International Review of Neurobiology Volume 4

INTERNATIONAL REVIEW OF

Neurobiology VOLUME 4

This Page Intentionally Left Blank

INTERNATIONAL REVIEW OF

Neurobiology Edited

by CARL C. PFEIFFER New Jersey Psychiatric Institute Princeton, New Jersey

J'OHN R. SMYTHIES Department of Psychological Medicine University of Edinburgh, Edinburgh, Scotland

Associate Editors V. Amassian J. A. Bain D. Bovet Sir Russell Brain Sir John Eccles

VOLUME

E. V. Evarts H. J. Eysenck F. Georgi G. W. Harris R. G. Heath

C. Hebb A. Hoffer

K. Killam S. Marten

4 1962

ACADEMIC PRESS

New York and London

Copyright0 1962 by Academic Press Inc. A L L RIGHTS RESERVED

N O PAHT OF THIS BOOK MAY B E REPRODUCED I N ANY FORM, BY PHOTOSTAT, MICROFILM, OR ANY OTHER MEANS, WITHOUT WRITTEN PERMISSION FROM T H E PUBLISHERS.

ACADEMIC PRESS INC. 111 FIFTHAVENUE NEW YORK 3, N. Y. United Kingdom Edition Published by ACADEMIC PRESS INC. (LONDON) LTD. BERKELEYSQUAREHOUSE,BERKELEYSQUARE,LONDON W. 1

Library of Congress Catalog Cord Number 59-13822

PRINTED I N T H E UNITED STATES OF AMERICA

CONTRIBUTORS L. G. ABOOD,Department of Psychiatry and Biological Chemistry, University of Illinois, College of Medicine, Chicago, Illinois

J. H. BIEL, Department of Psychiatry and Biological Chemistry, University of Illinois, College of Medicine, Chicago, Illinois A. BROSSI,Research Departments of F . Hofmann-La Roche G Co. Ltd., Basel, Switzerland K. F. GEY,Research Departments of F . Hofmann-La Roche G Co. Ltd ., Basel, Switzerland WILLIAMINA A. HIMWICH,Thudichum Psychiatric Research Laboratory, Galesburg State Research Hospital, Galesburg, Illinois A. HOFFER,Psychiatric Services Branch, Department of Public Health, and Department of Psychiatry, University of Saskatchewan, Saskatoon, Saskatchewan, Canada WERNERP. KOELLA, Worcester Foundation for Experimental Biology, Shrewsbury, Massachusetts

* F. LEMBECK,Department

of

Pharmacology, University of Graz,

Austria SIDNEYOCHS,Department of Physiology, Indiana University School of Medicine, Indianapolis, Indiana A. PLETSCHER, Research Departments of F . Hofmann-La Roche G Co. Ltd., Basel, Switzerland G. ZETLER, Deparbment of Pharmacology, University of Kiel, Germany

* Present address: Department of Pharmacology, University of Tubingen, Germany. V

This Page Intentionally Left Blank

PREFACE Volume 4 of the International Review of Neurobiology is now before you. The editors have continued to solicit reviews from each discipline which deals with the working of the nervous system. This is to them neurobiology in its broadest sense. Cross-fertilization of ideas is of utmost importance in any constructive approach to the many unsolved problems in this field. Techniques which may be known to the behavioral scientist may be clarified by the refined techniques of the anatomist working with tissue cultures. Any drug which has a selective locus of action may be profitably used by research workers in any discipline. Advances in basic biochemistry, psychopharmacology, or biophysics thus may give rise to useful developments in the clinical field, but mature judgement is required to select, from the vast detail of basic information, those specific bits which may apply to human disease. The review authors are supplying this mature judgement. The popularity of this approach is indicated by the necessity to publish an additional volume. This has been accomplished by the acceptance of two groups of manuscripts in the year 1962 with deadlines of June 1 and October 1 for Volumes 5 and 6, respectively. Hereafter the annual deadline for receipt of manuscripts will be October 1 of each year; Volume 7 will have a deadline of October 1, 1963. These reviews and summaries ordinarily are written by invitation. The editors, however, will be happy to review unsolicited manuscripts if these are submitted in complete or outline form. CARLC. PFEIFTER JOHNR. SMYTHIES

Spring 1962

This Page Intentionally Left Blank

CONTENTS CONTRIBUTORS PREFACE

.

.

.

. .

.

.

. .

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

V

vii

The Nature of Spreading Depression in Neural Networks SIDNEYOCHS I. 11. 111. IV. V. VI . VII. VIII.

Introduction . . . . . . . . . Phenomenological Aspects of Spreading Depression . Excitation of Spreading Depression . . . . Propagation . . . . . . . . . Cellular Changes during Spreading Depression . . Upper Cortical Layers and Release of Lower Layers . Spreading Depression and Higher Functions in Cortex . . . . . . . . . Conclusions References . . . . . . . . . .

.

.

.

.

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

2 3 14 22 33 47 56 63 65

Organizational Aspects of Some Subcortical Motor Areas WERNERP. KOELLA I. 11. 111. IV. V. VI.

Introduction: Fundamentals of Motor Organization . . Motor Effects Produced by Stimulation of the Diencephalon Motor Effects Produced by Stimulation of the Cerebellum . The Midbrain Tectum . . . . . . . . The Tegmental Reaction . . . . . . . . Discussion and Interpretation . . . . . . . References . . . . . . . . . . .

.

. . . . . .

71 76 86 96 100 101 114

Biochemical and Neurophysiological Development of the Brain in the Neonatal Period WILLIAMINA A. HIMWICH I. 11. 111. IV. V. VI.

Introduction . . . . . . Inherent Problems . . . . . Genesis of Behavior . . . . . Neuroanatomical Development . . Accumulation of Chemical Constituents . Development of Enzymatic Activity . ix

. .

. .

. .

. .

.

.

,

.

. . . .

.

.

.

. . . . .

. .

. . . .

117 119 122 125 129 138

X

CONTENTS

VII . Hematoencephalic Exchange in the Developing Brain . . . VIII . Neurophysiological Development . . . . . . . IX . Correlation of Anatomical. Chemical and Neurophysiological Factors . . . . . . . . . . . . X. Conclusions . . . . . . . . . . . References . . . . . . . . . . . .

141 142 147 154 155

Substance P: A Polypeptide of Possible Physiological Significance. Especially Within the Nervous System F. LEMBECK A N D G. ZELTER I. I1. 111. IV.

V. VI . VII .

Introduction . . . . . . . . . . . Chemical Characteristics of Substance P . . . . . Distribution of Substance P in the Organism . . . . . Relationship between Organ Function and Tissue Concentration of Substance P . . . . . . . . . . . Pharmacological Actions of Substance P . . . . . . Pharmacological Interactions with Drugs . . . . . Conclusions . . . . . . . . . . . References . . . . . . . . . . . .

160 161 170 183 191 205 209 210

Anticholinergic Psychotomimetic Agents L . G. ABOOD AND J . H. BIEL I. I1. 111. IV . V. VI . VII . ‘7111 . IX .

X. XI .

Introduction . . . . . . . . . . . The Concept of a Drug Receptor Site . . . . . . Chemical Nature of the Psychotomimetic Anticholinergic Agents . . . . . . . . . Experimental Methods Structure-Activity Relationships . . . . . . . Some Physical and Chemical Factors in Drug Action . . . Studies with Radioactive Labeled Piperidyl Glycolates . . Behavioral Effect of Anticholinergic Psychotomimetics . . Investigations into the Central Mechanism of Action of the . . . . . . . . . Piperidyl Glycolates Clinical Studies with the Glycolate Esters . . . . . Biochemical and Electrophysiological Studies with Piperidyl . . . . . . . . . . . Glycolates . References . . . . . . . . . . . .

218 219 221 221 226 232 241 244 249 256 260 271

Benzoquinolizine Derivatives: A N e w Class of Monamine Decreasing Drugs with Psychotropic Action A. BROSSI.AND K . F. GEY A. PLETSCHER. I . Introduction I1. Chemistry .

. .

. .

. .

. .

. .

. .

. .

. .

. .

. .

. .

273 277

xi

CON TENTS

111. IV . V. VI . VII

.

VIII .

Metabolism . . . . . . . . . . . I d u e n c e on Monamine Metabolism . . . . . . Pharmacology . . . . . . . . . . . Clinical Action . . . . . . . . . . . Relationship between Effects on Monamine Metabolism and Pharmacological Action . . . . . . . . . Summary . . . . . . . . . . . . . . . . . . . . . . . References .

281 282 291 300 300 302 303

The Effect of Adrenochrome and Adrenolutin on the Behavior of Animals and the Psychology of M a n A . HOFFER I . Introduction . . . . . . I1. Biochemistry of Adrenochromc . . 111. Action of Adrenochrome on Cells . . IV. Effect of Adrenochrome on Fish . . V. Effect of Adrenochrome on Spiders . VI . Effect of Adrenochrome on Pigeons . VII . Effect of Adrenochrome on Mammals . VIII . Effect of Adrenochrome on Electrograms IX . Effect of Adrenochrome and Adrenolutin on X . Mode of Action of Adrenochrome . . XI . Conclusions . . . . . . References . . . . . . . AUTHOR INDEX

SUBJECT INDEX

. .

. .

. .

. .

. .

. .

. .

.

.

.

.

.

. . . .

. . . .

. . . .

. . . .

. . . .

. .

. .

. .

. .

. .

.

.

.

.

.

Humans .

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

. .

. .

. .

. .

. .

307 311 314 315 316 316 320 335 337 361 365 365 373 386

This Page Intentionally Left Blank

THE NATURE OF SPREADING DEPRESSllON IN NEURAL NETWORKS By Sidney Ochs Department of Physiology. Indiana University School of Medicine. Indianapolis. Indiana

I. Introduction . . . . . . . . . . . . . . I1. Phenomenological aspects of Spreading Depression . . A . EEG Changes . . . . . . . . . . . . B . Steady-Potential Changes . . . . . . . . . C . Cortical Responses . . . . . . . . . . . D . Electrical Impedance . . . . . . . . . . E . Concomitant Vascular Changes . . . . . . . F. Subcortical Effects . . . . . . . . . . . G . Spreading Depression in Nonneocortical Tissue . . . H . Asphyxia1 Change . . . . . . . . . . . I11. Excitation of Spreading Depression . . . . . . . A . Modes of Excitation . . . . . . . . . . B . Spreading Depression in Chronic Preparations . . . C. Modification of Spreading Depression Excitability . . D . Refractoriness and Recovery Time . . . . . . IV. Propagation . . . . . . . . . . . . . . A . Characteristics of Propagation . . . . . . . . B . Mechanism of Spread . . . . . . . . . . C. Contiguity Theory of Transmission . . . . . . V . Cellular Changes during Spreading Depression . . . . A . Entry of Ions and Water into Cells . . . . . . B . Release of Interacellular Components . . . . . . C. Intercellular Space and Glia . . . . . . . . D . Metabolic events connected with Spreading Depression VI . Upper Cortical Layers and Release of Lower Layers . . A . Differential effect on Direct Cortical Response . . . B . Spreading Convulsion . . . . . . . . . . VII . Spreading Depression and Higher Functions in Cortex . . A . Effect of Spreading Depression on Conditioned Responses B . Transfer of a Learning Trace . . . . . . . . C. The Cortex and Its Subcortical Relations . . . . VIII . Conclusions . . . . . . . . . . . . . . References . . . . . . . . . . . . . . 1

. .

. . .

.

.

.

. . . . . . . . . .

. . . . . . . . .

.

. . . . . . . . . . .

.

. . . .

2 3 3 5 7 7 8 9 11 12 14 14 17 19 20 22 22 23 27 33 33 39 41 44 47 47 52 55 55 60 62 62 64

2

SIDNEY OCHS

I. Introduction

Since the discovery of spreading depression (SD) by Le5o ( 1944a), it has been intensively studied in a number of laboratories. Part of the fascination of SD is the slowness of its spread through the cortex. The rate of 2 to 6 mm/min has seemed too slow to be readily accounted for on the basis of known types of nervous conduction. Spreading depression was included by Roitbak (1955) in his book on electrophysiology of the cerebral cortex; Marshall's (1959) review of spreading depression is a comprehensive guide to the subject through 1958. Since then, new developments have clarified some long-standing conflicts concerning the basic cellular mechanisms involved. Spreading depression appears to be a neuronal reaction of the apical dendrites of pyramidal cells that is triggered by a component released from depressed neurons which in turn excites contiguous neurons. In the depressed cells an increased permeability occurs with a shift of electrolytes and water into the apical dendrites. Concomitant metabolic changes are found, in part connected with energetic requirements for reversion to the normal state. (We shall be concerned later with the evidence for this summary statement.) Fundamental aspects of SD will be dealt with in Sections I1 and 111. Emphasis will be placed on cellular mechanisms as follows: factors in propagation, Section IV; molecular changes within cells, Section V; and events occurring between parts of the cortex (particularly with respect to the closely related phenomenon, spreading convulsion), Section VI. Some special difficulties are associated with the study of SD. The phenomenon is of long duration, and the long-lasting changes induced in the tissues are difficult to handle experimentally. It is not surprising that SD has been unrecognized or variously described. Sloan and Jasper (1950) compared SD with suppression which was believed to originate from special suppressor strip areas (Fulton, 1955), and they concluded that suppression and SD are the same phenomenon. Difficulty in confirming the presence of suppressor strips has convinced most experimenters that suppressions of cortical activity are SD's and not suppression elicited from special suppressor-strip regions (Marshall, 1959). To those familiar with the phenomenon, the literature contains records which may represent unrecognized instances of SD. For this reason alone when working

SPREADING DEPRESSION IN NEURAL NETWORKS

3

on the cerebral cortex it is important to be alert to recognize SD when it appears. Empirical precautions against cortical damage probably have their basis in an obscure awareness of the prolonged changes in excitability produced by SD's. Some part of the difficulty experienced in recognizing SD is related to differences in the readiness with which it may be elicited in the various species. Of the common experimental animals it is regularly evoked in the rat and the rabbit, and has been observed in mouse, guinea pig, porcupine, marmoset, pigeon, sloth, macaque, and possibly also in man. A relationship between the computed movement of the cortical disturbance underlying a migrain escotoma and the slow rate of movement of SD was noted by Milner (1958) and the similarity between the rate of SD movement and that of Jacksonian March was also pointed out ( Marshall, 1959). Recently, SD has been used along with conditioned responses to study the higher functions of the cortex and to give another index of the effect of SD on cortical processes. The use of SD applied to studies of behavior has opened a new area of research. Localization of learning traces in one cortical hemisphere and even to parts of a cerebral cortex is accomplished with the aid of SD. These investigations and related ones concerning the functional connection of the hemispheres to one another and to subcortical structures will be discussed in Section VII. I I . Phenomenological Aspects of Spreading Depression

A. EEG CHANGES As first shown by Leg0 ( 1944a),mechanical, chemical, or electrical stimulation of the exposed cortex may induce SD. The appearance of SD is best shown with a series of bipolar recording electrodes set at successively greater distances from the stimulated site (Fig. 1). After stimulation, a latency occurs lasting a few minutes, and then the EEG is seen to gradually decrease in amplitude in the closest pair of recording electrodes and to remain depressed in amplitude for several minutes. The movement of SD in the cortex is observed by the successive decline of activity first in the electrode pair closest to the point stimulated to give SD; then with a delay corresponding to its slow rate of progression, there is a diminution in the next channel's record of EEG activity, etc. Within 3 or 4 minutes the ampli-

I

n

I.

t---Xz

C

I

D

I

E

I

F

H

Sm 0s

FIG. 1. Spre I of depression. In the inset to the right the rabbit's hemisphere is shown with recording electrodes placed in iine on its pia-arachnoid surface. S refers to the electrodes used to excite SD, and the electrode numbers shown on the brain correspond with the connections for each channel of EEG shown in different rows. Times in minutes and seconds shown in each lettered column of EEG samples. Soon after control EEG and stimulation ( A ) , the EEG is depressed first in channel 1-2 ( B ) , later in channel 23 further outward ( C ) ,etc. Recovery occurs gradually in the same order. Calibration in column G, vertical line 1 mv, horizontal line 1 sec. (From LePo, 1944a.)

SPREADING DEPRESSION IN NEURAL NETWORKS

5

tudes of the EEG gradually recover from their depression to attain control levels. Often, several large spike discharges are seen at the onset and during the depressed phase. These convulsive-like spikes occasionally may be so prominent as to replace the depression usually seen ( see Section VI, B ) . The spread is roughly similar to the outward ripples of a water wave, although irregularities are found at the advancing edge. If monopolar instead of bipolar electrodes are used, too wide an area is encompassed and the narrow region of passing depression may be missed (van Harreveld and Stamm, 1951).For this reason attempts to record SD by scalp electrodes through the skull have failed to show passage of SD in the underlying brain.

B. STEADYPOTENTIAL CHANGES A DC amplifier and appropriate nonpolarizable electrodes are used to record the steady potential (SP) of the brain. This DC potential is related to the polarization state of the uppermost part of the apical dendrites of pyramidal cells with relation to lower parts of the cell (Gerard and Libet, 1940; OLeary and Goldring, 1959). The SP of the cortex shows a characteristic variation when SD occupies the cortex under such an electrode ( LeHo, 1947). When SD enters an area of recording, the surface becomes negative in potential, and after 1 to 2 minutes the negative phase is succeeded by a smaller and longer-lasting positive phase before a return to the original level of resting potential (Fig. 2). The negative phase of the SP change is relatively large, with voltages of 5 to 15 mv usually recorded. Variants may occur in the pattern of depolarization. A prolonged negativity may be found that looks like a double negative phase, or the positive phase may be as great as, or even occasionally greater than, the negative phase. The most frequent type found is the pattern shown in Figs. 2 and 4.The relatively large .amplitude of the negative swing indicates that many of the apical dendrites are synchronously depolarized. The SP records the activity of the apical dendrites found in great concentration in the molecular layer of the cortex. The negative variation of the SP likely reflects a differential depolarization greater in the upper apical dendritic portions than in the lower dendritic and somatic portions of pyramidal cells. The smaller positive phase which occurs after the negative phase may be a positive electrogenesis of the apical dendrites, but more likely it

1

I....., ~ ~ l- - u.....-. l ? : . - r i l . l l ~ ~ . ..___._....___ *liOilid.nUl~~~Yi~~IJ.i.uliVL+a.r*.

-

-21

.---....-

--.--- ..... ........ ......____. -- .. ....

8'

c'

s

A

-29

...___ ....... ..

-

pBA-c

FIG.2. EEG, SP, and impedance change during SD. Traces from above down are: ( A ) the EEG, ( B ) steady potential (SP), ( C ) electrical conductance of the cortex. The insert shows the electrodes arrangement. At S the cortex is stimulated and after a latency of several minutes the EEG is diminished. SP changes in negative direction and at the same time conductance decreases. Return to control levels is indicated in the continuation of traces A', B', C' below. The negative variation of the SP is typically followed by a smaller positive variation shown below the dotted line with a slow return. Horizontal line 10 sec. Left vertical line 1 mv EEG, right vertical line 5 mv SP. (From van Harreveld and Ochs, 1957.)

SPREADING DEPRESSION I N NEURAL NETWORKS

7

reflects the presence of depolarization from the deeper layers of the cortex. Using semimicroelectrodes and recording SP changes from different depths within the cortex, LeHo (1951) found that the negative phase appears at successively later times when recording from deeper parts of the cortex. There is, therefore, a slow movement down into the depth of the cortex, which stops when the white matter is reached. C. CORTICAL~ P O N S E S Cortical responses evoked in a primary sensory area of the cortex as well as responses elicited by direct stimulation of the surface [direct cortical response ( DCR) ] are both depressed during SD. The decrease in DCRs during SD is shown in Figs. 12-14. The sensorily evoked responses are typically diphasic with a slow positive phase followed by a slow negative phase. The first of the smaller spikelike fast waves seen inscribed on the positive phase is due to activity of the entering afferent fibers ending within the cortex; the later fast waves are possibly due to intracortical cell discharge (Bishop and Clare, 1953). As SD invades an area of the cortex giving rise to an evoked response, the slow waves and then the later fast waves become depressed. Meanwhile the first fast wave representing activity in the afferent fibers is only briefly depressed. The significance of this temporary depression of the first fast wave will be discussed in Section IV. The slow negative component of the sensorily evoked response represents a discharge of the apical dendrites, and this .part of the response is most deeply depressed. The simple negative-wave DCR is also a response of apical dendrites and is depressed by SD (Grafstein, 1956a; Ochs, 1958). D. ELECTRICAL IMPEDANCE Electrical impedance measurements of the functioning cortex had not been accorded much interest until Leiio and Martins Ferreira (1953) showed a characteristic increase in the electrical impedance of the cortex occupied by SD. The increase of impedance found during SD is plotted as the inversely related decrease in conductance in the example given in Fig. 2. The measurements were recorded via electrodes placed on the surface of the brain using a bridge technique (van Harreveld and Ochs, 1956,1957). The bridge was balanced to a null using a frequency of lo00 cycles/sec, and

8

SIDNEY OCHS

changes in impedance during SD or cortical asphyxiation were measured by rebalance to the null or by the use of a rectifier and pen-recording technique. Fundamental work on measurements of cells in a conducting medium (Cole, 1940; Cole and Curtis, 1944) forms the basis for the interpretation of resistivity measurements and changes of resistivity of the cortex during SD. The cell membrane is characterized by resistive and capacitative elements. In a cell suspension at 1OOO cycles/sec, the measuring current passes mainly between the cells via the intercellular medium. This path is resistive, and the degree of resistivity is related to the electrolyte concentration of the intercellular space and its comparative volume with respect to the total volume of the tissue. Increasing the volume of the cells in the suspension relative to the whole volume decreases the intercellular compartment and the path available to current flow. Therefore, the resistance measured is greater. We shall refer to the evidence in more detail in Section V, but briefly state, the 17%increase in electrical resistance observed during SD is due to the entrance of NaCl and water from the intercellular compartment into the cells. It is the loss of electrolytes from the intercellular compartment that results in rise of resistivity. Another technique of impedance measurement was developed by Freygang and Landau (1955). A pulse of current was applied to the cortex via a surface electrode with the indifferent electrode placed elsewhere, or impedance was measured with current pulses passed through a pair of semi-microelectrodes inserted across a small thickness of cortex. They found a 10-205E: increase resistance during SD, which was most marked in the upper cortical layers. Under normal conditions the resistivity measured 220 ohm-cm, a value consistent with an intercellular compartment size of 2530%( Section V, C ) .

E. CONCOMITANT VASCULAR CHANGES LeHo (1944b) observed that the pial surface vessels undergo a marked dilation during SD, and he suggested that vessels are the medium of propagation of SD. A decreased amount of 0, during SD was found in blood samples taken from the sagittal sinus and by microspectrophotometic measurements of the surface ( LeHo and Morison, 1945). This evidence seemed to favor the concept of vessels as the medium of propagation. Volume changes of the cortex during the passage of a wave of SD also seemed to support the idea that vessels were involved (van Harreveld and Stamm, 1952). Such

SPREADING DEPRESSION I N NEURAL NETWORKS

9

volume changes, recorded optically by means of a small mirror placed on the surface, indicated vasoconstriction and at the same time showed a decrease in 0, pressure. A possible asphyxia1 change brought about by the constricted vessels was suggested. However, direct examination of the intracortical vessels showed that vasoconstriction was too brief to play a causal role (van Harreveld and Ochs, 1957). The intracortical vessels and changes in their diameter could be seen by means of a technique of fast-freeze and substitution-fixation. Ethyl alcohol cooled almost to its freezing point with liquid air was poured over the exposed surface to quick-freeze the brain; the frozen brain was then removed and placed in ethyl alcohol at -2OOC. Substitution-fixation takes place over a period of 3 days with the alcohol slowly entering the solid tissue, and as it enters a region the ice dissolves. As the alcohol passes deeper into the frozen tissue, cells are fixed without dislocative movements of water. By this means cell contents and their relations to one another retain their original positions. The tissue can then be handled with the usual histological techniques. A benzedine stain was used to stain blood cells in the vessels and thereby show variations in the diameter of the blood vessels of the cortex. The vessels appear in these sections as vertical stalks with smaller branches to the side supplying the parenchyma. With an SD wave caught in passage, a brief period of vessel constriction developed in the rabbit cortex simultaneously with the SP change, followed by a vasodilation in its wake. Only vasodilation accompanies SD in the cat cortex. Evidence for a dilation during SD was found by Freygang et al. (1954) and by Sonnenschein and Walker (1956) using a heated thermocouple inserted into the brain. A surface thermocouple technique was used by Bure;OVA (1957b) in her studies of vascular changes during SD, and she also found an increased blood flow. The vasodilation occurred 75 +- 15 seconds after the onset of the SP change of SD which is further evidence that the blood flow increase is a concomitant event and not a causal factor.

F. SUBCORTICAL EFFECTS When SD moves through the cortex, other parts of the brain neuronally connected with the cortex are secondarily involved. Winokur et al. (1950) recorded spontaneous activity from various thalamic nuclei and found alterations of EEG activity in those parts of the thalamus connected to the cortical region undergoing SD.

10

SIDNEY OCHS

Sloan and Jasper (1950) also reported decreased spontaneous activity in thalamic nuclei connected to that part of the cortex undergoing SD. The effect of SD’s on spontaneous EEG activity in the thalamus is most likely exerted primarily on the novspecific parts of the thalamus. The primary responses evoked in the thalamus by sciatic nerve stimulation was not depressed during SD (Weiss and Fifkovh, 1961) . While these primary responses were unaffected, Weiss (1961b) confirmed previous findings that spontaneous EEG activity recorded in the thalamus is depressed during SD. A further finding was that the spontaneous EEG activity recorded in the mesencephalic reticular formation was only infrequently depressed, in curarized animals. This finding suggests that the nonspecific thalamic regions are more closely connected with the cortex, at least insofar as spontaneous EEG activity is concerned. When barbiturate was administered, depression of mesencephalic EEG activity was found during SD’s. Other types of measurements show an effect of cortical SD’s on mesencephalic activity. Using a single-cell recording technique BureZ et al. (1961) reported that cortical S D s are associated mostly with increases in the rate of discharge of bulbopontine reticular formation units. A smaller proportion of units showed a decrease in rate or a longer-lasting increase, an effect found on both sides of the brain stem ( BureZ, 1959).A short-lasting desynchronization of the EEG in the cortex opposite to the one depressed was found (Weiss, 1961a). This crossed desynchronization was secondary to reticular formation involvement as indicated by the absence of this crossed effect when the brain stem was interrupted at the collicular level or when animals were narcotized with barbiturate. When SD moves into the sensorimotor area, it causes an elevation in the threshold of electrically evoked motor responses (Sloan and Jasper, 1950). Yet, the effect of SD in this area causes little evidence of postural or motor performance in the free-moving animal (BureZ and BureZovh, 196Ob; Ochs et al., 1961). Suppression is probably SD (Section I), and the excitability of the motor area to stimulation was reported as diminished during suppression, while knee-jerk reflexes were unaltered (Dusser de Barenne and McCulloch, 1941). During extinction (which also, in its long duration of effect, is similar to SD) a depression of knee jerks was found (Dusser de Barenne and McCulloch, 1939). Differences in motor effect

SPREADING DEPRESSION IN NEURAL NETWORKS

11

may be caused by different involvement of excitatory or inhibitory corticifugal neurons. Within the sensorimotor cortex, inhibitory regions can be found which inhibit knee-jerk reflexes (Ochs, 1955a). The path of this inhibitory region can be traced down into lower motor centers ( Hankinson et al.,1955). These regions are most likely a part of an integrative excitatory and inhibitory corticifugal system, the type of stimulation used determining excitatory or inhibitory effects. Possibly SD results in inhibition of lower motor centers; however, in chronic SD experiments (Section 111, B) peripheral motor effects were minimal. We should expect other subcortical regions directly interconnected with the cortex to show signs of release or other evidence of changed cortical control when SD occupies the part of the cortex connected to those regions. BureHovh (1957a) reported an action of SD on the hypothalamic-pituitary system. In rats with a water load, SD caused a decreased urinary output and, as indicated by creatinine clearance studies, an increased tubular reabsorption. This defect in urinary excretion was not seen in hypophysectomized animals and the inference made was that the hypothalamus is excited during SD to produce an increased antidiuretic hormone output from the pituitary. An effect on hypothalamic function was also indicated by the greater drop of body temperature in animals with SD in a cool environment as compared to a nondepressed group of rats (BureH and BureHovh, 1956). Weiss and Fifkovfi (1960) studied the effect of SD induced in the neocortex in relation to a possible spread to a limbic brain structure (hippocampus) and vice versa. They found that when SD was elicited in the neocortex, the dorsal hippocampus activity was unchanged with respect to spontaneous activity and externally induced synchronized activity. When SD was induced in the hippocampus, the spontaneous or sensorily evoked neocortical activity remained unaltered. These regions are independent as far as EEG activity is concerned. Hippocampal SD’s also did not affect the activity of bulbopontine reticular formation units in most of the units recorded (BureS et al., 1961).

G . SPREADING DEPRFSSION IN NONNEOCORTICAL TISSUE The neocortex of mammals has been investigated most extensively, but SD has been found in other neural tissues, Spreading depression can occur in the archipallium and in pigeon striatum, as

12

SIDNEY OCHS

first shown by LeHo (1944a). Martins Ferreira and LeHo (1958) reported the phenomenon in the cortex of alligators and in the caudate of mammals. The hippocampus appears to have as low a threshold for SD as the neocortex has (Bure;, 1959). Spreading depression in the pigeon striatum was studied in detail by Bure; et al. ( 1 9 6 0 ~ )The . threshold for SD was higher than in the neocortex of the rat, and failure of complete invasion to all parts of the surface was often noted. Unlike the mammalian cortex, the striatum is not limited to a relatively thin layer by an underlying compact layer of white matter. Recordings taken with microelectrodes at l-mm steps from successively deeper stations in the striatum showed that SD spreads successively downward to a depth of 5mm. The spread appears to proceed in the downward direction with the same slow velocity as on the surface. The size of the SP response in the depths is large, with amplitudes as great as 22 mv, and the shape of the SP change in the depths is similar to that found in surface recordings except for a relatively smaller positive phase. A slow-moving reaction that has the characteristics of SD has been found in the retina (Gouras, 1958). This phenomenon occurs periodically, appearing as a misty, grayish wave spontaneously starting from the cut edge of the tissue and spreading throughout the whole retina at a rate of 1to 2 mm/min. During the presence of this color change spontaneous electrical activity first increases and then is profoundly depressed so that intense stimulation cannot excite a response from these neural elements. The depression lasts 3 to 5 minutes, followed by gradual return of normal color and excitability. An SP negative change of 1 to 2 mv accompanies the spread. This remarkable observation was also reported by Gouras (1958) to have been seen by Hartline. Lipetz (commenting on Gouras’ paper) ruled out the special participation of nonneuronal pigment cells by his observation of the phenomenon in frog retinas lacking pigment epithelium. Presumably any neural tissue with sufficiently complex organization and density of elements can support the reaction.

H. ASPHYXIALCHANGE A l-minute period of anemia or asphyxiation by itself will not initiate an SD (LeHo, 1947; van Harreveld and Stamm, 1953c) although it will prolong the negative SP change of an SD which is already present ( LeHo, 1947). A prolonged period of asphyxiation

