone, 44-45 Tremorigenic agents electrophysiological effects, 3 16-323 mechanism of action, 324-327 neurochemical effects, 323-324 Tremorine, 315, 321 Tubulin, as antigens in brain, 257-258

V Vasopressin, 48-53 Vicarious functioning of brain, 204-206

This Page Intentionally Left Blank

CONTENTS OF RECENT VOLUMES Volume 12

Drugs and Body Temperature Peter Lomax Pathobiology of Acute Triethyltin Intoxication R . Torack, ,J. Gordon, and J . Prokop Ascending Control of Thalamic and Cortical Responsiveness M . Steriade Theories of Biological Etiology of Affective Disorders John M . Davis Cerebral Protein Synthesis Inhibitors Block Long-Term Memory Samuel H. Bnrondes

T h e Fine Structural Localization of Biogenic Monoamines in Nervous Tissue Floyd E. Bloom Brain Lesions and Amine Metabolism Robert Y. Moore Morphological and Functional Aspects of Central Monoamine Neurons Kjell Fwe, Tomas Hiikfelt, and Urban Unggerstedt Uptake and Subcellular Localization of Neurotransmitters in the Brain Solomon H . Snyder, Michael J . Knhar, Alan I. Green, Joseph T. Coyle, and Edward G. Shaskan Chemical Mechanisms of Transmitter-Receptor Interaction John T. Garland and Jack D u d

T h e Mechanism of Action of Hallucinogenic Drugs on a Possible Serotonin Receptor in the Brain J . R. Smythies, F. Benington, and R. D. Morin

T h e Chemical Nature of the Receptor Site-A Study in the Stereochemistry of Synaptic Mechanisms J . R. Smythies

Simple Peptides in Brain Isamu Sano

Molecular Mechanisms in Processing Georges Ungar

The Activating Effect of Histamine on the Central Nervous System M . Monnier, R. Sauer, and A . M . Hall Mode of Action of Psychomotor Stimulant Drugs Jacques M . van Rossum AUTHOR INDEX-SUBJECT INDEX

Volume 13

Of Pattern and Place in Dendrites Madge E. Scheibel and Arnold B. ScheihPl

Information

T h e Effect of Increased Functional Activity on the Protein Metabolism of the Nervous System B . Jakoubek and B. Semaginovskj Protein Transport in Neurons Raymond J . Larek Neurochemical Correlates of Behavior M . H . Aprison and J . N . Hingtgen Some Guidelines from System Science for Studying Neural Information Processing Donald 0. Walter and Martin F. Gardiner AUTHOR INDEX-SUBJECT INDEX

337

338

CONTENTS OF RECENT VOLUMES

Volume 14

T h e Pharmacology of Thalamic and Genicd a t e Neurons J . W . Phillis The Axon Reaction: A Review of the Principal Features of Perikaryal Responses to Axon Inquiry A. R. Lieberman C 0 2 Fixation in the Nervous Tissue Sze-Chuh ChPng

Chemistry and Biology of ‘I’wo Proteins, S-100 and 14-3-2, Specific to the Nervous System Blake W . Moore T h e Genesis of the EEG Rafael Elul Marhematical Identification of Brain Slates Applied to Classification of Drugs E. R. John, P . Walker, 11. Cnulood, M . R w h , and J . Gehrmunn AUTHOR INDEX-SUBJECT INDEX

Reflcctions on the Role of Receptor Systems for Taste and Smell J o h n G . Sinclair Central Cholinergic Mcchanisni and Behavior S . N . Pradhan and S. N . Dutta The Chemical Anatomy of Synaptic Mechanisms: Receptors J. R. Smythies AUTITOR INDEX-SUB.JECT INDEX

Volume 15

Projection of Forelimb Group I Muscle Afferents to the Cat Cerebral Col tex lngmur Rosin Physiological Pathways through the Vestibular Nuclei Vzrtor J . Wzkon Tetrodotoxin, Saxitoxin, and Related Substances: Their Applications in Neurobiology Martrn H . Euuru

Volume 16

Model of Molecular Mechanism Able to Generate a Depolariziation-Hyperpolarization Cycle Claru Turdu Antiacetylcholine Drugs: Chemistry, Stereochemistry, and Pharmacology T . D . Inch and R . W . Bnmblecombe Kryptopyrrole and Other Monopyrroles and Molecular Neurobiology Donald G . Irvine RNA Metabolism in the Brain Victor E . Shmhow A Comparison of Cortical Functions in