SPREADING DEPRESSION IN NEURAL NETWORKS

13

will produce, after a latency of 2.5 to 5 minutes, a depolarization and an accompanying electrical impedance increase (Fig. 3 ) that

~~~~, I000 500

,

'-2

-1

1 0

,

,

,/

I

2

3

4

5

6

7

8 9 10

FIG.3. Impedance and SP change during asphyxiation. Upper line shows the SP change after asphyxiation at A. Small drifts followed by a negative variation of SP 3.5 minutes after the onset of asphyxiation. After a dip, the negativity rises again. In the lower record, at the time of the SP change, electrical impedance shows a rapid rise followed by a slower continued increase in impedance. Vertical line 5 mv for SP; ordinate impedance change in ohms; time in minutes. (From van Harreveld and Ochs, 1957. )

has the same rapidity of onset found during an SD (van Harreveld and Ochs, 1956). The electrolyte changes that occur during the asphyxial change also occur during an SD (Section V, A ) , with chloride (and sodium) and water entering apical dendrites (van Harreveld and SchadB, 1959). Another indication of the similarity of the mechanisms giving rise to an SD and to the asphyxial change is that when SD's are not observed in an animal refractory to SD excitation, the animal also does not show the asphyxial change (Marshall, 1961). The same cellular reaction is triggered synchronously after several minutes of asphyxiation as occurs in successive regions of the cortex during SD. However, additional factors are required for SD, making this reaction difficult to obtain in some tissues-retrosplenial cortex ( Le5o and Morison, 1945; van Harreveld and Bogen, 1956) and cerebellum (van Harreveld, 1961 ) , The lack of EEG changes after the first 15 seconds of asphyxiation (Sugar and Gerard, 1938) and the lack of effect of one minute of asphyxiation on SD appearance was interpreted by Le5o and Morison (1945) to indicate that SD may not immediately depend on neurons. The fact that very deep levels of pentobarbital anesthesia

14

SIDNEY O C H S

do not block SD (van Harreveld and Stamm, 1953b) would seem to support this view. However, other types of cortical activity are less sensitive to asphyxiation or to anesthetic agents than is the EEG. The direct cortical response is resistant to high doses of pentobarbital (Ochs, 1956), the DCRs may last as long as 3 minutes after the onset of asphyxiation. They decrease in amplitude until they disappear during the rapid asphyxia1 change (Ochs, 1958). The DCR involves at least one and probably more synapses (Ochs and Booker, 1961). After asphyxiation of the rabbit cerebrum for durations of 25 to 70 minutes and a recovery period of 72 hours, SD’s were found with the usual velocity of spread, although with unusually long durations of 10 to 15 minutes (van Harreveld and Stamm, 1954a). There was a correlation between the density of neurons in the cortex surviving asphyxiation and the presence of the SP change. If neuronal destruction is too severe no SD’s are found, which is further evidence for belief in the neuronal nature of SD. Ill. Excitation of Spreading Depression

A. MODESOF EXCITATION 1. Mechanical Leiio (1944a) found that mechanical stimulation of the cortex by a tap or by stroking the surface is effective in eliciting SD. This cannot be repeated too quickly because of a supervening refractory period. Repeated mechanical stroking of the surface to elicit SD’s was used in the studies of Winokur et al. (1950) and in the metabolic studies of KIivhek (1958). A systematic study of the parameters of mechanical stimulation was made by Zachar and Zacharovii (1961b). Threshold was found to vary inversely with the size of the area struck by a plunger device. It would seem from their studies that a certain minimal number of neurons must be activated before SD is excited. Sometimes when the cortex had been superiicially damaged in preparation or otherwise mechanically injured, cycles of SD might appear. Apparently minor cortical damage is effective in producing such cycles of SD. However, the larger cuts required to make a cortical slab (Grafstein, 1956a,b), and the cuts made to produce cortical or pallid islands (Ochs, 1958), do not usually produce recurring cycles of SD’s.

SPREADING DEPRESSION I N NEURAL NETWORKS

15

2. Electrical Electrical excitation is extensively used to elicit SD, yet a complete study of the parameters of electrical excitation has not yet been reported. A continuous cathodal current from a 4- to 6-volt battery applied to the exposed surface for several seconds readily excites an SD in the rabbit or rat. When using bipolar stimulating electrodes placed 2 to 3 mm apart on the surface, an effective current is 0.4 to 2 ma. This depends in part on the amount of surface fluid which shunts the stimulating current. Marshall (1950) described the use of repetitive rectangular pulses at 50 cycles/sec applied for 5 seconds. A train of impulses lasting 5 to 10 seconds at a repetition rate of 10-20/sec was used by Grafstein (19%a, b ) to excite SD in the cortical slab preparation. One-millisecond pulses of 10 to 20 volts at 28/sec with a train duration lasting only several seconds were found effective in initiating SD in rabbits with implanted electrodes ( Ochs et al., 1961) . Studies of electrical stimulation suggest that the immediate cause of SD excitation is the intense localized neuronal activation brought about by these relatively localized stimulations. LeHo and Morison (1945) stimulated the white matter of one cerebral hemisphere to excite an SD in the cortex of the opposite hemisphere. This finding was repeated in LeHo’s laboratory ( Marshall, 1959). While strong electrical stimulation of callosal fibers is effective in exciting an SD on the opposite side, SD propagating on one side does not excite SD on the other side nor does it propagate to other areas on the same side via the underlying corticocortical fibers. Photic stimulation of the eye can excite SD in the visual cortex of rabbits sensitized with Metrazol (van Harreveld and Stamm, 1955), and a convulsive after-discharge can also excite an SD (van Harreveld and Stamm, 195413).

3. Chemical Potassium chloride applied to the cortex on small pieces of filter paper can initiate an SD ( LeHo, 1944a). Filter paper squares were used to apply KC1 to the exposed dura at one burr hole while recording the EEG or SP changes of SD at another part of the brain of rats (BureH and BureHovA, 196Oa). The concentration of KCl which would induce SD in these rats in half the trials (the ED5,,) was found to be approximately 0.6%.Burel and BureIovL (1956)

16

SIDNEY OCHS

compared SD's elicited by concentrations of 2% and 25%KC1. Usually a single SD was seen with the application of a 2%concentration of KC1, while 25%KC1 caused cycles of SD appearing at intervals of 6 to 10 minutes for as long as 3 to 4 hours. These cycles of SD were signaled by SP changes; the EEG remained depressed for several hours. Bureg et al. (1960a) classified substances which could evoke SD into three groups: ( a ) those influencing the membrane directly, ( b ) substances exciting through an impairment of the energetic metabolism of the cells, and ( c ) the group of SD-evoking amino acids. The group acting on the membrane directly includes KC1, RbC1, and NH,Cl. On a molar basis the ED,,'s for these substances are respectively 0.08, 0.052, and 0.124. The group acting on energetic metabolism (and ED,, in moles) are: 2,4-dinitrophenol (2.08), NaCN ( l S l ) , NaF (1.05), NaN, (2.50), and iodoacetic acid (IAA) (8.86). The relation of metabolism to excitation will be discussed in Section V, D. The group of SD-evoking amino acids are of particular interest since they are connected with the y-aminobutyric acid (GABA) system in the brain (Section V, D). The amino acids (and molar ED,,) are: L-asparagine (0.055), D-glutamine (0.017), L-glutamine (0.077), DL-glutamic acid ( 0.224), and L-aspartic acid (0.089). Very small amounts of L-glutamic and a as par tic acids are sufficient to excite an SD (van Harreveld, 1959), and this is of interest because of the presence of these amino acids in the cortex. The possibility was advanced by van Harreveld that one or the other, or both, may be a factor in affecting excitation and propagation of SD in the cortex (Section IV). An even smaller amount of the nonnaturally occurring D-isomer is required to excite SD. The D-isomer is reported to be transformed in the cortex to the L-form (Stern et al., 1949), which, if true, may permit a more prolonged action on the surface of the cell. The specificity of amines has been emphasized by Bur& et al. (1960a), who suggested an action of the amino acids on enzymatic processes. Decarboxylation changes glutamic and aspartic acids respectively into the inactive forms GABA and p-alanine (Section V, D ) . Isoglutamine is ten times less effective than L-glutamine, and glycylglutamine antagonizes the excitant effect of glutamine. These findings were taken to suggest that an enzymatic process is involved. In any case, a specific action of these amino acids to excite SD is indicated. Recently, Curtis and Watkins (1961) found that

SPREADING DEPRESSION IN NEURAL NETWORKS

17

N-methybaspartic acid was most active to excite SD, and D-homocysteic acid was also a highly active substance. Those studies further point to the importance of amino acids in the cortex and their likely participation in the appearance of SD.

B. SPREADING DEPRESSION IN CHRONIC PREPARATIONS Earlier work by Marshall (1950) suggested that exposure of the surface and dehydration were necessary factors in eliciting SD in the cat. That both exposure and dehydration are facilitatory, but not necessary, was shown in a number of studies where SD was elicited from unexposed cortexes. In the experiments of van Harreveld and Stamm (1951), SD was elicited in one burr hole and passed via skull-covered cortex to another burr hole, where changes typical of SD were recorded. Bure6 and Bureiovh (1956) excited SD from burr holes placed anterior to the site where needle electrodes were introduced into the skull to record the EEG changes. Magun and Ross (1954) used needle electrodes set through the skull to excite SD and 7 days after implantation SD’s were successfully elicited. Depressions were therefore propagated in brain which was covered and most likely normally hydrated. Chronically implanted electrode inserts were made in the skulls of rabbits, cats, and monkeys by van Harreveld et al. (1956). Five to 7 days after implantation SD’s were successfully elicited and recorded via the chronically implanted electrodes. Marshall ( 1959) postulated that the use of curarization and artificial respiration in those experiments facilitated the occurrence of SD. However, in rabbits with chronically implanted electrodes, SP changes indicative of SD were seen in situ, without anesthesia or its adjuncts, in the same animals on successive days (Ochs et al., 1961). The factors of curare and artificial respiration were thereby eliminated. SD’s were excited and typical SP changes (Fig. 4) could be repeatedly obtained from the silver-silver chloride electrodes of the assembly. The time between the appearance of the SP change in the recording electrode closest to the stimulated pair (lower trace in Fig. 4 ) and its appearance in the more distant electrode was used to compute conduction velocity. Values of 4 mm/min were obtained, typical of SD propagation rates found in the exposed cortex. While SD was progressing through the cortex in these chronically implanted animals, they gave little obvious outward sign of motor disturbance,

18

SIDNEY OCHS

H

FIG. 4. SP changes from chronically implanted electrodes. To the left the position of the chronically implanted electrode assembly on the rabbit’s hemisphere is shown by circle with dots indicating electrodes (silver-silver chloride) in contact with pia-arachnoid surface. S indicates stimulating electrodes with R,, the closest recording electrode indicating SP change on lower trace; Rz the further electrode on upper trace. After the SD stimulus indicated by the arrow a latency is followed by a negative SP change and smaller positive deflection first in R,, then in Rz.Horizontal line 3 minutes, vertical 10 mv. (From Ochs et al., 1961.)

confirming the findings of Burd et al. (1958). These animals could be subjected to repeated SD’s for as long as 2 weeks. There was little apparent cumulative effect on the threshold for SD and little apparent difference in their behavior. Damage to the cortex by the presence or elicitation of SD’s has been suggested by Marshall (1959). A delicate test for cumulative damage may be the effect of SD on conditioned behavior. Burei and Bureiovi (1956, 1960b) had shown that application of 25% KCl causes recurring cycles of SD for a period of 3 to 4 hours and that during this time conditioned responses are depressed. Using a chronic cup implant (Russell and Ochs, 1960, 1961), SD’s were repeatedly elicited over a period of several weeks in a study of operant conditioning. It was found that such repeated elicitations of S D s did not disturb learned behavior nor prevent the cortex subjected to repeated SD’s from acquiring a learning behavior (Section VII ) . These experiments indicate that the alterations produced

SPREADING DEPRESSION IN NEURAL NETWORKS

19

by SD are reversible; an accumulation of extensive damage over a period of weeks of repeated SD’s did not occur as indicated by this index of cortical function. C. MODIFICATION OF SPREADING DEPRESSION EXCITABILITY While exposure and dehydration are not necessary factors for producing SD, as indicated in the preceding part of this section, exposure facilitates the excitation of SD in the monkey (Marshall and Essig, 1951) and in the cat (Sloan and Jasper, 1950; van Harreveld et al., 1956). These species are ordinarily more refractory to SD than is the rabbit. In the rabbit, SD does not progress from the neocortex across the shallow retrosplenial sulcus into the medial strip of cingulate cortex. Exposure of the surface of the brain for periods of 8 hours resulted in propagation across the retrosplenial sulcus into the cingulate cortex (van Harreveld and Bogen, 1956). Other procedures known (from the work of Marshall and his associates) to facilitate SD in the cat and monkey also facilitated transmission across the retrosplenial sulcus into the cingulate cortex of the rabbit. Factors studied were: intravenous injection of concentrated sucrose solution (Marshall, 1950), cooling of the cortex with mineral oil at a temperature of approximately 10°C (Marshall et al., 1951), washing the cortex with isotonic sucrose or with solutions containing ten times the normal potassium concentration (Marshall et al., 1952). Zacharovi and Zachar (1961a) were even able to initiate S D s by local cooling at sufficiently low temperatures. Where several excitant factors are present at the same time, each in subliminal strength, they may add to reach threshold for an SD. An excitant presumably acts by increasing neuronal activity which reduces the threshold for SD. This was indicated by the work of van Harreveld and Stamm (1955) where the visual cortex was excited repetitively by means of strong light flashes to the eye. Those animals were sensitized with subconvulsive doses of pentylenetetrazolel; the light flashes to the eye excited SD which spread forward from the visual area to the anterior regions of the cortex. Subthreshold mechanical pulses will sum with KCl (Zachar and Zacharovi, 1961c) or with cooling of the surface (Zacharovi and Zachar, 1961a) to excite SD. As Metrazol.

20

SIDNEY OCHS

Applications of 0.2 to 0.4%KC1 were usually ineffective in exciting an SD. When a 45-second period of asphyxiation was produced after placing 0.2, or 0.4%KC1 on the brain, SD’s resulted. This period of asphyxiation produces a loss of EEG activity but does not, by itself, elicit an SD (van Harreveld and Stamm, 1953b). Summation of potassium with short periods of asphyxiation is not specific. Subthreshold amounts of NH,C1 can also be brought to threshold by a short period of asphyxiation (Burei and BureZovS, 1960a). Not all the various agents shown effective in exciting SD are indiscriminately effective in summating with other excitant agents. Pentylenetetrazole will lower the threshold of excitation for glutamate and aspartate but not for potassium ion (van Harreveld, 1959). The variety of agents which have been found effective to trigger SD probably do so through different types of mechanisms, as already suggested (Section 111, A). Potassium probably acts on the neuronal membrane, and magnesium and calcium are effective a t this site in blocking the action of potassium when the bivalent ions are added to the cortex (Burei and BureSovii, 1956).The ratio of calcium concentration required to block a given concentration of potassium was in direct proportion to the logarithm of the potassium concentration. The blocking action for bivalent cations was effective in the following order of decreasing effect: magnesium, calcium, strontium, and barium. The decrease in the blocking action was found to be inversely proportional to the atomic weight of the bivalent ions. A similar effect of bivalent ions in blocking mechanical excitation of SD was found by Zacharov6 and Zachar ( 1961b).

D. REFRACTORINESS AND RECOVERY TIME Two types of resistance to the production of spreading depression are found. If the stimulus is repeated at an interval less than 2 to 3 minutes after an SD has been elicited, the reaction cannot be repeated ( Whieldon and van Harreveld, 1950), a refractoriness most likely caused by the immediate effects of the SD process. Judging from EEG and SP changes, the refractory process seems to be over in approximately 6 to 10 minutes for the rabbit and 15 to 20 minutes for the cat (van Harreveld et ul., 1956). Using mechanical stimuli, Zachar and Zacharovii (1961a) found an absolute refractory period of 2 to 3 minutes, and a relative refractory period lasting 13.5

SPREADING DEPRESSION IN NEURAL NETWORKS

21

to 14.5 minutes. But still longer-lasting changes may follow an SD. BureS and BureSovA (1956, 1960b) found that a single wave of SD produced in rats by 2%KC1 could depress conditioned responses for 30 to 60 minutes. Similar long-lasting metabolic changes were found by KfivAnek (1958) to follow a single SD in the rat. A glycogen decrease of 28%was found, with a gradual return to control levels within a period of 30 to 40 minutes. These longer lasting metabolic changes (Section V, D ) may be associated with variations in stimulation threshold sometimes noted in animals susceptible to SD. Also relating to these longer-lasting metabolic changes is the difference in form of a first SP change with respect to succeeding SD’s. The negative phase of the first SP change is often longer lasting. Marshall (1959) has pointed out that during a first SD the first spikelike wave of the sensorily evoked cortical response is not affected, while later SD’s have a temporarily depressive effect on this sign of activity in specific afferent fibers, suggesting a deeper spread into the cortex by the later SD’s (Section V I ) , In BureS and BureBovh’s (1960a) work on subthreshold KC1 excitation, insulin had an effect to augment SDs, but this appeared only after repeated subthreshold applications of KC1. Sloan and Jasper (1950) noted that in some cats an SD could be elicited within several minutes after reflecting the dura; then subsequent stimulations might not be effective. In other cats, SD could be found only after several hours of exposure. An unnoticed SD during preparation followed by prolonged refractoriness could give rise to the impression that SD does not occur in a given species unless the brain has been exposed for a long time. This variability was also seen in the chronic implant experiments of van Harreveld et al. (1956), where some animals showed an SD upon first stimulation but failed to give another SD, although repeated attempts were made at 15 to 20 minute intervals for 2 to 3 hours. In their experiments, 19 out of 20 cats gave at least one SD during the first 3 hours. In work with chronically implanted electrodes ( Ochs et al., 1961), a greater irregularity of the length of the refractory period was also found as compared to the acute case. Refractory periods ranged from 3 to 20 minutes and cases were also found where only a first SD could be elicited, and later stimulations were ineffective for an hour or more. Prolonged refractoriness may be related to differences in the case with which SD can be elicited in the various species. The more pro-

22

SDNEY OCHS

longed refractoriness during recovery time may be connected with metabolic events, as indicated by the finding that in the cooled cortex only a first SD response may be found ( Marshall et al., 1951). The reduction in metabolism of the cooled cortex presumably interferes with an adequate recovery from the first SD, thus preventing another SD. Possibly the higher species may have a greater dependence on metabolism with more time required for full recovery. In addition, inhibiting mechanisms may have developed to a greater extent in the higher species. Further evidence is required for an evaluation on a comparative basis of the relative importance of recovery time or refractoriness and an inhibition mechanism for SD excitability. Excitation of an SD would require a local activation which attains a sufficient level of neuronal activity in order to propagate. If the level of local excitation is not sufficiently great to overcome inhibiting mechanisms normally present, then the local change slowly subsides; an example is seen in Fig. 2 in the publication of van Harreveld and Stamm (1954b). In that figure stimulation might only produce an SP change localized to the electrode closest to the stimulated pair. van Harreveld and Stamm suggested that refractoriness in the region being invaded could account for the failure of propagation. A subthreshold mechanical pulse stimulation is followed by a refractory period lasting several minutes (Zachar and Zacharovh, 1 9 6 1 ~ ) . In the cortical slab preparations of the cat used by Grafstein (1956a,b) SD’s are more readily produced than in the intact animal. The slab may, on this account, be considered abnormally excitable in this respect. Why cutting the lateral neuron interconnections permits SD’s to be more easily obtained in this preparation is of interest with regard to mechanisms controlling SD threshold. It is possible that inhibiting neuronal activity from other cortical areas is thereby removed. IV. Propagation

A. CHARACTERISTICS OF PROPAGATION In its propagation, SD behaves like an all-or-none rather than a graded phenomenon. Once set in motion, the spread within the cortex is roughly uniform in all directions outward from the point

SPREADING DEPRESSION IN NEURAL NETWORKS

23

stimulated. The degree of involvement of any particular area is indicated by the amount of EEG depression or the amplitude of the SP change, and this in turn depends upon the state of excitability of the cortex in that area. In some cases, SD may propagate more quickly; in other cases spread of the reaction may fail. Local excitability changes or possibly also inhibitory mechanisms operating within various areas form the basis for differences found in any given region (Section 111, D ) . Spread of SD does not appear to be related to the functional nature of the cortex. The spread, for example, does not appear to be affected when it moves from sensory to motor or to associational cortex. However, SD does not spread readily from one type of cortex to another. It does not spread in the rabbit from neocortex across the shallow retrosplenial sulcus into the cingulate mesocortex ( LeHo, 1944a; van Harreveld and Bogen, 1956). Weiss and Fifkovi (1960) have shown that SD remains in the neocortex and does not spread to the hippocampus. When initiated in the hippocampus, SD does not spread to the neocortex. That propagation is essentially intracortical was shown by LeHo and Morison (1945) when they cut the connections between cortex and subcortical nuclei and found that propagation of SD continued in its usual fashion in the cortex. No special neuronal loop down to the thalamus or other subcortical structure is required.

B. MECHANISMOF SPREAD It will be helpful to enumerate the theories that have been held at one time or another with respect to the mechanism of propagation. We may group those theories as: ( a ) vascular-anoxic, ( b ) pia-arachnoid, ( c ) electrical-field, ( d ) neuronal-synaptic, and ( e ) neuronal-contiguous. With respect to the first theory, the vascular-anoxic, the necessary intervention of an asphyxia1 step seems unlikely in view of the inability of periods of asphyxiation of 1 minute or longer to induce SD. A latency of 2 to 5 minutes is usually observed before the onset of the SP negative change and impedance increase ( LeHo, 1947; van Harreveld and Ochs, 1956).The evidence obtained of vessel participation (Section 11, A ) has served to indicate that vascular phenomena are concomitant events and not causal. In the pia-arachnoid hypothesis, suggested by the sensitivity of the upper layers in initiation of SD ( LeHo and Morison, 1945), ves-

24

SIDNEY OCHS

sels release a substance which excites nerves in the pial membrane. The suggestion that the pia-arachnoid membrane was a link in transmission seemed also to be supported by Marshall's (1950) evidence showing that alterations of the surface conditioned the appearance of SD in the cat and monkey. The varied experiments on the chronically implanted animal (Section 111, B ) relegate exposure to the status of a facilitating agent. In any case, when pallial cuts were made and SD block was produced (Ochs, 1958; Ochs and Hunt, 1960), the pia-arachnoid membrane and pial vessels were not cut. This is taken as evidence against a transmission in these nonneuronal structures. An electrical-field theory of spread was advanced by Sloan and Jasper (1950). They made complete cuts through the cortex, and in their studies it appeared that SD could still propagate across those cuts if the surfaces were closely apposed. Block developed if thin insulating materials were inserted. They believed that the electrical field of active cells across the cut surface is the agent in propagation of SD. Electrical fields of active neurons spread across cuts through the pallium in the experiments of Libet and Gerard (1939, 1941). Extensive efforts were made to prove or disprove this theory of electrical-field spread for the SD phenomenon. van Harreveld et al. (1956) studied transmission beyond acute and chronic cuts of the pallial made some days earlier. They reported a failure of SD transmission across a transection of cortex and underlying white matter. When recording the SP on the side across the cut, a small potential one-tenth the amplitude of the SP change on the other side was found. This small potential fell off in size within a few millimeters from the line of the cut, as expected from passive flow of electrical current from the active side in the cortex spreading across the line of the cut. These passive currents were ineBective in initiating an SD. Grafstein (1956a) made downward vertical cuts in the path of SD propagation in the slab preparation or, alternatively, made cuts upward from below. The studies showed that complete cuts through the cortex could block SD transmission, and her work failed to confirm passage via electrical fields. In similar studies (Ochs, 1955b) long cuts were made through the cortex and SD failed to pass across these cuts. In later studies (Ochs, 1958), cortical cuts were used to prevent spread of SD into or out of an island of cortex. A fine knife was used to cut from just under the pia down

SPREADING DEPRESSION IIJ NEURAL NETWORKS

25

to the underlying white, the cuts foiming a square (Fig. 5 ) . In these islands of cortex, the cut edges are :losely opposed; yet passage did not occur. Where only a single long; cut in the cortex was made, an SD could turn the end of the cut and invade the far side with a latency determined by the longer iiistance traversed. Or, in island studies, if a small bridge of tissuc: of sufficient size remains, the inside may become invaded, again ,vith a latency depending on the

FIG.5. Cuts and island. To the left :i cross section of cortex is shown with M indicating molecular layer; G, corticil gray; and W, underlying white. A superficial cut is indicated in ( A ) , an upper cortex cut ( B ) , and cuts sparing the molecular layer ( C ) and ( D ) wher :in D the underlying white matter is included. Cuts C and D are called mo1i:cular layer preparations. To the right the position of the islands made with cuts are shown. Three of the sides are made with complete cortical cuts. Line F indicates that this cut is one of the types shown on the left. Stimulation of S ’ l through electrodes SD and recording of SP changes in R1 outside and R2 inside the island. (From Ochs and Hunt, 1960.)

position of the bridge. This evidenoe indicated that both sides across such cuts could support SD’s. The island was also shown to support SD’s as well, by eliciting SD’s inside the island, Cuts of the upper half of the cortex blocked SD whei studied several hours later or in the chronic case when cuts were made several days before (Ochs and Hunt, 1960). Failure of passa ;e of SD in experiments with cut surfaces in good apposition and 5 e minimal damage observed is further evidence against the elect5cal-field theory of spread. The lack of evidence for a special pial reaction, a vascular effect or electrical fields as agents of transmission of SD caused recon-

26

SIDNEY OCHS

sideration of the concept that transmission involves neuronal mechanisms. Two types of neuronal transmission hypotheses may be differentiated: “synaptic” and “contiguous.” According to the first type of hypothesis, intense synaptic activity causes the cellular changes typical of SD. These cells in turn excite SD in other cells with which they synapse, etc. The second hypothesis holds that following intense neuronal activation, a change in membrane permeability takes place and a substance is released into the intercellular spaces. This substance diffuses to nearby or contiguous elements and in turn excites SD in those cells, etc. Partial cortical cuts have been used to throw light upon neuronal mechanisms of spread. Grafstein (1956a) cut the cortex downward from above and found transmission was not blocked until almost the bottom of layer V was interrupted. Cuts directed upward from below did not stop transmission until the bottom of layer I was reached. The conclusion derived was that transmission was possible in all but the uppermost molecular layer. Much evidence indicates a special involvement of the uppermost layers in SD. The apical dendrites undergo great changes during SD (Section V ) , and apical dendrites are densely packed in the molecular layer. Shallow cuts (Fig. 5) extending from the pial surface down through the upper few layers were shown to produce a block of SD transmission for hours (Ochs, 19%). Those experiments suggested that the damage of apical dendrites caused the block and that apical dendrites in the upper layers were preferentially involved in transmission. If cuts are made down to approximately the middle of the cortex (layer IV) then SD block was permanent. This was also seen in chronic preparations where cuts were produced several days beforehand (Ochs and Hunt, 1960). Those results further indicated that the upper cortical layers and particularly the uppermost layers where apical dendrites are in greatest amount are more crucially involved than are the lower layers. It was not difficult to make cuts from below which spared all but the molecular layer. In these preparations (Fig. 5 C, D) where all but the molecular layer was cut (and deep pallid cuts on the other three sides were made to complete an island) transmission via the molecular layer could be regularly observed (Ochs and Hunt, 1960). Either the cut was made that day or a chronic preparation was made several days beforehand. This molecular layer prepara-

SPREADING DEPRESSION I N NEURAL NETWORKS

27

tion simplifies the experimental attack on the problem of transmission because the relatively thin molecular layer remaining is accessible to diffusion of topically applied agents (cf. part C of this section). The evidence of transmission obtained from the molecular layer preparation is further evidence that the apical dendrites are closely connected with transmission. Additional support of this conclusion is the evidence presented by Le5o and Morison (1945) that thermocoagulation of uppermost layers or their destruction with a knife prevented the occurrence of SD within that area even though repeated attempts to elicit SD’s were made for as long as 6 hours afterward. However, Grafstein ( 1956b) thermocoagulated the upper layers of a slab arid found transmission beyond the area. Her result may possibly be explained by the curvature of the cat gyrus in the slab preparation and the propagation in portions of the cortex not sufficiently reached by the heat used to thermocoagulate. Possibly also the excitability for SD propagation is enhanced by the cutting required for the production of the cortical slab so that propagation can take place more readily in the deeper layers. The preferential role of the upper layers in propagation was also indicated by the effectiveness of topically applied agents that can block propagation of SD. Cocaine would not be expected to diffuse very readily into the cortex, and this substance was shown to block transmission by Le5o and Morison (1945). At that time they attributed its SD block to an action on vessels. Burei (1960) describes the blocking action on SD of topically applied divalent cations, calcium, and magnesium in concentrations and after times which would probably not allow deep penetration. A decrease in the amplitude of the SP change of SD was seen soon after an area had been treated with low concentrations of MgC1,; block occurred within 10 minutes with a concentration of 0.1%.

THEORY OF TRANSMISSION C. CONTIGUITY The molecular layer preparation, in which transmission occurs in relative isolation (Ochs and Hunt, 1960), was used to compare a synaptic or contiguous mechanism. In Fig. 6A horizontally oriented apical dendrites are shown interdigitating closely in the molecular layer. According to the contiguity theory, the release of some substance from depressed dendrites passing to contiguous dendrites could excite SD over the line of the cut. Afferent fibers are shown

28

SIDNEY OCHS

A

B

FIG.6. Neuronal contiguity and synapse passage in molecular layer preparation. In A, the hypothesis of contiguity passage over the cut is indicated by the interdigitating apical dendrites. In B, synapse on apical dendrites is pictured to show transmission by the contrasted synaptic hypothesis. (From Ochs and Hunt, 1960.)

synapsing on apical dendrites in the molecular layer in Fig. 6B. According to the synaptic theory excessive activity of these synapses could excite SD over the line of the cut. GABA (7-aminobutyric acid) and nicotine were two agents found effective in blocking SD transmission in this molecular layer preparation. The substance GABA was selected because of its dramatically rapid block of the negative wave DCR or the negative phase of the evoked responses. Pui-pura et al. (1957) believe that GABA is a specific synaptic blocking agent acting on excitatory synaptic sites of the apical dendrites, while Iwama and Jasper (1957) have suggested that it has more generalized action on the apical dendritic membrane. In any case, GABA was found to be an effective blocking agent of SD passage in the molecular layer, and this action was readily reversible. Transmission of SD was shown in the control records of Fig. 7 by the successive appearance of the usual SP and EEG changes recorded on either side of the cut producing the molecular layer preparation, and with the usual latency of SD propagation. A 1%solution of GABA was then applied by means of a strip of filter paper moistened in the solution and placed over the line of the cut. The blocking action of GABA was shown by the absence of SP changes and of typical EEG depression in the island across the line of the cut. Washing the cortex removed the block, and transmission was again found. Reapplication of GABA

SPREADING DEPRESSION IN NEURAL NETWORKS

29

I FIG.7. Blocking action of GABA on molecular layer preparation. Using electrode positions shown in Fig. 5 with R, below and Rz above, A shows passage first to electrode outside, then in upper trace inside an island via the molecular layer. After 1%GABA was placed over line of molecular layer preparation B, transmission into the island was blocked. In C after wash, recovery of transmission was found. I n D and E block is again seen after 1%GABA was reapplied. Horizontal bar 3 minutes, vertical bar 10 mv. (From Ochs and Hunt, 1960.)

again resulted in block, and cycles of block and recovery could be repeated three times or more. An interesting finding was that if GABA is applied to the intact brain along the path of transmission it does not have the blocking action demonstrable in the molecular layer preparation. The penetration of GABA is probably not very great; therefore, we may infer that the lower part of layer I or the layers immediately below layer I can support SD transmission in the intact brain. The molecular layer block with GABA shows that at least the major portion of the molecular layer is penetrated by GABA. Brinley et al. (19600a) found that GABA produced only a transient increase in the potassium efflux from the cortex, as measured in repeated efflux samples after first loading with radioactive potassium. This increase in efflux with GABA was only 1%of the increased potassium efflux seen during SD (Brinley et al., 1960b). These results indicated that GABA is not a depolarizing agent. It blocks electrogenesis, apparently by some generalized membrane action (Iwama and Jasper, 1957). A more selective synaptic blocking action is not excluded by these experiments, but this possibility does not seem likely in view

30

SIDNEY OCHS

of the experiments of Curtis et al. (1959). In the motoneuron of the spinal cord, a multi-barreled microelectrode technique was used to release GABA in the vicinity of a motoneuron cell membrane. GABA blocked not only the excitatory PSP's in that motoneuron but also inhibitory PSP's, and it depressed the spike potential as well. The direct electrical excitability of the motoneuron membrane was also decreased, while the resting membrane potential was not reduced. Those results and the later evidence of Jasper (1960) on DCR's indicate that GABA has a generalized depressing action on electrogenesis of neuron membranes. If this is the case, we may consider that GABA blocks SD transmission in the molecular layer preparation by a generalized action on the apical dendrites. In an attempt to evaluate a possible synaptic activation of apical dendrites during SD, nicotine was used to block SD transmission in the molecular layer preparation. It was an effective blocking agent for SD and its action was reversible as shown in Fig. 8. Nicotine was

F J L G L"L JA

-

p+-i---I-* FIG.8. Blocking action of nicotine. Transmission is one-way in this example of molecular layer passage. It does not occur in A but does occur from inside (upper line) to outside the island in B. Nicotine 0.1% applied over line of molecular layer cut blocks transmission in C. After wash, transmission was still blocked as shown in D and E. Recovery occurred as shown in F, 48 minutes later. In G control and then 0.1%nicotine was applied again for only 5 minutes to produce block in H and recovery occurred later in J. Horizontal bar 3 minutes, vertical bar 10 mv. (From Ochs and Hunt, 1960.)

used because Libet and Gerard (1938) indicated that it blocked synapses in the optic lobe of the frog and in the brain stem of the cat, Its action as a blocking agent in the cord was questioned by van

SPREADING DEPRESSION I N NEURAL NETWORKS

31

Harreveld and Feigen (1948); it appears that nicotine excites Renshaw cell activity which could give, by a generalized Renshaw inhibition, an appearance of synaptic blocking action ( Eccles, 1957). In ganglia it is an excitant or a blocking agent depending on concentration. If nicotine is a selective synaptic blocking agent, then its action in the molecular layer preparation would be evidence for the synaptic hypothesis of transmission. A parasympathetic synaptic transmitter action at the apical dendrites would also be indicated. Yet, at higher concentration nicotine may effect the cell membrane at extrasynaptic sites. It is probable that nicotine when applied to the cortex may be too high in concentration at the surface, and its action there may be generalized to the whole membrane; this may occur even if at some lower depth nicotine had a restricted synaptic blocking action. This type of argument makes it a hazardous procedure to directly interpret the block as existing at a synaptic site when using topical application of agents to the cortex. The contiguity theory has been strengthened by another type of evidence involving the response of the specific sensory afferent fibers. If transmission of SD takes place by means of the synaptic mechanism we should hardly expect to see a depression of specific afferent-fiber response activity. The first fast spike is considered to be a sign of activity of radiation afferent fibers (Marshall, 1950). An effect of SD was found in which it depresses the activity of the specific afferent fibers giving rise to the first spikelike wave of an evoked response. This radiation spike is not affected much by the first several SD’s (Marshall, 19SO), but in the cat conditioned with ten times the usual concentration of potassium in Tyrode’s solution on the cortical surface, the first spike may be diminished by SD (Marshall, 1959). The first spike wave produced by a strong direct cortical stimulation occluded that of the sensorily evoked first spike elicited by optic nerve stimulation (Ochs, 1959). This indicated that the specific afferents underlie the first spike responses to strong direct stimulation of sensory areas. During SD, the first spikelike response elicited by strong direct stimulation was found to show a brief period of depression. Figure 9, depicting studies on a rabbit in which SD’s were elicited from chronically implanted electrodes (Ochs et al., 1961), shows the first spike diminished in size for a brief time and then succeeded shortly by a long broad negative discharge before it reverted to its usual biphasic or triphasic form.

32

SIDNEY OCHS

7-

160

7170

90 I

100

-I180

-T- I 110

190

-r7 200

I

-rIJZ ?- 7- T 7120

130

210

-L-

220

c

FIG.9. Depression of DCR spike and slow-wave components. The responses are elicited from sensory cortex of rabbit with a chronically implanted electrode assembly. Animal behavior not visibly altered by DCR stimulations at 10-second intervals, shown by numbers to the left of each response. These DCR's were elicited with strong stimulations to give early fast-wave components before the slow negative wave. After a control response (Co.), SD was elicited by several seconds of stimulation at 28/sec. SD began 40 seconds later with decrease in both slow and fast components. In particular note the diminution of first spike at 50 to 120 seconds and its earlier recovery before the other fast and the slow components. First spike is believed to be sign of activity in sensory afferents. Horizontal bar 20 msec, vertical bar 1 mv. (From Ochs et al., 1961.)

These results, along with Marshall's, indicate that during SD some substance within the cortex is released which has a generally depressing effect on the excitability of neurons in the neighborhood of the involved cells (Ochs and Hunt, 1960). Potassium ion as the agent has been suggested by Grafstein (1956b). Potassium is effective in exciting SD upon topical application and its release has been shown to o'ccur during SD (Section V, B). Another more specific

SPREADING DEPRESSION IN NEURAL NETWORKS

33

candidate is L-glutamic acid or L-aspartic acid (van Harreveld, 1959). These substances, which are found in the cortex, can each elicit SD in lower concentration than potassium. The slow rate of passage of SD could be explained in part by the time taken for release and diffusion of the transmitter agent to nearby neurons. Other factors, e.g., neuronal activity, could condition the invasion of a region. Grafstein (1956b) showed that cathodal polarization could enhance the rate of propagation whereas anodal polarization decreases the velocity of SD in a region; an increase in unit activity was found at the front of a region of entry, When the upper cortical layers are depressed, the lower layers may show a release phenomenon (Section VI, A ) . Usually this release is not predominant and only a few convulsive spikes may appear during SD, but under certain conditions it may be so great that spreading convulsions rather than depressions are observed (Section VI, B ) . A further study of unit activity seems necessary to differentiate between the possibility of release and an increased neuronal activity which is a causal factor in invasion. That only some types of neuronal activity are involved in SD excitation and propagation is suggested by the fact that comparatively deep pentobarbital anesthesia does not interfere with SD (van Harreveld and Stamm, 1953b). The DCR is one such activity which is not blocked by high levels of pentobarbital (Ochs, 1956). The involvement of metabolic factors in transmission is indicated by the work of Bureg (195%); the studies showed that a reduced rate of SD conduction was caused by qhort periods of asphyxiation or by topical application of a metabolic inhibitor agent. Reduction of body temperature also slows the rate of SD propagation with a Qlo of 1.7 (BureS et al., 1957). These indications of metabolic events underlying transmission rate of SD would be expected from evidence showing that profound metabolic changes take place in cells occupied by SD (Section V, D ) . V. Cellular Changes during Spreading Depression

A. ENTRY OF IONSAND WATER INTO CELLS The increase in electrical resistivity accompanying SD ( Section 11, D ) results from an entry of electrolytes (NaC1) and water into cortical cells from the intercellular compartment. The loss of these ions from the intercellular space, where nearly all of the conducting

34

SIDNEY OCHS

current of the impedance measurements passes, would give rise to the increased electrical resistances found during SD (van Harreveld and Ochs, 1957). In order to analyze the cellular basis of this translocation of ions, the similar but larger resistance increase occurring after asphyxiation was investigated (van Harreveld and Ochs, 1956). After a latency of 2 to 6 minutes during asphyxiation, a rapid increase in resistance takes place (Fig. 3). At the time of the resistance increase, an SP change also occurs (top trace in Fig. 3 ) , showing the similarity of the asphyxia1 changes with the SD changes (Section 11, H). The rapid phase of resistance increase may be as high as 50%; it is succeeded by a slower increase to a rise of two times and more of the original level of electrical resistance. The changes in conductivity of a mixture of particles of low conductivity suspended in a conducting medium were formulated by Maxwell (Cole, 1940; Cole and Curtis, 1944) as follows:

Where r = resistivity of particles and medium, rl = resistivity of medium, r2 = resistivity of particles, p = ratio of particle volume to total volume, and X = particle-form factor. Measurements of cells of different shapes using a range of different concentrations (Cole, 1940; Cole and Curtis, 1944) were shown to give results in conformity with this equation. For conductivity measurements made in a complex tissue such as the cortex, the varied shapes of the cells might be thought to be too heterogeneous to give a worthwhile analysis. The dendrites approximate cylinders; the cell bodies approximate spheres. Fortunately, for these extremes of shape, the computed estimate of resistivity differs by less than 5% when the cortical cells are considered to be all spheres or all cylinders. The actual situation in the cortex is some uncertain proportion of spheres and cylinders, but little difference in the results would be expected from this factor of shape. The electrical impedance studies of van Harreveld and Ochs (1956) carried out on living cortex resulted in an estimate of 360 ohm-cm for the resistivity. Setting this value and an estimated 60 ohms for the intercellular fluid conductivity into the equation, gave an extracellular space of 25%. Shape factor was taken as 1.5, half spheres and half cylinders. Approximately 25 to 30% of the cortical cells are neurons (Sholl, 1956).

SPREADING DEPRESSION IN NEURAL NETWORKS

35

Blood vessels constitute approximately 5%of the volume of the cortex and the blood accounts for 10%of the conductivity (van Harreveld and Ochs, 1956). Taking into account the various factors involved, van Harreveld and Schadk (1960) compared the resistivities found by Freygang and Landau (1955), Tasaki et d. ( 1954), and van Harreveld and Ochs (1956) and showed the fair agreement between these various electrical measurements of the cortex for a resistivity of 220 to 300 ohm-cm. These all would indicate that the impedance-measuring current passes in an intercellular ccrmpartment which is roughly 5 3 0 %of the volume of the cortex. A higher resistivity of 630 ohm-cm, corresponding with an intercellular space of lo%,was found by Ranck and Crill (1960). A lower value of 230320 ohm-cm was later reported ( Ranck, 1962). The most important consideration in these impedance measurements is the assumption that the current passes between and not through the cells. The membrane resistance of nerve cell axons is approximately lo00 ohms/cm2 (Cole and Curtis, 1939; Hodgkin, 1951), and a similar value is likely for cell bodies and dendrites ( Eccles, 1957). An expression for the resistivity of cells in a conducting medium given by Cole and Curtis ( 1944) is as follows:

+

rz = r3 r d a , (2) where r2 = resistivity of a spherical particle, r3 = resistivity of inside solution (cytoplasm), r4 = resistance of membrane, a = radius. A cell with a 10 p radius and a membrane resistance of 1000 ohms/cm2 with an internal resistance of the cytoplasm of 100 ohms/cm2 would have a resistance >lo6 ohms. The resistivity of the membrane may decrease to 1/40 or less of its original value during excitation (Cole and Curtis, 1939; Hodgkin, 1951).In that case we might expect some current to enter those cells of the cortex which are excited and thus have a decreased membrane resistance. But setting a value of 10 ohms/cm2 for the membrane resistance of the cells into Eq. (2) gives a resistivity of > lo4 ohm-cm, and hence a small effect on total conductivity (van Harreveld and Schadk,l960). The capacitative component of the cell membrane does contribute to a conductance through the cell. However, the reactance of neurons is relatively high at lo00 cycles/sec (Cole and Curtis, 1939) and only a small part of the current would enter the cells. As shown

36

SIDNEY OCHS

by van Harreveld and Ochs ( 1956),the reactance component constitutes a relatively small part of the impedance of the cortex; therefore most of the impedance change with asphyxiation was resistive. Most of the electrical impedance measures given in the literature correspond to intercellular space comparable to that estimated from chloride determinations. The size of the intercellular space (in particular, the claim of a very small space made on the basis of electron microscope studies ) will be discussed later in this section (part C). The entry of electrolyte and water into the neurons from the intercellular space indicated by the impedance changes was shown in a more direct fashion by van Harreveld (1957). The diameters of apical dendrites were measured in cortices which had been rapidly frozen and substitution-fixed in one group of animals before and in another group after the asphyxial change had taken place. The increase in diameter of the apical dendrites after the asphyxial change had taken place was approximately 30%,amounting to a volume increase of 6%. In the cat, dendrites showed a comparable increase after the asphyxial change. The perikarya increased 11% in diameter, which represents a volume increase of 37%.This volume increase observed in the apical dendrites after asphyxiation may be compared to the increases found when SD occupies a region of the cortex (van Harreveld, 1958). The brain was fast-frozen when SD moved into the region of the recording electrodes; it was then substitution-fixed. The diameters of apical dendrites were measured in strips of the cortex taken at successive intervals along the path of spread, the opposite cortex serving as a control (Fig. 10). The diameter increase found in the dendrites was greatest at the region just occupied by SD while those parts of the cortex recovering from SD showed a gradual return to normal diameters. The dendrites showed a 17%average increase in diameter. A more direct study of the entry of electrolyte was made by the use of a histochemical technique for determining chloride distribution (van Harreveld and SchadB, 1959). In those experiments the brain was fast-frozen with isopentane at -lW°C, and substitutionfixation was accomplished at -25OC in 90% alcohol with silver nitrate present. The silver which had diffused into the tissue along with the alcohol was precipitated locally as silver chloride where sufficient chloride was present. The presence of silver chloride was shown after it was reduced by light exposure. In the nonasphyxiated

SPREADING DEPRESSION I N NEURAL NETWORKS

37

FIG.10. Volume changes in apical dendrites during SD. SD was excited on one hemisphere of rabbit’s brain (black dot), and when it reached impedance electrodes (open dots) it caused an increase of resistance. The brain was then quick-frozen and substitution-fixed. Sections were taken in strips on the SD side, the other side served as a control. The diameters of the apical dendrites were measured on both sides and plotted on the two curves. The increase in apical dendrite diameters on the SD side (solid points) is greatest in the strip reached by SD, and smaller diameters were found in sections after partial recovery had taken place. Multiply ordinate by 0.33 to get dendrite diameters in microns. (From van Harreveld, 1958.)

cortex the silver chloride present was diffusely distributed. After the asphyxial change had taken place, a strikingly large increase was seen in the silver chloride staining of the apical dendrites. Some part of the staining may be due to changes of cellular phosphates, although most of the staining was shown to be due to the chloride present in greater amounts within the dendrites after electrolyte entry. The locus of chloride within the dendrite is uncertain. It may be just on the inside of the apical dendrites near the membrane of the cell, as suggested from optical and preliminary electron microscope studies of chloride localization (van Harreveld and Schadi., 1959), although it is possible that this is a result of central core ice formation produced during the fast-freezing process. A possibility to be considered is that the endoplasmic reticulum communicating to pores on the outside may be taking up water and electrolytes; thus these constituents would not actually enter the cytoplasm. While a small amount of Nissl substance is found in dendrites, and strands of agranular endoplasmic reticulum are seen

38

SIDNEY OCHS

at dendritic roots (Palay, 1958), the endoplasmic reticular system is greatest within the soma. On the other hand, the uptake of electrolyte is greatest in the apical dendrites, lower in the soma, and not found to any measurable degree in basilar dendrites (van Harreveld and SchadB, 1959). A special participation of apical dendrites in these electrolyte shifts was also indicated by the greater increases in the electrical resistance of upper cortical layers as compared to lower layers during SD (Freygang and Landau, 1955). In the upper layers, apical dendrites are the preponderant constituent. The steps leading to the entry of electrolyte and water into the apical dendrites during the asphyxia1 change and SD are, at present, still conjectural. van Harreveld and Ochs (1956) and van Harreveld and SchadB (1960) have suggested that the initial step is an increased permeability to sodium ion. Chloride ion would accompany sodium ion because of electrical considerations and water would enter because of increased osmoticity. The increase in permeability to sodium would have to be prolonged for many seconds or minutes during SD rather than for the few milliseconds or so when the sodium permeability increases during an action potential ( Hodgkin, 1951). The peripheral axon shows a small increase in diameter upon stimulation; this is probably related to electrolyte exchange ( Hill, 1950; Tobias, 1952). Hill concluded that the water entered passively as the internal osmoticity was increased by sodium and chloride entering the fiber. The kinetics of water entry was studied and a half-time of 1.1 minutes was found. Calculations of an increased osmoticity brought about by metabolic changes associated with nerve activity were made, and this possibility was discarded. Yet, the metabolic rate of apical dendrites is much higher than that of axons (6. part D of this section). Breakdown of phosphate compounds and other metabolic changes described as occurring during SD could create an increase in internal osmotic pressure and an additional amount d water could enter on that basis. Robinson (1954) reviewed evidence for the hypothesis that metabolism is used by cells to supply an active “water pump” which moves water continually out of cells in the normal condition. Interference of metabolism would give rise to an increased content of cell water. Part of the evidence on which this concept was based is that swelling occurs when oxidative respiration is inhibited by 2,4-dinitrophenol, cyanide, or chilling to temperatures of 0 to 4OC. The con-

SPREADING DEPRESSION IN NEURAL NETWORKS

39

trary evidence of Conway and McCormack (1953) is cited in an addendum to Robinson’s article, and further contrary evidence was given in a later paper by Conway et al., ( 1955).If the experimental evidence does not support hyperosmoticity during normal cell life, then these experimental findings indicate a rapid breakdown of cell components ( ATP, hexose esters, creatine phosphate, glycogen ) when metabolism is interfered with; such a breakdown could contribute to cell hyperosmoticity and water entry. The SP negative variation occurring during SD, and the analogous changes during asphyxiation, would appear as normal polarization is lost and sodium and chloride ion enter the cells. The loss of polarization is indicated by decreased excitability of DCR’s in a region occupied by SD (Section VI) and by the great increase in outflux of potassium from the cortex ( Brinley et al., 1960b). Tschirgi et al. (1957) correlated pH changes with the SP change of SD’s. In their view SP changes are related to the concentration ratio of H’ across the pial surface rather than to depolarization of apical dendrites. In later studies, Rapoport and Marshall (1962) found that pH changes could not account for the SP change seen during SD.

B.

INTERCELLULAR COMPONENTS During the permeability increase and entry of electrolytes and water into cells, evidence of a release of potassium from cells has been obtained. In the technique used by Brinley et al. (196Ob), radioactive K42-enrichedRinger solution was applied to the exposed dura, with 2 hours allowed for loading K42 into the cortex. Then, washings were taken from the surface at regular intervals for a control period and during and after the time the brain was excited to give an SD. The effect of an SD on K42outflux is shown in Fig. 11. Potassium efflux is expressed in terms of the percentage increase of K42 above control amounts of outflux. The increase in K4’ outffux was directly related to changes in SP, shown in the lower part of the figure. A relationship was found between the degree of outflux and the magnitude of the SP change. The efflux of potassium during SD was studied by BureH and Kfivhnek (1960), who used a continuously perfused chamber cemented over the exposed dura. Rats were injected with large amounts of radioactive potassium or sodium, a certain portion of which passed into the circulation and through the dura. The specific activity in the washed perfusate of RELEASE OF

40

SIDNEY OCHS

2500 400

350

300

2 1 r c X rn

8 v)

m 250

z

R 73

200

X

2

g 0

m

it- 1000

z

150 --I

3 0

500 0

> €

z

0 -2 -4

-6 -8 -10

0

60 120 180 240 300 360 420 480 540

TIME IN SECONDS FIG. 11. Outflux of potassium during an SD. The cortex is &st loaded with K" by placing a solution containing K" over the pia-arachnoid surface for several hours. Then at intervals, increments of washing fluid are removed and outflux of K" from the cortex was determined. While outflux samples are measured, an SD was initiated and the rate of outflux shows a very great i n c r e a s e a s much as four times. The dots indicating sample activities are to be compared with the SP recorded at the same time and shown in the lower part of the figure. (From Brinley et al., 196Ob.)

brain was close to that found in the plasma. During asphyxiation, an increased efflux of potassium was found, because of an added contribution of potassium coming from the intracellular compartment of the cerebral cortex. Sodium efflux was decreased. Ktivbek and BureB (1960) found that during SD potassium efflux increased while sodium efflux was not altered. A decreased specific activity of potassium was found in the perfusate, as would be expected if the extra

SPREADING DEPRESSION I N NEURAL NETWORKS

41

potassium came from inside neurons not labeled in the short times involved. The increased potassium efflux was correlated with the time course of the SD’s. Possibly the use of perfusion techniques may also reveal the efflux of other cell constituents-e.g., glutamic acid and other amines which may be involved in propagation (Section IV, C ) .

C. INTERCELLULAR SPACEAND GLIA The movement of electrolyte and water into and out of cells, considered to account for changes seen in SD, assumes the presence of a space or compartment between the cells, an intercellular space like that which is found in muscle and other tissues. This assumption has been challenged for central nervous tissue by several electron microscopists (Wyckoff and Young, 1956; Schultz et al., 1957; Farquhar and Hartmann, 1957). Under the electron microscope sections of spinal cord and cerebral cortex appear to have all their interstices packed with finer glial and neuronal procesess. These finer processes are much below the resolution possible with the light microscope and appear to fill the space that was earlier considered to be taken up by the ground substance. The intercellular space remaining between the membranes of these finer processes is small. The electron microscope measurements of Horstmann and Meves (1959) resulted in a value of only 5 7 %of the total volume for this space. Against this low estimate is the evidence of an extracellular space amounting to 2530%derived from electrical measurements (see part A of this section). If the chloride in the cortex is present mainly in the intercellular space, as is the case in other tissues, then a corresponding volume for the intercellular compartment would be approximately 20-35% (Manery and Hasting, 1939; Aprison et al., 1960). Tower’s (1960) discussion on chloride spaces is valuable in this regard. Thiocyanate studies indicated a space of approximately 14%(Streicher and Press, 1959), although part of the thiocyanate space may be caused by its toxic effect on cells, Davson and Spaziani (1959) studied the diffusion of radioactive chloride and iodine into brain slices. The lower and upper limits of the intercellular space they computed are 14 and 22%. On the basis of the large and rapid electrical resistance increase (Fig. 3 ) , indicating electrolyte and water shifts several minutes after asphyxiation, an explanation of the discordancy of electrical

42

SIDNEY OCHS

impedance and anion space studies with that of the electron microscope pictures may be considered. The intercellular compartment normally present in the living cortex is lost within the few minutes required for taking the tissue from the animal or during the fixation process required for preparation of the tissues. When the process continues as long as 20 minutes or more, one finds the high values for resistivity of the brain reported by Crile et al. (1922). Fastfreezing of the functioning cortex and substitution-fixation to prevent dislocation of intercellular-cellular relationships, in order to show their normal relations in the electron microscope preparations, has not as yet proved satisfactory, because of disruptive ice formation. It is hoped that current efforts to overcome this defect will be successful. A relatively large intercellular space may then be found present in functioning cortex before it is changed in the process of fixation and preparation. The movement of electrolyte and water during asphyxiation takes place with only slight change in the overall thickness of the cortex (van Harreveld and Ochs, 1956); this is independent of the shrinkage which occurs in fixation. Therefore, measurements of myelin-layer spacing ( Horstmann and Meves, 1959) that show minimal shrinkage in the preparation of the electron microscope material are not directly related to the present problem. According to another concept derived from electron microscope studies, the intercellular space as measured electrically and by chloride space determinations is actually that volume of the brain occupied by neuroglial cell processes (Gerschenfeld et al., 1959; Luse, 1960). Using stains and the light microscope, identification of the various types of glial cells (Glees, 1955) has posed some difficulties, and in electron microscope studies the difficulties of identification appear to be even greater. Luse (1960) believes that the glial elements occupying the intercellular spaces are oligodendrocytes, but others (Gerschenfeld et d., 1959; Tower, 1960) have considered the astrocytes to be the chloride-containing glial component; until this problem is better resolved we shall use the noncommital terms neuroglia or glia. The hypothesis that neuroglial cells occupy the intercellular space requires that the cell membranes of those glial cells do not have the usual semi-impermeability, so that electrical impedance measurements would give a low value for the “glial space” (van Harreveld and Schad6, 1959, 1960). The glial cells

SPREADING DEPRESSION IN NEURAL NETWORKS

43

would also be relatively rich in chloride content, and ionic exchange between glial cells and neurons would have to take place with sufficient rapidity to account for the fast impedance change during asphyxiation, A further requirement of the glial cells to fit known facts is that the part of the cell membrane which faces the capillaries would have to have a relatively high impermeability to account for blood-brain barrier properties. As yet only a little is known concerning the electrical properties of glial cells, Resting membrane potentials and long-lasting action potentials have been recorded from glial cells in culture (Hild et d., 1958). This would indicate that neuroglial membranes have electrical properties similar to those of other cells and therefore cannot directly account for intercellular space. Potentials have been found in the cortex which are believed by Tasaki and Chang (1958) to have been taken from glial cells, and they and Galambos (1961) suggested that neuroglia plays a role in SD. Neuroglia takes up water during edematous states, as shown by the electron microscope studies of Luse (1960). The volume of the brain also increases and such neuroglial volume increases probably take place over a much longer period of time than the rapid interchange between intercellular and cellular compartments, which take place minutes after the onset of asphyxiation. In the case of the rapid interchange, the total volume is unchanged. The opposite procedure of shrinking the brain by perfusing hyperosmotic glucose solutions into a blood vessel is followed after a latency of 20 to 30 minutes by a decrease of cerebral cortex conductivity (van Harreveld et al., 1961). The shrunken brain has the same content of potassium ion as the control brains have; water and sodium and chloride ion is decreased. The results are consistent with a slow egress of water and electrolyte from an intercellular compartment of the brain with potassium held within the cells. The slow exchange of electrolytes, metabolites, and water between capillaries and neurons may be regulated by glia (Held, 1909; Farquhar and Hartmann, 1957). It is not unlikely that neuroglia may also participate along with neurons during the faster asphyxia1 change and SD. The participation of glia in the rapid shifts of water and electrolyte taking place in the cerebellum after asphyxiation was shown by van Harreveld (1961). The cerebellum does not show SD. but like the cerebrum it shows a similar decrease

44

SIDNEY OCHS

in conductivity after asphyxiation appearing after a latency of several minutes with a slower rate of development. In each experimental case the relative uptake of electrolyte by the Purkinje and glial cells (Bergmann fibers) differed. The variable participation of glia and neurons found may reflect a competition between neurons and neuroglial cells for water and electrolytes of the intercellular space during the rapid change after asphyxiation.

EVENTS CONNECTED WITH SPREADING DEPRESSION D. METABOLIC Substances interfering with intermediary metabolism-2,4-dinitrophenol; NaCN; NaF; NaN,; and IAA-can excite SD (BureB, 1959). BureB and BureHovh (1956) found that the excitation of SD by potassium is blocked by Krebs cycle intermediates: citrate, a-ketoglutarate, glutamate, succinate, and malonate. These observations formed the basis for the suggestion that one common means of triggering SD is through an effect on oxidative metabolism. In an SD-occupied area, changes indicating a profound interference with metabolism have been found: a decrease in glycogen (-36%) and glucose (-36%) and an increase in lactic acid (+lOZW) (BureB et al., 1960a). These changes last longer than the duration of a single SD. The glycogen decreases and then gradually returns to control levels over a period of 30 to 40 minutes. This is the time during which conditioned reflex responses are depressed ( Kiivhnek, 1958). The relationship of carbohydrate metabolism to SD is shown by the activation of a subthreshold topical application of KC1 2 hours after an insulin injection (BureS and BureBovh, 1960a). Insulin hypoglycemia results in similar decreases in brain glucose, glycogen, creatine phosphate, and ATP (Olsen and Klein, 1947) as takes place during SD. The more gradual metabolic changes are probably associated with the full restoration of normal excitability. Faster metabolic changes may have a more intimate connection with reversion processes. During SD a decrease in creatine phosphate was found which was closely related to the SP change (Kiivhnek, 1961). A corresponding increase of inorganic phosphate was found. A similar rapid fall and return of oxygen tension in the cortex during the SP change was also noted (van Harreveld and Stamm, 1952). The decrease and return of oxygen tension of the cortex correlated with the SP change has been confirmed in Marshall's laboratory (6.Fig. 1 in Marshall,

SPREADING DEPRESSION IN NEURAL NETWORKS

45

1959). This decrease, with evidence of an increased blood %OW (Section 11, E ) , suggests an increased oxygen utilization and metabolic events closely connected with the SP change of the SD. Oxygen and creatine phosphate changes possibly related to potential changes are analogous to the direct connection between creatine phosphate and the resting membrane potential of muscle fibers reported by Ling and Gerard ( 1949). Recent evidence of a relationship between arginine phosphate and sodium pump activity has been obtained in the giant fiber (Caldwell, 1960; Caldwell et al., 1960), where arginine phosphate plays the same role in the C ~ S tacean as creatine phosphate does, with respect to its energy-rich bond, in the mammal. The problem is whether breakdown of creatine phosphate causes a direct change in membrane permeability and excitability or an indirect change through an action on a sodium pump mechanism. In peripheral axon, permeability and resting and action potentials are not much altered for over 2 hours while the sodium pump is blocked by a metabolic blocking agent (Hodgkin and Keynes, 1954). Failure of potentials occurs at a later time. Interruption of a sodium pump could conceivably cause faster changes in membrane permeability and potentials in the apical dendrites which operate at a very much higher metabolic rate. High rates of metabolism in apical dendrites were indicated by the microhistochemical studies of Lowry et al. (1954). A vertical core of cortex was divided into thin layers so that key enzymes in those samples of cortex could be measured at different depths. A relatively high level of oxidative metabolism was found for the molecular layer and, therefore, for the apical dendrites. Maturational studies also indicated a relationship between the high level of metabolism and the apical dendrites. During the first 20 days post partum, the cortex of the rat thickens and the cells develop adult characteristics ( Sugita, 1918). During this time the apical dendrites, small at birth, increase in size and branch out multitudinously. Eayrs and Goodhead (1959) traced the growth of apical dendrites which appear after the sixth day in the rat cortex and reach their adult form on the eighteenth day. Associated with dendritic growth is a three fold increase of the mitochondrial content during the first 21 days (Samson et al., 1960). An increase of key oxidative metabolism enzymes during maturation was found in the molecular layer (Kuhlman and Lowry, 1956). The evidence

46

SIDNEY OCHS

seems highly suggestive that apical dendrites contain a high content of mitochondria, which accounts for a high level of oxidative metabolism. Further evidence that apical dendrites play a crucial role in SD is that the reaction cannot be initiated until the adult stage of dendritic maturation is reached. In the rat, SD’s can be elicited only after the fifteenth day of life ( BureS, 1957a). In the rabbit, maturation takes a little longer and SD’s can be elicited only after 24 to 30 days post parturn ( Schadk, 1959). L-Glutamic and/or L-aspartic acid has been suggested as the specific agent released by neurons to effect SD excitation (van Harreveld, 1959). These substances are normally present in the cortex and when topically applied they are effective in exciting SD’s (van Harreveld, 1959; BureS et al., 1960a).This hypothesis is of particular interest in view of the importance of amino acid metabolism in neurons. A number of amino acids are present in the brain in relatively large amounts. Glutamic acid and glutamine are metabolically interrelated with GABA, a key member of this metabolic system (Tower, 1959; Roberts, 1960). GABA and the enzyme glutamic acid decarboxylase, which transforms glutamic acid to GABA, are found exclusively in central nervous system tissues, and mainly in gray matter. During maturation of apical dendrites, a ninefold increase of glutamic acid decarboxylase and a threefold increase in GABA takes place (Baxter et al., 1960), indicating a high content of the system in the apical dendrites. A recent hypothesis developed from observations on the effect of semicarbazide in blocking pyridoxine, the m-carboxylase of glutamic acid decarboxylase (Killam and Bain, 1957; Pfeiffer, 1960), is that the excitability of the cell may be related to changes in its level of GABA. In this hypothesis, if the level of GABA is increased the cell excitability is lowered; if GABA is decreased the cell excitability is increased, leading to a convulsive discharge (Roberts et al., 1960). In accord with the hypothesis, when GABA is topically applied it does not excite SD (van Harreveld, 1959; BureS et d.,1960a). Concentrations up to 10%were ineffective (Ochs and Hunt, 1960).On the other hand, GABA does not diffuse deeply into the cortex; therefore, it does not block SD transmission in the intact cortex. GABA does block SD transmission (Fig. 7) if it is placed over the line of the cut of a molecular layer preparation. In this preparation transmission takes place in the rela-

SPREADING DEPRESSION IN NEURAL NETWORKS

47

tively thin remaining molecular layer (Ochs and Hunt, 1960). The depressive action of GABA on transmission of SD in the molecular layer preparation, therefore, corresponds with its depressive action on the DCR (Purpura et al., 1957). L-Glutamic and L-aspartic acids have been found to excite an SD (Section IV). These substances were also shown to be excitant for the motoneuron membrane when they were ionophoretically released in its vicinity (Curtis and Watkins, 1960). In all probability the GABA system is not selectively involved. The GABA system is closely connected to oxidative metabolism (Tower, 1960; Roberts et ul., 1960). Glutamine (which is connected with the GABA system) was shown some years ago by Krebs et al. (1951) to be required in the medium of respiring brain slices for active uptake of potassium, an ion closely associated with cell excitability and able to excite SD. VI. Upper Cortical Layers and Release of lower layers

A. DIFFERENTIAL EFFECTON DIRECT CORTICAL RESPONSE The special nature of the upper cortical layers with respect to excitation and propagation of SD has been indicated in several ways. Using Dusser de Barenne’s technique of laminar thermocoagulation of the uppermost layers, Le5o and Morison (1945) caused a block of SD propagation. Not only does topical application of calcium and magnesium block excitation by KCl, but a region treated with 0.1 to 0.2%solutions of these ions is not invaded by SD ( BureS, 1960). In the initial stages of divalent ion action, the negative part of the SP change accompanying SD may be blocked, leaving the positive phase for several minutes before a complete block of SD occurs. We may presume that these divalent ions do not diffuse very deep into the cortex. The inference is, therefore, that propagation takes place preferentially in the uppermost layers. This preferential involvement of the uppermost layers is supported by the studies of Le5o (1951) on the delayed appearance of the SP change in successively deeper layers when recording with a microelectrode from the depths of the cortex. Cuts of the cortex which extended from the pial surface downward (cut A, Fig. 5 ) could block transmission for several hours, when only the upper few layers were severed, and the

48

SIDNEY OCHS

block was permanent if the cut extended down through the upper half of the cerebral cortex (cut B, Fig. 5 ) (Ochs, 1958; Ochs and Hunt, 1960). The greater density of apical dendrites in the uppermost layer could account for SD changes appearing there first. The other layers of the cortex also have dendrites present in lesser amount (Sholl, 1956), and similar changes may be expected, though in lesser degree. The special involvement of the upper layers of the cortex was also indicated by the use of two types of DCR. The positive-negative (PN) type of DCR (Fig. 12) is similar in form to the sensorily evoked responses. To evoke this type of DCR the stimulating electrodes are widely spaced, and stronger strengths are used. The elements with the lowest threshold for excitation of this response are in the deeper layers of the cortex. With weaker stimulation or the use of closely spaced insulated wires with tips exposed, stimulation of the more superficial parts of the cortex occurs and the DCR found is a simple slow negative ( N ) wave (Figs, 13, 14). These two types of DCR were used to investigate the effect of SD on the region of the cortex excited and thereby to assess the effect of SD on superficial and deeper parts of the cortex (Ochs, 1958). In order to delimit the effect of SD to either the DCR stimulating site or the responding site, it was necessary to resort to cortical islands (Fig. 5 ) . These cortical islands were rectangles roughly 6 to 8 mm on each side and were produced by making cortical cuts from pia down to the underlying white matter. The cortex outside such an island was excited to give DCR’s with the electrodes placed similarly to those labeled SD in Fig, 5, and the DCR was recorded outside (R,) and from within (R,) the island. The neuronal elements transmitting excitation of the DCR from the site stimulated outside the island to the responding site inside were shown to be corticocortical axons passing in the white matter underneath the gray matter (Ochs, 1956; Ochs and Booker, 1961). The passage of PN or N types of DCR into the island is not blocked by such cuts, while SD cannot pass into the island and, when SD is elicited from within the island, cannot pass out of it, The effect of SD is, thereby, restricted to either the site stimulated to give either the PN or N types of DCR or to the responding site. When a PN type of DCR was elicited and an SD was produced outside the island, SD blocked the response outside the island, but not the response inside the island ( Fig. 12). The

SPREADING DEPRESSION I N NEURAL NETWORKS

49

FIG. 12. Positive-negative (PN) type of DCR and failure of SD to depress its excitation. The position of the stimulating and recording electrodes to obtain these DCR’s is similar to that shown in Fig. 5. In this case, SD would refer to the DCR-stimulating electrodes, R1 to the outside recording electrode on the lower trace, R2 to the electrode within the island on the upper trace. Cuts producing the island extended from pia down to underlying white. The DCR stimulation used outside an island was stronger to elicit the PN type of DCR. The path of DCR transmission is via corticocortical elements into the island. The cortical cut will not block DCR but will block SD. After control ( cont. ), SD was elicited outside the island and at times indicated to left of each frame. This caused depression of the DCR recorded outside the island with the usual slow recovery. The stimulated cortex is not completely depressed as shown by the DCR’s which are recorded in the island without much change. (Traces were drawn over in some parts.) Horizontal bar 40 msec, vertical bar 0.25 mv. (From Ochs, 1958.)

site excited to give the P N response is deeper in the cortex and was not effected by SD. When an SD was elicited outside the island where an N type of response was excited, then responses both inside and outside the island were depressed ( Fig. 13). These results agree with the idea that SD has a relatively superficial cortical locus. They also imply a superficial site of excitation for the N type of DCR.

FIG. 13. Negative-wave type of DCR and SD depression of stimulated and responding areas. With weaker DCR stimulation, a simple negative-wave type of DCR is elicited. The stimulating and recording electrodes were placed similarly to those described for Fig. 12. On the upper trace a response from within the island is shown. SD was produced outside the island and it depressed the outside response (lower trace) and also the DCR recorded from inside the island. This result indicates that the weaker stimulation and more superficial site excited to give this type of DCR is depressed by SD, which is mainly located in superficial layers. Recovery takes place with the usual time course. The displacement artifact is somewhat large in this example. Horizontal bar 40 msec, vertical bar 0.1 mv. (From Ochs, 1958.)

SPREADING DEPRESSION I N NEURAL NETWORKS

51

Both types of DCR are depressed by SD excited within the island. The negative phase of the PN type of DCR is depressed more deeply and for a longer time than the positive part of the response. The positive part of the PN type of DCR on a volume conductor basis is due to depolarization of deeper cortical elements (soma, lower dendrites, basilar dendrites ) ; the negative phase, to depolarization of the upper cortical elements (apical dendrites) (Ochs, 1956). These experiments on the effects of SD confined either to the stimulated or to the recording sites giving rise to these two types of DCR, is additional evidence that SD occupancy takes place mainly in the upper cortical layers. This upper cortical layer involvement affecting separately the stimulating and responding sites could also be shown without the use of cuts, by means of a dual site-stimulation arrangement (Fig. 14). In this case, two stimulating sites S, and S, were each excited to give an N type of DCR at a recording point midway between them. The distance from each of these sites to the recording electrode was 2 to 3 mm. Paired excitation of these sites at a 30 msec interval was used. The cortex was then stimulated to give rise to an SD. As SD moved first into the S, stimulated site, it blocked excitation, as shown by the depressed response to S, stimulation, while the response to S, remained. Then SD propagated into the recording electrode region and most of the response from S, was depressed. A small negative response remained, presumably from undepressed apical dendrites below the surface. Finally SD moved into the S2 stimulating region, and this small remaining S, response was eliminated. Recovery of DCR amplitudes occurred gradually in the next several minutes. At present the cellular nature of the DCR is not yet resolved (Ochs and Booker, M l ) , but the differential effect of SD on the PN and N type of DCR when using stronger and weaker stimulus strengths indicates at least two types of elements excited. More recently a micro-stimulation technique has been used in this laboratory to study the DCR evoked from the different cortical laminae. That study has confirmed that different elements are excited at the various cortical depths. Further, in addition to a cortico-cortical transmission path for DCR’s, those studies showed transmission of an N wave DCR via the molecular layer (Ochs, 1962). How those findings will add to or modify our picture of SD remains to be determined.

52

SIDNEY OCHS

FIG. 14. Effect of SD on negative-wave DCR’s using two stimulated sites. Negative-wave DCR‘s were elicited at a 30-msec interval from each of t w o stimulating sites S, and S, each placed 2.5 mni from a central monopolar recording electrode R as shown in the lower diagram. In the control ( C ) , a negativewave DCR is elicited first from S, and then 30 msec later from s,. An SD was initiated posterior to the Sl stimulating site as shown by the diagram. As SD moves forward in the cortex, the S1 site stimulated to give the right-hand DCR is depressed at 0.58 minute (numbers at left of each sample are in minutes); the Sa response remains. Then as SD moves into the recording site, only a smallamplitude S, response is seen remaining at 1.06 minutes. Finally as SD moves into the site stimulated to give the S, response at 1.84 minutes, this small remaining response is completely eliminated. Recovery occurs with gradual increase in response amplitudes. Sine-wave calibration, 100 cycles/sec, 0.37 mv.

B. SPREADING CONWLSION

When SD occupies an area of the cortex, all activity within that region does not cease. The presence of an occasional large spike may be seen during the onset and height of the depression, With repeated activations of SD, there is a tendency for the amount of convulsive spikes to increase until finally convulsions rather than SD’s

SPREADING DEPRESSION IN NEURAL NETWORKS

53

may slowly propagate in the cortex (Leilo, 1944a; van Harreveld and Stamm, 1953a). These convulsive discharges do not result in peripheral motor signs (cf. Section 11, F ) . The absence of signs of lower motor activity during these spreading convulsions has been the reason for referring to this activity as convulsoid (Whieldon and van Harreveld, 1950). Spreading convulsions were seen in the records taken from the chronically implanted animals with no outward signs of convulsion observed (Ochs et al., 1961). The use of the term “spreading depression” emphasizes propagation as the characteristic common to the two phenomena, giving rise in the one case to a depressed EEG, in the other case to convulsive spike discharges. The propagation rate of the spreading convulsion is essentially similar to SD, as shown by successive appearance of spike activity in the electrodes spaced along its path of spread (Fig. 15). The localized nature of the change in an area is indicated by the occasional recording of convulsions at one station and depression at a nearby station. During spreading convulsions the other indexes of SD involvement are present. An SP change with a similarly large negative phase is found, as is an increase in impedance and the depression of DCR’s. These results suggest that during spreading convulsions the same depression of the upper layers (apical dendrites) occurs as in SD. In the case of spreading convulsions there presumably is in addition a release of activity in the lower layers of the cortex. Some change in the state of the neurons in deeper layers of the cortex may be postulated to give rise to convulsive activity. Repeated production of SD’s can result in spreading convulsions. However this change is reversible; after waiting an hour, SD’s can again be elicited (Whieldon and van Harreveld, 1950). Acetylcholine topically applied to the cortex was shown by van Harreveld and Stamm (1953a) to convert SD into a spreading convulsive pattern, as shown in the two p a t s of Fig. 15. This effect disappeared after washing the cortex. Acetylcholine topically applied in higher concentrations and with eserine present can cause epileptic spike activity. At the concentrations used by van Hareveld and Stamm acetylcholine would be a subliminal excitant increasing the tendency for spreading convulsions. Pilocarpine acted similarly to give spreading convulsions. Concentrations at 7 to 15%carbon dioxide were also effective in changing SD’s to spreading convul-

54

SIDNEY OCHS

C

FIG.15. For legend, see facing page.

SPREADING DEPRESSION IN NEURAL NETWORKS

55

sions. Carbon dioxide has been reported to have convulsive (Gyarfas et d.,1949) and anti-convulsive (Gerard et al., 1936; Pollock et al., 1949) properties. The effect of carbon dioxide in converting SD's to spreading convulsion may be due to a selective increase, caused by carbon dioxide, in the excitability of the cells in the lower layers of the cortex. Possibly some neural mechanisms of inhibition are present in the cortex which normally prevent the cells from developing convulsive synchronization. Carbon dioxide and other agents effective in converting SD's to spreading convulsion may have a more selective action on these inhibiting cells. Some basis for an inhibition mechanism was also suggested by ScliadB's (1959) study of the maturing rabbit cortex. While the apical dendrites were growing into the pattern characteristic of the adult, SD's did not appear until 24 to 30 days after birth. When first elicited, spreading convulsions were found more often than SD's. Presumably the inhibiting mechanism takes a longer time to develop. VII. Spreading Depression and Higher Functions in Cortex

A. EFFECT OF SPREADING DEPRESSION ON CONDITIONED REXPONSES Until recently surgical ablation of the cortex was usually performed in an attempt to assess the role of the cortex with regard to behavior. The use of SD for this purpose, introduced by Burei (cf. review by Bureg, 1959), is of particular interest because SD results in a temporary decortication and the animal may be studied for the effects of cortical deprivation soon after SD initiation. A food-conditioned reflex and an avoidance-conditioned reflex was established in rats over a period of days. Then after making burr holes in the FIG.15. Spreading depression and \peatling convulsion. Traces A, B, C and their continuations A', B', C' and A", B", C" in Section I are EEG records taken from three different stations placed in line outward from an SD-excited point. After SD was initiated as shown by the largcl artifact disturbance, depression is seen to begin successively in A, B, and C records, with a return in that order in their continuation records. In Section 11, the brain has been treated with topically applied 5%acetylcholine; in D, E, F with the same electrode positions SD stimulation causes the appearance of convulsive activity successively in electrode sites D, E, F with return to normal EEG activity in that same order shown in D', E', F'. Horizontal bar 10 sec, vertical bar 1 mv. (From van Harreveld and Stamm, 1953a.)

56

SIDNEY OCHS

skull over each hemisphere of the brain and allowing several hours for recovery from the immediate traumatic effects of surgery, SD was initiated with either 2 or 25%KCl placed over the dura in each hole. After the application of 2%KCl, conditioned reflexes could be obtained in the next minute, but thereafter the conditioned reflexes were abolished and recovery of conditioned responses gradually took place in approximately an hour (BureS et al., 1958). While these conditioned reflexes were depressed, unconditioned motor performance of the animal (posture and usual motor behavior) was not depressed. The application of 25% KC1 caused a series of SD’s indicated by waves of SP changes, these cycles of SD occurring for 3 to 4 hours. The EEG was depressed for several hours. The depressive effect of 25%KC1 on conditioned reflexes was correspondingly lengthened. The simpler escape response using an electrified floor grid is not so easily depressed as compared to the more complex avoidance-response behavior where the animal must move from a potentially painful grid upon presentation of the unconditioned (tone signal) stimulus. In a simple type of avoidance reflex an animal must move to another part of the box to prevent painful shock through the floor grid. This reflex may be prevented by SD in only a proportion of the animals. A still more complex avoidanceconditioning reflex was arranged by adding a partition to the cage requiring the animal to move through a door in the partition to avoid painful shock. This conditioned reflex was abolished during SD ( BureS, 1959). The effect of SD on hypothalamic self-stimulation and tegmental escape response was studied by Olds and Travis ( 1960). A schedule was arranged for their rats so that at 2-minute intervals the animal could self-stimulate in the hypothalamus; during the next 2 minutes they could (but did not) self-stimulate the tegmentum. During the next 2 minutes hypothalamic stimulation was given which could be terminated by the animal (escape behavior). In the last 2 minutes tegmental stimulation was introduced and the animal could terminate the stimulus. When SD was evoked by applying 22% KC1 to both cortices, self-stimulation was depressed. Tegmental escape was depressed in the second hour and returned in the third hour. Only by the end of the fourth hour did hypothalamic self-stimulation return. These results showed that escape behavior was less easily

SPREADING DEPRESSION IN NEURAL NETWORKS

57

depressed than self-stimulation. Possibly the latter type of behavior has a subcortical component (see part C of this section). Rats can be shown to have a “handedness,” a preference in the use of the right or left paw in reaching for food offered in a narrow tube. The effect of SD induced with 2%KC1 placed on the dominant hemisphere (the cortex opposite to the preferred hand) is to depress the reaching for food for 30 to 40 minutes ( BureHovh et d.,1958). Evoking SD in the homolateral hemisphere did not have this depressing effect. Handedness was imposed on the animal by crushing the brachial plexus of one foreleg and then training it to reach with the other paw. After regeneration of the crushed nerves, the “habit” of reaching with the limb not denervated was retained and SD in the hemisphere opposite to that side effectively depressed its use in reaching for food for 40 minutes. Thus the motor performance organized within one hemisphere, either de novo or trained in that hemisphere by peripheral nerve crush, was depressed by a unilateral SD in the contralateral cortex. Unilateral organization of a learned behavior was found by the use of appropriate sectioning of the brain (Sperry, 1958). Myers (1955) showed this for the visual system where the optic chiasm was cut in the midline as well as the corpus callosum and the animal trained with one eye covered. In that case the habit was found in the trained homolateral hemisphere only, as shown when the eye of the trained side was covered and the eye on the untrained side uncovered. In that case behavior indicated the absence of the engram. Spreading depression was used to depress the cortex of one hemisphere during learning and thus to allow an engram to be laid down in the other hemisphere ( BureH and BureHovh, 1960b; Bureg, 1959). One hemisphere was depressed by 25%KCI, and learning within the time of depression was indicated when the animal had attained criterion ( 9 out of 10 trials in the learning of an avoidance behavior). The trained cortex was then depressed with KC1 on the next day and the same number of trials was required to reach criterion as for naYve animals. As indicated in the part labeled A in Fig. 16, a high rate was required on the second day when the trained side was depressed. Retention of the habit by the trained side was shown by the smaller number of responses required to reach criterion (part D of Fig. 16) when the trained side was operative on the second

58

SIDNEY OCHS

20. 10.

0.

5

0 B

0..

C

f.7

20 I0 0 D

0 E

FIG. 16. Unilateral conditioning and interhemispheric transfer. The rat’s hemisphere is unilaterally depressed and the number of conditioning trials to criterion shown in the bar graph below. SD was produced by 25% KC1 applied to the left hemisphere ( A ) and 15 trials were required for the rat to reach criterion. Then, next day the same number of trials was required when SD was initiated in the other cortex. In D, however, when SD was introduced on the same side, the next day the number of trials required to reach criterion was diminished. Learning had occurred on the undepressecl side. When 5 trials were permitted without SD ( B ) this had no effect on the next day’s test. However, 10 trials with both sides open ( C ) did cause transfer as shown when the trained side was depressed and fewer trials were required. E, as a control, shows effect of SD without training. (From Bures, 1959.)

day. The use of SD to localize a learning engram to one side was also shown by Russell and Ochs (1960). Operant techniques were used and the retention of unilateral localization over a period of several weeks was shown by the use of a chronic technique. Small plastic cups were implanted in rats over small holes drilled through

SPREADING DEPRESSION IN NEURAL NETWORKS

59

the skull, one over each hemisphere. SD could be induced in one hemisphere by placing 25% KC1 into a cup before a training or testing session. While one hemisphere was depressed, the animal learned to press a lever for food ( F R schedule; Ferster and Skinner, 1957) during 1-hour sessions. Learning took place after several days and was indicated by the increased rate of bar-pressing (Fig. 17). The learning engram was shown to be present on the side which was not

175 1 150 -125 -100 .7 5 .50 --

25 .n

D A Y S

FIG. 17. Unilateral engram obtained with SD in chronic studies. The rat was prepared with cup inserts so that SD could be induced on one side each day for up to several weeks. The animal was put in the box for 1-hour sessions with one side (shaded) depressed with 25% KCl. The animal is semi-starved and pressing the lever will give one food pellet. On days 1 and 2 the animal makes only a few accidental bar presses; on days 3 and 4 learning is indicated by the higher bar-press rates. On the day 5 the trained side was depressed and the rate fell to low levels. On days 6, 7 , 8, the learned side was again functional; a return to high rates was found. Again on day 9 the untrained side was tested without response; on day 10 the learned rate was not depressed and the rate was high. It was still high on day 11 without food reinforcement. The presence of a unilateral learning in one hemisphere is indicated. (From Russell and Ochs, 1961.)

60

SlDNEY O M S

depressed during training, when later the trained side was depressed and the responses fell to the untrained level. With SD in the originally depressed side, high rates were again found and this could be repeated a number of times. These experiments with chronically implanted animals indicate that a relatively complex habit can be localized within the cortex of one hemisphere. Engrams were shown to be localized to either anterior or posterior parts of one cortex by the use of topical application of MgC12, which prevents SD from entering an area so treated ( BureSovL and BureS, 1960). If MgC1, is placed on the sensorimotor area, KC1induced SD no longer inhibits reaching for food. To a lesser extent such protection with divalent salts prevents the interference by SD of conditioned responses (avoidance response with an acoustic signal). Partial protection against impairment by SD of visual discrimination behavior was found with topical application of MgCl, to the visual area. The application of MgCl, does not have a generalized depressant effect, as indicated by evoked responses obtainable from cortex treated with MgC1,. The partial protection by areal application of protecting solutions of MgC1, raises the question of how much of the cortex is involved in various learned responses. Are some complex responses multiply represented in the cortex, in the sense suggested by Lashley (1950)? Cortical and subcortical interactions may also have to be considered in the problem of the topology of the engram. OF B. TRANSFER

A

LEARNING TRACE

In the experiments of Myers (1955) a unilateral engram could be learned in one hemisphere by cutting the optic chiasm, the corpus callosum, and covering one eye in training. The habit was then localized to the side with the eye open in training. When the corpus callosum was not cut, it served to channel the learning engram from the input cortical locus across to the cortex on the other side (Myers, 1956). In that case the engram was found in the opposite side as well as in the trained side. In the SD experiments of BureS (1959) the corpus callosum was intact and yet the engram remained lateralized. The unilaterality of the habit to one side was shown even for periods of two weeks or more in the experiments of Russell and Ochs (1960, 1961).Transfer does not spontaneously occur during those parts of the day when recovery from SD has occurred. When the animal

61

SPREADING DEPRESSION I N NEURAL NETWORKS

trained on one side was permitted to make a number of responses, transfer was accomplished, as shown by decreased number of trials needed for the untrained cortex (Bure6, 1959). This is indicated in part C of Fig. 16, where 10 trials were permitted with both sides open before the trained side was depressed. The lower number of trials required on the second test day showed that transfer had been effected. Five trials did not cause transfer as indicated (part B of Fig. 16) by the fact that the same number of trials was required on the second day. Transfer of a bar-pressing operant-response performance from trained to naive side may be brought about if the animal is permitted to make only a single trial and to receive one reinforcement

I 3

4

5

6

1

FIG. 18. One-trial interhemispheric transfer of an engram. An engram is learned on one side after days 1 and 2 as shown by high rates on day 3. With SD on the learned side, failure to press bar shown on day 4. On day 5, high rates are found with the learned side even without food reinforcement. On day 6, transfer is achieved by letting the animal make one bar-press and receive a food pellet reinforcement with both trained and unlearned sides not depressed. Then 1 hour later the trained side was depressed and a high rate of nnreinforced responses was found indicating transfer had occurred. (From Russell and Ochs, 1961.)

62

SIDNEY OCHS

without SD present (Russell and Ochs, 1961). The results shown in Fig. 18 tend to confirm other types of recent psychological studies indicating that learning is single-trial (Estes, 1960). The results also suggest that specific behavior must be emitted for the learningengram transfer to occur. Neuronal activity of a special kind takes place soon after a learning experience and before the long-lasting type of memory trace is laid down (Gerard, 1955). This solidification period was found by Duncan (1949) to take place within 15 minutes to one hour. Solidification took place within an hour in the single-trial conditioning studies of mice by Abt et al. (1961), where anesthetics were used to interfere with this early process. It will be of interest to investigate the time period of consolidation for onetrial interhemispheric transfer.

C. THECORTEX AND ITS SUBCORTICAL RELATIONS By means of a cannula implanted into the skull so that KC1 could be introduced down the inside to excite SD from the hippocampus, Weiss and Fifkovh (196@)showed that while SD was present in the hippocampus, of the unanesthetized curarized rat, neocortical activity remained undepressed. Conversely, hippocampal activity remained unchanged during bilateral SD’s elicited in the neocortex. The two structures are, therefore, relatively independent with regard to the presence of SD either in the neocortex or in the hippocampus. BureB et al. (1960b) used chronically implanted cannulae to induce S D s bilaterally in the hippocampi of rats. Animals with hippocampal SD’s appeared stuporous with few spontaneous movements, and showed a reduced activity to nociceptive and other external stimulation. Posture and locomotion were not impaired. Emotional control is believed to be organized in the hippocampus (MacLean, 1959) and evidence of emotional changes was seen when two animals with hippocampal SD’s were placed together. They faced each other in aggressive fashion with forelegs and snouts touching. In rats trained to a conditioned-defense reflex (simple avoidance reaction), hippocampal SD produced a much more serious interference in conditioned behavior than did bilateral cortical SD’s. In cats trained to an alimentary conditioned reflex, bilateral hippocampal SD’s resulted in a deep disturbance of the reflex responses, which lasted several hours. The hippocampus may be related to

SPREADING DEPRESSIOX I N NEURAL NETWORKS

63

processes of immediate memory mechanism (cf. part B above) and BureS’ (1959) comment is of interest in this connection. He noted that when hippocampal SD’s were introduced immediately after learning, no retention was found. If hippocampal SD’s were introduced 12 hours after learning, then retention occurred. VIII. Conclusions

At the present time the evidence seems sufficient to conclude that SD is neuronal in nature, and there are many indications that the apical dendrites are preferentially involved. The participation of neuroglia is problematical, and changes in blood vessels appear to be concomitant events, The phenomenon may be found in tissues other than neocortex, including the retina; this indicates that no special neural organization is required. According to the contiguity theory of transmission, potassium or a special substance which may be glutamic acid is released from depressed cells to excite nearby neurons. Conditions are most favorable for SD excitation and spread where there is a close proximity of neurons, most predominant in the molecular layer where apical dendrites interdigitate. Transmission of SD appears normally to spread preferentially in the uppermost layers. Propagation of SD can also occur in layers of the upper half of the cortex, where dendrites and their interdigitation are present to a lesser extent. In the higher species, areas within the brain may be more autonomous with less interdigitation than in the lower species. This factor may account for the greater difficulty in experimentally eliciting SD in the higher species. However, special inhibitory mechanisms may be present which reduce the tendency for SD excitation in the cortex of higher species. When spreading convulsions appear, a concomitant increased excitability of the lower layers of the cortex is inferred. The specific change characteristic of SD is an increased permeability of the apical dendrites and to a lesser extent, the soma of pyramidal cells. Electrolyte (sodium and chloride) enters the cell accompanied by water to be pumped out again a few minutes later. Profound metabolic changes occur during SD. In part these changes may be related to energy requirements for the pumping out of ions and water. The initial use of SD as a tool in psychological investigations is both rewarding and exciting. Unlike surgical ablation, where the

64

SIDNEY OCHS

need for recovery from its effects must be allowed for, studies of behavioral changes may be made immediately after SD is introduced. A temporary decortication is in effect produced by this means. The localization of an engram within the cortex of one hemisphere has been shown with SD. Furthermore, the transfer of an engram from one cortex to the other can be studied. This technique gives great promise for clarification of the mechanisms of learning and memory. ACKNOWLEWGMENT The work of the author and associates was supported by grants from the National Institutes of Health (B-1993 and M-4815) and the National Science Foundation ( G-6237 ) . REFERENCES Abt, J. P., Essman, W. B., and Jarvik, M. E. (1961).Science 133, 1477. Aprison, M. H., Lukenbill, A., and Segar, W. E. (1960). J. Neurochem. 5, 150. Baxter, C. F., Schadh, J. P., and Roberts, E. (1960). I n “Inhibition in the Nervous System and Gamma-Aminobutyric Acid,” A Symposium (E. Roberts, ed.),p. 214. Pergamon, New York. Bishop, G. H., and Clare, M. (1953). J. Neurophysiol. 16,490. Brinley, F. J., Jr., Kandel, E. R., and Marshall, W. H. (1960a). J. Neurophysiol. 23, 237. Brinley, F. J., Jr., Kandel, E. R., and Marshall, W. H. (1960b).J . Neurophysiol. 23,246. Burex, J. (1957a). Ekctroencephulog. and Clin. Neurophysiol. 9, 121. Burex, J. (195%). Physiol. Bohemosloven. 6, 447. Burd, J. (1959). In “The Central Nervous System and Behavior” ( M . A. B. Brazier, ed.), p. 207. Josiah Macy Jr. Foundation, New York. Burex, J. ( 1960). Physiol. Bohemoslouen. 9, 202. Bure;, J., and BureSovft, 0. ( 1956). Physiol. B o h e m o s h . 5, Suppl. 4. J. Neurophysiol. 23,225. Burs, J., and BureSovP, 0. (”a). Burd, J., and Bures’ovi, 0. (1960b). Ekctroencephulog. and Clin. Neurophysiol. Suppl. 13. Bureg, J., and Kiivhek, J. (1960). Physid. Bohenwsbuen. 9,488. Bureg, J., BureSovP, O., and Zacharovi, D. (1957). Physiol. Bohemoslouzn. 6, 454. BureB, J., Burdovi, O., and ZQhorovi, A . (1958). J. Comp. and Physiol. Psychol. 51, 263. Burex, J., Buregovi, O., and KIivhek, J. ( 196Oa). In “Structure and Function of the Cerebral Cortex,” Proc. 2nd Intern. Meeting Neurobiologists (D. B. Tower and J. P. Schadb, eds.), p. 257. Elsevier, Amsterdam, Netherlands. Burs, J., BureSov6, O., and Weiss, T. (196Ob). Physiol. Bohemoslouen. 9, 219. BureB, J., Fifkovi, E., and Marsala, J. ( 1 9 6 0 ~ )J.. C m p . Neurol. 114, 1. Burs, J., BureBovA, O., and FifkovP, E. ( 1961). Arch. ital. biol. 99, 23.

SPREADING DEPRESSION IN NETJRAL NETWORKS

65

BureiovA, 0. (1957a). Physwl. Bohemosloven. 5, 12. BureSovi, 0. (195710). Physiol. Bohemosloven. 6, 1. BuregovA, O., and Burs, J. ( 1960). Physiol. Bohemoslouen. 9,210. BureHovi, O., Bureg, J., and Beran, V. (1958). Physiol. Bohemosloven. 7, 29. Caldwell, P. C. (1960). J. Physiul. (London) 152,545. Caldwell, P. C., Hodgkin, A. L., Keynes, R. D., and Shaw, T. I. (1960). J. Physiol. ( L o n d o n ) 152,561 and 591. Cole, K. S. (1940). Cold Spring Harbor Symposia Quant. BioE. 8, 110. Cole, K. S., and Curtis, H. J. (1939). J. Gen. Physiol. 22, 649. Cole, K. S., and Curtis, H. J. (1944). In “Medical Physics” (0.Glasser, ed.), p. 344. Year Book Publ., Chicago, Illinois. Conway, E. J., and McCormack, J. I. (1953). J. Physiol. 55,539. Conway, E. J., Geoghegan, H., and McCormack, J. I. (1955). J. Physiol. (London) 130,427. Crile, G. W., Hosmer, H. R., and Rowland, A. F. (1922). Am. J. Physiol. 60, 59. Curtis, D. R., and Watkins, J. C. (1960). J. Neurochem. 6, 117. Curtis, D. R., Phillis, J. W., and Watkins, J. C. (1959). J. Physiol. ( L o n d o n ) 146, 185. Curtis, D. R., and Watkins, J. C. ( 1961). Nuture 191, 1010. Davson, H., and Spaziani, E. (1959). J. Ph!ysiol. (London) 149, 135. Duncan, C. P. (1949).J. Comp. and Physiol. Psychol. 42, 32. Dusser de Barenne, J. G., and McCulloch, W. S. (1939). J. Neurophysiol. 2, 319. Dusser de Barenne, J. G., and McCulloch, W. S. (1941). J. Neurophysiol. 4, 311. Eayrs, J. T., and Goodhead, B. (1959). J. Anat. 93, 385. Eccles, J. C. (1957). “The Physiology of Nerve Cells.” Johns Hopkins Press, Baltimore, Maryland. Essig, C. F., and Marshall, W. H. (1950). PTOC.SOC. Exptl. Biol. Med. 75, 429. Estes, W. E. (1960). Psychol. Rev. 67, 207. Farquhar, M. G., and Hartmann, J. F. (1957). J. Neuroputhol. 16,18. Ferster, C. B., and Skinner, B. F. (1957). “Schedules of Reinforcement.” Appleton-Century-Crofts, New York. Freygang, W. H., Jr., Landau, W. M., and Marshall, W. H. (1954). Federation Proc. 13, 48. Freygang, W. H., Jr., and Landau, W. M. (1955). J. Cellular C o m p . Physiol. 45, 377. Fulton, J. F. ( 1955). In “A Textbook of Physiology,” 17th ed., p. 281. Saunders, Philadelphia, Pennsylvania. Galambos, R. (1961). Proc. N d l . Acud. Sci. U . S. 47, 129. Gerard, R. W. (1955). Science 122, 225. Gerard, R. W., and Libet, B. (1940). Am. J. Psychiat. 96,1125. Gerard, R. W., Marshall, W. H., and Saul, L. J. (1936). A. M . A. Arch. Neurol. and Psych. 36, 675. Gerschenfeld, H. M., Wald, F., Zadunaisky, J. A., and De Robertis, E. D. P. ( 1959). Neurology 9, 412.

66

SIDNEY OCHS

Glees, P. ( 1955). “Neuroglia, Morphology and Function.” Charles C Thomas, Springfield, Illinois. Gouras, P. (1958). Am. J. Ophthalmol. 46, 59. Grafstein, B. (1956a). J. Neurophysiol. 19, 154. Grafstein, B. ( 195613). J. Neurophysiol. 19, 308. Gyarfas, K., Pollock, G. H., and Stein, S. M. (1949). Proc. SOC. Expt. Biol. Med. 70, 292. Hankinson, J., Namin, P., and Ochs, S. ( 1955). Brain 78, 133. Held, H. ( 1909). Monatschr. Psychiat. Neurol. 26, 360. Hild, W., Chang, J. J., and Tasaki, I. (1958). Experientiu 14, 220. Hill, D. K. (1950).J. Physiol. (London) 111, 304. Hodgkin, A. L. (1951). Bwl. Revs. Cumbridge Phil. SOC. 26, 339. Hodgkin, A. L., and Keynes, R. D. (1954). Symposia Soc. Exptl. B i d . No. 8, 423. Horstmann, E., and Meves, H. (1959). Z . Zellforsch. u. mikroskop. Anut. 49, 569. Iwama, K., and Jasper, H. H. ( 1957).J. Physiol. (London)138, 365. Jasper, H. (1960). In “Inhibition In the Nervous System and Gamma-Aminobutyric Acid,” A Symposium (E. Roberts, ed.), p. 544. Pergamon, New York. Killiam, K. F., and Bain, J. A. (1957). J. Pharmucol. Exptl. Therup. 119, 255. Krebs, H. A., Eggleston, L. V., and Terner, C. ( 1951). Biochem. J. 48, 530. Ktivbek, J. (1958). J. Neurochem. 2, 337. Ktivhek, J. ( 1961 ) . J. Neurochem. 61, 183. Ktivhek, J., and Bure;, J. (1960). Physiol. Bohemoslowen. 9,494. Kuhlman, R. E., and Lowry, 0. H. (1956). J. Neurochem. 1,173. Lashley, K. S. (1950). Symposia SOC. Exptl. Biol. No. 4, 454. LeLo, A. A. P. (1944a). J. Neurophysbl. 7,359. Lelo, A. A. P. (194413). J. Neurophysiol. 7,391. LeZo, A. A. P. (1947). J. Neurophysid. 10,409. LeCo, A. A. P. (1951). Electroencephalog. and Clin. Neurophysiol. 3, 315. LeCo, A. A. P., and Martins Ferreira, H. (1953). An& acud. brasil. cienc. 25, 259. (Ref. Marshall, 1959.) LeLo, A. A. P., and Morison, R. S. (1945). J. Neurophysiol. 8, 33. Libet, B., and Gerard, R. W. (1938). POC. SOC. Exptl. BWZ. Med. 38, 886. Libet, B., and Gerard, R. W. (1939). J. Neurophysiol. 2, 153. Libet, B., and Gerard, R. W. ( 1941). J. Neurophysiol. 4, 438. Ling, G., and Gerard, R. W. (1949). J. Cellular C m p . Physiol. 34, 413. Lowry, 0. H., Roberts, N. R., Leiner, K. Y., Wu, M.-L., Farr, A. L., and Albers, R. W. (1954). J. Biol. Chem. 207,39. Luse, S. A. (1960). In “Inhibition in the Nervous System and Gamma-Aminobutyric Acid,” A Symposium ( E . Roberts, ed.), p. 29. Pergamon, New York. MacLean, D. (1959). I n “The Central Nervous System and Behavior” ( M . A. B. Brazier, ed.), p. 31. Josiah Macy Jr. Foundation, New York. Magun, R., and Ross, J. (1954). Deut. Z. Nerwenheilk. 171,403. Manery, J. F., and Hastings, A. B. (1939). J. Biol. C h m . 127, 657. Marshall, W. H. ( 1950). Electroeneephalog. and Clin. Neurophyswl. 2, 177.

SPREADING DEPRESSION IN NEURAL NETWORKS

67

Marshall, W. H. (1959). Physiol. Revs. 39, 239. Marshall, W. H. ( 1961). Personal communication. Marshall, W. H., and Essig, C. F. (1951). J. Neurophyswl. 14,265. Marshall, W. H., Essig, C. F., and Dubroff, S. J. (1951). J. Neurophysiol. 14, 153. Marshall, W. H., Essig, C. F., Landefrld, M., and Witkin, L. B. (1952). Electroencephalog. and Clin. Neurophysiol. 4, 244. Martins Ferreira, H., and Lezo, A. A. P. ( 1958). An& acad. b r a d . cienc. 30, No. 2, 1. Cited by Marshall ( 1 9 5 0 ) . Milner, P. M. ( 1958). Electroencephnlog. and Clin. Neurophysiol. 10, 705. Myers, R. E. (1955). J. C o m p . and Physinl. Psychol. 48, 470. Myers, R. E. (1956). Brain 79, 358. Ochs, S. (1955a). Am. J. Physiol. 182, 411. Ochs, S. (1955b). Science 122, 877. Ochs, S. (1956). J. Neurophysiol. 19, 513. Ochs, S. (1958). J. Neurophysiol. 21, 159. Ochs, S. (1959). J. Neurophysiol. 22, 2. Ochs, S. (1962). Federation Proc. 21, 642. Ochs, S., and Booker, H. (1961). Exptl. Ncurol. 4, 70. Ochs, S., and Hunt, K. (1960). J. Neurophysiol. 23, 432. Ochs, S., Hunt, K., and Booker, H. (1961). Am. J. Physiol. 200,1211. Olds, J., and Travis, R. P. (1960). Federation Proc. 19, 293. O’Leary, J. L., and Goldring, S. (1959). Physiol. Hand. (Am. Physiol. SOC., Washington, D. C.) 1, 315. Olsen, N. S., and Klein, J. R. (1947). Arch. Biochem. 13, 343. Palay, S. L. (1958). Exptl. Cell Research Suppl. 5, 275. Pfeiffer, C. C. (1960). I n “Inhibition in the Nervous System and GammaAminobutyric Acid,” A Symposium ( E . Roberts, ed.), p. 273. Perganion, New York. Pollack, G. H., Stein, S. N., and Gyarfas, K. (1949). Proc. Soc. Exptl. Biol. Med. 70, 291. Purpura, D. P., Girado, M., and Grunclfest, H. (1957). Science 125, 1200. Ranck, J. B., Jr., and Crill, W. E. (1960). Physiologist 3, 128. Ranck, J. B. (1962). Federation Proc. 21, 350. Rapoport, S. I., and Marshall, W. H. ( 1962). Federation Proc. 21, 352. Roberts, E. (1960). I n “Inhibition in the Nervous System and Gamma-Aminobutyric Acid,” A Symposium ( E . Roberts, ed.), p. 144. Pergamon, New York. Roberts, E., Baxter, C. F., and Eidelherg, E. (1960). In “Structure and Function of the Cerebral Cortex,” Proc. 2nd Intern. Meeting Neurobiologists ( D . B. Tower and J. P. Schadd, eds.), p. 392. Elsevier, Amsterdam, Netherlands. Robinson, J. R. (1954). Symposiu Soc. Expt. B i d . N o . 8, 42. Roitbak, A. I. (1955). “Bioelectric l’henoinena in the Cortex of the Cerebral Hemisphere.” ( I n Russian; translation available, Natl. Inst. Health Library, Bethesda, Md.) Acad. Sci. Georgian S.S.R., Tiflis. Russell, I. S., and Ochs, S. (1960). Physiologist 3, 152. Russell, I. S., and Ochs, S. ( 1961 ). Science 133, 1077.

68

SIDNEY OCHS

Samson, F. E., Balfour, W. M., and Jacobs, R. J. (1960). Federation PTOC. 19, 286. Schad6, J. P. (1959). J. Neurophysiol. 22, 245. Schultz, R. L., Maynard, E. A., and Pease, D. C. (1957). Am. J. A m t . 100,369. Sholl, D. A. (1956). “The Organization of the Cerebral Cortex.” Wiley, New York. Sloan, N., and Jasper, H. (1950). Electroencephalog. and Clin. Neurophysiol. 2, 59. Sonnenschein, R. R., and Walker, R. M. (1956). Federation Proc. 15, 175. Sperry, R. W. ( 1958). In “Biological and Biochemical Bases of Behavior” (H. F. Harlow and C. N. Woolsey, eds.), p. 401. Univ. Wisconsin Press, Madison, Wisconsin. Stern, J. R., Eggleston, L. V., Hems, R., and Krebs, H. A. (1949). Biochem. J. 44, 410. Streicher, E., and Press, G. D. (1959). Federation Proc. 18, 152. Sugar, O., and Gerard, R. W. (1938). J. Neurophysiol. 1, 558. Sugita, N. (1918). J. Comp. Neurol. 29, 61. Tasaki, I., and Chang, J. J. ( 1958). Science 128, 1209. Tasaki, I., Polley, E. H., and Orrego, F. (1954). J. Neurophysiol. 17, 454. Tobias, J. M. ( 1952). Cold Spring Harbor Symposia Quant. Biol. 17, 15. Tower, D. B. (1959). In “Biochemistry of the Nervous System,” 4th Intern. Congr. Biochem. (F. Briicke, ed.), p. 251. Pergamon, New York. Tower, D. B. (1960). In “Structure and Function of the Cerebral Cortex,’ Proc. 2nd Intern. Meeting Neurobiologists (D. B. Tower and J. P. Schadk, eds.), p. 411. Elsevier, Amsterdam, Netherlands. Tschirgi, R. D., Inanaga, K., Taylor, J. L., Walker, R. M., and Sonnenschein, R. R. (1957). Am. J. Physiol. 190,557. van Harreveld, A. (1957). Am. J. Physiol. 191,233. van Harreveld, A. (1958). Am. I. Physiol. 192,457. van Harreveld, A. (1959). J. Neurochem. 3, 300. van Harreveld, A. (1961). J. Cellular C m p . Physiol. 57, 101. van Harreveld, A., and Bogen, J. E. (1956). Proc. SOC. Exptl. Biol. Med. 91, 297. van Harreveld, A., and Feigen, G. A. (1948). J . Neurophysiol. 11, 141. van Harreveld, A., and Ochs, S. (1956). Am. J. Physwl. 187, 180. van Harreveld, A., and Ochs, S. (1957). Am. J. Physiol. 189, 159. van Harreveld, A., and Schadk, J. P. (1959). J. CelEular C m p . Physiol. 54, 65. van Harreveld, A,, and Schad6, J. P. (1960). In “Structure and Function of the Cerebral Cortex,” Proc. 2nd Intern. Meeting Neurobiologists. (D. B. Tower and J. P. Schad6, eds.), p. 2.39. Elsevier, Amsterdam, Netherlands. van Harreveld, A., and Stamm, J. S. (1951). Electroacephdog. und C2in. Neurophysiol. 3, 323. van Harreveld, A., and Stamm, J. S. (1952). J. Neurophysiul. 15,487. van Harreveld, A., and Stamm, J. S. (1953a). J. Neurophysiol. 16, 352. van Harreveld, A., and Stamm, J. S. (1953b). Am. J . Phy*l. 173, 164. van Harreveld, A., and Stamm, J. S. ( 1 9 5 3 ~ )Am. . J. Physiul. 173,171. van Harreveld, A., and Stamm, J. S. (1954a). Am. J. Phydo2. 178, 117.

SPREADING DEPRESSION I N NEURAL NETWORKS

69

van Harreveld, A,, and Stamm, J. S. (195413).J. Neurophysiol. 17, 505. van Harreveld, A., and Stamm, J. S. (19.55). Electroencephdog. and Clin. Neurophyswl. 7, 363. van Harreveld, A., Stamm, J. S., and Christensen, E. (1956). Am. J. Physiol. 184, 312. van Harreveld, A., Hooper, N. K., and Cusick, J. T. (1961). A.m. J. Physiol. 201, 139. Weiss, T. (1961a). Physiol. Bohernoslouen. 10, 21. Weiss, T. ( 1961b). Physiol. B o h m o s l m e n . 10, 109. Weiss, T., and Fifkovi, E. (1960). Electroencephalog. and Clin. Neurophysiol. 12, 841. Weiss, T., and Fifkovi, E. (1961). Arch. intern. physiol. ef b i o c h m . 69, 69. Whieldon, J. A,, and van Harreveld, A. (1950). Electroencephubg. and Clin. Neurophyswl. 2, 49. Winokur, G. L., Trufant, S. A,, King, R. B., and O’Leary, J. L. (1950). Electroencephulog. and CEin. Neurophysiol. 2, 79. Wyckoff, R. W. G., and Young, J. Z. (1956). PTOC.Roy. SOC. B144, 440. Zachar, J., and Zacharovi, D. (1961a). Ph!ysiol. Bohanosloven. 10, 341. Zachar, J., and Zacharovzi, D. (1961b). (In preparation.) . and Clin. NeuroZachar, J., and Zacharovi, D. ( 1 9 6 1 ~ ) Electroencephalog. physiol. 13, 896. Zacharovi, D., and Zachar, J. ( 1961a). Physiol. Bohemosbven. 10, 332. ZacharovA, D., and Zachar, J. (1961b). ( I n preparation).

This Page Intentionally Left Blank

ORGANIZATIONAL ASPECTS OF SOME SUBCORTICAL MOTOR AREAS By Werner

P.

Koella

Worcester Foundation for Experimental Biology, Shrewsbury, Massachusetts

I. Introduction: Fundamentals of Motor Organizations . . . 11. Motor Effects Produced by Stimulation of the Diencephalon 111. Motor Effects Produced by Stimulation of the Cerebellum .

. .

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

IV. The Midbrain Tectuni V. The Tegmental Reaction VI. Discussion and Interpretation References . . . . . .

. . . .

. . . . . . . . . . .

. . . . . . . . . . .

71 76 86 96 100 101 114

I. Introduction: Fundamentals of Motor Organization

Skeletomotor activity is the means by which the organism interacts with the surrounding physical world. It is the instrument which enables the organism to attack, defend, flee, secure food, and counteract disturbing factors such as wind and water turbulence; to move to different environments; to mate, play, communicate, and support autonomic functions such as respiration, micturition, defecation. A number of attempts have been made to classify motor activity. The most simple, and certainly most unsatisfactory, classification is one which tries to distinguish between voluntary and involuntary movements. In animals, this distinction fails for obvious reasons, and even in man this classification is useless when one considers such phenomena as compulsive neurotic behavior or the ability of some individuals voluntarily to induce myosis. Another classification distinguishes between pyramidal and extrapyramidal motor activity. Some authors assume that the voluntary activity is conveyed from the cerebral cortex to the anterior horn cell (or its homolog in the brain stem) via the pyramidal tract and that involuntary activity is organized by the extrapyramidal system. The negative characterization of the greater part of motor activity is one 71

72

WERNER P. KOELLA

important drawback of this classification. Misconceptions in the anatomical characterization constitute another disadvantage: the corticospinal tract is referred to as “pyramidal” because it passes through the medullary pyramid, and not because some of its fibers are axons of Betz’s giant pyramidal cells. A number of fibers originating in the classical precentral motor cortex never reach the medullary pyramid, as they make connections in the motor nuclei of the upper brain stem. The same is true of certain corticifugal fibers which interrupt in the reticular formation, bypass the medullary pyramid, and reach the spinal segment. Furthermore, there is no “pyramidal” motor activity which does not involve a large number of “extrapyramidal” structures as well as numerous sensory feedbacks. Lower animals have no pyramidal tracts and yet they are capable of producing movements which could be classified as homologs to the “pyramidal” activity in higher animals. Finally, young animals surgically deprived of the classical motor area still develop motor skills comparable to unoperated control animals. The most satisfactory and most physiological classification is the one introduced by W. R. Hess (1941, 1942a,b, 1943). This classification offers the great advantage of being entirely unencumbered by any anatomical considerations. It takes into account that a particular area of the neuromuscular system can be involved in different types and aspects of motor activity. Hess distinguishes between “teleokinetic” and “ereismatic” motor activity. The first type, whose name is derived from the Greek word “telos” (meaning “aim,” “goal”), includes all those movements, voluntary, automatized or involuntary, which deal with directed, aimed, skillful movements, and are usually under visual, proprioceptive, and tactile control. The second type, derived from “ereisma” (meaning support, foundation), encompasses all those reflex and, occasionally, voluntary motor activities which give “dynamic” support for the “teleokinetic” activity. The “ereismatic” activity ensures an adequate starting posture and position, and it compensates for all disturbing factors such as recoil, change in internal and external resistance, etc. The functional success of all motor phenomena depends on their precision in intensity, direction, and time. Precision depends upon proper adjustment of all force vectors involved. Since the muscles are the only independent “variables” in this mechanism, motor coordination can be referred to as the process which controls the

SUBCORTICAL MOTOR AREAS

73

contractile force of every muscle participating in a motor act. The task of the motor coordination is complicated by several factors: (1) The multiplicity and unequality of the muscular vectors. Every movement, even the most simple one, is the manifestation of activities in many muscles. The intensity-time component of each muscle differs from that of every other muscle. ( 2 ) The lack of constancy in the physical conditions at the beginning of the performance of a movement. For example, when lifting the arm in a sagittal plane, the necessary muscular forces involved change with a change of position of the body in the gravitational field. ( 3 ) The lack of constancy in the physical conditions during the movement. Variable external and internal resistances and inertia have to be overcome, and the vectorial effect of every muscle changes constantly with a change of topographical relation of origin, point of insertion, and center of rotation in the joint. (4) Every change in posture (i.e., relative change in the spatial relation of the different parts of the body) induces a shift of the center of gravity of the whole body which makes it necessary to introduce compensatory forces, changes in the supporting mechanisms, and changes in position. The degree of tension produced by a particular motor unit (i.e., the total number of muscle fibers innervated by one particular motoneuron) is controlled by the degree of activity in the anterior horn nerve cell (or its homologs in the brain stem). The total force produced by a particular muscle at any given moment is determined by the momentary integrated activity in the nerve cell pool innervating this muscle, and, reversely, the activity in this nerve cell pool at any given moment reflects the momentary force produced by this muscle. Consequently, the momentaiy activity in all the anterior horn cells (and their homologs) reflect the compound force (in its intensity-space distribution) of all the muscles of the body at this moment. In a topological sense, motor activity in its space-intensitytime stnicture is the reflection of the space-intensity-time structure of activity in all motoneurons. Since the activity patterns of motoneurons are largely dependent upon other segmental and suprasegmental neurons, one may generalize and state that any motor activity induced by muscular action is a reflection of a particular central nervous pattern of activity defined by its intensity and its spatial and temporal structure. We may place an electrode into a pool of nerve cells having

74

WERNER P. KOELLA

mono-, bi-, oligo-, or polyneuronal connections with a muscle or group of muscles, and apply an electric stimulus. This procedure produces a focus of excitation of a certain intensity, space, and time structure liable to induce an overt motor response. The distribution of the muscles thus activated will inform us about the functional connections between the area stimulated and these muscles. The nature of the movement induced allows us to learn about the physiological role of the stimulated structure in the realm of biomotor organization. A closer analysis of the time course of this motor effect (preferably with moving pictures) gives us the opportunity to differentiate between primary effects (for instance, bending of a leg) and secondary effects (for instance, compensatory movements and readjustment of the body equilibrium by vestibular reflexes ) . The classical concept of central motor representation is one of a point-to-point representation. If the precentral gyrus in man, or its homolog in lower forms, is stimulated, one finds indeed that each “point” (i.e., small area) appears to have a relatively direct connection to one particular muscle and that it controls quantitatively the action of this muscle. As destruction of parts or of the whole of this area (in adults) interferes permanently and severely with voluntary motor activity, it is generally thought that some “higher planning instance” uses this “keyboard to organize purposeful, skillful,, and goal-directed movements. A similar point-to-point principle has been found in the so-called supplementary motor areas of the cortex and in the cerebellar cortex (see Woolsey et al., 1950; Hampson et al., 1952). Such point-to-point organization also exists in the anterior horn of the spinal cord and in the homologous structures of the brain stem. As this principle of a point-to-point relation is easy to understand, it has been taken for granted that this principle is, or must be, generally realized throughout the central nervous system. Yet, there is ample, although not amply realized, evidence that there are central nervous areas which reveal different types of organization; types of organization which are recognized not so much by their clean-cut anatomical relationship as by their functional relationship. Such evidence stems from stimulation of loci in the central nervous system where the effect is measurable not in intensity of contraction of one particular muscle but rather in intensity, direction, and time course of compound movements (e.g., in degree and direction of head, eye, or whole body movements). Indeed there are examples

SUBCORTICAL hlOTOR AREAS

75

where it was found that the stimulus-induced activity may shift from one muscle group to another with a change in the momentary physical condition but that the end effect of the motor activity stays the same. For example, a stimulus in a particular area of the brain stem may induce a turning of the head to the side, but this movement is “taken over” by the eyeballs if the head is fixed in an immobile position. The functional result, a deviation of the optical axis to the side, is the same in both cases. Another example of such compound organization can be seen by stimulation in a particular area of the rostra1 brain stem, which not only induces urination but also the adequate positioning of the body. The fact that micturition alone (i.e., without adequate position) is produced from other areas strongly suggests that activation of urine release and adequate posture are represented as a “package deal,” involving a large number of muscular activities. There is no question but that a simple point-to-point relation is not the guiding principle. Finally, it should be mentioned that some parts of the central nervous system do not exert a control on movements directly, but rather indirectly in the sense that they may quantitatively modulate motor acts. We have come to refer to these structures ;is general inhibitory or facilitatory areas. It is the aim of this paper to describe the motor effects induced from a number of extracortical cerebral areas in which the organizational pattern deviates from that of the simple point-to-point representations. The results of central nervous stimulation will be described in the following sections according to anatomical subdivisions. This subdivision is, howcver, merely a temporary dissection, dictated by the classification made by the different authors who have chosen to investigate certain brain areas instead of certain types of movements or the “functional development” of motor phenomena while passing from one area to another. The last section oE this paper will demonstrate that for a majority of motor effects functional correlations and interrelations can be traced throughout almost the whole extent of the suprasegmental neuraxis. A final word in this introduction should refer to the methodological aspects of the studies. The various investigators have used a number of different techniques to elicit and to record their effects. Some have utilized electrical stimulation as a study tool while others have used the technique of focal destruction. Some have stimulated

76

WERNER P. KOELLA

the brain of anesthetized cats, others have used decerebrate animals. Only a few investigators have done the experimental work in nonanesthetized, freely moving animals with either chronically or acutely placed electrodes. It seems to us that it is the latter method that offers the best insight into the effect of any focal disbalancing procedure. There is only one disadvantage of this method, namely that a particular motor effect induced by a local electrical stimulus may call forth secondary and even tertiary motor effects which have no direct relation to the stimulus. Proper observation techniques utilizing moving pictures and g o d understanding of the dynamics of motor activity usually allow a clean-cut separation of such direct and indirect phenomena. In the final analysis, the disadvantage of the occurrence of, such “alien” phenomena is by no means as serious a source for erroneous conclusions as are the artifacts resulting from the “restraining effects of anesthesia, transections of the neuraxis, and the use of mechanical devices such as the Horsley-Clarke frame, etc. The primary value of these latter methods is to provide supplementary evidence such as that of nervous pathways (by electrical recordings) or the effect of the elimination of “higher centers,” etc. 11. Motor Effects Produced by Stimulation of the Diencephalon

Electrical stimulation of the diencephalon in the unanesthetized, freely moving animal elicits a number of motor phenomena. These effects are conveniently subdivided into two groups: ( 1 ) motor effects related to the processes of adjusting the position of the body in space and of adjusting the positions of the different parts of the body to each other (referred to here as posture; the effects of the facial muscles are also included), and ( 2 ) motor effects which are interpreted as auxilliary processes involved in a number of mechanisms of the autonomic sphere (referred to as supporting effects). The writer is convinced that he does no injustice to the work of many good investigators in this field, if he puts special emphasis on the work of W. R. Hess, since this author’s contributions are very extensive and most carefully analyzed. Hess’ findings were obtained by stimulation of approximately 4000 different points in the diencephalon and its surrounding areas during the course of about 400 experiments in cats. This work is described in a monograph (Hess, 1949), and in shorter publications (Hess, 1940; Hess and Weisscheddel, 1949; Hassler and Hess, 1954); the reader interested in

SUBCORTICAL MOl‘OR AREAS

77

more detail is referred to these original publications, With regard to Hess’ methodology, we may mention that d.c. pulses of 12.5-msec duration, variable voltage, at rates of from 2 to 15 pulses/sec, were used. Acutely inserted steel wires of 0.25-mm diameter served as electrodes. The following is a brief summary of Hess’ findings: One class of motor effects, which we may call “directional,” comprises movements of the head, accompanied as often as not by synergistic movements of the forebody or even of the whole body. Lowering of the head in a rhythmical movement synchronous with the stimulus frequency (between 2 and 15 pulses/sec) is one of these directional effects. The phasic component is superimposed upon a tonic one, as one observes the head to reach the level of the experimental table in 3 to 9 strokes. This lowering of the head is often, if not always, accompanied by a lowering of the forebody with bending of the forelegs. Often a “compensatory” raising of the hind quarters is seen under these conditions, which makes the lowering of the rostra1 parts even more pronounced. No rebound was ever reported to occur. A second type of effect is the exact mirror image of the first one. In a few “phasic strokes” the head ends up in a highly elevated position with pronounced (upward) rotation around the bitemporal axis. Often this head movement is accompanied by a rise of the forebody induced by an extreme stretching of the forelegs. Whenever this terminal position is reached, the head still shows a small synchronous (to stimulus frequency) up-and-down movement. With stronger stimuli, however, the up and backward tonic component may continue. The animal may then start to fall backward, but it performs a rotation of the whole body to land safely on its forefeet. This later phenomenon is evidently a secondaiy, corrective effect. Whereas these two types of movements constitute displacements in the midsagittal plane (or around a bitemporal axis), a third effect demonstrates displacement in a frontal plane or around a longitudinal axis. Uniformly upon stimulation the side of the head opposite to the stimulated side moves upward in this rotational effect. Again the synchronous phasic component is superimposed upon the tonic one. Frequently displacements between the forebody and the head of as much as 45O occur under these conditions. Also, as already described in the first two effects, the forebody may perform a synergistic rotation to such an extent that the head eventually

78

WERNER P. KOELLA

shows a total displacement of almost 180”. With slightly stronger stimuli, the animal may end up by rolling completely around the longitudinal axis (toward the stimulated side). The eyes, as may be easily observed in the slitlike pupils, maintain their proper position with respect to the head. If, however, the head is passively redressed from its rotated posture toward its normal position in space, the eyes do not follow this return rotation (as long as the stimulus is effective) but rather maintain their rotated position in space by performing a rotatory movement within the orbit. A fourth and a fifth type of effect constitute movements of the head in the horizontal plane or around a vertical axis. There are the turning movements to the stimulated (ipsiversive turning) and the one to the contralateral side (contraversive turning). In the ipsivemive turning effect the head is bent toward the stimulated side and the effect may include an extreme synergistic curvature of the body, so that the head touches the hind part of the body. With slightly stronger stimuli a narrow ‘‘manAge” movement may ensue. These ipsiversive horizontal movements differ in other aspects from those in the midsagittal and frontal planes. While the latter ones always start immediately with the beginning of the stimulus, the ipsiversive movement does so only after a delay of 1 to 3 seconds (as verified on movie film). Furthermore, the head-lowering, the head-raising, and the rotational eftects do not show an after-effect, whereas the ipsiversive turning effect typically shows a “positive after-discharge,” as the animal remains in the curved posture for several seconds after stopping of the stimulus. Finally, with frequencies of stimulation of 8 pulses/sec or even lower, the character of the ipsiversive movement is always smooth or “flowing” and no “phasic” synchronous strokes are observed as is the case in the sagittal and frontal rotation. The contrauersive turning shows typically an even longer latency of up to 30 seconds, and the positive after-effect is also prolonged as compared with that of the ipsiversive turning. Often these effects terminate in a (contraversive) mankge movement, whose radius is, however, considerably larger than that in the ipsiversive manbge. These movements are characterized also by their flowing progression with no evidence whatsoever for synchronous jerks. In addition to these “pure” phenomena, mixed effects are also observed, such as ipsiversive turning and lowering of the head. In

SUBCORTICAL MOTOR AREAS

79

all these mixed effects the characteristics of the single component, such as constant “flowing” of the turning component (to the side) and synchronous jerking superimposed on the tonic or flowing component are preserved. The comparison of a particular motor phenomenon and the location of the stimulating electrode ( as ascertained by histological examination) reveals a striking correlation between direction ( or plane) of movement and one particular area in the diencephalon and its adjacent areas (Fig. 1).Hess, in a first analysis, directed his interest mainly to gross “fields” with no further relation to particular nuclear or fibrous structures. On further analysis of the data Hess and Weisscheddel ( lY49), Hassler and Hess ( 1954), Hassler (1956a, b ) , and Jung and Hassler (1960) were able to correlate certain diencephalic (and paradiencephalic ) structures with certain stimulation effects as defined by their relation to the three-dimensional plan of the body. Thus it was found that all the points from which electrical stimulation induced lowering of the head (and, with increased stimulus intensity, of the forebody) were located in or close to the fiber tracts which run through the posterior commissure. The same functional system also was found to include elements in and around the interstitial nucleus of Cajal. On the other hand, a raising of the head, possibly accompanied by a raising of the forebody and often leading to a backward falling, was produced from points in the rubrospinal tract including its magnocellular area of origin in the red nucleus. Rotation of the head with a synergistic twisting of the forebody and a rolling over around the longitudinal axis (stimulated side downward) was obtained from points in and around the brachium conjunctivum. This fiber tract was interpreted as being the afferent system to the diencephalon for rotation, whereas the efferent, or descending, part of it was found to be the tractus interstitiospinalis. Stimulation of points located in the subthalamic area between the mammillothalamic tract and the medial border of the pedunculus led to contraversive turning, whereas stimulation of the f asciculus tegmentis produced ipsiversive turning. More will be said about these morphological-functional relations when the effects of stimulation in other areas (topographically close as well as remote to the diencephalon) are described. A second class of effects observed by Hess concerns the foreleg, contralateral to the stimulated site. Three different types of move-

80

WERNER P. KOELLA

ments can be distinguished: (1) A “pulling up” of the foreleg, demonstrating a coordinated cooperation of the shoulder, elbow, and wrist, The movement is of the “tonic” type, i.e., little or no synchronous jerking is observed. These effects are produced with stimulation in the pars arcuata of the thalamic posteromedial area. (2) A ‘lifting up” of the contralateral foreleg, exhibiting a marked adduction of the upper arm with positioning of the stretched forearm and paw horizontally forward. There is marked synchrony with the stimulus, as with each pulse the arm is moved upward while it drops again to the horizontal during the pulse interval. Studies in localization suggested that the ventromedial part of the ventral thalamic nucleus is involved. (3) A sideward “throwing-up” of the foreleg. As the rhythmicity of the movement is very pronounced, Hess compares it to the wing movements of a flying bird. The loci of stimulation in this type of effect are found in a similar area as in type ( 2 ) , i.e., in the ventromedial thalamus with an extension into Forrel’s H-fields and along the fiber tracts descending from there into the midbrain. A third class of movements concerm effects in the face mostly involving the pinnae, the eyelids, and the whiskers. A closer analysis reveals that the effects on the pinnae and vibrissae also exhibit “directional” trends. Thus, a typical effect is a backward bending of the contralateral combined with a foreward movement of the ipsilateral vibrissae. In other cases the whiskers of both sides are moved forward and downward or-in yet other experiments-hackward FIG. 1. Survey of responses related to head and body, and their location represented in four horizontal planes. The responsive areas are represented by solid lines, broken lines, or circles. The responses are indicated by symbols plotted in the center of the region from which they were evoked. Note the distribution of the fields, with partial overlapping in horizontal and frontal planes. Perpendicular solid lines : sinking; perpendicular broken lines : elevation; circles: rotation in the frontal plane; horizontal solid lines : ipsiversive deviation in the horizontal plane; horizontal broken lines: contraversive deviation in the horizontal plane. The sections lie one above the other with a distance of 1.5 mm between, enabling dimensional representation. Aq., aqueductus sylvii; C.U., commissura anterior; C.f.d., columna fomicis descendens; C.i., capsula interna; C.p., commissura posterior; I., infundibulum; N.e., Nucleus entopeduncularis; Nc.r., nucleus ruber; P.c., pedunculus cerebri; S .g.c., substantia grisea centraliq mesencephali; Tr. M., tractus Meynert; Tr.o., tractus opticus; V . d A . ,fasciculus Vicq d‘Azyr; III., ventriculus 111. (From W. R. Hess, 1949).

SUBCORTICAL MOTOR AREAS

I

O o $ 0 'O

0 - 0 0

0 0

81

82

WERNER P. KOELLA

and upward, Similar combinations are observed on the pinnae. An effect found very frequently is a narrowing or complete closing of the contralateral orbital fissure. All these effects in the face appear either as rhythmic (Hess refers to them as intermittent) movements, synchronous with the stimulatory pulses, or, in other cases, as tonic (Hess’ “summating” effect) activity. In the phenomena of the second type, but not of the first, a short (1 second) after-effect is usually observed. The stimulation points producing “summating” face effects are found in a lateral anterior area in the thalamic radiations, whereas the rhythmic effects are obtained from a more medial and ventral area of the thalamus extending from the bundle of Vicq d’Azyr and back toward the tegmentiim. The diencephalic and paradiencephalic effects described hitherto will be discussed in their functional significance in a later section. It should suffice here to mention that all these movements appear to be correlated directly to some interactive behavior of the organism with its surrounding physical world. This interpretation leads to the separation of this group of movements from a second group which may be referred to as auxiliary mechanisms involved in autonomic functions, or skeletomotor-autonomic integrations. The best example here (although usually not obtained from the diencephaIon proper) is the typical and adequate posture induced in those cases in which the stimulus elicits either micturition or defecation. Typically, the animal always assumes the correct posture shortly after the onset of stimulation, and only after a latency of about 10 seconds does it start to expel the feces or the urine. With respect to localization it is interesting to learn from Hess’ observations that micturition and defecation are produced from an area extending from the posterior lower septum pellucidum through the area preoptica, the anterior hypothalamus down to the central gray of the midbrain. As a rule, only the points in the septum yield the autonomic response together with the adequate posture, whereas points in the hypothalamic and adjacent areas elicit only the autonomic act. It may be noteworthy that it is not possible to make a distinction in localization between the bladder and the rectum effect. As a matter of fact, it is Hess’ impression that either or both of the effects are produced dependent upon the momentary degree of filling of the respective reservoir. A number of other skeletomuscular effects occurring together

SUBCORTICAL MOTOR AREAS

83

with autonomic mechanisms should be mentioned. Stimulation of points in the supraoptic area, in an area lateral to the mammillary bodies and, occasionally, in the ventral nucleus of the thalamus, evokes vomiting as well as the proper coordination of the skeletomotor apparatus, such as lowering of the forebody and the head and the contraction of the abdominal wall muscles. From points in the ventral nucleus of the thalamus stimulation induces licking movements which are often accompanied by chewing movements. A number of other authors have also investigated the motor function of the diencephalon although in less detail than Hess. Hunter and Jasper (1949) have also worked on unanesthetized and unrestrained animals which were prepared chronically. With the stimulating bipolar electrodes in the intralaminar nuclei, and with “moderate intensities” of stimulation, bilateral twitching of the face and “particularly” blinking of the eyes were found. If the stimulus intensity was elevated, bilateral clonic movements of the forelimbs and of the body would “sometimes occur.” Typically, the rhythm of the movements followed the pulsations of the stimulus at lower frequencies, whereas at higher stimulus frequencies the motor effects frequently performed their own slower beat. In the animals, in which the electrodes were placed in the center median or in the subthalamic region, tonic elevation of one forepaw and turning of the head and eyes were induced. At higher intensities the turning of the head was followed by a bending of the forebody and finally a circling movement. The authors stated that turning was sometimes ipsiversive but “more commonly” contraversive. These authors then confirmed with a different experimental method (bipolar versus monopolar stimulation, chronic preparation versus acute preparation) a number of Hess’ findings. Their results corresponded, to cite the authors’ statement, “closely to those of Hess.” Hinsey ( 1940) performed experiments on cats lightly anesthetized with Nembutal and suspended by strings passing beneath the supraspinous ligaments. Electrical stimulation in these studies was done by means of bipolar concentric electrodes guided by a HorsleyClarke instrument. In a number of experiments the animals had been subjected to hemi-decortication 2 months prior to the physiological testing. This interval was certainly more than sufficient for complete degeneration of all corticifugal fibers and thus for the

84

WERNER P. KOELLA

elimination of the possibility that the diencephalic stimulus would stimulate corticifugal fibers passing by or through diencephalic structures. A few characteristic examples from Hinsey’s findings on such hemi-decorticate preparations are given below. In one cat stimulation of the (left) medial part of the thalamic ventral nucleus caused a slight flexion of the right ( contralateral) forelimb. A stimulus applied to the mammillothalamic tract yielded a stronger flexion of the contralateral forelimb accompanied by an abduction of the hind limb. From a point in the H-field of Forel, adduction of the contralateral forelimb was obtained, whereas from the region of the thalamic fasciculus abduction and flexion of both contralateral extremities were elicited. In the same preparation stimulation of a point at the dorsal surface of the posterior hypothalamus in the H-field of Forel ( 2 mm backward from the one point mentioned above) produced flexion of both right ( contralateral) limbs, abduction of both left limbs with flexion of the left hind limb, and curving of the body to the left (left side concave). In another preparation extension of the ipsilateral forelimb and bilateral running movements were found as a result of stimulation of the right paraventricular nucleus. A similar effect was obtained from the right pei-ifornical area, and from the region of ventromedial hypothalamic nucleus and the lateral hypothalamic area. These points were situated on the right (unoperated) side. Stimulation in left entopeduncular nucleus caused flexion of the contralateral and moderate extension of the ipsilateral limbs. At the same time the body was curved to the left. Phasic bilateral running movements were obtained in this preparation upon stimulation of both the normal and hemi-decorticated sides. Although obtained under very diverse experimental conditions, there is ample evidence of similarity in Hess’ and Hinsey’s findings. Thus, there is raising of the contralateral foreleg from stimulation of the ventral nucleus of the thalamus and the mammillothalamic tract. Hess points to the qualitative difference and a participation of the hind leg in the second case. The almost entire lack of hind-limb effects in Hess’ material may be explained by the fact that these animals were using the hind limbs for support whereas Hinsey’s animals were suspended. In the H-fields of Forel, Hinsey produced in addition to limb effects a curving of the body to the ipsilateral side, which was seen in Hess’ material upon stimulation of the area

SUBCORTICAL MOTOR AREAS

85

lateral to Meynert’s tract in Forel’s tegmental fasciculus (Haubenbundel) . On the other hand, Ilinsey did not report rotational effects on the head nor any raising or lowering types of movements seen often in Hess’ material. This difference is understandable as evidently Hinsey’s animals were not able to move the head at all. The work of Ingram et nl. (1932) will be discussed in detail later on, when we deal with the so-called tegmental response. HOWever, these authors occasionally stimulated the subthalamus of etheranesthetized cats and found the following effect: Upon stimulation with faradic pulses, there was curving of the head, the neck, and the trunk toward the stimulated side. This activity was accompanied by a flexion of the ipsilateral forelimb and extension of the contralateral forelimb, while the hind legs performed “varying” movements. It will be seen later in this paper that this effect concurs very closely to the classical description of the “tegmental response.” It should be mentioned here that Hess observed ipsiversive turning from a similar area. Waller ( 1940), stimulating the subthalamus, observed locomotor activity, and it is of interest to read in a recent paper by Grossman (1958) that this activity elicited from the subthalamic nucleus is inhibited by simultaneous stimulation of the unspecific thalamic nuclei. Grossman emphasized that locomotor movements from the subthalamus were obtained only if stimulated at a lower level of this area (namely the H, and H2 fields of Fore1 and the zona incerta); if stimulated 1 to 2 mm above this level the typical “tonic” tegmental response was elicited. This latter one was found to be independent of concomitant stimulation in the unspecific thalamic nuclei. Godlowski ( 1938) stimulated various sites in the diencephalon (and mesencephalon) and found (while the heads of animals were fixed and the animals lightly anesthetized with ether) systematic conjugate eye movements. In a number of more recent papers, Hyde (1957), Hyde and Eason (1959), and Faulkner and Hyde (1958) reported about a series of experiments dealing with eye movements induced by electrical stimulation of the brain stem. In the present context it is of interest to make reference to one group of those authors’ findings: stimulation (in the enckphale is016 cat) of the posterior diencephalon ( Horsley-Clarke planes A-7 to A-10) “yielded many conjugate eye movements.” These were mostly contraversive

86

WERNER P. KOELLA

movements often combined with an upward component. In the zona incerta and the prerubral capsule, rotations of the eyeballs were obtained. Ill. Motor Effects Produced by Stimulation of the Cerebellum

The cerebellum has since long been recognized to play an important role in the motor behavior. The initial impetus for study of this area was given by the simultaneous yet independent observations of Sherrington (1897) and Lowenthal and Horsley (1897) that stimulation of the cerebellar anterior lobe induces a reduction of decerebrate rigidity, A similar effect of such stimuli was also observed upon phasic reactions of limb muscles ( Denny-Brown et al., 1929). A considerable number of studies followed these early investigations, culminating in the excellent and exhausting analysis by Moruzzi and his collaborators (Moruzzi, 1950a, b; Moruzzi and Pompeiano, 1956, 1957; Dow and Moruzzi, 19%). Such phenomena may have to do with a kind of a modulatory function of some parts of the cerebellum, a function similar to that established for some areas of the reticular formation, By their very nature these effects can be demonstrated best if one works on some “background activity, such as decerebrate rigidity ( i.e. exaggerated myotatic reflexes) or some phasic phenomena such as the knee jerk, the flexor reflex, the crossed extensor reflex, and many others. We turn our attention rather to a second group of effects which we may call “primary” and which consist of actual movements often initiated against the “normal” background of tonus distribution throughout the body. Whether there is an overlap between the group of the modulatory and that of the primary effects is hard to decide. The marked difference in the appearance of the evoked movement (natural, purposeful movement of the limbs, the head, and the body) in the second class, in contrast to the often generalized inhibition followed by a rebound, as described in the first class, certainly justifies a separation of these groups. We all remember the classical controversy as to whether the cerebellum works “as a whole” (Sherrington, 1900) or whether it shows a somatotopic organization (Bolk, 1906). It seems that the “primary” effects we are going to describe constitute a third principle of organization to be considered, since they suggest that at least parts 04 the cerebellum “think” in three dimensions.

SUBCORTICAL hiOTOR AREAS

87

Overt movements by localized stimulation have been observed by a number of investigators, in a variety of experimental conditions, and from different cerebellar structures. Some experimental work has been done in nonanesthctized unrestrained animals with either chronically or acutely implanted electrodes. Cortical as well as subcortical cerebellar structures were stimulated. The first to report on such studies was Lewandowski (1903). This author employed a technique developed by Ewald to stimulate the cerebellum in freely moving dogs. He described some body effects (incidental to other results of no interest in this contest) evidently produced from the paramedian posterior aspect of the cerebellum. For one of the really outstanding pieces of work on electrical stimulation of the cerebellar cortex in nonanesthetized animals we are indebted to Clark (1939). This author implanted electrodes to rest on the cerebellar cortex in chronically prepared cats. The (anatomically verified) sites of stimulation covered the whole of the posterior vermis, the paramedian lobes, crura I and I1 as well as the upper part of the anterior vermis. It is regrettable that the author used only a 60-cycle sine-wave current to stimulate, since the present author has demonstrated that not only voltage, but frequency, wave form, and pulse duration are of utmost importance for the determination of an effect produced by cerebellar stimulation ( Koella, 1955). Nevertheless, Clark‘s experiments and interpretation have added considerably to our understanding of the physiology of the cerebellum. Clark found that the motor phenomenon usually elicited by cerebellar cortical stimulation consisted of three phases: a first one coincident with the stimulus, a second (and usually larger) one of opposite directional sense (“rebound”), and a third one (“aftereffect”) which consisted of slow movements of the head, limbs, body, and tail lasting for several minutes. The form (i.e., direction and distribution over the different parts of the body) of the first and second phases, and to a lesser extent, of the third phase, depended upon the locus of stimulation. With stimulation in points in, or close to, the midline in the upper anterior lobe, the head was lowered and sometimes” protracted and in the rebound phase went up and occasionally backward. Back of the primary fissure the reverse picture was observed; head high during stimulation and low during the rebound phase. From points in the paramedian lobes “

88

WERNER P. KOELLA

a different movement was elicited. With the start of the stimulus the cat leaned toward the contralateral side and lifted the ipsilateral forepaw. A marked rebound or long after-effects were not found in these locations. From crus I, i.e., in its paravermal regions, the effects were similar to those obtained in the vermis itself, with some differences during the long after-effects. The same holds true for the effects from stimulation in crus 11. From the “‘most posterior exposed portion of the vermis” (evidently the pyramis ) and sites slightly lateral to the midline, stimulation produced leaning toward the ipsilateral side with “uncovering” and abduction of the contralateral hind limb. During the rebound the animal produced the “mirror-image” of the stimulation effect. Stimulation exactly in the midline of this area induced a sinking to the floor with the claws extended. Since the long after-effects observed by Clark (and others) are not particularly pertinent to the present review, we refrain from a detailed description and the reader interested in the topic is referred to Clark‘s original paper. McDonald (1953) studied the effect of stimulation of the anterior lobe of the cerebellum in cats which were chronically prepared in a manner similar to that of Clark. The stimulus parameters were also identical to those used by Clark. From most of the stimulated points studied, McDonald elicited effects on the thighs, legs, tail, neck, and face. Only in a few instances “directional” effects of the head and body were described. The author put much emphasis on a somatotopic arrangement in the anterior lobe where tail, hind limbs, forelimbs, neck, and face were found to be represented in an anterior-posterior direction of the anterior vermis (Fig. 2 ) . Furthermore he found that more proximal muscles were “localized in a more medial region, whereas distal muscles were represented in more lateral aspects of the vermis. McDonald confirmed (except for some minor points due mainly to differences in anatomical nomenclature) the work by Hampson et al. (1952), and also lent support to the findings of Adrian (1943), Dow (1939), and Snider and Stowell ( 1944), who found somatotopic sensory representation in this area. It is of interest to note the phenomenological difference between the effects induced (by similar methods) from stimulation of the anterior vermis and those from stimulation of the posterior vermis (including the uppermost parts of the anterior

SUBCORTICAL hlOTOR AREAS

89

FIG.2. Diagram of anterior lobe of cat’s cerebellum. This specimen had 12 folia posterior to fissura postcentralis. Symbols on diagram indicate region in which primary response took place when point indicated on cerebellar cortex was stimulated. The two points surrounded hy a circle, and numbered 10, 12, 13, and 14, indicate a bipolar electrode, whose tips were 1 mm apart. Folia are numbered consecutively from primary fissure forward. (From J. V. McDonald, 1953.)

lobe). While the latter are characterized by their “directional” aspects, i.e. up, down, and sideward effects, the former are characterized by their somatotropic organization. “Primary” motor effects have also been produced by electrical stimulation of the interior of the cerebellum (i.e., white matter and basal nuclei). The first investigator to perform such experiments in nonanesthetized (and anesthetized) normal animals was Chambers ( 1947). Cats were prepared chronically with indwelling tantalum electrodes (the exact locations of which were verified histologically) and a 60-cycle sine-wave current of adjustable strength was used as stimulus. Chambers subdivided the motor effects obtained in nonanesthetized animals into two general “patterns,” I11 and IV (pattern I and I1 referred to effects obtained under anesthesia). At threshold stimulus strength, pattern 111, obtained from the fastigial nuclei, the buried cortex, and the white substance of the vermis and

90

WERNER P. KOELLA

the medial parts of the nucleus interpositus, consisted of “one or more of the following: flattening (abdomen flat on flour with legs flexed) ; leaning or falling toward homolateral or contralateral side, or backward or forward with immediate recovery and, in some instances, movements opposite in direction while the stimulus was still being applied; oscillation of the head; swaying of the entire body.” The oscillations of the head were found to be in any plane. Pattern IV, also at threshold, was obtained from points in the lateral portion of the nucleus interpositus, the dentate nucleus, and the buried cortex and white matter of the parafloccular, ansiform, and paramedian lobes, as well as from three points in the vermis. This pattern was characterized by slow tonic movements which consisted of “one or all of the following: movements of the head toward the homolateral side, rotation of the head with the vertex down and toward the homolateral side and nose up and toward the contralateral side, flexion of the homolateral foreleg, and closure of the homolateral eye.” In contradistinction to pattern 111, no after effects were observed in pattern IV. Sprague and Chambers (1954) stimulated the interior of the cerebellum utilizing chronically implanted electrodes in nonanesthetized normal cats. They described their effects as follows: “The animal, if lying down, lifted its head and turned it toward the stimulated side, righted its fore quarters, then its hind quarters and circled toward the stimulus.” From these and other results the authors concluded that there were two “primary types” of cerebellar postural influences: a first one obtained from the vermian cortex of the anterior lobe, the fastigial nucleus, and the cortex of the pyramis and uvula, characterized by inhibition of the ipsilateral extensor muscles, facilitation of the contralateral extensors, and inhibition of the contralateral flexors; and a secmd one obtained from the lateral anterior lobe, the paramedian lobes, and the nucleus interpositus, characterized by facilitation of the ipsilateral extensors and the inhibition of the ipsilateral flexors. An extensive series of experiments involving stimulations of loci in the basal cerebellum of unrestrained unanesthetized cats was performed by the present writer (Koella, 1955). This series appears to be of particular value in the context of this review as the experiments were performed with a technique similar to that used by Hess, with respect to preparation of the animals, the positioning of

SUBCORTICAL MOTOR AREAS

91

the electrodes, and the shape, range of duration, amplitude, and rate of the electrical pulses. Thus these results seen in the cerebellum are directly comparable with those seen by W. R. Hess in the diencephalon. In these experiments a number of motor phenomena were observed whose character, intensity, and duration evidently depended upon the site, frequency (from 2 to 300 pulses/sec), intensity, and pulse duration of the stimulus. With the low frequency range ( 2 to 7 pulses/sec, comparable to Hess’ range ) the most impressive effects concerned the head and forebody. Four different types of effects could be distinguished (Fig. 3 ) . The first was a rotation of the head, i.e., a movement around the longitudinal axis, which consistently was performed with the stimulated side upward. In a second group of

FIG.3. Four types of effects induced by stimulation of loci in basal cerebellum of cats. Upper left: Stimulation with 0.6 volts, 20 cycles/sec, 10 msec pulse duration, in lower restiform body 2.4 mm from midline right side (coordinates: see C in Fig. 4 ) . Note isolated rotation of head right side up. Upper right: stimulation with 0.7 volts, 60 cycles/sec, 10 msec pulse duration, in upper restiform body 2 mm from midline on right side (coordinates: see C in Fig. 4 ) . Note rotation of head right side up and marked concomitant leaning of body to left side with stretching of right foreleg. Lower left: stimulation with 1.75 volts, 60 cycles/sec, 5 msec pulw duration in nodulus a t midline (coordinates: see A in Fig. 4). Note marked lowering of head, forefront touching table. Lower right: stimulation with 1 volt, 4 cycles/sec, 12.8 msec pulse duration, in lower part of anterior lobe, 2.8 mm to left (coordinates: see D in Fig. 4). Note marked ipsiversive turning of head. Cat’s body stays in normal position. (From W. P. Koella, 1955.)

92

WERNER P. KOELLA

animals the head movement consisted of a rhythmical lowering, i.e., a movement around the bitemporal axis. The third and fourth groups showed movements around the dorso-ventral axis leading to ipsiversive or contraversive turning. In some animals mixed effects were observed; thus, for instance, rotation would be combined with a turning or with a lowering of the head. In most of the animals stimulated with low frequencies, the effects on the head were phasic in character. In some other experiments, however, the rhythmical component of the movement was superimposed on a tonic change in the position of the head which occurred in the same direction as the phasic movement. The tonic component was more pronounced in the turning effects than in the rotatory or lowering effects. Furthermore, the “fusion frequency” was lower in the turning movements than in the other effects. Voltages close to the threshold and low frequencies usually produced head movements which appeared as isolated effects. Occasionally, however, the deviation of the head from its normal position was accompanied by a smooth change in the position of the “forebody” (shoulder girdle, chest, forelegs). For instance, a rotation of the head with its right side up would be accompanied by a smooth leaning of the forebody toward the left side and, in some cases, by a lifting of the ipsilateral (right) forepaw. Or a turning of the head to the right would be followed by a curving of the forebody to the right. These coordinated or “synergistic” movements were usually, but not always, related to strong head effects. The rotatory and lowering effects on the head were never followed by any after-effects. Positive after-eff ects, however, were rather common in the turning movements; the head had a tendency to stay in the sideward-turned position for from a few seconds to about half a minute after cessation of the stimulus. If the turning movements were smooth (i.e., in stimulations with slightly elevated frequencies), they usually were performed rather slowly and started only after a latency of several seconds. With a frequency range of 12.-25 pulses/sec, the head movements were completely smooth or “tonic” in nature. With these higher frequencies more and even stronger concomitant synergistic movements were observed. Rotation of the head and the sideward leaning of the body were often followed by a complete rolling around the longitudinal axis; turning of the head was accompanied

SUBCORTICAL MOTOR AREAS

93

by circling or mankge movements. In most cases the voltage threshold was slightly lower in this frequency range than in the 2 7 pulses/sec range. With frequencies of 40-60 pulses/sec and in the highest frequency range (2oO-3oO pulses/sec) smooth head movements around the three cardinal axis-i.e., rotation of the head with the stimulated side upward, lowering of the head, and ipsi- or contraversive turning-were still occasionally seen. Figure 4 illustrates the anatomical location of the electrodes producing these effects. It can be seen that a pure lowering of the

FIG. 4. Mapping of head movements. On left side, anatomical findings for effects are mapped out in four sagittal sections 0, 1, 2, and 3 mm from midline. Shaded areas in right side of figure illustrate fields from which pure effects were obtained. Note that lowering (vertical shading) is obtained from a large area in and close to midline including parts of uvular-nodular lobe, lingula, and fastigial nucleus. Rotation (diagonal shading) occupies a large field laterally to “lowering area.” Contraversive turning (horizontal full line shading) is produced from electrodes in a posterolateral location, whereas ipsiversive turning (broken line horizontal shading) is elicited from an anterolateral field. (From W. P. Koella, 1955).

94

WERNER P. KOELLA

head was almost exclusively elicited from a large area in and close to the midline. “Rotation” occupies a field 1-3 mm lateral of the lowering area, and ipsi- and contraversive turning are elicited from anterior and posterior paramedian fields respectively. In a number of animals, stimulation with frequencies between 2 and 26 pulses/sec produced movements of the facial muscles either isolated from, or in combination with, movements of the head. These effects appeared in the form of twitching of the pinnae, cheek, and vibrissae, and blinking of the eye, predominantly on the ipsilateral side. There seemed to exist a kind of synergism; thus, for example, a rhythmic turning movement of the head toward the right side was often combined with a twitching of the left side whiskers forward and the right side whiskers backward. Stimulation with frequencies between 4 and 10 pulses/sec evoked stimulus-synchronoustwitching of the foreleg, hind leg, skin over neck and chest, shoulder, or the hip, either in combination with the head effects or as isolated movements. If frequencies of 15-26 pulses/sec were applied, these twitchings changed to smooth effects, as did the head movements. Often under these stimulus conditions a lifting of the ipsilateral foreleg with stretched wrists and bent elbow was observed. In this investigation stimulation of many points in the area under study caused smooth movements of the hind leg (mostly abduction and stretching) either in combination with head movements or as isolated activities. In some animals stimulation evoked a sideward leaning of the back part of the body or a sideward falling, and it was observed that a change in frequency of stimulation often induced a change in the direction of the movement. For example, it was seen that a lower frequency (e.g., 20 pulses/sec) elicited a leaning to the right side, whereas a higher frequency (60 pulses/sec) elicited a leaning to the left side, or vice versa. Behavior revealing a strong inhibitory component was frequently observed when relatively high frequencies were used ( W O O pulsedsec) . The onset of this behavioral response began either with an initial lowering of the head, or with a sudden collapse of the whole cat. In either case, the cat was finally lying flat in prone or in side position. The four extremities were sometimes flaccid, sometimes rigid (extremely extended either in protraction, abduction, or retraction). Sometimes the animals remained motionless in this position, sometimes swayed the body or only the head, and some-

SUBCORTICAL LIOTOH AREAS

95

times performed helpless “rowing” movements with all four extremities. These “inhibitory” effects were often followed by strong rebound effects characterized by extreme stretching of the hind legs and raising of the back quarters of the body well above the shoulder level with the tail usually erected vertically. These rebound effects lasted between 10 and 150 seconds. Other of our animals showed typical inhibitory effects but no rebound phenomena. Ranson and his co-workers stimulated the interior of the cerebellum of lightly anesthetized monkeys and of anesthetized and decerebrate cats (Magoun et al., 1935; Hare et al., 1936, 1937). In the monkey these investigators found that stimulation produced gross effects on the body, head (including the eyes), tail, and limbs. Since these aimals were in a semi-restrained situation some finer details may have been lost in these observations. Depending upon the locus of stimulation, one of the following three types of responses was obtained: Stimulation of points within the extent of the cerebellar nuclei and from the adjacent white matter caused a marked conjugate deviation of the eyes and ;L slight turning of the head toward the stimulated side or a turning of head and eyes from a contralateral position toward forward gaze. Stimulation of the medial part of the cerebellum (the white matter bordering the fastigial nuclei and the medial edge of the emboliform nucleus) elicited movements which included the head and eyes and ipsiversive turning of the body which was followed by a rebound in the opposite direction. Finally, stimulation in the buried cortex of the anterior lobe, above and caudal to the fastigial, globose, and emboliform nuclei (the latter two identical with the nucleus interpositus of the cat) yielded flexion, or relaxation from extension, of the ipsilateral forearm, hand, and fingers, followed by a marked rebound extension of these parts after cessation of the stimulus. A second type of effect on the limbs was obtained from a similar area (not extending as far lateral as the first one) and consisted of a “relaxation” during stimulation, which involved all four limbs and was followed by a rebound extension. Very similar results were obtained in anesthetized as well as decerebrate cats. Evidently the upper brain stem (including the red nuclei) was not essential to elicit these reactions. It is of interest to quote a closing sentence of these authors (Hare et al., 1937): “The results of this investigation provide no support for the belief

96

WERNER P. KOELLA

that spec& muscle groups have either an exclusive or preponderant representation within any lobule of the cerebellum.” Sprague and Chambers (1954) in addition to their observations on chronic, normal animals also published some findings obtained in decerebrate cats. These authors found that stimulation of the medial (i.e,, medial of the paravermal vein) vermal cortex of the anterior lobe, the simplex, pyramis, and uvula, as well as the fastigial nuclei, produced ipsilateral flexion and contralateral extension of hind and forelimbs and turning of the head toward the stimulated side. After cessation of the stimulus a rebound of reverse direction took place. From the anterior lobe just lateral of the paravermal vein, from the paramedian lobes, and from the dorsomedial border of the nucleus interpositus, ipsilateral extension and contralateral flexion of the forelegs was obtained whereas during the rebound phase the reciprocal picture appeared. IV. The Midbrain Tectum

A third region of interest for the present discussion is the anterior colliculus of the midbrain. This region has been stimulated electrically and chemically, and experimental results are available from normal as well as from anesthetized animals. Hess et al. (1946) were the first investigators to make extensive and detailed studies on nonanesthetized unrestrained animals using the well-proven technique previously utilized in the studies on the diencephalon to stimulate the tectum of the midbrain. Direct observations (set down immediately) and motion pictures served as recording methods. Relatively low-frequency (2-12 pulses/sec) pulses of 12.5 msec duration were employed as stimuli. In all cases the functional results were supplemented by histological verifications of the electrode location and, in some cases, extended by Marcchi degeneration studies. The most striking effects obtained from stimulation in the anterior colliculus (and the adjacent pretectal regions) were synergistic postural changes of the head and the eyes. Contraversion, eleuatim, and lowering were the motor parameters most frequently observed, if the anterior colliculi were stimulated. Ipsiversive turning was seen only from stimulus loci located in the borderline region of the pretectal area and in the thalamus (identical with those effects described earlier in the section on the diencephalon). Rotations of

SUBCORTICAL MOTOR AREAS

97

the head were never observed from stimulation of the tectum and its borderline structures. The contraversive movement always began (at the onset of the stimulus) with a conjugate deviation of the eyes toward the contralateral side, followed almost immediately by the head; meanwhile the bulbi returned toward their primary position (with respect to their sockets). The eventual direction of gaze thus was determined by the angular deviation (often as large as 1 8 0 O ) of the head. Additional components of this postural change often consisted of a slight elevation of the head or rarely a slight lowering of the head. The effect was always followed by a “positive after-effect,” as the head (and eyes and thus the direction of gaze) remained in the stimulus-induced position for a few seconds after cessation of the electrical pulses. The movement often started with small rhythmical deviations, synchronous to the stimulus frequency, but usually ended in a smooth movement, which has been interpreted by the authors as a manifestation of central summation. Histological study revealed that contraversive turning was always elicited when the stimulating electrode tip was located in the lateral aspect of the colliculus, in particular in the brachitun colliculis superioris and in the lateral stratum opticum (Fig. 5). If the electrodes were situated in about the middle of the anteroposterior extension of the colliculi and in or close to the midline, stimulation would occasionally evoke elevation of the eye axis immediately followed by an elevation of the head. With more lateral displacement of the stimulating electrode a more and more pronounced contraversive component came into effect as was described previously (Fig. 5). A more frequent effect induced by stimulation of the anterior quadrigeminal hillock was lowering of the head and the eyeballs. This type of movement was very rapid and usually showed a marked rhythm which was synchronous to the stimulus rate, The forebody always participated in this movement, and in addition there were occasional synchronous and synergistic movements of the pinnae and vibrissae. Histological examination revealed that the electrodes were situated at the anterior margin of the tectum and extended into the pretectocommissural area limited by the caudal margin of Meynert’s tract (Fig. 5). Apter ( 1946) using pentobarbital-anesthetized cats produced small foci of local “disinhibition” by applying small crystals of

98

WERNER P. KOELLA

FIG.5. Diagram of a horizontal section through the colliculus superior, illustrating the tectally induced motor phenomena. Horizontal solid lines and arrow: Area from which contraversive turning is obtained. Vertical broken lines and arrow: pure elevation. Vertical solid lines and arrow: lowering. Horizontal broken lines and arrow: ipsiversive turning with lowering component (mostly in thalamic structures). The blank areas indicate no motor effects elicitable. Tr. M . tractus Meynert; c.p., commissura posterior; IZZ, 3rd ventricle. (From W. R. Hess et al., 1946.)

powdered strychnine sulfate to the (exposed) surface of the superior colliculus. Upon diffuse illumination of one eye, Apter observed conjugate movements of both bulbi, the direction and magnitude of which were correlated with the placement of the strychnine crystal on the superior colliculus. When the crystal was applied close to the anterior edge of the colliculus, little lateral displacement of the bulbi was obtained, whereas application close

SUBCORTICAL MOTOR AREAS

99

to the posterior collicular region produced a maximal deviation of the axis of gaze toward the contralateral (i.e,, with respect to the colliculus-treated) visual field. Upward components came into appearance if the focus was produced in the medial half of the colliculus and downward components from the lateral half. In general these observations are in concordance with Hess’ finding, with the exception that the latter found the “downward area to extend from the lateral margin onto the midline. Of interest from the viewpoint of comparative neurology is the study by Akert ( 1949) on the eye (and body) movements evoked by electrical stimulation of the tectum opticum of the trout. These animals were anesthetized with MS 222 (Sandoz), and partially immobilized by a head holder containing a water flow device feeding into the mouth and past the gills. The exposed optic tectum was stimulated by means of a springy silver ball electrode, using Hess’ pulse parameters (maximal rate: 15 pulses/sec, pulse duration: 12.5 msec, maximal voltage 2.5 volts). The movements induced by such stimuli again were compound in nature; they involved the eyes, head, body, and fins, and showed in their particular structure, direction, and time course a clean-cut relation to the site stimulated. Usually, after a latency of up to one second after initiation of the stimulus the phenomenon began with (bilateral) movements of the eyes, immediately followed by spreading and movements of the fins, and ending in a burst of tail movements. Akert interpreted these effects as being the manifestation of active goal-directed movements. While the direction of an actual swimming movement was rather hard to determine accurately, it was relatively easy to analyze the changes in gaze direction, and it was this latter parameter which the author used to “extrapolate” into the “future” behavior of the animal. For this reason the author restricted his analysis of the direction-to-locus-of-stimulation relations to the eye movements. Four types of ocular movements were distinguished: ( 1) contraversive conjugate turning, (2) ipsiversive conjugate turning, ( 3 ) pure convergence, and (4) convergence combined with elevation. Unilateral eye movements and lowering were not observed. The turning movements of the eyes were accompanied (after a short interval) by a synergistic bending of the axis of the body. Convergence (with and without elevation) was obtained from the anterior quadrant of the colliculus, whereas the

100

WFRNER P. KOELLA

turning movements were found to be “diffusely” distributed over the remainder of the area. Interestingly enough, ipsi- and contraversive turning were found to overlap completely and could be selectively elicited by proper choice of the stimulus intensity. Low voltages (up to 1.7 volts ) induced ipsiversive, and higher voltages contraversive, turning. Akert interpreted these findings to indicate that the two opposite effects were localized at different depths of the collicular “cortex.” V. The Tegrnental Reaction

According to the classical description, the tegmental response (TR), which is initiated by electrical stimulation of the midbrain tegmentum, consists of a flexion of the ipsilateral (with respect to stimulus site) foreleg, extension of the contralateral foreleg, and a (ipsilateral concave) slowly developing curvature of the body axis ( Fig. 6 ) . This reaction was first described in detail by Thiele ( 1905)

A

8

FIG.6 . Schematic characterizing classical tegniental response ( B ) as conipared to normal posture ( A ) . (From S. Biirgi, 1943.)

and then further investigated by Ingram et al. (1932), Hess (1940), and Monnier (1941).Burgi (1943) in an extensive study analyzed the TR in further detail and tried to arrive at a “biological” interpre-

SUBCORTICAL MOTOR AREAS

101

tation of this phenomenon. According to Burgi, one is indeed able, if using particular stimulus parameters and particular conditions, to elicit from the tegmentum reactions identical to the classical TR. However, if the experiments are performed in mechanically nonrestrained, normal animals (cats), and if one uses different stimulus parameters, one can obtain a number of different “quantities” of this phenomenon which range from simple ipsiversive head turnings to the classical TR as well as ipsiversive “manhge” movements. Furthermore, with different electrode placements, one may also observe contraversive turning. Finally, head rotation and vertical movements of the head are elicited from the same general area. Burgi then comes to the conclusion that the tegmental reaction is a special phenomenon belonging to a complex of three-dimensional representation in the midbrain tegmentum. The reason that that classical TR had been singled out has to be seen in the mechanical, surgical, and possibly chemical ( anesthesia) restraint forced upon the experimental animals of the earlier investigators. VI. Discussion and Interpretation

This rather cursory review has shown that electrical stimulation of some areas of the brain stem yields motor responses which are characterized by a complex and often a spatially oriented organization rather than by simple movements which are produced by the action of a single muscle or muscle group. In some structures of the brain stem there are small areas which appear to have functional connections to much of the body musculature, The organization of these connections is such that upon stimulation all muscles necessary to produce a “directionally” defined motor effect on the eyes, head, and trunk, or to produce well-coordinated leg movements, are thrown into action. For instance, stimulation of a point slightly off the midline in the anterior basal cerebellum causes contraction of the ipsilateral neck and trunk muscles of the ipsilateral abducens oculi, the contralateral internal rectus oculi, and the flexors of the ipsilateral foreleg. There can be little doubt that there also occurs a simultaneous relaxation of the antagonists to these muscles. In addition, an arrangement is available to quantify this directional effect, and it is likely that quantification is accomplished via adjustment of the degree of contraction (and relaxation) of the muscles involved and also by “recruitment,” i.e., activation of more (synergis-

102

WERNER P. KOELLA

tic) muscles with an increase in the central stimulus. The observation that weak stimuli induce a turning of the head whereas stronger stimuli bring about a “supporting” curvature of the trunk is very indicative of this “recruiting” principle. Furthermore, an *‘internalbalancing” system is also involved as is evidenced by the following observation: If the central stimulus induces a particular degree of deviation of the line of gaze, it is brought about via turning of the bulbi and of the head, Preventing the head from turning will automatically be compensated for by larger conjugate deviations of the eyes. Conversely, passive turning of the head in direction of the induced turning movement will quantitatively reduce the deviation of the eye bulbs within their sockets. A phenomenon also entering the picture (but not directly related to, nor specified for, these organizational patterns ) is seen in secondary effects. These can be statokinetic compensatory reflexes, coming into play if the stimulus-induced movements are apt to produce postures or displacements that are incompatible with normal behavior. The question arises as to what functional interpretation can be ascribed to these stimulatory effects. We may assume that the stimulating electrode in all these phenomena produced an excitatory focus somewhere in the course of central pathways which lead from the locus of first order projection of some sensory receptor area to the locus of origin of motoneurons. Seen in this manner the effects produced by such stimuli are interpreted as centrally induced correlates of reflex patterns or of parts thereof. The physical structure of most of these centrally induced motor phenomena suggests strongly that they constitute correlates of a number of postural reflexes which are normally called into action to adjust the bodily posture to, and to compensate for, changes in the physical structure of the surrounding space. With respect to the effects elicited from the diencephalon and adjacent areas we follow Hess’ own interpretation (1949). The vertical movements (i.e., raising and lowering of the head) as well as rotation of the head (stimulated side downward) are, in Hess’ view, manifestations of vestibular reflexes, in particular such reflexes which are physiologically induced by changes of the position of the head in the field of gravity. An important point, strongly supporting the view of the “vestibular” nature of these diencephalic movements,

SUBCORTICAL MOTOR AREAS

103

is illustrated by the fact that such motor phenomena are elicited in the thalamus from the terminal area of the brachium conjunctivum, this pathway (as it originates from the cerebellum) “being in close functional relation with the vestibular system.” It was also the striking similarity between vestibular reflexes induced by “physiological” stimuli and the diencephalic effects which brought Hess to this interpretation. Bartorelli and Wyss (1942) and others have shown that passive rotations of animals around any of the three cardinal spatial axes induce postrotatory motor phenomena consisting of eye, head, and body movements in the same spatial plane in which the animal had been passively rotated, namely: ( a ) lateral conjugate deviations and nystagmus of the eyes, turning and nystagmus of the head, body curving, and eventually manbge movements, after horizontal rotation (vertical axis ) ; ( b ) conjugate eye rotation, head rotation, rotational eye and head nystagmus, and longitudinal revolving movements of the body after rotation around the longitudinal axis; and ( c ) vertical eye and head movements, including nystagmus and raising and lowering of the body often leading to somersault-type movements after rotation around the bitemporal axis. It is of importance to note that this similarity between the physiologically induced vestibular reflex and the motor activity elicited from the diencephalon and its adjacent areas does not involve all types of reactions. As Hess points out, there is a striking dissimilarity between the postrotatory vestibular reflexes after angular decelerations around a vertical axis, on the one hand, and the head- and body-turning movements induced by diencephalic stimuli, on the other. The former, including eye and head deviations, eye and head nystagmus, body curving, and manBge movements, appear with a short latency. Their duration is closely related to the duration of the peripheral stimulus (i.e., the deflection of the cupola of the semicircular canal). The latter exhibit a latency of from 1 to 3 seconds (in the ipsiversive turning) and up to 30 seconds (in the contraversive turning). The electrically induced effects are also characterized by their long-lasting “postive after-effects.” These phenomenological differences led Hess to the interpretation that the “horizontal” diencephalic effects are rather manifestations of “complex” adversive and abversive movements, induced by tactile and

104

WERNER P. KOELLA

nociceptive stimuli respectively, impinging upon the face. In Hess’ view, vestibular reactions in the horizontal plane are organized in postdiencephalic structures of the brain stem. Two more details in Hess’ interpretation are of importance. The first refers to the olbservation that weak stimuli may induce a change in the posture of the eyes, the head, and possibly the body, but that the movement which leads to this change comes to a halt so that a “static,” maintained posture results. Evidently, the new posture represents a new equilibrium of central “forces,” manifested by an equilibrium in the peripheral muscular ( and additional “dead”) forces. Let us consider the example where the stimulus has induced a rotation of the head with the left side down. This deviation from the normal posture in the gravitational field induces a change in the afferent discharge from the vestibular apparatus as well as from the cervical proprioceptors. This discharge would normally call into action muscular forces which bring the head back to normal position. The “static” nature of a stimulus-induced head deviation is thus interpreted as an equilibrium between a stimulus-induced asymmetry in the central excitatory state and the asymmetry in the posture-controlling afferent impulses from the periphery. Another piece of interesting information in Hess’ work is derived from those experiments in which the electrode, previously used to induce a focal excitatory state, serves to produce small lesions in the same area. Under these conditions head, eye, or body deviations occur which are the mirror image of the previous stimulatory effects. Evidently, this experimental procedure leads to a central imbalance not by producing a focus of increased local excitation, but rather by decreasing or eliminating an excitatory state normally present in this locus. In other words, these lesion experiments very strongly suggest that the equilibrium in muscular forces, indicated by the normal position of the head, is the manifestation of a symmetrical, tonic output. Disequilibrium, resulting in deviation of the head in a particular direction is produced experimentally by either an increase of output on one side of the system or a decrease in output on the other side. We already pointed out that under natural conditions the dimcephalic and paradiencephalic motor effects involve three axes of rotation, but that only two of these three lead to a deviation of the vestibular apparatus in the gravitational field. Physiologically, such

SUBCORlICAL MOTOR AREAS

105

deviations are sensed primarily by the otolith part of the labyrinth, and suprathreshold stimulation of the semicircular receptors is not necessarily involved. Thus compensatory eye, head, and body reflexes are induced without any appreciable angular acceleratory component around a bitemporal or longitudinal axis. Viewed in this light, Bartorelli’s experiments (except those with horizontal rotations ) do not reflect the physiological situation very accurately. A further complication arises if we consider an animal not in primary (i.e., prone) but rather in a secondary (e.g., side) position. Under these conditions a passive deviation of the head to side, i.e., toward the shoulder, brings about a change in position in the gravitational field, whereas “lowering” of the head (i.e., the chin approaching the chest) would not represent an adequate stimulus for the gravitoreceptors. It would be of interest to learn whether under such conditions the diencephalic stimulus effect is different from that induced from the same locus, but while the head is in prone position. The motor phenomena induced by stimulation of the basal cerebellum (Koella, 1955) are surprisingly similar to those produced by Hess from the diencephalon. The similarity extends to even minor details such as double representation of the horizontal plane, long latency, long positive after-effects, and smoothness of movement in the ( ipsiversive and contraversive) turning effects. If one takes into account the crossing of the pathways leading from the cerebellum to the diencephalon (in all probability the brachium conjunctivum), the similarity also refers to the directional sense of the head rotations: cerebellar stimuli induce the ipsilateral-side-up rotation whereas the diencephalic stimulation brings the ipsilateralside-down. The only difference between the cerebellar and the diencephalic effects appears to concern the vertical movements. In the diencephalon proper, no head lowering was found (only the area from the posterior commissure backward appeared active in this direction). On the other hand, none of the points of the basal cerebellum probed in our experiments ever evoked a raising of the head. This similarity between our and Hess’ findings, the “directionspecific” organization of most of the motor phenomena and the intimate relation of the area stimulated with the cerebellar projection field of the eighth nerve, induced us to interpret these cerebellar effects upon the eyes, the head, and the trunk in a way similar

106

WERNER P. KOELLA

to that of Hess, namely, that these motor phenomena “constitute the efferent component of a vestibular reflex of higher (cerebellar) order.” This interpretation is compatible with respect to the rotatory and vertical head effects. As to the (horizontal) turning movements, an objection identical to the one raised in the discussion of the diencephalon has to be considered. Both the long latency and the tendency to outlast the electrical stimulus would be, in Hess’ view, rather uncharacteristic for vestibular reflex actions. An alternative interpretation, identical to that suggested for the diencephalic turnings, may be in place, especially in view of the massive inflow of tactile and proprioceptive impulses to the cerebellum as reported by Dow (1939), Adrian (1943), and Snider and Stowell (1944). Clark (1939), indeed, comes to such an interpretation for the “leaning movements” toward the ipsilateral side produced by stimulation of the lower posterior vermal cortex. From his description it appears that these movements occupy a position somewhat intermediate between the rotatory and the turning movements observed in our own experiments. Clark suggests that these movements “have the appearance of normal adjustments of the body to changes in the tension of groups of muscles, and, indeed, their counterpart can be produced by placing the hand on one side of a standing cat in the region of the thigh (or even of the shoulders) and pushing gently towards the opposite side.” There is, however, one aspect which leads us to accept, with some reservation, Hess’ as well as Clark‘s interpretation of the horizontal movements induced from the diencephalon and the cerebellum respectively. If the horizontal movements were correlates of adversive and abversive reactions to horizontally impinging tactil and nociceptive stimuli respectively, one would expect that phenomenologically similar movements should be elicitable in the vertical direction, representing reactions to such peripheral stimuli impinging on the head and body from above or from below. The probing of the diencephalon and the cerebellum does not provide any evidence of such movements, as only fast, rhythmical, and short-latency, vertical deviations are produced from those areas. Consequently, we tend to interpret the horizontal movements produced by stimulation of the diencephalon and the basal cerebellum as being the correlates of vestibular reflex actions. The phenomenological difference between the horizontal (turning) movements on the one hand, and the rotatory and vertical

SUBCORTICAL MOTOR AREAS

107

movements on the other, are satisfactorily explained in the following way: The rotatory and vertical movements (i.e., deviations around the longitudinal and the bitemporal axis) involve a change in position in the gravitational field. Physiologically the proper receptor for such changes is the otolith apparatus; the adequate stimulus is position per se. The flow of centripetal impulses depends at every moment upon the spatial relation between the head and the gravitational vector (if we disregard additional acceleratory vectors such as centrifugal forces, etc. ) . The “static” deviated position induced by the central stimulus is, as already mentioned, the manifestation of a new equilibrium between centripetal impulses and artificially induced central disbalance. In contrast, the turning movements do not involve a change in the gravitational field. Physiologically the proper receptor is the ampulla of the semicircular canal. There, the adequate stimulus is the acceleration, which is proportional to the force exerted upon the cupola. The centripetal flow of impulses depends upon the degree of deflection of the cupola in the ampulla, and the degree of deflection is a function of force, elasticity, viscosity of the endolymph, and time. The equilibrium condition between the central excitatory state induced by afferent impulses from the semicircular canal and the state induced by the central electrical stimulus is thus much more complex than the equilibrium between the afferent discharge from the otolith organs and the central electrically induced excitatory focus, The former is a dynamic equilibrium which depends upon time. The cupola, after an initial deflection, returns slowly toward the resting position as soon as no more acceleratory forces are present. This returning movement thus constitutes a kind of a mechanical “adaptive” process. As the afferent discharge from the ampulla counteracts the effect of the central electrical stimulus, it follows that during this adaptive process the central stimulus becomes more and more preponderant and eventually will “break through” to induce a turning movement. The situation is much more complex than described and can be understood only on the basis of a differential equation containing factors such as “turning power” of the central stimulus, angular acceleration, angular velocity, damping constant of the cupola, etc. Space and context of this paper do not allow discussion of further details in this extremely interesting field. But nevertheless this short discussion of vestibular physiology has demonstrated that the turn-

108

WERNER P. KOELLA

ing movements, too, may be interpreted as correlates of vestibular reflexes. The long latency, i.e., the slow start, and the positive aftereffects can be explained on the basis of the physical properties of the angular acceleroreceptors. Figure 5 in Clark‘s paper shows another stimulation effect (from the upper vennis) which is very similar to the turning effects seen in our experiments, with the exception that there was a marked rebound, whereas we have seen only positive after-effects. The rotational component of the effects induced from the cerebellar cortex is opposite in direction to one of the effects elicited from the basal cerebellum. This may indicate a crossing of the corticifugal fibers. There is, however, an alternate interpretation available. Clark observed that local distruction of a point previously stimulated invariably induced deviations similar to those induced by stimulation and not to those seen in the rebound phase (personal communication from Dr. Clark, 19%). This phenomenon suggests that the (relatively) high frequency of stimulation in Clark‘s experiments (60 cycles/sec) induced a local focus of reduced activity rather than one of excitation so that the disbalance was a consequence of reduction in tonus on the “stimulated” side. Our own experience, that an increase in stimulation frequency from 20 to 60 or 100 pulses/sec brought about a reversal in direction of a particular movement, supports this interpretation. Of interest is the story of the interpretation of the tegmental response. According to Biirgi (1943) the tegmental response observed is a special case of a number of three-dimensionally organized motor phenomena elicitable from the tegmental area. He thinks that the tegmental motor effects are very similar to those induced in the diencephalon and possibly involve the same pathways. Ingram et al. (1932), on the other hand, prefer a phylogenetic interpretation and consider the tegmental response (in its “classical” restrained form) as a remainder of motor phenomena which have lost their importance in the present level of evolution. They support their interpretation with observations by Coghill ( 1929), who described “tegmental-like” spontaneous movements in amblyostoma larvae. We tend to follow Biirgi’s interpretation and ascribe an important functional role to the tegmentum and to include this area into a whole group of subcortical motor structures which extend from the medillla oblongata to the basal ganglia. The observation

SUBCORTICAL MOTOR AREAS

109

that the movements induced by electrical stimulation of the tegmentum often are characterized by a certain lack of “natural” appearance is probably best explained by the fact that the stimulating electrode is hardly able to produce an excitatory focus in only one functional system and that it necessarily involves many other SYStems in this bottleneck of pathways and neuronal reticula. With the phenomenological character of the effects in mind it appears possible to interpret the diencephalic, the tegmental, and the (basal) cerebellar motor activity as manifestations of corrective or compensatory movements normally induced by disturbing external forces. In the obvious three-dimensional arrangement two dimensions appear to represent those bodily movements which are correlated with deviations of the position in the field of gravity, i.e., movements around axes which are at right angles to the gravitational vector and at right angles to each other. Obviously the statolith organs represent the sensory receptor system coming into play under normal conditions. The third dimension on the basis of the timing of the particular movements seems to be in functional connection with semicircular canal reflexes and possibly with reaction induced by forces impinging upon tactile or proprioceptive receptors. An entirely different interpretation is suggested with respect to the motor effects induced from the optic tectum. Apter ( 1945,1946), Hess ‘et al. (1946), and Akert (1949) agree that the sideward turnings, the raising or lowering of the eyes followed by synergistic movements of the head, are manifestations of optically induced adversive reflex movements of the direction of gaze toward an object appearing in the periphery of the visual field. Apter’s (1945) and Akert’s (1949) studies in the cat and the fish, respectively, show that there is a striking two-dimensional representation of the (twodimensional) visual field in the colliculus and that there is a correlation between the projection of a certain “point” of the retina on the colliculus, on the one hand, and the direction and degree of conjugate eye movement produced by stimulation of that collicular site, on the other. Apparently this movement is so discretely organized that through the motor act the external photic (or negative photic) punctiform stimulus, previously impinged upon the peripheral retina, is now focused onto its center, which possesses the highest degree of visual acuity and thus affords the optimal condition for

110

WERNER P. KOELLA

analysis. In these authors’ interpretation the colliculus constitutes a center for a relatively simple reflex action to which Hess refers as “visueller Greifreflex” ( visual grasp reflex). The comparison of tegmental and tectal motor activity leads Hess et al. (1946) to elaborate upon the differentiation between “ereismatic” and “teleokinetic” motor activity. The visual grasp reflex though extrapyramidal and involuntary, is a typical example of a “teleokinetic,” “goal-directed motor phenomenon; it does not compensate for, nor adjust to, external disturbing forces, and does not serve as a dynamic foundation upon which other movements can successfully develop, but is per se part of a motor activity which serves to reach a remote goal. The tegmental effects on the other hand are phenomena organized to interact with external disturbing forces, to bring about the stabilizing effect on the body which enables precise “teleokinesis.” The comparison of these biological stabilizers, with those used in battleships and “invented a few million years after Mother Nature has solved the problem, is not too farfetched. Up to this point we have described the motor phenomena elicited from a number of isolated extracortical areas and have treated these motor effects and their interpretation as if each of these areas represented an independent motor system. If one observes the similarity in the effects obtained from the various areas and takes into consideration the existence of neuronal pathways connecting these areas, as well as the lack of discontinuity in responses elicited from electrodes moving along these pathways, then it would seem that there is a high degree of functional interdependence between all the extracortical structures discussed earlier in this review. Recently Jung and Hassler (1960), in an excellent review article on the “extrapyramidal system,” attempted to integrate many of the findings discussed here. These authors subdivide a large portion of the brain stem into five neuronal systems subserving five direction-specific movements, namely ( a ) rotation around the longitudinal axis, ( b ) upward movements of the head and forebody, ( c ) downward movements of the head and forebody, ( d ) ipsiversive turning, and ( e ) contraversive turning. To each of these five types of movements they assigned a neuronal system consisting of nuclei and pathways which are represented diagrammatically in Fig. 7 and 8. There is only one point where we tend to disagree with Jung

SUBCORTICAL MOTOR AREAS

111

FIG. 7. Diagram of nuclei and tracts involved in posture and responsible for direction-specific motor effects of mesodiencephalic stimulation. Rotation movements are regulated by the interstitial nucleus ( 1 s t ) and its fiber systems (-). Raising movements are regulated by the praestitial nucleus ( P . s t ) and its fiber systems (-). This tonically active nucleus sends short fibers to the nucleus ruber magnocellularis ( Ru ) from which arises the rubrospinal tract ( ru.spi. ) . Lowering movements are rcguloted by the praecommissural nucleus ( P r . C o ) and its fiber systems ( ). The efferent fibers constitute the praecommissurotegniental tract. All descending fiber tracts of the direction-specific systems send collaterals to the nuclei of the ocular niiiscles ( n . 111, IV, and V I ) and the reticular formation. (From Jiing and Hassler, 1960.)

++

and Hassler’s interpretation; it is on the role of the cerebellum in the organization of these “direction-specific” movements. These authors do not ascribe much importance to this structure other than in connection with rotatory movements. In attempting to understand central organization of motor activity and to correlate certain neuronal structures with certain types of movements, we should include in such systems all structures from which such movements are elicitable, even though the probing electrode may show that by-passing pathways exist. The similarity between the (basal) cerebellar and the diencephalic motor effects, indeed, leads us to assume that these two areas are components of a larger unit which organizes much of the ereismatic activity. The brachium conjunctivum appears to be the major

112

WERNER P. KOELLA

cc

FIG. 8. Neuronal mechanism for turning movements in the horizontal plane to the left side. The apparatus for ipsiversive turning is schematically drawn on the left of the figure, that of contraversive turning on the right. From the vestibular nuclei ( Nc.ue) arises the ipsilateral vestibulorecticular thalamic tract which is called the dormlateral tegmental fasciculus (F.t.dZ) in its first part and, more rostrally, Forel's tegmental fasciculi ( F . F o ) . The fibers of this fasciculus terminate in the ventrointermediate nucleus of the thalamus ( V h ) . From there arise the cortical projections to the central region, probably area 3a. One of the contraversive turning systems starts from the anterior thalamic nucleus ( A . p r ) and reaches area 24 of the g y m s cinguli. This area and all adversive cortical areas ( 8,6,5,22) send fibers through the internal capsule ( C a i ) to the entopenduncular nucleus (Pa2Z.e).This nucleus also receives fibers from the caudate nucleus ( C d ). The efferent tract runs through the zona incerta (Z.i) near the subthalamic nucleus and crosses the midline in the anterior midbrain. It is connected with the reticular formation which contains the apparatus for all turning movements to the same side. Its efferent tract is mainly the reticulospinal tract ( Trt. s p ) . Another pathway for contraversive movements arises in area 18 and passes through the tectum opticum ( T a p ) where it is connected with the substrate of optic grasp reaction. The efferent path, the tectospinal tract ( Tt.sp), also mediates contraversive turning. It crosses the midline and descends as the predorsal fasciculus ( F p r d ) , (From Jung and Hassler, 1960).

SUBCORTICAL MOTOR AREAS

113