Man and the Other Primates R . E . Passingham and G . Ettlinger Porphyrid: Theories of Etiology and Treatment H. A. Peters, D.J. C n p p ~and , H . H . Reese SUBJECT INDEX

T h e Inhibitory Action of y-Aminobutyric Acid, A Probable Synaptic Transmitter Kunzhrko Obata

Volume 17

Some Aspects of Protein Metabolism of the Neuron Mei Satake

Epilepsy and y-Aminobutyric Acid-Medicated lnhibition B . S . Meldrum

CONTENlS OF RECENT VOLUMES

339

On Axoplasmic Flow Liliana Lubihcka

Synaptosomal Transport Processes Giulio Levi and Maurizio Raiteri

Schizophrenia: Perchance a Dream? 1. Christian Gillin and Richard,]. Wyatl

Glutathione Metabolism and Some Possible Functions of Glutathione in the Nervous System Marian Orlowski and Abraham Karkowsky

SUBJECT INDEX

Volume 18

Neurochemical Consequences o f Ethanol on the Nervous System Arun K . Rawat

Integrative Properties and Design Principles of Axons Stephen G . W a x m n

Octopamine and Some Related Noncatecholic Amines in Invertebrate Nervous Systems H . A . Robertson and A . V . Juorio

Biological Transmethylation Involving S-Adenosylmethionine: Development of Assay Methods and Implications for Neuropsychiatry Ross J . Baldessarini Synaptochemistry of Acetylcholine Metabolism in a Cholinergic Neuron Bertalan Csillik

Apormorphine: Chemistry, Pharmacology, Biochemistry F. C. Colpaert, W . F. M . Van Bever, and J . E . M . F. Leysen Thymoleptic and Neuroleptic Drug Plasma Levels in Psychiatry: Current Status Thomas B. Cooper, George M . Simpson, and J . Hillary Lee

Ion and Energy Metabolism of the Brain at the Cellular Level Leif Hertz a.nd Arne Schousboe

SUBJECT INDEX

Aggression and Central Neurotransmitters S . N . Pradhan

Volume 20

A Neural Model of Attention, Reinforcement and Discrimination Learning Stephen Grossberg

Functional Metabolism of Brain Phospholipids G. Brian Ansell and Sheila Spanner

Marihuana, Learning, and Memory Ernect I.. Abel

Isolation and Purification of the Nicotine Acetylcholine Receptor and Its Functional Reconstitution into a Membrane Environment Michael S. Briley and Jean-Pierre Changeux

Neurochemical and Neuropharmacological Aspects of Depression B. E . Leonard SUB,JECT INDEX

Volume 19

Do Hippocampal Lesions Produce Amnesia in Animals? Susan D. l v ~ r s ~ n

Biochemical Aspects of' Neurotransmission in the Developing Brain Jospph T. Coyle T h e Formation, Degradation, and Function of Cyclic Nucleotides in the Nervous System John W . Daly Fluctuation Analysis in Neurobiology Louic J . DeFelice

340

CONTENTS O F RECENT VO1,UMES

Peptides and Behavior Georges Ungar Biochemical Transfer of Acquired Inforrnatiori S . K . Mitchell, J . M . Beaton, and R. J . Bradley Aniinoti-ansferase Activity in Brain M . Benuck and A. Lajtha l h e Molecular Structure of Acetylcholine and Adrenergic Receptors: An All-Protein Modcl I . R.Smythies Structural Integration of Neuroprotease Activity Elena Gabnelrscu Lipotropin and the Central Nervous System W . H . Gispen, J . M . van Ree, and D. dr Wied

Presynaptic Inhibition: Transmitter and Ionic Mechanisms R. A. Nicoll and A. E. Alger Microquantitation of Neurotransmitters in Specific Areas of the Central Nervous System l u a n M . Saavedra Physiology and Glia: Glial-Neuronal Interactions R. Malcolm Stewart and Roger N . RoJenberg Molecular Perspectives of Monovalent Cation Selective Transmembrane Channels Dan W . Urty Neuroleptics and Brain Self-stimulation Behavior Albert Wauquier

Volume

l'issue Fractionation in Neurobiochemistry: An Analytical Tool or a Source of Artifacts Pierre Laduron Choline Acetyltransferase: A Review with Special Reference to Its Cellular and Subcellular Localization Jean Rossirr

22

Transport and Metabolism of Glutamate and GABA in Neurons and Glial Cells Arne Schousboe Brain Intermediary Metabolism in V i m Changes with Carbon Dioxide, Development, and Seizures Alexander I.. Miller