pathway between these two substations, and it is likely that the main flow of impulses occurs from the cerebellum t o the diencephalon. Jansen and Jansen (1955) showed that not only the lateral cerebellar nuclei but also the fastigial nuclei feed into the brachium. This pathway would include impulses from most of the area from which the present writer induced the “directional” effects. Cerebellomedullo-diencephalic pathways are likely to constitute additional channels. The cerebellum with its intensive supply of vestibular, proprioceptive, tactile, optic, and acoustic impulses certainly is a structure uniquely capable of integrating these afferent signals, all of which have paramount significance in postural adjustments. As a matter of fact there is good electrophysiological evidence that convergence of heteromodal afferent volleys does occur in the cerebellum. Bonnet and Bremer (1951), Bremer (1952), and Bremer and Bonnet 11951) have demonstrated ( facilitatory ) interaction of acoustic and tactile afferents, and Koella and Loeser (1959) and Levy et al. (1961) have shown that optic and acoustic afferent impulses in the cerebellar vermal cortex yield facilitatory or inhibitory interaction depending upon the interval in time of arrival of these volleys. An important parameter, the spatial relation of the visual and auditory source, has not yet been studied in these latter experiments. Work is in progress in our laboratory to elucidate the role of this factor, i.e., to investigate whether degree and sign of mutual interaction depends upon the angular relation between the optic and acoustic “beam.” Both modalities appear to have reflexogenic significance in that both modalities are able to induce a “grasp” or adversive reflex. Perhaps the posterior vermis is the locus where the struggle for preponderance of one or the other of the heterodirectional-heteromodal stimuli takes place. Studies on motor behavior in response to sensory stimuli of various modalities also strongly suggest that central interaction does indeed occur, but for none of the following examples have details of mechanisms been successfully investigated. Magnus (1924) in his classical work has demonstrated that vestibular, proprioceptive, and visual receptors all contribute to the determination of the position of the eye in its orbit and of the head. von Ilolst (1950a, b ) has done an excellent job in analyzing the compound influence of light and gravitation (both seen as vectors) upon the position of the fish

114

WERNER P. KOELLA

in running water. I have shown (Koella, 1947a, b, 1948) that the pattern of excitation in the semicircular canals in collaboration with afferent impulses from the otolith organs and proprioceptors in the neck, determines the plane of the nystagmic movements of the eye. Hess (1941, 1942) showed on the basis of three-dimensional geometry that the quantitative and directional effect of each of the six eye muscles changes with a change of the position of the eyeball in its socket. As a consequence, the innervation arrangement has to be different for any given movement depending upon whether this movement starts from the primary or a secondary position. Hess then “constructs” an apparatus, representing a “modulated reflex pattern” which would handle this organizational problem. The guiding principle again is interactive behavior between reflexinducing, retinofugal impulses and reflex-modulating, positionsignaling impulses from the eye “joint” and the external eye muscles. Eye and head movements, particularly the visual grasp reflex, pose another problem, Vestibular and proprioceptive influences keep the head and eyes “fixed in space. External forces liable to disturb this “equilibrium” are compensated for by internal (muscular) forces. Yet a few photons are able to move the head and the eyes away in any direction. Evidently, the movement-inducing volley of afferent impulses inhibits in a quantitatively and directionally specific way the vestibular and proprioceptive “fixation” mechanisms. It is likely that the organization of all of these interactive patterns in one way or another requires the participation of the structures discussed here. But how and where the complex task of switching is worked out has to be shown by future investigations. REFERENCES Adrian, E. D. ( 1943). Bruin 66, 289. Akert, K. ( 1949). Helv. Physiol. Actu 7, 112. Apter, J. T. (1945). J. Neurophysiol. 8, 123. Apter, J. T. (1946). J. Neurophysiol. 9, 73. Bartorelli, C., and Wyss, 0. A. M. (1942). Arch. ges. Physiol. Pfliiger’s 245, 511. Bolk, L. ( 19Of3). “Das Cerebellum der Saugetiere.” Bohn, Haarlem. Bonnet, V., and Bremer, F. (1951). J. Physiol. (London) 114, 54. Bremer, F. ( 1952). Rev. neurol. 87, 65. Bremer, F., and Bonnet, V. (1951). J. physbl. (Purk)43,665.

SUBCORTICAL MOTOR AREAS

115

Burgi, S. (1943). Helv. Physiol. Acta 1, 3. Chambers, W. W. (1947). Am. J. Anut. 80, 55. Clark, S. L. (1939). J. Neurophysiol. 2, 19. Coghill, G. E. (1929). “Anatomy and the Problem of Behaviour.” Cambridge Univ. Press, London and New York. Denny-Brown, D., Eccles, J. C., and Liddell, E. G. T. (1929). Proc. Roy. SOC., B104, 518. Dow, R. S. (1939). J. Neurophysiol. 2, 543. Dow, R. S., and Morruzzi, G. (1958). “The Physiology and Pathology of the Cerebellum.” Univ. of Minnesota Press, Minneapolis, Minnesota. Faulkner, R. F., and Hyde, J. E. (1958). J. Neurophysiol. 21,171. Godlowski, W. (1938). Z. Neurol. u. Psychint. 162, 160. Grossman, R. G. (1958). J. Neurophysibl. 21, 85. Hampson, J. L., Harrison, C. R., and Woolsey, C. N. (1952). Research Publ. Assoc. Research Nervous Mental Disease 30, 299. Hare, W. K., Magoun, H. W., and Ranson, S. W. (1936). Am. J. Physiol. 117, 261. Hare, W. K., Magoun, H. W., and Ranson, S. W. (1937). J. C m p . Neurol. 67, 145. Hassler, R. (1956a). Arch. Psychiat. 194, 456. Hassler, R. (1956b). Arch. Psych&. 194, 481. Hassler, R., and Hess, W. R. (1954). Arch. Psychiat. 192, 488. Hess, W. R. ( 1940). Arch. ges. Physiol. Pfliiger’s 243, 634. Hess, W. R. ( 1941). B i d . Zentr. 61, 545. Hess, W. R. (194%). Naturwissenschaften 30, 441, 537. Hess, W. R. (194213). NeTOenUTZt 15, 456. Hess, W. R. (1945). Helv. Physbl. Actu 1, C-62. Hess, W. R. ( 1949). “Das Zwischenhirn.” Schwabe, Basel, Switzerland. Hess, W. R., and Weisschedel, E. (1949). Helv. Physiol. Acta 7, 451. Hess, W. R., Burgi, S., and Bucher, V. (1946). Monutsschr. Psychiat. u. Neurol. 112,l. Hinsey, J, C. (1940). Reseurch Publ. A,woc. Research Nervous Mental Disease 20, 657. Hunter, J,, and Jasper, H. H. (1949). Electroencephubg. and Clin. Neurophyswl. 1, 305. Hyde, J. E. (1957). J. Comp. N ~ u T o108, ~ . 139. Hyde, J. E., and Eason, R. G. (1959). J. Neurophysiol. 22, 666. Ingram, W. R., Ranson, S. W., Hannett, F. I., Zeiss, F. R., and Terwilliger, E. H. (1932). A.M.A. Arch. Neurol. Psychiat. 28, 513. Jansen, J., and Jansen, J., Jr. (1955). J. Camp. Neurol. 102, 607. Jung, R., and Hassler, R. (1960). “The Extrapyramidal Motor System,” in Handbook of Physiology, Section 1, Vol. 11, pp. 863-927. Koella, W. P. (1947). Helv. Physiol. Acta 5, 154. Koella, W. P. ( 194713). Helv. Physiol. Actu 5, 430. Koella, W. P. (1948). Helv. Physiol. Actu 6, 280. Koella, w. P. ( 1955). J. NetLTophySiol. 18, 559.

116

WEXNER P. KOELLA

Koella, W. P., and Loeaer, J. (1959). Federation Proc. 18, 83. Levy, C. K., Loeser, J., and Koella, W. P. (1961). Electroencephlog. and Clin. Neurophysiol. 13, 235. Lewandowski, M. (1903). Arch. Anat. u. Physiol. 27,129. Lijwenthal, M., and Horsley, V. (1897). Proc. Roy. Soc. B61, 20. McDonald, J. V. ( 1953). J. Neurophysiol. 16, 69. Magnus, R. ( 1924). “Korperstellung.” Springer, Berlin. Magoun, H. W., Hare, W. K., and Ranson, S . W. (1935). Am. J. Physiol. 112, 329. Monnier, M. (1941). Schweiz. med. Wochschr. 71,598. Moruzzi, C. (1950a). Electroencephalog. and Clin. Neurophysiol. 2, 463. Moruzzi, G. ( 1950b). “Problems of Cerebellar Physiology.” Charles C Thomas, Springfield, Illinois. Moruzzi, C., and Pompeiano, 0. (1956). J. Comp. Neurol. 106, 371. Moruzzi, G., and Pompeiano, 0. (1957). Arch. ital. bwl. 95, 31. Sherrington, C. S. (1897). Proc. Roy. SOC.,B61, 243. Sherrington, C. S . (1900).In “Textbook of Physiology” ( E . A. Schafer, ed.), Vol. 2, p. 884. Pentland, London. Snider, R. S., and Stowell, A. { 1944). J . Neurophysiol. 7, 331. Sprague, J. M., and Chambers, W. W. (1954). Am. J. Physiol. 176,52. Thiele, F. H. (1905). J. Physiol. ( L d o n ) 32, 358. von Holst, E. (1950a). Z. vergl. Physiol. 32, 60. von Holst, E. (1950b). Naturwissenscheften 37,265. Waller, W. H. (1940). J. Neurophysiol. 3, 300. Woolsey, C. N., Settlage, P. H., and Meyer, W. (1950). Am. J. Physiol. 163, 763.

BIOCHEMICAL AND NEUROPHYSIOLOGICAL DEVELOPMENT OF THE BRAIN IN THE NEONATAL PERIOD By Williamina A. Himwich Thudichum Psychiatric Research Laboratory, Golesburg State Research Hospital, Galesburg, Illinois

I. Introduction . . . . . . . . . . . . . . . 11. Inherent Problems . . . . . . . . . . . . . . 111. Genesis of Behavior . . . . . . . . . . . . . IV. Neuroanatomical Development . . . . . . . . . . V. Accumulation of Chemical Constituents . . . . . . . . VI. Development of Enzymatic Activity . . . . . . . . . VII. Hematoencephalic Exchange in the Developing Brain . . . . VIII. Neurophysiological Development . . . . . . . . . A. Spontaneous EEG . . . . . . . . . . . . . B. Evoked Potentials . . . . . . . . . . . . . IX. Correlation of Anatomical, Chemical, and Neurophysiological Factors X. Conclusions . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . .

117 119 122 125 129 138 141 142 143 145 147 154 155

I. Introduction

Intensive study of the changes which occur as the brain matures in the young of various species of animals might be expected to yield pertinent information as to the relative rates at which the central nervous system develops in these species. Correlation of the anatomical, biochemical, and neurophysiological changes with the development of functional and behavioral patterns will indicate the necessary degree of maturation required for various functions. Basic information of this kind is essential if we are to study in experimental animals the neurological disorders that affect the human infant at birth or during development. It would be naive to assume that we could move in an easy progression from the development of a rodent with apparently low intelligence such as the guinea pig, 117

118

WILLIAMINA A. HIMWICH

through species of rodents with relatively high intelligence, to the more complex brains of the cat and the dog and thus obtain information applicable to man. However, data accumulated by our laboratory and by other workers, such as the Flexners, suggest that in terms of species comparison definite patterns can be ascertained. These patterns will be discussed in more detail in this chapter. In assessing the neurological maturation of the brain with the concomitant chemical and anatomical development we must also consider various factors in the total development of the animal. These factors include the rate of prenatal development, a correlary of which is the maturity of the animal at birth, and the rate of postnatal development. In no species on which data are available do these two rates coincide. Other important points are the complexity of the brain organization and also the total metabolic rate of the animal as a whole as well as that of the brain itself. The development of the young of any species is suggested by Anokhin (1956) to follow a pattern designed to permit maximum survival. For example, if vision is essential from the time of birth because of the habitat and living pattern of the species, the young must be born with eyes open and functional. On the other hand, in a species where the young are normally born in the dark and are kept there until they are able to protect themselves to some extent, the development of the visual tract is not of primary importance. It should be noted that the time at which the eyes open gives information only on the maturation of the visual system. Much more sophistication and understanding of the adaptive mechanisms and the influence of special patterns of living must be available before we can make accurate comparisons of young animals at various ages. This chapter, however, will be devoted to the thesis that we can now, within limits, make certain simple comparisons. Certain phases of the developing brain will not be considered in this chapter. Among these are the scattered data on the metabolic rate of the brain as it develops, the neurophylogenetic development of brain parts, and the rates of accumulation of various lipid fractions. Literature on the latter point although extensive is still in its infancy and many points require further clarification before they can be fitted into discussions of function. In the majority of instances the data used in this chapter were obtained on whole brain and reported on a wet weight basis. Where necessary, however, as in the

119

DEVELOPMENT OF BRAIN I N NEONATAL PERIOD

case of the guinea pig, data on cerebral cortex has been given. Unfortunately, little detailed anatomical and biochemical data are available to us on the various parameters of the vertical systems of the brain, e.g., the visual tracts, although these have been studied neurophysiologically. This field is a fertile one for further research. Investigations in this area, however, will be complicated by the different rates of maturation which even individual cells in a nucleus may show. Anokhin (1956) points out that those cells of the nuclei of the facial nerve in man, from which the nerve fibers necessary for sucking originate, develop first and those with fibers to the muscles of the forehead at a later period. II. Inherent Problems

The biggest problem in studying the developing brain is the wide variation in stage of maturity among animals of the same conceptual age. In the dynamic .and rapidly metabolizing fetal brain a few hours difference in age can produce a marked change in development. Although this difficulty arises in the study of any parameter, it can be illustrated easily in the glutamic acid and glutamine (G,a-G) content of the brains of littermate fetal guinea pigs (Fig. 1). The variations in content are probably not due to differences in growth of brain since the brain weights are often consistent among members of a litter unless there has been gross overcrowding in the uterus (Table 1) . Runts or giants can easily be TABLE I GUINEAPIG BRAIKSFROM DIFFERENT LITTERS AT 42 DAYS CONCEPTI.AI, .IOE

Litter no. 1

0.9685 0,9783 1.0429

111 149

2

0.8349 0.8544

96 72

3

0.8872 0.9368

68 57

107

120

WILLIAMINA A. HIMWICH

265 25.02

t 0

0 225-

0

A ?3 Mo--

A

E"

ss

175-

2

1

*

0

2 /so--

0

+

a u 125-

u

s 2 loo--

d

I5

A

0

8 A

75-

4

I ,

0

'35 34

40 42

'

I

4849

44 CONCEPTUAL AGE

1 1

53'55

54

A

L

6:

64

FIG. 1. Variation in glutamic acid and glutamine content of the brain of littermate guinea pigs at various times during gestation. At any given day the same symbol refers to members of the same litter.

eliminated by inspection but the variability among individuals remains even when only those of uniform size are studied. The differences between litters of the same conceptual age may be ascribed to genetic factors and/or to nutritional factors. Just before parturition physiological conditions in utero rapidly become untenable for the fetus. Birth itself is a physiological shock to the newborn and while it may be mitigated by good maternal care, the immature organism nonetheless suffers. The more mature the young animal is at birth, the more homeostatic mechanisms are

DEVELOPMENT OF BRAIN I N NEONATAL PERIOD

121

available to cushion its responses, These events all must affect the brain both biochemically and neurophysiologically. Anatomically they would seem to have little influence except to slow development. For these reasons data obtained on animals just before or after birth should be considered carefully. It is important to study animals in this period at as short intervals of time as possible so that variations due to birth trauma may be separated from those due to the normal course of development. After birth the technical problems of obtaining normal values are, if anything, multiplied, especially in young animals that are born relatively immature with high metabolic rates and large surface areas, such as the mouse and rat. Hypoglycemia, dehydration, and lowered body temperature all can quickly take their toll of brain constituents and be reflected in neurophysiological responses. The young selected for use must be known-not assumed-to be of average size and development and free of disease. Young mice from different litters have been observed in our laboratory to vary by 100% in weight. Divergent results on any species or age must first be considered in the light of possible environmental artifacts. As far as enzyme activity or accumulation of chemical constituents is concerned, it must be remembered that extractive and chemical methods which are adequate for one species may not be equally effective in another. This problem in regard to the guinea pig has been discussed elsewhere (Himwich and Petersen, 1959). The amount of sensory stimulation given the young by handling or deprivation of stimuli may also effect biochemical parameters of brain development (Levine and Alpert, 1959). The difficulties attendant on the study of human brain especially as regards analyses for chemical constituents or enzymatic activity are even greater than for experimental animals. These problems involve the condition of the fetus in utero and the chain of circumstances which rendered it unable to survive after birth (Himwich et al., 1959). Folch-Pi (1955), in his extensive study of mouse brain and the development of lipid fractions, pointed out that often several components appear deposited at essentially the same relative rate, but disparities often occur between respective rates of acquisition of components that have been accumulating at parallel rates at other points in the curve. These data, he believed, suggest that the deposition of constituents in the brain does not follow a smooth gradient.

122

WILLIAMINA A. HIMWICH

Inspection of almost any set of data on the developing brain can be used to illustrate this point. We have, however, felt that a much greater quantity of data collected at much smaller intervals in the life cycle was needed before this question could be answered. Therefore, in our studies of the developing brain the curves have been fitted by inspection and made as smooth as the data permit. Ill. The Genesis of Behavior

The basic need in studying the developing brain is to relate changes which occur in this organ or in its responses to the gradual acquisition of adult functions and patterns of behavior. It must be remembered, however, that the appearance of a behavioral response depends on the maturation not only of the central nervous system but also of the effector system as well. The great neurologist Tilney (1933-1934) emphasized this concept thirty years ago in his monograph on behavior and its relation to the development of the brain. The problem has intrigued many neurologists and chemists, e.g., Matilda and Waldemar Koch, who did pioneer work on the relation between brain chemistry and function. However, systematic studies have not appeared and it is only within the last few years that comparable studies on several species have been published by scattered laboratories throughout the world. Species differences in behavior at the time of birth reflect the stage of development of the brain. In general, the newborn animals can be divided into two large classes: ( 1 ) those which are born relatively mature and able to fend for themselves, and ( 2 ) those born relatively immature and helpless and requiring varying periods of parental care if they are to develop into adults. The young of the guinea pig, horse, sheep, goat, and cow are born able to move about freely with their eyes open and functional, and under favorable circumstances, to maintain themselves with relatively little parental attention. The amount of protection and care which they need varies, of course, from species to species, even in this group. As we will see later in the chapter, the electrophysiological, anatomical, and biochemical conditions of the brain of the newborn guinea pig are consistent with its behavior. In the rat, mouse, rabbit, dog, cat, and human being the young are born with little motor ability and varying degrees of immaturity of the sensory systems and require close parental care and supervision until development is complete.

~UCKING

/

CRAWLING ' APPROACH

~OSTURE

SYNERGISING

EYE OPENING SITTIN

,I

'~MKING EYE AND HEAD TURNlNGj WASHING

; \RUNNING

'

ESCA~

PAWING IN PLAY ~LIMBING SCRATCHING

FIG.2. Chronological succession of increments of behavior in the rat. ( From Tilney and Kubie, 1931. )

124

WILLIAMINA A. HIMWICH

The gradual acquisition of the characteristics of normal adult behavior was plotted in detail by Tilney and Kubie (1931) for rat, guinea pig, and cat (Figs, 2, 3, 4). It is unfortunate that similar studies are not available on the young animals of other species, STREAM OF BEHAVIOR I 2 \

DAYS

GUINEA PIG 3 4

5

6

7

-

20 0

I

SYNERGISING// EYE AND HEAD TURNING

I

I

I

I

I

~ L K I N G SITTING

FIG.3. Chronological succession of increments of behavior in the guinea pig. (From Tihey and Kubie, 1931.) CAT DAYS

I

2

3 4

5 6 7 8 9 10 II 12 13 14 15 16 17 18 19 20 21 22 23 2425 26 2 7 2 8 2 9 30

BACKING AWAY

FIG.4. Chronological succession of increments of behavior in the cat. (From Tilney and Kubie, 1931.)

DEVELOPMENT OF BRAIN I N NEONATAL PERIOD

125

IV. Neuroanatomical Development

A detailed discussion of the anatomical development of the nervous system in various species is beyond the scope of this chapter and the competence of the author. Some points, however, require enumeration in order to serve as a basis for the future correlation of the anatomical, biochemical, and neurophysiological events. The actual size of the brain increases rapidly during development. Brain weights at adulthood increase in the phylogenetic scale (Figs, 5 and 6 ) with a decreasing number of neurons per unit 4c )t

= DOG

ADULT LEVEL 8 0 . 0 GMS.

V = RABBIT

35 0 =

A

CAT

/'

4'

3c 25 Ln

I

d 0

20

15 I0

5

,

,*,

-30 -20 -10 BIRTH 10 AGE

20

30

m--.

40 3 MO. 5MO. 7MO. Al ULT

IN DAYS

FIG. 5. Growth of brain of dog, rabbit, and cat. (Reprinted by permission. From Himwich and Petersen, 1959. )

volume of cortex from mouse to dog brain (Tower and Elliott, 1952). In the guinea pig, brain weight increases most rapidly during the fetal period; in the other species the greatest increment of weight occurs in postnatal life. In the rat the differentiation of the cerebral cortex has barely commenced at birth [Riese (1944) felt that it was possible to distinguish the six cortical layers] and not until day 12 does it approach a mature type of organization (Eayrs and Goodhead, 1959). At this time nuclear volume in the neurons has reached the adult level

126

WILLIAMINA A. HIMWICH

45

~

V

= RAT

.- *--....

0 = GUINEA PIG

Ln

4

(z

301 25

0.

y'

0

t

151005

t -30

, ,m ,. AGE

IN

3 MO. 5MO. 7 MO. Al ILT

DAYS

FIG.6. Growth of brain of rat, guinea pig, and mouse.

(Brizzee and Jacobs, 1959a), Nissl (Sugita, 1918), and ground substances (Goodhead, 1957) are present. Relatively little work has been done on the mouse. Data from Isenchmid (Sugita, 1918) show that the cortex of the mouse brain, as far as thickness of that region is concerned, develops a little more slowly than in the rat, attaining almost adult width by day 12 (Fig. 7). We can, therefore, assume that the other anatomical characteristics follow the same temporal pattern. The ground substance, however, appears by day 5 or 6 (Hess, 1955a), and some dendritic growth is evident by day 8 and markedly increased by the day 15 (Ramon y Cajal, 1960). By contrast, in the guinea pig fetus at 45 days of conceptual age, the five cortical layers characteristic of the species are present (Peters and Flexner, 1950), ground substance is detectable ( Hess, 1955b), the neuronal volume has reached its maximum level (Peters and Flexner, 1950), the Nissl substance is only slightly less than in the adult (LaVelle, 1951), and dendritic processes are evident in the cortex (Peters and Flexner, 1950). In the rabbit by the day 23 of prenatal life there is initial differentiation of the pyramidal cells and of the future fifth layer of the

DEVELOPMENT OF BHAIN IN NEONATAL PERIOD

127

mm.

FIG.7. Changes in cortical thickness in the albino rat and in the mouse according to age. The data for the two species are taken from areas that approximately correspond. (From Sugita, 1918.)

cortex. The motor and visual cortexes increase in width, but not at a uniform rate, from day 20 of conceptual age. The first growth period is during the last third of prenatal life; the second begins on day 14 (Pentzik, 1937). The cortex as a whole shows a first growth phase at 0-10 days; during this time the “non-cell body” volume increased until 10-12 days. Ground substance is present by day 10 and Nissl substance is rapidly increasing at this time (SchadB, 195913). Growth of layers VI-VII begins after birth and is especially marked from day 7-8 in the visual cortex which between day 10-14 becomes similar to that of the adult (Pentzik, 1937). The newborn kitten has not only six distinct cortical layers but also regional differences in the cortex (Riese, 1944). In this species the mean nuclear volume increases rapidly until 20 days of age ( Brizzee and Jacobs, 1959b). Detailed studies of dendritic growth have been made by Scheibel (1962). Histological studies on the developing brain of the young dog are not available. From studies in progress in our laboratory, it appears that at 1 day of age dendritic processes show some development, and the cortex is beginning to be differentiated into layers ( Fig. 8 ) . Dr. Riese, in a personal

128 WILLIAMINA A. HIMWICH

DEVELOPMENT OF BRAIN I N NEONATAL PERIOD

129

communication, agrees that the cortex in the brain of the newborn dog is at the same stage of development as in the neonatal kitten. MacArthur and Doisy ( 1918-1919) have described three distinct periods of growth in the human brain. Cell division, the first phase, is probably complete in the 7-month-old fetus. Up to this time there is no evidence of any myelin development. From 7 months to about the time of birth the most important process is cell growth. The third period is that of myelination, which starts soon after birth and reaches its maximum a few months later. V. Accumulation of Chemical Constituents

The immature brain gradually acquires the chemical substances needed to serve as substrates for the functions that characterize the mature animal. Studies of the pattern of these acquisitions and the level of mature composition at which functions appear provide an opportunity to relate chemistry to function. Extensive studies are available on lipid deposition, protein accumulation, and changes in water and electrolyte and other substances as the brains of the young from several species mature. In this section of the chapter the complicated stoiy of changes in lipid composition as myelination proceeds will not be discussed, since the work although extensive, is still not definitive. Beginning with the early studies of the Kochs in 1913, the development of the rat brain has received more attention than that of any other animal, The changes which occur over the entire time of development are well documented, in most cases at 10-day intervals. Less is known about the day-to-day changes in the period before birth and in the first 10 days of life when the most rapid advances are being made. In the young rat the water content of the brain after birth rises slightly for a few days and then begins to fall (Fig. 9). The increases are small in whole brain (Donaldson and Hatai, 1931; Clouet and Gaitonde, 1956) and are of interest because of the changing patterns shown by other constituents during this time. Potassium content of whole brain appears to follow the same pattern as GA-G (Himwich et al., 1961; Millichap et al., 1958). Between day 4 and 7 of life brain sodium and chloride are high compared to older animals. During this period the level of carbon dioxide is very high, falling to the adult value at about 25 days of age (Millichap et aZ., 1958).

130

WILLIAMINA A. HIMWICH 89-

---+-----------.

, 88-

87,86 -

' 0

2

85-

'g.

84-

o-a A-- .A

83-

--

RAT MOUSE

i

'b

+----+ = RABBIT

82-

Before protein content of whole brain begins to increase, it decreases during the period 5 days before birth and the first days after parturition (Fig. 10). It follows in this fashion the changes in total solids (Clouet and Gaitonde, 1956). Glutamic acid taken alone or the sum of glutamic acid and glutamine (GA-G) has a high concentration in the fetal brain, decreases before birth, and by 10 days of age has risen again to approximately two-thirds of the adult level (Fig. 11) (Himwich and Petersen, 1959). The pyridine nucleotides increase rapidly until some time between day 10 and 21 and then are constant (Burton, 1957). From 10-20 days the deposition of lipid exceeds that of protein as myelination begins. The greatest increase of total lipids occurs between 10 and 20 days, and with a slight reduction of rate up to 40 days (Waelsch et al., 1941). Deoxyribonucleic acid (DNA) reaches maximal values in this species by day 16, an indication that cell proliferation is complete ( Mandel and Bieth, 1952). The mouse has also been studied to determine the pattern of brain development. The pattern of maturation in this animal is of

131

DEVELOPMENT OF BRAIN IN NEONATAL PERIOD

I

-40

o--o

-30

= GUlNEA PIG

-20

-10

0 10 AGE IN DAYS

20

. .

30

AWL

FIG. 10. Changes in per cent protein on a wet-weight basis in rat and guinea pig. [Data adapted from Flexner (unpublished observations) and Clouet and Gaitonde, 1956.1

particular importance in relation to that in other species because of the popularity of the mouse for neurochemical studies. As in other species water content falls from the level at birth to the adult value (Fig. 9). There is a slight rise in water content after birth and a subsequent return to the level present at birth. In this species nitrogen may slightly fall between birth and day 7 of life (May, 1948). Glutamic acid and glutamine (GA-G) start at a high level in the newborn and fall progressively to the adult level at 21 days (Fig. 12). (Although these results have been confirmed twice in our laboratory over a period of two years they do not appear reasonable. Some unsuspected artifact may be the answer.) It is difficult to tell from the data presented by Uzman and Rumley (1958) when the DNA of mouse brain reaches the adult level, because of the failure of the brains of their animals to increase regularly in weight during development and also because of the occurrence of some loss of brain cells. The Flexners (1951) have studied the nucleic acids in the developing cerebral cortex of the guinea pig from day 25 of gestation through term and have compared these data with the values ob-

132

WILLIAMINA A. HIMWICH

225 -

~

_--.--------~

,

- 80 - 70

200 -

175-

- 60

I50

- 50

s E" 126-

40

0" o 0

100 -

rn g % : ~

rng %:*--+=

= CONTROL

30

GLUTAMATE

-

75

50.

I

I1

21

31

41

61

CONCEPTUAL AGE

0

10

20

so

-

1

ADULT

AGE IN DAYS

FIG.11. Changes in glutamic acid and glutamine content of rat brain (as mg/100 gm) on a wet-weight basis with and without the intraperitoneal injection of glutamic acid; glutamic acid decarboxylase activity in homogenates reinforced with pyridoxal phosphate. For experimental details see Himwich and Petersen ( 1959), Himwich et al. ( 1957), and Himwich et al. (1961).

tained in the adult. Ribonucleic acid (RNA) shows a slight fall from day 25 of gestation to about day 47 and after that levels off at approximately the adult level. The DNA starts with very high levels at about day 25 of gestation and falls rapidly to about the same value as in the adult by day 45. These authors felt that there is an increase in the average amount of RNA per cell at the same time as cell processes begin to appear. The average concentration in the perikaryon appears to decrease, however, during and for some time after the period when the processes appear. This change may be owing to the volume of the perikaryon increasing more rapidly than does its content of DNA. In the guinea pig the amounts of water as well as of sodium and calcium, in both gray and white matter, decrease from just before birth ( 7 to 8 weeks gestation) until they reach the adult leveIs. Potassium shows little change. The decrease of sodium and calcium

133

DEVELOPMENT OF BRAIN I N NEONATAL PERIOD

200 -

175

-

150 -

8

F I25 100

-

MOUSE

75 -

I

mg % =V----7= CONTROL mg % =*-----a= GLUTAMATE

I

i 50

I

QCO,

=C-- 4= G A DECARBOXYLASE

I

I

I

0

10

I

20

I

30 40 AGE IN DAYS

--

ADULT

FIG.12. Changes in glutamic acid and glutamine content of mouse brain (as mg/100 gm) on a wet-weight basis with and without the intraperitoneal injection of glutamic acid; glutamic acid decarboxylase activity in homogenates reinforced with pyridoxal phosphate. For experimental details see Himwich and Petersen ( 1959) and Himwich et al. ( 1961) .

cannot be connected with myelination since it occurs to the same extent in gray and in white matter ( Wender and Hierowski, 1960). In the guinea pig whole brain, the content of free GA-G falls from a maximum of 174 mg/100 gm at 35 days of conceptual age to a low point at 44 days of conceptual age (Fig. 13). Shortly after birth the value is again high but never attains the level in the embryo (Himwich, in press). Similar changes have been observed in the total protein content of the guinea pig cortex (Fig. 10) (Flexner, unpublished observations). The rabbit brain develops apparently with much the same pattern as those of the mouse and the rat, but with different time

134

WILLIAMINA A. HIMWICH

t

1 7 ~ r

I

I

A

150.-

A/-

--- - - -

- -.

60

*.

_ _ - f

125-

3

E"

0" u

O\.

GUINEA PIG

100-

mg % A-O= 0

75.-

SO'

0

0

0

II

=&---A=

QC0 2 I

34

I

44

I

I

O-O=

I

54 64 74 CONCEPTUAL AGE I

1

0

10

CONTROL GLUTAMATE DECARBOXYLASE I

84

A

94

"V-L

20 30 AGE IN DAYS

ADULT

FIG. 13. Changes in glutamic acid and glutamine content of guinea pig brain (as ing/lOO gm) on a wet-weight basis with and without the intraperitoneal injection of glutamic acid; glutamic acid decarboxylase activity in homogenates reinforced with pyridoxal phosphate. For experimental details see Himwich and Petersen ( 1959) and Himwich et al. ( 1961).

relations. Water appears to remain constant from approximately day 15 before birth to the first day of postnatal life (Fig. 9 ) after which it decreases rapidly (Graves and Himwich, 1955). The percentage of nitrogen in the brain accumulates rapidly until day 30 after birth, after which time the change in concentration is slow (Kelley, 1956). As in the rat and guinea pig GA-G starts in the fetal brain of 20 days conceptual age at a level at least as high as that in the adult (Fig. 14), falls to a low point at birth, and rises rapidly to approximately the adult level at 30 days of postnatal life (Himwich and Petersen, 1959). DNA and RNA (Manukian, 1955) in the brain cortex fall progressively after fetal life and reach a constant level about 1 month after birth. The same is true of DNA of whole brain ( Mandel and Bieth, 1952). Few data are available on the biochemical changes in dog and cat brain as they grow (Figs. 15 and 16). GA-G levels in both

135

DEVELOPMENT O F BRAIN I N NEONATAL PERIOD

EO

225.

P 70

200

n 175

60

I50

50

125

40

I00

30

15

20

0" 0

8 m E

50

0

10 AGE

20 IN

30

3 M O . 5MO.

7MO.

ADULT.

DAYS

FIG. 14. Changes in glutamic acid and glutamine content of rabbit brain (as mg/100 gm) on a wet-weight b'isis with and without the intraperitoneal injection of glutamic acid; glutamic acid decarhoxylase activity in homogenates reinforced with pyridoxal phosphate. For experimental details see Himwich and Petersen ( 1959) and Himwich et al. ( 1961).

species start at a slightly higher fetal level than is present in the adult. The rise to the adult value is slow and probably does not occur until more than 30 days post partum. Sixty-seven per cent of the adult level is achieved by 1 day of age in the dog and in the cat at 3-5 days. In contrast to this early acquisition of glutamic acid, DNA content does not reach the adult level until 5 months after birth in the dog, and at 1 month in the cat (Mandel and Bieth, 1952). On considering the difficulties in obtaining material, an amazing number of studies of the human brain have been made. Although most of this work has been on adult samples, some data are available on the developing brain. The percentages of water and total ash of the cerebral hemispheres decrease progressively from the third month of conceptual life (Kimitsuki, 1955; MacArthu and Doisy,

136

WILLIAMINA A. HIMWICH

I

OCO, = C r . - ' - O = G A DECARBOXYLASE I

50

33

73 83 CONCEPTUAL AGE

L

I

I

0

10

20

: 7

I

1

63

53

43

93

103

I

30 40 AGE IN DAYS

;':

4Iw

ADULT

50

FIG.15. Changes in glutamic acid and glutamine content of dog brain (as mg/100 gm) on a wet-weight basis with and without the intraperitoneal injection of glutamic acid; glutamic acid decarboxylase activity in homogenates reinforced with pyndoxal phosphate. For experimental details see Himwich and Petersen ( 1959) and Himwich et al. ( 1961) . 160 140-

*ID E

-

120-

-= 100 SO I

a2

i 42

CONTIZOL

*--*= GLUTAMATE

*

62 72 82 CONCEPTUAL AGE

52

92

102

30

40

--.A2 0

10

20

ADULT

AGE IN DAYS

FIG. 16. Changes in glutamic acid and glutamine content of cat brain (as mg/100 gm) on a wet-weight basis with and without the intraperitoneal injection of glutamic acid. For experimental details see Himwich and Petersen (1959).

DEVELOPMENT OF BRAIN I N NEONATAL PERIOD

137

1918-1919) while total protein and lipids are increasing (Fig. 17). It is impossible to tell if the human brain shows a pattern similar to that seen in other species ( a high level of constituents which then falls to a minimum followed by a rise) before the third month of

A

AGE IN

MONTHS

FIG.17. Changes in composition of human brain during development. Figures combined from Tihey and Rosett (1931) and MacArthur and Doisy (1918-1919).

conceptual life. The flattening of the curve for GA-G content in the young cat and dog (Figs. 15 and 16), as compared to rat and rabbit, suggests that in more complex brains this phenomenon may not be seen. However, in the fetal cortex before 30 weeks of gestation potassium content is high, then falls and rises again before birth (Himwich, unpublished data). The total DNA reaches adult levels by one year after birth.

138

WLLLIAMINA A. HIMWICH

VI. Development of Enzymatic Activity

The time of appearance and the period of greatest increase in enzymatic activity are fertile fields in attempting to correlate function with biochemistry. The relation of enzyme activity to concentration of substrate is also important. Does substrate reach a given level before enzymatic activity becomes important or does the enzymatic activity stimulate the appearance of substrate? Unfortunately data are available now in only a few instances to attempt to answer these questions. The rat has been studied more intensively in this regard than any other species. Histochemical studies show postnatal increases in succinic dehydrogenase and in phosphorylase (Friede, 1958, 1959). In the last part of gestation the activities of succinic dehydrogenase and of ATPase are very low (Potter et al., 1945), the maximum rate of increase being at about day 10 of postnatal life with the adult level being attained at 30 days. Adenosine-5-phosphatase ( Naidoo and Pratt, 1954) appears, on the other hand, to increase at a nearly uniform rate throughout life. DPNase is low at birth (Burton, 1957) and rises rapidly to the adult level at 23 days of age, the maximum rate of increase being between 15 and 20 days of age. Carbonic anhydrase is low in activity until day 12 and then increases rapidly to the adult level at 28 days (Millichap, 1957; Ashby and Schuster, 1950). In general it might be said that of the enzymes concerned with respiratory metabolism only adenosine-5-phosphatase deviates from the general pattern. Carbonic anhydrase is the last of the enzymes studied to reach the adult level. Alkaline phosphatase is low from birth to 28 days; acid phosphatase is higher than the alkaline. Both enzymes show a slight tendency to be higher at 28 days than at earlier periods ( Himwich, unpublished data). Glutamic acid decarboxylase activity increases rapidly from birth to 30 days of age (Fig. 11). Both pseudocholinesterase and acetylcholinesterase per unit of weight begin to increase at 8 days of age and reach the mature level at 22 days of age (Elkes and Todrick, 1955). On the basis of assay of 100-mg samples of fresh brain, the enzymatic activity increases from birth to day 110 ( Nachmansohn, 1939). In studies on the mouse only two enzymes have been found

DEVELOPMENT OF BRAIN IN NEONATAL PERIOD

139

during the period of development, namely phosphatase and glutamic acid decarboxylase (Fig. 12). Alkaline and acid phosphatases have been studied both histochemically and by chemical assay. The high concentration of alkaline phosphatase appears to accompany early stages of morphological differentiation (Chiquoine, 1954). By 14 days of conceptual age the activity is decreasing in most of the central nervous system but is still on the increase in areas of the telencephalon. The adult animal shows low levels of this enzyme in all parts of the brain. Chemical assay of whole mouse brain from day 1 of life to day 28 shows a consistent low level for this enzyme with a suggestion of rise to a low peak at 10-17 days (Himwich, unpublished data). The extreme variability of the data makes it difficult to draw definite conclusions. The acid phosphatase activity is approximately double that of the alkaline during growth. The development of glutamic acid decarboxylase in this species follows a smooth pattern of increase (Roberts et al., 1951), being slightly higher in the animal at the beginning of maturity than in the adult. The most extensive studies on guinea pig have been conducted by the Flexners and their colleagues (1953, 1955) on the cortex of the brain. ATPase, succinic dehydrogenase, and cytochrome oxidase begin to increase at about 42 days of gestational age and rise rapidly so that at birth they are near the adult level. On a dry-weight basis acid phosphatase falls after birth to the low level characteristic of the adult. Alkaline phosphatase on a wet-weight basis in whole brain decreases from birth to adult life ( Himwich, unpublished data). The cholinesterase of the motor cortex (Kavaler and Kimel, 1952) on a fresh-tissue basis shows a rise at a steady rate from day 29 of gestation until 10 days post parturn. If cellular weight is taken as the basis for comparison, the activity begins to increase sharply at day 35 of gestation and continues until about day 57, after which it remains at approximately the same level until at least 7 days post purtum. The cholinesterase activity of whole guinea pig brain in terms of total activity rises relatively little between 7 and 19 days of age, the increase being due entirely to the change in the weight of the brain, not to enzyme activity ( Nachmansohn, 1939). Choline acetylase, on the other hand, develops at a rather constant rate and does not reach a peak in the cerebrum until 60 days after birth (Hebb, 1956). In the cerebellum the peak occurs at

140

WILLIAMINA A. HIMWICH

about 45 days of gestation. DPNase in the cortex is low until day 55 of conceptual age (Nemeth and Dickerman, 1960) and then rises rapidly to the adult level. Succinic oxidase measured in a homogenate showed greater activity than in a mince until about 45 days of gestation age (Flexner et al., 1953). This activity depends upon an enzyme chain some of whose components are succinic dehydrogenase, cytochrome oxidase, and cytochrome c. Carbonic anhydrase is at the adult level in this species (Ashby and Schuster, 1950). Glutamic acid decarboxylase starts at a high level in the fetus, falls, and then shows an increase to the previous level before birth (Fig. 13). The rabbit has been studied less extensively than the rat or guinea pig, Succinic oxidase activity reaches the adult level by 1% 15 days of age, and succinic hydrogenase begins to rise sharply at about day 10 and approaches the adult level shortly after day 20 ( Cassin, personal communication). The cerebrum and the cerebellum show markedly different patterns of the acquisition of choline acetylase activity. In the former, activity rises constantly from about day 20 of conceptual life to reach the adult level between 2 5 3 0 days of life (Hebb, 19%). In the cerebellum the peak of activity is reached 5-7 days after birth and values fall to about the adult level by about day 30 of life. In rabbit brain, the changes in glutamic acid decarboxylase (Fig. 14) parallel the concentrations of glutamic acid and glutamine (Himwich et al., 1961). The data available on cholinesterase in whole rabbit brain suggest that the adult level in whole brain is achieved by approximately 30 days of age (Nachmansohn, 1939). Cholinesterase activity in the 21-day-old cat (QChE) is considerably greater than in the newborn (Nachmansohn, 1939). Unfortunately data are not available for the adult cat brain. Carbonic anhydrase is low in both puppy (dog) and kitten brains (Ashby and Schuster, 1950). Because of the poor condition of the brain in human babies at death, it has been difficult to conduct enzyme studies. Carbonic anhydrase is low in the cerebrum (Ashby and Schuster, 1950), and succinic dehydrogenase in the cortex appears to increase with age ( Himwich, unpublished data). Glutamic acid decarboxylase develops in the same pattern as GA-G in the dog (Fig. 15).

DEVELOPMENT OF BRAIN I N NEONATAL PERIOD

141

VII. Hematoencephalic Exchange in the Developing Brain

A great number of studies on young animals with a wide variety of substances suggests that the so-called blood-brain barrier is not present for most materials in the immature animal. It is an attractive theory that the brain would accept material almost at random during the period that the nervous tissue is growing and developing. On this basis a given degree of maturity would automatically slow hematoencephalic exchange. The development of the adult type of exchange would, therefore, offer a possible correlation with other parameters of the maturation process in young animals. Pj2 (Fries and Chaikoff, 1941), potassium (Katzman and Leiderman, 1953), sodium ferrocyanide (Stern and Peyrot, 1927), and glutamic acid (Himwich et al., 1957) (Fig. 11) enter the brain of the rat readily until about 2 weeks after birth. On the other hand, Grazer and Clemente (1957) showed that even in the 12-day-old embryo trypan blue did not penetrate the central nervous system. In the developing mouse the areas known to be outside the blood-brain barrier, such as the area postrema and the tuber cinerium, appear larger than in the adult ( Behnsen, 1927). This difference, however, does not necessarily mean that the blood-brain barrier is less well developed in the newborn mouse than in the adult. Glutamic acid appears to enter freely until about day 21 after birth (Fig. 12) (Himwich et al., 1961). Ferrocyanide penetrates the brain and remains in the cerebrospinal fluid until day 12-15 (Stern and Peyrot, 1927). The guinea pig brain at the time of birth manages to exclude ferrocyanide (Stem and Peyrot, 1927) but continues to admit glutamic acid apparently even in adult life. The results of studies on glutamic acid shown in Fig. 13 may be due to the fact that socalled “free” glutamic acid and glutamine content are lower than the expected values in the adult guinea pig (Himwich and Petersen, 1959). If some mechanism blocks the extraction of these substances from the adult brain of this species and if the brain glutamate is exchanged on a 1:1 molecular basis as in the mouse (Roberts et nl., 1959), the recently entered glutamate may be more available for extraction. The rabbit brain shows a rapid acquisition of P3?, which

142

WILLIAMINA A. HIMWICH

diminishes from the fetal period to adult life (Manukian, 1955; Bakay, 1953).In general the adult response is evidenced by 30 days of age. Newborn rabbits are not susceptible to the lesions caused by penetration of bilirubin, apparently due to failure of bilirubin to enter the tissue (Rozdilsky and Olszewski, 1960). If glutamic acid enters the brain freely it is only in the perinatal period ( Fig. 14). In kittens it has been possible to study the blood-brain barrier to y-aminobutyric acid by following the blockade of evoked potentials caused by this material (Purpura and Carmichael, 1960). This type of experiment shows that the amino acid enters the brain freely from the blood. Glutamic acid, by contrast (Himwich et al., 1961), does not penetrate the brain even in the fetus (Fig. 16). The kitten also develops a true kernicterus when the bilirubin content of the blood is raised ( Rozdilsky and Olszewski, 1961) . Ferrocyanide is accepted by the brain from the blood until the kittens’ eyes open at about 2 weeks (Stem and Peyrot, 1927). Sodium (Bakay, 1960) rapidly exchanges between blood and brain and so accumulates in the brain of the fetal cat, Few data are available in this regard for the dog. Neither bilirubin (Rozdilsky and Olszewski, 1961) nor glutamic acid enters the brain readily in the newborn dog, nor does glutamic acid in the fetal dog (Fig. 15) (Himwich et al., 1961). In the human baby, evidence on the hematoencephalic exchange is available only from studies of kernicterus (Grontoft, 1954). Clinical experiences show that if the blood bilirubin rises to 20 mg/100 ml kernicterus will ensue (Diamond, 1952). The whole area of interpretation of hematoencephalic exchange is fraught with difficulty because we do not understand the mechanisms involved. Certainly no generalization can be drawn, and data cannot be assumed to apply to any species except the one studied and only under the specific experimental conditions observed. VIII. Neurophysiological Development

The tracing of the pattern of neurophysiological development is of special importance in attempting to relate results obtained on animals to man. In this area of research-perhaps more so than in any other-it is necessary for normal function that the animals be warm, well fed, not dehydrated, and not subjected to pain or dis-

DEVELOPMENT OF BRAIN I N NEONATAL PERIOD

143

comfort. In young animals pain may produce a prolonged, often inappropriate, response, usually after a relatively long latent period, probably owing to immaturity of the cortex. This type of response is similar to that of patients very lightly anesthetized with thiopental (stage 1 A ) (Himwich and Etsten, 1946). It must be remembered also that within a given period of time the nervous system of the adult is able to discriminate far more events than that of the immature animal ( Scherrer and Oeconomos, 1954). The anesthetic agent used, the depth of anesthesia, and the adequacy of oxygenation in the anesthetized or curarized animal affect the results obtained in immature animals even more than in the adult. Differences in technique such as these make it difficult to compare work from one laboratory with that of another. A. SPONTANEOUS EEG Crain (1952) was the first to describe the development of the spontaneous EEG in the rat. By day 7-10 of postnatal life there is a trend to rhythmicity and to continuous activity. The recordings are much like those in an adult on day 10. Other workers have suggested day 12 as the beginning of an adult-type EEG (RadoucoThomas, personal communication). The greatest increase of cortical polarization is between 5 and 10 days post partum according to Burei (1957), who also notes that in some brains spreading depression appears for the first time at 15 days, and that at 20 days it may be elicited regularly, In mouse the EEG during the first several days of life is characterized by low potentials and simplicity. Between the days 6 and 9 the EEG develops quickly and from day 10-14 the EEG in the well-nourished young mouse is similar to that in the adult (Kobayashi, personal communication). The guinea pig traces an adult-type EEG while still in utero (44-46 days) and at birth the young animal has definite cortical polarization and spreading depression (Jasper et al., 1937; Flexner et al., 1950; Marinesco et al., 1936). The developing EEG in the rabbit has been thoroughly studied ( Schadk, 1959a, b; Bradley et al., 1960). It assumes an adult type pattern 10-15 days after birth, and spreading depression appears 24-30 days after birth. Some electrical activity, however, can be recorded

144

WILLIAMINA A. HIMWICH

as early as day 31 of gestation and slight regional differences are apparent as early as day 5 postnatal life (Bradley et al., 1960). The tracings made from visual and motor areas of the cortex in the newborn are indistinguishable from one another according to Pentzik (1937). After day 7 the potentials of the visual zone become progressively larger in amplitude, whereas in the motor area the increase is less marked. By day 14 post parturn, the recordings from both areas do not differ essentially from those of the adult. The curarized kitten shows low-amplitude random activity at 2-6 days. By 25 days of age well-defined activity is present but it still lacks the fast components typical of the adult pattern (Libet et al., 1941; Marinesco et al., 1936). In the unanesthetized animal with chronically implanted electrodes, Scheibel ( 1962) found the adult type of EEG beginning at 10-14 days and well established by 3-5 weeks. Spindling could be seen as early as day 5-7. The early spontaneous activity of dog cerebral cortex has not been thoroughly explored. Two laboratories ( Charles and Fuller, 1956; Gokhblit, 1958) report no difference in sleep and wakefulness before 18-20 days and consider that EEG patterns comparable to those traced by the adult cortex are not obtained until at least 4-5 weeks of age. However in our laboratory, working with the curarized puppy, kept warm and well-fed, we have found what appear to be sleep spindles separated by low-amplitude irregular waves as early as the first postnatal day ( Petersen and Himwich, 1959) (Fig. 18). In the human infant many studies have been conducted on the development of the EEG. Dreyfus-Brissac (1956a, b) has reported initial potential variation in the premature infant. The activity, however, is probably of subcortical origin. As Ellingson (1958) has pointed out, many problems in human infant electroencephalography are due to the difficulties of recording satisfactory EEG’s during wakefulness. Differences during sleep and wakefulness have been reported in the newborn (Smith, 1938; Hughes et al., 1949; Ellingson, 1958). The closing of the eyes in the infant should not be considered as evidence of sleep since the eyes may be open while the EEG tracing indicates that the baby is asleep. In some newborn infants sustained rhythmical sequences appear in addition to bilateral synchrony ( Ellingson, 19%).

DEVELOPMENT OF HHAlN IN NEONATAL PEHIOD

145

ONE DAY OLD DOG RIGHT

LEFT

:zj €KG

i

&

-

t

wM 4.M.24-JCP-I

TWO DAY OLD DOG RlCHT

LEFT

y y

EKG

&. FIG. 18. The cortical EEG in thc young puppy. Spindling bursts are at the rate of lO-l2/sec.

B. EVOKED POTENTIALS Analysis of the evoked potentials gives an additional measure of the development of regional differences in the cortex of the young animal. The more complex the cortex the greater the variations in evoked potentials recorded from various areas at a given age. A long latent period and relatively large negative waves, fatiguability, and failure of repetitive responses characterize the evoked potentials in all immature brains. The rat will respond slowly with a strychnine spike as early as day 4 after term (Crain, 1952), but evoked potentials to sound do not appear until postnatal day 10-15 ( BureJ 1957). In the mouse, seizure patterns with pentylenetetrazole do not appear before day 14-16 (Kobayashi, personal communication). The guinea pig

146

WILLIAMINA A. HIMWICH

responds strongly to afferent stimuli before birth (60 days of conceptual age), and a strychnine spike first appears in the 46-day-old fetus ( Flexner et al., 1950). Bishop (1950) found that the rabbit shows a strychnine spike of extremely long duration on the first postnatal day, but long intervals appear between the spikes. The mature pattern does not appear until 20 days of age. The latency of evoked response to visual stimulation does not reach the adult level until 4 weeks of age (Hunt and Goldring, 1951). The optic nerve will respond to electrical stimulation by a cortical potential 24 hours after birth but the visual cortex will not respond to light during the first week of life. This response shows largely a surface negative component which becomes reduced in amplitude between 11 and 21 days. The initial surface positive component does not appear until later (Hunt and Goldring, 1951). Developmental patterns of the evoked potential have been studied extensively in the kitten and apparently neglected in the dog. Potentials have been recorded only from the somesthetic system in the newborn cat (Scherrer and Oeconomos, 1954). Other systems do not appear to mature sufficiently to give responses until 3-11 days post partum (compare Grossman, 1955). By day 10 visual potentials regularly appear and by day 20 the zone of recorded responses is comparable to that in the adult (Marty et al., 1958). Direct stimulation of the cortex by electrical current fails to show the dendritic component until day 16 after birth. The auditory system appears to mature before the visual system. Auditory evoked responses appear first at 5-10 days of age and are usually biphasic positive-negative at the beginning ( Ellingson and Wilcott, 1960). A small initial negative wave possibly of subcortical origin has been reported as early as 1 week of age (Rose et al., 1957). In the visual cortex normal latencies do not appear until about 2 weeks of age although some individuals show an earlier maturation ( Ellingson and Wilcott, 1960). Action potentials recorded from the primary sensory regions of the cortex following stimulation of the cutaneous or motor afferent nerves do not show a positive phase until day 3 (Malcolm, 1954). Individual variation in strength of response and in age of first appearance is marked in kittens. Scheibel (1962) has reported on the marked fatiguability of cortical responses to reticular formation stimulation.

DEVELOPMENT OF BRAIN I N NEONATAL PERIOD

147

Human infants offer an excellent opportunity to study evoked potentials. As in animals they show a long latency which does not reach the adult level until 9-12, weeks of age (Ellingson, 1960). Responsiveness to repetitive stimuli is very poor at birth but increases with age. The visual evoked responses show a great variability of form and amplitude. Ellingson (1960) suggests that the presence of a positive phase in an evoked potential may serve as a criterion of maturity of the responding system. He finds, in contrast to Kellaway (1957), that visual evoked responses appear earlier in human infants (week 28 after conception) than in most young animals. An initial negative response is seldom seen in the human infant and the response is of higher voltage than in the adult. This difference may be due to the position of the primary visual cortex which is buried in the adult brain and therefore does not permit optimal responses to be recorded at the surface of the cortex. IX. Correlation of Anatomical, Chemical, and Neurophysiological Factors

It seems appropriate that any attempted synthesis of the parameters discussed above should start with Flexner’s (1955) elegant discussion of the so-called “critical period in the guinea pig. This period is defined as that time during which the neurons approach maturity as judged by the nuclear volume, presence of Nissl substance, and dendritic processes. These anatomical changes are accompanied by a rapid increase in the activity of respiratory enzymes and by accumulation of protein and amino acids to more than half of the adult level. This degree of development in the guinea pig appears to furnish the anatomical structure and the biochemical substrate necessary for the adult type of cortical electroencephalogram which appears around day 4 4 4 6 of gestation (Fig. 19). The picture in this species is clear-cut and, except for data on the hematoencephalic exchange of glutamic acid discussed above, presents no obvious discrepancies. If a similar pattern can be fitted together for other species, then the appearance of a mature EEG in a young animal of any species can be assumed to indicate a similar degree of anatomical and biochemical maturity. The rat offers us the best opportunity to test this hypothesis. Jordan and his colleagues (1956) suggested, on the basis of anatomical data and the appearance of the EEG, that the critical period

I

C E L L PROLIFERATION I

C E L L GROWTH.

AXONS AND D E N D R I T E S GROWTH, MYELINATION

I AGE A F T E R BIRTH

I CONCEPTUAL AGE, DAYS.

. . .

.

34

l

.

.

.

39

.

l

.

.

.

.

44

I

.

.

.

49

.

l 54

.

.

.

.

l

.

.

59

.

.

I l

.

.

.

.

64

I

C 5 Cortical layers.

+ Ground s u b s t a n c e . + N e u r o n s max.

ANATOMY

size.

f Nissl substance.

+Dendritic p r o c e s s e s . 1 I

+Adult

C Adult

I

l e v e l RNA.

+H20 falls. ------- --•

l e v e l DNA

+N a . C a

BIOCHEMISTRY +Glut.

NEUROPHY SIOL.

d e c a r b . , falls.-

I

I

Adult l e v e l .

C Respiratory enzymes

I

BEHAVIOR

type E E G .

I

I

p r e a d i n g d e p r e s s ion.

C R e s p o n s e to d f e r e n t atirnuli.

<

>

Adult reactions.

Eyes open. ~~

Y

increase.

cnnine spirre.

~

- -*

fall.-------

~

FIG. 19. Correlations of various parameters in developing brain of guinea pig. Arrows pointing left indicate earliest age at which reported. Dashed arrows pointing right indicate age of complete development unknown or unspecified. Double-headed arrows indicate the range in age over which an observation has been recorded. For exact times and values and references, see text. Major subdivisions of brain growth are taken from McIlwain's (1955) periods of cerebral development. An open right end on a heading indicates that the evapt

enrl nF the nerinrl i c

I l n l m m x m nr

lotor thQn tho

onec

inolvlcleA

in

the r h a r t

1

I

CELL PROLIFERATION

C E L L GROWTH, AXONS AND DENDRITES MYELINATION, I

I 1D

i:

AGE A F T E R BIRTH

fa

k

n

m I

ONCEPTUALAGE DAYS.

%

L

,

I

I

#

1

1

1

1

0

%

%

fa

fa

Q

Q

4.

26

21 I

%

m

n 0

Ln

1

1

1

1

~

31

1

1

1

36

I

Cortical iffe rentiation begun.

1

N

.

1

1

1

1

N

.

I

I

+End +Mature

bcr'ease

I

BIOCHEMISTRY

I

>

-~ - - - - T o a d u l t level.--,

+Enzymes

increase.-

C S 1 . i n c r e a s e w a t e r followed by d e c r e a s e .

- - - - - - - - -- -- - - - - - - -j __--

I

(High Na&C1+

+Max.

C G r e a t i n c r e a s e lipids.

Low K

4 Rhythmical and NEUROPHYSIOL.

.+

I

Slow strychnine spike.

c--, Sucking Crawling

continuous EEG.

+r

cell pro1if:ration.

c o r t i c a l organization and n u c l e a r vol.

C Dendritic growth.

-3

46

41

I

N i s s l and ground sub.

+ F a l l GAG .& prot. +Low glut. decarb. act.

BEHAVIOR

GROWTH O F BRAIN

I " " ~ " " l " " l " " l " " l I

ANATOMY

0

0

YI

I

0

DNA.

>

Adult type EEG. j +Spreading 4 depression.

-

c Evoked potential j to sound.

E y e s open.

<,

>

Washing Climbing Running Walking I

FIG. 20. Correlations of various parameters in developing brain of rat. For meaning of symbols see Fig. 19; for exact times and values and references. see text.

.

CELL

PROLIFERATION

AXONS AND DENDRITES

C E L L GROWTH,

GROWTH, MYELINATION

1I

I

I

d

AGE AFTER BIRTH

1 ; : CONCEPTUAL AGE. DAYS

~

:

l

~

:

m

‘FI

Q

Ln

;

l

:

;

10

15

20

25

I

I

I

I

l

;

~

;

;

30

I

I

I

I I

+Glut.

I I

-

i

~

40

1

1 >

rises ----falls--+

b~ falls +rises+

I

decarb. r i s e s - - - - - . - - - - - - - - - - - - T o

-

NEUROPHYSIOL.

BEHAVIOR

I

;

I

I

BIOCHEMISTRY

N

;

C Cortex h a s adult thickness

f- Dendritic growth

+HzO

0

;

35 I

I Ground subst. appears

ANATOMY

Q

‘LI

In

d

.

I >.

i !

‘FI 0

;

I

>.

111

>.

x

.r(

:

I

I adult l e v e l

.-+

>

c

cb

E y e s open

FIG. 21. Correlations of various parameters in developing brain of mouse. For meaning of symbols see Fig. 19; for exact times and values and references, see text.

;

;

:

l

~

DEVELOPMENT OF BRAIN IN NEONATAL PERIOD

151

in this animal would be between day 4 and day 10 post partum. This age marks also the time when respiratory enzymes as well as glutamic decarboxylase activity are developing rapidly, and protein and amino acids are accumulating at a rapid rate (Fig. 20). The dendritic growth begins after day 6 and is especially rapid after day 12. At this time the mature type EEG has been observed by all investigators. In this species, maximum dendritic growth probably occurs at the time of or immediately after the development of an adult-type EEG. On the other hand, as in the guinea pig, the data on the hematoencephalic exchange does not fit into this period. The young mouse shows a rapidly developing EEG between day 6 and 9, which tends to become similar to the adult by day 10-14 of life (Fig. 21). In anatomical development the cortical thickness is the same as the adult by day 12. Dendritic growth is rapid between day 8 and 15. Few biochemical studies are available for correlation, but glutamic decarboxylase is increasing rapidly at this time, and the sum of glutamic acid and glutamine is approaching the adult level. It would seem that the “critical period” in this species might lie between day 10-15 post partum. The rabbit brain appears to be a transition between the simple brains of the mouse, rat, and guinea pig and the more complex ones of the higher mammalian species (Fig. 22). In this animal regional differences in electrical activity tend to develop early (Pentzik, 1937). Few data, however, are available for the rabbit insofar as enzymatic and chemical development is concerned, Examination of available biochemical data shows that protein as well as glutamic acid and glutamine are accumulating rapidly soon after birth. The respiratory enzymes are also increasing or have reached the adult level at this time. By day 10-15 the anatomical characteristics seem sufficiently developed to support the cortical EEG, and, as would be expected, an adult type of record is obtained at this time. A “critical period” therefore could be assigned to this species at 10-15 days post partum. The adult level of DNA, however, does not appear before day 30 post partum. In the more complex brains of dog, cat, and man (Fig. 2 3 ) it is more difficult to assign the “critical period.” In these brains regional differences are greater, and the rate of maturation of various cortical areas differs markedly. The young of all three species are born with the cortical layers well differentiated. In the dog and cat by 2 days

C E L L PROLIFERATION

I

C E L L GROWTH.

AXONS AND D E N D R I T E S

I

GROWTH,

MYELINATION

AGE AFTER BIRTH CONCEPTUAL AGE. DAYS

*

I

&Dirt. pyr. cells and 5th layer. C F i r s t growth period.

ANATOMY

C H Z O

> decreasea ------3

constant +N

+GAG

+Glut.

high d c c a r b . high.

C D e n d r i t i c3 growth. C V i s u d cortexj mature% + 2nd growth -+ C N i s s l and ground r u b ' s . period.

>

i n c r e m e s rapidly.. increase.

low

Add?leveI

BIOCHEMISTRY

Succinic dehydrogenase.

~

~~

+Firat

NEUROPHYSIOL.

electrical activity

#Some regional , E E G difference..

+Strychnine .pike long latency.

--

CVisual Cortex responds to electrical stimulation. BEHAVIOR

CRemponss to light.

-

A d d t level cholinestcrast

+Adult E E C viaual and m o t o r aream. m a t u r e tvpe+

I

m a t u r e s to 21.t

+Eye.

day.*

open3

FIG.22. Correlations of various parameters in developing brain of rabbit. For meaning of symbols see Fig. 19: for exact times and values and references, see text.

C E L L GROWTH,

AXONS AND DENDRITES GROWTH AND MYELINATION

I

AGE

AFTER BIRTH, CONCEPTUAL AGE, WEEKS

I

I

I

I

18 I

28

38

I

1

I

d

ID

m A

k

0

s

I

I

5

D

.I

8

I

I

I

5

;

ANATOMY

BIOCHEMISTRY

I

-

C K , high- - l o w - - - i n c r e a s e s

-

E E G probably of subcortical origin

Sustained rhythmical activity

NEUROPHYSIOL. Bilateral synchrony

Evoked visual potential

BEHAVIOR

+ - - -- - _ _ -- +

potentials

_ _ _ ____E ye s open, functional I

,

-j

------+

I

sucking

Turning Sitting Crawling

I

--- - ---. -- - - - --

FIG.23. Correlations of various parameters in developing brain of man. For meaning of symbols see Fig. 19; for exact times and values and references, see text.

154

WILLIAMINA A. HIMWICH

after birth well-developed dendritic processes can be seen although the intense arborization is not yet established. In dog and man differences in sleeping and waking EEGs can be elicited soon after birth. The spontaneous EEG in the dog does not begin to resemble the adult form before 28 days after birth at the earliest. In the dog the level of decarboxylase activity proceeds to rise rapidly after birth. The differences in glutamic acid and glutamine levels with development are not as marked in dog and cat as they are in simpler brains. Also, there does not appear to be even in fetal life of the cat and dog a time when glutamic acid accumulates easily in the brain. In these animals, as well as in man, the brain may develop in a slightly different fashion than in rat, rabbit, and guinea pig. It may be that some parameters will be found to apply to relatively simple brains but not to the more complex ones. Theoretically some other points should permit similar correlations although the data may be too scanty to do this for all species. The time when the glutamic acid and glutamine content reaches a low point probably has correlates in neuroanatomy that are still unknown. For the guinea pig, rat, and rabbit these times are 44 days of gestation, birth, and birth, respectively; for the rat and rabbits, these times fall before the “critical period.” The data are insufficient in the other species to attempt such an analysis. A similar correlation can be attempted at the time when any single constituent or group of chemical constituents reaches a maximum. In the rat the maximum values of GA-G and total protein after birth are achieved after 30 days of life. Similar ages for guinea pig and rabbit are approximately 55 days of conceptual life and 3 months after birth. The correlation of various phases of development (Figs. 19-23) suggests that in the rat and guinea pig events leading to maturity take up a short period of time either pre- or postnatal, whereas in the rabbit and man the maturation is a slower process. Adequate data are not yet available to make similar comparisons for the other species. X. Conclusions

Since our major interest lies in an attempt to relate development of the brain in experimental animals to that in man, an estimation of the comparable physiological ages in various species is necessary. Inspection of the summary charts suggests several considerations:

DEVELOPMENT OF BRAIN IN NEONATAL PERIOD

155

(1) Guinea pig, rat, and rabbit can probably be compared at the “critical period without undue oversimplification. The longer period required by the rabbit to reach maturity must be considered in any comparisons made at a later age. ( 2 ) More data are needed on developing mouse brain. If the divergent results noted in this chapter can be explained in terms of artifacts due to the environment, the “critical period of cerebral development would appear to correspond roughly to that in the rat. ( 3 ) The maturation of the brain in the cat and dog seems, from the limited data at our disposal, to be approximately the same. (4)The human infant is born more mature than the rat and rabbit (cf. cell proliferation, EEG, etc.) and on the basis of the visual evoked potential more mature than the cat. After birth, development in man appears to proceed more slowly than in the other species examined. It would not be wise to consider as final the neat correlations that are shown for rat and guinea pig. It may be that the apparent simultaneous development in several parameters is due to the greatly shortened period of development in these species. Many more data on the patterns of anatomical, biochemical, and neurophysiological maturation in the more complex brain are necessary before final conclusions can be drawn. REFERENCES Anokhin, P. (1956). Conference on the Problem of Individual Development. Kiev, U.S.S.R. Ashby, W., and Schuster, E. M. (1950). J. Biol. Chem. 184, 109. Bakay, L. (1953). A.M.A. Arch. Neural. Psychiat. 70, 30. Bakay, L. (1960). Neurology 10,564. Behnsen, G. (1927). Z. Zellforsch. u . mikroskop. Anat. 4, 515. Bishop, E. J. (1950). Ebctroencephalog. and Clin. Neurophysiol. 2, 309. Bradley, W., Kaiser, I., Morrell, F., and Nelson, E. (1960). In “Mental Retardation, Proc. 1st International Medical Conference (P. W. Bowman and H. V. Mautner, eds.), p. 98. Grune and Stratton, New York. Brizzee, K. R., and Jacobs, L. A. (1959a). Growth 23, 337. Brizzee, K. R., and Jacobs, L. A. (1959b). Acta Anut. 38,291. Bureb, J, (1957). Electroencephulog. and Clin. Neurophysiol. 9, 121. Burton, R. M. (1957). J. Neurochern. 2, 15. Charles, M. S., and Fuller, J. L. (1956). Electroencephulog. and Clin. Neurophysiol. 8, 645. Chiquoine, A. D. (1954). J. Cornp. Neutol. 100, 415. Clouet, D. H., and Gaitonde, M. K. (1956). J. Neurochem. 1,126. Grain, S. M. (1952). Proc. S O C . Exptl. Biol. Med. 81, 49.

156

WILLIAMINA A. HIMWICH

Diamond, L. K. ( 1952). Pediatrics 10,337. Donaldson, H. H., and Hatai, S. (1931). J. Comp. Neural. 53, 263. Dreyfus-Brisac, C., and Blanc, C. ( 1956a). Ence‘pluzk 45,205. Dreyfus-Brisac, C., Samson, D., and Monod, N. (1956b). Ebctroencepluzlog. and Clin. Neurophysiol. 8, 171. Eayrs, J. T. (1960). In “Structure and Function of the Cerebral Cortex” (D. B. Tower and J. P. SchadB, eds.), p. 43. Elsevier, Amsterdam. Eayrs, J. T., andGoodhead, B. (1959). J. Anat. 93, 25. Elkes, J., and Todrick, A. ( 1955). In “Biochemistry of the Developing Nervous System” (H. Waelsch, ed. ), p. 309. Academic Press, New York. Ellingson, R. J. (1958). Electroencephalog. and Clin. Neurophysiol. 10, 31. Ellingson, R. J. (1960). Electroencephalog. and CEZn. Neurophysiol. 12, 663. Ellingson, R. J., and Wilcott, R. C. (1960). J. Neurophysiol. 23, 363. Flexner, L. B. (1955). In “Biochemistry of the Developing Nervous System” (H. Waelsch, ed.), p. 281. Academic Press, New York. Flexner, J. B., and Flexner, L. B. (1951). J. Cellular Comp. Physiol. 38, 1. Flexner, L. B., Tyler, D. B., and Gallant, L. J. (1950). J. Neurophysiol. 13,427. Flexner, L. B., Belknap, E. L., Jr., and Flexner, J. B. (1953). J . Cellular Comp. Physiol. 42, 151. Folch-Pi, J. (1955). In “Biochemistry of the Developing Nervous System” (H. Waelsch, ed.), p. 121. Academic Press, New York. Friede, R. L. (1958). 2. physiol. Chem. Hoppe-SeyWs 310, 4. Friede, R. L. (1959). J. Neurochern. 4, 101. Fries, B. A., and Chaikoff, I. L. (1941). J. Biol. Cham. 141, 479. Gokhblit, I. I. (1958). B i d . Eksptl. B i d . Med. 47, 790. Goodhead, B. (1957). Acta Anat. 29,297. Graves, J., and Himwich, H. E. (1955). Am. J . Physiol. 180, 205. Grazer, F. M., and Clemente, C. D. (1957). Proc. Soc. Exptl. B i d . Med. 94, 758. Grontoft, 0. ( 1954). Actu Pdhol. Microbwl. Scand. 34, Suppl. 100, 8. Grossman, C. (1955). A.M.A. Arch. Neurol. Psychiat. 74, 186. Hebb, C. 0. (1958). J. Physiol. (London) 133, 566. Hess, A. (1955a). A.M.A. Arch. Neurol. Psychiat. 73, 380. Hess, A. (195513). J . Comp. Neurol. 102, 65. Himwich, H. E., and Etsten, B. (1946). J . Neruous Mental Disease 104, 407. Himwich, W. A. ( 1%). American Psychiatric Association Research Conference. ( In press.) Himwich, W. A., and Petersen, J. C. (1959). In “Biological Psychiatry” (J. Masserman, ed.), p. 2. Grune and Stratton, New York. Himwich, W. A,, Petersen, J. C., and Allen, M. L. (1957). Neurology 7, 705. Himwich, W. A., Benaron, H. B. W., Tucker, B. E., Babuna, C . , and Stripe, M. C. (1959). J. Appl. Physid. 6,873. Himwich, W. A., Petersen, J. C., and Graves, J. P. (1961). In “Recent Advance.~in Biological Psychiatry” (J. Wortis, ed.), Vol. 3, p. 218. Grune and Stratton, New York. Hughes, J. G., Ehemann, B., and Hill, F. S. (1949). Am. J. Diseases Children 77, 310.

DEVELOPMENT OF BRAIN Ih- NEONATA4L PERIOD

157

Hunt, W. E., and Goldring, S. (1951). Elcctroencephalog. und Clin. Neurophysiol. 3, 465. Jasper, H. H., Bridgman, C. S., and Ccirmichciel,L. (1937). J. Exptl. PsychoZ. 21, 63. Jordan, W. K., March, R., and Mes+g, R. A. (1956). I n “Progress in Neurobiology: I. Neurochemistry” ( S R. Korey and J. I. Nurnberger, eds.), p. 101. Hoeber, New York. Katzman, R. K., and Leidernian, P. H. L. (1953). Am. J. Physiol. 175,263. Kavaler, F., and Kimel, V. M. (1952). J. Comp. Neurol. 96, 113. Kellaway, P. (1957). Congr. intern. sci. neurol. 1st Brussels, 1957, 3rd Congr. intern. neuroputhol., Rappts. et discussions, p. 141. Kelley, B. (1956). Am. J. Physwl. 185, 299. Kimitsuki, M. (1955). Fukuoka Actu M c d . 46, 998. Koch, M. L. (1913). J. BioE. Chem. 14,267. Koch, W., and Koch, M. L. (1913). J. Biol. Chem. 15,423. LaVelle, A. ( 1951). J. Comp. Neurol. 94, 453. Levine, S., and Alpert, M. (1959). A.M.A. Arch. Gen. Psychiat. 1, 403. Libet, B., Fazekas, J. F., and Himwich, H. E. { 1941). Am. J. Physiol. 132,232. MacArthur, C. G., and Doisy, E. A. (1918-1919). J. Comp. Neurol. 30,445. McIlwain, H. (1955). In “Biochemistry and the Central Nervous System” p. 179. Little, Brown, Boston, Massachusetts. Malcolm, J. L. ( 1954). Electroencephalog. and Clin. Neurophysiol. 7, 143. Mandel, P., and Bieth, R. (1952). Compt. rend. SOC. biol. 235, 485. Manukian, K. G. (1955). Doklady Akad. Nauk S.S.S.R. 101, 1085. (Translation obtained privately.) Marinesco, G., Sager, O., and Kreindler, A. ( 1936). Bull. ueud. mSd. (Paris) 115, 873. Marty, R., Contamin, F., and SLherrer, J. (1958). Electroencephalog. and Clin. Neurophysiol. 10, 761. May, R. M. (1948). Reu. can. b i d . 7, 642. Millichap, J. G. (1957). PTOC.SOC. Exptl. Biol. Ned. 96, 125. Millichap, J. G., Baker, M., ‘and Hernandez, P. (1958). Proc. SOC.Exptl. Biol. Med. 99, 6. Nachmansohn, D. (1939). Bull. SOC. chim. biol. 21, 761. Naidoo, D., and Pratt, 0. E. (1954). Enxymologia 16, 298. Nemeth, A. M., and Dickerman, H. (1960). J. Biol. Chem. 235, 1761. Pentzik, A. S. (1937). Bull. biol. mSd. ex$. U.R.S.S. 4, 112. Peters, V. B., and Flexner, L. B. (1950). Am. J. Anat. 86, 133. Petersen, J. C., and Himwich, W. A. (1959). T h e Physiologist 2, 93. Potter, V. R., Schneider, W. C., and Liebl, G. J. (1945). Cancer Research 5,21. Purpura, D. P., and Carmichael, hl. MI. (1960). Science 131, 410. Ramon y Cajal, S. (1960). I n “Studies on Vertebrate Neurogenesis” pp. 327, 332. Charles C Thomas, Springfield, Illinok. Riese, W. ( 1944). Virginia Med. Monthly 71, 134. Roberts, E., Harman, P. J., and Frankel, S. (1951). Proc. Soc. Exptl. B i d . Med. 78, 799. Roberts, R. B., Flexner, J. B., and Flcwcr, L. B. (1959). J. Neurochem. 4, 78

158

WILLIAMINA A. HIMWICH

Rose, J. E., Adriin, H., and Santibjfiez, G. (1957). Acta neurol. latinoam 3, 133. Rozdilsky, B., and Olszewski, J. (1960). Neurology 10, 631. Rozdilsky, B., and Olszewski, J. ( 1961). J. Neuropathol. Exptl. NeuroE. 20, 193. SchadB, J. P. (1959a). J. Neurophysiol. 22,245. SchadB, J. P. (1959b). Koninkl. Ned. Akad. Wetenschap. Proc. Ser. C 62,445. Scheibel, A. B. (1962). In “Recent Advances in Biological Psychiatry” (J. Wortis, ed.), Vol. 4.Grune and Stratton, New York. In press. Scherrer, J., and Oeconomos, D. (1954). Etudes Nh-Natules. 3,199. Smith, J. R. (1938).J. Genet. Psychol. 53,431. Stem, L., and Peyrot, R. (1927). Compt. rend. SOC. biol. 96, 1124. Sugita, N. J. (1918).J. Comp. Neurol. 29, 248. Tilney, F. ( 1933-1934). Bull. Neurol. Inst. N . Y. 3,250. Tilney, F., and Kubie, L. S. (1931). Bull. Neurol. Inst. N . Y. 1,231. Tilney, F., and Rosett, J. (1931). Bull. Neurol. Inst. N . Y. 1, 28. Tower, D. B., and Elliott, K. A. C. (1952). Am. J. Physiol. 168, 747. Uzman, L., and Rumley, M. (1958). J. Neurochem. 3,170. Waelsch, H., Stoyanoff, V. A., and Sperry, W. M. (1991). 1. Biol. C h . 140, 885. Wender, M., and Hierowski, M. (1960). J. Neurochem. 5, 105.

SUBSTANCE P: A POLYPEPTIDE OF POSSIBLE PHYS IOLOGlCAL SIGNIFICANCE. ESPECIALLY WITHIN THE NERVOUS SYSTEM By F . Lembeck* and G. Zetler Deportment of Pharmacology. University of Grar. Austria. and Department of Pharmacology. University of Kiel. Germany

I . Introduction . . . . I1. Chemical Characteristics of A . Extraction . . . B . Estimation . . . C . Purification . . .

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

160

Substance P .

. . . D . Enzymatic Destruction .

. . . .

. . . .

. . . .

. . . . . . . 161 . . . . . . . . 161 . . . . . . . . 162 . .

. .

. .

. .

. .

. .

. .

. 163 . 165 E . Chemical Properties . . . . . . . . . . . . 167 F. Differentiation from Other Smooth-Muscle-Stimulating Substances. Especially Polypeptides . . . . . . . . 168 . . . 111. Distribution of Substance P in the Organism . . 170 A . Gastrointestinal Tract . . . . . . . . . . . . 170 B . Central Nervous System . . . . . . . . . . . 170 C . Peripheral Nervous System . . . . . . . . . . . 181 IV . Relationship between Organ Function and Tissue Concentration of Substance P . . . . . . . . . . . . . . 183 A . Intestine . . . . . . . . . . . . . . . 183 B . Nervous System . . . . . . . . . . . . . . 184 C . Degenerating Nerve . . . . . . . . . . . . . 185 D . Substance P and the Postdated “Transmitter Substance of . . . . . . . . . . . . . 186 Sensory Nerves” V . Pharmacological Actions of Substance P . . . . . . . . 191 A . Intestine . . . . . . . . . . . . . 191 B . Circulation . . . . . . . . . . . . . 194 C . Nervous System . . . . . . . . . . . 195 D. Tachyphylaxis . . . . . . . . . . . 203 VI . Pharmacological Interactions with Drugs . . . . . . 205 A . Intestine . . . . . . . . . . . . . . . . 205 B . Central Nervous System . . . . . . . . . . . 207 VII . Conclusions . . . . . . . . . . . . . . . . 209 References . . . . . . . . . . . . . . . . 210 * Present address: Department of Pharmacology. University of Tiibingen. Germany. 159

160

F, LEnlBECK AND G . ZLE'IZEH

I. Introduction

Substance P was detected 30 years ago by von Euler and Gaddum (1931), but more than 20 years elapsed before experimental results began to accumulate to an appreciable extent. By now enough work has been reported to justify the publication of this review. Substance P, being a polypeptide of possible neurobiological importance, deserves general interest, inasmuch as several polypeptides have already turned out to be active in amazingly low concentrations. It may be mentioned that the group of biologically active polypeptides comprises such potent substances as insulin, glucagon, oxytocin, vasopressin, angiotensin, ACTH, melanocyte-stimulating hormone, and bradykinin; the chemical constitution of these substances has already been established. Substance P (SP), although certainly a polypeptide, has only recently been obtained in pure form, and therefore it must be kept in mind that most of the results discussed below were obtained with very impure preparations. Nevertheless, they reveal some characteristics of activity, especially concerning the actions on the central nervous system. The relations of SP to neurophysiological and neuropharmacological problems will be described and discussed in detail, and in this respect the authors will present speculations and hypotheses, admitting the risk of being refuted, perhaps in the near future. Euler and Gaddum (1931) detected SP while working on the acetylcholine content of tissue extracts. The contraction of the isolated rabbit jejunum elicited by their extracts developed slower than that after acetylcholine and could not be prevented by pretreatment with atropine. The latter was also true for the fall of blood pressure after intravenous injection in rabbits. These observations permitted the conclusion that the extracts contained an active principle which was different from acetylcholine. One cannot forbear finding it admirable as well as satisfying that very simple pharmacological experiments in the hand of experienced and attentive observers can lead to the detection of a new substance, as happened with substance P. It may be appropriate here to explain the meaning of this name by describing its origin. Euler and Gaddum had several bottles containing substance P powder, labeled PI, P?, etc., meaning powder No. 1, powder No. 2, etc. In the course of time, the active

SUBSTANCE I’

161

substance in these powders acquired the private nickname of c‘substance P,” and this was first used publicly by Gaddum and Schild (1934). This name is free of suggested meaning, and Gaddum (1960) thinks that it is too late to change it now. I I . Chemical Characteristics of Substance P

A. EXTRACTION Von Euler and Gaddum (1931) detected SP in extracts from brain and intestine by mincing the tissue and suspending it in alcohol so that the final alcohol concentration was 72430%.After adding 1 ml of 1 N sulfuric acid to each 100 ml of alcohol this mixture was stirred for 1 hour at room temperature and filtered, and the alcohol in the clear filtrate was evaporated under reduced pressure. Fats were removed with ether, and the aqueous extract concentrated in vacuo, neutralized with sodium carbonate, and dried with anhydrous sodium sulfate, From this material the active principle was extracted with absolute alcohol; the resulting solution was evaporated and the residue was dissolved in water for biological assay. A simpler extraction method iised by several authors is to SUSpend the minced tissue in at least 2 volumes of distilled water acidified with sulfuric or hydrochloric acid to pH 3-4, boil the mixture for a few minutes, adjust the pH to 5.5 to precipitate proteins, and filter it. The extraction may be more complete if the tissue is extracted twice in this way. Amin et al. (1954) described a method for the separate extraction of SP and 5-hydroxytryptamine from the same piece of tissue. 5-Hydroxytryptamine is extracted from the minced tissue by addition of 20 volumes of acetone followed by 95%acetone; SP is then extracted from the finely ground dry precipitate by boiling with acid. Leach (1959) found that more SP can be extracted from acetone-dried nervous tissue if the dry powder is first treated with n-butanol. She thinks that SP is partly bound to insoluble material, possibly a lipoprotein. There is another indication that SP may normally exist in a free and a bound form. Zetler and Ohnesorge (1957) suspended ground mouse brains (about 0.35 gm without cerebellum) in 3 ml of distilled water, added 9 ml of ethanol (96%)and spun down

162

F. LEMBECK AND G. ZETLER

the precipitated proteins and tissue particles. The supernatant was evaporated on a boiling water bath and dissolved in Tyrode’s sdution for biological testing. The precipitated material was extracted three times more in the same way, thus delivering a total of 4 extracts with 56, 14, 5, and 2 units per gram fresh tissue, respectively. However, by then boiling the same remaining precipitate twice with diluted sulfuric acid at pH 3, still more activity could be extracted -14 and 1.8 units/gm in the first and the second extract, respectively, From these observations it was concluded that two fractions of SP exist in brain tissue, one of them being more easily extractable than the other. Finally, the findings of Euler and Lishajko (1961) and Zetler (1958) point to the possibility that SP is stored in central and peripheral nervous system in different ways (Section 111, C ) .

B. ESTIMATION The SP activity of tissue extracts and purified fractions can be tested using the fall of blood pressure in rabbits or the contraction of the isolated rabbit jejunum, rat uterus in estrus, or guinea pig ileum, the last being the most useful preparation. When testing impure tissue extracts, precaution has to be taken against histamine, acetylcholine, and 5-hydroxytryptamine which may be present as impurities. Sufficient amounts of an antihistaminic compound ( atropine ( l P ) ,and tryptamine ( 2 1e5) will keep the isolated guinea pig ileum insensitive to these three contaminants. Lysergic acid diethylamide or dihydroergotamine will abolish the effect of 5-hydroxytryptamine on the isolated rat uterus, which hardly responds even to large amounts of histamine (for details see Feldberg and Toh, 1953; Zetler and Schlosser, 1955; Amin et ul., 1954). According to Laszlo (1960) brain extracts with SP activity contain adenosine-.!?-phosphate (AMP) as a contaminant which interferes with SP by its inhibiting action on the isolated rabbit jejunum. Laszlo recommended enzymatic deamination by AMP-deaminase as the best way to remove AMP from the extracts without reducing their SP activity. Even after exclusion of the biogenic amines and AMP, extracts could still contain other smooth-muscle-stimulating polypeptides such as bradykinin, kallidin, or other “-kinins,” thus influencing the result. The chemical relation to SP would allow them to be carried even through several steps of purification, Posterior pituitary hormones hardly interfere with activity on the guinea pig ileum, because they are confined to hypothalamus tissue only and

x

SUBSTANCE P

163

do not stimulate the guinea pig ileum nor antagonize the effects of SP (Lembeck and Petschke, 1961). The most frequently used assay method employs guinea pig ileum in Tyrode’s solution at 32-37°C with addition of the above mentioned antagonists of acetylcholine, histamine, and serotonin. A small bath (2-5 ml) and an isotonic lever are used. Thirty seconds is usually time enough for contraction, and after a twofold wash the next dose should be added after a time interval of at least 3-4 minutes. The sensitivity vanes to some extent; 2 units/ml should give a good response. The unknown solution is matched against a standard solution by alternating addition of each, finding the “least larger” and the “least smaller” contraction of one of them between equal doses of the other. The superfusion technique of Gaddum (1953a) can be of great value when assaying very small amounts; however, since each dose has to be given with the same volume of fluid, it is more troublesome than the usual method where the amount of fluid added to the bath can be varied but should not exceed more than 10%of the bath volume. Recently, Gaddum (1961) designed a micromethod for determination of SP using a piece of goldfish gut suspended in 0.05 ml of 1:2 diluted Locke’s solution. The contraction of the gut is only 0.3 mm and cannot be recorded without an amplifying device. This preparation is sensitive to 0.01-0.001 units of SP. Also, the isolated fowl rectal caecum can be used for the assay of SP (Cleugh et ul., 1961 ) . This organ has the advantage of being very insensitive to other polypeptides, such as bradykinin. The error of a good assay using guinea pig ileum is about lo%, especially if designed as four-point-assay ( Gaddum and Lembeck, 1949). Most workers usually compare their extracts or SP samples with a standard preparation made by U. S. von Euler, the activity of which is expressed in units. One unit has the activity of 2-4 threshold doses on isolated rabbit jejunum in 30 ml of Tyrode’s solution (von Euler, 1942) or of 7-10 threshold doses on isolated guinea pig ileum in 3 ml of Tyrode’s solution (Pernow, 1953). About 8 0 3 0 mg of small intestine of the horse contains about one unit of SP.

C . PURIFICATION The .aqueous tissue extract is concentrated in z)ucuo and brought to pH 8 with NaOH. Large quantities of inactive material are then precipitated by adding 2 volumes of 96%ethanol, After evaporation

164

F. LEMBECK AND G . ZETLER

of the alcohol from the clear filtrate the active principle can be precipitated by 60-7M saturation with ammonium sulfate. In this way a dry powder will result with a biological activity of 5-15 units/mg (for details see von Euler, 1936b, 1942; Dalgliesh et al., 1953; Pernow, 1953; Amin et al., 1954; Zetler, 1956a). A further purification can be achieved by extraction of this powder with acetic acid and precipitation with ether (Lembeck, 1953) or by extraction with alcohol and precipitation with acetone (von Euler, 1936a). Ion exchange chromatography yielded a preparation with 30-70 units/mg (Matussek, 1959). Pernow (1953) purified SP on two aluminum oxide columns and then on a cellulose column which yielded a final product with 20003000 units/mg. The most advanced purification of SP has been described by Franz et al. (1961), who used ion exchange columns (Amberlite I R 4 5 and IRC-50, CMC-cellulose) and an aluminum oxide column. They achieved a probably pure preparation with a biological activity of 30,000-35,000 units/mg (see also Zuber and Jacques, 1962). Recently, Zetler (1961) analyzed a crude SP preparation made from cattle brain according to von Euler (1942) and achieved the following results: If one suspends the finely ground dry SP powder in carbon tetrachloride (CC1,; method of Behrens and Seydl, 1951), the inorganic salts, e.g., ammonium sulfate, sink to the bottom of the tube, whereas proteins and polypeptides float on the surface of CC1,. SP treated in this way yielded during chromatography on an A1,0, column three biologically active, i.e., gut-contracting, principles. Fraction “Fa” was not adsorbed by A1,0, and, therefore, came out from the column during elution with the starting solvent (7041: methanol). Fraction “ F b appeared during passage of distilled water, and fraction “Fc” was eluted by 0.1 N NaOH. The biological activity of SP powder which had not been in contact with CC1, was mainly due to fraction Fb, i.e. SP itself; however, Fa and Fc undoubtedly were also present, although in considerably smaller amounts than in the preparation treated with CC1,. By means of paper chromatography using n-butanol: acetic acid:water (40: 10:5 0 ) as the solvent system, the following Rf values were found: Fa, 0.22, Fb, 0.34; and Fc, 0.64.During paper electrophoresis in acetate buffer at pH 4.95, Fa and Fb migrated to the cathode and Fc to the anode. The biological activity of Fa and Fb disappeared during incubation with chymotrypsin, trypsin, and pepsin. Fc was destroyed by

SUBSlANCE P

165

chymotrypsin, pepsin, and papain, but not by trypsin. Fa and Fb elicited a fast contraction of isolated guinea pig ileum, but Fc, on the other hand, caused the ileum to contract very slowly, i.e., in a bradykinin-like manner. Despite its resistance to trypsin and its slow contracting activity, Fc is not bradykinin, for it migrated to the anode, induced contraction of the isolated rat duodenum, and was almost without effect on rabbit’s blood pressure (whereas bradykinin migrates to the cathode, caiises relaxation of the rat duodenum, and is strongly hypotensive in rabbits). Furthermore, Fc has R f values different from those of bradykinin. Thus it was concluded that Euler’s SP preparation made from bovine brain contains, in addition to SP, two hitherto unknown biologically active polypeptides (Fa and Fc). The amount of these two new polypeptides is considerably augmented by contact with CC14,which exerts a strong denaturating influence on protein material. Therefore, these two new polypeptides may perhaps be considered to be artifacts arising from unavoidable denaturating processes during the preparation of crude SP. Nevertheless, the possibility cannot yet he excluded that these two polypeptides are normal constituents of brain tissue similar to SP. Gaddum (1961), too, has observed that partially purified SP preparations can contain two other biologically active components, one of them being especially active on the isolated rat uterus and the other on the hen rectal caecum. These two principles migrate during paper electrophoresis at pH 3.4 to the cathode and can he separated from SP within 2 hours. At the present time, it is not known whether these components are polypeptides or whether they are identical with the peptides Fa and Fc described by Zetler (1961).

D. ENZYMATIC DESTRUCTION The biological activity of SP preparations is destroyed by chymotrypsin and trypsin. This has been demonstrated for the plainmuscle-stimulating and hypotensive activity ( von Euler, 1936a), and for the activity on the central nervous system (Euler and Pernow, 1956; Lechner and Lembeck, 1958; Zetler, 1956a, 1959). The SP-destroying activity of chymotrypsin is about 200 times greater than that of trypsin (Pernow, 1955a). The biological activity of highly purified SP was completely destroyed by chymo-

166

F. LEMBECK AND G . ZETLER

trypsin but only partially by trypsin, and it was resistant to carboxypeptidase (Franz et al., 1961). SP is also inactivated by pepsin, cathepsin, papain, and bacterial proteolytic enzymes (Eber and Lembeck, 1956). Recently, Matussek ( 1959) found that diamine oxidase inactivates SP. Furthermore, extracts of intestinal muscle, and brain (Gullbring, 1943), urinary bladder, ureter, and uterus (Pemow, 1955a) destroy SP; duodenum and jejunum show the greatest activity. According to Eber and Lembeck ( 1956), extracts of certain tissues inactivate SP; kidney and spleen are most potent in this respect; liver, pancreas, ileum, and lung are next; then placenta and brain, followed by heart and muscle; and finally, blood. These authors also showed that there is in different organs, especially brain parts, no relationship between the SP content of the tissue in question and its enzymatic activity and, moreover, that the SP-destroying principle is a catheptic protease and not a cholinesterase, amine oxidase, or glycolytic enzyme. The SP-inactivating enzyme from hog kidney was purified and chromatographically separated from diamine oxidase (Arvidsson et al., 1956). Blood serum of pregnant women destroys SP, in contrast to normal serum (Stem et al., 1961). The enzyme responsible for this effect is inhibited neither by lysergic acid diethylamide nor by the specific inhibitors of oxytocinase, vasopressinase, and diamine oxidase. Despite the existence of SP-inactivating enzymes in many organs, the substance is very resistant to autolysis (Gaddum and Schild, 1934). There was no difference in SP concentration between the 24- or 48hour-old bovine brains (Zetler and Schlosser, 1955). This postmortem stability of brain SP was also observed by Eber and Lembeck (1956), who, on the other hand, pointed out the inactivation of SP in tissue extracts contaminated with bacteria. Krivoy ( 1957) found that lysergic acid diethylamide (LSD) increases the sensitivity of the isolated guinea pig ileum to SP, and also that the enzymatic destruction of the polypeptide by brain extract but not by chymotrypsin was diminished by LSD (diluted W8).Eserine, morphine, mescaline, chlorpromazine, ergometrine, strychnine, and 2-bromo-LSD did not show this effect. 2-BromoLSD antagonized the protective action of LSD. Krivoy supposed that some of the well-known central effects of LSD may be related to its interaction with the metabolism of SP within the brain. This possibility has been refuted by Smith and Walaszek (1960)) since

SUBSTANCE P

167

LSD was found to potentiate also the gut-stimulating activity of other polypeptides (bradykinin and substance A ) which are thus far not thought to be of physiological significance for brain function. E. CHEMICAL PROPERTIES The papers of von Euler and Gaddum (1931), Gaddum and Schild (1934), and von Euler (1936a, b, 1942) revealed the following properties of SP: It is easily soluble in water, acetic acid, and in 1&2O% water containing organic solvents (methanol, ethanol, butanol, acetone). It is insoluble in ether and chloroform. Crude SP can be precipitated from alcoholic solutions with acetone or picric acid, and from watery solutions with mercuric chloride, phosphotungstic acid, or Reinecke acid, or by 70%saturation with ammonium sulfate. SP tolerates boiling for a t lea5t 20 minutes at pH 1-7 but is rapidly inactivated in alkaline environment. SP is not destroyed by deamination with HNO, (Gaddum and Schild, 1934; Vogt, 1955; Zetler and Schlosser, 19.55), a finding which has recently been refuted by Matussek ( 19.59). SP dialyzes quickly through parchment, Cellophane, and collodium membranes, and behaves like an ampholyte when subjected to Tiselius cataphoresis, migrating to the cathode at a pH < 6.4 and to the anode at pH 2 7.1 (von Euler, 1942). During paper electrophoresis, it travels to the cathode at pH < 10.5 (Pernow, 1955b). The distribution coefficient in n-butano1:acetic acid:water (40:10:50) is 0.48 (Pernow, 1955b), in the system n-butanol:0.2N NaOH the activity goes into the organic layer (Vogt, 1955). There was no physicochemical or biological difference between SP preparations made from brain and from intestine (Eliasson et al., 1956; Zetler, 19S6a). During paper chromatography with n-butanol-acetic acid: water (40:10:50), crude SP (5-6 units/mg; Zetler, 1961) as well as purified material (25oO-3500 unitdmg; Pernow, 1953) has the R f value 0.35-0.37. The R f value of the purest preparation thus far (30,00035,000 units/mg) was 0.7 when n-butano1:acetic acid:water (40:20:60) was used as solvent system, and this spot stained well with bromophenol blue but only very weakly with ninhydrin (Franz et d.,1961). These authors have also described the following properties of their practically pure material: During high-voltage electrophoresis on paper it migrates at pH 1.9 as fast as glutamic acid ( E 1.9 = 1.0 Glu) and at pH 5.8 slower than histidine ( E 5.8 = 0.8

168

F. LEMBECK AND G . ZETLER

His ) toward the cathode. The total hydrolysis releases arginine, proline, and leucine/isoleucine. After treatment with chymotrypsin, the substance gives two spots, one of them being ninhydrin-positive and bromophenol blue-negative and the other ninhydrin-negative and bromophenol blue-positive. There is no release of arginine. Trypsin does not change the position or stainability of the original bromophenol blue-positive spot. The authors assume that SP is a linear peptide which, in contrast to the similarly linear nonapeptide bradykinin, does not have the terminal group Phe-Arg-OH. From the above-mentioned findings (Section 11, C-E) it can be concluded that SP is a small polypeptide. This molecule probably has arginine amide or lysine amide groups (inactivation by trypsin ) , no S-S bonds (resistance against thioglycolic acid: Gaddum, 1955a), no free amino groups (resistance against HNO, and formaldehyde: Zetler and Schlosser, 1955; pink spot with ninhydrin), and no free carboxylic acid (migration from alkaline medium into butanol). Therefore, SP seems not to be a split product of a larger protein molecule, After strong acid hydrolysis of substance P preparations, histamine appears in the hydrolyzate (Swingle et al., 1956, Matussek, 1959). Therefore, Matussek ( 1959) thinks that histamine is a part of the SP molecule. F. DIFFERENTIATION FROM OTHER SMOOTH-MUSCLE-STIMULATING SUBSTANCES, ESPECIALLY POLYPEPTIDES

SP shares its smooth-muscle-stimulating activity with many other compounds normally present in tissue extracts. It is, nevertheless, very easy to differentiate SP from histamine, acetylcholine, and 5-hydroxytryptamine by pretreating the testing preparation, e.g., the isolated guinea pig ileum, with an antihistamine, atropine, and one of the antagonists of 5-hydroxytryptamine (LSD, BOL-148, dihydroergotamine, dibenamine) . Under these conditions, the remaining activity must be attributed to a polypeptide-perhaps SPif it disappears during incubation of the extract with a protease like trypsin or chymotrypsin. Resistance against chymotrypsin would mean that the active principle is probably a lipid-soluble acid such as acetalphosphatidic acid, an unsaturated fatty acid, or a similar compound (reviewed by Vogt, 1958). Unfortunately, there is still no specific antagonist of SP as, e.g., atropine against acetylcholine (Section VI, A ) . However, it is possible to desensitize the isolated

SUBSTANCE P

169

guinea pig ileum specifically to SP by suspending the gut in Tyrode’s solution with a high concentration of SP ( Gaddum, 1953b). NOWthe gut still responds normally to histamine, acetylcholine, S-hydroxytryptamine, and lipid-soluble acids, but is completely insensitive to formerly active low doses of SP. The differentiation of SP from some other biologically active polypeptides appearing in the body offers greater difficulties. There are three groups of polypeptides to be compared with SP: (1) vasopressin and oxytocin, ( 2 ) angiotensin (hypertensin, angiotonin: Braun-Menkndez and Page, 1958; Skeggs, 1960), and ( 3 ) the socalled kinins of which only bradykinin (Rocha e Silva, 1960) may be considered here as representative of the whole group since its general properties are fairly well known (Konzett and Stiirmer, 1960); moreover, it has recently been synthesized (Boissonnas et d., 1960). Perhaps the simpliest way to discriminate SP from the polypeptides just mentioned is to use the isolated rat duodenum (Horton, 1959). Here, only SP elicits a contraction, whereas bradykinin, oxytocin, and vasopressin cause relaxation, angiotensin being ineffective. The blood pressure reacts to vasopressin and angiotensin with a rise, to oxytocin, SP, and bradykinin with a fall. Even if two polypeptides exert qualitatively the same actions they can be quantitatively distinguished from each other by means of parallel assays on different tissues which show unequal sensitivity to one of the two principles in question. Such experiments deliver an “index of discrimination” ( Gaddum, 19551)) which clearly shows that, e.g., bradykinin and SP are different. Mostly, it can be recognized prima facie that a tissue extract contains bradykinin or another kinin instead of SP since contraction and relaxation of the isolated guinea pig ileum develop after the former much slower than after the latter (Pernow and Rocha e Silva, 195.5). All polypeptides discussed here are easily destroyed by chymotrypsin, whereas trypsin leaves only bradykinin intact ( Werle et al., 1950; Rocha e Silva, 1951). Therefore, incubation experiments with trypsin are most helpful in differentiating between SP and bradykinin-like substances. Treatment with thiosorbitol abolishes the activity of vasopressin and oxytocin but not that of SP (Gaddum, 1955a). The mode of preparation and the biological actions of the inhibitory factor I (Florey, 1954) allow to dissociate it from SP.

170

F. LEMBECK AND C. ZETLER

In contrast to the latter, solutions of factor I lose their biological activity very fast. Ill. Distribution of Substance P in the Organism

A. GASTROINTESTINAL TRACT Extracts made from the intestines of mammals (von Euler and Gaddum, 1931), fishes (von Euler and Ostlund, 1956a; Dahlstedt et al., 1959), and frogs (Lembeck and Petschke, 1961) contain SP. The distribution in the wall of the mammalian alimentary tract has been studied by Bjurstedt et al. (1940), Douglas et al. (1951), and Pernow (1951). Feldberg and Toh (1953) pointed out that these authors did not take precautions to avoid errors owing to the presence of 5-hydroxytryptamine in the extracts, and that therefore several very high values for SP were wrong. However, according to Gaddum ( 1960) the 5-hydroxytryptamine content of the intestines is not high enough to interfere generally with the SP values when extracts are being tested on the isolated guinea pig ileum. The concentration of SP is low in the esophagus and stomach, is high in the duodenum and jejunum, and decreases progressively towards the caecum. The rectum has about half the maximum concentration. The amount which can be extracted from small intestines differs greatly among several animal species, being highest in the monkey ( approximately 50 units/gm ) and lower-in decreasing order-in the dog, horse, bovine, rat, guinea pig, sheep, cat, and pig ( 7 units/@). All layers of the intestines contain SP, but the highest values were found in the muscularis mucosae. This finding leads to the speculation that SP is connected with Meissner’s plexus and influences the rhythmic motility of the gut (von Euler, 1936a; Pernow, 1951). Ehrenpreis and Pernow (1952) confirmed this view by demonstrating that in Hirschsprung’s disease the aganglionic and inactive part of the rectosigmoid contains significantly less SP than control pieces, whereas the proximal, hyperactive part shows abnormally high values. B. CENTRAL NERVOUS SYSTEM

The brains of all species of vertebrates so far investigated contain SP. It can be seen from Table I that the concentration of this polypeptide decreases with increasing differentiation of the central

171

SUBSTANCE P

TABLE I SUBSTANCE P CONCENTRATION I N THE BRAINOF DIFFERENT SPECIES Subst,ance P concentration (units/gm)

Species

No.0

Whole brain

Fore- Brain brain stem Referencesc ~

Man Dog Cat Rabbit Guinea pig Rat Rat Mouse Duck Chicken Chicken Sea gull Owl Pigeon Lark Frog (Rana esculenta) (Rana esculenta) Toad (Bujo vulgaris) (Bombinator puch ypus) Lizard Turtle Ringed snake (English: “grass snake”) (Natrzx = Tropidonotus natrix) Carp Smooth hound fish (Mustelus mustelus) Porgy (Box salpa) Dogfish (Squalus acanthias) European (gray)skate (Raja batis) Hagfish (Myzine ylutinosa) Cod

2 3 2 1 5 G 8 14 2 1 4 6 1 5 10 10 350 10

1-P 15 36b 75h 81h 8gb 20

8gh 9Zb

113b 5 15 12 103b 10 817* 250 6OOh 40 200 60 190 15 200 G 100

15 8 2 3 _

173 -

40 71b 25 30 _ 4-10 3 3 G

12

55

_ _ _ _

1 2 1 1 1 1 2 3 1 1 2 2 2 1 2 1 2 1 2 2 2 2

_

_

2

65

_ _

71

_ _

1 2 2

50

-

-

-

4 4 4

_

_

21 40 39 40

77 160 159 238

80 100

133 133

_ _

_ _

- _ - - _ 77

-

500

-

400

_ _ _

_ -

150

_

1000

_

667

_ -

5

Numhcr of brains investigated. Without, cerebellum. c References: 1. Grabner et al. (1959) 4. Dahlstedt et al. (1959) 2. Correale (1959) 5. von Euler and 6stlund (195610) 3. Zetler and Ohncsorge (1957)

a

172

F. LEMBECK AND G. ZETLER

nervous system. Table I also shows in detail the results of Grabner et al. (1959). These authors separated forebrain from brain stem and demonstrated clearly that the difference in SP concentration between these two parts is definitely less marked in primitive brains than in mammalian (including human) brains. This finding is explained by the fact that in fishes, amphibia, and birds the forebrain consists almost entirely of basal ganglia which are known to contain a high concentration of SP (see below). In mammals, however, the basal ganglia belong to the brain stem, whereas the forebrain contains increasing amounts of cortex and white matter which are not as rich in SP. Grabner et al. (1959) concluded that SP is chiefly concentrated in phylogenetically older parts of the brain. In the nervous system of invertebrates ( Cephalopoda, Echinoidea, Hdothurioidea), Lembeck and Petschke ( 1961) could not find appreciable amounts of SP, but their results are qualified by the fact that the amount of nervous tissue available for extraction was extremely small, Euler and Gaddum (1931) in their first paper on SP pointed out that this polypeptide is not equally distributed within the brain. In the last few years several papers have been devoted to this problem in the hope of obtaining a clue as to the physiological significance of SP. Kopera and Lazarini (1953) referred the biological activity of their brain extracts to a standard different from that used by other authors; also, they did not take precautions against 5-hydroxytryptamine, which is present in crude brain extracts. The latter difficulty applies also to the investigations of Pernow ( 1953). Nevertheless, the main results given in these two papers are in reasonable accordance with the findings of Lembeck ( 1953), Amin c?t al. (1954), Zetler and Schlosser ( 1955),and Paasonen and Vogt ( 1956). Figure 1 gives a survey of the distribution of SP in mammalian brain. Generally, the white matter contains less SP than the gray substance. Extracts made from white matter have in fact such low activity that only well-reacting guinea pig ilea respond to them with satisfactory contractions. In the spinal cord, however, there are white areas with definitely higher concentrations of SP (see below), Unfortunately, only six areas of brain cortex have been investigated so far (the hippocampus of dogs with 15 units/gm is not shown in Fig. 1). Nevertheless, it is evident that the distribution

SUBSTANCE P

173

of SP in the cortex is unequal. In man (Zetler and Schlosser, 1955), as well as in dog (Pernow, 1953), there was no difference between the precentral and postcentral gyrus, the value found for the olfactory cortex being of the same magnitude (Amin et d.,1954). It is, therefore, surprising that the visual cortex of the dog (Amin et al.,

FIG.1. Parasagittal section through mammalian brain showing the concentration of substance P in different parts. Condensed from Amin et at. (1954: dog), Zetler and Schlosser ( 1955: bovine, man) and Paasonen and Vogt (1956: dog, bovine, pig). Pr.g., precentral gyrus; Po.g., postcentral gyrus; V.C.,visual cortex; A.c.g., anterior cingulate gyrus; O.C., olfactory cortex; C.C., corpus callosum; Th., thalamus; O.b., olfactory bulb; A.h., anterior hypothalamus; P A . , posterior hypothalamus; M.b., mammillary body; C.g.m., central gray matter; P., pons; Py., pyramid; OZ., olive; N.c., nucleus cuneatus; N.g., nucleus gracilis; C.( c . ) , cerebellum (cortex); C . (zoh.), cerebellum (whole); P.c., posterior colliculi; A x . , anterior colliculi.

174

F. LEMBECK AND G . ZETLER

1954) contains hardly more SP than the white matter, e.g., corpus callosum. In human brain (Zetler and Schlosser, 1955),the cortical tissue of the anterior cingulate gyrus contains 85 A 7 units/gm, which is significantly ( p < 0.01, n = 10) more than the concentration found for the precentral (43 5.3 units/gm) and postcentral gyrus (39 2 4.9 unitdgm). In contrast to its abundance in nerve cells the cerebellar cortex contains small amounts of SP (13 units/gm: Paasonen and Vogt, 1956),and concerning the cerebellum as a whole most investigators (cat, bovine: Kopera and Lazarini, 1953; dog: Amin et al., 1954; bovine: Zetler and Schlosser, 1955) agree that there is not more substance P than in the white matter. The brain stem represents great differences in SP concentration even in closely neighboring areas (Fig. 1; see also Figs. 2 and 3, P

A.c.

c.c

FIG. 2. Substance P concentration (units per gram tissue) in several parts of human mesencephalon ( right half) ; means of 10 brains ( Zetler and Schlosser, 1955). Ax., anterior colliculus; C.U., cerebral aqueduct; C.g.m., central gray matter; R.n., red nucleus; M.b., mammillary body; S.n., substantia nigra; C.C., crus cerebri; O.t., optic tract; L.g.b., lateral geniculate body; M.g.b., medial geniculate body; P., pulvinar.

which give means of 10 human brains). Here, again, the white matter shows the lowest values, no matter whether it is intermingled with nerve cells (pons, olives) or not (crus cerebri). Especially high concentrations are present in the floor of the fourth ventricle, in the central gray matter, and in the anterior hypothalamus and mammillary body. The highest concentration of SP so far found

175

SUBSTANCE P

in human brain is present in the substantia nigra, which differs remarkably from the red nucleus (Fig. 2 ) . As to Figs. 2 and 3 it

E to.-

db.

'\".

FIG.3. Substance P concentration (units per gram tissue) in several parts of human diencephalon (right half of coronal section). Means of 10 brains according to Zetler and Schlosser ( 1955). F., fornix; la.,interthalamic adhesion; M.b., mammillary body; H . , hypothalamus; Cr.c., crus cerebri; G.p., globus pallidus; P., Putamen; I.c., internal capsule; Th., thalamus; C.n., caudate nucleus; C.C., corpus callosuni.

may be noted that the original results of Zetler and Schlosser (1955) give the standard error for each value and therefore permit the application of statistics. Thus it can be pointed out that the differences between globus pallidus and putamen, mammillary body and thalamus, thalamus and globus pallidus or putamen, central gray matter and red nucleus or anterior colliculi, are statistically significant. Another example of the neurohumoral differentiation of mesencephalic structures is given by Table 11, which demonstrates clearly that in man each geniculate body has less SP than the corresponding colliculus, and that the parts belonging to the auditory system contain definitely more SP than those which are elements of the visual system. In 10 brains of bovines, too, there was significantly ( p < 0.01) more SP in the posterior (122 7.3 units/ gm) than in the anterior colliculi ( 92 -t 4.4 units/gm) . The medulla oblongata as a whole contains 25 units/gm (dog: Amin et al., 1954), but it embodies some areas with considerably higher concentrations of SP. The mean value for the floor of the fourth ventricle is 52 units/gm [derived from the figures of Amin et al. (1954)for the dog, bovine, and pig], but the highest concen-

*

176

F. LEMBECK AND G. ZETLER

TABLE I1 MEAN SUBSTANCE P CONCENTRATION^ IN SOMEMESENCEPHALIC AREAS BELONGING TO T H E V I S U A L A N D AUDITORY S Y S T E M OF 10 HUMAN BRAINS* Visual ___----__A__-_-__-_

Anterior colliculi 55

T p

I 141

?C

5.2t-

Lateral geniculate body p

< 0.0027-

5

T

< 0.0027 k 10.4+--

--t

* 1.0

p p

< 0.0027 -437

< 0.0027

1

* 3.0

Posterior colliculi

Medial geniculate body ...----_-_-_____------/ Auditory

(1

b

Concentration in units/mg rt standard error. From Zetler and Schlosser (1955).

trations were found in structures located at its lower end (rhomboid fossa) : ala cinerea with 486 units/gm in bovines or 284 units/gm in man (Zetler and Schlosser, 1955) and 334 units/gm in bovines ( Paasonen and Vogt, 1956). Amin et al. ( 1954) found 290 units/gm in the area postrema of dogs, and this finding has been confirmed by Paasonen and Vogt (1956) for the dog (457 u/gm), bovine (143 u/gm), and pig (167 u/gm). The closely neighboring trigonum hypoglossi, however, shows relatively low values, namely 34 units/gm in bovines or 37 units/gm in man (Zetler and Schlosser, 1955; means of 10 brains each), and 70 units/gm in bovines (Paasonen and Vogt, 1956; l brain). As regards the area postrema, some controversy has arisen from the finding of Zetler and Schlosser (1955) that this region in bovines is almost completely devoid of SP. We think that the reason for the lack of the polypeptide in this case is not so much a mistake in dissecting but rather the fact that the brains were removed from the heads, not immediately after death as done by other authors, but after 24 hours’ storage in the cold room. It is possible that during this time SP disappeared in this area by enzymatic destruction and/or diffusion. However, these two factors have obviously only a negligible influence on the postmortem fate of SP present in normal brain tissue, and this would make it necessary to postulate special conditions for the area postrema. In fact, this structure has been shown to have an ex-

177

SUBS'I'ANCE 1'

tremely high permeability (Wislocki and Putnam, 1924), to be abundantly vascularized (Cammermeyer, 1944), and to behave toward pharmacological stimuli, such as morphine, in a way which differs greatly from that of other brain parts, e.g., hypothalamus (M. Vogt, 1954). The SP concentration in several parts of the spinal cord is shown in Fig. 4. Here again the gray matter shows definitely higher values

(0-25 J

,r Choline A cotyhse

& (x)

750-12 cool

FIG.4. Substance P concentration (iinits/gni; Amin et al., 1954) and choline acetylase activity ( p g ACh/gm/hr; Hebb and Silver, 1956) in the spinal cord of the dog. F.g.c., funiculi gracilis and cuneatus; D.T.,dorsal root; P.t., pyramidal tract; V.T.,ventral root.

than the white matter, e.g., the pyramidal tract. However, the funiculi gracilis and cuneatus are an obvious exception to this rule as they contain relatively much SP (Kopera and Lazarini, 1953; Amin et al., 1954) despite consisting of medullated nerve fibers, i.e., white matter. A similarly great difference exists between anterior and posterior roots, the latter showing much higher SP values than the former (Kopera and Lazarini, 1953; Lembeck, 1953; Pernow, 1953; Amin et al., 1954). The connecting link between these two findings of a high SP concentration in white matter is the fact that many of the nerve fibers entering the spinal cord via the posterial roots from the funiculi gracilis and cuneatus. Therefore it is perhaps not surprising that the nuclei cuneatus and gracilis to which these sensory nerves run are particularly rich in SP (dog: 110 units/gm; Amin et al., 1954). The posterior roots, on the other hand, contain neither acetylcholine ( Loewi and Hellauer, 1938) nor the acetylcholineforming enzyme choline acetylase (Feldberg and Mann, 1946; Hebb and Silver, 1956) and in this respect differ conspicuously from the anterior roots (see Fig. 4). These differences between the anterior

178

F. LEMBECK AND G . ZETLER

and posterior roots led to the hypothesis that SP may be the transmitter substance of the first sensory neuron (Lembeck, 1953). The inverse ratio of SP concentration to choline acetylase activity does not seem to be restricted to the anatomical structures just mentioned. Kopera and Lazarini (1953) compared SP values with the choline acetylase activity of different brain parts (Feldberg and Vogt, 1948) and reached the conclusion that high SP concentrations are typical for brain areas with low acetylcholine synthesizing activity. Zetler and Schlosser (1955) determined the choline acetylase activity of 18 different parts of 3 human brains by applying the method of Feldberg and Vogt (1948) and came to the same result. Hebb and Silver (1956), using a much more sensitive method, reached practically the same principal conclusions about characteristic differences in choline acetylase activity between neighboring brain areas. Their figures as well as those of Feldberg and Vogt (1948) and of Zetler and Schlosser (1955) show, e.g., that the thalamus synthesizes more acetylcholine than the hypothalamus, or the anterior colliculi more than the posterior colliculi. Comparing these relations with those for the SP content of the same parts of brain (see above) it can be seen that here, too, SP concentration is in inverse ratio to choline acetylase activity. This would mean that SP is located in neurons which are noncholinergic, according to the hypothesis of Feldberg and Vogt ( 1948). Of course, it could be that the relatively low choline acetylase activity of brain areas with high SP content is a reflection of interference by the polypeptide with the enzymatic process. In the in vitro experiments of Zetler and Schlosser (1955) SP did not exert any influence on the choline acetylase activity of acetone-dried brain powder. However, it may be worth repeating these experiments with purified SP and a refined choline-acetylase system. The unequal distribution of SP in the central nervous system is not confined to mammals, as shown by Dahlstedt et d.( lS59), who investigated different brain parts of dogfish (Squalus acanthias) and ray (Raja butis). In dogfish, telencephalon (50 u/gm) and medulla oblongata (45u/gm) contained about twice as much SP as midbrain with cerebellum (21 u/gm) and spinal cord (18 u/gm). In ray, telencephalon ( 5 u/gm) and midbrain with cerebellum (8.5 u / p ) showed definitely less activity than medulla oblongata (39 u/gm) . Furthermore, in this species the dorsal part of the spinal cord was

179

SUBSTANCE P

found to contain a higher concentration of SP (39 u/gm) than the ventral part ( 18 u/gm) . This finding deserves special attention as it closely suggests corresponding observations concerning mammalian spinal cord (see above). Table I11 gives information on the intracellular distribution of TABLE I I r INTRACELLULAR DISTRIBUTION OF SUBSTANCE P ACTIVITY IN BRAINTISSUE Substance P concentration (units/gm brain tissue)

Brain Cat Mouse

Microsomes Control Cell and (whole nuclei Mitochondria cytoplasms brain) 14 14

30 58

(10)" 12

54 88

References Lembeck and Holasek (1960) Zetler (1958)

a This value is not based on an estimation, since the concentration in the supernatant fluid was too small. It is the calculated difference, valid only under the assumption that there was no loss of siibst,arice P.

SP in brain tissue. Two thirds of the total activity can be spun down with the mitochondrial fraction which according to Lembeck and Holasek (1960) contains 0.7 units/mg protein, in contrast to the nuclear (0.38 u/mg) and cytoplasmatic fractions (0.46 u/mg) It may be concluded that in brain tissue SP is predominantly stored in the mitochondria. Gaddum ( 1961) prepared subfractions of guinea pig brain mitochondrial fraction by means of centrifugation in a sucrose gradient and found SP in the layer which also contains acetylcholine, histamine, and 5-hydroxytryptamine. The experiment shown in Fig. 5 (Zetler, 1958) demonstrates that SP can be easily extracted from the mitochondrial fraction by boiling at p H 3, and also that the same amount of biological activity is very slowly released if the corpuscles are suspended without pretreatment in the bath fluid. It can be assumed that SP is present in brain mitochondria in an active form and is released during the breakdown of the mitochondrial membrane in Tyrode's solution at 32OC. Is SP located in nerve cells or in glial cells? This crucial question is widely neglected as far as other neurohormones are concerned. The significance of the problem arises from the fact that in

180

F. LEMBECK AND G. ZETLER

I I I I I I I 1 1 1 1 1 1 1 1 111 I I I

I I I

n 1 1 I I I I I I I I I I I I I I 1 1 I I

I I 1 1 1 II

FIG.5. Isolated guinea pig ileum, bath volume 3 ml, temperature 32°C (atropine lo-‘, mepyramine lo-‘, tryptamine 2 x lo4). S, : 0.2 units substance P. S2 : 0.24 units substance P. M, : 0.1 ml mitochondrial fraction of mouse brain, S minutes boiled at pH 3 (HC1) and neutralized with NaOH; M2the same dose of mitochondria1 fraction, but neither acidified nor boiled. The doses of MI and M2correspond to about 3.5 nig mouse brain each (the cerebellum was removed before mincing). The time marker indicates 10 seconds.

brain there are about 10 times more glial cells than nerve cells (Haug, 1953; HydCn, 1958).Therefore, it could well be that one or another physiological finding applies to glial tissue rather than to true nerve tissue. Grabner and Lembeck (1960) approached this problem by determining the SP content of different brain neoplasms taken from patients during operation or from post-mortem dissections. The tumors were derived from neuroglia ( 5 astrocytomas, 4 multiform glioblastomas, 2 malignant spongioblastomas, and 1 cerebellar gliosis),from nervous tissue ( 1 ganglioneuroma, 3 medulloblastomas, and 1 retinoblastoma), and from the pia mater ( 3 meningiomas) . Nineteen out of these 20 neoplasms contained no SP or only negligible amounts; the extract of one irradiated multiform glioblastoma exhibited an activity of 13 unitdgm. The extracts of one glioblastoma and of an astrocytoma showed a “slow contracting” activity which is not typical for SP and was therefore attributed to another active principle. Grabner and Lembeck (1960) interpreted these observations by the assumption that if normal glial tissue were capable of synthesizing SP the glial neoplasms under observation could be expected to contain SP too. This assumption seems to be justified by the finding of cholinesterase and amine oxidase activity

SUBSTANCE P

181

even in immature glial tumors (Bulbring et al., 1953). Neoplasms of nervous tissue, however, did not contain SP, and this was attributed to the immature state of these tumorous tissues. Grabner and Lembeck (1960) therefore concluded per exclusionem that SP is located in nerve cells and not in glial cells. Their hypothesis is supported by the finding of Holton (1958) that after cutting a peripheral nerve the SP concentration decreases in the peripheral part and increases in the central stump (cf. Section IV, C ) . Further support arises from the fact that the retina, which contains very little glial tissue, is rich in SP. This last finding was made by DunCr et d. (1954): the cow retina contained 7 units of SP, i.e., 13 units/@ wet weight or 93 units/gm dry weight. The corresponding values for the dog retina were 20-28 units/gm wet weight or 130200 units/@ dry weight. Stern and Kocib-Mitrovib (1958) found for the cow retina 2 units/gm wet weight which is only one-sixth of the value given by Dun& et al. ( 1954).

C. PERIPHERAL NERVOUS SYSTEM

SP can be extracted from peripheral nerves too (Table I V ) , but there is generally less activity than in the central nervous system. Nevertheless, it is evident that SP is unequally distributed in peripheral nerves; this is analogous to the difference between anterior and posterior roots which has been mentioned above. With respect to the sympathetic nervous system, Table IV shows a contrast between truncus sympathicus cervicalis and plexus lienalis ( Pernow, 1953). [We do not agree with Pernow (1953) that this difference arises from the purely preganglionic nature of the former and purely postganglionic nature of the latter sympathetic nerves. Truncus symp. cervicalis contains both pre- and postganglionic fibers and plexus lienalis contains also parasympathetic fibers.] Motor nerves show the lowest activity (n. phrenicus, n. hypoglossus), but sensory nerves (n. opticus, n. acusticus, n. saplienus) do not contain substantially higher values, although this could perhaps have been expected from the findings with posterior roots. The high SP activity found in the spinal ganglia, however, may stem from the same neurons that form the dorsal roots. Von Euler and Lishajko (1961) examined the subcellular distribution of SP in the sciatic and brachial nerve of dogs. In this tissue, about 50% of the total biological activity was concentrated in the

182

F. LEMBECK AND G . ZETLER

TABLE I V SUBSTANCE P CONCENTRATION IN PERIPHERAL NERVES Units/gm tissue Nerve

Bovinea

N. opticus N. acusticus N. saphenus Ganglion spinale N. hypoglossus N. phrenicus N. ischiadicus N. vagus Truncus sympathicus cervicalis Truncus sympathicus thoracalis N. splanchnicus Plexus lienalis Plexus mesentericiis Ganglion stellare

7 (5-10) 4 (3-5) 12 (10-15) 40 (3047) 8 (7-10) 8 (5-12) 12 (10-15) 22 (18-25) 40 (25-50) 17 (12-20) 10 (8-15) 7 (5-10) 6 (4-9) 19 (15-22) 12 and 2OC 12 (10-18)

Ganglion nodosum Ganglion cervicale Nervi accelerantes Pooled sympathetic ganglia a

b

Dogb

9( 7-10) -

4.4 0 21 and 26"

From Pernow (1953). From Amin et al. (1954). From Paasonen and Vogt (1956).

microsomal fraction. This fraction was inactive when added to the bath fluid of the isolated gut, but showed typical SP activity after treatment with sulfuric acid. The authors concluded that in peripheral nerves, SP is stored in the microsomes in a conjugated, inactive form. This finding of von Euler and Lishajko (1961) contrasts sharply with that of Zetler (1958) with brain mitochondria (see Fig. 5). Thus, SP seems to be stored in central nervous system and peripheral nerves in different ways: in the former it is predominantly located in mitochondria and already active; in the latter, in microsomes as an inactive complex. The small fraction of SP present in mouse brain which cannot be extracted by mild methods (Zetler and Ohnesorge, 1957; Section 11, A of this paper) may also be stored in such a way as suggested by von Euler and Lishajko (1961).

183

SUBSTANCF: P

IV. Relationship between Organ Function and Tissue Concentration of Substance P

A. INTESTINE

The amount of SP in different segments and layers of the intestine of various species has been described in Section 111, A. The differences between species are much more pronounced than those between different segments. The results do not give any information concerning a relation between the propulsive motility, the kind of food intake or other factors, and the SP content of the gut. It should be mentioned, however, that there exists some correlation between the SP content of the gut and its sensitivity to it (Pernow, 1951, 1953). The SP content of bovine fetus was found to be very low until some time before birth, when normal amounts already were present (Fig. 6 ) (Lembeck and Petsclike, 1961).

1

4

6

0 10

;RzI 2832 4250

weight of brain in gm

., I&

c 0

FIG.6. Substance P content of forebrain A, brain stem and small of bovine fetus. C = calf, 0 = ox, X = whole brain; A , 0 means intestine that the amount was too small for estimation and lower than that indicated by the signs.

184

F. LEMBECK AM) G . ZETLER

The gall bladder with its abundance of ganglion cells contains no SP and does not respond to it (Pernow, 1953; Hultman, 1955). The ureter, containing ganglion cells only in its vesical portion (Hryntschak, 1925) contains SP in all parts and responds to SP much better than to acetylcholine (Pernow, 1953). The aganglionic segment of the colon in Hirschsprungs disease has a low content of SP (Ehrenpreis and Pernow, 1952), which suggests that SP is located in intestinal nervous structures. The submucosa contains high amounts of SP but not choline acetylase (Feldberg and Lin, 1950), and it is tempting to assume that the SP there is confined to the submucosal nervous fibers. A correlation between SP and some, but not all, nervous structures of the gut therefore seems likely to be possible. Von Euler (1936a) suggested that SP could be an essential factor for the spontaneous activity of the intestine; acetylcholine or histamine are not essential in this respect, since spontaneity of the intestine is hardly affected by atropine or antihistamine.

B. NERVOUS SYSTEM There are few instances in which we know the relation between the concentration of a substance normally occurring in the brain and the functional state of the brain; changes in excitability with respect to the glucose content could be given as an example. Certainly, changes in the content of acetylcholine during anesthesia (Richter and Crossland, 1949; Herken and Neubert, 1953) or of norepinephrine and serotonin after reserpine have been found, but these changes are to be attributed to interactions of drugs rather than to a changed functional state per se. Lembeck and Petschke (1961) estimated the SP content in the forebrain, brain stem, and intestine of bovine fetus at various ages (Fig. 6); it was found that a relatively high content of SP occurred very early in the brain stem whereas the forebrain did not contain detectable amounts before birth. Since the fetal movements during the later period of pregnancy already originate from the brain stem (Barcroft and Barron, 1939) and the cortex does not begin to function before birth, it seems possible that SP appears at about the same time as the earliest functioning. In this connection it is interesting to note that in the cerebral cortex of guinea pigs the acetylcholine esterase begins to rise rapidly in the middle (day 35) of the fetal development (Kavaler and Kimel, 1952). In rabbits the choline

SUBSTANCE P

185

esterase activity of the gray matter of motor cortex was found to be low during gestation and rose to maximal values some time after birth (Himwich and Aprison, 1955). Koci6-Mitrovib (1959) in her experiments used a part of the central nervous system which can be subjected relatively easily to changes in functional state, namely the retina, which is rich in SP (Dun& et al., 1954). In retinae of cow eyes illuminated for 1.5 to 2 hours, 2.5 units/gm were found, whereas the retinae of previously closed eyes contained 5 units/gm. Excised retinae kept for the same time in light or in the dark respectively at 37OC did not show differences in SP content. Unfortunately, the SP content found by Koci6-Mitrovi6 (1959) is lower than that found by Dun& et al. (1954). It would be interesting to know whether these findings have any bearing on the observations of De Robertis and Franchi (1956), who found changes in the size and distribution of the synaptic vesicles of the retina under similar conditions of light or dark. Koci6Mitrovid (1959) found that the cerebral concentration of SP in rabbits and rats which were kept in the dark for 7 days prior to decapitation was lower than that of the control group kept in light. Total brains were used for extraction in these experiments and it would be interesting to know whether the differences are more pronounced if the contents in distinct areas are compared. Walaszek et al. (1958) and Walaszek (1960) twice injected rabbits with 2 ml of serum of schizophrenic patients, with an interval of 4 days. They found in the hypothalamus of the rabbits treated with schizophrenic serum much more epinephrine and norepinephrine, some more histamine, equal serotonin, and less SP; the basic mechanism of these findings is difficult to explain. Nothing is known about the SP content of the brain under pathological conditions, except that all brain tumors so far investigated do not contain SP (Grabner and Lembeck, 1960). The recent observations of Ehringer and Hornykiewicz ( 1960), who found decreased dopamine in the neostriatum of patients who suffered from parkinsonism, emphasize the need for similar investigations with regard to substance P. C. DEGENERATING NERVE During nerve degeneration, acetylcholine, choline acetylase, and true cholinesterase decrease rapidly to low concentrations. Some

186

F. LEMBECK AND G. ZETLER

other substances, such as thiamine (von Muralt and Wyss, 1944; Umrath and Hellauer, 1949; Umrath, 1951a), neutral fat (Johnson et aZ., 1950), and pyridine nucleotides (Greengard et d.,1954), decrease also at about the same rate but not to the same low levels as the former group. The proximal stump of a severed nerve contains increasing amounts of thiamine and acetylcholine (Umrath and Hellauer, 1949), choline acetylase (Hebb and Waites, 1956), and true cholinesterase (Sawyer, 1946). Andrews and Holton ( 1958) in very carefully performed experiments found a decrease of SP in the distal stump of the severed sciatic and auricular nerves of rabbits to 10-3(4%of normal values 7 to 14 days after the dissection, after which a slow increase took place, possibly due to regeneration. The decrease of SP is quite similar to that of acetylcholine or choline acetylase. In the proximal end of the cut nerve SP concentration rose to about 600%of the controls at about the fifth day and thereupon declined. This accumulation is comparable to the change in choline acetylase activity in retrograde degeneration (Hebb and Waites, 1956). These results demonstrate that SP behaves like other substances associated with transmission of nerve impulses. Umrath ( 195313) found a similar disappearance of the “Erregungssubstanz sensibler Nerven” (see Section IV, D ) after degeneration of the sciatic nerve in guinea pigs and concomitantly a decrease of SP. Holton (1960) concluded: “. . . it seems that substance P behaves more like the substances associated with transmission from nerve endings than those which are not.”

D. SUBSTANCE P AND THE POSTULATED “TRANSMITTER SUBSTANCE OF SENSORY NERVES” Loewi and Hellauer (1938) found little or no acetylcholine in sensory nerves; this observation was confirmed by Macintosh ( 1941). Umrath and Hellauer (1951) showed that acetylcholine was increased in the central end of the previously cut optic nerve, which means that its acetylcholine belongs to cholinergic fibers arising from the anterior colliculus. The choline acetylase activity of sensory nerves, e.g., the posterior roots or the optic nerve, was also found to be very low (Feldberg and Vogt, 1948; Hebb, 1955). The assumption that the mode of transmission at central sensory nerve endings is the same as that in cholinergic or adrenergic nerves includes the concept of the existence of another transmitter sub-

SUBSTANCE P

187

stance in sensoiy fibers. Eccles (1957) discussed in detail the necessity of assuming at least one more transmitter substance besides acetylcholine within the central nervous system. Based on their studies on the localization of choline acetylase in the central nervous system, Feldberg and Vogt (1948) postulated that the central nervous pathways consist of alternating cholinergic and noncholinergic fibers. Hellauer and Umrath (1947, 1948) were the first to try a direct experimental approach to this question. They injected extracts of ventral and dorsal roots of the spinal cord intracutaneously into the dorsal side of the denervated rabbit ear. Around the point of injection a circular region of hyperemia arose, extracts from dorsal roots being more effective than those from ventral roots. Since addition of atropine to the extracts only reduced the effect of ventral roots, acetylcholine could be ruled out as a possible cause of the reddening by dorsal root extracts. Following a suggestion by Dale (1935) that the antidromic vasodilatation of the so-called “axon reflex” might be produced by release of the same substance that mediates the centripetal transmission of the impulse, they put forward the hypothesis that the vasodilatation due to the dorsal root extract is produced by the “transmitter substance of sensory nerves.” Since the attempts to prove the existence of this “Erregungssubstanz sensibler Nerven” either chemically or by means of specific biological tests have so far not been completed, we will do better to speak in terms of a sensory transmitter factor (“STF”). In most of their experiments Hellauer (1953), Hellauer and Umrath (1947, 1948), Umrath (1951b, 1953a, b, 1956), and Umrath and Hellauer (1948) used boiled aqueous extracts of nerves and brain and tested the activity on the denervated rabbit ear in the manner as described above, which allows some rough quantitative estimation. In search of a more reliable preparation for the test, Lembeck (1953) compared the effects of ventral and dorsal root extracts on the isolated guinea pig ileum. By this method a comparatively large amount of SP in dorsal roots was found. This observation opened the question of whether SP is identical with the STF. In the following, a speculative comparison of SP and the STF will be tried. SP produces reddening of the rabbit ear after intradermal injections of solutions of 0.75 to 2.0 unitdm1 (Umrath, 1953a, b; Hellauer, 1953). STF has not yet been purified and all

188

F. LEMBECK AND G. ZETLER

STF-containing extracts used contain SP too. Hellauer (1953) compared dorsal root extracts with SP but did not find profound differences, except that SP was more resistant to alkali. Umrath (1961) found that the effect of SP preparations on his rabbit-ear test was not abolished by incubation with trypsin, in contrast to its action on the guinea pig ileum. Hellauer (1953) mentioned that the equal smooth-muscle-stimulating effects of ventral and dorsal root extracts were different on various segments of the guinea pig gut under atropine and antihistamines. It is the present reviewers’ opinion that this difference may be explained by the presence of at least one other stimulating or inhibiting substance in the extracts. Recent findings of Laszlo (1960) indicate the possibility that the content of AMP in these spinal root extracts could be responsible for the different effects on various segments of gut; however, this has not yet been proven. Hellauer and Umrath (1947, 1948), and Umrath and Hellauer (1948), showed that STF activity was abolished by incubation in fresh saline extracts of nervous tissue. Fresh saline extracts of dorsal roots were about three times as active as extracts of ventral roots. They assumed a specific “STF-destroying enzyme.” Furthermore, Umrath (195313) found that incubation of these saline extracts for 16-18 hours at 33OC under sterile conditions increased their STFdestroying activity. The STF-destroying enzyme was found to be inhibited by picrotoxin and strychnine and some other substances ( brucine, metrazol, atropine, caffeine, santonine ) in rather high concentrations (Umrath, 1953a). He also described a close correlation between the convulsion-producing dose of various substances in frogs and the threshold concentration of these substances required to inhibit the STF-destroying enzyme. The STF-destroying enzyme has been found to be active in the pH range between 5 and 8 (Umrath, 1953a). Eber and Lembeck (1956) showed that bacterial growth in saline homogenates of nervous tissue used as “enzyme suspension” can lead to inactivation of SP by the proteolytic activity of bacteria. Even the ultrafiltrate of a culture of bacteria taken from an unsterile “incubated enzyme suspension” had a high SP-destroying activity. The STF-destroying enzyme has been found in cerebrospinal fluid and in the aqueous humor (Umrath and Hellauer, 1956). It is not known whether it occurs in different concentration in various parts of the brain. There is some SP-destroying capacity

SUBSTANCE P

189

in brain tissue, but it is difficult to say that it can be attributed to an enzyme specific for SP (Eber and Lembeck, 1956). Umrath (1951b) described a burning sensation after applying dorsal root extracts to the human cornea, and this effect could be abolished by alkali treatment or enzymatic destruction of STF in the extract, It must be mentioned that SP preparations in concentrations of 60-80 unitdm1 also elicited some sort of burning sensation on the eye (Hellauer, 1953; Umrath, 195313). Acetylcholine, which in higher concentrations also produces a burning sensation on the eye (Hellauer, 1950), could not be responsible for this effect, since its concentration in root extracts was not high enough, The present reviewers think that the specificity of this effect on the cornea by all means must not be overemphasized. Umrath (1953b) showed that in the degenerating sciatic nerve of guinea pigs there is a decrease of STF to 10%of the normal value within 10 days, a decrease of the STF-destroying enzyme beginning at day 6 and becoming almost complete after 15 days, and a fall of the SP content to about 60%; the last result has been confirmed and extended by Holton ( Section IV, C ). Umrath reached the following conclusions about the relation of SP to the STF: 1. He assumes from the rate of enzymatic destruction by his STF-destroying enzyme that SP of dorsal and ventral roots is different, the former being able to split off the STF, whereas the SP in ventral roots plays the role of a “carrier” of acetylcholine ( Umrath, 19%). 2. The above mentioned enzyme seems to destroy SP in nerves somewhat quicker than SP in intestine, and Umrath (1953b) concludes from this that they are two different substances. This is in disagreement with the detailed investigations of Eliasson et al. (19%), who could not find any differences in the physical or chemical properties of partially purified preparations of SP from brain and from intestine. Using his rabbit-ear test and a phototactic reaction test on bees, Umrath (1961) recently came to the conclusion that SP as a polypeptide would carry various transmitter substances as ‘‘prosthetic groups” which can be liberated by enzymatic processes or extraction procedures. Table V compares the properties and actions of SP and of the

190

F. LEMBECK AND G. ZETLER

COMPARISON OF

THE

TABLE V PROPERTIES AND ACTIONS OF SUBSTANCE P A N D STF Substance Pa

Properties and actions Higher concentration in dorsal than in ventral roots Presence in brain Presence in intestine Polypeptide characteristics Precipitation by ammonium sulfate Precipitation by ferrous hydroxide Dialysis through cellophane membrane Stable at 100" C in weak acid Destroyed by hot alkali Treatment with oxygen Treatment with hydrogen Solubility in ethanol Solubility in ether Adsorption on aluminum oxide Decrease after degeneration of nerves Vasodilatation on intradermal injection into the denervated rabbit ear Painful on instillation into the eye Inactivation by trypsin Inactivation by a specific enzyme Inhibition of the enzymatic inactivation By strychnine By other convulsive substances ~_______

a

~

STFe

+ (5) + (15) + (13) ? ?

+ (4) + (4) + (4) ++ (4) Destroyed (4) No effect (4)

+ (4) + (4) + (13)

0

0 (7, 13) ?

~~

Numbers in parentheses indicate references:

1. Eber and Lembeck (1956). 2. von Euler and Gaddum (1931), von Euler (1Y36a,b, 1942). 3. Gaddum and Schild (1934). 4. I-Iellauer (1953). 5. Hellauer and Umrath (1947, 1948). 6. Holton (1960). 7. Krivoy (1957).

8. Lembeck (1953), Amin et al. (1954). 9. Pernow (1953). 10. Pernow (1955). 11. Umrath (1951b). 12. Umrath (1953a). 13. Umrath (1953b). 14. Umrath (1961). 15. Umrath and Hellauer (1948).

STF. It must be mentioned that the distinction between two chemical and pharmacological closely related substances of known structure as, e.g., between epinephrine and norepinephrine, is quite easy by bioassay procedures. To find enough evidence for or against a difference between two substances of unknown structure can still

SUBSTANCE P

191

be relatively easy when the purification has reached a certain level and reliable assays are available, SP and bradykinin, although both stimulate the intestinal muscle and decrease the blood pressure, can be distinguished easily by a qualitative pharmacological analysis on several test preparations (Section 11, F). But the STF has not yet been purified at all; the tests available offer great difficulties for accurate quantitative estimation, and react to substance P and probably to other compounds too. Conclusions about relations between SP and the postulated STF would be premature. Conclusive assays as well as purified preparations of STF are needed in order to decide whether they are identical or different. Nevertheless, Hellauer and Umrath commendably have taken the first experimental step toward finding a substance which fits the concept of a sensory transmitter substance. V. Pharmacological Actions of Substance P

A. INTESTINE 1. Isolated Gut It was found that SP is effective in causing contraction of all preparations of isolated intestine, uterus, ureter, and urinary bladder, except the gall bladder. For assay (Section 11, B ) the guinea pig ileum and the rabbit jejunum have been preferred, since both give accurately reproducible contractions and are very sensitive. The effect of SP on various preparations of smooth muscle has been summarized in Table VI. The time-effect relation of contraction and relaxation is slower than that of histamine or carbachol, as shown on the guinea pig ileum or rat colon (Pernow and Rocha e Silva, 1955; Friedman and Jamieson, 1958). The response to pure SP is, however, quicker than to bradykinin (Franz et al., 1961). Libonati and Segre (1960) found an equal degree of activity of SP in acid ( p H 4.7) and alkaline ( p H 8.0) environments, whereas the effect of bradykinin was increased at acid pH. The automaticity of the movements of the intestinal villi is increased after intravenous administration of SP and remains after hexamethonium, which abolishes the action of villikinin ( L u d h y et al., 1960). Since the action of SP is not inhibited by atropine, hexamethonium, nicotine, cocaine (Pernow, 1953), antihistamines (Douglas et al., 1951), and serotonin antagonists (Feldberg and

192

F. LEMBECK AND G. ZETLER

ACTION

OF

Species

SUBSTANCE P

ON

TABLE VI ISOLATED ORGANS OF VARIOUS SPECIES

Organ

Teleosts Elasmobranches Goldiish

Intestine Intestine Intestine

Rabbit

Guinea pig Monkey Dog Cat Mouse, rat Frog Hen Rat Man Man Man

Esophagus, strips of gastric wall, small intestine Jejunum Strips of gastric wall Ileum Small intestine Small intestine Small intestine Small intestine Small intestine Rectal caecum Colon Duodenum Jejunum Ileum

Guinea pig Rabbit Rat Dog Dog Dog Guinea pig Guinea pig

Uterus Uterus Uterus Ureter Urinary bladder Gall bladder Gall bladder Bronchial muscle

Rabbit Guinea pig

Reaction

Referencesu

Strong and sustained 5 Contraction 5 Contraction, most sensitive 12 preparation

Contraction Contraction, very sensitive

9

Contraction Contraction, very sensit'ive Contraction Contraction Contraction Contraction Contraction Contraction, very sensitive Contraction Contraction Contraction Slow contraction after initial relaxat#ion Contraction, low sensitivity Contraction, low sensitivity Contraction Contraction, low sensitivity Contraction, low sensitivity No effect No effect Contraction

9 4, 9 9 9 9 3, 9 9 2, 9, 10 2, 6 8 8

1, 2, 3, 4, 9

8

3, 4, 9 3 3 9 9 799 7 11

,. References: 1. 2. 3. 4. 5. 6.

Blair and Clark (1956). Dalgliesh et al. (1953). von Euler (1936a). von Euler and Gaddum (1931). yon Euler and Ostlund (1956b). Friedman and Jamieson (1958).

7. 8. 9. 10. 11.

Hultman (1955). Liljedahl et a2. (1958). Pernow (1953, 1961). Pernow and Roche e Silva (1955). Stern and Vukdevi6 (1960). 12. Gaddum (1961).

Toh, 1953; Gaddum, 1953b), the receptor site has been assumed to be located in the muscle fiber (Pernow, 1953). Cocaine and nicotine showed inhibitory effects only in very high concentrations

SUBSTANCE P

193

which also inhibited the effects of other substances, e.g., histamine (Pernow, 1953). Blair and Clark (1956) concluded from experiments on the longitudinal muscle of the rabbit jejunum that SP does not stimulate the intrinsic ganglia. Deprivation of oxygen or glucose at 37"C, keeping the gut for 19 to 24 hours at low temperature without oxygen, or reducing the temperature of the bath to 30"C, depressed the effects of SP, barium, and physostigmine much more than that of nicotine. Blair and Clark (1956) think that SP acts through a rather labile mechanism and has probably a trophic influence on the amplitude of spontaneous contractions. The longitudinal contraction of the guinea pig ileum due to distension was inhibited by cooling to 25"-17"C and the action of nicotine, serotonin, barium, and SP was also inhibited by lowering the temperature. Morphine inhibited the effects of nicotine, serotonin, and barium, but not of SP; with regard to the first three substances, Innes et al. (1957) explain this by an action on nervous elements, but they did not include SP in their discussion. Lewis (1960) found that the inhibition of the action of SP by morphine on the guinea pig ileum was of the same size as that of other directly stimulating substances, whereas drugs which act by stimulation of nervous structures are much more depressed by morphine. Beleslin and Varagii: (1960) showed that in small doses SP potentiated the response of the isolated guinea pig ileum to nicotine, and inhibited it in large doses. SP also potentiated the response to acetylcholine in most of the experiments; this potentiation could be inhibited by hexamethonium. They explain this potentiation by an action on nervous structures. 2. Peristaltic Reflex Gernandt ( 1942), using the Trendelenburg techniques, showed that the peristaltic reflex could be elicited at a lower threshold when SP was in the bath. SP enhanced the number and amplitude of the peristaltic waves. Beleslin and Varagii: (195Sa) applied SP intraluminally and found also an increase of the peristaltic activity especially in fatigued intestine. Hexamethonium was able to inhibit the stimulating effect of SP on the peristaltic reflex, whereas it did not inhibit the action of SP on the longitudinal muscle (Pernow, 1953). SP failed to initiate peristaltic waves in preparations from

194

F. LEMBECK AND G. ZETLER

which the mucous membrane had previously been removed. SP (30-100 units) when acting from the outside of the guinea pig ileum abolished the peristaltic reflex and reduced the stimulating action of nicotine (Beleslin and Varagib, 195813). These authors (1958a, b) put forward the concept of two mechanisms of SP action, one directly on the muscle and the other trough the nervous elements of the peristaltic reflex arc. Medakovib and Radmanovik (1959) found that intraluminally introduced SP cannot restore the peristaltic reflex previously abolished by morphine-like analgesics, and furthermore that the stimulation of the peristaltic reflex by SP was inhibited by these analgesics.

3. I n Vivo Preparations Gernandt (1942) reported an increased motility of the small intestine after intravenous injection or local application of SP to the mucosa in rabbits; the colon, however, did not respond. Grabner and Lembeck (1961) found an increase in tonus of the ileocaecal valve in cats after intra-aortal injection of 50 units/kg of SP. Liljedahl et ul. (1958) infused 600-1000 units SP intravenously in patients during a period of 20 minutes. The cineroentgenographic recording showed a marked increase of the segmental and peristaltic movements of the small intestine beginning about 3 minutes after the start of the infusion of 60 unitdminute and lasting for about 20 minutes following the end of the infusion. In three cases of intestinal paralysis a transient motility was initiated by the infusion. Prostigmine influenced mainly the local segmental contractions, whereas SP showed a more pronounced effect on peristalsis. B. CIRCULATION In the first publication about SP, von Euler and Gaddum (1931) noted a fall of blood pressure in the atropinized rabbit after intravenous injection. The effect is due to peripheral vasodilatation (von Euler, 1936a). In the rabbit, the fall of blood pressure is rapid and cannot be inhibited by atropine, antihistamines, or ganglionic blocking agents (Pernow, 1953). In the cat much higher doses are necessary and after a slow return a secondary rise in blood pressure may follow (Pernow, 1953); the effect is greater in cats anesthetized with chloralose than with urethane (Bjurstedt et al., 1940). The blood pressure of the rat is rather insensitive. When purifying SP,

SUBSTANCE P

195

Pernow (1953) noted a change in the ratio between the intestinal and the hypotensive effect in the different fractions. Franz et uZ. (196l), however, found in the rabbit’s blood pressure, the guinea pig ileum, and the hen rectal caecum an equal increase of activity during the purification up to the pure preparation of 30,000 to 35,000 units/mg. Holton and Holton (1952, 1954) observed a vasodilatation following intra-arterial injection of SP in the rabbit ear previously deprived of its sympathetic nerve supply; at the time of application the preparation had regained its vasoconstrictor tone. On the innervated, isolated perfused rabbit ear Lembeck (1957) saw only negligible or no vasodilatation after intra-arterial injection of 9 units of SP. No other isolated vessels or perfused organs have been used to investigate the vasodilator effect of SP and nothing is known about an interaction with drugs. In man, intravenous infusion of SP (25-75 unitdminute) leads to a fall in blood pressure, accompanied by a rise in pulse rate, which is probably a natural reflex resulting from the hypotension. A bright red flush in the face was noticed and the patients felt a throbbing sensation in their heads (Dun& and Pernow, 1960). The cardiac output was slightly increased, and the arteriovenous oxygen difference (as calculated from the brachial and pulmonary blood) was slightly decreased. The splanchnic and upper arm blood flow increased (Pernow, 1961). These effects disappeared during continuous infusion even when the rate of infusion was subsequently raised (see also Section V, D ) ; the hemodynamic effects were seen again, however, when the infusion was repeated after 15 to 30 minutes. Amin & al. (1954) found a vasodilator action of 0.25-0.375 units of SP on the isolated perfused rabbit ear, but the preparation was not very suitable for routine assay. On the perfused frog leg the SP preparation used had a vasoconstrictor effect.

C. NERVOUS SYSTEM Central effects of substances are often difficult to determine, because the blood-brain barrier may hinder their entry from the circulatory system, or because peripheral effects are prominent. Acetylcholine, for example, although present and of high physiological significance within the central nervous system, cannot be applied peripherally in doses high enough to enter the brain in sufficient amounts to produce effects there. But the central actions of nicotine,

196

F. LEMBECK AND G. ZETLER

physostigmine, or atropine, which do enter the brain, offer some general views about central actions of acetylcholine. Direct injection into the brain is possible only by microinjection into small areas or into the ventricles, which can somehow be regarded as a pharmacological “back entrance” to the brain. There is some evidence that SP enters the brain from the circulation. However, the evaluation of effects observed thereafter should be done cautiously, since SP has powerful peripheral actions on the intestine and blood vessels. Thus effects which may look like central ones could be of reflex origin. Earlier experiments using crude samples of SP showed sedative effects in mice (Zetler, 1956a) and in wild hares (Stem and Milin, 1959), inhibition of the aggressivity in Betta splendens (Stem and Hucovi6, 1958), and prolongation of the hexobarbital sleeping time (Zetler, 1956a); SP also induced antagonistic effects against tetanus toxin (Stem and Hucovi6, 1956), against strychnine and harmine (Zetler, 1956a), and against the central excitatory action of iminodipropionitrile in mice (Stern and Dobri6, 1957). Other work revealed some synergism to mephenesin and meprobamate (Stem et d.,1957, Stern et al., 1958) and potentiation of bulbocapnine catatonia (Zetler, 1956a, b). Hucovib et al. (1961) could exclude the effect of the AMP contamination of SP preparations with regard to the increased strychnine tolerance, the prolonged hexobarbital sleeping time, and the morphine antagonism observed after injection of SP into mice, since SP preparations in which AMP was enzymatically destroyed showed the same results. AMP itself was found to enhance the morphine analgesia. In recent experiments, Stern and Hucovib (1960) showed that partially purified samples (100-270 units/mg) of SP did not antagonize convulsions caused by strychnine, whereas samples of a purity of 20 units/mg were effective. In contrast to the crude sample, the purified one did not prolong the hexobarbital sleeping time in mice. Partially purified samples still had an antagonistic effect against morphine analgesia at a dose of lo00 units/kg, but not at the large dose of 10,OOO units/kg. These recent results throw some doubt on many earlier experiments which used much less purified preparations of SP and which need to be repeated with a pure preparation. 1. Intrauentricular Administration Von Euler and Pernow (1954, 1956) injected 1%25 units of SP into the cystema magna and the lateral and third ventricles of

SUBSTANCE P

197

rabbits and cats. The behavior of the animals showed an inhibition of spontaneity, and licking, swallowing, yawning, stupor, changes in blood pressure, and long-lasting (30 minutes) increase in respiratory rate; one cat also became very aggressive. Since a purer preparation (1013300 units/mg) was used, these results clearly demonstrate that SP can exert central effects, There was a latency period of about 10 minutes before the onset of the central effects of SP, in contrast to the action of acetylcholine which was given in the same way and produced similar symptoms almost immediately. Diuresis was not influenced by SP in these experiments. These results indicate that SP, in an amount less than that present in 1 gm of brain tissue, exerts a central neurotropic action when introduced into the cerebrospinal fluid. 2. Spinal Reflems Angelucci (1956) perfused the spinal cord of frogs via the vertebral canal and recorded the flexor reflex. Whereas the reflex was potentiated by acetylcholine, nicotine, and strychnine, it was inhibited by ether, chlorbutol, mephenesin, chlorpromazine, reserpine, and morphine. SP did not influence these effects at all. Similar results were obtained by Kissel and Domino (1959) ; intravenous injection of SP in doses up to 18 mg (units/mg not stated) did not affect spinal reflexes in spinal cats, There remains some doubt whether this evidence is conclusive in ruling out spinal effects of SP, since the actual dose that reaches the spinal cord in these experiments might have been too low; a close arterial injection might render more decisive results. Stern et al. (1957) found an inhibition of the homolateral flexor reflex and the contralateral extensor reflex in anesthetized or spinal cats after injection of 500 units of SP, but no influence on the patellar tendon reflex. Conditioned reflexes, discrimination tests, and the maze performance of rats were not influenced by intraperitoneal injection of 2300 units/kg of SP (Stem, 1959).

3. Release from the Central Nervous System Angelucci (1956) perfused the spinal cord of frogs and found after electrical stimulation of both feet that the collected perfusate stimulated the guinea pig ileum. The active principle in the perfusion fluid was not identified but he speculated that it was “more likely , . . substance P than 5-hydroxytryptamine.” Hilton and

198

F. LEMBECK AND G . ZETLER

Schain (1961) perfused the ventricular system of anesthetized cats and searched for pharmacologically active substances in the effluent from aqueduct and cisterna. No SP was found, either before or after strong electrical stimulation of the brain with electrodes implanted in the hypothalamus or midbrain, or with external electrodes at occiput and maxilla. Catechol amines, 5-hydroxytryptamine, histamine, and oxytocin could not be detected either, whether the brain was stimulated or not.

4. Release of ACTH Swingle et al. (1956) found that samples of SP prepared from horse intestine are rich in a component releasing ACTH from pituitary gland in uitro. After strong acid hydrolysis all the SP was destroyed, but the hydrolyzate still contained all the ACTH-releasing activity and also a ‘histamine-like substance.”

5. Effect on the EEG Crepax and Infantellini (1956) investigated the effect of SP on the isolated cortex of the cat after local application (100 units) or injection into the carotid artery of the contralateral side (40 to 100 units). They observed an increased excitability of the cortical neurons, prolongation of rhythmic activity (which was elicited by repeated electrical stimulation), and the appearance of a continuous or augmented electrical activity in previously inactive or slightly active preparations. Lechner and Lembeck (1958) injected 30 to 100 units of SP (purity 5.6 units/mg) centrally into the carotid artery of slightly anesthetized rabbits and found a decrease in amplitude and an increase in frequency of cortical activity; frequency and synchronization of the hippocampus activity were increased. It was concluded from these findings that SP stimulates the activating systems of the reticular formation. Caspers and Stem (1961) applied SP (purity 6.5-20 units/mg) locally to the cortex of unanesthetized rats with previously implanted electrodes allowing stimulation. They recorded the spontaneous rhythm, the direct cortical response (dendritic potential), and the d.c. component of the neocortex. On local application of SP (2-10 units), the surface-negative dendritic potential was increased in voltage and duration (Fig. 7). The d.c. potential of the cortical surface became more positive at the side of application

SUBSTANCE P

199

FIG. 7. Dendritic potential before ( N ) and after (a-e) epicortical application of 5 units of substance P. a: Recording 2 minutes after the application, intervals between the following recordings ( a - e ) 10 minutes each. The strength of stimulation was kept constant during the experiment (Caspers and Stem, 1961) .

(Fig. 8); this positive shift of the steady potential was associated with a strong reduction of the surface-negative d.c. displacements which are released by various natural stimuli. Furthermore, an EEG pattern similar to that of sleep developed earlier on the side of application of SP. This effect on the cortical macropotentials could be reproduced and increased with an anodic polarization or decreased by cathodic polarization. Local application of y-aminobutyric acid antagonized the effect of SP. Samples of SP inactivated by tryptic digestion were found to be inactive in this respect. The authors reached the conclusion that SP exerts a hyperpolarizing action in the upper layer of the cerebral cortex. In a recent thorough study Caspers ( 1961) applied crude SP preparations (6.7, 16.5, and 20 units/mg) in doses of 500-10,000 units/kg intraperitoneally to

200

F. LEMBECK AND G. ZETLER

SP

1 10

20

30

40m in 50

FIG. 8. Changes in cortical d.c. potential after local application of substance P. The potential found at the beginning has been plotted as 0 and changes observed are referred to it. K = changes in d.c. potential in untreated controls, SP = changes in d.c. potential after 6-8 units substance P. Vertical lines express the threefold standard deviation (Caspers and Stem, 1961).

conscious, freely moving rats with implanted electrodes. SP greatly diminished spontaneity and abolished the electrocortical equivalents of wakefulness in the same manner as seen with topical application. The reactivity of the animals to acoustic stimulation was very easily depressed by SP, while the reactivity to tactile stimulation was much more resistant, and the electrocortical equivalent of the conditioned alarming reaction showed the greatest resistance. Furthermore, SP was found to decrease the normal activity level of the neurons in the midbrain reticular formation as well as the response of these structures to tactile stimulation (Fig. 9). This effect lasts at least 30 minutes. Caspers concludes that SP evokes sedation in part by depressing the cortical generator structures and in part by inhibiting the ascending reticular formation. These results do not correspond with those of Lechner and Lembeck (1958), and the question was raised whether the arousal reaction found by the latter was caused by carotid sinus stimulation; this should be re-

SUBSTANCE P

201

FIG.9. The action of substance Y on the unit discharges of neurons in the midbrain reticular formation of a conscious rat, evoked by tactile stimulation of the vibrissae. The moment of stimulation in each tracing is indicated by the arrow R. Both the level of the background activity and the normal stimulation response (N)of reticular neurons, recorded with a 50-p electrode, are strongly reduced after the intraperitoneal application of 8OOO units/kg. The normal response has recovered after approximately 60 minutes (Caspers, 1961).

investigated. In any case, the described results give further evidence for the characteristic electrocortical activity of SP. 6. Dorsal Root Potentials

Krivoy (1961), using the method of Lloyd and McIntyre (1949), found an increase of potential no. IV of the dorsal root complex (DR IV) in the cat after intravenous injection of 10 pg/kg LSD. Smaller doses of LSD ( 5 pg/kg) did not alter by themselves the DR IV potential but showed an enhancement of the DR IV potential by SP ( 8 units/kg intravenously) which by itself had no effect. He regards LSD as a specific inhibitor of the enzyme which destroys SP (Krivoy 1957) and is, therefore, able to unmask the action of SP. In this respect it must be mentioned that Smith and Walaszek (1960) reached another conclusion about the influence of LSD on

202

F. LEMBECK AND G . ZETLER

the effect of SP on the guinea pig ileum. They found that 0.03-0.1 pg/ml LSD potentiated not only the action of substance P but also that of substance A and bradykinin. This LSD effect is therefore not specific and the explanation of Krivoy (1957) is perhaps not the only one possible.

7. SuperZor Cervical Ganglion Trendelenburg (1956) showed that a number of substances of different chemical nature potentiate the ganglionic transmission. It is interesting to note that SP has similar effects when injected intraarterially in doses of 10 to 30 units ( Beleslin et al., 1960). Furthermore, SP as well as nicotine-like substances in high doses depress ganglionic transmission, SP in small doses potentiates and in large doses inhibits the stimulation of the ganglion by acetylcholine ( Beleslin et al., 1960). The stimulating action suggests similar observations on the peristaltic reflex (Beleslin and Varagih, 1958a, b, 1960) and on peripheral nerve endings (Lembeck, 1957). Beleslin and Radmanovii: (1961) used the isolated perfused superior cervical ganglion and found that a dose of only 5 units of SP inhibited the effect of preganglionic stimulation or of acetylcholine injection and diminished the acetylcholine output from the ganglion. This is in some contrast to the results on the superior cervical ganglion left in the normal circulation.

8. Peripheral Nerve Endings Lembeck (1957), using the method of Mietkiewski (1956), investigated the effect of SP on peripheral sensory nerve endings. The preparation consisted of an isolated perfused rabbit ear connected with the body only through the intact auricular nerve; blood pressure and respiration were recorded. Stimulation of the nerve endings by pinching the ear caused a reflex fall in systemic blood pressure and an increase in respiration. Intra-arterial injection of epinephrine into the perfusion fluid of the ear produced vasoconstriction in the ear but did not have an influence on the systemic blood pressure. Acetylcholine and SP ( 9 units) when injected into the perfusate of the ear elicited a reflex fall in blood pressure and an increase in respiration which was abolished after injection of cocaine into the perfusate. After hexamethonium only the stimulating effect

SUBSTANCE P

203

of acetylcholine was blocked; the effect of SP was not blocked (see also Potter et al., 1962). 9. Allergy Stern and VukCevii: (1960) evoked an anaphylactic shock in guinea pigs by letting the animals breathe a spray of protein against which the animals were sensitized previously. The anaphylactic reaction could be prevented by injection of 150 to 200 units (50 mg) of SP prior to the inhalation of the spray, No protective effect of SP could be observed against histamine or serotonin asthma. SP did not influence the Dale-Schultz reaction of the guinea pig ileum. In perfused lungs, SP was found to produce a bronchoconstriction. Stern and VukCevib (1960) briefly discuss a central nervous effect of SP which would be able to inhibit the anaphylactic reaction. This phenomenon is difficult to explain and much further evidence is needed.

D. TACHYPHYLAXIS Tachyphylaxis can occur with several smooth-muscle-stimulating substances; i.e., the effect of these agents decreases when high doses are given at short intervals. This is known to be the case for the effect of histamine on the guinea pig ileum and, more particularly for that of serotonin on the rat uterus. Giving low or intermediate doses with adequate intervals between the applications prevents the rise of tachyphylaxis in the usual bioassay. SP (Fig. 10) also shows this phenomenon ( Gaddum, 195313). The tachyphylaxis can be used for identification of SP, since a contraction produced by an unknown principle can be attributed to SP when it is inhibited after giving a large dose of SP. The tachyphylaxis may influence the accuracy of the four-point assay when after two large doses a small dose is given which thus will be smaller than expected. The tachyphylaxis can hardly be seen on the blood-pressure effect of SP in the rabbit. Lechner and Lembeck (1958) also recorded decreased effects on the EEG after 4 to 5 repeated injections of the same dose of SP into the carotid artery. von Euler and Pernow (1956) injected SP into the ventricles of rabbits and cats. They found that repeated injections had much weaker, if any, effects on blood pressure and respiration, which also could be explained by tachyphylaxis. Dunkr and

204

F. LEMBECK AND G . ZETLER

.

rr

i I t o 2

t

0

4

t

8

o

~

16

o

32

t

o

L ~

0o 0

~

o

64 128

FIG. 10. Tachyphylaxis on the guinea pig ileum. At open circles addition of 10 units substance P. Intermediate, at the arrows raising amounts of substance P (units) are given. A decline of the effect of 10 units is seen after intermediate addition of 64 units of substance P.

Pernow (1960) infused highly purified preparations of SP in man, and observed peripheral vasodilatation with flushing and an initial and transitory fall of blood pressure which did not appear again when the rate of infusion was subsequently increased. The tachyphylaxis was short lasting, however, and infusions repeated after 15 to 30 minutes gave the same hemodynamic effects as before. The hyperalgesic and morphine-antagonistic effect of SP in mice can disappear after repeated subcutaneous injections of the polypeptide (Zetler, 1956a). In these experiments, the tolerance or tachyphylaxis developed within 4 days after the last normally effective dose (16,500units/kg) and was still present 3 and 5 days later. On the sixth day the brains of the animals were homogenized and dried with acetone. The powdered brains of the treated animals revealed a 50%higher SP-destroying enzyme activity than the brains of an untreated control group. Assuming a specific SP-destroying enzyme, this would mean an acquired tolerance owing to an increased enzyme activity.

SUBSTANCE P

205

VI. Pharmacological Interactions with Drugs

A. INTESTINE The lack of an interaction, namely the failure of atropine to block its action on the gut, led to the discovery of SP (von Euler and Gaddum, 1931). Pernow (1953) first investigated the interactions between SP and atropine, antihistamines, ganglionic blocking agents, and cocaine on the isolated guinea pig ileum and found all these drugs free from antagonistic activity in adequate concentrations (Section V, A). Only adrenochrome inhibited the action of SP and of serotonin on the guinea pig ileum, but not that of histamine and barium (Stern et al., 1956). This effect of adrenochrome can hardly be regarded to be specific. AMP present in tissue extracts also interferes with the action of SP on the rabbit jejunum (Section 11, B ) . Stem and Hucovib (1961) carefully tested a large number of compounds for antagonism against SP using the PA, method of Schild ( 1947). They found that only cystine-di-p-naphthylamide and Arfonad had a greater pa,-value against SP than against histamine, serotonin, and acetylcholine. Many other compounds (listed in Table VII) were found to be inactive or had only an unspecific inhibitory action. The question of receptors on the smooth muscle can usually be answered by antagonists. If a drug cannot be inhibited by antagonists which block all other stimulating substances, one more receptor is generally assumed which allows the drug in question to “hit the membrane.” Since no specific antagonists of SP on the longitudinal muscle are known, this concept would lead to the assumption of a specific receptor for it. It may be mentioned that also no specific antagonists of other smooth-muscle-stimulating polypeptides are known, with the one exception of a weak antagonist of oxytocin (Konzett, 1957). There are, however, antagonists against the actions of SP on the nervous elements of the peristaltic reflex (Section V, A, 2 ) and the central nervous system ( Section VI, B ) . Lembeck and Petschke (1961) investigated the influence of neostigmine, atropine, and morphine on the SP content of rat small intestine. The animals were killed 30 minutes after the administration of the drugs; the gut was extracted and the extracts tested for SP. Despite the obvious increase of gut motility caused by neostigmine, it had no influence on the SP content of the gut; atropine and

206

F. LEMBECK AND G. ZETLEn

TABLE VII SUBSTANCES F o m u TV l ? ~IXEFFECTIVE OR UNSPECIFIED INHIBITORS OF ACTION O F SUBSTANCE P ON GUINEAP I G ILEUM" Adcnosine (7) Adermine ( 7 ) yAminobutyric acid (7) p-Aminobenzoic acid (7) I-AMP, 2-AMP (7) Aneurine (7) Atropine (3, 6) ATP (7) Antihistamines ( 1 ) Avacan (7) Azulen (7) Bcnzoquinone ( 7 ) Biotin (7) Bromo-LSD (7) Buscopan (7) Bradykinin (5) Caffeine (7) Catechol (7) Citric acid ( 7 ) Cocaine (6) Codcine ( 7 ) Cysteine ( 7 ) Dehydrocholate (sodium) ( 7 ) Deoxyribonucleic acid (7) Dih ydroergotamine (4) Dilatol (7) Glutathione (7) Heparin (7) Histamine (7) Hexamethonium (6) 1. 2. 3. 4.

Lactoflavin (7) LSD ( 7 ) Iproniazid (7) Mephenesin (7) Meprobamate (7) 1,2-Naphthoquinone (7) 1,4-Naphthoquinone (7) 0-Naphthylamine (7) 8-Naphthylamine (7) N-(Naphthy1)-methylcnediaminc (7) Narceine (7) Nepresol ( 7 ) Nicotine (6) Oxytocin (7) Patulin (2, 7) Phenurone (7) Polyethylene sulfate (7) Protamine (7) Quinine (7) 8-Hydroxyquinoline (7) Renin (7) Scopolamine ( 7 ) Segontin (7) Semicarbazide (7) Sparteine ( 7 ) Strychnine (7) Thiosemicarbazide (7) Trimethadione (7) Urea (7) Vasopressin (7)

a Numbers in parentheses indicate references: 5. Lembeck, unpublished observation. Douglas et al. (1951). 6. Pernow (1953). Eliasson (1958). 7. Stern and Hukovid (1961). von Euler and Gaddum (1931). Gaddum (195313).

morphine, on the other hand, were also without effect (Table VIII) . It could be assumed that either a very slow and steady synthesis or release exists, or a rapid resynthesis follows a higher rate of release. Investigations on the tissue concentration of SP have not proceeded far enough to discriminate between these two possibilities.

207

SUBSTANCE P

TABLE VIII INFLUENCE OF ATROPINE,NEOSTIGMINE, AND MORPHINEON THE CONCENTRATION OF SUBSTANCE P IN SMALLINTESTINE OF RAT^

Drug Atropine Neostigmine Morphine a

Dose (mdkg, subcut.)

70 70 20

Number of animals

Concentration of substance P (units/gm)

9 9 9

56.1 k 8.8 57.5 rt 8.8 62.3 f 8.8

From Lembeck and Petschke (1961).

Beleslin and Varagi6 (1959) found an interaction between serotonin and SP. Intraperitoneal injection of 60 to 120 units/kg of SP in rats led to an increase of the serotonin content of the stomach and the ileum but not of the spleen. Intraluminal application of 60 to 80 units of SP into isolated, 15-20-cm-long pieces of guinea pig ileum resulted in a doubled content of serotonin in the gut when extracted after 20 to 35 minutes and compared with the control group. These results show that SP augments the synthesis or inhibits the release of serotonin either directly or by smooth-muscle stimulation. Since Bulbring and Lin (1958) found an increase of the serotonin release after raising the intraluminal pressure, the results of Beleslin and Varagi6 (1959) could be interpreted as showing that SP increases the intraluminal pressure, thus stimulating the release of serotonin and subsequently its resynthesis.

B. CENTRAL NERVOUS SYSTEM Central effects of substances have often and successfully been classified by interactions with drugs. Moreover, the central activity of several substances could be much better elucidated by such interactions than by direct observations of the behavior, affectivity, or other reactions of animals not under the influence of another centrally acting drug. Zetler (1956a, 195613) was the &st to investigate interactions of SP with drugs in the central nervous system. Some of these results have already been discussed in Section V, C since several central actions could not have been clarified without the use of interactions with other drugs. According to the recent investigations of Franz et al. (1!361), chemically pure SP has an

208

F. LEMBECK AND G. ZETLER

activity of 30,000 to 35,OOO units/mg, which means that a sample of 30 units/mg still contains 99.9% of other material which possibly could have other actions. Zetler (1956a, 1959), aware of this possibility, used as controls an SP-free protein fraction obtained in the course of preparation of SP, or SP treated with trypsin. Matched against these controls, he found an antagonism of SP against the excitatory effects of harmine, strychnine, picrotoxin, pentylenetetrazole, methamphetamine, and morphine in mice, a prolongation of the hexobarbital sleeping time and the bulbocapnine catatonia, but no effect on seizures due to electroshock, ammonium acetate, caffeine, or nicotine. There was a lowered threshold to painful stimuli and an antagonism against morphine analgesia. Neither the “control protein” nor SP inactivated by trypsin had any of these effects. Bradykinin, in contrast to SP, was without effect on the harmine tremor or the bulbocapnine catatonia (Zetler 1956b). Bonta et al. (1961) confirmed the depressant action of SP on an exploratory behavior test and the strychnine titration test in rats, but did not observe a prolongation of the hexobarbital sleeping time or a morphine antagonism; they mention, however, that a strychnine antagonism was also seen with an SP-free subfraction of hog brain. According to Stem and Hucovib (1960) preparations of SP with a purity of 100-270 units/gm are devoid of the strychnine antagonistic and hexobarbital synergistic action, but are still able to antagonize the morphine analgesia. Prii6 (1961) investigated drug effects on writhing in mice. Morphine depressed the writhing, and AMP did not; the effect of morphine was reduced by SP, but not by AMP. Zetler and Ohnesorge (1957) estimated the SP content of mouse brain after treatment with various centrally acting substances. Substances which are centrally exciting in mice, such as benzedrine or morphine, resulted in an increased, and chloroform, urethan, and phenobarbital in a lowered, total content of SP in brain; lysergic acid diethylamide was ineffective. By differentiating between an “available” and a “bound” form of SP (Section 11, A ) , an influence of such drugs could be revealed which did not modify the total amount of SP. Picrotoxin increased the available SP and lowered the bound SP, whereas LSD lowered the bound form leaving the available form unchanged. Paasonen and Vogt (1956) did not find an influence of dl-amphetamine, ephedrine, insulin, p-tetrahydronaphthylamine, caffeine, and reserpine on the SP content of hypo-

SUBSTANCE P

209

thalamus and nucleus caudatus of anesthetized dogs. Stem and Kocik-Mitrovik (1959) described a three- or four-fold increase of the SP concentration in rat brain after reserpine, but not after phenobarbital, chlorpromazine, meprobamate, mephenesin, or syrosingopine. Zetler and Ohnesorge (1957) discussed a possible explanation of the discrepancy between their results and those of Paasonen and Vogt ( 1956) : the latter authors anesthetized the dogs before killing them and this probably abolished the state of excitement that had been achieved by the centrally stimulating drugs. Concluding this chapter, the present reviewers feel that other unknown substances present in impure SP preparations rather restrict our present knowledge about the central actions of SP. Certain findings about central actions or interactions of SP are still a matter of discussion and need confirmation before they can be regarded as being significant. Despite these limitations, there has been accumulated a good number of results which evidently show central effects of preparations with typical SP activity. We are far from reaching a general statement about mode of action or physiological significance of this polypeptide. However, it must be kept in mind that substance P is one of the few substances which possess very high biological activity and are normally present in the nervous system. VII. Conclusions

Some good evidence has been achieved that SP is a small polypeptide present primarily in the nervous system and the intestinal tract. Under normal conditions the tissue of these organs very probably contains SP in its definite form; there are no hints that this polypeptidelike bradykinin-is enzymatically released from inactive precursors. Detailed information exists about the distribution of SP in various parts of the central and peripheral nervous system and of the intestine. The kind of distribution seems to indicate that the physiological role of SP is connected with function rather than with structure. It has been assumed that it is located in nerve cells and not in glial tissue. There are some findings which show changes of the SP content in brain under experimental conditions and a decrease in nerve after degeneration. The peripheral actions consist of a stimulation of intestinal and some other plain muscles, as well as a vasodilatation. As to the

210

F. LEMBECK AND G . ZETLER

action of SP on the brain, a number of pronounced effects have been observed. Unless these experiments have been repeated with pure SP their significance remains a matter for discussion since contaminants of polypeptide nature in the SP preparations which were used did probably contribute to some of these effects. It is very well possible, however, that a peptide of extremely high activity on the gut (there acting in a concentration of less than gm/ml despite being a larger molecule than acetylcholine, serotonin, or histamine) can act on cells of the central nervous system too. We are still far from reaching the conclusion that SP can be regarded as a central transmitter substance. There are a few results in favor and, until now, no results against such an assumption. At any rate, it is not probable that a substance having a high biological activity and belonging like histamine and acetylcholine to the group of “local hormones” should be without any physiological significance on the organ where it is normally present. Assuming the central effects observed with crude SP preparations cannot be attributed to SP itself, the question of what other active constituents are present in the impure material remains important. REFERENCES

Amin, A. H., Crawford, T. B. B., and Gaddum, J. H. (1954). J. Physid. (London) 126, 596. Andrews, T. M., and Holton, P. (1958). J . PhyhZ. (L&) 143, 45P. Angelucci, L. (1956). Brit. J . P h u r m o l . 11, 161. Arvidsson, U. B., Bernow, B., and Swedin, B. (1956). Actu Physiol. S c u d . 35, 338. Barcroft, I., and Barron, D. H. (1999). Ergeb. Physiol. u. exptl. Phurmakol. 42, 107. Behrens, M., and Seydl, G. ( 1951). Arzneimittel-Forsch. 1, 228. Beleslin, D., and Radmanovib, B. Z. (1961). Intern. Symposium on Substance P, Sarajevo, Yugoslavia, p. 61. Beleslin, D., and Varagib, V. (1958a). Brit. J. P h u m c o l . 13, 321. and Varagib, V. (1958b). J. Pharm. and Phumacol. 11, 99. Beleslin, D., Beleslin, D.,and Varagib, V. ( 1959). Experientiu 15, 186. Beleslin, D.,and Varagii., V. (1960). Arch. intern. pharmacodynamie 126, 321. Beleslin, D.,Radmanovi6, B., and Varagii., V. (1960). Brit. J. P h u m c o l . 15, 10. Bjurstedt, H., von Euler, U. S., and Gernandt, B. (1940). Skund. Arch. Physiol. 83, 257. Blair, M. R., and Clark, B. B. (1956). J. Phurmucol. Exptl. Therup. 117, 467.

SUBSTANCE P

211

Boissonnas, R. A,, Guttmann, St., and Jaqucnoud, P. A. (1960). Helu. Chim. Acta 43, 1349. Bonta, I. L., Wijmenga, H. G., and Hoheiisee, F. (1961). Acta Physiol. et Phurmucol. N e e d 10, 114. Braun-Menbndez, E., and Page, I. H. (1958). Science 127,242. Bulbring, E., and Lin, R. C. Y. (1958). J. Physiol. (London) 140, 381. Bulbring, E., Philpot, F. J., and Bosanquet, F. D. (1953). Lancet 264, 865. Cammemeyer, J. (1944). Skrifter Norske Videnskaps. Akal. Oslo I Mat.Naturv. KZ. No. 12. Caspers, H. ( 1961). Intern. Synipoaiuin on Substance P, Sarajevo, Yugoslavia, p. 29. Caspers, H., and Stem, P. (1961). Arch. ges. Physiol. Pfliig. 273, 94. Cleugh, J., Gaddum, J. H., Holton, P., and Leach, E. (1961). Brit. J. Phurmucol. 17, 144. Correale, P. ( 1959). Arch. Intern. phurmacodymrnie 119, 435. Crepax, P., and Infantellini, F. ( 1956). Arch. sci. biol. ( Bologna) 40, 297. Dahlstedt, E., von Euler, U. S., Lishajko, F., and iistlund, E. (1959). Acta Physiol. Scand. 47, 124. Dale, H. H. (1935). Proc. Roy. SOC.Mecl. 28, 319. Dalgliesh, C. E., Toh, C. C., and Work, T. S. (1953). J. Physiol. (London) 120, 298. De Robertis, E., and Franchi, C. M. (1956). J. Biophys. Biochem. Cytol. 2, 307. Douglas, W. W., Feldberg, W., Paton, W. D. M., and Schachter, M. ( 1951). J. Physiol. ( L o n d o n ) 115,163. Dun&, H., and Pemow, B. (1960). Acta Physiol. Scand. 49, 261. Dun&, H., von Euler, U. S., and Pernow, B. (1954). A c t a Physiol. Scand. 31, 113. Eber, O., and Lembeck, F. (1956). Arch. eyptl. P u t h d . Pharmakol. NaunynSchmiedeberg’s 229, 139. Eccles, J. C. (1957). “The Physiology of Nerve Cells.” Oxford Univ. Press, London and New York. Ehrenpreis, T., and Pemow, B. ( 1952). Acta Physiol. Scand. 27,380. Ehringer, H., and Homykiewicz, 0. (1960). Klin. Wochschr. 38, 1240. Eliasson, R. ( 1958). Experientia 14, 460. Eliasson, R., Lie, L., and Pemow, B. (1956). Brit. J. P h r m a c o l . 11, 137. Feldberg, W., and Lin, R. C. Y. (1950). J. Physiol. (London) 111, 96. Feldberg, W., and Mann, T. ( 1946). J. Physiol. (London) 104, 411. Feldberg, W., and Toh, C. C. ( 1953). J. Ph!/,\iol. (London) 119, 352. Feldberg, W., and Vogt, M. (1948). J . Physiol ( L o n d o n ) 107, 372. Florey, E. ( 1954). Arch. intern. Physiol. 62, 33. Franz, J., Boissonnas, R. A., and Stiinner, E. (1961). Helv. Chim.Acta 44, 881. Friedman, S. M., and Jamieson, J. D. (1958). Proc. SOC. Exptl. B i d . Med. 97, 767. Gaddum, J. H. (1953a). Brit. J. PhurmacoE. 8, 321. Gaddum, J. H. (195313).J . Physiol. (London) 119,363. Gaddum, J. H. ( 1955a). Personal communication to Lembeck.

212

F. LEMBECK AND G. ZETLEX

Gaddum, J. H. ( 1955b). In “Polypeptides Which Stimulate Plain Muscle” (J. H. Gaddum, ed. ), p. 130. Livingstone, London. Gaddum, J. H. (1960). In “Polypeptides Which Affect Smooth Muscles and Blood Vessels” (M. Schachter, ed.), p. 163. Pergamon, London. Gaddum, J. H. (1961). Intern. Symposium on Substance P, Sarajevo, Yugoslavia, p. 7. Gaddum, J. H., and Lembeck, F. (1949). Brit. J. P h u m o l . 4, 401. Gaddum, J. H., and Schild, H. ( 1934). J. Physiol. (London) 83, 1. Gernandt, B. (1942). Actu Physiol. S c u d . 3,270. Grabner, K., and Lembeck, F. (1960). Arch. exptl. Pathol. Phurmukol. Nuunyn Schniedeberg’s 240, 157. Grabner, K., and Lembeck, F. ( 1961 ). Unpublished results. Grabner, K., Lembeck, F., and Neuhold, K. (1959). Arch. exptl. Pathol. Phul.makol. Nuunyn-Schmiedeberg’s 236, 331, Greengard, P., Brink, F., Jr., and Colowick, S. P. (1954). J. CelluZur Comp. Physiol. 44,395. Gullbring, B. (1943). Actu Physiol. S c u d . 6, 246. Haug, H. (1953). Actu Anat. 19,239. Hebb, C . 0. (1955). Quart. J. Exptl. Physiol. 40, 176. Hebb, C. O., and Silver, A. (1956). J. Physiol. (London) 134, 718. Hebb, C. O., and Waites, G. M. H. (1956). J. Physiol. (London) 132, 667. Hellauer, H. ( 1953). Arch. exptl. Puthol. Phurmakol. Naunyn-Schrniedeberg’s 219, 234. Hellauer, H. F. (1950). Z. uergleich. Physwl. 32,303. Hellauer, H. F., and Umrath, K. (1947). J. P h y k l . (London) 106, 20P. Hellauer, H. F., and Umrath, K. (1948). Arch. ges. Physiol. Pfliiger’s 249, 619. Herken, H., and Neubert, D. ( 1953). Arch. exptl. Puthol. Phurmakol. NaunynSchmiedeberg’s 219,223. Hilton, S. M., and Schain, R. J. ( 1961 ). J . Physiol. (London) 157, 46P. Himwich, H. E., and Aprison, M. H. (1955). In “Biocheniistry of the Developing Nervous System” (H. Waelsch, ed.), p. 301. Academic Press, New York. Holton, F. A., and Holton, P. ( 1952). J. Physiol. (London) 118, 310. Holton, F. A., and Holton, P. (1954). 1. Physiol. (London) 126, 124. Holton, P. ( 1958). J. Physiol. (London) 143,45F’. Holton, P. (1960). In “Polypeptides Which Affect Smooth Muscles and Blood Vessels” ( M. Schachter, ed.), p. 192. Perganion, London. Horton, E. W. (1959). Brit. J. Phurmucol. 14, 125. Hryntschak, T. (1925). Z. Urol. 18, 86. HukoviQ S., KoBak, R., and Stem, P. (1961).Intern. Symposium on Substance P, Sarajevo, Yugoslavia, p. 15. Hdtman, E. H. (1955). Actu Chern. S c u d . 9,1042. H y d h , H. (1958). 4th Intern. Congr. Bimhem. ( V h n u ) Symposium 111, reprint 4. (Suppl. to Intern. Abstr. Biol. Sci.) Pergamon Press, London. Innes, I. R., Kosterlitz, H. W., and Robinson, J. A. (1957). J . Physiol. (London) 137, 396. Johnson, A. C., McNabb, A. R., and Rossiter, R. J. (1950). A.M.A. Arch. Neurol. Psychiat. 64, 105.

SUBSTANCE P

213

Kavaler, F., and Kimel, V. M. (1952). J. C o m p . Neurol. 96, 113. Kissel, J. W., and Domino, E. ( 1959 ) . J. Phurmacol Exptl. Therap. 125, 168. Kocib-MitroviQ D. ( 1959). Arch. exid. Pathol. Pharmakol. Naunyn-Schmiedeberg’s 236, 37. Konzett, H. (1957). Helv. Physiol. Actu 15, 419. Konzett, H., and Stiirmer, E. (1960). Brit. J. Pharmacol. 15,544. Kopera, H., and Lazarini, W. ( 1953). Arch. exptl. Pathol. Phnrtnakol. NaunynSchmiedeberg’s 219, 214. Krivoy, W. A. (1957). Brit. J. Pharmucol. 12, 361. Krivoy, W. A. (1961). Brit. J. Pharnucol. 16, 253. Laszlo, I. ( 1960). J . Physiol. (London) 153, G9P. Leach, E. ( 1959). 1. Physiol. (London) 149, 34P. Lechner, H., and Lembeck, F. (1958). Arch. exptl. Pathol. Phurmukol. NaunynSchmiedeberg’s 234, 419. Lembeck, F. ( 1953). Arch. exptl. Pathol. Pharnukol. Naunyn-Schmiedeberg’s 219, 197. Lembeck, F. ( 1957). Arch. exptl. Pathol. Pharmukol. Naunyn-Schmiedeberg’s 230, 1. Lembeck, F., and Holasek, A. ( 1960 ) . Arch. exptZ. Pathol. Pharmakol. NuunynSchmiedeberg’s 238, 542. Lembeck, F., and Petschke, B. ( 1961) . Unpublished results. Lewis, G. P. (1960). Brit. J. Pharmucol. 15, 425. Libonati, M., and Segre, G. (1960). Arch. exptl. Puthol. Phurmakol. NaunynSchmiedeberg’s 240, 14. Liljedahl, S. O., Mattsson, O., and Pernow, B. (1958). S c a d . J. Clin. 6 Lab. Inuest. 10, 16. Lloyd, D. P.’C.,and McIntyre, A. K. ( 1949). J. Gen. Physiol. 32,409. Loewi, O., and Hellauer, H. (1938). Arch. ges. Physiol. Pfiiger’s 240, 769. Ludiny, G., GLity, T., Rig6, J., aiid Szab6, H. (1960). Arch. ges. Physiol. Pfliiger’s 270, 499. Macintosh, F. C. ( 1941). J. Physiol. (London) 99,436. Matussek, N. (1959). 2.physiot. Chem. Hoppe-Seyler’s 316, 241. Medakovi6, M., and Radmanovib, R. (1959). Arch. intern. pharmacodynumie 122, 428. Mietkiewski, E. ( 1956). Arch. Intern. phnrnLucodynamie 104, 373. Paasonen, M. K., and Vogt, M. (1956). J. Pli!/sioZ.(London) 131, 617. Pernow, B. ( 1951). Acta Physiol. Scnnd. 24, 07. Pernow, B. (19’53).Acta Physiol. Scnntl. 29, Suppl. 105. Pernow, B. ( 1955a). Acta Physiol. Scund. 34, 295. Pernow, B. (1955b). Z. Vitamin- Hormon- u. Fermentforsch. 7,59. Pernow, B. ( 1961). Intern. Symposium on Substance P, Sarajevo, Yugoslavia, p. 45. Pernow, B., and Rocha e Silva, M. (1955). Actu Physiol. Scund. 34, 59. Potter, G. D., Gurman, F., and Lim, R. K. S. (1962). Nature 193,983. Pr%, R. ( 19’61) . Intern. Symposium on Substance P, Sarajevo, Yugoslavia, p. 71. Richter, D., and Crossland, J. (1949). Am. J. Phyriol. 159, 247.

214

F. LEMBECK AND G . ZETLER

Rocha e Silva, M. ( 1951).Arch. intern. p h u m o d y n u m i e 88,271. Rocha e Silva, M. (1960). In “Polypeptides Which Affect Smooth Muscles and Blood Vessels” (M. Schachter, ed.), p. 210. Pergamon, London. Sawyer, C. H. (1946). Am. J. Physiol. 146, 246. Schild, H. 0. (1947). Brit. J. Pharmacol. 2, 251. Skeggs, L. T. (1960). In “Polypeptides Which Affect Smooth Muscles and Blood Vessels” ( M. Schachter, ed. ), p. 106. Pergamon, London. Smith, C. M., and Walaszek, E. J. (1960). Federation Proc. 19, 264. Stem, P. (1959). Acta Physiol. et Phurmacol. Neerl. 8, 10. Stern, P., and Dobri6, V. ( 1957). Natuwissenschaften 44, 517. Stern, P., and Hukovii., S . ( 1956). Naturwissenschaften 43, 538. Stern, P., and Hukovi6, S. ( 1958). Naturwbsenschuften 45,626. Stem, P., and Hukovib, S. (1960). Med. Erptl. 2, 1. Stem, P., and Hukovik, S. (1961). Intern. Symposium on Substance P, Sarajevo, Yugoslavia, p. 83. Stem, P., and Kocib-Mitrovi6, D. (1958). Naturwbsenschuften 45, 213. Stem, P., and Kocih-Mitrovi6, D. (1961)). Arch. exptl. Puthol. P h u r d o l . Naunyn-Schmiedeberg’s 238,57. Stem, P., and Milin, R. (1959). Proc. SOC. Exptl. Biol. Med. 101, 298. Stern, P., and VukEevib, S. (1W).Acta Allergol. Suppl. 7 , 370. Stem, P., Hukovib, S., and Muaeevii., C. (1956). Naturwissenschuften 43, 162. Stem, P., Dobrii., V., and Mitrovi6-Koci6, D. (1957). Arch. intern. p h u m codynumie 112, 102. Stern, P., Misirlija, A., Dobrik, V., and Kocib-Mirovik, D. (1958). Acta med. ingosl. 12, 153. Stem, P., GaHparovib, I., and Kovai., J. ( 1961). Intern. Syinposium on Substance P, Sarajevo, Yugoslavia, p. 139. Swingle, W. W., Parlow, A. F., Brannick, L. J., and Barrett, W. (1956). Proc. SOC. Exptl. Biol. Med. 92, 594. Trendelenburg, U. ( 1956). J. Physiol. ( London) 132, 529. Umrath, K. (1951a). 2. Vitamin- Hormon- u. Femntforsch. 4, 19. Umrath, K. (1951b). Z. uergleich. Physiol. 33, 457. Umrath, K. ( 195Sa ) . Arch. exptl. Pathol. Pharmakol. Naunyn-Schmiedeherg’s 219, 148. Umrath, K. ( 195313). Arch. ges. Physiol. Pfliiger’s 258, 2.30. Umrath, K. ( 1956).Arch. ges. Physiol. Pfliiger’s 262, 368. Umrath, K. ( 1961). Intern. Symposium on Substance P, Sarajevo, Yugoslavia, p. 23. Umrath, K., and Hellauer, H. F. ( 1948). Arch. ges. Physiol. Pfliiger’s 250, 737. Umrath, K., and Hellauer, H. F. (1949). 2. Vitamin- H m n - u. Fermentforsch. 2, 421. Umrath, K., and Hellauer, H. F. (1951). Deut. 2. Neruenheilk. 165, 409. Umrath, K., and Hellauer, H. (1956). Arch. ges. Physiol. Pfliiger’s 262, 121. Vogt, M. (1954). J. Physiol. (London) 123,451. Vogt, W. (1955). Naturwissenschuften 42, 607. Vogt, W. ( 1958). Arzneimittel-Fmsch. 8, 253.

SUBSTANCE 1’

215

von Euler, U. S. ( 1936a). Arch. ex$. Pathol. Phurmakol. Naunyn-Schmiehberg’s 181, 181. von Euler, U. S. (193613).Skand. Arch. Physiol. 73, 142. von Euler, U. S. (1942). Acta Phiysiol. S c a d . 4, 373. von Euler, U. S., and Gaddum, J. H. ( 1931 ). J. Physiol. (London) 72, 74. von Euler, U. S., and Lishajko, F. (1961). Intern. Symposium on Substance P, Sarajevo, Yugoslavia. ( I n press.) von Euler, U. S., and Ostlund, E. (1956a). Brit. J. Pharmacol. 11, 323. von Euler, U. S., and iistlund, E. ( 195613).Acta Physiol. Scand. 38, 364. von Euler, U. S., and Pernow, B. (1954). Nature 174, 184. von Euler, U. S., and Pernow, B. (1956). Actu Physiol. Scand. 36, 265. von Muralt, A., and Wyss, F. ( 1944). Elelv. Physiol. et Phurmmol. Acta 2, 445. Walaszek, E. J. (1960). Intern. Rev. Neurobiol. 2, 137. Walaszek, E. J., Smith, C. M., and hlinz, B. (1958). Federation Proc. 17, 416. Werle, E., and Kehl, R., and Koebke, K. (1950). Biochem. Z. 321, 210. Wislocki, G. B., and Putnani, T. J. (1924).Anat. Record 27, 151. Zetler, G. ( 1956a). Arch. exptl. Pathol. Pharmakol. Naunyn-Schmiedeberg’s 228, 513. Zetler, G. ( 1956b). Arch. exptl. Pathol. Pharmakol. Naunyn-Schmiedeberg’s 229, 148. Zetler, G. ( 1958). Unpublished results. Zetler, G. ( 1959). Arch. exptl. Pnthol. Pharmakol. Naunyn-Schmiedeberg’s 237, 11. Zetler, G. ( 1961) . Arch. exptl. Pathol. Phurmakol. Naunyn-Schmiedeberg’s 242, 330. Zetler, G., and Ohnesorge, G. (1957). Arch. exptl. Pathol. Pharmakol. NuunynSchmiedeberg’s 231, 199. Zetler, G., and Schlosser, L. ( 1955). Arch. exptl. Pathol. Phurmakol. NuunynSchmiedeberg’s 224, 159. Zuber, H., and Jacques, R. (1962). Angew. Chem. 74,216.

This Page Intentionally Left Blank

ANTICHOLINERGIC PSYCHOTOMIMETIC AGENTS

. .

By 1. G . Abood and J H BieP Departments of Psychiatry and Biological Chemistry. University of Illinois. College of Medicine. Chicago. Illinois

I. Introduction . . . . . . . . . . . . . . . . A. Definition of Term “Psychotoiniiiietic” . . . . . . . B. Scope of Present Review . . . . . . . . . . . I1. The Concept of a Drug Receptor Site . . . . . . . . I11. Chemical Nature of the Psychotoniimetic Anticholinergic Agents . IV. Experimental Methods . . . . . . . . . . . . . A . Evaluation of CNS Effccts of Agent\ in Rats . . . . . . B . Evaluation of Psychotomimetic Potency in Humans . . . . V . Structure-Activity Relationships . . . . . . . . . . A. Piperidyl Glycolates . . . . . . . . . . . . B . Structure-Activity Studies with Piperazine Esters OF Glycolic Acid Derivatives VI . Some Physical and Chemical Factors in Drug Action . . . . A. Spatial-Geometric Considerations . . . . . . . . . B. Steric Hindrance . . . . . . . . . . . . . C. Strength of the Acid Group . . . . . . . . . . D Cationic Charge . . . . . . . . . . . . . E . Drug Interaction with Phospholipids . . . . . . . . VII . Studies with Radioactive Labeled Piperidyl Glycolates . . . . VIII . Behavioral Effect of Anticholinergic Psychotomimetics . . . . A. Gross Effects in Rats . . . . . . . . . . . . B. Sequential Response Device . . . . . . . . . . C . Behavioral Effects of the Anticholinergic Psychotomimetic Agents in Humans D. Neurological-Autonomic Effects of the Psychotomimetic Piperidyl Glycolates . . . . . . . . . . . . . IX . Investigations into the Central hlechanism of Action of the Piperidyl Glycolates A Electrophysiological Studies with Piperidyl Glycolates . . .

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

.

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

.

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

218 218 219 219 221 221 224 224 226 226 227 232 232 233 238 238 238 241 244 244 244 245 247 249 249

‘This research was supported by grants from the Army Chemical Center and the Teagle Foundation. and by a contract with the Office of Naval Research . The authors are indebted to Lakeside Labs., Milwaukee, for supplying many of the agents used . ’Aldrich Chemical Co., Milwaukee. Wisconsin . 217

218

L. G. ABOOD AND J. H. BIEL

B. The Relationship of Hyperactivity to Other CNS Functions of the Piperidyl Glycolates . . . . . . . . . . . C. Temperature Regulation Centers and Activity . , . D. Antagonists to the Piperidyl Glycolates . . . . . X. Clinical Studies with the Glycolate Esters . . . . . A. Piperidyl Glycolates . . . . . . . . . . . B. THA: An Antagonist to Ditran . . . . . . . . . C. Clinical Studies with Piperazinoalkylglycolates . . . . XI. Biochemical and Electrophysiological Studies with Piperidyl Glycolates . . . . . . . . . . . . . . . . . . A. Effect on Enzyme Systems . . . . . . . . B. Metabolic Effects in Frog Nerve and Muscle . . . . . . C. Possible Mechanism for Biochemical Effects . . . . . . D. Effects on the Contractile and Excitable Systems of Frog Sar. . . . . . . . . . . . . . torius Muscle E. Applicability of Piperidyl Glycolates to Studying BiochemicalFunctional Relationships . . . . . . . . . . . References . . . . . . . . . . . . . . . .

251 251 254 256 256 258 259 260 260 261 264 267 271 271

I. Introduction

Interest in drugs capable of producing aberrations or bizarre distortions in human behavior and affect, simulating those seen in schizophrenia and related conditions, has received considerable impetus in the last decade as a result of the introduction of lysergic acid diethylamide (LSD). The concept of a drug-induced model psychosis, however, had been proposed over 30 years ago by Kliiver ( 1928), while the knowledge regarding agents capable of greatly altering psychic behavior has been in existence for at least two millenia. Kluver suggested, as a result of his classical studies with mescaline, that this agent be used to produce a model which would not only provide an experimental approach to the problem of mental disease but also aid in the understanding of brain function. With the introduction of LSD, this concept has been revived and greatly developed. A. DEFINITION OF TERM “PSYCHOTOMIMETIC” Considerable confusion exists in the literature with regard to the nature of the term “psychotomimetic,”and as to how extensively the definition is to apply to drugs acting upon the central nervous system; cf. discussions by Jarvik (1958), Hoch et al. (1953), and Osmond (1957).

ANTICHOLINERGIC PSYCI-IOTOMIMETIC AGENTS

219

For the purpose of this paper, the operational definition used for a psychotomimetic will be: an agent of a relatively high therapeutic index which produces, in the majority of subjects to whom it is administered, transient states, the psychopathological condition of which resembles to a greater or lesser extent those signs and symptoms of severe mental illnesses. The type and variety of agents fulfilling this definition is, indeed, extensive, involving almost any agent acting upon the central nervous system. If, however, the scope were to be limited to agents whose predominant effects were psychotomimetic, while the effective dose was considerably less than the toxic dose, then the number of candidates is greatly restricted. In the search for clues in the pharmacological literature which may lead to the development of new psychotomimetic agents, interest was focused on the naturally occurring anticholinergic agents such as those of the belladonna group. For an excellent review of the central actions of the belladonna alkaloids and related substances, the reader is referred to De Boor, 1956. The diversity and number of anticholinergic agents synthesized has been very extensive; see Bovet and Bovet-Nitti (1948) for review.

B. SCOPEOF PRESENT REVIEW As part of a program to develop effective spasmolytics, a group of piperidyl glycolate esters were synthesized (Biel et al., 1955) in an effort to develop agents exhibiting functional specificity on various parts of the gastrointestinal tract. A number of these agents, upon being reexamined for possible CNS effects, were found to produce hallucinations and other profound behavioral disturbances in humans (Abood et al., 1958, 1959a). With the discovery of the psychotomimetic properties of thcse agents, an extensive program was undertaken in an effort not only to investigate the relationship of psychotomimetic activity to chemical constitution, but to explore their mode of action upon the nervous system from the biochemical, pharmacological, and neurophysiological points of view. Particular emphasis will be placed upon the structure-activity relationships of this class of agents and upon their possible mode of action upon neuronal and other excitable systems. II. The Concept of a Drug Receptor Site

The notion of an antimetabolite or pharmacological antagonist is

220

L. G. ABOOD AND J. H. BIEL

similar in principle to that of the suhstrate-enzyme complex of biochemistry. With Ehrlich's "key and lock" analogy of drug action, it has been possible to formulate clearer concepts. According to this theory, the biological site of attachment or interaction of a drug consists of a configuration analogous to the chemical structure of the drug. A substance such as acetylcholine was presumed to have a corresponding receptor site located at nerve endings and elsewhere in the nervous system. Atropine, on the other hand, exerted its effect by blocking the action of acetylcholine. It is apparent from a comparison of the chemical formulas of atropine and acetylcholine that a structural similarity exists:

w Atropine

CHI

I

CHaNCH2CH202CCH3 I

CH3 Acetylcholine

Both molecules possess a cationic site (nitrogen) and an esteratic site ( C = 0), which constitute the points of attachment at the protein receptor site. The steric configuration of this moiety must conform to certain chemical, as well as spatial, requirements for activity (Pfeiffer, 1948; Schueler, 1953; Abood et al., 1959b). Presumably, anticholinergic agents such as atropine, although complementing the receptor site, are unable to activate it, while at the same time they interfere with the action of the endogenous cholinergic agent (i.e., acetylcholine). As will be discussed in detail later, the action of the anticholinergic agents upon the central nervous system cannot be explained on the basis of cholinergic blockade, and must be considered as acting directly upon certain receptor sites within excitable tissues. Acetylcholine is an ester of an amino alcohol and a relatively weak acetic acid (see Table I ) . If the aliphatic portion of the amino alcohol is increased in length or branched, the cholinergic potency decreases precipitously. If, on the other hand, the amino alcohol is

ANTICHOLINERGIC PSYCHOTOMIMETIC AGENTS

221

held constant, and the acid strength is increased (by increasing the length of the aliphatic acid chain), the ester is no longer cholinergic, but instead becomes anticholinergic. Further increase of the acid strength by the introduction of aromatic groups (tropic) or glycolic acid (particularly with aromatic residues as in benzilic acid) greatly increases anticholinergic potency. Converting the amino alcohol from a quaternary to a tertiary ammonium derivative, now imparts CNS properties to the esters. Dimethylaminoethyl benzilate is a powerful anticholinergic agent with marked CNS properties. Ill. Chemical Nature of the Psychotomimetic Anticholinergic Agents

The agents described in the present review can be classified into the following groups: Group I: esters of substituted glycolic acid derivatives and heterocyclic amino alcohols; group 11: esters of substituted glycolic acid derivatives and substituted piperazino alcohols; group 111: miscellaneous analogs of I and 11. The chemical synthesis of these agents is described elsewhere (Biel et d.,1955,1959). After it had been established that the piperidyl glycolates were psychotomimetic as well as potent anticholinergic agents, a study was undertaken to investigate the interrelationship of chemical constitution to the two major properties of this series. Although the structure-activity relationships regarding anticholinergic potency have been investigated for a variety of structural classes [see Bovet and Bovet-Nitti (1948) for review], until recently no attention has been devoted to either the structural requirements for psychotomimetic characteristics or the relationship of anticholinergic to psychotomimetic properties.

IV.

Experimental Methods

The agents in the present study were examined for their anticholinergic and central properties. Anticholinergic potency was determined by the ability of the agent to block the contraction of the isolated guinea pig ileum by acetylcholine. The results were expressed as the EDs0,i.e., the concentration of the agent necessary to produce a 50% blockade of a contraction induced by acetylcholine. Mydriatic potency was determined by introducing an aqueous solution of the agent directly into the eye of the rat; the results were expressed in terms of the minimal solution necessary to produce the degree of mydriasis recorded. The results were an

ACID

INFLUENCE O F

STRENGTH A N D

TABLE I AMINO ALCOHOL ON CHOLINERGIC. ANTICHOLINERGIC, A N D CNS

~

R-Y-Acid

R

Acid

Relative potencya

Y

Acetic acetic Acetic Isovaleric Allylisopropyl Phenylacetic Tropic Bensilic Beneilic Bensilic Benzilic”

STIMULATION

~~

Cholinergic Anticholinergic

Relative CNS stimulation

2000 200 200

0 0 0 0 0 0 0 0 100 300

400

200

-

1000

1000

-

500

300

100 0.1 0.02

1 10 100

200

0H

Beneilicc

Beneilicc

-

ANTICHOLINERGIC PSYCHOTOMIMETIC AGENTS

223

2.24

L. G . ABOOD AND J. H. BIEL

average of determinations based on three separate rats employing both eyes. A. EVALUATION OF CNS EFFECTS OF AGENTS IN RATS Since hyperactivity or increased motor activity was related to the CNS action of the agents, it was desirable to quantitatively measure this response, employing a device which was suitable to the peculiar activity produced. The instrument used for recording hyperactivity of rats was a modification of that described by Chappel et al. (1957). It consisted of a circular cage suspended from a cantilever beam of spring steel by means of a ball and socket joint. Movements of the cantilever were then electrically recorded by means of pulse counters or continuously by means of a rectilinear ink-recorder. By use of an appropriate gauge of spring steel and by means of an adjustable contact device it was possible to adjust the sensitivity of the device so that practically any type of body movement encountered with the agents could be accurately recorded. The cantilever also provided an excellent dampening system so that the movements of the cage were in phase with those of the animals. Since a wide variety of environmental factors, such as background noise, illumination, and particularly, environmental temperature, were known to influence activity it was found desirable to perform hyperactivity measurements in a controlled climatic chamber (Lipman, 1961). The effect of environmental temperature on the activity of rats with and without a piperidyl glycolate is presented in Fig. 1.

B. EVALUATION OF PSYCHOTOMIMETIC POTENCY IN HUMANS The evaluation of psychotomimetic potency was made largely upon normal human subjects. A large number of psychiatric patients, including schizophrenics, psychoneurotics, and depressives, also received a variety of the agents, but it was often difEcult to evaluate the psychotomimetic action of the agent on such patients, as will be explained later. Although methods for the quantitative psychological evaluation of psychotomimetic effectiveness are available and have been employed on occasions with some of the piperidyl glycolates (Lebovitz et d.,196Oa, b ) , their applicability to the study of a large number of agents would be difficult. Psychotomimetic potency was, therefore, determined relatively on the basis

225

ANTICHOLINERGIC PSYCHOTOMIMETIC AGENTS

5000

c

4000 W *

'' s c

P .*

3000

2000

Y .r(

1000

I

I

10

20

I

I

I

30

40

50

Ambient temperature in "C

FIG. 1. Relationship of environmental temperatures to hyperactivity. Albino male rats weighing 140-180 gm were injected with 5 mg/kg intraperitoneally of N-methyl-3-piperidyl benzilate and the activity measured for a period of 1 hour. (See text for details.) Each value represents an average of six animals and the agreement between any two values was 5%.Lower curve represents control and upper curve the drug-treated animals.

of the evaluation of trained observers consisting of psychiatrists, psychologists, medical internists, and psychiatric nurses. Since most of the agents were tested on no more than three or four subjects and the variation in individual responsiveness to the agents was often considerable, the evaluation was further complicated. By selecting humans with related personalities and vocational interests (nurses, medical and dental students, laboratory personnel, and clinicians), these difficulties could in part be circumvented. In the case of each agent, the evaluation was made independently by two independent observers, and occasionally by as many as four observers. Since the spectrum of behavioral and affective disturbances produced by the agents is extensive, and perhaps without limitations, a number of criteria had to be selected for the purpose of assessing the drug effectiveness. The extent and variety of the CNS effects will be subsequently described (see Tables X and XI later).

226

L. G . ABOOD AND J. H. BIEL

Among the CNS effects in humans included in the evaluation of psychotomimetic effectiveness were the following: extent of confusion, disorientation, hallucinations, illusory experiences, speech difficulties, mood changes, muscle weakness, and other neurological disorders. A scale of 0 to 5+ was used to record each one of the indices; and the final evaluation of effectiveness was based on the sum. Finally, the agents themselves were scored on the basis of an arbitrary scale of 0 to 5+. V. Structure-Activity Relationships

A. PIPERIDYL GLYCOLATES The structure-activity relationships of the piperidyl glycolates are presented in Tables I1 through V. On the basis of these results the following generalizations could be made: 1. R, must be a lower alkyl for potent CNS effects, but not necessarily for anticholinergic potency. In general, it can be stated that with regard to R, substitution the psychomotimetic action correlates very well with the degree of hyperactivity produced in rats. The outstanding exception to this rule was compound BA, where replacement of alkyl by H greatly diminished psychotomimetic potency without appreciably affecting hyperactivity (or anticholinergic) action. 2. R, must be a hydroxyl group for maximal CNS and anticholinergic potency. Compounds with a hydrogen or an isosteric methyl group are inactive centrally while possessing relatively weak spasmolytic properties. In the fluorenyl series, replacement of OH by C1 also greatly diminishes CNS and spasmolytic potency. If the OH group is transferred to the beta position of a cycloalkyl ring, CNS as well as anticholinergic activity was preserved. The presence of a hydroxyl group in R, as well as the beta position of the cyclopentyl group diminishes anticholinergic but not CNS activity (compare A F and AG in Table 11). 3. R, should be a phenyl group, particularly for maximal CNS activity. Replacement of R, by a propyl group greatly diminished psychotomimetic potency without affecting anticholinergic action; however, the decrease in hyperactivity was considerably less. 4. R, must be either a phenyl, thienyl, cycloalkyl, or 1-hydroxy alkyl group. In general, compounds where R, = phenyl and R, =

ANTICHOLINERGIC PSYCHOTOMIMETIC AGENTS

227

cycloalkyl possessed the most potent CNS stimulating properties. If R, is OH and R, is a 1-hydroxycycloalkyl group, CNS activity was markedly diminished. When both R, and R, were replaced by a fluorenyl (diphenylmethylene) group, both CNS and anticholinergic effectiveness were diminished. Of particular interest is the fact that the cyclopropyl ( AR) and cyclobutyl (AS) congeners were almost identical in all of their properties to the benzilate derivative. The most potent and long-lasting psychotomimetic derivative occurred where R, was cyclopentyl (AD ) . 5. The position of the ester side chain or the size of the cycloalkylene imine ring (piperidyl or pyrrolidyl), for the most part did not greatly affect CNS activity, provided that Y was either a chemical bond or a methylene group. Lengthening or branching of the side chain Y markedly decreased both hyperactivity and psychotomimetic effectiveness. The 3-piperidyl esters yielded compounds with the most potent CNS properties, with the 2-piperidyl esters ranking second and the 4-derivatives last in relative effectiveness. 6. Substituents such as chloride, methyl, methoxy, 3,4-dioxymethyl, or phenyl in the phenyl ring of the diphenylglycolates, completely destroyed the psychotomimetic as well as anticholinergic effectiveness of the series. 7. Although only compounds which possessed potent anticholinergic properties (in the range of atropine) were effective CNS stimulants, the correlation between the two activities was incomplete. In the case of R, substitution, a number of potent anticholinergic agents (BA, BD, and BG) were almost devoid of CNS properties (Table 111).