SUBJECT INDEX

Volume 21

Relationship of the Actions of Neuroleptic Drugs to the Pathophysiology of Tardive Dyskinesia Ross J . Baldeuanni and DanieL Tarsy Soviet Literature on the Nervous Systeni and Psychobiology of Cetacea Thedorr H. Bullock and Vladimir S. Gurevich Binding and lontophoretic Studies on Centrally Active Amino Acids-A Search f o r Physiological Receptors F. V . DeFeudk

N,N-Dimethyltryptarnine: An Endogenous Hallucinogen Steven A. Barker, John A . Monti, and Samurl T. Christian Neurotransmitter Receptors: Neuroanatomical Localization through AutoradioWPhY L. Charles Murrin Neurotoxins as Tools in Neurobiology E . G . McGeer and P . L. McCeer Mechanisms of Synaptic Modulation William Shain and David 0. Carpenter Anatomical, Physiological, and Behavioral Aspects of Olfactory Bulbectomy in the Rat B . E. Leonard and M . Tuite

CONTENTS OF KECENT VOLUMES

The Deoxyglucose Method for the Measurement of Local Glucose Utilization and the Mapping of Local Functional Activity i n the Central Nervous System Louis Sokoloff

34 1

Sleep Mechanisms: Biology and Control of KEM Sleep Dennis J . McCinty arid RpnC R . Drucker-Colin INDEX

INDEX

Volume 24 Volume 23

Chemically Induced I o n Channels in Nerve Cell Membranes David A. Mathers and JeJury L. Barker Fluctuation of Na and K Currents in Excitable Membranes Berthold Neunicke Biochemical Studies of the Excitable Menibrane Sodium Channel Robert L. Barchi Benzodiazepine Receptors in the Central Nervous System Phil Skoolnlck mad Stuven M. Pnul Rapid Changes in Phospholipid Metabolism during Secretion and Receptor Activation F . T. Crewr

Antiacetylcholine Receptor Antibodies and Myasthenia Gravis Bernard W.Ful,9iw l’harniacology of Barbiturates: Electrophysiological and Neurocheinical Studies Mox Willow c4nd Graham A. R . Johrwton Inimunodetection of Endorphins and Enkephalins: A Search for Reliability Alejnndro Bayon, William J . Shoemaker, Jocqueline F. McCinty, and Floyd Bloom On the Sacred Disease: T h e Neurochemistry of Epilepsy 0.Carter Snmd III Biochemical and Electropliysiological Characteristics o f Mammalian GABA Receptors Salvcitore J . Enna and Joel P. Gallugher

Glucocorticoid Effects on (kntral Nervous Excitability and Synaptic Transmission Edward D . Hall

Synaptic Mechanisms and Circuitry Involved in Motoneuron Control during Sleep Michael H . Chcise

Assessing the Functional Significance of Lcsion-Induced Neuronal Plasticity Oswald Steward

Kecent Developments in the Structure and Function of the Acetylcholine Receptor F . J . Barrantes

(:haracterization of a,-and a2-Adrenergic Dopamine Receptors in the Central NcrKeceptors vous System Dnvid B. Hylund and David C . U’Prichard Ian Cresse, A. Leslie Morrow, Stuart E . I.
342

CONTENTS OF KECENT VOLUMES

Volume 25

Guanethidine-Induced Destruction Sympathetic Neurons Eugene M . Joh71so7~,Jr., and Pamela Toy Manning

of

Dental Sensory Receptors Margaret R. Byers Cerebrospinal Fluid Proteins in Neurology A . Lowenthal, R. Crols, E. De Schulter, J . Gheuens, D. Karcher, M . Noppe, and A . Tasnier Muscarinic Rcceptors in the Central Nervous System Mordechai Sokolovsky l’eptides and Nociception Daniel Luttanger, Daniel E . Hernandez, Charles B . Nemerofi, and Arthur J . Prange, J r .

Opioid Actions on Mammalian Spinal Neurons W . Zieglgansberger Psychobiology of Opioids Albert0 Oliveria, Claudio Cmtellano, and Stefano Publisi-Allegra Hippocampal Damage: Effects on Dopaminergic Systems of the Basal Ganglia Robert L . Isaacson Neurochemical Genetics v. cscinyi The Neurobiology of Some Dimensions of Personality M a n m Zuckerman. James C. Rnllenger, and Robert M . Post INDEX