B. STRUCTURE-ACTIVITY STUDIESWITH PIPERAZINE ESTERS OF GLYCOLIC ACID DERIVATIVES In general there were many similarities but some outstanding differences in the structure-activity relationships between the piperazine and piperidine series of esters (Table VI through VIII) . The piperazine esters were usually completely devoid of psychotomimetic activity so that the CNS effects of this group of agents could only be evaluated in terms of hyperactivity produced in rats. The following generalizations could be made regarding the relationship of anticholinergic and CNS stimulating properties to chemical constitution:

TABLE I1 INFLUENCE OF ACID DISTRIBUTIONON ANTICHOLINEBCIC AND CENTRAL ACTIONOF PIPERIDYL GLYCOLATES~

CHs (or CzHs)

Agent*

Rz

AA

OH

AB AC AD AE AF AG AH A1 AJ

OH OH OH

AK ,4L AM AN A0

OH

+ 4

+ @ +

OH OH

4

H H

+

CHa

H c1 OH H CHaCOO

R4

Ra

+ 4 4 4 4

+ ProPYl cyhex wen thienyl 1-OH-cyhex l-OH-cypen 1-OH-cypen 1-OH-cyhex 4 4 4 fluorenyl fluorenyl fluorenyl

EDw vs AChc (1:104)

300 20 392 375 1000

40 111 526 476

10 4 100 0.1 40

Mydriasisd

Activitye

%

mm

mdkg

0.005

5 5 5 6

5 5 5 5

0.001 0.01 0.001 0.001

5 6

10 10 10 10

0.005

4

0.1 0.1

0

0.005 0.005 0.005

5

5 0

5 20 20

C/5'

1000 500

1000 1010 500 500 650

620 1000 110 50

-

-

-

-

0.05

6 1

5

400 340 270

0.1 0.1

4

10 10

Psychotomimetic 4+ 2+ 4+ 5+

3+ 3+ 3 f

3+ 0 0

-

3+ 2+ 2+

AP A& AR AS AT

CHa C1 OH

OH

fluorenyl fluorenyl cypro cybut Atropine

+ +

20 30 250

300 450

0.1 0.1 0.005 0.005 0.001

1 5 5

4 5

10 10 5 5 5

170 200

+ +

800

4+

900 500

4+ 2+

Abbreviations. cypent: cyclopentyl, cyhex: cyclohexyl, cypro: cyclopropyl, cybut: cyclobutyl, 6:phenyl. The agent was administered intraperitoneally, and activity measured 5 minutes later, after a prior 15-minute control period. c EDm, which is a measure of anticholinergic potency, is the dilution of the drug (parts per million) capable of inhibiting acetylcholine-induced spasms by 50%, e.g., a value of “300” refers to the dilution of a one part/million solution. Mydriasis is expresaed as the % solution of agent needed to produce pupillary dilation indicated (in mm) when placed directly in the rat’s eye. Activity is expressed in cage oscillations per 5-minute period. Control value was 50-100. See text for explanation of psychotomimetic evaluation, which was made on human volunteers. a b

INFLUENCE OF

TABLE I11 N-SUBSTITUTION ON ANTICHOLINERGIC AND CENTRAL ACTIONO F PIPERIDYL GLYCOLATES

0

OzCC(OH)+z

EDso v8 ACh

R,

Agrrrt

hlydriasia

(1:lo-s)

%

158 300 200 303 0.2 2.R 313 2.7 57 21 23.8 1.0 7.7 2.0 18.2 14.3 81

0.005 0.005 0.005 0.005 0.1 0.1 0.1 0.1 0.1 0.05 0.1 0.1 0.05 0.1 0.1 0.1 0.1

-

-

-

-

BA BIi BC BD BE BF BO BH BI BJ BK BL 13M BN BO

BQ I3R

ns

BT

Activity

mm Dose 4 5 ' 4 4

10 5

4

5

4 3

10

0

20 10 10 20 10 10 10 10 10 10 10 10 10 10

0 0

0 4 4

0 5 0 0 0 4

-

-

10

627 1000 800 110 113 50 85 116 40 200 170 100 400 80 200 100 600 160 320

Psychotomimetic 2-t 4+

2+ 0 0 0 0 0

f

0 0

-

0

2+ 0

+

TABLE IV INFLUENCE OF AROMATIC RINGSUBSTITUTION ON ANTICHOLINERGIC AND CENTRAL ACTIONSOF PIPERIDYL GLYCOLATES

CH, Mydriasis

Rs

Agent

Re

EDso vs ACh

%,

Activity

mm Dose

c/5'

~

BB CA CB CC CD CE CF CG CH

p-C1 p-CHr ni-CHa o-CH~ p-OCHa

-

-

p-phenyl

-

p-OCH,

m-OCHr CHiOz

3,4-

300 -

0.5 2 20 10 30 15 5

0.005 0.1 0.1 0.1 0.1 0.1

0.1 0.1 0.1

230

4 0

1 1 0 0 1

0 0

Psychotomimetic ~~~

5

10 10 10 10

10 10 20 10

1000 24 125 140 50 40 180 50 30

4+

0 0 0 0 0

0 0

+

TBBLE V INFLUENCE OF RING SIZEA N D POSITION ON ANTICHOLINERGIC A N D CENTRAL ACTIONSOF GLYCOLATES

I

R4

Agent

DS DB BB

DC DE DF DG DH DI DJ DK

R1

R3 9 9 9 9 9 @

9 9 9 9 4 Atropine

R*

n

2 2 2 1 2 1 2 1 2 1 2

Ring Position

3

a 3 2 2 2 3 3 2 3 4

Y

ED50 vs ACh (1: 10-6)

Mydriasis ‘lC

Activity

mm

mg/kg

c/5‘

Psychotomimetic

158 14 350 139 155 348 476 200 260 435 835

0.005 0.1 0.005 0.005 0.01 0.001 0.001 0.001 0.001 0.001 0.001

1 6 4 3 3 6 6 5 4 6 5

10 10 5 5 5 5 5 5 10 5 5

626 150 1000 490 400 1010

800 850 750 900

3+

450

0.001

5

10

800

2+

700

2+

+

4+

a+

+

4+ 5+ 5+

k-

w3

v)

232

L. G. ABOOD AND J. H. BIEL

1. When R was a lower alhyl or benzyl the hyperactivity effect was optimal, although in certain instances a benziloxyethyl substituent had equal influence. When R was hydrogen anticholinergic activity was maximal while the CNS stimulation was minimal. Such N-unsubstituted derivatives were generally effective muscle relaxants as well as potent anticholinergics. The presence of a phenylpropyl group in R greatly augmented anticholinergic potency and CNS stimulation (Table VII). 2. The influence of varying the aliphatic chain length ( Y ) on either anticholinergic or CNS activity was relatively slight, all other things being equal. 3. As in the piperidine series, substitution of a cycloalkyl group in R, greatly enhanced both anticholinergic and CNS stimulating activity. Replacement of a fluorenyl group for the two phenyl groups in the acid moiety generally diminished both anticholinergic and CNS properties. Replacement of phenyl with a 1-OH cycloalkyl substituent likewise enhanced anticholinergic as well as CNS activity. A fluorenyl group in place of a diphenyl resulted in a diminution of both properties. 4. Unlike the situation in the piperidine esters, the addition of a halide substituent to the aromatic rings of the acid moiety did not diminish CNS stimulation in spite of a diminution of anticholinergic potency. Substitution of a fluoride in the 4-position, as in compound EG, resulted in a compound with the most potent CNS stimulation of the piperazine series. 5. The presence of two or more methyl substituents in the piperazine ring tended to decrease both anticholinergic and CNS stimulating properties. 6. Replacement of an aromatic glycolic acid derivative by a benzilamido group diminished both anticholinergic and CNS stimulating properties. As a rule the benzilamido derivatives tended to produce muscle relaxation and counteracted the effect of the stimulating piperazinoglycolates. VI. Some Physical and Chemical Factors in Drug Action

A. SPATIAL-GEOMETRIC CONSIDEIWTIONS An understanding of drug interaction with a biological substance must ultimately deal with the peculiar structural and chemical

ANTICHOLINERGIC PSYCHOTOMIMETIC AGENTS

233

properties of the agent whose significance emerges from structureactivity considerations. With regard to the spatial configuration of the cholinergic molecules Pfeiffer (1948) and Schueler ( 1953) have referred to the importance of the intramolecular distance between the amino alcohol nitrogen and the carbonyl oxygen of the esteratic acid moiety. Both authors suggest that a critical distance between the two active sites is a prerequisite for cholinergic and, consequently, anticholinergic action. For optimal activity this critical distance must be somewhere between the maximal possible distance of about 9A and the minimal of about 3A; the potent anticholinergic numbers in the present series appear to fulfill this requirement. If one were to assume in the case of the piperidyl glycolates that the cationic N and the carbonyl 0 are the points of attachment to the “receptor site,” it would be expected that such factors as the electrical charge characteristics of the reactive groups would influence activity of the agents. In addition there are other factors to be considered such as lipophilic-hydrophilic balance, steric hindrance, other reactive groups, surface active properties, and other types of intramolecular forces. Only a few of the more obvious characteristics will be discussed in connection with the present series. B. STERICHINDRANCE The effect of steric hindrance on the properties of the piperidyl glycolates becomes strikingly evident when substitution takes place in the phenyl rings of the benzilate molecule. Substitution of alkyl or halide groups in the mtho and para positions of the ring restricts the free rotation of the phenyl groups as a consequence of van der Waals’ repulsion [see Ferguson ( 1952 ) for discussion]. One consideration would be the freedom of access of the large aromatic structure as it becomes attached to the receptor site, since it may, because of spatial barriers, interfere not only with attachment of the cationic nitrogen, but with the lipophilic association of the aromatic rings themselves. Coplanarity of the aromatic rings in the benzilate series would tend to favor high biological activity. With regard to the aromatic nucleus, substituents on the phenyl ring could conceivably affect pharmacological properties of the esters by affecting lipid affinity, although the partition coefficient (ch1oroform:water)

TABLE VI EFFECT OF AROMATIC RINGREPLACEMENT A K D N-SUBSTITUTIOX

EDso vs ACh Compound EA EB

EC ED EF EG EH EI EJ

EK EL EM EN EO EP EQ

ER ES

T

R1

(1:

4

4 4 4 4 4 4 cypen cmen 1-OH-cypen cphex a-thien yl 1-OH-cypen 4 4

50.

0.36

4.0 263. 83. 83. 2.0

208.

$

-

4

40 -

+

Mydriasis

Activity

2 8 0 4 0 2

200 550 260 380 100 200

0

100 1000 200 700 700 320 600 150 300 200 350 125

8 0

8 8 3 8

0 8 8 2 4

ET EU EV EW EX EY EZ FA FB FC

FD FE

0.12 2.2 -

4.4

0.50 16.7 14.3

3 3 4

2 4 4 I 0

4 0 G -

300 200 210 100 500 500

150 100 50 75 170 270

TABLE V I I

THEEFFECT OF AROMATIC RINGSUBSTITUTION

A

CH3N

N-Y-R

LJ Compound

Ra

Y

GA GB GC GD GE GF GG GH GI GJ GKb GL

2-Methbenziloxy 2-Methoxybenziloxy 3-Chlorbenziloxy PChlorbenziloxy 2-Chlorbenziloxy 4Fluorbenziloxy 3-Trifluormethbenziloxy 3,4-methylenedioxybenziloxy 1-OH-fluorenecarboxylyl 1-OH-fluorenecarboxylyl

(CH2)3 CzH, CzH4 (CHZ)3 (CH2)3 (CHd3 (CH2)3 (CH2)3 CHzCH(CH3) (CI%h CHzCH(CH3) COCH (OH) CHCHz

GM a b

+ 4

$J 1-OH-cypen

Benz: benzilate. 1-N-Unsubstituted.

EDao vs ACh (1:

Mydriasis

Activity

0.12 0.10 0.10

4 8 8 8

70 180 200 190 360 800 230 250 150 120 90 250 300

-

6.8 2.2

-

-

-

7 8 3 2 0 3 6 0 5

TABLE VIII THEEFFECTOF SUBSTITUTION IN THE PIPERAZINE RINGAND VARIATION IN

THE

ACIDMOIETY

R4NYN-y-R,

80

Rj

EDw vs hCh

Compound HA HB HC HDu HEu HF

HG HHb HI HJ

HK HL HM a

R3 Benzilamido Benzilamido Benzilamido Benzilamido Benziloxy Phenylcyclopentylglycoloxy Benziloxy Benziloxy Benzilamido Benziloxy Benziloxy Phenylcyclopentylglycoloxy 1-Hydroxyfluorenecarboxylyl

Homopiperazine. 1,2,6-Transtrimethyl isomer.

R4 CH3 CH3 CH3 CH3 CHa CHa CH3 CHs CH3 +CH*

H 4(CH2)3 +(CH2)3

1-

R6

2-CH3 -

2-CH3 2-CH3 2-CH3 2-CH3 2-CH3 2-CHa -

-

6-meth 6-meth 5-meth -

-

(1: 106) -

(CH2)3 CHzCH(CH3) 0.02 (CHd3 (CWa (CHzh (CH2)3 80.7 (CH2)3 (CH& (CH2)3 (CH2)3 (CH2)3 (CH2)3 212. (CH2)3 666.

Mydriasis Activity

5E

n

v

0 2 0 0 0

8 0 0 0 2 8 6 8

250 100 200 150 300 900 130 150 200 230 250 170 650

CA

ce n

m

8z

EM =I

n k-

@2 t.3

w

-4

238

L. G. ABOOD AND J. H. BIEL

of such phenyl substituted esters is not appreciably different from that of the unsubstituted congeners. OF THE ACIDGROUP C. STRENGTH

With increasing acid strength, both the anticholinergic and psychotomimetic effectiveness of the amino alcohol-acid esters tend to increase. While acetylcholine, the ester of a weak acid and choline, is actually cholinergic, increasing the acid strength by extending the aliphatic length of the acid chain diminishes cholinergic potency. With an even greater increase in acid strength, as in the case of glycolic acid and, particularly, aromatic glycolic acid, anticholinergic activity becomes manifest. Whether or not the hydroxyl group of the glycolic acid is capable of forming an additional point of attachment to the receptor site (Pfeiffer, 1959) is not certain. It has not been possible, however, to demonstrate intramolecular hydrogen binding between this hydroxyl group and the piperidyl nitrogen in the case of the piperidyl glycolates (Cannon, 1960).

D. CATIONIC CHARGE It is probable that the cationic nitrogen of the piperidine moiety constitutes the main point of attachment of the ester to its active site, which, presumably, is a carboxylic, phenolic hydroxy, or some other available acidic polypeptide group. The magnitude of the charge is of primary importance in determining both psychotomimetic and anticholinergic effectiveness of the piperidyl glycolates. Alkyl substituents in the 2 and/or 4 position of the piperidine ring diminish the charge of N, while at the same time diminishing psychotomimetic and anticholinergic potency. Quaternization of the N, on the other hand, increases the charge of N, but while the peripheral anticholinergic action is augmented, the quaternary derivative is devoid of central action, presumably because it does not enter the central nervous system.

E. DRUGINTERACTION WITH PHOSPHOLIPIDS One of the primary considerations in attempting to' elucidate the mechanism of action of the drug is the nature of the molecular interaction of the agent with components of the biological system in question. There would appear to be little doubt that the site of

ANTICHOLINERGIC PSYCHOTOMIMETIC AGENTS

239

action of the piperidyl glycolates lies somewhere in neural or other excitable tissues. There are a variety of ways in which these agents can interact with various biological substances, such as lipoprotein complexes, nucleic acids, and mucopolysaccharides: by ionic binding; ion-dipole bonds; dipole-dipole interaction ( hydrogen binding). With respect to the glycolates, the possibility of all these interactions with biosystems may exist. 1. Binding Studies with Piperidyl Glycolates

Experimental evidence, particularly that obtained from the use of tritiated materials, suggests that the glycolates interact with the lipid portion of cellular membranes, an observation predictable from the high lipoid solubility and affinity of the compounds within the biological pH range (6-8.5). The glycolates appear to be bound to the cytoplasmic particulates, such as mitochondria and microsomes, where the phospholipid content is relatively high (Abood and Rinaldi, 1959). Insofar as the agents cannot readily be removed from the particulates by a variety of procedures such as treatment with mild detergents, prolonged dialysis, and repeated freezing and thawing, they would appear to be bound rather tightly. Preliminary determinations of the partition coefficient of the agents between chloroform and water bear out the lipophilic property of the glycolates. Furthermore, some degree of correlation obtains between the psychotomimetic effectiveness of the agents and their lipophilic characteristics. In order to investigate the nature of the binding of the piperidyl glycolates, studies were conducted on isolated frog sartorius muscles, using tritiated N-ethyl-2-pyrrolidylmethyl cyclopentylphenyl glycolate (PMCG). Muscles were incubated with gentle agitation in Ringer’s solution at 3 4 O C with varying concentrations of EDTA and phosphate containing 10 pg of labeled PMCG. As the concentration of orthophosphate was increased in Ringer’s solution, the degree of binding decreased rapidly at first, and then gradually leveled off (Fig. 2 ) . A similar curve was obtained with EDTA, except that the rate of decline in hinding was greater than in the case of phosphate. The interpretation of these findings is not entirely clear, but they do indicate that chelating agents interfere with the binding of the piperidyl glycolates. Numerous other chelating agents, such as

240

L. C. ABOOD AND J. H. BIEL

0

0 5

10

15

20

uM

FIG. 2. Effect of orthophosphate and EDTA on binding of tritiated PMCG by isolated frog sartorius muscle. Muscles were immersed in Ca-free Ringer’s solution containing 10 mg/ml PMCG for a period of 30 minutes. After muscles were washed in 100 nil of Ringer’s for 2 minutes, they were soaked in 5% formalin in order to remove the drug. Radioactivity was measured as described elsewhere (Abood and Ftinaldi, 1960). Abscissa is in terms of micromoles or EDTA ( 0 )in Ringer’s solution. orthophosphate (0)

oxidized glutathione and dithiocarbamate, are likewise inhibitory. Evidently, the availability of cations such as calcium and/or magnesium is essential for the binding of the agents, but the manner in which they are involved is not clear at present, It is conceivable that the permeability characteristics of the muscle membrane are dependent upon such divalent cations (particularly calcium), and that chelating agents may be expected to alter permeability. 2. Structure of Membrane of Excitable Tissues Nerve membranes are believed to consist of repeating pairs of bimolecular leaflets with phospholipid molecules sandwiched between monolayers of protein. The ionic groups of the phospholipid (e.g., phosphate and amino) are attached to the hydrated proteins while the fatty acid and other lipophilic moieties are intertwined in ordered configurations (Schmitt et al., 1941; Finean, 1957; Fern&dez-Morh, 1957). The cationic nitrogen of the piperidyl glycolates could be expected to form an ionic bond with the acidic phosphate

ANTICHOLINERGIC PSYCHOTOMIMETIC AGENTS

241

of the phospholipid, while the tail end of the acid moiety containing the highly lipophilic aromatic and cycloalkyl groups would be attracted to the fatty acid moiety of the leaflets. Such agents would probably form a micelle with the phospholipid of the membrane and, therefore, be held firmly, It is probable that the highly organized molecular configuration of the lipoprotein complex is a critical factor in the regulation of the changes in membrane permeability associated with the phenomena of excitability. It is postulated that some ordered structure of the bimolecular leaflet is responsible for the selective accumulation of potassium ions (to the exclusion of sodium) ; so that an agent which is capable of disarranging this organized complex, or interfering with its rearrangement after excitation or depolarization, would be expected to influence the excitable characteristics of the membrane involving sodium-potassium exchange. 3. Possible Mlolecular Interactions of Glycolates

The manner in which the piperidyl glycolates affect the excitable properties of the membrane is rather vague, and the major difficulty, of course, is the lack of understanding of the molecular events associated with excitation. One of the most perplexing problems concerns the mechanisms associated with repolarization of the excited membrane, during which time the ionic balance and molecular rearrangement of the lipoprotein complex are restored. Present-day concepts implicate the role of calcium as a fundamental one in the regulation of membrane permeability associated with excitation [Brink, 1954; see Shanes (1958) for review]. Calcium is presumed to be involved in the critical structural arrangement of the lipoprotein complex, by ionic complexing with the phosphates of contiguous phospholipid molecules. Electrical currents may result in excitation by displacing calcium ions from such sites within the membrane, thereby, influencing the permeability of the membrane to external ions. VII. Studies with Radioactive labeled Piperidyl Glycolatcs

Knowledge regarding the fate and distribution of an agent in a biological organism is essential to an understanding of its mode of action. It is of particular importance to be able to determine the distribution of a neurotropic agent in various parts of the central

2/42

L. C . ABOOD AND J. H. BIEL

nervous system. The availability of isotopic techniques has begun to make such investigations possible. An earlier study (Abood and Rinaldi, 1959) had reported on the distribution of a tritium-labeled piperidyl glycolate ( N-ethyl-3piperidyl benzilate) where the material had been prepared by tritium bombardment of the parent compound according to the method of Wilzbach (1957). Since this material was labeled throughout the molecule where hydrogen is normally present, there was apt to be some spontaneous exchange of tritium for hydrogen in water such as in the case of the acidic hydroxy group. Nonetheless, the material was sufficiently stable to permit a study of the body-wide distribution of the material, its rate of excretion, and, to some extent, its distribution within the central nervous system. More than 95% of the drug was excreted by way of the kidneys within 2 hour after administration. About 0.1%of the total injected material was found in the central nervous system within an hour after injection, the greatest concentration being in such structures as the caudate nucleus, hypothalamus, and corpus callosum. In order to obtain material of greater spec& activity and chemical stability, tritiation was accomplished by reduction of the 3-hydroxypyridine ethyl bromide derivative to the corresponding piperidine alcohol: % : O H

C*Hh+Br3-Hydroxypyridine ethyl bromide

+ HBr

CzHs 3-Hydroxypiperidine

The amino alcohol was esterified with various glycolic acids to form the respective esters. This material possessed the specific activity of 20 millicuries per gram. The incorporation of compound AA ( N ethyl-3-piperidyl cyclopentylphenyl glycolate) in various parts of the rabbit brain was studied after rabbits were killed at various times, after receiving a dose of 15 mg/kg of the tritiated material intravenously. Within % hour after injection the greatest concentration of the material was found in such regions as the caudate nucleus, medial thalamus, corpus callosum, and neocortex, with the greatest activity in such areas as the pyriform cortex, tegmentum, and lateral thalamus (Table IX) . The activity after 1 hour was, in general, slightly lower than it was at the ? hour i point; but in some

243

ANTICHOLINERGIC PSYCHOTOMIMETIC AGENTS

TABLE 1X DISTRIBUTIOX OF TRITIUM-T~AREI,EI>N-ETIIYL-3-PIPERIDYL CYCLOPENTYLPHEXYL GLYCOLATE I N VARIOUS PARTS OF RABBITBRAW Activity (cpm) after:

3.i Hour

1 Hour

3?.i Hours

Telencephalon Anterior neocortex Posterior neocortex Pyriform cortex Hippocampus White matter Corpus callosum Corpus striaturn Caudate nucleus

1071 1027 23 1 668 318 1135 874 1197

832 757 2545 122 968 207 247 775

25 35 47 21 124 39 10 83

Diencephalon Lateral thalamus Medial thalamus Hypothalamus

242 1215 414

295 1309 188

170 66 74

MesencephaIon Tegmentum Foot of midbrain

30 1 358

120 180

56 97

642

351 268 125

30 134 125

Brain part and derivative

Rhombencephalon Cerebellum Pons Medulla

50'2

449

a Rabbits were given 15 mg/kg of agent by slow intravenous infusion (3 minutes) and killed at time indicated. The brains were then perfused with 50 ml of saline through the carotid artery, after puncture of the jugular vein. The tritiated material had an activity of 3000 cpm/fig.

structures, such as the corpus callosum, and corpus striaturn, the activity decreased rather sharply. After 3 hours, the activity in most areas of the rabbit brain was but a small fraction of the activity after !k hour, although such areas as the cerebral white matter, lateral thalamus, pons, and medulla still exhibited relatively high activity. The tritiated material was found to be present largely in the particulate fraction, containing predominantly mitochondria, when the brains of animals receiving the drug were fractionated and the particulates separated by conventional procedures. When this mito-

244

L. G. ABOOD AND J. H. BIEL

chondrial particulate fraction was further resolved by gradient centrifugation in hypotonic sucrose, it could be shown that the labeled material was partly associated with the true mitochondria and to some extent with fragmented structures which under the electron microscope appeared to consist of osmophilic double membranes. Whether this latter material was a fragment of neuronal and other processes of the central nervous system is unknown. Some preliminary studies have been done with the distribution of acetylcholine in these same particulate fractions, and, to a great extent, there was a pattern of distribution similar to the piperidyl glycolates. This finding would indicate that the piperidyl glycolates exhibited an afsnity for much the same receptor sites as acetylcholine. VIII. Behavioral Effect of Anticholinergic Psychotomimetics

A. GROSSEFFECTS IN RATS One of the chief difficulties confronting the investigator in working with psychotomimetic agents is the evaluation of their behavioral effects upon animals. Eventually, the development of operant procedures utilized by the experimental psychologists will undoubtedly provide the solution. For the present, at least, these procedures are not of predictive value and serve only to corroborate what is observable with a psychotropic agent in humans. This class of agents produces in rats a number of gross behavioral changes which are evident to the trained investigator. At a dose of 5-10 mg/kg intraperitoneally in rats, a centrally active piperidyl glycolate produces the following gross behavioral changes: head bobbing, head swaying, increased activity, jerky and hesitant body movements, exploratory head movements while standing on hind limbs, hyperreadivity, hypersensitivity, and spontaneous squealing. While at lower doses (1-10 mg/kg) there is generalized increased body tone, at larger doses (10-30 mg/kg) there is a marked decrease in body tone, accompanied by hind limb prostration, generalized muscle weakness, and ataxia.

B. SEQUENTIAL RESPONSEDEVICE After unsuccessfully employing a wide variety of objective behavioral and operant procedures in an effort to measure behavioral changes in rats which would be comparable to the psychotomimetic reactions in humans, a device for evaluating sequential

ANTICHOLINERGIC PSYCHOTOMIMETIC AGENTS

245

response was developed and described in detail (Polidora and Abood, 1961). The technique is based upon the proposition that one of the most advanced and critically controlled response patterns in animals and man is the chain of responses or decisions made sequentially, both in time and space. It was reasoned that since such delicate behavioral patterns can be readily influenced by external circumstances (distractibility) they may also be liable to endogenous ones, such as those produced by a psychotomimetic agent. The automated testing apparatus consisted of a cylinder one meter in diameter and containing four equidistant response locations to which the animal made responses sequentially so that a reward was automatically made available at the fourth. Any possible combination of these three locations can be required, thereby permitting a regulation of the degree of complexity of the task. It was possible to disrupt a learned sequential response in a given rat with doses of the drug as low as 0.25 to 0.5 mg/kg intraperitoneally-doses which, in themselves, barely produced any outward changes in behavior, such as the characteristic head movements and hyperactivity. The degree of disruption produced by a given drug was a function of both the dose and the complexity of the sequence to be performed. Thus, an animal (which had previously learned and performed a given series of sequential responses), on being given n dose of 1 mg/kg, performs normally when performance is contingent upon it relatively simple sequence: when given a similar dose but recjnired to perform a relatively complex sequence, the animal performs us if completely disorganized. Other agents, such as chlorpromazine and CNS depressants in general, will interfere with performance in ‘1 comparable situation, but unlike the case with the anticliolinergic agents, where the animals are hyperactive and make many unsuccessful attempts at performance, the animals under the influence of the depressants are generally inactive and make few, il my, attempts at a correct response. This type of behavioral apparatus offers considerable potentiality for the evaluation of tlw action of psychotomimetic agents.

c. BEHAVIORAL EFFECTS OF TIII.1 A 1 ’ 1 ICIIO1,INERGIC PSYCIIOTOMISrEl IC AGENTSIN HUMANS The behavioral effects observed with the piperidyl glycolates in human subjects are extremely diverse and complex, depending not

246

L. G . ABOOD AND J. H. BIEL

only on the dose of the agent used, but on the particular psychological idiosyncracies of the subject. The psychological-behavioral effects are described in Table X, where the type of response is given,

PSYCHOLOGIC.4L-BEHAVIORAL

Type

TABLE X EFFECTSOF ANTICHOLINERGIC HALLUCINOGENS Description

Stupor

Spatial and temporal awareness impaired; slowed mental functions; slowed responsiveness; disinterest; incoherence; somnolence

Deleriuni

Confusion assoc,iated with restlessness; illusions; apprehension; bewilderment; anxiety

Hallucinations

-4uditory; visual; tactile; olfactory; gustatory

Confusion

Spatial, temporal disorientation; misinterpretation of questioning and environment in general

Emotional

Negativistic; sudden anger and other emotional outbursts; paranoid reactions; occasional anxiety, fear, and evcn panic in highly emotional individuals

Amnesia

Lack of recall of psychotomimetic experiences

Dreamlike states

Flash-back memory; aimless wandering

Aphasia, agnosia

Senseless, rambling talking; loss of trcnd of thought; complete aphasia

Apraxia

Impaired motor, amnestic, ideational, and idcomotor abilities

Intellectual impairment,

Reduced speed and accuracy in problem solving; inability to understand questioning

with a description of thc symptomatology associated with it. Investigators, such as Meduna and Abood (1959) and Hoffer and Osmond ( 1960), have characterized the psychotomimetic action of these agents as a delirium since many of the psychotomimetic mamifestations seem to be associated with a delirium or state of marked confusion and disorientation. The nature and characteristics of the hallucinations produced by the piperidyl glycolates are among the most dramatic and interesting effects of the agents. Although the hallucinations are predominantly visual, they are frequently auditory, tactile, olfactory, and gustatory. The visual hallucinations usually involve people, animals, and in general clearly defined

ANTICHOLINERGIC PSYCHOTOMIMETIC AGENTS

247

objects. Very often, subjects see a variety of colors such as those seen in brightly colored, ornate oriental rugs or in vivid oil paintings. There are frequently illusions and distortions of visual images: a small spot on the wall, for example, may suddenly appear to the subject as moving and take on characteristics of some known living object, such as an insect, animal, or even a human face. The nature of the visual hallucinations may be understood better in terms of the so-called “hallucinatory constants” ( Kluver, 1942) which are assignable to three levels: ( a ) the level of “form constants,” where objects of definite shape are envisaged; ( b ) the level of alterations in number, size, and shape (polyopia, dysmegalopsia, dsymorphopsia); ( c ) the level of changes in spatio-temporal relations. Kluver regarded these constants as being applicable to all types of hallucinations; and their manifestations were particularly common with the piperidyl glycolates. Auditory hallucinations usually involve musical instruments such as drums or violins; and frequently the subject will insist that they hear a radio playing music, either nearby or at a distance. Subjects often hear voices, usually familiar ones, and will carry on extended conversations with such phantom voices. As the patients become increasingly more confused and begin losing contact with the environment, they often begin to react with their hallucinations and are unable to distinguish reality from phantasy. They may often pick up phantom objects, such as cigarettes, sandwiches, cups of coffee, and glasses of beer, and actually perform the appropriate oral movements associated with the utilization of such objects. An alcoholic subject will behave as if he were having delirium tremens. He may pound on the table vociferously demanding an alcoholic beverage and actually begin to drink from imaginary glasses. Some subjects will often describe tastes and odors associated with such phantom experiences and later, after recovery from the effects of the drug, may recall the experience and describe it vividlv. As a rule, however, when a subject becomes so confused that he -loses contact with his environment, amnesia sets in so that he is able to recall only a small fraction of the total experience.

D. NEUROLOGICAL-AUTONOMIC EFFECTSOF PIPERIDYL GLYCOLATES

THE

PSYCHOTOMIMETIC

The autonomic effects associated with the agent typify what is seen with the anticholinergic agents in general, such as atropine.

248

L. G. M OOD AND J. H. BIEL

The neurological, like the psychological-behavioral effects, are complex?depending on the dose and, in some instances, on the particular subject. In Table XI the neurological effects are classified under TABLE XI NEUROLOGICAL-AUTONOMIC EFFECTS OF AXTICHOLINERGIC HALLUCINOGENS Autonomic (peripheral)

M ydriasis Dryness of mouth Skin flushing (hyperemia) Tachychardia, hypertension

Motor-somatic

Slowness of speech Slowness of gait Festination Tremors Rigidity

Sensory (thalamic)

Astereognosis Ataxia Loas of tactileand-thermal sensations Over-reactive to pain

Telencephalic

Hypotonia Fine tremor Somnolence Excitemci~t~ Aphasia Apraxia Agnosia (auditory and visiial) Anarthria, dysarthria Dizziness Nystagmus Restlessness

motor-somatic, sensory ( thalamic) and telencephalic categories. These effects in some individuals are extremely severe and prolonged, lasting often 2 to 4 days.

E. PSYCHOLOGICAL STUDIESIN HUMANS In an attempt to quantitate the effects of the piperidyl glycolates, a variety of psychological tests were given to normal human subjects to whom the drug was administered (Lebovitz et al., 1960a, b ) . Medical students were used in this series, and the response to N -

ANTICHOLINERGIC PSYCHOTOMIMETIC AGENTS

249

ethyl-3-piperidyl benzilate ( NPB ) was compared with the response to other psychotomimetic agents such as LSD. Among the various tests used were the Minnesota Multiphasic Personality Inventory (MMPI), Jarvik questionnaire, Clyde Mood Scale, and the Rorschach test. According to the MMPI, both NPB and LSD induced a pessimistic outlook which affected self-evaluation, self-trust, and psychological functioning; in addition, LSD produced irritability and brooding. NPB elicited greater preoccupation with physical functioning and body, and tended to produce a condition characterized by a decreased impunitive attitude. Both drugs induced feelings of alienation, depersonalization, lack of “ego” mastery, confusion, and disorganization. In essence, although differing in the magnitude of their effect, both drugs induced almost identical profiles on the MMPI. With regard to the classical Rorschach test, the DeVos scores were within the normal range, but were significantly correlated with the amount and kind of psychopathology exhibited on the tape interviews. Anxiety and body preoccupation responses were significantly related to the behavioral effects produced by NMP, whereas hostile and neutral responses were related to LSD effects. NMP induced a significant decrease in the Earnest, Contented, Energetic, and Friendly scales of the Clyde Mood Scale, and a significant increase in the Fearful and Desperate scales. The Adventurous scale scores showed a significant increase 3 to 9 days after administration of NMP as compared with scores during the drug period. All the other aforementioned scales tended to be higher in the post-drug repetition than in the pre-drug assessment. These changes chronologically paralleled a hypomanic state which began shortly after the administration of NMP. IX. Investigations into the Central Mechanism of Action of the Piperidyl Glycolates

A. ELECTROPHYSIOLOGICAL STUDIESWITH PIPERIDYL GLYCOLATES Although there is extensive literature on the electroencephalographic (EEG) effects of atropine and related derivatives (Wikler, 1957), the attempt to correlate these changes with the behavioral effects of the drug has been somewhat confusing and not very rewarding. Many years ago, a close correlation was demonstrated

350

L. G . ABOOD AND J. H. BIEL

between EEG patterns and behavior manifestations of sleep and wakefulness (Gibbs and Gibbs, 1941), low-voltage fast activity being generally associated with wakefulness, while liigh-voltage slow activity was related to sleep. The EEG pattern of an atropinized dog exhibited high-voltage slow waves accompanied by “spindle bursts,” whether the animal lay quietly with closed eyes or was awake and trying to escape. As a result of this disagreement between EEG sleep patterns and overt behavior, the term “dissociation” was coined. More recent studies have shown that atropine decreased the EEG “arousal response” elicited either by sensory stimulation or by stimulation of the reticular formation (Rinaldi and Himwich, 1955; Bradley and Elkes, 1957). Again, the gross behavioral response to such stimuli was unaffected by atropine. It appears from studies by these investigators that modification in the EEG response to sensory afferent stimulation induced bv atropine could be attributed to depression of the mechanism responsible for the maintenance of spontaneous cortical activity, upon which the ascending reticular activating system presumably acts. The depression of the EEG arousal response may result from the action of atropine at lower levels of integration in the reticular system occurring independently of the effects modifying spontaneous EEG rhythm. Working with a variety of anticholinergic compounds, including NPB and N-methyl-3-piperidyl benzilate, in psychiatric patients, Fink (1960) attempted to correlate EEG changes with behavioral and other pharmacological effects. With the occurrence of hallucinations, fantasies, illusions, or tremors, the existing EEG pattern tends to shift in the direction of desynchronization. With the centrally active anticholinergic agents a correlation existed, while in the case of centrally weak anticholinergic agents, such as atropine, no such association was apparent. At the height of the EEG desynchronization and the marked CNS disturbances, intravenous doses of chlorpromazine tended to reverse both the EEG and behavioral effects. The observations are in conformity with the concept that shifts of the EEG pattern in the direction of desynchronization occurred in association with anxiety, hallucinations, fantasies, etc., and in the direction of synchronization with symptoms such as euphoria, relaxation, or drowsiness.

ANTICHOLINERGIC PSYCHOTO\fIhIETIC AGENTS

251

B. THERELATIONSHIPOF HYPERACTIVITY TO OTHERCNS FUNCTIONS OF THE PIPERIDYL GLYCOLATES The most characteristic, noticeable effect of the piperidyl glycolates on overt behavior is the hyperactivity response which seems to be associated both qualitatively and quantitatively with the psychotomimetic actions of the drug in humans. This correlation between hyperactivity and psychotomimetic effectiveness, however, is applicable only to the piperidyl glycolate series and does not apply to other chemical groups, such as the piperazinoglycolates. An extensive study has been made as to the relationship of hyperactivity response to other effects of the drug on the autonomic and higher central nervous system ( Lipman, 1961 ) .

c. TEMPERATURE

REGULATION CENTERS .\ND ACTIVITY It is known that regions concerned with the regulation of body temperature are limited in the posterior part of the lateral hypothalamus, while those centers involved in heat loss functions, such as sweating and panting, are situated in the preoptic region with pathways from this center extending backward through the lateral hypothalamus ( Ranson and Clark, 1959). The center for preventing heat loss by vasoconstriction and for increasing heat production by shivering is situated in the hypothalamus proper, and its descending pathway also runs backward through the lateral hypothalamus. If drugs, such as the piperidyl glycolates, were to interfere with the hypothalamic vasoconstrictive mechanisms, then the ability of the animal to prevent heat loss would be impaired. The marked vasodilation produced by the glycolates, which is presumably of central as well as peripheral origin, may result in excessive heat loss (Lipman, 1961). The increased motor activity resulting from the drug may be compensatory in an effort to offset the increased heat loss due to vasodilation. Some of the other heat loss mechanisms other than vasodilation that may be influenced by the piperidyl glycolates are respiratory and heart rates, piloerection, and sweating. Whereas a control animal will show a diminution in spontaneous activity as the environmental temperature is increased from Oo to 4OoC, Lipman (1961) has demonstrated that an animal receiving an anticholinergic psychotomimetic exhibits an elevated but constant

252

L. G . ABOOD AND J. H. BIEL

rate of activity between Oo and 2OoC, which, thereafter, falls rather precipitously as the temperature is increased to 3OoC (see Fig. 1). Since the colonic temperature of the animal decreases gradually in both the drug-treated and control animals from Oo to 21°C, it would appear that in the case of the drug-treated animals there was no correlation between environmental and colonic temperature. Furthermore, at a given environmental temperature there was no significant decrease in the colonic temperature of a given animal before and after the administration of a piperidyl glycolate. It was concluded from these studies that the environmental temperature was a more important factor in determining body temperature than was the general activity; furthermore, despite the large increases in activity there was an actual decrease in the colonic temperature with decreasing environmental temperature. The relationship of activity to environmental temperature, in the absence of the drug, correlates well with the relationship of basal metabolism rate to environmental temperature; however, this correlation does not seem to obtain in the drug-treated animals at the temperature range of 0 to 21OC. Evidently in this temperature range, internal compensatory mechanisms such as piloerection, huddling, and other autonomic responses may offset heat loss owing to lowered temperature. By recording cerebral temperatures with implanted thermocouples, it could be shown that a rise in cerebral temperature accompanied increased skeletal muscle activity resulting from the electrical stimulation of cats (Chatonnet et al., 1960). The most consistent correlation of body temperature with activity was observed when the thermocouples were inserted into the anterior region of the hypothalamus. It was also possible to demonstrate that a drop in the brain temperature resulting from artificial cooling led to a drop in the skin temperature of dogs as a result of vasoconstriction. It was concluded that the temperature-sensitive regulator in the hypothalamus is able to maintain body temperature by direct regulation of muscle activity and by means of autonomic control of peripheral structures. It is known that numerous agents are capable of exerting a direct action on this thermoregulatory system of the hypothalamus. One of the areas in the rat brain showing a large concentration of piperidyl glycolates is the frontal lobe, an area which is known to be involved in the regulation of hypothalamic functions. Lesions

ANTICHOLINERGIC PSYCHOTOMIMETIC AGENTS

253

in the orbital surface of the frontal lobe (area 13) have been known to result in marked disturbances in motor activity in man and monkeys (Bailey and Sweet, 1940; Langworthy and Richter, 1939; Ruch and Shenkin, 1943). Area 13 has been regarded as an autonomic sensory zone, insofar as stimulation of the central end of the vagus nerve gave rise to electrical disturbance in this part of the brain. Direct stimulation of this region resulted in autonomic discharge characterized by a rise in blood pressure and altered tonus in the gastric musculature. This region has also been shown to exercise a control over respiratory movements, while stimulation of this area in dogs or monkeys caused thermal responses resulting in as much as 3O to 5°C rise in the temperature of the extremities (Delgado and Livingston, 1948). Animals with frontal lesions may show so much ceaseless movement that they literally run themselves to death (Langworthy and Richter, 1939) while removal of area 13 without destruction elsewhere gave rise to hyperactivity in monkeys (Ruch and Shenkin, 1943). It is of particular interest that another area with a relatively large concentration of piperidyl glycolates is the corpus callosum. A variety of disturbances, such as apathy, drowsiness, and memory defect, are known to be associated with the so-called “corpus callosum syndrome” which results from tumor-induced destruction in the cingulate gyrus in man (Voris and Adson, 1935). Furthermore, tumors in the general area of the corpus callosum produced marked changes in personality with disorientation and confusion. Of further interest is the large concentration of drug in the approximate area of the dorsal medial nucleus of the thalamus which comprises a relay system between the hypothalamus and the frontal cortex (Meyer et d.,1947). There is actually a point-topoint relationship between different parts of the frontal cortex and corresponding regions of this thalamic nucleus. It has been said that the greater part of the cortex of the frontal lobe must be regarded as a projection area receiving products of activity of the hypothalamus, in much the same way that the visual cortex is the projection area for retinal activities. In turn, the frontal cortex sends projections to the hypothalamic nuclei and into the brain stem. These studies involving the relationship of area 13 of the frontal lobe to the hypothalamus and other components of the autonomic nervous system have led to the suggestion that the somatic and

254

L. G . ABOOD AND J. H. BIEL

autonomic functions are, to a considerable extent, involved in the maintenance of homeostasis of the organism (Green and Hoff, 1937; Kaada, 1951). The central effects of the piperidyl glycolates strongly suggest that this kind of interrelationship between the higher CNS and hypothalamic-limbic system is functional.

D. ANTAGONISTS TO THE PIPERIDYL GLYCOLATES Apart from the intrinsic interest in developing drugs capable of producing a model psychosis, there are a number of other reasons for pursuing such a project. One of the interesting facets of this type of study is related to the synthesis of analogs which might act as antagonists to the psychotomimetic effects of the agents. If such antagonists would eventually be of value in the treatment of certain aspects of psychoses, the concept of a drug-induced model psychosis could be of considerable theoretical as well as practical interest. Within the past decade, the rapid development of the area of psychotherapeutic drugs has made it increasingly imperative that rational techniques be developed for designing such agents. Since the piperidyl glycolates were much more effective psychotomimetic agents than agents such as LSD and mescaline, it was felt that they would more effectively serve as a model for the development of possible therapeutic agents. Since the psychotomimetic potency of the piperidyl glycolates was related to the hyperactivity produced by these agents in rats, i t might be possible to employ the hyperactivity response as a screening device for antagonists. A number of existing psychotropic agents were therefore tested for their ability to block this hyperactivity response of an agent such as N-methyl-3piperidyl benzilate (Table XII). The procedure was to administer a given dose of the test drug 5 minutes befo,re the administration of 5 mg/kg of the glycolate ester. The more potent phenothiazines were found to almost completely block the hyperactivity response, while other muscle relaxants and tranquilizers such as meprobamate, reserpine, and piperidine were a little less effective. Chlordiazepoxide was an effective blocking agent at a dose of 5 mg/kg. A number of the piperazinoalkylglycolates known to produce muscle relaxation in animals were also found to be antagonistic. (Tetrahydroaminoacridine ( T H A ) , in doses known to antagonize the central actions of the piperidyl glycolates, was also effective in blocking the hyper-

ANTICHOLINERGIC PSYCHOTOMIMETIC AGENTS

255

rr.m,I+:XII HYPERACTIVITY 1 1 RATS VARIOUSNEUROPHARRIACOLOGICAL AGENTS=

h T A C 0 S I S R l OF

BY

Activity antagonism

(%I Chlorpromazine Fluphenazine Perphenazine Trifluoperazine Prochlorperaxine Reserpineb Meprobamate Piperidine Chlordiazepoxide

HR HE ED d-Tu bocurarine d-l'ubocurarine Phenobarbital" Primidone Control

10 10

10 10

I0 5 100 100 5 10 10 10 0.1 0.2 20 25 -

83 80 85 90 83 50 80 88 75 40 20 20 52 90 20 25 -

a Test agent was administered (intriiperitoiically) 5 minutes before 5 mg/kg of N-methyl-3-piperidyl benzilate, which alone gave a value of 1000 oscillations prr 5-minute period. * Compound given 30 minutes aftcbr test drug.

activity response. Soporifics and nonselective depressants, such as the barbiturates, had a relatively slight effect as antagonists. Although d-tubocurarine appeared to be an effective agent, its antagonism was directly related to its paralyzant action. It would appear from this study that such a screening device might be of considerable use in the development of psychotherapeutic agents. Without exception, the more powerful tranquilizers or selective CNS depressants were capable of antagonizing hyperactivity in rats. The mechanism by which this drug antagonism is produced is obscure at present, but it obviously is of central origin. It would appear as if some aspects of the hyperactivity response of the piperidyl glycolates are related to the psychotomimetic effects of the agents, although, as indicated previously, the two are not directly related.

256

L. G. ABOOD AND J. H. BIEL

X. Clinical Studies with the Glycolate Esters

A. PIPERIDYL GLYCOLATES In the course of investigating the psychotomimetic potency of the various glycolate esters in normal human subjects, it was observed that some of the subjects developed a change in their basic mood and drive within a period of a few days following a single dose of the agent ( Abood and Meduna, 1958).The newly emerging modulation of mood could be described as slightly hypomanic and was further characterized by a feeling of euphoria, increased drive, and more aggressiveness. These observations indicated that the drug might be useful in the treatment of psychiatric states in which the outstanding symptom is a depressed mood ( Abood and Meduna, 1958; Meduna and Abood, 1959). Although this property of the glycolates was believed to be a general one for the more potent members of the series, it was particularly noticeable with Ditran. Although originally believed to be N-ethyl-3-piperidyl cyclopentylphenyl glycolate ( I ) , Ditran later was shown to contain 70%N-ethyl2-pyrrolidylmethyl cyclopentylphenyl glycolate ( 11):

uozcQ OHQ

OHQ

L, N J c H z o z c Y ~ , C2H5

C2H5

(1)

(11)

Ditran was not only one of the most potent psychotomimetic members of the series but it had a very prolonged action, which in some subjects was apparent after 3 days. In collaboration with Professor Meduna, the drug was administered to a series of depressed psychiatric patients in single oral doses ranging from 10 to 20 mg. About 75%of these patients responded with the normal psychotomimetic symptoms, including hallucinations, confusion, disorientation, and a general dreamlike state lasting for periods of 12 to 48 hours. In most of the patients who responded in this manner, there followed a period of elation and increased activity lasting for a period of days. In over two-thirds of these patients, many signs of the depression disappeared either immediately or within a period of

ANTICHOLINERGIC PSYCHOTOMIMETIC AGENTS

257

a few weeks. Since all of these patients had received almost every known form of treatment, including electroshock therapy, and a variety of antidepressant drugs, without effect, there was no question but that the improvement observed was directly attributable to the drug. Finkelstein (1961) conducted an intensive study on a group of 100 psychiatric patients which included depressives as well as schizophrenics and psychoneurotics. In many instances, electroshock therapy was combined with Ditran. Most of the depressed patients responded effectively to treatment with Ditran alone, while psychoneurotics showed sufficient improvement so that they responded more freely and with greater accessibility to psychotherapy. In many instances it was observed that ego-alien, obsessive-compulsive ideas lost their significance, while neurotic patients with dissociative reactions gained insight and appeared to be more immune to anxiety and tension situations. In those patients who did not respond to a single dose of Ditran, repeated administration combined with electro-shock therapy was applied with considerable success. English (19f3l) conducted a comparable study on a group of some 40 patients and noted that the majority of patients responded to a single dose or to repeated administration of the drug. Gershon (1960) administered this same agent in doses of 5 to 10 mg orally per day to chronic schizophrenics over a period of 2 to 4 weeks. As would be expected, during the period that the patients received the drug, there was an exacerbation of most of their psychotic-like symptoms. Among the effects observed were increased withdrawal, apathy, asocial behavior, and inactivity, while patients who were previously on work assignment were unable to resume them. These effects were particularly noticeable in paranoid schizophrenics. Upon chronic administration of the agent, tolerance to the autonomic effects developed rather quickly, while tolerance to the central actions was relatively slower and in many patients not apparent. As part of a greater study to explore the physiological and psychodynamic differences between potentially “labile” and “fixed schizophrenics, Apter (1957) had employed a wide variety of autonomic and psychotherapeutic agents. Ditran was administered in single doses to 10 female and 10 male chronic schizophrenics ranging all the way from extremely labile to extremely fixed reactors

258

L. G . ABOOD AND J. H. BIEL

(Apter and Abood, 1962). Analysis of the data of such a study led to the following conclusions: The range of responsiveness to the anticholinergic or autonomic effects of Ditran in chronic schizophrenic patients is limited when compared to the responses of neurotic patients and “normals.” The more labile of the nonschizophrenic patients respond to relatively smaller amounts of the drug in a manner similar to the normal controls, while the “fixed” reactors respond least like the controls. Furthermore, in the fixed subjects there appears to be a dissociation between the intensity of autonomic and central responses to the drug. It appeared from this study that a good correlation existed between the responsiveness of a given schizophrenic to Ditran and the general lability of his illness. As a rule, the more labile patients responded more effectively to psychotherapeutic agents and other therapeutic measures.

B. THA: AN ANTAGONISTTO DITRAN In an effort to develop antidotes or antagonists to the piperidyl glycolates, Gershon (1960) tested a variety of possible agents in schizophrenic subjects at the height of a psychotomimetic episode produced by Ditran. After failing to influence the psychotomimetic or autonomic effects of Ditran with such agents as succinate and a variety of central depressants, including phenothiazine tranquilizers,

l,‘,:~,~-Tetmhydr0-5-amiiioacriclint.[TIIAI)

1,2,3,4-tetrahydro-5-aminoacridine( THA ) was tried. Upon the administration of 60 mg of THA by slow intravenous injection at the height of a mild or severe psychotomimetic episode, it was possible almost to interrupt the episode, while at the same time reversing the peripheral autonomic effects of the drug. On the other hand, it was not possible to influence the mild psychotomimetic and other central actions of agents such as LSD and 1-(l-phenylcyclohexyl) piperidine ( Sernyl) . The antidotal action of THA, therefore, is quite specific. Pharmacologically, THA is an analeptic, which has found use as an antagonist to barbiturates and morphine, and is also a decurarizing agent (Shaw et al., 1957; Feldberg, 1955). It also

ANTICI-IOLINERGIC PSYCHOTOMIMETIC AGENTS

259

possesses marked anticholinesterase properties. Insofar as other more potent anticholinesterase agents such as physostigmine, prostigmine, and diisopropylfluorphosphate do not reverse the effects of the piperidyl glycolates, it would appear that the anticholinesterase effect is not directly related to its antidotal action.

C. CLINIC 4~ STUDIESWITH PIPERAZINOALKYLGLYCOLATES One of the objectives in developing a wide variety of piperazine glycolate homologs was the finding of a psychotherapeutic agent which possessed the CNS stimulating properties of the piperidyl glycolates but not their psychotomimetic actions. An attempt was made to introduce structural alterations into the piperazinoalkylglycolates in order to accentuate their muscle relaxant property, since it had been observed in the piperidine series that certain structural variations resulted in the development of powerful muscle relaxants. In the course of the last five years some piperazinoalkylglycolates have been subjected to clinical trial in a variety of psychiatric patients, including depressives, psychoneurotics mentally retarded, and schizophrenics of all types. One of the very early studies conducted by Jackman ( 1958) with l-ethyl-4-~-piperazinoethylbenzilate on a series of psychoneurotics indicated that this type of compound might be of benefit in the treatment of some types of psychoneurotic and depressed patients. Various congeners of this compound, including l-methyl-4-( 7-piperazinopropyl ) benzilate and 1,4-bis-( p-benziloxyethyl ) -piperazine have been tried by Jackman and have proved to be similarly effective. This group of agents appeared to have not only a calming and sedative action on disturbed psychoneurotics, but an antidepressant effect as well. An extensive study on a group of 32 severe chronic schizophrenics was conducted by Tourlentes ct al. (1960) with JB 8035 ( 1,2-dimethyl-4-~-piperazinopropylbenzilate). Other forms of therapy, including the newer psychotropic drugs had failed to produce any lasting salutary effects in this group of patients. All of these patients were hospitalized for a period of about eight years and were characterized by an idle apathetic regressive ward adjustment. This study with the test agent was conducted on a triple-blind double-placebo rotating medication schedule employing a similar anticholinergic agent (Piptal) with no CNS properties and an inert placebo. After a period of 3 months all the investigators agreed that

260

L. G . MOOD AND J. H. B1F.L

the patients generally showed more disturbed and regressed behavior, characterized by greater degrees of idleness, withdrawal, and autism; a particularly noteworthy effect of the agent was the production of increased sexual activity manifested by open and excessive masturbation, homosexual acts, and overtures toward female personnel. The most striking effect of the drug in this group of patients was observed in the psychological test results. A significant increase in intellectual efficiency and speed of responsiveness had occurred, as shown by a variety of psychological tests. A significant reduction of the Rorschach “M” and “Y” responses occurred, and this may be interpreted as indicating a tendency toward strengthening of ego resources and reduction of social anxiety. Despite the difficulty in interpreting the psychotherapeutic effectiveness of this agent in this group of patients, it was quite apparent from the study that the drug was acting as an effective stimulant. The increased symptomatology noticed in the patients was probably a direct consequence of this stimulating action of the agent. XI. Biochemical and Electrophysiological Studies with Piperidy I GIyco I at es

A. EFFECT ON ENZYME SYSTEMS

In an attempt to elucidate the mechanism of action of the agents, their effect on a variety of enzyme systems of nervous and other excitable tissues was examined. Among the enzymes studied were numerous oxidases, glycolytic enzymes, components of the electron transport scheme, phosphatases, and esterases (including acetylcholinesterase and a variety of nonspecific esterases) ( Abood et al., 1959a). In addition, over-all oxidative phosphorylation of rat brain mitochondria was studied, using a variety of substrates. The agents, even at concentrations in excess of 5 x M , were without a direct effect on any of the enzymes and enzyme systems examined. In an effort to determine whether the agents might be indirectly affecting intermediary metabolism through an action on cellular permeability, their effects were studied in respiring brain and liver slices. Regardless of substrate utilized (which included glucose, amino acids, and components of the Krebs cycle), the agents in no way influenced either the oxidative or glycdytic metabolism of tissue slices. With the use of radioactive P”-orthophosphate and

ANTICHOLINERGIC PSYCHOTOMIMETIC AGENTS

261

C14-labeled glucose, the incorporation of these isotopes into phospholipids, proteins, and nucleic acids of brain mitochondria and brain and liver slices was examined. It was found that the piperidyl glycolates produced a slight but significant increase in the turnover of P3, in such phospholipids as diphosphoinositide and phosphatidyl choline. The incorporation of C"-glucose into lipids, amino acids, and proteins, however, was unaffected by the agents.

B. METABOLIC EFFECTS IN FROG NERVEAND MUSCLE When isolated sciatic nerves and ventral nerve roots of the bullfrog were exposed to the drug, a number of biochemical effects were noticed, depending on the incubation system employed. In a calcium-free Ringer's solution containing 5 mM ethylenediaminetetracetic acid (EDTA), it was possible to demonstrate a 20 to 30% inhibition of lactate production in the presence of 10-G M concentration of one of the centrally active piperidyl glycolates. This effect on lactate production, however, was not consistent-a fact which might be attributable to the relative impermeability of the nerve fiber to the drug and other constituents of the external media. In a modified Ringer's solution containing calcium (112 mM NaCl,, 2 mM KCl, 2 mM NaHCO,, 2 mM CaCl,) however, the piperidyl glycolates at this concentration produced instead a 20 to 30% increase in over-all lactate production. Although external calcium concentration did not affect over-all glycolysis in itself, it did seem to influence the drug effect on lactate production. The most constant and striking biochemical effects of the piperidyl glycolates could be demonstrated in the isolated frog (Runu pipiens) sartorius muscle. As can be seen in Table XIII, a drug concentration of 10-GMPMCG produced a 40% increase of lactate production in a Ringer's solution (containing 1.8 mM calcium). On the other hand, in the absence of calcium, the agent produced a 40%inhibitionof lactate production, while in a calciumfree solution containing EDTA there was as much as a 70%inhibition of glycolysis. When the external potassium concentration was increased to 20 mM, the lactate production was increased 20% in the presence of the drug. The addition of 0.5 pM of 2,4-dinitrophenol to a calcium-free system resulted in a fourfold increase in lactate production, but in the presence of this inhibitor the piperidyl glycolate no longer exerted its action.

262

L. G . ABOOD AND J. H. BZEL

TABLE XI11 EFFECT OF PMCG ON LACTIC ACID PRODUCTION IN FROG SARTORIUS MUSCLE^ System Ringer

Control PMCG

0.85 k 0.32 1.40 + 0 . 3 0

+40

0.001

Ca-free

Control PMCG

1.00 k 0 . 2 8 0.60 f 0 . 3 2

-40

0.001

Ca-free plus EDTA

Control PMCG

2.50 k 0 . 2 5 0.80 k 0 . 2 2

-

0.001

-69

K-rich Ringer

Control PMCG

1.33 k 0.18 1.60 k 0.20

+20

Ca-free EDTA plus 0.5 p M DNP

Control PMCG

3.20 3.15

0.20

k 0.15

0.01

-

-

0

a Frog sartorius muscles were incubated aerobically a t 34°C for 1 hour with gentle shaking. Lactate determinations were performed on the external solution and expressed in pmoles/gm wet weight muscle ( f standard deviation). Concentration of EDTA was 5 mM; potassium-rich Ringer’s solution contained 20 mM potassium. Conccntration of PMCG was 10-6 M . P = Probability by “t” test.

Other anticholinergic agents were also tested for their effect on muscle lactate production in a Ca-free EDTA system (Table XIV). In general the psychotomimetic piperidyl glycolates were inhibitory, and there appeared to be a correlation between metabolicinhibitory and psychotomimetic potencies. Quaternization of the N group or replacement of the acidic OH by H tended to diminish both actions. Atropine and procaine were without any action even at relatively high concentrations. An investigation was undertaken to determine the relationship of certain cations to the inhibitory action of the glycolate esters on lactate production of frog sartorius muscle. In a Ca-free solution, the lactate production of untreated frog sartorius muscle remained fairly constant as the concentration of EDTA was increased to 15 mM (Fig. 3); however, in the presence of lo-” M PMCG lactate production decreased gradually with increasing concentration of

263

ANTICHOLINERGIC PSYCHOTOMIMETIC AGENTS

TABLE XIV EFFECT OF VARIOUS AGENTS O N LACTATEPRODUCTION, SPONTANEOUS TWITCHING, A N D CONTRACTION OF FROG SARTORIUS MCJSCLE~ Isometric contraction

Lactate production

74 Agent Control DF AA BC

Q” AK Atropine Procaine

Inhibition of twitchb

(MI

Conc.

pMoles

Inhibition

-

1.3 0 3 0.6 0.7 1.2 1.3 1.2 1.3

77 54

5

x

46 8 0

5

x

10-6

10-6 10-6 10-5 10” 10-5

10-5

8 0

10-7 10-6

% M

Increase

-

50 30 25 20 0 20 0

10-3

10-6 10-6 10-6 10-5 10-4

10-5

10-5

10-4

10-5

10-6

10-6

Lactate production is expressed in twms of pmoles lactate/100 mg muscle/hr (aerobically). Minimal concentration neressary to inhibit spontaneous twitching of frog sartorius muscle in a Ca-free EDTA Ringer’s solution. c N-Methyl-3-piperidyl bcnzilatr mctthobromide, a quaternary derivative.

EDTA. The inhibition by the drug was about 15%in the absence of EDTA and almost 75%at 15 mhl EDTA. As the concentration of external K was increased to 20 mM the rate of lactate production increased linearly to a value 4 times greater than in its absence (Fig. 4 ) . In the presence of PMCG, however, the rate of lactate production, which was over 30% greater initially, remained constant with increasing K concentration. In order to determine whether the stimulatory effect of K on glycolysis either lacked specificity or was merely osmotic, sodium chloride was substituted for potassium chloride. There was no stimulatory action of Na even at concentrations of 200 mh-l. In a Ca-free EDTA system, the lactate production of the control muscles was found to increase about 30%when the external Mg concentration was 2 mM while the PMCG-treated muscles showed a slight decrease (Fig. 5 ) . As the concentration of Mg increased to 6 mM, the lactate production of the control muscles dropped about

264

L. C. ABOOD AND J. H. BIEL

1

I

I

5

10

15

mM E D T A

FIG. 3. Effect of EDTA on inhibition of lactate production by PMCG. Frog sartorius muscles were immersed in 2 ml of Ca-free Ringer's solution concontrol muscles; (0) with taining varying concentrations of EDTA. Key: ).( lO-'M PMCG. Muscles were incubated aerobically with gentle shaking at 34°C for 1 hour, and lactate determined on the external solution. The ordinate expresses activity in terms of 1 gram of tissue,

75%, and thereafter remained constant as the Mg concentration reached 10 mM. The lactate production in the PMCG-treated muscles also decreased as the concentration of Mg was increased above 2 mM. In Ringer's solution, the lactate production of the PMCG-treated muscles increased about 40% as the Mg concentration was increased from 0 to 4 mM, and then decreased to below the rate of lactate formation in the absence of Mg when the Mg concentration was increased to 10 mM. The control muscles behaved in similar manner, except that the magnitude of increase in lactate production was considerably less at 4 mM Mg.

MECHANISMFOR BIOCHEMICAL EFFECTS C. POSSIBLE The mechanism by which the piperidyl glycolates affects lactate production is at present somewhat obscure. This effect, however, does appear to be related to the external calcium concentration.

ANTICHOLINERGIC PSYCHOTOMIMETIC AGENTS

265

x 5

10

15

20

m M KCI

FIG. 4. Effect of KC1 on the lactate production of frog sartorius muscle. key: (0) control; ( 0 )with 10-'M PMCG. The KCl was added to normal Ringer's solution and an equivalent amount of NaCl removed to maintain isotonicity .

which in turn may be influencing the effect of the agent itself, either on membrane permeability or upon some critical metabolic factor within the muscle. In a calcium-free system, the drug can be shown to increase the uptake of glucose and orthophosphate, while at the same time decreasing the degradation of the intermediate glycolytic metabolites. Since the agents do not seem to affect oxidative metabolism, and since their action upon lactate production is demonstrable anaerobically, the locus of action of the agent can be neither in the electron-transport scheme nor in the Krebs cycle. It would appear as if the agent were influencing the binding characteristics of a cation such as calcium or magnesium in a structuralized metabolic system where the concentration of either of these ions may be critical for metabolism. The absence of the calcium and presence af EDTA evidently depletes the muscle of all excess calcium and magnesium so that now the remaining cations are critical for glycolytic metabolism. If, in such a system, an agent could interfere with the mechanisms responsible for making a cation such

266

L. G. A B W D AND J. H. BIEL

4

I

1

I

2

6

I

10 m M Mg

FIG. 5. Effect of magnesium on inhibition of lactate production by PMCG. Frog sartorius muscles were immersed in 2 1111 of test solution containing varying concentrations of magnesium. Dotted lines refer to experiments in normal Ringer’s solution and solid lines to experiments carried on in Ca-free EDTA with M PMCG. Muscles were solution. Key: ( 0 )control muscle; (0) incubated aerobically with gentle shaking at 34°C for 1 hour.

as magnesium available for glycolysis, then the over-all rate of lactate production would be influenced. The fact that structural organization is essential in order to demonstrate an effect of the piperidyl glycolates on muscle glycolysis becomes evident when one attempts to reproduce the effects on damaged or minced muscle. In such preparations it is possible to demonstrate an increased lactate production with the agents either in a Ringer’s or Ca-free medium, but it is impossible to demonstrate the inhibitory effect of the agent in a Ca-free EDTA system. This observation would suggest that the structural organization of the excitable tissue is a critical factor in the regulation of muscle glycolysis and that the piperidyl glycolates act by somehow affecting the interchange of some critical cofactors such as magnesium or calcium within this complex. The role of calcium in the electrochemical phenomena of excitable tissues is becoming increasingly more apparent [see Shanes (1958) for review].

ANTICHOLINERGIC PSYCAOTOMIMETIC AGENTS

267

D. EFFECTS ON CONTRACTILE AND EXCITABLE SYSTEMS OF FROGSARTORIUS MUSCLE

In order to understand the metabolic, functional relationships of an excitable tissue such as muscle, it is important to examine the metabolic components in connection with both contractile and excitatory events within the muscle, The piperidyl glycolates have not only a very striking effect upon the functional characteristics of skeletal muscle, but may also serve as an interesting and effective tool in studying the connecting link between excitation and contraction. 1. Contractile Efects in Calcium-free Solutions When a frog sartorius muscle is placed in a calcium-free system (with or without EDTA), the muscle soon begins to twitch spontaneously. It can be shown that the muscle twitches in such a system as long as calcium is diffusing out; and when the membrane potential reaches a level of about -30 mv the muscle ceases to twitch (Koketsu and Koyama, 1961).The addition of the PMCG at a concentration of M will completely inhibit the spontaneous twitching in a calcium-free system. The agent does not seem to be having a direct action upon the contractile system itself, since the drug, even at very large concentrations, does not influence the contraction of actomyosin threads or glycerated muscle fibers, either upon the addition of ATP or when they are electrically excited. It would, therefore, appear as if the piperidyl glycolate were acting directly upon the excitable membrane or upon the so-called “coupling system” between excitation and contraction. In view of the effect of the piperidyl glycolate on the spontaneous twitching of muscle in a calcium-free system, it might be inferred that the drug were indirectly inhibiting lactate production by affecting spontaneous twitching which in turn would accelerate lactate production. This inference seems unwarranted, however, insofar as the lactate production of muscle in a Ringer’s solution is the same as, if not higher than, in a potassium-rich system, even with the muscle in a relaxed state. 2. Changes in Excitatory Characteristias

By studying the effect of the piperidyl glycolates on the electrical

268

L. G. ABOOD AKD J. H. BIEL

characteristics of excitable tissue, it was possible to demonstrate a direct action of the agent on the excitable membrane (Koketsu and Koyama, 1961). By use of the isolated bullfrog spinal ganglion preparation, in which recordings are made from single spinal ganglion cells after stimulation of the sciatic nerve, it was possible to eliminate completely the positive after-potential, while greatly prolonging the negative after-potential of the spike with PMCG at a concentration of 10-6M. In the action potential of single muscle fibers in the frog sartorius muscle after direct electrical stimulation, a marked prolongation of the negative after-potential was demonstrable ( Fig. 6). Upon supramaximal electrical stimulation of the 1

2

FIG. 6. The effect of PMCG on the action potential of a single fiber of isolated frog sartorius muscle. Recordings were made intracellularly after stimulation of the attached sciatic nerve. Record 1; control in Ringer’s solution; 2, after immersion in 5 pg/ml of PMCG for 30 minutes. Calibration is 50 mv; time marker is 500 cycle per second.

frog sartorius muscle, a concentration of M PMCG accounted for as much as 50% increase in the isometric contraction of the muscle (Fig. 7). It is possible, therefore, that the increase in the negative after-potential produced by the agent may account for the increase in isometric contraction.

ANTICHOLINERGIC PSYCHOTOMIMETIC AGENTS

2

1

269

3

FIG. 7. The effect of PMCG on the isometric contraction of the isolated frog sartorius muscle. Supramaximal stimulation was applied to the attached sciatic nerve. Record 1 was taken after immersion in Ringer’s solution; 2, after immersion in Ringer’s containing 5 Pg/ml PMCG; 3, after reimmersion in Ringer’s.

PMCG was found to have a striking effect on the resting potential of single frog sartorius muscles after immersion in Ca-free EDTA (Fig. 8). Immediately lipon immersion in Ca-free EDTA. A

... .. I00

:. .... .. .. ..

. . ’.

B

I

.. .

. . . ..: :.

.,

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

> 50

..

.

10

.

. . . .

,

20

10 MIN

20

MIN

FIG. 8. The effect of PMCG on the membrane potential of single frog sartorius muscle fibers in a Ca-free EDTA sjstem. Frog niuscles were exposed to 10-5M PMCG in Ringer’s solution for % hour and then transferred to a Ca-free EDTA solution. Recordings were made from numerous muscle fibers before and immediately after transfer into Ca-free EDTA (indicated by arrow). A, resting potential in absence of drug; B, in presence of drug. Ordinate is in millivolts ( m v ) and abscissa in niinutcs in Ca-free EDTA solution.

the resting potential of single muscle fibers decreased to values of 50 to 6S mv; and by 10 minutes the majority of muscle fibers had

270

L. G. ABOOD AND J. H. BlEL

potentials of less than -25 mv (see A in Fig. 8). If the muscles had been exposed to PMCG prior to transfer into Ca-free EDTA, the resting potentials showed a considerably smaller decrease (see B ) , Many of the fibers still had potentials of about -50 mv even 20 minutes after immersion in Ca-free EDTA. TABLE XV SITES OF ACTION OF PIPERIDYL GLYCOLATES IN VARIOUS TISSUEA N D ORGANSYSTEMS System

Effects

Brain mitochondria

Site of binding Increase in P32-phospholipidturnover Increase in membrane permeability No effect on oxidat.ion-phosphorylation

Brain slices, minces

No met(abo1iceffects Increase in permeability

Tissue culture neurons

Increased movement of cytoplasmic granules (mitochondria, liposomes) in vicinity of Nissl substance, dendritic and axonal endings

Smooth muscle

Anticholinergic Antiserotonic Antihistaminic

Isolated frog sartorius

Increases isometric contraction in Ringer’s solution Decreases spontaneous twitching in Ca-free solution M) Prolongs negative after-potential Increases lactate production in Ringer’s solution Decreases lactate production in Ca-free EDTA solution Increases permeability

Isolated frog spinal ganglia

Increases negative after-potential (5 X Blocks action potential (10-5-10-6 M ) Increases oxidative-glycolytic metabolism

Intact nervous system

Spindle slow-wave patterns (sleeplike) Inhibition of alerting response in reticular form Desynchronization of EEG with blocking of postconvulsive delta activity

M) M) M)

M)

ANTICHOLINERGIC PSYCIiOTOMIMETIC AGENTS

271

The effect of the drug on membrane depolaiization could conceivably be the result of interference with the movement of ions, such as Ca and K. It is conceivable that the lipophilic-cationic property of PMCG permits competition with Ca for phospholipid binding sites within the excitable membrane, and that, PMCG may, in part, substitute for Ca. Studies underway indicate that after the resting potential of a frog sartorius muscle drops to about -50 mv in a Ca-free system, it can be restored to almost the normal level by the addition of 2 x M PMCG to the Ca-free medium. Furthermore, the muscle fibers once again become excitable.

E. APPLICABILITYOF PIPERIDYL GLYCOLATES TO STUDYING BIOCHEMICAL-FUNCTIONAL RELATIONSHIPS It is obvious from the present studies that the piperidyl glycolates offer an important tool in exploring the chemical-functional interrelationship of excitable tissues (see Table XV for summary of drug actions). The results of attempts to link metabolic changes with either excitation or contraction of muscle have been somewhat meager and complex. There seems little question, however, that the piperidyl glycolates must be affecting the link between excitation and contraction in such a system as frog sartorius muscle, and that the manner in which it affects this link must in turn have some bearing on certain metabolic changes either directly or indirectly associated with the phenomenon. Although these studies have been demonstrated only on muscle, it is quite likely that the drug is having a similar metabolic effect on nervous tissue. Evidently, the conditions of an excitable frog sartorius muscle in a calcium-free system must be critical enough so that the drug effect can be measured. The primary action of the drug, however, is upon the nervous system, and one must ultimately demonstrate the action of the agent on similar biochemical-functional relationships in neuronal tissue. ACKNOWLELXMENT

The authors would like to thank the following men for their part in the chemical synthetic program: W. K. H o p , E. P. Sprengeler, and H. A. Leiser of Lakeside Laboratories; Dr. J. G. Cannon, University of Wisconsin; Dr. F. Blicke, University of Michigan. REFERENCES

Abood, L. G. (1959). J . Neuropsychint. 1,92.

272

L. G . ABOOD AND J. H. BIEL

Abood, L. G. (1961). J. Med. Phamn. Chem. 3, 469. Abood, L. G., and Meduna, L. J. (1958). J. Nervous Mentd Diseuse 127, 546. Abood, L. B., and Rinaldi, F. (1959). Psychophumuzcologia I, 117. Abood, L. G., Ostfeld, A. M., and Biel, J. H. (1958). Proc. SOC. Expptl. Biol. Med. 97, 483. Abood, L. G., Ostfeld, A. M., and Biel, J. H. (1959a). Arch. intern. p h u m codynamie 120, 186. Abood, L. G., Biel, J. H., and Ostfeld, A. M. (1959b). In “Neuro-Psychopharmacology” (P. Bradley, ed.), p. 433. Pergamon, New York. Apter, N. S. (1957). In “Congress Report on 2nd International Congress for Psychiatry,” p. 156. Apter, N. S., and Abood, L. G. (1962). In preparation. Bailey, P., and Sweet, W. H. (1940). J. Neurophysiol. 3, 276. Biel, J. H., Sprengeler, E. P., Leiser, H. A., Homer, J., Drukker, A., and Friedman, H. L. (1955). J. Am. Chem. SOC. 77,2250. Biel, J. H., Hoya, W. K., and Leiser, H. A. (1959). J. Am. Chem. SOC. 81, 2527. Biel, J. H., Abood, L. G., Hoya, W. K., Leiser, H. A., Nuhfer, P. A,, and Kluchesky, E. F. (1962). J. Org. Chem. 26,4096. Bovet, D., and Bovet-Nitti, F. ( 1948). “Medicaments du Systkme Nerveux Vkg6tatif.” Karger, Basel, Switzerland. Bradley, P. B., and Elkes, J. (1957). In “Metabolism of the Nervous System” (Richter, D., ed.), p. 515. Pergamon, New York. Brink, F. (1954). Pharmmol. Revs. 6,243. Cannon, J. G. ( 1960). Personal communication. Chappel, C. I., Grant, G. A., Archibald, S., and Paquette, R. (1957). J. Am. Phumn. Assoc. Sci. Ed. 46, 497. Chatonnet, J. M., Tanche, M., and Cabanac, J. L. (1960). J. physiol. ( P a r i s ) 52, 48. De Boor, W. ( 1956). “Pharmakopsychologie und Psychopathologie.” Springer, Berlin. Delgado, J. M. R., and Livingston, R. B. (1948). J. Neurophysiol. 11, 39. English, D. ( 1961 ) . J. Neuropsychiat. ( In press. ) Feldberg, W. S. (1955). Lancet 2, 900. Ferguson, L. N. ( 1952). “Electronic Structures of Organic Molecules.” PrenticeHall, New York. Fernhndez-MorLn, H. ( 1957). In “Metabolism of the Nervous System” ( D. Richter, ed.), p. 1. Pergamon, New York. Finean, J. B. (1957). In “Metabolism of the Nervous System” ( D . Richter, ed.), p. 52. Pergamon, New York. Fink, M. (1960). Eledroencephulog. and CEin. Neurophysiol. 12,359. Finkelstein, B. A. ( 1961). J. Neuropsychiat. 2, 144. Gershon, S. (1960). J. Neuropsychiat. 1, 283. Gibbs, F. A., and Gibbs, E. (1941). “Atlas of Electroencephalograpl?~,” Vol. I. Cummings, Cambridge, Massachusetts. Green, H. D., and Haff, E. C. (1937). Am. J. Physiol. 118,641. Hoch, P. H., Pennes, H. H., and Cattell, J. P. (1953). Assoc. Reyearch Nervous Mental Diseuse Research Pub., p. 33.

ANTICHOLINERGIC PSYCHOTOMIMETIC AGENTS

273

Hoffer, A., and Osmond, H. (1960). “The Chemical Basis of Psychiatry.” Charles C Thomas, Springfield, Illinois. Jackman, A. ( 1958). Personal communication. Jarvik, M. ( 1958). “Progress in Neurobiology.” Hoeber-Harper, New York. Kaada, B. R. (1951). Acta Physiol. S c a d . 24, Suppl. 83. Kliiver, H. (1928). “Mescal: The Devine Plant.” Kogan Paul, London. Kliiver, H. ( 1942). In “Studies in Personality” (Quinn McNemar and Maud A. Merrill, eds.), p. 189. McGraw-Hill, New York. Koketsu, K., and Koyama, K. (1961). Personal communication. Langworthy, 0. R., and Richter, C. P. (1939). Am. J. Physiol. 126, 158. Lebovitz, B. Z., Visotsky, H. M., and Ostfeld, A. M. (196Oa). Arch. Gen. Psychiat. 2, 390. Lebovitz, B. Z., Visotsky, H. M., and Ostfeld, A. M. (1960b). Arch. Gen. Psychiat. 3, 176. Lipman, V. ( 1961). Ph.D. Dissertation. Illinois Inst. Technology, Chicago, Illinois. Meduna, L. J., and Abood, L. G. (1959). J. Neuropsychiat. 1, 1. Meyer, A. E., Beck, E., and McLardy, T. (1947). Brain 70, 18. Osmond, H. (1957). Ann. N . Y. Acarl. Sci. 66, 418. Ostfeld, A. M., Abood, L. G., and hlarcus, D. (1958). Arch. Nmrol. Psychiat. 79, 317. Ostfeld, A. M., Visotsky, H., Abood, L. G . , and Lebovitz, B. Z. (1959). Arch. Neurol. Psychiat. 81, 256. Pfeiffer, C. C. (1948). Science 23, 94. Pfeiffer, C. C. (1959). Intern. Rev. Neurobiol. 1, 195. Polidora, V., and Abood, L. G. ( 1961). J. Exptl. Psychol. ( I n press.) Ranson, S. W., and Clark, S. L. (1959). “The Anatomy of the Nervous System,” 10th ed. Saunders, Philadelphia, Pennsylvania. Rinaldi, F., and Himwich, H. E. (1955). Arch. Neurol. Psychiat. 73, 387. Ruch, T. C., and Shenkin, H. A. (1943). J. Neurophysiol. 6, 349. Schmitt, F. O., Bear, R. S., and Palmcr, K. J. (1941). J . Cellukzr Comp. Physiol. 18, 31. Schueler, F. W. (1953). Arch. intern. pharmacodynamie. 93, 417. Shanes, A. (1958). Pharmucol. Revs. 10, 59. Shaw, F. H., Gershon, S., and Bentley, G. A. (1957). J. Pharm. and Phamnacol. 9, 666. Tourlentes, T., Axiotis, A,, Hunsicker, A., Hurd, D., Vassilon, G., and Abood, L. G. (1960). J. Neuropsychiat. 2, 49. Voris, H. C., and Adson, A. W. (1935). Arch. Neurol. Psychiat. 34, 965. Wikler, A. ( 1957). “The Relationship of Psychiatry to Pharmacology.” Williams & Wilkins, Baltimore, Maryland. Wilzbach, K. E. (1957). J. Am. Chem. SOC. 79, 1013.

This Page Intentionally Left Blank

BENZOQUINOLIZINE DERIVATIVES: A NEW CLASS OF MONAMINE DECREASING DRUGS WITH PSYCHOTROPIC ACTION

.

. .

By A Pletscher. A . Brossi. and K F Gey Research Departments of

F.

.

Hoffmann-La Roche & Co Ltd., Basel. Switzerland

I. Introduction . . . . . . I1. Chemistry . . . . . . . A . Ketones . . . . . . . . . . B . Secondary Carbinols C . Tertiary Carbinols . . . . I11. Metabolism . . . . . . . A . Tissue Distribution and Excretion B . Metabolites . . . . . . IV . Influence on Monamine Metabolism A . Central Nervous System . . B . Peripheral Organs . . . . . . . C . Urinary Metabolites D. Mechanism of Action . . . V . Pharmacology . . . . . . A . Nervous System . . . . . B. Cardiovascular System . . . C . Gastrointestinal Tract . . . D . Other Actions . . . . . E . Toxicity . . . . . . . VI . Clinical Action . . . . . . VII . Relationship between Effects on . . . Pharmacological Action A . Sedation . . . . . . . B . Blood Pressure . . . . . C . Gastrointestinal Tract . . . VIII . Summary . . . . . . . References

. . . . . . . . . 275 . . . . . . . . . 277 . . . . . . . . . 278 . .

. .

. .

. .

. .

. .

. .

. .

. 278 . 279

. . . . . . . . . 281 . . . . . . . . . 281

. . . . . . . . .

.

.

.

.

.

.

.

.

281

. 282

. . . . . . . . .

282

. . . . . . . . . 287 . . . . . . . . . 288 . . . . . . . . . 288

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . .

. . . . . . . . .

. . . . . . . . .

291 291 296 297 298 299 300

Monamine Metabolism and

. . . . . . . . . 300 . . . . . . . . . 301 . . . . . . . . . 301 . . . . . . . . . 302 . . . . . . . . . 302

.

I Introduction

The discovery by Brodie and his group that reserpine liberates 5-hydroxytryptamine from brain and intestine stimulated new ideas 275

276

A. PLETSCHER, A. BROSSI, AND K. F. GEY

concerning the mechanism of drug action and the physiological role of monamines (Pletscher et al., 1955, 1956; Shore et al., 1955). Among a wide variety of drugs, only Rnuzoolfia alkaloids with sedative action depressed the content of 5-hydroxytryptamine and norepinephrine in brain. In consequence, the hypothesis was put forward that the sedative effect of reserpine might be mediated through changes of cerebral monamines. In 1956, during studies in the emetine field, a new group of drugs was synthesized, the 1,2,3,4,6,7hexahydro-llbH-benzo[a]quinolizines ( Brossi et al., 1958a). Various derivatives of this class depressed 5-hydroxytryptamine and norepinephrine in the brain of different animal species and caused reserpine-like sedation, This was of particular interest because the benzoquinolizines belong to a new class of compounds chemically unrelated to indole alkaloids such as reserpine. The benzoquinolizines might therefore be useful to elucidate further the aforementioned hypothesis. The purpose of this paper is to review the present knowledge on chemistry, metabolism, biochemistry, pharmacology, and clinical effects of benzoquinolizines (see tabulation). Furthermore, the relation between changes of monamine metabolism and pharmacological action will be discussed.

Code no.

Srihstitm t.ioii

RO 1-9288 2-Hydroxy-3-n-butyl-9,lO-dimethoxyRO 1-9564 2-Hydroxy-3-ethyl-9,lO-methplendiox;vRO 1-9569 2-0xo-3-isobutyl-9,lO-dimethosy-

Formula

Va Va IVb

( = tetrabenazine,Nitoman)

RO 1-9571 2-Hydroxy-3-isohutgl-C3,10-dimethoxy-

Va

Ro 4-0786 2-Hydroxy-2-ethiny1-3-ethyl-9,lO-dimethoxyRo 4-1284 2-Hydroxy-2-ethyl-3-isoh~ityl-~,l0-dirnethoxyRO 4-1286 2-Hydroxy-2-ethyl-3-n-buty1-9,lO-dimethosy-

VIa VIa VIa VIh VIh

Ro 4-1374 RO 4-1376 RO 4-1398 RO 4-1632 RO 4-1860

2-Hydroxy-2-allyl-3-isohutyl-9,l0-dimethoxy2-Hydroxy-2-isobutyl-3-ethyl-9,10-dimethoxyosp2-Hydroxy-2-(3-methoxyhutyl)-3-isobuty1-9,10-dimeth 2-0x0-3-(2-cthylhuty1)-9,lO-dimethoxy2-Hydroxy-2-ethyl-3-(2-ethylhut~yl)-9,lO-dimethoxy-

VIh IVb VIa

BENZOQUINOLIZINE DERIVATIVES

277

II. Chemistry

All compounds to be reviewed are derivatives of 1,2,3,4,6,7-hexahydro-llbH-benzo[a]quinolizine ( I ) . This ring system is a structural feature of certain alkaloids which belong to the protoberberine (e.g., I1 is tetrahydropalmatine) and ipecac groups (e.g., I11 is emetine) .

( 111 1

The ketones ( IVa) and ( IVb), the secondary carbinols (Va), and the tertiary carbinols ( VIa) and ( VIb) are the most interesting derivatives. They contain at least two asymmetric centers and therefore different stereoisomers must be considered. The compounds to be reviewed have the relative configuration shown in their formu1a.l Benzoquinolizine derivatives the pharmacological action of which is not known in detail will not be considered in this survey, e.g., decahydrobenzoquinolizines,2 benzoquinolizines with basic substituents (Naike, 1959; Suzuta, 1959; Yamazaki, 1959), azabenzoThe benzoquinolizines discussed are optically inactive, therefore only the relative configurations are indicated in the formulas. Various decahydrobenzoquinolizines were prepared by Drs. M. Walter and 0. Schnider from the Chemical Research Department of F. Hoffniann-La Roche & Co. Ltd., Basel (compare, e . g , Brossi et al., 1958a).

278

A.

PLETSCHER, A.

BROSSI, AND K. F. GEY

quinolizines (Lombardino at d.,1959, 196l), and other benzoquinolizine derivatives (compare for instance Ban et d.,1959; Battersby and Garratt, 1959; Burgstahler and Bithos, 1960; Griissner et al., 1959). A. KETONES The ketones ( IVa), substituted in the aromatic ring with alkoxy, hydroxy, or methylenedioxy groups can be synthesized by two main methods (Brossi et al., 1958a; Belg. Patent No. 574,911, The Wellcome Foundation Ltd., England). For the synthesis of the pharmacologically more interesting ketones ( IVb), substituted with an alkyl chain at C-3, several different approaches have been worked out (Battersby et al., 1953; Brossi et al., 1958a, 1960; Itoh and Sugasawa, 1959; Osbond, 1961; Belg. Patent No. 574,911, The Wellcome Foundation Ltd., England). It is possible to resolve the ketones ( IVb) and to racemize the unwanted optically active isomer under a variety of condition^.^ The most interesting derivatives prepared in this series are Ro 1-9569 (tetrabenazine, Nitoman) and Ro 4-1632. R

(ma)

(Nb)

IVb, RO 1-9569: R=-CHz--CH(CHs)z IW, Ro 4-1632: R = -CHrCH(CzHs)z

Quaternization, aromatization of ring C, and elimination of the carbonyl group4 lead to compounds of minor interest. €3. SECONDARY CARBINOLS

By reduction of the ketones ( IVb), the secondary carbinols ( Va) or mixtures of (Va) and (Vb), respectively, are obtained (Brossi et a Belgium Patent No. 604,163. The Wellcome Foundation Ltd., England. ‘Ready elimination of the carbonyl function is possible by way of the dithioketal. The preparation of deoxo bases is also mentioned in the publication by Osbond (1961)and in the paper of Tani and Ishibasi (1967).

279

BFNZOQUINOLIZINE DERIVATIVES

al., 1958b). The stereoisomer ( V a ) , which can also be prepared in a different way (Osbond, 1961), predominates. The n-butylcarbinol Ro 1-9288, the isobutylcarbinol Ro 1-9571, and the ethylcarbinol Ro 1-9564 belong to this group:

%q :-

R,

i

R,

R,

OH Wa)

OH

(Vb)

Va, Ro 1-9288: R, = -((CH,),-CH, , R, = R, = -OC% R, = R, = -O-CH2-oVa, Ro 1-9564: R, = -CzH5, Va, Ro 1-9571: R, = - CH,- CH -(CHs)2, R, = R, = -OCH,

Again quaternization results in loss of activity. Esterification of the carbinol grouping yields new and interesting derivatives ( Brossi et al., 1958b). C , TERTIARY CARBINOLS

Compounds of formula (VIa) can be prepared from (IVb) by condensation with acetylene and hydrogenation of the ethynyl group.

(VJW

The tertiary carbinol (VIb) is obtained from ( IVb) by a Grignard r e a ~ t i o n .In ~ both experiments mixtures of two stereoisomers are obtained. Some interesting compounds found in this series are shown in Table I. A publication on this class of compounds is in preparation.

SOME

COMPOIJSlX3 I N

TABLE I TERTIARY CARBINOL

THE

Compound

Ro

Configuration

4-0786 4-1284 4-1286 4-1860

VIa

4-1374 4-1376 4-1398

VTb

R,

Ra

Level of brain 5-HTo

8m

Dose (mg/kg)

(% of controls)

B

50 50 5

36 k 2 109 _+ 2 32 k 1

U

H

-C-CH

-CZ& -CzHb --C,H,

-C€Iz--CH=CH2 -CHZ-CII (CH3)2 --CH,-CH,-CH-OCHa

I

c& a

*

SERIES

One-half hour after intraperitoneal injection into mice.

-CH?-CH(CH3jz -C*& --CH?-CH(CHa)z

5 F w

we

BENZOQUINOLIZINE DERIVATIVES

281

Ill. Metabolism

A. TISSUE DISTRIBUTION AND EXCRETION

In animals tetrabenazine has a relatively short biological halflife in plasma and organs; it is between 1%and 3%hours in guinea pigs after intraperitoneal injection and is even shorter in rabbits after intravenous administration (Quinn et al., 1959; Schwartz et al., 1960). The compound has special affinity for body fat, where its concentration is up to 40 times higher than in plasma. Studies with tritium-labeled tetrabenazine show no preferential affinity for special brain structures in mice (Stumpf et al., 1961). Because of the short half-life time no cumulation of tetrabenazine occurs in the organism. After repeated administration only small amounts of the drug are present in plasma and tissues of rabbits 24 hours after the last dose. Some discrepancies in the concentration of tetrabenazine in the organs as found by various authors may be due to differences in animal species and analytical methods (Quinn et ul., 1959; Schwartz and Rieder, 1961) . In the urine of rabbits less than 2%of intravenously administered tetrabenazine is excreted unchanged within the first 24 hours (Quinn et aZ., 1959). These experiments indicate that almost all the drug undergoes metabolic transfonnation. In humans the radioactivity of blood drops to minimal levels within 10 hours after intravenous injection of tritium-labeled tetrabenazine. Three days later no further activity can be detected. In the urine 40% and in the feces 2.5% of the total activity is found within 24 hours; after 48 hours 54%of the total activity has been excreted. In the cerebrospinal fluid no radioactivity appears at all (Stumpf et al., 1961). B. METABOLITES By thin layer chromatography, unchanged drug and nine metabolites have been detected in the urine of humans and rabbits, as well as in liver homogenates of guinea pigs incubated with tetrabenazine. Some metabolites are present as glucuronides. Up to now, only (VII) and (VIII) have been identified by comparison with synthetic compounds.b ‘Experiments carried out by Drs. D E. Schwartz and J. Rieder, Medical Research Department of F. Hoffmann-La Roche & Co. Ltd., Baqel.

282

A . PLETSCHER, A. BROSSI, AND K. F. GEY

(VIII) IV. Influence on Monamine Metabolism

A. CENTRAL NERVOUS SYSTEM A characteristic property of certain benzoquinolizines consists in lowering the mommine content of the brain (Figs. 1 and 2 ) . The first compound found to have this effect was tetrabenazine. After administration of 20-40 mg/kg, a marked and rapid decrease of the 5-hydroxybyptamine and norepinephrine content occurs in various animal species, e.g., mice, rats, guinea pigs, rabbits. Tetrabenazine thus has an action similar to that of reserpine; the two drugs differ, however, in the following respects : 1. Tetrabenazine is 1&20 times less potent than reserpine. For instance, the dose which causes 50%depression of S-hydroxytryptamine (EDs0) in rabbit brain is about 20 mg/kg for tetrabenazine and 2 mg/kg for reserpine. Maximal amine depression obtained with high doses is less marked with tetrabenazine (50-7058 for 5hydroxytryptamine, 7%85%for norepinephrine) than with reserpine ( approximately 90%for 5-hydroxytryptamine and norepinephrine). 2. Maximal depression of the amines occurs more rapidly after tetrabenazine than after reserpine. 3. The duration of action of tetrabenazine (16-24 hours) is shorter than that of reserpine ( several days).

283

BENZOQUINOLIZINE DERIVATIVES

'a.

r

0

.. ........................................ . *

I

1

8

16

4-

24

140

FIG. 1. Decrease of 5-hydroxytryptamine in rabbit brain by benzoquinolizines and reserpine. Abscissa: time in hours after intravenous administration of the drugs; ordinate: 5-hydroxytryptamine in pg/gm fresh brain. Curves: tetrabenazine, 40 mg/kg; - - - - - - Ro 1-9571, 30 mg/kg; - - - RO 1-9288, 30 mg/kg; reserpine, 2 mg/kg. The points represent averages from double determinations [spectrophotofluorometric ( Bogdanski et al., 1956)1 in one animal. (From Pletscher et al., 1958a).

The tetrabenazine-induced decline of 5-hydroxytryptamine differs somewhat from that of norepinephrine. Thus, 5-hydroxytryptamine diminishes rapidly after injection of the drug, minimum levels being obtained after about 4 hours. Furthermore, in rabbits the minimum levels which can be reached are 15-25% of controls for norepinephrine and 3049% for 5-hydroxytryptamine. The recovery of 5-hydroxytryptamine after tetrabenazine and other benzoquinolizines proceeds at a somewhat higher speed than that of norepinephrine. Thus, in the rabbit, after a dose of 40-50 mg/kg tetrabenazine, brain 5-hydroxytryptamine regains its normal level after about 16-24 hours, whereas the norepinephrine content is restored only to about 7040% after this time interval. This might be due to the fact that 5-hydroxytryptamine has a faster turnover than norepinephrine (Pletscher, 195713; Pletscher and Gey, 1960; Pletscher et ul., l958a, b; Quinn et al., 1958, 1959) . According to recent experiments, tetrabenazine, like reserpine,

284

A. PLETSCHER, A. BROSSI, AND K. F. CEY

- ... ............................ *

'

II 1

8

16

24'

140

FIG. 2. Decrease of norepinephrine in rabbit brain by benzoquinolizines and reserpine. Abscissa: time in hours after intravenous administration of the drugs; ordinate: norepinephrine content of the central parts of the brain stem in pg/gm fresh weight. Curves: -tetrabenazine, 40 mg/kg; - - - - - - Ro 1-9571, 30 nig/kg; - * - - * Ro 1-9288, 30 mg/kg; reserpine, 2 mg/kg. The points represent averages from double determinations [spectrophotofluorometric (Shore and O h , 1958)] in one animal. (From Pletscher et al., 1958).

-

also causes a decrease of the dopamine content of brain.7 The effect of the drug on other amines is not yet known. Various other benzoquinolizines also depress the 9hydroxyTABLE I1 DOSE OF VARIOUSBENZOQUINOLIZINES CAUSING50y0 DEPRESSION O F BRAIN5-HYDROXYTRPPTAMINE (EDSO)' Compound

EDso

Fiducial limits

Tetrabenazine

3.6 0.5

2.6-5.2 0.3-0.8 0.3-1.8 1.7-2.8

RO 4-1860

RO 4-1284 RO 4-1633 a

1 .o 2.2

One hour after intraperitoneal injection into rats.

BENZOQUINOLIZINE DERIVATIVES

285

tryptamine and norepinephrine content of the brain. The compounds differ in duration and intensity of action. None acts longer than tetrabenazine (Figs. 1 and 2 ) . Several substances, e.g., Ro 4-1284, Ro 4-1632, and Ro 4-1860 have, however, a 2-7 times smaller ED,, than tetrabenazine' (Table 11). Controversial results have been reported as to whether Ro 4-1398 has a preferential effect on brain 5-hydroxytryptamine. Some authors found that in mice Ro 4-1398 and Ro 4-1284 depressed 5-hydroxytryptamine to the same degree, whereas the norepinephrine decrease was less marked after Ro 4-1398 than after Ro 4-1284 (Pletscher et al., 1959b). This might, however, be due to the fact that the more potent Ro 4-1284 has a quicker onset and shorter duration of action than the less effective Ro 4-1398 (Brodie, 1960; Brodie et al., 1960; Quinn and Brodie, 1959). Other benzoquinolizines (e.g., Ro 1-9564, Ro 4-1376, Ro 4-0786) have no influence at all on the brain amines, even in high doses7 (see Table I). The effect of benzoquinolizines on the amine content in different parts of the brain has not been investigated in detail. In preliminary experiments comparing brain stem and pallium of rabbits no differences could be shown.' The benzoquinolizine-induced depression of amines in the brain is counteracted by monumine oxidme inhibitors. In rabbits, for instance, pretreatment with iproniazid diminishes the decrease of 5hydroxytryptamine and norepinephrine caused by tetrabenazine (Table 111) (Pletscher et al., 1958a). An analogous antagonism has been shown between monamine oxidase inhibitors and reserpine (Brodie et al., 1956; Pletscher, 195713; Shore and Brodie, 1957). Furthermore, there is interaction between amine-depleting benzoquinolizines and reserpine. Thus, 24 hours after administration of reserpine in tetrabenazine-pretreated rabbits the 5-hydroxytryptamine content of brain is still as high as 50% of the norm, whereas reserpine alone depletes 5-hydroxytryptamine almost completely. The reserpine-induced norepinephrine decrease cannot be influenced by pretreatment with tetrabenazine (Quinn et al., 1958, 1959). If brain 5-hydroxytryptamine has been depleted by reserpine, subsequent administration of tetrabenazine is without effect on the levels and the recovery rate of the amine.'

' Experiments carried out by the authors

INHIBITION OF

1 -4mine

THE

TABLE 111 TETRABENAZINE-INDUCED MONAMINE DECREASEI N RABBITBRAINBY I P R O N I A Z I W ~ ~

Kithout iproniazid pretreatment"

Controls

5-Hydroxytryptamine

Norepinephrine

1

0.i1;)0.04

3 Hours after tetrabenazine

/I

With iproniazid pretreatment"

75 Decrease

Controls

3 Hours after 96 tetrabenazine Decrease

I/ Significance

0.18 f 0.015 (5)

1.05

16 k 3 . 0 (B)

(A): (B) p (C):(D) p

< 0.01 < 0.01

0.20 f 0.01 (4)

1.12 k 0 . 0 7 0.60 k 0 . 0 7 46 f 6 . 2 (5) (6) (D)

(B):(D) p

< 0.01

k 0.05 0.88 (9)

0.03 (5)

From Pletscher et al. (195%).

* Pretreatment with 100 mg/kg iproniazid intravenously 16 hours prior to administration of 40 mg/kg

tetrabenaxine intravenously. CThe figures are averages in pg/gm wet tissue (whole brain for 5-hydroxytryptamine, central parts of the brain stem for standard error. The number of animals is given in parentheses. The monamines were determined by spectronorepinephrine) photofluorometric methods (Bogdanski et al., 1956; Shore and Olin, 1958).

287

BENZOQUINOLIZINE DERIVATIVES

B. PERIPHERAL ORGANS The effect of benzoquinolizines on the monamine content of peripheral organs is generally less pronounced than in brain, Thus, in rabbits and guinea pigs relatively high doses of tetrabenazine do not cause a marked 5-hydroxytryptamine decrease in platelets and small intestine. The norepinephrine content of the heart in rabbits is not depressed either, whereas in guinea pigs a norepinephrine decrease of about 50%can be observed. This depression of norepinephrine is rather less marked than in the brain (Fig. 3). Reserpine, in contrast, depresses norepinephrine in the heart more markedly than in the brain. In the adrenals of rabbits, large doses of tetrabenazine also cause a moderate decrease of catechol amines

1

1 at

5c

0 ‘fa 1

2

1

12 7%

FIG.3. Effect of tetrabenazine on the monamine content of several organs of guinea pigs and of whole blood of rabbits. Abscissa: time in hours after administration of tetrabenazinemethansulfonate ( corresponding to 20 mg/kg base), intraperitoneally in guinea pigs, intravenously in rabbits; ordinate: monamine content in per cent of controls (controls = 100%). Curves: - - 5-hydroxytryptamine of whole brain; ___ norepinephrine of brain stem; - - - - - - norepinephrine of heart; - -- 5-hydroxytryptamine of small intes5-hydroxytryptamine of whole blood. The points represent mean tine; values of four experiments i standard error. Spectrophotofluorometric determinations of 5-hydroxytryptarnine ( Bogdanski et al., 1956) and norepinephrine (Shore and Olin, 1958). (From Schwartz et al., 1960).

-

288

A. PLETSCHER, A. BROSSI, AND K. F. GEY

which is less pronounced than after reserpine (Carlsson et d.,1957; Quinn et al., 1959; Schwartz et al., 1960). Other benzoquinolizines (Ro 4-1284, Ro 1-9288) depressing cerebral monamines also have little effect on the 5-hydroxytryptamine and norepinephrine content of peripheral organ^.^ In conclusion, tetrabenazine and other benzoquinolizines exert a preferential action on monamines of the brain. In contrast to its effect in the brain, tetrabenazine does not influence the reserpine-induced depression of 5-hydroxytryptamine in peripheral organs. Thus, pretreatment with tetrabenazine in rabbits cannot prevent the action of reserpine on 5-hydroxytryptamine in platelets and intestine (Quinn & al., 1959). The effect of tetrabenazine on 5-hydroxytryptamine is not limited to the intact organism. The drug lowers the 5-hydroxytryptamine content of rabbit platelets in vitro (Paasonen and Pletscher, 1959; Quinn d al., 1959) (Fig. 4 ) . Tetrabenazine, as expected from the effects in duo, is less potent than reserpine. Concentrations of 0.3 pg/ml (and possibly less) reserpine cause a 5hydroxytryptamine decrease of 40-50%,whereas for a similar effect 10 pg/ml tetrabenazine are needed ( Quinn et al., 1959). C. URINARY METABOLITES In rabbits tetrabenazine, like reserpine, causes a significant increase in the urinary elimination of 5-hydroxyindolylacetic acid, a major metabolic end-product of 5-hydroxytryptamine ( Pletscher, 1957a). This rise is probably a consequence of the 5-hydroxytryptamine depletion in the tissues. It cannot be decided which organ is the major source of the increased 5-hydroxyindolylacetic acid because of differences in content and turnover of 5-hydroxytryptamine. For instance, brain contains only about 1%of the total 5hydroxytryptamine of the body but has a higher turnover than the major 5-hydroxytryptamine depots (e.g., intestine and blood) ( Udenfriend and Weissbach, 1958). The effect of benzoquinolizines on monamine metabolites other than 5-hydroxyindolylacetic acid remains to be investigated.

D. MECHANISMOF ACTION 1. Interference with the Storage Capacity of Monamirtes Interference with the storage capacity constitutes the most likely mechanism for the monamine lowering effect of benzoquinolizines.

BENZOQUINOLIZINE DERIVATIVES

289

The evidence for this hypothesis is based on the following observations and considerations: ( a ) Benzoquinolizines presuinably do not interfere with the synthesis of monamines. At least tetrabenazine does not inhibit the last step of 5-hydroxytryptamine formation, i.e., decarboxylation of 5-hydroxytryptophan, in vitro and in ~ i v o . ~ ( b ) The decrease of 5-hydroxytryptamine levels in the organs parallels the urinary excretion of 5-hydroxyindolylacetic acid. This is compatible with the view that the amine, no longer protected by tissue storage, is rapidly metabolized, e.g., by monamine oxidase. ( c ) The interaction between tetrabenazine and reserpine (see Sections IV, A and IV, D, 3 ) suggests that these two types of drugs have similar mechanisms of action, For reserpine evidence exists that the storing capacity of the tissues for monamines is impaired (Pletscher et al., 1961). ( d ) Release of 5-hydroxytryptamine by tetrabenazine occurs also in vitro. Thus, tetrabenazine decreases 5-hydroxytryptamine in rabbit platelets (see Section IV, B ) and simultaneously increases free 5-hydroxytryptamine in the incubation medium, provided that monamine oxidase has been inhibited ( Paasonen and Pletscher, 1959) (Fig. 4). The storage mechanism of monamines is only poorly understood. In consequence it is difficult to explain the mode of action of substances like reserpine and tetrabenazine which impair this mechanism. A continuous exchange between free and stored monamines probably occurs in vivo and possibly also in vitro, whereby uptake and release are in a dynamic equilibrium. It has not been possible to decide whether in vivo tetrabenazine and reserpine act by increasing the liberation of monamines from the storage sites (i.e., increase of efflux) or by diminishing monamine uptake (i.e., reduction of influx). A further unexplained problem arises from some dissimilarities between benzoquinolizines and reserpine. Benzoquinolizines cannot decrease Shydroxytryptamine in brain to the same extent as reserpine, and their duration of action is shorter. These findings suggest certain differences in the mode of action of the two classes of drugs. 2. Inhibition of the Benzoquinolizine-lnduced5-Hydroxytryptmnine and Norepinephrine Decrease hj Monamine Oridase lnhibitors In the brain 5-hydroxytryptamine and catechol amines are metab-

290

A. PLETSCHER, A . BROSSI, AND K. F. GEY

k

1

2

3

FIG. 4. Liberation of 5-liydroxytryptamine from platelets of rabbit blood by tetrabenazine in vitro. Abscissa: time in hours after addition of tetrabenazine (final concentration of 20 fig/ml) to a platelet suspension in plasma; ordinate: 5-hydroxytryptamine in &ml corresponding to the amount of platelets and plasma respectively contained in 1 ml of the platelet suspension. Curves (from top to bottom) indicate: no tetrahenazine incubation, platelets; tetrahenazine incubation, platelets; tetrabenazine incubation, plasma; no tetrabenazine incubation, plasma. The animals were pretreated with the monamine oxidase inhibitor isocarboxazid [ 1-benzyl-2- ( 5-methyl-3-isoxazolylcarbonyl ) hydrazine, Marplan] (20 mg/kg intraperitoneally) 15 hours before removing the arterial blood. Spectrophotofluorometric determination of 5-hydroxytryptamine (Bogdanski et al., 1956). (From Paasonen and Pletscher, 1959).

olized to a considerable extent by monamine oxidase. Inhibition of monamine oxidase, together with impairment of monamine storing by benzoquinolizines, might cause an accumulation of free (nonstored) and a decrease of stored monamines. Therefore, after combined treatment with monamine oxidase inhibitors and benzoquinolizines, the total content of monamines would decrease less than after benzoquinolizines alone. The assumption of an accumulation of free monamines is supported by certain pharmacological effects seen after combined treatment with iproniazid and tetrabenazine (see Section V, A ) . This hypothesis has not been proven for benzoquinolizines. There exists, however, some experimental evidence that after pretreatment with monamine oxidase inhibitors reserpine

BENZOQUINOLIZINE DERIVATIVES

291

increases free monamines in the brain (Giarman and Schanberg, 1958). 3. Interaction of Tetrabenasinc with Reserpine

This interaction can probably be explained by assuming that tetrabenazine occupies the same receptor sites as reserpine in the system which is responsible for storage and release of 5-hydroxytryptamine. The benzoquinolizine, if administered first, would prevent the attachment of reserpine which in consequence probably disappears by metabolic breakdown and excretion. This hypothesis is supported by the finding that benzoquinolizines causing no depletion of brain amines also have no effect on the reserpine-induced depression of 5-hydroxytryptamine.' It has not been possible to explain why tetrabenazine counteracts only the reserpine-induced depletion of 5-hydroxytryptamine and not that of norepinephrine ( see Section IV, A ) .

V.

Pharmacology

A. NERVOUS SYSTEM The benzoquinolizines which lower cerebral monamines influence the behavior of animals in a manner similar to that of reserpine. These compounds decrease locomotor activity and aggressiveness in monkeys. Mice and rats show the typical attitude of reserpinized animals (curved backs). Even after high doses of the drugs the animals can still be aroused. Benzoquinolizines differ in intensity and duration of action, but none is as potent or as long acting as reserpine (Cahn and Herold, 1960; Leusen, 1960; Leusen et al., 1959a; Pletscher et al., 1958a; Quinn et al., 1959). In cats and dogs, for example, tetrabenazine in oral doses of 10-20 mg/kg has a sedative effect lasting 5-6 hours; Ro 4-1286 is about twice and Ro 4-1284 about five times as potent, but their duration of action is shorter. Ro 1-9288 in oral doses up to 20 mg/kg causes only weak sedation in cats and dogs.' The decrease in motor activity caused by benzoquinolizines can be confirmed by objective measurements, e.g., with the jiggle cage, the tread-wheel, or photographic methods. Spontaneous activity in rats and mice, as well as hyperactivity induced by caffeine and

292

A. PLETSCHER, A. BROSSI, AND K. F. GEY

metamphetamine, is depressed by benzoquinolizines. Tetrabenazine, Ro 4-1284, Ro 4-1286, and Ro 4-1860 are among the most effective compounds, their potency being about to l/lothat of reserpine, according to the method used.8 Monamine-depressing benzoquinolizines also induce ethanol potentiation and barbiturute narcosis in mice (Leusen, 1960; Leusen et al., 1959a; Pletscher et al., 1958a). Potency and duration of action differ, but arc in any case inferior to those of reserpine (Fig. 5 ) .

FIG. 5. Ethanol potentiation by benzoquinolizines and reserpine in mice. A dose of 4 gm/kg ethanol was injected intraperitoneally at various intervals after subcutaneous administration of the compounds. Abscissa : interval in hours between injection of the compounds and of ethanol; ordinate: sleeping tetrabenazine, 40 nig/kg; - - - - - - Ro 1-9571, time in minutes. Curves; 30 mg/kg; - - - Ro 1-9288, 30 mg/kg; reserpine, 2 mg/kg. Each point means the average of measurements in 20 animals. With ethanol alone the sleeping time did not exceed 1 minute. (From Pletscher d al., 1958a).

-

~

Lysergic acid diethylamide antagonizes the barbiturate potentiation caused by tetrabenazine (Cahn and Herold, 1960). Benzoquinolizines which do not lower cerebral monamines either cause no sedation and narcosis potentiation, or produce these effects only in high doses. Thus, 50-100 mg/kg Ro 4-1376, Ro 1-9564, and Ro 4-0786, intraperitoneally, produce moderate sedation in mice; the animals, in contrast to those treated with monamine depleting benzoquinolizines or reserpine, lie flaccidly extended.? * Experiments carried out by Dr. F. Steiner, Medical Research Department of F. Hoffmann-La Roche & Co. Ltd., Basel.

BENZOQUINOLIZINE DERIVATIVES

293

Apomorphine-induced emesis is counteracted by tetrabenazine in dogs. The effect of 40 mg/kg lasts for several days (Forster and Kunze, 1961; Leusen et al., 19594. Body temperature is depressed by various benzoquinolizines. In the rabbit, Ro 4-1284 and Ro 4-1286 have a definite effect in doses as low as 0.5-2.0 mg/kg; tetrabenazine is less potent. In mice, the latter and Ro 1-9288 cause hypothermia at room temperature as well as at elevated environmental temperature (32OC)? Fever induced by injection of Escherichia coli into rabbits or by yeast into rats is also depressed by tetrabenazine, Ro 4-1284, Ro 4-1286, and Ro 1-9288 in doses between 10 and -SO mg/kg (Pletscher et al., 19584.;, Doses of 20-100 mg/kg tetrabenazine enhance the convulsant efiect of pentylenetetrazole (as Metrazol ) . The sensitivity to electroshock in cats, however, does not seem to be changed by tetrabenazine (Lessin and Parkes, 1959), 3 Seizures in rats hypersensitive to audiogenic stimuli are not influenced by 20-40 mg/kg tetrabenazine and are diminished by very high doses (80-150 mg/kg) of the drug (Leusen et al., 1959a); in mice, benzoquinolizines (e.g., Ro 4-1284, Ro 4-1398) potentiate audiogenic seizures (Busnel and Lehmann, 1960; Lehmann and Busnel, 1959; Lehmann et al., 1957). In rats, tetrabenazine and other benzoquinolizines inhibit oonditionqd avaidance reactions in doses which have minimal effect on the unconditioned escape response (Heise, 1960a, b; Knoll, 1959; Knoll and Knoll, 1959; Pletscher et al., 1959a) (Fig. 6 ) . Furthermore, tetrabenazine depresses the rate of responding in a nondiscriminated avoidance procedure developed by Sidman ( 1953) (Heise, 1960a). The compounds are active for 3-5 hours in doses (subcutaneous or intraperitoneal ) of 0.2-4.0 mg/kg. In the electroencephalogram (EEG) of rabbits, 40 mg/kg tetrabenazine cause a pattern of slow waves with high voltage interrupted by bundles of spikes (Cahn and Herold, 1960; Mathis and Fischgold, 1960). Also in humans, chronic oral administration of tetrabenazine slows spontaneous EEG activity and increases the amplitude ( Bente, 1960). In patients with posttraumatic cerebral damage, tetrabenazine counteracts the depressed alpha rhythm. In 'Experiments carried out by Drs. M. W. Parkes and A. W. Lessin, Medical Research Department of Roche Products Ltd., London.

294

A. PLETSCHER, A . BROSSI, AND K. F. GEY

0

2

L

6

0.1

a2

a5

1

2

5

10

FIG. 6. Effect of benzoquinolizines on conditioned and nonconditioned escape reactions of rats. Stimulus for the unconditioned reaction: electrical shock. Stimulus for the conditioned reaction: sound of bell and simultaneous flash of light. Abscissa (left-side graphs) : hours after intraperitoneal injection of 4 mg/kg tetrabenazine or 0.25 mg/kg Ro 4-1284; (right-side graphs): dose of the drugs. Measurements: 2 hours after their injection. Ordinate: suppression of the responses of the animals. Upper graphs: tetrabenazine. Lower graphs: Ro 4-1284. Curves: _ _ conditioned reflexes; - - _ - - - unconditioned reflexes. (From Pletscher et al., 1959a).

this respect tetrabenazine is superior to other drugs (hypnotics, tranquilizers, etc.) ( Mathis and Fischgold, 1960). The spontaneous activity of the reticuhr formation of the brain stem in cats and dogs is not influenced by tetrabenazine (10 and 20 mg/kg). Ro 4-1284 (0.1 and 1.0 mg/kg intravenously) even has a stimulant effect in cats.l0 The hypertensive action of epinephrine applied locally to the cerebral cortex is increased by tetrabenazine (Minz and Walaszek, 1959). Furthermore, the drug diminishes the transport of sodium through the blood-brain barrier in rats but increases the phosphate transfer ( Quadbeck, 1960). Tetrabenazine and other benzoquinolizines (e.g., Ro 4-1284, Ro 4-1286, Ro 4-1860) also influence the autonomic nervous system. lo Experiments carried out by Dr. W. Schallek, Biological Research Department of Hoffmann-La Roche Inc., Nutley, N. J.

BENZOQUINOLIZINE DERIVATIVES

295

They cause miosis, narrowing of the palpebral fissure, and lacrimation, and also enhance the reflex which induces active closure of the eyelids and miosis in light (Bogdanski et al., 1961; Cahn and Herold, 1960; Leusen, 1960; Leusen et al., 1959a b) . Monamine oxidase inhibitors counteract or reverse the effect of monamine-depleting benzoquinolizines on the central nervous system. Thus, after pretreatment with iproniazid, tetrabenazine no longer produces sedation but may even cause excitation, mydriasis, pilo-erection, hyperthermia, etc. (Fig. 7 ) (Pletscher et al., 1958a).

FIG. 7. “Reversal” of tetrabenazine action by iproniazid. Above: intravenous injection of 40 mg/kg tetrabenazine. Below: intravenous injection of 40 mg/kg tetrabenazine 16 hours after 100 mg/kg iproniazid intravenously. ( From Pletscher 1959).

Iproniazid also counteracts ethanol potentiation due to tetrabenazine. Various benzoquinolizines which have a hypothermic action in rabbits induce hyperthermia after pretreatment with monamine oxidase inhibitors such as hydrazines, tranylcypromine,

296

A. PLETSCHER, A. BROSSI, AND K. F. GEY

harmaline (Cahn and Herold, 1960; Heise, 196Oa; Leusen, 1960; Pletscher et al., 1958a). These effects suggest an .accumulation of free monamines (e.g., catechol amines) in the brain (Pletscher et al., 1961).The pharmacological actions of benzoquinolizines on the nervous system mostly resemble those of reserpine, but are generally less intensive and of shorter duration. Tetrabenazine also decreases several pharmacological effects of reserpine. Thus, rats and rabbits given tetrabenazine followed by reserpine show little or no sedation 15 or 24 hours later. Animals treated with reserpine alone still are markedly sedated after this time (Leusen, 1960; Quinn et al., 1959).

B. CARDIOVASCULAR SYSTEM

1. Blood Pressure Some benzoquinolizines do not markedly influence blood pressure; others cause hypotension. One of the most hypotensive compounds found thus far in the benzoquinolizine series is Ro 1-9288. In anesthetized dogs and cats, 1-2 mg/kg induce a fall in blood pressure of 20-40 mm Hg. In anesthetized rabbits and in conscious cats and rabbits, EL10 mg/kg Ro 1-9288 also decrease arterial pressure. The effect in rabbits is less marked than that in dogs and cats. Ro 1-9288 differs from reserpine in that it produces hypotension almost immediately after intravenous injection. The duration of action in cats is about 23 hours. In animals with different types of experimental hypertension (unilateral nephrectomy in cats, the remaining kidney being wrapped in a Perlon capsule; ligation of the carotid arteries in rabbits; rats treated with DOCA, renin, or spermine), Ro 1-9288 decreases blood pressure in doses ranging from 2 to 20 mg/kg.ll. l 3 The mechanism of action of Ro 1-9288 is not fully understood. Since the drug decreases the carotid sinus reflex, it probably acts on the central nervous system. In addition, the compound affects the "Experiments carried out by Dr. B. Pellnlont, Medical Research Department of F. Hoffinann-La Roche & (20.Ltd., Basel. Experiments carried out by Dr. L. 0. Randall, Biological Research Department of Hoffman-La Roche Inc., Nutley, N. J. '3Experiment~carried out by Dr. H. P. Bachtold, Medical Research Department of F. Hoffmann-La Roche & Co. Ltd., Basel.

BENZOQUINOLIZINE DERIVATIVES

297

peripheral arteries, since a hypotensive effect is also seen in decapitated animals, Furthermore, Ro 1-9288 increases blood flow in the isolated ear and hind leg of the rabbit, indicating peripheral vasodilation. Ganglionic transmission is not inhibited. The effects of epinephrine and acetylcholine on blood pressure are not markedly influenced. In cats, however, Ro 1-9288 strongly antagonizes the 5-hydroxytryptamine-induced rise in blood pressure.11*l2 Ro 4-1284, Ro 4-1286, and Ro 4-1860 also decrease blood pressure,lj their potency and duration of action being similar to, or slightly less than, those of Ro 1-9288.11 Tetrabenazine has a less potent hypotensive effect than Ro 1-9288 and reserpine. In anesthetized dogs and cats, intravenous doses up to 5 mg/kg cause no consistent fall in blood pressure and doses of 10-20 mg/kg are only moderately hypotensive. In rats with renal hypertension administration of 10 mg/kg tetrabenazine daily for 1 week has no effect on blood pressure. The drug also causes no direct vasodilation in the isolated rabbit ear.", 12, As with reserpine, the rise in blood pressure due to epinephrine tends to be enhanced by tetrabenazine in cats.'l In bivagotomized dogs, however, tetrabenazine decreases the pressor effect of epinephrine and norepinephrine ( Leusen et al., 1959a). 137

2. Cardiac Rhythm Ro 1-9288 decreases frequency and amplitude of the isolated perfused heart of the cat only in concentrations as high as 100 pg/ml. Concentrations of 0.1-10.0 pg/ml tetrabenazine increase frequency and amplitude of the heartbeat.l' In dogs the heart rate remains uninfluenced or is only slightly accelerated by slow intravenous injection of 5 mg/kg tetrabenazine ( Leusen et al., 1959a)." C. GASTROINTESTINAL TRACT

Benzoquinolizines, in contrast to reserpine, have little effect on the gastrointestinal tract. A dose of 50 mg/kg tetrabenazine does not induce diarrhea in rabbits (Quinn et al., 1959). In conscious dogs, however, the same dose sometimes causes defecation ( Leusen et al., 1959a). In the intestine in situ (anesthetized rabbits), high doses of these drugs slightly increase tone and amplitude. In the isolated l4 Experiments carried out by Dr. H. Besendorf, Medical Research Department of F. Hoffmann-La Roche & Co. Ltd., Baael.

298

A. PLETSCHER, A . BROSSI, AND K. F. GEY

ileum of rabbits, tetrabenazine and Ro 4-1284 inhibit the amplitude of spontaneous contractions as well as acetylcholine- and bariuminduced spasms, but only in concentrations as high as 100 &ml. In anesthetized cats, ga&& secretion (total volume, acidity, and pepsin production) is moderately enhanced by tetrabenazine and Ro 4-1284. The number of experimental gastric ulcers produced in rats fed exclusively with glucose cannot be increased by 10 mg/kg of tetrabenazine; 1 mg/kg of reserpine, however, causes a significant increase. Tetrabenazine in mice causes less gastric hemorrhage than reserpine ( Blackman et al., 1959; Zbinden et al., 1959).I4

D. OTHERACTIONS In homognates of mouse brain, 1-2 x lo-* moles/liter tetrabenazine causes 50% inhibition of peroxide formation as well as of oxidation of endogenous substrates (Eberhard et al., 1961). In oioo, various effects of biogenic amines are counteracted by tetrabenazine, e.g., the bronchoconstrictor action of 5-hydroxytryptamine and histamine in guinea pigs and the local edema produced by injection of 5-hydroxytryptamine in the rat's paw.13 Ro 4-1284 injected intraperitoneally, like reserpine, increases lactic acid in blood of rats, with levels reaching a maximum after 2

20'.

0

1

2

3

L

FIG. 8. Increase of lactic acid in rat blood by tetrabenazine. Abscissa: hours after intraperitoneal injection of 30 mg/kg tetrabenazine; ordinates: left, mg/100 ml lactic acid in blood serum; right, per cent increase in comparison to controls. Each point represents average values f standard error of four estimations ( Barker and Summerson, 1941 ). ( From Gey and Pletscher, 1961).

299

BENZOQUINOLIZINE DERIVATIVES

hours (Gey and Pletscher, 1961). This effect is possibly due to liberated monamines, such as 5-hydroxytryptamine, catechol amines, which are known to raise lactic acid levels in the blood (Fig. 8 ) . Detailed investigations of other metabolic effects have not been carried out.

E. TOXICITY The acute toxicity of various benzoquinolizines in different animal species is considerably less than chat of reserpine (Table IV) .I1 LDso (MG/KG) Animal Mice, intravenousb subcutaneous oral

OF

TABLE IV VARIOUSBENZOMIJINOLIZINES~

TetraRo Ro benazine 1-1Y288 4-1284 150 400 550

TO 210 350

140 580 710

Ro 4-1286 100 250 290

Rabbits, intravenousb subcutaneous oral

500

650

Rats, subcutaneous

250

400

(I

50

Ro

Ro

4-1860 4-1398 200 250 500

50

40

>200

For mcthod, see Miller and Taintcr (1944). Injection during one minute.

Chronic oral administration of tetrabenazine in daily doses of 8 and 15 mg/kg for 13 weeks in growing rats and of 50 mg/kg five times weekly for 4 weeks in rabbits is well tolerated. These doses do not depress normal growth in rats. No pathological changes in the blood picture are observed and histological examination of various organs (liver, kidney, adrenals, spleen, pancreas, testes, epididymis, heart, lungs, bone marrow ) reveals no abnormalities. On chronic oral administration in rats and rabbits the tolerance of Ro 4-1284, Ro 4-1286, and Ro 4-1860 is similar to that of tetrabenazine or better.15 '6Experiments carried out by Drs. G. Zbinden, A. Studer, E. Lauppi, and M. Piller, Medical Research Department of F. Hoffmann-La Roche & Co. Ltd., Basel.

300

A. PLETSCHER, A. BROSSI, AND K. F. GEY

VI. Clinical Action

Tetrabenazine, in average daily doses of 75-150 mg, is used in the treatment of psychiatric disorders. According to various authors, the compound is especially beneficial in relieving hallucinations and paranoid symptoms of schizophrenia. In this respect it seems to be superior to neuroleptics of the reserpine-chlorpromazine type. Furthermore, the compounds exert a sedative effect in psychomotor states of excitation, though here its action is less pronounced and less constant than that of reserpine or chlorpromazine. Tetrabenazine generally takes effect more rapidly than reserpine (Espinosa, 1960; Heinze, 1960; Lende, 1960; Nieto and Castellanos, 1959; Smith, 1960; Stockhausen, 1960a,b; Stumpf, 1960; Ulett, 1960; Voelkel, 1958, 1960; Voelkel and Dresler, 1959; Weckowicz et al., 1960). Remarkable results have also been achieved in severe chronic schizophrenia resistant to other types of treatment (insulin and electroshock, chlorpromazine, reserpine). Of 134 patients, 41 (31%) improved markedly; 27 (20%) of them could even be discharged from the hospital. Tetrabenazine seems to be most effective in euphoric hebephrenia with superimposed mania ( von Brauchitsch, 1962a, b ) . Tetrabenazine has also been found beneficial in Huntington’s chorea ( Sattes, 1960; Brandrup, 1960). Side g e c t s of tetrabenazine are in general similar to those of reserpine but are rather less severe. With higher doses parkinsonism may occur, but in most cases this disappears promptly when the drug is withdrawn or the dose reduced, Significant drop in blood pressure is rare and in general appears only with high doses. Other side effects reported are insomnia, hyperactivity, turbulence, drowsiness, mental depression, spontaneous lactation, and menstrual disturbances. In severe schizophrenia the optimal dose seems to be rather close to the dose which causes side effects. Preliminary results with Ro 4-1284 and Ro 4-1860 indicate that these drugs also have a sedative action in agitated and paranoid psychoses (Voelkel, 1960). VII. Relationship between Effects on Monamine Metabolism and Pharmacological Action

It has been postulated that some pharmacological effects (e.g., sedation, hypotension, gastrointestinal disturbances ) arise from

BENZOQUINOLIZINE DERIVATIVES

301

changes in monamine metabolism, but definite proof is lacking. It may therefore be of interest to inquire whether benzoquinolizines give additional information.

A. SEDATION The following findings support the assumption that sedation is causally related to changes in cerebral monamine metabolism: 1. Numerous benzoquinolizines, such as tetrabenazine, Ro 4-1284, Ro 4-1860, produce reserpine-like sedation although they are chemically unrelated to Rauwolfia alkaloids. Like reserpine, the benzoquinolizines decrease 5-hydroxytryptamine, norepinephrine, and dopamine in the brain. 2. Benzoquinolizines are less effective than reserpine with respect to both sedation and depletion of cerebral monamines. Within the series of 2-oxobenzoquinolizines, sedation and potentiation of narcosis parallel the extent and duration of monamine depletion in brain. No derivative has been found which lowers monamines in brain without producing sedation. 3. Pretreatment with monamine oxidase inhibitors counteracts the fall in cerebral monamines which normally occurs after administration of a benzoquinolizine such as tetrabenazine. Correspondingly, the sedative effect of monamine-depleting benzoquinolizines is abolished or reversed after pretreatment with monamine oxidase inhibitors. Here again, benzoquinolizines and reserpine behave similarly. Observations on the antagonism of tetrabenazine and reserpice might possibly indicate that depression of cerebral norepinephrine is not essential for sedation. Tetrabenazine has no influence on the reserpine-induced depletion of norepinephrine, but it counter-acts the reserpine-induced decrease of 5-hydroxytryptamine, as well as reserpine sedation (Quinn et nl., 1959). It remains to be proven whether the parallelism between sedation and 5-hydroxytryptamine is coincidental or causal, and whether sedation is due to a decrease of total, or an increase of free 5hydroxytryptamine. B. BLOODPRESSURE

Monamine depression in the brain is unlikely to be the major cause of the hypotensive effect of benzoquinolizines. Tetrabenazine markedly decreases 5-hydroxytryptamine, norepinephrine, and dopamine in the brain of various animals, but it has little effect on

302

A. PLETSCHER, A.

BROSSI,

AND K. F. GEY

blood pressure in rabbits, cats, and rats. In contrast, Ro 1-9288, which induces only a slight decrease in cerebral monamines, lowers blood pressure to a greater extent than tetrabenazine. The norepinephrine decrease in the heart is probably not the main mechanism of hypotension either, since the two effects are not closely correlated. Investigations to determine the influence of benzoquinolizines on the catechol amine content of the arterial system (e.g., arterioles), autonomic ganglia, and other organs, would be desirable.

C. GASTROINTESTINAL TRACT Experiments with benzoquinolizines have not undermined the hypothesis that changes in endogenous 5-hydroxytryptamine influence the physiology of the gastrointestinal tract. In contrast to reserpine, benzoquinolizines have only a slight effect on both intestinal S-hydroxytryptamine and gastrointestinal functions. VIII. Summary

Chemistry: Among the derivatives of 1,2,3,4,6,7-hexahydro-llbHbenzo[a]quinolizines the ketones ( IVb ), the secondary carbinols (Va), and the tertiary carbinols (VIa and VIb) show the most interesting biological properties, Drug metabolism: Tetrabenazine has a biological half-life of a few hours, varying according to the organ and the animal species. The drug shows a special affinity for body fat. Two metabolites have been identified. E8ect on Monamine Metabolism: Several benzoquinolizines, e.g., tetrabenazine, Ro 4-1284, Ro 4-1632,Ro 4-1860, and Ro 4-1286, decrease the 5-hydroxytryptamine and catechol amine contents preferentially in the brain. Compared to reserpine they are less potent but of shorter duration and faster onset of action. The monamine depression is counteracted by pretreatment with monamine oxidase inhibitors. In brain, tetrabenazine prevents the reserpine-induced decrease of 5-hydroxytryptamine, but not that of norepinephrine. In addition, tetrabenazine increases S-hydroxyindolylacetic acid in the urine. The effect of benzoquinolizines on monamine metabolism is probably due to an impairment of monamine storing in the tissue. Pharmacology: Monamine-depressing benzoquinolizines intluence the central and the autonomic nervous system, e.g., they induce

BENZOQUINOLIZINE DERlVATIVES

303

sedation, narcosis potentiation, hypothermia, enhancement of pentylenetetrazole convulsions, decrease of conditioned reflexes, and prevalence of parasympathetic nerve tone. These actions parallel the depression of cerebral monamines and are of shorter duration than those of reserpine. Monamine oxidase inhibitors antagonize or reverse several of these effects. Tetrabenazine prevents the reserpine-induced sedation. Some benzocjuinolizine derivatives cause moderate hypotension, probably by central and peripheral mechanisms unrelated to monamine depression in brain and heart. The effects of benzoquinolizines on gastrointestinal functions are less pronounced than with reserpine. Clinical efects: Tetrabenazine exerts a sedative action and has beneficial effects on various symptoms of schizophrenia as well as in Huntington’s chorea. The results obtained with the benzoquinolizine support the hypothesis that certain effects on the central nervous system, like sedation, might be mediated through changes in cerebral monamines. ACKNOWLEDGMENTS

We wish to thank Dr. A. Jenni for his valuable criticism and help in preparing this paper. We are also grateful to Dr. J. Osbond and Dr. 0. Schnider for their cooperation. REFERENCES

Ban, Y., Terashima, M., and Yonemitsu, 0. (1959). Chem. Ind. (London), p. 568. Barker, S. B., and Summerson, W. H. (1941). J. Biol. Chem. 138, 535. Battersby, A. R., and Garratt, S. (1959). J. Chem. Soc., p. 3512. Battersby, A. R., Openshaw, H. T., and Wood, H. C. S. (19’53).J. Chem. Soc., p. 2463. Bente, D. (1960). Psychkt. et Neurol. 140, 51. Blackman, J. G., Campion, D. S., and Fastier, F. N. (1959). Brit. J. P h a ~ m a ~ o l . 14, 112. Bogdanski, D. F., Pletscher, A., Brodie, B. B., and Udenfriend S. (1956). J. Phamuzcol. Exptl. Therap. 117, 82. Bogdanski, D. F., Sulser, F., and Brodie, B. R . ( 1961). J . Phamnacol. Exptl. Therap. 132, 176. Brandrup, E. ( 1960). Report at the meeting of the Dansk Neurologisk Selskab, January. Brodie, B. B. ( 1960). Diseases of Neroow System 21, Suppl. 107. Brodie, B. B., Pletscher, A., and Shore, P. A. (1956). J. Pharmacol. Exptl. Therap. 116, 9.

304

A. PLETSCHER, A. BROSSI, AND K. F. GEY

Brodie, B. B., Finger, K. F., Orlans, F. B., Quinn, G. P., and Sulser, F. (1960). J . Pharmucol. Exptl. Therap. 129, 250. Brossi, A., Lindlar, H., Walter, M., and Schnider, 0. (1958a). Helv. Chim. Acta 41, 119. Brossi, A., Chopard-dit-Jean, L. H., and Schnider, 0. (1958b). Helv. Chim. Acta 41, 1793. Brossi, A., Chopard-dit-Jean, L. H., Wiirsch, J., and Schnider, 0. (1960). Helv. Chim. Acta 43, 588. Burgstahler, A. W., and Bithos, Z. J. (1960). J. Am. C h . Soc. 82, 5466. Busnel, R. G., and Lehmann, A. (1960). J. physiol. (Park) 52,37. Cahn, J., and Herold, M. (1960). Psych&. et Neur01. 140, 210. Carlsson, A., Rosengren, E., Bertler, A., and Nilsson, J. (1957). In “Psychotropic Drugs” (S. Garattini and V. Ghetti, eds.), p. 372. Elsevier, Amsterdam, Netherlands. Eberhard, F., Wilke, G., and Ansorg, W. (1961). Psychophumologiu 2, 1960. Espinosa, J. (1960). Rev. esputi. Oto-neuro-oftalmul. Neurocir. No. 107, 3. Forster, W., and Kunze, M. (19f3l). Arch. exptl. Pathol. P h u d o l . NuunynSchmiedeberg’s 241, 195. Gey, K. F., and Pletscher, A. ( 1961).Expe-rientia 17,25. Giarman, N. J., and Schanberg, S. (1958).Biochem. PharmucoE. 1,301. Grussner, A., Jaeger, E., Hellerbach, J., and Schnider, 0. (1959). Helv. Chim. Acta 42, 2431. Heinze, H. (1960). Psychiat. et Neurol. 140, 57. Heise, G. A. (1960a). J. Phurmcol. Exptl. Therap. 129, 155. Heise, G. A. (1960b). Diseases of Nervous System 21,111. Itoh, N., and Sugasawa, S. (1959). J. Org. Chem. 24, 2042. Knoll, J. ( 1959). Arch. exptl. Pathol. Phurmukol. Nuunyn-Schmiedeberg’s 236, 92. Knoll, J., and Knoll, B. (1959). Arzneimittel-Forsch. 9, 633. Lehmann, A., and Busnel, R. G. (1959). Arch. sci. physbl. 13, 193. Lehmann, A., Halpern, B. N., and Busnel, R. G. (1957). J. physiol. (Puris) 49, 265. Lende, N. ( 1960). Diseases of Ne-rvous System 21, Suppl. 118. Lessin, A. W., and Parkes, M. W. (1959).Brit. J. Pharmucol. 14, 108. Leusen, I. ( 1960).Psychiat. et Neurol. 140, 154. Leusen, I., Lacroix, E., and Demeester, G. ( 1959a ). Arch. intern. Pharmacodynm i e 119,225. Leusen, I., Lacroix, E., and Demeester, G. (1959b). Actu Physiol .& Pharmacol. N m l . 8, 92. Lombardino, J. G., Bodin, J. I., Gerber, C. F., McLamore, M. W., and Laubach, G. D. (1959). Abstracts of meeting of the American Chemical Society, Atlantic City, September. Lombardino, J. G., Bodin, J. I., Gerber ,C. F., McLaniore, M. W., and Laub a d , G. D. ( 1961).J. Med. Phm. Chem. 3,505. Mathis, P., and Fischgold, H. ( 1960). Psychiat. et Neurol. 140, 41. Miller, L. D., and Tainter, M. L. (1944). Proc. SOC. Exptl. Biol. Med. 57, 261.

BENZOQUINO1,IZINE DERIVATIVES

305

Minz, B., and Walaszek, E. J. ( 1959). Ann. N. Y. A c d . Sci. 80, 617. Naike, Y. (1959). J. Pharm. Soc. Japan 79, 1519. Nieto, D., and Castellanos, C. (1959). Neurol. Psiquiat. 1, 32. Osbond, J. (1961). J. Chem. SOC. 1961,4711. Paasonen, M. K., and Pletscher, A. ( 1959). Experientiu 15, 477. Pletscher, A. ( 1957a). Science 126, 507. Pletscher, A. ( 1957b). Schzoeiz. mecl. Wochschr. 87, 1532. Pletscher, A. ( 1959). In “Psychopharmacology Frontiers” (N. J. Kline, ed.), p. 395. Little, Brown, Boston, Massachusetts. Pletscher, A., and Gey, K. F. (1960). Psychiat. et Neurol. 140, 165. Pletscher, A., Shore, P. A., and Brodie, B. B. (1955).Science 122, 374. Pletscher, A., Shore, P. A., and Brodie, B. B. (1956). J. Phamtacol. Exptl. Therap. 116, 84. Pletscher, A., Besendorf, H., and Bachtold, H. P. (1958a). Arch. exptl. Pathol. Phamnukol. Naunyn-Schmiedeberg’s 232, 499. Pletscher, A., Besendorf, H., and Biichtold, H. P. (195813). Cmfinia Newol. 18, 165. Pletscher, A., Steiner, F., Voelkel, A. ( 1959a). In “Neuro-Psycho-Pharmacology” (P. B. Bradley, P. Deniker, and C. Radouco-Thomas, eds.), p. 138. Elsevier, Amsterdam, Netherlands. Pletscher, A., Besendorf, H., and Gey, K. F. (1959b). Science 129, 844. Pletscher, A., Gey, K. F., and Zeller, P. (1961). I n “Advances in Drug Research (E. Jucker, ed.), p. 417. Birkhiiuser, Basel, Switzerland. Quadbeck, G. ( 1960). Psychiat. et Neurol. 140, 29. Quinn, G. P., and Brodie, B. B. (1959). Pharmacologist 1, 74. Quinn, G. P., Shore, P. A., and Brodie, €3. B. ( 1958). Federation PTOC. 17, 404. Quinn, G. P., Shore, P. A., and Brodie, B. B. (1959). J. Pharmucol. Exptl. Therap. 127, 103. Sattes, H. (1960). Psychiut. et Neurol. 140, 13. Schwartz, D. E., and Rieder, J. (1961). Clin. Chim. Acta 6, 453. Schwartz, D. E., Pletscher, A,, Gey, K. F., and Rieder, J. (1960). Helu. Physwl. et Pharmacol. Acta 18, 10. Shore, P. A., and Brodie, B. B. (1957). PTOC.Soc. Exptl. Biol. 94, 433. Shore, P. A., and Olin, J. S. (1958). J. Pharmacol. ExptE. Therap. 122, 295. Shore, P. A., Silver, S. L., and Brodie, B. B. (1955).Science 122, 284. Sidman, M. (1953). Science 118, 157. Smith, M. E. (1960). Diseases of Nervous System 21, Suppl. 120. Stockhausen, F. C. (1960a). Psychiat. et Nezrrol. 140,61. Stockhausen, F. G. (1960b). Disease of Nervous System 21, Suppl. 115. Stumpf, W. (1960). Psychiat. et Neurd. 140, 63. Stumpf, W., Graul, E. H., and Hundshagen, H. ( 1961).ArzneimittebForsch. 11, 47. Suzuta, Y. (19859). J. Phurm. SOC. Japan 79, 1314, 1319. Tani, C., and Ishibasi, K. (1957). 1. Pharm. Soc. Japan 77, 324. Udenfriend, S., and Weissbach, H. (1958). Proc. SOC. Exptl. Biol. 97, 748. Ulett, G. A. ( 1960 ) . Abstracts of 2nd meeting of the Collegium Intemationde Neuro-Psycho-Pharmacologicum, Basel, July.

306

A . PLETSCHER, A. BROSSI, AND K. F. GEY

Voelkel, A. ( 1958). Confiniu Neurol. 18,144. Voekel, A. (1960). Psychid. et Neural. 140, 36. Voelkel, A., and Dresler, J. C. ( 1959). Nervenarzt 30, 322. von Brauchitsch, H. ( 1962a). Nervenarzt 33, 60. von Brauchitsch, H. ( 1962b). ( I n press.) Weckowicz, T. E., Ward, T., and Hoffer, A. (1960).Am. J. Psychiat. 116, 930. Yamazaki, T. hl. (1959). J. Pharm. SOC.Japan 79,1003,1008, 1014. Zbinden, G., Pletscher, A., and Studer, A. (1959). Schweiz. m d . Wochschr. 89, 289.

THE EFFECT OF ADRENOCHROME AND ADRENOLUTIN O N THE BEHAVIOR OF ANIMALS AND THE PSYCHOLiOGY OF MAN By A . Hoffer Psychiatric Services Branch. Department of Public Health. and Department of Psychiatry. University of Saskatchewan. Sorkatoon. Saskatchewan

I . Introduction . . . . . . . . . . . . . . I1. Biochemistry of Adrenochrome . . . . . . . . I11. Action of Adrenochrome on Cells . . . . . . . . IV . Effect of Adrenochrome on Fish . . . . . . . . V. Effect of Adrenochrome on Spiders . . . . . . . VI . Effect of Adrenochrome on Pigeons . . . . . . . VII . Effect of Adrenochrome on Mammals . . . . . . A. Mice . . . . . . . . . . . . . . . B.Rats . . . . . . . . . . . . . . . C. Rabbits . . . . . . . . . . . . . . D . Cats . . . . . . . . . . . . . . . E . Dogs . . . . . . . . . . . . . . . F. Monkeys . . . . . . . . . . . . . . VIII . Effect of Adrenochrome on Electrograms . . . . . IX . Effect of Adrenochrome and Adrenolutin on Humans . . A. Adrenolutin Double Blind Experiments . . . . . B. Prolonged Reactions to Adrenochrome and Adrenolutin C . Effect of Adrenochrome on Humans as Reported by Laboratories . . . . . . . . . . . . . D . Adrenochrome Given Sublingually . . . . . . E . Potentiation of the Action of Adrenochrome . . . . F. Discussion . . . . . . . . . . . . . X . Mode of Action of Adrenochrome . . . . . . . . XI . Conclusions . . . . . . . . . . . . . . References . . . . . . . . . . . . . .

. . . . . .

. . . . . .

307 311

314 315 316 316 . . 320 . . 320 . . 320 . . 324 . . 325 . . 333 . . 334 . . 335 . . 337 . . 337 . . 348 Other . . 353 . . 357 . . 359 . . 360 . . 361 . . 365 . . 365

.

I Introduction

Experimental psychiatry has been handicapped because the diseases of man most characteristically psychiatric cannot be reproduced in animals. The inner experience of the mentally ill patient 307

308

A. HOFFER

as transmitted to another person distinguishes the practice of psychiatry from the other medical specialties. Research psychiatrists, lacking experimental animals have been forced to use human models. Man has never lacked chemical substances growing all about him which have allowed him to enter strange states of mind. But for thousands of years, these were sought for purposes other than to reproduce models of the natural psychiatric diseases of man (Osmond, 1958). Despite repeated hints and direct suggestions, the use of psychotomimetic chemicals has developed slowly. Lewin (1931), de Jong (1945,1955,1956), Kluver (1928), Stockings (1940), Ellis (1898), and Mitchell ( 1896) had shown quite clearly how the ingestion of chemicals or plant extracts produced unnatural or unusual states, The similarities between these new states of mind and schizophrenia seemed quite remarkable to these workers. There has been notable resistance by many psychiatrists to using these chemicals for producing models of certain diseases. Perhaps this was due to the current preoccupation with psychological theories of etiology-especially of the so-called functional psychoses. Probably it was due to the natural reluctance of psychiatrists to produce a model of a disease which was so mysterious and baffling as schizophrenia. Most likely it was due to the primitive type of thinking which characterized psychiatry several decades ago so that differences stuck out much more clearly than similarities (see Bartlett, 1958). It seemed quite unlikely that experimentally produced psychiatric states, model psychoses, psychotomimetic experiences, or psychedelic experiences could advance our knowledge of the great functional psychoses. By the end of the first half of our century, Bleuler (1956) was secure in his belief that the experimental psychoses produced by lysergic acid diethylamide (which he had been one of the first to study) had nothing to offer to the student of schizophrenia. The discovery of Stoll and Hofmann (1943) of the hallucinatory action of d-lysergic acid diethylamide (LSD) quickened interest in the use of models. d-Lysergic acid amide, a compound very similar to LSD, is the active fraction of ololiuqui (Hofmann, 1960; Hofmann and Tscherter, 1960; Osmond, 195%). More recently, Hofmann et al. (1958) added psilocybine, the active principle of the Mexican

EFFECT OF ADRENOCHROME AND ADRENOLUTIN

309

hallucinogenic mushroom, to the growing list of psychotomimetic chemicals. Mescaline will produce schizophrenic-like states in man, but between 300 and 500 mg is needed. This quantity is easily detected in body fluids. For this reason, it was thought unlikely that natural animal substances with a similar order of activity could be present and escape detection. But when 100 p g of LSD was shown to be equally active, it became easier to believe that similar quantities of active natural substances could be present. After studying mescaline, Osmond and Smythies (1952) suggested that there might be present in the schizophrenic patient substances with the psychological properties of mescaline and with a structure similar to that of epinephrine, since the reactions of normal volunteers to mescaline resembled schizophrenia. In the course of our work on schizophrenia, Dr. Humphry Osmond and I also became interested in epinephrine derivatives which were indoles, as is LSD, rather than sympathomimetic amines like epinephrine or mescaline. In 1952, we presented our epinephrine metabolite hypothesis ( adrenochrome) to the Dementia Praecox Committee, Scottish Rite Masons, New York; and in 1954, our first formal report on adrenochrome appeared (Hoffer et al., 1954). The purest preparations of adrenochrome available between 1954 and 1957 were bright red or black powders which were very unstable even when stored under optimal conditions at low temperature and devoid of oxygen. Adrenochrome in aqueous solution deteriorated in minutes and formed insoluble brown-black melanins. Heacock et al. (1958) removed traces of silver ion from the preparations of adrenochrome and formed stable crystals which could be stored at room temperature. Later Heacock and Mahon (1958) synthesized stable adrenolutin and other reduction compounds of adrenochrome (Heacock and Scott, 1959; Heacock and Laidlaw, 1958). A comprehensive chemical review of adrenochrome was made by Heacock (1959a). These chemical studies have greatly facilitated the study of the psychological properties of adrenochrome and its derivatives. The adrenochrome hypothesis of schizophrenia has been discussed in some detail by Hoffer et al. (1954). Hoffer (1957a, b, c, d, 1958a, b, c, 1959a, b ) , Hoffer and Callbeck (1960), Hoffer and

310

A. HOFFER

Osmond ( 1955, 1959, 1960), Osmond (1955a, 1957), and Osmond and Hoffer (1958, 1959), and therefore will not be discussed further in this review. The behavioral changes produced in animals are interesting from two points of view. Little is known about the action in the body of animals of the adrenochrome degradation products of epinephrine. The fact that adrenochrome produces marked changes is interesting because it enlarges our understanding of the function of epinephrine and its derivatives. Furthermore, clear and definitive evidence that adrenochrome changes animal behavior would provide corroborative support for its psychological activity in man. This would then make an adrenochrome hypothesis of schizophrenia stronger. On the other hand, if these compounds were inert when given to animals, it would make the adrenochrome hypothesis weak. Thus, these studies, while not crucial in themselves to the adrenochrome hypothesis, can strengthen or weaken it. Crucial evidence, of course, would be the demonstration that adrenochrome or adrenolutin is present in the human body and in greater quantities in patients ill with schizophrenia. The evidence for the presence of adrenochrome has been adequately reviewed elsewhere in some detail. As I have previously stated (Hoffer, 1960), in an in vizho system containing substrate, i.e. epinephrine and its degradative enzymes, it is inherently probable that adrenochrome may occur. The following investigators have suggested or implied as a result of their own researches that adrenochrome was a metabolite of epinephrine: Bacq (1949), Blaschko and Schlossman (1940), Braines et al. ( 1959), Bullough ( 1952, 1955), Fellman ( 1958), Foley and Baxter ( 1958), Greig and Gibbons (1957), Grewal (1952), Iordanis and Kuchino ( 1959), Kisch ( 1947), Korzoff and Kuchino ( 1959), Kuchino ( 1959), Langemann and Koelle ( 1958), Meirowsky ( 1940), Roston ( 1960), Takahashi and Akabane ( 1960). These studies are substantial but more impressive than conclusive. The following investigators have concluded that adrenochrome is present, have provided evidence for the presence in living beings of oxidized derivatives of epinephrine, i.e., adrenochrome and/or adrenolutin, or have shown how adrenochrome can be transformed in vivo into adrenolution or 5,6-dihydroxy-N-methylindole: Altschule (1960), Bell et al. (1959), Fischer and Landtsheer (1950), Fischer and Lecomte ( 1951), Gershenovich et al. ( 1955), Golden-

EFFECT OF ADRENOCHROME AND ADRENOLUTIN

311

berg et al. (1950), Green et al. (1956), Hoffer and Kenyon (1957), Hoffer and Payza (1958), Jantz (1956), Kaufman and Koch (1959a, b), Lecomte and Fischer ( 1951), Maslova (1959), Nova1 et al. (1959a), Osinskaya (1957), Payza and Mahon (1959), Pickworth (1952), Rigdon ( 1940), Senoh and Witkop ( 1959), Sohler et al. (1961), Sulkowitch and Altschule (1959), Sulkowitch et al. ( 1957), Utevsky and Osinskaya ( 1957), Veech et d. (1960). The kind of scheme suggested by Gerard (1960) will be followed in describing the activity of adrenochrome and some of its derived compounds. The effect of adrenochrome upon simple systems will be considered first, then upon the more complex systems, upon simple animals, and finally, upon the most complex animal, man. The animals that have been given adrenochrome range from spiders, fish, and pigeons to the mammals-rats, cats, dogs, monkeys, and man. The sections on cats and on man will include much original data. Activity in animals has been reported so frequently and consistently that it can no longer be doubted that adrenochrome is indeed active when given to animals. Activity in man has received less corroboration but all the published accounts (as against references to unpublished works) are corroborative. It would be remarkable for a compound so active in lower animals and mammals to be inactive in man. 11. Biochemistry of Adrenochrome

The structural formulas of adrenochrome and some of its derivatives are shown in Fig. 1. The instability of adrenochrome, adrenolutin, and 5,6-dihydroxy-N-methylindole in aqueous solution leads to great difficulties in working with these substances. Biochemists, unaware of this difficulty, have consistently failed to demonstrate the presence of adrenochrome in body fluids. Since adrenochrome is decomposed by many solvents used in paper chromatography, e.g., propanolammonia-water and n-butanol-acetic acid-water, this is not surprising. Distilled water and 2% acetic acid are suitable solvents (Heacock, 1959b). Reducing agents discharge the red color of adrenochrome solutions and cause an internal rearrangement of the molecule, forming quantities of 5,6-dihydroxy-N-methylindole and adrenolutin. The

312

A. HOFFER

H: f - - - J J +o

/

0

I CH,

5,6-Dihydroxy- N me thylindole

Adrenochrome

Adrenolutin

o+J-J---:H /

0

I

,k H,C H

A

CH,

H,C H CH,

N-Isopropyinoradrenochrome

5,B-Dihydroxy-Nisopropylindole

FIG. 1. Formula of some adrenochrome derivatives.

relative amount of each depends upon the nature of the reducing agent and the conditions of the reaction. Blood plasma, which is adjusted to optimal conditions of pH, ionic strength, catalysts, etc., quickly oxidizes epinephrine to adrenochrome. Some of the adrenochrome is rapidly transformed into adrenolutin (Leach and Heath, 1956; Hoffer and Kenyon, 1957). Ascorbic acid and glutathione, two natural constituents of blood, influence this reaction. Apparently adrenolutin and 5,6-dihydroxyN-methylindole are relatively stable in plasma ( Melander, 1957). The same reactions occur in vivo when adrenochrome is injected. Fischer and Lecomte (1951) found that most of the injected adrenochrome was converted to adrenolutin in rabbits, dogs, and cats. Fischer and de Landtsheer ( 1950) found adrenochrome disappeared rapidly from blood and was converted by liver and kidney into

EFFECT OF ADRENOCHROME AND ADRENOLUTIN

313

adrenolutin. Jantz (1956) reported that blood serum obtained from alcoholic patients changed adrenochrome into adrenolutin more quickly than did normal blood serum. Hoffer and Osmond (1960) reported that adrenochrome injected into patients was rapidly removed from the plasma. In normal people the original control levels were reached after 30 minutes, but in schizophrenic patients, the plasma adrenochrome levels were still much above the preinjection value at 30 minutes. This also occurred in normal volunteers pretreated with LSD (but not when pretreated with bromo-LSD). Recently Sohler et al. (1961) found from radioactive tracer studies that adrenochrome injected in rats is partially converted to adrenolutin and 5,6-dihydroxy-N-methylindole. It is not surprising that these changes can occur, since blood does contain substances, such as hemoglobin, which catalyze these reactions. The natural enzymes of blood which oxidize epinephrine to adrenochrome are not well characterized. Leach et al. (19%) believed that ceruloplasmin was the enzyme which oxidized epinephrine. Epinephrine is readily autoxidized in pure aqueous solution to adrenochrome, but it is not likely that autoxidation plays a major role in blood. Blood contains substantial quantities of proteins, and reducing substances which inhibit autoxidation. This fact and the fact that the oxidation of epinephrine in plasma is greatest under optimal conditions, support the suggestion the oxidation is enzymatic. Ceruloplasmin oxidizes epinephrine but not as quickly as epinephrine oxidase. The properties of ceruloplasmin, the enzyme which oxidizes p-phenylene diamine (PPD), are quite different from the enzyme which oxidizes epinephrine. Ceruloplasmin or PPD oxidase is strongly inhibited by amine oxidase inhibitors (iproniazid, semicarbazide, hydroyuinone ) and by epinephrine and adrenolutin, whereas epinephrine oxidase is activated by semicarbazide and iproniazid. Epinephrine, which inhibits PPD oxidase, is the substrate for epinephrine oxidase. A comparison of these two enzymes is given in Table I. The effect of adrenochrome on many mammalian enzyme systems was summarized by Hoffer and Osmond (1960). Briefly, adrenochrome inhibits glycolysis in brain tissue under aerobic and anaerobic conditions, probably by inhibiting hexokinase and/or uncoupling oxidative phosphoiylation ( Cohen and Hochstein, 1960; Karzoff and Kuchino, 1959; Meyerhof and Randall, 1948; Park et al.,

314

A. HOFFER

TABLE I COMPARISON OF PROPERTIES OF CERULWLASMIN (p-PHENYLENE OXIDASE" DIAMINEOXIDASE)AND EPINEPHRINE Variable

PPD oxidase

Epinephrine oxidase

Substrate

p-Phenylene diamine Epinephrine

Epinephrine

Activators

Copper Sulfanilamide Hemoglobin

Copper Semicarhaside Iproniazid

Irihibitors

EDTA Semicarbazide Epinephrine Adrenolutin Iproniazid

Cysteine Ascorbic acid Tris buffer

Optimum p H Heat Optimum salt concentration

5.0 Inactivates 0.04 M

6.8 Inactivates 0.05 M

0

From Payaa and Zaleschuk (1959) and Payza and Hoffer (1959).

1956a, b; Radsma and Golterman, 1954; Randall, 1946; Walaas and Walaas, 1956; Woodford, 1959). Adrenochrome markedly inhibits decarboxylation of glutamic acid in brain tissue (Holtz and Westermann, 1956), oxidizes simple amino acids, and is polymerized to brownish melanin pigments in brain, intestinal mucosa, and skin. It is an antagonist of serotonin ( Stern et al., 1956). However, its action is not always inhibitory or toxic. Derouaux and Roskam (1949) found that sympathetic nerves in the rabbit's ear did not fatigue as rapidly in the presence of adrenochrome. On the other hand Marrazzi (1957) and Hart et al. (1956) reported adrenochrome inhibited synaptic transmission as did epinephrine. Ill. Action of Adrenochrome on Cells

It is not surprising that adrenochrome interferes with the growth and function of intact cells. Substances which inhibit respiratory reactions and glycolysis should be toxic for cells. Thus, Lettrk and Albrecht (1941), LettrQ (1954), and Frederic (1954) found that

EFFECT OF ADRENOCHROME AND ADRENOLUTIN

315

adrenochrome and adrenolutin inhibited mitosis of cells. Bullough (1952, 1955) showed that adrenochrome, both in vivo and in vitm, inhibited mitosis in mouse epidermis. The effect on chromosomes in vitro was confirmed by Gelfant ( 1960). Schwarzenbach ( 1957) reported that adrenochrome was a very strong inhibitor of spore germination of some fungi. Hoffer (1954) found that crude adrenochrome acted as a plant hormone for Avenu sutiva (oat) seedlings, i.e., it inhibited the rate of growth of rootlets relative to shoot growth. Plant hormones may be indolic, such as indolylacetic acid. Geiger (1960) demonstrated that adrenochrome is a very powerful toxin for cerebral neurons in pure culture. A concentration of 0.001 pg/ml ( 6 x moles/liter ) induced much more rapid and drastic changes in neurons than did either epinephrine or norepinephrine in the same concentrations. Recently, Dr. Ruth Geiger was good enough to show me a film she had made in which this action of adrenochrome on living neurons was demonstrated. The normal neurons pulsated slowly and rhythmically. When a small quantity of adrenochrome was added to the culture the cells began to pulsate more quickly and vigorously. Each cell appeared to develop contortions or convulsions in slow motion. After some time, the neurons rounded up in a spherical structure. Then the membrane must have ruptured for the cell disappeared leaving a spherical ring of dark fragments and pigmented material. LSD and serotonin also influenced pulsatile behavior but did not kill the cells (Geiger, 1960). Schizophrenic serum was also toxic. IV. Effect of Adrenochrome on Fish

The effect of adrenochrome on fish behavior has received little study. Abramson (1955) added epinephrine to an aquarium containing Siamese fighting fish. In time, the water turned pink indicating that there had been some oxidation of epinephrine to adrenochrome, but no changes in behavior were observed. In contrast, Abood (1957) added small quantities of pure adrenochrome made from epinephrine by phenolases and did find changes in the behavior of the guppy. These divergent results probably are owing to either a species difference or to the different compositions of the adrenochrome. A deteriorated solution of epinephrine would contain adrenochrome and other degradation products of epinephrine, such as H202,which might antagonize the action of adrenochrome.

316

A. HOFFER

V. Effect of Adrenochrome on Spiders

In 1954, Witt reported that the spider Zilla r-notGta was very sensitive to adrenochrome. (The adrenochrome had been made by a colleague in Bern. ) As little as 200 pg of adrenochrome in fresh or old solution could be biologically identified and differentiated by the effect of web building if it was given less than 10 hours earlier. [Photographs of the adrenochrome effect on webs are shown by Witt (1954, 1958).] The web was substantially disorganized, which is in contrast to the effects of LSD, which made the pattern more precise, and of mescaline, which disorganized it slightly. Adrenoxyl supplied by Labaz, Brussels, was not active. This is not surprising, since adrenoxyl is not hydrolyzed to adrenochrome in the body (Fischer and Lecomte, 1949). In December 1958, Witt tested adrenochrome and adrenolutin made by Heacock in our laboratory (Witt, 1961). The adrenochrome was given at the same time and in the same quantity as before but to Araneus diadematus, which responded like ZilZu x-notutu. Adrenochrome semicarbazone was inactive. Dr. Witt reported the following action of adrenolutin: “We tested the adrenolutin from you in and solution. From 7 spiders which had had the high dose only one built a web the following day. This might indicate that this dose has an effect on building drive. Twelve spiders received the lower dose of adrenolutin equal to the effective adrenochrome dose; 6 of these spiders built the following day but their webs showed no significant changes.” On the basis of this small number of experiments, Witt concluded that high doses of adrenolutin inhibited web building but that lower doses were not as active as adrenochrome. As a matter of interest, the spider web disorganization produced by adrenochrome seems quite specific. LSD and mescaline produce different patterns. However, serum from catatonic patients disorganized the web of Zilla x-notata in a way very similar to adrenochrome. This can be seen by examining the pictures of webs after treatment with adrenochrome and schizophrenic serum ( Bercel, 1960; Buehler, 1960). VI. Effect of Adrenochrome on Pigeons

About three years ago, I made a series of observations, assisted by Dr. H. Wojcicki, on the effect of adrenochrome and adrenolutin

EFFECT OF ADRENOCHROME AND ADRENOLUTIN

317

on the behavior of some pigeons. These were racing pigeons bred by Dr. Wojcicki, a psychiatrist and a pigeon fancier. Two pairs of birds were used, each weighing about 650 gm. These experiments were conducted over a 4-day period in October 1957 in my office. Each pair was kept in one cage, and the cages were placed side by side on a table. A. MALEPIGEONA Male pigeon A was a very fine proud bird. He was very solicitous for his mate, female A. He firmly attacked any of the two pigeons B if they were placed in his cage, and he clearly dominated male B. If male A was removed from the cage and kept a few minutes out of sight of his mate, he went through a vigorous and consistent courting behavior when he was returned. When placed on the edge of the cage door, he would quickly hop in and then would vigorously coo and strut around his female for a short time. At 4:04 PM, male A was injected intraperitoneally with 5 mg of d-adrenochrome synthesized by Heacock et al. (1958). When returned to the cage, he responded normally to his mate. Five minutes later, there was no change. Eight minutes later, female A was removed and a few minutes later returned to the cage. Male A did not court nor coo. At 10 minutes, male B was placed in the cage. Male A did not attack him as he would have done normally. But 2 minutes later, h e did fight and peck at male B. There was thus a delayed reaction to male B but there was no further evidence for any abnormality. Twenty minutes after the injection ( 4 : 2 4 ) , another 10 mg of adrenochrome was injected. One minute later, the bird was listless and his feathers were droopy and bedraggled. Four minutes later, he was passive and disinterested. He allowed H.W. to pull his beak with no protest. Six minutes later, he remained passive and did not court the female when she was returned to the cage after an absence. When male B was placed in the cage, h e did not attack him and appeared quite indifferent. Ten minutes later, male A was placed in cage B containing pair B. Normally, he would attack male B and court female B. This time, he did not attack male B and ignored female B for the several minutes he was left in the cage. When at 12 minutes, he was returned to female A, it was noted he was aphonic. His coo was a very weak and tremulous gurgle. Twenty minutes later, he was still able to fly normally but he sat on H.W.’s hand one minute before he flew away. In general, male A became indifferent to his environment. He was disinterested in females and did not follow his routine courting behavior with his mate which he had always done before the injection. The next morning male A was found dead. At autopsy, no pathological changes were found in the abdomen. It was apparently death due to adrenochrome [15 mg] and not to mechanical trauma caused by the injection.

B. FEMALE PIGEON A October 23, 1957, this bird was given 5 mg adrenochrome intraperitoneally at 4:13 PM. There was no observable change in behavior. The next morning,

318

A. HOFFER

she was normal. At 1O:OO AM she was injected with 10 mg of adrenochrome. Eight minutes later, she was normally combative when a hand was placed in the cage, but there was much less startle reaction to a loud noise. This was the only change observed. Four hours later, she was normal. October 30 at 9:35AM, she was given 10 mg of adrenolutin. Fifty minutes later, she was less aggressive and did not fight the hand when she was teased in her cage. Fifteen minutes later, she required a great deal of provocation before she would respond although she appeared alert. At 11:OO AM, she refused completely to fight back. At 11:05, she suddenly flew sharply against the side of her cage toward the adjacent cage B where male B was standing. But when male B was placed in her cage, they ignored each other. Four hours after the injection, female A was normal and aggressively drove out the attacking hand from her cage. November 1, 1957, she was given 25 mg adrenochrome at 2:05 PM. We wished to find out how much adrenochrome would kill this bird. Three minutes later, her feathers were fluffed and droopy and she vomited. For the next 12 minutes, she continued to retch and vomit and would not fight. Her eyes became glazed. One hour and 15 minutes after injection, she preened herself and began to recover. We estimated that her total reaction to 25 mg adrenochrome was less than her reaction to the 10 mg of adrenolutin. November 4, she was injected with 30 mg adrenochrome at 1:19 PM. She quickly became very ill as before. She could not fly and when released did not fly but fell down. At 4:OO PM, she was dead. Thus she required twice as much adrenochrome for death as her mate.

C. MALE PIGEONB Pair B were nesting two eggs. The nest was in one wrner of the cage and built of grass. This male was dominated by male A but was normally aggressive and fond of female B. He was devoted to her and dutifully did his share of brooding on the eggs. On October 24, male B was given 10 mg adrenochrome at 2:07 PM. Twenty-three minutes later, he vomited and had a tremor during which he stood and rocked back and forth. He blinked his eyes slowly and appeared sleepy, Three minutes later, his fighting behavior was normal but he had a slight tremor of his wing. Forty minutes after injection, his vocal ability was diminished and his voice became feminine, sad, and dull. Male pigeons lose their voices only when moulting. He did not court with enthusiasm and could hardly coo. One hour later, female A was placed in his cage. Normally male B would have driven her out but this time he ignored her. The next morning, he was normal. Male B was then given 20 nig adrenochrome together with 100 mg ascorbic acid. One hour later, there was a slight and fleeting change; a few minutes later, he was normal. One day later, he was still normal, but 2 days later he was found dead.

D. FEMALE PIGEON B Female B was brooding two eggs and was very motherly and protective. She fiercely drove off a hand reaching toward the nest. This bird was injected

EFFECT OF ADRENOCHROME AND ADRENOLUTIN

319

with 10 mg of adrenolutin. Five minutes later, she stood over her eggs but did not set on them. She then walked away from her eggs. Ten minutes after the injection she retched, then sat beside her eggs and blinked her eyes. Five minutes later, she was sick and her feathers drooped. She no longer retreated from the hand. Twenty minutes after the injection, H.W. placed her on his hand and walked from the office down a long corridor and back. She made no attempt to fly away but clung to his hand. This test we termed the Baruk test because he often demonstrated the effect of schizophrenic bile on pigeons in this way (Baruk, 1957). This pigeon was able to fly when she was released but was apparently not interested in flying. One hour later she had not returned to her nest. The next day, the bird was studied again. At 9:OO AM she appeared disinterested and did not set on her eggs. By 2:OO PM she was much more alert and responsive. She fought against the hand vigorously and was better groomed. Four days later, she appeared normal. She had again started to set on the dead eggs in the normal way. At 1:57 PM, she was given 20 mg adrenochrome. Three minutes later she was ill, rocked, and once more refused to set her eggs. Four minutes after injection she vomited. One and a half hours later she had apparently recovered. She was now very alert and vicious when the hand was placed near her. She came toward the hand to attack. Late that night she was dead.

E. DISCUSSION It is not possible to draw firm conclusions from this study on only four birds, but each bird had several injections and the results were definite. A larger series would be required in order to complete this kind of study. The following tentative conclusions may be made: ( 1 ) Male pigeons are apparently more susceptible to death from adrenochrome than female pigeons. ( 2 ) Adrenochrome in nonlethal doses (10 mg) alters behavior, including courting, protecting the nest, and brooding. ( 3 ) Adrenolutin is more effective in altering pigeon behavior than adrenochrome. In one case, catatonia was produced. Lehrman (1956a, b, 1958, 1959) and Lehrman and Wortis ( 1960) have examined the factors which influence cyclical breeding behavior in ring doves (Streptopelia risoria). This does not normally occur in birds kept in isolation from each other, According to these workers the presence of one bird stimulates the pituitary gland of the other; the pituitary hormones change courting behavior to nest-building, then to incubation. When birds were placed in cages with eggs they would normally begin to nest in 4 to 10 days. When progesterone was injected the birds nested almost immediately. Birds treated with estrogens sat after 1-3 days or after a

320

A. HOFFER

period of 11 days. Birds injected with prolactin appeared to be under considerable tension. In birds with no breeding experience (such as feeding squabs) the restlessness was not directed toward any specific response, whereas in birds with previous breeding experience the restlessness (tension) aroused approaches to the young. According to Lehrman, the hormones induce vascularity and tension in the birds’ brood patch which thus becomes a source of irritation. This is reduced by setting on the smooth, cool, hard surface of the eggs. It is possible that adrenochrome and adrenolutin interfered with the cyclical changes in the pigeons by markedly reducing tension. Hoffer and Osmond (1960) reviewed the evidence which suggested that adrenochrome and adrenolutin reduced tension in humans, which they ascribed to an anti-epinephrine action. Therefore, giving adrenochrome and adrenolutin to the pigeon which was brooding eggs might immediately reduce the tension and hence the urge to set. The fact that several days later, when the adrenolutin effect had worn off, the female again began to sit on her eggs supports this conclusion. Presumably the tension induced by the estrogens had by now returned. VII. Effect of Adrenochrome o n Mammals

A. MICE Not much has been written about the effect of adrenochrome on mice. Laborit et al. (1957a, b) found that both adrenochrome and adrenochrome semicarbazide increased the tendency of mice for convulsions. LSD had no effect. Serotonin protected the mice against convulsions. Glutamic acid gave some protection. Eade (1954) found that adrenochrome produced sedation in mice, and before death the sedation quickly passed off and was replaced by progressive spasmodic clonic convulsions beginning in the hind limbs. Walking was uncoordinated. When disturbed, the mice hopped about, often leaping in the air and falling over backward.

B. RATS 1. Review of Literature Eade (1954) gave adrenochrome to some rats in order to measure the lethal dose. Cause of death was respiratory paralysis

EFFECT OF ADRENOCHROME AND ADRENOLUTIN

321

preceded by dyspnea, micturition, clonic convulsions, and exophthalmos. Later, Noval et al. (1959a) showed that impure adrenochrome could produce a “de Jong type” catatonia in rats. A dose of 10 mg/kg of crystalline adrenochrome in saline injected intravenously caused convulsions and death in half the treated rats in less than 15 minutes. After 8 mg/kg their physical activity was greatly reduced for many hours, On observing some of these rats given adrenochrome, it was noted that for the first few minutes the animals appeared sick but this soon passed. After that, they were disinclined to move about; when placed alongside a cold Bunsen burner, the animals grasped the burner with their forelegs and clung until they sank slowly to the table, doubled up, apparently exhausted, but still clinging, Purer adrenochrome was one fourth as toxic and appeared to be incapable of inducing the clinging behavior ( Noval et al., 1959b); this suggested to these investigators that decomposition products of adrenochrome may have been responsible for the behavioral effect. Vallbo (1957) gave 20 mg/kg of adrenochrome to rats and saw no effect. However, 100 mg/kg of adrenolutin produced marked catatonia. The rear legs were quite relaxed. Neither chlorpromazine, promazine, perphenazine, nor acepromazine protected the rats against adrenolutin, but KABI-HdA8 did protect them completely. 2. The Efect of Adrenochrome on Conditioned Responses of A1bin0 Rats

Grof et al. (1961), studied the response of Wistar rats to adrenochrome. They weighed about 200 gm, and received 6 and 8 mg/kg. The control group received physiological saline. Two kinds of observations were made: ( a ) a test of general irritability and activity, and ( b ) the effect on conditioned reflexes. Rats given adrenochrome were significantly different from rats given saline in the orientation-searching experiments, There was a marked decrease in the duration of erect reactions ( t o 20% of the controls) and a marked increase in periods of immobility (which were doubled). Rats which were of a markedly inhibitive type showed enhanced excitation whereas those with medium or low inhibition became less irritable as the depth of inhibition was greater. The maximum inhibition of activity occurred 7 to 9 minutes after the injection. For the conditioning experiments, rats were trained in two

322

A. HOFPER

groups of 6 animals each. The unconditional impulse was an electric shock applied to the floor. The animal avoided shock by running up a net. The conditioning signal was an optical signal of low intensity. Each group of rats was trained for seven sessions with 61 trials. In the eighth session, one group received adrenochrome, the other saline. With the adrenochrome group, there was a significant ( P < 0.01) increase in the latency periods, which were doubled. There was also a decrease of intersignal reactions to one fifth of the control value ( P < 0.01). These authors concluded that adrenochrome acted as an inhibitor of higher nervous activity in most animals but produced excitation in extremely inhibited animals.

3. A Study of the Efject of Adrenochrome on Conditioning in Albino Rats' a. Effect on Acquisition. Wistar-strain albino rats (age 3 to 5 months) were used. A Mowrer-type shuttle box was used with a buzzer as the conditioning signal and an electric shock applied to the floor as the unconditioned stimulus. Adrenochrome was synthesized by Heacack. Freshly prepared solutions were injected at a dose of 25 mg/kg. The criterion of learning was 10 consecutive avoidance responses. The median number of trials with saline was 38.5 and with adrenochrome 98.5 ( P < 0.02) using 12 rats in each group. This experiment was replicated later with new animals. All saline-treated rats performed to criterion in less than 140 trials, whereas 8 out of the 12 rats given adrenochrome had not learned by 140 trials. Using 90 trials as a median with saline, 10 were below, whereas with adrenochrome only 2 were below ( P < 0.01 ) . A dose response study was later made with 10 rats in each group, using a randomized block design. The rats learned to criterion with saline and 6.25 mg, 12.5 mg, and 25 mg/kg of adrenochrome in 40, 52, 64, and 80 trials. For this range of dosage, the number of trials required to reach criterion of acquisition of the conditioned avoidance response (CAR) was a linear function of the logarithm of the dose given. When adrenochrome was given in a very large dose of 80 mg/kg, about one-third of the rats died. The survivors when tested 14 to 34 days later disclosed no residual effect in that they learned as well as Data of T. Weckowicz ( 1961).

EFFECT OF ADRENOCHROME AND ADRENOLUTIN

323

did rats not receiving adrenochrome. At autopsy, no changes were seen in the central nervous system. b. Egect on Performance. Two groups of 10 animals were used. One group received adrenochrome first and saline 1 week later. The other group received these substances in the reverse order. Before the experiment, the rats were trained to a criterion of 18 out of 20 trials. One hour after the injection, they were tested again and were given 20 trials in 15 minutes. Overall, the rats given adrenochrome made significantly fewer responses ( P < 0.01), owing to the poor performance of the group receiving adrenochrome first. Adrenochrome had much less effect on the group receiving saline first. The only difference was that animals which received saline first had received more training before being tested under adrenochrome. The difference in performance of the two adrenochrome groups was significant (0.05 < P < 0.01). Thus the more training the rats received, the less susceptible were they to adrenochrome. This is a very important observation, for it proves that the effect of adrenochrome was not due to marked physical weakness of the rats. In this experiment, it was also observed that rats given adrenochrome had many fewer anticipatory responses ( P < 0.01). With 12.5 mg/kg, there was no effect on performance, but acquisition of the CAR was decreased with this low dose. The ability of rats to jump away was tested by another method. An electric timer was connected to the stimulator which delivered 1.25 ma to the grid of the box. Both were then started and both stopped automatically when the rat escaped. Well-trained rats were used. Ten received adrenochrome, 25 mg/kg, and ten saline; all were tested 1 hour later. The mean speeds of escape were respectively 1.26 and 1.40 seconds. Thus, it is clear that gross motor impairment and/or complete lack of motivation may be excluded as the cause of lack of acquisition of the CAR. c. Efect on Extinction. Twenty rats trained to criterion (18 out of 20 trials) were used. Ten received adrenochrome and 10 saline. Extinction trials were started in blocks of 20 followed by 15 minutes rest in the cage. The criterion was 10 consecutive non-responses to the conditioning stimulus-the buzzer. The score was the number of trials necessary to reach the criterion. For the 10 rats given saline, the following number of trials were required: 10, 172, 12, 210, 10, 39, 70, 130, 50, and 50. For adrenochrome, they were 10,

324

A. HOFFER

10, 10, 10, 10, 11, 10, 13, and 20. Chi-square for a median of 115 was 5.00 (0.05 < P < 0.01). Adrenochrome significantly accelerated extinction of CAR. In most of the adrenochrome rats, it was extinguished at once, but of the animals given saline 3 rats required more than 100 responses. d. Eflect on Driue. When trained rats were used and the intensity of the electric shock was varied, there was a slight trend for saline rats to escape at a lower intensity of current, i.e., at 0.22 vs. 0.25 ma, but the difference was statistically insignificant. However, when untrained rats were used and observed for startle responses (including jumps) the saline threshold was 0.14 and adrenochrome 0.19 (0.05 < P < 0.01). The threshold for naive animals was lower than for sophisticated animals. The intensity of current was set at 1.25 ma for one group and 2.50 for the other. Animals with adrenochrome at a higher drive level performed slightly less well. It was concluded that level of drive (or arousal) was not involved in the way adrenochrome affected CAR. e. Eflect on Bar-Pressing. The rats were tested in a Skinner box with food and water as the reward. They had to run to one end, press the bar, and return to the other end for their reward. After a constant rate of bar-pressing was obtained, they were injected and tested for 13 minutes. Adrenochrome significantly ( P < 0.01) reduced the rate of bar-pressing. The difference was greater when the rats were less well-trained, i.e., received adrenochrome first, f. Conclusion. Adrenochrome decreased the acquisition of the conditioned avoidance response and markedly accelerated extinction. The prolongation of acquisition was related in a linear way to the logarithm of the dose. In a Skinner box, the rate of bar-pressing for food or water was markedly reduced.

C. RABBITS Krupp and Monnier (1960) studied the effect of adrenochrome on 12 rabbits. They observed these animals for changes in behavior, for changes in frequency of respiration and pulse, and for brain electrical activity. They injected 3.5 to 6 mg/kg of adrenochrome by vein. Behavior was influenced slightly, The animals reacted more strongly when stimulated by light or noise. There were slight changes in spontaneous motility of the free animals and a transient but slight decrease in pulse rate.

EFFECT OF ADRENOCHROME AND ADRENOLUTIN

325

Vallbo (1957) found that adrenochrome 20 mg/kg produced no change when given intravenously (50 mg per animal); however, adrenolutin at 50 mg per animal (20 mg/kg) produced a stupor lasting 30 to 90 minutes. It may well be that the first in vivo demonstration of the conversion of epinephrine into adrenochrome was made by Rigdon (1940) using rabbits. They were shaved 24 hours before the experiment, and epinephrine was injected intradermally. This blanched the skin for several hours. If the surface of the skin was treated with xylol either before or after the injection, the skin became reddish brown in color. This appeared 15 to 30 minutes after the application of xylol. The only reddish brown pigment known to be related to epinephrine is adrenochrome. Several years ago, an injection of adrenochrome subcutaneously in my left arm formed a small brown pigmented area which remained over 3 months, Meirowsky (1940) showed that the production of pigment in human skin is highly increased by adrenochrome. Recently I injected 1 mg of adrenochrome intradermally into shaved rabbit skin. A faint brownish color remained in the skin for many hours. These observations suggest that under certain conditions epinephrine can be converted into adrenochrome by epidermis. Walaszek (1960) reported that LSD, bufotenin, and d-adrenochrome potentiated in rabbits the pressor response which follows the topical application of epinephrine to the brain cortex.

D. CATS

1. Literature Review Schwarz et al. (1956b) made the first study of the effect of adrenochrome and adrenolutin on cats. (Their paper should be consulted for more detailed description of the behavioral effects. ) Pure adrenochrome (freshly prepared solution, pH 7.7), in doses ranging from 0.125 to 1 mg, was placed in the lateral ventricles, using permanent indwelling canulae. Adrenolutin (also freshly prepared solution, pH 7.6) was similarly injected, using doses of 300-650 pg. “Stupor and catatonia, as described by de Jong, were observed after the administration of adrenochrome and adrenolutin. After an initial gradual reduction in motor activity, the cats would finally sit in odd positions with their eyes wide open for long periods and howl. Although, for various reasons, the term ‘waxy flexibility’

326

A. HOFFER

cannot be used as it would be in referring to human beings, the cats permitted their limbs to be placed passively in unnatural positions without immediately correcting them, or even resisting such positions. In the presence of a clear sensorium, other bizarre forms of behaviour were noted to follow the injection of adrenochrome and adrenolutin. Frequently, the cats were in a deep trance, with concomitant electroencephalographic changes; but they could always be alerted readily.” (This description is from the study of Schwarz et al., 1956b.) Rice and McColl (1957, 1960), using a similar technique, gave a series of cats 0.6 mg/kg of adrenochrome. This resulted in a higher incidence of sympathetic rather than parasympathetic stimulation. Four minutes following the injection, animals began growling, meowing, and spitting. Eight minutes after injection a severe tonic seizure developed, which became tonic-clonic in nature. Following the convulsions the animals were immobile with flaccid muscles and they appeared semistuporous. It resembled the action of mescaline, rather than LSD. Schwarz et al. also observed this. All the cats given adrenochrome showed changes in habits, as did those given mescaline. None of the cats given LSD showed any change in habits. These authors observed that mescaline and adrenochrome appeared to affect predominantly various regions of the hypothalamus, diencephalon, medulla, higher brain stem, and motor cortex. Vallbo (1957) reported that 100 mg adrenochrome per cat produced little change. However after 50 to 100 mg, intravenous adrenolutin ( 10-30 mg/kg ) produced stupor and diminished interest in surroundings and in food; this lasted 15 to 30 minutes. According to Melander and MHrtens (1958, 1959), pretreatment with taraxein or LSD (15 to 20 pg per cat intravenously) markedly potentiated the effect and 2 to 3 mg/kg intravenously of adrenolutin produced drowsiness and muscle relaxation. Acetyl-LSD also potentiated adrenolutin but bromo-LSD did not. Sherwood (1957) gave one cat 0.5 mg of crystalline adrenochrome. During the next 3 hours, the cat showed inappropriate behavior. It walked into its bowl of milk and did not lick its paws. It let itself be placed in unusual positions without protest. It preferred to lie down and was disinterested in grooming. But throughout, it remained affectionate toward Dr. Shenvood and it seemed fully aware of its surroundings. Capek et al. (19.60), as summarized

EFFECT OF ADRENOCHROME AND ADRENOLUTIN

327

by Grof et al. (1961), also gave cats 1 to 2 mg of adrenochrome intraventricularly. The adrenochrome ( Light's ) produced changes similar to those described by Schwarz et al., whereas their own adrenochrome produced changes similar to those described for mescaline.

2. A Comparison of the Effect of Some Indoles on Cut Behavior Dr. J. D. McColl, of Frank Horner and Company, Montreal, placed indwelling permanent canula into the lateral ventricles of some cats using the procedure pioneered by Feldberg and Sherwood ( 1954),described by Rice and McColl (1960). The cats were delivered by air express from Montreal to Saskatoon and all arrived in good condition.2 Each cat was housed in its own cage, near each other in a special room where there were no other animals. The first time the cats were used, they were allowed to become familiar with each other since it was planned to work them in pairs in order to study their interaction. After that, the usual procedure was to inject one cat and then release the other so that both were free in the experimental room. Each experiment was a type of field experiment. No conditioning experiments were made at this time although in future work this will be done for a few of the chemical compounds. I was primarily interested in the qualitative effect of these compounds upon cat behavior and not on the quantitative relationship between dose and activity. After the injection, both cats were observed for 2 or more hours. The chemicals were all synthesized by Dr. Heacock and his staff in our biochemical research section and were freshly prepared in saline just before the injection, which was temporarily painful. There was no residual effect of the injection and after 2 or 3 minutes the cats seemed normal. Each compound was given to each of the 2 cats once. Adrenochrome with heparin was given to an additional 2 cats. In addition to adrenochrome, adrenolutin, 5,6-dihydroxy-Nmethylindole, and N-isopropylnoradrenochrome (see Fig. 1) , the following compounds were used: '1 wish to express to Dr. McColl my gratitude for his helpfulness in this phase of our work. This kind of cooperation compensates for the many frustrations one encounters in the normal course of research.

328

A. HOFFER

I

CHI Adrenochrome methyl ether

Itubcseririe

I

CH? II

CH, .V-Ethylnoradrenochrorne

3,4-Methylenedioxy-aaminomethylbenzyl alcohol oxalate

The effects of adrenochrome and its congeners administered intraventricularly in cats are summarized in Tables I1 and 111. These results were obtained from 2 cats; “BB,” a medium-sized male cat of nervous excitable disposition, and “Big,” a larger male cat, very friendly and affectionate toward observers. In addition, two other cats were the subjects of less extensive experimentation: “Sandy” was a very affectionate quiet cat, and his reaction to 2 mg adrenochrome was studied alone and with heparin in order to test the limit of the protective action of heparin. August 23, 1960, he was given 2 mg in 0.5 ml saline. In 1 minute, he was quiet ,and indifferent. In 5 minutes, he walked in a clumsy way with jerky movements. He was restless, yowled several times, then became very unsteady and began to crawl backward. Then he began to circle around himself yowling all the time. His hind legs were not paralyzed but he seemed to have lost control of them. At 9 minutes, he no longer responded to stimuli. At 15 minutes, his respiration was very fast; he panted, yowled, and then urinated. At 60 minutes, he had been yowling and scratching his head for 40 minutes. Ten minutes later, his righting reflex was gone. At 4 hours, he could not walk at all. There was no recovery at 6 hours. Two days later, he was still unable to walk. At 3 days when he was examined, he had a series of convulsions. When he was removed from his cage and placed on the floor, a series of convulsions and convulsive flailings of his front legs were set off. He was friendly to observers but appeared very ill. He also had waxy flexibility. On the fourth day he was still very weak but he had begun to eat. However, on the fifth day he was dead. The brain was removed for examination. Gross

RESPOXSE OF CAT “BB”

TO

TABLE I1 ADREKOCHROME AND SOME SIMILAR COMPOUNDS

Personality reaction Compound

To observer To other cat

Alertness

Motor activity

Autonomic changes

Gait

e B

2 A

Saline

Hostile

Adrenochrome, >6 mg Heparin, 1 mg

Hostile

Normal

Normal

Normal

Less hostile Indifferent

Much decreased

Much decreased

Awkward, unsteady No

Hostile

Normal

Normal

Normal

NO

Slight decrease

Normal

YO

Decreased

Normal

N O

Much increased

Much increased

Normal

NO

Decreased

Much decreased

Normal

No

Hostile

Adrenochrome, 56 Hostile Hostile Slight decrease mg, and heparin, 1 mg -4drenochrome Very hostile Very hostile hTormal methyl ether, 1mg 5,6-Dihydroxy-NHostile methylindole, 1 mg Rubeserine, 1 mg

Hostile

Less hostile Less hostile

N O

9

3 k

&Epinephrine, 1 mg Hostile

Hostile

Decreased

Much decreased

Weak, unsteady

Retched, vomited

Metanephrine, 1 mg Hostile

Less hostile

Slight decrease

Slight decrease

Normal

No

1 2=i Z

Z

wco

8

TABLE I11 RESPONSE OF CAT “BIG”

TO

0

ADRENOCHROME A N D SOMESIMILARCOMPOUNDS

Personality reaction Compound Saline Adrenochrome,

35 mg

Heparin, 1 mg Adrenochrome, 35 mg, and heparin, 1 mg Adrenochrome methyl ether, 1 mg 5,6-Dihydroxy-N-methyl-

To observer Friendly Hostile

To other cat

Alertness

Motor activity

Indifferent Indifferent

Normal Much decreased

Normal Much decreased

Gait Normal Weak, incoordination

L4utonomic changes None Licking, retching, vomiting

No change from saline No change from normal Much decreased

Much decreased

p

Friendly

Indifferent

Very friendly

Very hostile Increased

Normal

Weak, incoordination Normal

Friendly Friendly

Indifferent Indifferent

illTormal Normal

Normal None Slight weakness None

indole, 1 mg Rubeserine, 1 mg N-Ethylnoradrenochrome, 1 mg Isopropylnoradrenochrome, 1 mg Adrenolutin, 36 mg Z-Epinephrine, 35 mg

Hostile Hostile

Hostile Hostile

Decreased Normal

Decreased Much decreased

Metanephrine, 1 mg

Hostile

Hostile

Normal

Decreased

Normal Normal

Rapid respiration None

No changes were observed Normal Weak, incoordination Incoordination

None Retching, vomiting Normal

5:

$

::

EFFECT OF ADRENOCHROME AND ADRENOLUTIN

331

examination showed the canula placed inside the lateral ventricle and there were no visible areas of destruction or hemorrhage. The canula had not been displaced. Histological examination by Dr. R. Altschul, Professor of Anatomy, University of Saskatchewan, revealed no evidence of any injury. The cause of death remains unknown. “Dracula” was a nervous, hostile, and suspicious cat. He never came to the cage door to be released and resisted being removed from the cage. When released into the room, he would immediately leap from the floor back into his cage if the door was left open and would crouch in it. August 25, 1W0, he was given 1 mg adrenochrome in 0.5 ml saline. At 17 minutes, he became unsteady. He tried to jump back into his cage but was not able to leave the floor. He was placed on a stool in front of his cage and had only to walk across a small gap. But in doing so, he fell off the stool. At 31 minutes, he could hardly walk. He defecated. At 67 minutes, his hind limbs were still very weak. He was the same at 2 hours. During this experiment, he was able to defend himself well against attack from the other cat. One week later, he was given 1 mg adrenochrome with 1 mg heparin. For the next 2 hours, no change at all was seen. He remained normally active, had no difficulty jumping into his cage and eventually he became mildly affectionate to the observers.

3. Discussion

The changes observed in these cats were due to chemical action and not to the increase in cerebrospinal fluid pressure, since injection of saline, of heparin, and of several of the other compounds produced no change. This series of experiments was not extensive enough to determine which chemicals were most active. However, certain findings may stand. As a general rule adrenochrome and the compounds which could be formed from it were more active than the unnatural derivatives. Thus, 5,6-dihydroxy-N-methy1indole7 which is formed in vivo from adrenochrome (Sohler et al., 1961; Fischer and Lecomte, 1951), produced a condition characterized by increased alertness, increased activity, and increased aggressivity. But this increase in activity, which can be compared to a human hypomanic state, was appropriate and directed. Hostility was directed against the other cat and merely accentuated natural antagonism. The cats remained normally friendly or indifferent to the observers. Adrenolutin in one cat decreased alertness and activity, and produced marked weakness and incoordination of the hind limbs. The cat became abnormally hostile to the observer and to the other cat. Two adrenochrome homologs very similar in structure (N-isopropylnoradrenochroe and N-ethylnoradrenochrome) produced hardly any change in behavior. The integrity of a single methyl

332

A. HOFFER

group on the pentavalent nitrogen seems important since an ethyl or isopropyl group attached here removed activity. It is clear the effect of adrenochrome placed in the brain ventricle is not due to an unspecific effect, whatever that may be. Adrenochrome meth$ ether seems to resemble adrenochrome in activity more closely than any other compound. This compound, according to Heacock (1959b)) will on reduction be changed into a dihydroxyindole, not a trihydroxyindole. It should therefore resemble both adrenochrome and 5,6-dihydroxy-N-methylindole. From the changes induced by adrenochrome and 5,6-dihydroxy-N-methylindole, it can be assumed adrenochrome will have certain properties (assuming an equivalence in dose response). It would be expected to produce in “Big” an attitude toward the observers which varies between moderate friendliness and neutrality; toward the other cat, neutrality and moderate hostility. Alertness would be normal, and motor activity decreased; weakness and incoordination would be present, and there would be some autonomic changes. These predicted changes are fairly close to what really did happen. If we use a similar extrapolation with “BB,” the adrenochrome methyl ether would be expected to produce hostility toward observers and toward the other cat. Alertness and motor activity would be normal. Gait would be altered somewhat and there would be no autonomic changes. This prediction again is not far off from what did happen. In “Big,” it produced a marked decrease in motor activity, incoordination, and hind limb weakness. “Big,” a friendly cat, became hostile to the observers and to “BB.” This corroborates the suggestion that some of the adrenochrome methyl ether is altered to a dihydroxyindole in uiuo. Heparin had a clear and marked action in protecting the cats against adrenochrome. This was found in all 3 cats where the combination was used. The reasons for this protection are not clear, but it is possible heparin could bind the adrenochrome and keep it away from certain brain receptors. Rubeserine which is somewhat similar in structure to adrenochrome had practically no effect on “Big” but did alter “BB.” It is not as strong a acetylcholine esterase poison as is eserine. Epinephrine produced marked weakness of the hind limbs, autonomic changes, and increased hostility in the cats. These changes are not unexpected as they reproduce some of the findings of Leimdorfer et al. (1947)) Leimdorfer and Metzner (1949), and

EFFECT OF ADRENOCHROME AND ADRENOLUTIN

333

Leimdorfer (19%). Metanephrine was not inert in these experiments. Evarts (1958) and Evarts et al. (19%) had concluded that normetanephrine had no central activity in cats and was not active either psychologically or physiologically. Metanephrine decreased alertness in BB and decreased motor activity. Evarts (1958) and more recently Bacq and Renson (1961) found some sympathomimetic activity for this substance. It is possible metanephrine is not inert centrally as is normetanephrine; this question should be examined further. In a parallel way, norepinephrine has little central activity as compared to epinephrine. One mg of 3,4-methylenedioxy-a-aminomethylbenzyl alcohol oxalate placed on the ventricles of two cats did not produce any behavioral changes. Alles (1959), on the other hand, found that two similar substances, 3,4-methylenedioxy phenethylamine and 3,4-methylenedioxy phenisopropylamine, were psychotomimetic when studied in a few self experiments. E. DOGS The only record of the effect of adrenochrome on dogs was reported in a series of studies from the Institute of Psychiatry, Academy of Medical Sciences, USSR (1959). They found that adrenochrome produced changes in behavior in dogs. When injected by vein, it alerted the fluorescence characteristics of plasma so that it resembled the fluorescence of natural schizophrenic blood plasma. Braines et al. (1959) reported that when normal human or dog serum was irradiated with an excitatory beam of light, it fluoresced. The fluorescent spectrum had two main peaks, one at about 2750 A and a less intense one at 2440 A. When schizophrenic serum was examined in an identical manner, there was no peak at 2440 A. The second maximum is due to the emission of light energy absorbed by the peptide bonds and emitted by tryptophan molecules. When the dog was given adrenochrome by injection, the 2440 A maximum vanished and the fluorescence curve resembled that found in schizophrenic serum. Schizophrenic serum given to dogs by vein produced a similar alteration in the fluorescence spectrum. Kuchino (1959) observed that doses of adrenochrome less than 0.1 mg/100 gm body weight did not produce catatonic-like states in dogs but abolished conditioned reflexes. Braines et al. (1959) studied the combined effect of fear and adrenochrome on dogs. In

334

A. HOFFER

1941, he had shown that mild states of fear insufficient to produce catatonic symptoms intensified the effect of small doses of bulbocapnine (insufficient to produce catatonia) with the production of catatonic symptoms. More recently, he found that when a dose of adrenochrome which was not su5cient by itself to produce catatonic-like symptoms was introduced into an animal in a mild environmental stress it produced catatonic symptoms. Furthermore, adrenochrome even in small doses which do not produce catatonia in dogs affected the conditioned sterotypes, i.e., complexes of conditioned reflexes, although it did not affect individual simple conditioned reflexes. Complex habits were affected by adrenochrome before simple habits.

F. MONKEYS The most extensive investigation of adrenochromes effect on monkey behavior also comes from the Institute of Psychiatry, Academy of Sciences, Moscow. Kuchino ( 1959) reported, without giving details of his experimental method, that small quantities of adrenochrome (less than 0.1 mg/100 gm) abolished conditioned reflexes. Iordanis and Kuchino (1959) reported their studies of adrenochrome in greater detail. A monkey “Mashka” learned to obtain food by performing a complex sequence of steps; e.g., when a white light flashed, it had to pull a lever, press a button, pull a handle, and press a pedal, in that order. When a red light was flashed the monkey was not reinforced. Adrenochrome in doses of 0.7 to 1.2 mg affected learning habits for 2 to 5 hours. “Mashka” reacted to food and took it from the basket but did not react to the stimuli signifying food. In some experiments there were short sleeplike states after injection of adrenochrome. The adrenochrome does produce changes in conditioned reflexes very similar to 1 mg doses of chlorpromazine and to injections of schizophrenic serum. This did not occur after injection of normal human serum. Melander and Mhtens (1958, 1959) reported that adrenolutin exerted only slight pharmacodynamic activity when given intravenously to monkeys in the dose range of 20 to 25 mg/kg. After premedication with taraxein or LSD, as little as 2 to 3 mg/kg produced drowsiness and muscular relaxation. Vallbo (1957) found that 25 to 50 mg per animal of adrenochrome produced activity. With adrenolutin, 25 to 100 mg given by vein (10 to 40 mg/kg)

EFFECT OF ADRENOCHROME AND ADRENOLUTIN

335

diminished muscle tone, produced catatonia for a brief time, and diminished interest in their surroundings. With one monkey, 25 mg intravenously made him catatonic 15 minutes. He appeared stuporous with eyes closed. Heath obtained some adrenochrome from Stockholm and injected it into monkeys. No change was observed either in the EEG or in their behavior. Neither was there any change after the injection of adrenolutin. However, Heath et d. (1959) found that in one monkey pretreatment with taraxein sensitized the animal to adrenolutin; it died after injection of 100 mg. It is clear that very large quantities of adrenochrome are required to produce gross changes in behavior. Sensitization with LSD or taraxein decreases the amount required. But when very sensitive and refined behavioral tests are used, only 1 mg of adrenochrome (for a monkey weighing nearly 2 pounds) markedly altered complex learned habits. KABI-HdA8 given to the monkey protected it against 25 mg and later 50 mg of adrenolutin. VIII. Effect of Adrenochrome on Electrograms

Slocombe (1956) and Slocombe et al. (1956) measured the changes produced in spontaneous and evoked electrical potentials in albino rats by serotonin, LSD, epinephrine, norepinephrine, and adrenochrome. All the compounds flattened spontaneous activity at the cortical and subcortical sites. The most effective was serotonin. The rest were effective in decreasing order of activity as listed. The changes were profound with thiopental anesthesia, but there was no change when ether was used. These authors believed the action was nonspecific on lower centers which have cortical and subcortical projections. Krupp and Monnier (1960) injected adrenochrome into rabbits. There was a change in the EEG right after the injection. The amplitude of the spontaneous activity of the neocortex decreased, and slow wave activity disappeared. Simultaneously there was an increase in synchronicity in the hippocampus and thalamus. There was thus a typical arousal pattern. After intercollicular decerebration, adrenochrome produced neocortical desynchronization with a slight increase in hippocampal synchronicity. After adrenochrome, there was an increase in the arousal reaction to sensory stimulation. The excitability of the hippocampus and its connections to the

336

A. HOFFER

neocortex was increased, These authors concluded that LSD, mescaline, and psilocybine altered spontaneous electrical activity in a similar way to adrenochrome. All produced a sharp arousal pattern of activity. LSD caused desynchronization of the subcortex but the other three substances intensaed subcortical synchronicity. LSD had no action on the “encbphale isol6” or the “cerveau isole?‘ but adrenochrome had a slight effect and the other two compounds a marked effect. Both adrenochrome and mescaline activated the hippocampus but did not release spontaneous discharges. LSD released spontaneous discharge and psilocybine decreased hippocampal activity. Schwarz et al. ( 1956b), using the ventricular cannula technique of Feldberg and Shenvood (1954), found that after the injection of 1 mg adrenochrome the cats were drowsy for 24 hours. The deep EEG showed occipital 4 cycles/sec slow waves with low-voltage spike components spreading to the frontal region and then diffusely over the brain. Painful stimulation caused inconsistent arousal. An arousal pattern appeared when the animal was drinking milk. These authors described the EEG changes as a trance pattern. Heath (1959) observed no change in the EEG pattern of monkeys. However, one monkey pretreated with taraxein was sensitized to adrenolutin and died after 100 mg. EEG changes then were present. Hoffer et al. (1954) reported that adrenochrome produced pathological changes in the electrogram of some epileptic patients. A detailed report was presented by Szatmari et al. (1955). A few volunteers with normal electrograms were given 10 to 25 mg by vein. There was no change in the electrograms. Epileptic patients were given 10, 25, or 50 mg of adrenochrome. Five patients had a high-voltage, diffuses, paroxysmal abnormality with bilateral hypersynchrony and diffuse high-voltage 5/sec activity. Adrenochrome produced a marked increase of the dysrhythmia and an increased sensitivity to hyperventilation. In two cases, the threshold for convulsions was lowered. Another group of 15 patients had focal activity showing spike, sharp-wave, and irregular delta activity in all cases, and diffuse bilateral slow activity in 10 of them. After adrenochrome, there was an increase of dysrhythmia, an increase in voltage and decrease in frequency, a marked increase in all cases of focal activity during

EFFECT OF ADRENOClIROZlE AND ADRENOLUTIN

337

hyperventilation along with a spread of pathological activity in the opposite homologous cortical area, and in 4 cases a spontaneous increase in irritability of the focus. Schwarz et al. (1956a) measured changes in depth electrogram induced by adrenochrome. Patient 1, a chronic paranoid schizophrenic, on three occasions was given 50 mg, 60 mg, and 75 mg of adrenochrome by vein. There was a moderate increase in bitemporal paroxysmal discharge of 2 to 7 cycles/sec, and increased persistence and amplitude of the focal temporal sharpwave discharge from the depths, in all three instances. Subject 2 was given 50 and 60 mg on two occasions. There was a moderate increase in paroxysmal activity. Patient 5, who had psychosis with epilepsy, was given 50 mg of adrenochrome. This produced high-voltage waves 2 to 3 cycles/sec, associated with drowsiness. The depth electrograms of patient number 1 resembled that reported for a schizophrenic by Sem-Jacobsen et al. (1955), but after adrenochrome, the focal sharp wave activity of maximum amplitude from the temporal region was persistent. The authors concluded “administration of mescaline, LSD, and adrenochrome can cause striking changes in the depth electrogram.” Grof et al. (1961) gave 6 subjects 20 mg adrenochrome sublingually. In 5, there were marked changes between 30 and 90 minutes. Alpha activity was slightly disintegrated. Theta waves appeared with spikes and the EEG became hypersensitive to hyperventilation. There was no correlation between the EEG changes and the intensity of the psychological changes. IX. Effect of Adrenochrome and Adrenolutin on Humans

Hoffer et uZ. (1954) observed that adrenochrome produced psychological changes in man. Since then, several additional studies have been recorded but not in great detail, Only the methods used and summaries of results have been given, due to the difficulty in having journals accept long clinical reports. In this review, results of experiments with adrenochrome and adrenolutin will be outlined in greater detail. A. ADRENOLUTIN DOUBLE BLINDEXPERIMENTS Although the influence of faith or lack of faith cannot be eliminated, it may be minimized by using the double blind design. This

338

A. HOFFER

research was briefly recorded in 1957 (Hoffer, 1957c, d ) ; the reader is referred to those outlines for a description of the research design. Normal intelligent subjects were used. They included graduate students, medical students, and graduate nurses. The volunteers were paid for their time. They were told that two new but psychologically active compounds were being used, neither one being LSD or mescaline and that any changes which did occur would be mild and of short duration. Volunteers with a known history of severe physical or mental disease were excluded. All subjects received the compound at 6.00 PM. The observers were one psychiatrist and one psychologist. They knew that the subject would get one of four possible combinations as follows: ( a ) adrenolutin followed by adrenolutin, ( b ) adrenolutin and placebo, ( c ) placebo and placebo, or ( d ) placebo and adrenolutin. Each experiment was run in the evening and repeated 1week later. We hoped that the observers would record their observations free of bias. Secondly, they attempted to predict after the two experiments were completed whether the subject had received placebo (riboflavin) or adrenolutin. In testing the predictions of the observers against what the subjects had really taken, a very harsh criterion was used. The assumption was made that all subjects given adrenolutin would react in a clear and noticeable manner and that no subjects given placebo would react. Even with well known hallucinogens like LSD, this would not be true. I have seen many subjects have no reaction to 100 pg of LSD and a few have failed to react with anything but increased tension to 500 p g . Anxiety will induce some volunteers to have some LSD-like reactions. But inasmuch as there was no way of knowing what proportion of subjects would react, there was no other way of testing this. The first thirteen subjects were tested between October 4 and December 19, 1955, using fresh adrenolutin which had been synthesized by Pfizer and Company, New York. The second series of twelve subjects was tested between January 9 and April 13, 1956. By this time, the adrenolutin, which when fresh had been yellow-green, was dark green in color. There was no doubt it had deteriorated. The observers were able to predict the drug given to the first series at the 5%level of confidence. But this was not possible with the second series. This provides some evidence that autoxidized derivatives of adrenolutin are less active psychotomimetics than is pure adrenolu-

339

EFFECT OF ADRENOCHROME AND ADRENOLUTIN

tin. It might be argued that the series should have been discontinued as soon as it was noticed the chemical was changing. But this would have biased our data in favor of a positive result. Furthermore, it was then impossible to get any more. Each sample sent to us by various manufacturers was as dark and deteriorated on arrival as our own stock had become. Another way of examining the data statistically is to examine the recorded data for certain changes. These as recorded by Hoffer (1957c,d ) were summarized from records made by the observers and by the subjects before the compound given was known. The code was not broken for each subject until all the written material which forms the raw data was complete. I have reexamined the reports and from data published there (Hoffer, 1957c) scored each subject for the presence or absence of definite changes. This is shown in Table IV. In all categories but anxiety, the evidence of TABLE I V EFFECT OF ADRENOLUTIN A N D RIBOFLAVIN O N MENTALSTATUS OF VOLUNTEERS I N A DOUBLE BLINDSTUDY^ Number of subjects showing this change Type of change Perception Thought Mood Anxiety Personality Carry-over, next day Total number subjects

Receiving ndrenolut,in

12 14 3

1 3 8 20

Receiving placebo I 1 1 7 1 1 14

From Hoffer (1957~).

abnormality was much greater in the subjects given adrenolutin, but they were singularly free from anxiety. This kind of report provides merely the dry bones of a study. For this reason, I will now give in clinical detail a few typical cases which were recorded many years ago. It might be argued that rather than report data obtained with adrenolutin now known to be less pure than it could have been, it would be desirable to repeat the study with stable crystalline adrenolutin. The reasons we have

340

A. HOFFER

not done so involve economics of research, priorities, and faith. It is doubtful whether several repetitions of similar experiments would be any more convincing than the original ones, Skeptics are not convinced by over-selling one’s point of view, What would be more convincing would be reproduction of similar work by other investigators. One independent corroboration is worth several repetitions by the same worker. Fortunately several research groups have begun to provide some corroboration as will be shown later. The following accounts include 2 subjects who had both riboflavin (placebo) and adrenolutin, 1 subject who had placebo twice, and 2 who had adrenolutin twice. 1. Subjects Receiving Adrenolutin and Placebo a. Miss B. X . (Riboflavin). Miss B. X., a graduate nurse, was tensc for thc first half hour and felt her speech and thinking were slow. At the end of the first hour, no change was seen. Two hours after starting, no changes had occurred. She remained alert and interested, and felt stimulated by the questions hurled at her to test her thinking. At 10:00, she was sleepy and very tired. For a moment, the printed word appeared altered. She slept well. The next morning, she was normal. One week later, she received adrenolutin. There was no change the first half hour. Then she noted she was much less anxious than she had been the previous week at this time. At 7:00, she was relaxed and cheerful and felt it was amusing she did not do as well on the tests. While doing the Bender Gestalt test, she could hardly keep from laughing. She was not able to do the 100-minus-7 test correctly nor make simple conversions of grains to milligrams. These are routine for nurses and she had done very well the previous week. The investigator seemed much funnier to her. Her converbation was flippant and for her inappropriate. At 8 : 0 ,she was indifferent to pain or the proceedings and found everything amusing. When the stroboscope was flashing before her closed eyes at 12 per second, she saw a bright white light rimmed with red. This turned into a large moss rose. At 9:30, she was depressed but not tired. The experience of the testing, etc., appeared very silly to her and she could not understand why she had been so free in her conversation. Later, she had a frontal headache. That evening, she arrived home elated and very talkative. She was amused by the research and remained hilarious to an unusual degree. The next day, she was normal. She reported that she had felt unusually disinhibited socially and had made remarks she would have made only to a close friend or sister and not to strangers. She had been much more relaxed with the second EEG run. There was less fatigue during the first experiment and the whole evening had passed more quickly. b. Dr. L. J. (Adrenolutin). Dr. L. I,, a very intelligent intern, wrote the following account after his first session:

EFFECT OF ADRENOCHROME AND ADRENOLUTIN

341

“From the period of 5:30 to 6:00, I had a vague feeling of apprehension and some very very slight anxiety not knowing quite what to expect but knowing that the experiment, generally speaking, was considered boring by several of my friends who had taken the test. At 6:00, I took the drug and from 6:OO to 7:00, I noticed several interesting features. The first thing I noticed was that I was mildly to moderately anxious of what was happening in spite of the fact that I was in fairly familiar surroundings and knew the psychiatric personnel who were examining me. I was made particularly anxious by some of the questions that were asked by the psychiatrist. I found that I was completely unable to answer fairly simple questions. I felt a definite frustration and some humility and also a quite marked antagonism Cowards the psychiatrist because he insisted on asking questions I was unable to handle. I felt that I was definitely putting on a very poor performance to begin with and I assumed that if I had had a drug that in the first hour it would probably have little effect and that most of my slowness in performing these questions was either due to anxiety or to sheer stupidity. I was quite apologetic about not answering the questions smartly. I had definite symptoms of moderate anxicty-my hands were very cold and extremely sweaty. I felt upset, I had marked tachycardia and a feeling of discomfort in the epigastrium which is extremely uncommon for me. From the period 6:30 to 7:00, I noticed an increasing inability to perform the questions. I could remember the facts presented to me quite quickly in these mathematical questions, but I lacked the energy, the initiative to carry the solution through to a final answer. I was almost incapable of setting up the equations in any manner. Towards the latter part of the questioning, drawing close to 7:OO I believe, I could hardly even entertain the questions and this I attributed to my marked anxiety. I found this most frustrating and upsetting experience for I seemed to lack ability to localize my thoughts in a forward direction in handling these relatively simple arithmetic problems which I should imagine would be at grade eight level. I, towards 7:00, also began to feel a tremendous lethargy and apathy towards things in general and this I attributed to the fact that I was overtired when the experiment began which on looking back is probably not exactly true. Some time after 7:00, I found it extremely difficult to concentrate on the questions being asked me because during the application of the electrodes portions of my hair were being snipped off and also acetone and compressed air were used during this procedure, all of which I found fairly distracting. At about this time which I would imagine would be 7:30, although I could not b e sure, my attitude seemed to b e changing somewhat and I had lost much of my initial anxiety and this was being rapidly replaced by a type of apathy which seemed to be increasing to tremendous fatigue. Accompanying these feelings, I began to lose my feeling of initial humility and also much of my initial hostility. These feelings were being replaced by a feeling of disinterestedness. Also a feeling that I was rather above these frivolous questions that were being asked of me. My examiners no longer seemed quite as friendly as they had formerly and I felt rather like they were young schoolboys wasting the government’s money asking rather stupid and superfluous questions. At one point, during this procedure, I was asked to interpret some proverbs, Because of the distractions going on around my head,

3 49

A. HOFFEX

I adopted a policy of answering the questions as quickly as possible and with as much dispatch as possible and sloughed off the answers in an effort to get them behind me as quickly as possible. Although the questions did not irritate me, I felt them rather frivolous and somewhat senseless and my main object was to get that part of the questioning over with as quickly as possible and likewise get the evening over with as quickly as possible because the whole thing by this time had become rather boring. I thought my questions to the proverbs being asked were sometimes rather clever. Likewise when I was asked such questions as “why are people taxed” and “what is the function of govemment” etc., I thought my answers to these were quite concise and astute and at least would be sufficiently good to satisfy this young group of upstarts that were attempting to question me. Towards 8:00, I found the results of the stroboscopic examination were not too disagreeable although rather tiresome. The f€icker fusion test following the stroboscopic examination required a good deal more concentration for at this time I felt very fatigued, very apathetic and very often during the flicker fusion test, I would just take a rough stab at what I estimated was the fusion point. I had lost my enthusiasm completely for the experiment but considered that I would play the game and continue with the experiment just co please my examiners who I felt were carrying on in a rather frivolous manner. “The culmination of the tests came when I was given the critical thinking test which I found extremely difficult. It seemed to never end and I must have taken well over an hour to perform it. Not only did I find the questions difficult but I found the instructions di6cult to understand and I believe most of my time was spent trying to figure out what was required of me in the test rather than getting on and doing the actual questions. I found that I had to read over the questions perhaps two or three times and even then was not entirely clear as to what was intended. I was fairly discouraged and rather depressed, and thoroughly fatigued, and yet I had a supreme apathy much as a lotus eater must have had. I was very thankful when the test was finally finished. Following the experiment I returned to my ward where I had several duties to carry out. Although I felt very very tired and certainly ready to go to bed, I felt that I was in my right mind actually. I had to start an intravenous injection and to my chagrin I had great difficulty, making four or five attempts. However, this did not particularly upset me although it must have upset the patient somewhat and I continued to attempt to start this intravenous with some abandon. Later that night at 4:OO in the morning, I was required to get up to give another intravenous injection on the ward on which I worked. For this duty which was to be carried out on a ward some considerable distance from where I sleep, I decided I would get out my bicycle which was kept in a cupboard close by my room and ride down to the ward, which I felt was a very practical thing to do at that hour. I had not done this before, although that evening I thought it would be a very sensible thing to do. On reaching the ward, I again experienced considerable difficulty giving an intravenous injection but again this did not worry me very much. This is perhaps unusual for me for when I miss intravenous injections, I normally become quite disturbed when I miss them for the second time. However, at that hour I didn’t

EFFECT OF ADRENOCHROME AND ADRENOLUTIN

343

think it mattered if it took three or four or even five trys which I again carried out with the same abandon as previously. The next morning I awoke and I was quite tired, still very very apathetic, and with a rather dull headache. I was unable to pick up any speed in my work throughout the morning and even into the early afternoon and although I had a fairly heavy schedule of work ahead of me, I was unable to muster the necessary energy and initiative to carry out these various duties. The fact that I couldn’t seem to carry out these duties didn’t really worry me too much at any time. By the following evening, however, I felt I had considerably more energy, I was more awake and more able and willing to carry out the necessary daily tasks. The following day I felt completely normal in all respects. On looking back over the whole evening, I might say that I had received a depressant drug which made me very apathetic, rather depressed and did not seem to completely abolish my anxiety. Although I felt anxious, I didn’t seem to be really able to do much about it because of this tremendous sense of fatigue. Generally speaking, I felt I had done only moderately well or rather poorly on the tests, particularly on the first few and especially on the critical thinking test. I regarded the evening as rather unpleasant and one which placed a considerable strain on me. As a result, I was not particularly looking forward to the next week‘s experiment, which I was afraid might have a somewhat similar effect.” One week later, Dr. L. J. was given riboflavin. At 6:30, there was no change. At 6:45,he was able to do calculations better than the previous week. At 7:00,he did pretty well but failing to give a correct answer to a problem made him upset and anxious. At 7:45, he seemed more relaxed, somewhat elated, and full of good humor. This he described as his normal personality. The rest of the evening, there were no changes. The following day, he was sure he had placebo the second time because he had been wide awake, more attentive, and much more alert. The questions had been easier, there had been no headache, no fatigue, and the memory for the evening had remained clear. He concluded the second drug was either a placebo or an euphorient. His sown account follows: “On this Monday, in spite of the fact that I had again stayed out rather late Sunday evening, the day before the test, I was in moderately good spirits when I embarked upon the test. I felt somewhat more confident about what was going to happen. I knew that the test would b e somewhat similar and therefore I had no particular apprehension. At 5:30 to 6:00,I had the usual flicker test which I did with some enthusiasm as before, knowing that they would be rather boring but that it could be well tolerated. From 6:OO to 7:OO my reactions were considerably different from the previous week and I didn’t feel nearly the anxiety that I’d had. My hands remained warm and dry, I was attentive and relaxed. The psychiatrist fired a similar type of question at me involving the rather simple arithmetic questions. Unlike the previous week, I did not have a certain sense of panic and confusion but was able to make at least an effort at working out an equation. I was making a sincere and somewhat enthusiastic effort to at least make a good stab at the question even though I might not be able to get the answer. The fact that I couldn’t get the answer didn’t particularly worry me as it had the week before. I seemed

344

A. HOFFER

generally able to give more answers this week in a quick and more accurate manner. 1 did not feel the humility nor was I as apologetic when I made mistakes on this occasion. My examiners did not arouse the antagonism and hostility in me as they had done previously. I felt in a quite good humour from the hours of 6:OO to 7:OO. I seemed to be able to muster the necessary energy and initiative and enthusiasm to tackle the questions even though I probably would not get the right answer. I was not particularly discouraged or depressed at all and was in an exceptionally good mood when we moved on to the second part of the experiment which again consisted of the EEG and stroboscopic examination. . . . I think I did a better job on the proverbs although there were one or two which I seemed to have a mental block in explaining. On the questions of abstract thinking , . . my performance was better than the week before. On the EEG and stroboscopic examination, I think the results were markedly different from the previous week. At no time was I really aware of geometry and symmetry to the field that I saw; there were no definite centres of light particularly and no straight radiating lines coming off the centre of light. . . . The critical thinking test, I noticed a very marked subjective difference. I was able to understand the instructions immediately on reading them over. I found the questions quite difficult and still not clear today, but I was able to make what I thought was an astute try at the questions and on the whole, I felt much better about the test when I finished than the week before. I felt that I was clearer in my mind, my judgment was better, and I had definitely done the test in a more precise and much more rapid manner. I believe that I had bettered my time by perhaps twenty minutes to half an hour. I felt no particular sensations other than one of extremely well being. I was much more awake at that hour than I usually am. “. . . In retrospect, looking back over the evening, I felt the whole experience was much more pleasant the second evening and I think that I actually enjoyed it. I was able to maintain my interest throughout it and I seemed able to bring my powers of concentration to bear on all the questions so that I could give them all a good try. I didn’t feel any anxiety and was in extremely good spirits throughout the evening. I felt quite kindly towards my psychiatrists who I felt were my colleagues and generally felt that the experiment on the second evening was well worthwhile and that the tests were astute ones. As a final conclusion therefore I think that on the first evening I was given a depressant drug, which made a very unpleasant evening for me. On the second evening, I was given an euphorient drug, a stimulant type of drug and the evening was quite passable. However, the drugs are of such a subtle nature that following the first evening, I would not be sure that I had had anything and even following the second evening, taken by itself, I could not be s u e that I had had anything. However, in comparing the two evenings, there was a very very marked difference in the way I felt and therefore I must have had contradictory types of drugs on these occasions.” After the two sessions had been completed the subject was very concerned. For most of the week after having had adrenolutin he was very disturbed because of his unusual paranoid ideation during the first experimental night, which he had not disclosed to the investigators. On the second day after the

EFFECT OF ADRENOCHROME AND ADRENOLUTIN

345

first experience he reported to me that he had not told the two investigators everythmg that had occurred as he had been instructed to do. What had happened was now worrying him since he was quite certain he had had placebo. I asked him to come for an interview after he had finished the second half of the experiment because I then did not know what he had had. After the experiments were over, he reported what he had neglected to tell the two observers. During the first experiment, he was asked his opinion of socialized medicine. He immediately felt very strongly that the two observers were both communists and were trying to draw him out. He thought he would play along with their little game and tried to draw them out in order to obtain evidence that they really were communists. This he proceeded to do by agreeing with e v e r y h n g they said about socialized medicine. He watched them carefully for evidence. He felt B. was a very sinister person although he had known him before, Finally he saw B. pick up a pencil marked “Government of Saskatchewan.” This immediately confirmed his suspicions, for who but a communist would get a pencil from the Government of Saskatchewan? He then became preoccupied with the EEG procedure. The next morning, he could not understand how he could have had these ridiculous ideas. He felt so guilty over this he did not wish to tell anyone. For the next few days, he was very quiet and subdued on the wards, quite unusual for him.

2. Subject Receiving Adrenolutin Twice Miss H . G. This graduate nurse was given 50 mg adrenolutin. Before taking the chemical, she was exceedingly tense and her hands were wet with perspiration. At 6:30, she suddenly noted a marked sense of relaxation and some lightheadedness. At 6:40, her head felt funny and she felt her thinking was fuzzy. She had difficulty following conversation. The observers’ faces seemed distant from her and she became suspicious of one. She was strongly aware of being watched. At 7:00, she seemed to be small and the two observers towered over her and looked down at her. She considered these visual changes foolish. Words hopped up and down on a page and she had difficulty concentrating. Her thinking was very sluggish and she could not solve elementary nursing arithmetical conversions. At 7:40, there was no more fuzziness and she was relaxed. Words still moved u p and down on the page. At 8:00, her hands were dry and this surprised her as she felt she should have been anxious and was not. At 9:00, after the EEG test, she was uneasy and felt oppressed by the stroboscopic light and thought things were closing in on her. At 9:35, she was relaxed again and there were no more visual changes. The next day, she wrote the following account: “After the third test, it appeared to me that perhaps I was being looked down upon and the letters of a psychology book had a dancing movement up and down. It was difficult to convince myself that the words were moving when I knew they should not be. “I was quite disturbed at not being able to do my calculations and thought Dr. Hoffer would think I was very incompetent as a nurse. During the EEG, I was very disturbed at the technician who was just talking too much and not paying attention to his work. My head hurt and I could have screamed out but

346

A. HOFFER

I told Dr. Hoffer I was all right. The flashing light gave me a suffocating feeling and I felt like running away or telling them to stop, but couldn’t. Later, I went to the bathroom and felt like getting away from the experiment. There were some people in the hall. I thought they were looking at me and thinking I was foolish. “The next three tests seemed very foolish but I still maintained I was interested. I did not care how I had done. Before going home I believed nothing had happened and that I had been normal. “At home I decided not to tell my friends anything. I slept well. The next morning I was tired but my head was clear. I was able to figure out the math problem in my head easily [the one she could not solve the previous night]. I realized I had been uncommunicative last night but was able to talk freely now. I realized that I had not reacted normally last night but today I’m thinking more clearly and critically.” Two weeks later, she was again given 50 mg adrenolutin. She was more relaxed this time. The first half hour, she discussed the three things which had impressed her about the first experience. These were ( a )her feeling she must not communicate; ( b ) her feeling of suspiciousness; ( c ) her difficulty in thinking. At 6:30,she again felt she must not talk, became suspicious, and felt again that we were looking down at her. She also felt quiet. She described one observer’s face as having cruel eyes, an accentuated mouth and eyebrows. At 6:45, she again saw words move but less than the first experiment, and they were blurred. At 6:50,these visual changes were gone. At 7:00, she felt more relaxed and was more sure of herself. She tried the serial seven test four times and failed each one but it did not disturb her. On proverbs and on the Wechsler-Bellevue test, she was somewhat better. At 8:00, there was no tension. She felt quiet. There was less paranoid feeling about the observers this time. At 10:00, she was relaxed and felt she had done much better this evening. There was actually very little difference on the scores. The next day she reported: “At 5:15 p.m. I was not apprehensive, and was looking forward to the evening. I was curious to know what would happen. After the first test, I took the drug. After a few minutes, we went back to the first room-repeating the first test to which I felt I responded fairly well. After this, Dr. Hoffer gave me a book and asked me if I saw the same thing as last week (movement in page)-I did not. I then experienced the feeling of being looked down upon by A. more than Dr. Hoffer. A.’s eyes appeared somewhat cruel, his eyebrows were pronounced as was his mouth; just Dr. Hoffer’s eyes and eyebrows were pronounced. This lasted only momentarily and was not as intense as last week. After this we went out to do the “distance test.” At this time, I realized that my mood was definitely calm, that I was not suspicious as on the previous test and that I would probably be more truthful in my comments but still not too talkative. My reaction towards the resident psychiatrist was friendly and not critical realizing he knew his work. The test passed quickly; I was sure of my measurement and did not feel as though I was underestimating myself. Interest did not seem to fade during the test. I believe at this time I was somewhat relieved to think I was not having the same reactions. Dr. Hoffer began his

EFFECT OF ADRENOCHROME AND ADRENOLUTIN

347

math quiz again. I thought I could do better than I did, however. I did have several different answers to the 93-7 test; felt I did better on the math but the fact that I had made some errors did not greatly concern me. I felt that Dr. Hoffer and A. were friendly and kindly towards me; also was not as worried about not being able to answer questions but felt perhaps could still do better. During the EEG, I felt very relaxed and calm-could easily have gone to sleep. The time of the test did not annoy me, the hyperventilation part of the testhad some headache, but not as severe as last week. I was finding the evening more interesting even though the tests were the same-perhaps I enjoyed the calm, relaxed feeling I was experiencing in contrast to the previous evening. The Rorschach test also was more enjoyable-I still saw some bats and crawling animals although not as sinister or frightening; and I was able to see more pleasant things-clouds, melting snow, rock and tree reflections in water, just modern art painting for my imagination could not see anything but color. The colors were softer, warmer blue. “February 25. This morning I was still tired-somewhat more than usual. Talked to the Director-was not the least bit annoyed with her this morning either-a feeling which is different from the annoyance I have felt towards her ( P.s., this is a feeling that is experienced not only by myself but also my fellow workers). I still feel relaxed and calmer I believe than what I normally am. “February 26. I believe the calming effect lasted well into Saturday evening. I was going out Saturday evening and had none of my usual apprehensive manifestations-an excessive perspiration in axilla and palms of the hands. I am still somewhat tired this morning but I am unable to be definitely certain that this is a result of the medication. The experience still remains enjoyable and is beginning to repress the first evening from my mind.”

3. Subjects Receiving Placebo Twice Miss B . C. This subject received riboflavin twice. There was no change evident during either experiment; this was typical of the placebo runs. The following account was submitted by the subject on the day following the first session : “I can remember everything that happened last night quite clearly. I think I was at ease all evening and not exceptionally nervous. I recall that my eyes seemed to be jumping with the lights flashing on and off and flickering but I had no headache. While walking to the EEG lab, I did feel slightly unsteady and not too sure of how I was walking. When I was being asked questions at EEG, I felt a little slow in my thinking partly because of the work being done on my head. It didn’t annoy me much but did distract me some. I felt quite relaxed and sleepy after the questioning and when the EEG first started. With the light flashing in my eyes, I saw various colors and designs-all symetrical and moving but no images of any kind, For a while I was dizzy, with my eyes closed. The bed seemed to b e turning under me and I was turning around in the opposite direction-similar to a ride at the Exhibition. But it did not last too long, cleared some when I opcned my eyes. It stopped without leaving any different feeling-no headache or nausea. After the EEG and the lights, I felt

348

A. HOFFER

rather tired and not interested in any more problems. But on walking back to the office, I wakened up again and felt quite normal. However, I was beginning to get hungry. I think I did and said the same things on the tests then as I would have without any drug. After I returned home, I felt quite normal, ate and went to bed and to sleep right away.”

4. Summary A summary of the fourteen subjects who received adrenolutin and placebo is shown in Table V. A score was derived by giving the subject one point for an abnormality and zero for normality under the headings shown in the table. Thus the presence of changes in thought is scored one. I consider it normal for the subject to have had anxiety during these experiments. Its absence is scored as one. The mean score of 14 adrenolutin runs was 4.43, with a range of 1 to 7. For placebo, it was 0.93 with a range of 0 to 3. In 11pairs of results the scores for placebo were less. The subjective accounts were then examined for the presence of residual changes in the days following the experiments. Out of 18 subjects given adrenolutin, there were 9 prolonged reactions. Out of 32 placebo experiments, only 4 were prolonged (Chi-square = 6.4, P < 0.01). The subjects were not asked regularly what they thought they had received. But several spontaneously remarked that they had had something active or inactive. Of 9 subjects each of whom received both adrenolutin and a placebo, only 1 subject erred by calling adrenolutin inactive; in all instances the placebo was termed inactive. (For this distribution, with one df.,the value of chi-square is 10; therefore P < 0.001.) This sample of series is of course biased in that people guessing what they had may have been different, and it is possible that if all subjects had been asked to make these predictions, the results would have been different. The work reported by Grof d d.( 1961) makes this suggestion quite unlikely. Their subjects were able to detect adrenochrome very well. B. PROLONGED REACTIONS TO ADRENOCHROME AND ADRE”

1. Adrenochrome Some of the changes produced by adrenochrome may persist several days, and in some cases the effects nearly led to disastrous

EFFECT OF ADRENOCHROME AND ADRENOLUTIN

349

results. Two cases of prolonged reactions will be discussed. These experiences with adrenochrome have made us quite cautious with this drug which seems to be so mild in its action but which can be SO dangerous because of the lack of insight it induces in some subjects. a. Miss F . M . This young female, age 16, had been well until age 8. For the next 2 years, she had unusual sensations in her chest and nose every night. A few weeks before this investigation, she developed changes in perception (her mother’s face seemed strange and altered as she watched; the visual field was covered with a checkerboard pattern, her body image was different, time moved very slowly, and houses looked like cornflake boxes). There was no thought disorder but she was very apprehensive. For several years, she had been given dilantin and phenobarbital for mixed petit and grand ma1 epilepsy. At the time of investigation, she was normal. The EEG showed a very marked left anterior temporal focus characterized by high voltage, slow waves, and an occasional saw-tooth wave with single spikes (Szatmari et al., 1955, p. 607). The left temporal focus increased with hyperventilation. At the beginning of the EEG test, she was happy, cheerful and friendly. She was given 50 mg of adrenochrome by vein. After 10 minutes, she developed a feeling of estrangement and fear and her nose itched. The pathological activity of the temporal focus increased and dysrhythmia became generalized. She was now morose, quiet, and depressed. When she was urged to describe how she felt she cried. Then she reported that the strange feelings and disturbances in body image which had been present 2 weeks before had returned. She was given 500 mg of nicotinic acid by vein. Within 15 minutes, the focus was less intense, the dysrhythmia had disappeared, and she was mentally normal. That evening at home, she was moody and quiet. One week later, she was very disturbed and 2 weeks after the treatment required committal to a mental hospital. On admission, she suffered from perceptual changes (all faces were strange; there were marked ddjd vu, feelings of estrangement and unreality, and visual hallucinations). She had thought disorder (her present life was merely a show and a replay of a previous period in her life; she was confused, rambling, and almost incoherent) and referential ideas with delusions of guilt, and she was paranoid. Her mood was flat. The staff were not aware she had received adrenochrome. On admission, the diagnosis of schizophrenia was entertained but with her history of epilepsy, she was finally diagnosed as having an epileptic psychosis. A few days after admission, her psychosis cleared and a few weeks later, she had typical grand ma1 convulsions. These were controlled by anti-convulsant medication. She was in the hospital about 6 months. Over the next 6%years, she has shown no schizophrenic-like symptoms. b. M r . D. S. This patient, age 18, was first treated June 7 to September 29, 1957. He was diagnosed as having a neurotic hysterical reaction with marked depression. Schizophrenic features were present. He was given regular psychotherapy and 20 ECT and was discharged much improved. Two weeks

8 4

1 2

ou sa.C

OU

ou ou sail

ou ou

ou ou

1 E

ou ou

ou sah

9

ou saS

ou ou

Z Z

ou

0 9

Z

1 2

0

L

ou

ou sail

ou sail

ou sail

ou ou

sail ou sa-i

ou sail ou saL ou sail ou

on ou sah ou

ou ou

ou ,sail

ou ou ou Sah

ou sail

ou sail

ou ou

sab ou

ou sail

sa6

,sad< .sail

sail ou

sail ou

ou ou

ou ou ou ou

ou .sail

saA sail sah

ou ou

ou sail

ou sall

ou on ou sail ou ou

ou

ou

ou

ou

ou sail

ou sail

d Q

:3

d Q

:,$

d

Q

:3

d

v

:a

d

v d ti

:3

:a

d

v

:\-

imz

P

5

5 2

E

ou ou

1 I

sail sail

E 57

ou ou OU

ou

ou ou ou sail

ou ou ou ou

ou sab ou sail

sail sail

ou ou

ou sail

ou ou

y~oydn3 .passaJdaa

d V

:N

d

v

:lN

4

!

9

0 P

0

8

sad( sad

on sad

ou sa6

0 9

B

sai ou

ou saiC

ou sa,C

0

O

T

ou salC

ou sail

ou

P

ou

I

ou sa6

on ou

0 9

ou sail

sail ou sail ou

ou ou

,sail ,,sa.{ ou .sail

ou .sail ou .sail

ou ou

ou

ou sail

ou ou

ou sa6

ou sail ou sail ou sail

ou ou ou

ou sail ou sail

ou saL

ou sail ou sail ou sail

ou ou

d

Y d

v d 8

:7

:x :r

d

v

:I

d

v

:€I

352

A. HOFFER

later, he remained well. He was readmitted February 21 to March 21, 1958, in the same condition as on his first admission. A better history was taken and he was found to have had perceptual changes. He felt people were staring at him, thought his nose was getting bigger; he felt small in stature compared to others, and he had visual hallucinations of shadows and people. He had thought disorder and paranoid delusions, and was extremely tense. He restlessly paced up and down in his room for many hours. He was diagnosed as an early schizophrenic. Three days after admission, he was given 10 mg of d-adrenochrome by vein. Almost immediately h e became more relaxed but developed vivid hallucinations. He could see his hands growing larger and smaller, he could no longer estimate distance of people from himself, and pictures appeared very vivid. One hour later, while looking in a mirror, he saw his face divided into two halves, one white and one black. For the next 6 days, this recurred whenever he became very tense.

2. Adrenolutin

There have been many prolonged reactions to adrenolutin. Some of these changes have already been described in the case histories (cE. Section IX, A ) . Other reactions lasted more than 1 day after a single administration of adrenolutin, and reactions up to 1 week have occurred. M r . 1. K . Mr. I. K. was given 25 mg of adrenolutin at 6:OO PM on January 27, 1955. Over the next few hours, a few visual changes were present; he was relaxed and felt his mind was clear but his thinking was altered. He had difficulty in following conversation and felt other people about him were unimportant to him. Between 8:30 and 1O:OO P M voices seemed unnaturally loud. The next day he suffered from a slight headache all morning, and he was listless and tired, On March 3, 1955, he was given 50 mg of adrenolutin at 6:OO PM. There was little change in him. Two hours later, he was quieter and spoke with difficulty. Later, he developed an irritating headache over his left eye. W e concluded that there had been no reaction and he was allowed to go home at 1O:OO PM. His wife reported he looked drawn and tired. He was unusually unresponsive and did not tell his wife anything about the evening. That night he slept fitfully and he was less responsive to the fussing of his two young children. The next morning, he was irritable and seemed disinterested in his mail. Episodes which normally did not bother him were causes for irritation. That night, he was angry about his car being stuck in the snow near his house. The second morning, he seemed more normal. When he could not start his car, he came rushing in and shouted at his wife that he could not start the car and that it was all her fault. This degree of anger and this type of accusation was most unusual for him. When asked how it was her fault, he stated "if she had not wakened him yesterday, he wouldn't have been able to help her sister get her car out and therefore she wouldn't have had to drive home again and so get stuck in the same hole and that therefore he wouldn't have had to drive

EFFECT OF ADRENOCHROME AND ADRENOLUTIN

353

his car around to the front and leave it there to freeze on the road and ruin the starter.” That evening when Mrs. I. K. was feeding a young child, he suddenly without warning shouted her name in a violent manner and then said she was harming the child by her concern over the feeding. When Mrs. K. asked him why he shouted instead of just quietly telling her, he glowered and muttered something. That evening, he appeared haggard and ill. He was unreasonable and uncommunicative. The third day, he awoke cheerful and normal until late in the afternoon when his irritability, restlessness, and peevishness returned. The rest of the day be was alternately normal and withdrawn and irritable.

Much has been made of placebo reactions, and they are indeed very powerful. But the incidence of prolonged reactions was very unusual with our placebo subjects. Most of them stated they were normal on the day following the experiment. The records of 25 subjects were reexamined and only their subjective accounts used. In these accounts they described how they felt the following day. Thirty-two subjects had placebo. Four stated there was some residual effect which consisted of some dizziness in three and euphoria in one. Most subjects received Tuinal 200 mg at bedtime. Eighteen subjects received adrenolutin. Of these, nine had prolonged reactions. Chi-square for this is about 6.4. There is less than 2%chance that this difference is due only to chance. Furthermore, not a single placebo subject had any residual reaction lasting more than one-half a day. Yet many adrenolutin subjects were not normal by noon and some were clearly abnormal for several days; there was one reaction that lasted 2 weeks.

C. EFFECTOF ADRENOCHROME ON HUMANS AS REPORTED BY OTHERLABORATORIES In order that investigators can judge a report, it must include a description of the subjects used, the nature of the chemical, and the results which were found. Using these criteria, there have so far been recorded corroborative results from three independent investigators; there are no adverse papers, although apparently sporadic trials by some have proven negative. Rinkel et al. (1954) showed that adrenochrome semicarbazone did not produce changes in human volunteers. This is the derivative of adrenochrome which is used for hemostasis. It is not hydrolyzed in the body to adrenochrome; neither is it active in producing changes in spiders. The first group of workers to corroborate our work was Schwarz

354

A. HOFFER

et al. (1956a). On three occasions, they gave 50, 60,and 75 mg of adrenochrome by vein to subject 1in their series. He suffered body image disturbances and had a loosening of associations. Subject 2 was given 50 and 60 mg. He developed a pleasant smile and marked relaxation. He experienced catalepsy on both occasions which persisted for more than 30 minutes. At these times, he held his arms in unnatural positions for periods of time which could not be matched by controls. He had never shown this before, and it was not induced by mescaline or LSD. The epileptic patient was very relaxed and drowsy after adrenochrome but there were no other changes. Szatmari et al. (1955) also reported that little change occurred in chronic deteriorated epileptics until 50 mg of adrenochrome was given. This is not surprising; apparently, they react minimally to mescaline (Denber and Merlis, 1955~). The second group of workers to find activity in adrenochrome was Taubmann and Jantz (1957). Taubmann explained the increased toxicity of Novocain when applied sublingually to venous anastomosis between the buccal mucosa and the cerebral cortex. Man, for many years, has taken his euphorients by his buccal mucosa, e.g., coca, betel, hashish, tobacco (snuff), and cocaine. By this route, decomposition of the active principle by blood or liver is avoided. These authors therefore gave adrenochrome sublingually and reported marked psychological activity in their subjects. Grof ( 1960) and Grof et al. (1961) summarized a year’s research with adrenochrome. The adrenochrome was synthesized by Dr. V. Vitek according to Feldstein (1958) or purchased from L. Light and Company. They carried out double blind studies on 15 volunteers using subjects very similar to those used in our studies in Saskatchewan, i.e., intelligent, educated, normal subjects as well as some psychiatric patients. Many of their subjects were sophisticated in psychological experiments, having taken LSD, mescaline, or psilocybine. The placebo was a red dye; the dose of adrenochrome varied between 15 and 30 mg sublingually. In the double blind design, out of 15 subjects given placebo, only one thought he had received an active compound. Out of 15 subjects given adrenochrome, only 4 believed they had received placebo. [Chi-square for 1 d.f. is over 11 ( P < 0.001).]

Clinical Changes Perception. There were no perceptual changes in 6 subjects.

EFFECT OF ADRENOCHROME .4ND ADRENOLUTIN

355

Changes in body image including depersonalization and derealization occurred in 4 subjects. Of these one had a disorder of body image and derealization and felt his legs were short. Visual perceptual changes occurred in 5 subjects and ranged from increased sensitivity to color, to illusions, pseudo-hallucinations, and hallucinations. Auditory changes were reported by 4 subjects and included increased acuity for sound to clear auditory hallucinations of mysterious messages in telegraphic code coming from the universe. Tactile hallucinations occurred in 3 subjects. Taste and olfactory hallucinations were not reported. Eight had alterations in perception and estimation of time. Very few and vague changes were reported by 3 placebo subjects. They included transitory derealization, and minor undulations in the visual field. Thought. There were no clinical changes in thought in 5 subjects. Some of these showed marked changes on association tests. Paranoid and other delusions were present in 7 subjects. Changes in tempo of thinking such as flight of ideas, difficulty in concentrating, blocking, and speech alterations occurred in 5. One showed negativism, ambivalence, and splitting of personality. Another developed inappropriate behavior such as sitting in a wastebasket or creeping along the floor. Two subjects had no insight into the fact that their condition had been changed. None of the placebo subjects showed any pathological changes in thinking. The most sensitive method for demonstrating the central effects of adrenochrome was the word-association experiment. There was a high frequency of disturbed associations compared to the placebo experiments. It was significantly different, at the 1%level at 30 minutes, 2-5% level at 60 minutes, and at the 5%level at 120 minutes. The latency period, i.e., the time between stimulus word and response, was prolonged significantly by adrenochrome at the 5%level at 2 hours. In 11 subjects given 330 verbal stimulus words, there were 81 disturbed associations (25%). The most frequent were clang associations. There were only 6 to 7% disturbed associations with placebo. The quantity of disturbed associations is about the same as for schizophrenic patients. The authors concluded that in many cases the subjects formed answers before they understood the meaning of the stimulus word. But for other cases, the origin of the disturbed association was not known. In a few subjects disturbed associations carried on until the next placebo experiment although they had been normal before. This they had never observed with

356

A. HOFFER

LSD, mescaline, or psilocybine. They finally concluded that the changes in thinking induced by adrenochrome were similar to those observed in schizophrenia. Adrenochrome caused an elective inhibition of the process which determines the content of associative thinking. This occurred in doses which did not heighten lability of basic processes, did not reduce excitation, and did not loosen temporary connections as was the case with LSD. Sommer et al. (1960) used the Kent-Rosanoff word-association test for testing the hypothesis that schizophrenics had a specific language which had been suggested by some psychoanalytic writers. They found that a group of 49 schizophrenic patients gave about 15% uncommon responses. This degree of unusual responses is very similar to the 25%disturbed associations found by Grof et al. ( 1961). The nonschizophrenic group of 69 subjects gave only 7%uncommon responses. Furthermore, schizophrenics were less stable on repetition of the test ( P < 0.02) and fewer patients “thought alike” ( P < 0.001). Mood. Eight subjects reported or demonstrated no changes in affect. Euphoria and silly laughter or giggling occurred in six. Three subjects had anxiety, one was fearful and one became hostile and depressed. Very often early tension or anxiety was replaced by euphoria and relaxation. Comparbon to Other Psychotomimetic Experiences. Most of the subjects had not taken other hallucinogens and so had no basis for comparison. Of the group that did, two compared it to mild psilocybine experiments and three to mild LSD reactions but in each instance without the autonomic changes. General. The other tests used were in agreement with the clinical observations. The changes in 3 subjects varied from no reaction to severe schizophrenic-like states. Nine subjects received doses of 30 mg sublingually. Four suffered endogenous Bonhoeffer type psychosis, 3 schizophrenic-like psychosis, and in one the reaction was doubtful. One failed to react. There were thus 7 out of 9 reactors, or nearly 8Q%.When 15 mg was given, there were 6 definite reactions ( 1 toxic, 1 schizophrenic-like, and 4 neurotic), i.e. nearly 40%. Five subjects had uncertain reactions and 5 were without reaction. I have reviewed this work in some detail because of its importance. It is the first double blind study with adrenochrome on humans and fully corroborates the Saskatchewan findings.

EFFECT OF ADRENOCHROME AND ADRENOLUTIN

357

D. ADRENOCHROME GIVENSUBLINGUALLY The Taubmann and Jantz method for giving adrenochrome seems most useful. We therefore ran a few trials with crystalline adrenochrome. Since two adrenochromes are known, made from either d- or I-epinephrine both forms and the dl mixture were tested. 1. Adrenochrome f r o m l-Epinephrine In several studies 6 mg of d-adrenochrome given sublingually produced no change. In one instance, 10 mg produced a change which was very clear and obvious for about 2 days. This case is unusually well documented. The usual adrenochrome changes in perception and thinking resulted. The subject was depressed and irritable, and suffered a marked change in personality. In another case, 3 mg did produce a marked change. After 30 minutes, there was a flush. One and a half hours later, the subject developed strong feelings of isolation. The room became unclear visually and there seemed to be much movement about him. The rest of the day passed without having any sensation of time. In the evening, while in bed, he momentarily hallucinated the face of a person close to him and he became very anxious. In the morning, he was normal.

2. Adrenochrome f r o m d-Epinephrine This was synthesized by R. A. Heacock in 1958. Three mg was given sublingually to two subjects. Subject 1 observed some difficulty in reading and focusing at 7 minutes. At 10 minutes, he was light-headed. At 12 minutes, far objects seemed very far away, At 24 minutes, he was euphoric and could not estimate time. At 35 minutes, time had seemed almost stationary. At 45 minutes, colors were very bright and vivid. At 80 minutes, he was very active in speech and movements and abrupt with people. At 5% hours, he felt normal. However, for the next 24 hours, people’s faces and other objects would become alternately small and large in size. In one instance, a speaker’s face appeared to move away and toward him. Objects moving toward him increased in size too quickly. Subject 2 also received 3 mg. At 4 minutes, printing on a page became blurred and he had difficulty grasping the meaning of words. At 9 minutes, he became tired and his vision was blurred. At 30 minutes, his thinking was fuzzy. He was apathetic and could not concentrate. The outline of his hand seemed blurred, At 39 minutes, curtains in the room appeared to shimmer. At 52 minutes, he was withdrawn. For the next 30 hours, he was depressed, withdrawn, and disinterested.

358

A. HOFFER

3. Adrenochrome from dl-Epinephrine The same subject who had taken 3 mg d-adrenochrome (from 1-epinephrine) took 3 mg dl-adrenochrome sublingually. Slight changes in perception ( dizziness, light-headedness, increased brightness of room, changes in size of far objects) occurred. There was no change in thought and there was slight euphoria. After 2 hours, he found the experience unpleasant and took 1 gm of nicotinic acid by mouth. That evening, he was irritable, restless, and without ambition. He was bothered by the odor of new wax on the floor, and later by insomnia.

The following subjects received 6 mg of dl-adrenochrome sublingually . Subject 1 took his adrenochrome at 4:OO PM. Ten minutes later, he had an anesthetic area over both cheeks and he had diEculty in focusing. At 4:15, he was very quiet and appeared sad but denied this. He underestimated the size of objects about 20% Three minutes later, he was dizzy as if he would faint. At 4:25, he could no longer estimate passage of time. He thought he had been in all afternoon. His limbs became very light. His hands changed in size as he looked at them and the observer’s face changed in size. The rest of the hour, he found paintings unusually vivid. At the end of the hour, he had a headache in the occipital area and felt indifferent. Because of his discomfort, he was given 1 gm of nicotinic acid. In 10 minutes as he began to flush, the perceptual changes vanished and he felt normal. That night, he slept lightly and was not sure whether he had been awake or asleep (twilight sleep). The next morning, he was very tired and considered not coming to work. At work, he was irritable all day. Subject 2 had had much experience with LSD and was skilled at introspective observation. At 2:OO PM, he received the adrenochrome. In 5 minutes, he became aware that colors and detail were more distinct. People in pictures seemed more lifelike and larger. In 10 minutes, he had a marked frontal headache. He was able to read but could not make sense out of what he read. At 20 minutes, the visual changes were very clear. He estimated 30 seconds as 45 seconds (mean of 3 trials). At 25 minutes, he looked older in the mirror. He was relaxed and disinterested. His headache was almost gone but he felt clumsy when moving. One hour after starting he was depressed and irritable. His face was flushed. He was withdrawn and indifferent. While lined up in a cafeteria for coffee, the other people appeared to be puppets. When he drank his coffee, he complained about the noisiness. He felt the people around him were puppetlike, lacked understanding. They annoyed him but he stated he was superior to them. They seemed empty people. At 3:15, his facial flush was gone. The white uniforms of nurses in the cafeteria annoyed him. At 3:40, he thought 2 hours had elapsed since taking adrenochrome. He markedly overestimated size of objects ( 12 trials). At 2 hours, he felt music was being played at half speed. He likened the experience to the initial symptoms of LSD. It

EFFECT OF ADRENOCHROME AND ADRENOLUTIN

359

seemed like 3%hours since taking it. His headache was now gone. H e reported that the most pleasant part of the experience was that it was wearing off. The next hour, he was easily confused and still could not estimate time correctly. After that he was normal.

E. POTENTIATION OF THE ACTION OF ADRENOCHROME Melander and MBrtens (1958) found that lysergic acid diethylamide (15 to 30 @kg) and taraxein when given ahead of adrenolutin markedly potentiated its effect. Thus 20-25 mg/kg when given by vein produced only slight changes in cats. But pretreated cats showed a marked response of drowsiness and muscle relaxation after 2-3 mg/kg. Acetyl-LSD also was a potentiator but bromo-LSD was not. MBrtens et al. ( 1 9 5 9 ~ )reported that LSD and taraxein sensitized cats to acetylcholine, epinephrine, atropine, chlorpromazine, histamine, mescaline, and serotonin. They therefore made the sensible suggestion that taraxein, LSD, and dl-acetyl lysergic acid diethylamide ( ALD ) increase the permeability of these substances through the blood-brain barrier, i.e., they “have the property of enabling certain intravenously injected drugs to act on selected brain centers not normally accessible to them.” Hoffer (1959b) and Hoffer and Osmond (1960) found that humans reacted to the combination of LSD followed by adrenochrome or adrenolutin in the same way. There was a marked potentiation of the adrenochrome effect, This was especially notable in human subjects who reacted to LSD primarily by the production of severe tension and anxiety. Visual and psychedelic changes were minimal if present at all. In these subjects, the injection of 10 mg adrenochrome or adrenolutin produced a certain relaxation from tension and the usual LSD experience of marked visual and other changes. Adrenochrome can be used in this way to help break across the tension barrier into the psychedelic experience, which is helpful in treating alcoholics. Heath et al. (1958) postulated that ceruloplasmin formed part of a protective system. Its function would be to protect the body against amines or their metabolites liberated during stress. MBrtens et al. (1959a, b ) provided powerful evidence in support of this idea. Ceruloplasmin irreversibly binds adrenolutin ( Melander, 1957) and histamine (Mh-tens et al., 19594. When animals were pretreated with ceruloplasmin, they were protected against the psychotomimetic properties of LSD alone or LSD followed by other com-

360

A. HOFFER

pounds listed by these authors above. It also decreased the toxicity of histamine. Further support for the protective role of ceruloplasmin were the interesting therapeutic responses of schizophrenic patients to ceruloplasmin, reported by MArtens & al. ( 1959a,b ) .

F. DISCUSSION The basic issue is whether or not adrenochrome and compounds which can be derived from it in vivo, perhaps adrenolutin and 5,6dihydroxy-N-methylindole, are active hallucinogens in man. The evidence that these compounds exist in the body is growing more substantial. The fact that these compounds are active in producing changes in animals can no longer be denied, and indeed it was concluded by Kety (1959) that there was no doubt it did produce changes in animals. Adrenochrome will therefore be most interesting to physiologists who are following the properties of metabolic products of epinephrine. The fact that adrenochrome and adrenolutin have produced changes in the perception, thinking, and feeling of humans makes them very interesting for psychiatrists and physiological psychologists, Although Smythies (1960) denies that adrenochrome is active in humans, the literature reviewed in this paper, as well as the detailed outline of new data from our research, makes clear that it is no longer sufficient merely to state that adrenochrome is not active. If it were true that adrenochrome is indeed inactive in humans, we would then have the curious situation of a chemical which is active in many species of animals, including the monkey, being inactive in man. In general, it seems to be true that when more sophisticated tests of animal behavior are used, smaller quantities of adrenochrome are effective in producing change. This is most clearly established for monkeys. Thus, Heath, giving 100 mg of adrenolutin to monkeys fixed in a chair, and using the animal’s rage as an index of activity, saw no effect unless the animal was pretreated with taraxein. I n sharp contrast, Iordanis and Kuchino (1959) using very refined conditioned reflex techniques showed that 1 mg of adrenochrome abolished sequential conditioned responses. The monkey reacted appropriately to the sight of food but did no longer respond to the symbols signifying food. It, in effect, became very concrete in its thinking. Similarly for rats, the refined experiments of Weckowicz

EFFECT OF ADRENOCHROME AND ADRENOLUTIN

361

(1961) clearly showed a remarkable effect of adrenochrome in decreasing learning and in increasing extinction of acquired conditioned reflexes. X. Mode of Action of Adrenochrome

Theoretically, chemicals could affect the mind by interfering with its media of input or output. If perceptual stimuli were distorted by some defect in the retina of the eye, then there might be some disorder of those aspects of mind which depend upon the accurate reception of sensory data. In fact, Fogel and Hoffer (1961) have produced a large series of models of psychiatric syndromes merely by producing simple alterations in perception. Psychological methods only were used. Adrenochrome and adrenolutin are chemically very reactive substances and react with an amazing variety of constituents of the body. They could therefore somehow interfere with sensory perception. However, the evidence gathered in this review suggests clearly that the brain is the chief target. This evidence is that (1) given into the ventricles of the brain, the compounds are much more active than when given by vein; ( 2 ) pretreatment with taraxein, LSD, and ALD potentiates their action; ( 3 ) pretreatment with ceruloplasmin protects the animal against their toxic effect; (4) adrenochrome produces hypothermia in normal and adrenalectomized rats ( Hutcheon et al., 1956) without decreasing the consumption of oxygen (Eade, 1954); and (5) anxiety increases the susceptibility of normal human subjects to adrenolutin (Hoffer, 1 9 5 7 ~ )In . contrast, however, is our finding that severely tense alcoholics and endogenous depressives react very little to adrenochrome given by vein. Braines et al. (1959) found that fear potentiated the reaction of dogs to adrenochrome. Recently, Maze1 and Bush (1961) found that epinephrine brought about a great increase in the rate of entry of barbital into mouse brain. Norepinephrine had little effect. The concept of the blood-brain barrier has been reviewed by Aird ( 1956). ( 6 ) Adrenochrome alters the EEG pattern from the depths or from the surface in animals and on the surface in man. Adrenochrome may affect the brain by interfering with the blood-brain barrier. Thus, Greig and Gibbons (1959) found that adrenochrome decreased the penetration of glucose, labeled with CI4, into mouse brain; bulbocapnine, bufotenine, and mescaline were much more

362

A. HOFFER

active. The effect of LSD was variable, while adrenolutin and iproniazid had no effect. They further found that human serum pseudocholinesterase was inhibited 50%by the following concentrations of LSD, bulbocapnine, bufotenine, adrenolutin, and adrenochrome: 9 x lo-;, 2 x 1 e 5 , 4 x lo-*,3.8 x le3, and 3.3 X le3M. They therefore suggested that hallucinogens could act by decreasing transfer of glucose into brain. In the brain substance there are many enzyme systems which would be inhibited. These include ( 1 ) hexokinase ( Bullough, 1952; Gelfant, 1960; Walaas and Walaas, 1956; Takahashi and Akabane, 19f30) and in general (2) the glycolytic cycle (Cohen and Hochstein, 1960; Hochstein and Cohen, 1960; Korzoff and Kuchino, 19%; Meyerhof and Randall, 1948; Radmsa and Golterman, 1954; Randall, 1946; Woodford, 1959). Cohen and Hochstein found that adrenochrome as well as some other quinones inhibited the production of lactic acid from glucose by mouse-brain homogenate. They suggested that these compounds might have an in vivo role in regulating energy production from glucose in the central nervous system. Hochstein and Cohen found that brain tissue was more sensitive to adrenochrome than was liver tissue. The mitochondria seemed to be most sensitive. Liver supernatant protected brain tissue against inhibition by M adrenochrome. Liver mitochondria were sensitized by brain supernatant. Frederic (1954) had shown that mitochondria of living cells were inhibited by adrenolutin. ( 3 ) Oxidative phosphorylation is uncoupled by adrenochrome (Park et d., 1956a, b) . ( 4 ) ATPase is inhibited (Inchiose and Freedberg, 1961). ( 5 ) Glutamic acid decarboxylase is inhibited (Holtz and Westermann, 1956). (6) Coenzyme A is deactivated (Roston, 19f30). The effect of adrenochrome on neurons (Geiger, 1960) has already been described. It also has some inhibitor action on synaptic transmission (Marrazzi, 1957; Hart et al., 1956) and some antifatigue effect on sympathetic nerves (Derouaux and Roskam, 1949). The pigments in the cells of the central nervous system are apparently not derived from tyrosine. Foley and Baxter (1958) examined the brains of 2 albino human subjects. In both cases, the intensity and pigmentation in the cells of the locus caeruleus and substantia nigra appeared normal. One patient also had several pigmented cells in the dorsal motor nucleus of the vagus. In contrast neither brain had any melanin pigment present in the pial melano-

EFFECT OF ADRENOCHROME AND ADRENOLUTIN

363

phores. They therefore concluded that “melanins” of the brain stem differ fundamentally from the melanin of skin, choroid, and pia in that they are not formed by the action of the tyrosinase complex. Albinos lack tyrosinase. Other sources of melanin-like pigment include epinephrine, norepinephrine, hydroxytyramine, dopa, and tryptophan. Fellman ( 1958) found that substantia nigra contained enzymes which oxidized epinephrine. He observed argentophilic granules not unlike those seen in the adrenal medulla and in other chromaffin tissue. Epinephrine and norepinephrine might be the main source of these pigments. The evidence for this view is as follows: (1) Brain tissue is very rich in epinephrine oxidases (Payza and Hoffer, 1959) and contains argentophilic cells ( Fellman, 1958). ( 2 ) Brain contains inhibitors of epinephrine oxidation ( Walaas and Jervell, 1958). One would expect both activators and inhibitors to be present in tissue in which epinephrine is oxidized to adrenochrome. Brain is rich in ascorbic acid, which tends to inhibit oxidation of epinephrine to adrenochrome. ( 3 ) The increase in density of brain pigments parallels closely the increase in secretion of epinephrine during fetal and postnatal development. The medulla contains chiefly norepinephrine at a late stage of intrauterine life. After birth there is a rapid reorganization of the medulla and the proportion of epinephrine increases rapidly for about 2 years. After 3 years the Zuckerkandl bodies show very little activity (West et al., 1951, 1953). Foley and Baxter found brown-black granules in cells of the locus caeruleus as early as the fifth month of gestation, after which their number increased rapidly. Similar granules were not present in the substantia nigra until 18 months, after which they increased. But the locus caeruleus always had had more until puberty, Since epinephrine is oxidized to adrenochrome more readily than norepinephrine is to noradrenochrome, it is possible, that the formation of the pigment depends upon the production of epinephrine. (4) Although methoxylation of phenolic hydroxyls may be a main pathway of epinephrine and norepinephrine degradation, it is not yet clearly established whether or not this is also the main pathway in brain. Thus, Weil-Malherbe et al. (1961) found that pyrogallol, an inhibitor of catechol-O-methyl transferase, does not affect concentration of brain norepinephrine unless it is combined with an m i n e oxidase inhibitor. Borovitz and Merritt ( 1961) reported that the

:364

A. HOFFER

iusual studies of epinephrine metabolism refer to liver metabolism. 'They found that heart, brain, and liver handled epinephrine in #different ways. Spector et d. (1960) provided evidence that methoxylation is not the main enzyme for inactivating norepinephrine in brain. Pyrogallol did not block metabolism of brain amines released by reserpine. They suggested that monamine oxidase was responsible for metabolism of norepinephrine in tissues where it regulates levels of stored amines, while catechol-O-methyl transFerase inactivated catechol amines after they were released into the circulation. (5) In Wilson's disease, there is an excess accumulation of copper combined with a deficiency of ceruloplasmin (DennyBrown, 1953). The excess of copper would favor the oxidation of epinephrine to adrenochrome and the lack of ceruloplasmin would intensify the pathological effects. The chief pathological changes are in the brain and liver. Clinically, there are severe neurological changes including tremor and marked mental changes. Barbeau (1960) found evidence for an abnormality of catechol amine metabolism in basal ganglia diseases, and Domer and Feldberg (1960) found that minute quantities of epinephrine placed in the brain ventricles had marked antitremor properties. I postulate that in Wilson's disease a rapid conversion of epinephrine to adrenochrome occurs (due to copper deposition). This would account for the tremor (due to a lack of epinephrine), progressing to increased deposition of brain pigment ( a copper adrenochrome melanin complex), psychosis (due to excess adrenochrome and adrenolutin combined with lack of ceruloplasmin ), and finally destruction of neurons by adrenochrome as was found by Geiger (1960). These ideas are highly conjectural; perhaps newer techniques for staining adrenochrome and adrenolutin-like pigments will reveal whether they are indeed present. If ceruloplasmin does play a role in protecting the brain against toxic amines and indoles as has been suggested, then serum ceruloplasmin levels need to be considered as a factor determining the response of animals given adrenochrome and adrenolutin. Pregnant animals, especially just before term, when the production of ceruloplasmin by the placenta is at a maximum, should be quite resistive to the action of LSD or adrenochrome, but Peck (1960) indicates that no papers are available which report tests of this hypothesis. A woman 6 months pregnant reacted in the usual way to a large dose

EFFECT OF ADRENOCHROME AND ADRENOLUTIN

3%

of LSD but at this stage ceruloplasmin levels have not increased very much. It is perhaps more than coincidence that toxemia of pregnancy is more common in schizophrenics than in controls ( Wiedorn, 1954), and that epinephrine levels are elevated just before parturition (and apparently do little harm) but decrease very quickly during labor (Ritzel et al., 1957). Puerperal psychosis came on most frequently in the few weeks after delivery. During this time, ceruloplasmin levels decrease very quickly. Linn (1941) reported that 59 cases of post pnrtum psychosis out of a series of 76 developed within 2 weeks. Of these, 42 occurred within 7 days. The third day was apparently a critical one. Paffenbarger et al. (1961) found that from a series of 125 women developing puerperal psychosis, 96 came on in the first month and 109 in the first 2 months (87%). XI. Conclusions

Adrenochrome and adrenolutin produce marked changes in behavior in spiders, pigeons, rats, cats, dogs, and monkeys. The finer aspects of behavior appear to be altered first. Gross changes are produced by larger quantities. Activity is also seen in human subjects. The types of change produced mimic in many ways the changes seen in schizophrenia. The kind of visual hallucinations seen with mescaline, LSD, psilocybine, and other substances is not produced. The findings of our group have been corroborated in man by three independent research centers, while no research paper has reported details of failure to corroborate. REFERENCES Abood, L. G. (1957). Conference on Biochemistry and Mental Illness, University of British Columbia, Vancouver. Abramson, H. A. ( 1955). Personal communication. Aird, R. B. (1956). Sci. American 194, 101. Alles, G . A. (19159). In “Neuropharnicicology” ( H . A. Abramson, ed.), p. 181. Josiah Macy, Jr. Foundation, New York. Altschule, M. D. (1960). J. Neuropsychiat. 2, 71. Bacq, Z. M. (1949). J. Phurmucol. Exptl. Therap. 95, 1. Bacq, Z. M., and Renson, J. (1961). Arch. intern. pharmucodynumie 130, 385. Barbeau, A. ( 1960). Neurology 10,446. Bartlett, F. (1958). “Thinking, An Experimental and Social Study.” Allen and Unwin, London. Baruk, H. ( 1957). 2nd International Congress of Psychiatry, Zurich. Personal communication.

:366

A. HOFFER

Bell, G. H., Davidson, J. N., and Scarborough, H. (1959). “Textbook of Physiology and Biochemistry.” Livingstone, London. Bercel, N. A. ( 1960). A.M.A. Arch. Gen. Psychiat. 2, 189. Blaschko, H., and Schlossmann, H. ( 1940). J. Physiol. ( L o n d o n ) 98, 130. Bleuler, M. (1956). Deut. med. Wochschr. 27, 1078. Borowitz, J. L., and Merritt, J. H. ( 1961). Federation Prm. 20, 340. Braines, C. H., Konev, C. B., Golubieva, G. P., Kuchino, E. B., and Kobrinckaio, 0. J. ( 1959). “Biophysical Research in Psychiatry,” collected papers. Laboratory of Experimental Pathology, Institute of Psychiatry, Academy of Sciences, Moscow. Buehler, K. ( 1960). International Congress of Psychology, Bonn University, Germany. Bullough, W. S. (1952). J. Endocrind. 8,265. Bullough, W. S. (1955). I n “Vitamins and Hormones” (R. S. Harris and K. V. Thimann, eds.), Vol. 13, p. 261. Academic Press, New York. Capek, R., Vojtechovskj., M., and Vitek, V. (1960). Actio. Neru. Super. 2, 333. Cohen, G., and Hochstein, P. ( 1960). Federation Proc. 19, 265. de Jong, H. H. ( 1945). “Experimental Catatonia.” Williams & Wilkins, Baltimore, Maryland. de Jong, H. H. (1955). Trans. Soc. B i d . Psych&. p. 24. de Jong, H. H. (1956). J. Clin. Exptl. Psychoputhol. 17,388. Denber, H. C . B., and Merlis, S. (1955a). Psychiat. Quart. 29,421. Denber, H. C . B., and Merlis, S. ( 195513). Psychiat. Quart. 29,430. Denber, H. C. B., and Merlis, S. ( 1 9 5 5 ~ )Psychiat. . Quart. 29,433. Denny-Brown, D. ( 1953). In “Metabolic and Toxic Diseases of the Nervous System” ( H . H. Merritt and C. C. Hare, eds.), Vol. 32, p. 190. Williams & Wilkins, Baltimore, Maryland. Derouaux, G., and Roskam, J. (1949). J. Physiol. (London)108, 1. Domer, F. R., and Feldberg, W. (1960). Brit. J. Phurmacol. 15, 578. Eade, N. R. (1954). Ph.D. Thesis, Graduate School, University of Saskatchewan, Saskatoon, Canada. Ellis, H. ( 1898). Contemporary Reo. 73, 130. Evarts, E. V. (1958). Proc. SOC. Exptl. Biol. M e d . 98, 74. Evarts, E. V., Gillespie, L., Fleming, T. C., and Sjoerdsma, A. (1958). Proc. SOC. Exptl. Bwl. M e d . 98, 74. Feldberg, W., and Shenvood, S. (1954). J . Physiol. (London) 123, 148. Feldstein, A. (1958). Science 128, 28. Fellman, J. H. (1958). J. Neuro2. Neurosurg. Psych&. 21, 58. Fischer, P., and de Landtsheer, L. ( 1950). Experentia, 6, 305. Fischer, P., and Lecomte, J. (1949). Arch. intern. physiol. 36, 327. Fischer, P., and Lecomte, J. (1951). BUZZ. soc. chim. biol. 33, 569. Fogel, S., and Hoffer, A. (1962). J. Clin. 6 Exptl. Psychopath. 23, 24. Foley, J. M., and Baxter, D. (1958). J. Neuropathol. Exptl. Neurol. 17, 586. Frederic, J. (1954). Ann. N . Y. Acad. Sci. 58, 1246. Geiger, R. S. (1960). J. Neuropsychiat. 1, 185. Gelfant, S. ( 1960). Exptl. Cell Research 21, 603. Gerard, R. W. (1960). I n “Disease and the Advancement of Basic Science”

EFFECT OF ADRENOCHROME AND ADRENOLUTIN

367

(H. K. Beecher, ed.), p. 234. Harvard Univ. Press, Cambridge, Massachusetts. Gershenovich, Z. S., Krichevskaya, A. A., and Kalnitsky, G. (1955). Ukruine Biokhim Zhur. 27,3; Chem. Abstr. 49, 10470. Goldenberg, M., Aranow, H., Smith, A. A., and Faber, M. (1950). A.M.A. Arch. Internal. Med. 86, 823. Green, S., Mazur, A., and Shorr, E. (1956). J . Biol. C h m . 220, 237. Greig, M. E., and Gibbons, A. J. (1957). Federation Proc. 16, 302. Greig, M. E., and Gibbons, A. J. (1959). Am. J . Physiol. 196, 803. Grewal, R. S. (1952). Brit. J . PhumcoE. 7, 338. Grof, S . (1960). Actiu. N e w . Super. 2, 426. Grof, S., Vojtechovsk9, M., Vitek, V., and Frankova, S. (1961). Abstract of paper. International Congress Psychiatry, Montkal, June. Hart, E. R., Langfitt, T. W., and Marrazzi, A. S. (1956). Federation Proc. 15, 88. Heacock, R. A. (1959a). Chem. Reus. 59, 181. Heacock, R. A. (1959b). Chem. d? Ind. (London), 1959, 752. Heacock, R. A., and Laidlaw, B. D. ( 1958). Nature 182,526. Heacock, R. A., and Mahon, M. (1958). Can. J. Chem. 36,1550. Heacock, R. A., and Scott, B. D. (1959). Can. J . Biochem. and Physiol. 37, 1087. Heacock, R. A., Nerenberg, C., and Payza, A. N. (1958). Can. J. Chem. 36, 853. Heath, R. G. (1959). In “Neuropharniacology” ( H . A. Abramson, ed.), pp. 37-151. Josiah Macy, Jr. Foundation, New York. Heath, R. G., Leach, B. E., Byers, L. W., Mlrtens, S., and Feigley, C. A. (1958). Am. J. Psychiat. 114, 683. Heath, R. G., Leach, B. E., and Cohen, M. (1959). In “The Effects of Pharmacologic Agents on the Nervous System” (F. J. Braceland, ed.), p. 37. Williams & Wilkens, Baltimore, Maryland. Hochstein, P., and Cohen, G. ( 1960 ). Federation Proc. 19, 265. Hoffer, A. ( 1954). Unpublished observations. Hoffer, A. (1957a). J . Clin. Exptl. Psychopathot. 18,27. Hoffer, A. (1957b). In “Psychotropic Drugs” (S. Garattini and V. Ghetti, eds.), p. 10. Elsevier, Amsterdam, Netherlands. . “Hormones, Brain Function, and Behavior” (H. HoagHoffer, A. ( 1 9 5 7 ~ )In land, ed.), pp. 181-194. Academic Press, New York. Hoffer, A. (1957d). In “Tranquilizing Drugs” ( H . E. Himwich, ed.), p. 73. Am. Assoc. Advancement Science, Washington, D. C . Hoffer, A. (1958a). Conference on Amine Oxidase Inhibitors. New York Academy of Science, November. Hoffer, A. (1958b). Am. J. Psychiut. 114,752. . Neuropsiqukit. Arg. 4, 355. Hoffer, A. ( 1 9 5 8 ~ )Acta Hoffer, A. ( 1959a). In “Molecules and Mental Health” (F. A. Gibbs, ed.), p. 44. Lippincott, Philadelphia, Pennsylvania. Hoffer, A. ( 195913). In “Biological Psychiatry” (J. H. Masserman, ed.). Grune and Stratton, New York.

I168

A. HOFFER

Hoffer, A. ( 1960). Diseases of Nervous S y s t m 21, 79. Hoffer, A., and Callbeck, M. J. (1960). J. Mental Sci. 106, 138. Hoffer, A., and Kenyon, M. (1957). A.M.A. Arch. Neurol. Psychiat. 77,437. Hoffer, A., and Osmond, H. (1955). J. Nervous Mental Disease 122,448. Hoffer, A., and Osmond, H. (1959). J. Nervous Mental Disease 128, 18. Hoffer, A., and Osmond, H. (1960). “Chemical Basis of Clinical Psychiatry.” Charles C Thomas, Springfield, Illinois. Hoffer, A., and Payza, A. N. (1958). Annual Meeting, American Psychiatric Assoc., San Francisco, California. Hoffer, A,, Osmond, H., and Smythies, J. (1954). 1. Mental Sci. 100, 29. IIofmann, A. (1960). International Symposium on the Chemistry of Natural Products, Canberra, Australia. Hofmann, A., and Tscherter, H. (1960). Experentia 16,414. Hofmann, A., Heim, R., Brack, A,, and Kobel, H. (1958). Experentia 14, 107. Holtz, P., and Westermann, E. ( 1956). Naturwissenschuften 43, 37. Hutcheon, D. E., Lowenthal, J., and Eade, N. R. (1956). Arch. intern. p h m m o d y n u m i e 106, 90. lnchiose, M. A., Jr., and Freedberg, A. S. (1961). Federation Proc. 20, 298. lordanis, K. A., and Kuchino, E. B. (1959). In “Problems of Experimental Pathology” ( C. H. Braines, ed. ), collected papers. Laboratory of Exptl. Pathol., Institute of Psychiatry, Academy of Sciences, Moscow. Jantz, H. (1956). Z. ges. Neurol. Psychiat. 137, 141. Kaufmann, H., and Koch, E. (1959a). Presse m‘d. 67, 1141. Kaufmann, H., and Koch, E. (195913). Arch. des MaEadies du Coeur. Rev. Ateroscler. (Paris), 1, 290. Kety, S. (1959). Meeting of Eastern Psychiatric Research Assoc., New York. Panel discussion following Hoffer (1960). Kisch, B. (1947). Exptl. Med. Surg. 5, 166. IUuver, H. (1928). “Mescal, the ‘Divine’ Plant and Its Psychological Effects.” Kegan Paul, Trench, Trubner, London. Korzoff, B. A., and Kuchino, E. B. (1959). I n “Problems of Experimental Pathology” (C. H. Braines, ed.), collected papers. Laboratory of Exptl. Pathol., Institute of Psychiatry, Academy of Sciences, Moscow. IGupp, P., and Monnier, M. ( 1960). Experentia 16, 206. I
EFFECT OF ADRENOCHROME AND ADRENOLUTIN

369

Lehrman, D. S. (1956a). In “L’instinct dans le comportement des animaur et de l’homme.” Masson, Paris. Lehrman, D. S. (195613). Ann. Rev. Physiol. 18, 527. Lehrman, D. S. (1958). J. Comp. and Physiol. Psychol. 51, 142. Lehrman, D. S. (1959). Trans. N . Y. Acad. Sci. 21,682. Lehrman, D. S., and Wortis, R. P. ( 1960). Science 132, 1667. Leimdorfer, A. (1950). J. P h a m c o l . 98, G2. Leimdorfer, A., and Metzner, W. R. T. (1949). Am. J. Physiol. 157, 116. Leimdorfer, A., Arana, R., and Hack, M. H. (1947). Am. J. Physiol. 150, 588. LettrB, R., and Albrecht, M. (1941). 2. physiol. Chem. Hoppe Seyler’s 271, 200. Lett&, R. (1954). Ann. N . Y. Acad. Sci. 58, 1085. Lewin, L. (1931). “Phantastica: Narcotic and Stimulating Drugs: Their Use and Abuse.” Kegan Paul, Trench, Trubner, London. Linn, L. (1941). Diseases of Nervous System 2, 290. Marrazzi, A. (1957). Ann. N . Y. Acad. Sci. 66,496. MHrtens, S., Vallbo, S., and Melander, B. (1959a). Intern. Rev. Neurobiol. 1, 333. Mlrtens, S., Vallbo, S., and Melander, B. ( 195913). In “Biological Psychiatry” ( J . H. Masserman, ed.), p. 273. Grune and Stratton, New York. . Psychiat. Mlrtens, S., Vallbo, S., Anderson, K., and Melander, B. ( 1 9 5 9 ~ )Acta Neurol. S c a d . 34, 361. Maslova, A. F. (1959). Biokhhitya 24, 181. Mazel, P., and Bush, M. T. (1961). Federution Proc. 20, 306. Meirowsky, E. (1940). Brit. J. Dermutol. Syphilis. 52, 205. Melander, B. ( 1957). Personal communication. Melander, B., and Mlrtens, S. (1958). Diseases of Nervous System 19, 478. Melander, B., and Mlrtens, S. (1959). Acta Psychiat. Neurol. S c a d . 34, 344. Meyerhof, O., and Randall, L. 0. ( 1948). Arch. Biochem. 17, 171. Mitchell, S. W. (1896). Brit. Med. J. 2, 1625. Noval, J. J., Brande, B. L., and Sohler, A. (1959a). Federation Proc. 18, 428. Noval, J. J., Brande, B. L., and Sohler, A. ( 1959b). Unpublished results. Osinskaya, V. 0. (1957). Biochemistry (U.S.S.R.) 22, 496. (English ed. of Biokhimiya, 22, 537.) Osmond, H. (1955a). Diseases of Nervous System 16, 2. Osmond, H. ( 1955b). J. Mental Sci. 101, 526. Osmond, H. (1957). Ann. N . Y. Acad. Sci. 66, 418. Osmond, H. ( 1958). In “Chemical Concepts of Psychosis” (M. Rinkel and H. C. B. Denber, eds.), p. 3. McDowell, Obolensky, New York. Osmond, H., and Hoffer, A. (1958). J. Mental Sci. 104, 302. Osmond, H., and Hoffer, A. (1959). Can. Med. Assoc. J.80,Ql. Osmond, H., and Smythies, J. (1952). J. Mental. Sci. 98, 309. Paffenbarger, R. S., Steinmetz, C. H., Pooler, B. G., and Hyde, R. T. (1961). J . Chronic Diseases 13, 161. Park, J. H., Meriwether, B. P., and Park, C. R. (1956a). Fedemtion Proc. 15, 141.

370

A. HOFFER

l’ark, J. H., Meriwether, B. P., Park, C. R., Mudd, S. H., and Lipmann, F. ( 195613). Biochim. et Biophys. Acta, 22, 403. l’ayza, A. N., and Hoffer, A. (1959). Bull. fuc.d d . Istanbul, 22, 1096. Payza, A. N., and Mahon, M. E. (1959). Anal. C h m . 31, 1170. l’ayza, A. N., and Zaleschuk, J. (1959). Bull. fuc. w ’ d . Istanbul 22, 1523. Ileck, T. T. ( 1960). In “The Use of LSD in Psychotherapy” (H. A. Abramson, ed.), p. 89. Josiah Macy, Jr. Foundation, New York. l?ickworth, F. A. (1952). “New Outlook on Mental Disease.” Williams and Wilkins, Baltimore, Maryland. IRadmsa, W., and Golterman, H. L. (1954). Biochim. et Biophys. Actu 12, 80. Randall, L. 0. (1946). J. BWZ. Chem. 165,733. Rice, W. B., and McColl, J. D. (1957). Proc. Can. Physiol. SOC., Can. Rev. de Biol. 16, 507. Rice, W. B., and McColl, J. D. (1960). Arch. intern. phurmucodynumk 128, 249. Rigdon, R. H. (1940). Surgery 8, 839. Rinkel, M., Hyde, R. W., and Solomon, H. C. (1954). Diseases of Neruous System 15, 259. Ritzel, G., Staub, H., and Hunzinger, W. A. (1957). Germun Medical Monthly 2, 110. Roston, S . (1960). J. Biol. Chem. 235, 3315. Schwarz, B. E., Sem-Jacobsen, C. W., and Petersen, M. C. (1956a). A.M.A. Arch. Neurol. Psychiat. 75, 579. Schwarz, B. E., Wakim, K. G., Bickford, R. G., and Lichtenheld, F. R. (1956b). A.M.A. Arch. Neurol. Psychiat. 75, 83. Schwarzenbach, F. ( 1957). Second International Congress of Psychiatry, Zurich. Sem-Jacobsen, C. W., Petersen, M. C., Lazarte, J. A., Dodge, H. W., Jr., and Holman, C. B. (1955). Amm. J. Psychiat, 112, 278. Senoh, S., and Witkop, B. (1959). J. Am. Chem. SOC. 81,6231. Shenvood, S. ( 1957). Private demonstration. Slocombe, A. G. (1956). Federation Proc. 15, 172. Slocombe, A. G., Hoagland, H., and Tozian, L. S. (1956). Am. J. Physiol. 185, 601. Smythies, J. R. (1960). Lancet 1, 1287. Sohler, A., Noval, J. J., Stackhouse, S., and Byran, A. ( 1961). Federation ROC. 20, 173. Sommer, R., Dewar, R., and Osmond H. (1960). Arch. Gen. Psych&. 3, 665. Spector, S., Kuntzman, R., Shore, P. A,, and Brodie, B. B. (1960). J . Phurmucol. Exptl. Therap. 130, 256. Stern, P., Hukovi6, S., and Muacevic, G. (1956). Naturwissenschaften 43, 162. Stockings, G. T. (1940). J. Mental Sci. 86, 29. Stoll, A., and Hofmann, A. ( 1943). Helv. Chim. Acta 26, 944. Sulkowitch, H., and Altschule, M. D. (1959). A.M.A. Arch. Gen. Psychiot. 1, 108. Sulkowitch, H., Perrin, G. M., and Altschule, M. D. (1957). Proc. SOC. Exptl. BWZ. Med. 95, 245.

EFFECX OF ADRENOCHROME AND ADRENOLUTIN

371

Szatmari, A., Hoffer, A,, and Schneider, R. (1955). A,m. J. Psych&. 111, 603. Takahashi, Y., and Akabane, Y. (1960). Arch. Gen. Psychiut. 3, 674. Taubmann, G., and Jantz, H. (1957). Neroenurzt 20, 485. Utevsky, A. M., and Osinskaya, V. 0. (1957). Intern. Abstr. Biol. Sci. 6, 192. Vallbo, S. ( 1957). Personal communication. Veech, R. L., Altschule, M. D., Salkowitch, H., and Holliday, P. D. (1960). Arch. Gen. Psychiut. 3, 642. Walaas, E., and Jervell, K. F. (1958). Intern. Abstr. B i d . Sci. 10, 108. Walaas, E., and Walaas, 0. (1956). Biochim. ef, Biophys. Actu 20, 77. Walaszek, E. J. (1960). Intern. Reu. Neurobiol. 2, 138. Weckowicz, T. ( 1962). Ph.D. thesis ( Psychology), University of Saskatchewan, Saskatoon, Canada. Weil-Malherbe, H., Posner, H. S., and Bowles, G. R. ( 1961). Federation Pfoc. 20, 315. West, G. B., Shepherd, D. M., and Hunter, R. B. (1951). Lancet 2,966. West, G. B., Shepherd, D. M., and Hunter, R. B. (1953). Clin. Sci. 12, 317. Wiedorn, W. S. (1954). J. Neruoiis Mentul Diseuses 120, 1. Witt, P. N. (1954). Helo. Physiol. Acta. 12, 327. Witt, P. N. (1958). In “Psychopathology: A Source Book” ( C . F. Reed, ed.), p. 660. Harvard Univ. Press, Cambridge, Massachusetts. Witt, P. ( 1961). Personal communication. Woodford, V. R. ( 1959). Can. J. Biochem. und Physiol. 37, 873.

This Page Intentionally Left Blank

AUTHOR INDEX Numbers in italic indicate the pages on which the references are listed. Bain, J. A., 46, 67 Bakay, L., 142, 155 Abood, L. C., 219, 220, 239, 242, Balfour, W. M., 45, 68 245, 246, 256, 258, 259, 260, Baker, M., 129, 157 271, 272, 273, 315, 365 Ban, Y.,278, 303 Abramson, H. A., 315, 365 Barbeau, A., 364, 365 Abt, J. P., 62, 65 Barcroft, I., 184, 210 Adrian, E. D., 88, 108, 114 Barker, S. B., 298, 303 Adrih, H., 146, 158 Barrett, W., 168, 198, 214 Adson, A. W., 253, 273 Barron, D. H., 184, 210 Aird, R. B., 361,365 Bartlett, F., 308, 365 Akabane, Y., 310, 362, 371 Bartorelli, C., 103, 114 Akert, K., 99, 109, 114 B a d , H., 365 Albers, R. W., 45, 67 Battersby, A. R., 278, 303 Albrecht, M., 314,369 Baxter, C. F., 46, 47, 65, 68 Allen, M. L., 132, 141, 156 Raxter, D., 310, 362, 366 Alles, G. A., 365 Bear. R. S.. 240, 273 Alpert, M., 121, 157 Beck, E., 253, 273 Altschule, M. D., 310, 311, 365, 370, Behnsen, G., 141, 155 371 Behrens, M., 164, 210 Amin, A. H., 161, 162, 164, 172, 173, Beleslin, D., 193, 194, 202, 207, 210 174, 175, 176, 177, 182, 195, 210 Belknap, E. L., Jr., 139, 140, 156 Anderson, K., 359, 369 Bell, G. H., 310, 366 Andrews, T. M., 186, 210 Benaron, H. B. W., 121,156 Angelucci, L., 197, 210 Bente, D., 293, 303 Anokhin, P., 118, 119, 155 Bentley, G. A., 258, 273 Ansorg, W., 298, 304 Reran, V., 57, 65 Aprison, M. H., 41, 65, 185, 212 Bercel, N. A., 316, 366 Apter, N. S.,257, 258, 272 Bernow, B., 166,210 Apter, J. T., 97, 109, 114 Bertler, A., 288, 304 Arana, R., 332, 369 Besendorf, H., 283, 284, 285, 286, Aranow, H., 311, 367 291,292, 293,295, 305 Archibald, S.,224, 272 Bickford, R. G., 325, 326, 336, 370 Arvidsson, U. B., 166, 210 Biel, J. H., 219, 220, 221, 260, 272 Ashby, W., 138, 140, 155 Bieth, R., 130, 134, 135,157 Axiotis, A., 259, 273 Bishop, E. J., 146, 155 Bishop, C. H., 7, 65 B Bithos, Z. J., 278, 304 Bjurstedt, H., 170, 194, 210 Babuna, C., 121, 156 Blackman, J. C., 298, 303 Bacq, Z. M., 310, 333,365 Bachtold, H. P., 283, 284, 285, 286, Blair, M. R., 192, 193, 210 Blanc, C., 156 291,292,293,295,305 Blaschko, H., 310, 366 Bailey, P., 253, 272 373

A

374

AUTHOR INDEX

Bleuler, M., 308, 366 Bodin, J. I., 278, 304 Bogdanski, D. F., 283, 286, 287, 290, 295, 303 Bogen, J. E., 13, 23, 69 Boissonnas, R. A., 164, 166, 167, 169, 191, 195, 207, 211 Bolk, L., 86, 114 Bonnet, V., 113, 114 Bonta, I. L., 208, 211 Booker, H., 10, 14, 15, 17, 18, 21, 31, 32, 48, 51, 53, 68 Borowitz, J. L., 363, 366 Bosanquet, F. D., 181, 211 Bovet, D., 219, 221, 272 Bovet-Nitti, F., 219, 221, 272 Bowles, G. R., 363, 371 Brack, A., 308, 368 Bradley, P. B., 250, 272 Bradley, W., 143, 144, 155 Braines, C. H., 310, 333, 361, 366 Brande, B. L., 311, 321, 369 Brandrup, E., 300, 303 Brannick, L. J., 168, 198, 214 Braun-MenBndez, E., 169, 211 Bremer, F., 113, 114 Bridgman, C. S., 143, 157 Brink, F., 241, 272 Brink, F., Jr., 186, 212 Brinley, F. J., Jr., 29, 39, 40, 65 B,rizzee, K. R., 126, 127, 155 Brodie, B. B., 276, 281, 283, 285, 286, 287, 288, 290, 291, 295, 296, 297, 301, 303, 304, 305, 364, 370 Bsrossi, A., 276, 277, 278, 279, 304 B,roussolle, B., 320, 368 Bncher, V., 96, 98, 109, 110, 115 Buehler, K., 316, 366 B'iilbring, E., 181, 207, 211 Biirgi, S., 96, 98, 100, 108, 109, 110, 115 Bullough, W. S., 310, 315, 362, 366 Bures, J., 10, 11, 12, 15, 16, 17, 18, 20, 21, 27, 33, 39, 40, 44, 46, 47, 56, 57, 58, 59, 60, 61, 63, 65, 67, 143, 145,155

BureSovi, O., 9, 10, 11, 15, 16, 17, 18, 20, 21, 33, 44, 46, 56, 57, 58, 60, 63, 65 Burgstahler, A. W., 278, 304 Burton, R. M., 130, 138, 155 Bush, M. T., 369 Busnel, R. G., 293, 304 Byers, L. W., 359, 367 Byran, A., 311, 313, 331, 370

C Cabanac, J. L., 252, 272 Cahn, J., 291, 293, 295, 296, 304 Caldwell, P. C., 45, 65 Callbeck, M. J., 309, 368 Cammermeyer, J., 177, 211 Campion, D. S., 298, 303 Cannon, J. G., 238,272 Capek, R., 326,366 Carlsson, A., 288, 304 Carmichael, L., 143, 157 Carmichael, M. W., 142,157 Caspers, H., 198, 199, 200, 201, 211 Castellanos, G., 300, 305 Cattell, J. P., 218, 272 Chaikoff, I. L., 141, 156 Chambers, W. W., 89, 90, 96, 11.5, 116

Chang, J. J., 43, 69 Chappel, C. I., 224,272 Charles, M. S., 144, 155 Chatonnet, J. M., 252, 272 Chiquoine, A. D., 139, 155 Chopard-dit-Jean, L. H., 278, 279, 304 Christensen, E., 17, 19, 20, 21, 24, 69 Clare, M., 7, 65 Clark, B. B., 192, 193, 210 Clark, S. L., 87, 115, 251, 273 Clemente, C. D., 141, 156 Cleugh, J., 163, 211 Clouet, D. H., 129, 130, 131, 155 Coghill, G. E., 108, 115 Cohen, G., 313, 362, 366, 367 Cohen, M., 313, 335, 367,368 Cole, K. S., 8, 34, 35, 65 Colowick, S. P., 186, 212

375

AUTHOR INDEX

Contamin, F., 146, 157 Conway, E. J., 39, 65 Correale, P., 171, 211 Crain, S. M., 143, 145, 155 Crawford, T. B. B., 161, 162, 164, 172, 173, 174, 175, 176, 177, 182, 195, 210 Crepax, P., 198, 211 Crile, G. W., 42, 65 Crill, W. E., 35, 68 Crossland, J., 184, 213 Curtis, D. R., 16, 30, 47, 68 Curtis, H. J., 8, 34,35, 65 Cusick, J. T., 43, 69

D Dahlstedt, E., 170, 171, 178, 211 Dale, H. H., 187, 211 Dalgliesh, C. E., 164, 192, 211 Davidson, J. N., 310, 366 Davson, H., 41, 66 DeBoor, W., 219, 272 de Jong, H. H., 308, 366 de Landtsheer, L., 310, 312,366 Delgado, J. M. R., 253, 272 Demeester, G., 291, 292, 293, 295, 297, 304 Denber, H. C. B., 354,366 Denny-Brown, D., 86,115, 364, 366 DeRobertis, E. D. P., 42, 66, 185, 211 Derouaux, G., 314, 362, 366 Dewar, R., 356, 370 Diamond, L. K., 142,156 Dickerman, H., 140, 157 Dobri6, V., 196, 197, 214 Dodge, H. W., Jr., 337, 370 Doisy, E. A., 129, 135, 137, 157 Domer, F. R., 364, 366 Domino, E., 197, 213 Donaldson, H. H., 129, 130, 156 Douglas, W. W., 170, 191, 206, 21 1 Dow, R. S., 86, 88, 106, 115 Dresler, J. C., 300, 306 Dreyfus-Brisac, C., 144, 156 Drukker, A., 219, 221, 272 Dubroff, S. J., 19, 22, 67 Duncan, C. P., 62, 66

Dun&, H., 181, 185, 195, 204, 211 Dusser de Barenne, J. G., 10, 66

E Eade, N. R., 320, 361,366,368 Eason, R. G., 85,115 Eayers, J. T., 45, 66, 125, 156 Eber, O., 166, 188, 189, 190, 211 Eberhard, F., 298, 304 Eccles, J. C., 31, 35, 66, 86, 115, 187. 211 Eggleston, L. V., 16, 47, 67, 68 Ehemann, B., 144, 156 Ehrenpreis, T., 170, 184, 211 Ehringer, H., 185, 211 Eidelberg, E., 46, 47, 68 Eliasson, R., 167, 189, 206, 211 Ekes, J., 138, 156, 250, 272 Ellingson, R. J., 144, 146, 147, 156 Elliott, K. A. C., 125, 158 Ellis, H., 308, 366 English, D., 257, 272 Espinosa, J., 300, 304 Essig, C. F., 19, 22, 66, 67 Essman, W. B., 62, 65 Estes, W. E., 62, 66 Etsten, B., 143, 156 Evarts, E. V., 333, 366

F Faber, M., 311, 367 Farquhar, M. G., 41, 43, 66 Farr, A. L., 45, 67 Fastier, F. N., 298, 303 Faulkner, R. F., 85, 115 Fazekas, J. F., 144, 157 Feigen, G. A., 31, 69 Feigley, C. A., 359, 367 Fcldberg, W., 162, 170, 177, 178, 184, 186, 187, 191, 192, 206, 211, 258, 272, 327, 336, 364, 366 Feldstein, A., 354, 366 Fellman, J. H., 310, 363, 366 Ferguson, L. N., 233, 272 Fernlndez-Morln, H., 240, 272

376

AUTHOR INDEX

Ferster, C. B., 58, 66 FifkovA, E., 10, 11, 12, 23, 63, 65, 69 Finean, J. B., 240, 272 Finger, K. F., 304 Fink, M., 250, 272 Finkelstein, B. A,, 257, 272 Fischer, D., 310, 311, 312, 316, 331, 366, 368 Fischgold, H., 293, 294, 304 Fleming, T. C., 333, 366 Flexner, J. B., 131, 139, 140, 141, 156, 157 Flesner, L. B., 126, 131, 139, 140, 141, 143, 116, 147,156, 157 Florey, E., 169, 211 Forster, W., 293, 304 Fogel, S., 366 Folch-Pi, J., 121, 156 Foley, J. M., 310, 362, 366 Franchi, C. M., 185, 211 Frankel, S., 139, 157 Frankova, S., 321, 327, 337, 348, 354, 356, 367 ]?ram, J., 164, 166, 167, 191, 195, 207, 211 Frederic, J., 314, 362, 366 !Freedberg, A. S., 362, 368 Freygang, W. H., Jr., 8, 9, 35, 38, 66 Friede, R. L., 138, 156 Friedman, H. L., 219, 221, 272 Friedman, S. M., 191, 192, 211 Fries, B. A., 141, 156 Fuller, J. L., 144, 155 IWton, J. F., 2, 66

G Caddun, J. H., 160, 161, 162, 163, 164, 165, 166, 167, 168, 169, 172, 173, 174, 175, 176, 177, 179, 182, 190, 192, 194, 195, 203, 205, 206, 210, 211, 212, 215 Gaitonde, M. K., 129, 130, 131, 155 Galambos, R., 43, 66 (Gallant, L. J., 143, 146, 156 Garratt, S., 278, 303 Ca8parovi6, I., 166, 214 (Xty, T., 191, 213

Geiger, R. S., 315, 362, 364, 366 Gelfant, S., 315, 362, 366 Geoghegan, H., 39, 65 Gerard, R. W., 5, 13, 24, 30, 45, 62, 66, 67, 69, 311, 366 Gerber, C. F., 278, 304 Gernandt, B., 170, 193, 194,210,212 Gerschenfeld, H. M., 42, 66 Gershenovich, Z. S., 310, 367 Gershon, S., 257, 258, 272, 273 Gey, K. F., 281, 283, 285, 287, 288, 296, 298,299,304,305 Giorman, N. J., 291, 304 Gibbons, A. J., 310, 361, 367 Gibbs, E., 250, 272 Gibbs, F. A., 250,272 Gillespie, L., 333, 366 Girado, M., 28, 48, 68 Glees, P., 42, 66 Godlowski, W., 85, 115 Gokhblit, I. I., 144, 156 Goldenberg, M., 311, 367 Goldring, S., 5, 68, 146, 157 Golterman, H. L., 314, 362, 370 Golubieva, G. P., 310, 333, 361, 366 Goodhead, B., 45, 66, 125, 126, 156 Gouras, P., 12, 66 Grabner, K., 171, 172, 180, 181, 185, 194, 212 Grafstein, B., 7, 14, 15, 22, 24, 26, 27, 32, 33, 66 Grant, G. A,, 224, 272 Graul, E. H., 281, 305 Graves, J., 130, 134, 156 Graves, J. P., 129, 132, 133, 134, 135, 136, 140, 141, 142,156 Grazer, F. M., 141, 156 Green, H. D., 254, 272 Green, S., 311, 367 Greengard, P., 186, 212 Greig, M. E., 310, 361, 367 Grewal, R. S., 310, 367 Grontoft, O., 142, 156 Grof, S., 321, 327, 337, 348, 354, 356, 367 Grossman, C., 146, 156 Grossman, R. G., 85, 115

AUTHOR INDEX

377

Hill, F. S., 144, 156 Hilton, S. M., 198, 212 Himwich, H. E., 130, 134, 143, 144, 156,157, 185,212,250,273 Himwich, W. A., 121, 125, 129, 130, 132, 133, 134, 135, 136, 140, 141, 142, 144,156,157 H Hinsey, J. C., 83, 115 Hack, M. H., 332, 369 Hoagland, H., 335, 370 Hoch, P. H., 218, 272 Halpern, B. N., 293, 304 Hampson, J. L., 74, 88, 115 Hochstein, P., 313, 362, 366, 367 Hankinson, J., 11, 66 Hodgkin, A. L., 35, 38, 45, 65, 66 Hannett, I. I., 85, 100, 108, 115 Hoff, E. C., 254, 272 Hare, W. K., 95, 115, 116 Hoffer, A,, 246, 273, 300, 306, 309, Harman, P. J., 139, 157 310, 311, 312, 313, 314, 315, Harrison, C. R., 74, 88, 115 320, 336, 337, 338, 339, 349, Hart, E. R., 314, 362, 367 354, 359, 361, 363, 366, 367, Hartmann, J. F., 41, 43, 66 368, 369, 370, 371 Hassler, R., 76, 79, 110, 111, 112, 115 Hofmann, A., 308, 368, 370 Hastings, A. B., 41, 67 Hohensee, F., 208, 21 1 Hatai, S., 129, 130, 156 Holasek, A., 179, 213 Haug, H., 180, 212 Holliday, P. D., 311, 371 Heacock, R. A,, 309, 311, 317, 332, Holman, C. B., 337, 370 IIolton, F. A., 181, 195, 212 357, 367 Heath, R. G., 312, 313, 335, 336, Holton, P., 163, 186, 190, 195, 210, 359, 367, 368 211,212 Hebb, C . O., 139, 140, 156, 177, 178, Holtz, P., 314, 362, 368 186, 212 Hooper, N. K., 43,69 Heim, R., 308,368 Horner, J., 219, 221, 272 Heise, G. A., 293, 296, 304 Hornykiewicz, O., 185, 211 Heinze, H., 300, 304 Horsley, V., 86, 116 Held, H., 43, 66 Horstmann, E., 41, 42, 66 Hellauer, H., 177, 188, 189, 190, 212, Horton, E. W., 169, 212 213 Hosmer, H. R., 42, 65 Hellauer, H. F., 186, 187, 188, 189, Hoya, W. K., 221,272 190, 212,214 Hryntschak, T., 184, 212 Hellerbach, J., 278, 304 Hughes, J. G., 144, 156 Hems, R., 16, 68 Hukovib, S., 196, 205, 206, 208, 212, Herken, H., 184, 212 214, 314, 370 Hultman, E. H., 184, 192, 212 Hernandez, P., 129, 157 Herold, M., 291, 293, 295, 296, 304 Hundshagen, H., 281, 305 Hess, A., 126, 156 Hunsicker, A,, 259, 273 IIess, W. R., 72, 76, 79, 80, 96, 98, Hunt, K., 10, 15, 17, 18, 21, 24, 25, 100, 102, 109, 110, 114, 115 26, 27, 28, 29, 30, 31, 32, 46, Hierowski, M., 133, 158 47, 48, 53, 68 Hild, W., 43, 66 Hunt, W. E., 146, 157 Hill, D. K., 38, 66 Hunter, J., 83, 115

Criissner, A., 278, 304 Crundfest, H., 28, 48, 68 Cullbring, B., 166, 212 Gurman, F., 203, 213 Guttmann, St., 169, 211 Cyarfas, K., 56, 66, 68

:378

AUTHOR INDEX

Hunter, R. B., 363, 371 Hunzinger, W. A., 365, 370 Hurd, D., 259, 273 Hutcheon, D. E., 361, 368 Hyde, J. E., 85, 115 IHyde, R. T., 365,369 Hyde, R. W., 353,370 IHydin, H., 180, 212

I Ihanaga, K., 39, 69 Ihchiose, M. A., Jr., 362, 368 Ihfantellini, F., 198, 211 Ingram, W. R., 85, 100, 108,115 Innes, I. R., 193, 212 lordanis, K. A., 310, 334, 360,368 Ishibasi, K., 278, 305 l~toh,N., 278, 304 Iwama, K., 28, 29, 67

J ],ackman, A., 259, 273 Jacobs, L. A., 126, 127, 155 Jacobs, R. J., 45, 68 Jacquenoud, P. A., 169, 211 Jacques, R., 164, 215 Jaeger, E., 278, 304 Jamieson, J. D., 191, 192, 211 ],ansen,J., 113, 115 ][ansen,J,, Jr., 113, 115 J/antz,H., 311, 313, 354, 368, 371 ]a&, M. E., 62, 65, 218, 273 Jasper, H. H., 2, 10, 19, 21, 24, 28, 29, 30, 67, 68, 83, 115, 143, 157 Iervell, K. F., 363, 371 Ilohnson, A. C., 186, 212 Jordan, W. K., 147,157 lung, R., 79, 110, 111, 112,115

K Kaada, B. R., 254, 273 Kaiser, I., 143, 144, 155 Kalnitsky, G., 310, 367 Kandel, E. R., 29, 39, 40, 65 Katzman, R. K., 141, 157 Kaufmann, H., 311, 368

Kavaler, F., 139, 157, 184, 213 Kehl, R., 169, 215 Kellaway, P., 147, 157 Kelley, B., 134, 157 Kenyon, M., 311, 312, 368 Kety, S . , 360, 368 Keynes, R. D., 45, 65, 66 Killiam, K. F., 46, 67 Kimel, V. M., 139,157, 184, 213 Kimitsuki, M., 135, 157 King, R. B., 9, 14, 70 Kisch, B., 310, 368 Kissel, J. W., 197, 213 Klein, J. R., 44, 68 Kluchesky, E. F., 272 Kliiver, H., 218, 247, 273, 308, 368 Knoll, B., 293, 304 Knoll, J., 293, 304 Kobel, H., 308, 368 Kobrinckaio, 0. J., 310, 333, 361, 366 Koch, E., 311, 368 Koch, M. L., 129, 157 Koch, W., 129,157 Kocib-Mitrovii., D., 181, 185, 196, 197,209, 213,214 Koebke, K., 169,215 Koella, W. P., 87, 90, 91, 93, 105, 113, 114, 115, 116 Koelle, G. B., 310, 368 Konev, C. B., 310, 333, 361, 366 Konzett, H., 169, 205, 213 Koketsu, K., 267, 268, 273 Kopera, H., 172, 174, 177, 178, 213 Korzoff, B. A., 310, 313, 362, 368 Kogak, R., 196, 212 Kosterlitz, H. W., 193, 212 KovaE, J., 166, 214 Koyama, K., 267, 268, 273 Krebs, H. A., 16, 47, 67, 68 Krichevskaya, A. A,, 310, 367 Kriendler, A., 143, 144, 157 KiivLnek, J., 14, 16, 39, 40, 44, 46, 65, 67 Krivoy, W. A., 166, 190, 201, 202, 213 Krupp, P., 324, 335, 368 Kubie, L. S., 123, 124, 158

AUTHOR INDEX

379

Lcusen, I., 291, 292, 293, 295, 296, 297, 304 Levine, S., 121, 157 Levy, C. K., 113, 116 Lewandowski, M., 87, 116 Lewin, L., 308, 369 Lewis, G. P., 193, 213 1 Libet, B., 5, 24, 30, 66, 67, 144, 157 Laborit, H., 320, 368 Libonati, M., 191, 213 Lacroix, E., 291, 292, 293, 295, 297, Lichtenheld, F. R., 325, 326, 336, 370 304 Liddell, E. C. T., 86, 115 Laidlow, B. D., 309, 367 Lie, L., 167, 183, 211 Landau, W. M., 8, 9, 35, 38, 66 Liebl, G. J., 138, 157 Landefeld, M., 19, 67 Liljedahl, S. O., 192, 194, 213 Langemann, H., 310, 368 Lim, R. K. S., 203, 213 Langfitt, T. W., 314, 362, 367 Lin, R. C. Y., 184, 207, 211 Langworthy, 0. R., 253, 273 Lindlar, H., 276, 277, 278, 304 Lashley, K. S., 61, 67 Ling, G., 45, 67 Laszlo, I., 162, 188, 213 Linn, L., 365, 369 Laubach, G. D., 278,304 Lipman, V., 224, 251, 273 Lipmann, F., 314, 362, 370 La Velle, A., 126, 157 Lazarini, W., 172, 174, 177, 178, 213 Lishajko, F., 162, 170, 171, 178, 181, Lazarte, J. A., 337, 370 182, 211, 215 Leach, B. E., 312, 313, 335, 339, Livingston, R. B., 253, 272 367, 368 Lloyd, D. P. C., 201, 213 Leach, E., 161, 163, 211, 213 Loeser, J., 113, 116 LeEo, A. A. P., 2, 3, 4, 5, 7 , 8 , 12, Liiwenthal, M., 86, 116 13, 14, 15, 23, 27, 47, 53, 67 Loewi, O., 177, 186,213 Lebovitz, €3. Z., 224, 248, 273 Lombardino, J. C., 278, 304 Lechner, H., 165, 198,200, 203, 213 Lowenthal, J., 361, 368 Lecomte, J., 310, 311, 312, 316, 331, Lowry, 0. H., 45, 67 Ludhy, G., 191, 213 366, 368 Lukenbill, A., 41, 65 Lehmann, A., 293, 304 Lehrmann, D. S., 319, 369 Luse, S. A., 42, 43, 67 Leiderman, P. H. L., 141, 157 M Leimdorfer, A., 332, 333, 369 MacArthur, C. G., 129, 135, 137, 157 Leiner, K. Y., 45, 67 McColl, J. D., 326, 327, 370 Leiser, H. A,, 219, 221, 272 Lembeck, F., 163, 164, 165, 166, McCormack, J. I., 39, 65 170, 171, 172, 177, 178, 179, McCulloch, W. S., 10, 66 180, 181, 183, 184, 185, 187, McDonald, J. V., 88, 89, 116 188, 189, 190, 194, 195, 198, McIlwain, H., 148, 157 200, 202, 203, 205, 207, 211, Macintosh, I?. C . , 213 McIntyre, A. K., 201, 213 212, 213 McLamore, M. W., 278, 304 Lende, N., 300, 304 McLardy, T., 253, 273 Lessin, A. W., 293, 304 MacLean, D., 63, 67 LettrB, R., 314, 369

Kuchino, E. B., 310, 313, 333, 334, 360, 361, 362, 366,368 Kuhlman, R. E., 45, 67 Kuntzman, R., 364, 370 Kunze, M., 293, 304

i380

AUTHOR INDEX

IMcNabb, A. R., 186, 212 IMagnus, R., 113, 116 IMagoun, H. W., 95,115,116 Magun, R., 17, 67 Mahon, M., 309, 367 Mahon, M. E., 311, 370 ,Malcolm, J. L., 146, 157 IMandel, P., 130, 134, 135, 157 IManery, J. F., 41, 67 IMann, T., 177, 211 Manukian, K. G., 134, 142,157 March, R., 147, 157 Idarcus, D., 273 IMarinesco, G., 143, 144, 157 IMarrazzi, A., 314, 362, 369 IMarrazzi, A. S . , 314, 362, 367 Ivlarsala, J., 12, 65 Marshall, W. H., 2, 3, 9, 13, 15, 17, 18, 19, 22, 24, 29, 31, 39, 40, 45, 65, 66, 67, 68 IMBrtens, S., 313, 326, 334, 359, 360, 367, 368, 369 Martins, Ferreira, H., 7, 12, 67 Marty, R., 146, 157 Maslova, A. F., 311, 369 Mathis, P., 293, 294, 304 Mattsson, O., 192, 194, 213 Matussek, N., 164, 166, 167, 168, 213 May, R. M., 131, 157 Maynard, E. A., 41, 68 Ivlazel, P., 369 Mazur, A., 311, 367 Medakovik, M., 194, 213 IMeduna, L. J., 246, 256, 272, 273 IMeirowsky, E., 310, 325, 369 Melander, B., 312, 326, 334, 359, 360, 369 Meriwether, B. P., 314,362, 369,370 !Merlis, S., 354, 366 IMerritt, J. H., 363, 366 Messing, R. A., 147, 157 Metzner, W. R. T., 332, 369 Meves, H., 41, 42, 66 Meyer, A. E., 253, 273 Meyer, W., 74, 116 Meyerhof, O., 313, 362, 369 IMietkiewski, E., 202, 213

Milin, R., 196, 214 Miller, L. D., 299, 304 Millichap, J. G., 129, 138, 157 Milner, P. M., 3, 67 Minz, B., 185, 215, 294, 305 Misirlija, A., 196, 214 Mitchell, S. W., 308, 369 Monnier, M., 100, 116, 324, 335, 368 Monod, N., 156 Morison, R. S., 8, 13, 15, 23, 27, 47, 67 Morrell, F., 143, 144, 155 Morruzzi, G., 86, 115, 116 Muacevic, G., 205, 214, 314, 370 Mudd, S. H., 314, 362, 370 Myers, R. E., 58, 61, 67

N Nachmansohn, D., 138, 139, 140,157 Naidoo, D., 138, 157 Naike, Y., 277, 305 Namin, P., 11, 66 Nelson, E., 143, 144, 155 Nemeth, A. M., 140, 157 Nerenberg, C., 309, 317, 367 Neubert, D., 184, 212 Neuhold, K., 171, 172, 212 Nieto, D., 305, 305 Nilsson, J., 288, 304 Noval, J. J., 311, 313, 321, 331, 369, 370 Nuhfer, P. A., 272

0 Ochs, S., 6, 7, 9, 10, 11, 13, 14, 15, 17, 18, 21, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 38, 42, 46, 47, 48, 49, 50, 51, 53, 58, 60, 61, 62, 66, 67, 68, 69 Oeconomos, D., 143, 146, 158 Ostlund, E., 170, 171, 178, 192, 211, 215 Ohnesorge, G., 161, 171, 182, 208, 209, 215 Olds, J., 57, 68 O’Leary, J. L., 5, 9, 14, 68, 70 Olin, J. S . , 284, 286, 287, 305

381

AUTHOR INDEX

Olsen, N. S., 44, 68 Olszewski, J., 142, 158 Openshaw, H. T., 278, 303 Orlans, F. B., 304 Orrego, F., 35, 69 Osbond, J., 278, 279, 305 Osinskaya, V. O., 311, 369, 371 Osmond, H., 218, 246, 273, 308, 309, 310, 313, 320, 336, 337, 356, 359, 368,369,370 Ostfeld, A. M., 219, 220, 224, 248, 260, 272, 273

P Paasonen, M. K., 172, 173, 174, 176, 182, 208, 209, 213, 288, 289, 290, 305 Paffenbarger, R. S., 365, 369 Page, I. H., 169, 211 Palay, S. L., 38, 68 Palmer, K. J., 240, 273 Paquette, R., 224, 272 Park, C. R., 314, 362, 369, 370 Park, J. H., 314, 362, 369, 370 Parkes, M. W., 293, 304 Parlow, A. F., 168, 198, 214 Paton, W. D. M., 170, 191, 206, 211 Payza, A. N., 309, 311, 314, 317, 363, 367, 368, 370 Pease, D. C., 41, 68 Peck, T. T., 364, 370 Pennes, H. H., 218,272 Pentzik, A. S., 127, 144, 157 Pernow, B., 163, 164, 165, 166, 167, 169, 170, 172, 173, 177, 181, 182, 183, 184, 185, 189, 190, 191, 192, 193, 194, 195, 196, 203, 204, 205, 206, 211, 213, 215 Perrimond-Trouchet, R., 320, 368 Perrin, G. M., 311, 370 Peters, V. B., 126, 157 Petersen, J. C., 121, 125, 123, 130, 132, 133, 134, 135, 136, 140. 141, 142, 144,156,157 Petersen, M. C., 337, 354, 370

Petschke, B., 163, 170, 172, 183, 184, 205, 207, 213 Peyrot, R., 141, 142, 158 Pfeiffer, C. C., 46, 68, 220, 233, 238, 273 Phillis, J. W., 30, 66 Philpot, F. J., 181, 211 Pickworth, F. A., 311, 370 Pletscher, A., 276, 281, 283, 284, 285, 286, 287, 288, 289, 290, 291, 292, 293, 294, 295, 296, 298, 299, 303,304,305,306 Polidora, V., 245, 273 Pollack, G . H., 56, 68 Polley, E. H., 35, 69 Pollock, G. H., 56, 66 Pompeiano, O., 86, 116 Pooler, B. G., 365, 369 Posner, H. S., 363, 371 Potter, G. D., 203, 213 Potter, V. R., 138, 157 Pratt, 0. E., 138, 157 Press, G. D., 41, 68 Priii., R., 208, 213 Purpura, D. P., 28, 48, 68, 142, 157 Putnam, T. J., 177, 215

Q Quadbeck, G., 294, 305 Quinn, G. P., 281, 283, 285, 288, 291, 296,297, 301,304,305

R Radmanovih, B., 202, 210 Radmanovi6,B. Z., 202, 210 Radmanovii., R., 194, 213 Radmsa, W., 314, 362, 370 Ramony Cajal, S., 126, 157 Ranck, J. B., 35, 68 Ranck, J. B., Jr., 35, 68 Randall, L. O., 313, 314, 362, 369, 370 Ranson, S. W., 85, 95, 100, 108, 115, 116, 251, 273 Rapoport, S. I., 39, 68

382

AUTHOR INDEX

Scherrer, J. 143, 146, 157, 158 Schild, H., 161, 166, 167, 190, 212 Schild, H. O., 205, 214 Schlosser, L., 162, 166, 167, 168, 172, 173, 174, 175, 176, 178,215 Schlossmann, H., 310, 366 Schmitt, F. O., 240, 273 Schneider, R., 336, 349, 354, 371 Schneider, W. C., 138, 157 Schnider, O., 276, 277, 278, 279, 304 Schueler, F. W., 220, 233, 273 Schultz, R. L., 41, 68 Schuster, E. M., 138, 140, 155 Schwartz, D. E., 281, 287, 288,305 Schwarz, B. E., 325, 326, 336, 337, 354, 370 Schwarzenbach, F., 315, 370 Scott, B. D., 309,367 Segar, W. E., 41, 65 Segre, G., 191, 213 Sem-Jacobsen, C. W., 337, 354, 370 Senoh, S., 311, 370 Settlage, P. H., 74, 116 Seydl, G., 164, 210 Shanes, A., 241, 266,273 Shaw, F. H., 258, 273 Shaw, T. I., 45, 65 Shenkin, H. A., 253, 273 Shepherd, D. M., 363,371 Sherrington, C. S., 86, 116 Shenvood, S., 326, 327,336,366,370 Sholl, D. A., 34, 48, 68 Shore, P. A., 276, 281, 283, 284, 285, 286, 287, 288, 291, 296, 297, Sager, O., 143, 144, 157 301,303,305,364,370 !Samson, D., 156 Shorr, E., 311, 367 Samson, F. E., 45, 68 Sidman, M., 305 Santibhiiez, G., 146, 158 Silver, A,, 177, 178, 212 Sattes, H., 300, 305 Silver, S. L., 276, 305 !Saul, L. J., 66 Sjoerdsma, A., 333, 366 !Sawyer, C. H., 186, 214 Skeggs, L. T., 169, 214 !Scarborough, H., 310, 366 Skinner, B. F., 58, 66 !3chachter, M., 170, 191, 206, 211 !
llenson, J., 333, 365 Rice, W. B., 326, 327, 370 Richter, C. P., 253, 273 Richter, D., 184, 213 llieder, J., 281, 287, 288, 305 lliese, W., 125, 127, 157 lligdon, R. H., 311, 325, 370 llig6, J., 191, 213 llinaldi, F., 239, 242, 250, 272, 273 llinkel, M., 353, 370 llitzel, G., 365, 370 Roberts, E., 46, 47, 65, 68, 139, 157 Roberts, N. R., 45, 67 Roberts, R. B., 141, 157 Robinson, J. A., 193, 212 Robinson, J. R., 38, 68 llocha e Silva, M., 169, 191, 192, 213, 214 lloitback, A. I., 2, 68 Rose, J. E., 146, 158 Hosengren, E., 288, 304 llosett, J., 137, 158 lloskam, J., 314, 362, 366 lloss, J., 17, 67 Rossiter, R. J., 186, 212 Roston, S., 310, 362, 370 llowland, A. F., 42, 65 llozdilsky, B., 142, 158 lluch, T. C., 253, 273 llumley, M., 130, 131, 158 llussell, I. S., 18, 58, 60, 61, 62, 68

s

383

AUTHOR INDEX

Smith, M. E,, 300, 305 Smythies, J., 309, 336, 337, 368, 369 Smyhes, J. R., 360, 370 Snider, R. S., 88, 106, 116 Sohler, A., 311, 313, 321, 331, 369, 370

Solomon, H. C., 353, 370 Sommer, R., 356, 370 Sonnenschein, R. R., 9, 39, 68, 69 Spaziani, E., 41, 66 Spector, S., 364, 370 Sperry, R. W., 58, 68 Sperry, W. M., 130, 158 Sprague, J. M., 90, 96, 116 Sprengeler, E. P., 219, 221, 272 Stackhouse, S., 311, 313, 331, 370 Stamm, J. S., 5, 8, 12, 14, 15, 17 19, 20, 21, 22, 24, 33, 44, 53, 55, 69 Staub, H., 365, 370 Stein, S. M., 56, 66 Stein, S. N., 56, 68 Steiner, F., 293, 294, 305 Steinmetz, C. H., 365,369 Stern, J. R., 16, 68 Stern, L., 141, 142, 158 Stem, P., 166, 181, 192, 196, 197,

Swedin, B., 166, 210 Sweet, W. H., 253,272 Swingle, W. W., 168, 198, 214 Szab6, H., 191, 213 Szatmari, A., 336, 349, 354, 371

T

Tainter, M. L., 299, 304 Takahashi, Y., 310, 362, 371 Tanche, M., 252, 272 Tani, C., 278, 305 Tasaki, I., 35, 43, 69 Taubmann, G., 354, 371 Taylor, J. L., 39, 69 Terashima, M., 278, 303 Terner, C., 47, 67 Terwilliger, E., 85, 100, 108, 115 Thiele, F. H., 100, 116 Tilney, F., 122, 123, 124, 137, 158 Tobias, J. M., 38, 69 Todrick, A., 138, 156 Toh, C. C., 162, 164, 170, 192, 211 Tourlentes, T., 259, 273 Tower, D. B., 41, 42, 46, 47, 69, 125, 158 Tozian, L. S., 335, 370 198, 199, 200, 203, 205, 206, Travis, R. P., 57, 68 208, 209, 211, 212, 214, 314, 370 Trendelenburg, U., 202, 214 Stockhausen, F. G., 300, 305 Trufant, S. A., 9, 14, 70 Stockings, G. T., 308, 370 Tscherter, H., 308, 368 Stoll, A., 308, 370 Tschirgi, R. D., 39, 69 Stowell, A., 88, 106, 116 Tucker, B. E., 121, 156 Stoyanoff, V. A., 130, 158 Tyler, D. B., 14.3, 146, 156 Streicher, E., 41, 68 Stripe, M. C., 121, 156 U Studer, A., 298, 306 Udenfriend, S., 283, 286, 287, 288, Stunner, E., 164, 166, 167, 169, 191, 290, 303, 305 195, 207, 211, 213 Ulett, G. A,, 300, 305 Stumpf, W., 281, 300, 305 Umrath, K., 186, 187, 188, 189, 190, Sugar, O., 13, 69 212, 214 Sugasawa, S., 278, 304 Utevsky, A. M., 311, 371 Sugita., N., 45, 69 Uzman, L., 130, 131, 158 Sugita, N. J., 126, 127, 158 Sulkowitch, H., 311, 370, 371 V Sulser, F., 295, 303, 304 Vallbo, S., 321, 325, 326, 334, 359, Summerson, W. H., 298, 303 360, 369, 371 Suzuta, Y.,277, 305

384

AUTHOR INDEX

van Harreveld, A., 5, 6, 7, 8, 9, 12, 13, 14, 15, 16, 17, 19, 20, 21, 22, 23, 24, 31, 33, 34, 35, 36, 37, 38, 42, 43, 44, 46, 53, 55, 69 Varagie, V., 193, 194, 202, 207, 210 Vassilon, G., 259, 273 Veech, R. L., 311, 371 Visotsky, H., 273 Visotsky, H. M., 224, 248, 273 Vitek, V., 321, 326, 327, 337, 348, 354,356,366,367 Voelkel, A., 293, 294, 300, 305, 306 Vogt, M., 172, 173, 174, 176, 177, 178, 182, 186, 187, 208, 209, 211, 213, 214 Vogt, W., 167, 168, 214 Vojtechovsk?, M., 321, 326, 327, 337, 348, 354,356,366,367 von Brauchitsch, H., 300, 306 von Euler, U. S., 160, 161, 162, 163, 164, 165, 167, 170, 171, 172, 178, 181, 182, 184, 185, 190, 192, 194, 196, 203, 205, 206, 210, 211, 215 von Holst, E., 113, 116 von Muralt, A., 186, 215 Voris, H. C., 253, 273 Vuk6evi6, S . , 192, 203, 214

W Waelsch, H., 130, 158 Waites, G. M. H., 186, 212 Wakim, K. G., 325, 326, 336,370 Walaas, E., 314, 362, 363, 371 Walaas, O., 314, 362, 371 Walaszek, E. J., 166, 185, 201, 214, 215, 294, 305, 325, 371 Wald, F., 42, 66 Walker, R. M., 9, 39, 68, 69 Waller, W. H., 85,116 Walter, M., 276, 277, 278, 304 Ward, T., 300, 306 Watkins, J. C., 16, 30, 47, 66 Weckowicz, T., 322, 361, 371

Weckowicz, T. E., 300, 306 W&l-Malherbe, H., 363, 371 Weiss, T., 10, 11, 23, 63, 65, 69 Weissbach, H., 288, 305 Weisschedel, E., 115 Weisscheddel, 76, 79 Wender, M., 133, 158 Werle, E., 169, 215 West, G. B., 363, 371 Westermann, E., 314, 362, 368 Whieldon, J. A,, 20, 53, 69 Wiedorn, W. S., 365, 371 Wijmenga, H. G., 208,211 Wikler, A., 249, 273 Wilcott, R. C., 146, 156 Wike, G., 298, 304 Wilzbach, K. E., 242, 273 Winokur, G. L., 9, 14, 70 Wislocki, G. B., 177, 215 Witkin, L. B., 19, 67 Witkop, B., 311, 370 Witt, P., 316, 371 Witt, P. N., 316, 371 Wood, H. C. S., 278,303 Woodford, V. R., 314, 362, 371 Woohey, C. N., 74, 88,115,116 Work, T. S., 164, 192, 211 Wortis, R. P., 319, 369 WU, M.-L., 45, 67 Wursch, J., 278, 304 Wyckoff, R. W. G., 41, 70 Wyss, F., 186, 215 Wyss, 0. A. M., 103, 114

Y Yamazaki, T. M., 277, 306 Yonemitsu, O., 278, 303 Young, J. Z., 41, 70

Z Zachar, J., 14, 19, 20, 22, 70 ZacharovL, D., 14, 19, 20, 22, 33, 65, 70 Zadunaisky, J. A., 42, 66

AUTHOR INDEX

Ztihorovi, A., 18, 56, 65 Zaleschuk, J., 314, 370 Zbinden, G., 298, 306 Zeiss, F. R., 85, 100, 108, 115 Zeller, P., 296, 305

385

Zetler, G., 161, 162, 164, 165, 166, 167, 168, 171, 172, 173, 174, 175, 176, 178, 179, 182, 196, 204,207, 208,209,215 Zuber, H., 164, 215

SUBJECT INDEX A Adrenochrome biochemistry of, 311314 action of on cells, 314, 315 modes of, 3 6 1 4 6 5 potentiation of, 359 effects of on electrograms, 33-37 on fish, 315 on humans prolonged reactions, 348-352 reported by other laboratories, 353-357 on mammals cats, 325333

psychotominiectic potency, 224-226 on rats behavioral, 244, 245 central nervous system (CNS), 224

B Benzoquinolizine derivatives chemistry carbinols, ccondary and tertiary, 278-280 ketones, 278 actions clinical, 300 on monamine metabolism central nervous system, 282dogs, 333, 334 mice, 320 286 monkeys, 334,335 mechanism, 288-291 metabolites, urinary, 288 rabbits, 324, 325 organs, peripheral, 287, 288 rats, 3 2 0 3 2 4 pharmacological, effects on monaon pigeons, 316-320 mine metabolism and, 300302 on spiders, 316 sublingually given, 357359 metabolism, 281 Adrenolutin tissue distribution and excretion, 281 effects on humans phmacology double blind experiments, 337348 cardiovascular system, 296, 297 gastrointestinal tract, 297, 298 prolonged reactions, 352, 353 Anticholinergic psychotomimectic nervous system, 291-296 agents toxicity, 299 in the neonatal period Brain chemical nature, 221 behavior, genesis of effects chemical constituents, accumulation drug action, factors affecting of, 129-137 acid group, strength of, 238 enzymatic activity, development of, cationic charge, 238 138-140 with phospholipids’ 238-241 hematoencephalic exchange, 141, spatial-geometric, 232, 233 142 steric hindrance, 233-238 neuroanatomical development, 125on humans 129 behavioral, 245-247 neurophysiological development, psychological studies, 248, 249 142, 147 386

387

SUBJECT INDEX

electroencephalogram ( EEG ) , spontaneous, 143-145 evoked potentials, 145-147

C Cortex higher functions in, see Spreading Depression learning trace, transfer of, 61 and subcortical relations, 63 Cortical layers upper, and, release of lower, 47-56 Cortical response, direct differential effect on, 47-52

D Drug receptor site, 219-221

G Glycolate esters, see Piperidyl glycolates

M Monamine decreasing drugs, see B m zoquinolizine derivatives

N Neural networks spreading depression in, see Spreading depression

P Piperazine esters of glycolic acid derivatives . structure activity studies with, 227-

232 Piperazinoalkylglycolates see Piperidyl glycolates Piperidyl glycolates antagonists, 254, 255 biochemical and electrophysiological studies biochemical effects, mechanism, 264-266 biochemical-functional relationships, 271

enzyme systems, effect on, 260, 261 frog nerve and muscle, metabolic effects, 261-264 frog sartorius muscle, 267 central mechanism of action electrophysiological studies, 249, 250 hyperactivity, relation to other CNS functions, 2.51 temperature regulation centers and activity, 251-254 glycolate esters, clinical studies, 256-260 tetrahydroaminoacridine ( THA ) , antagonist t o Ditran, 258, 259 piperazinoalkylglycolates, clinical studies, 259, 260 psychotomimetic, neurological-autonomic effects, 247, 248 radioactive labeled, 241-244 structure-activity relationships, 226227 Polypeptide, see Substance P Psychotomimetic (definition), 218, 219 Psychotomimetic agents, see Anticholinergic psychotomimectic agents

S Spreading convulsion, 52-58 Spreading depression cellular changes during intercellular components, release of, 39-41 intercellular space and glia, 41-44 ions and water, entry into cells, 33-39 in chronic preparations, 17-19 conditioned responses, effect on, 56-61 and cortex, higher functions in, 56-63 excitation of modes of, 14-17 modification of, 19-20

388

SUBJECT INDEX

metabolic events connected with, 44-47 in nonneocortical tissue, 11, 12 phenomonological aspects asphyxia1 change, 12-14 EEG changes, 3-5 electrical impedance, 7, 8 cortical responses, 7 steady potential changes, 5-7 subcortical effects, 9-11 vascular changes, 8, 9 propagation of characteristics of, 22, 23 mechanism of spread, 23-27 transmission, contiguity theory of, 2733 refractoriness and recovery time, 20-22 Subcortical motor areas midbrain tectum, 96-100 motor effects cerebellum stimulation, 86-96 diencephalon stimulation, 76-86 motor organization, 71-76 tegmental reaction, 100, 101 Substance P chemistry

characteristics, 167, 168 differentiation, 168-170 enzymatic destruction, 165-167 estimation, 162, 163 extraction, 161, 162 purification, 163-165 distribution in the organism central nervous system, 170-181 gastrointestinal tract, 170 peripheral nervous system, 181, 182 organ function and tissue concentration degenerating nerve, 185, 186 intestine, 183, 184 nervous system, 184, 185 pharmacological actions circulation, 194, 195 intestine, 191-194 nervous system, 195-203 tachyphylaxis, 203, 204 pharmacological interactions central nervous system, 207-209 intestine, 205-207 and “transmitter substance of sensory nerves,” 186, 191